Treatment of Cancer
Fifth Edition
Edited by
Pat Price MA MD FRCP FRCR Professor of Radiation Oncology Academic Depar...
51 downloads
1875 Views
20MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Treatment of Cancer
Fifth Edition
Edited by
Pat Price MA MD FRCP FRCR Professor of Radiation Oncology Academic Department of Radiation Oncology School of Cancer and Imaging Sciences Christie Hospital and University of Manchester Manchester, UK
Karol Sikora MB PhD FRCP MRCR Medical Director Cancer Partners UK and Visiting Professor of Cancer Medicine Imperial College Hammersmith Hospital London, UK
Tim Illidge BSc PhD MRCP FRCR Professor of Targeted Therapy and Oncology School of Cancer and Imaging Sciences Christie Hospital and University of Manchester Manchester, UK
PART OF HACHETTE LIVRE UK
First published in Great Britain in 1982 by Chapman & Hall This second edition published in 2008 by Hodder Arnold, an imprint of Hodder Education, part of Hachette Livre UK, 338 Euston Road, London NW1 3BH http://www.hoddereducation.com
©
2008 Edward Arnold (Publishers) Ltd
All rights reserved. Apart from any use permitted under UK copyright law, this publication may only be reproduced, stored or transmitted, in any form, or by any means with prior permission in writing of the publishers or in the case of reprographic production in accordance with the terms of licences issued by the Copyright Licensing Agency. In the United Kingdom such licences are issued by the Copyright Licensing Agency: Saffron House, 6–10 Kirby Street, London EC1N 8TS. Whilst the advice and information in this book are believed to be true and accurate at the date of going to press, neither the author[s] nor the publisher can accept any legal responsibility or liability for any errors or omissions that may be made. In particular (but without limiting the generality of the preceding disclaimer) every effort has been made to check drug dosages; however it is still possible that errors have been missed. Furthermore, dosage schedules are constantly being revised and new side-effects recognized. For these reasons the reader is strongly urged to consult the drug companies’ printed instructions before administering any of the drugs recommended in this book. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress ISBN-13 978 0 340 912218 1 2 3 4 5 6 7 8 9 10 Commissioning Editor: Gavin Jamieson Project Editor: Francesca Naish Production Controller: Andre Sim Cover Designer: Helen Townson Typeset in 10/12 pt Minion by Charon Tec Ltd (A Macmillan Company) Chennai, India Printed in India
What do you think about this book? Or any other Hodder Arnold title? Please visit our website: www.hoddereducation.com
Contents
Colour plates appear between pages 728 and 729 Contributors
vi
Foreword
xiv
Preface
xv
List of abbreviations used
xvi
Evidence scoring
xxiii
Reference annotation
xxiii
PART ONE: PRINCIPLES 1 2 3 4 5 6 7 8 9 10 11 12 13
Introduction Karol Sikora Molecular Biology Hani Gabra, Euan Stronach, Rohini Sharma Clinical radiobiology Michele I Saunders, Stanley Dische Mathematical modelling and its application in oncology Roger G Dale, Bleddyn Jones Principles of chemotherapy and drug development Jeffry Evans, Meenali M Chitnis, Denis C Talbot Tumour imaging in oncology Anju Sahdev, Rodney H Reznek Interventional radiology Tarun Sabharwal, Anne P Hemingway, Andreas Adam Vaccination strategies for malignant diseases Deepak P Assudani, Stephanie McArdle, Murrium Ahmad, Geng Li, Robert C Rees, Selman A Ali Biological therapies: cytokines and adoptive cell therapy Fiona C Thistlethwaite, Peter L Stern, Robert E Hawkins Radioimmunotherapy Tim Illidge, Mike Bayne Monoclonal antibodies and treatment of cancer Hossein Borghaei, Liat Binyamin, Igor Astsaturov, Louis M Weiner Angiogenesis as a target for the treatment of cancer Srinivasan Madhusudan, Daniel Patterson, Adrian L Harris Gene therapy Kate Relph, Kevin Harrington, Hardev S Pandha
1 3 23 40 58 75 112 146 167 186 203 226 251 269
PART TWO: PRACTICE
285
14
287
Central nervous system Roy Rampling
iv Contents
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
41 42
Ocular and adnexal tumours John L Hungerford, P Nicholas Plowman Head and neck cancer Alastair J Munro, Nicholas D Stafford Thyroid Masud Haq, Clive Harmer Endocrine cancer Maria Gueorguiev, Ashley B Grossman, P Nicholas Plowman Breast cancer Navita Somaiah, John Yarnold Bronchus Vanessa A Potter, Nicholas Thatcher, Penella J Woll Oesophageal cancer David B Smith, Brian J Haylock Liver Daniel H Palmer, Philip J Johnson, Pancreas Patrick G Johnston, Martin M Eatock Biliary tract Hemant M Kocher, Ajit T Abraham, Satya Bhattacharya Stomach Marianne C Nicolson, Kenneth GM Park, Leslie M Samuel Bladder cancer Robert Huddart, Paula Wells, Alan Horwich Prostate cancer Malcolm Mason Colorectal cancer Richard HJ Begent, Astrid Mayer, Christopher H Collis, Russell N Moule Anus Bernard J Cummings Germ-cell cancers of the testis and related neoplasms Graham M Mead Renal cell cancer Tom R Geldart Ovary and fallopian tubes Daniela D Rosa, Andrew R Clamp, Gordon C Jayson Uterus Bleddyn Jones Cervix Peter Blake Carcinoma of the vagina and vulva Sadaf Ghaem-Maghami, W Pat Soutter Gestational trophoblastic tumours Peter Schmid, Michael J Seckl Non-melanoma skin cancer Stephen L Morris, Sean Whittaker, Margaret Spittle Malignant melanoma Susana Banerjee, Martin Gore Bone Jeremy S Whelan, Rob C Pollock, Anna M Cassoni Soft tissue sarcomas Thomas F DeLaney, Andrew E Rosenberg, David C Harmon, Frances J Hornicek, Sam Yoon, David G Kirsch, Henry J Mankin, Daniel Rosenthal Leukaemias Tariq I Mughal, John M Goldman Hodgkin’s lymphoma Beate Klimm, Andreas Engert
320 343 408 438 468 502 539 559 573 585 602 614 645 666 695 706 729 748 764 777 796 809 835 879 905 925
981 1027
Contents v
43 44 45 46 47
Non-Hodgkin’s lymphoma Adrian J C Bloor, David C Linch Multiple myeloma Jennifer M Bird, Cathy Williams Introduction to haematopoietic stem cell transplantation Effie Liakopoulou, David I Marks Paediatric oncology Stephen Lowis, Eddy Estlin, Keith Sibson, AIDS-related malignancy Mark Bower, Tom Newsom-Davis
1045 1073 1098 1118 1167
PART THREE: MANAGEMENT
1187
48
1189
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 Index
Medical care Clare E Green, Richard H J Begent Interstitial brachytherapy J Roger Owen, David Bottomley Principles of external beam radiotherapy planning techniques Stephen L Morris, H Jane Dobbs, Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy Vincent Khoo Concomitant chemo-radiotherapy principles and management Syed A Hussain, Anjali Zarkar, Nicholas D James Palliative care Anne Naysmith, Karol Sikora Communications with the cancer patient Justin Stebbing, Maurice L Slevin Clinical cancer genetics Lisa J Walker, Ros A Eeles Large-scale randomized evidence: trials and overviews Richard Gray, Rory Collins, Richard Peto, Keith Wheatley The economics of cancer care Nick Bosanquet Medical audit Amit K Bahl, Gareth J G Rees, The organisation of cancer services: a UK perspective Ian Kunkler New drug development George Blackledge Late effects of cancer therapy Susan E Davidson Surgical oncology David Merrilees, David Neal Cancer prevention strategies William P Steward, Andreas Gescher
1210 1233 1254 1280 1290 1305 1318 1338 1352 1365 1377 1407 1415 1423 1433 1445
Contributors
Ajit T Abraham MS MA FRCS Consultant Surgeon Hepato-biliary and Pancreatic Surgery Unit The Royal London Hospital London, UK
Mike Bayne BSc MBChB MRCP FRCP DM Consultant Clinical Oncologist Dorset Cancer Centre, Poole Dorset, UK
Andy Adam MBBS (Hons) FRCP FRCS FRCR FFRRCSI (Hon) Professor of Interventional Radiology Department of Radiology St Thomas’ Hospital King’s College London London, UK
Richard H J Begent MD FRCP FRCR FMedSci Department of Oncology Royal Free and University College Medical School University College London London, UK
Murrium Ahmad BSc MSc PhD Post-Doctoral Research Fellow/ Research Project Manager School of Science and Technology Nottingham Trent University Nottingham, UK
Satya Bhattacharya MS MPhil FRCS Consultant Hepato-biliary Surgeon Hepato-biliary and Pancreatic Surgery Unit The Royal London Hospital London, UK
Selman A Ali PhD BVM&S Senior Lecturer School of Science and Technology Nottingham Trent University Nottingham, UK
Liat Binyamin PhD Postdoc Associate Fox Chase Cancer Centre Philadelphia, USA
Deepak Assudani MBBS MSc PhD Post-Doctoral Research Fellow Moffitt Cancer Centre and Research Institute Florida, USA Igor Astsaturov Fox Chase Cancer Centre Philadelphia, USA Amit K Bahl MD MRCP FRCR FFR RCSI Consultant in Clinical Oncology Bristol Haematology and Oncology Centre Bristol, UK Susana Banerjee MA MBBS MRCP Clinical Fellow The Breakthrough Breast Cancer Research Centre at the Institute of Cancer Research Chester Beatty Laboratories London, UK
Jennifer M Bird MD FRCP FRCPaTH Avon Haematology Unit Bristol Haematology and Oncology Centre Bristol, UK Peter Blake MD FRCR Consultant Clinical Oncologist The Royal Marsden Hospital London, UK George Blackledge Hospira UK Ltd Warwickshire, UK
MB MD PhD FRCP
Adrian J C Bloor PhD MRCP MRCPath Consultant Haematologist Christie Hospital Manchester, UK
Contributors vii
Hossein Borghaei DO MS Medical Oncology Fox Chase Cancer Centre Philadelphia, USA
Susan E Davidson MD FRCR FRCP(Glas) BW Consultant and Honorary Senior Lecturer in Clinical Oncology Christie Hospital NHS Foundation Trust Manchester, UK
Nick Bosanquet BA Msc Professor of Health Policy Imperial College London, UK
Thomas F DeLaney MD Associate Professor of Radiation Oncology Harvard Medical School Associate Radiation Oncologist Department of Radiation Oncology Medical Director Francis H Burr Proton Therapy Center Massachusetts General Hospital Boston, USA
David Bottomley MBBS MRCP FRCR Consultant Clinical Oncologist Cookridge Hospital Leeds, UK Mark Bower PhD FRCP FRCPath Consultant Medical Oncologist Chelsea and Westminster Hospital London, UK Anna M Cassoni BSc FRCR FRCP Consultant Clinical Oncologist University College Hospital London, UK Meenali M Chitnis BmedSci MBBS MRCP Specialist Registrar in Medical Oncology Churchill Hospital Oxford, UK Andrew R Clamp MRCP PhD Senior Lecturer and Honorary Consultant in Medical Oncology Cancer Research UK Department of Medical Oncology Christie Hospital Manchester, UK Rory Collins FRCP FMedSci Co-director, Clinical Trial Service Unit (CTSU) University of Oxford Oxford, UK Christopher H Collis MB BChir Clinical Oncologist Royal Free and University College Medical School London, UK Bernard J Cummings MBChB FRCPC FRCR FRANZCR Department of Radiation Oncology Princess Margaret Hospital University of Toronto Toronto Ontario, Canada Roger G Dale MSc PhD FlnstP FIPEM FRCR (Hon) Director Department of Radiation Physics and Radiobiology Professor of Cancer Radiobiology Hammersmith Hospitals NHS Trust Faculty of Medicine Imperial College London, UK
Stanley Dische MD FRCR FACR Marie Curie Research Wing Centre for Cancer Treatment Mount Vernon Hospital Middlesex, UK H Jane Dobbs MA FRCP FRCR Consultant in Clinical Oncology Department of Clinical Oncology Guy’s and St. Thomas’ NHS Foundation Trust London, UK Martin M Eatock MbChB FRCP(Lond) FRCP(Glasg) Consultant/Honorary Senior Lecturer in Medical Oncology Northern Ireland Cancer Centre Belfast City Hospital Belfast, UK Ros A Eeles MA MRCP FRCP PhD Translational Cancer Genetics Team Institute of Cancer Research and Cancer Genetics Unit Royal Marsden NHS Foundation Trust Sutton and London, UK Andreas Engert MD Professor for Internal Medicine Hematology and Oncology University Hospital of Cologne Department of Internal Medicine I Cologne, Germany Eddy Estlin BSc(Hons) PhD MRCP(UK) FRCPH Macmillan Consultant in Paediatric Oncology Department of Paediatric Oncology Royal Manchester Children’s Hospital Manchester, UK Jeffry Evans MBBS MD MRCP(UK) FRCP(Glasgow) FRCP(London) Professor of Translational Cancer Research University of Glasgow Honorary Consultant Medical Oncologist Beatson West of Scotland Cancer Centre Honorary Group Leader Beatson Institute for Cancer Research Glasgow, UK
viii Contributors
Hani Gabra PhD FRCPE FRCP Professor of Medical Oncology Ovarian Cancer Action Research Unit and Section of Molecular Therapeutics Department of Oncology, Imperial College London, UK
Masud Haq BSc MRCP MD Locum Consultant in Diabetes & Endocrinology Chelsea & Westminster Hospital NHS Foundation Trust London, UK
Tom R Geldart BSc MBBS MRCP DM Consultant in Medical Oncology Poole and Royal Bournemouth Hospital NHS Tusts Dorset, UK
Clive Harmer FRCP FRCR Consultant Oncologist Former Head of the Thyroid Unit Royal Marsden Hospital London, UK
Andreas Gescher BSc PhD DSc Professor of Biochemical Toxicology Leicester Royal Infirmary Leicester, UK Sadaf Ghaem-Maghami PhD MRCOG Senior Lecturer and Honorary Consultant in Gynaecological Oncology Imperial College Faculty of Medicine Division of Surgery, Oncology, Reproduction and Anaesthetics Department of Reproductive Biology Hammersmith Hospitals Trust London, UK John M Goldman DM FRCPath Hematology Branch, National Heart Lung and Blood Institute National Institutes of Health Bethesda, USA Martin Gore PhD FRCP Professor of Cancer Medicine The Royal Marsden Hospital London, UK Richard Gray MA MSc Professor of Medical Statistics Director, Birmingham Clinical Trials Unit Robert Aitken Institute University of Birmingham Birmingham, UK Clare E Green Bsc MRCP MBBS Department of Oncology Southampton General Hospital Southampton, UK Ashley B Grossman BA BSc MD FRCP FMedSci Department of Endocrinology St Bartholomew’s Hospital London, UK Maria Gueorguiev MD Department of Endocrinology St. Bartholomew’s Hospital London, UK
David C Harmon MD Massachusetts General Hospital Division of Hematology & Medical Oncology Boston, USA Kevin Harrington FRCP FRCR The Institute of Cancer Research Royal Marsden Hospital London, UK Adrian L Harris BSc Hons MB ChB MA DPhil FRCP Director Cancer Research UK Molecular Oncology Laboratories Weatherall Institute of Molecular Medicine John Radcliffe Hospital Oxford, UK Robert E Hawkins MBBS PhD FRCP Professor and Director of Medical Oncology Cancer Research UK Department of Medical Oncology Christie Hospital NHS Foundation Trust and University of Manchester Manchester, UK Brian J Haylock BSc Hons Pharm MB MRCP DRCOG FRCR FRCP(C) Consultant Clinical Oncologist Clatterbridge Centre for Oncology Bebington Wirral, UK Anne P Hemingway BSc MBBS MCRP FRCR FRCP Department of Diagnostic Radiology Hammersmith Hospital Imperial College NHS Trust London, UK Francis Hornicek MD PhD Department of Orthopedics Massachusetts General Hospital Boston, USA
Contributors ix
Alan Horwich PhD MRCP FRCR Professor of Radiotherapy Academic Radiotherapy Unit The Institute of Cancer Research The Royal Marsden Hospital Surrey, UK Robert Huddart PhD MRCP FRCR Reader in Urological Oncology Academic Radiotherapy Unit The Institute of Cancer Research The Royal Marsden Hospital Surrey, UK John L Hungerford Ocular Oncology Service Department of Ophthalmology St Bartholomew’s Hospital London, UK Syed Hussain Clinical Lecturer in Oncology Cancer Research UK Institute for Cancer Studies University Hospital Birmingham Birmingham, UK Tim Illidge BSc PhD MRCP FRCR Professor of Targeted Therapy and Oncology School of Cancer and Imaging Sciences Christie Hospital and University of Manchester Manchester, UK Nicholas D James SSc MBBS FRCP FRCR PhD Professor of Clinical Oncology Cancer Research UK Institute for Cancer Studies University of Birmingham Birmingham, UK Gordon C Jayson FRCP PhD Professor of Medical Oncology Cancer Research UK Department of Medical Oncology Christie Hospital Manchester, UK Philip J Johnson MD FRCP Clinical Head Professor of Oncology Director, Clinical Trials Unit Institute of Cancer Studies The University of Birmingham Birmingham, UK Patrick Johnston Centre for Cancer Research & Cell Biology Queen’s University Belfast University Floor Belfast City Hospital Belfast, UK
Bleddyn Jones MA MSc MD FRCP FRCR med FIPEM FBIR Consultant in Oncology and Applied Radiobiology University Hospital Birmingham Birmingham, UK Vincent Khoo MBBS FRACR FRCR MD Consultant & Honorary Senior Lecturer in Clinical Oncology Royal Marsden Hospital London, UK David G Kirsch MD PhD Department of Radiation Oncology Duke University Medical Center North Carolina, USA Beate Klimm, MD Department I of Internal Medicine University Hospital Cologne Cologne, Germany Hemant M Kocher MS MD FRCS Department of Health National Clinician Scientist Senior Lecturer and Hon Consultant Surgeon Tumour Biology Laboratory Cancer Research UK Clinical Centre Barts and The London School of Medicine and Dentistry London, UK Ian Kunkler MA MBBChir DMRT FRCR FRCPE FRSA Consultant Clinical Oncologist Edinburgh Cancer Centre Western General Hospital University of Edinburgh Edinburgh, UK Geng Li, MBBS, MSc, PhD Senior Clinical Scientist Central Labs Alconbury Cambridgeshire, UK Effie Liakopoulou Consultant in Haematology and Stem Cell Transplantation Christie Hospital NHS Trust Honorary Senior Lecturer in Cancer Studies University of Manchester Manchester, UK David C Linch FRCP FRCPath FMedSci Professor of Haematology Royal Free and University College Medical School London, UK
x Contributors
Stephen Lowis BA (Hons) BM BCh Phd MRCP(UK) FRCPCH Macmillan Consultant in Paediatric and Adolescent Oncology Department of Paediatric Oncology Bristol Royal Hospital for Children Bristol, UK Srinivasan Madhusudan MBBS MRCP PhD Clinical Associate Professor and Consultant in Medical Oncolgy School of Molecular Medical Sciences Academic Unit of Oncology University of Nottingham Nottingham, UK Henry J Mankin MD Orthopaedic Oncology Massachusetts General Hospital Boston, USA David I Marks Consultant in Haematology and Stem Cell Transplantation Director Adult Bone Marrow Transplant Unit United Bristol Health Trust Honorary Clinical Reader in Haematopoietic Cell Transplantation University of Bristol Bristol, UK Malcolm Mason MD FRCP FRCR Cancer Research Wales Professor of Clinical Oncology Department of Oncology and Palliative Medicine Cardiff University Velindre Hospital Cardiff, UK Astrid Mayer PhD Senior Lecturer in Medical Oncology Hampstead Site Royal Free and University College Medical School London, UK Stephanie McArdle BSc PhD Senior Post-Doctoral Research Fellow School of Science and Technology Nottingham Trent University Nottingham, UK Graham M Mead DM FRCP FRCR Department of Medical Oncology Royal South Hants Hospital Southampton, UK David Merrilees FRACS (Urol) Auckland Hospital Auckland New Zealand
Stephen L Morris MBBS MRCP FRCR MacMillan Consultant Clinical Oncologist Skin Tumour Unit St Johns Institute of Dermatology St Thomas’ Hospital Clinical Oncology Department London, UK Russell N Moule Department of Oncology Royal Free and University College Medical School University College London London, UK Tariq I Mughal MB MD MRCP MRCS FACP CRC Department of Medical Oncology Royal Preston Hospital Preston, UK Alastair J Munro Tayside Cancer Centre NHS Tayside Ninewells Hospital Dundee, UK Anne Naysmith FRCP Pembridge Unit St. Charles Hospital London, UK David Neal FMed Sci MS FRCS Professor of Surgical Oncology Honorary Consultant Urological Surgeon University of Cambridge/Addenbrooke’s Hospital Cambridge, UK Tom Newsom-Davis BSc MRCP Specialist Registrar Medical Oncology Department of Oncology Hammersmith Hospitals NHS Trust London, UK Marianne C Nicolson BSc MD FRCP ANCHOR Unit – Clinic D Aberdeen Royal Infirmary Aberdeen, UK J Roger Owen MBBS FRCP FRCR Cheltenham General Hospital Cheltenham, UK Daniel H Palmer BSc MBChB MRCP PhD Cancer Research UK Clinician Scientist and Honorary Consultant in Medical Oncology Cancer Research UK Institute for Cancer Studies University of Birmingham Birmingham, UK
Contributors xi
Hardev Pandha FRCP FRACP PhD Department of Oncology Postgraduate Medical School University of Surrey Guildford, UK Kenneth GM Park MB ChB MD FRCS (Ed) Consultant Surgeon/Honorary Senior Lecturer Aberdeen Royal Infirmary Aberdeen, UK Sir Richard Peto FRS Co-director, Clinical Trial Service Unit (CTSU) University of Oxford Oxford, UK P Nicholas Plowman MA MD FRCP FRCR Ocular Oncology Service Department of Ophthalmology St Bartholomew’s Hospital London, UK Rob Pollock FRCS Consultant Orthopaedic Surgeon Royal National Orthopaedic Hospital Stanmore Middlesex, UK Vanessa A Potter BSc MRCP PhD Consultant Medical Oncologist Nottingham City Hospital Nottingham, UK Pat Price MA MD FRCP FRCR Professor of Radiation Oncology Academic Department of Radiation Oncology School of Cancer and Imaging Sciences Christie Hospital and University of Manchester Manchester, UK Roy Rampling PhD FRCR FRCP MBBS Professor of Neuro-Oncology Beatson West of Scotland Cancer Centre Glasgow, UK Gareth J G Rees FRCP FRCR Consultant in Clinical Oncology Bristol Haematology and Oncology Centre Bristol, UK Robert Rees BSc PhD Dean of School School of Science and Technology Nottingham Trent University Nottingham, UK Kate Relph BSc PhD Department of Oncology Postgraduate Medical School University of Surrey Guildford, UK
Rodney Reznek Academic Department of Radiology St Bartholomew’s Hospital London, UK Daniela D Rosa MD, PhD Medical Oncologist Cancer Research UK Department of Medical Oncology Christie Hospital Manchester, UK Andrew E Rosenberg MD Department of Pathology – WRN 2 Massachusetts General Hospital Boston, USA Daniel Rosenthal Harvard Medical School Boston, USA Tarun Sabharwal MBBCH FRCSI FRCR Consultant Interventional Radiologist And Honorary Senior Lecturer Guy’s and St Thomas’ Hospital London, UK Anju Sahdev MBBS MRCP FRCR Consultant Radiologist St Bartholomew’s Hopsital Barts and the London Trust London, UK Denis C Talbot BSc MB Bchir MA PhD FRCP Consultant Medical Oncologist and Reader in Cancer Medicine Cancer Research UK Experimental Cancer Centre University of Oxford Oxford, UK Leslie M Samuel BSc MSc FRCP (Edin) FRCR (Lond) Macmillan Consultant Oncologist ANCHOR Unit Aberdeen Royal Infirmary Aberdeen, UK Michele I Saunders MD FRCP FRCR Windeyer Professor of Oncology University College London, UK Peter Schmid MD PhD Consultant Medical Oncologist and Senior Clinical Lecturer Director, Hammersmith Early Clinical Trials Unit Charing Cross Campus, Imperial College NHS Trust Imperial College London London, UK
xii Contributors
Michael J Seckl PhD FRCP Head of Section of Molecular Oncolgy Director of the Charing Cross Gestational Trophoblastic Disease Centre Charing Cross Campus of Imperial College NHS Trust London, UK Rohini Sharma FRACP Clinical Fellow Ovarian Cancer Action Research Unit and Section of Molecular Therapeutics Department of Oncology Imperial College London, UK Keith Sibson MB ChB MRCP Haematology Department University College Hospital London, UK Karol Sikora MB PhD FRCP FRCR Medical Director Cancer Partners UK and Visiting Professor of Cancer Medicine Imperial College Hammersmith Hospital London, UK Maurice L Slevin MD FRCP Department of Medical Oncology St Bartholomew’s Hospital London, UK David B Smith MB FRCP Consultant in Medical Oncology Clatterbridge Centre for Oncology Merseyside, UK Navita Somaiah MBBS MD FRCR Clinical Research Fellow Academic Radiotherapy Department Royal Marsden Hospital and Institute of Cancer Research Surrey, UK
Nicholas D Stafford MB FRCS Professor of Otolaryngology & Head and Neck Surgery Postgraduate Medical Institute University of Hull Hull Royal Infirmary Hull, UK Justin Stebbing Imperial College School of Medicine Department of Medical Oncology Hammersmith Hospital London, UK William P Steward MB ChB PhD FRCPC FRCP(Lon, Gla) Head of Department of Cancer Studies and Molecular Medicine University of Leicester Leicester Royal Infirmary Leicester, UK Peter L Stern BSc PhD Head of Cancer Research UK Immunology Group Paterson Institute for Cancer Research University of Manchester Manchester, UK Euan Stronach PhD Senior Research Associate Ovarian Cancer Action Research Unit and Section of Molecular Therapeutics Department of Oncology, Imperial College London London, UK Herman D Suit MD DPhil Department of Radiation Oncology Massachusetts General Hospital Boston, USA Denis C Talbot ICRF Medical Oncology Unit The Churchill Hospital Oxford, UK Fiona C Thistlethwaite MB BChir PhD MRCP Clinical Fellow Cancer Research UK Department of Medical Oncology Christie Hospital NHS Foundation Trust Manchester, UK
W Pat Soutter MSc MD FRCOG Honorary Reader in Gyanecological Oncology Department of Obstetrics and Gynaecology Imperial College Faculty of Medicine Hammersmith Hospital London, UK
Nicholas Thatcher Professor of Medical Oncology CRC Department of Medical Oncology Christie CRC Research Centre Christie Hospital NHS Trust Manchester, UK
Margaret Spittle MSc FRCR FRCP Consultant Clinical Oncologist Meyerstein Institute of Oncology The Middlesex Hospital London, UK
Lisa J Walker MRCPCH Dphill Consultant in Cancer Genetics Department of Clinical Genetics The Churchill Hospital Oxford, UK
Contributors xiii
Louis M Weiner MD Vice President, Translational Research Chairman, Department of Medical Oncology G Morris Dorrance, Jr Endowed Chair in Medical Science Fox Chase Cancer Center Philadelphia, USA Paula Wells PhD MRCP FRCR Consultant Clinical Oncologist Department of Radiotherapy St Bartholomew’s Hospital London, UK Keith Wheatley DPhil Professor of Medical Statistics Director, Birmingham Clinical Trials Unit Robert Aitken Institute University of Birmingham Birmingham, UK Jeremy Whelan MD FRCP Consultant Medical Oncologist University College Hospital London, UK Sean Whittaker MD FRCP Consultant Dermatologist Skin Tumour Unit St Johns Institute of Dermatology St Thomas’ Hospital London, UK
Cathy Williams MRCP FRCPath Consultant Haematologist Centre for Clinical Haematology Nottingham University Hospitals Nottingham, UK Penella J Woll MBBS PhD FRCP Professor of Medical Oncology Weston Park Hospital Sheffield, UK John Yarnold MBBS BSc MRCP FRCR Professor of Clinical Oncologist and Honorary Consultant Academic Radiotherapy Department Royal Marsden Hospital Surrey, UK Sam Yoon MD Division of Surgical Oncology Massachusetts General Hospital Harvard Medical School Boston, USA Anjalia Zarkar MBBS MRCP MD FRCR Consultant Clinical Oncologist The Cancer Centre University Hospital Birmingham Birmingham, UK
Foreword
HISTORICAL FOREWORD Cancer has been present since time immemorial, before man himself had developed; sarcoma have been seen in bones of dinosaurs and of our predecessor Pithecanthropus erectus. We have feared cancer especially after it was first recognized by Hippocrates (c. 460–370 BC), and named by Galen of Pergamon (AD 129–216). Surgery was the natural first treatment, and was usually extensive and often mutilating, even more so after anaesthesia came into use in the nineteenth century; but unfortunately for the patient, recurrence was frequent. The many drug and medicines available were of little avail, even when caustics were applied. Then X-rays were discovered in 1895, and radium in 1898, and soon each began to be used with some success. Nevertheless, the energy of the rays was low, in kilovolts only, and their use in cancer treatment was of lasting benefit against no more than basal and squamous cell skin cancers. Radium, however, was soon found to have remarkable value when inserted into tumours of the uterus and then implanted into superficial and accessible cancers. Replication of treatments necessarily required definition and measurement of units – the ro˝ntgen, rad and gray for dose; and the curie and becquerel for activity. Methods and results were steadily improved up to the 1930s. Therapy units had then become available to produce X-rays of energy as high as 500 kV, and two of these were constructed in series to make the first millionvolt machine, installed in St. Bartholomew’s Hospital in 1938; ‘Megavoltage X-ray therapy’ had begun, but only in one privileged centre. The Second World War intervened for six long years, from 1939 to 1945, with peace coming in the Far East only after the atomic bombs were dropped. Artificial radioactive isotopes then became freely available, not only those such as iodine-131 and phosphorus-32 but more importantly cobalt-60, emitting megavoltage γ-rays. Sources were constructed with activities of tens and hundreds of curies, far more than even the largest radium ‘bombs’. These were, first of all, placed into teleradium housings with great benefit, and soon into purpose-built ‘cobalt units’, operating at 80 or 100 cm source-skin distances. In 1952, an even better unit
arrived – the linear accelerator, operating at 4 Mev ore more. This was truly revolutionary – with the help of wedge filters and good planning one could now treat deep tumours anywhere within the body; there was also no skin reaction, which had long been an unpleasant side-effect limiting prescribed dosage. The ‘LA’, as it became known, had also been achieved as a result of wartime research – at the heart of the linear accelerator was a high-power electron source, the magnetron, which had been developed at the Malvern Telecommunications and Radar Establishment for use in radar equipment. Within a few years megavoltage treatment became available everywhere. One other revolutionary discovery had also come from the war: nitrogen mustard had been shown to be sensationally effective against Hodgkin’s disease about 1948, and then bleomycin in 1972 against squamous carcinoma. More drugs followed, together with more hormones, particularly tamoxifen. They began to be used in combinations, and chemotherapy joined surgery and radiotherapy as another ally against cancer. The first edition of Treatment of Cancer was published in 1982; at this time it was sorely needed. After the Second World War, two texts had been published, from David Smithers in London in 1946 describing X-ray therapy, and Ralston Paterson in Manchester in 1948 (second edition in 1963) comprehensively covering both X-ray and radium treatment. These were followed in 1955 by Rock Carling, Windeyer and Smithers’ British Practice in Radiotherapy, and Ronald Raven’s Cancer in 1960. By the 1970s these were beginning to become out of date. I began planning my own book in 1977. The first edition was well received and a second edition was requested after I had retired and had begun work for 2 years in Hong Kong. Karol Sikora kindly took over as joint editor for the second edition, published in 1990, bringing in molecular biology, oncogenes, sensitizers, hypothermia and AIDS. Pat Price joined him as editor for the third edition in 1995, by which time cost-efficiency, genetics and audit needed further space. Yet more is to come. I am looking forward with much pleasure to the publication of the fifth edition. Keith Halman, FRSH
Preface
Cancer is an increasingly common problem; by the year 2020, it is likely that one in two of the global population will develop the disease at some time. This striking increase is mainly due to an increasingly ageing population. The treatment of cancer is therefore a vital component of modern healthcare provision. Oncology crosses the traditional boundaries in which medicine is taught, researched and practiced, thereby providing a great challenge for those involved in its treatment. It is a truly multidisciplinary subject. One of the current dogmas, almost a cliché, is the importance of integration and teamwork between those involved in its management: surgeons, radiotherapists, medical oncologists and a myriad of other organ-based specialists. But the most exciting discoveries of the past two decades have been in the laboratory. Our understanding of the process of growth control and its deregulation has gone through a remarkable transformation, with the discovery of oncogenes, tumour suppressor genes and their encoded products. Coupled with the tremendous advances in imaging and the likely explosion in our understanding of the genome, we are set for a dramatic improvement in cancer therapy over the next decade. This book attempts to synthesize in one place scientific discoveries together with the best of current practice. Tumours of all kinds, common and esoteric, are considered so that authoritative help on how bet to manage an individual patient can be found rapidly. Our authors come from many disciplines, to paint as broad a canvas as possible of oncological knowledge. Many are associated in some way with Hammersmith Hospital, but we have also invited colleagues from the UK and abroad to share with us their special expertise. In many cases, we have teamed a surgeon with an oncologist to give a complete overview of the management of an individual tumour type. The book is divided into three parts. The first section considers the principles of cancer treatment and the likely developments that will take place over the next decade. The contributors here are active participants in current research programmes in their area. The second and main
part – practice – is an extensive series of chapters covering every individual class of tumour. They represent the distillation of considerable experience from senior clinicians. For many tumours, such as early carcinoma of the larynx, treatment has become standardized as well as successful. For others, such as breast cancer, there is considerable controversy. We have given more space to those tumours where controversy reigns, for it is here that difficulties in management are most likely to arise. We have used some diplomatic editing, with the agreement of the authors, to make their management plans widely applicable. In this way, we hope the book will provide sufficient guidance to point the way to the best contemporary treatment protocols for individual patient care throughout the world. The final section – management – gives further essential detail on the general problems of caring for the cancer patient. Here we consider medical problems that commonly arise, and also the development of continuing care philosophies, which have revolutionized the management of those patients for whom no cure is yet available. We consider radiotherapy planning techniques as well as new developments in conformal planning. We review medical audit and clinical trials, both essential tools of progress in an emerging discipline. We also consider complementary medicine and cancer, for the first time, we believe, in an orthodox oncology text. We have been greatly helped by the expertise and writing skills of our authors who have produced truly excellent manuscripts. We hope that this unique, multiauthor text will continue to educate a new generation of oncologists at a time of considerable progress. Delivering optimal care requires far more teamwork than ever before and the integration of many specialities. The success of modern cancer treatment is transforming the lives of many of our patients. We hope that this contribution will help to do this. Pat Price, Christie Hospital, Manchester Karol Sikora, Hammersmith Hospital and Cancer Partners UK, London
List of abbreviations used
AAF 17-AAG AAVs ABC ABVD AC ACE ACNU ACT ACTH ACTION ADC ADDC ADEPT AEs AF AFP AIM AIN AIs AJCC ALA ALL ALP AML AP APC APCs APL APT ara-CTP ARCON ASCO ASCT ASH ASTEC AT ATG
IFN-alpha-activated factor 17-allylaminogel-danamycin adeno-associated viruses Adjuvant Breast Cancer trial doxorubicin, bleomycin, vinblastine, dacarbazine Audit Commission doxorubicin, cyclophosphamide Adult Comorbidity Evaluation 1-(4-amino-2-methylpyrimidine-5-yl)methyl3-(2-chloroethyl)-3-nitrosurea adoptive cell transfer adrenocorticotrophic hormone Adjuvant Cytotoxic Chemotherapy In Older Women (trial) apparent diffusion coefficient antibody dependent directed cytotoxicity antibody-directed enzyme prodrug therapy adverse events accelerated fractionation alpha-fetoprotein doxorubicin, ifosfamide, mesna anal intra-epithelial neoplasia aromatase inhibitors American Joint Committee on Cancer 5-aminolevulinic acid acute lymphoblastic leukaemia alkaline phosphatase acute myeloid leukaemia antero-posterior fields adenomatous polyposis coli antigen presenting cells promyelocytic leukaemia Antiplatelet Trialists Collaborative Group cytarabine triphosphate accelerated radiotherapy, carbogen and nicotinamide American Society of Clinical Oncology autologous stem-cell transplantation American Society of Hematology A Study in the Treatment of Endometrial Cancer trial ataxia telangiectasia anti-thymocyte globulin
AT/RTs ATAC ATBC ATLAS ATLL ATP ATRA aTTom AUC BAL BCG BCIRG BCNU BDSs BEACOPP BEAM BEC BED BEP BER BIG 1–98 BIPSS BNCT BOLD bp BrdUrd BRT BSO BU C-HDT C225 CA CAD CALG cAMP CAR CARET CASH
atypical teratoid/rhabdoid tumours Arimidex or Tamoxifen Alone or in Combination trial Alpha-Tocopherol, Beta-Carotene study Adjuvant Tamoxifen Longer Against Shorter trial adult T-cell leukaemia/lymphoma adenosine phosphate all-trans retinoic acid adjuvant Tamoxifen Treatment offer more trial area under time/concentration curve bronchoalveolar lavage bacillus Calmette—Guerin Breast Cancer International Research Group 1,3-bis-(2-chloroethyl)-1-nitrosurea (carmustine) beam direction shells bleomycin, doxorubicin, cyclophosphamide, prednisone, gemcitabine BCNU, etoposide, cytarabine, melphalan bleomycin, etoposide, carboplatin biologically effective dose bleomycin, etoposide, platinum base excision repair Femera-Tamoxifen Breast International Group (FEMTA) bilateral inferior petrosal sinus sampling boron neutron capture therapy blood oxygen level-dependent base pair bromodeoxyuridine brachytherapy bilateral salpingo-oophorectomy busulphan conventional high-dose therapy cetuximab carbohydrate antigen; carbonic anhydrase computer-aided diagnosis Cancer and Leukaemia Group cyclic adenosine monophosphate coxsackie and adenovirus receptor Beta-Carotene and Retinol Efficacy Trial Cancer and Steroid Hormone study
List of abbreviations used xvii
CaSR CAV CCNU CCT CD138 CD CDC CDKs CEA CED CEUS CEV CF CGH CHART CHF CHI/NAO CIN CIRs CIS CLDR CLIP CLL CM CMF CML CMV CNS CNS CODE COG COX-2 CPR CPT-11 CR3-DCC CR CRC CRE CRM CRP CRu CSCG CSF CTAs CTD CTLs CTV
calcium-sensing receptor cyclophosphamide, doxorubicin and vincristine lomustine combination chemotherapy syndecan cytosine deaminase complement-dependent cytotoxicity cyclin-dependent kinases carcinoembryonic antigen convection-enhanced delivery contrast enhanced ultrasound etoposide, cyclosphosphamide and vincristine conventional fractionation comparative genomic hybridization continuous hyperfractionated accelerated radiotherapy congestive heart failure Commission for Health Improvement/ National Audit Office conjunctival (or corneal) intraepithelial neoplasia chimeric immune receptors carcinoma-in-situ continuous low-dose-rate Cancer of the Liver Italian Programme chronic lymphocytic lymphoma complete hydatidiform mole disease cyclophosphamide, methotrexate, 5-fluorouracil chronic myeloid leukaemia cisplatin, methotrexate, vinblastine; cytomegalovirus central nervous system coagulase negative Staphylococcus cisplatin, vincristine, doxorubicin and etoposide Clinical Outcomes Group cycloxygenase-2 cardio-pulmonary resuscitation irinotecan complement-receptor-3-dependent cellular cytotoxicity complete response Cancer Research Campaign cumulative radiation effect continual reassessment method C-reactive protein complete remission unconfirmed Cancer Services Coordinating Group cerebrospinal fluid cancer—testis antigens cyclophosphamide, thalidomide, dexamethasone cytotoxic T lymphocytes clinical target volume
CVB CVC CVF CU CUPI Cy CYP D&C DACH DAHANCA DALM DC, NOS or NST DCE-MRI DCIS DCs DES 5-DFCR DFS 5-DFUR DHFR DISC-HSV DLBCL DLIs DLTs DNES DNP DPD DR DRE DRs DTCH dThdPase DTIC DTPA dTTP dUMP DUS DWI EAG EASL EBCTCG EBRT EBV ECD ECM ECMV ECOG ECST EF-RT EFS
cyclophosphamide, etoposide, BCNU central venous catheter cobra venom factor 1-(2-chloroethyl)-3-cyclohexyl1-nitrosurea (lomustine) Chinese University Prognostic Index cyclophosphamide cytochrome P450 dilatation and curettage 1, 2-diaminocyclohexane Danish Head and Neck Cancer Study dysplasia-associated lesion or mass Ductal Carcinoma, Not Otherwise Specified or of No Special Type dynamic contrast-enhanced MRI ductal carcinoma-in-situ dendritic cells diethylstilbestrol 5-deoxy-5-fluorocytidine disease-free survival 5-deoxy-5-fluorouridine dihydrofolate reductase disabled infectious single cycle-herpes simplex virus diffuse large B-cell lymphoma donor lymphocyte infusions dose-limiting toxities diffuse neuroendocrine system dinitrophenyl dihydropyrimidine dehydrogenase duration of remission digital rectal examination duration of responses delayed-type cutaneous hypersensitivity thymidine phosphorylase dacarbazine diethylenetriamine penta-acetic acid deoxynucleotide, deoxythymidine triphosphate deoxyuridylate colour Doppler ultrasound diffusion-weighted imaging Expert Advisory Group European Association for Study of the Liver Early Breast Cancer Trialistsí Collaborative Group external-beam radiotherapy Epstein—Barr virus extracellular domain extracellular matrix etoposide, cyclophosphamide, methotrexate and vincristine Eastern Cooperative Oncology Group European Carotid Surgery Trial extended field radiotherapy event-free survival
xviii List of abbreviations used
EGF EGP-2 ELISA ELVIS EM EMA EMEA EMP EMR EORTC EPO EPIC ER ERCP ESR ETT EUA EUD EUROCARE EUROMEN EUS EV 18 F-FDG PET 18
F-FLT FAC FAMMM FAP FASG FBC fCT FDA FDG FdUMP FdUTP FEC FEMTA FEV FFTF FGF FHDR FIGO FIHP FISH FLAIR FLR FLT-3 FMTC FNAC FPGS
epidermal growth factor epithelial cell adhesion molecule Enzyme-Linked ImmunoSorbent Assay Elderly Lung Cancer Vinorelbine Study trial electron microscopy etoposide, methotrexate and actinomycin European Medicines Agency extramedullary plasmacytoma endoscopic mucosal resection European Organisation for Research and Treatment of Cancer erythropoietin European Prospective Investigation into Cancer and Nutrition oestrogen receptor endoscopic retrograde cholangiopancreaticography erythrocyte sedimentation rate epithelioid trophoblastic tumour examination under anaesthetic equivalent uniform dose Survival of Cancer patients in Europe study 172 European Multiple Endocrine Neoplasia study group endoscopic ultrasound etoposide and vincristine fluorodeoxyglucose-positron emission tomography fluorine-18 fluorothymidine 5-fluorouracil, doxorubicin, cyclophosphamide familial atypical mole/malignant melanoma familial adenomatous polyposis French Adjuvant Study Group full blood count functional CT Food and Drug Administration (US) fluorodeoxyglucose 5-fluoro-deoxyuridine monophosphate 5-fluoro-deoxyuridine-5-triphosphate 5-fluorouracil, epirubicin, cyclophosphamide Femara-Tamoxifen Breast International Group forced expiratory volume freedom from treatment failure fibroblast growth factor fractionated high-dose-rate International Federation of Gynecology and Obstetrics familial isolated hyperparathyroidism in-situ hybridization fluid-attenuated inversion recovery future liver remnant FMS-like tyrosine kinase-3 familial medullary thyroid carcinoma fine-needle aspiration cytology folypoly-gamma-glutamate synthetase
FSE FSH FSRT FT FTT 5-FU FUDR FUTP GARFT
fast spin echo follicle stimulating hormone fractionated stereotactic radiotherapy farnesyl transferase Fibrinolytic Trialists Collaborative Group 5-fluorouracil 5-fluoro-2-deoxyuridine fluorouridine triphosphate glycinamide ribonucleotide formyl transferase GBM glioblastoma multiforme GC gemcitabine/cisplatin doublet GCSF granulocyte colony-stimulating factor GCT granulosa cell tumour GCV gangiclovir Gd-BOPTA gadobenate dimeglumine Gd-EOB-DTPA gadoxetate GDEPT gene-directed enzyme prodrug therapy GDNF glial-cell-derived neurotropic factor GEP gastro-enteropancreatic, gene expression profiling GERD gastro-oesophageal reflux disease GFAP fibrillary acidic protein GFR glomerular filtration rate GH growth hormone GHSG German Hodgkin Study Group GISTs gastrointestinal stromal tumours GITSG Gastrointestinal Tumour Study Group GM CSF granulocyte—macrophage colony-stimulating factor GMP Good Manufacturing Practice GnRH gonadotrophin-releasing hormone GOG Gynaecologic Oncology Group GRE gradient-echo pulses (RF pulses) GTD gestational trophoblastic disease GTT gestational trophoblastic tumours GTV gross tumour volume GVHD graft-versus-host disease GVL graft-versus-leukaemic effect GVM graft-versus-myeloma GVT graft-versus-tumour effect HAART highly active anti-retroviral therapy HACA human anti-chimeric antibody HAMA human anti-mouse antibody HATS histone acetyltransferases HAV hepatitis A virus HB-EGF heparin-binding EGF-like growth factor HBV hepatitis B virus HCC hepatocellular carcinoma hCG human chorionic gonadotrophin HCV hepatitis C virus HDAC histone-deacetylase HDCT high-dose chemotherapy HDM high-dose melphalan HDR high dose rate HDT high-dose therapy her herceptin
List of abbreviations used xix
HERA HGF HGPRT 5-HIAA HIFU HL HLA HM HMO HNSCC HNPCC hPL HPT-JT HPV HR HRT HSCs HSPs HSV HSVtk 5-HT HTLV HU 131 I-BC8 IAP IARC IBC IBCS IBIS ICE ICRU IDC IDM IES IF-RT IFA IFM IFN IFIs IFL IgH IGF IgG1 IL-2 ILC ILP IMC IMRT INT101 IP IR ISGF3
herceptin in adjuvant breast cancer, Herceptin Adjuvant (trial) hepatocyte growth factor hypoxanthine-guanine phosphoribosyl transferase 5-hydroxyindole acetic acid high-intensity focused ultrasound Hodgkin’s lymphoma human leukocyte antigen hydatidiform mole Health Maintenance Organization squamous cell carcinoma of head and neck hereditary non-polyposis colon cancer syndrome human placental lactogen hyperparathyroidism—jaw tumour syndrome human papillomavirus Hazard Ratio; homologous recombination hormone replacement therapy haematopoietic stem cells heat shock protein(90) herpes simplex virus HSV thymidine kinase 5-hydroxytrypamine human T-lymphotropic virus Hounsfield units murine anti-CD45 mAb inhibitors of apoptosis proteins International Agency for Research on Cancer inflammatory breast carcinoma International Breast Cancer Study Group International Breast Cancer Intervention Study ifosfamide, carboplatin and etoposide International Commission for Radiation Units and Measurements invasive ductal carcinoma intermediate-dose melphalan Intergroup Exemestane Study trial involved field radiotherapy incomplete Freund’s adjuvant Intergroupe Francais du Myelome interferon invasive fungal infections irinotecan, 5-FU, leucovorin immunoglobulin heavy chain insulin-like growth factor immunoglobulin G1 interleukin-2 invasive lobular carcinoma interstitial laser photocoagulation; isolated limb perfusion internal mammary chain intensity-modulated radiation therapy Intergroup 101 trial irinotecan and cisplatin interventional radiology IFN-stimulated gene factor-3
ISIS IVUs JACIE kb kDa KGF KIT LAKs LCIS LD LDH LDHRS LET LGAs LH LHRH LI LIMIT LOD score LOH LR LQ LQC LTRs LTCRs MA-170 MAbs MAC MAGE MAHA MAID MALT MAP MCA MCL MCM MCNs MDAH MDM2 MDR MDS MDTC MeCCNU MEN MEPs MESNA MFs MGMT MGUS MHC MIBG
International Study of Infarct Survival intravenous urograms Joint Accreditation Commitee for ISCT and EBMT kilobase kiloDalton keratinocyte growth factor kinase inhibitor therapy lymphokine-activated killer cells lobular carcinoma-in-situ latissimus dorsi (flap) lactate dehydrogenase low-dose hyper-radiosensitivity linear energy transfer low-grade diffuse astrocytomas luteinizing hormone luteinizing hormone releasing hormone labelling indices Leicester Intravenous Magnesium Intervention Trial logarithm to base 10 of the odds loss of heterozygosity local recurrence linear quadratic last qualifying chemotherapy long terminal repeats long-term reconstituting cells Letrozel and Tamoxifen trial monoclonal antibodies membrane attack complex melanoma antigen microangiopathic haemolytic anaemia mesna, doxorubicin, ifosfamide, and dacarbazine mucosa-associated lymphoid tissue mitogen-activated protein methylcholanthrene mantle cell lymphoma mini-chromosome maintenance Managed Clinical Networks MD Anderson Hospital mouse double minute 2 multi-drug resistance myelodysplastic syndromes multi-detector computed tomography semustine multiple endocrine neoplasia motor-evoked potentials sodium-2-mercaptoethane multiplying factors methyl guanine methyl transferase monoclonal gammopathy of undetermined significance major histocompatability complex 123 I-meta-iodobenzylguanidine
xx List of abbreviations used
MINDACT MIRD MLC MM MMC MMF MP MMPs MP-T MPNST Mn-DPDP MOPP 6-MP MRA MRC MRD MRCP MRI MRP MRS MSKCC MTC MTD MUC-1 MVAC MVD MVEC MVP NAA NAC NAT nBCC NCAM N-CAM NCCTG NCI NCIC NCRI NCRN NDA NEAT NER NF-κ-B NHL NI NICaN NICE NIHPILCO
NK NMDA NMPs NMS
Microarray In Node negative Disease may Avoid Chemotherapy Trial Medical Internal Radiation Dosimetry multi leaf collimator multiple myeloma mitomycin C mycophenolate mofetil melphalan and prednisolone matrix metalloproteinases melphalan, prednisolone plus thalidomide malignant peripheral nerve sheath tumours mangafadipir trisodium mustine, vincristine, procarbazine, prednisone 6-mercaptopurine magnetic resonance angiography Medical Research Council minimal residual disease magnetic resonance cholangio-pancreatography magnetic resonance imaging MDR-associated protein magnetic resonance spectroscopy Memorial Sloan Kettering Cancer Centre medullary carcinoma of the thyroid maximum tolerated dose mucin-1 methotrexate, vinblastine, doxorubicin, cisplatin micro-vessel density methotrexate, vinblastine, epirubicin, cisplatin mitomycin C, vinblastine and cisplatin N-acetyl acetate neoadjuvant chemotherapy N-acetyl transferase nodular basal cell carcinoma neural cell adhesion molecule neural cell adhesion molecule North Central Cancer Treatment Group National Cancer Institute National Cancer Institute of Canada National Cancer Research Institute National Cancer Research Network new drug application national breast cancer study of epirubicin nucleotide excision repair nuclear factor-κ-B non-Hodgkin’s lymphoma Nottingham Index Northern Ireland Cancer Network National Institute for Health and Clinical Excellence National Institute of Health, Prostate and Lung, Colorectal and Ovarian Cancer Screening Trial natural killer cell N-methyl D-aspartate nuclear matrix proteins New Member States
NMSG NOS NPV NSABP P-1 NSAIDs NSCLC NSE NST NTCPs OECD OLT ONJ OPSI OPG ORR OS PACS PAM PARP PBD PBSCT PBPCs PBSCs PCCL PCG/T PCI PCNA PCNSL PCNU PCP PCR PCTs PCV PD PDGF PDT PECAM PE PEG PEIT PERCHE PES PFS PGE2 PGL PHPT PIAF PICC PKA PLL PLAP PLG
Nordic Myeloma Study Group not otherwise specified negative predictive value National Surgical Adjuvant Breast and Bowel Project P-1 non-steroidal anti-inflammatory drugs non-small-cell lung cancer neurone-specific enolase no special type normal-tissue complication probabilities Organization for Economic Cooperation and Development orthotopic liver transplantation osteonecrosis of the jaw overwhelming post-splenectomy syndrome osteoprotegerin overall response rate overall survival Programme Adjuvant Cancer du Sein primary acquired melanosis poly ADP-ribose polymerase percutaneous biliary drainage peripheral stem cell transplantation peripheral blood progenitor cells peripheral blood stem cells Primary Care Cancer Lead Primary Care Group/Trusts prophylactic cranial irradiation proliferating cell nuclear antigen primary CNS lymphoma 1-(2-chloroethyl)-3-(2,6-dioxo-3-piperidyl)-1nitrosurea Pneumocystis carinii pneumonia polymerase chain reaction Primary Care Trusts procarbazine, CCNU and vincristine progressive disease platelet-derived growth factor photodynamic therapy platelet endothelial cell adhesion molecule platinum and etoposide polyethylene glycol percutaneous ethanol injection therapy Premenopausal Endocrine-Responsive Chemotherapy post-embolization syndrome progression-free survival prostaglandin E2 paraganglioma primary hyerparathyroidism cisplatin, rINF-alpha, doxorubicin, and 5-fluorouracil peripherally inserted central catheters protein kinase A prolymphocytic leukaemia placental alkaline phosphatase poly-lactide-coglycolide microspheres
List of abbreviations used xxi
PM PNETS POEMS PPAR gamma PPV PR PR-negative PSA PSC PSTT PTC PTCLu PTEN PTH PTV PVE PWI QALY QOL QUASAR R-CHOP RARs RANKL RB RBE RCA RCC RCHOP RCR RE RECIST RES RF RFA RFC RFLPs RFS rhTSH RIA RIC RISC RIT RMSs RNAi ROI ROTI RR RT RT-PCR RTK RTOG RXRs
partial hydatidiform mole disease primitive neuro-ectodermal tumours polyneuropathy, organomegaly, endocrinopathy, and skin changes syndrome peroxisome proliferator-activated receptor gamma positive predictive value partial remission; partial response progesterone-receptor-negative prostate-specific antigen primary sclerosing cholangitis placental-site trophoblastic tumour percutaneous transhepatic cholangiogram peripheral T cell lymphoma unspecified phosphatase and tensin homologue parathyroid hormone planning target volume portal vein embolization perfusion weighted imaging quality adjusted life year quality of life Quick and Simple and Reliable rituximab, cyclophosphamide, doxorubicin, vincristine, prednisolone retinoid acid receptors receptor activator of nuclear factor kappaB ligand retinoblastoma relative biological effect replication-competent adenovirus renal cell carcinoma rituximab plus cyclophosphamide, doxorubicin, vincristine and prednisolone Royal College of Radiologists relative effectiveness response evaluation criteria in solid tumours reticuloendothelial system repopulation factor radiofrequency ablation reduced folate carrier restriction fragment length polymorphisms relapse-free survival recombinant TSH stimulation radio-immunoassay reduced intensity conditioning RNA-induced silencing complex radioimmunotherapy rhabdomyosarcomas RNA interference therapy region of interest related organ or tissue impairment relative risk; ribonucleotide reductase radiotherapy reverse transcriptase polymerase chain reaction receptor tyrosine kinase Radiation Therapy Oncology Group retinoid X receptors
sBCC SBP SCAN SCF SCF SCLC SCT SD SDH SDT SE SEER SF SF2 SFV SHAs SHH SIGN SIN siRNA SNB sNDA SNPs SOBP SOFT SOS SPD SPECT SRS STR STS START SUPREMO SUV SVCO SWOG T1/T2W TAA TACE TAH Taxol TBI TCCs TCDD TCH TCNU TCP TCR TD TDF TE TEXT 6-TG TGF
superficial based cell carcinoma solitary plasmacytoma of bone South East Cancer Network supraclavicular fossa stem cell factor, supraclavicular fossa small-cell lung cancer stem cell transplantation stable disease succinate dehydrogenase gene skin directed therapy spin echo (RF pulses) Surveillance, Epidemiology, and End Results (NCI) scatter factor surviving fraction Semliki forest virus Strategic Health Authorities sonic hedgehog pathway Scottish Intercollegiate Guidelines Network sinbis virus small inhibitory RNAs sentinel node biopsy supplementary new drug application single nucleotide polymorphisms spread-out Bragg peaks Suppression of Ovarian Function Trial sinusoidal obstruction syndrome sum of products of greatest diameters single photon emission computed tomography stereotactic radiosurgery short tandem repeat soft tissue sarcoma STandardization of breast Radiotherapy Trial Selective Use of Postoperative Radiotherapy aftEr MastectOmy standard uptake value superior vena caval obstruction South West Oncology Group study weighted MRI image effects tumour-associated antigen transarterial chemoembolization total abdominal hysterectomy taxanes paclitaxel total body irradiation transitional cell carcinomas tetrachlorodibenzo-dioxin taxotere, carboplatin, herceptin tauromustine tumour cure probability T cell receptor total physical dose time—dose factor time to echo Tamoxifen and Exemestane Trial 6-thioguanine transforming growth factor
xxii List of abbreviations used
TH1 TILs TME TMP TNF TNF-alpha TNM TP TR TRAIL TRAM TREGS TRUS TRM TS TSH TSP TTF-1 TTP TUR TURBT TURPS TUUS TVUS UAPI UC UCAs UCNT UCSF
T helper 1 cytokines tumour-infiltrating lymphocytes total mesorectal excision tumour marker production tumour necrosis factor tumour necrosis factor-alpha tumour, node, metastasis thymidine phosphorylase time to repeat TNF-related apoptosis-inducing ligand transverse rectus abdominis myocutaneousfree flap regulatory T cells transrectal ultrasound transplant related mortality thymidylate synthase thyroid-stimulating hormone thrombospondin thyroid transcription factor-1 thrombocytopenic purpura, time to progression transurethral resection transurethral resection of bladder tumour transurethral resection of the prostate transurethral ultrasound transvaginal ultrasound uterine artery pulsatility index ulcerative colitis ultrasound contrast agents undifferentiated carcinomas of the nasopharynx University California San Francisco
UFT UICC UKCTOCS USPIO V-DMSA VAD VAIN VATS VDAs VDR VEE VEGF VHL VI VICE VIN VIP VLPs VNPI VNTRs, VOD VUDs VP-16 WAGR WBD WBRT WL WW Z-Dex
uracil/tegafur Union Internationale Contre Cancer UK Collaborative Trial of Ovarian Cancer Screening ultra-small particles of iron oxide pentavalent dimercapto succinic acid vincristine, adriamycin, dexamethasone vaginal intra-epithelial neoplasia video-assisted thoracic surgery vascular disrupting agents vitamin D receptor Venezuelan equine encephalitis vascular endothelial growth factor von Hippel—Lindau syndrome vascular invasion vincristine, ifosfamide, carboplatin and etoposide vulval intra-epithelial neoplasia vasoactive intestinal peptide virus-like particles Van Nuys prognostic scoring index system variable number of tandem repeats veno-occlusive disease volunteer unrelated donors etoposide Wilms—Aniridia—Gentitourinary— Retardation syndrome whole-body dose whole-brain radiotherapy window level window width dexamethasone
Evidence scoring
★★★ ★★
★
systematic review or meta-analysis one or more well designed randomized controlled trials nonrandomized controlled trials, cohort study etc.
Reference annotation
The reference lists are annotated, where appropriate, to guide readers to key primary papers and major review articles as follows: ● ◆ ★
Key primary papers as indicated by Major review articles as indicated by Papers that represent the first normal publication of a management guideline are indicated by
We hope that this feature will render extensive lists of references more useful to the reader and will help to encourage self directed learning among both trainees and practising physicians.
This page intentionally left blank
PART
1
PRINCIPLES
1 Introduction Karol Sikora 2 Molecular biology Hani Gabra, Euan Stronach, Rohini Sharma 3 Clinical radiobiology Michele I Saunders, Stanley Dische 4 Mathematical modelling and its application in oncology Roger G Dale, Bleddyn Jones 5 Principles of chemotherapy and drug development Jeffrey Evans, Meenali M Chitnis, Denis C Talbot 6 Tumour imaging in oncology Anju Sahdev, Rodney H Reznek 7 Interventional radiology Tarun Sabharwal, Anne P Hemingway, Andreas Adam 8 Vaccination strategies for malignant diseases Deepak P Assudani, Stephanie McArdle, Murrium Ahmad, Geng Li, Robert C Rees, Selman A Ali 9 Biological therapies: cytokines and adoptive cell therapy Fiona C Thistlethwaite, Peter L Stern, Robert E Hawkins 10 Radioimmunotherapy Tim Illidge, Mike Bayne 11 Monoclonal antibodies and treatment of cancer Hossein Borghaei, Liat Binyamin, Igor Astsaturov, Louis M Weiner 12 Angiogenesis as a target for the treatment of cancer Srinivasan Madhusudan, Daniel Patterson, Adrian L Harris 13 Gene therapy Kate Relph, Kevin Harrington, Hardev Pandha
3 23 40 58 75 112 146 167 186 203 226 251 269
This page intentionally left blank
1 Introduction KAROL SIKORA
Cancer’s timeline Epidemiology Prevention Screening Diagnosis Surgery
3 4 6 8 10 10
This book is written by many authors around one common theme – the optimal treatment of cancer. The problem at first seems relatively simple. There are about 1013 cells in the human body. From the fertilized egg to death in old age, a human being is the product of 1016 cell divisions. Like all complex systems, growth control can go wrong, resulting in the loss of normal territorial restraint, producing a family of cells that can multiply indefinitely. But it is not just the local growth of tumour cells that makes them so lethal. It is their spread, directly through invasion and by metastases to other sites of the body. Tumours that remain localized can usually be cured by surgery or radiotherapy, even when enormous. Patients with large, eroding basal cell skin cancers, for example, can be treated successfully, as these tumours seldom invade deep into the skin or spread to lymph nodes. Yet a breast lump less than 1 cm in diameter, which causes the patient no problems and is picked up in a screening clinic, can be lethal if metastases have already arisen from the primary site. It is this spread that provides the plethora of clinical problems. Just as no two individuals are alike, no two tumours behave in exactly the same way, although we can make some broad generalizations from clinical experience. The physical and psychological interactions of a patient with a growing cancer require careful analysis and action by those involved in the patient’s care. Cancer is not universally fatal. Tremendous advances have been made in the treatment of leukaemia, lymphoma, testicular cancer, choriocarcinoma and several other rare tumours, for which cure of even widespread disease is now common. Even with lung cancer, the most common
Radiotherapy Chemotherapy The future – getting innovation into practice Conclusion References
11 11 13 20 21
single tumour type throughout the world, about 8 per cent of patients survive for many years and die of other causes. However, although there are some pointers, we do not understand why this 8 per cent should be spared. If they can be cured, why not the rest? Vast sums of money are currently spent worldwide on research and yet for most common tumours there has been little change in overall cure rates over the last 30 years. As an intellectual problem to the scientist, malignant disease has always appeared eminently soluble. After all, it would seem a relatively straightforward task to identify the differences between normal and malignant cells and devise a selective destruction process. Yet we still do not know precisely the first biochemical step that takes a cell down the road to neoplasia. The recent advances in molecular biology seem poised to rectify this and to give us new avenues for clinical exploitation, but we have to treat our patients now – providing for them the best of today’s technology with the skill of the caring physician.
CANCER’S TIMELINE The first recorded reference to cancer was in the Edwin Smith Papyrus of 3000 BC, in which eight women with breast cancer are described. The writings of Hippocrates in 400 BC contain several descriptions of cancer in different sites. But our understanding of the disease really began in the nineteenth century with the advent of cellular pathology. Successful treatment by radical surgery became possible in the later part of that century thanks to advances in
4 Introduction
anaesthetics and antiseptics. Radical surgery involved the removal of the tumour-containing organ and the draining of its lymph nodes in one block. Halstead in Johns Hopkins was the main protagonist of the radical mastectomy, Wertheim of the hysterectomy, Trotter of the pharyngectomy and Miles of abdomino-perineal resection of the rectum. These diverse surgical procedures all followed the same principles. The twentieth century ended with the conservation of organs by minimizing the destruction caused by surgery and replacing it with radiotherapy and, for some sites, effective adjuvant therapy with drugs (Box 1.1).
Box 1.1 Cancer’s timeline 3000 BC 400 BC 1880 1896 1898 1899 1946 1953 1958 1999 2000 2006
First recorded description Hippocrates describes six cancer types Successful radical mastectomy Oophorectomy for breast cancer Discovery of radium Discovery of X rays First publication on successful chemotherapy for cancer Double-helical structure of DNA elucidated Successful use of combination chemotherapy First molecularly targeted therapy approved for use Human genome mapped More than ten targeted therapies available
Radiotherapy has come a long way since the first patient with a nasal tumour was treated in 1899, only a year after the discovery of radium by Marie Curie. Although radiobiology developed as a research discipline, it has really contributed little to clinical practice. The rationale behind modern fractionated radiotherapy comes as much from empirical trial and error as from experimental results. Radiotherapy is remarkably successful for certain areas of the body. Increasing sophistication in equipment coupled with dramatic strides in imaging have led to great precision in the planning and execution of treatment, thus sparing critical normal tissues and increasing the dose to the tumour. The sinking of the US battleship John B Harvey in Bari Harbour in Italy by the Germans in 1942 led to the development of effective chemotherapy. The warship was carrying canisters of mustard gas for use in chemical warfare. Survivors developed leucopenia and this led Goodman and others back in the USA to experiment with halogenated alkylamines in patients with high white cell counts – lymphomas, leukaemias and Hodgkin’s disease. From the first publication in 1946, the field has blossomed, with more than 200 drugs now available in our global pharmacopoeia. But as with radiotherapy, our clinical practice is based mainly
Table 1.1 Chemotherapy for advanced cancer High CR/high cure 5% HD ALL Testis Choriocarcinoma Childhood BL
High CR/low cure
Low CR/low cure
40% AML Breast Ovary SCLC Sarcoma Myeloma
55% NSCLC Colon Stomach Prostate Pancreas Glioma
CR, complete response; HD, Hodgkin’s disease; ALL, acute lymphoid leukaemia; BL, Burkitt’s lymphoma; AML, acute myeloid leukaemia; SCLC, small-cell lung cancer; NSCLC, non-small-cell lung cancer.
on empiricism. Most currently used drugs were found serendipitously from plants or fungi – paclitaxel, vincristine, doxorubicin – and not by rational drug design. Although very successfully used in combination for lymphoma, leukaemia, choriocarcinoma, testicular cancer and several childhood cancers, results in metastatic common solid tumours have been disappointing, with little more than palliative benefit (Table 1.1). The advent of molecularly targeted drugs promises to change this dramatically.
EPIDEMIOLOGY The global incidence of cancer is soaring due to the rapid increase in the number of elderly people in most countries. By the year 2020, there will be 20 million new cancer patients each year, and 70 per cent of them will live in countries that collectively will have less than 5 per cent of the world’s resources for cancer control (Box 1.2). We have seen an explosion in our understanding of the disease at a molecular level and are now poised to see some very significant advances in prevention, screening and treatment.
Box 1.2 The global cancer burden Current population 6 billion 10 million new cancer patients – 6 million deaths 50% in developing countries with 5% resources 2020 population 8 billion 20 million new cancer patients – 12 million deaths 70% in developing countries
Dramatic technological change is likely in surgery, radiotherapy and chemotherapy, leading to increased cure rates, but at a price.1 The completion of the Human Genome Project will almost certainly bring sophisticated genetic risk assessment methods requiring careful integration into existing screening programmes.2 Preventive strategies
p.c. GNP US$ (1995)
Epidemiology 5
45000 40000 35000 30000 25000 20000 15000 10000 5000 0 0
20
40
60
80
Life expectancy (1997)
could considerably reduce the global disease burden at low cost, and palliative care to relieve pain and suffering should be a basic right of all cancer patients. The next 25 years will be a time of unprecedented change in the way in which we will control cancer. However, the optimal organization of prevention and detection programmes as well as of treatment services is a universal problem in all economic environments. The world is in a health transition. Infection as a major cause of suffering and death is giving way to new epidemics of non-communicable disorders such as cardiovascular disease, diabetes and cancer.3 Different countries are in different stages of this transition depending on their age structure and economy. Some countries are faced with a double burden, with increasing infection problems compounded by surging cancer rates. This is fuelled in part by the globalization of unhealthy lifestyles.4 Cancer is often thought to be the problem of rich countries. The 1998 World Health Organisation World Health Report5 life expectancy data for 1997 and the World Bank data on per capita gross national product expressed in US dollars (pcGNP$) is available for 155 countries. Cancer in Five Continents6 and the Electronic Database for Cancer provide incidence figures for 1990 and those predicted for 2020.7 The incidences for men and women were analysed separately. The ratio of cancer incidence per 100 000 population in 2020 to 1990 can be compared and correlated to the wealth of the population. Many relatively poor countries with already inadequate treatment facilities will see an explosion in cancer incidence.
Longevity and wealth Figure 1.1 examines the relationship between life expectancy at birth for both men and women and wealth of the 155 countries. There is a clear relationship between increasing GNP and longer life. There are relatively large gains for small increases in pcGNP$ in the poorer countries, reflecting reduced infant and childhood mortality. Above a pcGNP$ of 1000, the proportional gain in longevity is markedly reduced. This almost certainly reflects the importance of basic measures such as vaccination, good water supply,
100
Figure 1.1 Longevity and wealth.
Table 1.2 Longevity and wealth Longevity lower than expected pcGNP$ 2000 Longevity 60 years Namibia, Botswana, Gabon
Longevity higher than expected pcGNP$ 1000 Longevity 65 years Egypt, Trinidad, Honduras, Nicaragua, Vietnam, Mongolia, Indonesia, China, Surinam, Kyrgistan, Sri Lanka, Tajikstan, Turkmenistan, Uzbekistan, Armenia, Georgia, Azerbaijan, Albania, Macedonia, Solomon Islands
pcGNP, per capita gross national product.
improved health education and access to simple medical care. After this, longevity continues to increase with wealth, but the increase is slow, reflecting the biological determinants that cause disease and death in all human populations. There are two interesting clusters (Table 1.2). The first is those countries where longevity is significantly less than expected for their relative wealth, with a pcGNP$ of above 2000 but a longevity of less than 60 years. These are three African countries – Namibia, Botswana and Gabon. The high level of disease related to human immunodeficiency virus (HIV) is the responsible factor. The second cluster is those states with a higher than expected longevity of greater than 65 years but a pcGNP$ of below 1000. Common factors are efficient public health systems, low infant and childhood mortality and an integrated primary care system: A further confounding factor is the relatively recent reductions in pcGNP$ in these countries caused by external factors and political change. Clearly there is a long incubation period between the factors responsible for longevity and the outcome. Major changes over the last decade will have considerable impact over the next 25 years.
Wealth and cancer incidence Figure 1.2 shows the relationship between wealth and cancer. There is a clear correlation between increasing wealth
6 Introduction
45000 p.c. GNP US$ (1995)
40000 35000 30000 25000 20000 15000 10000 5000 0 0
100
200 300 Cancer incidence per 100,000
400
500
Figure 1.2 Cancer incidence and wealth.
Table 1.3 Cancer incidence and wealth Incidence lower than expected pcGNP$ 5000 Incidence 150 per 100 000 Kuwait, Quatar, Saudi Arabia, Bahrain, UAE
Incidence higher than expected pcGNP$ 5000 Incidence 250 per 100 000 Bulgaria, Czech Republic, Hungary, Poland, Romania, Russian Federation, Kazakstan, Belarus, Slovakia, Ukraine, Estonia, Croatia, South Africa, Albania
pcGNP, per capita gross national product.
45000 p.c. GNP US$ (1997)
40000 35000 30000 25000 20000 15000 10000 5000 0 0
1
2
3
4
5
Ratio of cancer incidence 2020 : 1990
and cancer incidence. This is almost certainly due to the influence of tobacco and dietary factors as well as other more complex lifestyle factors together with increased longevity of the population. Exceptions include a cluster with a pcGNP$ of greater than 5000 and a cancer incidence of less than 150 per 100 000. These are all Arabian Gulf states (Table 1.3). This almost certainly reflects the benefit of the traditional lifestyle maintained by the majority of the population. The second cluster is the former socialist countries of Europe, certain former Soviet republics and South Africa, where the cancer incidence exceeds 250 per 100 000 but the pcGNP$ is less than 5000. This reflects increased longevity due to good public health and efficient healthcare systems, a Western lifestyle and again a reduction in real pcGNP$ due to political factors. Figure 1.3 shows the ratios of cancer incidence in 2020:1990 correlated to relative wealth. The largest changes
6
Figure 1.3 Change in cancer incidence by 2020.
in incidence are clearly predicted for poorer countries, with a good correlation between poverty and greatly increased incidence. Countries with the greatest increase will have the least facilities to deal with the healthcare problems posed by the disease. In many parts of the world patients usually present with late-stage disease that is not amenable to simple surgery.
PREVENTION Tobacco Optimal use of current knowledge could reduce the overall cancer incidence by at least 3 million. Tobacco control is the most urgent need.8 We need to look for longterm solutions here. The politics of tobacco is a complex
Prevention 7
conspiratorial web of industrialists, farmers, manufacturers, politicians and the pensions business, all looking after their own interests.9 Reduce cigarette consumption in many countries and the economy simply collapses. Governments are naturally cautious. In democracies they are subject to intense lobbying. In less democratic societies corruption, using the massive profits generated by the industry, usually achieves the desired endpoint. Advertising blatantly exploits the young of the developing world, associating images of sex, success and wealth with cigarettes as a lifestyle marker. The solutions are complex and require considerable political will. But with forceful and concerted international action against cigarette promotion, we could reduce cancer incidence by 20 per cent by the year 2020.
Diet Dietary modification could result in a further 30 per cent reduction across the board. The problem is refining the educational message and getting it right in different communities. Changing our current high-fat, low-fibre diet with a low fruit and vegetable intake is a common theme for cancer prevention. But many features of the modern Western diet are now being adapted globally as branded fast-food makers seek out new markets. Again, political will is necessary to reduce the costs to the public of healthy foods. We need to obtain more data so that we can make firmer recommendations. The European Prospective Investigation into Cancer and Nutrition (EPIC) study currently in progress is a good example of how painstaking data and serum collection from 400 000 Europeans could, over the years, provide a vast resource for investigating prospectively the complex inter-relationships between diet and cancer.10 Cancer incidence varies enormously across Europe, providing an excellent natural laboratory for such studies. Interventional epidemiology using rigorously controlled studies could produce the evidence that could lead to major changes. The current problem is the difficulty in making dietary advice specific and in some countries affordable. Although several groups have produced guidelines, there are so far few data about their uptake or significance in large populations. Box 1.3 provides a summary of the main consensus from several sources.
Box 1.3 Common dietary guidelines for cancer prevention ● ● ● ● ● ●
Avoid animal fat Increase fibre intake Reduce red meat consumption Increase fruit and vegetable intake Avoid obesity and stay fit Avoid excess alcohol
Table 1.4 Infection and cancer Hepatitis B virus Human papilloma virus Helicobacter pylori Epstein-Barr virus Human immunodeficiency virus Schistosomiasis Liver fluke
Hepatoma Cervix, anus Stomach Lymphoma, nasopharynx Kaposi’s sarcoma, lymphoma Bladder Cholangiocarcinoma
Infection Infection causes around 15 per cent of cancer worldwide and is potentially preventable (Table 1.4). This proportion is greater in the developing world, where an estimated 22 per cent of cancer has an infectious cause.11 Hepatitis B immunization in children has significantly reduced the incidence of infection in China, Korea and West Africa. Shortly we will see if it has reduced the incidence of hepatoma, which begins in endemic regions by the third decade of life. The unconfirmed trends are already encouraging.12 Cancer of the cervix, the commonest women’s cancer in parts of India and South America, is clearly associated with certain subtypes of human papilloma virus. Vaccines are now becoming available and entering trial.13 Helicobacter pylori is associated with stomach cancer. Here, without any intervention, there has been a remarkable downward trend in incidence worldwide. Dissecting out the complex factors involved, including food storage, contamination, preparation and content, is a considerable challenge. Other cancer-causing infections are schistosomiasis, the liver fluke, the human T-cell leukaemia virus and the ubiquitous hepatitis B virus. Although geographically localized, their prevention by lifestyle changes and vaccination programmes is a realistic short-term goal. Clearly, the effectiveness of any infection control or immunization programme at reducing the cancer burden will depend on many factors and require careful research and field evaluation.
Targeting The key to success in cancer prevention is careful targeting. Figure 1.4 shows estimates of the three main reversible cancer-causing factors, which between them are responsible for 7.5 million out the 10 million new patients annually. Targeted prevention programmes are very cost effective and can be shared by different countries with similar cancer patterns, and therefore countries with limited resources need not keep reinventing the wheel. Prevention packages can be tailored and adapted widely. To do this we need good data of incidence in relation to geography. Descriptive epidemiology provides a fertile hunting ground for patterns of carcinogenesis. Relating genetic changes in cancer to their cause and geography – the emerging discipline of
8 Introduction
lung orophx larynx bladder kidney pancreas TOBACCO
Table 1.5 Advantages and disadvantages of screening Advantages Better outcome Less radical therapy
stomach colon oesop breast liver orophx prostate DIET stomach cervix liver npx NHL bladder INFECTION
Figure 1.4 Global causes of cancer: (a) tobacco; (b) diet; (c) infection. Orophx, oropharynx; Oesop, oesophagus; Npx, nasopharynx; NHL, non-Hodgkin’s lymphoma.
molecular epidemiology – will complete the circle and point the way to specific interventions. The future of prevention will almost certainly be about using such techniques carefully to target preventive strategies to those who would benefit most. In the post-genomic era it is likely that cancer prevention programmes, at least in developed countries, will be completely individualized: a combination of environmental and lifestyle data will be used to construct very specific personalized messages.
Reassurance for those with negative results Psychological benefit to population Attractive to politicians Savings because therapy is less complex
Disadvantages Longer morbidity if prognosis unaltered Over-treatment of borderline abnormalities False reassurance for false negatives Unnecessary investigation Risks of screening test Resource costs of screening system
application of a test to individuals who have not sought medical attention. It may be opportunistic (offered to patients consulting their doctors for other reasons) or population based (covering a predefined age range, with elaborate call and recall systems). The risk of dying from a cancer increases with its degree of spread or stage; thus, the aim of screening is to detect cancer in its early, asymptomatic phase. The problem is that many screening tests are relatively crude, and cancers may have metastasized before they are detected. Sensitivity varies between tests. A 100 per cent sensitive test detects all cancers in the screened population. The most rigorous means of calculating sensitivity is to determine the proportion of expected cancers not presenting as interval cases between screens. Good cancer registration is essential when making this calculation. Specificity is the proportion of negative results produced by a test in individuals without neoplasia. A 100 per cent specific test gives no falsepositive results. Investigation of patients without cancer is a major factor in the cost of screening.
SCREENING Cancer screening is one of the great controversies of medicine. At the interface between public health and specialist care, economics creates tensions between professional groups, politicians and the public: a screening test may be cheap, but applying it to a population (with rigorous quality control and effective processing of patients with abnormal results) creates a huge workload and therefore cost. Screening can also have psychological effects on individuals with false-positive results who require investigation but are eventually found not to have cancer. Unless screening can be shown to reduce the mortality from a specific cancer, the money used is better spent on improving care, and this has led to a disparity in screening recommendations between countries. The Human Genome Project is likely to provide new approaches to cancer risk assessment and will bring new challenges to this complex area. Cancer screening is defined as the systematic
Advantages and disadvantages of screening The advantages and disadvantages of screening (Table 1.5) must be considered carefully; they vary between cancers and tests. The three main problems in assessing the benefit of any screening test for cancer are lead-time bias, length bias and selection bias, all of which impair the effectiveness of screening as a method of reducing cancer mortality. Lead-time bias advances the diagnosis but does not prolong survival, as occurs when the disease has already metastasized although the primary tumour is still small – patients die at the same time as they would if the disease had not been detected early. Length bias results in the diagnosis of less aggressive tumours. Rapidly growing cancers with a poorer prognosis present in the screening interval, reducing the value of the screening process. Selection bias occurs even in the best
Screening 9
organized healthcare systems. Worried but healthy individuals (who would present with cancer symptoms early) comply with screening, whereas less well-educated and socially disadvantaged individuals do not. In the UK National Health Service (NHS) breast cancer screening programme, compliance rates vary between communities depending on their relative deprivation.
Developing a screening programme Rational decision-making about cancer screening requires a detailed analysis of factors that may vary between populations.14 The cancer should be common and its natural history should be properly understood. This allows a realistic prediction of the probable value of the proposed test. The test should be effective (high sensitivity and specificity) and acceptable to the population. Cervical smears, for example, are difficult to perform in many Islamic countries, where women prefer not to undergo vaginal examination, and the take-up rate for colonoscopy is low in asymptomatic individuals because it is uncomfortable and sometimes unpleasant. The healthcare system must be able to cope with patients who produce positive results and require investigation. This may be a particular problem at the start of a population-based study. Ultimately, screening must improve the survival rate in a randomized controlled setting. The natural history of many cancers (including incidence and mortality) may change over time for reasons that are poorly understood. In Europe, the incidence of stomach cancer has decreased dramatically over the last few decades, whereas breast cancer deaths reached a peak in the UK in 1989 and have decreased slightly each year since then. Lobby groups often exercise political pressure to implement screening programmes (even when their effectiveness is undemonstrated), and manufacturers of equipment or suppliers of reagents may exercise commercial pressure. In fee-for-service-based provider systems, there is a financial inducement for doctors to investigate because doing nothing earns no money. The launch of the NHS breast screening service by the UK government in 1989 was viewed by many as a pre-election vote-winning exercise rather than a rational public health intervention, and there are now similar pressures to introduce prostate cancer screening, though uncertainty remains about the management of men with slightly elevated prostate-specific antigen (PSA). Many groups (e.g. governmental organizations, medical charities, health maintenance organizations, professional bodies) have produced guidelines on cancer screening. These guidelines vary widely between countries, reflecting bias in the interpretation of evidence and cultural values in the practice of medicine. For example, annual PSA testing and digital rectal examination in men over 50 years of age are recommended by the American Cancer Society, but are not advocated in most other countries. The incidence of a particular cancer in a particular country and the economics of screening must be considered carefully – the cost
of the technology required must correspond with the gain. Low-cost, direct inspection techniques for oral and cervical cancer by non-professional health workers seem attractive in achieving tumour downstaging and hence better survival results, but cervicoscopy programmes in India and China have shown surprisingly poor overall effectiveness.15 It remains to be seen whether intra-vital staining with acetic acid can enhance specificity at little extra cost. A major cost in instituting any screening procedure is informing the public and then developing the logistics, often under difficult geographical conditions. Cultural barriers may be insurmountable without better education, particularly of girls, who as mothers will become responsible for family health. Low-technology tests have low specificity; as a result, hard-pressed secondary care facilities are inundated with patients with non-life-threatening abnormalities. Detailed field assessment, preferably in a randomized setting, is essential before firm recommendations can be made, but political factors often interfere with this process.16 The wellmeaning charitable donation of second-hand mammography units to some African countries has led to haphazard introduction of breast screening in populations in which the incidence of breast cancer is low and where there are few resources to deal with abnormal results.
Assessing the benefits of screening programmes The ultimate measure of success in a screening programme is a demonstrable reduction in mortality in the screened population. This needs large numbers of individuals, however, and at least 10 years’ assessment for most of the common cancers. Although randomized studies may show conclusive benefit, it must be remembered that the expertise and professional enthusiasm available to a study population may be considerably greater than those achievable under subsequent field conditions. Quality of mammography interpretation and investigation of breast abnormalities are good examples of this, and may explain the relatively disappointing results of breast screening in practice. Case-control studies using agematched individuals from the same population and nonrandomized comparison between areas providing and not providing screening may provide useful indicators, but are not as conclusive as randomized trials. Surrogate measures of effectiveness can be used to assess a programme with relatively small numbers of patients soon after its implementation, but are insufficient to prove that screening saves lives. When a population is first screened, a higher than expected incidence of cancer should be seen because screening is detecting cancer that would not present with symptoms for several years. Subsequent rounds of screening are less productive. Tumour downstaging is a second measure of impact. An increase in early-stage cancer detection and, consequently, reduction in advanced disease are expected over 3–5 years. The third, short-term evaluation
10 Introduction
is a comparison of the survival of screen-detected patients with that of patients presenting symptomatically. Success in terms of these three indices may not necessarily be translated into a useful screening programme. In the 1970s, a study of routine chest radiography and sputum cytology to detect lung cancer showed a 5-year survival of 40 per cent in screendetected patients compared with an overall figure of 5 per cent, but a reduction in mortality from lung cancer in large populations has not been seen.
DIAGNOSIS Cancer presents with a myriad of symptoms depending on the site, size and growth pattern of the tumour. Although some symptoms alarm patients more than others, there is tremendous variability in the speed at which cancer can be diagnosed. A lump can be biopsied, but many deep-seated tumours present late, long after they have already spread: most patients have actually been harbouring the cancer for several years before it becomes apparent. Trying to speed up the diagnostic process and to get on with definitive treatment makes good sense. But delays plague all healthcare systems. In Britain, the current obsession is for all patients with cancer-related symptoms to be seen within 2 weeks. This was politically inspired to show something could be achieved quickly. The problem is defining what constitutes a cancer-related symptom – there are just so many. Studies show that having two queues for entry into the hospital system – one urgent and one not – leads to either excess system capacity or serious delays in the slow queue. Forming a unified entry system and shortening it makes more sense. A far bigger problem is getting a complex series of investigations performed with a reasonable start time for definitive therapy. Attempts to do this have been hampered by poor information technology systems, which are fragmented, non-communicative and primitive. In an age when a WAP cellphone can be used to book instantly a complex travel itinerary including hotels and opera tickets, it is a huge indictment that general practitioners (GPs) in many parts of the world cannot fix a hospital appointment for a potential cancer patient without posting a letter. The two drivers of the improvement of cancer diagnosis are imaging and biomarkers. The last two decades have seen a massive rise in the use of computed tomography (CT) and magnetic resonance imaging (MRI) scans to outline beautifully and in great detail the anatomy of a cancer and its surrounding normal structures. Positron emission tomography (PET), in which a molecule is labelled with a radioactive marker, allows us to examine the living biochemistry of the body. The future of imaging is coupling high-definition structural information with real-time functional change. In this way the precise effects of drug or other treatment can be monitored in three dimensions. It is also likely that the telecom revolution will produce new devices for examining the interior compartments of the body without causing distress to the patient.
Biomarkers are biochemical changes produced by the presence of a cancer. They may be synthesized directly by the cancer, such as PSA, or represent a complex change in an organ system, such as abnormal liver function tests caused by liver metastases. As we understand more about the molecular abnormalities that lead to cancer through the science of genomics and proteomics, novel biomarkers will be identified. These will give us the ability not only to diagnose cancer at an earlier stage, but also to predict the probable natural history of the cancer – whether it will spread rapidly or invade neighbouring structures. This information will be essential for planning optimal care. The basic tests are likely to be converted to kits sold in pharmacies. It is possible that a cancer screening kit for the four major cancers will be on sale within the next decade. Already there is great variation in the practice of cancer screening in different countries, and it is likely that the availability of commercial kits will increase consumerism. There will be a rise in cancer screening and prevention clinics in the private sector, almost certainly attached to the ‘cancer hotels’ of the future. Looking further forward, it is likely that continuous monitoring for potentially dangerous mutations will be possible. Up-market car engines have systems to measure performance against baseline, sending a signal to the driver if a problem arises. Implanted devices to identify genomic change and signal abnormalities to a home computer may allow the detection of cancer well before any metastasis. It will be essential to carry out careful outcome research on such new diagnostic and screening techniques to validate their benefits.
SURGERY Cancer surgery has been a dramatic success. Effective cancer surgery began in the late nineteenth century when it was realized that tumours could be removed along with their regional lymph nodes. This enhanced the chances of complete cure, as it had the greatest possibility of avoiding any spread of the cancer. Surgery still remains the single most effective modality for cancer treatment. Increasingly, it has become far more conservative, able to retain organs and structures and in turn to maintain good function in many parts of the body. Breast cancer is an excellent example. The radical mastectomy performed until 30 years ago left women with severe deformity of the chest wall. This was replaced first by the less mutilating simple mastectomy and now by simple excision followed by radiotherapy, the breast remaining fully intact. New technology permits minimally invasive (keyhole) surgery for many cancer types. The science of robotics will allow completely automated surgical approaches with enhanced effects and minimal damage to surrounding structures. Ultimately, it is likely that surgery will disappear as an important treatment and become confined simply to biopsy performed under local anaesthetic with image guidance to check that the correct sites are biopsied (Box 1.4).
Chemotherapy 11
Box 1.4 Future of surgery ● ● ● ● ● ● ●
Organ conservation Minimally invasive surgery Robotic surgery Distance surgery Tailored adjuvant approaches Biopsy only for many cancers All fast tracked – next-day service
RADIOTHERAPY Radiotherapy was first used for cancer treatment over 100 years ago. Originally crude radium was used as the radiation source, but we now have a variety of sophisticated techniques available. Modern linear accelerators – the workhorses for radiotherapy – allow precise dose delivery to the shape of the tumour. Conformal therapy aims to deliver a high dose just to the tumour volume in three dimensions, killing the cancer cells and avoiding sensitive normal surrounding tissue. Novel computer-based imaging techniques have revolutionized our ability to understand the precise anatomy of cancer in a patient and therefore to deliver far more effective radiotherapy. The future of radiotherapy is about further computerization with multimedia imaging and optimized conformal planning. We have also learnt to understand the biological differences between different tumours in patients and can begin to plan individualized treatment courses to optimize selective destruction. With remarkable technological changes in imaging and computerization, continued development is essential (Box 1.5). Radiotherapy in many parts of the world is the Cinderella of cancer care.
Box 1.5 Future of radiotherapy ● ● ● ● ●
Multi-media imaging Robotic set-up Optimized conformal planning Biological optimization Designer fractionation
CHEMOTHERAPY After the sinking of the battleship John B Harvey during the Second World War, naval physicians treated survivors suffering from lymphoma and leukaemia with nitrogen mustard. In 1946, 67 patients were reported to have a good, but brief, response to injections of this drug16 – and a new era of cancer care had begun. The current position of chemotherapy for advanced cancer is shown in Table 1.1. Essentially, there are three groups
of cancers, in the first of which we can achieve a high complete response rate and a high cure rate. This first group includes diseases such as Hodgkin’s disease, childhood leukaemia and testicular cancer. Unfortunately, this group of cancers that can be successfully treated represents less than 5 per cent of the global cancer burden. At the other end of the spectrum, we have a group with a low complete response and low cure rate, such as lung cancer, colon cancer and stomach cancer. So far, chemotherapy has made few inroads into their treatment, although some useful palliation and prolongation of survival, sometimes for months, can be achieved. In the middle we have a group of diseases with a high complete response but a low cure rate. These cause problems to those involved in rationing cancer care. The use of taxanes in breast and ovarian cancers is a classic example. High-cost drugs can achieve extension of life by several months for many patients and when deciding on priorities, we have to assess how much we are willing to pay for a month of reasonable quality of life. We are at the beginning of a revolution in cancer care. The pharmaceutical industry has taken on the new challenge, and is now going through a massive transition from an era of classical chemotherapy drugs (not too dissimilar to nitrogen mustard) that were discovered by screening programmes for their potential to destroy cells, to a molecular targeted approach. Currently there are 370 molecules in clinical development by 43 pharmaceutical companies. It is likely that fewer than 30 will actually make it to the marketplace, and fewer than 5 will make a significant impact on cancer care. Increasing consolidation in the industry has resulted in a shrinking of the total number of key players in cancer drug development. However, there has been a dramatic increase in research into molecular therapies. The Human Genome Project has created a dictionary of the genome, but we can now interrogate it through sophisticated bio-informatic systems. Not only do we have the library, we also have the search tools. We can now predict the three-dimensional structural biology of many proteins and create images of drugs in silico using computers to design small molecules that can then be synthesized in the laboratory to check their activity. A platform approach to drug discovery is creating a massive increase in new candidate molecules for cancer therapy. One of the problems currently is the large numbers of cellular targets that have been identified and to which new drugs can be developed. These targets include growth factors, cell-surface receptors, signal transduction cog molecules, transcription factors, apoptosis-stimulating proteins and cell-cycle-control proteins. Which one to target and invest research funds into is a difficult decision. The total cost of bringing an anti-cancer drug to market exceeds £300 million. Well-defined targets are the starting point on the road to our future treatments. It is likely that classical cytotoxic drugs will continue to be used for the next 25 years, although they will have a declining share of the total marketplace. By 2015, successful molecular targeted approaches will overtake cytotoxics and transform cancer
12 Introduction
medicine (Fig. 1.5). These new drugs will be individualized, chosen on the basis of molecular measurements of the patient’s tumour and normal cells, and taken orally for long periods of time. The classical way in which we develop cancer drugs is split into three phases. In phase I, maximally tolerable doses are determined by gradually escalating the dose in patients with cancer. From this we can determine a workable dose that patients can tolerate and yet is likely to have
a therapeutic effect based on animal studies. We then carry out phase II studies, in which a series of patients with cancers that can be easily measured by X-rays or photographs is given the drug to see what effect it has on their cancer. This allows us to determine the response rate. Phase III is the last and longest phase, in which patients are randomized to receive either the new drug or the best available treatment and their long-term survival is determined (Fig. 1.6).
Base case launch years in the US Breast 2000
2005
2010
2015
2020
2000
2005
2010
2015
2020
2000
2005
2010
2015
2020
2000
2005
2010
2015
2020
Lung
Colorectal
Prostate
Key MAbs Vaccines Anti-angiogenesis Kinase inhibitors
Apoptosis inducers Anti-sense Gene therapy
Figure 1.5 Predicted new drug application (NDA) dates for molecular therapies in the USA. The years 2005–2010 will see an explosion of novel therapies coming into clinical use outside a research setting. PD endpoint on downstream biomarker and MTD determined
Molecular target clinical assay
IHC screen as criterion for entry into phase II/III
I 30 pts
Short term surrogate response for randomization entry using second biopsy or serum test
II 60 pts selected by molecular pathology
III 400 pts selected by molecular pathology and short term surrogates
Mechanism of action and downstream biomarkers Diagnostic kits for patient selection and surrogates via specialist CRO
sNDA approval on surrogate IV alone sNDAs based on molecular pathology and short term response surrogates
Clinical research
Laboratory research
Figure 1.6 In future cancer drugs will be administered to patients accompanied with effective biomarkers. These will lead to novel surrogate markers of response. This will change the dynamic of the phase I, phase II and phase III studies currently used in cancer drug development. PD pharmacodynamic; sNDA supplementary new drug application; CRO contract research organisation; MTD maximum tolerable dose
The future – getting innovation into practice 13
This traditional approach may not be appropriate for many of our new agents. Toxicity may be minimal and effectiveness may be greatest well below the maximally tolerated dose. Furthermore, tumours may not actually shrink but just become static, so no responses are seen. As the new agents have been discovered by measuring their effect on specific molecular targets in the laboratory, it should be feasible to develop the same assay for use in patients. This gives us a short-term pharmacodynamic endpoint and tells us that we are achieving our molecular goals in a patient. Genomic technology has come to our aid. Gene chips allow us to examine the expression of thousands of genes simultaneously before and after administration of the drug. If a second biopsy can be obtained for the tumour, we can compare gene expression patterns in both tumour and normal cells in the same patient after exposure to a new drug. This enables us to get the drug to work in the most effective way. A particularly intriguing approach for the future is to use gene constructs, which signal tiny light pulses when their molecular switches are affected by a drug. We would also like to obtain information about how a drug distributes itself within the body, and ideally to get a picture of the changes it causes in a tumour. Functional imaging allows us to do just this. The aim is to understand the living biochemistry of a drug in the body: we label the drug with a radioactive tracer and then image using PET. Such techniques promise to revolutionize our ability to understand drug activity and to select and improve the way in which we choose anti-cancer drugs for further development. The next decade is likely to be a new golden age for cancer drug discovery, with many novel targeted molecules coming into the clinic. These agents will eventually transform cancer care forever.
THE FUTURE – GETTING INNOVATION INTO PRACTICE The age of the world’s population is rising dramatically. This will increase the total burden of cancer, with many patients living with considerable co-morbidity. At the same time, new technology in many areas of medicine is bringing improvements to the quality and length of life. Major innovations in the following six areas are likely to have the greatest impact on cancer. 1. Molecularly targeted drugs with associated sophisticated diagnostic systems to personalize care. 2. Biosensors to detect, monitor and correct abnormal physiology and to provide surrogate measurements of cancer risk. 3. Our ability to modify the human genome through systemically administered, novel, targeted vectors. 4. The continued miniaturization of surgical intervention through robotics, nanotechnology and precise imaging.
5. Computer-driven interactive devices to help with everyday living. 6. The use of virtual reality systems which, together with novel mood-control drugs, will create an illusion of wellness. Over the last 20 years, a huge amount of fine detail of the basic biological processes that become disturbed in cancer has been amassed. We now know the key elements of growth-factor binding, signal transduction, gene transcription control, cell-cycle checkpoints, apoptosis and angiogenesis. These have become fertile areas to hunt for rationally based anti-cancer drugs. This approach has already led to a record number of novel compounds currently being in trials. Indeed, targeted drugs such as rituximab, trastuzumab, imatinib, sunitinib, sorafenib, bevacizumab and cetuximab are now all in routine clinical use. Over the next decade there will clearly be a marked shift in the types of agents used in the systemic treatment of cancer. Because we know the precise targets of these new agents, there will be a revolution in how we prescribe cancer therapy. Instead of defining drugs for use empirically and relatively ineffectively for different types of cancer, we will identify a series of molecular lesions in tumour biopsies. Future patients will receive drugs that target these lesions directly. The Human Genome Project provides a vast repository of comparative information about normal and malignant cells. The new therapies will be more selective, less toxic and given for prolonged periods of time, in some cases for the rest of the patient’s life. This will lead to a radical overhaul of how we provide cancer care.17 Investment in more sophisticated diagnostics is now required (Table 1.6). Holistic systems such as genomics, proteomics, metabolomics and methylomics provide fascinating clues as to where needles can be found in the haystack of disturbed growth. By developing simple, reproducible and cheap assays for specific biomarkers, a battery
Table 1.6 Cancer diagnostics in drug development Diagnostic Predisposition screen Screen for presence of cancer Pharmacodynamic biomarker Surrogate marker of clinical efficacy Predictive reclassification of disease Patient-specific toxicity prediction
Value Identify patients for chemo prevention Increase in patients with earlier disease Establish pharmacological dose of drug Early indication of proof of concept Target therapy to those likely to respond Avoid adverse events, adjust dose
14 Introduction
of companion diagnostics will emerge.18 It is likely that for the next decade these will be firmly rooted in tissue pathology, making today’s histopathologists essential in moving this exciting field forward. Ultimately, the fusion of tissue analysis with imaging technologies may make virtual biopsies of any part of the body – normal and diseased – a possibility.19 Individual cancer risk assessment will lead to tailored prevention messages and a specific screening programme to pick up early cancer and will have far-reaching public health consequences. Cancer preventive drugs will be developed that will reduce the risk of further genetic deterioration. The use of gene arrays to monitor serum for fragments of DNA containing defined mutations could ultimately develop into an implanted gene chip. When a significant mutation is detected, the chip would signal the holder’s home computer and set in train a series of investigations based on the most likely type and site of the primary tumour. There will be an increase in the total prevalence of cancer as a result of improved survival, as well as change in cancer types to those of older age groups such as prostate cancer which has a longer survival. This will create new challenges in terms of assessing risks of recurrence, designing care pathways, use of information technology (IT) and improving access to services. There will be new opportunities for further targeting and development of existing therapies as experience grows with risk factors over the longer term. Careful monitoring of patient experiences could help in improving results. Cancer could soon be a longterm management issue for many patients who would enjoy a high quality of life even with a degree of chronic illness.20 The funding of cancer care will become a significant problem.21 Already we are seeing inequity in access to the taxanes for breast and ovarian cancer and gemcitabine for lung and pancreatic cancer. These drugs are only palliative, adding just a few months to life. The emerging compounds are likely to be far more successful and their long-term administration considerably more expensive. Increased consumerism in medicine will lead to increasingly informed and assertive patients seeking out novel therapies and bypassing traditional referral pathways through global information networks. It is likely that integrated molecular solutions for cancer will develop, leading to far greater inequity than at present. Cost-effectiveness analyses will be used to scrutinize novel diagnostic technology as well as therapies. Within 20 years, cancer will be considered a chronic disease, joining conditions such as diabetes, heart disease and asthma, conditions that impact on the way people live but will not inexorably lead to death. The model of prostate cancer – many men dying with it rather than from it – will be more usual. Progress will be made in preventing cancers. Even greater progress will be made in understanding the myriad causes of cancer. Our concepts will be different to those of today, and the new ways in which cancer will be detected, diagnosed and treated will be crucial to understanding in the future.
When a cancer does develop, refinements of current technologies and techniques – in imaging, radiotherapy and surgery – together with the availability of targeted drugs will make it controllable. Cure will still be sought, but will not be the only satisfactory outcome. Patients will be closely monitored after treatment, but fear that cancer will definitely kill, which is still prevalent in the early years of the twenty-first century, will be replaced by an acceptance that many forms of cancer are a consequence of old age. Looking into the future is fraught with difficulties. Who could have imagined in the 1980s the impact of mobile phones, the Internet and low-cost airlines on global communication. Medicine will be overtaken by similarly unexpected step changes in innovation. For this reason, economic analysis of the impact of developments in cancer care is difficult. The greatest benefit will be achieved simply by assuring that the best care possible is on offer to most patients, irrespective of their socio-economic circumstances and of any scientific developments. But this is unrealistic. Technologies are developing fast, particularly in imaging and the exploitation of the human genome. Wellinformed patients, with adequate funds, will ensure that they have rapid access to the newest and the best – wherever it is in the world. More patients will benefit from better diagnosis and newer treatments, with greater emphasis on quality of life.22 Innovation will bring more inequality to health. The outcome of the same quality of care differs today between socio-economic groups and will to continue to do so. Clinicians in Europe will continue to be dependent on technologies primarily designed for the major health market in the world – the USA, which currently consumes nearly 55 per cent of cancer medication but contains less than 5 per cent of the world’s population. European legislation covering clinical trials could bring research in the UK to a grinding halt, while ethicists – zealously interpreting privacy legislation – could impose restrictions on the use of tissue. Targeted niche drugs will be less appealing to industry, as the costs of bringing each new generation of drugs to market will not be matched by the returns from current blockbusters. The delivery of innovation will be underpinned by patient expectation. The well informed will be equal partners in deciding the health care they will receive, much of which will take place close to their homes using mechanisms devised by innovative service providers.23 This has huge implications for the training of health professionals and the demarcations between specialties. Emerging technologies will drive the change. Intraprofessional boundaries will blur – doctors from traditionally quite distinct specialties may find themselves doing the same job – and clinical responsibilities will be taken up by health professionals who will not be medically qualified.24 All professionals are likely to find challenges to their territory hard to accept. Box 1.6 shows the challenges that need to be addressed in order to deliver most health benefit.
The future – getting innovation into practice 15
Box 1.6 The challenges of cancer care ● ●
●
●
●
●
●
Increasing the focus on prevention. Improving screening and diagnosis and the impact of this on treatment. New targeted treatments – how effective and affordable will they be? How expectations of patients and their carers will translate into care delivery. Reconfiguration of health services to deliver optimal care. The impact of reconfiguration on professional territories Will society accept the financial burden of these opportunities?
Prevention and screening At the beginning of the twenty-first century, 10 million people in the world develop cancer each year.25 The cause of these cancers is known in roughly 75 per cent of cases: 3 million are tobacco related; 3 million are a result of diet; and 1.5 million are caused by infection. In the UK, 120 000 people die from cancer each year, even though many of these cancers are preventable, a third being related to smoking. But cancer prevention absorbs only 2 per cent of the total funding of cancer care and research. Anti-smoking initiatives are considered to be successful, although it has been 50 years since the association between smoking and cancer was first identified. In the 1960s, 80 per cent of the population smoked; by 2005 the average was under 30 per cent. This masks real health inequality: the percentage of smokers in the higher socio-economic classes is in low single figures, whereas the percentage amongst socio-economically deprived classes is still about 50 per cent in parts of the country. Despite the known risks, if friends and family smoked and there was no social pressure to stop, there was no incentive to do so. Banning smoking in public places will lead to a further drop of about 4 per cent. Increases in tax were a powerful disincentive to smoke, but the price of a packet of cigarettes is so high that smokers turn to the black market: as many as one in five cigarettes smoked is smuggled into the country. Lung cancer, for example, is a rare disease in higher socio-economic groups – it is a disease of poverty. Lessons from anti-smoking initiatives will be instructive for prevention in the future. Although the link between poor diet, obesity and lack of exercise, and cancer has not been confirmed, there is sufficient circumstantial evidence to suggest that strong associations will be found. There will be bans on advertising for crisps, sweets and soft drinks on television, the introduction of a health tax on these products and a ban on the sponsorship of any public event by manufacturers of these products. By 2010, obesity among
the middle classes will be socially unacceptable, but it will remain common among the economically disadvantaged. Creating meaningful, imaginative incentives for people to adopt healthy lifestyles will be a major challenge. The future prevention picture will be coloured by postgenomic research. In 2005, it was accepted that about 100 genes are associated with the development of a whole range of cancers. The detection of polymorphisms in lowpenetrance cancer-related genes – or a combination of changed genes – will identify people at increased risk. Within 20 years, most people will be genetically mapped and the information – gained from a simple blood test – will be easily stored on a smart-card. Legislation will be required to prevent this information being used to determine an individual’s future health status for mortgage, insurance and employment purposes. However, the process of mapping will reveal that every person who has been screened will carry a predisposition to certain diseases – and people will learn to live with risk. Today, the average age of diagnosis of cancer is 68. Improvements in screening, detection and diagnosis will reduce this. A predisposition for some cancers that manifests itself in a patient’s seventies or eighties will be found in young adult life and detected and corrected successfully in the patient’s thirties. Increasing age will remain the strongest risk predictor. Little of what has been described is not happening already in some form, but the computing power of the future will bring accurate calculation of risk, and predictions will take place on an unimaginable scale. Screening programmes will be developed on a national basis if they are simple, robust and cheap. Patients will expect the screening to take place at a venue that is convenient for them – for example in shopping malls – and not be painful or overly time consuming. Health professionals will demand that any programme is accurate and does not give misleading results, and governments will demand that its costs will lead tomore effective use of other resources. Novel providers of risk assessment services are likely to emerge (Box 1.7).
Box 1.7 Balancing cancer risk ●
●
●
●
●
●
Great health inequity exists in smoking-related diseases. Novel prevention strategies are likely to lead to similar inequity. Creating meaningful incentives to reduce risk will be essential. Individually tailored messages will have greater power to change lifestyles. Biomarkers of risk will enhance the validation of cancer preventive drugs. Novel providers of risk assessment and correction will emerge.
16 Introduction
Detecting cancer Cancers are fundamentally somatic genetic diseases that result from several causes: physical, viral, radiation and chemical damage. There are other processes implicated, for example chronic inflammatory change, immuno-surveillance and failure of apoptosis. In the future, cancer will no longer be understood as a single entity; it will be considered to be a cellular process that changes over time. Many diseases labelled as cancer today will be renamed, as their development will not reflect the new paradigm. Patients will accept that cancer is not a single disease and will increasingly understand it as a cellular process. Many more old people will have increased risk or a pre-cancer. This has huge implications for cancer services. Today, most diagnoses of cancer depend on human interpretation of changes in cell structures seen down a microscope. Microscopes will be superseded by a new generation of scanners to detect molecular changes. These scanners will build up a picture of change over time, imaging cellular activity rather than just a single snapshot. We will have the ability to probe molecular events that are markers for early malignant change. This dynamic imaging will lead to more sensitive screening and treatments: imaging agents that accumulate in cells exhibiting tell-tale signs of pre-cancer activity will be used to introduce treatment agents directly.26 Imaging and diagnosis will be minimally invasive and enable the selection of the best and most effective targeted treatment (Box 1.8). Even better imaging will be able to pick up pre-disease phases and deal with them at a stage long before they are currently detectable. These techniques will also be crucial in successful follow-up. A patient who has a predisposition to a certain cancer process will be monitored regularly and treatment offered when necessary. However, not all cancers will be diagnosed in these earliest of stages – some patients will inevitably fall through the screening net. Nevertheless, there will be opportunities to offer less invasive treatment than at present. Surgery and radiotherapy will continue, but in greatly modified form as a result of developments in imaging. Most significantly, surgery will become part of integrated care. The removal of tumours or even whole organs will remain necessary on occasion. However, the surgeon will be supported by three-dimensional imaging, by radiolabelling techniques to guide incisions and by robotic instruments. Although many of the new treatments made possible by improved imaging will be biologically driven, there will still be a role for radiotherapy – the most potent DNA-damaging agent – to treat cancer with great geographical accuracy. The targeting of radiotherapy will be greatly enhanced, enabling treatment to be more precise. In addition to the reconfiguration and merging of the skills of clinicians, the delivery of care will also change. Minimally invasive treatments will reduce the need for long stays in hospital. As more patients are diagnosed with cancer, the provision of care close to where patients live will be both desirable and possible and, as this report will show later, expected. The prospect of highly sophisticated
Box 1.8 Innovation in diagnostics ● ●
●
●
● ●
Radiology and pathology will merge into cancer imaging. Dynamic imaging will create a changing image of biochemical abnormalities. Cancer changes will be detected prior to disease spread from primary site. Greater precision in surgery and radiotherapy will be used for pre-cancer. Molecular signatures will determine treatment choice. Cost control will be essential for healthcare payers to avoid inefficient diagnostics.
scanning equipment and mobile surgical units being transported to where they are required is not unrealistic. Technicians, surgical assistants and nurses would provide the hands-on care, while technical support would be provided by the new breed of clinician – a disease-specific imaging specialist working from a remote site. Cost control will be an essential component of the diagnostic phase. Healthcare payers will create sophisticated systems to evaluate the economic benefits of innovative imaging and tissue analysis technology.
New treatment approaches Future cancer care will be driven by the least invasive therapy consistent with long-term survival. Eradication, although still desirable, will no longer be the primary aim of treatment. Cancers will be identified earlier and the disease process regulated in a way similar to that for chronic diseases such as diabetes. Surgery and radiotherapy will still have a role, but the extent of their role will depend on the type of cancer a patient has and the stage at which the disease is identified, as well as on how well the drugs being developed today perform in the future. Cancer treatment will be shaped by a new generation of drugs (Box 1.9). What this new generation will look like will critically depend on the relative success of agents currently in development. Over the next 3–5 years, we will understand more fully what benefits compounds such as kinase inhibitors are likely to provide. It is estimated that there are about 500 drugs currently being tested in clinical trials. Of these, around 300 inhibit specific molecular targets.27 But this number is set to rise dramatically: 2000 compounds will be available to enter clinical trials by 2007 and 5000 by 2010. Many of these drug candidates will be directed at the same molecular targets, and industry is racing to screen those most likely to make it through to the development process. Tremendous pressures are coming from the loss of patent protection from the majority of high-cost chemotherapy drugs by 2008. Unless new
The future – getting innovation into practice 17
premium-priced innovative drugs are available, cancer drug provision will come from global generic manufacturers currently gearing up for this change.
Box 1.9 Drivers of molecular therapeutics ● ● ● ● ● ● ●
Human Genome Project and bioinformatics Expression vectors for target production In-silico drug design Robotic high throughput screening Combinatorial chemistry Platform approach to drug discovery Huge increase in number of molecular targets
So what will these drug candidates look like? Small molecules are the main focus of current research, most of which are designed to target specific gene products that control the biological processes associated with cancer such as signal transduction, angiogenesis, cell-cycle control, apoptosis, inflammation, invasion and differentiation. Treatment strategies involving monoclonal antibodies, cancer vaccines and gene therapy are also being explored. Although we do not know exactly what these targeted agents will look like, there is growing confidence that they will work. More uncertain is their potential overall efficacy at prolonging survival. Many could just be expensive palliatives. In future, advances will be driven by a better biological understanding of the disease process. Already we are seeing the emergence of drugs targeted at a molecular level – trastuzomab, directed at the HER2 protein, imatinib, which targets the Bcr-Abl tyrosine kinase, and gefitinib and erlotinib, directed at epidermal growth factor receptor (EGFR) tyrosine kinase. These therapies will be used across a range of cancers. What will be important in future is whether a person’s cancer has particular biological or genetic characteristics. Traditional categories will continue to be broken down and genetic profiling will enable treatment to be targeted at the right patients. Patients will understand that treatment options are depen-dent on their genetic profile, and the risks and benefits of treatment will be much more predictable than today. Therapies will emerge through our knowledge of the human genome and the use of sophisticated bio-informatics. Targeted imaging agents will be used to deliver therapy at screening or diagnosis. Monitoring cancer patients will also change as technology allows the disease process to be tracked much more closely. Treatment strategies will reflect this, and drug resistance will become much more predictable. Biomarkers will allow those treating people with cancer to assess whether a drug is working on its target. If it is not, an alternative treatment strategy will be sought. Tumour regression will become less important as clinicians look for molecular patterns of disease and its response.
There will be more of a focus on therapies designed to prevent cancer. A tangible risk indicator and risk-reducing therapy along the lines of cholesterol and statins would allow people to monitor their risk and seek intervention. Delivering treatment early in the disease process will also be possible because subtle changes in cellular activity will be detectable. This will lead to less aggressive treatment. The role of industry in the development of new therapies will continue to change. Smaller, more specialized companies linked to universities will increasingly deliver drug candidates and innovative diagnostics to the large commercially driven multinational pharmaceutical companies who will market them globally. People will be used to living with risk and will have much more knowledge about their propensity for disease. Programmes will enable them to determine their own predisposition to cancer. This in turn will encourage health-changing behaviour and will lead people to seek out information about the treatment options available to them. Patients will also be more involved in decision-making as medicine becomes more personalized. Indeed, doctors may find themselves directed by well-informed patients. This, and an environment in which patients are able to demonstrate choice, will help drive innovation towards those who will benefit. However, inequity based on education, wealth and access will continue (Box 1.10).
Box 1.10 The uncertainty of novel drugs for cancer ●
●
●
●
●
●
Will the new generation of small molecule kinase inhibitors really make a difference or just be expensive palliation? How will big pharma cope with most high-value cytotoxics becoming generic by 2008? Can expensive late-stage attrition really be avoided in cancer drug development? How will sophisticated molecular diagnostic services be provided? Will effective surrogates for cancer preventive agents emerge? Will patient choice involve cost considerations in guiding therapy?
Barriers to innovation Innovation in cancer treatment is inevitable. However, there are certain prerequisites for the introduction of new therapies. First, innovation has to be translated into usable therapies. These therapies must be deliverable, to the right biological target, and to the right patient in a way that is acceptable to the patient, healthcare professional and society. Innovation must also be marketed successfully
18 Introduction
Table 1.7 Marketed targeted therapies Drug Herceptin Mabthera Glivec Erbitux Avastin Tarceva Iressa
Generic
Manufacturer
Yearly cost
Traztuzumab Rituximab Imatinib Cetuximab Bevacizumab Erlotinib Gefitinib
Roche Roche Novartis BMS Genentech Roche AZ
£60 K £40 K £50 K £60 K £70 K £65 K £40 K
so that professionals, patients and those picking up the cost understand the potential benefits. Those making the investment in research will inevitably create a market for innovation even if the benefits achieved are minimal. The explosion of new therapies in cancer care is going to continue, and pricing of these drugs will remain high. The cost of cancer drugs in 2005 is estimated to be $24 billion globally, of which $15 billion is spent in the USA. If effective drugs emerge from the research and development pipeline, the cancer drug market could reach $300 billion globally by 2025, with this cost spreading more widely around the world (Table 1.7). But parallel to this explosion in therapies and increase in costs, a number of confounding factors will make markets smaller. The technology will be available to reveal which patients will not respond to therapy, so making blockbuster drugs history. Doctors will know the precise stage of the disease process at which treatment is necessary, and as cancer transforms into a chronic disease, people will have more co-morbidities, which will bring associated drug–drug interactions and an increase in care requirements (Box 1.11).
that pharmaceutical companies will stop developing drugs for cancer and focus instead on therapeutic areas where there is less individual variation and therefore more scope for profit. Furthermore, development costs are rising. Ten years ago, the average cost of developing a new cancer drug was around $400 million; now it is $1 billion. At this rate of growth, the cost of developing a new drug could soon reach $2 billion, an amount unsustainable in a shrinking market. With this in mind, the process of developing drugs needs to be speeded up. However, instead of research being made simpler, changes in legislation concerned with privacy and prior consent are making it more difficult. The EU Clinical Trials Directive will make quick hypothesis-testing trials impossible. Other challenges exist as well, such as the need to obtain consent for new uses of existing human tissue, following political anxiety when consent for removing and storing tissues had not been obtained in the early years of the twenty-first century. However, surveys have shown that patients who gave consent for tissue to be used for one purpose were happy for it to be used for another; they do not wish to be reminded of their cancer years later. To overcome these constraints, regulators will have to start accepting surrogate markers rather than clinical outcomes when approving therapies. Outcome studies may well move to post-registration surveillance of a drug’s efficacy, similar to cholesterol-lowering agents today. The rise of personalized medicine will mean that the temptation to over-treat will disappear. Doctors and patients will know whether a particular treatment is justified. The evidence will be there to support their decisions. As a consequence of this, treatment failure – with all its associated costs – will be less common.
The patient’s experience Box 1.11 Barriers to innovation ●
●
●
●
●
●
The drug industry will continue to compete for investment in a competitive, capitalist environment. Blockbuster drugs drive profit – niche products are unattractive in today’s market. Personalized therapies are difficult for today’s industry machine. Surrogate endpoints will be essential to register new drugs. Novel providers will emerge providing both diagnostic and therapy services. Payers will seek robust justification for the use of high-cost agents.
How do we balance this equation? The pharmaceutical companies will not necessarily want to do the studies to fragment their market; research leading to rational rationing will need to be driven by the payers of health care. There is a risk
Two separate developments will determine the patient’s experience of cancer care in future. Increasing expectations of patients as consumers will lead health services to become much more responsive to the individual, in the way that other service industries have already become. Targeted approaches to diagnosis and treatment will also individualize care. People will have higher personal expectations, be less deferential to professionals and more willing to seek alternative care providers if dissatisfied. As a result, patients will be more involved in their care; they will take more responsibility for decisions rather than accepting a paternalistic ‘doctor knows best’ approach. This will be fuelled partly by the Internet and competitive provider systems. By 2025, the overwhelming majority of people in their seventies and eighties will be familiar with using the Internet to access information through the massive computing power that they will carry personally. With patients having access to so much health information, they will need someone to interpret the huge volumes available and to help them assess the risks and benefits as
The future – getting innovation into practice 19
well as to determine what is relevant to them. These patient brokers will be compassionate but independent advocates who will act as patients’ champions, guiding them through the system. They will be helped by intelligent algorithms to ensure patients understand screening and the implications of early diagnosis, and they will spell out what genetic susceptibility means and guide patients through the treatment options. Patients and health professionals will have confidence in computer-aided decision-making because they will have evidence that the programs work. How the service will be designed around patients’ needs and expectations will be determined by the improved treatments available and their individualization (Box 1.12). When cancer centres developed in the mid-twentieth century, the diseases were relatively rare, and survival was low. Although distressing for patients when they were referred to a centre, their existence concentrated expertise. Cancers will be commonly accepted chronic conditions and therefore even when inpatient care is required, patients will be able to choose from many places in the world where they will receive care at a ‘cancer hotel’. But for many patients even that option will not be necessary: most new drugs will be given orally, so patients will be treated in their communities.28 However, this approach to cancer and other concomitant chronic conditions will place a huge burden on social services and families. Systems will be put in place to manage the ongoing control of these diseases and conditions – psychologically as well as physically. Pain relief and the control of other symptoms associated with cancer treatment will be much improved.
Box 1.12 Experiencing cancer in future ●
● ● ●
●
Patient brokers will guide people with cancer through the system. Choice will be real and will involve cost decisions. Patients will make a contribution to their care costs. Complementary therapies will be widely available and well regulated. Themed death chosen by patients will be possible.
Today, 70 per cent of the cancer budget in the United States is spent on care associated with the last 6 months of people’s lives. Although many recognize that such treatment has more to do with the management of fear than with the management of cancer, medical professionals have relatively few treatment options available and there has been limited awareness of which patients would benefit. There is also an institutional reluctance to destroy patients’ hopes, which led to confusion between the limits of conventional medicines, and a reluctance to face the inevitable – by patients and their families and doctors. There is a widespread perception that if patients are continuing to be offered anti-cancer treatment, there is a possibility that their health might be restored.
With better treatments, consumers of services will be able to focus on quality of life, and much of the fear now associated with cancer will be mitigated. Demand for treatments with few side effects or lower toxicity will be high, even if there are only quite modest survival gains. The transition between active and palliative care is often sudden, but in future, because patients will be in much greater control of their situation, the change in gear will not be as apparent.29
Professional reconfiguration One of the greatest challenges to providing the best cancer care in future will be having the right people in the right jobs. It will be essential not to continue to train people for jobs that will no longer exist. Policy makers have begun to grasp some of the workforce difficulties that lie ahead, and there are moves to ensure that healthcare professionals have responsibilities commensurate with their level of education and professional skills. Nurses and pharmacists are being encouraged to take over some responsibilities that have been held firmly by doctors, such as prescribing, while their traditional roles have been handed on to technicians and other support staff.30 The appropriate skill mix will become even more critical (Box 1.13). Barriers between healthcare professions will have to be broken down in order for the new approaches to the care of patients with cancer and many other diseases to be delivered. The work of pathologists and radiologists will become one, as their traditional skills are augmented by the new generation of diagnostic and treatment devices. Oncologists will find that many forms of chemotherapy will be delivered with the aid of the new technology, and surgeons will be using robots to enable them to operate. Fewer of the most highly trained specialists will be required, since much of their responsibility will be delegated to specialist technicians and nurses working to protocols. In addition, the most highly trained individuals will be able to work at a number of sites on the same day, since the technology will be mobile and skills will be used remotely. The balance between skills will be driven by a number of factors: the size of the medical workforce and the capacity of the system to provide care, as well as the availability of trained support staff.31
Box 1.13 The right person for the right job – key challenges ● ● ●
● ●
Manpower planning for new technology. Doctors and other healthcare specialists. Prescribing cancer drugs by nurses, pharmacists and others. Training carers for elderly people with co-morbidities. Making patients equal partners in decision-making.
20 Introduction
CONCLUSION Cancer will become incidental to day-to-day living. Cancers will not necessarily be eradicated, but that will not cause patients the anxiety that it does today. People will have far greater control over their medical destinies. Patients in all socio-economic groups will be better informed. In addition, surgery and chemotherapy will not be rationed on grounds of age, since all interventions will be less damaging – psychologically as well as physically. How true this picture will be will depend on whether the technological innovations emerge. Will people, for example, really live in ‘smart houses’ where their televisions play a critical role in monitoring their health and well-being? It is also dependent on healthcare professionals working alongside each other, valuing the input of carers who, even more than today, will provide voluntary support because of the number of people in older age groups compared with those of working age. The reality for cancer care may be rather different: the ideal may exist for a minority of patients, but the majority may not have access to the full range of services. Old people, having been relatively poor all their lives, may suffer from cancer and a huge range of co-morbidities that will limit their quality of life. Looking after them all – rich and poor – will place great strains on younger people: will there be enough of them to provide the care? As with all health issues, the question of access will be determined by cost and political will. In 2005 a cancer patient consumes about £25 000 worth of direct medical care costs, with 70 per cent being spent in the last 6 months of life. Conservatively, with patients living with cancer, rather than dying from it, and with access to new technologies, this could reach £100 000 per patient per year by 2025. Table 1.7 shows the annual cost of currently marketed targeted therapies. In theory, cancer care could absorb an ever-increasing proportion of the healthcare budget. Would this be a reflection of what patients want? Probably ‘yes’. Surveys reveal that
three-quarters of the UK population believe cancer care should be the NHS priority, with no other disease area coming even a close second. But to achieve that expenditure – and assuming that part of the health service will be funded from taxation – the tax rate might have to rise to 60 per cent. Inevitably, there will be conflicting demands on resources: the choice may be drugs or care costs. How are the costs computed? Although the technology will be expensive, it will be used more judiciously since it will be better targeted. Another argument suggests that when patients are empowered they use less and fewer expensive medicines, in effect lowering the overall costs. An extension of that argument is that although costs will increase for treating each individual patient, the overall costs will decrease because more care will be delivered at home. But because people will live longer, the lifetime costs of cancer care will rise along with co-morbidity costs. Politicians will be faced with a real dilemma: if the prevalence of cancer increases, the cost of delivering innovative care could be massive. Will cancer care need to be rationed in a draconian way? One dilemma for the future will be the political power of old people. More will be living longer and their chronic problems will not necessarily incapacitate them physically or mentally. This educated gerontocracy will have high expectations that will have been sharpened through the first two decades of the twenty-first century and they will not tolerate the standards of care now offered to many old people. They will wield considerable influence. Will a tax-based health system be able to fund their expectations? Politicians will have to consider the alignment between patients’ requirements and the wishes of taxpayers and voters. Fewer than 50 per cent of voters now pay tax, and the percentage of tax-paying voters is set to fall as the population ages. Will the younger taxpayers of the future tolerate the expensive wishes of non-taxpayers? The interests of voters may be very different from the interests of taxpayers. It seems
INNOVATION
SOCIETY
prevention screening diagnosis surgery radiotherapy drugs supportive care
willingness to pay expectation economy selfishness spirituality family integrity ethics political ideology
DELIVERY hospital-hotel specialist-primary care-DTC professionals role public v private globalization
THE CANCER FUTURE
FINANCE self pay co–payment optional insurance mandatory insurance state insurance HMO NHS charity
Figure 1.7 The four components of the future of cancer are innovative technologies, societal changes, the delivery infrastructure and the financial mechanisms. DTC direct to consumer; HMO Health maintenance organisation; NHS National Health Service (UK)
References 21
likely, therefore, that the days of an exclusively tax-funded health service are numbered. Co-payments and deductibles will be an inevitable part of the new financial vocabulary. Figure 1.7 shows the four components of cancer’s future: innovation, delivery, finances and society. Whatever system is put in place, there is the prospect of a major socio-economic division in cancer care. A small percentage of the elderly population will have made suitable provision for their retirement, in terms of both health and welfare, but the vast majority will not be properly prepared. Policy-makers need to start planning now, as they are doing for the looming pensions crisis. The most productive way forward is to start involving cancer patient and health advocacy groups in the debate, to ensure that difficult decisions are reached by consensus. Societal change will create new challenges in the provision of care. A decline in hierarchical religious structures, a reduction in family integrity through increasing divorce, greater international mobility and the increased selfishness of a consumer-driven culture will leave many lonely and with no psychological crutch to lean on at the onset of serious illness. There will be a global shortage of carers – the unskilled, low-paid but essential component of any health delivery system. The richer parts of the world are now harnessing this from the poorer, but eventually the supply of this precious human capital will evaporate. New financial structures will emerge with novel consortia from the pharmaceutical, financial and healthcare sectors enabling people to buy into the level of care they wish to pay for. Cancer, cardiovascular disease and dementia will be controlled and will join today’s list of chronic diseases such as diabetes, asthma and hypertension. Hospitals will become attractive health hotels run by competing private sector providers. Global franchises will provide speciality therapies through these structures similar to the internationally branded shops in today’s malls. Governments will have long ceased to deliver care. Britain’s NHS, one of the last centralized systems to disappear, will convert to UK Health – a regulator and safety net insurer – by the end of this decade. The ability of technology to improve cancer care is assured. But this will come at a price: the direct costs of providing it and the costs of looking after the increasingly elderly population it will produce. We will eventually simply run out of things to die from. New ethical and moral dilemmas will arise as we seek the holy grail of compressed morbidity. Living long and dying fast will become the mantra of twenty-first-century medicine. Our cancer future will emerge from the interaction of four factors: the success of new technology, society’s willingness to pay, future healthcare delivery systems and the financial mechanisms that underpin them (Fig. 1.7).
REFERENCES 1 Sikora K (ed.). The future of cancer care. Expert Rev Anticancer Ther 2004; 4. 1–78
2 Holtzman NA, Shapiro D. Genetic testing and public policy. Br Med J 1998; 316:852–6. 3 World Health Organisation. The World Health Report. Geneva: WHO, 2000. 4 Murray CT, Lopez AD. The Global Burden of Disease. Boston: Harvard University Press, 1996. 5 WHO Executive Board, EB 102. WHO: Geneva, 1998. 6 Parkin D, Whelan S, Ferlay J, Raymond L, Young J. Cancer in Five Continents, Vol. VII. Lyon: International Agency for Research on Cancer, Scientific Publications, 1997. 7 Cancer in Five Continents. Electronic Database for Cancer. Lyon: IARC/WHO, 1998. 8 Doll R, Peto R. The Causes of Cancer. Oxford: Oxford University Press, 1985. 9 Taylor P. Smoke Ring – The Politics of Tobacco. London: Bodley Head, 1984. 10 Riboli E, Kaaks R. European perspective investigation into cancer and nutrition. Int J Epidemiol 1997; 26(Suppl. 1):6–14. 11 Pisani P, Parkin DM, Munoz N, Ferlay J. Cancer and infection: estimates of the attributable fraction in 1990. Cancer Epidemiol, Biomarkers and Prevention 1997; 6:387–400. 12 Chang M-H, Chen CJ, Lai MS, et al. Universal hepatitis B vaccination in Taiwan and the incidence of hepatocellular carcinoma in children. Taiwan Childhood Hepatoma Study Group. N Engl J Med 1997; 336:1855–9. 13 Monsonego J, Franca E. Cervical Cancer Control. General Statements and Guidelines. Paris: EUROGIN, 1996. 14 Sikora K. Cancer screening Cancer Screening Med 1999; 27:35–9. 15 Sankaranarayanan R, Wesley R, Somanthan T, et al. Visual inspection of the uterine cervix after the application of acetic acid in the detection of cervical carcinoma and its precursors. Cancer 1998; 83:2150–6. 16 Goodman L, Wintrobe M, Dameshek W, Goodman M, Gilman A, Mclennan M. Nitrogen mustard therapy. JAMA 1946; 132:126–32. 17 2020 Vision. Our Future Healthcare Environments. Norwich: The Stationery Office, 2003. 18 Nicolette CA, Miller GA. The identification of clinically relevant markers and therapeutic targets. Drug Discovery Today 2003; 8:31–8. 19 Adam BL, Qu Y, Davis JW, et al. Serum protein fingerprinting coupled with a pattern-matching algorithm distinguishes prostate cancer from benign prostate hyperplasia and healthy men. Cancer Res 2002; 62:3609–14. 20 Tritter JQ, Calnan N. Cancer as a chronic illness? Reconsidering categorisation and exploring experience. Eur J Cancer 2002; 11:161–5. 21 Bosanquet N, Sikora K. The Economics of Cancer Care. Cambridge: Cambridge University Press, 2006. 22 Symonds RP. Radiotherapy. BMJ 2001; 323:1107–10. 23 Wanless D. Securing Good Health for the Whole Population. London: Department of Health, 2003. 24 Laing A. Meeting patient expectations: healthcare professionals and service re-engineering. Health Services Management Res 2002; 15:165–72. 25 World Cancer Report. Lyons: IARC Press, 2003.
22 Introduction
26 Watters JW, McLeod HL. Cancer pharmacogenomics: current and future applications. Biochim Biophys Acta 2003; 1603:99. 27 Blackledge G. Cancer drugs: the next ten years. Eur J Cancer 2003; 39:273. 28 Brumley RD. Future of end of life care: the managed care organisation perspective. J Palliat Med 2002; 5:263–70.
29 Melzer D. My Very Own Medicine: What Must I Know? Cambridge: Cambridge University Press, 2003. 30 Locock L. Redesigning health care: new wine from old bottles? J Health Serv Res Policy 2003; 8:120–2. 31 Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, National Academy Press, 2001.
2 Molecular biology HANI GABRA, EUAN STRONACH AND ROHINI SHARMA
Introduction Molecular biology: the central dogma A molecular genetic basis to cancer
23 24 25
INTRODUCTION Molecular biology is no longer a revolution in medicine. It is now the very fabric of clinical practice and this is particularly so in the practice of oncology. Understanding of the molecular processes underlying cancer, whether for prevention, diagnosis, treatment or palliation, is essential for the practitioner. The purpose of this chapter is twofold: first, to describe the concepts and language underlying this discipline, and second, to integrate these concepts into a clinical overview of where molecular biology currently impacts cancer treatment. Oncogenesis is a process of molecular dysfunction from the first initiation of inherited or somatic events through clonal selection and expansion by accumulated multi-step progression of genomic, transcriptional, translational and post-translational events to the clinical entity that we recognize as cancer. Reductionism and scientific method are the cornerstones of progress in this disease, and whilst psychosocial issues are important, reduction of the incidence of and mortality from cancer will require a continued scientific approach grounded in a molecular understanding of the disease. Cancer can be regarded as a disease of genes. The initiating genetic reduction can be inherited in a Mendelian fashion or acquired, and these principal events initiate cancer as a process. Subsequently, the process of oncogenesis is driven (promoted) by further somatic (non-inherited) events, which may be modified by heritable components (modifier polymorphisms) that are subject to non-Mendelian complex genetics that interact with the environment. Clonal development of cancer is essentially a Darwinian process of natural selection.
Molecular profiling Future impact of molecular biology in cancer References
29 35 36
The molecular biology of cancer demonstrates that many fundamental cellular processes are disrupted, and so the biology of cancer has been hugely informative for elucidating biology itself and therefore for informing other areas of human health. However, even as we engage in reductionism and analysis, an opposite synthetic process is occurring to inform our understanding of cancer. This process has been driven by evolving technologies that allow us to look at whole biological compartments simultaneously during a change of state or phenotype in the cancer. Thus it is possible to look at the entire genome (genomics), the entire RNA compartment (transcriptomics), the entire protein compartment (proteomics), or the sum of metabolic endpoints in the cancer state (metabonomics or metabolomics). These ‘omics’ technologies represent a huge challenge in cancer research, but the potential for diagnostics, prognostics and therapeutics is profound. However, the problem of the meaningful analysis of huge datasets brings enormous challenges for the field of bio-informatics, and it will take empirical development of these (and novel) analysis methodologies in rigorously defined large prospective patient cohorts to make sense and full use of this approach. Systematic analysis of particular biochemical compartments is also very informative, particularly for molecules involved in cell signalling (kinome) and for cell-surface molecules altered by aberrant sugar residues (glycomics). The rebuilding of these complex networks to demonstrate the impact of cellular changes and perturbations in biology and cancer is known as systems biology, and this is expected to be an area of great growth in the biology, and especially the molecular biology, of cancer.
24 Molecular biology
This chapter summarizes the principal components of the molecular biology of cancer and gives examples of the importance of these components in clinical practice.
MOLECULAR BIOLOGY: THE CENTRAL DOGMA The description of the structure of DNA in 1953 by Watson and Crick began the most profound revolution in our understanding of biology. The recognition that DNA replicates semi-conservatively suddenly opened up our understanding of the genetic basis of inheritance and of life. Another profound insight came from the understanding of transcription – that is, how the process of gene expression commences by the generation of messenger RNA, along with ribosomal RNA and transfer RNA to form basic components of the ribosome where the translation of RNA into protein occurs. This process of DNA replication, RNA transcription and protein translation is referred to as the central dogma and forms the most important basic principle of biology.
DNA and its replication DNA is composed of a long double-stranded chain of a sugar, deoxyribose, linked by phosphate on its 5’ carbon to the next deoxyribose on its 3’ carbon. This phosphatelinking process extends the sugar backbone into long chains. The chain can pair with another DNA chain utilizing one of four nitrogenous bases – adenine (A), thymine (T), guanine (G), cytosine (C) – that connect to deoxyribose via the 1’ carbon. These bases recognize a complementary base on the opposite strand, thus allowing a double helix to form. Non-covalent bonding between adenine with thymine and guanine with cytosine occurs to efficiently create this double-helix structure. The two strands of the double helix run in opposite orientations: 5’ to 3’ in each direction. Since DNA is extremely long, it is compacted into nucleosomes, protein complexes around which DNA is wound. Nucleosomes are composed of histone proteins, with around 146 bases of DNA wound around each histone H1 protein with a linker of variable size of up to 100 nuceotides.
RNA and its transcription RNA is typically a single-stranded molecule with a ribose (rather than deoxyribose) chain. The main forms of RNA are messenger RNA (mRNA), which carries the genetic information for proteins, transfer RNA (tRNA), which mediates the attachment of amino acids during translation, and ribosomal RNA (rRNA), which helps form the ribosome. The synthesis of mRNA is followed by a cap on the 5’ end and a polyA tail at the 3’ end. This approach extends to the removal of introns and splicing together of exons to produce the mature mRNA. Messenger RNA is the RNA
generally considered in the central dogma, and it is this type of RNA that communicates the information content from the gene in the DNA to its translation as a protein. mRNA makes up only a small proportion of the cell’s RNA content; more RNA is actually committed to rRNA and tRNA. Ribosomal RNA is a crucial component, together with ribosomal proteins of the ribosomes where translation of mRNA into protein occurs. A single precursor RNA is processed into three ribosomal RNAs, designated by size as 28S, 18S and 5.8S. Transfer RNAs are small RNAs with a high degree of secondary structure and which have an amino acid attached to them. Transfer RNAs recognize the triplet codons on mRNA via an anticodon complementary triplet and therefore bring the correct amino acid to the ribosomal machinery at the correct position to extend the protein chain correctly. Transcription is initiated by RNA polymerases on the DNA template to form RNA transcripts. The efficiency of RNA transcription is determined by transcriptional promoters, usually located upstream of transcriptional initiation sites (although not invariably so). These promoter regions are GC rich, and in cancer can be subject to aberrant methylation, which results in transcriptional repression. This so-called somatic methylation is an epigenetic modification in cancer that is a powerful and common mechanism of activating tumour suppressor genes or negative regulators of cancer.
Protein and its translation Proteins are the effector molecules of life. They are the functional industrial end product of the information content encoded by nucleic acids. They deliver almost all enzymatic and structural processes, including storage, motion and structural support (along with lipids and sugars). Their functional diversity is generated by alternative sequence combinations of amino acids, of which there are 20. The amino acids are linked together during translation by peptide bonds between the carboxyl group of one amino acid and the amino group of the next amino acid. The primary sequence structure of a protein gives rise to secondary structure of folding of defined regions called domains in sheets or helices. These then take up an overall conformation of the protein, the so-called tertiary structure. Proteins can then associate into complexes, making up quaternary structures. The process of translation involves distinct phases of initiation of the polypeptide chain, then its elongation, and finally termination. Initiation at the AUG (start) codon locates an initial methionine, and subsequent codon recognition incorporates amino acids into the growing chain. The tRNA separates from its amino acid and the next tRNA occupies the site of nascent synthesis. Finally there is a stop codon to terminate the protein chain. The central dogma is therefore a rich and complex system process of high fidelity based on information content recognition and flow. Mistakes in this process represent the primary lesion in cancer, and so in that sense cancer
A molecular genetic basis to cancer 25
can be considered as a disease of dysfunctionality of the central dogma, usually in some way either genetically or epigenetically disrupting DNA itself.
A MOLECULAR GENETIC BASIS TO CANCER The major future advances in the treatment of cancer will almost certainly centre around strategies that take our current understanding of cancer biology and turn it upon the disease in the form of targeted therapeutics either singly or, more likely, sequentially or in combination with other conventional or targeted therapeutic approaches. To understand how such strategies are working now and where this approach may lead us we must view cancer at the molecular level. Although it is now clear that cancer is the result of defects in the genetic machinery within the cells that make up our bodies, this conclusion was slow to emerge. Epidemiological studies identified strong associations between certain cancers and particular exposures or lifestyles. For example, a high incidence of scrotal cancer was observed in men who had worked as chimney sweeps; the use of X-rays was associated with skin cancer and leukaemia; and smoking was linked to lung cancer.1–3 This and other evidence suggested an environmental cause for cancer as opposed to a random tissue breakdown, as had been thought previously. This theory was strengthened by work that showed that chemicals present in the soot to which chimney sweeps were exposed caused tumours in animals. Later it was shown that both X-rays and certain carcinogenic chemicals caused heritable damage, leading to the suggestion that cancer may be caused by alterations to the base structure of DNA. This was further substantiated by the Ames test, which revealed that previously identified carcinogens were also potent inducers of DNA mutation in bacteria.4 The definitive proof that carcinogens caused genetic mutations resulting in cancer would, however, require the mutant genes to be found in tumour samples. Interestingly, this evidence came from research suggesting that cancer was the result of infection with tumour-causing viruses – research that at the time seemed to be in opposition to the carcinogen hypothesis.
Oncogenes Peyton Rous demonstrated in 1909 that viral particles from the tumour of one chicken could induce new tumours in other chickens, and subsequent research identified several other such examples of tumour-causing viruses.5 These viruses were linked by their ability to induce uncontrolled growth in the infected cells of the host. It was later revealed that the tumour virus genomes all contained an additional gene when compared to their non-tumour-causing counterparts and that these genes, when isolated, promoted cell growth.6 This seemed to contradict the carcinogen theory,
which suggested that damage to the endogenous genome was responsible for tumour formation and not the addition of a foreign gene. These two theories were ultimately united by the observation that the growth-promoting viral ‘oncogenes’ from the viruses were mirrored by normal genes already present in our own genomes.7 A new theory was thus proposed, for which there is now good evidence, that the genomic mutations cased by carcinogens may affect the behaviour of these endogenous ‘proto-oncogenes’, causing them to promote uncontrolled cellular proliferation in exactly the same manner as their viral counterparts. Sequencing analysis of one of these genes, the ras oncogene, implicated in bladder cancer, revealed that the normal proto-oncogene and the cancer-causing oncogene differed only by a single nucleotide base, changing one critical amino acid in the translated protein.8 The result of this is a constitutive activation of the gene product and consequent acceleration of the cell cycle. Other oncogenes have been shown to be amplified (e.g. myc, PI3K) or genomically rearranged such that they are under the control of the wrong genetic elements (e.g. Bcr–Abl gene fusion).9,10 A comprehensive review of known oncogenes is beyond the scope of this text; however, the function of a few, in the context of intracellular signalling pathways aberrant in cancer, will be discussed later. The discovery of oncogenes was a revelation in our understanding of cancer at the molecular level. However, it rapidly became clear that oncogenic mutations were not sufficient to explain the complexity of cancer.
Tumour suppressor genes Epidemiological studies revealed that the incidence of most cancers rises exponentially with age, which suggests a multi-step process in which several distinct events must occur for a cancer to result. Consequently, the simple idea that a single oncogenic mutation could explain the development of cancer had to be modified. However, mutations in multiple oncogenes within a single tumour were very rarely encountered, suggesting that further cancer genes remained to be identified. A second set of cancer genes, called tumour suppressor genes, were postulated as a result of experiments in which normal and tumour cells were fused together. Surprisingly, these experiments revealed that the fusion cells did not grow as tumours, as was expected, but rather that some component of the normal cells’ genome, presumably absent from the tumour cells, acted to prevent this inappropriate growth.11 Ongoing research has now identified several tumour suppressor genes that have been found to be mutated or inactivated, by varied and unexpected mechanisms, in well-described hereditary cancers and in sporadic tumours (Table 2.1). The study of these genes has provided a deeper understanding of fundamental cancer biology and, indeed, of the normal processes that regulate growth, differentiation and cellular biology.
26 Molecular biology
KNUDSON’S TWO-HIT HYPOTHESIS
The identification of tumour suppressor genes allowed some peculiarities surrounding sporadic and inherited tumours to be explained and led Alfred Knudson to propose his famous ‘two-hit hypothesis’. It was known that in rare, sporadic cases of retinoblastoma, single, unilateral tumours tended to be seen, whereas the inherited form of the disease is typified by multiple, bilateral tumours. Knudson reasoned that in the familial form of retinoblastoma, one of the two cellular copies of a tumour suppressor gene carried an inherited mutation and as such was mutated in every cell in the body. He suggested that a tumour would arise subsequent to any retinal cell sustaining damage to the second, normal copy of the gene – an event that could occur relatively frequently, thus leading to multiple bilateral tumours. Conversely, in the sporadic form of the disease, both copies of the gene would have to be mutated de novo in a single cell in order for a tumour to form. This would be a far less likely scenario and therefore the observed rare unilateral tumours would be the expected phenotype.12 The tumour suppressor gene responsible for retinoblastoma, RB, has since been identified and its protein product, Rb, has been shown to have a pivotal role in the control of the cell cycle, functioning as an inhibitor of another gene, E2F, which has powerful growth stimulatory effects.13 This two-hit model of cancer predisposition has since been found to hold true for a large number of other tumour suppressors and cancer syndromes (see Table 2.1).
Stability genes The final broad category of cancer genes that have been described to date are the so-called stability genes. Unlike the oncogenes and tumour suppressor genes, which tend to have functions in the control of cell growth and differentiation, the normal function of the stability genes is in the maintenance of genomic integrity. Exposure to a wide variety of cellular and environmental agents such as reactive oxygen species, ultraviolet radiation or cigarette smoke causes considerable damage to our genomic DNA. In addition, the basic process of DNA replication, which must occur at
each cell division, is inherently error prone, and therefore without complex DNA-repair mechanisms we would rapidly accumulate mutations within our cells. Studies of several cancer syndromes such as hereditary non-polyposis colon cancer (HNPCC), ataxia telangiectasia and xeroderma pigmentosum revealed that the predisposing genes were not typical growth-controlling tumour suppressor genes or oncogenes, but stability genes, the absence of which results in an increased rate of genetic mutation.14 HNPCC results from defects in any one of a number of DNA mismatch repair genes; these genes control the process by which mispaired stretches of DNA, resulting from errors in DNA replication, are repaired.15 The consequence of such defective repair mechanisms is, of course, an increased rate of accumulation of mutations, which in time will result in defects affecting growth-controlling oncogenes and tumour suppressor genes – the end result being an increased incidence of tumours.
Cancer cell biology The identification of the many and varied cancer genes outlined very briefly earlier has been critically important to our understanding of cancer. However, in isolation the information that can be obtained from any one gene is limited. Through understanding the pathways these genes are involved in and the subtle interplay that occurs between pathway components, we are truly beginning both to realize fully the complexity of the problem and to identify new and unexpected ways to combat it. THE CELL CYCLE
The cell cycle describes the typical states in which a viable, somatic cell can exist and follows the orderly sequence: gap phase 0/gap phase 1 (G0/G1), synthesis phase (S), gap phase 2 (G2) and mitosis (M). Non-dividing cells tend to exist in a G0 state. However, when stimulated to do so, either in a normally controlled manner or in an aberrant manner, following oncogene activation/tumour suppressor gene inactivation, the cells will undergo DNA replication (S phase) followed by a (usually short) G2 phase prior
Table 2.1 A selection of tumour suppressor genes associated with hereditary cancer syndromes and the molecular pathways they impinge on. A more complete overview can be found elsewhere14 Gene CDKN2A RB1 TP53 WT1 PTEN STK11 TSC1, TSC2
Cancer syndrome
Tumour types
Molecular pathway
Familial malignant melanoma Retinoblastoma Li-Fraumeni syndrome Familial Wilms’ tumour Cowden syndrome Peutz–Jeghers syndrome Tuberous sclerosis
Melanoma, pancreas Eye Adrenal, brain, breast, ovary etc. Wilms’ tumour Hamartoma, glioma, uterus Intestine, ovary, pancreas Hamartoma, kidney
RB RB p53 p53 PI3K/AKT PI3K/AKT PI3K/AKT
A molecular genetic basis to cancer 27
to the mitotic process of cell division (Fig. 2.1). The critical nature of this pathway is highlighted by the discovery that several of the oncogenes and tumour suppressor genes identified, without prior functional knowledge, were revealed to have roles in the control of various ‘checkpoints’ throughout the cell cycle. The retinoblastoma gene product (Rb), as mentioned earlier, is a key regulator of the G1/S phase transition and acts by binding to and repressing the activity of another protein, E2F.12 In the normal situation, upon appropriate mitogenic stimulation, Rb is phosphorylated and consequently releases bound E2F protein.16 The E2F protein acts as a transcription factor and once freed from Rb mediates expression of genes that drive the cell cycle.16 Hence, in a cancer situation in which both copies of the RB gene have been inactivated by mutation, the control over E2F transcriptional activity is absent and therefore the cell cycle is driven through G1/S phase. However, the situation is more complex still, as the phosphorylation, and thus inactivation of Rb, is catalysed by a cyclin/cyclindependent kinase (CDK) dimer, which is in turn activated by mitogenic signalling, resulting, for example, from growth factor stimulation.17 Specifically, cyclin D/CDK4 and cyclin E/CDK2 activation drives the cell cycle from G1 to S phase by inhibiting Rb. These cyclin/CDK molecules have been reported to be hyperactive in certain cancers.18
p16ink4a
p21Cip1
p27Kip1
CyclinD
CyclinE
Cdk4/6
Cdk2 P
Finally, cyclin/CDK activity is under the control of CDK inhibitors such as p16Ink4a, which inhibits cyclinD/CDK4, and p21Cip1/p27Kip1, which inhibits cyclinE/CDK2.17. These genes have tumour suppressor functions and indeed inactivating mutations in p16 are associated with familial malignant melanoma.19 The hierarchy of proteins controlling every aspect of the cell cycle is evidently complicated and these proteins have been found to be the target of mutations in several cancers.
Control of cell fate Circumventing the control mechanisms inherent to the cell cycle is fundamental to the carcinogenic process. However, our cells have another safety feature built in, which must also be bypassed if a cell is to survive and proliferate to form a tumour. Apoptosis is a term that describes a complex and varied set of events by which a cell can be induced to die or ‘commit suicide’.20 Checkpoints exist whereby if damage is detected, the cell cycle can be halted to allow the repair process to proceed, or if the damage is too great, the cell can be eliminated so as to prevent an accumulation of mutated cells. Inherent to this process is the best known tumour suppressor gene of them all, p53, which is often referred to as ‘the guardian of the genome’.21 It is estimated that p53 is mutated in 50 per cent of all solid tumours22 and in fact the pathways it governs are altered in
DNA damage
P
ATM/ATR
Rb
Rb
E2F
CHK1/2 p53
E2F
G1
Puma
p21cip1
S
MDM2
Bax Noxa G1 M
M
G2
S G2
Cell cycle arrest
Figure 2.1 Cell-cycle control: the G1/S checkpoint
Apoptosis
Figure 2.2 Control of cell fate following DNA damage
28 Molecular biology
almost all cancers.23 At the most basic level, the p53 protein functions as a transcription factor controlling the expression of a plethora of genes involved in cell-cycle arrest, cell senescence, apoptosis and DNA repair (Fig. 2.2).24 It is this functionality that makes p53 such a pivotal molecule in determining cell fate following damage and which makes it such a central target for mutation if cancer is to progress. Indeed, it has been found that almost all known DNA tumour viruses encode proteins that inactivate the p53 and Rb pathways.25 When DNA is damaged, sensor assemblies activate the effectors ataxia telangiectasia mutated (ATM) and ATM and Rad3 related (ATR).26 As the name reflects, ATM is the tumour suppressor gene mutated in the familial cancer syndrome ataxia telangiectasia. ATM and ATR lead to activation of checkpoint kinases CHK2 and CHK1, respectively,27 CHK2 having been recently found to harbour inherited mutations in a small percentage of hereditary breast cancer families.28* CHK2 and ATM elicit their cellular responses through p53, which subsequently induces expression of the CDK inhibitor p21Cip1 which, as described previously, inhibits the cyclin E/CDK2 complex preventing G1/S phase transition of the cell cycle. Additionally, pro-apoptotic factors, including PUMA, Bax and NOXA, are induced by p53. Interestingly, p53 also induces expression of its own negative regulator, murine double minute 2 (MDM2), which ubiquitinates p53, targeting it for proteasomal degradation.29,30 Unsurprisingly, hyperactivity of MDM2 is a recognized feature of many cancers, and amplification of the MDM2 locus has been described in certain sarcomas.31,32 Once initiated, apoptosis can occur by two broad mechanisms, the extrinsic pathway (also known as the death receptor pathway) and the intrinsic pathway (also known as the mitochondrial pathway).33 The extrinsic pathway is activated by ligand binding of extracellular death receptors such as TRAIL, Fas or tumour necrosis factor (TNF) receptor. Receptor activation mediates the formation of a death-inducing signalling complex (DISC) comprised of oligomerized receptor, an adaptor protein and the initiator caspase, caspase-8. Autoactivation of the caspase-8 protease leads to cleavage-induced activation of effector caspases, caspase-3, caspase-6 and caspase-7.33 The intrinsic apoptosis pathway can be initiated by a wide variety of stimuli, including the detection of DNA damage, and is defined by characteristic events focused at the mitochondria. The mitochondrial outer membrane becomes permeable, under the control of the Bcl-2 family of proteins, allowing cytoplasmic release of cytochrome c, which in turn binds apoptosis-protease-activating factor 1 (Apaf-1) and procaspase-9 to form the ‘apoptosome’. Activation of caspase-9 within the apoptosome triggers a cascade of caspase activation, leading ultimately to activation of caspase-3.34 The final steps of apoptosis, the formation of apoptotic bodies containing the contents of the dead cells and the clearance of these bodies by phagocytosis, are common to both pathways.
Signal transduction pathways The alteration in the actively maintained balance between progress though the cell cycle and cell elimination by induction of apoptosis is of central importance in the accumulation of a tumour cell mass and can be hugely influenced by the transduction of signals from the cell surface into the cell. Cell signalling is a highly active area of research, and many diverse cell signalling pathways have been discovered, several of which extensively cross-talk with other pathways, resulting in an enormously complex network of molecular interactions. In the interests of brevity and simplicity, this chapter will describe one of the best established pathways: the phosphatidylinositol-3-kinase (PI3K)/AKT pathway (Fig. 2.3), which recent estimates have suggested is mutated, in at least one of its components, in up to 30 per cent of all cancers.35 Interest in PI3K arose in the 1980s, when it was discovered that proteins such as the src kinase enzyme, the protein product of the oncogene isolated from the Rous sarcoma virus, and polyomavirus middle T antigen were physically and functionally associated with PI3K.36 It has since been shown that both the non-receptor tyrosine kinase, src, and transmembrane receptor tyrosine kinases (RTKs), activate PI3K.37 PI3K itself is composed of two subunits, a catalytic subunit (p110) and a regulatory subunit (p85) and upon activation of RTKs by growth-factor binding, the receptors become phosphorylated on their intracellular side, thus promoting the binding of the p85 subunits of PI3K. It is thought that this binding alleviates a repressive effect of p85 on the catalytic subunit and thus increases PI3K activity.38 The primary catalytic function of PI3K is the phosphorylation of the cell membrane phospholipid, phosphoinositol-4,5-diphosphate (PIP2) to produce
PIP2 P P
P P
FKHR
p85
p110
BAD
P
Apoptosis
P
PIP3
PTEN P
GSK3β P
P
PDK1/ PDK2
MDM2 P
Cell cycle
PIP3 Akt P P
mTOR P
Cell growth
Figure 2.3 Intracellular signalling: the PI3K/AKT pathway
Molecular profiling 29
phosphoinositol-3,4,5-triphosphate (PIP3). The fundamental importance of PI3K in cancer was further highlighted by the discovery that the tumour suppressor gene PTEN, which is associated with Cowden syndrome, catalyzes the reverse reaction to PI3K – that is, the conversion PIP3 to PIP2.39 The formation of the so-called second messenger PIP3 within the cell membrane allows it to act as a ligand, recruiting proteins containing a specific sequence of amino acids known as a plecstrin homology (PH) domain.37 One such critical protein is the serine/threonine kinase AKT/ protein kinase B (PKB), which has three family members, AKT1, AKT2 and AKT3. Upon binding to PIP3, the AKT protein is phosphorylated at two residues, one event catalysed by the protein phosphoinositide-dependent protein kinase-1 (PDK1), which also has a PH domain, and the other by PDK2, the precise identity of which is unclear.37 AKT is activated by phosphorylation and in turn phosphorylates a huge number of downstream targets which can either promote or repress their normal functions. The net effect of AKT activation is a pro-survival phenotype resulting either from inhibition of apoptosis, promotion of cell growth and protein translation or from increased rate of proliferation.40 The number and variety of targets that have been found to be under the control of AKT make the subversion of the PI3K/AKT pathway an important event in the carcinogenic process. Various alterations in this pathway have been described, including PI3K p110α subunit amplification;41 PI3K p85α subunit mutations42 elevated AKT1 activity;43 AKT2 amplification and over-expression;44 and PTEN loss/mutation.45 Downstream effects of AKT activity include: MDM2 activation, which, as discussed earlier, leads to the degradation of the p53 tumour suppressor and consequent release from its cell-cycle arrest/pro-apoptotic properties; inhibition of the forkhead transcription factor, FKHR, which, when active, positively regulates the expression of proapoptotic genes;46 and inhibition of glycogen synthase kinase3β (GSK3β), which, when active, phosphorylates cyclin D, targeting it for proteasomal degradation and thus abrogating its effect in driving the G1/S phase cell-cycle checkpoint.47 What has been outlined above is merely a glimpse at the complexity of the intracellular signalling involved in the tumorigenic process. A more in-depth history of cancer molecular biology can be found elsewhere.48 Our knowledge and understanding of this signalling biology and of the many and varied means by which it is subverted in human cancer are rapidly deepening. As will be seen later, this understanding is already being used to develop rationally designed therapeutics aimed at treating patients with a particular subset of tumour mutations.
advances are the family of approaches that have grown up around the genome sequencing projects of the 1990s, and these have given rise to an explosion in so-called ‘omic’ research. Broadly, these refer to the rapid study of large numbers of a particular type of molecule and as such are often referred to as high-throughput technologies. There are currently several tens of omic technologies available for the study of ever-more specialized subsets of molecules; however, the most well recognized and broadly applicable are transcriptomics (concerned with gene expression levels), genomics (concerned with varied DNA level analyses), proteomics (concerned with protein level) and metabonomics (concerned with levels of metabolites). Owing to space limitations, only transcriptomics will be discussed here. Transcriptomics describes the study of the profile of genes expressed within the system under investigation, be it a cell, tissue or organ. To date this has been the most widely publicized of the ‘genome-wide’ approaches to studying the biology of cancer. Typically transcriptomics, also known as gene expression profiling, requires that RNA from the sample under investigation, representing the range and level of genes expressed at the time of collection, is labelled with a detectable reagent. The RNA is then selectively bound to a solid matrix containing precisely defined detectors, or probes, with a large number of gene sequences: this technology is often referred to as a microarray. The binding of target RNA sequences in the sample to the probes on the microarray, which are simply immobilized nucleic acid sequences homologous to their gene target, occurs by the process of hybridization. When a labelled RNA sample is hybridized to a microarray slide, the transcripts in the sample bind to their homologous probes such that when the microarray is scanned using a suitable detector, the presence and level of expression of each gene in the sample can be determined. The applications of this technology are almost unlimited and it has been used extensively to examine global changes in transcription following various treatments of cultured cells; changes in gene expression throughout a time course; the profile of genes expressed in different tissues; the differences between ‘normal’ and ‘cancer’ tissue; the differences between clinicopathological subtypes of a particular cancer, for example histological subtype or tumour grade;49,50 and the correlation between particular gene expression profiles and patient outcome, such as response to chemotherapy, progression-free survival or overall survival.51,52 This last example has led to a very exciting use of this technology: the prediction of outcome for patients with a particular cancer based on the gene expression profiles of their tumours.
MOLECULAR PROFILING
Molecular classification of cancer
There have been many technological advances in molecular biology and these advances frequently allow new avenues of research to be opened. Among the most significant recent
In a paper by Alizadeh and colleagues,53 molecular profiling was used to analyze diffuse large B-cell lymphoma (DLBCL), a histologically indivisible subtype of non-Hodgkin’s
30 Molecular biology
lymphoma that carries a highly unpredictable outcome following treatment. The authors hypothesized that this variability in outcome may reflect an underlying molecular heterogeneity. In their study, they generated transcript profiles for 42 DLBCL patients and used a mathematical technique called hierarchical clustering to look for similarities within the gene expression profiles of the various tumours. This technique works by defining each sample on the basis of its measured levels of gene expression and then comparing each sample to all the others until the two most similar are identified. These two samples are then grouped together as a ‘cluster’ and the process is allowed to continue iteratively until eventually all samples are clustered as one. In the DLBCL study this was done using a subset of all the genes studied and revealed two fundamentally different types of DLBCL. Interestingly, when the authors compared the survival outcome data for the patients belonging to each group, they were found to be very different: 76 per cent versus 16 per cent 5-year survival. It appeared therefore that the careful application of gene expression profiling could provide new clarity in previously heterogeneous disease populations. Many other such examples of cancer classification by expression profiling have also been described. 49,50,54
Prediction of outcome In 2001 a paper appeared in the journal Nature describing a more elaborate, and more far reaching, application of microarray-based expression profiling.55 In their paper the authors reported that by measuring gene expression in primary breast tumours with known clinical outcome it was possible to identify a ‘signature’ within the transcripts that could be used to predict the outcome of an ‘unknown’ patient sample. The rationale of the study was that the current practice of treating women with breast cancer with chemotherapy or hormonal therapy in order to reduce the risk of metastasis is unnecessary in some cases as the women would survive equally well without it. However, predicting who would and who would not benefit is very difficult by conventional means. Expression profiles from 78 primary breast carcinomas were therefore generated using microarray technology and analyzed for their ability to predict a short interval to metastasis, defined in the study as less than 5 years. First, the 25 000 transcript measurements for each tumour were reduced to a more manageable number by elimination of those transcripts that did not vary among the tumour set. The resulting subset of genes was next ordered in terms of the strength of association between gene expression level and outcome. The genes were then tested for their ability to predict the prognosis of an unknown sample. By this process the authors arrived at a final set of 70 genes from the initial 25 000 which, when considered together, made the most accurate possible predictions of the metastatic potential of a primary tumour. The success of this approach led the authors to propose that, because there are genes expressed in the primary tumour
which can predict whether or not the cancer will metastasize, primary tumours appear to be pre-programmed for metastasis at an early time point. Furthermore, the predictive power of this gene classifier suggested that many women could be spared the ill-effects of unnecessary therapy if it is found that their tumours do not contain the metastatic gene expression signature. The validity of this bold suggestion is the subject of the MINDACT (Microarray In Nodenegative Disease may Avoid ChemoTherapy) clinical trial. Several further examples of the use of gene expression profiling to predict clinical outcome have been described.51,52,56,57
MOLECULAR BIOLOGY IN THERAPEUTIC DEVELOPMENT As can be seen from the preceding section, there has been startling progress in the molecular biology of cancer since the last edition of this textbook. These advances in molecular biology have led to a new generation of molecular therapeutics that target specific pathways or molecules involved in proliferative signal transduction, cell-cycle transit, apoptosis/survival, invasion, angiogenesis and metastasis (Table 2.2, 58–61). There is increasing clinical interest in these molecules, with the regulatory approval of a number of targeted therapies including trastuzumab, a humanized monoclonal antibody for the treatment of ErbB2-positive breast cancer; imitanib mesylate, a Bcr-Abl and c-Kit inhibitor effective in chronic myeloid leukaemia and gastrointestinal stromal tumour; cetuximab for the management of colorectal cancers that over-expresses EGFR; and bevacizumab, a monoclonal antibody directed towards VEGF for the management of advanced colorectal cancer.62–66* Unlike traditional agents, the targeted therapies are not necessarily cytotoxic in that they can result in inhibition of tumour growth or prevention of metastases. Because of this, there is a need to incorporate measures of anti-tumour behaviour other than an objective measurable reduction in tumour size into clinical practice. Furthermore, these novel compounds are often characterized by a lack of the typical clinically significant organ toxicities associated with conventional chemotherapy, traditionally the determinant of the starting dose for early clinical trials. Therefore, the use of toxicity as a surrogate endpoint for activity in these agents may be unnecessary or unachievable. There is, therefore, a growing interest in alternatives derived from molecular biology instead of objective tumour response and toxicity as surrogate endpoints both in clinical practice and in early trial design.
Molecular biology and clinical study endpoints Despite the number of molecular targeted therapies used in routine practice, clinically meaningful and stringent endpoints remain unclear. Objective tumour response is felt not to be the best possible endpoint, and different endpoints
Molecular biology in therapeutic development 31
Table 2.2 A selection of targeted agents currently under clinical development Drug ANTI-ANGIOGENIC TARGETS Monoclonal antibody Bevacizumab IMC-1121B 2C3 Receptor tyrosine kinase inhibitors PTK-787 AEE788 ZD6474 AZD2171 SU112248 AG13925 AG013736 CEP-7055 CP-547,632 GW786024 Bay 43-9006 AMG706 SU5416 SU6668 SU11248 SU14813 Soluble receptor chimeric protein VEGF-Trap Inhibitors of endothelial cell proliferation ABT-510 Angiostatin Thalidomide Inhibitors of integrin activity Medi-522 EMD12194 (Cilengitide) Vascular targeting agents Combretastatin A4 AVE8062A ZD6126 AS1404 Matrix metalloproteinase inhibitors Marimastat Prinomastat BMS 275291 Neovastat Other CDP-791 EGFR-TARGETED THERAPIES Receptor tyrosine kinase inhibitors Quinazoline tyrosine kinase inhibitors Gefitinib Erlotinib Lapatinib Canertinib EKB-569 PD153035
Target
Clinical Development
VEGF-A VEGFR-2 VEGF-A
Approved Phase I Pre-clinical
VEGFR-1, VEGFR-2 VEGFR-2, EGFR VEGFR-1, VEGFR-2, VEGFR-3, EGFR VEGFR-1, VEGFR-2 VEGFR-1, VEGFR-2, PDGFR VEGFR-1, VEGFR-2 VEGFR-1, VEGFR-2 VEGFR-1, VEGFR-2, VEGFR-3 VEGFR-1, VEGFR-2, EGFR, PDGFR VEGFR-1, VEGFR-2, VEGFR-3 VEGFR-1, VEGFR-2, PDGFR VEGFR-1, VEGFR-2, VEGFR-3 VEGFR-1, VEGFR-2 VEGFR-2 VEGFR, PDGFR, kit, FLT3 VEGFR, PDGFR, kit, FLT3
Phase III
VEGF-A, PIGF
Phase I
Endothelial CD36 Various Reduction of TNF-alpha production
Phase I/II Phase I Approved
Integrin alphaV Integrin alphaV
Phase I/II Phase I/II
Endothelin tubulin Endothelin tubulin Endothelin tubulin Induction of TNF-alpha;
Phase I/II Phase I Phase I Phase I
MMP-1, MMP-2, MMP-3, MMP-7, MMP-9 MMP-2, MMP-9 MMP-1, MMP-2, MMP-8, MMP-9, MMP-13, MMP-14 MMP-2, MMP-9, MMP-12, VEGF
Phase III Phase III Phase III
Phase II Phase I Phase II/III Phase II Phase I Phase I/II Phase I Phase I Phase III (discontinued) Phase I Phase III Phase I
Phase III
VEGFR-2
erbB1 erbB1 erbB1/2 erbB1/2 erbB1/2 erbB1/2
Approved Approved Phase II/III Phase I/II Phase II Pre-clinical
32 Molecular biology
Table 2.2 (Continued) Drug Pyrrolotriazine tyrosine kinase inhibitors BMS 599626 AEE788 PKI-166 Pyridopyrimidine tyrosine kinase inhibitors ARRY-334543 PD158780 CP-724,714 TAK165 Monoclonal antibodies Cetuximab Panitumumab Trastuzumab Pertuzumab Nimotuzumab Matuzumab MDX-447 TheraCIM Mab 806 IMC-225 ABL, SRC TARGETED THERAPIES Receptor tyrosine kinase inhibitors Imatinib mesylate SKI-606 BMS354825 AZD0530 AP23464 CGP76030 AMN107 Monoclonal antibodies Panitumumab EMC 72000 MDX-447 TheraCIM Mab 806
Target
Clinical Development
erbB1/2
Phase I Pre-clinical/phase I Phase I
erbB1/2 erbB1/2 HER-2 HER-2
Pre-clinical/phase I Pre-clinical
EGFR EGFR HER-2 HER-2 EGFR EGFR EGFR/CD64
Approved Phase III Approved Phase II PhaseI/II Phase II Phase I/II
EGFRvIII EGFR
Pre-clinical Phase III
BCR-ABL Abl, Src Abl, Src Abl, Src Abl, Src Src PDGF, Abl, kit
Approved Phase I Phase II Pre-clinical Pre-clinical Pre-clinical Phase I-II Phase II
FARNESYL TRANSFERASE INHIBITORS BMS-214662 Tipifarnib Lonafarnib FTI-277 L-744832 FTI-276 FTI-2148 L-739,750 BZA-2B
CAAX CAAX CAAX Farnesyl transferase Farnesyl transferase Farnesyl transferase Farnesyl transferase Farnesyl transferase Farnesyl transferase
Phase II Phase III Phase III Pre-clinical Pre-clinical Pre-clinical Pre-clinical Pre-clinical Pre-clinical
PROTEASOME INHIBITORS Bortezomib
Proteasome
Phase III
Molecular biology in therapeutic development 33
Table 2.2 (Continued) Drug
Target
Clinical Development
Phase III
PD184352 PD0325901 ARRY-142886 CGP 69846A
Raf-1 kinase, B-Raf, VEGFR-2, VEGFR-3, c-Kit, PDGFR-β MEK 1/2 MEK 1/2 MEK 1/2 C-Raf
M-TOR INHIBITORS Temsirolimus Everolimus AP23573
mTOR mTOR mTOR
Phase III Phase I/II Phase II
AURORA KINASE INHIBITORS ZM447439 Hesperadine VX680
Aurora-B Aurora-A, Aurora-B Aurora-A, Aurora-B, Aurora-C
INHIBITORS OF ERK PATHWAY Sorafenib
have therefore been proposed, such as time to disease progression and stabilization of disease. An improvement in overall survival is also being increasingly reported as a standard by which these agents can be evaluated in the first-line setting. For example, in the development of trastuzumab, a 25 per cent improvement in overall survival was documented in those patients randomized to receive trastuzumab and chemotherapy versus chemotherapy alone.62* Pre-clinical/phase I data therefore play a pivotal role in determining potential biomarkers. Furthermore, it is possible that the development and validation of a predictive assay of response from pre-clinical/early trial data may enable the selection of those patients who are likely to benefit from targeted therapy.67
Molecular biology and pre-clinical/ phase I endpoints In classic phase I trials with investigational drugs, the agents are administered to patients starting with a very low dose based on mouse toxicology data. Dose escalation is then performed to determine the highest dose that a patient can reasonably tolerate. This is based on the underlying assumption that the higher the dose, the greater the likelihood of drug efficacy.68 In addition to the relationship between dose and anti-tumour response, cytotoxic agents also exhibit a dose–toxicity relationship. Thus, dose-related toxicity is regarded as a surrogate for efficacy and is the pharmacodynamic (PD) endpoint employed in early clinical trial design. It can be questioned whether this is relevant for targeted agents that predominantly act as biochemical modulators. For pre-clinical and clinical development of a targeted therapy it is essential that (i) adequate or optimal exposure to the drug is being obtained in the target tissues of the patient, (ii) the molecular target is being appropriately modulated, and (iii) the desired biological effect is obtained.
Phase II Phase I/II Phase I Phase II
It is important to make these observations as part of the phase I/II trials, particularly in phase I trial design where, as described, the maximally tolerated dose and dose-limiting toxicity may not be sufficient to determine the optimal starting dose for phase II study.69*** The optimal biological dose is defined as the dose that reliably inhibits a drug target or achieves a target plasma concentration.70 The optimal dose is based on either pharmacokinetic (PK) endpoints or direct evidence of biochemical effect on the target molecule. In the case of the PK endpoint, plasma protein binding, which determines the amount of free drug available to bind with the receptor, as well as inter-individual variation in drug absorption and metabolism, need to be accounted for.70 The rationale behind PD studies is that by sequentially analyzing the effects of a given therapy on a patient’s tissue, it may be possible to determine the expression level of the target and monitor the effects of the drug on its molecular target. In addition, if the study agent results in the modulation of expression of genes that are related to the target, PD studies could also be instrumental in the identification of the subpopulation of patients that may derive benefit from treatment. The ideal tissue on which to perform PD studies is the tumour tissue itself. However, given the difficulty of obtaining biopsies, the use of surrogate tissues is being explored. In choosing appropriate surrogate tissues, a series of conditions needs to be met: (i) the tissue must express the target and/or markers downstream of the target, (ii) the target and/or additional pathways have to be affected by the experimental therapy, and (iii) in pre-clinical studies a tight correlation has to exist between the optimal therapeutic effect and the observed changes in the selected biomarker. Furthermore, it is preferable that the surrogate tissue be readily accessible for repeat biopsy. In this setting, peripheral blood cells, buccal mucosa and skin are ideal. In the early studies with EGFR inhibitors, the skin was proposed as a surrogate tissue for EGFR inhibition
34 Molecular biology
in vivo.71–73 Skin was proposed not only because of ease of access, but also because of the established role of EGFR in the regeneration of the dermis.74,75 In normal adult skin, EGFR is strongly expressed in keratinocytes and in the cells of the eccrine and sebaceous glands. In initial studies with the EGFR inhibitor gefitinib, sequential skin biopsies demonstrated changes in phosporylation of EGFR, mitogen-activated protein kinase (MAPK) and STAT-3, as well as the levels of CDK inhibitor, proliferation marker Ki67 and skin maturation markers.72* These changes were noted at 150 mg/day, well below the maximal tolerated dose of 700 mg/day. These results have been mirrored with other anti-EGFR agents, including OSI-774, PKI-166, CI-1003, and the monoclonal antibody EMD72000.76–79* What remains unclear is whether PD changes in the skin correlate with the inhibition of EGFR at tumour sites and whether qualitative changes in these biomarkers at either site predict tumour sensitivity. It is possible that blood flow may be different in the skin and the heterogeneous, and at times poorly vascularized tumour, and this may affect drug delivery to the tumour tissue. Furthermore, it is possible that downstream effects of target inhibition may be different within the tumour tissue. As an example, tumours frequently exhibit mutations that result in the activation of the AKT pathway which could render the cells independent of upstream EGFR inhibition.37 A number of studies have been conducted using tumour tissue during phase I development.80* In a dose-finding study of lapatinib in heavily pre-treated patients with metastatic cancers over-expressing ErbB2 and/or ErbB1, the biological effects of treatment on tumour growth and survival pathways were assessed by serial tumour biopsies.81* The study reported variable levels of inhibition of p-ErbB1, p-ErbB2, pErk1/2, p-AKT, cyclin D1 and transforming growth factor alpha in those patients who responded to treatment. Furthermore, the study reported increased tumour apoptosis (TUNEL) in patients with tumour regression compared with non-responders. Studies such as this serve to generate hypotheses regarding the use of biomarkers that can be used to guide the clinical development of the drug. This process of obtaining serial tumour biopsies requires careful patient selection in terms of biopsy-accessible tumours, and close collaboration with institutional interventional radiologists. Whilst most tumour types may be amenable to computed tomography (CT)-guided biopsy, certain histology types, for example colon cancer with live metastases, are more amenable to this approach. Another approach has been the preoperative administration of the drug and the subsequent analysis of the drug concentration and target activity in the tumour at timed, planned surgical intervention. As many tumours are initially managed with preoperative biopsies, a comparison is then possible with non-drug-treated controls.80* Although measurement of a molecular target effect is optimal, it is associated with a number of difficulties. First, given the complexity of cellular pathways and signalling processes, it may be difficult to define the appropriate measure of achieved target effects for a specific drug.69***
Second, restricting patient enrolment to those with accessible disease for assessment of the drug effect on tumour decreases the eligible numbers and increases the complexity of the conduct of the trial. Even if patients consent to serial biopsies, serial tumour biopsies are invasive and associated with sampling errors resulting from the heterogeneous tissue composition of cancer. It is not always possible to measure the direct target of the drug within tumour tissues or in surrogate tissues; there is a role therefore for the identification of other proteins that may be affected by the drug, and which can then be analyzed as potential biomarkers. To facilitate clinical drug development, pharmacogenomics and proteomic approaches are being applied to identify gene or protein biomarkers that could predict the responsiveness of patients to drug treatment. It has been shown that certain proteins are potentially involved in the resistance mechanism of cetuximab therapy,82 where the different expression levels of proteins may serve as biomarkers predicting resistance; this same approach has been used to identify a set of gene and protein biomarkers predicting response to cetuximab.83 A wide range of invasive techniques is available to determine PK and PD endpoints and, as discussed, these procedures are important in early clinical trials. However, because of logistic and ethical considerations, there is an increasing need for non-invasive techniques.84 The role of functional imaging is therefore increasing in the setting of PK/PD trial design. Imaging biomarkers allow non-invasive serial studies of the entire tumour. This is in contrast with tumour biopsies, where only small parts of the tumour can be evaluated at any one time, and intra-tumoral heterogeneity may therefore confound the analysis of potential biomarkers. Serum biomarker studies, whilst non-invasive, may be altered by metabolism and excretion of tumour-derived biomarkers. By allowing the direct assessment of the tumour tissue, positron emission tomography (PET) scans and other imaging tools are likely to complement these biomarker studies without the interference of metabolic processes in plasma or normal organs.
In-vivo molecular biology: magnetic resonance techniques Magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) are increasingly being used in anti-cancer drug trials. MRI is routinely used in the initial evaluation of the anatomical appearance of the malignancy, and changes in this can be used to assess and quantify the PD effect of a drug. Dynamic contrast enhancement-MRI (DCE-MRI) involves serial evaluations of images obtained every few seconds following tracer injection. These serial evaluations can then be used for assessment of the kinetics of the contrast agent in the tumour. The kinetic parameters derived from this approach depend on tissue perfusion and on the permeability and surface area of tumour blood vessels. Therefore
Future impact of molecular biology in cancer 35
DCE-MRI is of particular use in the evaluation of tumour vasculature before and after treatment. New blood vessels within tumours are structurally abnormal – leaky, with fragmented supportive structures (pericytes and basement membranes). A successful anti-angiogenic therapy would therefore result in stabilization of the vasculature, and reduction in permeability and levels of interstitial fluid hypertension and thus the size of the leakage space.85 Through PK modelling of the behaviour of a low-molecular-weight contrast agent, DCE-MRI allows measurement of these parameters: permeability, Ktrans and leakage space (Ve).86*,87 In a study of patients with liver metastases from colorectal cancer, PTK787/ZK 222584 (PTK/ZK), an orally active inhibitor of VEGF-R, the PD effects of PTK/ZK were evaluated by assessing changes in contrast-enhancement parameters of metastatic liver lesions using DCE-MRI in dose-escalating phase I studies.88 A correlation was demonstrated between a reduction in Ki (equivalent to Ktrans) and of both dose and plasma levels of the study compound; DCE-MRI was seen to be a useful biomarker for defining the pharmacological response and this approach has been employed in a number of other phase I clinical trials of anti-angiogenic agents.89,90 Magnetic resonance spectroscopy (MRS) can be used for minimally invasive monitoring of anti-cancer drug uptake and metabolism. It is the only non-invasive method of chemically distinguishing between, and measuring, the concentrations of drugs and their metabolites. Instead of producing an anatomic image, MRS data are usually visualized as spectra, the peaks of which correspond to different chemicals.91 MRI can then be used to define tumour volume and MRS can be used to measure the concentration of drugs within that volume in real time.92 MRS can also be used for PK studies and concentrations of the drug can be monitored in target tumour and organs. This procedure alleviates the need for invasive drug sampling or radiation exposure. The use of MRS for PK studies, however, is limited by the lack of sensitivity and in general only drugs given in quantities of 0.5 mg/m2 can be detected. The use of MRI in PK/PD studies is likely to increase with improvements in imaging and availability of biologically targeted contrast agents.93,94
and those with specific biological endpoints.92 Studies that utilize generic endpoints are predominantly employed to assess the effects of novel therapeutic agents on biological systems such as cellular proliferation, which can be assessed using 11C-thymidine, and blood volume, with 15O-CO.98*,99* Studies that utilize specific biological endpoints are conducted to provide proof of principle for a proposed mechanism of action of existing and novel therapies, for example the detection of VEGF or its receptor expression can be assessed using 124I-labelled antibodies or peptides.100,101* Although PET imaging is a useful adjunct in drug development, it has some important limitations. First, not all compounds can be radiolabelled, and each compound needs to be considered individually. Second, the anatomic resolution of PET images is low and it is common practice to align images from PET with computed tomography (CT) or MRI data. PET is also limited by a lack of chemical resolution – that is, PET tracers cannot distinguish between the parent radiotracer and its labelled or unlabelled metabolites. This is a concern when investigating drugs that are extensively metabolized, and can complicate the interpretation of PET data. For some positron-labelled compounds, the combination of a short physical half-life and a short biological half-life can limit the type of pharmacological information that can be attained. Finally, because of the costs of labelling the compounds, quality control, radiographers’ time and modelling of the data, PET scanning is expensive and most centres can only focus on one or two tracers. Understanding the impact of the therapy on the tumour target and identification of the patient population most likely to benefit are critical steps in the drug development process. Despite this, only a minority of pre-clinical and early clinical trials incorporate non-traditional endpoints such as measures of molecular drug effects in tumour or surrogate tissue or functional imaging studies.69*** This increases the risk that these agents will fail in the clinical setting. It is therefore important to define the optimal clinical dose and regimen, which should be determined by the use of an array of biomarkers, including target exposure biomarkers and patient selection biomarkers, in combination with clinical response and toxicity profiles.
In-vivo molecular biology: positron emission tomography
FUTURE IMPACT OF MOLECULAR BIOLOGY IN CANCER
PET scanning enables dynamic, non-invasive measures of the three-dimensional distribution of positron-labelled compounds within the body. It can be used to evaluate intratumoral and normal tissue PK in patients prior to phase I studies or as part of phase I and II studies. This can be achieved by labelling the drugs of interest with positron-emitting isotopes such as 11C, 18F, 124I or 13N. A number of drugs have been evaluated in this manner, including 18F-5-fluorouracil (5-FU), 11C-temozolamide and 18F-tamoxifen.95–97* Pharmacodynamic studies performed by PET can be divided into two different types: those with generic endpoints
As can be seen, the molecular sciences permeate into every part of clinical cancer science. They inform and are embedded in the design of most rigorous diagnostic, prognostic and therapeutic developments in cancer care. The approaches encompass reductionism, analytical science, synthetic hypothesis formation and complete descriptions of biology of systems. This understanding is essential for future progress in the wet laboratory, for in-silico studies, for the biology of imaging, and for drug development. Molecular biology remains the horizontal language that unifies these apparently disparate areas. As we have shown, understanding of
36 Molecular biology
the fundamental molecular biology of cancer is starting to be mirrored by a steep increase in the number of available drugs that are targeted to these processes. Understanding the role of these targeted agents fully will be necessary to exploit them most appropriately. It is the combination of these drugs with each other and with chemotherapy that is most likely to help us with the redundancy of pathway inhibition. The molecular basis of drug action in the context of understanding of the molecular biology of cancer are therefore inseparable approaches.
◆3
●4
◆5
6 ●7
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
Molecular diagnostics and therapies are now part of mainstream clinical practice in oncology. The study of gene expression by transcriptomics is beginning to influence therapy choice for patients with common solid tumours. Downstream signalling proteins and their phosphorylated counterparts are promising biomarkers with which to assess the mechanism of a drug’s action. The burgeoning global cancer drug pipeline needs better mechanism for prioritising which drugs to take forward to large scale trail. It is simply not financially viable to take all promising candidates forward. Measuring the optimal biological dose rather than just the maximally tolerated dose is essential in phase I trials of molecularly targeted therapies. Determining pharmacodynamic endpoints for new drugs involves the discovery and utilisation of validated biomarkers of a drug’s target effect. The need for serial biopsies to study tissue biopsies is limited by ethical considerations. There is also the potential of selective sampling. Imaging biomarkers are becoming available that allow ethically acceptable, non invasive serial studies of the entire tumour. New imaging technologies are likely to help our understanding of gene and protein expression changes as well as the physiological processes that change after therapeutic intervention. The close interaction of clinical and laboratory investigators is vital to optimise the potential gain to patients from translational medicine.
REFERENCES 1 Waldron HA. A brief history of scrotal cancer. Br J Ind Med 1983; 40(4):390–401. 2 Shu XO, Jin F, Linet MS, et al. Diagnostic X-ray and ultrasound exposure and risk of childhood cancer. Br J Cancer 1994; 70(3):531–6.
●8
9
10
●11
◆12
13
◆14
15 16
17 18
19
●20
21 22
Doll R. Fifty years of research on tobacco. J Epidemiol Biostat 2000; 5(6):321–9. Ames BN, Durston WE, Yamasaki E, Lee FD. Carcinogens are mutagens: a simple test system combining liver homogenates for activation and bacteria for detection. Proc Natl Acad Sci U S A 1973; 70:2281–5. Rous P. Viruses and tumour causation. An appraisal of present knowledge. Nature 1965; 207(996):457–63. Martin GS. The hunting of the Src. Nat Rev Mol Cell Biol 2001; 2(6):467–75. Stehelin D, Varmus HE, Bishop JM, Vogt PK. DNA related to the transforming gene(s) of avian sarcoma viruses is present in normal avian DNA. Nature 1976; 260(5547):170–3. Reddy EP, Reynolds RK, Santos E, Barbacid M. A point mutation is responsible for the acquisition of transforming properties by the T24 human bladder carcinoma oncogene. Nature 1982; 300(5888):149–52. Mills GB, Lu Y, Fang X, et al. The role of genetic abnormalities of PTEN and the phosphatidylinositol 3-kinase pathway in breast and ovarian tumorigenesis, prognosis, and therapy. Semin Oncol 2001; 28(5 Suppl. 16):125–41. Kurzrock R, Kantarjian HM, Druker BJ, Talpaz M. Philadelphia chromosome-positive leukemias: from basic mechanisms to molecular therapeutics. Ann Intern Med 2003; 138(10):819–30. Harris H, Bramwell ME. The suppression of malignancy by terminal differentiation: evidence from hybrids between tumour cells and keratinocytes. J Cell Sci 1987; 87 (Pt 3):383–8. Knudson AG. A personal sixty-year tour of genetics and medicine. Annu Rev Genomics Hum Genet 2005; 6:1–14. Bagchi S, Weinmann R, Raychaudhuri P. The retinoblastoma protein copurifies with E2F-I, an E1A-regulated inhibitor of the transcription factor E2F. Cell 1991; 65(6):1063–72. Vogelstein B, Kinzler KW. Cancer genes and the pathways they control. Nat Med 2004; 10(8):789–99. Bodmer WF. Cancer genetics: colorectal cancer as a model. J Hum Genet 2006; 51(5):391–6. Sherr CJ, Roberts JM. CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 1999; 13(12):1501–12. Murray AW. Recycling the cell cycle: cyclins revisited. Cell 2004; 116(2):221–34. Dickson C, Fantl V, Gillett C, et al. Amplification of chromosome band 11q13 and a role for cyclin D1 in human breast cancer. Cancer Lett 1995; 90(1):43–50. Haluska FG, Tsao H, Wu H, Haluska FS, Lazar A, Goel V. Genetic alterations in signaling pathways in melanoma. Clin Cancer Res 2006; 12(7 Pt 2):2301s–7s. Kerr JF, Wyllie AH, Currie AR. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 1972; 26(4):239–57. Lane DP. Cancer. p53, guardian of the genome. Nature 1992; 358(6381):15–16. Hainaut P, Hernandez T, Robinson A, et al. IARC Database of p53 gene mutations in human tumors and cell lines:
References 37
◆23
24 25 ◆26
27 28
●29
30
●31
32
◆33
34
35
36
◆37
38
●39
40
41
updated compilation, revised formats and new visualisation tools. Nucleic Acids Res 1998; 26(1):205–13. Vogelstein B, Lane D, Levine AJ. Surfing the p53 network. Nature 2000; 408(6810):307–10. Liu G, Chen X. Regulation of the p53 transcriptional activity. J Cell Biochem 2006; 97(3):448–58. Zur Hausen H. Oncogenic DNA viruses. Oncogene 2001; 20(54):7820–3. Sancar A, Lindsey-Boltz LA, Unsal-Kacmaz K, Linn S. Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints. Annu Rev Biochem 2004; 73:39–85. Massague J. G1 cell-cycle control and cancer. Nature 2004; 432(7015):298–306. Meijers-Heijboer H, van den Ouweland A, Klijn J, et al. Low-penetrance susceptibility to breast cancer due to CHEK2(*)1100delC in noncarriers of BRCA1 or BRCA2 mutations. Nat Genet 2002; 31(1):55–9. Haupt Y, Maya R, Kazaz A, Oren M. Mdm2 promotes the rapid degradation of p53. Nature 1997; 387(6630):296–9. Honda R, Tanaka H, Yasuda H. Oncoprotein MDM2 is a ubiquitin ligase E3 for tumor suppressor p53. FEBS Lett 1997; 420(1):25–7. Oliner JD, Kinzler KW, Meltzer PS, George DL, Vogelstein B. Amplification of a gene encoding a p53-associated protein in human sarcomas. Nature 1992; 358(6381):80–3. Leach FS, Tokino T, Meltzer P, et al. p53 mutation and MDM2 amplification in human soft tissue sarcomas. Cancer Res 1993; 53(10 Suppl.):2231–4. Jin Z, El-Deiry WS. Overview of cell death signaling pathways. Cancer Biol Ther 2005; 4(2):139–63. Slee EA, Harte MT, Kluck RM, et al. Ordering the cytochrome c-initiated caspase cascade: hierarchical activation of caspases-2, -3, -6, -7, -8, and -10 in a caspase-9-dependent manner. J Cell Biol 1999; 144(2):281–92. Luo J, Manning BD, Cantley LC. Targeting the PI3K-Akt pathway in human cancer: rationale and promise. Cancer Cell 2003; 4(4):257–62. Whitman M, Kaplan DR, Schaffhausen B, Cantley L, Roberts TM. Association of phosphatidylinositol kinase activity with polyoma middle-T competent for transformation. Nature 1985; 315(6016):239–42. Vivanco I, Sawyers CL. The phosphatidylinositol 3-kinase AKT pathway in human cancer. Nat Rev Cancer 2002; 2(7):489–501. Yu J, Zhang Y, McIlroy J, Rordorf-Nikolic T, Orr GA, Backer JM. Regulation of the p85/p110 phosphatidylinositol 3kinase: stabilization and inhibition of the p110alpha catalytic subunit by the p85 regulatory subunit. Mol Cell Biol 1998; 18(3):1379–87. Stambolic V, Suzuki A, de la Pompa JL, et al. Negative regulation of PKB/Akt-dependent cell survival by the tumor suppressor PTEN. Cell 1998; 95(1):29–39. Bader AG, Kang S, Zhao L, Vogt PK. Oncogenic PI3K deregulates transcription and translation. Nat Rev Cancer 2005; 5(12):921–9. Shayesteh L, Lu Y, Kuo WL, et al. PIK3CA is implicated as an oncogene in ovarian cancer. Nat Genet 1999; 21(1):99–102.
42 Philp AJ, Campbell IG, Leet C, et al. The phosphatidylinositol 3-kinase p85alpha gene is an oncogene in human ovarian and colon tumors. Cancer Res 2001; 61(20):7426–9. ●43 Sun M, Wang G, Paciga JE, et al. AKT1/PKBalpha kinase is frequently elevated in human cancers and its constitutive activation is required for oncogenic transformation in NIH3T3 cells. Am J Pathol 2001; 159(2):431–7. 44 Bellacosa A, de Feo D, Godwin AK, et al. Molecular alterations of the AKT2 oncogene in ovarian and breast carcinomas. Int J Cancer 1995; 64(4):280–5. 45 Saito M, Okamoto A, Kohno T, et al. Allelic imbalance and mutations of the PTEN gene in ovarian cancer. Int J Cancer 2000; 85(2):160–5. 46 Brunet A, Bonni A, Zigmond MJ, et al. Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor. Cell 1999; 96(6):857–68. 47 Diehl JA, Cheng M, Roussel MF, Sherr CJ. Glycogen synthase kinase-3beta regulates cyclin D1 proteolysis and subcellular localization. Genes Dev 1998; 12(22):3499–511. 48 Weinberg RA. One Renegade Cell: The Quest for the Origins of Cancer. New York, Phoenix Press, 1998. 49 Alon U, Barkai N, Notterman DA, et al. Broad patterns of gene expression revealed by clustering analysis of tumor and normal colon tissues probed by oligonucleotide arrays. Proc Natl Acad Sci U S A 1999; 96(12):6745–50. 50 Perou CM, Jeffrey SS, van de Rijn M, et al. Distinctive gene expression patterns in human mammary epithelial cells and breast cancers. Proc Natl Acad Sci U S A 1999; 96(16):9212–17. 51 Beer DG, Kardia SL, Huang CC, et al. Gene-expression profiles predict survival of patients with lung adenocarcinoma. Nature Med 2002; 8(8):816–24. 52 Sorlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A 2001; 98(19):10869–74. ●53 Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000; 403(6769):503–11. 54 Schwartz DR, Kardia SL, Shedden KA, et al. Gene expression in ovarian cancer reflects both morphology and biological behavior, distinguishing clear cell from other poor-prognosis ovarian carcinomas. Cancer Res 2002; 62(16):4722–9. ●55 van’t Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002; 415(6871):530–6. ●56 Golub TR, Slonim DK, Tamayo P, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999; 286(5439):531–7. 57 Eschrich S, Yang I, Bloom G, et al. Molecular staging for survival prediction of colorectal cancer patients. J Clin Oncol 2005; 23(15):3526–35. 58 Garrett MD, Workman P. Discovering novel chemotherapeutic drugs for the third millennium. Eur J Cancer 1999; 35(14):2010–30.
38 Molecular biology
59 Gibbs JB. Mechanism-based target identification and drug discovery in cancer research. Science 2000; 287(5460):1969–73. 60 Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100(1):57–70. 61 Kaelin WG Jr. Choosing anticancer drug targets in the postgenomic era. J Clin Invest 1999; 104(11):1503–6. ●62 Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344(11):783–92. ●63 Druker BJ, Sawyers CL, Kantarjian H, et al. Activity of a specific inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N Engl J Med 2001; 344(14):1038–42. ●64 Heinrich MC, Corless CL, Demetri GD, et al. Kinase mutations and imatinib response in patients with metastatic gastrointestinal stromal tumor. J Clin Oncol 2003; 21(23):4342–9. ●65 Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecanrefractory metastatic colorectal cancer. N Engl J Med 2004; 351(4):337–45. ●66 Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004; 350(23):2335–42. ◆67 Arteaga CL, Baselga J. Clinical trial design and end points for epidermal growth factor receptor-targeted therapies: implications for drug development and practice. Clin Cancer Res 2003; 9(5):1579–89. 68 Skipper HE, Schabel FM Jr, Mellett LB, et al. Implications of biochemical, cytokinetic, pharmacologic, and toxicologic relationships in the design of optimal therapeutic schedules. Cancer Chemother Rep 1970; 54(6):431–50. ●69 Parulekar WR, Eisenhauer EA. Phase I trial design for solid tumor studies of targeted, non-cytotoxic agents: theory and practice. J Natl Cancer Inst 2004; 96(13):990–7. 70 Adjei AA. What is the right dose? The elusive optimal biologic dose in phase I clinical trials. J Clin Oncol 2006; 24(25):4054–5. 71 Albanell J, Codony-Servat J, Rojo F, et al. Activated extracellular signal-regulated kinases: association with epidermal growth factor receptor/transforming growth factor alpha expression in head and neck squamous carcinoma and inhibition by anti-epidermal growth factor receptor treatments. Cancer Res 2001; 61(17):6500–10. 72 Albanell J, Rojo F, Averbuch S, et al. Pharmacodynamic studies of the epidermal growth factor receptor inhibitor ZD1839 in skin from cancer patients: histopathologic and molecular consequences of receptor inhibition. J Clin Oncol 2002; 20(1):110–24. 73 Workman P. Challenges of PK/PD measurements in modern drug development. Eur J Cancer 2002; 38(16):2189–93. 74 Stoll SW, Benedict M, Mitra R, Hiniker A, Elder JT, Nunez G. EGF receptor signaling inhibits keratinocyte
75
76
77
78
79
80
81
82
83
84
◆85
86
87
apoptosis: evidence for mediation by Bcl-XL. Oncogene 1998; 16(11):1493–9. Jost M, Kari C, Rodeck U. The EGF receptor – an essential regulator of multiple epidermal functions. Eur J Dermatol 2000; 10(7):505–10. Malik SN, Siu LL, Rowinsky EK, et al. Pharmacodynamic evaluation of the epidermal growth factor receptor inhibitor OSI-774 in human epidermis of cancer patients. Clin Cancer Res 2003; 9(7):2478–86. Hoekstra R, Dumez H, Eskens FA, et al. Phase I and pharmacologic study of PKI166, an epidermal growth factor receptor tyrosine kinase inhibitor, in patients with advanced solid malignancies. Clin Cancer Res 2005; 11(19 Pt 1):6908–15. Nemunaitis J, Eiseman I, Cunningham C, et al. Phase 1 clinical and pharmacokinetics evaluation of oral CI-1033 in patients with refractory cancer. Clin Cancer Res 2005; 11(10):3846–53. Vanhoefer U, Tewes M, Rojo F, et al. Phase I study of the humanized antiepidermal growth factor receptor monoclonal antibody EMD72000 in patients with advanced solid tumors that express the epidermal growth factor receptor. J Clin Oncol 2004; 22(1):175–84. Dowlati A, Haaga J, Remick SC, et al. Sequential tumor biopsies in early phase clinical trials of anticancer agents for pharmacodynamic evaluation. Clin Cancer Res 2001; 7(10):2971–6. Spector NL, Xia W, Burris H 3rd, et al. Study of the biologic effects of lapatinib, a reversible inhibitor of ErbB1 and ErbB2 tyrosine kinases, on tumor growth and survival pathways in patients with advanced malignancies. J Clin Oncol 2005; 23(11):2502–12. Skvortsov S, Sarg B, Loeffler-Ragg J, et al. Different proteome pattern of epidermal growth factor receptorpositive colorectal cancer cell lines that are responsive and nonresponsive to C225 antibody treatment. Mol Cancer Ther 2004; 3(12):1551–8. Luo FR, Yang Z, Dong H, et al. Prediction of active drug plasma concentrations achieved in cancer patients by pharmacodynamic biomarkers identified from the geo human colon carcinoma xenograft model. Clin Cancer Res 2005; 11(15):5558–65. Collins JM. Functional imaging in phase I studies: decorations or decision making? J Clin Oncol 2003; 21(15):2807–9. Jain RK. Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science 2005; 307(5706):58–62. Galbraith SM, Rustin GJ, Lodge MA, et al. Effects of 5,6-dimethylxanthenone-4-acetic acid on human tumor microcirculation assessed by dynamic contrast-enhanced magnetic resonance imaging. J Clin Oncol 2002; 20(18):3826–40. Choyke PL, Dwyer AJ, Knopp MV. Functional tumor imaging with dynamic contrast-enhanced magnetic resonance imaging. J Magn Reson Imaging 2003; 17(5):509–20.
References 39
88 Morgan B, Thomas AL, Drevs J, et al. Dynamic contrastenhanced magnetic resonance imaging as a biomarker for the pharmacological response of PTK787/ZK 222584, an inhibitor of the vascular endothelial growth factor receptor tyrosine kinases, in patients with advanced colorectal cancer and liver metastases: results from two phase I studies. J Clin Oncol 2003; 21(21):3955–64. 89 Evelhoch JL, LoRusso PM, He Z, et al. Magnetic resonance imaging measurements of the response of murine and human tumors to the vascular-targeting agent ZD6126. Clin Cancer Res 2004; 10(11):3650–7. 90 Galbraith SM, Maxwell RJ, Lodge MA, et al. Combretastatin A4 phosphate has tumor antivascular activity in rat and man as demonstrated by dynamic magnetic resonance imaging. J Clin Oncol 2003; 21(15):2831–42. 91 Leach MO. Magnetic resonance spectroscopy (MRS) in the investigation of cancer at The Royal Marsden Hospital and The Institute of Cancer Research. Phys Med Biol 2006; 51(13):R61–82. ◆92 Workman P, Aboagye EO, Chung YL, et al. Minimally invasive pharmacokinetic and pharmacodynamic technologies in hypothesis-testing clinical trials of innovative therapies. J Natl Cancer Inst 2006; 98(9):580–98. 93 Golman K, Ardenkjaer-Larsen JH, Petersson JS, Mansson S, Leunbach I. Molecular imaging with endogenous substances. Proc Natl Acad Sci U S A 2003; 100(18):10435–9.
94 Lanza GM, Winter P, Caruthers S, et al. Novel paramagnetic contrast agents for molecular imaging and targeted drug delivery. Curr Pharm Biotechnol 2004; 5(6):495–507. 95 Saleem A, Yap J, Osman S, et al. Modulation of fluorouracil tissue pharmacokinetics by eniluracil: in-vivo imaging of drug action. Lancet 2000; 355(9221):2125–31. 96 Galldiks N, Kracht LW, Burghaus L, et al. Use of 11 C-methionine PET to monitor the effects of temozolomide chemotherapy in malignant gliomas. Eur J Nucl Med Mol Imaging 2006; 33(5):516–24. 97 Inoue T, Kim EE, Wallace S, et al. Preliminary study of cardiac accumulation of F-18 fluorotamoxifen in patients with breast cancer. Clin Imaging 1997; 21(5):332–6. 98 Eary JF, Mankoff DA, Spence AM, et al. 2-[C-11]thymidine imaging of malignant brain tumors. Cancer Res 1999; 59(3):615–21. ❊99 Gupta N, Price PM, Aboagye EO. PET for in vivo pharmacokinetic and pharmacodynamic measurements. Eur J Cancer 2002; 38(16):2094–107. 100 Collingridge DR, Carroll VA, Glaser M, et al. The development of 124Iiodinated-VG76e: a novel tracer for imaging vascular endothelial growth factor in vivo using positron emission tomography. Cancer Res 2002; 62(20):5912–19. ●101 Jayson GC, Zweit J, Jackson A, et al. Molecular imaging and biological evaluation of HuMV833 anti-VEGF antibody: implications for trial design of antiangiogenic antibodies. J Natl Cancer Inst 2002; 94(19):1484–93.
3 Clinical radiobiology MICHELE I. SAUNDERS AND STANLEY DISCHE
Introduction Radiosensitivity at the cellular level The radioresponsiveness of tumour and of normal tissues Time and radiation response Time, fractionation and the implications for current programming of radiotherapy Nutrition, environment and radiation response
40 40 41 45 47 47
INTRODUCTION All those caring for patients with cancer strive to attain the greatest degree of tumour control with the lowest level of treatment-related morbidity. Essential to the good and safe practice of radiation therapy is knowledge of the biology of tumours and of the normal tissues together with an understanding of the way in which they are affected by ionizing radiation.
RADIOSENSITIVITY AT THE CELLULAR LEVEL The immediate response In most normal tissues and tumours, little change can be demonstrated in the first few days after beginning a course of conventional radiotherapy. However, changes in the DNA are induced that become manifest in time.1 There are some normal tissues, such as the bone marrow and the salivary glands, that do show evidence of this damage during the first 24 hours after administration of a dose as low as 2 Gy, with steep falls in lymphocyte counts and steep rises in the serum amylase concentration respectively.
The differences between normal and tumour cells In early work, the inherent radiosensitivity of cells derived from animal and human tumours, as measured in the
The combination of chemotherapy with radiotherapy The combination of surgery with radiotherapy The biological basis for special approaches in radiotherapy Biological features of the tumour important to management Significant points References
49 51 51 52 52 53
laboratory by the survival of clonogenic cells after high radiation doses, did not appear to differ greatly from that of cells derived from normal tissues. However, Fertil and Malaise2 demonstrated (in cell lines derived from a range of animal tumours) that the surviving fractions of cells irradiated using a much lower and more clinically relevant dose of 2 Gy (SF2) did differ widely from each other and from normal cells. The results could in broad terms predict the radiocurability as determined when the tumours were grown in mice, with lymphomas the most sensitive and sarcomas most resistant to treatment. The responsiveness of types of human tumours as revealed in experiments using cell lines grown in nude mice was shown broadly to reflect that observed clinically.3
Extreme radiosensitivity Patients with the rare inherited ataxia telangiectasia show an extreme sensitivity to radiation that appears to be at a level several-fold greater than that normally exhibited. Patients with ataxia telangiectasia, Cockayne’s, Gardner’s, Down’s and basal cell naevoid syndromes and those with Fanconi’s anaemia may also show unusually severe reactions.1 Some patients with dermatomyositis and other collagen disorders may also show an increased sensitivity. In cell culture, fibroblasts taken from the skin of patients with ataxia telangiectasia have been shown to possess a high radiosensitivity. Cultures of fibroblasts from other patients exhibiting unexpectedly severe reactions have also been shown to demonstrate high radiosensitivity.4,5*
The radioresponsiveness of tumour and of normal tissues 41
Normal range of radiosensitivity It is important to identify patients with high degrees of radiosensitivity, preferably before embarking upon a course of treatment, but fortunately they represent only a fraction of 1 per cent of all patients requiring treatment. Patients given identical radiotherapy show a range of normal tissue reactions, both early and late. In a series of patients previously given a course of radiotherapy a correlation has been obtained between the radiosensitivity of their fibroblasts cultured in vitro and that of the reactions observed in their normal tissues.6* However, the technique is not a practical one to apply in everyday care, as the result may not be available for several weeks. There is an interest in the radiosensitivity of peripheral blood lymphocytes because these can easily be obtained. Some correlation has been reported,7 but there is no consensus as to the value of studies of lymphocyte sensitivity as a predictor of radiation response.
Planning treatment according to radiosensitivity of the individual tumour Using the techniques described above, attempts have been made to determine the SF2 prior to radiotherapy in individual cases. West and colleagues8* showed that in carcinoma of the cervix there was a close correlation between the SF2 and tumour control and survival. In contrast,9 no such correlation could be demonstrated in squamous cancer in the head and neck region. The demonstration of a relationship between inherent sensitivity and tumour response in one common tumour in humans is important in radiobiology, but as this relationship has not been confirmed in other types of tumour and because the results of an assay cannot be read for 4 weeks, this approach is, as yet, unpromising in terms of advancing the care of individual patients. A more immediate change can be seen in the micronuclei of radiation-damaged cells using in-situ hybridization (FISH).10 So far, however, no convincing evidence for a clinical application has been established.
avoid unacceptable morbidity. The demonstration of such differences on an individual case basis will demand the use of methods of observing normal tissue reaction that are sufficiently sensitive; however, there is currently no evidence that such a sensitive system is available. Of relevance to this approach is the relationship between the inherent sensitivity of the normal tissues and of the tumours in an individual patient. There are case descriptions that report a similar high sensitivity of the tumour in sensitive patients, but it is suspected that this is not a general rule.11
THE RADIORESPONSIVENESS OF TUMOUR AND OF NORMAL TISSUES The early and late reactions The pattern of response seen in a tumour can range from an obvious shrinkage observed within 24 hours to an absence of change for a number of weeks with reduction in size delayed for some months after the completion of treatment. Lymphomas and undifferentiated tumours such as oat-cell carcinomas show the earliest changes, whereas adenocarcinomas and well-differentiated sarcomas may exhibit the longest time course. In contrast, many normal tissues, including the skin and the mucosa of the upper aerodigestive tract, exhibit early reactions during and immediately after a course of treatment. These tend to be transitory and, provided the very highest doses have not been administered, are followed by healing. The same tissues may show late changes after some months or years, which, once established, do not readily resolve and may show steady progression. Other tissues, such as the spinal cord, liver, kidney, fat and subcutaneous tissues, show little early change but may show late changes. Late changes in the early-responding tissues are often secondary to effects on the supporting connective tissue. It has long been traditional to set 90 days as the border between early and late responses, but it is now considered that this is too rigid an approach.12 It is also recognized that severe early reactions may progress to late reactions without healing, which is called consequential damage.
Planning treatment according to the radiosensitivity of an individual patient Whole or part-organ irradiation The determination of the radiosensitivity of normal tissues in the management of an individual case remains a subject of research in clinical oncology.11 Among methods undergoing study is the determination of a relationship between the response of normal tissues, early and late, and molecular change. Genetic patterns may emerge to associate with radiosensitivity. It has been proposed that patients with relatively resistant normal tissues may be safely given higher doses of radiation, while those at the sensitive end of the spectrum may have their treatment modified so as to
The effect of part-organ or whole-organ irradiation differs according to the tissue concerned. Loss of a considerable portion of the liver or part (or even the whole) of one kidney may lead to little functional disability and be an acceptable consequence of radiotherapy if the clinical situation justifies it. With other structures, such as the intestine and the central nervous system, loss of, or severe damage to, a small part may have a vital and life-threatening effect.
42 Clinical radiobiology
Table 3.1 Guide to organ tolerance to fractionated radiotherapy in 2-Gy daily doses14,15 A (Gy)
B (Gy)
Eye Ovary Liver Oesophagus Stomach Intestine Rectum
– – 50 60 60 50 –
18 6 35 55 50 40 60
Optic nerve, chiasma Brachial plexus Brain Spinal cord Heart Lung Kidney
50 62 60 50 55 30 30
50 60 45 47 40 17.5 23
Common severe complication Cataract requiring surgery Sterility Liver failure Stricture/perforation Ulceration Ulceration/stricture Severe proctitis/necrosis Fistula/stenosis Blindness Plexopathy Necrosis Myelitis Pericarditis and pancarditis Pneumonitis/fibrosis Nephrosclerosis
A Tolerance to partial-organ (approx. 1⁄3) irradiation. Dose giving 5% incidence of severe morbidity in 5 years. B Tolerance to whole-organ irradiation. Dose giving 5% incidence of severe morbidity in 5 years.
Setting the level of normal tissue tolerance
The balance of the relationship between the responses of tumours and normal tissues Dose–response curves derived from work with animal models and from observation of patients have a sigmoid or S shape with a threshold level before the curves depart
Tumour Response
A broad guideline of normal tissue tolerance, based on the work of Rubin13 and Emami et al.14 is shown in Table 3.1. Considerable variations will occur with different dose fractionation regimes of radiotherapy and when treatment is combined with other modalities, such as cytotoxic chemotherapy. The table must be regarded as an approximate guideline and applied with extreme care. The acceptable risk of morbidity of any measure of treatment must be related to the purpose for which the treatment is being given and the result likely to be achieved. The functional loss of a considerable portion of a lung will give some disability, which could well be justified in the control of an advanced bronchial carcinoma; on the other hand, any impairment of respiratory function is not acceptable following radiotherapy given after lumpectomy for an early carcinoma of the breast in a young patient. Even in the management of advanced and inoperable tumours, level of risk must be low in certain vital tissues because of the dire consequences that may follow, which can be as distressing as uncontrolled tumour growth. An example would be radiation myelitis leading to paraplegia, for which a risk greater than 1 per cent can rarely be justified.
100
Normal tissues
0 Dose
Figure 3.1 A diagram showing the probability of tumour control and the incidence of late effects in the normal tissues. The actual curves in an individual case will depend on the characteristics of the tumour and of the normal tissues irradiated, as well as on the regime of radiotherapy.
significantly from the baseline (Fig. 3.1). In normal tissues, there is generally a higher threshold dose, but this is followed by a steeper dose–response curve. The steepness is described by the ‘gamma value’,15 which gives the increase in response in percentage points for a 1 per cent increase in dose. The commonly demonstrated gamma values for the cure of a squamous cancer of the head and neck region are in the range 1.5–2.5. With a gamma value of 2, a 10 per cent increase in dose can be expected to elevate the tumour cure probability by 20 per cent; this may mean an elevation in cure rate of from 60 to 80 per cent. Animal tumour models usually show much higher gamma values – commonly
The radioresponsiveness of tumour and of normal tissues 43
exceeding 10. The greater heterogeneity of human tumours and inhomogeneities of radiation dose, together with inevitable inaccuracies in staging and inadequacies in follow-up, probably account for the much lower values found in the human situation. At dose levels high in the therapeutic range, the gamma values for severe late reactions in normal tissues are greater, and emphasize the importance in the planning of radiotherapy of avoiding maximum doses to significant areas in excess of that prescribed to the tumour.16 Consideration of dose–response relationships leads to the important conclusion that once the therapeutic range is reached, very small differences in dose can considerably affect the chance of cure and also the incidence of normaltissue injury.17
Fractionation and total dose In the early years of radiotherapy, benefit in terms of increased tumour control, with acceptable tissue morbidity, was achieved by moving from a single (or a few) large dose fraction to small doses given daily over a number of weeks. To achieve this benefit, it was found that the total dose given in many small fractions needed to be increased. When, for example, treatment was changed from one single to 30 daily treatments over 6 weeks, the total dose needed to be multiplied by a factor of approximately 3. Strandqvist in 194418 proposed a mathematical formula to relate total dose and overall time to skin tolerance. Ellis in 196919 importantly suggested that the effects of fraction number and overall time needed to be separately included in a formula. The Ellis formula was widely used and was found to be very helpful in clinical practice. However, it later became apparent that when the number of fractions was reduced below 12, the incidence of normal-tissue effects was greater than predicted.20
The linear quadratic formula Laboratory studies have shown that at clinically relevant doses, tumours and early-reacting tissues respond to ionizing radiation, dominantly with a linear relationship between dose and effect – the α or linear element. In the late-reacting tissues, however, it is dominantly related to the square of the individual dose – the β or quadratic element. In 1982, Barendsen21 proposed the linear quadratic (LQ) formula that brought together this knowledge. The formula predicts that by giving radiotherapy in many small doses there is sparing of the damage in late-reacting tissues but little alteration in the response of the early-reacting normal tissues and in the responsiveness of the tumour. Clinical experience has given support to the LQ model, with the predicted increased damage in late-reacting tissues, such as the spinal cord, with the use of high doses per fraction and reduced damage when low doses per fraction are
employed. In a number of randomized controlled clinical trials using hyperfractionation with low doses per fraction, the total dose given in a standard period of time has been increased, with a modest and tolerable increase in acute morbidity but with significant improvement in tumour control (Table 3.2).22
α/β ratios for tumours and for normal tissues The relative importance of the two elements in the LQ formula for a given response in a given tissue is expressed by the α/β ratio.23 In animal studies, the ratio is in the range of 7–20 Gy for acutely responding tissues, whereas for the late-responding tissues it ranges from 0.5 to 6 Gy. In humans, values calculated for early reactions have ranged from 8 to 15 Gy and for late reactions from 0.8 to 4.0 Gy, giving good concordance with the laboratory data.23 High α/β ratios in excess of 10 have been recorded for animal tumours and similar values have long been accepted for human tumours. In carcinoma of the prostate, however, some evidence has accumulated to suggest that the α/β ratio may be as low as 1.0 Gy, but there is considerable controversy on the subject, with one study suggesting a much higher value of 8.3 Gy.24 If a low value is correct, the best radiotherapy might be with a few large fractions, and certainly such treatment is easier for the patient and less costly for the community. Randomized controlled studies of hypofractionation are underway and these may well give us an answer.24,25 Carcinoma of the breast is another situation in which it has been suggested that, in contrast to the generality of tumours, the α/β ratio may be low. However, hard evidence is lacking.26
Low-dose hyper-radiosensitivity Until recently, there was no reason for clinicians to be interested in the use of radiation doses below 1 Gy, but with the development of new techniques, research has shown an excess of cell kill at doses below 1 Gy relative to that predicted by the LQ model.27 This phenomenon is termed low-dose hyper-radiosensitivity (LDHRS) and has been shown to be present in many human tumour cell lines, its effect being generally greater in those that are radioresistant.28 The effect has been demonstrated in many laboratory tumour models and can be exhibited to a lesser extent in normal tissues such as skin and kidney, but not in spinal cord. Modelling suggested that there would be an increase in the therapeutic ratio.27 Clinical work in which volunteer patients underwent radiation of normal skin and of skin deposits of tumour provided proof, in principle, that LDHRS could be used to advantage in the management of human tumours.28 However, the prospect of treating two or three times daily for periods of 5 or 6 weeks or more has so far inhibited progress.
Table 3.2 Fractionation studies in head and neck cancer Normal Tissue reaction
Dose/F (Gy)
No.
Total dose (Gy)
Conventional Radiotherapy
2
33
66
45
24
Hyperfractionated EORTC 2279154 (RTOG 900356)
1.15
70
80.5
47
8
Acceleration but with split course EORTC61
1.6
42 or 45
67.2 or 72
47
3
Acceleration with shorter split EORTC 2285155
1.6
46
72
33
4
Acceleration with shorter split and reduction in total dose – WANG57 (RTOG 900356)
1.6
40
64
40
4
Concomitant boost Ang et al (1990)58 (RTOG 900356)
1.18 and 1.5
30 and 10
69
40
3–6
Highly accelerated GORTEC98 9402
2
31–32
62 64
21
8
Continuous Hyperfractionated accelerated with reduction in total dose CHART63
1 .5
36
54
12
6
Moderately accelerated DAHANCA 6&758
2
33
66 68
38 39
24 (and 8)
Radiotherapy schedule
total time (days)
Time between F and hours
Early
Late
Locoregional control
1
2
3
4
5
6
7
slight increase
equal
improved
benefit
slight increase
equal
equal
no benefit
increase
increase
improved
morbidity too great
Week
8
Overall assess.
slight increase
equal
equal
no benefit
slight increase
equal
improved
benefit
increase
increase
improved
concerns regarding morbidity
slight increase
reduced
equal
slight benefit
slight increase
equal
improved
benefit
In addition to the trial considered representative of the type of study, other similar trials are shown in brackets. F fractionation
Time and radiation response 45
TIME AND RADIATION RESPONSE The mitotic cycle and radiosensitivity The phase in mitosis at the time of therapy is important in determining radiosensitivity and chemosensitivity.1 With ionizing radiation, the maximum resistance is in S phase and the greatest sensitivity in M and late G2; there may be a factor of two in the relative sensitivity. Cytotoxic drugs show a number of patterns, but generally S phase is the most sensitive period. However, with some drugs, restingphase cells have been reported to be more sensitive than those that are proliferating. The initial treatment in a fractionated course of radiotherapy may, after the period of mitotic arrest, lead to cell-cycle synchrony, and if the next treatment is given when the cells are in a sensitive phase of the cycle, an increase in cell kill may occur. So far, these experiments have been confined to the laboratory and there is no immediate application in clinical radiotherapy.1 With the multi-fraction techniques employed for cure in the clinic, it is difficult to see synchrony maintained beyond a second treatment. This makes it improbable that a manipulation of the cell cycle can be employed to improve the results of radiotherapy.
Repair of sublethal injury After exposure to radiation, some of the damage sustained within the cell can be repaired.1 The rate of repair falls in an exponential fashion, and in laboratory models the time for the rate to fall to half has been calculated to range from 30 to 120 minutes. Some variation is seen from one tissue to another, and two or more components of repair with different half-times may be present.29 It is important to note that the proportion of damage that is recoverable is smaller in tumours and in acutely responding tissues, such as skin and mucosa, than in the late-responding tissues, such as kidney and connective tissue. With treatments spaced 24 hours or more apart, we can assume that repair has been completed between fractions. However, when more than one treatment is given each day, the duration of time between fractions must be chosen with care in order to allow as much repair as possible to occur in the normal tissues. In the early clinical studies of hyperfractionation, intervals of 4 hours or less were often employed. Acute and late reactions were rather greater than expected, particularly when intervals of 2 hours or less were used. With intervals of 6–8 hours these were acceptable.30–33** In the CHART (continuous hyperfractionated accelerated radiotherapy) studies, particular care was taken to measure acute and late reactions, and when the comparison was made with the control arm treated conventionally, it was possible to calculate the halftimes of repair in human normal tissues. These were found to be 4 hours or more and there was evidence that in neural tissue the half-time was significantly greater, so accounting for
the unexpected incidence of a number of cases of radiation myelitis in the pilot study of CHART.34* We can conclude that in general the interval between fractions should never be less than 6 hours; when brain and spinal cord are irradiated, the interval should be 12 hours.
Tumour-cell proliferation When there is an opportunity to observe the natural progression of a carcinoma or sarcoma, several months may pass before the tumour doubles its size. When serial observations have been possible, volume doubling times have ranged from 30 to more than 300 days, with median values often greater than 100 days.1 These observations, together with knowledge that there is a period of mitotic arrest after exposure to ionizing radiation, led radiotherapists to the view that growth during a course of radiation therapy lasting 4–7 weeks was not likely to be important in influencing the result. Advances in knowledge have caused this opinion to be revised. In the past, it has only been possible to study growth at the cellular level by administering radiolabelled thymidine. Only patients with advanced disease could therefore be investigated, and the procedure is a complex one, requiring autoradiography of biopsy samples for the result to emerge. Begg and his colleagues35 were able to determine cell kinetics for tumour by giving a single injection of bromodeoxyuridine (BrdUrd) and performing a biopsy 4–6 hours later. With the use of a cell sorter, the proportion of cells preparing for division (the labelling index) and the duration of the S phase could be determined. From these values the potential cell doubling time was calculated. This is a measure of the proliferative activity of tumour cells, taking into account the presence of dividing and nondividing cells, but assuming the absence of cell loss.36 A wide range of values has been obtained in extensive studies in human tumours using this technique.37 In squamous cell cancer in the head and neck region and in the uterine cervix, median values lie between 4 and 5 days, whereas median values a little higher are seen in other tumours (Fig. 3.2). Spontaneous cell loss, due to apoptosis, necrosis, nutritional deprivation and differentiation, accounts for the 10–20-fold differences between cellular doubling times and volume doubling times observed clinically.38 With an immunohistochemical technique, the cells that have taken up the BrdUrd can be seen clearly under the microscope, and considerable variation in the proportion of labelled cells can be observed from field to field. At the growing edge of many squamous cell carcinomas, up to 60 per cent of the cells may appear to contain BrdUrd and therefore be in DNA synthesis.39 The advantage of the immunohistochemical technique over flow cytometry is that tumour cells can be readily differentiated from normal cells. This is particularly important in the case of diploid tumours for which it is not possible, using the flow cytometer, to separate the cells. In squamous cell cancers in the head and
46 Clinical radiobiology
Breast Melanoma Lung 7
Oesophagus
6
0
2
4
6
8
Awaad
Antognoni
0
Foster
1
Corvo
2
Hoyer
3
Bourhis
4
Wilson
Colorectal
5
Zackrisson
Cervix
Begg
Median Tpot (days)
Head and neck
10
Median Tpot (days)
Figure 3.2 Median values for potential doubling time (Tpot) measured in a variety of different tumours (data provided by George Wilson). The inset shows published data from different groups for head and neck tumours only.
neck region, the labelling indices determined by immohistochemistry have been combined with the duration of S phase determined by flow cytometry to calculate the potential doubling time. When the maximum labelling index, as seen microscopically, is used, half the tumours show extremely rapid potential cell doubling times of less than 2 days.39 The evidence showing that tumour cells may proliferate at such rapid rates has important clinical implications. When tumour cells are destroyed by radiotherapy or cytotoxic chemotherapy, it is likely that the pattern of cell division and loss will be altered considerably.40–42 It is probable that the high spontaneous cell loss occurring in the unperturbed tumour will be greatly reduced and the surviving tumour cells will realize their full reproductive potential. It is even possible that cellular division may be accelerated with an increased growth fraction of the surviving tumour cells, and the cell-cycle time may be reduced so that repopulation occurs at a rate even faster than that suggested by the cell kinetics performed prior to treatment.40,43 Evidence to support the view that repopulation is an important factor in determining the result of a course of radiotherapy has been derived from analyses of clinical data.44 Prolongation of a course of radiotherapy has been reported to result in the reduction of the probability of tumour cure at a number of sites, including uterine, cervix, breast and head and neck.45 The prolongation of overall duration of treatment for head and neck cancer by 1 week may reduce tumour control rates by 13 per cent.46,47** Most of this work is the result of retrospective analysis of consecutively treated patients, where treatment may have been prolonged for many reasons, including some that may be associated with unfavourable features of the tumour, but convincing data have been obtained from an
analysis of the results of a randomized trial of split-course therapy in head and neck cancer.48
Onset of repopulation In a range of different animal tumour models, the time to commencement of repopulation has shown a wide variation, with intervals ranging from 7 to 21 days. Withers et al.,40 following an analysis of the reported results of radiotherapy in head and neck cancer, found evidence that the repopulation of squamous cell cancer in the head and neck region begins after a lag period of 3–5 weeks after the first treatment. The influence of growth factors in determining the time of onset and extent of cellular repopulation has been studied and the influence of epidermal growth factor (EGF) has received much attention.49,50 Over-expression of the EGF receptor (EGFR) is now considered to be associated with decreased local tumour control after radiotherapy, particularly when overall treatment times have been long.49–51 A considerable research effort has been made to inhibit EGFR in combination with radiotherapy. The blocking antibody C225 with radiotherapy has been shown in animal tumour models to be more effective than either radiotherapy or C225 alone.52 In a randomized phase III trial in head and neck cancer, the addition of cetuximab (C225) gave significantly improved loco-regional tumour control and survival compared with radiotherapy alone.53** Some increase in acute skin reactions was noted in the experimental arm. Benefit has not been clearly achieved in other studies and it would seem that the EGFR mechanism is a complex one and much effort will be required to exploit this approach fully.49
Nutrition, environment and radiation response 47
TIME, FRACTIONATION AND THE IMPLICATIONS FOR CURRENT PROGRAMMING OF RADIOTHERAPY Unconventional fractionation in the head and neck site Knowledge concerning cellular division, the time course of repopulation and the implications of the LQ equation led to a questioning of conventional, five times weekly, 1.8–2.0 Gy doses to a total of 64–70 Gy as the standard best treatment when cure is the objective. Clinicians have been concerned about the possibility that towards the end of the standard course of radiotherapy there is much loss of effect because of tumour cell repopulation. The use of many small fractions to increase tolerance was attractive. Many trials have taken place, including a large number of randomized controlled studies. Squamous cell cancer of the upper aerodigestive tract has been the tumour most extensively studied (see Table 3.2).54–62** At least for such tumours, we can conclude the following: ●
●
●
The overall duration of time planned for the course of treatment should not be prolonged. Gaps will lead to a lowering of the expectation of tumour control. Shortening of the overall duration, other aspects remaining constant, will elevate tumour control. Acute reactions are likely to be enhanced by acceleration, but it is possible with hyperfractionation employing small doses per fraction and with adequate inter-fraction interval to achieve a higher total dose and with it a greater expectation of tumour control.
Unconventional fractionation in other sites The situation with regard to other tumours is less clear; however, it seems likely that squamous cell cancer at other sites such as lung, oesophagus and uterine cervix will benefit from the same approach. Experience with the CHART randomized controlled trial in lung cancer gives this support.63** There is evidence to suggest that it is the well-differentiated and moderately differentiated squamous tumours that show benefit from acceleration.64*,65** It would seem that such tumours have retained the characteristics of the parent normal tissue and its ability to respond with rapid proliferation after injury. It is possible that other tumours showing rapid proliferative characteristics, such as some malignant lymphomas, oat-cell carcinoma and some tumours occurring in children, may also benefit from acceleration; however, clear evidence is lacking.
The balance of tumour control and normal-tissue injury As in all applications of radiobiology to clinical practice, great care must be taken not to cause increase in normal-tissue
damage. Much can be learnt from laboratory study; however, the time course of clinical radiotherapy invariably differs from that in the laboratory, where the characteristics of both the tumour and of the host animal differ from those of the human. A high standard of clinical science is necessary to exploit the promise of benefit. The situation has been further complicated by the wide adoption of combined chemo-radiotherapy and the difficulties in interpretation of interactions between the elements of the treatment planned for the patient.
NUTRITION, ENVIRONMENT AND RADIATION RESPONSE The environment of tumour and normal tissues and the importance of oxygen All biological processes may be influenced by the nutritional support of the system. In determining radiation response, the dominant influence is that of the concentration of oxygen available at the time of treatment. The relationship between oxygen tension and radiation response has long been recognized in radiobiology. In 1935, Mottram gave a clear account of the importance of radioresistance due to hypoxia, and it was his colleague, Hal Gray, who gathered the evidence together and brought it to wide attention.66 There is good evidence to show that hypoxia commonly exists in human tumours and that hypoxic tumour cells are resistant to radiotherapy. A process of re-oxygenation has been shown to occur in tumours during a fractionated course of radiotherapy: as sensitive oxic tumour cells are destroyed, hypoxic ones move nearer to the blood supply released and so become sensitive. There is good circumstantial evidence, however, to suggest that this process is incomplete in some tumours, and that hypoxia is a cause for failure to achieve cure.67
Methods applied clinically to overcome hypoxia BREATHING OXYGEN OR CARBOGEN
Breathing 100 per cent oxygen or carbogen (95 per cent oxygen, 5 per cent carbon dioxide) has been the subject of phase I and II studies but has never reached phase III.67,68 HYPERBARIC OXYGEN AT 3 ATMOSPHERES ABSOLUTE
This technique, pioneered by Churchill Davidson in 1955, did proceed to randomized clinical trials and 15 have been reported in carcinoma of the cervix, bladder, lung and head and neck cancer. Some studies contained relatively fewer patients; however, margins of benefit were shown in nine, and statistical significance was reached in three.69
48 Clinical radiobiology
There was some increase in late normal-tissue effects. The difficulties of treating patients in hyperbaric chambers led to interest in simpler methods of achieving sensitization of hypoxic tumour cells. CHEMICAL SENSITIZING AGENTS
Chemical sensitizing agents such is misonidazole and etianidazole have been extensively studied in human tumours but nearly all showed no advantage.70,71** However, in a series of studies in oropharyngeal and supraglottic tumours performed by the Danish oncology centres (DAHANCA) a significant advantage was shown, first with misonidazole and then with nimorazole, which, unlike misonidazole, did not induce peripheral neuropathy.72**,73 Nimorazole is included in standard management for patients with such tumours in Denmark but its use in other countries has been limited. Despite demonstration of benefit in a large, well-conducted randomized controlled clinical trial, the report of so many other trials of chemical sensitizing agents that proved negative seems to have undermined its establishment and interest has moved to other methods to advance radiotherapy. AGENTS TO OVERCOME TRANSIENT HYPOXIA
In addition to chronic hypoxia, evidence has accumulated to show that acute hypoxia due to temporary occlusion of small vessels may be important during radiotherapy. Vasoactive drugs such as pentoxifylline and nicotinamide have been introduced to overcome this problem.74 It has not been felt that, used alone, sufficient numbers of patients would benefit and the application has been essentially in combination with other methods used to overcome chronic hypoxia.75 ARCON (ACCELERATED RADIOTHERAPY, CARBOGEN AND NICOTINAMIDE)
It was proposed by the late Juliana Denekamp that a tumour that is rapid in proliferation is also one that outgrows its vascular supply to become hypoxic.75 ARCON brought together carbogen to overcome chronic hypoxia, nicotinamide to overcome transient hypoxia, and acceleration to deal with cellular proliferation. In an extensive pilot study at Nijmegen, very promising results were reported and a margin of benefit was achieved with minimal increase in normal-tissue effects; a pilot study in bladder cancer performed at Mount Vernon Hospital has also given promise.76*,77* Randomized controlled clinical trials are underway in carcinoma of the bladder and in pharyngolaryngeal cancer. HYPOXIC CYTOTOXICITY
The work to develop hypoxic cell radiosensitizers has also produced compounds that are in themselves cytotoxic to hypoxic cells. A drug developed by the Stanford group,
tirapazamine has shown promise in phase I/II clinical trials.78 Such a drug needs to be given in combination with radiotherapy or cytotoxic chemotherapy, so that the oxic component of the tumour can be dealt with.79* The drug may also enhance the cytotoxic effect of cis-platinum, and remarkably good results have recently been reported from a pilot study of the combination in advanced head and neck cancer.79* Randomized trials of the use of tirapazamine are now underway and the results are awaited. COMBINATION OF VASCULAR-ACTING AGENTS WITH RADIOTHERAPY
The importance of the blood supply to the growth and advance of tumours led to the development of drugs that cause a shutdown of vascular supply and the induction of massive necrosis.80 Although promising agents have been developed that can achieve short-term palliation, a longer effect can only be expected when they are used in combination with either radiotherapy or cytotoxic chemotherapy so that the surviving tumour cells at the periphery of the tumour may be controlled.81,82 Efforts so far have concentrated on the combination with cytotoxic chemotherapeutic agents; however, studies are currently underway of the combination of combretastatin and radiotherapy.83–85
Overview Overgaard and his colleagues have performed an overview of trials of methods employed clinically to overcome the resistance associated with hypoxia, and have gathered data related to more than 10 600 patients. A margin of 5 per cent of improvement in loco-regional control (P 0.001) was demonstrated, and with it a 3 per cent improvement in survival (P 0.04).86** The overview is important in that it demonstrates that there is an oxygen effect and that methods of overcoming it can be successful. However, the margin is small and it is obvious that the normal process of re-oxygenation, which occurs during radiotherapy, is commonly successful. Before treatment is commenced, we need to identify those cases in which the process is unlikely to be successful and in which a method to overcome hypoxia should be employed.
Determining where hypoxia matters We know that once tumours have reached clinically detectable size hypoxia probably exists in practically all. Despite this, a large number of patients are cured and therefore the radiotherapy is successful in eradicating all tumour cells. Patients likely to benefit from methods to overcome hypoxia can therefore only comprise a proportion of those attending for treatment.86 The identification
The combination of chemotherapy with radiotherapy 49
of such patients continues to be a major area of research activity. Techniques employed have included measuring oxygen tension using sophisticated probes, radiolabelled nitroimidazoles to identify hypoxic areas, and measurements of vascular perfusion, often using magnetic resonance (MR) techniques, besides many other approaches.86,87 Among recent particularly promising lines of research has been the use of positron emission tomography (PET) with a range of molecules to allow for the imaging of hypoxia.88–91 All this research has been important for the development of our understanding of human tumours, but has not, so far, yielded a technique that can be readily applied to identify patients in whom hypoxia is a major problem and for whom a method to overcome it is indicated.
for women – levels in the lower range of normality.92 The demands on a blood transfusion service to achieve this in clinical practice may be considerable. Erythropoietin is effective in stimulating bone marrow function and can, in moderately anaemic patients, bring the haemoglobin concentration to within normal levels in 10–14 days. A randomized controlled clinical trial in patients undergoing radiotherapy for head and neck cancer demonstrated very satisfactory elevation of haemoglobin levels; however, there was no evidence of clinical benefit.94**,95** In this study, a margin in favour of the controls does raise the possibility of a direct adverse influence on tumour progression and the possibility that erythropoietin stimulates both bone marrow function and tumour growth.95** Considerable caution would seem appropriate for the use of erythropoietin in patients with potentially curable cancer undergoing treatment.
Hypoxia and future directions Molecular biology has now shown the importance of hypoxia in the regulation of tumour growth and progression. Hypoxia will induce genetic changes that stimulate tumour growth and facilitate dissemination of metastatic disease. There is thus a further stimulus to the understanding of the importance of hypoxia and its mechanism and the discovery of ways to overcome it to benefit the patient with cancer.
Anaemia In 25 papers reporting the clinical experience of radiation response and anaemia, 23 reported an adverse radiation response in anaemic patients. The interpretation of retrospectively derived data of this type must always be done with caution.92 There is no doubt that there is, in many tumour sites, an association between anaemia and advancement of disease. Even taking this into consideration, however, the evidence does suggest a real association between anaemia and response to radiation. A randomized controlled clinical trial of the use of blood transfusion in carcinoma of the cervix showed an improvement in primary tumour control when the haemoglobin was brought to 125 g/L or above, compared with the control group in which the haemoglobin level lay between 100 g/L and 125 g/L during therapy.93** However, in the DAHANCA head and neck cancer studies, a sub-randomization to test the value of transfusion showed only a small margin of advantage, which was not of statistical significance.86** Considering all the evidence, anaemia does seem of importance, and the implication for clinical practice is that in the curative situation the haemoglobin should be brought to normal levels before radiation therapy is commenced. It is noteworthy that in the Danish studies in which anaemia was associated with impaired response in head and neck tumours, the levels of haemoglobin chosen to divide the cases were 9 mmol (144 g/L) for men and 8 mmol (128 g/L)
Radioprotectors Radioprotectors were developed in the USA in an attempt to give some protection in atomic warfare; the compound that has reached clinical testing in oncology is amifostine. The initial work in the laboratory with amifostine suggested that the drug does not accumulate in tumours; however, the total reliability of this remains uncertain.96 The progress of this drug in clinical trial has been extremely slow,97 but its use in patients with head and neck cancer with a view to reducing late effects on salivary gland function has yielded a highly significant benefit,98** although the side effects can be troublesome.99 The drug is now marketed for this purpose and there may be further application for amifostine may have value in combination with cytotoxic drugs, such as cyclophosphamide, for the protection of bone marrow; evidence to support this has been demonstrated in clinical trials.100,101**,102**
THE COMBINATION OF CHEMOTHERAPY WITH RADIOTHERAPY To enhance tumour control within the irradiated volume The combination may be employed to enhance tumour control within the irradiated volume. The combination of local effect may be due to a simple addition of cell kill or to a true radiosensitization – processes often difficult to separate.103 Possible mechanisms of radiosensitization include the following: ●
●
Compounding of radiation-induced DNA breaks and so reducing repair. Drugs capable of doing this include cisplatin, bleomycin, adriamycin and hydroxyurea. Selective killing of cells in radioresistant phases of the cell cycle (late S and G1).
50 Clinical radiobiology
These mechanisms may be important to clinical practice, but so far there has been no way of scientifically directing inhibition of repair and selective cell cycle kill so as to maximize effect. ●
Selective killing of hypoxic cells. This promising approach has already been discussed above.
A simple addition of cell kill may in itself lead to advance in local tumour control. However, this relies on independent toxicities and no increase in radiation damage, otherwise a simple increase in radiation dose would achieve the same result.
Spatial co-operation Eradication of the primary tumour by radiotherapy and/or surgery can be complemented by chemotherapy for micrometastatic disease. In the management of the bulk of carcinomas, chemotherapy is most effective when used towards the achievement of cure by the eradication of micrometastases.103
Importance of time The giving of combined therapy concurrently, or with only a few days’ separation, does commonly lead to an enhancement of effect on the tumour. However, this may also be seen in the normal tissues. Reduction in total radiation dose may be necessary, and this complicates the interpretation of results. When chemotherapy and radiotherapy are separated, an increased effect in normal tissues is less commonly seen,103 but clinical experience now suggests that benefit is also lost.104*** With growing awareness that cellular repopulation may occur in gaps between treatments, whether they be surgery, radiotherapy or chemotherapy, the trend has been for concurrent radiotherapy and chemotherapy.
Debulking Preliminary (neoadjuvant) chemotherapy can be expected to reduce the tumour-cell number and so increase the possibility of cure, and may also allow a reduction in the volume of normal tissue to be irradiated as the tumour volume is smaller. Timing is critical: in order to minimize the morbidity of the use of two modalities and to see shrinkage of the tumour, a considerable separation is desirable. However, while shrinkage may be obviously visible, rapid cellular repopulation may already be underway and may negate the benefit. The approach is best established in the management of lymphomas.
Novel molecular targets Many of the cellular growth factors and signalling pathways for the control of proliferation, differentiation and angiogenesis may be activated in tumours and not in normal tissues, so giving rise to the possibility of drug targeting. Targets modifying radiosensitivity have been identified.10 Of great interest is the EGFR, which is highly expressed in many head and neck tumours and is associated with aggressive behaviour. This has already been discussed in the section dealing with repopulation.49,53** Other targets that are the subjects of current research include the ErbB-2 receptor in breast cancer and COX2 in colon cancer.4
Interactions between radiation and chemotherapy In addition to a potentiation or simple addition of the effects on tumour, there is, in some circumstances, the possibility of a reduction in radiotherapeutic response. As discussed, radiation response varies according to position in the cell cycle and it follows that if the chemotherapeutic agent significantly alters mitotic activity before the radiotherapy is given, the response may be reduced. Careful laboratory experiment can give some guidance, but as conditions for clinical application may differ considerably from those in the laboratory, careful planning and study in the clinic are essential. The potentiation of radiation effects in the normal tissues has already been discussed.
Current situation A vast clinical experience has now been recorded. The design of the studies has not come from laboratory work but from practical considerations related to programming and tolerance. There has been progression from phase I through to phase III, and with some tumours to adoption as ‘best practice’. Considerable margins of benefit have been demonstrated in the management of anal and oesophageal cancers, for which the combination therapies have become standard. In head and neck and non-small-cell lung cancer, margins have been smaller and overall results variable. Overviews and meta-analyses have demonstrated small but certainly significant gains, concurrent therapies appearing more successful.104***–107*** These have led to fairly wide adoption of combination chemotherapy in these conditions. There must, however, be concern, for in many of the individual studies, and certainly in the overviews and meta-analyses, data concerning morbidity, especially late morbidity, have not been included.104*** When consideration has been given to them, the data available have not been such as to be sensitive in evaluating morbidity. There is the necessity for large, randomized trials, where
The biological basis for special approaches in radiotherapy 51
tumour effects and those in normal tissues are given equal priority in order to assess the true therapeutic benefit.
THE COMBINATION OF SURGERY WITH RADIOTHERAPY Tumour cell kinetics in the postoperative field Advances in surgery have led to an increase in the number of patients with the more advanced tumours undergoing resection. Commonly the pathologist reports that tumour cells have reached the margin of the surgical specimen or that margins are narrow. Postoperative radiotherapy has been shown to reduce the high level of recurrence in subsequent months. In contrast to the situation when a late recurrence is managed in a surgically scarred neck, the tissues are very vascular.108** Tumour cells in the operated area are likely to proliferate rapidly and be stimulated by growth factors generated within the operative field. It is therefore important that the interval between surgery and onset of radiotherapy be as short as possible so as to allow the minimum time for the tumour cells to proliferate. The volumes of tissue at risk may be extensive. There is considerable risk of post-radiation morbidity being increased because of the recent surgery and any postoperative complication, together with the large volume commonly requiring treatment and the many vital structures in the area. The margin of benefit now demonstrated by the addition of chemotherapy suggests that it should now be incorporated, so the hazard of increased morbidity may be further accentuated.109**
Radiation dose required In the postoperative state, such as after lumpectomy for primary carcinoma of breast, where any nodule of residual tumour is likely to be truly microscopic, a total radiation dose of 46–50 Gy given in 2-Gy fractions may be adequate to achieve a high degree of probability of elimination of all tumour. However, after surgery for an advanced squamous cell carcinoma in the head and neck, the risk of residual disease is greater and the tumour-cell burden is likely to be higher. In those circumstances a dose of 60 Gy or greater may be indicated.
THE BIOLOGICAL BASIS FOR SPECIAL APPROACHES IN RADIOTHERAPY
with 250 kV X-rays. Values between 2 and 10 may be obtained, according to the energy of neutron beam and, importantly, the size of dose per fraction, for there is an inverse relationship between dose per fraction and RBE.110 The biological advantages rest upon a lessened influence of oxygen tension and the position in the mitotic cycle together with, in animal models and human tumour transplants, a more uniform response among tumour types.110 The results of phase II and phase III clinical trials have been difficult to interpret because of case selection, unusual dose fractionation regimes, a variable handling of morbidity and, in some trials, the use of a combination of photons and neutrons. A place was suggested in the management of squamous cancer in the head and neck region, but this was not confirmed in further trials. With evidence based on small numbers, there may be a particular advantage in advanced parotid and prostatic carcinoma.111 It can, however, be concluded that the considerable resources needed to pursue neutron therapy might give a greater benefit if employed in other ways to advance radiotherapy.
Charged particles The ion beams of charged particles increase their rate of energy emission as they slow down, finally stopping and releasing energy in an intense burst of ionization (the Bragg peak). The beam produced may be used to deliver high doses to small volumes and spare adjacent sensitive structures, so bringing advantages over photon therapy in certain situations, such as in the treatment of tumours in the base of the skull.112 When employed clinically, photon and helium ion beams have a Linear Energy Transfer (LET) similar to that of photons and 250 kV X-rays and present, therefore, no special biological advantage.110 If deeply placed tumours are to be treated with proton beams, high-energy cyclotrons are required. The cost and complexity of the apparatus required for the production of a heavy-ion beam are even greater than with neutrons.110 The advantage of proton ion beams has, however, been well demonstrated in the treatment of tumours closely related to the spinal cord and brainstem,113 and a strong justification has been made for the establishment of national centres where such therapy may be available for those patients likely to benefit.
Brachytherapy and the biological advantages of low dose rate
Neutrons The efficiency of any ionizing radiation increases with the density of the ionization along its track. X-rays and γ -rays are sparsely ionizing radiations, but neutron beams and heavy-ion beams have densities that may be several hundred times greater per unit length. The relative biological effectiveness (RBE) is normally determined by comparison
Interstitial and intra-cavitary applications of radioactive sources have an important place in radiotherapy. A high cure rate is obtained with accessible, well-defined tumours of small or moderate size, and such techniques may also be used to give a boost to the area of gross tumour after external-beam treatment. With the original low-activity sources, applications were for 1–7 days and the dose rates
52 Clinical radiobiology
were commonly less than 1 Gy/h. This can be compared with conventional external-beam therapy, in which 2 Gy may often be given in 1–2 minutes. Sublethal injury is important in determining the late effects of external-beam therapy using photons, but it has now been shown in the laboratory that there is complete recovery of all sublethal injury at the low dose rates used for brachytherapy.114 This does accord with the excellent normal tissue tolerance with limited late changes associated with brachytherapy. There is a further advantage in that with overall treatment times in brachytherapy ranging over 1–7 days, treatment is accelerated compared with conventional external-beam therapy and little or no time is allowed for repopulation. Finally, it has also been shown in the laboratory that the oxygen enhancement ratio is considerably reduced and that re-oxygenation appears to be very effective under low-dose therapy, even though overall times are relatively short. In addition to the considerable biological evidence to support the efficacy of brachytherapy, there is the well-localized dose distribution.
Afterloading and high dose rate The hazard of radiation exposure to staff has led to the introduction of hollow applicators and subsequent afterloading of the radioactive sources; this is most safely performed mechanically. However, the long period of exposure to radiation with traditional brachytherapy techniques is difficult to reproduce with machines for afterloading. With the advent of small, high-intensity sources, high-dose-rate techniques have been introduced and exposure times reduced from days to less than 30 minutes. The dose rate thus becomes similar to that of conventional external-beam therapy, and similar biological characteristics may be expected. As one or a few large doses are commonly given, there is not the benefit of multiple fractions usually associated with conventional externalbeam therapy. Reductions in dose have been made, but in some centres these were initially inadequate and morbidity was considerably increased. Fractionated brachytherapy is obviously a useful approach, easier to achieve with a boost rather than with definitive therapy. Treatment units have now been constructed to allow the introduction of very highactivity sources for a few minutes every hour. Calculations based on laboratory experiments have suggested that this approach will allow safe brachytherapy with biological effects similar to those of continuous low-dose-rate techniques.115
BIOLOGICAL FEATURES OF THE TUMOUR IMPORTANT TO MANAGEMENT Tumour volume and clonogenic tumour-cell number A ‘tumour mass’ will contain some normal tissue, stromal elements and necrotic tissue as well as tumour cells, therefore similar sized tumour masses may contain varying
numbers of viable tumour cells. There is, again, a variation in the proportion of those tumour cells that are clonogenic. In some neoplasms, such as the lymphomas, it is believed that clonogenic cells are relatively few in number, thus accounting, in part, for their relative radiosensitivity.41 The clonogenic tumour-cell burden is an important parameter in determining the response, and this has been clearly shown in laboratory experiments. We can expect in humans that in any one type of tumour, the number of clonogenic cells within a nodule 5 mm in diameter may be 1000 times fewer than the number within a mass 5 cm in diameter.41 In a range of clinical situations using standard programming of radiotherapy, very different total doses are required to achieve a 90 per cent probability of elimination of all tumour cells. In the postoperative state, as has already been discussed, doses may range from 46 Gy to 60 Gy according to the burden of risk. For a primary squamous cell carcinoma of 20 mm diameter in the head and neck region, the dose given in 2-Gy fractions may need to be elevated to 66–70 Gy to give an equal probability of loco-regional control of disease. However, with a big tumour mass of 5 cm diameter, 75 Gy may control at best only 50 per cent. Other factors, such as hypoxia, inherent radiosensitivity and proliferation characteristics, will all modify and complicate the situation. Nevertheless, the tumour-cell burden is, itself, an important biological parameter and justifies the use of a range of doses in clinical practice.
Precision in dosimetry and its biological significance Because of the steepness of the curve relating dose to tumour eradication, an elevation or depression of dose of 5 per cent at the 50 per cent control point can be expected to elevate or depress tumour control in the human situation by approximately 10 per cent. In clinical practice, there is commonly a 10 per cent variation in dose across the target volume at the centre of the fields of treatment, and when a three-dimensional view is taken, variations up to 15 per cent or even 20 per cent may be encountered. These dose variations within the area irradiated will also have important biological implications for the normal tissues. Where the total dose is lowered, normaltissue effects will be diminished further by the lower dose fraction. On the other hand, where the dose is elevated, the effect will be increased further by the higher dose per fraction. Using a conventional 2-Gy increment and a 66-Gy total dose, an increase of 10 per cent in part of the irradiated volume is likely to lead to a 13 per cent increase in biological effect, because of the increase in dose per fraction. This has been described by Rodney Withers as ‘double trouble’.116
SIGNIFICANT POINTS Laboratory and clinical research has greatly extended our knowledge of the processes that influence tumour control
References 53
and normal-tissue injury after radiotherapy. Brachytherapy and the fractionation of external-beam therapy, which were developed as a result of a continued process of clinical trial and observation, have now been given a firm scientific basis. Modification of the fractionation has now given guidance as to how the benefit of radiotherapy can be enhanced by programming treatment for the management of tumours in certain clinical situations. The complexity of the biology of human tumours, varying from site to site and within a site from tumour to tumour, combined with the many interacting factors that influence the result of a course of radiotherapy add to the clinicians’ burden as they attempt to give individual patients the most appropriate treatment. The complexity has to be recognized and tackled by high-standard research, which must be interpreted carefully and applied to the individual case. Finally, the chosen schedule will only be successful if biological knowledge is applied to good planning and delivery of treatment. In every aspect of clinical research designed to improve the results of treatment, the importance of observation of the normal-tissue effects, particularly in the long term, must be emphasized and is essential to the evaluation of any approach to improve the care of patients by radiotherapy. Morbidity must be measured not only by the incidence of severe episodes but also by the incidence of moderate or even mild effects together with its duration. The total burden to the patient must be measured and recognized. The established knowledge concerning the factors influencing radiation response in tumours and in the normal tissues is now being extended using new techniques developed from advances in molecular biology. There is great promise that this will lead to further improvement of radiotherapy. Clinical oncologists must keep in close contact with their laboratory colleagues and understand the tumour biology and radiobiology important to their work. A high standard of clinical science must then be brought to the bedside so that advances originating in the laboratory can be exploited to the benefit of the patient with cancer.
REFERENCES ◆1
●2
◆3
◆4
●5
●6
Begg AC, Steel, GG. In: Steel GG (ed.) Basic Clinical Radiobiology, 3rd edn. London: Arnold, 2002 pages 8-22 Fertil B, Malaise EP. Inherent cellular radiosensitivity as a basic concept for human tumor radiotherapy. Int J Radiat Oncol Biol Phys 1981; 7(5):621–9. Steel G. Growth Kinetics of Tumours: Cell Population Kinetics in Relation to the Growth and Treatment of Cancer. Oxford: Oxford University Press, 1977. McMillan JT. Genetic control of the cellular response to ionizing radiation. In: Steel G (ed.) Basic Clinical Radiobiology. London: Arnold, 2002, 89–90. Burnet N, Nyman J, Turesson I, Wurm R, Yarnold J, Peacock J. Prediction of normal-tissue tolerance to radiotherapy from invitro cellular radiation sensitivity. Lancet 1992; 339:1570–1. Johansen J, Bentzen S, Overgaard J, Overgaard M. Relationship between the in vitro radiosensitivity of skin
7
●8
9
◆10
◆11
12
◆13
◆14
●15
●16
17
●18
●19
◆20
●21
◆22
◆23
fibroblasts and the expression of subcutaneous fibrosis, telangiectasia, and skin erythema after radiotherapy. Radiother Oncol 1996; 40:101–9. West CML, Davidson SE, Elyan SAG, et al. Lymphocyte sensitivity is a significant prognostic factor for morbidity in carcinoma of the cervix. Int J Radiat Oncol Bio Phys 2001; 51:10–15. West C, Davidson S, Roberts S, Hunter R. The independence of intrinsic radiosensitivity as a prognostic factor for patient response to radiotherapy of carcinoma of the cervix. Br J Cancer 1997; 76(9):1184–90. Brock W, Brown B, Goefpert H, Peters L. In vitro radiosensitivity of tumor cells and local tumor control by radiotherapy. In: Dewey W, Edington M, Fry R, Hall E, Whitmore G (eds) Radiation Research: a Twentieth Century Perspective. San Diego: Academic Press, 1992, 696–9. McMillan TJ, Steel G. DNA damage and cell killing. In: Steel G (ed.) Basic Clinical Radiobiology, 3rd edn. London: Arnold, 2002, 71–83. Begg AL. Individualization of radiotherapy. In: Steel G G (ed.) Basic Clinical Radiobiology, 3rd edn. London: Arnold, 2002, 241–50. Trotti A, Bentzen SM. The need for adverse effects reporting standards in oncology clinical trials. J Clin Oncol 2004; 22(1):19–22. Rubin P. Law and order of radiation sensitivity – absolute versus relative. Front Radiat Ther Oncol 1989; 23:7–40. Emami B, Lyman J, Brown A, et al. Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 1991; 21:109–22. Brahme A. Dosimetric precision requirements in radiation therapy. Acta Radiol Oncol 1984; 23:379–91. Bentzen SM, Overgaard M. Relationship between early and late normal-tissue injury after postmastectomy radiotherapy. Radiother Oncol 1991; 20:159–65. Bentzen SM. Steepness of the clinical dose-control curve and variation in the in vitro radiosensitivity of head and neck squamous cell carcinoma. Int J Radiat Biol 1992; 61(3):417–23. Strandqvist M. Studien Uber Die Kumulative Wirkung Der Rontgenstrahlen Bei Franktionierung. Erhfahrungen Aus Dem Radiumhemmet An 280 Haut Und Lippenkarzinomen. Acta Radiol 1944; Suppl. 55:101–4, 108–9, 120–1, 287–92. Ellis F. Dose, time and fractionation: a clinical hypothesis. Clin Radiol 1969; 20:1–7. Thames H, Bentzen S, Turesson I, Overgaard M, Van den Bogaert W. Time – dose factors in radiotherapy: a review of the human data. Radiother Oncol 1990; 19:219–35. Barendsen GW. Dose fractionation, dose rate and iso-effect relationships for normal tissue responses. Int J Radiat Oncol Biol Phys 1982; 8:1981–97. Bentzen S. Dose – response relationships in radiotherapy. In: Steel G (ed.) Basic Clinical Radiobiology, 3rd edn. London: Arnold, 2002, 94–104. Joiner MC, Bentzen SM. Time – dose relationships: the linear-quadratic approach. In: Steel G (ed.) Basic Clinical Radiobiology, 3rd edn. London: Arnold, 2002, 120–33.
54 Clinical radiobiology
24 Bentzen SM, Ritter MA. The α/β ratio for prostate cancer: what is it, really? Radiother Oncol 2005; 76(1):1–3. 25 Fowler JF, Ritter MA. A rationale for fractionation for slowly proliferating tumors such as prostatic adenocarcinoma. Int J Radiat Oncol Biol Phys 1995; 32(2):521–9. 26 Yarnold J, Ashton A, Bliss J, et al. Fractionation sensitivity and dose response of late adverse effects in the breast after radiotherapy for early breast cancer: long-term results of a randomised trial. Radiother Oncol 2005; 75(1):9–17. ◆27 Joiner MC, Marples B, Lambin P, Short SC, Turesson I. Lowdose hypersensitivity: current status and possible mechanisms. Int J Radiat Oncol Biol.Phys 2001; 49(2):379–89. 28 Saunders M, Shah N, Joiner M. Low dose hyperradiosensitivity – is it a clinical reality and can it influence the radiation schedules used in the treatment of gliomas? Int J Radiat Oncol Biol Phys 1999; 45(3; Supplement: Proceedings of the American Society for Therapeutic Radiology and Oncology 41st Annual Meeting):267–8. ◆29 Joiner MC. Models of radiation cell killing. In: Steel G G (ed.) Basic Clinical Radiobiology, 3rd edn. London: Arnold, 2002, 65–70. ●30 Horiot JC, Bontemps P, van den Bogaert W, et al. Accelerated fractionation (AF) compared to conventional fractionation (CF) improves loco-regional control in the radiotherapy of advanced head and neck cancers: results of the EORTC 22851 randomized trial. Radiother Oncol 1997; 44(2):111–21. 31 Horiot JC, Le Fur R, N’Guyen T, et al. Hyperfractionation versus conventional fractionation in oropharyngeal carcinoma: final analysis of a randomized trial of the EORTC cooperative group of radiotherapy. Radiother Oncol 1992; 25(4):231–41. 32 Horiot J, Le Fur R, N’Guyen T, et al. Hyperfractionated compared with conventional radiotherapy in oropharyngeal carcinoma: an EORTC randomised trial. Eur J Cancer 1990; 26(7):779–80. 33 Horiot JC. Controlled clinical trials of hyperfractionated and accelerated radiotherapy in otorhinolaryngologic cancers. Bull Acad Natl Med 1998; 182(6):1247–60; Discussion 1261. ●34 Bentzen SM, Saunders MI, Dische S. Repair halftimes estimated from observations of treatment-related morbidity after CHART or conventional radiotherapy in head and neck cancer. Radiother Oncol 1999; 53(3):219–26. ●35 Begg AC, McNally NJ, Shrieve DC, Karcher HA. A method to measure the duration of DNA synthesis and the potential doubling time from single sample. Cytometry 1985; 6:620–6. ◆36 Begg AC. Derivation of cell kinetic parameters from human tumours after labelling with bromodeoxyuridine or iododeoxyuridine. In: McNally NJ (ed.) The Scientific Basis of Modern Radiotherapy. London: British Institute of Radiology, 1989, 113–19. ●37 Wilson GD, McNally NJ, Dische S, Bennett MH. Cell proliferation in human tumours measured by in vivo labelling with bromodeoxyuridine. Br J Radiol 1988; 61:419–22.
◆38
●39
●40
◆41
●42
43 44
45
46
●47
●48
◆49
●50
●51
52
●53
Denekamp J. Cell Kinetics and Cancer Therapy. Springfield, IL: Charles C Thomas, 1982. Bennett MH, Wilson GD, Dische S, Saunders MI, Martindale CA, O’Halloran A. Tumour proliferation assessed by combined histological and flow cytometric analysis: implications for therapy in squamous cell carcinoma in the head and neck. Br J Cancer 1992; 65:870–8. Withers H, Taylor J, Maciejewski B. The hazard of accelerated tumour clonogen repopulation during radiotherapy. Acta Oncol 1988; 27:131–46. Tubiana M. Repopulation in human tumors. Acta Oncol 1988; 27:83–8. Dische S, Saunders MI. The rationale for continuous, hyperfractionated, accelerated radiotherapy (CHART). Int J Radiat Oncol Biol Phys 1990; 19:1317–20. Dische S. The clinical science of radiation oncology. Regaud Lecture, Malmo, 1992. Radiother Oncol 1993; 28:93–107. Bentzen SM, Thames HD. Clinical evidence for tumour clonogen regeneration: interpretations of the data. Radiother Oncol 1991; 22:161–6. Maciejewski B, Withers HR, Taylor JM, Hliniak A. Dose fractionation and regeneration in radiotherapy for cancer of the oral cavity and oropharynx: tumor dose – response and repopulation. Int J Radiat Oncol Biol Phys 1989; 16(3):831–43. Fowler JF. Apparent rates of proliferation of acutely responding normal tissues during radiotherapy of head and neck cancer. Int J Radiat Oncol Biol Phys 1991; 21:1451–6. Fowler JF, Lindstrom MJ. Loss of local control with prolongation in radiotherapy. Int J Radiat Oncol Biol Phys 1992; 23(2):457–67. Overgaard J, Alsner J, Eriksen J, Horsman MR, Grau C. Importance of overall treatment time for the response to radiotherapy in patients with squamous cell carcinoma of the head and neck. Rays 2000; 25(3):313–19. Baumann M, Krause M. Targeting the epidermal growth factor receptor in radiotherapy: radiobiological mechanisms, preclinical and clinical results. Radiother Oncol 2004; 72(3):257–66. Baumann M, Krause M, Zips D, et al. Selective inhibition of the epidermal growth factor receptor tyrosine kinase by BIBX1382BS and the improvement of growth delay, but not local control, after fractionated irradiation in human FaDu squamous cell carcinoma in the nude mouse. Int J Radiat Biol 2003; 79(7):547–59. Akimoto T, Hunter NR, Buchmiller L, Mason K, Ang KK, Milas L. Inverse relationship between epidermal growth factor receptor expression and radiocurability of murine carcinomas. Clin Cancer Res 1999; 5(10):2884–90. Krause M, Hessel F, Zips D, Hilberg F, Baumann M. Adjuvant inhibition of the epidermal growth factor receptor after fractionated irradiation of FaDu human squamous cell carcinoma. Radiother Oncol 2004; 72(1):95–101. Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006; 354(6):567–78.
References 55
54 Horiot J, Le Fur R, N’Guyen T, et al. Hyperfractionation versus conventional fractionation in oropharyngeal carcinoma: final analysis of a randomized trail of the EORTC cooperative group of radiotherapy. Radiother Oncol 1992; 25:231–41. 55 Horiot J, Le Fur R, Schraub S, et al. Status of the EORTC Cooperative Group of radiotherapy with hyperfractionated and accelerated radiotherapy regimes. Semin Rad Oncol 1992; 2:34–7. ●56 Fu KK, Pajak TF, Trotti A, et al. A Radiation Therapy Oncology Group (RTOG) phase III randomized study to compare hyperfractionation and two variants of accelerated fractionation to standard fractionation radiotherapy for head and neck squamous cell carcinomas: first report of RTOG 9003. Int J Radiat Oncol Biol Phys 2000; 48(1):7–16. 57 Wang C, Suit H, Blitzer P. Twice-a-day radiation therapy for supraglottic carcinoma. Int J Radiat Oncol Biol Phys 1985; 12:3–7. ●58 Ang KK, Peters LJ, Weber RS, et al. Concomitant boost radiotherapy schedules in the treatment of carcinoma of the oropharynx and nasopharynx. Int J Radiat Oncol Biol Phys 1990; 19(6):1339–45. ●59 Overgaard J, Hansen HS, Specht L, et al. Five compared with six fractions per week of conventional radiotherapy of squamous-cell carcinoma of head and neck: DAHANCA 6 and 7 randomised controlled trial. Lancet 2003; 362(9388):933–40. ●60 Dische S, Saunders M, Barrett A, Harvey A, Gibson D, Parmar M. A randomised multicentre trial of CHART versus conventional radiotherapy in head and neck cancer. Radiother Oncol 1997; 44(2):123–36. ●61 Van den Bogaert W, van der Schueren E, Horiot JC, et al. Early results of the EORTC randomized clinical trial on multiple fractions per day (MFD) and misonidazole in advanced head and neck cancer. Int J Radiat Oncol Biol Phys 1986; 12(4):587–91. 62 Van den Bogaert W, van der Schueren E, Horiot JC, et al. The EORTC randomized trial on three fractions per day and misonidazole in advanced head and neck cancer: prognostic factors. Radiother Oncol 1995; 35(2):100–6. 63 Saunders M, Dische S, Barrett A, Harvey A, Griffiths G, Palmar M. Continuous, hyperfractionated, accelerated radiotherapy (CHART) versus conventional radiotherapy in non-small cell lung cancer: mature data from the randomised multicentre trial. CHART Steering Committee. Radiother Oncol 1999; 52(2):137–48. 64 Dische S, Saunders M, Barrett A, et al. A randomised multicentre trial of CHART versus conventional radiotherapy in head and neck cancer. Radiother Oncol 1997; 44:123–36. 65 Overgaard J, Hansen HS, Jorgensen K, Hjelm Hansen M. Primary radiotherapy of larynx and pharynx carcinoma – an analysis of some factors influencing local control and survival. Int J Radiat Oncol Biol Phys 1986; 12(4):515–21. ●66 Gray LH, Conger AD, Ebert M. The concentration of oxygen dissolved in tissues at the time of irradiation as a factor in radiotherapy. Br J Radiol 1953; 26:638–48.
67 Dische S, Saunders M. Hypoxic cell radiosensitisation – clinical relevance in 1998. In: Kogelnik H, Sedlmayer F (eds) Progress in Radio-Oncology VI. Bologna: Monduzzi Editore International Proceedings Division, 1998, 205–21. ●68 Dische S, Rojas A, Rugg T, Hong A, Michael BD. Carbogen breathing: a system for use in man. Br J Radiol 1992; 65:87–90. 69 Dische S. The clinical use of hyperbaric oxygen and chemical hypoxic cell sensitizers. In: Steel A, Peckham MJ, (eds.) The Biological Basis of Radiotherapy. Philadelphia]: Elsevier Science Publishers, 1983, 225–37. ◆70 Dische S. Radiotherapy and radiosensitizers. Br J Cancer 1982; 45:628–9. 71 Dische S. The clinical use of hyperbaric oxygen and chemical hypoxic cell sensitizers. In: Steel A, Peckham MJ, (eds.) The Biological Basis of Radiotherapy. Philadelphia: Elsevier Science Publishers, 1983, ●72 Overgaard J, Sand Hansen H, Lindelov B, et al. Nimorazole as a hypoxic radiosensitizer in the treatment of supraglottic larynx and pharynx carcinoma. First report from the Danish Head and Neck Cancer Study (DAHANCA) protocol 5–85. Radiother Oncol 1991; 20(Suppl. 1):143–9. 73 Overgaard J Hanson SH, Overgaard M, Bastholt L, et al. Randomized double-blind phase III study of nimorazole as a hypoxic cell radiosensitizer of primary radiotherapy in supraglottic larynx and pharynx carcinoma. Results of the Danish Head and Neck Cancer Study (DAHANCA) protocol 5–85. Radiother Oncol 1998; 46:S31. ◆74 Horsman M, Hoyer M, Honess D, Dennis I, Overgaard J. Nicotinamide pharmacokinetics in humans and mice: a comparative assessment and the implications for radiotherapy. Radiother Oncol 1993; 27(2):131–9. ◆75 Denekamp J, Fowler JF. ARCON – current status: a summary of a workshop on preclinical and clinical studies. Acta Oncol 1997; 36(5):517–25. 76 Hoskin PJ, Saunders MI, Phillips H, et al. Carbogen and nicotinamide in the treatment of bladder cancer with radical radiotherapy. Br J Cancer 1997; 76(2):260–3. ●77 Kaanders JHAM, Pop LAM, Marres HAM, et al. ARCON: experience in 215 patients with advanced head and neck cancer. Int J Radiat Oncol Biol Phys 2002; 52(3):769–78. ◆78 Brown JM, Giaccia AJ. Tumour hypoxia: the picture has changed in the 1990s. Int J Radiat Biol 1994; 65(1):95–102. ●79 Rischin D, Peters L, Fisher R, et al. Tirapazamine, cisplatin, and radiation versus fluorouracil, cisplatin, and radiation in patients with locally advanced head and neck cancer: a randomized phase II trial of the Trans-Tasman Radiation Oncology Group (TROG 98.02). J Clin Oncol 2005; 23(1):79–87. ●80 Denekamp J. Vascular attack as a therapeutic strategy for cancer. Cancer Metastasis Rev 1990; 9(3):267–82. 81 Denekamp J. Anti-vascular cancer therapy: is it a dream or reality? Eur J Surg Suppl 1991(561):21–6. 82 Denekamp J. The tumour microcirculation as a target in cancer therapy: a clearer perspective. Eur J Clin Invest 1999; 29(9):733–6.
56 Clinical radiobiology
83 Zhao D, Jiang L, Hahn EW, Mason RP. Tumor physiologic response to combretastatin A4 phosphate assessed by MRI. Int J Radiat Oncol Biol Phys 2005; 62(3):872–80. 84 Cooney MM, Ortiz J, Bukowski RM, Remick SC. Novel vascular targeting/disrupting agents: combretastatin A4 phosphate and related compounds. Curr Oncol Rep 2005; 7(2):90–5. 85 Horsman MR, Murata R. Combination of vascular targeting agents with thermal or radiation therapy. Int J Radiat Oncol Biol Phys 2002; 54(5):1518–23. ◆86 Horsman MR, Overgaard J. The Oxygen Effect and Tumour Microenvironment in Basic Clinical Radiobiology, 3rd edn. London: Arnold, 2002 pages 8-22 87 Corry J, Rischin D. Strategies to overcome accelerated repopulation and hypoxia – what have we learned from clinical trials? Semin Oncol 2004; 31(6):802–8. 88 Hicks RJ, Rischin D, Fisher R, Binns D, Scott AM, Peters LJ. Utility of FMISO PET in advanced head and neck cancer treated with chemoradiation incorporating a hypoxiatargeting chemotherapy agent. Eur J Nucl Med Mol Imaging 2005; 32(12):1384–91. 89 Piert M, Machulla HJ, Picchio M, et al. Hypoxia-specific tumor imaging with 18F-fluoroazomycin arabinoside. J Nucl Med 2005; 46(1):106–13. 90 Rasey JS, Koh W-J, Evans ML, et al. Quantifying regional hypoxia in human tumours with positron emission tomography of [18F]-fluoromisonidazole: a pretherapy study of 37 patients. Int J Radiat Oncol Biol Phys 1996; 36(2):417–28. 91 Sorensen M, Horsman MR, Cumming P, Munk OL, Keiding S. Effect of intratumoral heterogeneity in oxygenation status on FMISO PET, autoradiography, and electrode PO2 measurements in murine tumors. Int J Radiat Oncol Biol Phys 2005; 62(3):854–61. 92 Dische S. Radiotherapy and anaemia – the clinical experience. Radiother Oncol 1991; 20:35–40. 93 Bush RS, Jenkin RDT, Allt WEC, et al. Definitive evidence for hypoxic cells influencing cure in cancer therapy. British Journal of Cancer 1978; 37:302–6. ●94 Henke M, Laszig R, Rube C, et al. Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial. Lancet 2003; 362(9392):1255–60. ●95 Lavey RS, Dempsey WH. Erythropoietin increases hemoglobin in cancer patients during radiation therapy. Int J Radiat Oncol Biol Phys 1993; 27(5):1147–52. 96 Lindegaard JC, Grau C. Has the outlook improved for amifostine as a clinical radioprotector? Radiother Oncol 2000; 57:113–18. 97 Kligerman M, Turrisi A, Urtasun R, et al. Final report on phase I trial of WR-2721 before protracted fractionated radiation therapy. Int J Radiat Oncol Biol Phys 1988; 14:1119–22. 98 Bourhis J, Crevoisier RD, Abdulkarim B, et al. A randomised study of very accelerated radiotherapy with and without amifostine in head and neck squamous cell carcinoma. Int J Radiat Oncol Biol Phys 2000; 46(5):1105–8.
99 Demiral AN, Yerebakan O, Simsir V, Alpsoy E. Amifostineinduced toxic epidermal necrolysis during radiotherapy: a case report. Jpn J Clin Oncol 2002; 32(11):477–9. ◆100 Brizel DM, Overgaard J. Does amifostine have a role in chemoradiation treatment? Lancet Oncol 2003; 4(6):378–81. ● 101 Brizel DM, Wasserman TH, Henke M, et al. Phase III randomized trial of amifostine as a radioprotector in head and neck cancer. J Clin Oncol 2000; 18(19): 3339–45. 102 Wasserman TH, Brizel DM, Henke M, et al. Influence of intravenous amifostine on xerostomia, tumor control, and survival after radiotherapy for head-and- neck cancer: 2year follow-up of a prospective, randomized, phase III trial. Int J Radiat Oncol Biol Phys 2005; 63(4):985–90. ◆103 Steel GG, Peckham MJ Terminology in the description of drug-radiation interactions. Int J Radiat Oncol Biol Phys 1979; 5:1145-50 104 Pignon JP, Bourhis J, Domenge C, Designe L. Chemotherapy added to locoregional treatment for head and neck squamous-cell carcinoma: three meta-analyses of updated individual data. MACH-NC Collaborative Group. Metaanalysis of chemotherapy on head and neck cancer. Lancet 2000; 355(9208):949–55. 105 Burdett S, Stewart L. Postoperative radiotherapy in nonsmall-cell lung cancer: update of an individual patient data meta-analysis. Lung Cancer 2005; 47(1):81–3. 106 De Ruysscher D, Pijls-Johannesma M, Vansteenkiste J, Kester A, Rutten I, Lambin P. Systematic review and meta-analysis of randomised, controlled trials of the timing of chest radiotherapy in patients with limitedstage, small-cell lung cancer. Ann Oncol 2005, Dec. 12, 17543–552 107 Pignon JP, Stewart LA. Randomized trials of radiotherapy alone versus combined chemotherapy and radiotherapy in stages IIIa and IIIb nonsmall cell lung cancer: a metaanalysis. Cancer 1996; 77(11):2413–14. 108 Ang KK, Trotti A, Brown BW, et al. Randomized trial addressing risk features and time factors of surgery plus radiotherapy in advanced head-and-neck cancer. Int J Radiat Oncol Biol Phys 2001; 51(3):571–8. ◆109 Bernier J, Cooper JS, Pajak TF, et al. Defining risk levels in locally advanced head and neck cancers: a comparative analysis of concurrent postoperative radiation plus chemotherapy trials of the EORTC (#22931) and RTOG (# 9501). Head Neck 2005. 27:843–850 ◆110 Joiner M. Particle beams in radiotherapy. In: Steel G G (ed.) Basic Clinical Radiobiology, 3rd edn. London: Arnold, 1997, 204–16. 111 Wambersie A, Richard F, Breteau N. Development of fast neutron therapy worldwide. Radiobiological, clinical and technical aspects. Acta Oncol 1994; 33(3):261–74. 112 Wambersie A, Hendry J, Gueulette J, Gahbauer R, Potter R, Gregoire V. Radiobiological rationale and patient selection for high-LET radiation in cancer therapy. Radiother Oncol 2004; 73(Suppl. 2):S1–14.
References 57
◆113
Suit H, Goitein M, Munzenrider J, et al. Evaluation of the clinical applicability of proton beams in definitive fractionated radiation therapy. Int J Radiat Oncol Biol Phys 1982; 8(12):2199–2205. ◆114 Steel GG. Basic Clinical Radiobiology, 3rd edn. London: Arnold, 2002. pp 192–204 ◆115 Brenner D, Huang Y, Hall E. Fractionated high dose-rate versus low dose-rate regimens for intracavitary
brachytherapy of the cervix: equivalent regimens for combined brachytherapy and external irradiation [review]. Int J Radiat Oncol Biol Phys 1991; 21(6):1415–23. ●116 Withers H, Taylor J, Maciejewski B. Treatment volume and tissue tolerance. Int J Radiat Oncol Biol Phys 1988; 14:751–9.
4 Mathematical modelling and its application in oncology ROGER G. DALE AND BLEDDYN JONES
Introduction The basis and routine application of radiobiological models More advanced considerations
58 59 66
INTRODUCTION Clinical experience and knowledge of the results of clinical trials can sometimes provide little more than tentative guidance on how to manage difficult clinical problems. It is useful for the treating physician to have a quantitative assessment of the likely clinical impact of more subtle changes in treatment, when dose per fraction, overall time, dose rate etc. are varied. Difficult cases can be formally assessed by using radiobiological models: from a numerical appraisal of the likely biological outcomes, the best available option can be selected or the radiation prescription can be modified in a direction compatible with safety of outcome. The amount of mathematics required for simple radiobiological assessments is well within the grasp of radiation oncologists, who should be aware of radiation effect models, their usefulness and limitations. Although computers and mathematical software packages are necessary for more advanced modelling, for a wide range of clinically relevant calculations the use of a pocket calculator is sufficient. Practice of these calculations is advised, as there are inherent subtleties. Radiation oncology is concerned with dose–effect relationships and their clinical optimization. With so many controllable variables (overall time, dose fractionation, choice of modalities, etc.) and inter-patient variations (in the biological and clinical parameters that influence outcomes), predictive models can sometimes be used to supplement clinical judgement, e.g. when an urgent decision is necessary. Randomized clinical trials can only provide
Conclusion Acknowledgement References
72 72 73
answers to situations in which the protocol is adhered to strictly and the information they provide only refers to those conditions. Modelling assessment can be used in the design of clinical trials, e.g. to define an equivalent radiotherapy fractionation schedule to be tested alongside a conventional schedule. Modelling assessments are already invaluable in special circumstances, e.g. treatment delivery errors (under-dosage or over-dosage) or if treatment is unintentionally interrupted, be this during external fractionated radiotherapy or brachytherapy. Where higher than usual doses of radiation are used, any local dose variations assume increased significance and modelling offers a valuable aid in the assessment of the resultant biological effect. Also, in the reporting of clinical data that comprise multiple different fractionation schedules, the quantification of the expected biological effect provides a method for inter-patient and inter-departmental comparisons. Modelling can be used to: ● ● ● ●
● ●
identify and modify dangerous treatment schedules, provide a standard way of reporting clinical data, optimize radiotherapy treatment outcomes, analyse clinical results from departments that use different fractionation or dose-rate schedules, aid in the design of better clinical trials, and test medico-legal issues.
This chapter examines the application of radiobiological models to a range of practical problems by means of many
The basis and routine application of radiobiological models 59
worked examples. The last part of the chapter considers more advanced examples that could improve the design of future radiotherapy schedules.
THE BASIS AND ROUTINE APPLICATION OF RADIOBIOLOGICAL MODELS
i.e.: E αd βd 2
(4.2)
For n such fractions, delivered several hours apart, both types of damage are n times greater, i.e.: E nαd nβd 2 n(αd βd 2)
(4.3)
Types of model Earlier radiobiological models were based on logarithmic power-law models similar in form to those used in physics to predict the behaviour of gases under adiabatic conditions. Such formulations were empirical, and significant deviations from the predicted results were observed when they were applied at doses (or fractional doses) that lie outside the range of values from which the empirical relationships were originally derived. Thus, at very low and high doses per fraction, errors were found. Additionally, when power laws were first introduced, there was little awareness that acute-reacting tissues behaved differently from latereacting tissues, although tissue-specific line fitting did overcome this problem to a limited extent. No further discussion is required here about power-law models, but practising oncologists should be aware that until quite recently, many patients received treatments that had been calculated using the Ellis, cumulative radiation effect (CRE) or time–dose factor (TDF) formulations. Thames and Hendry (1987) have written a detailed history of these models.1 More reliable radiobiological models have been derived from consideration of the biophysical events that govern radiation effects. Mathematically awkward multi-target and multi-hit models were initially developed, but the target was undefined before DNA structure was discovered. Currently the best and most commonly used radiobiological model is the linear-quadratic (LQ) formulation, which can be derived from consideration of physical lesions in double-stranded DNA and their repair and mis-repair.2–4 In LQ methodology, lethal radiation damage consists of two components, A and B. Type A lethal damage is caused by a single ionizing event, the amount increasing linearly with dose with a proportionality constant of α (unit Gy1). Type A damage is not influenced by changes in the pattern of fractionation. Type B lethal damage is a consequence of there being an interaction between two sub-lethal damage components, each created in different ionizing events with a proportionality constant of β (unit Gy2). Unlike Type A damage, the amount of Type B damage is dependent on the pattern of fractionation and on the dose rate. For radiation fractions delivered in short irradiation times, the Type B damage is proportional to the square of the dose per fraction. Thus, for a single acute fractional dose of d (Gy), the total effect may be written as the summation of the Type A and Type B components: Effect (E) Type A Type B
(4.1)
It is clear from the dose-squared terms in Equations 4.2 and 4.3 that high doses per fraction create proportionately more effect than lower doses per fraction, the extent of this effect being governed by the relative magnitude of the β parameter compared to α. The LQ formulation is reliable over the range of dose per fraction encountered in conventional radiotherapy, but there are indications that it is less accurate at large (6 Gy) and low (1.8 Gy) values.
The concept of biologically effective dose The biologically effective dose (BED) is a useful measure of biological effect that allows comparison of different radiation schedules. The BED (in units of Gy) for a conventional course of fractionated radiotherapy is found by dividing both sides of Equation 4.3 by α and re-arranging, i.e.: BED
⎛ E βd 2 ⎞⎟ ⎟⎟ n ⎜⎜ d ⎜⎝ α α ⎟⎠
i.e. ⎡ d ⎤⎥ BED nd ⎢1 ⎢ α /β ⎥⎦ ⎣
(4.4)
The bracketed term in Equation 4.4 is called the Relative Effectiveness (RE) per Unit Dose and essentially converts physical dose to biological dose. The α/β ratio (often called the fractionation sensitivity) is a measure of how a particular tissue will respond to changes in fractionation or dose rate. Consequently, any calculated BED is also tissue specific, and normally written with the α/β factor as a suffix to the dose unit, e.g. a BED of 100 Gy derived using α/β 3 Gy is written as 100 Gy3. As the product nd in Equation 4.4 is the total physical dose (TD) delivered by the treatment: biological dose (BED) total physical dose(TD) relative effectiveness (RE) (4.5) The relationship summarized in Equation 4.5 applies also to radiotherapy that does not involve well-spaced fractions and to brachytherapy. In such cases, the form of the RE factor is more complex.
60 Mathematical modelling and its application in oncology
What does the BED represent? The BED is the (hypothetical) dose required for a given biological effect when delivered by the gentlest form of radiotherapy, i.e. using very small doses per fraction or, in the case of continuous irradiation, very low dose rates. To understand this, consider Equation 4.3 in situations in which d is progressively decreased, but with n correspondingly increased in order to maintain an iso-effect. The βd 2 term then becomes very small relative to αd and E 艑 nαd 艑 αD, where D is the total physical dose. Thus, the BED (i.e. E/α) corresponds to the iso-effective physical dose when delivered using very low dose per fraction or very low dose rate. The strength of the BED concept is that it can be calculated for any practical situation in radiotherapy, and a particular BED value may be achieved by a wide variety of schedules. It then follows that the biological effects of quite different schedules can be inter-compared using this single radiobiological yardstick. Also, because BED is tissue specific, it is a ‘one-number’ representation of either tumour or normaltissue response at the point of calculation. Caution is required in that a single value of BED cannot represent an entire volume of tissue unless further, more complex, calculations are made. Finally, it should be noted that BEDs are additive, thus a single BED can be used to represent a radiotherapy schedule consisting of two or more components (e.g. external radiotherapy plus brachytherapy) providing the doses are taken for the same point of interest.3 Two fractionation schedules of total doses D1 and D2 and dose per fractions of d1 and d2 respectively are said to be iso-effective on a particular tissue if: ⎡ ⎡ d ⎤ d ⎤ D1 ⎢1 1 ⎥ D2 ⎢1 2 ⎥ ⎢ ⎥ ⎢ α /β ⎦ α /β ⎥⎦ ⎣ ⎣
sometimes prudent to select lower values for situations in which normal-tissue tolerance may be compromised due to age, concomitant medical conditions, etc. A case in point is spinal cord, for which the fractionation sensitivity is around 1.7 Gy.5 Generic values of α/β derived from animal studies may not be representative for clinical use because of natural variation in α and β values in humans. Even for tumours of similar histology, radiosensitivity variations in heterogenous clinical data are a potentially significant factor affecting the reliability of BED-based predictions. It is clear that fractionation sensitivities need to be more accurately determined in a wider range of human tumours and normal tissues.
Worked example What are the tumour and late-responding BEDs associated with 45 Gy/25 fractions at the 100 per cent prescription surface and at a point where the iso-dose surface is given as 107 per cent? Using generic α/β values of 10 Gy for tumour and 3 Gy for late-responding normal tissue we have: ⎡ 1.8 ⎤ ⎥ 53.1 Gy10 BEDtumour 45 ⎢1 ⎢⎣ 10 ⎥⎦
and: ⎡ 1.8 ⎤ ⎥ 72 Gy 3 BEDlate 45 ⎢1 ⎢⎣ 3 ⎥⎦
(4.6)
Equation 4.6 may thus be utilized to design a fractionated treatment which is iso-effective to a first, or it may be used to estimate α/β ratios if parameters D1, D2 and d1 and d2 are already known from existing iso-effect treatments that are similar in their effects on a specific tissue. FRACTIONATION SENSITIVITIES: (α/β) RATIOS
These may be derived from data sets where different fractionation schedules have been used to achieve a common clinical endpoint. Typical α/β value ranges are 10–30 Gy for squamous-cell carcinomas and 4 Gy for breast cancer.5 There is a continuing debate about the value of α/β in prostate cancer; the value may be as low as 1.5 Gy.6 In the case of melanoma, it may also be small.1 A generic value of 10 Gy is often used for the tumour α/β when the exact value is not known, but a range of realistic values should be used wherever possible. Normal-tissue α/β ratios are generally smaller (1–4 Gy) than those for experimental cancers and many human tumours. A generic value of 3 Gy is often used, but it is
The fact that BEDlate is greater than BEDtumour does not mean that the late-responding tissue is receiving a greater biological dose than the tumour. Rather, the two numbers represent how much dose would be required for the same respective effects on the two tissues if a very gentle treatment (very low dose per fraction) could be used. For the 107 per cent surface we introduce a factor of 1.07 to the total dose and dose per fraction to give: ⎡ 1.8 1.07 ⎤ ⎥ 57.4 Gy10 BEDtumour 45 1.07 ⎢1 ⎢⎣ ⎥⎦ 10
and ⎡ 1.8 1.07 ⎤ ⎥ 79.1 Gy 3 BEDlate 45 1.07 ⎢1 ⎢⎣ ⎥⎦ 3 It can be seen that the relative increment in BED is greatest for the late reacting-tissue.
The basis and routine application of radiobiological models 61
Table 4.1 Fractionation schedules summarized and compared in terms of BED values (modified from Fowler (1989) with permission). Schedule (a) 68 fractions 1.2 Gy 81.6 Gy (b) 35 fractions 2 Gy 70 Gy (c) 30 fractions 2.0 Gy 60 Gy (d) 20 fractions 2.8 Gy 56 Gy (e) 36 fractions 1.5 Gy 54 Gy (f) 16 fractions 3.38 Gy 54 Gy
BED2
BED3
BED4
130.6 (3) 140.0 (2) 120.0 (5) 134.4 (4) 94.5 (6) 145.5 (1)
114.2 (3) 116.7 (1) 100.0 (5) 108.3 (4) 81.0 (6) 115.0 (2)
106.1 (1) 105.0 (2) 90.0 (5) 95.2 (4) 74.3 (6) 99.8 (3)
For explanation, see text.
Table 4.1 – modified from Fowler, 1989, with permission7 – gives some examples of how a selection of disparate fractionation schedules can be summarized and compared in terms of BED values. In the left-hand column the schedules are ranked in order of their total physical dose. In columns two, three and four are listed the respective later-responding BEDs corresponding to α/β values of 2, 3 and 4 Gy. The numbers in bold show, for each value of α/β, the BED lateeffect rankings of each of the schedules. The table illustrates two important points relating to radiobiological assessments. First, total dose alone is a poor predictor of radiobiological effect, as evidenced in particular by schedule f. Second, the ‘hotness’ ranking is influenced by the particular α/β selected to perform the BED calculation. Thus, where specific values are not known, it is prudent to use two or more α/β values in order to obtain a more balanced view. The occurrence of regions of high dose that exceeds the prescription dose can be a problem in radiotherapy. Although it may be beneficial within a tumour, if a critical volume of late-reacting normal tissue is included, increased toxicity may occur because small increments in dose per fraction will yield greater increments in effect due to the influence of the quadratic relationship. Withers coined the term ‘double trouble’ for this phenomenon. With further increases in the prescribed dose per fraction, i.e. as occurs in hypofractionated radiotherapy, a further increment in effect occurs when BED is estimated in high-dose regions. This additional penalty has been called ‘treble trouble’.8 The corollary of this is that additional care in treatment planning is necessary with hypofractionated radiotherapy of large volumes where dose homogeneity is common, as, for example, in breast radiotherapy.
The consequences of tumour repopulation Tumour cells originated during treatment reduce overall treatment efficacy and need to be accounted for in calculations. If repopulation occurs at a continuous (exponential) rate throughout treatment, the net effect depends on the treatment duration (T) and the effective tumour doubling
Worked example Compare the BEDs of (a) 50 Gy in 25 fractions and (b) an iso-effective dose given in 6 fractions to a volume of normal tissue (assumed α/β 2 Gy) that receives 120 per cent of the prescribed dose. What might the expected increase in complication rate be if the slope of the dose–response curve is approximately 1 per cent per Gy3? The iso-effective dose in 6 fractions is found from the solution of d in: ⎛ ⎛ 2⎞ d⎞ 50 ⎜⎜1 ⎟⎟⎟ 6d ⎜⎜1 ⎟⎟⎟ ⎜⎝ ⎜⎝ 3 ⎟⎠ 3 ⎟⎠ d 5.13 Gy. The BEDs at 120 per cent will be: (a) 50 1.2(1 1.2 2/3) 108 Gy3 (b) 6 1.2 5.13(1 1.2 5.13/3) 112.7 Gy3. The increment in BED due to hypofractionation (treble trouble) is 112.7 108 4.7 Gy3. This would imply a 4–5 per cent increase in late tissue complications.
time teff (in units of days). The consequence of this is that the expression for biological effect (E) in Equation 4.3 is modified to: E n(αd βd 2 ) 0.693
T t eff
(4.7)
To express this in terms of BED, it is necessary to divide E by α, as occurred in the derivation of Equation 4.4, i.e.: ⎡ d ⎤⎥ T BED nd ⎢1 0.693 ⎢ ⎥ α / β α t eff ⎣ ⎦
(4.8)
62 Mathematical modelling and its application in oncology
If the subtractive repopulation factor in Equation 4.8 is simply referred to as RF, the aide-memoir for calculating BED [Equation 4.5] is then changed to: BED TD RE RF
detectable.9 To allow for this, a simple time time-delay factor (TK can be incorporated into the LQ equations, i.e. Equation 4.8 becomes: ⎡ (T TK ) d ⎤⎥ BED nd ⎢1 0.693 ⎢ ⎥ α / β αt eff ⎣ ⎦
(4.9)
where RF (units of Gy) is a measure of the biological dose ‘wasted’ in combating repopulation. It will be noted from Equation 4.8 that inclusion of an allowance for repopulation in the calculation of a tumour BED requires a value for α to be assumed. This potential disadvantage may be overcome by using clinically derived RF values, e.g. in head and neck squamous-cell cancers, the average daily repopulation may be equivalent to as much as 0.6 Gy/day, i.e. over a 6-week (42-day) treatment, the RF factor (i.e. the ‘wasted’ dose) is 42 0.6 25.2 Gy.
Worked example Comparison of tumour and late-normal tissue BEDs delivered by three schedules: (a) total dose of 60 Gy in 2-Gy fractions over 40 days (b) total dose of 50 Gy in 2.5-Gy fractions over 26 days (c) total dose of 52.5 Gy in 2.625-Gy fractions over 26 days. Assumed tumour parameters: α/β 10 Gy, α 0.35/Gy, teff 5 days. The assumed α/β for the normal tissue (which is assumed to receive the same dose schedule as the tumour) is 3 Gy. The results of applying Equations 4.5 and 4.8 are summarized in Table 4.2.
The table indicates how biological effects are governed by dose per fraction, treatment duration and total dose. Particularly noteworthy is the fact that schedules (a) and (c) are nearly identical in terms of their effects on both tumour and late-responding tissue, yet the total dose delivered in the latter schedule is 7.5 Gy (12.5 per cent) less than in the former.
Delayed repopulation For some tumours (squamous-cell cancers), there is an apparent time delay before significant repopulation is
and can be simplified as: ⎡ d ⎤⎥ BED nd ⎢1 K (T TK ) ⎢ α /β ⎥⎦ ⎣
(a) 60 Gy/30 fractions/40 days (b) 50 Gy/20 fractions/26 days (c) 52.5 Gy/20 fractions/26 days
(4.11)
where K is now the required dose equivalent of repopulation per day (units of BED Grays per day). Clearly the repopulation factor in Equation 4.11 is only required in cases where TK is less than T. TK values of 21–28 days are often recommended, although 21 days is probably most appropriate, as there is evidence of a time effect for tumour control beyond 3 weeks. For times that are less than TK, used in hypofractionated schedules, we advise that a smaller value of K be used, e.g. around 0.1–0.2 Gy10 per day. This is because it is more realistic to assume a slow rate of repopulation at short treatment times than to assume that there is no repopulation at all. The above considerations have also been applied to considering the implications of on-going tumour proliferation on overall tumour cure probability when there are delays in the initiation of treatment.10
Loss of control with extended treatment time: the effect of unscheduled treatment interruptions Tumour repopulation during treatment effectively wastes a significant amount of the delivered dose. It follows that for rapidly growing tumours, the presence of unintended gaps in the treatment schedule will have the effect of wasting even more dose. It has been estimated that for head and neck tumours, the loss of control amounts to 1–2 per cent per day of treatment extension.11 However, this figure is from data sets that included a wide range of parameter variations within the treated population and may seriously
Table 4.2 The results of applying equations 4.5 and 4.8.
Schedule
(4.10)
Tumour BED (Gy10) [via Equation 5.8]
Normal tissue BED (Gy3) [via Equation 5.5]
72.0 15.8 56.2 62.5 10.3 52.2 66.3 10.3 56.0
100.0 91.7 98.4
The basis and routine application of radiobiological models 63
underestimate the loss of control in individual patients. Tumour loss of control with increasing time depends on the expected tumour cure probability (TCP) and is most significant when the TCP is in the middle range, for example between 30 per cent and 60 per cent. In such situations, the values may be as high as 5 per cent per day for individuals. The potential seriousness of unscheduled gaps in treatment is such that the Royal College of Radiologists (RCR) has issued guidelines on how to deal with the issue.12,13 There are many technical, clinical and social reasons why individual patients might miss one or more fractions during a treatment course. It is important to appreciate that the treatment gap is less of a problem if the prescribed treatment can still be completed within the originally stipulated treatment time. More significant radiobiological problems arise if it is necessary to extend the overall treatment time, as may be the case for gaps occurring late during the treatment. There are essentially three methods that may be employed to offset the effects of an unscheduled treatment gap once it has occurred.14 1. Maintain the original prescription time by treating the ‘missing’ fractions at weekends and using the prescribed dose per fraction. 2. Maintain the original prescription time by employing twice-daily fractionation (using the prescribed dose per fraction) with a minimum of 6-hour intervals between daily fractions. 3. Devise a radiobiological compensation employing altered fractional doses. Methods 1 and 2 (either individually or in combination) are the preferred options, as they maintain the number of fractions and dose per fraction as originally prescribed. However, if the unscheduled gap occurs late in the schedule, it can be impossible to provide dose compensation without increasing the treatment time, so method 3 is used.
Worked example A patient prescribed a schedule of 66 Gy in 33 fractions over 6.5 weeks misses the whole of the second week of treatment (five fractions). How can this unscheduled gap be compensated for? Answer: After the gap, i.e. at the beginning of the third week of treatment, there are still 28 fractions to be delivered. If weekend working is possible then the ‘lost’ fractions can be treated on five of the six remaining weekend days, the rest of the treatment proceeding as prescribed. If treatment cannot be performed at weekends then, on five of the remaining treatment days, two fractions per day should be delivered, with a minimum 6h gap between those particular fractions.
Worked example A patient prescribed a schedule of 70 Gy in 35 fractions over 7 weeks misses the whole of the final 2 weeks’ treatment (10 fractions). How can this unscheduled gap be compensated for? Answer: This is a much more difficult case, as it is impossible to treat the ‘lost’ fractions without overrunning the original treatment time. If it is not possible to treat at weekends, then, in order to restrict the ‘over-run’ as much as possible, the only viable possibility is to treat twice daily on 5 days after the unscheduled break, taking the overall treatment time to 8 weeks (53 days). If the dose-equivalent of daily repopulation is 0.85 Gy per day and is operational after 25 days, the generic BEDs expected from the treatment as prescribed are: ⎡ 2⎤ BEDtumour 70 ⎢1 ⎥ 0.85 (46 25) ⎢⎣ 10 ⎥⎦ 66.2 Gy10 ⎡ 2⎤ BEDlate 70 ⎢1 ⎥ 116.7 Gy 3 ⎢⎣ 3 ⎥⎦ The extra dose required to offset the repopulation in the 1-week extension of treatment time is 7 0.85 5.95 Gy, and this is the amount by which the tumour BED will be reduced if no compensation is made, i.e. BED10 will be 60.2 Gy10 rather than the expected 66.2 Gy10. Increasing the dose per fraction can restore the BED10, but at the cost of a BED3 that is greater than that prescribed. Delivery of more than 10 fractions, with a view to decrease dose per fraction, is a possible solution, but this extends treatment into the ninth week, which will require more dose to offset yet more repopulation. This is the essential dilemma involved in difficult cases, and the inevitable result is that there must be some compromise to the therapeutic ratio. The prescribed late BED can only be respected at the cost of a reduced tumour BED; achievement of a tumour BED that is close to that expected from the uninterrupted treatment can only be attained with some increase in the late BED. Usually it is necessary to consider several possible scenarios in order to identify the most acceptable compromise. For example, if we allow the late BED to increase by around, say, 3.5 Gy3, then if we give the remainder of the treatment in accelerated form using 12 b.i.d. fractions over 6 days (assuming that treatment can be completed on a Saturday), we find the late BED given at the time of interruption 50(1 2/3) 83.3 Gy3. To deliver a late BED of 116.7 3.5 120.2 Gy3, the remaining BED is 120.2 83.3 36.9 Gy3, and to find the
64 Mathematical modelling and its application in oncology
requisite dose per fraction (d ) to deliver this, we solve for d in: ⎛ d⎞ 12d ⎜⎜1 ⎟⎟⎟ 36.9 , so that d 1.89 Gy. ⎜⎝ 3 ⎟⎠ The tumour BED for the entire course if completed in 54 days will then be 50(1 2/10) 1.89 12(1 1.89/10) 0.85(54 25) 62.3 Gy10. This is 66.2 62.3 3.9 Gy10 less than the original prescription. The alternative approach would be to calculate the full compensatory dose required to maintain equal tumour control but allow normal-tissue side effects to increase (the reader is left to calculate these). The patient may also have views on which is the best alternative. It must be remembered that normal-tissue dose–effect increment slopes are usually of the order of 1–3 per cent per Gy.
Worked example A treatment prescription for 60 Gy in 30 fractions includes the optic chiasm. Suppose there is an air cavity inhomogeneity that results in an 8 per cent increase in dose across a region of interest in the brain. What does this dose increase imply? For an α/β ratio of 2 Gy (applicable to the optic chiasm, a fractionation-sensitive structure), the prescribed BED is: ⎡ 2⎤ 60 ⎢1 ⎥ 120 Gy2 ⎢⎣ 2 ⎥⎦
Because of the 8 per cent increase, the fractional dose is 2 1.08 2.16 Gy, instead of the prescribed 2 Gy. Over 30 fractions, the total dose will be 30 2.16 64.8 Gy. The true BED is then: ⎡ 2.16 ⎤ ⎥ 134.8 Gy2 64.8 ⎢1 ⎢⎣ 2 ⎥⎦
Methods of presenting modelling results: the equivalent dose delivered in 2 Gy fractions Several methods exist for presenting the results of modelling assessments, e.g. comparison of BED values or their derived TCPs or normal-tissue complication probabilities (NTCPs). The results for routine clinical application can be summarized in a form that relates to the prescribing practice of the radiation oncologist. It is possible to stipulate any treatment in terms of a 2 Gy/fraction equivalent. This concept is easy for most clinicians: the calculated equivalent dose can be related to previous clinical experience of treating with 2 Gy fractions to form a judgement of both tumour control and normal-tissue complication risk.
Thus, although the dose increase is 8 per cent, the resultant increase in BED2 is 134.8 120 14.8 Gy2. What does this mean in clinical terms? This can be assessed by calculating how many 2 Gy fractions would give a similar increase in BED. To find the equivalent total dose given in n 2 Gy fractions, we solve for total dose D as if 2 Gy fractions had been given to the increased BED of 134.8 Gy2: ⎛ 2⎞ BED 134.78 D ⎜⎜1 ⎟⎟⎟ ⎜⎝ 2 ⎟⎠ from which D 67.4 Gy, the equivalent dose if given in 2-Gy fractions.
Brachytherapy For high-dose-rate brachytherapy, the LQ equations are essentially the same as those required for teletherapy. For low-dose-rate or medium-dose-rate continuous treatments, the delivery of a radiation dose is protracted such that it takes more than a few minutes, and thus there is an opportunity for sub-lethally damaged cells to repair during treatment delivery. The amount of lethal damage caused by interaction between sub-lethally damaged entities is thus reduced due to a smaller amount of Type B (βmediated) lethal damage than occurs in acute radiation delivery. For a given dose, the overall treatment time is determined by the dose rate, and the Type B damage calculation must include an allowance for this, together with the repair rate of sub-lethal damage. Type A damage, because it is caused only by instantaneous lethal events, is unaffected
by extension in irradiation time and is therefore independent of dose rate. A realistic radiobiological assessment of brachytherapy applications is always more difficult than in the case of teletherapy, principally because of the non-homogeneous doses and the rapid fall of dose with distance from the source(s).15 In particular it should be remembered that the critical normal tissue might receive a lower dose than that prescribed to the tumour, and this should be allowed for in the calculation of the normal-tissue BED. Also, the dose gradients within any tissue that contains sources causes the effective BED to that tissue volume to be higher than that calculated at the reference iso-dose surface. This point is returned to later in the chapter.
The basis and routine application of radiobiological models 65
be summed to obtain a measure of the total biological effect, i.e.:
BED calculations in brachytherapy For fractionated high-dose-rate (FHDR) brachytherapy, the BED is calculated using Equation 4.4 or 4.8, the latter being required in cases in which it is necessary to include an allowance for tumour repopulation. For continuous low-dose-rate (CLDR) brachytherapy, the required equation is: ⎡ 2 R ⎤⎥ BED RT ⎢1 ⎢ μ(α /β) ⎥⎦ ⎣
(4.12)
where R is the dose rate (Gy/hour), T (hours) the duration of brachytherapy, and μ the DNA sub-lethal damage repair rate constant, related to the repair half-life via: μ
0.693 T1/ 2
Equation 4.12 is a simplified form of a more complex equation,3 but is nevertheless reliable in most cases where the CLDR treatment time is greater than about 12 hours. For shorter exposures, the longer equation should be used. In cases in which brachytherapy is combined with teletherapy the individual BEDs for each component may
Worked example A CLDR gynaecological treatment delivers 40 Gy in 48 hours to Point A. If an FHDR treatment involving six fractions is to be used instead, what total dose is required in order to maintain the normal tissue iso-effect? For α/β of 3 Gy and repair half-life of 1.5 hours, i.e. μ 0.693/1.5 0.46/hour. The dose rate (R) at Point A is 40/48 0.83 Gy/h. Therefore, from Equation 4.11:
BED (total) BED (external beam) BED (brachytherapy) (4.13)
Worked example A pelvic teletherapy regime of 45 Gy/25 fractions is combined with an FHDR brachytherapy regime of 2 7 Gy prescribed to Point A. The bladder and rectum each receive 100 per cent of both the teletherapy dose and the brachytherapy dose. What is the BED for the bladder/rectum, assuming an α/β value of 3 Gy? ⎡ ⎡ 1.8 ⎤ 7⎤ ⎥ 14 ⎢1 ⎥ 118.7 Gy 3 BED 45 ⎢1 ⎢⎣ ⎥ ⎢ 3 ⎦ 3 ⎥⎦ ⎣ If geometrical sparing of the bladder/rectum during brachytherapy achieves 80 per cent of the prescribed Point A dose, what is the new BED in these tissues? ⎡ ⎡ 1.8 ⎤ 7 0.8 ⎤ ⎥ 14 0.8 ⎢1 ⎥ BED 45 ⎢1 ⎢⎣ ⎥ ⎢ 3 ⎦ 3 ⎥⎦ ⎣ 104 .1 Gy 3 In general, the severe side effects associated with pelvic radiotherapy manifest after BED values that significantly exceed 110 Gy3. This example therefore demonstrates how, in principle, a modest reduction of the brachytherapy dose and dose per fraction to the critical organ can bring about a reduction in the likelihood of serious toxicity.
⎡ 2 0.83 ⎤ ⎥ 88.1 Gy 3 BED3 40 ⎢1 ⎢⎣ 0.46 3 ⎥⎦ To match this BED value with six FHDR fractions requires a dose per fraction of d Gy, where: ⎡ d⎤ 6 d ⎢1 ⎥ 88.1 ⎢⎣ 3 ⎥⎦ i.e.: 2d2 6d 264.3 0 i.e. the required dose per fraction (d) is 5.3 Gy.
Amelioration of an incorrectly delivered radiotherapy treatment It is sometimes said that no amount of radiobiological manipulation will correct a treatment that is incorrectly prescribed or executed. Whilst this is correct in cases involving a geometrical miss or significant over/under-dosing, there are occasions when changes in time–dose-fractionation may significantly ameliorate the potential damage, although it may be necessary to accept a reduction in control probability if total dosage has to be reduced. The problem is essentially one of obtaining the best benefit: risk ratio in a particular set of circumstances.
66 Mathematical modelling and its application in oncology
Worked example During the first four fractions of a treatment, a 30 per cent overdose to the spinal cord occurred due to the use of an incorrect wedge filter. The prescribed cord dose was 46 Gy in 23 fractions. How can the overdose be corrected over the remaining treatment course if it is decided to complete the treatment in a further 20 fractions rather than 19? The spinal cord α/β is assumed to be 2 Gy. The intended spinal cord BED was to be: ⎡ 2⎤ 46 ⎢1 ⎥ 92 Gy 2 ⎢⎣ 2 ⎥⎦ In the first four fractions, the BED actually delivered is: ⎡ 2 1.3 ⎤ ⎥ 23.9Gy2 4 2 1.3 ⎢1 ⎢⎣ 2 ⎥⎦ The remainder of the treatment course should give a BED2 and not exceed 92.0 23.9 68.1 Gy2. For a further 20 fractions, allowing one extra fraction to be given beyond that originally prescribed, the required dose per fraction (d ) is given by the solution of d in: ⎡ d⎤ 20 d ⎢1 ⎥ 68.1 ⎢⎣ 2 ⎥⎦ from which d 1.79 Gy. It is therefore decided to give 20 fractions of 1.79 Gy. If the tumour receives the same dose as the spinal cord and possesses an α/β value of 15 Gy, how is the tumour BED affected by the change to the schedule? The prescribed tumour BED was: ⎡ 2⎤ 46 ⎢1 ⎥ 52.1 Gy15 ⎢⎣ 15 ⎥⎦
The addition of conventional external-beam and focal methods Worked example A patient is treated to a tolerance dose equivalent to 70 Gy in 35 fractions to a small volume by: 1. initial wide-field radiotherapy to 40 Gy in 22 fractions followed by 2. reduced field sizes to a further 16 Gy in eight fractions and then 3. focal radiotherapy given in four fractions. Calculate the dose per fraction of the focal therapy (assume α/β 2 Gy). The total BED will be: ⎡ 2⎤ 70 ⎢1 ⎥ 140 Gy2 ⎢⎣ 2 ⎥⎦ We know that the following condition must hold: BEDphase1 BEDphase2 BEDphase3 140 Gy2 So that: i.e.: ⎡ ⎡ ⎡ 1 .8 ⎤ 2⎤ d⎤ ⎥ 16 ⎢1 ⎥ 4 d ⎢1 ⎥ 140 40 ⎢1 ⎢⎣ ⎥ ⎢ ⎥ ⎢ 2 ⎦ 2⎦ 2 ⎥⎦ ⎣ ⎣ d2 2d 15.8 76 32 4d 2d2 140 d 3.1 Gy. The above equations and examples are sufficient for most practical clinical problems.
whereas the true value will be: ⎡ ⎡ 2 1 .3 ⎤ 1 .79 ⎤ ⎥ 20 1 .79 ⎢1 ⎥ 4 2 1 .3 ⎢1 ⎢⎣ ⎢⎣ 15 ⎥⎦ 15 ⎥⎦ 52.3 Gy15
which is scarcely different from that prescribed. However, an extra fraction would have involved an extra treatment day, with possible loss of tumour control, depending on the operative tumour repopulation dose equivalent per day. Thus, the treatment error has been effectively overcome by the altered fractionation and use of a lower – more forgiving – dose per fraction.
MORE ADVANCED CONSIDERATIONS The discussion and examples have so far focused on what might be called general modelling: the parameters used are intrinsically assumed to apply throughout individual tumour volumes and throughout whole patient populations. For example, use of the radiosensitivity factor α assumes a uniform radiosensitivity throughout the tumour, i.e. in hypoxic and oxic zones, and is also inclusive of clones that show variation in intrinsic radiosensitivity. Likewise, the Teff value is generalized, being a one-figure representation of a more complex pattern of repopulation
More advanced considerations 67
throughout very heterogeneous tumours. Even the ubiquitous α/β ratios are averages, which probably incorporate the effects of multiple radiobiological processes such as reoxygenation (in the case of tumours), cell-cycle redistribution and repopulation. Thus, although the LQ model has a sound individual biophysical basis, the variations inherent in the critical parameters place a limit on how accurate modelling can be in individual cases. The alternative approach is to attempt to model the processes in greater detail whilst accepting the associated drawback that multiple parameter assumptions are necessary. Some more advanced approaches are briefly discussed below.
If the number (C) of clonogenic cells is known, the Poisson model may be used to estimate TCP: (4.14)
The slope of the dose response generated by the Poisson distribution (i.e. TCP plotted against BED or total dose) is much steeper than found in data sets obtained from large numbers of patients. This is to be expected because, as the Poisson distribution reflects the probability of achieving a cure from the elimination of all cells in a single tumour, it is essentially useful only for individual tumours. The alternative logistic function is applicable to population data and has a less steep dose–response curve.
⎡ 2⎤ BED 60 ⎢1 ⎥ 0.5 39 52.5 Gy10 ⎢⎣ 10 ⎥⎦
TCP exp[108 exp(0 .35 52 .5)] 0 .35(35 %)
If the tumour is treated without weekend breaks, i.e. is treated with 2 Gy fractions on 7 days of each week, the overall time falls to 29 days. The BED then increases to: ⎡ 2⎤ BED 60 ⎢1 ⎥ 0.5 29 57.5 Gy10 ⎢⎣ 10 ⎥⎦ For this BED, the associated TCP is:
TCP exp[108 exp(0 .35 57 .5)] 0 .83 (83 %)
Optimization Clinically effective radiotherapy requires a balance to be maintained between several potentially conflicting requirements. For example, use of small fractional doses is more likely to minimize NTCP, but this may require the use of so many fractions that treatment time is greatly extended, thereby allowing more tumour repopulation, and thus a reduced TCP. As the LQ model includes these effects, the model can be extended to identify schedules that are optimized for a given set of circumstances. By consideration of how tumour cell kill varies with changes in dose per fraction, overall time and the mean inter-fraction interval (f ), calculus may be used to derive an optimum dose per fraction that maximizes the TCP whilst respecting any given normal-tissue effect.15–17 Essentially, the optimum tumour dose per fraction (z) is given by the solution for z of: ⎞ ⎛ ⎜⎜ g LATE (α /β) ⎟⎟ ⋅ z 2 2 ⋅ f ⋅ g ⋅ k ⋅ z ⎟⎟ ⎜⎜ ⎝ TUM (α /β) ⎠ LATE (α /β) ⋅ f ⋅ k 0
A particular tumour has a dose equivalent of repopulation of 0.5 Gy/day and consists of 108 clonogens. It is treated with 60 Gy/30 fractions over 6 weeks (39 days). What is the expected TCP? (Assume α/β 10 Gy.) The repopulation dose equivalent is subtracted from the given BED:
In order to calculate TCP from Equation 4.12 it is necessary to assume a value for α. If this is taken to be 0.35/Gy:
Estimation of tumour cure probabilities from BEDs
TCP exp[Cexp(αBED)]
Worked example
(4.15)
This example clearly demonstrates how overall treatment time can have a strong influence on TCP in individual patients.
where g is the normal-tissue sparing factor such that d gz, where d is the dose per fraction actually received by the normal tissue and K is the tumour repopulation factor. The suffixes LATE and TUM apply to the α/β ratios of late-reacting normal tissues and tumour respectively. The term f is the mean inter-fraction interval (days), which is approximated by seven days divided by the number of treatments given per week. Where there is a delayed onset to the initiation of tumour repopulation during treatment, it may be necessary to use non-analytical, reiterative techniques to determine the optimum fraction size.18 In conjunction with accurate predictive assays, or with better information as to the averaged values of the key parameters during radiotherapy for specific cancers,
68 Mathematical modelling and its application in oncology
Estimate the optimum fraction size for treating a tumour with average repopulation rate during treatment of 0.5 Gy/day, assuming treatment is to be given five times per week and the critical normal tissue receives 80 per cent of the full dose. Assume α/β 10 Gy (tumour) and 3 Gy (normal tissue). In this case, f 7/5 1.4 and g 0.8. Using Equation 4.15, optimum dose per fraction ( 0.8 3/10)z 2 2 1.4 0.8 0.5 z 3 1.4 0.5. The solution for z is 3.46 Gy per fraction. This is the fractional dose to be prescribed to the tumour; the normal tissue would receive 3.46 0.8 2.77 Gy per fraction. In this case the model is predicting the use of a relatively large fraction size and this is therefore an example of where the decision over whether to use that or a reduced value needs to be additionally guided by clinical experience.
the developments described could be an important step towards ‘tailored’ radiotherapy.
Incorporation of cell-loss factors An alternative explanation for the apparent appearance of delayed acceleration is that tumour repopulation rates are initially slow but may progressively increase during the course of treatment. When analysing the effect in a range of data sets, such a phenomenon could easily be mistaken for a two-component process (Fig. 4.1). An explanation for an asymptotically increasing repopulation rate may be developed from consideration of the role of the potential doubling time (Tpot) and the cell-loss factor θ, which is the probability that newly born cells will die due to causes such as hypoxia and insufficiency of nutrients and growth factors. The cell-loss factor is likely to decrease during radiotherapy due to improved tumour microvascular perfusion. The probability that newborn cells will survive can be represented by 1 θ, and this term can be used for BED assessments when there is a slow exponential reduction in θ (at a rate constant v) due to improved blood perfusion during treatment.19 Relatively rapid changes in cell-loss factor are then effectively followed by a delayed elevation in repopulation rate. The formulation is applicable at any time duration, including ultra-short schedules.
The problem with closely spaced fractions The delivery of an acute dose of radiation results in the creation of both lethal and sublethal damage. By definition, the lethal damage is unrepairable, but the sub-lethal damage component is able to repair pseudo-exponentially with
Clonogen repopulation rate
Worked example
Treatment duration
Figure 4.1 A schematic representation demonstrating how a continuously increasing repopulation process (solid line) can be mistakenly interpreted as a two-component linear process with accelerated repopulation beginning part-way through treatment (dashed line).
a repair half-time typically in the range 0.5–3 hours. When the fractional doses are well spaced (e.g. as 24-hour gaps), the residual sublethal damage after any particular fraction is effectively fully repaired before delivery of the next fraction. If the fractions are delivered at much closer intervals, any unrepaired sublethal damage remaining after any one fraction can be compounded to extra lethal damage by fractions delivered later in the schedule. The amount of extra damage created depends on several factors, but the inter-fraction intervals and the repair rates are of prime importance and, within the LQ formulation, the extra damage due to closely spaced fractions appears in the form of a modified β term in Equation 4.2. Any increased damage due to this effect is obviously beneficial in a tumour, but will be of major concern if it occurs in a critical normal tissue. The equations for dealing with closely spaced fractions are complex, but, using published tables,1,5 a simpler alternative may be used. The tables provide tabulated values of an incomplete-repair factor, hM, which allows for different numbers of treatments per day, repair half-times and inter-fraction intervals. When hM is incorporated into the basic BED equation, the latter is modified to: ⎡ d(1 hM ) ⎤ ⎥ BED nd ⎢1 ⎢ α /β ⎥⎦ ⎣
(4.16)
More advanced considerations 69
Worked example A dose of 60 Gy is to be delivered in 30 fractions (2 Gy per fraction) over 10 consecutive days by treating three fractions per day at 4-hour intervals. Assuming T1/2 1.5 hours and α/β 3 Gy, what is the lateresponding BED? From Table 6.3 of Thames and Hendry (1987),1 hM 0.2265. Therefore, from Equation 4.16: ⎡ 2 (1 0.2265) ⎤ ⎥ 109.1 Gy 3 BED 30 2 ⎢1 ⎢⎣ ⎥⎦ 3 The BED3 for 60 Gy delivered as daily fractionation is 100 Gy3. Therefore three fractions per day increases this figure by approximately 9 per cent.
New repair models Most radiobiological models assume that sublethal damage repair is a mono-exponential or multi-exponential process, the latter being required when there is the apparent presence of separate fast and slow repair rates. Fowler (1999) suggested a different description of repair.20 In-vitro DNA repair data are consistently better fitted by assuming that the fraction of unrepaired breaks decreases in proportion to the reciprocal of the time elapsed (t) since their production, i.e. proportional to 1/(1 zt), where z is an appropriate constant. This postulation of a ‘reciprocal-time’ pattern of repair is consistent with the view that there is total saturation of repair enzymes at clinically relevant doses. The practical advantage is that the observed presence of separate components of fast and slow repair in some cell lines can be covered by an equation involving only one rate constant (z), rather than several exponential rate constants. The reciprocal-time model may be incorporated into the LQ model,21 where it predicts the breakdown in tissue tolerance when the central nervous system is irradiated by three fractions per day. This is in contrast with monoexponential models of repair, which cannot satisfactorily explain the clinical problems observed in the early continuous hyper-fractionated and accelerated radiotherapy CHART trials. This alternative model of repair predicts the following: ●
●
Repair in normal tissues may remain incomplete at much longer intervals of time, such as at 18–24 hours, than had previously been thought. It is only over a weekend ‘gap’ that late-reacting normaltissue repair may be complete.
●
For long exposures, as occur in low-dose-rate brachytherapy, the apparent rate of repair after irradiation appears to be slower for longer irradiation times.
Limitation in the scope of BED inter-comparisons The earlier worked examples demonstrate how the BED parameter may be used to inter-compare existing treatments or to design a new treatment that is iso-effective to an existing treatment. It is nevertheless important to appreciate that, as conventionally used, BED is a measure of radiobiological effect at one particular point within the treated volume. BED inter-comparisons thus lose much of their usefulness if they are used to compare treatments in which the geometrical details (e.g. field shapes, number of applied fields, irradiated volumes, etc.) are not held constant. Indeed, the application of LQ methodology in cases in which these conditions are not met may produce some very misleading results. This point also needs to be borne in mind when using BED calculations retrospectively to assess the results of clinical trials for which the various treatment schedules do not share common geometrical field arrangements, or where there are significant dose inhomogeneities. In such cases, the concept of a radiobiologically derived equivalent uniform dose (EUD) is of major interest.22
The particular problem caused by brachytherapy dose gradients For brachytherapy, the above observations are all the more important, principally because of the non-homogeneous doses and the rapid fall of dose with distance from the source(s).15 The full radiobiological significance of brachytherapy dose gradients is discussed elsewhere,23 from which some general guidelines have been established. In cases in which there is a continuous dose gradient from the reference (or prescription) surface (at which the dose is prescribed) to the sources, as occurs in most gynaecological treatments and line-source applications, the effective BED throughout the volume enclosed by that surface can be derived using tabulated multiplying factors (MFs). The BED at the surface is first calculated using the simple equations outlined above and then corrected by an MF appropriate to the particular treatment set-up. In the case of an FHDR brachytherapy application, the MF value will be mostly dependent on the number of fractions and the dose per fraction, but only marginally so on the radiobiological factors (α, α/β, etc.). The treatment geometry (e.g. line source, point source) also has little influence. For CLDR applications, the MFs are mostly influenced by total dose and dose rate.
70 Mathematical modelling and its application in oncology
Worked example An FHDR gynaecological treatment involves delivering 6 5 Gy to Point A. What is the effective BED to the tissues within the iso-dose surface that contains Point A? First, the BED3 is calculated in the usual manner: ⎡ 5⎤ BED3 6 5 ⎢1 ⎥ 80 Gy 3 ⎢⎣ 3 ⎥⎦
Inter-comparison of treatment alternatives on the basis of their net costs
From the table look-up data given by Dale et al. (1997) (21a), the appropriate MF value for this particular combination of dose per fraction and fraction number is 1.185. Therefore the true effective BED throughout the enclosed volume is 80 1.185 94.8 Gy3.
Tumour volume changes during radiotherapy Tumour regression generally follows a negative exponential function. In such cases, if V0 and Vt are the tumour volumes at times 0 and t respectively: Vt V0 ezt
treatment volume are often modelled empirically using clinical dose–response data of limited usefulness. The result is that most of the existing NTCP models have little biological basis; thus the use of simpler BED iso-effect calculations backed by clinical judgement remains reasonable until better predictive models are found.
(4.17)
where z is the regression rate coefficient. Consequently, if the regression rate is accurately measured, one can predict the volume at a future time point. This information may be used to plan additional treatment, such as conformal therapy, where the reduced volume allows a higher radiation dose to be delivered.
Worked example Serial imaging during the first part of a course of radiotherapy shows that a tumour, originally of volume 95 cm3, has a volume regression rate of 3 per cent per day. Estimate the volume of tumour 28 days after completion of a 35-day course of treatment. From Equation 4.17: Vt 95 ⋅ e 0.03 63 14.4 cm3
The volume effect and difficulties in relation to normal tissues Whilst it is clinically well established that large volumes of normal tissue will tolerate less radiation than small volumes (the so-called volume effect), accurate modelling of the effect is a difficult and controversial issue. The volume effect is strongly influenced by variations in tissue physiology and anatomical location, yet attempts at formulating the phenomenon tend to concentrate almost exclusively on physical and radiobiological considerations. A further difficulty is that the changes in radiation tolerance with
The net costs of treatment can be estimated by inclusion of the fixed costs (treatment planning, treatment delivery, etc.) together with the notional costs associated with treatment failure. The former costs can take account of the treatment complexity, whilst the latter are based on the cost of additional medical/surgical care in the event of failure to control the primary tumour. For a given set of radiobiological parameters, cost-optimized treatments will often differ from radiobiologically optimized treatments, this being especially true of those more complex treatments that offer relatively small gains in TCP. Such modelling may have a useful role to play in identifying the situations in which the more complicated (and hence more costly) methods of treatment can be most effectively employed.24
Radiotherapy combined with chemotherapy, surgery and other molecular-based treatments For concomitant chemo-radiotherapy or the use of chemotherapy in close temporal approximation to radiotherapy, the equivalent BED of the chemotherapy can be estimated from some clinical data sets, particularly from randomized control trials.25 For normal-tissue sensitization, similar concepts apply.26 The distinction between cytotoxic and cytostatic adjuvant therapies can also be accommodated in calculations that estimate the changes in TCP by the use of adjuvant therapies.27 Pure dose sensitization of the radiotherapy is another method, whereby the dose per fraction is enhanced by factor s, which is 1.06 for the effect of the widely used CMF (cyclophosphamide, methotrexate and 5-fluorouracil) chemotherapy regime, in causing subcutaneous fibrosis in breast cancer and is equivalent to a BED3 value of approximately 6.5 Gy3 extra dose; the effect of age in increasing shoulder fibrosis after breast and gland field radiotherapy is approximately 3.6 Gy3, and when laparotomy is used in intra-abdominal lymphoma, the subsequent bowel radiation tolerance is reduced by approximately 17.7 Gy3. There are also several literature discussions on how to model gene and other novel molecular-based therapies in association with radiotherapy.27–32
Palliative radiotherapy In principle, radiation-induced growth delay is amenable to BED quantification, providing the α/β and repopulation
More advanced considerations 71
dose equivalents (K) are known for a particular tumour type. Interestingly, the average K values can also be derived from re-growth delay times. For example, if a slow-growing adenocarcinoma (α/β 4 Gy assumed) has recurred to its previous dimensions with recurrence of symptoms following a remission period of 300 days after 20 Gy in five fractions was given, for restoration of the original tumour volume, the given BED should equate with the total BED equivalent of repopulation over the observed re-growth time (TR), i.e. the average K value is determined from: ⎛ 4⎞ 20 ⎜⎜1 ⎟⎟⎟ KTR ⎜⎝ 4 ⎟⎠ i.e. K
40 0.13 Gy 4 per day. 300
⎡ RBE 2 ⎢ MIN .d BED D ⎢ RBE MAX α /β ⎢ ⎣⎢
Serum tumour markers The serum marker concentration is assumed to reflect the total number of cells present, and although marker levels fall during or after therapy, the absolute levels during a period of such change do not reflect the absolute number of cells. If the half-life (T1/2) of the marker is known, the tumour marker production rate (TMP), which will better represent the absolute number of cells, can be calculated as in the equation of Price (1989):33 TMP C2 C1 ⋅ e
0.693(t 2 t1 ) T1/ 2
(4.18)
where C1 and C2 are the marker concentrations at times t1 and t2 respectively.
Worked example Consider a pre-treatment situation in which the marker is slowly rising, so that in 14 days the level changes from 95 units to 100 units, and T1/2 is 12 hours (0.5 days). From Equation 4.18, the average TMP, expressed as units per day, in this period is: TMP 100 95 ⋅ e
0.693(14 ) 0.5
100 0 100 day1
Now, if in a 2-day interval near to the completion of radiotherapy the levels fall from 75 to 60 units, the new marker production rate falls to: TMP 60 75 ⋅ e
0.693(2) 0.5
particles or ions (e.g. protons and carbon ions). The increased biological effects of high-LET radiations are found mainly at the Bragg peaks and have usually been quantified as a relative biological effect (RBE), where, RBE doseLOW/doseHIGH, where LOW and HIGH refer to the radiation LET quality. Experimental studies in vitro and in vivo have confirmed that there is an inverse relationship between RBE and dose per fraction in the case of radiations with high LET, although the effect is difficult to find in proton beam experiments since the RBE is relatively small, being of the order of 1.1. For neutrons, and ions, larger RBEs are operative and here the BED equations can be modified to account for the very high RBEs of around 8 (the RBEMAX) at very low dose to values of 1–1.3 at very high doses per fraction (the RBEMIN). It can be shown that:34
60 4.69 55 .3 day1
High linear energy transfer radiations There is increasing clinical interest in high linear energy transfer (LET) radiations delivered by high-energy atomic
⎤ ⎥ ⎥ ⎥ ⎥⎦
(4.19)
It must be noted that the α/β used is that of the low LET, and the RBE factors essentially convert this to the fraction sensitivity of the high-LET radiation.35 The use of this equation allows the overall RBE, when high-LET and low-LET radiations are inter-compared, to vary between its maximum and minimum values. This then avoids underestimation of dose at low dose per fraction and overestimation of dose at high dose per fraction in comparison to the use of a fixed RBE, which is the present international method. A comparison of the two methods is shown in Figure 4.2.
Worked example A carbon-ion beam delivers a physical dose of 12 Gy to a small squamous-cell cancer. What would be the true equivalent dose using X-rays (RBE 1) assuming that, for the carbon ions, RBEMAX 8 and RBEMIN 1.2? From Equation 4.19, this physical dose would deliver a tumour BED of: 12
8 (1.22 12) 116.7 Gy10 10
The equivalent single dose (d) of X-rays required to provide the same BED is given as the solution for d in: d
1 d 116.7 10
the solution for which is d 29.53 Gy.
It can also be shown that low-LET repopulation factors can be used with the above equation.36 Essentially, the dosecorrection factor for compensation for missed treatments for high-LET radiation is close to K (the low-LET dose equivalent repopulation factor) divided by RBEmax.
72 Mathematical modelling and its application in oncology
RBEmax8, RBEmin1.25, Generic RBE3
Relative biological effect
10 Variable RBE
8
Generic RBE
6 4 2
2.5
5
7.5
10
12.5
15
17.5
20
High LET physical dose per #(Gy)
Figure 4.2 Relationship between RBD and dose per fraction of a high LET radiation using Eq. (18). It can be seen that use of a fixed (generic) RBE value of 3 will underestimate the RBE at low dose per fraction and overestimate the RBE at high dose per fraction.
Relative biological effects may also be highly significant in the biological dosimetry of radionuclides used for permanent implant brachytherapy and for targeted radiotherapy.37,38 In such cases, the enhancing effect of using a high-LET radionuclide needs to be assessed alongside a number of other significant physical and biological considerations (radionuclide half-life, biological half-life, shrinkage rates, repopulation rates, etc.), and the associated methodology may become complex.39–42
Commercial software packages for radiobiological bio-effect planning Despite the advances in radiobiological understanding, treatment planning in radiotherapy continues to be based on the computation and assessment of physical dose distributions. Where radiobiological features are available on commercial systems, two approaches have been adopted. The first is to transform physical dose distribution to a BED distribution, a method that, in principle, allows the biological effects of various radiation treatments, e.g. external-beam therapy and brachytherapy, to be graphically added. The second approach is to produce estimates of TCP and NTCP in specific volumes of interest. The TCP calculations are based on clonogenic cell survival and may take into account non-uniform dose delivery. The NTCP calculations are based on models of normaltissue functional organization in which the sub-units are not regarded as independent for response purposes. Despite the growing acceptance of the potential importance of bio-effect planning, progress in this area remains slow because of difficulties in identifying user requirements, limitations of the models (particularly in relation to normal-tissue biology) and concerns relating to the clinical interpretation of results.
CONCLUSION Radiobiological models can provide a useful complement to clinical judgement in a variety of circumstances. In the
routine practice of radiotherapy there are many situations (e.g. geometrical difficulties, dosimetric errors, treatment interruptions, etc.) in which modelling can allow compensatory treatment schedules to be estimated. With the necessary caveats already mentioned, modelling can often guide or inform clinical judgement by means of ranking a series of alternative treatments in order to reject the less satisfactory options. For radiation oncologists in training, familiarity with the simpler methods of quantifying radiobiological effect should not be underestimated, because even the simplest radiobiological estimates can provide a useful assessment of disparate treatments. In research and development, quantitative appraisal of proposed new techniques or clinical trials should always be the subject of a modelling review. Similarly, in order to allow quantitative comparisons between new and existing treatment schedules, all published works should record at least the generic BEDs associated with the reported treatments. Technological developments presently abound in radiation oncology. However, whilst some technological advances (e.g. the ability to reduce irradiated normal-tissue volumes through the use of conformal techniques) provide seemingly obvious benefits, the full potential of modern radiotherapy will be realized only if commensurate attention is given to the radiobiological aspects that determine treatment outcome. Radiation oncology thus faces an interesting future, with radiobiological modelling providing more pivotal guidance on how to make optimal use of the emerging technology.
ACKNOWLEDGEMENT The authors wish to thank Elena Antoniou for her excellent help in updating this manuscript.
KEY LEARNING POINTS ●
●
●
●
●
●
Biological effect is not related solely to the total physical dose delivered in a radiotherapy treatment – factors such as dose per fraction and dose rate are also fundamentally important. Modern radiobiological models take account of a range of practical and radiobiological factors that are known to influence treatment outcome. Simple radiobiological assessments can provide a useful supplement to clinical judgement in a variety of ways. More advanced modelling ideas, when coupled with the results of reliable predictive assays, will allow more appropriate use of the emerging technologies. Radiobiological models already have an important role to play in the teaching and understanding of clinical radiotherapy. In order to facilitate easier comparison between clinical studies, reports and publications should routinely record the BEDs.
References 73
REFERENCES ◆1
●2
●3
●4
◆5
6
◆7
◆8
●9
●10
●11
✸12
✷13
✷14
●15
●16
17
18
Thames HD, Hendry JH. Fractionation in Radiotherapy. London: Taylor and Francis,1987. Barendsen GW. Dose fractionation, dose rate and iso-effect relationships for normal tissue responses. Int J Radiat Oncol Biol Phys 1982; 8:1981–97. Dale RG. The application of the linear quadratic dose–effect equation to fractionated and protracted radiotherapy. Br J Radiol 1985; 58:515–28. Thames HD. An ‘incomplete-repair’ model for survival after fractionated and continuous irradiations Int J Radiat Biol 1985; 47:319–39. Joiner MJ, Van der Kogel AJ. The linear-quadratic approach to fractionation and calculation of isoeffect relationships. In: Steel GG (ed.) Basic Clinical Radiobiology. London: Arnold, 1997, 106–22. Fowler JF. The radiobiology of prostate cancer including new aspects of fractionated radiotherapy. Acta Oncol 2005; 44:265–76. Fowler JF. The linear quadratic formula and progress in fractionated radiotherapy. Br J Radiol 1989; 62: 679–94. Jones B, Dale RG, Deehan C, Hopkins KI, Morgan DAL. The role of biologically effective dose (BED) in clinical oncology. Clin Oncol 2001; 13:71–81. Withers HR, Taylor JMG, Maciejewski B. The hazard of accelerated tumour clonogen repopulation during radiotherapy. Acta Oncol 1988; 27:131–46. Wyatt RM, Beddoe AH, Dale RG. The effect of delays in radiotherapy treatment on tumour control. Phys Med Biol 2003; 48:139–55. Fowler JF, Lindstrom MJ. Loss of local control with prolongation in radiotherapy. Int J Radiat Oncol Biol Phys 1992; 23:457–67. Guidelines for the Management of the Unscheduled Interruption or Prolongation of a Radical Course of Radiotherapy. Document BFCO(96)4. London: Royal College of Radiologists, 1996. Guidelines for the Management of the Unscheduled Interruption or Prolongation of a Radical Course of Radiotherapy. Document BFCO(02)5. London: Royal College of Radiologists, 2002. Dale RG, Hendry JH, Jones B, Deehan C, Sinclair J, Robertson G. Practical methods for compensating for missed treatment days in radiotherapy, with particular reference to head and neck schedules. Clin Oncol 2002; 14: 382–93. Dale RG, Jones B. The radiobiology of brachytherapy (Review). Br J Radiol 1998; 71:465–83. Jones B, Tan LT, Dale RG. Derivation of the optimum dose per fraction from the linear quadratic model. Br J Radiol 1995; 68:894–902. Jones B, Dale RG. Estimation of optimum dose per fraction for high-LET radiations: implications for proton radiotherapy. Int J Radiat Oncol Biol Phys 2000; 48:1549–57. Armpilia CI, Dale RG, Jones B. Determination of the optimum dose per fraction in fractionated radiotherapy
when there is delayed onset of tumour repopulation during treatment. Br J Radiol 2004; 77:765–7. 19 Jones B, Dale RG. Cell loss factors and the linear quadratic model. Radiother Oncol 1995; 37:136–9. ●20 Fowler JF. Is repair of DNA strand break damage from ionizing radiation second-order rather than first-order? A simpler explanation of apparently multiexponential repair. Radiat Res 1999; 152:124–36. 21 Dale RG, Fowler JF, Jones B. A new incomplete-repair model based on a ‘reciprocal-time’ pattern of sub-lethal damage repair. Acta Oncol 1999; 38:919–29. 21a Dale RG, Coles IP, Deehan C, O`Donoghue JA. Calculation of integrated biological response in brachytherapy. Int J Radiat Oncol Biol & Physics, 1997; 38, 633–42 ●22 Niemierko A. Reporting and analysing dose distributions: a concept of equivalent uniform dose. Med Phys 1997; 24:103–10. ●23 Dale RG, Coles IP, Deehan C, O’Donoghue JA. Calculation of integrated biological response in brachytherapy. Int J Radiat Oncol Biol Phys 1997; 38:633–42. 24 Jones B, Dale RG. Radiobiologically based assessments of the net costs of fractionated focal radiotherapy. Int J Radiat Oncol Biol Phys 1998; 41:1139–48. ●25 Jones B, Dale RG. The potential for mathematical modelling in the assessment of the radiation dose equivalent of cytotoxic chemotherapy given concomitantly with radiotherapy. Br J Radiol 2005; 78:939–44. 26 Jones B, Gaya A, Dale RG. Linear quadratic modelling of increased late normal tissue effects in special clinical situations. Int J Radiat Oncol Biol Phys 2006; (in press). ●27 Jones B, Dale RG. Inclusion of molecular biotherapies with radical radiotherapy: modelling of combined modality treatment schedules. Int J Radiat Oncol Biol Phys 1999; 45:1025–34. 28 Tucker SL, Geara FB, Peters LJ, et al. How much could the radiotherapy dose be altered for individual patients based on a predictive assay of normal tissue radiosensitivity? Radiother Oncol 1996; 38:103–14. 29 Wheldon TE, Mairs RJ, Rampling RP, Barrett A. Modelling the enhancement of fractionated radiotherapy by gene transfer to sensitize tumour cells to radiation. Radiother Oncol 1998; 48:5–13. ●30 Jones B, Dale RG. Mathematical models of tumour and normal tissue response. Acta Oncol 1999; 38:883–93. 31 Keall PJ, Lammering G, Lin PS, et al. Tumor control probability predictions for genetic radiotherapy. Int J Radiat Oncol Biol Phys 2003; 57:255–63. 32 Jones B, Cominos M, Dale RG. Application of Biological Effective Dose (BED) to estimate the duration of symptomatic relief and repopulation dose equivalent in palliative radiotherapy and chemotherapy. Int J Radiat Oncol Biol Phys 2003; 55:736–42. 33 Price PM. Individualisation of cancer treatment by predicting response to therapy. MD Thesis, University of Cambridge, 1989. 34 Carabe-Fernandez A, Dale RG, Jones B. The incorporation of the concept of minimum RBE (RBEmin) into the linearquadratic model and the potential for improved radiobiological analysis of high-LET treatments. Int J Radiat Oncol Biol Phys 2007; 83:27–39.
74 Mathematical modelling and its application in oncology
●35
36
37
38
●39
Dale RG, Jones B. The assessment of RBE effects using the concept of biologically effective dose. Int J Radiat Oncol Biol Phys 1999; 68:894–902. Jones B, Carabe-Fernandez A, Dale RG. Calculation of highLET radiotherapy dose required for compensation of overall treatment time extensions. Br J Radiol 2006; 79:245–57. Ling CC, Li WX, Anderson LL. The relative biological effectiveness of 125I and 103Pd brachytherapy sources. Int J Radiat Oncol Biol Phys 1995; 32:373–8. Dale RG. Use of the linear-quadratic radiobiological model for quantifying kidney response in targeted radiotherapy. Cancer Ther Radiopharm 2004; 19:363–70. Dale RG. Radiobiological assessment of permanent implants using tumour repopulation factors in the linear-quadratic model. Br J Radiol 1989; 62:241–4.
40 Antipas V, Dale RG, Coles IP. A theoretical investigation into the role of tumour radiosensitivity, clonogen repopulation, tumour shrinkage and radionuclide RBE in permanent brachytherapy implants of 125I and 103Pd. Phys Med Biol 2001; 46:2557–69. ●41 Armpilia CI, Dale RG, Coles IP, Jones B, Antipas V. The determination of radiobiologically optimized half-lives for radionuclides used in permanent brachytherapy implants. Int J Radiat Oncol Biol Phys 2003; 55:378–85. 42 Abou-Jaoud W, Dale RG. A theoretical radiobiological assessment of the influence of radionuclide half-life on tumour response in targeted radiotherapy when a constant kidney toxicity is maintained. Cancer Ther Radiopharm 2004; 19:308–21.
5 Principles of chemotherapy and drug development JEFFRY EVANS, MEENALI M. CHITNIS AND DENIS C. TALBOT
Basic principles Principles of treatment Drug development Classification of cytotoxic drugs, mode of action, toxicities and clinical utility
75 77 79
Drug resistance Novel anti-cancer therapies References
90 97 105
81
BASIC PRINCIPLES Chemotherapy drugs are developed for their potential to cause a greater proportion of cell death among neoplastic as opposed to normal cells. Differences exist between normal and malignant cells that result in the latter being more susceptible to anti-cancer drugs by virtue of their biological and proliferation characteristics.
Cancer cell kinetics CHARACTERISTICS OF THE TUMOUR CELL
The proliferation of tumour cells is not entirely autonomous and there is increasing evidence of local control by autocrine and paracrine factors produced by the tumour cells and the stroma. The rate of proliferation during the lifetime of a tumour is not constant. In experimental tumours in the early stages, growth is exponential and the growth fraction is high. As the tumour enlarges, the growth rate slows and the growth fraction falls. A decreased rate of growth is commonly observed for transplantable tumours in animals and probably results, in part, from decreasing tumour vascularity and cellular nutrition leading to slowing of cell proliferation, and also from cell loss due to death or differentiation. The smallest tumour that is likely to be clinically detectable (either by physical or radiological assessments) will be approximately 1 cm in diameter and will contain 108–109 tumour cells, depending on the contribution of
stroma and other elements to tumour bulk; growth of these small, clinically detectable tumours follows a Gompertzian pattern. Such a tumour will have undergone approximately 30 doublings in cell number if it is clonally derived from a single transformed cell, and will usually weigh about 1 g. Growth to a potentially lethal mass of 1 kg of tumour requires only a further 10 doublings of cell number. Thus the period of tumour growth that is clinically apparent is only a relatively short period in the total life history of a tumour, and clearly the potential exists for micro-metastases to develop before detection of the primary tumour. The effect of chemotherapy on a tumour is influenced by some of the features of its growth pattern: ●
●
●
●
Response to chemotherapy is proportional to the number of cells synthesizing DNA. The shorter the doubling time at the onset of treatment (i.e. the more rapid is tumour growth), the better the response to chemotherapy, since more cells will be synthesizing DNA. As the tumour grows, the disease becomes less easy to cure. As the tumour shrinks with treatment, the growth rate increases because of the Gompertzian growth pattern.
CELL CYCLE
All proliferating cells go through a series of events that comprise the cell cycle. The division of the cell cycle into
76 Principles of chemotherapy and drug development
discrete phases followed the demonstration that DNA synthesis took place during a defined time interval, rather than continuously during interphase. After mitosis (M), the cell spends a variable resting period (G1) during which DNA synthesis does not occur but RNA and protein are produced. Entry to the S phase is heralded by an increase in RNA synthesis followed by doubling of the DNA content. The G2 phase follows as DNA synthesis ceases, and is followed by mitosis. The total duration of the cell cycle depends mainly on the duration of the G1 phase, which may be 0–30 hours. The durations of the S phase (6–8 hours), M phase (less than 1 hour) and G2 phase (2–4 hours) are fairly constant both in normal and malignant cells. Consequently, the cycle of a malignant cell may last between 9 and 43 hours. For aggressive, highly proliferative tumours, the cell cycle will be at the shorter end of the time span, but for more indolent, low-grade tumours, the cell cycle will be significantly longer. The mean cycle time of cells within human tumours is typically much shorter than the mean volumedoubling time of the tumours for two main reasons: a high rate of cell death, and a high proportion of non-proliferating cells. The term G0 is applied to cells that are out of cycle. The proportion of cells within a population that is undergoing active proliferation in the cycle is termed the growth fraction. Estimates of growth fraction calculated by comparing the measured proportion of cells in S phase with that predicted from the phase distribution of cycling cells are consistently of the order of 20–30 per cent. This is particularly relevant, as most anti-cancer drugs do not cause cell death during the G0 phase. Furthermore, although higher proportions of S-phase cells are found in some rapidly growing tumours such as high-grade lymphomas, most tumours do not have a higher proportion of S-phase cells than some normal highly proliferative tissue such as bone marrow and intestinal crypt cells. Mathematical models have been developed to describe the interaction of cytotoxic chemotherapy and tumour growth kinetics and may be used to evaluate hypothetical strategies for cancer treatment.1
ANTI-CANCER DRUGS AND THE CELL CYCLE
Cytotoxic chemotherapy agents have traditionally been classified as phase or non-phase specific, depending on the effect on the cell cycle (Table 5.1). In-vitro models demonstrate that phase-dependent drugs kill cells exponentially at lower doses but reach a plateau when given at a higher dose because they are only able to kill cells in a specific part of the cell cycle. Non-phase-dependent drugs kill cells exponentially with increasing dose and are equally toxic both for cycling cells and those in G0. The practical value of this classification is somewhat limited in that chemotherapy regimens designed on kinetic principles have so far shown no advantage over those derived empirically.
Table 5.1 Cytotoxic drugs and the cell cycle Predominantly non-phase-specific agents Nitrogen mustards Cyclophosphamide Melphalan Chlorambucil Busulphan Thiotepa 5-Fluorouracil Doxorubicin Mitomycin-C Dacarbazine Actinomycin-D
Predominantly phase-specific agents Methotrexate Cytosine arabinoside 6-Mercaptopurine 6-Thioguanine Vincristine Vinblastine Bleomycin Etoposide Procarbazine
SKIPPER HYPOTHESIS FOR CELL KILL BY CYTOTOXIC AGENTS
In the early 1960s, Skipper et al.2 formulated some principles of tumour cell kill on the basis of experiments using the L1210 murine leukaemia model: ●
●
●
●
The survival of an animal is inversely related to the tumour burden. A single leukaemic cell is capable of multiplying to kill the host. For most drugs there is a clear relationship between dose of drug and eradication of tumour cells. Cell destruction by anti-cancer drugs follows log kill kinetics. That is, a given dose of drug kills a constant fraction of cells and not a constant number. Thus if a particular dose of an individual drug kills 3 logs of cells and reduces the tumour burden from 1010 to 107 cells, the same dose used at a tumour burden of 105 will reduce the tumour burden to 102 cells. The cell kill is therefore proportional regardless of tumour burden.
These principles established that there is an inverse relationship between cell number and curability and imply that tumours are best treated when they are small in volume. Furthermore, if drug treatment is discontinued as soon as the tumour is no longer clinically detectable, at least 109 tumour cells remain and relapse is inevitable. However, these observations should be considered in the context of the growth differences between this murine leukaemia model and human cancers. For example, L1210 leukaemia is a rapidly growing tumour, with a high percentage of cells in S phase and a growth fraction of 100 per cent, giving a consistent and predictable cell cycle. In contrast, the cell cycles of human tumours are heterogeneous, prolonged and with a smaller growth fraction. Overall, the evidence favours a Gompertzian growth pattern with a growth fraction that is not constant and with an exponential decrease in growth rate as the tumour enlarges.
Basic principles 77
Nevertheless, Gompertzian kinetics also support the notion that chemotherapy is more likely to be effective in eradicating a small tumour burden. When the tumour burden is small (such as when no longer clinically detectable), its growth fraction would be at its largest, and the proportional cell kill would be larger. This is one of the principles that form the basis of adjuvant chemotherapy strategies. NORTON–SIMON HYPOTHESIS
In tumours that show Gompertzian-type growth curves, the rate of re-growth increases as the tumour shrinks with therapy. Thus the level of treatment necessary to initiate a regression may be insufficient to maintain the regression and produce cure. Norton and Simon3 hypothesized that to overcome the slowing rate of regression in a tumour responding to therapy it was necessary to increase the intensity of treatment as the tumour became smaller. This can be achieved in one of two ways: 1. increase the dose intensity of the chemotherapy agents used to induce remission; 2. switch to alternative cytotoxic agents in an aggressive schedule. Dose intensity is commonly used in leukaemia, for which agents such as cytosine arabinoside are used in highdose pulses after the induction of remission. In addition, high-dose chemotherapy with bone-marrow transplantation or peripheral blood stem-cell harvesting with growth factor support is another means of achieving dose intensity. The concept of dose intensity is discussed more fully later in this chapter. Alternatively, the use of other cytotoxic agents in hybrid regimens such as MOPP-ABVD for Hodgkin’s disease4 not only exposes the tumour to drugs that are different from those used to achieve induction or remission, but also attacks residual populations of cells that are biochemically resistant to the initial combination of drugs. GOLDIE–COLDMAN MODEL
Spontaneous mutation is a basic property of DNA. There is also evidence that tumour cells may be more genetically unstable than normal cells. In 1979, Goldie and Coldman proposed a model to explain the genetically determined resistance to cancer chemotherapy based on this principle.5 They proposed that populations of cells within a tumour are capable of randomly mutating and becoming resistant to the cytotoxic agents. These spontaneous mutations occur at population sizes of less than 106 tumour cells, which is less than the clinically detectable level. As such mutations occur at frequencies of 106 or higher, a clinically detectable tumour of 109 cells is likely to have several drug-resistant clones. However, the absolute number of resistant cells would be relatively small and these tumours would probably respond initially to treatment with a
complete or partial remission, only to relapse and reappear when the resistant clone(s) expand, a clinical picture that is familiar in oncology practice. Some tumours are resistant to cytotoxic chemotherapy agents even when they present with a relatively small tumour volume. However, such slow-growing tumours may have considerable cell loss through cell death – up to 90 per cent of the tumour volume. Therefore what appears to be an early tumour may have gone through many more cell doublings than expected in order to compensate for cell loss in achieving that size. Consequently these cells may have undergone a higher frequency of spontaneous mutations, leading to a tumour that consists predominantly of drugresistant clones. However, there are many other mechanisms of drug resistance, including decreased uptake due to changes in drug-specific transport mechanisms, decreased activation of pro-drugs, alterations in cellular metabolism and repair mechanisms, increased inactivation of drugs, target alterations and acquisition of a multi-drug-resistance phenotype. Drug resistance is discussed more fully in a later section of this chapter.
Principles of treatment ROUTE OF ADMINISTRATION
The route of administration of a cytotoxic drug is determined by the stability, size, molecular charge and sclerosant characteristics of that drug. Traditionally, the oral route has been used infrequently because of the unpredictability of patient compliance and variable absorption of the drug. However, the desire to move cancer treatment from a predominantly hospital-based, inpatient system into the ambulatory setting, together with a growing body of information showing higher anti-tumour activity, or lower systemic toxicity, with dosing regimens that produce prolonged exposure to some cancer agents suggest that oral cancer chemotherapy may be increasingly of interest in the future. This has already led to the evaluation of oral administration of anti-cancer drugs that have been available for many years (e.g. etoposide, idarubicin) as well as strategies for oral use of anti-cancer drugs traditionally given intravenously, for example the oral fluoropyrimidines.6 Subcutaneous and intramuscular routes are rarely employed because large volumes of diluent are frequently required to dissolve the drug, or because, once dissolved, they are often highly toxic to tissues in the extravascular compartments. For the majority of cytotoxic drugs, the intravenous route is the optimal means of delivery for most indications, and the drug can be administered by bolus injection, short infusions or protracted venous infusion, usually through an indwelling central venous catheter in the last case. Other routes that can be used for specific indications include the intrathecal administration of non-sclerosant water-soluble
78 Principles of chemotherapy and drug development
drugs such as methotrexate or cytosine arabinoside to palliate or prevent meningeal disease; intra-arterial administration of drugs via the hepatic artery to achieve high doses of drug to be delivered locally or in chemo-embolization strategies; intra-arterial administration in isolated limb perfusion in melanoma or sarcoma; intravesical administration in reducing recurrence rates of superficial bladder cancers; and intraperitoneal administration to small tumour nodules on the peritoneal surface.
COMBINATION CHEMOTHERAPY
Based on the cancer cell kinetics data, it is reasonable to assume that even small tumours, when clinically detectable, already have resistant clone(s) of cells. The use of cytotoxic agents of different groups in combination should give a broader range of cover of resistant cell clones in a heterogeneous tumour population, and prevent or slow the development of new resistant cell clones, thereby giving maximum cell kill that is not possible with a singleagent regimen. There are four general principles that guide the selection of drugs for use in effective combination chemotherapy regimens: 1. Each drug should have activity against the tumour when used alone, those drugs with maximal efficacy being most preferable. 2. Drugs should have different mechanisms of action. 3. Drugs should have minimal overlapping toxicities. When several drugs of a class are available, a drug should be selected with minimal overlapping toxicity, which, although leading to a greater range of side effects, decreases the risk of potentially life-threatening cumulative toxicity to the same organ system. 4. Each drug in a combination should be used in its optimal dose and schedule. However, it should also be noted that omission of a drug from an effective combination may allow overgrowth of a cell clone that is sensitive to that drug alone and resistant to the other drugs in the combination. Conversely, adding too many drugs to a regimen with the aim of increasing efficacy may reduce the doses of some, or all, of the component drugs of the combination below the threshold of efficacy, thereby decreasing activity of the regimen. THE DOSE-RESPONSE RELATIONSHIP
Within experimental systems, it is easy to demonstrate the dose–response effect of anti-cancer drugs on cancer cells. This relationship is more difficult to demonstrate and quantify in patients due to the heterogeneity of tumour responses and also because of the variability of drug absorption, binding, distribution, metabolism, excretion and delivery of the drug to the tumour. Nevertheless, a positive dose–response
relationship has been demonstrated in retrospective studies,7 and also in prospective studies in advanced ovarian, breast and colon cancers, and in the lymphomas.8**,9–11 In experimental models, the dose–response relationship is steep and linear and the principles of chemotherapy include exploiting the differences between the dose– response curves of normal and tumour tissue. Reduction of doses in this linear phase of the dose–response curve in experimental models results in a decreased cure rate before there is a reduction in the response rate. Although complete remissions are still observed in these animal tumours, a few residual cells will not be killed and will ultimately repopulate, leading to relapse. It has been suggested that in these experimental models, a dose reduction of 20 per cent will lead to a loss in the cure rate in excess of 50 per cent.9,12 Conversely, in high growth fraction tumours, a two-fold increase in dose often leads to a ten-fold increase (1 log) in tumour cell kill. This not only re-enforces the desire to minimize the reductions in recommended doses, but also justifies efforts to increase dose to aim for increased cure rates in chemosensitive tumours. The principal restriction to dose escalation is toxicity of normal tissues. When myelosuppression is the dose-limiting toxicity, this can be overcome by the use of recombinant bone-marrow growth factors such as G-CSF, by bone-marrow transplantation, or by haemopoietic support using peripheral blood progenitor cells. Ultimately, dose escalation of these agents (and those agents that have non-bone-marrow dose-limiting toxicities) will be restricted by toxicities to tissues such as the myocardium, lungs and kidneys, for which there is no currently available means to overcome toxicity. Consequently, the drugs that can be used in dose-escalation strategies are few in number, as are the tumour types that demonstrate a dose response to these agents. SCHEDULING OF CYTOTOXIC CHEMOTHERAPY
The concentration of a cytotoxic agent in plasma or in tissue does not depend on total dose alone, but also on the schedule of administration; that is, it is believed that ‘drug exposure’ as measured by the area under the concentration–time curve (AUC) is a crucial determinant of drug activity in many situations. In a series of classical experiments with the L1210 murine leukaemia model, Goldin and Schabel13 demonstrated that methotrexate was more effective and less toxic when administered to mice in an intermittent rather than a daily regimen, and this observation has also been noted in clinical practice. Scheduling is also a determinant of achieving dose intensity. Dose intensity has been defined as the amount of drug delivered per unit time, expressed as mg/m2 per week. Relative dose intensity is the amount of drug delivered per unit time relative to an arbitrarily chosen standard single drug, or, for a combination regimen, the fraction of the ratio of the test regimen to the standard regimen. In this way, treatment delays are given equal consideration to dose reductions when calculating dose intensity. Clearly if the
Drug development 79
time interval between successive courses of chemotherapy is too prolonged, this can allow repopulation of cancer cells, which may well have acquired resistance to the cytotoxic agents to which they have been exposed. If the time interval is too short, toxicity in normal tissues may be unacceptable. Consequently, scheduling may influence outcome by affecting toxicity, allowing greater doses to be administered over the same time frame. In this regard, accurate scheduling is also required to rescue normal tissues from the toxic side effects of certain chemotherapy drugs; for example higher doses of methotrexate can be administered to patients during 24 hours when a sufficient flow of alkaline urine is maintained and by rescue of normal tissues by giving folinic acid titrated to the plasma concentration of methotrexate. Similarly, scheduling of individual cytotoxic agents can enhance response rates, e.g. when 5-fluorouracil (5-FU) is given as a continuous intravenous infusion rather than by intermittent boluses,13**,14 and also in exploiting potential interactions between cytotoxic drugs to achieve optimal synergistic effects.
DRUG DEVELOPMENT Although many of the agents currently in use were developed by a combination of science and serendipity, the acquisition and evolution of novel agents are generally the result of both carefully thought-out screening processes and our increasing knowledge of the molecular basis of cancer development and progression, and cellular metabolism. Criteria for selecting new anti-cancer drugs can include: ● ● ● ● ● ●
novel chemical entity novel mechanism of action selective for solid tumours in-vivo activity in solid tumour models evidence for a positive therapeutic index feasibility of supply.
The serendipitous discovery of active anti-cancer agents has resulted in the development of many important drugs. In the mid-1960s, Rosenberg was studying the effects of electric currents on the growth of bacterial cultures and noted that the application of an alternating voltage reduced the growth rate of the bacteria, and that this effect was due to the formation of platinum complexes in the growth medium resulting from dissolution of platinum from the electrodes. Neutral platinum complexes exhibited a relatively selective effect on rapidly dividing bacterial cells, and therefore these compounds were tested as anticancer agents. Platinum was thus found to have significant anti-tumour activity in these tests and subsequently also in a wide variety of animal models. Another way in which new agents are appropriated is by chemically modifying an existing active agent that may otherwise have restricted utility because of toxicity. The
term ‘structure–activity relationship’ describes the link between varying specific chemical substructures on the parent compound and differential anti-tumour activity. Analogue development is aimed primarily at producing: ●
● ●
a drug as effective as the parent compound, with significantly less toxicity, a more potent drug, or a drug with a different spectrum of anti-tumour activity.
Examples of successful analogues are epirubicin (from doxorubicin), carboplatin (from cisplatin) and topotecan or CPT-11 (from camptothecin). It is expected that more active agents will be developed as a result of the increasing understanding of the biochemical and biological behaviour of cancer cells. Certainly, focusing on a defined molecular target and using computers to design and construct molecules that would interact and antagonize these targets would increase the potential selectivity for such an agent. However, empirical drug screening for novel agents will continue to play a major role in the twenty-first century.
Acquisition of drugs SCREENING
Large-scale, empirical screening for the discovery of new anti-cancer agents was initiated by the National Cancer Institute (NCI) of the USA in 1955. During the next 20 years, nearly half a million potential anti-cancer agents of natural or synthetic origin were screened, producing more than 50 per cent of all clinically useful agents. This screen used a tumour panel with two models being involved primarily – murine leukaemia L1210 and P388. In the mid1980s it was decided to change this leukaemia-based strategy because: ●
●
●
not many drugs were discovered with broad activity against common solid tumours, agents inactive against P388 were often active in other pre-clinical models, and vice versa, many active agents, e.g. bleomycin, were inactive against P388.
The NCI disease-oriented in-vitro screen has been used regularly since 1991. Here, human tumour cell lines derived from seven common cancers have been selected and maintained for use as a test-bed for novel-agent sensitivity. Such a screen could theoretically speed up the process of clinical development for agents found to be differentially sensitive to a particular tumour type. Secondary in-vitro studies may be used in order to examine the effects of schedule, exposure and potential mechanisms of action. Finally, in-vivo testing is carried out, generally using the most sensitive human tumour cell line as a subcutaneous xenograft in a nude mouse.
80 Principles of chemotherapy and drug development
FORMULATION
An intravenous formulation is generally preferred for preclinical studies, in order to exclude variations in bioavailability and to enable promising drugs to be used at their maximum tolerated dose (MTD). Solubility is also very important, as limited solubility is regularly a reason to discontinue overall development of an agent. Various approaches and experimental methods for solubilization are being used, in addition to techniques such as liposomal encapsulation, conjugation to polymers and other delivery systems that could empower the clinical evaluation of a promising agent. However, these means of solubilizing the drug may alter the toxicity profile and/or the anti-tumour efficacy. TOXICOLOGICAL AND PHARMACOLOGICAL TESTING
Unlike other drugs in clinical use, cytotoxic agents have very narrow therapeutic windows, i.e. the most biologically active doses are often close to the lethal doses. It is therefore necessary to try to predict the potential toxic effects, and also to define a safe (but still potentially active) starting dose prior to human studies. Animal studies are often performed using two species. In Europe, an accepted method for obtaining the safe dose for human clinical trials is defined as one-tenth the dose that is lethal to 10 per cent of mice (called the LD10), providing this dose is safe in another species, e.g. rat. However, it is important to note that animal toxicities do not always predict similar toxicity in humans. It is rarely necessary to perform additional tests on larger animals (e.g. dogs, monkeys), although some data suggest that tests in these species are more able to predict certain human toxicities. Additional toxicological studies are undertaken in order to evaluate more fully any organ toxicity demonstrated in the animals, and to investigate any relationships with dose, schedule and whether any documented effects are potentially reversible. Single-dose and multiple-dose studies are undertaken in order to mimic more accurately the anticipated human trial protocols. Finally, pharmacological studies derive knowledge about bioavailability, metabolism, route of excretion etc. and allow rational schedule design for the clinical testing of the agent.
Clinical testing For many anti-cancer drugs, a dose–response relationship is observed in pre-clinical studies, and this knowledge is subsequently integrated into the design of the initial clinical studies. There are three stages (phases) of clinical studies used for the evaluation of a novel compound. PHASE I TRIALS
Phase I studies are concerned with the first exposure of a novel agent to humans. The primary objective is to establish
the MTD by an escalating dose protocol, and to document the toxicity profile of the agent at the chosen dose, schedule and route of administration. However, one of the goals of pre-clinical studies is to try to predict a phase I starting dose close to the agent’s therapeutic window. Despite this, it has been noted that clinical response rates in phase I trials are less than 5 per cent overall. This raises ethical dilemmas, particularly with respect to informed consent. However, these studies are conducted in a patient population deemed to have disease that is refractory to standard chemotherapy, or disease for which there is no known effective treatment. Anti-cancer efficacy is therefore necessarily a secondary endpoint, and unfortunately the first patients are often treated at sub-therapeutic doses. Research on study designs that decrease the number of patients treated at sub-therapeutic doses is ongoing, and includes the accelerated titration design and continual reassessment method (CRM).15 The former design allows for one patient per dose level in the early stages, provided no significant toxicity is encountered. The CRM allows for intra-patient dose escalation. Both designs may be useful, provided that data on toxicity are always as up to date as possible. In most trials, however, patients (generally three) are treated at the first dose level and monitored for toxicity. Subsequent patients are entered at the next dose level (cohort) and so on, until significant toxicity is encountered. Historically, the degree of dose escalation is based on a modified Fibonacci scheme (e.g. dose level 2 is twice dose level 1; level 3, 167 per cent of level 2; level 4, 150 per cent of level 3). Some studies have used a pharmacokinetically guided scheme, involving target AUCs. All studies incorporate pharmacokinetic and pharmacodynamic aspects, and assessments of biological endpoints are often included (e.g. inhibition of a target enzyme). Dose-limiting toxicities (DLTs) are defined, and once the MTD has been reached at a particular dose level, the previous lower dose level is generally expanded to a larger cohort of patients and becomes the recommended dose for further studies. Any anti-tumour activity documented may also be useful in planning subsequent phase II studies.
PHASE II TRIALS
These studies are conducted using the dose and schedules from phase I studies considered to be optimal in terms of pharmacology and therapeutic index (i.e. near to the MTD). The stated objective of phase II trials is to determine a level of efficacy for the drug in question, and therefore patients with measurable disease are selected. In addition, patients are less heavily pre-treated in order to select a patient population with the best chance of demonstrating anti-tumour activity, if it exists. Phase II studies may be directed towards specific tumour types, based on any responses seen in phase I studies, or towards a well-defined spectrum of pre-clinical activity. Further information on drug toxicity is gathered, and specific data on cumulative toxicity are acquired, because more patients
Classification of cytotoxic drugs, mode of action, toxicities and clinical utility 81
are likely to receive multiple cycles of treatment. Phase II trials are designed to allow early study termination if a drug has no, or minimal, efficacy. Designs such as the Gehan two-stage model allow a number of patients to be accrued (generally 14), and if one or more response is documented, recruitment proceeds to a total of 25 patients. If no responses are documented in these first 14 patients, the trial terminates. A response rate of 20 per cent may be considered worthy of further evaluation; however, this should relate to the patient population being tested. If performed in patients with a relatively chemoresistant malignancy, pre-treated with ‘standard’ chemotherapy, the likelihood of detecting activity may be prejudiced. Another design is to randomize patients to the experimental treatment and a drug with known activity in the disease (e.g. the parent compound of a novel analogue), or even to another experimental agent. Such studies will not allow the determination of efficacy differences, but may give a baseline response rate for the patient population being selected. Finally, new active agents are often subsequently added to other standard therapies in two-drug or three-drug combination regimens. These need to be piloted as feasibility/phase I type studies to ensure that toxicity is neither additive nor synergistic. PHASE III TRIALS
The next stage of testing for a novel agent is to compare against the existing standard therapy. The most scientifically valid method is by a prospective, randomized clinical trial, in which patients eligible for the protocol are randomly assigned to one of the treatment arms. Suitable endpoints for a phase III study include progression-free survival (PFS), overall survival (OS) and overall response rate (ORR). Data on quality of life and comparative toxicities are also collected in order to determine the overall clinical benefit of a treatment. Stratification for known prognostic factors is carried out to ensure that the arms of the study are statistically balanced. Such procedures do not ensure that the study will include a representative sample of the general population with the disease, but can provide evaluation of the relative merits of the treatments being compared. Variations on the simple randomized study format include: ●
●
the crossover design, which allows patients relapsing on one treatment to receive the other drug, therefore using the patient as his/her own control, the 2 2 factorial design, wherein the first factor may be a comparison between two treatments (e.g. the novel agent versus standard therapy), and the second factor interventions consequent on the first factor (e.g. maintenance treatment or not).
Phase III trials are generally large, consisting of hundreds of patients, because the differences between treatments may be small and therefore increasing sample size increases the
chance of small differences in outcome being statistically significant.
CLASSIFICATION OF CYTOTOXIC DRUGS, MODE OF ACTION, TOXICITIES AND CLINICAL UTILITY Historically, anti-cancer agents have been divided into groups according to the mechanism of cytotoxicity. However, the increasing number of new drugs, and their diverse and novel mechanisms of action, make such divisions increasingly difficult and arbitrary.
DNA alkylators The alkylating agents were the first compounds identified to have activity against neoplastic diseases, and have been in use now for more than 50 years. Early studies used the nitrogen mustard mechlorethamine, following observations from the mustard gas chemical warfare programme that soldiers exposed to this agent developed aplasia of the bone marrow. Mechlorethamine was subsequently found to induce significant tumour regression in patients with lymphoma in clinical trials. Mechlorethamine is ideally suited to illustrate the chemistry of alkylation, having the simplest structure of the class. The CH2CH2Cl linked to nitrogen is labelled the mustard group, and in the case of mechlorethamine there are two such groups, thus giving the term ‘bifunctional nitrogen mustard’. Following administration, the drug undergoes an internal cyclization reaction and loses a chloride ion to form an electron-deficient, positively charged aziridinium ion. These highly reactive species are able to react and form covalent bonds (adducts) with electronrich (nucleophilic) sites in DNA, such as the 7-nitrogen group of guanine on the major groove. Other biological macromolecules, e.g. proteins, are also targeted, but it is the reactions involving the nitrogenous bases in DNA that are critical for the anti-cancer action of the alkylating agents. Adduct formation with two separate bases on the DNA, especially across the two anti-parallel strands – the inter-strand crosslink – is thought to be the most lethal interaction. The results of DNA alkylation involve interference with the fidelity of replication and transcription by abrogating the functions of DNA and RNA polymerases. It follows that alkylating agents are most toxic to rapidly cycling cells; tumours with a high fraction of cells in S phase are more vulnerable, possibly as they have less damagerepair time. In addition, adduct formation leads to structural lesions, which include ring opening and base deletions. Cellular repair processes attempt to restore the integrity of the DNA, but if incomplete can result in further damage, such as the creation of apurinic sites or strand breaks. Furthermore, it appears that these repair processes can be saturated by higher doses of alkylating agents, thus providing
82 Principles of chemotherapy and drug development
a rationale for extending the use of these agents into highdose chemotherapy techniques. MECHLORETHAMINE (NITROGEN MUSTARD, MUSTINE)
Mechlorethamine is administered intravenously, and has a half-life of approximately 3 minutes due to rapid hydrolysation. It has extremely vesicant properties, and can also cause severe phlebitis and sclerosis of the vein used. The major use is in the treatment of Hodgkin’s disease, as part of the combination MOPP (nitrogen mustard, vincristine, procarbazine and prednisolone). It is also occasionally administered by direct intra-cavity infusion for the treatment of malignant effusions. MELPHALAN
Melphalan is another bifunctional alkylating agent, and is a phenylalanine derivative of mechlorethamine. It is active against a broad cross-section of tumours (lymphomas, breast and ovarian cancers, multiple myeloma) and can be administered orally, intravenously or intraperitoneally. Although the oral bioavailability is variable, with 20–50 per cent being excreted in the stool, the ease of oral administration makes this the most common route employed. It routinely causes myelosuppression, with nadir counts occurring at 4–5 weeks following a short, 7-day, oral course. Doses require to be adjusted according to the level of myelosuppression. It is also used in high-dose chemotherapy regimens for haematological malignancies and multiple myeloma, due to exponential cell killing demonstrated in cell culture. Occasionally, it is used intraperitoneally for relapsed ovarian cancer, although efficacy data are not well established. CHLORAMBUCIL
This bifunctional alkylating agent is a benzene butanoic derivative of mechlorethamine, and is also a close structural congener of melphalan. It is almost completely absorbed when given by the oral route, and is used either continuously or intermittently for long periods in lowgrade lymphoma, chronic lymphocytic leukaemia and multiple myeloma. It can also be given to elderly patients with ovarian cancer who are unable to tolerate more aggressive chemotherapy. The toxicity is predictable myelosuppression, which makes dose adjustments fairly straightforward. However, stem-cell damage is cumulative and irreversible, leading to problems of severe myelosuppression with subsequent cytotoxic drug use. Longer-term use (e.g. 1 year) has been associated with pulmonary toxicity and the development of second malignancies.
to develop cytotoxicity. It undergoes a complex multi-step activation process, initially being metabolized by the cytochrome P450 system in the liver and eventually converted to a variety of active metabolites, of which phosphoramide mustard is thought to be the major cytotoxic derivative. One of the main metabolites, acrolein, is excreted in the urine and can cause a chemical (haemorrhagic) cystitis in up to 10 per cent of patients. Adequate hydration and the concurrent administration of sodium-2-mercaptoethane (MESNA), which inactivates acrolein in the urine, can prevent this toxicity. Cyclophosphamide has good oral bioavailability of around 90 per cent and therefore is often utilized by this route. The main toxicity is myelosuppression, with the nadir occurring 10–21 days after intravenous administration. Doses of 750 mg/m2 are common in regimens such as intravenous CMF (cyclophosphamide, methotrexate and 5-FU), whereas higher doses may be given as part of high-dose stem-cell transplantation. Such high doses can lead to problems such as hyponatraemia, haemorrhagic carditis, cardiomyopathy and pulmonary fibrosis. Like all alkylating agents, there is a risk of developing second malignancies. The major use of cyclophosphamide is as a component of combination regimens for cancers of the breast, small-cell lung and haematological malignancies. The CMF regimen is still one of the mainstays of chemotherapy for breast cancer, with variations existing that deliver different dose intensity and toxicity. It has recently fallen out of favour in ovarian cancer treatment, with the introduction of newer, more active combinations. Ifosfamide is a structural analogue of cyclophosphamide, which exhibits a similar spectrum of activity but different pharmacological properties and toxicity profile. It is generally administered as prolonged infusions, with concurrent MESNA and hydration, due to a higher incidence of urothelial toxicity. In addition, ifosfamide can cause a severe but reversible neurological syndrome, characterized by altered mental state, cerebellar dysfunction, cranial neuropathies and epileptiform seizures. Risk factors for the development of neurotoxicity are impaired renal or hepatic function. Both ifosfamide and cyclophosphamide can cause impairment of gonadal function. The usefulness of ifosfamide is evidenced by its incorporation into chemotherapy regimens that have been shown to be curative in germ-cell malignancies. It is also used in the treatment of sarcomas and lymphomas and has demonstrated activity in ovarian cancer. As part of the ICE regimen (ifosfamide, carboplatin and etoposide) or combined with vincristine, VICE, it has been used in the treatment of small-cell lung cancer. NITROSOUREAS (BU, CU)
OXAZAPHOSPHORINES (CYCLOPHOSPHAMIDE AND IFOSFAMIDE)
Cyclophosphamide differs from the previously described alkylating agents in that it is a pro-drug, requiring activation
BU (1,3-bis-(2-chloroethyl)-1-nitrosourea,carmustine) and CU (1-(2-chlorothyl)-3-cyclohexyl-1-nitrosourea, lomustine) are important agents in that they exhibit only partial cross-resistance with the other alkylating agents. They are
Classification of cytotoxic drugs, mode of action, toxicities and clinical utility 83
highly lipid soluble and easily cross the blood–brain barrier. Clinical activity has been observed in lymphoma, melanoma, myeloma and malignant glioma. Major toxicities are emesis and delayed myelosuppression, which can be severe. BU is given intravenously, whereas CU can be administered orally. ALKYLAKLANE SULPHONATES (BUSULFAN, TREOSULFAN)
Busulfan is an orally available bifunctional alkylating agent which, when hydrolysed, binds extensively to nucleophilic sites on DNA to form crosslinks. It is used mainly in haematological malignancies, and is an integral component of many high-dose chemotherapy regimens preceding stem-cell transplantation. Myelosuppression can be severe and long lasting after excessive dosing. Other notable toxicities include ‘busulfan lung’, a form of interstitial pulmonary fibrosis, and increased pigmentation in an Addisonian distribution. Treosulfan was synthesized in 1961 as dihydroxybusulphan and is a bifunctional alkylator, structurally related to busulfan. It is a pro-drug, activated to the reactive epoxide in a non-enzymatic, first-order, pH-dependent process. It has demonstrated specific activity in ovarian cancer and is available in intravenous and oral form. The predominant toxicity is mildto-moderate myelosuppression. AZIRIDINYL DRUGS (THIOTEPA, MITOMYCIN C)
Thiotepa (N,N,N-triethylenethiophosphoramide) is a complex drug that acts as a monofunctional DNA-alkylating agent. It is thought to act as a pro-drug, being metabolized to highly reactive aziridine moieties via hydrolysis, following diffusion into the cell. It is delivered by the intravenous or intra-cavity route and readily crosses the blood–brain barrier. It has a role in the palliation of refractory ovarian cancer as an intraperitoneal agent, and recently there has been a resurgence of interest in its use as part of high-dose chemotherapy due to its relative lack of non-myelogenous toxicity. Mitomycin C is related to the anthracycline anti-tumour antibiotics, being derived from Streptomyces species, but differs substantially in that it is the prototype bioreductive agent, undergoing preferential activation in the hypoxic environment found in solid cancers. Once activated, mitomycin C performs bifunctional alkylation, crosslinking DNA with the adduct in the minor groove and causing strand breaks. The utility of mitomycin C is limited by delayed myelosuppression, similar to that of the nitrosureas. Other notable toxicities include renal failure and cardiomyopathy, both related to the total cumulative dose of mitomycin C administered. The elucidation of the mechanism of action of mitomycin C has evoked great interest due to the potential of tumour selectivity through preferential reductive activation in hypoxic solid tumour masses. A range of aziridinyl agents, including E09, has been developed and is undergoing clinical evaluation.
N-METHYLTRIAZINES AND MELAMINES
Dacarbazine (DTIC) was initially thought to function as an anti-metabolite, given that its genesis was as an analogue of 5-amino-imidazole-4-carboxamide, a purine precursor. However, it is now thought to be hepatically activated to function as an alkylating agent. It is decomposed by exposure to light and is administered by intravenous infusion. It is active against a broad spectrum of tumours, but the main clinical utility is in the treatment of malignant melanoma, lymphoma and sarcomas. Hexamethylmelamine has an uncertain mechanism of action, but is likely to act as a DNA-methylating agent, with crosslink formation. However, it is incompletely cross-resistant with classical alkylating agents such as cyclophosphamide. It is administered by the oral route, but bioavailability is erratic due to variable first-pass metabolism. It is used in the treatment of ovarian cancer, but emesis can be problematic. In addition, it can produce neurological toxicity in some patients. Procarbazine is metabolically activated in the liver microsomes into a DNA-methylating species. It is generally administered orally in the treatment of Hodgkin’s disease and is also useful in the treatment of brain tumours due to its ability to penetrate well into the cerebrospinal fluid. Side effects are usually not severe, but care should be taken with co-administration of other drugs, due to the inhibitory effect of procarbazine on monoamine oxidase.
Platinum analogues Platinum, transition metal, possesses eight electrons in its outer d shell, which are highly polarizable and able to form covalent bonds. Cisplatin is a square planar molecule, which has two chloride and two ammonia ligands in the cis-configuration. This stearic conformation is important for anti-tumour effect, as the trans-isomer has no activity. The pharmacological behaviour and activation to a cytotoxic species are determined by the aquation reaction, in which a chloride ion is replaced by a water molecule. This reaction mainly occurs in the cellular cytoplasm, as the low chloride-ion concentration favours driving the formation of the active species. Once in the cell, platinum complexes are able to react with nucleophiles, such as DNA bases, RNA and proteins, to form adducts. These adducts can bridge the strands of DNA in a similar way to the crosslinks formed by melphalan and chlorambucil. The actual pathway from DNA damage to cell death has not been fully elucidated, but involves many steps, resulting ultimately in apoptosis or programmed cell death. Cisplatin is generally given intravenously and has an initial half-life of 40 minutes and a terminal half-life of greater than 24 hours. It is highly protein bound, and renally excreted, resulting in nephrotoxicity due to a
84 Principles of chemotherapy and drug development
mechanism thought to be related to persisting adducts in renal tissue DNA. Large volumes of fluid are required to ameliorate this renal damage, but significant cumulative doses of cisplatin invariably lead to an irreversible reduction in glomerular filtration rate. Renal wasting of magnesium and potassium can occur, and therefore supplementation is provided during the hydration phases. In addition, cumulative cisplatin dosing produces neurotoxicity, which usually manifests as a sensory peripheral neuropathy or hearing loss. No interventions are known to abrogate this toxicity, and when present it is thought to be only partially reversible over many months or years. Other significant toxicities include emesis, which is routinely managed by prophylactic use of 5-HT3 antagonists in addition to corticosteroids. Myelosuppression is mild to moderate. Cisplatin has a wide spectrum of activity against solid tumours, and is an integral component of curative regimens for testicular and ovarian cancer. It is also used in the treatment of upper gastrointestinal cancers, head and neck cancer, cervical and endometrial cancer, lung cancer (nonsmall cell and small cell), bladder cancer and osteosarcomas. Intraperitoneal cisplatin has been utilized in malignancies such as ovarian cancer and gastrointestinal cancer, with demonstrated activity and a decreased frequency of systemic toxicity. However, optimal pharmacokinetic and pharmacodynamic considerations mean that intraperitoneal tumour masses of 1 cm are required to ensure adequate delivery of drug to target. Carboplatin is a cisplatin analogue that is less potent but more stable, with a longer half-life. Mechanistically, the ultimate reaction products of carboplatin are thought to be chemically identical to those of cisplatin, and the drugs have similar, if not identical, spectra of activity. However, toxicity profiles are different, with carboplatin being much less nephrotoxic and neurotoxic, but causing more bonemarrow suppression. Clearance of carboplatin is also renal, with the majority of the administered dose appearing in the urine during the first 24 hours. Dosing is most accurately performed using the Calvert formula, wherein the required AUC is chosen and the dose in milligrams is calculated by: (glomerular filtration rate 25) desired AUC. Carboplatin has virtually replaced cisplatin in combination chemotherapy for ovarian cancer, having demonstrated equivalent efficacy in prospective, randomized trials. However, carboplatin has been shown to be less effective than cisplatin in testicular cancer; therefore cytotoxic equivalency extrapolations between tumour types is not recommended. Oxaliplatin is a third-generation platinum analogue, from the 1,2-diaminocyclohexane (DACH) platinum family. Pre-clinical studies have demonstrated at least equivalent potency when compared with cisplatin but, more interestingly, a degree of non-cross-resistance with other platinum compounds. In addition, data from the US NCI COMPARE programme – a screen for functional families of cytotoxics – identified DACH platinum compounds as
mechanistically different from cisplatin and carboplatin. Although oxaliplatin engages DNA in a similar way to the other platinum compounds, it seems likely that the specific cellular target and exact mechanism of action are different. Tumour types sensitive to oxaliplatin include ovarian cancer and colorectal cancer, and there is evidence for synergy with 5-FU. An unusual type of peripheral neurotoxicity has been documented, mainly characterized by reversible paraesthesias or cold-related dysaesthesias. Haematological toxicity is mild.
Anti-metabolites Anti-metabolites interfere with key steps in normal cellular metabolism due to their similarity of structure to certain RNA and DNA precursors. They can act as substrates for key enzymes, or inhibit enzymic reactions crucial to the synthesis of RNA and DNA, and are therefore S-phase specific.
ANTI-FOLATES
Methotrexate primarily inhibits dihydrofolate reductase (DHFR), an enzyme that functions to catalyse the conversion of dihydrofolate to tetrahydrofolate, which, in turn, is converted to a variety of co-enzymes involved in reactions in which the carbon atom is transferred in the synthesis of thymidylate, purines, methionine and glycine. By abrogating thymidylate monophosphate synthesis, methotrexate inhibits RNA and DNA synthesis. Folic acid (also known as leucovorin), given orally or intravenously, is converted to the tetrahydrofolate co-enzymes that are needed for the function of thymidylate synthase (TS). This is able to bypass the blocking activity of methotrexate to prevent systemic toxicity. It can also be given locally as a mouthwash or as eye drops. Methotrexate is well absorbed orally at low doses, but higher doses are given parenterally. An initial fast half-life is followed by a prolonged phase of renal excretion and a long terminal half-life. This is responsible for methotrexate toxicity to the bone marrow, mucous membranes and gastrointestinal tract – areas of high cell turnover and active DNA synthesis. Because methotrexate can accumulate in third spaces and be slowly released into the circulation, its use should be avoided in patients with effusions because of the risk of severe toxicity. Methotrexate also penetrates the blood–brain barrier at high doses, and can be given intrathecally for meningeal disease. With high doses, adequate diuresis should be obtained. It is also 50 per cent albumin bound and can be displaced by other protein-bound drugs, leading to higher systemic levels of free methotrexate and increased toxicity. However, it is generally well tolerated, with few side effects unless the risk factors are not taken
Classification of cytotoxic drugs, mode of action, toxicities and clinical utility 85
into account. Severe toxicity is manifested by myelosuppression, oropharyngeal ulceration and diarrhoea, with renal and hepatic failure and pneumonitis seen less commonly. Indications for use are as part of combination regimens in breast cancer, haematological malignancies, osteosarcoma and choriocarcinoma. Methotrexate is also known to act partly through inhibition of TS, which catalyses the methylation of deoxyuridylate (dUMP) to thymidylate, which is then incorporated into DNA. More specific inhibitors of TS have been developed, which target the folate-binding site of the enzyme. Raltitrexed (Tomudex) acts as a direct and specific TS inhibitor which, once transported into cells, is extensively polyglutamated to chemical entities that are even more potent inhibitors of TS. Such polyglutamation increases the duration of TS inhibition, which in theory could improve anti-tumour activity. Pemetrexed targets TS, but also inhibits dihydrofolate reductase and glycinamide ribonucleotide formyl transferase (GARFT), folatedependent enzymes involved in purine synthesis. Again, once inside the cell, pemetrexed is an excellent substrate for folylpolyglutamate synthase. FLUOROPYRIMIDINES
5-Fluorouracil (5–FU) is an analogue of uracil that is converted by multiple alternative biochemical pathways to several cytotoxic forms. It is converted to 5-fluoro-2deoxyuridine (FUDR) by the enzyme thymidine phosphorylase, and subsequent phosphorylation by thymidine kinase results in the formation of 5-fluoro-deoxyuridine monophosphate (FdUMP). In the presence of reduced folate, FdUMP forms a stable covalent complex with TS, which is also a key enzyme in the de-novo synthesis of the pyrimidine deoxynucleotide, deoxythymidine triphosphate (dTTP), a direct precursor for the synthesis of DNA. This inhibition of TS is considered to be the main mechanism for the action of 5-FU, although nucleotides of 5-FU can also be incorporated directly into both DNA (5-fluoro2-deoxyuridine-5-triphosphate, FdUTP) and RNA (fluorouridine triphosphate, FUTP). The TS–FdUMP complex is slowly dissociable, with a half-life of 6 hours, and the expression of TS is cell-cycle dependent, with high activity during the S phase. Also, as the presence of reduced folate is critical for TS–FdUMP complex formation, depletion of intracellular reduced folates impairs the maintenance of TS inhibition. Co-administration of leucovorin has been shown to increase the duration of TS inhibition and enhance the cytotoxic effect of 5-FU. The pharmacology of 5-FU is complex and characterized by erratic oral bioavailability, non-linear elimination pharmacokinetics and significant intra/inter-patient variability. Clearance is rapid, especially when given by bolus injection, with a half-life of 15 minutes. Several randomized trials have demonstrated the advantage of continuous infusion over bolus injection, but this is at the expense of considerable patient inconvenience (with the requirement
of an indwelling central venous catheter) and increased cost. The main uses are in the treatment of breast and gastrointestinal cancers. Toxic effects include nausea, diarrhoea, mucosal inflammation and moderate myelosuppression. Infusional schedules also increase the incidence of plantar–palmar erythrodysaesthesia (hand–foot syndrome). The development of oral fluoropyrimidines is proceeding apace, with the aim to deliver optimal 5-FU to the tumour in a convenient and controlled fashion while minimizing the variability and clearance. Capecitabine is an orally available tumour-selective fluoropyrimidine carbamate. Following administration, it is bioactivated to 5-FU by a cascade of three enzymatic reactions. After oral administration it passes unchanged through the gastrointestinal tract and is metabolized in the liver by carboxylesterase to 5-deoxy-5-fluorocytidine (5-DFCR). Then it is converted to 5-deoxy-5-fluorouridine (5-DFUR) by cytidine deaminase in liver and also tumour tissues. Further metabolism of 5-DFUR occurs selectively within tumours by thymidine phosphorylase (dThdPase) to 5-FU, thus minimizing the exposure of normal tissues to systemic 5-FU. Side effects resemble those seen with infusional 5-FU and are reversible; severe (grade III/IV) toxicities were shown to be infrequent and manageable with subsequent dose modification. The events observed most often were diarrhoea, nausea, hand–foot syndrome, vomiting, fatigue and stomatitis. Capecitabine is used in gastrointestinal and breast cancers. The enzyme dihydropyrimidine dehydrogenase (DPD) is the rate-limiting step for the catabolism of 5-FU, and converts more than 85 per cent of clinically administered 5-FU into inactive metabolites. It is primarily responsible for the high systemic clearance and short half-life of this drug, and therefore limits the amount of 5-FU available for conversion to the most active cytotoxic metabolite, 5-FUTP. DPD is found in many human tumours, but also in normal tissues, including the liver and the intestines, where it is largely responsible for the erratic bioavailability of 5-FU. Pharmacological inhibition of DPD may therefore increase the therapeutic index and efficacy of oral 5-FU, by allowing consistent dose delivery, eliminating hepatic clearance and producing more predictable drug clearance through the renal tract. In addition, over-expression of DPD in tumour cells may result in rapid intracellular fluoropyrimidine destruction and increased resistance to 5-FU. It is theoretically possible that the oral administration of fluoropyrimidines together with DPD inhibitors may overcome clinical drug resistance. In contrast to capecitabine, which is absorbed as an inactive pro-drug, other novel oral fluoropyrimidine formulations are co-administered with inhibitors of DPD in order to produce an improved pharmacokinetic profile for 5-FU. These compounds include UFT (uracil/Tegafur), eniluracil, S-1 and BOF-A2. These novel agents all deliver a therapeutic advantage from DPD modulation, and permit safe and effective 5-FU administration.
86 Principles of chemotherapy and drug development
PURINE ANALOGUES
6-Mercaptopurine (6-MP) is an analogue of the natural purine base hypoxanthine. It is converted by the enzyme hypothine-guanine phosphoribosyl transferase to the active nucleotide 6-mercaptopurine ribose phosphate, which inhibits de-novo purine synthesis. The triphosphate nucleotides also incorporate into DNA, causing strand breakage. 6-MP is absorbed well orally, and broken down by hepatic xanthine oxidase to inactive metabolites, with half the dose excreted within 24 hours. Allopurinol can inhibit this enzyme, and therefore if both drugs are co-administered, care is needed in order to reduce the risk of increased toxicity. Toxicity includes emesis, myelosuppression and a hepatic toxicity characterized by cholestatic jaundice. 6-Thioguanine (6-TG) is an analogue of guanine and has a similar mechanism of action to 6-MP. However, xanthine oxidase is not involved in its metabolism, and therefore there is no interaction with allopurinol. Again, oral administration is used, and the main indications are in haematological cancers.
PYRIMIDINE ANALOGUES
Cytarabine (cytosine arabinoside; Ara-C) is an analogue of deoxycytidine isolated from the sponge Cryptothethya crypta. It follows the same metabolic pathways of its physiological counterpart, and thus requires to be transported to the cell for activation. Cytarabine triphosphate (araCTP) is the cytotoxic metabolite of cytarabine, and acts via inhibition of DNA replication and repair and by incorporation into the DNA. Because of this, it is considered as an S-phase-specific drug, although it is active at other phases of the cycle. The main use is in the treatment of lymphoma and leukaemia, and it is given by intravenous injection because it is affected by first-pass metabolism if given orally. Intrathecal cytarabine is administered in the treatment of meningeal leukaemia and carcinomatosis. Toxicities include myelosuppression, emesis and diarrhoea. Syndromes of pulmonary toxicity and neurological toxicity occur rarely. Gemcitabine (2,2-difluorodeoxycytidine) is a pyrimidine analogue structurally similar to cytarabine. The metabolism differs in that accumulation of its active cytotoxic metabolite is higher than ara-CTP, and its elimination is much more prolonged. The mechanism of activity is similar, consisting of incorporation into DNA and inhibition of DNA synthesis. However, gemcitabine can also be incorporated into RNA. Toxicity is relatively low, consisting of myelosuppression, lethargy, flu-like symptoms and a skin rash. A rare pulmonary toxicity is also thought to be implicated. Gemcitabine is active in many pre-clinical solid-tumour models, and has demonstrated clinical activity against ovarian, gastrointestinal, breast, bladder and nonsmall-cell lung cancers. Furthermore, pre-clinical evidence for synergism has been demonstrated with several other cytotoxic agents, including cisplatin, etoposide and mitomycin C.
Anti-tumour antibiotics ANTHRACYCLINES
Doxorubicin hydrochloride is one of the most widely used anti-cancer drugs and has the broadest spectrum of activity of all chemotherapeutic agents. Daunorubicin was first isolated from Streptomyces in the 1960s, and was found to have activity against a variety of cancers. Mutations of Streptomyces led to a new strain from which doxorubicin was isolated. These two first-generation anthracyclines differ structurally by a single hydroxyl group, which results in a considerable difference in their anti-tumour activity. The mechanism of action has not been completely determined, but seems to involve DNA intercalation, free-radical formation, covalent DNA binding and inhibition of the enzyme topoisomerase II. Daunorubicin is mainly used for the treatment of acute non-lymphocytic leukaemia, whereas doxorubicin has a much wider spectrum of activity and is particularly used in the treatment of lymphomas, small-cell lung cancer, breast cancer, upper gastrointestinal cancer, sarcomas and ovarian cancer. Doxorubicin is administered by the intravenous route and is rapidly metabolized by the liver, to be excreted in bile. Caution must be used in hepatic dysfunction, and it (and all anthracyclines) is highly vesicant. Other toxicities include emesis, myelosuppression, oropharyngeal ulceration, diarrhoea and alopecia. The major long-term complication is a cumulative, dose-limiting cardiotoxicity, which is irreversible and may be fatal. Doses are usually up to 75 mg/m2 as a single agent every 3 weeks, but cardiotoxicity becomes increasingly frequent at cumulative doses of 450 mg/m2. ‘Cardioprotectors’ such as bisdioxopiperazine have been developed in an attempt to circumvent this cardiotoxicity, and are thought to work by chelating iron required by doxorubicin to produce the free radicals proposed to initiate the membrane damage evident in cardiac cells. Developmental chemistry has produced more than 1000 structural anthracycline analogues in an attempt either to increase activity of the drug or decrease toxicity, especially cardiac. Most do not proceed to clinical testing, and many get to trials only to demonstrate no clear advantage over doxorubicin or different toxicities. Epirubicin was synthesized in 1975 and has a similar spectrum of activity to doxorubicin while exhibiting less chronic cardiotoxicity. Idarubicin was shown to have a high affinity for lipids, and therefore can be administered effectively by the oral route. Liposomal drug-delivery systems (e.g. daunorubicin encapsulated by distearoylphosphatidylcholine/cholesterol, DaunoXome, and polyethylene glycol (PEG)-coated liposomal encapsulation of doxorubicin, Caelyx) have been extensively investigated as carriers for anti-cancer agents. Both DaunoXome and Caelyx have longer half-lives, higher AUCs and lower clearances than either free daunorubicin or doxorubicin, and also appear to be free of significant cardiotoxicity, and generally have decreased anthracycline-like
Classification of cytotoxic drugs, mode of action, toxicities and clinical utility 87
toxicities. However, myelosuppression is still an important feature of both formulations at recommended doses, with neutropenia occurring at grade III/IV in 50–60 per cent of patients. In addition, plantar–palmar erythrodysaesthesia was noted to be dose limiting at higher cumulative doses of Caelyx. Liposomal drug-delivery systems have been found to be especially useful in the treatment of autoimmune deficiency syndrome-associated Kaposi’s sarcoma (AIDS-KS), as conventional treatment with anthracyclines tends to be limited by cumulative toxicities (especially cardiac) preventing prolonged courses of chemotherapy.
Kaposi’s sarcoma. The main toxicities are myelosuppression, mucositis, diarrhoea and alopecia. In addition, drug extravasation can lead to severe tissue necrosis. Actinomycin D appears to inhibit DNA repair after radiation damage, and therefore the combination of both modalities enhances the risk of toxicity.
Plant-derived agents Many of the anti-cancer drugs that are used in clinical medicine are of natural origin.
MITOXANTRONE
Mitoxantrone is based on the anthracenedione structure. It is completely synthetic and was designed with the aim of retaining anthracycline anti-tumour activity with less toxicity. It undergoes DNA intercalation, in a similar fashion, and also inhibits topoisomerase II. It is given by intravenous injection and has a long terminal half-life of up to 40 hours. It is used in leukaemia, breast cancer and in combination regimens for the treatment of lymphomas. The reduced toxicity profile makes it an attractive drug to use in the elderly breast cancer patient. In addition, cardiac toxicity is less than that due to doxorubicin, probably as a result of its decreased free-radical production. BLEOMYCIN
Bleomycin is a mixture of low-molecular-weight glycopeptides isolated from the fungus Streptomyces verticullus and has both anti-bacterial and anti-cancer activity. The mechanism of cytotoxicity appears to relate to DNA binding and the production of strand breakage, with ferrous iron being essential to this mechanism. It is administered by parenteral injection and is renally excreted, with an initial half-life of 30 minutes and a later elimination phase of 2–9 hours. It is generally administered either by intravenous infusion or intramuscular injection, and also has utility via the intra-cavity route to control pleural or pericardial effusions. It has virtually no myelosuppressive toxicity or gastrointestinal side effects. Chronic administration can produce pneumonitis, which can lead to an irreversible and occasionally fatal interstitial fibrosis. Fevers, chills and flu-like symptoms are common, and prolonged administration also leads to skin pigmentation. It is used mainly in combination regimens for the treatment of germ-cell tumours and lymphomas. ACTINOMYCIN D
This compound is also isolated from Streptomyces and acts in a similar fashion, by DNA intercalation and the induction of strand breaks. It is eliminated almost unchanged in the bile and urine. Actinomycin D is usually administered intravenously. It is active against choriocarcinoma, Wilms’ tumour, Ewing’s sarcoma, embryonal rhabdomyosarcoma and, to a lesser extent, testicular cancer, lymphoma and
VINCA ALKALOIDS
Compounds with marked anti-tumour activity have been extracted from the periwinkle plant. Vinblastine and vincristine have been widely used, both as single agents and in combination with other drugs. Vinca alkaloid analogues have been prepared either by functional transformation (vindesine, desacetylvinblastine-amide) or, more recently, by hemisynthesis (vinorelbine, 5-noranhydrovinblastine). Although these compounds are all chemically related, differences have been observed in anti-tumour activity and toxicity. The mode of action of vinca alkaloids is yet to be completely understood, but they act as mitotic spindle poisons, which impair chromosomal segregation during mitosis. Microtubules are essential for normal cellular function and are involved in the maintenance of cell shape, mobility, adhesion and intracellular integrity, as well as having a role in the formation of the mitotic spindle during proliferation. Vinblastine is active in haematological cancers and in testicular and breast cancer. Vincristine is also active in these tumours, in addition to Wilm’s tumour, Ewing’s sarcoma, neuroblastoma, hepatoblastoma and embryonal rhabdomyosarcoma. The anti-tumour activity of vindesine is similar to that of vinblastine and vincristine, whereas vinorelbine is particularly active in non-small-cell lung cancer, breast cancer, ovarian cancer and Hodgkin’s disease. Vincristine and vindesine administration can cause neurological toxicity characterized by a decrease in the deep tendon reflexes, paraesthesias, constipation, myalgias, muscle weakness and paralytic ileus. Vinblastine generally produces haematological toxicities. Vinorelbine neurotoxicity is usually very mild, with only rare cases of paraesthesia or paralytic ileus being reported. Experimental evidence suggests that this may be due to the capacity of vinorelbine to bind to mitotic microtubules, rather than axonal microtubules. Clinical pharmacokinetics of vinca alkaloids are characterized by large distribution volume, high systemic clearance and long terminal half-life, although there are significant differences between analogues. TAXANES
Paclitaxel (Taxol) is a chemically complex molecule first isolated from the bark of the Pacific Yew tree Taxus brevifolia in
88 Principles of chemotherapy and drug development
the early 1970s. Its unique mechanism of action was not elicited for almost a decade and its importance as a major advance in the treatment of malignant disease was not recognized until 1989. Paclitaxel interferes with cell division by manipulating the molecular regulation of the cell cycle. In the presence of paclitaxel, polymerization of subunits of tubulin occurs, and the formed microtubules resist disassembly, thus shifting the equilibrium towards microtubule formation. Disruption of this equilibrium interferes with cell division and normal cellular activities involving microtubules. The pharmacokinetic behaviour of paclitaxel was studied during the early-phase trials in the 1980s. These studies used infusional schedules of between 1 and 24 hours and optimally modelled the concentration/time profile using a triphasic model. In general, administered dose was proportional to the area under the time/concentration curve (AUC). However, shorter infusions of paclitaxel (3 hours) seem to demonstrate non-linearity, and the important implication for this feature is that small dose de-escalations may result in a disproportionate lowering of the AUC, with subsequently decreased anti-tumour activity. Paclitaxel is 90 per cent bound to plasma proteins, but this is readily reversible and results in rapid elimination of the drug. Renal excretion is negligible, and no renal metabolites have been identified. There may be significant hepatic metabolism, and various studies are in progress to look at the effects that other drugs that use the cytochrome P450 pathway may have on the pharmacological behaviour of paclitaxel. Administration of paclitaxel is associated with allergic hypersensitive reactions ranging from acute anaphylaxis and hypotension to flushing, rashes and urticaria. Indeed, development of this agent was almost halted in the early stages due to the severity of these reactions. However, the instigation of a premedication regimen consisting of corticosteroids and H1/H2 blockers has significantly reduced the incidence of allergic reactions from 20 per cent to 3 per cent. Although the formulation of paclitaxel in Cremophor EL® (due to its limited aqueous solubility) may be responsible for the hypersensitivity phenomena, some contribution of paclitaxel itself is possible, as hypersensitivity reactions also occur with the semisynthetic analogue docetaxel formulated in Tween-80. Leukopenia is the dose-limiting toxicity, and thrombocytopenia is rare. Myelosuppression occurs early and is associated with a rapid recovery, allowing treatment to proceed on a 3-week cycle. Neurotoxicity occurs with higher cumulative doses of paclitaxel, and with higher doses per course. This is mainly a sensory neuropathy and manifests in a ‘glove and stocking’-type distribution. It is at least partially reversible with time. Many cardiac arrhythmias have been reported with paclitaxel administration, ranging from asymptomatic bradycardia to various degrees of heart block and atrial/ventricular tachyarrhythmias. Other frequently reported toxicities include alopecia (universal), fatigue and arthralgia/myalgia. Indications for use
include breast cancer (anthracycline-pretreated), ovarian cancer, and non-small-cell lung cancer. Docetaxel (Taxotere) is another member of the taxoid family. There is pre-clinical evidence to suggest that docetaxel may be superior to paclitaxel. Docetaxel has been shown to be more potent (up to five-fold) in vitro than paclitaxel with regard to the promotion of tubulin polymerization and inhibition of depolymerization. In a direct comparison with paclitaxel in a large number of freshly explanted tumours (including breast, ovarian, lung and colorectal tumours), docetaxel was found to have at least equivalent cytotoxicity, but with incomplete crossresistance. Docetaxel also had a longer residence time, accumulated at higher concentrations within cells and demonstrated a superior therapeutic index in vivo. Clinical studies have confirmed the presence of incomplete crossresistance with paclitaxel, reporting activity for docetaxel in paclitaxel-resistant metastatic breast cancer. Docetaxel pharmacokinetics following a 1-hour infusion demonstrate that the plasma elimination at the highest doses follows a triphasic decay, with a terminal half-life of 13.5 hours and a plasma clearance of approximately 21 L/h per m2. The AUC increases linearly with the dose and correlates with the percentage decrease of neutrophils. Less than 10 per cent of the administered dose is excreted unchanged in the urine. The recommended dose as a single agent is 100 mg/m2 given as a 1-hour intravenous infusion. Patients that have been heavily pretreated or have abnormal hepatic function should receive a reduced dose. The most frequent toxicity is an early, reversible, noncumulative neutropenia, which may be associated with concurrent infections. Anaemia has also been observed. Hypersensitivity reactions occurred in 25 per cent of patients who were unpremedicated with corticosteroids, and these generally occurred on the first course, within a few minutes of the start of the infusion. Premedication with corticosteroids has reduced the incidence of hypersensitivity reactions to less than 5 per cent. Cutaneous reactions occur, characterized by a rash, mainly on the extremities (but also localized eruptions on arms, face or thorax), and are occasionally associated with pruritus. These are usually transient and resolve prior to the next course of treatment. Less frequently, desquamation is observed. Nail changes (onycholysis, hypo/hyperpigmentation) are also seen. A fluid-retention syndrome, characterized by asymptomatic weight gain and/or oedema (and, less often, pleural, peritoneal or pericardial effusions), has been observed. This generally starts in the lower extremities and at cumulative doses of 400 mg/m2. The pathogenesis of this condition is incompletely understood, but appears reversible on cessation of docetaxel treatment and can be significantly reduced by premedication with dexamethasone. Mild degrees of this syndrome can be treated adequately by diuretic therapy, but it is recommended that patients with severe oedema should not continue with docetaxel. Clinically significant cardiac events occurred in less than 2 per cent of all patients treated, and are not clearly
Classification of cytotoxic drugs, mode of action, toxicities and clinical utility 89
related to docetaxel administration. Other significant toxicities encountered are alopecia, hepatotoxicity (mild, reversible transaminase elevations), mucositis, diarrhoea, arthralgia/myalgia and peripheral neuropathy (sensory and motor). Emesis is generally mild and easily treated with 5HT3 antagonists.
Topoisomerase I and II inhibitors DNA topoisomerase I and II are enzymes that bind to supercoiled DNA, forming a ‘cleavable complex’ and, through strand breakage, passage and re-ligation, allow a wide variety of essential DNA metabolic reactions, including replication and repair, to take place. These enzymes are functionally related, work together and appear to be essential to maintain cellular viability throughout the cell cycle. Since it was shown in the 1980s that the cleavable complex could be stabilized by known cytotoxic drugs such as doxorubicin, etoposide and camptothecin, resulting in interference with the strand breakage–re-ligation catalytic cycle and subsequent cell death, much research has taken place into the development of agents that exploit this novel nuclear target. Established inhibitors of topoisomerase II include the anthracycline antibiotics (e.g. doxorubicin) and the epipodophyllotoxins (etoposide), which were not developed on the basis of rational drug design against this specific cellular target, and are in fact not ‘pure’ topoisomerase II inhibitors. Specific topoisomerase I inhibitors with antitumour activity include the heterocyclic alkaloid camptothecin and its analogues. Isolated in 1966 from the tree Camptotheca acuminata, early studies with camptothecin demonstrated unpredictable and severe toxicity with limited efficacy. However, the development of water-soluble synthetic/semi-synthetic analogues of camptothecin, such as irinotecan (CPT-11) and topotecan, and the discovery that topoisomerase I levels were higher in some tumours compared to normal tissues have led to renewed interest in the topoisomerases as important targets for anti-cancer agents. Topotecan exerts its cytotoxic effects by stabilizing the covalent DNA–topoisomerase I cleavable complex, thus blocking DNA repair. When DNA replicates in the presence of this complex, double-strand breaks occur and the resulting DNA fragmentation causes cell death. The major side effect of topotecan is myelosuppression, which can be severe, is schedule dependent and may be associated with infectious complications. Thrombocytopenia and anaemia are also common, and non-haematological toxicities are otherwise mild – emesis, diarrhoea, oropharangeal ulceration and alopecia. Significant activity has been observed in refractory ovarian cancer, small-cell lung cancer, breast cancer, lymphoma, and head and neck cancer. It has only modest activity against gastrointestinal tumours, unlike CPT-11 (irinotecan). Combination studies of topotecan with other cytotoxic agents, including
platinum and paclitaxel, demonstrate dose-limiting myelosuppression, generally requiring dose reductions of both agents. CPT-11 was developed in Japan and is used for the treatment of 5-FU-relapsed colorectal cancer. In addition to colorectal cancer, activity has been observed in ovarian cancer, lung cancer (small cell and non-small cell), gastric cancer, oesophageal cancer and cervical cancer. It is converted by hepatic carboxylesterases to a major metabolite, SN-38, which is up to 2000-fold more effective at topoisomerase I inhibition than the parent compound. Like topotecan, the formation of a cleavable complex effects damage to the DNA through double-strand breaks. After intravenous infusion, CPT-11 concentrations decline in a multi-exponential manner, with a mean terminal half-life of 6 hours, whereas the equivalent SN-38 half-life is 10 hours. SN-38 is much more highly protein bound than CPT-11. The main toxicities are myelosuppression, which can be severe, and diarrhoea. CPT-11 can induce both early and late diarrhoea, which appears to be mediated by different mechanisms. Early diarrhoea (within 24 hours) is cholinergic in nature and can be avoided by atropine administration. Late diarrhoea can be prolonged and severe and requires prompt instigation of loperamide as it can quickly lead to dehydration and electrolyte imbalance. ETOPOSIDE
Podophyllin and podophyllotoxin are derived from the mandrake root, and act as mitotic poisons in a similar fashion to vinca alkaloids. Early clinical trials were stopped because of significant toxicities, and researchers looked at analogue development as a way of proceeding. VP-16 (etoposide) is an epipodophyllotoxin analogue that is not only less toxic, but also has a different mechanism of action from the parent compound. It interferes with the re-ligation of topoisomerase II, causing stabilization of DNA–topoisomerase II complexes that are cleavable and result in DNA strand breaks. Another possible contributory mechanism of cytotoxicity is the generation of free radicals. Oral VP-16 bioavailability is approximately 50 per cent (range 17–100 per cent), and approximately 50 per cent of the VP-16 dose is eliminated as unchanged drug or glucuronide within 24 hours of administration. Protein binding is approximately 90 per cent. VP-16 clearance appears to be related to renal clearance, and therefore increased toxicity may be expected in patients with poor renal function. Bone-marrow and gastrointestinal toxicity dominate the side effect profile, but other common side effects include alopecia, nausea, vomiting and diarrhoea. VP-16 is active in small-cell lung cancer, germ-cell tumours, ovarian cancer, choriocarcinoma and haematological cancers. Dosing schedules may involve intravenous administration, but many studies have demonstrated that chronic oral dosing is feasible and safe.
90 Principles of chemotherapy and drug development
Miscellaneous agents AGENTS DERIVED FROM MARINE ORGANISMS
Marine organisms have survived for up to 700 million years through natural selection and the release of potent chemicals that offer protection in a hostile milieu without the need for bioactivation. It follows that there is extraordinary potential among these toxic chemicals for antitumour specificity. Over the past 10–15 years, a wide range of potent chemicals with different cellular targets has been isolated. Many have been evaluated in clinical trials, and some are listed briefly here. Bryostatin 1 is the prototype of a novel class of structurally related compounds isolated from the marine bryozoan Bugula neritina. It is a macrocyclic lactone, which exhibits many biological effects mediated through modulation of protein kinase C, a family of enzymes crucial in cellular signalling pathways controlling proliferation and differentiation. It has demonstrated broad-spectrum preclinical anti-tumour activity, and has also been shown to induce differentiation, enhance the immune response and inhibit the production of members of the matrix metalloproteinase family thought to be essential for angiogenesis. Dolastatin 10 was isolated from a herbivorous mollusc in the Indian Ocean, and acts by halting microtubule assembly. Exceptional potency has been described in preclinical models, and it is rapidly and extensively metabolized, with high protein binding. Ecteinascidin 743 (ET-743; Yondelis) was isolated from a Caribbean tunicate in 1990 and forms covalent adducts with DNA in the minor groove, with sequence specificity. In addition, effects on the microtubular spindle are postulated, and down-regulation of transcription factors. A broad spectrum of pre-clinical anti-cancer activity has been demonstrated, in addition to a lack of cross-resistance with other agents, including anthracyclines, taxoids and platinum. Target organs for toxicity are the liver and bone marrow. This agent has been evaluated extensively in clinical trials, with activity in soft-tissue sarcoma.16* Aplidine is a potent cyclic depsipeptide isolated from a Mediterranean marine tunicate, Aplidium albicans. The mechanism of action is not completely understood, but studies are in progress to identify the potentially unique target involved in its anti-tumour activity.
DRUG RESISTANCE Resistance to chemotherapeutic agents remains one of the predominant reasons for cancer treatment failure. Tumours can be intrinsically resistant, showing no significant response to first-line agents; resistance may also be induced such that, after an initial response to drug therapy, the tumour acquires drug resistance to initial and/or subsequent therapy. A defining feature of all cancer types is the presence of an unstable genome. The high mutation rate of
cancer cells results in the characteristic phenotype of uncontrolled cell growth and enhanced survival seen in tumours. By the time the cancer becomes clinically detectable, the original clone of cells has diversified into a multi-clonal tumour capable of adapting to environmental stresses such as anti-tumour agents. The Goldie and Coldman hypothesis proposes that numerous subpopulations of cells within a tumour are likely to possess mutations that render certain cells resistant to treatment. Although chemotherapy-sensitive clones are destroyed, resistant clones continue to evolve. The rationale behind the use of combination chemotherapy lies in this inherent genetic diversity that the tumour possesses. Mechanisms of drug resistance are multiple and diverse. In any given tumour type, it is usual for combinations of different mechanisms to contribute to the overall resistance pattern seen. The major advances in the field of drug resistance have been in the understanding of the molecular pathways by which tumour cells acquire this resistance. As our knowledge of the intracellular pathways of cell survival, division and cell death broadens, the importance of cell-cycle-specific mechanisms of drug resistance is being increasingly recognized. Ways in which this might occur are varied and complex and by no means completely elucidated. This section of the chapter highlights some important and clinically relevant examples of how drug resistance may occur and current therapeutic advances that are designed to overcome this obstacle (Fig. 5.1).
Cytokinetic drug resistance Various models have been proposed to predict cell kill in the presence of cytotoxic agents, as outlined earlier in the chapter. There exists a broad relationship between the rate of cellular proliferation within a tumour and subsequent response to chemotherapy. The difference between curable tumours such as lymphomas and the common solid tumours is the higher proportion of cycling cells in lymphomas compared with the relatively small percentage (2–3 per cent) of proliferating cells in the latter. The proportion of cycling cells may vary considerably, even within tumours of the same histological type, due to genetic heterogeneity and tumour microenvironment.17 Certain cytotoxic agents are cell-cycle specific, having anti-tumour effects only on cells in cycle. Of this group of drugs, some only have activity in cells within certain stages of the cell cycle, for example 5-FU and the anthracyclines targeting cells in the S phase. It follows that in cells proliferating slowly, a short period of exposure to an S-phase-specific drug will result in a smaller proportion of cell kill than for rapidly proliferating cells. In order to overcome ‘cytokinetic resistance’ to a cellcycle-phase-specific agent, a prolonged dosing schedule can prove more effective than bolus regimens. This enables a higher proportion of cells in the vulnerable phase of their cell cycle to be exposed to the chemotherapeutic agent (Fig. 5.2).
Drug resistance 91
Reduced delivery of cytotoxic agent Tumour angiogenesis
Decreased drug uptake
Increased drug efflux
Decreased drug activation Inhibition of apoptosis
Increased DNA repair
Alteration in target protein
Increased metabolism Alteration in target protein
Figure 5.1 Cellular mechanisms of drug resistance. Original tumour mass
Cells entering S-phase of cell cycle Cells not proliferating
Tumour mass after short exposure to S-phase agent
Tumour mass decreases further as result of prolonged exposure to S-phase agent
Figure 5.2 Prolonged exposure to cell-cycle-phase-specific agents may allow cytokinetic resistance to be overcome.
For example, several randomized trials have shown the superiority in tumour response rates and overall survival in colorectal cancer with the use of a prolonged infusion of 5FU compared with bolus regimens.18***,19**
resistance. There is also the potential to exploit tumour microenvironment to enhance the targeting of cancer therapy. HYPOXIA
Tumour microenvironment Cells in normal tissues exist in a homeostatically controlled environment. A wide variation exists, however, in the physiological microenvironment of tumour cells, even between different regions of a single solid tumour. Such differences in the microenvironment between tumour cells and normal cells, as outlined below, may account for drug
Rapid tumour proliferation results in tumour cells outgrowing their existing blood supply, with resultant lack of nutrients, hypoxia and eventual necrosis. Hypoxia is one of the characteristic features of cancer, and is a cause of resistance to chemotherapy and radiotherapy. A key reason for this effect is that hypoxia reduces the rate of cell proliferation, decreasing the effectiveness of cell-cycle-specific drugs. A second factor is that hypoxic cells are those often
92 Principles of chemotherapy and drug development
furthest away from an adequate blood supply and therefore exposed to a lower concentration of the drug. Tumour hypoxia may be used to direct therapy towards cancer cells while sparing normal, aerobic tissues. Drugs that undergo reductive activation under hypoxic conditions can be used as pro-drugs of cytotoxic agents. Once activated, these agents act as cytotoxins of tumour cells by direct or indirect (bystander) mechanisms, and are specifically targeting those hypoxic cells that would normally be resistant to traditional cytotoxins. A clinical example of how this can be applied is the use of tirapazamine, a benzotriazine-di-N-oxide, which is selectively activated under hypoxic conditions to DNA-damaging reactive radicals that synergize with cisplatin and radiation. A phase III trial has shown a survival benefit of patients with advanced non-small-cell lung cancer treated with tirapazamine and cisplatin compared with cisplatin alone,20** and comparison of tirapazamine and cisplatin with cisplatin and etoposide in a second phase III trial21** has shown similar response rates between the two groups. Several other hypoxically activated drugs are under investigation, including AQ4N, an anthraquinone-di-N-oxide, that is activated intra-tumorally by CYP3A, a member of the cytochrome P450 family of enzymes, to the topoisomerase II inhibitor AQ4. Tumour expression of CYP3A is thought to be upregulated under hypoxic conditions22 and may enhance the selectivity of AQ4N. Clinical studies of AQ4N are currently underway.
therapy.23 In combination with traditional cytotoxic agents, anti-angiogenic agents have been shown to behave synergistically. For example, the use of bevacizumab (Avastin) in combination with standard platinum-based chemotherapy for non-small-cell lung cancer has proved effective in enhancing tumour response to chemotherapy and resulted in a statistically and clinically significant survival advantage in a phase III evaluation.24**
Control of intracellular drug concentrations Cell membranes are biological barriers composed of proteins embedded in a cholesterol and phospholipid bimolecular matrix, which selectively inhibit the passage of drug molecules. Drugs may cross this biological barrier by passive diffusion, facilitated passive diffusion, active transport or pinocytosis (Fig. 5.3). Pinocytosis probably plays a minor role in drug transport, except for protein drugs. It follows that any alterations to the structure or function of
ANGIOGENESIS
Once tumours have outgrown their existing blood supply, in order to continue to grow they must develop their own blood supply by means of neovascularization or angiogenesis. Angiogenesis is stimulated by hypoxia and tumour secretion of angiogenic factors. The resultant tumour ‘neovasculature’ possesses characteristic structural and physiological differences to the normal cellular micro-environement. Vessels are generally more tortuous, with increased vascular permeability. The process of neo-angiogenesis can result in drug resistance in a number of ways. Whereas traditional cytotoxic agents will be effective against sensitive clones of tumour cells, drug-resistant tumour populations supported by a new blood supply will continue to grow. Increased vascular permeability accompanied by reduced lymphatic clearance within the tumour environment results in an increased interstitial pressure and hence reduced delivery of cytotoxic therapy to the tumour. The mechanisms and implications of angiogenesis are covered in detail in a separate chapter (Chapter 12). The use of anti-angiogenic agents directed against vascular endothelial cells can result in regression of tumour blood vessels, blood supply and tumour size. The relative genetic stability of endothelial cells compared with tumour cells renders them far less prone to the development of drug resistance with repeated courses of anti-angiogenic
ATP
A
B
C
Cell membrane
Carrier component of membrane
Drug molecule
Transport protein
Figure 5.3 (A) Passive diffusion: transport across a cell membrane depends primarily on the concentration gradient of the solute, but also on the molecule’s lipid solubility, degree of ionization and size, and on the area of the absorptive surface. (B) Facilitated passive diffusion: a carrier component is thought to combine reversibly at the cell membrane exterior with a selective substrate molecule of a relatively specific molecular configuration. This results in rapid diffusion of the carrier–substrate complex across the membrane, releasing the substrate at the interior surface. The process does not require energy expenditure, and transport against a concentration gradient does not occur. (C) Active transport involves membrane transport proteins, of which the ATP-binding cassette family is the most extensively studied. Active transport is characterized by selectivity and saturability, requires energy expenditure by the cell, and may result in accumulation of substrates intracellularly against a concentration gradient.
Drug resistance 93
proteins involved in drug transport will affect the amount of the drug reaching its intracellular target, and hence lead to increased drug resistance. DECREASED DRUG UPTAKE
Folates are critical in DNA and RNA synthesis by acting as co-factors in a number of reactions, including the synthesis of purines and thymidylate de novo. Drugs targeting the folate-dependent enzymes, such as the dihydrofolate reductase inhibitor methotrexate and the thymidylate synthetase inhibitor 5-FU, are widely used as cytotoxics in the treatment of cancer. The passage of anti-folate agents across the cell membrane occurs predominantly via the widely expressed reduced folate carrier (RFC) system. In cell lines in which a defective RFC protein is present, 250fold higher concentrations of methotrexate are required for equivalent cell kill as compared with cell lines in which the transport protein has retained normal function. In the clinical setting, decreased transport of methotrexate across the cell and resistance to methotrexate were demonstrated in children with acute lymphoblastic leukaemia (ALL), in whom the presence of a particular polymorphism in the RFC gene (RFC1A80) resulted in higher levels of methotrexate in the plasma and a shorter event-free survival.25 One strategy to overcome this form of drug resistance is the use of anti-folate agents selectively targeted to folate transport proteins other than the RFC, such as the α-folate receptor (α-FR). The α-FR is often over-expressed in ovarian and other epithelial malignancies.26 Conventional antifolate agents such as methotrexate have a low affinity for the α-FR. Novel anti-folate inhibitors of thymidylate synthetase have been synthesized with low affinity for the RFC and high affinity for the α-FR. For example, BGC 945 has been studied in vitro in α-FR over-expressing tumour cell lines and has been identified as a potential candidate for phase I studies.27 FPGS (folylpoly-γ-glutamate synthetase) catalyses the attachment of glutamate residues to all naturally occurring folates and folate analogues such as methotrexate. Absence, decreased expression and mutations resulting in decreased activity of FPGS may all result in impaired polyglutamylation and the inability of cells to retain folates and their analogues after cell-membrane transport.28 This may lead to an intrinsic resistance to anti-folates such as methotrexate. Evidence of defective polyglutamylation resulting in resistance has been seen in rodent tumour cell lines and in human leukaemia, breast and other tumour types.29–31 In ‘good-risk’ subgroups of children with ALL, the ability of blast cells to accumulate methotrexate and form long-chain methotrexate polyglutamates is an important determinant of outcome, with the accumulation of higher levels of methotrexate and its polyglutamates correlating with a more favourable event-free survival outcome.32 Defective polyglutamylation as a mechanism of resistance can be overcome by the development of newer anti-folates that may not require polyglutamylation for
retention and efficacy. For example, trimetrexate, an analogue of methotrexate, is retained in cells in high concentrations despite its inability to be polyglutamylated, although limited activity as a single agent in phase II studies in cancer and dose-limiting toxicities have restricted its clinical use.33 ZD9331, which does not require polyglutamylation for its activity, has been the subject of numerous phase I and II studies, as monotherapy and in combination, and promising activity has been seen in platinum-refractory relapsed ovarian, pancreatic and gastric cancers.34 INCREASED DRUG EFFLUX AND MULTI-DRUG RESISTANCE
Tumour cells exposed to one chemotherapeutic agent may often demonstrate resistance to other structurally unrelated agents with differing mechanisms of action, a phenomenon known as multi-drug resistance (MDR). One proposed mechanism to account for this is the presence of transport proteins that actively pump drugs out of cells. This is of particular relevance to hydrophobic agents such as the taxanes, anthracyclines and vinca alkaloids that enter cells by means of passive diffusion. One such protein is a member of the ATP-binding cassette family of membrane transporters, P-glycoprotein (P-gp), which is expressed at high levels in several normal tissues and over-expressed in certain solid tumours and haematological malignancies. P-gp knockout mice have increased sensitivity to chemotherapeutic agents, as drug elimination is decreased and cellular levels of the cytotoxic agent are increased.35 Alterations in drug efflux secondary to expression of P-gp have also been demonstrated in vitro. In clinical practice, data exist that would support a role for P-gp in mediating MDR. In soft-tissue sarcoma of childhood, detectable levels of P-gp appear to be an important adverse prognostic factor. The probability of a longer relapse-free and overall survival was significantly greater in chemotherapy-treated patients with P-gp-negative tumours than in those whose tumours contained detectable levels of P-gp.36 In acute myeloid leukaemia or breast cancer, patients with P-gp activity at presentation were significantly more likely to have chemotherapyresistant disease.37***,38 Clinical trials of agents that inhibit proteins such as P-gp (cyclosporine, verapamil) have been disappointing, and these agents are not routinely employed in clinical practice. Their contribution to clarifying the clinical relevance of P-gp and other proteins potentially implicated in MDR, such as the MDR-associated protein (MRP) family, thus remains limited. Phase II trials of more specific third-generation P-gp inhibitors such as Tariquidar in conjunction with chemotherapy are ongoing.39* If P-gp plays a direct role in mediating MDR, one would expect to see improved, prolonged or even restored drug sensitivity in these patients as a result of effective inhib-ition of P-gp.
94 Principles of chemotherapy and drug development
Drug metabolism DECREASED DRUG ACTIVATION
Many anti-cancer agents are pro-drugs, agents that require metabolic activation by key enzymes prior to exerting cytotoxic effects. Consequently, any structural or functional abnormalities in these key enzymes could result in resistance to the effects of the drug, as a result of a decrease in the level of enzymatic activity and consequent decrease in the amount of active drug metabolite within the cell. Examples of this method of drug resistance include the absence or reduction in expression of salvage enzymes in the activation of the pyrimidine and purine anti-metabolites cytosine arabinoside and 6-mercaptopurine respectively. Ara-C requires activation by deoxycytidine kinase to ara-CTP, a competitive inhibitor of DNA polymerase-α. The rate of formation of ara-CTP is a critical factor in the response of acute leukaemia to ara-C therapy. Similarly, 6-MP and 6-TG require activation to their monophosphates by hypoxanthine-guanine phosphoribosyl transferase (HGPRT) before they are able to exert anti-tumour activity (inhibition of synthesis of purine nucleosides de novo). Reduced expression of, or decreased affinity for, 6MP or 6-TG by HGPRT results in drug resistance. The enzyme carboxylesterase is responsible for the conversion of irinotecan to its active metabolite SN-38, and reduced tumour expression of the enzyme may be responsible for inducing resistance. Strategies for overcoming resistance to irinotecan include gene therapy to transfer the carboxylesterase gene directly into tumour tissue prior to the administration of irinotecan. This approach has been used in cell lines and mouse models with some success.40,41
INCREASED DRUG METABOLISM OR DETOXIFICATION
Pathways of drug metabolism are traditionally classified as either phase I reactions (e.g. oxidation, reduction and hydrolysis) or phase II, conjugation reactions (e.g. acetylation, glucuronidation, sulphation and methylation). Most drugs are metabolized by several different enzymes. Enzymes of the cytochrome P450 super-family of drugmetabolizing enzymes are the most important enzymes that catalyze phase I drug metabolism. Agents responsible for the induction of these enzymes can result in striking variations in the pharmacokinetics and physiological effects of individual chemotherapeutic drugs. For example, the use of the anticonvulsant phenytoin in patients with central nervous system tumours resulted in increased clearance of irinotecan due to the induction of hepatic cytochrome P450 enzymes.42* Similarly, polymorphisms in the CYP enzymes can result in resistance to cytotoxic agents. Inhibition of CYP1B1 has recently been the focus of targeted drug therapy because this member of the CYP family is over-expressed in ovarian, breast and testicular tumours but not in normal tissues.43 CYP1B1 is involved
in the metabolism of various anti-cancer agents, resulting in reduced efficacy. Inhibition of this enzyme is therefore an attractive proposition for attempting to reverse drug resistance.44 Rapid drug detoxification within a cell can result in a decreased amount of active product available to bind to its intracellular target. Glutathione, an anti-oxidant, binds to highly reactive chemical groups generated by anthracyclines and platinum agents, and promotes drug efflux through ABC transporter proteins in the cell membrane. In tumour cell lines, the level of glutathione itself, as well as levels of other enzymes involved in the process of glutathione conjugation, correlates with the degree of resistance observed to platinum agents.45–47
Altered expression of target proteins Many chemotherapeutic agents ultimately exert their effect by interacting with one or more key target proteins. Mutations in target proteins may result in the inability of the drug to bind to its target and exert its effect; they may also result in a lower affinity of binding. This is exemplified by the taxanes, which exert their effects on the cellular microtubule apparatus during mitosis. Microtubule depolymerization to tubulin monomers is an important step in the restructuring of the microtubule apparatus for mitosis. The taxanes bind to and stabilize the protein tubulin, thus blocking the process of mitosis and eventually triggering cell death via apoptosis. Resistance to taxanes is seen in cell lines expressing mutations in β-tubulin that interfere with tubulin stabilization (48). Of clinical relevance, in a study sequencing β-tubulin genes in non-small-cell lung carcinoma patients treated with singleagent paclitaxel, the presence of β-tubulin mutations was found to be a predictor of paclitaxel resistance.49 If a drug exerts its effect by inhibition of a target protein, over-expression of the target proteins may confer resistance, simply due to the greater amount of target requiring inhibition for the same concentration of active drug. For example, the anti-folate agent 5-FU inhibits the enzyme TS. A high level of TS has been associated with a significantly poorer survival in Duke’s stage A–D colorectal cancers50 and appears to be a powerful prognostic indicator independent of disease stage.51 In-vitro studies have demonstrated that high levels of TS expression are associated with resistance to 5-FU.52 In the clinical setting, a statistically significant association has been demonstrated between the degree of TS expression in colorectal cancer metastases and response to infusional 5-FU,53 although the response appears to vary according to the particular schedule of 5-FU used. Specific TS inhibitors currently in development, either alone or in combination with traditional anti-folate agents, may prove effective in enhancing response to treatment in tumours with high levels of expression of TS. Agents such as irinotecan exert their cytotoxic effects by inhibiting the enzyme topoisomerase I involved in the
Drug resistance 95
relaxation and re-ligation of supercoiled DNA during DNA replication. This results in S-phase-specific cell kill. Human cancer cell lines that are resistant to irinotecan have been shown to exhibit increased levels of topoisomerase I, topoisomerase I mutations, or down-regulation and reduced activity of the enzyme. The true relevance of these changes to the mechanism of irinotecan resistance remains unclear due to the lack of well-designed prospective clinical studies. A combination chemotherapeutic approach has been shown to be of some benefit in overcoming resistance to irinotecan. Mitomycin C has been shown to increase topoisomerase I catalytic activity and, not surprisingly, has synergistic effects with irinotecan in pre-clinical studies. Sequential administration of this combination chemotherapy has shown promising activity in phase I studies in heavily pre-treated gastric, oesophageal, breast and non-small-cell lung cancer patients, with major responders showing a greater induction of topoisomerase I when compared with non-responders.54 Phase II studies continue to evaluate this combination regime.55* Alterations in target proteins conferring resistance are also demonstrated by the epidermal growth factor receptor (EGFR) family of transmembrane receptors, which include EGFR, HER-2, HER-3 and HER-4. The receptors exist as monomers spanning the plasma membrane, and require dimerization in response to the appropriate ligand in order to become functional. They play a critical role in tumour proliferation and survival by virtue of possessing tyrosine kinase activity. Abnormal signal transduction or ligandindependent activation of the EGFR receptor, often caused by mutation in the receptor extracellular domain, is thought to correlate with increased resistance to conventional chemotherapy and a poorer prognosis.56,57 The use of agents that interfere with EGFR signalling, such as the HER-2 inhibitor traztuzumab, and the EGFR tyrosine kinase inhibitors gefitinib and erlotinib, has provided a method to overcome resistance to cytotoxic therapy. The EGFR inhibitors exert their effect by binding the ATP pocket within the catalytic domain of the receptor. Traztuzumab has shown synergism with cytotoxic agents such as the anthracyclines. The co-localization of the topoisomerase II-α gene (the target of the anthracyclines) and the HER-2 gene on the same chromosomal region and potential co-amplification in tumours might be one explanation for this synergism.58 Similarly, the EGFR inhibitor cetuximab demonstrates synergism with irinotecan in colorectal carcinomas.59**
Tumour cell apoptosis A large number of chemotherapeutic agents act by the induction of programmed cell death, or apoptosis. A variety of enzymes, namely the caspases (intracellular cysteine proteases), mediate apoptosis in a cascade that leads to the systematic de-structuring and death of the cell. There are two main apoptotic pathways that activate the effector
caspases: an extrinsic apoptotic pathway involving the triggering of cell-surface death receptors, including tumour necrosis factor-α (TNF-α ) receptor and CD-95 (Fas), and an intrinsic pathway involving increased mitochondrial permeability and disruption of function. There is extensive cross-talk between the pathways. Two families of proteins are involved in apoptotic regulation. The Bcl-2 family of proteins act at the mitochondrial level and include promoters of apoptosis such as Bax and Bak, and inhibitors of apoptosis such as Bcl-2 and BclXL. A second family comprise the IAP (inhibitors of apoptosis proteins), survivin being one such example, which act as endogenous caspase inhibitors. The relative proportion of these proteins within a cell determines whether apoptosis is triggered in that individual cell. In cells undergoing DNA damage, the p53 family of proteins play a key role in triggering apoptosis, partly by regulating levels of Bax protein, as is seen in DNA damage occurring secondary to platinum drugs such as cisplatin. Drugs such as the taxanes that do not result in direct DNA damage are also able to trigger apoptosis by inducing phosphorylation of Bcl-2, reducing its binding and thus resulting in a higher proportion of pro-apoptotic proteins within the cell. It follows that accumulation of mutations resulting in the absence or under-expression of pro-apoptotic proteins, or over-expression of anti-apoptotic proteins, will not only result in tumorigenesis, but can also play a part in clinical drug resistance as a result of reduced apoptosis and cell death in response to drug-induced cellular damage. Invitro studies have shown Bcl-2 over-expression increases chemo-resistance, whereas Bax over-expression increases chemosensitivity.60,61 Ways of manipulating the process of apoptosis in order to overcome drug resistance have included the development of small-molecule inhibitors. For example, ABT-737, an inhibitor of the anti-apoptotic proteins Bcl-2, Bcl-XL and Bcl-w, has shown promise in animal models, cell lines and primary patient-derived cells.62 The use of techniques such as gene therapy in trying to restore p53 expression and of antisense oligonucleotides in attempting to reduce Bcl-2 expression, survivin expression or any specific gene of interest has also been investigated. Although some promising results have been seen in early trials, these techniques need continued development and refinement. More recently, attention has been paid to the importance of the protein ubiquitin in targeting cellular proteins for proteosomal degradation. Ubiquitin is a 76-aminoacid protein that, in conjunction with a three-enzyme ubiquitination complex, forms a polyubiquitin chain on a target protein. This in turn targets the protein for degradation via the 26S proteasome.63 With the knowledge that a number of tumour suppressor genes and oncogenes are integral to or degraded via the ubiquitin–proteasome pathway comes the realization that disruptions to this pathway may lead to alterations in the balance between proteins signalling cell death and those signalling survival. Examples include von Hippel Lindau (VHL) gene mutations in renal
96 Principles of chemotherapy and drug development
cell carcinoma resulting in altered function of the VHL ubiquitin–ligase complex. This results in reduced degradation of hypoxia-inducible factors and increased angiogenesis.63 Mutations in the adenomatous polyposis coli (APC) gene in colorectal carcinoma leads to reduced phosphorylation, ubiquitination and degradation of β-catenin, which is a transcription factor for genes such as cyclin D1, and MYC.63 Similarly, increased proteosomal degradation of the inhibitor I-κB results in increased levels of the transcription factor NF-κB and hence increased expression of the anti-apoptotic protein Bcl-2. The end result in all these cases is the inhibition of apoptosis and the potential for resistance to chemotherapeutic agents that are inherently relying on apoptosis for therapeutic effect. Inhibition of the 26S proteosome may be a mechanism to overcome drug resistance by allowing the process of DNA damage and apoptosis initiated by other chemotherapeutic agents to proceed. A number of agents have been identified and have been shown to overcome drug resistance in vitro in myeloma, prostate and ovarian cancer cells.64,65 One such agent is bortezomib, which has now been approved for use in the treatment of relapsed multiple myeloma. The effects of bortezomib appear to be related to reduced proteasomal degradation of the inhibitor I-κB, with subsequent decreased levels of the transcription factor NF-κB and hence reduced expression of the anti-apoptotic protein Bcl-2,66–68 as illustrated in Figure 5.4. There are initial data supporting a role for bortezomib in sensitizing cells to chemotherapy-induced apoptosis in solid tumours such as non-small-cell lung carcinoma.69
3) I-κB stabilises NF-κB and prevents nuclear translocation and transcription
1) Unbound NF-κB is a nuclear transcription factor
BORTEZOMIB can also reduce efficiency of DNA repair mechanisms following DNA damage
Enhanced DNA repair The majority of chemotherapeutic agents act by causing either direct or indirect damage to cellular DNA. There are cellular mechanisms in place for detecting DNA damage and initiating one of a number of potential responses, including apoptosis, modulation of cell-cycle progression via DNA damage checkpoints, and initiation of DNA repair. DNA damage can occur in many different ways, and multiple DNA-repair pathways exist involving a series of enzymatic steps. If chemotherapeutic agents act by causing DNA damage, an important mechanism of drug resistance is the efficiency of DNA-repair mechanisms within a cell. Genetic mutations within the DNA of repair genes can alter the efficiency of repair. Recently, increasing attention is being paid to the role of single nucleotide polymorphisms (SNPs) of DNA repair genes in predicting prognosis and response to chemotherapy. The SNPs are single base differences in the DNA between individuals, and when present in certain genes, such as those involved in DNA-repair cascades, they may alter the efficiency of those genes. Novel therapeutic agents targeting DNA-repair mechanisms may be effective treatment strategies to overcome resistance. Cisplatin induces cellular damage primarily by the formation of bulky intra-strand platinum–DNA adducts. Removal of these adducts is mediated by the nucleotide excision repair (NER) pathway, a crucial element of which is the excision cross-complementing 1 (ERCC1) gene. It has previously been shown that over-expression of the ERCC1 gene is associated with increased removal of platinum–DNA adducts
4) I-κB targeted for degradation BORTEZOMIB inhibits 26S proteasome
26S proteasome
2) Expression of anti-apoptotic proteins such as Bcl-2 and suppression of apoptotic cascade induced by chemotherapy
Figure 5.4 Proposed mechanism of action of NF-κB and its inhibitor I-κB in inhibiting apoptosis. Bortezomib inhibits this process by inhibiting the 26S proteasome. Bortezomib may also have some direct action on altering DNA-repair mechanisms within the damaged cell.
Novel anti-cancer therapies 97
and relative platinum resistance, as seen in human ovarian and gastric tumours.70,71 There are known SNPs of the ERCC1 gene conferring differences in the levels of ERCC1 mRNA synthesis, and stability of the ERCC1 protein. Polymorphism C8092A is shown to affect overall survival negatively in patients with non-small-cell lung cancer previously treated with platinum-based chemotherapy (13.4 months median survival versus 22.3 months, p 0.006).72 Although predicting survival, it remains to be determined whether these polymorphisms directly impact on response to platinum-based chemotherapy. The ability of tumour cells to utilize an alternative repair pathway where the normal pathway of repair is deficient can account for some of the resistance patterns observed with chemotherapeutic regimes. However, if the cells were to be prevented from utilizing the alternative DNA-repair pathway, by targeted inhibition, this might prove an effective way of overcoming drug resistance. PARP (poly [ADP-ribose] polymerase) is involved in the base excision repair (BER) pathway for repair of singlestrand DNA breaks. Where PARP is deficient or lost, it is thought that accumulating defects in DNA that would normally be repaired by BER are alternatively repaired by homologous recombination (HR). As BRCA1 and BRCA2 are essential components of DNA repair by HR, inhibition of PARP in carriers of BRCA1 and BRCA 2 mutations, i.e. in cells already deficient in the HR repair pathway, has been shown to result in continued accumulation of DNA defects, chromosomal instability, cell-cycle arrest and apoptosis.73
Future directions in drug resistance The Human Genome Project and the resultant progress in the field of gene expression profiling of human tumours have opened the door to the concept of individualized cancer therapy. Traditional approaches have relied on grouping tumours into histopathological categories to make treatment decisions, usually resulting in patients receiving a standard combination chemotherapy regimen. However, it is increasingly obvious that two patients with the same pathological grouping can have remarkably different clinical outcomes to standard treatment. This is likely to be determined by the individual gene expression profile of that particular tumour. It is hoped that gene profiling will allow for a more individualized molecular approach to tumour staging, determination of molecular predictors of outcome and predicting response/resistance patterns to therapies. Similarly, with an increase in the understanding of the complex and varied mechanisms by which a tumour cell can become resistant to chemotherapeutic agents will come newer, more targeted and less toxic alternatives to traditional chemotherapeutic agents. Some advances in these areas have been highlighted in this chapter. In order to translate gene expression signatures of individual tumours into reliable clinical benefit we will need to evaluate prospectively, using
validated methods, in randomized clinical trials. Only then will we be able to progress to a stage at which clinicians can reliably identify a group of patients for whom treatment can be tailored for optimum efficacy and with the prospect of avoiding drug resistance.
NOVEL ANTI-CANCER THERAPIES The development of the current generation of novel anticancer therapies is based on exploiting our increasing understanding of the molecular basis of cancer. The identification of putative molecular targets, allied to high through-put screening strategies of potential lead compounds, and utilizing the advances in biotechnology (particularly in the field of drug delivery) give rise to the exciting possibility of a number of novel therapies, many of which are already undergoing clinical evaluation. However, the preclinical and early clinical evaluation of these novel therapeutic strategies targeted to specific molecular pathways presents new challenges that will require an integrated approach from both laboratory and clinical scientists. Preclinical evaluation will require the demonstration of reproducible biological effects in experimental systems at concentrations of drug comparable to those that can be achieved in the clinic. In addition to the conventional endpoints of toxicity and pharmacokinetics, early clinical evaluation will also require the demonstration of desired biological activity, which is likely to be particularly pertinent with those agents that are likely to have a cytostatic effect, in order to determine the optimal biologically active dose for subsequent clinical trials. Furthermore, these agents may not have objective evidence of anti-tumour activity by classical tumour response criteria in patients with bulk disease, which usually make up the patient population in whom new agents are evaluated. Consequently, evaluation of these agents will require identification of appropriate candidate patients (e.g. the presence of molecular target in biopsy material), demonstration of desired biological effect (usually in tumour biopsy material or by assessment of surrogate biological endpoints) and identification of the appropriate clinical scenario for evaluation (e.g. as maintenance therapy after ‘debulking’ chemotherapy, or as adjuvant therapy for agents with a proposed cytostatic effect).
Signal transduction inhibitors The processes of normal cell growth, proliferation, differentiation and death are controlled by signals that balance their activation and inhibition. Disruption of normal cellular signalling enables malignant cells to proliferate and/or survive when normal cells would not. The process of signal transduction typically involves ligand binding to, and activation of, a specific receptor. This, in turn, initiates a cascade of enzymatic and biochemical reactions that
98 Principles of chemotherapy and drug development
allow proliferation signals to be transmitted from the cell surface, through the cytoplasm, to the nucleus.74,75 Several signal transduction pathways have been proposed as potential targets for cancer therapy, and among the most prominent examples are the tyrosine kinase inhibitors. INHIBITORS OF TYROSINE KINASE
Tyrosine kinases are a family of enzymes that catalyse the phosphorylation of the phenolic moiety of tyrosine residues.76 Abnormal constitutive ‘activation’ of this group of signalling proteins has been implicated in malignant growth and progression.77 Common to the structure of all tyrosine kinases are a substrate-binding domain, an ATP-binding domain and a catalytic or kinase domain. Furthermore, there are two classes of tyrosine kinase – receptor (RTK) and nonreceptor. The RTKs are further organized into three distinct sections: an extracellular ligand-binding region, a hydrophobic transmembrane portion, and a cytoplasmic domain containing the active tyrosine kinase site.78 The various types of RTK include the EGFR, platelet-derived growth factor receptor (PDGFR), vascular endothelial growth factor receptor (VEGFR) and the fibroblast growth factor receptor (FGFR). The mechanisms by which RTKs and non-RTKs activate downstream signalling pathways have been extensively reviewed.79 Several strategies have been proposed to inhibit tyrosine kinases, including inhibiting receptor–ligand interactions, inhibiting the tyrosine kinase domain of RTKs, inhibiting non-receptor tyrosine kinases, and antisense oligonucleodies against RTK mRNA. The most striking example of the potential use of these approaches in clinical practice is the use of imatinib in chronic myeloid leukaemia (CML). The Philadelphia chromosome is the result of a t(9;22) reciprocal translocation; it is present in more than 90 per cent of patients with CML and results in the juxtaposition of DNA sequences from the BCR and ABL genes. BCR-ABL encodes a protein, p210BCR-ABL, with dysregulated tyrosine kinase activity, which is necessary and sufficient for leukaemogenesis. Imatinib mesylate (Glivec) is a potent competitive inhibitor of the tyrosine kinases associated with ABL and impedes the interaction of ATP with these proteins and thereby inhibits their ability to phosphorylate and activate proteins downstream. Initial phase I and phase II studies with imatinib in CML showed that it was well tolerated, and 60 per cent of the patients had a major cytogenetic response (defined by the finding that no more than 35 per cent of cells in metaphase were Ph-positive), with 41 per cent having a complete cytogenetic response and 95 per cent of the patients still alive after a median follow-up of 18 months.83**,84** Imatinib was subsequently compared with interferon-alpha combined with lowdose cytarabine in newly diagnosed chronic-phase CML [n 1106].85** After a median follow-up of 19 months, the estimated rate of a major cytogenetic response at 18 months was 87 per cent in the imatinib group, compared with 35 per cent in the interferon-α plus cytarabine group
(p 0.001). The estimated rates of complete cytogenetic response were 76 per cent and 15 per cent respectively (p 0.001). Furthermore, imatinib was also better tolerated than combination therapy, and had a significantly superior progression-free rate (p 0.001) than combination therapy.85** Imatinib mesylate is also active in inhibiting other tyrosine kinases, including the transmembrane receptor KIT. Gastrointestinal stromal tumours are mesenchymal neoplasms that express the cell-surface transmembrane receptor KIT, which is the product of the KIT proto-oncogene. These tumours have frequent gain-of-function mutations of KIT, and KIT activation occurs in almost all cases of gastrointestinal stromal tumour, regardless of the mutational status of KIT.86,87 Gastrointestinal stromal tumour is a universally fatal disease that is resistant to most conventional cytotoxic chemotherapy agents, with a response rate of less than 5 per cent with single-agent doxorubicin. Imatinib induced a sustained objective response in 53 per cent of patients in a multi-centre trial [n 147].88* Interestingly, early resistance to imatinib was noted in 13 per cent of patients, and this is likely to be the next major challenge in this disease. The role of imatinib in the management of gastrointestinal stromal tumours is discussed more fully in Chapter 40.
MONOCLONAL ANTIBODIES
One example of an approach to inhibit receptor–ligand interactions of RTKs is to use monoclonal antibodies against the RTK. The most prominent example of this approach is trastuzumab (herceptin), a highly purified, recombinant, DNA-derived, humanized monoclonal antibody that binds with high affinity and specificity to the extracellular domain of the erbB-2 receptor. Amplification of erbB-2, which occurs in approximately 30 per cent of breast cancers, is associated with poor prognosis and has been linked to resistance to both tamoxifen and alkylating agent chemotherapy drugs.89 Although the precise function of erbB-2 is unclear, its over-expression is associated with constitutive tyrosine kinase activity and may play a role in unregulated cell growth. Trastuzumab can downregulate HER-2 and angiogenic proteins, such as VEGF, and induce antibody-dependent cellular cytotoxicity. The rates of response to trastuzumab given as a single agent range from 12 per cent to 40 per cent, depending in part on the method used to determine the HER-2 status and the prior treatment received.80*,91,92* Furthermore, the combination of chemotherapy with trastuzumab produces higher response rates and increased survival compared with chemotherapy alone in patients with advanced breast cancer.93** More recently, it has been demonstrated that the addition of trastuzumab to standard adjuvant chemotherapy regimens for patients with HER-2-positive early breast cancer can improve disease-free survival, and can reduce the risk of death by 33 per cent.94**,95**
Novel anti-cancer therapies 99
Antibodies have also been developed to target other signal transduction pathways, including bevacizumab and cetuximab. Bevacizumab is a humanized antibody that targets VEGFR, one of the most important pro-angiogenic growth factors, and has proven efficacy in phase III studies in advanced colon cancer and also promising activity in renal cancer, non-small-cell lung cancer and a range of other tumour types.96 This agent is discussed more fully in the chapter on angiogenesis (Chapter 12). Similarly, cetuximab is one of the agents that targets the EGFR pathway.97 KINASE INHIBITORS (e.g. EGFR TYROSINE KINASE INHIBITORS)
A further strategy for inhibiting RTKs is to target the kinase domain directly, either by competitive substrate inhibitors or by competitive inhibition of the ATP-binding site. Several inhibitors of the RTK kinase domain have entered clinical trials, including agents that target the EGFR signalling pathway. EGFR is a transmembrane receptor tyrosine kinase of the Erb (also known as HER) family that is abnormally activated in many epithelial tumours. Several mechanisms lead to aberrant receptor activation, including receptor over-expression, gene amplification, activating mutations, over-expression of receptor ligands and/or loss of their regulatory mechanisms. Receptor activation leads to recruitment and phosphorylation of several intracellular substrates, which, in turn, engage mitogenic signalling and other tumour-promoting activities. In addition to monoclonal antibodies directed to the extracellular domain of the receptor, a second class of anti-EGFR agents are those that compete with ATP binding to the tyrosine kinase domain of the receptor, which then inhibits tyrosine kinase activation and subsequently leads to blockade of EGFR signalling pathways. There are a number of this class of agents currently undergoing clinical development, including gefitinib (Iressa, ZD1839), erlotinib (Tarceva, OSI774) and lapatinib.98 The clinical development of some of these agents highlights the opportunities and challenges in the development of kinase inhibitor therapy for cancer. The failure of gefitinib, when combined with chemotherapy, to show superiority to standard treatment in advanced non-small-cell lung cancer demonstrates some of the pitfalls in the clinical evaluation of novel agents. In pre-clinical studies, gefitinib showed activity against a broad panel of tumour cell lines expressing EGFR, either when given alone or in combination with chemotherapy or radiation. Consequently, early studies were not restricted to including patients with a specific molecular abnormality, and clinical activity was observed in lung and colonic cancers.99*,100* Subsequently, significant responses, including complete remissions, were observed in patients with late-stage lung cancer, but only in a small number.112** However, the combination of gefitinib with chemotherapy showed no survival advantage over chemotherapy alone in two large, randomised, phase III lung cancer trials.102**,103**
There has been renewed interest in the field following the results of a placebo-controlled, randomized study of single-agent erlotinib in patients with non-small-cell lung cancer after first-line or second-line chemotherapy. Statistically significant and clinically relevant differences were observed for both progression-free and overall survival.98 In all of these studies, there was a strong indication that there was a subset of patients with non-small-cell lung cancer who seemed to benefit from treatment with erlotinib or gefitinib. Patients with bronchioalveolar carcinoma, never-smokers, females and Japanese patients seemed to have a higher response rate and clinical benefit. These clinical findings have now been followed by the discovery of somatic mutations in exons 18 through 21 encoding the tyrosine kinase domain of the EGFR, and the close association between these mutations and clinical responses to these agents.104,105 Current research strategies are directed at selecting the optimal dose and schedule and identifying factors of response and resistance beyond EGFR gene mutations and/or amplifications, and also to integrate inhibitors of EGFR signalling with conventional therapies and with other molecular targeted therapies, including other antireceptor therapies, receptor-downstream signalling inhibitors and targeted approaches interfering with other essential drivers of cancer such as angiogenesis. INHIBITION OF RAS, RAF, MET AND OTHER SIGNALLING PATHWAYS
The ras family of oncogenes is one of the most frequently activated groups of dominant transforming genes in both human and experimental cancers.106 The ras family of genes encodes highly similar proteins with molecular weights of 21 kDa which are thought to play a key role in signal transduction. One strategy for regulating ras gene expression is by the use of an antisense oligonucleotide directed against H-ras expression. An alternative approach is by inhibiting protein farnesylation. Ras proteins carry an essential lipid moiety – a farnesyl group – at theirCOOH termini. Inhibition of ras farnesylation blocks ras localization to the plasma membrane and, as a consequence, ras fails to interact with regulatory and effector molecules.107 Inhibition of this step alone may be sufficient to stop the cell-signalling and transforming function of constitutively activated ras in tumour cells. Therefore, farnesyl transferase (FT), the enzyme that catalyses this reaction, has become an interesting target for the design of anti-cancer drugs. Initially, inhibitors of FT were regarded as specific and sensitive inhibitors of ras-mediated cellular proliferation.108,109 However, it has become apparent that the critical target of FT inhibitors may not be ras proteins or may include other polypeptides in addition to ras.110 More than 100 proteins have been identified that possess a ‘CAAX’ sequence that can potentially be farnesylated,110 and up to 20 of these have been shown to undergo farnesylation, including rho B, laminins A and B, transducin,
100 Principles of chemotherapy and drug development
CENP-E and CENP-F. Currently, at least three proteins have been identified, inhibition of which may be implicated in the cytotoxic actions of FT inhibitors, and these include rho B, which regulates cytoskeleton re-organization,111 the centromeric proteins CENP-E and EP-F, which interact with microtubules,112 and proteins associated with the phosphoinositide 3-OH kinase AKT pathway.113 Thus, the molecular targets of FT inhibitors remain unclear, but are likely to include several key proteins and possibly some or all of the ras isoforms. Several FT inhibitors have entered clinical trials, including R115777 (Zarnestra), SCH-66336 (Sarasar), L-778,123 and BMS-214662,114 either as single agents or in combination with standard therapies. Promising results have been observed in single-agent studies in breast cancer,115* CML116 and recurrent malignant glioma.117 Pre-clinical experiments have suggested that FT inhibition may have synergistic antitumour effects when administered in combin-ation with several chemotherapy, hormone therapy or targeted therapy agents.114 Furthermore, the combination of SCH-66336 with paclitaxel has demonstrated responses in patients whose disease was refractory to taxanes.114 However, in phase III studies, the addition of Zarnestra to gemcitabine has no bene-fit in patients with advanced pancreatic cancer,118** and further studies are awaited to define the role of FT inhibitors in cancer therapy. One strategy to overcome the challenge inherent in developing therapeutics against signalling elements situated in redundant pathways is to target downstream of convergence points of critical signalling modules. The MAPK pathway integrates a wide array of proliferative signals initiated by RTKs and G protein-coupled receptors. Therapeutic strategies directed at the MAPK pathway are targeting the cascade of downstream effector proteins including raf, MAPK kinase (also called MEK or ERK kinase) and ERK. The raf/MEK/ERK module of the MAPK pathway, which is immediately downstream of ras, may be less redundant and innately resistant to therapeutic manipu-lation than ras. The potential of raf as a therapeutic target has been comprehensively reviewed.119 In brief, the raf family consists of three genes: A-raf, B-raf and C-raf. All the raf proteins are serine/threonine kinases capable of activating the MAPK cascade, but with distinct downstream phosphorylation targets and with unique roles in signalling. Raf activation occurs immediately downstream of membrane and cytoplasmic receptors that relay mitogenic signals.120 Although principally activated by ras, raf may also be activated by ras-independent elements, including the soluble non-RTK Src, protein kinase C, retinoids, anti-apoptotic proteins (e.g. bcl-2) and interferon-beta. The activation status of raf depends on the integration of both activating and inhibitory stimuli, the net result of which determines the downstream messages. Initial efforts at identifying raf mutations in human cancer focused on C-raf, but it would appear that activating B-raf mutations are the predominant genetic aberrations. Most
B-raf mutations are caused by thymidine-to-adenine transversions at nucleotide position 1796 in exon 11 or 15, which encode a valine-to-glutamic acid substitution at amino acid 599 in the kinase domain of the protein. B-raf mutations have been discovered in 60–70 per cent of malignant melanomas and are essentially mutually exclusive with ras mutations in melanoma. There have been many attempts to develop therapeutics against raf, including antisense oligonucleotides, small-molecule kinase inhibitors and dominant interfering DNA constructs. However, most have been directed at C-raf rather than B-raf. The most prominent example of a potential therapeutic against raf is sorafenib, which is a potent competitive inhibitor of ATP binding in the catalytic domains of C-raf, wild-type B-raf and B-raf mutant.121 Sorafenib is also a potent inhibitor of RTKs involved in tumour progression and angiogenesis, including VEGFR-2, VEGFR-3, PDGFR, flt-3, c-KIT and FGFR-1. Sorafenib has been evaluated in the clinic through to phase III studies. It improved the progression-free survival from 12 weeks to 24 weeks compared with placebo (p 0.0000001) in patients with advanced renal cancer who had received at least one prior systemic therapy.122** In contrast, sorafenib has little anti-tumour activity in patients with melanoma in the studies reported so far.123*,124* The results of additional studies of sorafenib, including combination studies with other anti-cancer agents, are awaited. The proto-oncogene C-met is also an interesting target for cancer therapy. C-met encodes the high-affinity receptor for hepatocyte growth factor (HGF) or scatter factor (SF). C-met and HGF are each required for normal mammalian development and are important in cell migration, morphogenic differentiation, cell growth and angiogenesis.125 Both C-met and HGF have been shown to be deregu-lated, and to correlate with poor prognosis, in a number of human cancers. A number of approaches to the therapeutic inhibition of C-met and HGF are currently undergoing evaluation.125 Other signalling pathways currently being investigated as potential targets for cancer therapies are the Src family kinases and the P13K/Akt/mTOR pathway. Src was one of the first proto-oncogenes to be identified and plays an important role in cellular proliferation, adhesion and motility, and is also involved in cell survival and intracellular trafficking in various specialized cell types.126 Several inhibitors are currently in early-phase clinical trials. The downstream effector of phosphatidylinositol 3-kinase (PI3K), AKt, is frequently hyperactivated in human cancers. A critical downstream effector of AKt, which contributes to tumorigenesis, is mTOR. In the P13K/Akt/mTOR pathway, AKt is flanked by two tumour suppressors: PTEN activating as a brake upstream of AKt, and TSC1/TSC2 heterodimer, acting as a brake downstream of AKt and upstream of mTOR.127 The rapamycin derivatives CCl-779, RAD001 and AP23573 are currently being investigated in clinical trials as cancer therapies.128 Finally, angiogenesis is one of the most attractive targets in developing cancer
Novel anti-cancer therapies 101
therapies. The mechanisms of new blood vessel formation and its role in malignant transform-ation, tumour growth, metastasis and its inhibition are discussed in Chapter 12.
Targeting the cell cycle Dysregulation of the cell cycle is a hallmark of malignancy. The biochemical events involved in cell-cycle progression have been reviewed in detail.129,130 Cyclin-dependent kinases (CDKs) regulate the transition from one cell-cycle phase to the next by phosphorylating key structural and regulatory substrate molecules. This activity is regulated by the orderly appearance of cyclins, prompted by environmental factors and by post-translational modification of CDKs through phosphorylation of key stimulating and inhibitory sites in the catalytic subunit. Altered expression of CDK components has been demonstrated in malignant cells with consequent loss of regulation of cell-cycle progression, and as such are targets for developing novel therapeutic strategies. Many different strategies for targeting the cell cycle have been described. Potential strategies have included targeting mitosis (by targeting tubulin or mitotic kinases such as Aurora kinases); chemical inhibitors of CDK catalytic activity; inhibition of the interaction between cyclins and cdks; decreasing cyclin expression; promoting the degradation of cyclins by increasing their phosphorylation; and restoration of endogenous cdk inhibitor function. The first generation of cdk inhibitors lacked specificity, with flavopiridol, staurosporine and its analogue U-01, and E7070 being non-selective inhibitors of not only cdks but also many other targets. Second-generation inhibitors are more selective, with many of the compounds developed to target selected cdks. Flavopiridol has several mechanisms of anti-cancer activity131 and is a broad-spectrum cdk inhibitor targeting cdks 1, 2, 4, 6 and 7, and also inhibits the cdk9–cyclin T complex, repressing transcription and decreasing cyclin D1 mRNA expression. A number of schedules of flavopiridol have been evaluated in phase I and phase II studies, with some indication of anti-tumour activity in patients with advanced mantle-cell lymphoma.132* The use of flavopiridol in combination studies is also being pursued. U-01 also has several mechanisms of action, including CDK1 and CDK2 inhibition, induction of cell-cycle arrest and apoptosis, abrogation of the G2 checkpoint in response to DNA damage and the induction of p53-independent apoptosis.133–135 This agent has also been evaluated in a phase I clinical trial.136* E7070, a chloroindolyl sulphonamide, has been evaluated in phase I and phase II trials, and the purine analogue R-roscovitine (CYC202) and 2-aminothiazole BMS387032 are also in early-phase clinical trials.137 Aurora is the name given to a family of serine/threonine protein kinases that regulate many processes during cell
division. Aurora kinases are involved in the control of the centrosome and nuclear cycles, and have essential functions in mitotic processes such as chromosome condensation, spindle dynamics, kinetochore–microtubule interactions, chromosome orientation and establishment of the metaphase plate. They are also required for the proper completion of cytokinesis.138 The family members have been designated Aurora A, B and C and have very distinct localizations and functions. Aurora A over-expression is correlated with tumour progression, is mutated in certain cancers and behaves, at least in certain circumstances, as an oncogene. Survivin, part of the Aurora B complex, may be a key protector against apoptosis and/or mitotic catastrophe.139 Given these observations, and the known roles of Aurora A and B kinases in so many fundamentally important aspects of a range of cell-cycle events, these kinases are likely to be relevant targets for designing novel anti-cancer drugs. Information on at least three Aurora kinase inhibitors have been fully published, including ZM447439,140 hesperadin,141 and VX-680.142 The results of clinical studies are awaited.
Apoptotic pathways – targets for drug design Apoptosis (programmed cell death) is a set of ordered events that enables the selective removal of cells from tissue and is essential for homeostasis and proper function of multi-cellular organisms. Components of this signalling network, which includes ligands such as CD45, TNF and TNF-related apoptosis-inducing ligand (TRAIL), as well as downstream molecules such as caspases, bcl-2 family members and inhibitor-of-apoptosis proteins, which trigger and regulate apoptosis, are crucial targets for anti-cancer drug development.143 The cloning of the human TNF cDNA, as reported in 1984, allowed the generation of recombinant soluble protein for pre-clinical and clinical studies. In these trials, it was found that TNF was unsuitable for systemic administration at clinically relevant doses due to toxicity associated with its strong pro-inflammatory activity, including fever, lung or liver failure, hypotension and coagulopathy.144* Subsequent studies have shown that TNF can be administered safely and effectively by isolated limb perfusion, for example in melanoma, in which the protein does not enter the systemic blood circulation.145 Another death-ligand family member, TRAIL, has been cloned and characterized.146 TRAIL can induce apoptosis in transformed cells with little detectable cytotoxic effect in normal and non-transformed cells. Several receptors for TRAIL have been identified, including two death receptors (DR4 and DR5, also called TRAIL-R1 and TRAIL-R2) and two decoy receptors (TRAIL-R3, TRAIL-R4).147 Monoclonal antibodies that functionally engage DR4 or DR5 have demonstrated anti-tumour activity in pre-clinical models,148 and the results of clinical evaluation are awaited.
102 Principles of chemotherapy and drug development
MANIPULATION OF THE TUMOUR SUPPRESSOR P53
P53 is a transcription factor that plays a critical role in the cellular response to stress. Functional p53 is vital for tumour growth suppression in vivo,149 and this function is lost by mutation in more than 50 per cent of cancers, the remainder appearing to have defective p53 signalling pathways.150 Under normal, non-stress conditions, p53 is maintained at very low levels. However, on exposure to the stress of DNA damage, hypoxia, alterations in redox potential and deregulated oncogene expression, p53 is stabilized and activated through post-translational modification.151 Activated p53 induces growth arrest and/or apoptosis through the induction or repression of specific target genes. P53-directed pathways of apoptosis involve both transcriptional regulation of multiple p53 target genes as well as transcription-independent mechanisms. P53 targets have been identified in both the extrinsic (death receptor) and intrinsic (mitochondrial) pathways of apoptosis.152 Through these mechanisms, p53 eliminates cells with damage that cannot be repaired. Several strategies have been proposed in the manipulation of p53 in cancer therapy,151 including relieving p53 from inhibition by Mdm2; accumulating p53 by proteosome inhibitors; targeted eradication of mutant p53 by HSP90 inhibitors; reactivation of p53 by small molecules or adaptor proteins; and genetic therapies, by either re-expressing wildtype p53 or selectively eradicating cells with a mutant p53 using an oncolytic virus [Onyx-015].153 Heat shock protein 90 (HspP90) is a molecular chaperone whose association is required for the stability and function of multiple mutated, chimeric and over-expressed signalling proteins that promote the growth and/or survival of cancer cells. Hsp90 client proteins include mutated p53, bcr-abl, raf-1, Akt, erbB2, and HIF-1α. Hsp90 inhibitors, by interacting specifically with a single molecular target, cause the destabilization and eventual degradation of Hsp90 client proteins, and have shown promising anti-tumour activity in pre-clinical models.154 The Hsp90 inhibitor 17-allylaminogeldanamycin (17AAG) has been evaluated in early-phase clinical studies.155* Myelosuppression was not a dose-limiting toxicity, supporting the notion that this agent may be combined with other anti-cancer agents in combination studies. The ubiquitin–proteosome system also plays a critical role in protein homeostasis and in regulating normal and cancerrelated cellular processes. Ubiquitin-mediated degradation is a complex process that comprises of well-defined steps involving ubiquitin-activating enzymes, ubiquitin-conjugating enzymes and ubiquitin ligases, and is a rich source of potential molecular targets for therapeutic intervention.156 The first in class proteosome inhibitor bortezomib (Velcade) inhibits NF-κB, and can overcome conventional drug resistance in both in-vitro and in-vivo models of human myeloma in the bone-marrow microenvironment.157,158 Following promising results in initial studies, bortezomib was compared with high-dose dexamethasone in patients
with relapsed multiple myeloma who had received one to three previous therapies. Bortezomib had a statistically significant superior response rate, time to progression and median overall survival compared to high-dose dexamethasone in these patients with previously treated multiple myeloma, and the results of combination therapy studies are awaited.159**
Differentiation agents Included amongst the newer strategies in oncology practice is the evaluation of a number of agents designed to induce differentiation in tumour cells, thereby leading to inhibition of cellular proliferation and apoptosis. The most prominent of these so-called differentiation agents are the retinoids and vitamin D analogues. RETINOIDS
Vitamin A and its biologically active derivatives retinal and retinoic acid together with a large repertoire of synthetic analogues are collectively referred to as retinoids. Naturally occurring retinoids regulate the growth and differentiation of a wide variety of cell types and play a crucial role in the physiology of vision and as morphogenic agents during embryonic development. Retinoids exert most of their effects by binding to specific receptors and modulating gene expression. The nuclear retinoid receptors are members of the steroid/thyroid hormone superfamily of receptors,160 with which they share common structural and functional properties. The diversity of retinoid-induced signalling pathways is mediated by at least six retinoid receptors, which fall into two subfamilies, retinoic acid receptors (RARs) α, β and γ, and the retinoid X receptors (RXRs) α, β and γ.161 The mechanism of action of retinoid receptors and the cellular consequences of retinoid stimulation have been extensively reviewed.162 Briefly, the RARs bind all-trans retinoic acid (ATRA) with high affinity,163 whereas the stereoisomer 9-cis retinoic acid is a bifunctional ligand that can bind to and activate both RARs and RXRs.164 Despite these similarities, the RXRs belong to a subgroup of nuclear receptors distinct from the RARs, suggesting that these two groups of retinoid receptors have distinct roles in retinoid signalling. Both negative and positive effects on transcription can occur in the absence of ligand, and these bimodal transcriptional properties of retinoid receptors are mediated, in part, by the ability of these receptors to associate with various co-activators and co-repressors such as SMRT and N-CoR.165 Transcriptional regulation by receptors would therefore seem to be controlled by selective recruitment of co-activators and co-repressors in response to hormone and, in turn, control of activity of a target promoter. It may be that the role of ligand binding is to cause a conformational change in the receptor, and as a result of this a co-repressor protein is dissociated from the receptor
Novel anti-cancer therapies 103
and a co-activator binds to the receptor, thereby initiating transcription.166 The diversity of dimer complexes that can occur increases the complexity of retinoid signalling mechanisms. Both RARs and RXRs can bind response elements as homodimers, albeit at high concentrations, although heterodimerization of RARs and RXRs enables high-affinity binding of RARs to response elements. Furthermore, RXRs also serve as promiscuous partners in a multitude of other hormonal response systems, including vitamin D signalling pathways. Therefore a large number of different receptor complexes can be formed, controlling distinct pathways.167 Consequently, a range of possible effects is possible by disruption of these pathways by pharmacological agents. At the cellular level, activation of the retinoid receptors can inhibit cell proliferation, induce differentiation and induce apoptosis in normal and transformed cells in tissue culture. Although there is a difference between various cellline models in the receptor which mediates these processes, it appears in some models that activation of RARs alone is sufficient to induce differentiation, but activation of RXRs is essential for the induction of apoptosis.168 Numerous putative mechanisms have been proposed for the induction of apoptosis by retinoids, including activation of the AP-1 complex, for which activation of the receptors is not necessary,169 but the exact mechanisms remains unknown. However, retinoids can cause growth regression in in-vivo xenograft models of experimental cancer169 as well as having anti-promotion activity in several animal models of carcinogenesis.170 Most of the clinical trials of retinoids as chemoprevention agents focus on individuals at an increased risk of developing cancer, such as patients with pre-malignant lesions or patients who have been successfully treated for an early-stage carcinoma and have a high risk of developing a second primary cancer. None of these studies has as yet demonstrated significant chemopreventive effect with acceptable toxicity, that can be maintained after retinoid withdrawal, and which could justify routine use in clinical practice.162 The most prominent example of retinoids as differentiating agents in oncology practice is the remarkable activity of ATRA in patients with acute promyelocytic leukaemias (APLs). Numerous phase II studies have confirmed that ATRA induces complete remission in the vast majority of patients, with rapid resolution of the characteristic life-threatening coagulopathy. The duration of remission with ATRA alone is usually brief, and post-remission chemotherapy is required to diminish the likelihood of relapse. A randomized study has confirmed that ATRA as induction or maintenance treatment improves disease-free and overall survival as compared with chemotherapy alone and should be included in the treatment of APL.171** The increasing understanding of retinoid-induced signalling pathway should lead to the design of combination therapies with other agents acting on steroid hormone receptors, agents which inhibit intracellular signalling pathways, and raises the intriguing possibility of enhancing the sensitivity
of tumours to cytotoxic agents and of overcoming drug resistance by adjusting the apoptotic set point.162 VITAMIN D ANALOGUES
In addition to its role in calcium homeostasis, vitamin D can also promote cellular differentiation, inhibit proliferation and induce apoptosis in cancer cells as well as inhibiting tumour-induced angiogenesis172 and the invasive potential of breast cancer cells in vitro.173 The vitamin D hormone mediates its action through the activation of the vitamin D receptor, where the receptor–ligand complex functions as a transcription factor, and binds with DNA by interacting with vitamin D response elements leading to either activation or suppression of target gene transcription. Vitamin D3 is limited in its potential clinical application because of the induction of hypercalcaemia at therapeutic doses. A number of analogues have been synthesized with the aim of decreasing this calcaemic effect and enhancing its anti-proliferative actions. MC903, as a topical application, can stabilize locally advanced and cutaneous metastatic breast cancer.174 EB1089 can induce regression of colon and breast cancers in animal models without causing significant hypercalcaemia and has been evaluated in a phase I study in these two types of cancer.175* EB1089 has also subsequently been evaluated in phase II studies in pancreatic176* and hepatocellular177* cancers. In both studies, the most commonly observed toxicity was dose-related hypercalcaemia that resolved with dose interruption. Although stable disease was seen at the first disease assessment (12 weeks of treatment) in a small number of patients with pancreatic cancer,176* there were no objective responses. In contrast, objective responses were observed in a small number of patients with hepatocellular cancer (3–5 per cent). Another novel vitamin D analogue, ILX23-7553, has also entered phase I clinical trials.178*
Epigenetic therapies Epigenetic mechanisms regulate the expression of genetic information. Epigenetic modifications of DNA and histones are stable and heritable, but also reversible.179 They include covalent modifications of bases in the DNA and of amino-acid residues in the histones. DNA methyltransferases are a family of enzymes that methylate DNA at the carbon-5 position of cytosine residues.180 Methylated DNA can then be bound by methyl-binding proteins that function as adaptors between methylated DNA and chromatinmodifying enzymes (e.g. histone deacetylases and histone methyltransferases) by recruiting histone-modifying enzymes to stretches of methylated DNA.181 Histone-modifying enzymes then covalently modify the amino-terminal residues of histones to induce the formation of chromatin structures that repress gene transcription.182 DNA methylation is a crucial mechanism associated with epigenetic regulation. It has a comparatively simple binary
104 Principles of chemotherapy and drug development
pattern (i.e. methylated versus non-methylated bases) compared with the highly complex pattern of histone modifications. Changes in the pattern of DNA methylation, either increased (hypermethylation) or decreased (hypomethylation), have been identified in all types of cancer cells examined so far. Genomic tumour DNA is characterized by distinct methylation changes, termed epimutations. At the global level, the DNA is often hypomethylated, particularly at centromeric repeat sequences, and this hypomethylation has been linked to genomic instability. Local hypomethylation of individual genes may also occur, and has been associated with aberrant gene silencing.183 The reversibility of epigenetic modifications makes them attractive targets for possible therapeutic intervention. Epigenetic mutations must be actively maintained, and so inhibition of certain epigenetic modifications could lead to a change in gene expression patterns and consequently in cellular characteristics. DNA METHYLTRANSFERASE INHIBITORS
DNA methyltransferase inhibitors have been tested in phase I–III clinical trials, and include both nucleoside and non-nucleoside inhibitors. The DNA methyltransferase inhibitor 5-azacytidine is a derivative of the nucleoside cytidine. It is incorporated into DNA and leads to methyltransferase being trapped and inactivated in the form of a covalent protein–DNA adduct. Consequently, cellular DNA methyltransferase is depleted and genomic DNA is demethylated as a result of continued DNA replication. 5azacytidine is a ribose nucleoside and must be modified to a deoxyribonucleoside triphosphate to be incorporated into DNA. In contrast, 5-aza-2 -deoxycytidine (decitabine) is a deoxyribose analogue of 5-azacytidine which does not need to be modified to a deoxy form and can be directly incorporated into DNA. Decitabine has greater inhibition of DNA methylation and anti-tumour activity than 5-azacytidine in experimental models. Decitabine has single-agent activity in myeloid malignancies and myelodysplastic syndrome.184,185 However, it also has significant haematological toxicity.186* Another derivative of 5-azacytidine, zebularine,187 has not yet been evaluated in the clinic. Its mechanism of action is similar to that of the aza-nucleoside inhibitors, and so the demethylating activity of zebularine may be difficult to separate from the toxic effects of DNA methyl-transferase depletion that result from covalent enzyme trapping. Some non-nucleoside agents can also inhibit DNA methyltransferase activity. These compounds include ECGC, the main polyphenol component in green tea,188 RG108,189 4-aminobenzoic acid derivatives, psammaplins, and oligonucleotides such as MG98.190 Of these, MG98 is currently in clinical trials. HISTONE-DEACETYLASE (HDAC) INHIBITION
Gene expression can be regulated by the remodelling of chromatin and altered chromatin conformation. In an open
state, chromatin is conducive to transcriptional activation, whereas closed chromatin confirms transcriptional repression. Chromatin consists of nucleosomes comprising DNA wound around a histone octomer. Post-translational modifications of the N-terminal ‘tails’ of histones, through acetylation, methylation, phosphorylation and ubiquitination, can result in chromatin remodelling, altered chromatin conformation and regulated transcription.182 The opposing activities of histone acetyltransferases (HATs) and histone deacetylases (HDACs) are potent regulators of gene expression.191,192 HDACs counteract the activity of HATs and catalyse the removal of acetyl groups from lysine residues in histone N-termini, leading to chromatin condensation and transcriptional repression.191,192 To date, 18 HDACs have been identified in humans.192 HDACs are over-expressed, or associated with oncogenic transcription factors due to chromosomal translocations or aberrant physical associations, in a number of cancers, making them attractive targets for anti-cancer therapies.192 A number of HDAC inhibitors have been developed, and can be subdivided into five groups: 1. hydroxamic acid-derived compounds, including trichostatin A, PXD101, suberoylanilide hydroxamic acid (SAHA); 2. cyclic peptides, e.g. depsipeptide; 3. short-chain fatty acids, such as valproic acid; 4. benzamides, e.g. MS-275 and CI-994; 5. ketones.194 HDAC inhibitors show selective cytotoxicity against tumour cells, and can activate a number of molecular pathways to mediate a range of biological events that can influence tumour-cell development, growth and survival.194 At least nine different HDAC inhibitors are currently in earlyphase clinical trials, either as single-agent therapy or as part of combination therapy with other agents.195
Senescence, the telomere structure and the telomerase enzyme Telomerase is an enzyme which, although thought to be required during development, is largely repressed in adult somatic tissues.196 Telomerase replicates the terminal sequences of eukaryotic chromosomes, namely the telomere.197 The absence of telomerase activity from normal somatic cells has led to the proposal that telomere shortening may be a molecular clock that contributes to the onset of cellular senescence in normal cells.198 Conversely, the reactivation or expression of telomerase may be a major mechanism by which cancer cells overcome normal cellular senescence.199 Indeed, telomerase activity may be present in more than 80 per cent of tumour biopsies, yet absent or reduced in normal somatic tissues.200 The realization that activation of telomerase can co-operate with a limited
References 105
number of other ‘genetic hits’ leading to malignant transformation, allied to the high levels of expression of the telomerase enzyme in specific cancers, reinforces the notion that the telomerase enzyme and the telomeric structure are exciting targets as potential anti-cancer strategies. Numerous potential therapeutic strategies have been proposed, including telomerase-interactive compounds (peptide nucleic acids, antisense oligonucleotides, ribozymes, reverse transcriptase inhibitors) and telomere-interactive compounds, such as the G-quartet interactive agents.201 One potential drawback of these strategies is the ‘phenotype lag’, i.e. the number of cell divisions necessary before inhibition of telomerase leads to sufficient shortening of the telomere to give rise to a phenotypic effect, suggesting that these agents are likely to be active in minimal disease states. However, the most successful approaches that have been evaluated in pre-clinical studies have exploited the selectively of telomerase gene expression within cancer cells to develop gene therapy strategies.202
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
●
Smaller tumours have more rapid growth rates and are more likely to be responsive to chemotherapy; cure rates are likely to be higher with a small tumour burden. The route to administration of a cytotoxic drug is determined by the stability, size, molecular change and sclerosant characteristics of the drug. Combination chemotherapy aims to prevent or slow the development of resistant tumours; drugs used in such regimens have activity as single agents, different mechanisms of action and minimal overlapping toxicities. Drug resistance is the most important reason for cancer treatment failure. Genetic instability and the multi-clonal nature of tumours enable them to adapt to environmental changes. Hypoxia reduces the sensitivity of tumour cells to many drugs. Tumours often have mutations that make them resistant to chemotherapy-induced programmed cell death. DNA-damage repair is important in resistance to many alkylating agents. Understanding the biological mechanisms of drug resistance allows the design of strategies to overcome drug resistance, with potential to improve the efficacy of cytotoxic drugs. Novel anti-cancer agents are being developed based on exploiting our increasing understanding of the molecular and cellular bases of cancer development and progression.
Genes, vaccines and drug delivery Strategies to target the underlying genetic lesions of the cancer cell are often considered under the umbrella of ‘gene therapy’, and include antisense oligonucleotides, recombinant vaccines as immunotherapy, transducing a drug-resistance gene into bone-marrow stem cells to protect the bone marrow during chemotherapy, and the use of expression vectors to convert inactive pro-drugs into active drugs. These approaches are discussed more fully in Chapters 8 and 13. Many of the existing vectors for gene delivery and expression have limitations, and improvement in drug delivery mechanisms and vector design remains a crucial area in order to increase the efficiency of expression and the precision of targeting and to reduce toxicity. The possibility of specific gene targeting and possible synergy with other, existing, cancer therapeutics is a promising goal for future development.
REFERENCES 1 Birkhead BG, Rankin EM, Gallivan S, et al. A mathematical model of the development of drug resistance to cancer chemotherapy. Eur J Cancer Clin Oncol 1987; 23:1421–7. ●2 Skipper HE, Schabel FH, Wilcox ES, et al. Experimental evaluation of potential anticancer agents XII: on the criteria and kinetics associated with ‘curability’ of experimental leukaemia. Cancer Chemother Rep 1964; 35:1–11. 3 Norton L, Simon R. Tumour size, sensitivity to therapy and design of treatment schedules. Cancer Treat Rep 1997; 61:1307–17. 4 Glick JH, Young ML, Harrington D, et al. MOPP/ABV hybrid chemotherapy for advanced Hodgkin’s disease significantly improves failure-free and overall survival: the 8-year results of the intergroup trial. J Clin Oncol 1998; 16:19–26. ●5 Goldie JH, Coldman AJ. A mathematical model for relating the drug sensitivity of tumours to the spontaneous mutation rate. Cancer Treat Rep 1979; 63:1727–33. ◆6 McLeod HL, Evans WE. Oral cancer chemotherapy: the promise and the pitfalls. Clin Cancer Res 1999; 5:2669–71. 7 Hryniuk W, Levine MN. Analysis of dose intensity for adjuvant chemotherapy trials in stage II breast cancer. J Clin Oncol 1986; 4:1162–70. ●8 Kaye SB, Lewis CR, Paul J, et al. Randomised study of two doses of cisplatin with cyclophosphamide in epithelial ovarian cancer. Lancet 1992; 340:329–33. 9 De Vita VT, Hubbard SM, Longo DL. The chemotherapy of lymphomas: looking back, moving forward. The Richard and Hilda Rosenthal Foundation Award Lecture. Cancer Res 1987; 47:5810–24. 10 Hryniuk WM. Average relative dose intensity and the impact of design on clinical trials. Semin Oncol 1997; 14:65–74. 11 Levin L, Hrynuik WM. Dose intensity analysis of chemotherapy regimens in ovarian carcinoma. J Clin Oncol 1987; 5:756–67.
106 Principles of chemotherapy and drug development
12 Skipper H. Data and analyses having to do with the influence of dose intensity and duration of treatment (single drugs and combinations) on lethal toxicity and the therapeutic response of experimental neoplasms. Sern Research Institute Booklets 1986; 13, 1986 and 2–13, 1987. ●13 Goldin A, Schabel FM. Clinical concepts derived from animal chemotherapy studies. Cancer Treat Rep 1981; 65(Suppl. 3):11–19. 14 Lokich JJ, Ahlgren JD, Gullo SJ, et al. A prospective randomised comparison of continuous infusion fluorouracil with a conventional bolus schedule in metastatic colorectal carcinoma. A Mid-Atlantic Oncology Program study. J Clin Oncol 1989; 7:425–32. ◆15 O’Quigley J, Zohar S. Experimental designs for phase I and phase I/II dose-finding studies. Br J Cancer 2006; 80:1–8. 16 Garcia-Carbonero R, Supko JG, Maki RG, et al. Ecteinascidin743 (ET-743) for chemotherapy-naïve patients with advanced soft tissue sarcomas: multicenter phase II and pharmacokinetic study. J Clin Oncol 2005; 23:5484–92. ●17 Brown JM, Giaccia AJ. The unique physiology of solid tumours: opportunities (and problems) for cancer therapy. Cancer Res 1998; 58:1408–16. ●18 Meta-analysis group in cancer. Efficacy of intravenous infusion of fluorouracil compared with bolus administration in advanced colorectal cancer. J Clin Oncol 1998; 16:301–8. 19 de Gramont A, Bosset J, Milan C, et al. Randomized trial comparing monthly low-dose leucovorin and fluorouracil bolus plus continuous infusion for advanced colorectal cancer: a French Intergroup study. J Clin Oncol 1997; 15:808–15. 20 von Pawel J, von Roemeling R, Gatzemeier U, et al. Tirapazamine plus cisplatin versus cisplatin in advanced non-small-cell lung cancer: a report of the international CATAPULT I study group. Cisplatin and Tirapazamine in subjects with advanced previously untreated non-small-cell lung tumors. J Clin Oncol 2000; 18(6):1351–9. 21 Shepherd F, Loschel G, von Pawel J, et al. Comparison of tirazone (tirapazamine) and cisplatin vs. etoposide and cisplatin in advanced non-small cell lung cancer (NSCLC): final results of the international phase III CATAPULT II Trial. Lung Cancer 2000; 29(Suppl. 1):28(Abstr #87). 22 Patterson LH, McKeown SR, Robson T, et al. Antitumour prodrug development using cytochrome P450 (CYP) mediated activation. Anticancer Drug Des 1999; 14(6):473–86. 23 Kerbel RS. A cancer therapy resistant to resistance. Nature 1997; 390:335–6. 24 Sandler AB, Gray R, Brahmer J, et al. Randomized phase II/III trial of paclitaxel (P) plus carboplatin (C) with or without bevacizumab (NSC # 704865) in patients with advanced non-squamous non-small cell lung cancer (NSCLC): an Eastern Cooperative Oncology Group (ECOG) trial – E4599. J Clin Oncol 2005; 23, Abstract 4. 25 Laverdiere C, Chiasson S, Costea I, et al. Polymorphism G80A in the reduced folate carrier gene and its relationship to methotrexate plasma levels and outcome of childhood acute lymphoblastic leukemia. Blood 2002; 100:3832–4.
26 Wu M, Gunning W, Ratnam M. Expression of folate receptor type alpha in relation to cell type, malignancy, and differentiation in ovary, uterus, and cervix. Cancer Epidemiol Biomarkers Prev 1999; 8:775–82. 27 Gibbs DD, Theti DS, Wood N, et al. BGC 945, a novel tumourselective thymidylate synthase inhibitor targeted to α -folate receptor-overexpressing tumours. Cancer Res 2005; 65:11721–8. 28 McGuire JJ. Antifolate polyglutamylation in preclinical and clinical antifolate resistance. In: Jackman AL (ed.) Antifolate Drugs in Cancer Therapy., New Jersey: Humana Press, 1999, 339–63. 29 Barredo JC, Synold TW, Laver J, et al. Differences in constitutive and post-methotrexate folylpolyglutamate synthetase activity in B-lineage and T-lineage leukemia. Blood 1994; 84:564–9. 30 Cowan KH, Jolivet J. A methotrexate-resistant human breast cancer cell line with multiple defects, including diminished formation of methotrexate polyglutamates. J Biol Chem 1984; 259:10793–800. 31 McCloskey DE, McGuire JJ, Russell CA, et al. Decreased folylpolyglutamate synthetase activity as a mechanism of methotrexate resistance in CCRF-CEM human leukemia sublines. J Biol Chem 1991; 266:6181–7. 32 Whitehead VM, Rosenblatt DS, Vuchich MJ, et al. Accumulation of methotrexate and methotrexate polyglutamates in lymphoblasts at diagnosis of childhood acute lymphoblastic leukemia: a pilot prognostic factor analysis. Blood 1990; 76:44–9. 33 Seitz DE. Trimetrexate: a critical appraisal of the phase II clinical trial experience: evidence of drug discovery–clinical development disjunction. Cancer Invest 1994; 12:657–61. 34 Benepal TS, Judson I. ZD9331: discovery to clinical development. Anticancer Drugs 2005; 16:1–9. 35 Schinkel AH, Smit JJ, van Tellingen O, et al. Disruption of the mouse mdr1a P-glycoprotein gene leads to a deficiency in the blood–brain barrier and to increased sensitivity to drugs. Cell 1994; 77:491–502. 36 Chan HS, Thorner PS, Haddad G, et al. Immunohistochemical detection of P-glycoprotein: prognostic correlation in soft tissue sarcoma of childhood. J Clin Oncol 1990; 8:689–704. 37 Clarke R, Leonessa F, Trock B. Multidrug resistance/Pglycoprotein and breast cancer: review and meta-analysis. Semin Oncol 2005; 32(6 Suppl. 7):S9–15. ●38 Leith CP, Kopecky KJ, Chen IM, et al. Frequency and clinical significance of the expression of the multidrug resistance proteins MDR1/P-glycoprotein, MRP1, and LRP in acute myeloid leukemia: a Southwest Oncology Group study. Blood 1999; 94:1086–99. 39 Pusztai L, Wagner P, Ibrahim N, et al. Phase II study of tariquidar, a selective P-glycoprotein inhibitor, in patients with chemotherapy-resistant, advanced breast carcinoma. Cancer 2005; 104:682–91. 40 Kojima A, Hackett NR, Crystal RG. Reversal of CPT-11 resistance of lung cancer cells by adenovirus-mediated gene transfer of the human carboxylesterase cDNA. Cancer Res 1998; 58:4368–74.
References 107
41 Kojima A, Hackett NR, Ohwada A, et al. In vivo human carboxylesterase cDNA gene transfer to activate the prodrug CPT-11 for local treatment of solid tumors. J Clin Invest 1998; 101:1789–96. 42 Friedman HS, Petros WP, Friedman AH, et al. Irinotecan therapy in adults with recurrent or progressive malignant glioma. J Clin Oncol 1999; 17:1516–25. 43 Murray GI, Taylor MC, McFadyen MC, et al. Tumor-specific expression of cytochrome P450 CYP1B1. Cancer Res 1997; 57:3026–31. 44 Chun YJ, Kim S. Discovery of cytochrome P450 1B1 inhibitors as new promising anti-cancer agents. Med Res Rev 2003; 23:657–68. 45 Cullen KJ, Newkirk KA, Schumaker LM, et al. Glutathione S-transferase pi amplification is associated with cisplatin resistance in head and neck squamous cell carcinoma cell lines and primary tumors. Cancer Res; 63:8097–102. 46 Sakamoto M, Kondo A, Kawasaki K, et al. Analysis of gene expression profiles associated with cisplatin resistance in human ovarian cancer cell lines and tissues using cDNA microarray. Hum Cell 2001; 14:305–15. 47 Siddik ZH. Cisplatin: mode of cytotoxic action and molecular basis of resistance. Oncogene 2003; 22:7265–79. 48 Wang Y, O’Brate A, Zhou W, et al. Resistance to microtubule-stabilizing drugs involves two events: betatubulin mutation in one allele followed by loss of the second allele. Cell Cycle 2005; 4:1847–53. ●49 Monzo M, Rosell R, Sanchez JJ, et al. Paclitaxel resistance in non-small-cell lung cancer associated with beta-tubulin gene mutations. J Clin Oncol 1999; 17:1786–93. ●50 Edler D, Kressner U, Ragnhammar P, et al. Immunohistochemically detected thymidylate synthase in colorectal cancer: an independent prognostic factor of survival. Clin Cancer Res 2000; 6:488–92. 51 Edler D, Hallstrom M, Johnston PG, et al. Thymidylate synthase expression: an independent prognostic factor for local recurrence, distant metastasis, disease-free and overall survival in rectal cancer. Clin Cancer Res 2000; 6:1378–84. 52 Berger SH, Jenh CH, Johnson LF, et al. Thymidylate synthase overproduction and gene amplification in fluorodeoxyuridine-resistant human cells. Mol Pharmacol 1985; 28:461–7. ●53 Leichman CG, Lenz HJ, Leichman L, et al. Quantitation of intratumoral thymidylate synthase expression predicts for disseminated colorectal cancer response and resistance to protracted-infusion fluorouracil and weekly leucovorin. J Clin Oncol 1997; 15:3223–9. 54 Villalona-Calero MA, Kolesar JM. Mitomycin as a modulator of irinotecan anticancer activity.Oncology 2002; 16(8 Suppl. 7):21–5. 55 Giuliani F, Molica S, Maiello E, et al. Irinotecan (CPT-11) and mitomycin-C (MMC) as second-line therapy in advanced gastric cancer: a phase II study of the Gruppo Oncologico dell’ Italia Meridionale (prot. 2106). Am J Clin Oncol 2005; 28:581–5. 56 Hirata A, Ogawa S-I, Kometani T, et al. ZD1839 (Iressa) induces antiangiogenic effects through inhibition of
57
58
●
59
60
61
62
63 64
65
66
67
68
69
70
71
epidermal growth factor receptor tyrosine kinase. Cancer Res 2002; 62:2554–60. Hirsch FR, Varella-Garcia M, Bunn PA Jr, et al. Epidermal growth factor receptor in non-small-cell lung carcinomas: correlation between gene copy number and protein expression and impact on prognosis. J Clin Oncol 2003; 21:3798–807. Bhargava R, Lal P, Chen B. HER-2/neu and topoisomerase IIa gene amplification and protein expression in invasive breast carcinomas: chromogenic in situ hybridization and immunohistochemical analyses. Am J Clin Pathol 2005; 123:889–95. Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecanrefractory metastatic colorectal cancer. N Engl J Med 2004; 351:337–45. Miyashita T, Reed JC. bcl-2 gene transfer increases relative resistance of S49.1 and WEHI7.2 lymphoid cells to cell death and DNA fragmentation induced by glucocorticoids and multiple chemotherapeutic drugs. Cancer Res 1992; 52:5407–11. Sakakura C, Sweeney EA, Shirahama T, et al. Overexpression of bax sensitizes breast cancer MCF-7 cells to cisplatin and etoposide. Surg Today 1997; 27:676–9. Oltersdorf T, Elmore SW, Shoemaker AR, et al. An inhibitor of Bcl-2 family proteins induces regression of solid tumours. Nature 2005; 435:677–81. Mani A, Gelmann EP. The ubiquitin–proteasome pathway and its role in cancer. J Clin Oncol 2005; 23:4776–89. Frankel A, Man S, Elliott P, et al. Lack of multicellular drug resistance observed in human ovarian and prostate carcinoma treated with the proteasome inhibitor PS-341. Clin Cancer Res 2000; 6:3719–28. Hideshima T, Richardson P, Chauhan D, et al. The proteasome inhibitor PS-341 inhibits growth, induces apoptosis, and overcomes drug resistance in human multiple myeloma cells. Cancer Res 2001; 61:3071–3076. Bold RJ, Virudachalam S, McConkey DJ. Chemosensitization of pancreatic cancer by inhibition of the 26S proteasome. J Surg Res 2001; 100:11–17. Fahy BN, Schlieman M, Virudachalam S, et al. AKT inhibition is associated with chemosensitisation in the pancreatic cancer cell line MIA-PaCa-2. Br J Cancer Ju 2003; 89:391–397. Yang Y, Ikezoe T, Saito T, et al. Proteasome inhibitor PS-341 induces growth arrest and apoptosis of non-small cell lung cancer cells via the JNK/c-Jun/AP-1 signaling. Cancer Sci 2004; 95:176–180. Mortensen MM, Schlieman MG, Subbulakshmi V, et al. Reduction in BCL-2 levels by 26S proteasome inhibition with bortezomib is associated with induction of apoptosis in small cell lung cancer. Lung Cancer 2005; 49:163–170. Dabholkar M, Vionnet J, Bostick-Bruton F, et al. Messenger RNA levels of XPAC and ERCC1 in ovarian cancer tissue correlate with response to platinum-based chemotherapy. J Clin Invest 1994; 94:703–708. Metzger R, Leichman CG, Danenberg KD, et al. ERCC1 mRNA levels complement thymidylate synthase mRNA levels in
108 Principles of chemotherapy and drug development
72
73
◆74
75 76
●77
●78
79
80
●81
●82
●83
●84
●85
86
87
●88
predicting response and survival for gastric cancer patients receiving combination cisplatin and fluorouracil chemotherapy. J Clin Oncol 1998; 16:309–316. Zhou W, Gurubhagavatula S, Liu G, et al. Excision Repair Cross-Complementation Group 1 Polymorphism Predicts Overall Survival in Advanced Non-Small Cell Lung Cancer Patients Treated With Platinum-Based Chemoptherapy. Clin Cancer Res 2004; 10:4939–4943. Farmer H, McCabe N, Lord CJ, et al. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 2005; 434:917–920. Aaronson SA. Growth factors and cancer. Science 1991; 254:1146–1153. Cantley LC, Auger KR, Carpenter C, et al. Oncogenes and signal transduction. Cell 1991; 64:281–13. Al-Obeidi FA, Wu JJ, Lam KS. Protein tyrosine kinases: structure, substrate, specificity and drug discovery. Biopolymers 1998; 47:197–223. Gullick WJ. Prevalence of aberrant expression of the epidermal growth factor receptor in human cancers. Br Med Bull 1991; 47:87–98. Ullrich A, Schlessinger J. Signal transduction by receptors with tyrosine kinase activity. Cell 1990; 61:203–212. Hao D, Rowinsky EK. Inhibiting signal transduction: recent advances in the development of receptor tyrosine kinase and ras inhibitors. Cancer Invest 2002; 20:387–404. Buchdunger E, Zimmermann J, Mett H, et al. Inhibition of the Abl protein-tyrosine kinase in vitro and in vivo by a 2-phenylaminopyrimidine derivative. Cancer Res 1996; 56:100–104. Druker BJ, Tamura S, Buchdunger E, et al. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-abl positive cells. Nat Med 1996; 2:561–566. Schindler T, Bornmann W, Pellicena P, et al. Structural mechanism for STI-571 inhibition of abelson tyrosine kinase. Science 2000; 289:1938–1942. Druker BJ, Talpaz M, Resta DJ, et al. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukaemia. N Engl J Med 2001; 344:1031–1037. Kantarjian H, Sawyers C, Hochlaus A, et al. Hematologic and cytogenetic responses to imatinib mesylate in chronic myelogenous leukaemia. N Engl J Med 2002; 346:645–652. O’Brien SG, Guilhot F, Larson RA, et al. Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukaemia. N Engl J Med 2003; 348:994–1004. Lux M, Rubin BP, Biase TL, et al. KIT extracellular and kinase domain mutations in gastrointestinal stromal tumours. Am J Pathol 2000; 156:791–795. Rubin BP, Singer S, Tsao C, et al. KIT activation is a ubiquitous feature of gastrointestinal stromal tumours. Cancer Res 2001; 61:8118–8121. Demetri GS, von Mehren M, Blanke CD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumours. N Engl J Med 2002; 347:472–450.
89 Ross JS, Flectcher JA. The her2/neu oncogene in breast cancer: prognostic factors, predictive factor and target for therapy. Stem Cells 1998; 16:413–428. ●90 Baselga J, Tripathy D, Medelsohn J, et al. Phase II study of weekly intravenous recombinant humanized anti-p185HER2 monoclonal antibody in patients with HER2/neuoverexpressing metastatic breast cancer. J Clin Oncol 1996; 14:737–744. ●91 Vogel CL, Cobleigh MA, Tripathy D, et al. Efficacy and safety of trastuzumab as a single agent in first-line treatment of HER2-overexpressing metastatic breast cancer. J Clin Oncol 2002; 20:719–726. 92 Cobleigh MA, Vogel CL, Tripathy D, et al. Multinational study of the efficacy and safety of humanized anti-HER2 monoclonal antibody in women who have HER2overexpressing metastatic breast cancer that has progressed after chemotherapy for metastatic disease. J Clin Oncol 1999; 17:2639–2648. ●93 Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344:783–792. ●94 Romond EH, Perez EA, Bryant J, et al. Trastuzumub plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 2005; 353:1673–1684. ●95 Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 2005; 353:1659–1672. ◆96 Midgley R, Kerr D. Bevacizumab – current status and future directions. Ann Oncol 2005; 16:999–1004. 97 Pal SK, Pegram M. Epidermal growth factor receptor and signal transduction: potential targets for anti-cancer therapy. Anti-Cancer Drugs 2005; 16:483–494. ◆98 Baselga J, Arteaga CL. Critical update and emerging trends in epidermal growth factor receptor targeting in cancer. J Clin Oncol 2005; 23:2445–2459. ●99 Baselga J, Rischin D, Ranson M, et al. Phase I, safety, pharmacokinetic, and pharmacodynamic trial of ZD1839, a selective oral epidermal growth factor receptor tyrosine kinase inhibitor, in patients with five selected solid tumour types. J Clin Oncol 2002; 20:4292–4302. ●100 Lorusso PM. Phase I studies of ZD1839 in patients with common solid tumours. Semin Oncol 2003; 30:21–29. ●101 Fukuota M, Yano S, Giaccone G, et al. Multi-institutional randomised phase II trial of gefitinib for previously treated patients with advanced non small lung cancer. J Clin Oncol 2003; 21:2237–2245. ●102 Giaccone G, Herbst RS, Manegold C, et al. Gefitinib in combination with gemcitabine and cisplatin in advanced non-small-cell lung cancer: A phase III trial – INTACT 1. J Clin Oncol 2004; 22:777–784. ●103 Herbst RS, Giaccone G, Schiller JH, et al. Gefitinib in combination with paclitaxel and carboplatin in advanced non-small-cell lung cancer: A Phase III trial – INTACT 2. J Clin Oncol 2004; 22:785–794. ●104 Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth factor receptor underlying
References 109
●105
◆106
107
108
109
◆110
◆111
●112
113
◆114
●115
116
117
●118
◆119
◆120
responsiveness of non-small cell lung cancer to gefitinib. N Engl J Med 2004; 350:2129–2139. Paez JG, Janne PA, Lee JC, et al. EGFR mutations in lung cancer: Correlation with clinical response to gefitinib therapy. Science 2004; 304:1497–1500. Bos JL, Ras oncogenes in human cancer: A review. Cancer Res 1989; 49:4682–4689. Holmlund JT, Monia BP, Kwoh TJ, Dorr FA. Towards antisense oligonucleotide therapy for cancer. ISIS compounds in clinical development. Curr Opin Mol Ther 1999; 1:372–385. Kato K, Cox AD, Hisaka MM, et al. Isoprenoid addition to ras protein is the critical modification for its membrane association and transforming activity. Proc Natl Acad Sci USA 1992; 89:6403–6407. Sepp-Lorrenzo L, Ma Z, Rands E, et al. A peptidomimetic inhibitor of farnesylprotein transferase blocks the anchorage-dependent and – independent growth of human tumor cell lines. Cancer Res 1995; 55:5302–5309. Adjei AA. Protein farnesyltranferase as a target for the development of anticancer agents. Drugs of the Future 2000; 25:1069–1079. Lebowitz Lebowitz PF, Prendergast GC. Non-ras targets of farnesyltransferase inhibitors: focus on Rho. Oncogene 1998; 17:1439–1445. Ashar HR, James L, Gray K, et al. FTIs block the farnesylation of CENP-E and CENP-F and alter the association of CENP-E with the microtubules. J Biol Chem 2000; 275:30451–30457. Jiang K, Coppola D, Crespo NC, et al. The phosphoinositide 3-OH kinase AKT pathway as a critical target for farnesyltransferase inhibitor-induced apoptosis. Mol Cell Biol 2000; 20:139–148. Mazieres J, Pradines A, Favre G. Perspectives on farnesyltransferase inhibitors in cancer therapy. Cancer Lett 2004; 206:159–167. Johnston SR, Kelland LR. Farnesyltransferase inhibitors – a novel therapy for breast cancer. Endocr Rel Cancer 2001; 8:227–235. Cortes JE, Albitar M, Thomas D, et al. Efficacy of the farnesyltransferase inhibitor R115777 in chronic myeloid leukaemia and other haematological malignancies. Blood 2003; 101:1692–1697. Cloughesy T, Kuhn J, Wen P, et al. Phase II trial of R115777 (Zarnestra) in patients with recurrent glioma not taking enyme inducing antiepileptic drugs (EIAED): a North American Brain Tumor Consortium (NABTC) report. Proc Am Soc Clin Oncol, 2002; A317. Van Cutsem E, Van de Velde H, Karasek P, et al (2004). Phase III trial of gemcitabine plus tipifarnib compared with gemcitabine plus placebo in advanced pancreatic cancer. J Clin Oncol 2004; 22:1430–1438. Beeram M, Patnaik A, Rowinsky EK. Raf: A strategic target for therapeutic development against cancer. J Clin Oncol 2005; 23:6771–6790. Chong H, Vikis HG, Guan KL. Mechanisms of regulating the raf kinase family. Cell Signal 2003; 15:463–469.
121 Wilhelm S, Chien DS. BAY43-8006: Pre-clinical data. Curr Pharm Des 2002; 5:2255–2257. 122 Escudier B, Szczylik C, Eisen T, et al. Randomized phase III trial of the raf kinase and VEGFR inhibitor sorafenib (BAY43–9006) in patients with advanced renal cancer (RCC). J Clin Oncol 2005; 23:350s, abstr LBA4510 ●123 Ratain MJ, Flaherty K, Stadler WM, et al. Preliminary antitumor activity of BAY43–9006 in metastatic renal cell carcinoma and other advanced refractory solid tumors in a phase II randomised discontinuation trial (RDT). J Clin Oncol 2004; 22:381S, abstr 4501. ◆124 Ahmad T, Marais R, Pyle L, et al. BAY43–9006 in patients with advanced melanomas. The Royal Marsden experience. J Clin Oncol 2004; 22:708s, abstr 7506. ◆125 Christensen JG, Burrows J, Salgia R. C-Met as a target for human cancer and characterization of inhibitors for therapeutic intervention. Cancer Lett 2005; 225:1–26. ◆126 Susva M, Missbach M, Green J. Src inhibitors: drugs for the treatment of osteoporosis, cancer or both. Trends in Pharmacol Sci 2000; 21:489–495. 127 Hay N. The Akt-mTOR tango and its relevance to cancer. Cancer Cell 2005; 8:179–183. 128 Vignot S, Faivre S, Aguirre D, Raymond E. mTOR-targeted therapy of cancer with rapamycin derivatives. Ann Oncol 2005; 16:525–537. ◆129 Morgan DO Cyclin dependent kinases: Engines, clocks and microprocessors. Ann Rev Cell Dev Biol 1997; 13:261–291. ◆130 Sherr CJ. Cancer cell cycles. Sciences 1986; 274:1672–1677. ◆131 Senderowicz AM, Sausville EA. Preclinical and clinical development of cyclin-dependent kinase modulators. J Natl Cancer Inst 2000; 92:376–387. ●132 Kouraukis CT, Belch A, Crump M, et al. Flavopiridol in untreated or relapsed mantle-cell lymphoma: results of a phase II study of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2003; 21:1740–1745. 133 Wang Q, Fan S, Eastman A, et al. U-01: a potent abrogator of G2 checkpoint function in cancer cells with disrupted p53. J Natl Cancer Inst 1996; 88:956–965. 134 Busby EC, Leistritz DF, Abraham RT, et al. The radiosensitizing agent 7-hydroxystaurosporine (U-01) inhibits the DNA damage checkpoint kinase hChk1. Cancer Res 2000; 60:2108–2118. 135 Yu Q, LaRose J, Zhang H, et al. U-01 inhibits p53 upregulation and abrogates gamma-radiation-induced G(2)M checkpoint independently of p53 by targeting both of the checkpoint kinases Chk2 and Chk1. Cancer Res 2002; 62:5743–5748. 136 Sausville EA, Arbuck SG, Messmann R, et al. Phase I trial of 72-h continuous infusion U-01 in patients with refractory neoplasms. J Clin Oncol 2001; 19:2319–2333. ◆137 Benson C, Kaye S, Workman P, et al. Clinical anticancer drug development: targeting the cyclin-dependent kinases. Br J Cancer 2005; 92:7–12. ◆138 Carmeno M, Earnshaw WC. The cellular geography of aurora kinases. Nat Rev Mol Cell Biol 2003; 4:842–854. ◆139 Andrews PD. Aurora kinases: shining lights on the therapeutic horizon. Oncogene 2005; 24:5005–5015.
110 Principles of chemotherapy and drug development
140 Ditchfield C, Johnson VL, Tighe A, et al. Aurora B couples chromosome alignment with anaphase by targeting BuR1, Mad2 and Cenp-E to kinetochores. J Cell Biol 2003; 161:267–280. 141 Hauf S, Cole RW, La Terra S, et al. The small molecule Hesperadin reveals a role for Aurora B in correcting kinetochore-microtubule attachment and in maintaining the spindle assembly checkpoint. J Cell Biol 2003; 161:281–294. ●142 Harrington EA, Bebbington D, Moore J, et al. VX-680, a potent and selective small-molecule inhibitor of the Aurora kinases, suppresses tumour growth in vivo. Nat Med 2004; 10:262–267. ◆143 Los M, Burek CJ, Stroh C, et al. Anticancer drugs of tomorrow: apoptotic pathways as targets for drug design. Drug Discov Today 2003; 8:67–77. 144 Hersh EM, Metch BS, Muggia FM, et al. Phase II studies of recombinant human tumor necrosis factor alpha in patients with malignant disease: a summary of the SouthWest Oncology Group Experience. J Immunother 1991; 10:426–431. 145 Eggermont AMM. TNF registered in Europe: does TNF get a second chance. J Immunother 2000; 23:505–506. 146 Wiley SR, Schooley K, Smolak PJ, et al. Identification and characterization of a new member of the TNF family that induces apoptosis. Immunity 1995; 3:673–682. ◆147 Ashkenazi A. Targeting death and decoy receptors of the tumour-necrosis factor superfamily. Nat Rev Cancer 2002; 2:420–430. ●148 Ichikawa K, Liu W, Zhao L, et al. Tumoricidal activity of a novel anti-human DR5 monoclonal antibody without hepatocyte cytotoxicity. Nat Med 2001; 7:954–960. ●149 Jimenez GS, Nister M, Stommel JM, et al. A transactivationdeficient mouse model provides insight into Trp53 regulation and function. Nat Genet 2000; 26:37–43. ◆150 Vogelstein B, Kinzler KW. P53 function and dysfunction. Cell 1992; 70:523–526. ◆151 Haupt S, Haupt Y. Manipulation of the tumor suppressor p53 for potentiating cancer therapy. Sem Cancer Biol 2004; 14:244–252. ◆152 Haupt S, Berger M, Goldberg Z, Haupt Y. Apoptosis – the p53 network. J Cell Sci 2003; 116:4077–4085. ●153 Heise C, Sampson-Johannes A, Williams A, et al. ONYX015, an E1B gene-attenuated adenovirus causes tumorspecific cytolysis and antitumoral efficacy that can be augmented by standard chemotherapeutic agents. Nat Med 1999; 3:639–645. ◆154 Whitesell L, Lindquist SL. Hsp90 and the chaperoning of cancer. Nat Rev Cancer 2005; 5:761–772. ●155 Banerji N, O’Donnell A, Scurr M, et al. Phase I pharmacokinetic and pharmacodynamic study of 17-allylamino, 17-demethoxygeldanamycin in patients with advanced malignances. J Clin Oncol 2005; 23:4152–4161. ◆156 Burger AM, Seth AK. The ubiquitin-mediated protein degradation pathway in cancer: therapeutic implications. Eur J Cancer 2004; 40:2217–2229. 157 Hideshima T, Richardson P, Chauhan D, et al. The proteosome inhibitor PS-341 inhibits growth, induces
158
●159
◆160
◆161
◆162
163
164
165
◆166
●167
168
169
◆170
171
172
173
174
apoptosis, and overcomes drug resistance in human multiple myeloma cells. Cancer Res 2001; 61:3071–3076. LeBlanc R, Catley LP, Hideshima T, et al. Proteosome inhibitor PS-341 inhibits human myeloma cell growth in vivo and prolongs survival in a murine model. Cancer Res 22002; 62:4996–5000. Richardson PG, Sonneveld P, Schuster MW, et al. Bortezomib or high-dose dexamethasone for relapsed multiple myeloma. N Engl J Med 2005; 352:2546–2548. Evans RM. The steroid and thyroid hormone superfamily. Science 1988; 240:889–895. Chambon P. The molecular and genetic dissection of the retinoid signalling pathway. Rec Prog Horm Res 1995; 50:317–332. Evans TRJ, Kaye SB. Retinoids: present role and future potential. Br J Cancer 1999a; 80:1–8. Giguere V, Ong ES, Segui P, Evans RM. Identification of a receptor for the morphogen retinoic acid. Nature 1987; 330:624–629. Allenby G, Bocquel M-T, Saunders M, et al. Retinoic acid receptors and retinoid X receptors: interactions with endogen-ous retinoic acids. Proc Natl Acad Sci USA 1993; 90:30–34. Horwitz KB, Jackson T A, Rain DL, et al. Nuclear hormone receptor co-activators and co-repressors. Mol Endocrinol 1996; 10:1167–1177. Perlmann T, Evans RM. Nuclear receptors in Sicily: all in the famiglia. Cell 1997; 90:391–397. Nagpal S, Saunders M, Kastner P, et al. Promoter context – and response element-dependent experimental neoplasms. Southern Research Institute Booklets 1986; 13, 1986 and 2–13, 1987. Nagy L, Thomazy VA, Shipley GL, et al. Activation of retinoid X receptors induces apoptosis in HL-60 cell lines. Mol Cell Biol 1995; 15:3540–3551. Schadendorf D, Kern MA, Artuc M, et al. Treatment of melanoma cells with the synthetic retinoid CD437 induces apoptosis via activation of AP-1 in vitro and causes growth inhibition in xenografts in vivo. J Cell Biol 1996; 135:1889–1898. Lotan R. Retinoids in cancer chemoprevention. FASEB J 1996; 10:1031–1039. Tallman MS, Andersen JW, Schiffer CA, et al. All-trans retinoic acid in acute promyelocytic leukaemia. N Engl J Med 1997; 337:1021–1028. Majewski S, Szmurlo A, Marczak M, et al. Inhibition of tumour-cell induced angiogenesis by retinoids, 1,25-dihydroxyvitamin D3 and their combinations. Cancer Lett 1993; 75:35–39. Hansen CM, Frandsen TL, Brunner N, et al. 1α 25dihydroxyvitamin D3 inhibits the invasive potential of human breast cancer cells in vitro. Clin Exp Metastasis 1994; 12:195–202. Bower M, Colston KW, Stein RC, et al. Topical calcipotriol treatment in advanced breast cancer. Lancet 1991; 337:701–702.
References 111
●175
176
177
178
◆179
180 181
◆182
◆183
184
185
●186
187
188
Gulliford T, English J, Colston KW, et al. A phase I study of the vitamin D analogue EB1089 in patients with advanced breast and colorectal cancer. Br J Cancer 1998; 70:6–13. Evans TRJ, Colston KW, Lofts FJ, et al. A Phase II trial of the vitamin D analogue seocalcitol (EB1089) in patients with inoperable pancreatic cancer. Br J Cancer 2002; 86:680–685. Dalhoff K, Dancey J, Astrup L, et al. A phase II study of the vitamin D analogue seocalcitol in patients with inoperable hepatocellular carcinoma. Br J Cancer 2003; 89:252–257. Wieder R, Novick SS, Hollis BW, et al. Pharmacokinetics and safety of ILX23-7553, a non-calcaemic vitamin D3 analogue, in a phase I study of patients with advanced malignances. Invest New Drugs 2003; 21:445–452. Bird A. DNA methylation patterns and epigenetic memory. Genes Dev 2002; 16:6–21. Goll MG, Bestor TH. Eukaryotic cytosine methyltranferase. Ann Rev Biochem 2004; 74:481–514. Hendrich B, Tweedie S. The methyl-CpG binding domain and the evolving role of DNA methylation in animals. Trends Genet 2003; 19:269–277. Jenuwein T, Allis CD. Translating the histone code. Science 2001; 293:1074–1080. Jones PA, Baylin SB. The fundamental role of epigenetic events in cancer. Nat Rev Genet 2002; 3:415–428. Santini V, Kantarjian HM, Issa JP. Changes in DNA methylation in neoplasia: pathophysiology and therapeutic implications. Ann Intern Med 2001; 134:573–586. Leone G, Voso MT, Teofili L, Lubbert M. Inhibitors of DNA methylation in the treatment of haematological malignances and MDS. Clin Immunol 2003; 109:89–102. Issa JP, Gharibyan V, Cortes J, et al. Phase II study of lowdose decitabine in patients with chronic myelogenous leukaemia resistant to imatinib mesylate. J Clin Oncol 2005; 23:3948–3956. Zhou L, Cheng X, Connolly BA, et al. Zebularine: a novel DNA methylation inhibitor that forms a covalent complex with DNA methyltransferases. J Mol Biol 2002; 321:591–599. Fong MZ, Wang Y, Ai N, et al. Tea polyphenol(-)epigallocatechin-3-gallate inhibits DNA methyltransferase and re-activates methylation – silenced genes in cancer cell lines. Cancer Res 203; 63:7563–7570.
189 Brueckner B, Garcia BR, Siedlecki P, et al. Epigenetic reactivation of tumor suppressor genes by a novel smallmolecule inhibitor of human DNA methyltransferases. Cancer Res 2005; 65:6305–6311. ◆190 Lyko F, Brown R. DNA methyltransferase inhibitors and the development of epigenetic cancer therapies. J Natl Cancer Inst 2005; 97:1498–1506. ◆191 Roth SY, Denu JM, Allis CD. Histone acetyltransferases. Ann Rev Biochem 2001; 70:81–120. ◆192 Thiagalingam S, Cheng KH, Lee HJ, et al. Histone deacetylases: unique players in shaping the epigenetic histone code. Ann NY Acad Sci 2003; 983:84–100. ◆193 Johnstone RW. Histone-decaetylase inhibitors: Novel drugs for the treatment of cancer. Nat Rev Drug Discov 2002; 1:287–299. ◆194 Lindemann RK, Gabrielli B, Johnstone RW. Histonedeacetylase inhibitors for the treatment of cancer. Cell Cycle 2004; 3:779–788. ◆195 Kelly WK, O’Connor OA, Marks PA. Histone deacetylase inhibitors: from target to clinical trials. Expert Opin Investig Drugs 2002; 11:1695–1713. ◆196 Harley CB, Villeponteau B. Telomeres and telomerase in ageing and cancer. Curr Opin Gen Dev 1995; 5:249–255. 197 Morin SG The implication of telomerase biochemistry for human disease. Eur J Cancer 1997; 33:750–760. ◆198 Hurley CB, Sherwood SW. Telomerase, checkpoints and cancer. Cancer Surv 1997; 29:263–284. ◆199 Parkinson EK, Newbold RF, Keith WN. The genetic basis of human keratinocyte immortalization in squamous cell carcinoma development: the role of telomerase reactivation. Eur J Cancer 1997; 33:727–734. 200 Shay JW, Bacchetti S. A survey of telomerase activity in human cancer. Eur J Cancer 1997; 33:781–791. ◆201 Sharma S, Raymond E, Soda H, et al. Preclinical and clinical strategies for development of telomerase and telomere inhibitors. Ann Oncol 1979; 8:1063–1074. ◆202 Keith WN, Bilsland A, Hardie M, Evans TRJ. Drug Insight: cancer cell immortality – telomerase as a target for novel cancer gene therapies. Nat Clin Practice Oncol 2004; 1:88–96.
6 Tumour imaging in oncology ANJU SAHDEV AND RODNEY H. REZNEK
Introduction Imaging modalities Clinical application of imaging techniques to specific tumours
112 112
Imaging treatment response References
139 140
124
INTRODUCTION Imaging plays an integral and diverse role in oncology, including in diagnosis, staging, planning treatment, surveillance, assessing response to treatment and its complications and identifying recurrent disease. It also helps to target therapies such as radiofrequency/thermal/cryotherapy ablation and embolization of tumours. It plays a vital part in monitoring response to treatment, often acting as a surrogate endpoint for drug trials. Recent advances in functional and molecular imaging have widened the scope and application of imaging by combining the exquisite anatomical detail provided by cross-sectional imaging with information concerning tissue and cellular activity obtained by scintigraphy and molecular imaging. The role of imaging is specific to each cancer type and site. Radiologists with a special interest in imaging patients with cancer need to develop a detailed knowledge of its natural history and response to treatment while keeping abreast of new developments in imaging. For all these reasons, imaging now has a central role in the practice of multidisciplinary teams caring for cancer patients. This chapter provides a brief description of basic principles and recent advances in commonly used imaging modalities. It discusses the optimal use of imaging as applicable to individual cancer types, with particular emphasis on detection (including screening and characterization), staging, planning and monitoring treatment and identifying recurrent disease.
IMAGING MODALITIES Conventional radiographic techniques The basic principle used in medical imaging is that the degree of X-ray absorption (attenuation) of a beam is based
on the atomic density and thickness of material through which it is passed. Dense materials like bone and contrast agents (iodine and barium) attenuate most, whereas air attenuates few of the X-rays within the applied beam. Once X-rays have travelled through the material they are ‘collected’ by X-ray film. X-ray film is photographic film consisting of a photographically active or radiosensitive emulsion coated by plastic. It contains silver halide, that on exposure to light or X-rays, is converted to silver, which appears black. Therefore, in conventional radiographs, less dense materials such as air appear black, as most X-rays travel through, exposing the film and converting silver halide to black silver. The converse is true for dense materials, which will appear white, as fewer X-rays are available to expose the film. Conventional techniques include plain films, fluoroscopy, intravenous urograms (IVUs) and barium studies.1 The main advantage of these techniques is their availability worldwide. The techniques require minimal technical training, support or expertise and can often provide adequate diagnostic information. For example, the International Federation of Gynaecology and Obstetrics (FIGO) staging of cervical cancer includes an IVU and barium enema for the assessment of bladder and rectal invasion respectively.2 Intravenous urograms are also important in the detection of transitional cell carcinomas (TCCs) of the urinary tract. The main disadvantage of plain films is poor sensitivity, as the inherent contrast resolution is too poor to allow separation of soft-tissue structures; they perform well at sites where the tumour lies against air or bone. However, here too, lesions have to be significantly larger in size for detection on plain films compared to cross-sectional imaging. Due to good contrast between fat, calcification and tumour, conventional X-rays in the form of mammography play an important role in breast cancer. Barium studies, particularly double-contrast barium studies, utilize the contrast between
Imaging modalities 113
barium-coated mucosa and air within the bowel lumen for the detection of upper and lower gastrointestinal tumours. These are increasingly being replaced by endoscopy, endoscopic ultrasound (EUS) and computed tomography (CT). DIGITAL RADIOGRAPHY
The basic principle of digital imaging is the same as that of conventional radiography, but it utilizes digital apparatus. For digital radiography, the radiograph is produced on a special phosphor screen and then read by laser, converting the image either onto laser film or displayed on television monitors. This can be done in real time as in fluoroscopy, or the images can be stored in local digital networks or PACS systems by modified digital converters. The benefits of digital radiography are clearly the varied forms of image output that can be achieved and the long-term storage and distribution of images. As digital images can be manipulated and post-processed, technically poor images do not require repeat X-rays, thus saving patients unnecessary radiation exposure. Post-processing also allows threedimensional and multi-planar reconstruction, image subtraction, as in digital subtraction angiography, and dual energy subtraction techniques used for screening and characterizing lung masses.3–4
Ultrasound Sound waves are longitudinal waves which once generated continue in their original direction until they are reflected, refracted or absorbed. In medical imaging, ultrasound waves (US) – sound frequencies between 2.5 and 20 Megahertz (MHz) – are generated by transducers that can convert an electrical signal into US and, once reflected from tissues, back into electrical signal. A thin layer of acoustic jelly is placed over the area to be scanned in order to obtain effective acoustic coupling between the skin and transducer. The most important component of the transducer is a thin (approximately 0.5 mm) piezoelectric crystal element located near the surface of the transducer. When the transducer is in contact with a skin surface and a voltage pulse is applied across the transducer, the piezoelectric crystal vibrates, generating sound waves that are transmitted through the body. A backing block dampens the vibrations to prime the transducer for its second function, which is to detect reflected sound waves from the body. The returning sound waves induce a voltage in the transducer and are converted into a grey-scale image. The reflection of sound waves is greatest where there is a large difference in acoustic impedance (density and elasticity) of two tissues. Soft tissues reflect more echoes than fluid and appear brighter (or hyperechoic), whereas fluid appears dark (or hypoechoic). The elapsed time for the wave to return allows estimation of distance or depth, providing spatial information in the construction of the image. Bone and air reflect all sound and therefore structures beyond them cannot be imaged.
The Doppler effect is the shift in frequency of a wave when the source moves relative to the receiver, e.g. blood flow in vessels. When blood flows towards the transducer, the reflected sound wave has a higher frequency than the transmitted frequency. Conversely, when blood flows away, the reflected wave has a lower frequency. This frequency shift can be measured and blood-flow velocity and direction can be calculated. When pulsed Doppler is applied, the transducer emits bursts of sound, between which it remains quiet to receive reflected sound. This allows the combination of an US image and a Doppler trace (duplex scan), providing depth, site and velocity information, which is accomplished by transmitting a sound beam and recording the returning waves. The difference in frequencies of the emitted and returning waves usually falls in the audible range. Spectral analysis of the returning Doppler signal allows evaluation of blood flow in normal and stenotic vessels.5 INTRACAVITY ULTRASOUND
Intracavity US is performed using small transducers that can be introduced into body cavities. The commonest clinical applications are transvaginal (TVUS), transrectal (TRUS), transoesophageal and endoscopic US (EUS). Intracavity US utilizes ultra-small curved transducers that have a small field of view (a few centimetres only) but provide high-resolution images. Endoscopic US was introduced in the early 1980s. Its uses have since expanded to evaluate upper and lower gastrointestinal tracts, pancreato-biliary tumours, diagnosis and staging oesophageal and gastric tumours and evaluation of mediastinal nodes in lung cancer. Although this has limitations and is greatly dependent on the skill of the operator, one of its main advantages over other cross-sectional techniques is that it allows simultaneous tissue biopsy of detected lesions and offers the potential of localized treatment options. ULTRASOUND CONTRAST AGENTS
Ultrasound contrast agents (UCAs) are micro-gas bubbles that when exposed to an US beam reflect sound several times more than blood. The UCAs currently used in diagnostic US have a micro-bubble structure consisting of inner gas bubbles stabilized by an outer protein shell. They remain within the intravascular space and are therefore blood-pool agents that enable the display of parenchymal microvasculature with real-time enhancement. There are currently three approved agents, one micro-air bubble and two lowsolubility gas bubbles. The contrast agents are administered to the patient via a peripheral vein and US of the target organ is performed immediately and/or at timed intervals. After injection, they traverse even the smallest capillaries and therefore can be used to detect lesions not seen on conventional US (Fig. 6.1). The assessment of micro-bubble enhancement requires contrast-specific US imaging mode software, which is generally available and compatible with newer US machines. Normal tissues produce linear signals,
114 Tumour imaging in oncology
metastases within the homogeneously enhanced background liver parenchyma can improve their detection and also improve differentiation from hepatocellular carcinoma or benign focal lesions. All malignant lesions generally show negative enhancement or washout during the extended portal venous phase, and this pattern is useful to differentiate them from benign lesions.6 HIGH-INTENSITY FOCUSED ULTRASOUND
(a)
High-intensity focused ultrasound (HIFU) is a completely non-invasive treatment technique that utilizes a highly focused US beam to coagulate a well-defined tissue volume by heating it to above 50°C. This temperature is maintained for 1–2 seconds, resulting in cell death. A single treatment results in cell death of 0.5 mL of tissue. There is a linear demarcation between coagulated and normal tissue. This method has been used to treat small malignant lesions of the prostate, breast, liver and kidneys. Treatment efficiency is monitored by magnetic resonance imaging (MRI), as US cannot differentiate between treated and untreated tumour or between coagulated and normal tissue. Improvements in technique and technology will make this a promising technique in the future.8
Computed tomography
(b)
Figure 6.1 Contrast enhanced ultrasound. (a) 49-year-old woman with breast cancer. Conventional ultrasound of the liver showing a diffuse heterogenous echotexture of the liver detected during staging for breast cancer. No definite local lesions can be appreciated. (b) After administration of Sonovue™ (Bracco, Milan, Italy) multiple focal hypoechoeic lesions consistent with multiple liver metastases are seen. The normal liver is hyperechoeic and enhances after contrast administration.
but micro-bubbles produce a non-linear signal. The specific US imaging modes are based on separation of the linear from the non-linear signal enhancing the tissue images containing micro-bubble contrast agents.6,7 Most benign liver masses show characteristic features on contrast enhanced US (CEUS), allowing an accurate diagnosis. Hepatocellular carcinoma is typically characterized by increased arterial flow with frequent dysmorphic tumour vessels and decreased portal venous flow. However, negative enhancement in the portal phase is often not obvious until late (2 minutes). On the other hand, metastasis shows prompt brief arterial hypervascularity, with either a rim or diffuse pattern and rapid washout, seen as perfusion defects during the portal venous phase. This pattern of complete rapid washout of
Computed tomography was first described by G.N. Hounsfield, a research scientist at EMI Limited, England, in 1972 at the Annual Congress of the British Institute of Radiology. It is currently the most commonly used imaging tool in oncology. Computed tomography is similar to conventional radiography in that the image is based on differential X-ray absorption within the body. The amount of X-ray absorption is dependent on the tissue atomic density and thickness. The difference is that the X-ray beam is narrow and passes through a thin section of the body at a time. The attenuated beam is collected at the other side by detectors that are capable of differentiating very subtle differences in tissue density. Therefore CT has much greater contrast resolution than plain radiography; it has the ability to separate 2000 or more densities, compared to about 20 on plain films. The CT images are generated by computers from information collected by the detectors. The images are a grey-scale representation of anatomy, and the computer allocates electronic CT values to all structures within the scanned area and arranges them as a spatial image. These CT numbers, called attenuation values, are measured in Hounsfield units (HU) and are calculated based on the linear X-ray absorption coefficient of tissues compared to water. To generate a grey scale, water has a standardized Hounsfield number of 0 HU, air 1000 HU, and bone 1000 HU. Fat has between 80 and 100 HU. The normal pre-contrast liver measures between 40–60 HU, but this decreases with increasing fatty replacement in the liver. Blood density varies with the age of the blood; acute
Imaging modalities 115
haematomas measure 50–60 HU, with a fall in the density value as the blood ages. Although the tissue contrast of CT is good, it can be improved further by the use of oral and intravenous contrast media. Oral contrast outlines the lumen of bowel, allowing better separation of normal bowel from mesenteric, serosal or peritoneal disease. Intravenous contrast medium is usually administered into a large peripheral vein, usually the antecubital fossa, at variable rates. Intravenous contrast is used for angiographic detail and characterization at different phases of the circulation. Traditionally, the term ‘enhancement’ refers to the appearance following intravenous injection of contrast media. Post-processing and manipulation of the grey-scale images allow optimal visualization of different tissues of interest to detect focal lesions. This is done by selecting the mean value closest to the attenuation value of the tissue of interest (window level) and selecting a narrow range of attenuation values around the mean value (window width). Tissue densities outside this range will not be visualized and will be either white or black. For example, optimized settings for the routine visualization of soft tissues will be best performed at a window width (WW) of 350 HU and a window level (WL) of 40 HU. For the lung parenchyma, the WW is 1500 HU and the WL is 700 HU, for bone the WW is 2000 HU and the WL is 500 HU, for liver the WW is 250 HU and the WL is 40 HU. Post-processing also allows the direct measurement of tissue attenuation values on CT, useful when characterization of masses is required. This is done by placing an elliptical or circular electronic cursor over the region of interest (ROI) and the computer measures the mean attenuation value in the ROI. On unenhanced scans (non-contrast), this allows characterization, as the attenuation values of soft tissue, fat, water and blood will vary. Measurements of ROI values before and after the administration of intravenous contrast media allow interrogation of the enhancement properties of tissues, providing information on their vascularity. This has a wide application in tumour imaging, as enhancement properties of liver lesions allow characterization of haemangiomas, hepatocellular carcinomas and metastases. Adrenal lesions are characterized by their CT attenuation values and contrast washout properties. Tumour angiogenesis evaluation also relies on measurements before and after different phases of contrast enhancement. The major advantage of CT is that it provides very high quality images of different organ systems with little or no overlap. It separates normal from abnormal tissues and can provide limited physiological information by means of quantitative measurements of perfusion, blood volume and vessel permeability using intravenous contrast agents. It is cost effective, widely available and an easily reproducible technique. Its major disadvantages are the radiation burden – CT accounting for the greatest proportion of medical radiation exposure in the Western world – and the fact that intravenous iodine-based contrast agents are nephrotoxic and can cause mild to fatal anaphylactic reactions.9
Figure 6.2 The principle of spiral CT. The patient on the CT table travels through the scanner at a pre-determined rate. The CT gantry containing the x-ray tube and detectors rotates continuously around the patient from the beginning to the end of the scan area, thereby obtaining a volume acquisition of information. The movement of the gantry forms a spiral or helix providing the term ‘spiral CT’.
CONVENTIONAL SINGLE-SLICE CT
Conventional scanners acquire data on a ‘slice-by-slice’ method, the X-ray tube rotating around the patient and providing information from a single slice; the X-ray stops and the patient moves into position for the next scan. This continues until all the slices in the ROI are obtained. The slice thickness is set at the start of the scan and, to minimize the radiation burden to the patient, a gap between the slices is introduced. Slice-by-slice scanning is time consuming and prone to respiratory and motion artefacts, as patients are required to be still and in suspended respiration during each slice acquisition. During image generation, the computer has to average the densities at tissue and slice interfaces, introducing the partial volume effect, which may appear as false lesions. SPIRAL (HELICAL) CT
Since its introduction, spiral CT has dramatically altered the approach to CT scanning. The most direct impact has been in the increased speed of scanning. The revolutionary difference lies in the introduction of a slip-ring gantry mechanism that no longer requires a delay between scans. In contrast to conventional CT, in which the gantry rotates around a stationary patient, in helical CT the patient moves through a continuously rotating gantry that continuously produces X-rays. The gantry consists of multiple parallel rings of X-ray detectors and the X-ray tube. Whereas conventional CT requires multiple views of precisely the same structure through at least 180° (slice-by-slice technique), spiral CT relies on a 360° segment of the patient associated with one gantry rotation and is therefore volume acquisition (Fig. 6.2). After mathematical interpolation, the data from the detectors are displayed as axial images, similar to conventional CT.10 The most important clinical advantage of spiral CT is the speed of data acquisition, which allows the chest and abdomen to be scanned in a single breath hold, compared to several minutes on conventional CT. Rapid scanning
116 Tumour imaging in oncology
image reconstruction, virtual endoscopic procedures (virtual CT colonography, bronchoscopy, cystoscopy) and angiography (Fig. 6.4). This additional information assists in providing more accurate and early lesion detection and localization and aids pre-surgical and radiotherapy treatment planning for patents with cancer.12,13 FUSION/CO-REGISTRATION TECHNIQUES
Figure 6.3 The principle of multidetector CT. The CT gantry and the patient movement of multidetector scanners are similar to the principle of single row detector spiral scanners shown in figure 6.2. In multidetector CTs the detectors form up to 128 rows. This allows data from more than one slice position to be collected, simultaneously dramatically reducing the scan times.
Developments in CT software have allowed co-registration of other imaging modalities onto CT images. This is particularly useful in providing functional images when co-registered with nuclear medicine techniques and positron emission tomography (PET). In oncology, this provides useful information in treated lymphoma and teratoma patients with residual masses in the chest and retroperitoneum by differentiating active from sterile residual masses.14 Fusion of CT and MRI is particularly useful in mapping gynaecological, particularly cervical, cancer and prostate tumours that are best demonstrated on MRI but receive CT-guided radiotherapy. Accurate co-registration allows targeted radiotherapy and avoids side effects of radiotherapy to the bowel, bladder and other pelvic structures.15 COMPUTER-AIDED DIAGNOSIS
allows organs to be scanned at differing phases of contrast enhancement, which in turn is very important in characterizing lesions. In the liver, hypervascular lesions may not be detected at all in the equilibrium phase on conventional CT, but may be detected on arterial spiral scanning.11 Rapid scanning also reduces cardiac pulsation artefacts, improving lesion detection on chest CT. The reduction in respiratory artefact allows improved three-dimensional image reconstruction and two-dimensional multi-planar image reformatting. Post-processing of the acquired data allows reconstruction of scan slices thinner than the original acquired scan thickness. This clearly has a benefit of lower radiation burden for the patient and improves spatial resolution.
MULTI-DETECTOR CT
Recently, the ability to acquire data from more than one slice simultaneously has been developed using parallel rows of detectors. Scanners are now capable of obtaining 4–128 slices in a single X-ray-tube rotation (Fig. 6.3). Data are therefore obtained much faster than with a single-slice spiral scanner. The advantages of multi-detector CT include fast scanning times, which are particularly useful in the paediatric patient, for whom sedation or general anaesthesia may be avoided or used for significantly less time. The entire chest, abdomen and pelvis can be scanned in less than 2 minutes. Much thinner slices can be obtained, providing greater spatial and temporal resolution and a reduction in partial volume effects. Post-processing of the data allows exquisite three-dimensional reconstruction, multi-planar
Computer-aided diagnosis (CAD) can be defined as diagnosis made by a radiologist using computer-aided image analysis. The computer acts as a ‘second reader’ of the images and lists the detected abnormalities. It helps increase the sensitivity by detecting lesions that may be missed by the radiologists. The main clinical applications are in reporting screening mammography in the detection of breast cancer, identifying pulmonary nodules, and CT pneumocolonography for the detection of small polyps. The CAD systems are especially helpful where a large volume of low-incidence examinations needs to be screened rapidly.16 FUNCTIONAL CT AND TUMOUR ANGIOGENESIS
The combination of fast CT scanning techniques using helical and multi-detector computed tomography (MDCT) and commercially available CT perfusion software has allowed CT to explore tumour vasculature and tumour biology. Computed tomography assesses vascular tumour physiology by measuring the temporal changes in X-ray attenuation occurring in major vessels and tissues after intravenous iodine contrast administration. It is possible to derive timeattenuation curves that depict changes in attenuation and iodine concentration and hence measure perfusion, relative blood volume, vascular permeability and relative extravascular space. The term functional CT (fCT) is applied to the measurement of these values using CT.17 Tumour angiogenesis is characterized by increased numbers of small blood vessels. This increased vessel density increases tumour perfusion and blood volume. The vessels also have defective basement membranes that allow
Imaging modalities 117
(a)
(d)
(c)
(b)
Figure 6.4 Multidetector CT and post-processing and reconstruction images. (a) Coronal oblique reformatted image demonstrating a left renal carcinoma (arrow) and an adjacent simple cyst (arrow head). (b) Surface volume rendered image showing the relationship of the ribs and underlying kidney and tumour. (c) Surface volume rendered image with the ribs removed demonstrating renal surface and the relationship of the renal cyst and carcinoma. (d) CT angiogram demonstrating the renal artery and vein.
macromolecules to leak and increase the extravascular space. Thus values obtained by fCT can be correlated with microvascular changes in tumour angiogenesis. Clinical applications of fCT include assessment of liver perfusion to detect metastases in colorectal cancer, and characterization of pulmonary nodules where the microvasculature density in malignant lesions is much higher than in benign nodules. Tumour angiogenesis has also been studied with contrast enhanced US, dynamic MRI and PET, but fCT maintains the best and most direct correlation between tumour angiogenesis and functional measurements. The main disadvantages of fCT are the limited area that can be interrogated and the high radiation exposure.18
Magnetic resonance imaging Magnetic resonance imaging is a non-ionizing-radiationbased, cross-sectional imaging modality that utilizes weak signals emitted by body tissues when the body is placed in a strong magnetic field and radiofrequency (RF) pulses are applied. These emitted signals are then converted into greyscale images. The signals are predominantly associated with protons from water and, to a lesser extent, fat. The principle underlying MRI is the weak magnetic property of hydrogen protons in water. Within very strong magnetic fields (such as large circular MRI magnets), the spinning hydrogen protons align themselves with the external magnetic field. When a short RF pulse with the same frequency as hydrogen protons is applied, the magnetization direction is
altered and all the protons align themselves in the same direction (in-phase). A weak signal is emitted by the aligned protons, and is detected and amplified by a receiver coil (antenna) placed around the patient. As soon as the RF pulse is stopped, the signal begins to decay. This loss of signal depends on two factors: (a) the protons return to their original alignment in the static magnetic filed (spin–lattice relaxation), and (b) interactions of nearby protons disrupt local proton magnetic fields (spin–spin relaxation). The spin–lattice and spin–spin relaxation signals are converted into T1 and T2 relaxation times respectively by means of a mathematical conversion. The strength of the signals depends on the T1 and T2 relaxation properties of the tissue, the number of free protons (proton density) and the number of moving protons, especially flowing blood or cerebrospinal fluid. T1 relaxation is short in heavy molecules such as fat and protein, whereas smaller molecules with more free protons have long T1 relaxation. T2 relaxation depends on the dissipation of magnetic energy into the local environment. Solids, bone and large rigid molecules do not move and have very short T2 relaxation times, unlike free water protons, which have a long T2 relaxation. Image generation relies on varying the time of applied RF pulses and the time to collect returning signals. Signals can be obtained at times when the T1 or the T2 effects predominate (T1/T2 weighted). Electromagnetic signals emitted by the body tissues decay rapidly. They can be prolonged sufficiently to allow capture by applying spin-echo (SE) or gradient-echo (GRE) RF pulses for image generation. In SE systems the signal is flipped by 90° or 180° several times
118 Tumour imaging in oncology
Table 6.1 MRI tissue characteristics on T1-weighted and T2-weighted images Tissue
T1 signal intensity
T2 signal intensity
Bone and calcium
Low
Low
Air Flowing blood Flowing blood Fat
Low Low High High
Water Blood Deoxyhaemoglobin (acute blood) Methaemoglobin (subacute blood) Haemosiderin (chronic blood) Gadolinium-based contrast agents
Low Mildly low High
Low Low High Intermediate High High Very low Low
High Intermediate to low High
Calcium in some states may be bright on T1 sequences
High, with thin, hypointense rim Low with thick, very low signal rim Reflect host tissue signal
(time of repeat (TR)) and the returning signal echoes are collected at variable times (time to echo (TE)). In GRE, the applied RF pulses flip the pulse signal by less than 90° by variable angles (flip angles). In general, GRE sequences are faster to acquire than SE sequences. SE sequences with a short TR and TE produce images where the T1 effect of tissues is dominant (T1W). A long TR and long TE result in images where the T2 effects are pronounced (T2W). Long TR and short TE produce proton density images that are influenced by T1, T2 and proton density effects of the tissue.19 Recent advances in receiver coil technology and rapid sequences in both SE and GRE have allowed faster imaging times and improved image quality. The major advantages of MRI are its lack of radiation, lack of iodine-based contrast media and its superior tissue contrast resolution. Its multi-planar imaging abilities were an advantage, but this importance has diminished with the advent of multidetector CT, which also provides very good multi-planar imaging. Magnetic resonance imaging has the ability to provide functional information (fMRI), echoplanar imaging, MR perfusion and spectroscopy. These are important, mainly in neuroradiology where correlation of anatomical and functional information assists in planning surgical resection and radiation of brain tumours aiming to preserve vital sensory and motor areas.20 The main disadvantages of MRI include longer imaging times than CT, less widespread availability and a need for specialist training. The combination of a long period in a closed space and noise generated by loud radiofrequency pulses can result in claustrophobia in adults and the studies are difficult for children to tolerate. The performance of MRI is also highly protocol and technique dependent. To achieve reproducibility, imaging protocols need to be standardized.
Spin-echo sequences Gradient-echo sequences Spin echo sequences Fast spin-echo sequences
Intracellular (early subacute blood) Extracellular (late subacute blood)
TISSUE CHARACTERISTICS
On T1W images, fat, complex proteins, heavy metals (gadolinium) and melanin have high T1 signal intensity. Water, air and bone (calcium) have low signal intensity. On T2W images, water has high signal intensity. Fat has low signal intensity on SE T2W images but high signal intensity on fast SE and some GRE sequences. Air and bone have low signal T2W sequences. Blood has very complex appearances on MRI. A summary of the tissue characteristics is provided in Table 6.1. PERFUSION IMAGING
Perfusion MRI measures blood flow through an organ or tissue. Gadolinium contrast agents are intravenously injected rapidly (5 mL/s) and the MRI signals are acquired rapidly. This is done using T2* and echoplanar imaging and analysing the images pixel by pixel. The increase in signal intensity is related to the rate of passage and amount of gadolinium passing through the tissue of interest. Signal intensity–time curves are used to derive blood-flow volumes, mean transit time and overall blood flow. These reflect tissue perfusion, microvessel permeability and extracellular leakage. The clinical applications include characterization of brain and breast tumour vascularity, follow-up of treatment response and study of stroke. At present, there are limited clinical data related to the application of these techniques.21
DIFFUSION-WEIGHTED IMAGING
Diffusion-weighted imaging (DWI) is more sensitive than conventional MRI sequences in discriminating tissues. The
Imaging modalities 119
(a)
(b)
(c)
Figure 6.5 Diffusion weighted imaging (DWI) of thick walled cystic intracranial mass. 32-year-old woman presenting with fever, headaches and focal neurology. Previous history of cervical cancer stage 2B treated with radiotherapy 3 years ago. (a) Axial T2-weighted image showing a large right sided cystic mass with mass effect and surrounding oedema (arrow). (b) Coronal T1-weighted post gadolinium image showing a thick irregular enhancing wall with no central enhancement (arrow). This appearance may represent either a cerebral abscess or cystic metastases. (c) DWI shows central low signal intensity suggesting no restriction of diffusion (arrow). The central tissue therefore does not contain long molecules seen in abscesses and the mass was a cystic metastasis on biopsy and treated by radiotherapy.
degree of mobility of water molecules is dependent on the surrounding tissue characteristics. The motion of water molecules is restricted by high viscosity fluids and limited by cell membranes. Water molecules that are restricted will not respond to gradient pulses applied during the diffusion-weighted scan and thereby retain signal on DWI images. Tissues with highly diffusing water molecules will quickly lose signal (low signal intensity on DWI images) with the application of gradient pulses. The loss of signal as a function of diffusion gradient can be quantified by the apparent diffusion coefficient (ADC) of water and presented as a ‘map’ or MRI image. On ADC maps, areas of high diffusion have high signal intensity and areas of low diffusion have low signal intensity. Although DWI and ADC maps are most commonly applied to stroke imaging, they have been used for differentiating tumours from cerebral abscesses and for monitoring tumour response after treatment (Fig. 6.5). Abscesses and untreated tumours have bound water molecules, and therefore have low diffusion (high signal intensity), whereas treated tumours, due to cellular necrosis, have more free water molecules with more diffusion (lower signal intensity).22 MAGNETIC RESONANCE SPECTROSCOPY
Magnetic resonance spectroscopy (MRS) differs from MRI in that it detects signals from chemical compounds rather than water to evaluate in-vivo biochemistry. A conventional MRI signal is obtained from water and fat. By selecting signals from other metabolites and measuring these
relative to water, a spectrum can be generated that provides important clinical information. This provides a twodimensional spectral trace with the plot of frequency on the horizontal axis and intensity of resonance on the vertical axis, which can be presented as a colour-coded trace or superimposed onto conventional images (MRS imaging or chemical shift imaging). The spectral trace is obtained from a small region of interest by placing a rectangular block over the tissues to be studied within a conventional image. The signals are obtained at variable echo times and can be altered to obtain signals from molecules of interest. Only certain atomic nuclei have biological significance (1H, 31P, 13 C, 19F, as they are constituents of biological molecules. 1H spectroscopy is the most frequently utilized, as the atomic nucleus provides the strongest signal and its detection does not require additional hardware.23 In the brain, N-acetyl acetate (NAA) and choline are molecules of particular interest, as NAA is a structural component of intact neural tissue and choline is a component of cell membranes that results in increased cell turnover. In tumours, NAA decreases and choline increases in concentration. The ratio of NAA:choline confers prognostic information, with a lower NAA:choline ratio being related to a worse prognosis (Fig. 6.6).24 The presence of lactate and lipids in a spectral trace is always pathological but not specific for tumours. Normal prostate tissue contains choline, creatinine and citrate. Citrate is produced by normal glandular tissue, but in prostate cancer the normal production of citrate ceases, thereby elevating the choline:citrate ratio. This is useful in separating benign from malignant processes in the prostate.
120 Tumour imaging in oncology
(a)
(b)
Figure 6.6 50-year-old woman with a glioma in the left frontal lobe. (a) Axial T2-weighted image showing the large glioma with surrounding oedema (arrow). A small square voxel is placed on the mass to acquire a spectral trace of the mass. (b) Spectral trace of the mass demonstrates an elevated choline and N-acetyl-acetate peak but has a high NAA/choline peak in keeping with a low grade glioma.
In patients treated by radiotherapy or anti-androgens, normal tissue production of citrate stops and there is a spurious elevation of the choline:citrate ratio. Magnetic resonance spectroscopy cannot reliably discriminate between recurrent disease and normal treated prostatic tissue.25
Tissue-specific contrast agents MAGNETIC RESONANCE LYMPHOGRAPHY (LYMPH-NODE-SPECIFIC AGENTS)
Nodal metastases have a significant negative prognostic impact in patients with cancer. Although 18F-fluorodeoxyglucose (FDG)-PET has been shown to have high sensitivity and specificity in the detection of nodal metastases in thoracic malignancies, its application in pelvic and abdominal cancers is limited. Cross-sectional imaging modalities have low sensitivities in detecting lymph-node metastases in normal-sized nodes. High-resolution MRI (scan thickness 3 mm) using ultra-small particles of iron oxide (USPIO), which act as lymph-node-specific agents, allows the detection of small lymph-node metastases in patients with primary pelvic cancer. The use of lymph node-specific agents involves conventional imaging of the pelvic tumour and high-resolution GRE T2* sequences in sagittal and coronal oblique planes (parallel to the psoas muscles) prior to administration of USPIO. The USPIO is then administered as an intravenous injection and is taken up by normal macrophages and transported to lymph nodes. Deposited USPIO reduces the signal intensity within normal nodes (Fig. 6.7). In nodes or parts of nodes replaced by tumour metastases, the signal intensity is not reduced, as there is no
normal uptake and deposition of USPIO. Twenty-four hours after contrast administration, normal nodes and malignant tissues have different signal intensities, normal nodes appearing dark and malignant tissue appearing bright on GRE T2* images. Hence MR lymphography requires two scans, the first prior to and the second 24 hours after the administration of intravenous USPIO.26–28 The accuracy of USPIO MR lymphography has been best evaluated in prostate, endometrial and cervical cancers. The threshold for detecting metastases is 2–3 mm in 5–10-mm diameter nodes. The sensitivity and specificity of USPIO for the detection of metastases are 90.5 per cent and 97.8 per cent respectively in prostate cancer. In endometrial and cervical cancers, the sensitivity and specificity are 93 per cent and 97 per cent respectively. For the detection of nodal metastases, the use of USPIO increases the sensitivity without loss of specificity. These results show the excellent potential of MR lymphography as a non-invasive technique in diagnosing nodal metastases not only in pelvic cancers but also in other anatomical sites.29 LIVER-SPECIFIC AGENTS
Liver-specific contrast agents comprise a heterogeneous group that can be divided with respect to their target tissues into hepatocyte-selective or Kupffer-cell-selective (reticuloendothelial system, RES) agents. Hepatocyte-specific agents are ionic metal chelates with weak protein binding (manganese-DPDP, gadolinium-BOPTA, gadoliniumEOB-DTPA. After an intravenous injection, the contrast agents distribute in the extracellular space and are transported to the liver, where they are taken up by hepatocytes or Kupffer cells. On the basis of the paramagnetic properties of gadolinium and manganese, all hepatocyte-specific
Imaging modalities 121
(a)
(b)
Figure 6.7 24-year-old woman with Stage IB2 cervical cancer. (a) Magnified T2*-weighted MRI of the right pelvic side wall showing two small lymph nodes measuring 6 mm and 2 mm in their short axis with high signal before administration of USPIO contrast agents. (b) T2*-weighted image acquired 24 hours after intravenous administration of USPIO. The lymph nodes have uniformly lost signal intensity indicating normal lymph nodes filled with macrophages and with no metastastic deposits.
contrast agents increase the T1 signal intensity of the liver. In contrast, the Kupffer cell-selective agents, which are iron based, cause a strong reduction in signal intensity on T2-weighted images and increase signal intensity on T1weighted images.30,31 Mangafadipir trisodium (Mn-DPDP) reaches peak liver enhancement in 10–15 minutes, with an imaging window of several hours. By increasing the signal intensity of the normal liver on T1-weighted sequences, it increases the conspicuousness of liver lesions that do not contain normal hepatocytes and therefore do not take up the contrast
agent. On post-contrast images, cysts, haemangiomas and most metastatic lesions appear of low signal intensity compared to the surrounding liver. Benign and malignant hepatocellular lesions have a variable appearance depending on the hepatocyte cellular differentiation and content. Mn-DPDP improves the detection rate of liver lesions compared to contrast-enhanced CT and gadolinium-enhanced MRI. Although it offers reliable differentiation between hepatocellular and non-hepatocellular lesions, in clinical practice the differentiation between benign and malignant lesions is of greater importance. This is particularly relevant in hepatocellular lesions, where uptake is demonstrated in regenerating nodules, hepatocellular carcinomas, adenomas and focal nodular hyperplasia. Similar uptake has also been demonstrated in metastases from neuroendocrine tumours.32 Gadobenate dimeglumine (Gd-BOPTA) demonstrates both perfusion properties and hepatocellular selectivity. This enables MR angiography and dynamic perfusion imaging after intravenous administration of Gd-BOPTA. It reaches peak liver enhancement after 10 minutes, followed by a plateau phase of 120 minutes. In this phase the differential uptake into hepatocytes and non-hepatocyte lesions is maximal. The combination of dynamic contrastenhanced MRI and delayed imaging has been shown to increase lesion conspicuity and detectability. Again, differentiation between hepatocyte-predominant benign and malignant lesions is limited.33 Gadoxetate (Gd-EOB-DTPA) differs from Gd-BOPTA in its time to reach peak enhancement in the liver. This takes 10–25 minutes after intravenous injection, and maximum liver-to-lesion contrast is achieved in 20–45 minutes, compared to 60–120 minutes with Gd-BOPTA. The other pharmokinetic properties are similar. Care should be taken with hepatocyte-specific agents, as delayed images alone will miss hepatocyte-predominant lesions, which have signal intensity similar to that of the normal liver. Optimal imaging should include sequential information from dynamic contrast-enhanced and delayed images.34 Kupffer-cell-specific contrast agents are colloidal compounds composed of an iron-oxide core. The core iron particles are typically between 3 nm and 10 nm in size, but the whole particle with the outer protein coat is 50–150 nm. After intravenous administration, the agents are taken up by the RES in the liver (80 per cent) and the spleen (12 per cent). The particles are eliminated in lymph nodes and bone marrow. Although the half-lives are in minutes, the paramagnetic effects persist for several days. Optimal imaging is performed several hours after administration. The molecules are degraded within the macrophages of the RES, and free iron is incorporated into the cellular iron metabolic pathways. The presence of iron in Kupffer cells results in pronounced reduction in signal intensity of normal liver on T2-weighted images (fast spin echo (FSE) and T2* images). This results in an ‘inverse’ contrast effect, lesions not containing Kupffer cells retaining their normal signal but the surrounding normal liver becoming of very low signal intensity on T2 images.35
122 Tumour imaging in oncology
Kupffer cells are present in lesions of focal nodular hyperplasia, hepatocellular adenomas, regenerating nodules and well-differentiated hepatocellular carcinomas and adenomatosis hyperplasia. Hence these lesions lose signal intensity to varying degrees along with normal liver tissue. As with hepatocyte-specific agents, Kupffer-cell agents increase lesion conspicuity and detection rate, but complete characterization remains elusive and there are currently not enough data and experience to eliminate the need for a diagnostic biopsy for certain liver lesions.36 Metastatic deposits in the liver do not contain normal liver hepatocytes and Kupffer cells (except for HCC metastases). Therefore metastases remain of low signal intensity with hepatocyte-specific agents and of high signal intensity with Kupffer-cell agents. The blood-pool phases usually reveal some peripheral increase in contrast uptake due to the variable angiogenesis in liver metastases. Characterization of metastases in patients with a known primary cancer is relatively simple, and the use of liver-specific contrast agents has been shown to increase the number of detected metastases compared to CT and gadolinium-enhanced MRI.30,33,35
surgical clips up to 6 weeks after surgery. Joint prostheses are fixed and generally acceptable but may cause susceptibility artefacts in the images, and occasionally patients may experience discomfort due to heating of the metallic prosthesis. Pregnancy, particularly the first trimester, is a relative contraindication, as there are no studies demonstrating the safety of MRI in vivo in the human fetus. Intravenous administration of gadolinium is also avoided, and its safety in pregnancy has not been confirmed. However, no fetal developmental adverse effects have been reported. The most significant side effect in pregnancy can occur at any magnetic field strength. Supine scanning of a pregnant woman, particularly in the third trimester, may result in aorto-caval compression by the fetus and reduce maternal blood flow resulting in dizziness in the mother and a reduction in placental blood flow to the fetus. At higher field strengths (higher than 2T), in-vitro experiments have demonstrated a small increase in amniotic fluid temperature; the significance of this has not been evaluated in humans. Despite these limitations, MRI and US are the preferred imaging modalities, rather than ionizing radiation (CT), in all stages of pregnancy.40,41
ENDOLUMINAL MRI
Endoluminal MRI has been used most extensively in rectal, prostatic and cervical cancers. It involves placing transmitter/receiver MRI coils in the rectum and vagina. The use of endoluminal coils allows very high-resolution images of the organ of interest. The baseline sequences are T2-weighted images in several planes. In early cervical cancer it provides detailed information about the size and extent of cervical lesions and parametrial spread, optimizing the surgical approach to treatment. Patients with small-volume invasive disease without deep cervical stromal invasion can be treated successfully by radical removal of the cervix, thereby conserving their fertility. In prostate cancer, transrectal MRI offers better loco-regional staging than transabdominal MRI. This is used in patients suitable for radical treatment or localized tumour treatment such as cryotherapy or thermal ablation. The reported accuracies of transrectal MRI staging in prostatic cancer are variable, between 51 and 85 per cent, the results being highly dependent on local radiology and histopathology expertise. Transrectal MRI has been used in benign (inflammatory bowel disease, diverticulitis and adenomatous disease) and malignant lesions of the rectum. In rectal carcinoma it has been used to delineate the tumour and assess invasion into the surrounding mesorectal fascia prior to total mesorectal excision (TME) surgery. In rectal, prostatic and cervical cancers, concurrent imaging of the pelvis is recommended using pelvic phase-array coils to image the pelvic sidewalls for nodal disease.37–39 SAFETY OF MRI
Magnetic resonance imaging is contraindicated in patients with metallic intra-ocular foreign bodies, non-MRIcompatible cerebral aneurysm clips, cardiac pacemakers, neurostimulator devices, cochlear implants and metallic
Nuclear medicine Nuclear medicine imaging has contributed significantly to the diagnosis, treatment planning and evaluation of response to treatment in patients with cancer since the development of modern techniques. It utilizes the body’s own normal and abnormal biochemical and physiological pathways, making it a form of functional imaging. The patient is injected with a gamma-emitting labelled tracer and the distribution of tracer after a period of time is detected by modified gamma cameras containing scintillation crystals that are read by photomultiplier tubes. The patient lies under the gamma detector system (camera) for 20–60 minutes and the whole-body distribution of the tracer is mapped into an image by applying mathematical algorithms. The image is most commonly a planar image, but if tomographic techniques are applied, single-photon emission CT (SPECT) and PET images can be obtained. Table 6.2 summarizes the current common uses of nuclear medicine tests and tracers in oncology.42 POSITRON EMISSION TOMOGRAPHY
Positron emission tomography is a nuclear medicine modality that provides quantitative tomographic images allowing non-invasive functional imaging. It has taken on an increasingly important role in oncology for diagnosis, staging during and after completion of treatment, and detection of recurrent disease. It utilizes biochemical metabolites (the commonest being 2-deoxy-D-glucose), labelled with beta-emitting radiotracers (18FDG), which are injected intravenously. Both glucose and deoxyglucose enter cells via the cell glucose transporters and undergo phosphorylation, but whereas glucose undergoes further
Imaging modalities 123
Table 6.2 Current nuclear medicine tests and tracers in oncology Test
Tracer
Indication
Bone scan
99m
Gallium scan MIBG scan Octreotide scan
67
Thallium scan Monoclonal antibodies
201
Staging bone disease in prostate, breast, lung cancers; detecting symptomatic bone disease in other cancers Staging and treatment response in lymphoma Localization and staging of neuroendocrine tumours Localizes and stages tumours with somatostatin receptors (pancreas, carcinoids, medullary thyroid cancer, neuroblastoma) Brain tumours and osteosarcoma To stage and detect tumour containing tumour antigen (lung, colorectal, ovarian and prostate cancers)
Tc-methylenediphosphate
Ga-chloride I-labelled or 123I-labelled MIBG 111 In-octreotide 131
TI-chloride In-labelled or 99mTc-labelled tumour antibodies
111
enzymatic breakdown, deoxyglucose becomes trapped in the intracellular compartment. Cancer cells over-express glucose membrane transporters and intracellular phosphorylation enzymes, thereby taking up and trapping more glucose and deoxyglucose than normal cells. It must be remembered that FDG is not a specific marker for cancer cells, and activated macrophages are also known to exhibit high tracer uptake. This has significant implications in the analysis of PET studies, as active infection and cancer can behave similarly. The 18F component undergoes beta decay in the body and produces a positron and a neutrino. Once a positron is slowed down, it produces two gamma-rays, which are emitted back-to-back and have energy of 511 KeV each. These decay events are detected by coincidence registration of gamma quanta. Therefore it is not the position of positron emission but the position of annihilation into gamma-rays that is detected by PET. As the positron only travels small distances, in clinical practice the locations of positron emission and annihilation are close together. However, this explains the ‘flare’ effect, in which the tissue source appears to have ill-distinct margins. This is registered by two opposing external detector units, and images are constructed after applying mathematical algorithms (18FDG PET imaging). Detectors for PET imaging are similar to those for conventional nuclear medicine, consisting of scintillation crystals (the commonest is bismuth germinate), which are read out by photo-multiplier tubes. The images created represent the metabolic activity of underlying tissues.42 Positron emission tomography has an advantage over conventional nuclear medicine in that it is possible to measure the effects of attenuation of photons within the body, providing image correction for better qualitative images. It also allows measurement of tracer concentrations within tissues, providing the most commonly used clinical measurement index, the standard uptake value (SUV), which compares lesion activity concentration to average activity concentration of the whole body. The SUV values have been utilized to differentiate between benign and malignant lesions and to prognosticate survival in lung cancer and head and neck squamous cell cancers – in general, the
higher the SUV, the poorer the outcome. Changes in SUV before and after chemotherapy have been compared to assess the effectiveness of the chemotherapy agent.43,44 18 F-Fluorodeoxyglucose-PET does not reliably distinguish between inflammation/infection and malignancy, as FDG accumulates in granulocytes and macrophages. This can lead to limitations in clinical practice, as differentiation between residual or recurrent disease and post-radiation inflammatory changes may be difficult. There are, however, several established clinical applications of PET, including the pre-surgical work-up of patients with small-cell lung cancer, oesophageal cancer, colorectal cancer and metastatic melanoma, for which the addition of PET has been shown to decrease surgical procedures in patients with CT-occult metastases. 18F-Fluorodeoxyglucose-PET has also been used to improve radiation treatment planning by defining more accurate biological target volume, especially in intensitymodulated radiotherapy (IMRT). Its use is also being investigated to predict response to chemotherapy both during and at the completion of treatment.45–47 PET-CT
In clinical oncology, the diagnosis and staging of cancer are traditionally based on CT imaging that relies on detecting anatomical alteration of normal tissues. However, cancer originates as a functional change that precedes anatomical alteration. Positron emission tomography has a high sensitivity for the detection of this functional change, but its spatial resolution for anatomical localization is limited. Over the past decade, software-based algorithms have been developed that allow the fusion and combined display of CT and PET images. However, this requires the superimposition of data acquired retrospectively, and normal variations in the position and activity of bowel, respiration and cardiac function contribute to correlation difficulties.48 The alternative solution of hardware fusion has created PET-CT. This allows the acquisition of CT and PET in a single session without re-positioning the patient between scans. In clinical practice, this involves injecting patients with 18F-FDG-PET tracers at least 1 hour prior to the start
124 Tumour imaging in oncology
of the combined PET-CT. A topogram is then acquired to define the area of interest. Contiguous helical CT scans are acquired through the area of interest. After completion of the last helical CT image, the patient is advanced for the PET emission scan acquired in a caudal–cranial direction and taking up to 1 hour. Unlike conventional CT, due to the long duration of the combined acquisition, patients have their arms by their sides and are allowed to breathe shallowly during the scan. Once the collection of CT data is complete, they are sent to the PET console for attenuation and scatter correction. Once corrected, the emission and CT data can be viewed separately, side by side or fused with the PET images superimposed on the CT images. Early results show that combined PET-CT improves diagnostic accuracy compared with CT and PET alone. At present, even with the best PET technology, imaging is slow and the hardware expensive.49 Future developments in PET include hardware improvements that will allow combined use with multi-detector CT by increasing the acquisition rate of emission data. Faster acquisition would overcome respiratory and cardiac limitations, increasing its application in cardiac viability imaging. In oncology, PET has been used to evaluate gene expression and offer prognostic information. It can be used in combination with numerous tracers that image key molecules and molecular based events. Other tracers, particularly labelled chemotherapeutic agents, have been used to predict the uptake of the agent into primary and metastatic lesions, thereby predicting the effectiveness of the chemotherapeutic agent.50
lesions, tumoral calcification and bony destruction, which help to characterize the lesions.52 Although CT is sufficient for diagnosis in most cases, MRI is without doubt a more sensitive technique, visualizing the posterior cranial fossa and leptomeningeal space better than CT. It is the imaging modality of choice in the diagnosis and pre-surgical planning of brain tumours. It has a higher spatial (2 mm) and tissue contrast resolution and delineates and separates tumour from surrounding oedema better than CT. The presence and degree of necrosis detected on MRI have been shown to be associated with higher-grade tumours.53 The MRI technique is variable, but usually includes T2-weighted, T1-weighted, fluid-attenuated inversion recovery (FLAIR) sequence and gadolinium-enhanced T1 sequences. The imaging planes are also variable, but should include axial and coronal images, ensuring the complete cranial fossa is included. Both CT and MRI have high sensitivities but lack specificity for differentiating tumours from other infective and inflammatory masses and predicting tumour grade. Fluorodeoxyglucose-PET is rarely helpful, as it lacks the sensitivity and spatial resolution to challenge MRI as the investigation of choice. SURVEILLANCE AND RECURRENCE
Tumours of the brain account for 1.6 per cent of all cancers in England and Wales and are the twelfth and fifteenth commonest type of cancer in men and women respectively. Between 1995 and 1997, 2170 new tumours were reported in men and 1780 in women.51 Improvements in detection have probably led to the observed increase in incidence of brain tumours in both men and women, particularly in the 55–64-year age group. The brain is also a common site for metastatic disease from other tumours. However, this discussion focuses on primary brain tumours.
Magnetic resonance imaging is the modality of choice for postoperative surveillance and for the detection of recurrent disease. A routine postoperative study is performed 4–6 weeks after initial surgery.54 If radiotherapy is also administered to the patient, differentiating radiotherapy changes from recurrent disease can be difficult. Both radiation necrosis and recurrent disease occur at the postoperative bed and both can demonstrate similar appearances on conventional and contrast-enhanced MRI. Magnetic resonance spectroscopy is most commonly used to differentiate between tumour recurrence and radiation effects: there is marked elevation of choline in tumour recurrence, whereas marked reduction of choline levels is observed in radiation necrosis. Positron emission tomography demonstrates a reduction of FDG uptake in areas of radiation necrosis and an increase in uptake in recurrent disease. It has a reported accuracy of 85 per cent for detecting tumour recurrence.55 Thallium, SPECT and perfusion weighted imaging (PWI)201 have also been suggested for detecting tumour recurrence; the ultimate diagnosis still requires a brain biopsy.56,57
DIAGNOSIS AND STAGING
RECENT ADVANCES
The introduction of CT 30 years ago revolutionized the detection of brain tumours. It still remains the most frequently used imaging modality to detect space-occupying lesions in the brain. The CT technique should include images acquired before and after the administration of intravenous contrast. The advantages of CT over plain radiography and MRI include better assessment of bone-based
To improve the specificity of CT and MRI in characterizing intracranial masses, newer MRI sequences, MRS, fCT and fMRI, MRI perfusion (PWI) and MRI DWI, have been applied. Diffusion-weighted imaging has been used for lesion characterization and can usually differentiate between abscesses and necrotic tumours, which is not possible by conventional CT and MRI (see Fig. 6.5). It distinguishes between
CLINICAL APPLICATION OF IMAGING TECHNIQUES TO SPECIFIC TUMOURS Brain tumours
Clinical application of imaging techniques to specific tumours 125
Squamous cell carcinomas of the nasopharynx, oropharynx and larynx account for 80–90 per cent of all head and neck cancers.
main advantage is the ability to examine lymph nodes in the neck and, combined with fine-needle aspiration (FNA) of suspicious nodes, to provide cytology. It has a sensitivity of 73 per cent and a specificity of 100 per cent in the detection of nodal metastases in normal-sized nodes, significantly better than for CT and MRI.61 Nodal staging with FDG-PET has also been shown to be more sensitive and accurate than CT or MRI. FDG-PET has a sensitivity of 90 per cent and a specificity of 94 per cent, compared to 82 per cent and 85 per cent respectively for CT and 88 per cent and 79 per cent respectively for MRI.62 It can detect nodal metastases in lymph nodes down to 4 mm, but has the propensity to miss smaller disease. In a small minority of patients (5 per cent), lymph-node metastases are noted in the neck without a known primary source. In this group, FDG-PET (in association with CT) has been shown to identify the occult primary in 40 per cent of the patients (Fig. 6.8).63
DIAGNOSIS AND STAGING
SURVEILLANCE AND RECURRENT DISEASE
Most cancers at presentation are mucosal lesions and undetected on cross-sectional imaging. They are diagnosed and characterized on direct visualization and biopsy. These tumours have the tendency to submucosal spread, and assessment of their extension to the skull base, infratemporal and pterygopalantine fossa cannot be evaluated clinically and requires cross-sectional imaging. Metastatic lymph-node involvement is the most important prognostic factor affecting patient survival, and loco-regional node involvement decreases the 5-year survival to nearly 50 per cent.59 Imaging is used to identify subclinical, retropharyngeal and paratracheal lymph-node involvement. Other clinically important nodal groups are the submandibular, submental, lateral cervical, posterior triangle and supraclavicular chains. Both CT and MRI have similar performance in staging head and neck tumours and should be regarded as complementary tools. On the whole, for high tumours (nasopharyngeal and paranasal) with suspected involvement of the skull base, MRI detects early intracranial extension, perineural spread and marrow involvement better than CT. Computed tomography demonstrates nodal involvement and early cortical bone invasion of the skull base, orbits and nasal sinuses better than MRI.60 Both modalities require thin-section scans (2–3 mm) through the head and neck. Computed tomography has the advantages of performing additional scans during the modified Valsalva manoeuvre and obtaining three-dimensional volumetric data sets that can be reconstructed for planning treatment. Contrast enhancement is essential for both CT and MRI. A significant disadvantage of MRI is its long scanning time, which introduces significant motion artefacts from coughing and prolonged swallowing, frequent in these patients. Ultrasound is commonly used in the examination of the neck and its main contribution is in the detection of primary cancers of the salivary glands and thyroid. Its role in the detection of oral and pharyngeal tumours is limited. Its
After radiotherapy, tissue changes become evident on CT and MRI within 2 weeks and are most pronounced during the first few months, resolving slowly over time. These changes include symmetric thickening of the skin, subcutaneous tissue and platysma, oedema in the retropharyngeal space, symmetric thickening and enhancement of the laryngeal and pharyngeal walls, atrophy of nodes in the Waldeyer’s ring, and atrophy and sialadenitis of the salivary glands. The post-surgical changes include the introduction of reconstructive tissue and material grafts. Immediately after surgery, muscle grafts appear as normal muscle but, gradually undergoing atrophy and fatty replacement due to denervation. Areas of residual fibre-like muscle tissues can be confused with recurrent disease. Due to these post-radiation and post-surgery changes, baseline post-treatment imaging is suggested 3–6 months after completion of treatment.64 Studies using CT have shown that baseline scans can be used to predict long-term outcomes and patients, who demonstrate expected post-radiotherapy changes only, with complete resolution of primary tumour have the best prognosis. CT features of a residual mass 1 cm in maximum diameter and post-radiotherapy reduction of tumour volume by 50%, are predictive of future local disease recurrence.65,66 Fluorodeoxyglucose-PET has been shown to have better sensitivity and significantly better specificity than CT and MRI in detecting residual and recurrent disease. The sensitivity and specificity of FDG-PET are between 88 and 100 per cent and 75 and 100 per cent respectively, compared with 70 and 92 per cent and 50 and 57 per cent for CT and MRI.67 Recommendations have been made to assist in the selection of imaging modalities in patients being investigated for residual or recurrent disease. FluorodeoxyglucosePET should be used when no clinically obvious mass is palpable in patients with non-specific symptoms of recurrence or in patients with a low clinical suspicion of recurrence with advanced cancer at presentation. Computed
densely cellular tumours such as lymphoma and gliomas, but it cannot separate tumour infiltration from oedema, and the appearances of high-grade and low-grade tumours overlap. Perfusion-weight imaging provides information concerning tissue vascularity and can be used to grade tumours into low and high grades. Low-grade tumours have low perfusion, whereas high-grade tumours have increased perfusion. Functional MRI is increasingly used to maximize surgical margins whilst attempting to conserve unaffected brain tissue to retain normal function.58
Head and neck tumours
126 Tumour imaging in oncology
(a)
(b)
(c)
Figure 6.8 67-year-old male patient presenting with cervical lymphadenopathy. (a) Non-contrast enhanced image of the neck demonstrating large volume left sided cervical lymphadenopathy (arrow). On CT alone the primary tumour is not clearly visualised. (b) 18F-FDG PET image showing tracer uptake in the left sided lymphadenopathy and the left oro-pharynx (arrow). (c) CT-PET fusion image accurately delineates the primary tumour lying in the left Waldeyer’s fossa (arrow) (See Plate Section.).
tomography or MRI is advocated when there is a positive or equivocal FDG-PET or to image the extent of disease in biopsy-proven recurrence or a suspicious palpable mass. Early-stage disease with low clinical suspicion precludes imaging studies.68 RECENT ADVANCES
Improvements in imaging have concentrated on differentiating malignant from non-malignant nodal disease in an attempt to improve surgical outcomes but also to avoid extensive neck dissection in NO disease. Dynamic contrast enhancement in MRI and functional imaging techniques in CT and MRI have been evaluated but are not currently in routine clinical use. Early experience with USPIO MRI lymphography suggests that this technique will improve the diagnostic accuracy in the detection of nodal metastases. Colour Doppler ultrasound (DUS) is another method investigated in the diagnosis of cervical metastatic lymphadenopathy that enhances the predictive value of conventional B-mode US. The presence of cortical thickening, structural inhomogeneity, compressed, displaced or nonvisualized hilum and high-resistance arterial flow in the nodal hilum increases the likelihood of metastatic involvement of nodes. Extracapsular extension, which is a significant negative prognostic feature, can also be detected on DUS. There is currently a limited number of studies evaluating DUS and no studies comparing DUS with FDG-PET.69,70
Lung cancer Lung cancer was a relatively rare disease at the beginning of the twentieth century but is currently the commonest cancer
in the world; nearly 90 per cent of cases are caused by tobacco smoking. In 1997, in the UK, there were 21 000 new cases of lung cancer in men, in whom it was the most common cancer, and 13 200 new cases in women, in whom it was the third commonest cancer.51 Imaging provides information about disease extent by defining the extent of local invasion as well determining mediastinal and distant spread. Lung cancers are divided into small-cell (SCLC) and non-small-cell lung cancer (NSCLC), reflecting the marked difference in natural history, therapeutic strategies and response to treatment between the two groups. DIAGNOSIS AND STAGING
Small-cell lung cancer is usually disseminated at presentation in 60–80 per cent of patients.71 Plain radiographs and contrast-enhanced CT of the chest and abdomen are the standard imaging tests.72 In SCLC, staging divides tumours into limited (involving one hemithorax, unilateral or contralateral mediastinal and supraclavicular nodes) or extensive disease (metastatic thoracic and extrathoracic disease). Patients with limited disease receive radiotherapy, whereas extensive disease is treated with systemic chemotherapy. Staging of NSCLC is more complex. Involvement of hilar, mediastinal lymph nodes, chest wall invasion and fissural extension and extra-thoracic disease are crucial features to stratify patients for surgery, radiotherapy or systemic chemotherapy. Despite discrepancies between CT and surgical TNM staging in up to 40 per cent of cases, CT is widely used as a guide for the use of EUS-guided biopsy, mediastinoscopy, MRI and exploratory thorocotomy (Fig. 6.9).73 Both CT and MRI show extensive mediastinal tumour extension and chest-wall invasion equally well, but lesser
Clinical application of imaging techniques to specific tumours 127
(a)
(b)
Figure 6.9 45-year-old male smoker with NSCLC. (a) Non-contrast enhanced CT of the mediastinum demonstrating lymph nodes at the upper limit of normal size in the sub-carinal site (arrow). This is a difficult site to access at mediastinoscopy. (b) Trans-oesophageal ultrasound demonstrates the lymph nodes (arrow) and offers the ability to biopsy these nodes aiding management of the patient.
degrees of invasion are less reliably diagnosed or excluded.74 No other imaging modality, including FDGPET, has proved superior to CT for diagnosing mediastinal invasion.75 For chest-wall invasion, MRI was shown to have a sensitivity of 85 per cent and a specificity of 100 per cent, but when compared to CT in a prospective study, no significant difference was seen.76 The one exception, where MRI is superior to CT, is in the detection of brachial plexus and chest wall involvement in superior sulcus or Pancoast tumours.77 Nodal involvement on CT is based on increase in size. The normal sizes of nodes in the mediastinum are 10–11 mm in the subcarinal and lower paratracheal regions and 7 mm in the upper paratracheal regions. Using 10 mm as the upper limit of normal nodes, CT has a sensitivity and specificity of 50–65 per cent for metastatic involvement;78 MRI does not perform better than CT. The most comprehensive comparative study demonstrated sensitivity and specificity of CT as 52 per cent and 69 per cent respectively, compared to 48 per cent and 64 per cent respectively for MRI.76 Fluorodeoxyglucose-PET is being increasingly used for staging lung cancer; it consistently shows better sensitivity (79 per cent) and specificity (91 per cent) for detecting nodal involvement.79 Optimal staging is achieved by the combined use of FDG-PET and CT, which confirms tumour involvement of enlarged nodes and detects metastases in unenlarged nodes and unsuspected extra-thoracic disease (Figs 6.10 and 6.11). Combined use has a negative predictive value between 92 and 100 per cent and is therefore advised.80 More than 50 per cent of all patients with lung cancer present with distant metastases, with the probability increasing with advancing local stage. Currently the most common strategy to detect extra-thoracic disease is additional CT of the brain and abdomen at the same time as the staging thoracic CT. This is performed with intravenous
contrast enhancement to identify brain, liver and adrenal metastases. With increasing availability of FDG-PET, this is likely to change, as PET has been shown to have higher sensitivity and specificity and is cost effective compared to CT in the detection of all extra-thoracic disease except for in the brain. It has been shown to alter management in 24–65 per cent of patients staged by CT, and prevents unnecessary thorocotomy in 20 per cent. Fluorodeoxyglucose-PET has also been shown to have performance superior to that of 99m Tc bone scans, with sensitivity and specificity of 92 per cent and 99 per cent respectively, compared to 92 per cent and 50 per cent respectively for 99mTc bone scans.81,82 SURVEILLANCE AND RECURRENT DISEASE
There is limited literature and no generally accepted guidelines on imaging for local recurrence in lung cancer. For asymptomatic patients, the routine work-up is a plain chest radiograph only, although CT has been shown to be significantly more sensitive.83 The only imaging modality to have acceptable sensitivity and specificity for the detection of recurrent disease is FDG-PET performed 3–6 months after treatment.84 The false-positive rate is as high as 40 per cent if the imaging is performed close to treatment, due to scarring, fibrosis and radiation changes.82 Recurrent tumour is almost never amenable to cure, and the survival benefit of routine imaging has not been demonstrated.
RECENT ADVANCES
Population screening for lung cancer The use of low-dose CT has been advocated to screen an at-risk population. The initial results demonstrate a diagnostic yield of lung cancer of 2.7 per cent on the first round of screening. Most cancers detected are stage IA, fulfilling
128 Tumour imaging in oncology
(a)
(b)
(c)
Figure 6.10 56-year-old male smoker presenting with a large cavitating central NSCLC. (a) Coronal reformatted non-contrast enhanced CT demonstrating the left sided lung cancer and a right sided supraclavicular enlarged lymph node (arrow). (b) 18F-PET image shows tracer uptake centrally in the lung cancer within viable tissue, in the right supraclavicular lymph node and also in an unsuspected right humeral metastasis (arrow). (c) PET-CT fusion provided anatomical and functional correlation of the active tumour sites (See Plate Section.).
the screening criteria of diagnosing asymptomatic earlystage disease. However, 90 per cent of all detected lung nodules are benign and therefore imaging protocols have been suggested to minimize false-positive thoracotomies. In subsequent rounds of screening, the false-positive rate and the true-positive rate are lower, but remain significant. The implementation of screening is currently being debated and its long-term benefit of reducing mortality and morbidity from lung cancer is still undecided.85,86
Breast Breast cancer is the commonest female cancer in the West. In the UK, there are approximately 41 000 new cases annually.51 Imaging plays an important role in the detection of primary breast cancer, in the characterization of palpable breast masses and in the detection of recurrent disease.
DIAGNOSIS AND STAGING
Breast-cancer screening programmes exist in many countries. Mammography is the screening modality of choice. The screening programmes have shown a significant reduction in mortality of around 25–30 per cent among screened groups compared to non-screened populations.87 However,
the age at which screening commences, the screening interval and the age at which it ceases all vary in different countries and there is currently no consensus as to the optimal strategy. In the UK, screening commences at the age 50 years and is performed every 3 years up to the age of 70 years. The most common mammographic manifestations of carcinoma are stellate breast mass, poorly defined irregular mass, and clustered, multiple, polymorphic and branching breast calcification. Screen-read mammograms are categorized into seven categories: 0 needs additional imaging, 1 normal, 2 benign, 3 probably benign, 4 suspicious, 5 highly suggestive of malignancy, and 6 proven malignancy. Carcinomas are present in 23–34 per cent of category 4 and 81–97 per cent of category 5 mammograms.88,89 Mammography is supplemented by US, biopsy and MRI in difficult cases. Ultrasound is useful to separate benign cystic from malignant solid lesions, and FNA or biopsy can be performed in non-palpable masses. Magnetic resonance imaging is also emerging as a useful tool in the detection of breast cancer. It is especially useful in patients with previous breast cancer and post-treatment distorted breast tissue, in patients with breast prosthesis and in young high-risk women below the national screening age.90 The staging of breast cancer is based on the TNM classification. It is largely clinical, but imaging is increasingly used to delineate the tumour. Ultrasound is more readily available in UK for this purpose, but has been shown to
Clinical application of imaging techniques to specific tumours 129
sequences has been advocated as the most accurate technique for the measurement of primary tumour.92 Computed tomograhy is used to look for lymph-node involvement (in the axilla, internal mammary, supraclavicular and mediastinal nodes) and to identify pulmonary metastases. A liver US is performed to exclude liver metastases. Radioisotope bone scans are performed to exclude bony metastases preoperatively in patients with clinical stage III disease.
SURVEILLANCE AND RECURRENT DISEASE (a)
(b)
Mammography is the standard modality for the follow-up of patients after breast surgery. It identifies microcalcification and in the long term can detect non-palpable ipsilateral or contralateral breast cancers. The Royal College of Radiologists recommends mammography should be performed every 2 years after surgery, whereas The British Association of Surgical Oncology recommends a baseline at 6 months followed by annual mammography for 5 years after surgery. Localized recurrence occurs at 1 per cent per year.93 Magnetic resonance imaging has been used for surveillance, particularly in patients with breast prosthesis and post-surgical scarring.94 Metastatic breast cancer is unusual at the time of presentation in early-stage disease, but can present between 2 and 5 years after diagnosis. Metastatic disease can involve any organ, with nodal disease, bone, liver and lung deposits being the most common. No clinical advantage has been demonstrated in imaging distant disease in asymptomatic patients, and the choice of imaging is dictated by the patient’s presenting symptoms. RECENT ADVANCES
(c)
Figure 6.11 54-year-old male smoker with SCLC. (a) Noncontrast CT of the chest demonstrating left upper lobe collapse (arrow). A separate mass is difficult to define. 5-7 mm lymph nodes are also noted in the left paratracheal and pre tracheal sites (arrow). (b) 18F-PET image demonstrates tracer uptake in a central tumour without uptake in the distal collapsed lung (arrow). (c) The PET-CT fusion image provides anatomical correlation and demonstrates that the lymph nodes have no increased tracer uptake, thus excluding mediastinal metastases (See Plate Section.).
underestimate size due to the irregular margins of the tumour.91 Magnetic resonance imaging has been shown to delineate tumour margins accurately, assess the contralateral breast and detect multi-focal disease not appreciated clinically or by mammography. Magnetic resonance imaging with gadolinium enhancement and fat-suppressed
Over the past few years there has been a marked increase in the use of MRI and 18F-FDG-PET imaging of the breast. Their use covers the spectrum of cancer detection, staging, response to treatment and detecting recurrent disease. In 2004 the International Breast MRI Multicenter Consortium published a large study (n 821) evaluating the role of MRI in patients recommended for biopsy based on mammographic, clinical or US findings. This demonstrated that despite the inclusion of dynamic enhancement techniques, MRI missed 12 per cent of cancers (particularly well-differentiated ductal and lobular carcinomas) detected by other modalities.94 The study concluded that MRI cannot substitute for biopsies in patients with palpable or imagingdetected abnormalities. The main role of MRI is in assessing the extent of disease, multi-centricity and bilaterality of breast cancers at the time of detection. It has also been shown to detect occult breast cancer in 75–85 per cent of patients presenting with axillary lymphadenopathy with an unknown source.95 In screening high-risk patients, although MRI has a higher sensitivity than clinical examination and mammography (80 per cent versus 18 per cent and 33 per cent respectively), its specificity is highly variable.
130 Tumour imaging in oncology
Magnetic resonance imaging has not been tested for screening the general population. In detecting breast cancer, FDG-PET has no advantage over MRI and has a lower sensitivity than mammography for lobular carcinoma, small carcinomas and carcinomas with low metabolic activity.96 The lack of evidence to demonstrate a clear advantage over other techniques and its high cost have limited the use of FDG-PET in the routine diagnosis of primary breast cancer. Early studies reported very high sensitivities (79–100 per cent) and accuracy of more than 90 per cent of FGD-PET in detecting axillary nodal metastases, but more recent prospective studies are less convincing (sensitivities between 20 per cent and 61 per cent).97,98 The current data therefore do not support the routine use of FDG-PET to predict axillary metastases. The main strength of FDG-PET lies in the detection of recurrent breast cancer and distinguishing it from post-radiation and post-surgical changes.99 It has a higher sensitivity than bone scans for the detection of bone metastases. New radio-pharmaceuticals in PET which bind to oestrogen receptors (ER) have been utilized to assess metastatic disease. As the ER status is almost always known from the histopathology specimen, early studies indicate that the potential role of these ligands is to assess recurrent breast cancer in women with ER-positive primary tumour and potentially predict response to second-line chemotherapy.100 Fluorine-18 fluorothymidine (18F-FLT) is a new PET agent that measures cellular proliferation and has been used for the assessment of therapeutic effectiveness. A single available study concludes that this tracer seems to be suitable for the diagnosis of primary breast cancer and loco-regional metastases, but clearly further studies are required.101
Colorectal cancer Colorectal cancer accounts for 13 per cent of all cancer in the UK and is the second most common cancer in women and the third most common cancer in men. In 1997, there were 28 900 new cases, of which two-thirds were in the colon and a third in the rectum.51 Imaging has an established role in detecting, staging, surveillance, identifying recurrent disease and, more recently, in screening for colorectal carcinomas.
DIAGNOSIS AND STAGING
The mode of detection of colorectal carcinomas is determined by the clinical risk factors. Patients with moderate/ high risk usually undergo colonoscopy, whereas those with low risk undergo barium-enema studies. In the elderly who are unable to tolerate a barium enema, CT of the abdomen and pelvis will exclude large lesions only. The technique of CT pneumocolon, which involves prior bowel preparation similar to that for barium enemas and rectal air insufflation at the time of CT, has been proposed as a substitute for
barium enemas for screening in colorectal cancer. Its role is discussed below. The usefulness of CT in local tumour staging of both the colon and rectum is limited.102 The spatial resolution of CT cannot separate the layers of the bowel wall, and the limits of the mesorectal fascia are also not well demonstrated on CT. Nonetheless, preoperative CT of the chest, abdomen and pelvis is performed to assess nodal and distant metastases and to identify surgically significant complications of the primary tumour. These complications include intestinal obstruction, local tumour perforation, pericolic abscess, intussusception and acute appendicitis, the presence of which confers an unfavourable prognosis.103,104 The performance of CT for the detection of nodal metastases is also poor. The sensitivity, specificity and accuracy are 38 per cent (rectal), 56 per cent (colon), 88 per cent and 62 per cent respectively.105 Distant metastases from colorectal cancer are most commonly to the liver, lungs and bone. The sensitivity and specificity of intravenous contrast-enhanced CT in the detection of liver metastases are 77 per cent and 94 per cent respectively.106 Computed tomography arterioportography – which involves rapid scanning of the liver in arterial and portal phases whilst contrast is administered directly into the hepatic circulation via a previously placed catheter in the superior mesenteric artery or splenic artery – has been shown to have better sensitivity and specificity than conventional CT. However, this technique is invasive, and recent CT advances, particularly multi-detector CT, are likely to improve the results of conventional CT. Magnetic resonance imaging has also been shown to perform as well as CT arterioportography; particularly with the use of hepatocyte-specific agents, it has higher sensitivity and specificity than CT in the demonstration of lesions below 1 cm in size.107,108 Magnetic resonance imaging now has a central role in the local tumour staging of rectal cancers for patients being considered for TME surgery. High-resolution, thin-slice (2–3 mm) pelvic MRI can be used to measure accurately the depth of extramural spread (Fig. 6.12). It has been shown to predict accurately the histological status of the circumferential resection margin, with a positive predictive value of 92 per cent, and to identify negative prognostic features such as extramural venous invasion and peritoneal invasion.109,110 For nodal staging, however, MRI, like CT, relies mainly on size criteria and only provides limited morphological information. Its sensitivity for detecting nodal metastases is similar to that of CT. It may be possible to overcome this in the future with the use of MRI lymphnode-specific agents. Currently FDG-PET is not used routinely in the staging of colorectal carcinoma. Small studies have compared histology, sensitivity, specificity and CT and FDG-PET showing a accuracy for detection of tumour of 90 per cent, 66 per cent and 87 per cent with FDG-PET, compared to 60 per cent, 100 per cent and 65 per cent with CT.111 Its main role is in the identification of occult liver metastases in
Clinical application of imaging techniques to specific tumours 131
(a)
(b)
Figure 6.12 67-year-old female patient with stage T3 rectal carcinoma. (a) Sagittal T2-weighted image demonstrating a large circumferential mid rectal carcinoma (arrow). The exact site, length of the tumour and anterior bladder and posterior sacral involvement can be assessed on the sagittal images. (b) Axial-oblique T2-weighted image obtained perpendicular to the long axis of the tumour provides a true cross-section of the tumour. This allows evaluation of the depth of tumour invasion and the extent of mesorectal fascial invasion. Lymph nodes in the mesorectum are also evaluated (arrow).
patients being considered for hepatic resection. There is evidence that FDG-PET loses sensitivity when the lesion size is below 1 cm, and combined use with multi-detector CT may result in improved accuracy.112
soon after surgery or radiotherapy is useful, but a positive scan may be due to recurrent disease or treatment changes.111 In patients with a rising CEA and a negative CT, FDG-PET is able to locate recurrence accurately in about 67 per cent of cases.115
SURVEILLANCE AND RECURRENT DISEASE
The total incidence of recurrence in colorectal cancer is 15–50 per cent of which 75 per cent recur within 5 years.113 As surveillance, patients undergo clinical assessment and carcinoembryonic antigen (CEA) measurements, with colonoscopy at regular intervals. Imaging is of vital importance in surveillance and currently consists of CT of the chest, abdomen and pelvis at 3 months after surgery for baseline postoperative appearances, followed by 3-monthly CT examinations for 3 years and annually thereafter.114 It is important to detect early recurrent disease, as salvage surgery may be possible. Magnetic resonance imaging performs as well as CT, but both are unable to differentiate between postoperative scar tissue and radiation fibrosis and necrosis. Also, MRI does not routinely image the chest, and therefore metastases in the lung, a common site of recurrent disease, are undetected. Currently, MRI is not superior to CT in the surveillance of colorectal cancer. 18 F-Fluorodeoxyglucose-PET is evolving as an important problem-solving tool, and as many as 70 per cent of cases that are equivocal on CT may be resolved on FDG-PET.111 One weakness of FDG-PET is the uptake of tracer in postsurgical scar tissue and radiation fibrosis up to 6 months after primary therapy, with appearances indistinguishable from those of recurrent disease. Negative FDG-PET imaging
RECENT ADVANCES
Screening Screening programmes are currently being evaluated in the USA and Europe. The American Cancer Society guidelines recommend screening for men and women over the age of 50 years; high-risk groups enter the screening programme earlier. High-risk groups include individuals with (a) a strong family history of colorectal cancer or polyps, (b) a family with hereditary colorectal cancer syndromes, (c) a previous history of polyps or colorectal cancer, and (d) a history of chronic inflammatory bowel disease. There are five possible screening protocols that are in use: annual testing for faecal occult blood (alone or in association with 5-yearly flexible sigmoidoscopy); 5-yearly flexible sigmoidoscopy alone; colonoscopy every 10 years; and double contrast barium enema every 5 years. A positive result for faecal occult blood, sigmoidoscopy or barium enema should be followed by colonoscopy.116 Computed tomography colonography has been evaluated in the role of screening and has been shown to be accurate in the detection of polyps down to 10 mm.117,118 The potential benefits and limitations of screening are currently being debated and it is too early to assess the difference it makes to the mortality rates from colorectal cancer.
132 Tumour imaging in oncology
Prostate cancer Prostate cancer is now the second commonest cancer in men in the UK. There were 18 300 new cases in 1997 in England and Wales, with the peak age distribution in men in their seventies.51 The combination of digital rectal examination (DRE), serum prostate-specific antigen (PSA) level and prostatic biopsy forms a very effective means of detecting prostate cancer. Appropriate imaging is highly controversial and continually evolving as treatments for early disease unfold. There is currently little evidence to support imaging of low-risk (clinically stage I or II), Gleason score 7, PSA 10 ng/mL) patients prior to radical treatment by surgery or radiotherapy, but imaging is appropriate in patients with intermediate (stage I-II), Gleason score 5–7, PSA 10–20 ng/mL) and high risk of extra-prostatic (clinically stage T3 or higher, Gleason score 8 and PSA 20 ng/mL) disease.
DIAGNOSIS AND STAGING
Imaging has no role in detecting prostate cancer except for guiding transrectal prostatic biopsies.118 It is widely used in staging to stratify patients into organ-confined (stage I and II) and non-organ-confined (stage III and IV) disease for the purpose of treatment selection. Digital rectal examination is simple, inexpensive and central in staging prostate cancer, but is highly subjective and underestimates local extent of disease in 30–60 per cent of patients.119 Transrectal ultrasound of the prostate examines the prostate in a transverse and sagittal plane. Its most important role is in guiding prostate biopsies and measuring prostate size prior to brachytherapy and cryotherapy. The tumours of the peripheral prostate gland appear as hypoechogenic foci, but these appearances are non-specific and mimicked by inflammatory and infective processes of the prostate. TRUS also has a low positive predictive value (18–52 per cent) for the diagnosis of cancer.120 For extracapsular extension, the sensitivity and specificity of TRUS are equivalent to those of DRE and are between 50 and 92 per cent and 46 and 91 per cent respectively.121 For seminal vesicle involvement, the sensitivity and specificity of TRUS are highly variable, ranging between 20 and 92 per cent and 65 and 100 per cent respectively.122 In patients at high risk of seminal vesicle involvement, seminal vesicle biopsies can be performed on TRUS. Computed tomography does not confer any information concerning the local organ-confined tumour. In asymptomatic patients with PSA 20 ngmL, CT is not recommended due to the low probability of a positive study. Advanced extracapsular extension into the peri-prostatic fat and presacral space can be demonstrated. Its main role is in nodal staging, where using a cut-off value of 7 mm as the upper limit for normal nodes, CT has sensitivity and specificity of 78 per cent and 100 per cent respectively.123
Figure 6.13 82-year-old male with clinically T2b prostate cancer. The Axial T2-weighted image demonstrates the low signal intensity tumour centrally in both right and left peripheral zones at the bases (arrow heads). On the right there is tumour extension through the capsule into the peri-prostatic fat (arrow) indicating a T3a tumour.
Magnetic resonance imaging provides the best visualization of tumour in the prostate gland, the prostatic capsule, peri-prostatic tissue and the seminal vesicles. The performance of MRI in the available studies has been highly variable. A meta-analysis of several studies shows maximum sensitivity and specificity of MRI for extracapsular extension of 64 per cent and 69 per cent respectively, and for seminal vesicle involvement of 82 per cent and 80 per cent respectively (Fig. 6.13).124 As for CT, conventional MRI relies on nodal enlargement to detect metastases. Using the criterion of rounded nodes larger than 8 mm in the pelvis, MRI has a reported maximum accuracy of 90 per cent and a positive predictive value of 94 per cent in the detection of nodal metastases.125 Distant metastases are seen in less than 1 per cent of lowrisk patients. Imaging for distant metastases is recommended in patients with high-risk disease. Bone is the commonest site for metastases. Radioisotope 99mTc bone scans and plain films are used to detect bony metastases. Magnetic resonance imaging is helpful in equivocal cases and in patients with known bony metastases and symptoms suggestive of cord compression to guide radiotherapy. Metastases to liver, spleen and lungs are detected during the staging CT examinations. SURVEILLANCE AND RECURRENT DISEASE
Routine imaging of asymptomatic patients is not recommended. For patients who have undergone radical prostatectomy or radiotherapy, imaging is performed if the
Clinical application of imaging techniques to specific tumours 133
serum PSA is greater than 2 ng/mL. Local recurrence is best imaged by MRI, whereas distant nodal and metastatic disease is best assessed by CT and 99mTc bone scans.126 RECENT ADVANCES
Magnetic resonance spectroscopy has recently been evaluated in prostate cancer. In prostate cancer there is an elevation of choline and a reduction in citrate when compared to normal peripheral zone tissue of the prostate. The combination of MRI and MRS improves the localization, specificity and detection of extracapsular extension. There is an increase in accuracy from 53 per cent to 75 per cent in detecting prostatic cancer. The high specificity of MRS aids in the distinction of post-biopsy haemorrhage and other benign abnormalities from tumour.127
73–96 per cent.130 Magnetic resonance imaging is superior to CT in demonstrating peri-vesical extension, due to the high inherent tissue contrast of the bladder wall and peri-vesical fat. Overall, when early tumours are included in the staging accuracy, no significant difference is demonstrated between the performance of CT and MRI.131 For nodal metastases, MRI relies on the same criteria as CT, and both are equally accurate in the detection of nodal metastases. Distant organ metastases occur in advanced invasive local disease, with liver, lung and bone metastases occurring most frequently. Computed tomography of the chest and abdomen is included in the staging investigations of invasive bladder cancer. Technetium-99m bone scans are not used in routine staging, but are reserved for patients with bone pain. SURVEILLANCE AND RECURRENT DISEASE
Bladder cancer Bladder cancer is the fourth commonest cancer in men and the ninth commonest cancer in women in the UK, with 12 100 new cases reported in 1997.51 Cystoscopy is the main mode of tumour detection and staging, and is highly accurate in the evaluation of superficial tumours. However, in invasive tumours, clinical staging is inaccurate in 25–50 per cent of cases, and CT and MRI are used for staging. DIAGNOSIS AND STAGING
For local tumour staging, the key issue is whether the tumour is confined to the bladder wall or whether there is extension beyond into perivesical fat, adjacent structures or pelvic and abdominal walls. Cystectomy is offered to patients with invasive disease confined to the bladder, and very early T3 disease may be down-staged by radiotherapy prior to surgery. Advanced extravesical disease receives palliative chemoradiotherapy. For optimal CT technique, the patient should have a full bladder during the time of imaging; intravenous contrast enhancement is mandatory and oral contrast may be helpful. With optimal CT technique the local tumour staging accuracy of CT has been reported to be between 72 and 89 per cent.128 The best accuracy is achieved in early superficial tumours and advanced invasive disease. Distinction between stage T2a and T2b is not possible. Under-staging on CT results from its inability to detect early peri-vesical invasion. In the detection of nodal metastatic disease using an increase in size, rounded shape and enhancement after contrast administration, CT has sensitivity, specificity and accuracy of 75 per cent, 90 per cent and 90 per cent respectively for nodal metastases.129 The optimal MRI technique includes axial T1, axial, coronal and sagittal T2 images, and T1 images with dynamic contrast enhancement and fat saturation. Using these optimal techniques, the accuracy of MRI in staging bladder cancer is
Surveillance CT of the abdomen and pelvis is controversial, although there is evidence that CT can detect up to 30 per cent of recurrent disease prior to the development of symptoms.132 It is currently unclear whether it should be offered to all patients who have undergone radical treatment or should be reserved for patients at high risk (highgrade, high-stage, extra-vesical tumour, nodal metastases, lymphatic and vascular invasion prior to treatment). The frequency of the CT imaging has been proposed at 6, 12 and 24 months, as 70 per cent of all recurrent disease occurs within 2 years of completion of treatment.133 Imaging has an important role in detecting and defining sites of recurrent disease. Pelvic soft-tissue masses, nodal disease at the common iliac and internal iliac groups and retroperitoneal nodal disease are the commonest manifestations of relapse. Computed tomography of the chest, abdomen and pelvis is the imaging modality of choice, with MRI used for imaging the vertebral column in suspected cord compression. Both CT and MRI also have a role in monitoring treatment response in patients offered chemoradiotherapy to identify progression and assess response. This may be offered to down-stage patients prior to surgery or as palliative treatment. With both modalities there may be difficulty in differentiating radiotherapy changes from residual or recurrent tumour. The bladder wall is thickened and enhances in the presence of tumour and inflammatory changes secondary to radiation. Fluorodeoxyglucose-PET cannot be used to assess local bladder disease, as normal excretion of the tracer into the bladder lumen masks abnormalities of the bladder wall and immediate perivesical tissue.
Ovarian cancer Ovarian cancer is the fourth commonest cancer in women in England and Wales, with 6100 new cases registered in 1997.51 Imaging performs poorly in accurate detection,
134 Tumour imaging in oncology
pre-treatment diagnosis and staging, and thus ovarian cancer remains a major challenge in oncological imaging. These tumours often present late with widespread dissemination throughout the abdomen and pelvis. Imaging has an important role at presentation for diagnosis, preoperative assessment of extent of disease, postoperative assessment of residual disease and assessing response to adjuvant chemotherapy.
DIAGNOSIS AND STAGING
Ultrasound, particularly transvaginal ultrasound (TVUS), is the commonest modality used to identify and characterize pelvic ovarian masses at presentation. TVUS is usually performed during investigations for pelvic pain, menstrual disturbances, as part of investigations of a raised serum CA125 level or in patients with ascites and a palpable pelvic mass. Criteria for characterizing malignant masses on US have been described and include masses larger than 10 cm, the presence of soft-tissue nodules (soft-tissue components) and vegetations, thickened, irregular walls, thickened nodular septae, presence of ascites and peritoneal nodules. The sensitivity and specificity of US, including the addition of DUS, have been reported as between 85 and 97 per cent and 56 and 95 per cent respectively.134 Contrastenhanced CT relies on the same criteria as TVUS and has an accuracy of detection of ovarian masses of up to 95 per cent and in distinguishing benign from malignant disease of between 66 and 94 per cent.135 Magnetic resonance imaging with Gd enhancement has also been shown to have a high sensitivity (95 per cent) and an overall better specificity (60–93 per cent) and accuracy (91 per cent) than US for malignancy in ovarian masses.136 In the presence of a suspicious ovarian mass, elevated CA125 and a high clinical suspicion of ovarian carcinoma, FIGO-based staging is performed. This is based on a formal staging laparotomy sampling the omentum and peritoneal reflections, chest radiograph, intravenous urogram and barium enema. Surgery includes hysterectomy and bilateral salpingooopherectomy, providing local staging and tissue for histology. Surgical staging is the gold standard, but is inaccurate in 30–40 per cent of cases, with upper abdominal disease being the most common site of occult disease.137,138 Intravenous and oral contrast-enhanced CT of the abdomen and pelvis is the commonest imaging modality used for radiological preoperative staging or as baseline imaging prior to chemotherapy after debulking surgery. Computed tomography is useful in detecting ascites, pleural effusions, tumour involvement of intra-abdominal viscera, mesentery, peritoneal reflections and omentum (Fig. 6.14). The major limitation of CT is its inability to detect peritoneal deposits below 1 cm; the staging accuracy of CT is between 70 and 90 per cent.139 Magnetic resonance imaging with Gd enhancement has similar accuracy to CT in detecting peritoneal disease and does not improve the detection of peritoneal disease below 1 cm. It has better
Figure 6.14 56-year-old female with Stage 3 ovarian cancer. The contrast enhanced CT image demonstrates ascites, omental thickening (short arrow) and peritoneal nodules (arrows) in keeping with stage 3 disease. Note the calcified peritoneal nodule (arrowhead).
accuracy detecting implants on bowel serosa, and invasion of the sigmoid colon, bladder and rectum.140 Although cross-sectional imaging cannot replace surgical staging, it is valuable in patients with unresectable tumour and can identify disease missed at laparotomy. SURVEILLANCE AND RECURRENT DISEASE
Response to treatment may be assessed by non-invasive imaging such as CT and MRI, by invasive laproscopy or laparotomy or serum CA125 levels. Computed tomography is most commonly used to document treatment response. Localized disease in the pelvis at the vaginal vault and the pouch of Douglas is better assessed by MRI. Both imaging modalities miss small (1 cm) peritoneal, omental and serosal nodules. Although more sensitive (62–91 per cent versus 40–67 per cent), MRI has a significantly lower specificity (40–93 per cent versus 93–100 per cent) than CT in the restaging of recurrent disease.141 RECENT ADVANCES
Radionuclide imaging Indium-111-labelled monoclonal antibodies binding to tumour cell-surface antigens have been used experimentally to improve the detection of peritoneal and omental disease that is missed by other techniques. The site of antibody uptake is detected by SPECT or intra-operatively using handheld gamma cameras. This technique has a reported sensitivity of 68 per cent for detecting primary and recurrent disease.142 Its main strength is identification of tracer uptake in peritoneal seedlings 2–5 mm in size. However, the poor
Clinical application of imaging techniques to specific tumours 135
spatial resolution of SPECT does not allow accurate localization of the disease. This technique is currently not in routine clinical use and its role is poorly defined. 18
F-FDG-PET
Fluorodeoxyglucose-PET has recently been used in staging primary disease and detecting recurrent disease. The literature reports no improved sensitivity of FDG-PET alone or in conjunction with CT compared to conventional CT alone. In fact, FDG-PET is inferior to CT in the demonstration of small peritoneal deposits. False-negative FDGPET results have also been reported early after completion of chemotherapy.143 Population screening for ovarian carcinoma Ovarian cancer can be detected in asymptomatic patients using several methods, of which pelvic clinical examination, serum CA125 and TVUS are the most commonly used. The effectiveness of screening the general population of postmenopausal women using the above methods is currently under investigation by two randomized control trials: the UK-based UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) and the USA-based National Institute of Health Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial (NIH PLCO) study. The UKCTOCS is a 7-year prospective study measuring ovarian cancer mortality rates in controls, patients screened by TVUS alone and patients undergoing multi-modality screening. The results from these studies are currently awaited.
Endometrial cancer In the UK, endometrial cancer is the fifth commonest female malignancy and 5600 new cases and 1200 deaths were reported in 2002.51 Carcinoma of the endometrium is primarily a disease of the elderly, but 25 per cent of cases occur in pre-menopausal women; 90 per cent are histologically adenocarcinomas. Conventional treatment is by hysterectomy and as 75 per cent of tumours are confined to the uterus at presentation, this is a curative procedure. Although the mainstay of treatment is hysterectomy with bilateral salpingo-oopherectomy, pelvic lymphnode sampling or lymphadenectomy is required in a subset of patients who have a higher risk of nodal metastases. The Gynaecologic Oncology Group (GOG)144 has identified those groups of patients in whom it recommends paraaortic and pelvic lymph nodes should be sampled: ●
● ● ● ●
myometrial invasion greater than 50 per cent, regardless of tumour grade; extension of tumour into the internal os or cervical stroma; adnexal or other extra-uterine metastases; visibly enlarged lymph nodes; serous, undifferentiated, clear cell or squamous histology.
A significant number of women with endometrial cancer have various co-morbidity factors increasing the risks
of anaesthesia and surgery. These factors include obesity, diabetes mellitus, hypertension, ischaemic heart disease and AIDS. It is prudent therefore to perform the most appropriate and expeditious procedure after consultation between the surgeon and anaesthetist. Preoperative imaging will allow staging and planning so that a straightforward hysterectomy and salpingo-oopherectomy is performed without other surgical planning prolonging the procedure. FIGO recommends pre-treatment imaging evaluation of endometrial cancer with conventional imaging including barium enema, chest X-ray and intravenous urography. Recently, TVUS for endometrial cancer, CT, MRI and FDG-PET have expanded the role of imaging in the management of endometrial cancer. Imaging has become an important adjuvant to clinical evaluation in tumour staging, treatment planning, assessing treatment response, monitoring complications and during surveillance. DIAGNOSIS AND STAGING
Transvaginal ultrasound and endometrial biopsy establish the diagnosis of endometrial cancer. In endometrial cancer, TVUS shows an increase in endometrial width, heterogeneous echotexture and an irregular, poorly defined edge. As part of staging, TVUS has been used to evaluate the depth of myometrial and cervical invasion in endometrial cancer. The depth of myometrial invasion is an important prognostic factor, the incidence of nodal metastases increasing sharply from 3 per cent for stage IB tumours to 40 per cent for stage 1C tumours.145 Transvaginal ultrasound, CT and MRI have all been used to assess myometrial invasion. The performance of TVUS is variable and operator dependent. The reported accuracies for myometrial invasion vary between 77 per cent and 91 per cent.146–148 Computed tomography, including newer multi-detector CT, has poor sensitivity and specificity for myometrial and cervical invasion. The sensitivity and specificity are 83 per cent and 42 per cent respectively for the detection of deep myometrial invasion, and 25 per cent and 70 per cent respectively for the detection of cervical invasion.149 The overall staging accuracy of CT is between 58 and 76 per cent.150,151 Computed tomography is therefore of limited value for local staging and unlikely to affect management in early endometrial cancer. In more advanced disease with parametrial and pelvic sidewall disease, CT is highly accurate in detecting local spread beyond the uterus.152 Due to its excellent tissue contrast resolution, MRI has established its role in identifying deep myometrial invasion, and thereby contributes to the preoperative decision regarding the need for lymphadenectomy (Fig. 6.15). It is highly accurate in local staging and is recommended for patients in whom there is clinical suspicion of advanced disease and in select histological subtypes, listed earlier. With regards to myometrial invasion, the reported sensitivity and specificity are between 84 and 87 per cent and 91and 94 per cent respectively. The positive predictive value (PPV) is 87 per cent and the negative predictive value (NPV) is 91 per cent for the identification of myometrial invasion greater
136 Tumour imaging in oncology
with overall accuracy, sensitivity and specificity of 92 per cent, 92 per cent and 80 per cent respectively.158 Fluorodeoxyglucose-PET has been evaluated in small studies, which report promisingly high sensitivity of 96 per cent and specificity of 57 per cent. Correlation of FDG-PET with tumour markers and CT minimizes false-positive rates, resulting in a higher specificity of 88 per cent.159 As FDGPET is a whole-body imaging technique, all comparative studies report a higher accuracy than for both CT and MRI in the detection of para-aortic disease and lung metastases. RECENT ADVANCES
Figure 6.15 70-year-old woman with Endometriod Stage 1A carcinoma of the endometrium. Sagittal T2-weighted image demonstrating the large tumour in the endometrial cavity with intermediate signal intensity (star) and complete preservation of the underlying junctional zone which indicates no myometrial invasion (arrow). The endocervical canal and cervical stroma are normal. A synchronous ovarian carcinoma is also noted (arrowhead).
than 50 per cent.153,154 The use of Gd improves the ability to assess depth of myometrial invasion with MRI. Magnetic resonance imaging also performs particularly well in the detection of cervical invasion, for which its sensitivity, specificity, accuracy, PPV and NPV are 80 per cent, 96 per cent, 92 per cent, 89 per cent and 93 per cent respectively.155 Several investigators have reported that macroscopic cervical invasion, detectable on MRI, imparts a worse prognosis than microscopic invasion. Therefore preoperative assessment for cervical involvement may help in planning surgery and radiotherapy. For the detection of cervical invasion, FSE T2-weighted images are well suited. The detection of cervical invasion utilizes the natural contrast between the low T2 signal intensity of normal cervical stroma and the hyperintense T2 signal intensity of the invading endometrial carcinoma. Also as part of staging, extra-uterine disease, including peritoneal deposits in the pouch of Douglas, omental disease and pelvic or retroperitoneal lymphadenopathy, is also readily detected on MRI. SURVEILLANCE AND RECURRENT DISEASE
Magnetic resonance imaging has also been advocated in the detection of recurrent disease to evaluate surgical resectibility.151 Recurrent endometrial cancer may present as a pelvic mass in the hysterectomy bed or as pelvic or retroperitoneal lymphadenopathy. Less commonly, it may manifest as peritoneal carcinomatosis. Distant metastases and early recurrent disease are usually associated with high-grade tumours and advanced stage at presentation.156,157 Computed tomography performs well in the detection of recurrent pelvic disease,
Although endometrial cancer is primarily a disease of postmenopausal women, 25 per cent of tumours occur in younger pre-menopausal women, with 3–5 per cent under 40 years of age. In this subset of patients, the issue of fertility preservation poses a therapeutic dilemma both for patients and surgeons. Recent studies have shown grade 1, stage 1A tumours can be safely treated conservatively with hormonal therapy. Patients are treated with a high-dose progesterone regimen, with endometrial sampling every 3 months until complete regression of the tumour is documented. In these patients, MRI is performed during initial staging to exclude myometrial and cervical invasion, during surveillance, and after completion of treatment.160–162 Magnetic resonance imaging lymphography with USPIOs has been used to improve the detection of lymph-node metastases from gynaecological malignancies independent of node size. The sensitivity, specificity, PPV and NPV by size criteria were 27 per cent, 94 per cent, 60 per cent and 79 per cent respectively, and by USPIO criteria were 100 per cent, 94 per cent, 82 per cent and 100 per cent respectively.29 The USPIOs increase the sensitivity of MRI in the prediction of lymph-node metastases without loss of specificity. This will greatly improve preoperative treatment planning in endometrial carcinoma.
Non-Hodgkin’s lymphoma Non-Hodgkin’s lymphoma (NHL) is the commonest of the leukaemias and lymphomas. It is a collection of malignancies whose clinical behaviour, prognosis and management vary according to the subtype. The International Prognostic Index was developed by the International Collaborative Group to compare the efficiency of treatment regimens by implementing a standardized prognostic system.163 Imaging plays an important role in staging NHL, and in particular stage III and IV disease, and in identifying sites of extranodal disease, which are poor prognostic features, surveillance and detecting disease relapse. High-risk groups are treated with more aggressive therapies. DIAGNOSIS AND STAGING
The objectives of imaging are to define the extent of symptomatic disease and to detect occult disease. Non-Hodgkin’s
Clinical application of imaging techniques to specific tumours 137
lymphoma is commonly disseminated at presentation and although the majority of patients present with superficial lymphadenopathy, visceral involvement is common. The ability of CT to demonstrate enlarged lymph nodes throughout the body and to detect visceral pathology together with its easy availability and reproducibility have made it the imaging modality of choice for staging NHL. Patients undergo routine intravenous contrast-enhanced CT of the neck, chest, abdomen and pelvis. In the neck, lymph nodes larger than 1 cm with minimal or contrast enhancement and a rounded shape are considered to be involved. Nodal involvement in the mediastinum is seen in 25–40 per cent of cases of NHL, and the commonest sites are the superior mediastinum, subcarinal and hilar nodes. The nodes may be discrete (1 cm in short axis), matted together, have cystic changes or fine calcification. Other important sites of nodal disease in the chest are the paracardiac, intramammary, diaphragmatic, axillary, posterior mediastinal and the bronchopulmonary nodes.164,165 Within the abdomen and pelvis, mesenteric nodes are involved in 50 per cent, retroperitoneal nodes in 45–55 per cent and pelvic nodes in 40–45 per cent of patients.166 Other important sites of disease that may be overlooked are celiac, hepatic porta, splenic hilum and the porto-caval node (node of Winslow). In NHL, nodal involvement is non-contiguous, bulky and often associated with extra-nodal disease. About 40 per cent of NHL arises in extra-nodal sites. Secondary involvement may occur by direct invasion from adjacent nodal disease. The presence of extra-nodal disease is an adverse prognostic feature.167 As for nodal disease, CT is excellent for the depiction of extra-nodal disease, particularly in the lungs and abdomen. In NHL, pulmonary infiltrates may be present without thoracic lymphadenopathy in up to 50 per cent of patients. Pulmonary involvement occurs as nodules, rounded or segmental consolidation, ill-defined alveolar shadowing and cavitation.168 Pleural effusions may occur due to pleural involvement or to lymphatic obstruction resulting in a chylothorax. Chest-wall involvement is uncommon in NHL, but large masses may occur in the soft tissues of the chest wall. The margins and extent of the chestwall disease are better demonstrated on MRI than CT, as the former is more sensitive to bone-marrow infiltration and has better soft-tissue contrast than CT. Similarly, breast and cardiac involvement is also better assessed on MRI.169 Thymic infiltration may be indistinguishable from the frequently co-existent superior mediastinal lymphadenopathy. Thymic hyperplasia occurs in response to treatment and can be difficult to distinguish from recurrent disease on CT, MRI, gallium-67 (Ga-67) and FDG-PET, and follow-up studies may be necessary.170 Within the abdomen, the spleen is involved in up to 40 per cent of patients, but the detection of the disease is poor with CT, as the spleen may be of normal size and the disease in the form of lesions less than 1 cm in size.171 When lesions are larger than 1 cm, they are seen on CT and MRI and detection is improved by the administration of intravenous contrast media and multi-detector CT.
Splenomegaly indicates disease involvement; MRI does not perform better than CT in the detection of lesions smaller than 1 cm. The addition of USPIOs may improve sensitivity, and recent FDG-PET studies have demonstrated better sensitivity for the detection of splenic infiltration than CT and gallium-67.171,172 The liver is involved in 15 per cent of cases of NHL, with a higher predilection in children and recurrent disease. As for the spleen, the disease is usually microscopic, confined to the portal triad and consequently difficult to detect on CT and MRI. Non-Hodgkin’s lymphoma of the gastrointestinal tract is the most common site of primary extra-nodal NHL (stomach 51 per cent, small bowel 33 per cent, large bowel 16 per cent and oesophagus 1 per cent). As the disease is submucosal, EUS and barium contrast studies (barium meal, small-bowel follow-through and enemas) along with CT are particularly valuable. Local staging is best performed by EUS, but the frequently co-existing distant disease is best staged on CT.173 Although involvement of the genitourinary tract is rare at presentation, it is involved in more than 50 per cent of end-stage NHL. Histologically, more than 90 per cent are high-grade NHL, 40 per cent at recurrence, and renal function is usually normal. Multiple masses occur in 60 per cent of cases and solitary masses in 10–20 per cent; no associated retroperitoneal lymphadenopathy is present in 50 per cent. In 25 per cent there is direct invasion by retroperitoneal nodal disease and in a minority a peri-renal soft-tissue mass encasing the kidney without CT evidence of parenchymal invasion is seen (Figs 6.16 and 6.17).174 Extra-nodal disease of the central nervous system (particularly leptomeningeal disease), musculoskeletal system, bone marrow and head and neck lymphomas are poorly demonstrated and staged on CT. With its better tissue contrast, MRI improves both the detection and depiction of extent of disease. In the central nervous system, MRI is also very useful in detecting spinal extension of retroperitoneal or pre-sacral nodal disease and to exclude spinal cord compression in vertebral marrow disease.175 Bonemarrow disease occurs in 20–40 per cent of NHL and is considered stage IV disease and confers a poor prognosis. The involvement of the marrow is patchy but disseminated and, not surprisingly, imaging modalities such as wholebody MRI and FDG-PET are more sensitive than bonemarrow aspiration and biopsy for the detection of bone-marrow involvement. Although their exact role in staging bone-marrow disease is as yet undefined, they can be used to guide and increase the positive yield of bonemarrow aspiration for cytology.176,177 SURVEILLANCE AND RECURRENT DISEASE
Imaging is important in monitoring response to treatment and, once completed, regular periodic surveillance studies detect relapse early, allowing salvage treatment. Computed tomography is used for chest, abdominal and pelvic disease, but localized central nervous system, bone marrow, musculoskeletal system and head and neck disease is monitored
138 Tumour imaging in oncology
Figure 6.16 45-year-old woman presenting with an incidental left renal mass. Coronal reformatted contrast enhanced CT in the arterial phase demonstrates a mass associated with the inferior pole of the left kidney (arrows). The mass encases the renal parenchyma without invading or distorting the renal parenchyma. These appearances are typical of primary renal lymphoma.
Figure 6.17 32-year-old woman presenting with NHL. Sagittal T2-weighted mass involving the perineum, vulva, urethra, vagina and cervix (arrows). The mass enlarges all the affected tissues but no destruction of the tissue planes is present.
Table 6.3 Radiological criteria for diagnosing response in lymphoma Response
Features
Complete remission (CR)
1. No residual abnormality at original sites of disease 2. Nodes and nodal masses regressed to 1.5 cm in greatest transverse diameter for nodes 1.5 cm before therapy 3. Nodes regressed to 1 cm in greatest transverse diameter for nodes 1.1–1.5 cm before therapy 4. Spleen if enlarged by CT criteria must have regressed
Complete remission, unconfirmed (CRu)
Residual nodal mass 1.5 cm maximum diameter that has regressed by 75% of sum of products of greatest diameters (SPD)
Partial remission (PR)
1. 50% decrease in SPD of largest nodes/masses 2. No increase in size of liver, spleen and lymph nodes 3. Hepatic and splenic nodules decreased by 50%
Stable disease (SD)
Less than PR and no evidence of progressive disease
Progressive disease (PD)
1. 50% increase from nadir in SPD of previous abnormal nodes for PRs 2. Appearance of any new lesion during or at the end of treatment
Relapse (from CR or CRu)
1. Appearance of any new lesion 2. Increase by 50% in size of previous disease 3. Increase by 50% in greatest diameter of any previously identified node 1 cm in short axis or in SPD of more than one node
on MRI. The criteria for diagnosing radiological response are listed in Table 6.3.178 The optimal imaging intervals for assessing treatment response and for surveillance are not established. Strategies for surveillance vary according to the extent of primary disease, presence of residual masses, site of disease and grade of NHL. For patients who attain complete remission, functional imaging with Ga-67 and FDG-PET have been shown to have high sensitivity for
detecting recurrent disease prior to the development of symptoms.178 Ga-67 has been shown to detect relapse an average of 7 months prior to the onset of symptoms. Routine CT surveillance has limited value, as relapse can occur at sites distant from the primary disease and there is a significant time lag between microscopic relapse and the development of a mass detectable on cross-sectional imaging.
Imaging treatment response 139
Table 6.4 Definition of response based on World Health Organization (WHO) and Response Evaluation Criteria in Solid Tumours (RECIST) criteria WHO
RECIST
Response
Change in sum of the products of two longest perpendicular diameters
Change in sum of the longest diameter of each target lesion
Complete response at 4 weeks
Disappearance of all lesions with no residual disease
Disappearance of all target lesions
Partial response at 4 weeks
50% decrease in tumour area (product of the two longest perpendicular diameters) in single lesion or 50% decrease in the sum of the product of perpendicular diameters in multiple lesions, without 25% increase in size of one or more measurable lesions
30% decrease in the sum of the longest diameter of target lesions (using baseline sum of the longest diameters as the reference)
Stable disease
Absence of criteria of partial response or progressive disease
Absence of criteria of partial response or progressive disease
Progressive disease
25% increase in tumour area or appearance of new lesions
20% increase in sum of the longest diameter of target lesions (compared to smallest sum of the longest diameter recorded since start of treatment) or appearance of one or more new lesions
RECENT ADVANCES
Fluorodeoxyglucose-PET has been extensively investigated in lymphoma. It has a higher sensitivity than cross-sectional imaging and Ga-67 in staging lymphoma at presentation. Its strengths lie in detecting active disease in normal-sized lymph nodes and extra-nodal disease, particularly the liver and spleen, and detecting bone-marrow and skeletal involvement indicating tumour burden. It has been shown to upstage 10–20 per cent of patients compared to CT and Ga-67. Fluorodeoxyglucose-PET has also been evaluated in detecting relapse and restaging; it has a higher PPV for relapse than CT and Ga-67.179,180 A positive FDG-PET study at the end of therapy is strongly associated with an early relapse in NHL.181
IMAGING TREATMENT RESPONSE Imaging treated cancer is an expanding role of radiology. Imaging plays an increasingly central role in monitoring tumour response to treatment after standard therapies. It is increasingly used as a surrogate marker for treatment response, particularly in cancer treatment trials, which require more frequent follow-up and strict adherence to protocols. The choice of the most appropriate technique depends on the cancer type, cost, local availability of the imaging equipment and expertise. To assess response, a baseline study should be available prior to commencing treatment using the most appropriate imaging technique for follow-up. The frequency and
duration of follow-up differ for each type of cancer, when assessing patients for radical or palliative treatments, for type and interval of radiotherapy or chemotherapy, and for drug trials. Generally, CT remains the major modality for assessing treatment response for most solid cancers and lymphoma, although conventional techniques such as US and plain radiography can be used. Newer imaging techniques such as dynamic MRI, FDG-PET and MRS are under evaluation for the assessment of tumour response. Response to treatment can be measured objectively using the World Health Organization (WHO) criteria and the Response Evaluation Criteria in Solid Tumours (RECIST).182,183 The RECIST were developed specifically to standardize the evaluation of response in drug trials. A measurable lesion is a solid lesion whose limits can be clearly visualized due to the contrast with surrounding tissues. Target lesions chosen under the RECIST should be more than 2 cm in longest diameter or longer than 1 cm on spiral CT. Lesions smaller than 2 cm, bone disease, pericardial effusions, ascites, pleural disease, leptomeningeal disease, inflammatory breast lesions, cystic disease, and lymphangitis and carcinomatosis have been declared non-measurable diseases. Table 6.4 summarizes the definitions of response in WHO and RECIST terms. There are many problems with the application of these criteria, including the variability of methodologies, quality of reproducibility of the radiological examination, choice of target lesions and operator dependence. Despite these limitations, the WHO and RECIST are the most widely used assessment criteria in treatment response.
140 Tumour imaging in oncology
KEY LEARNING POINTS ●
●
●
●
●
Plain film radiography, mammography, IVU, barium studies, CT, nuclear scintigraphy, PET and PET-CT are ionising radiation based imaging modalities. Ultrasound and MRI are non-ionising imaging modalities. Imaging is part of the management of all patients with cancer. Cross-sectional imaging is important in diagnosis and staging of many cancers, assessing treatment options, surveillance and diagnosing recurrent disease. CT, including MDCT, is the most widely used cross-sectional imaging modality. It is the technique of choice in the evaluation of lung parenchyma, mediastinal, abdominal and pelvic pathology. It is widely used in staging, monitoring response, surveillance and assessing cancer recurrence in most malignancies. MDCT has the advantages of faster scanning time, thinner slice acquisition and exquisite 3D reconstruction. MRI is the imaging modality of choice in brain, head and neck and pelvic cancers. It has a superior tissue contrast than CT which is vital for imaging in the above cancers. MRI also serves as a problem solving tool in liver, adrenal and bone lesions. FDG-PET provides non-invasive functional tumour imaging. It has an established role in staging lung, oesophageal, malignant melanoma and head and neck cancers. Currently PET is an integral part of management in colorectal cancer and lymphoma. PET alone provides limited anatomical information. The combination of PET-CT adds the anatomical information and serves to reduce the false positive rate of PET alone.
REFERENCES 1 The radiographic image. In: Curry TS, Dowdey JE, Murry RC Jr Christensen’s Physics of Diagnostic Radiology, 4th edn. Philadelphia: Williams & Wilkins, 1990, 196–219. 2 Tarlowska L, Lukawska K, Mielcarzewicz Z, et al. Comparison of the FIGO and TNM staging systems for uterine cervix cancer based on classification of 6193 cases. Gynecol Oncol 1976; 4:270–7. 3 Digital radiography. In: Curry TS, Dowdey JE, Murry RC Jr. Christensen’s Physics of Diagnostic Radiology, 4th edn. Philadelphia: Williams & Wilkins, 1990, 392–432. 4 Branspetter BF. Basics of imaging informatics: Part 2. Radiology 2007, 244:78–84. 5 Ultrasound. In: Curry TS, Dowdey JE, Murry RC Jr. Christensen’s Physics of Diagnostic Radiology, 4th edn. Philadelphia: Williams & Wilkins, 1990, 323–72.
6 Jakobsen JA. Ultrasound contrast agents: clinical applications. Eur Radiol 2001; 11:1329–37. 7 Whittingham T. New and future directions in ultrasonic imaging. Eur Radiol 1999; 9(Suppl. 3):S307–311. 8 Visioli AG, Rivens IH, ter Haar GR, et al. Preliminary results of a phase I dose escalation clinical trail using focused ultrasound in the treatment of localised tumours. Eur J Ultrasound 1999; 9:11–18. 9 Computed tomography. In: Curry TS, Dowdey JE, Murry RC Jr. Christensen’s Physics of Diagnostic Radiology, 4th edn. Philadelphia: Williams & Wilkins, 1990, 289–323. 10 Armstrong PA, Hawnaur JM, Reznek RH, et al. Imaging techniques. In: Armstrong PA, Wastie ML (eds) A Concise Textbook of Radiology. New York: Arnold, 2001, 1–17. 11 Kanematsu M, Kondo H, Goshima S, et al. Imaging liver metastases: review and update. Eur J Radiol 2006; 58:217–28. 12 Foley WD. Special focus session: multidetector CT: abdominal visceral imaging. Radiographics 2002; 22:701–19. 13 Hu H, He HD, Foley WD, et al. Four multidetector-row helical CT: image quality and volume coverage speed. Radiology 2000; 215:55–62. 14 von Schulthess GK, Steinert HC, Hany TF. Integrated PET/CT: current applications and future directions. Radiology 2006; 238:405–22. 15 Parker CC, Damyanovich A, Haycocks T, et al. Magnetic resonance imaging in the radiation treatment planning of localized prostate cancer using intra-prostatic markers for computed tomography co-registration. Radiother Oncol 2003; 66:217–24. 16 Summers RM, Yao J, Pickhardt PJ, et al. Computed tomographic virtual colonoscopy computer-aided polyp detection in a screening population. Gastroenterology 2005; 129:1832–44. 17 Jeswani T, Padhani AR. Imaging tumour angiogenesis. Cancer Imaging 2005; 5:131–8. 18 Miles KA. Functional computed tomography in oncology. Eur J Cancer 2002; 38:2079–84. 19 Magnetic resonance imaging. In: Curry TS, Dowdey JE, Murry RC Jr. Christensen’s Physics of Diagnostic Radiology, 4th edn. Philadelphia: Williams & Wilkins, 1990, 470–505. 20 Uematsu H, Maeda M. Double-echo perfusion-weighted MR imaging: basic concepts and application in brain tumors for the assessment of tumor blood volume and vascular permeability. Eur Radiol 2006; 16:180–6. 21 Rollin N, Guyotat J, Streichenberger N, et al. Clinical relevance of diffusion and perfusion magnetic resonance imaging in assessing intra-axial brain tumors. Neuroradiology 2006; 48:150–9. 22 Erdogan C, Hakyemez B, Yildirim N, Parlak M. Brain abscess and cystic brain tumor: discrimination with dynamic susceptibility contrast perfusion-weighted MRI. J Comput Asst Tomogr 2005; 29:663–7. 23 Alger JR. Magnetic resonance spectroscopy in cancer. In: Padhani AR, Choyke PL (eds), New Technologies in Oncologic Imaging. New York: Taylor & Francis, 2006, 193–212.
References 141
24 Stadlbauer A, Gruber S, Nimsky C, et al. Preoperative grading of gliomas by using metabolite quantification with high-spatial-resolution proton MR spectroscopic imaging. Radiology 2006; 238:958–69. 25 Mueller-Lisse UG, Vigneron DB, Hricak H, et al. Localized prostate cancer: effect of hormone deprivation therapy measured by using combined three-dimensional 1H MR spectroscopy and MR imaging: clinicopathologic casecontrolled study. Radiology 2001; 221:380–90. 26 Heesakkers RA, Futterer JJ, Hovels AM, et al. Prostate cancer evaluated with ferumoxtran-10-enhanced T2*-weighted MR imaging at 1.5 and 3.0 T: early experience. Radiology 2006; 239:481–7. 27 Will O, Purkayastha S, Chan C, et al. Diagnostic precision of nanoparticle-enhanced MRI for lymph-node metastases: a meta-analysis. Lancet Oncol 2006; 7:52–60. 28 Carot C, Robert P, Idee JM, Port M. Recent advances in iron oxide nanocrystal technology for medical imaging. Adv Drug Deliv 29 Rockall AG, Sohaib SA, Harisinghani MG, et al. Diagnostic performance of nanoparticle-enhanced magnetic resonance imaging in the diagnosis of lymph node metastases in patients with endometrial and cervical cancer. J Clin Oncol 2005; 23:2813–21. 30 Kim YK, Lee JM, Kim CS, et al. Detection of liver metastases: gadobenate dimeglumine-enhanced three-dimensional dynamic phases and one-hour delayed phase MR imaging versus superparamagnetic iron oxide-enhanced MR imaging. Eur Radiol 2005; 15:220–8. 31 Kim MJ, Kim JH, Chung JJ, et al. Focal hepatic lesions: detection and characterization with combination gadolinium- and superparamagnetic iron oxide-enhanced MR imaging. Radiology 2003; 228:719–26. 32 Chung JJ, Kim MJ, Kim KW. Mangafodipir trisodiumenhanced MRI for the detection and characterization of focal hepatic lesions: is delayed imaging useful? J Magn Reson Imaging 2006; 23:706–11. 33 del Frate C, Bazzocchi M, Mortele KJ, et al. Detection of liver metastases: comparison of gadobenate dimeglumineenhanced and ferumoxides-enhanced MR imaging examinations. Radiology 2002; 225:766–72. 34 Bluemke DA, Sahani D, Amendola M, et al. Efficacy and safety of MR imaging with liver-specific contrast agent: U.S. multicenter phase III study. Radiology 2005; 237:89–98. 35 Reimer P, Tombach B. Hepatic MRI with SPIO: detection and characterization of focal liver lesions. Eur Radiol 1998; 8:1198–204. 36 Reimer P, Jahnke N, Fiebich M, et al. Hepatic lesion detection and characterization: value of nonenhanced MR imaging, superparamagnetic iron oxide-enhanced MR imaging, and spiral CT-ROC analysis. Radiology 2000; 217:152–8. 37 Gilderdale DJ, Larkman DJ. Technical considerations, design and development. In: deSouza NM (ed.), Endocavitary MRI of the Pelvis. Amsterdam: Harwood Academic Publishers, 2001, 1–15. 38 Shepherd JH. Uterus-conserving surgery for invasive cervical cancer. Best Pract Res Clin Obstet Gynaecol 2005; 19:577–90.
39 Trabulsi EJ, Merriam WG, Gomella LG. New imaging techniques in prostate cancer. Curr Urol Rep 2006; 7:175–80. 40 Shellock FG, Curtis JS. MR imaging and biomedical implants, materials, and devices: an updated review. Radiology 1991; 180:541–50. 41 Levine D, Zuo C, Faro CB, Chen Q. Potential heating effect in the gravid uterus during MR HASTE imaging. J Magn Reson Imaging 2001; 13:856–61. 42 Ziegler S. Physical principles, dedicated/coincidence PET. In: Oehr P, Biersack HJ, Coleman RE (eds), PET and PET-CT in Oncology. Heidelberg: Springer-Verlag, 2004, 3–25. 43 Pottgen C, Levegrun S, Theegarten D, et al. Value of 18Ffluoro-2-deoxy-D-glucose positron emission tomography/computed tomography in non-small-cell lung cancer for prediction of pathologic response and times to relapse after neoadjuvant chemoradiotherapy. Clin Cancer Res 2006; 12:97–106. 44 Pio BS, Park CK, Pietras R, et al. Usefulness of 3-[F18]fluoro-3-deoxythymidine with positron emission tomography in predicting breast cancer response to therapy. Mol Imaging Biol 2006; 8:36–42. 45 van Baardwijk A, Baumert BG, Bosmans G, et al. The current status of FDG-PET in tumour volume definition in radiotherapy treatment planning. Cancer Treat Rev 2006; 22: [Epub ahead of print] 46 Goshen E, Davidson T, Zwas ST, Aderka D. PET/CT in the evaluation of response to treatment of liver metastases from colorectal cancer with bevacizumab and irinotecan. Technol Cancer Res Treat 2006; 5:37–43. 47 Choi HJ, Roh JW, Seo SS, et al. Comparison of the accuracy of magnetic resonance imaging and positron emission tomography/computed tomography in the presurgical detection of lymph node metastases in patients with uterine cervical carcinoma: a prospective study. Cancer 2006; 106:914–22. 48 Seemann MD, Meisetschlaeger G, Gaa J, Rummeny EJ. Assessment of the extent of metastases of gastrointestinal carcinoid tumors using whole-body PET, CT, MRI, PET/CT and PET/MRI. Eur J Med Res 2006; 11:58–65. 49 von Schulthess GK, Steinert HC, Hany TF. Integrated PET/CT: current applications and future directions. Radiology 2006; 238:405–22. 50 Oehr P. Introduction to experimental PET in oncology. In: Oehr P, Biersack HJ, Coleman RE (eds), PET and PET-CT in Oncology. Heidelberg: Springer-Verlag, 2004, 67–71. 51 National statistics online. www.statistics.gov.uk/statbase. Cancer trends 1950–1999. 52 Hart BL, Benzel EC, Ford CC (eds). Fundamentals of Neuroimaging. Philadelphia: WB Saunders, 1997. 53 Cao Y, Sundgren PC, Tsien CI, et al. Physiologic and metabolic magnetic resonance imaging in gliomas. J Clin Oncol 2006; 24:1228–35. 54 Atlas SW, Lavi E, Fisher PG. Intraxial brain tumours. In: Atlas SW (ed.), Magnetic Resonance Imaging of the Brain and Spine, 3rd edn. Philadelphia: Lippincott Williams & Wilkins, 2002, 627–760.
142 Tumour imaging in oncology
55 Wang SX, Boethius J, Ericson K. FDG-PET on irradiated brain tumour: ten years’ summary. Acta Radiol 2006; 47:85–90. 56 Barai S, Rajkamal, Bandopadhayaya GP, et al. Thallium-201 versus Tc99m-glucoheptonate SPECT for evaluation of recurrent brain tumours: a within-subject comparison with pathological correlation. J Clin Neurosci 2005; 12:27–31. 57 Floeth FW, Pauleit D, Sabel M, et al.18F-FET PET differentiation of ring-enhancing brain lesions. J Nucl Med 2006; 47:776–82. 58 Pirotte B, Goldman S, Dewitte O, et al. Integrated positron emission tomography and magnetic resonance imagingguided resection of brain tumors: a report of 103 consecutive procedures. J Neurosurg 2006; 104:238–53. 59 Ishikawa M, Anzai Y. MR imaging of lymph nodes in the head and neck. Magn Reson Imaging Clin North Am 2002; 10:527–42. 60 Mendenhall WM, Mancuso AA, Parsons JT, et al. Diagnostic evaluation of squamous cell carcinoma metastatic to cervical lymph nodes from an unknown head and neck primary site. Head Neck 1998; 20:739–44. 61 Atula TS, Grenman R, Varpula MJ, et al. Palpation, ultrasound, and ultrasound-guided fine-needle aspiration cytology in the assessment of cervical lymph node status in head and neck cancer patients. Head Neck 1996; 18:545–51. 62 Adams S, Baum RP, Stuckensen T, et al. Prospective comparison of 18F-FDG PET with conventional imaging modalities (CT, MRI, US) in lymph node staging of head and neck cancer. Eur J Nucl Med 1998; 25:1255–60. 63 McGuirt WF, Greven K, Williams D 3rd, et al. PET scanning in head and neck oncology: a review. Head Neck 1998; 20:208–15. 64 Hermans R. Tumours of the pharynx, tongue and mouth. In: Husband JE, Reznek RH (eds), Imaging in Oncology, 2nd edn. New York: Taylor & Francis, 2004, 607–38. 65 Pameijer FA, Hermans R, Mancuso AA, et al. Pre- and postradiotherapy computed tomography in laryngeal cancer: imaging-based prediction of local failure. Int J Radiat Oncol Biol Phys 1999; 45:359–66. 66 Hermans R, Pameijer FA, Mancuso AA, et al. Laryngeal or hypopharyngeal squamous cell carcinoma: can follow-up CT after definitive radiation therapy be used to detect local failure earlier than clinical examination alone? Radiology 2000; 214:683–7. 67 Kostakoglu L, Agress H Jr, Goldsmith SJ. Clinical role of FDG PET in evaluation of cancer patients. Radiographics 2003; 23:315–40. 68 Mukherji SK, Wolf GT. Evaluation of head and neck squamous cell carcinoma after treatment. AJNR 2003; 24:1743–6. 69 Shah GV, Fischbein NJ, Patel R, Mukherji SK. Newer MR imaging techniques for head and neck. Magn Reson Imaging Clin North Am 2003; 11:449–69. 70 Chikui T, Yonetsu K, Nakamura T. Multivariate feature analysis of sonographic findings of metastatic cervical lymph nodes: contribution of blood flow features revealed by power Doppler sonography for predicting metastasis. AJNR 2000; 21:561–7.
71 Abrams J, Doyle LA, Aisner J. Staging, prognostic factors, and special considerations in small cell lung cancer. Semin Oncol 1988; 15:261–77. 72 Quint LE, Francis IR. Radiologic staging of lung cancer. J Thorac Imaging 1999; 14:235–46. 73 Lewis JW Jr, Pearlberg JL, Beute GH, et al. Can computed tomography of the chest stage lung cancer? Yes and no. Ann Thorac Surg 1990; 49:591–6. 74 Musset D, Grenier P, Carette MF, et al. Primary lung cancer staging: prospective comparative study of MR imaging with CT. Radiology 1986; 160:607–11. 75 Lowe VJ, Naunheim KS. Current role of positron emission tomography in thoracic oncology. Thorax 1998; 53:703–12. 76 Webb WR, Gatsonis C, Zerhouni EA, et al. CT and MR imaging in staging non-small cell bronchogenic carcinoma: report of the Radiologic Diagnostic Oncology Group. Radiology 1991; 178:705–13. 77 Freundlich IM, Chasen MH, Varma DG. Magnetic resonance imaging of pulmonary apical tumors. J Thorac Imaging 1996; 11:210–22. 78 Watanabe Y, Shimizu J, Tsubota M, Iwa T. Mediastinal spread of metastatic lymph nodes in bronchogenic carcinoma. Mediastinal nodal metastases in lung cancer. Chest 1990; 97:1059–65. 79 Dwamena BA, Sonnad SS, Angobaldo JO, Wahl RL. Metastases from non-small cell lung cancer: mediastinal staging in the 1990s – meta-analytic comparison of PET and CT. Radiology 1999; 213:530–6. 80 Erasmus JJ, McAdams HP, Patz EF Jr. Non-small cell lung cancer: FDG-PET imaging. J Thorac Imaging 1999; 14:247–56. 81 Marom EM, McAdams HP, Erasmus JJ, et al. Staging nonsmall cell lung cancer with whole-body PET. Radiology 1999; 212:803–9. 82 Shon IH, O’Doherty MJ, Maisey MN. Positron emission tomography in lung cancer. Semin Nucl Med 2002; 32:240–71. 83 Gorich J, Beyer-Enke SA, Flentje M, et al. Evaluation of recurrent bronchogenic carcinoma by computed tomography. Clin Imaging 1990; 14:131–7. 84 Hicks RJ, Kalff V, MacManus MP, et al. The utility of (18)F-FDG PET for suspected recurrent non-small cell lung cancer after potentially curative therapy: impact on management and prognostic stratification. J Nucl Med 2001; 42:1605–13. 85 Henschke CI, McCauley DI, Yankelevitz DF, et al. Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet 1999; 354:99–105. 86 Henschke CI, Yankelevitz DF, Naidich DP, et al. CT screening for lung cancer: suspiciousness of nodules according to size on baseline scans. Radiology 2004; 231:164–8. 87 Kerlikowske K, Grady D, Rubin SM, et al. Efficacy of screening mammography. A meta-analysis. JAMA 1995; 273:149–54. 88 Liberman L, Freeman HP, Chandra S, et al. Carcinoma detection at the breast examination center of Harlem. Cancer 2002; 95(1):8–14.
References 143
89 Brown J, Coulthard A, Dixon AK, et al. UK MRI Breast Screening Study Advisory Group. Protocol for a national multi-centre study of magnetic resonance imaging screening in women at genetic risk of breast cancer. Breast 2000; 9(2):78–82. 90 Pain JA, Ebbs SR, Hern RP, et al. Assessment of breast cancer size: a comparison of methods. Eur J Surg Oncol 1992; 18(1):44–8. 91 Davis PL, McCarty KS Jr. Technologic considerations for breast tumor size assessment. Magn Reson Imaging Clin North Am 1994; 2(4):623–31. 92 Kurtz JM, Amalric R, Brandone H, et al Local recurrence after breast-conserving surgery and radiotherapy. Frequency, time course, and prognosis. Cancer 1989; 63(10):1912–17. 93 Whitehouse GH, Moore NR. MR imaging of the breast after surgery for breast cancer. Magn Reson Imaging Clin North Am 1994; 2(4):591–603. 94 Bluemke DA, Gatsonis CA, Chen MH, et al. Magnetic resonance imaging of the breast prior to biopsy. JAMA 2004; 292(22):2735–42. 95 Orel SG, Weinstein SP, Schnall MD, et al. Breast MR imaging in patients with axillary node metastases and unknown primary malignancy. Radiology 1999; 212(2):543–9. 96 Avril N, Menzel M, Dose J, et al Glucose metabolism of breast cancer assessed by 18F-FDG PET: histologic and immunohistochemical tissue analysis. J Nucl Med 2001; 42(1):9–16. 97 Adler LP, Crowe JP, al-Kaisi NK, Sunshine JL. Evaluation of breast masses and axillary lymph nodes with [F-18]2-deoxy2-fluoro-D-glucose PET. Radiology 1993; 187,(3):743–50. 98 Avril N, Rose CA, Schelling M, et al. Breast imaging with positron emission tomography and fluorine-18 fluorodeoxyglucose: use and limitations. J Clin Oncol 2000; 18(20):3495–502. 99 Kamel EM, Wyss MT, Fehr MK, et al. [18F]Fluorodeoxyglucose positron emission tomography in patients with suspected recurrence of breast cancer. J Cancer Res Clin Oncol 2003; 129(3):147–53. 100 Mintun MA, Welch MJ, Siegel BA, et al. Breast cancer: PET imaging of estrogen receptors. Radiology 1988; 169:45–8. 101 Smyczek-Gargya B, Fersis N, Dittmann H, et al. PET with [18F]fluorothymidine for imaging of primary breast cancer: a pilot study. Eur J Nucl Med Mol Imaging 2004; 31:720–4. 102 Mehta S, Johnson RJ, Schofield PF. Staging of colorectal cancer. Clin Radiol 1994; 49:515–23. 103 Fielding LP, Phillips RK, Fry JS, et al. Prediction of outcome after curative resection for large bowel cancer. Lancet 1986; 2:904–7. 104 Shepherd NA, Baxter KJ, Love SB. The prognostic importance of peritoneal involvement in colonic cancer: a prospective evaluation. Gastroenterology 1997; 112:1096–102. 105 Zerhouni EA, Rutter C, Hamilton SR, et al. CT and MR imaging in the staging of colorectal carcinoma: report of the Radiology Diagnostic Oncology Group II. Radiology 1996; 200:443–51.
106 Valls C, Andia E, Sanchez A, et al. Hepatic metastases from colorectal cancer: preoperative detection and assessment of resectability with helical CT. Radiology 2001; 218:55–60. 107 Soyer P, Levesque M, Elias D, et al. Preoperative assessment of resectability of hepatic metastases from colonic carcinoma: CT portography vs sonography and dynamic CT. AJR 1992; 159:741–4. 108 Mann GN, Marx HF, Lai LL, Wagman LD. Clinical and cost effectiveness of a new hepatocellular MRI contrast agent, mangafodipir trisodium, in the preoperative assessment of liver resectability. Ann Surg Oncol 2001; 8:573–9. 109 Brown G, Richards CJ, Newcombe RG, et al. Rectal carcinoma: thin-section MR imaging for staging in 28 patients. Radiology 1999; 211:215–22. 110 Brown G, Radcliffe AG, Newcombe RG, et al. Preoperative assessment of prognostic factors in rectal cancer using high-resolution magnetic resonance imaging. Br J Surg 2003; 90:355–64. 111 Schiepers C, Penninckx F, De Vadder N, et al. Contribution of PET in the diagnosis of recurrent colorectal cancer: comparison with conventional imaging. Eur J Surg Oncol 1995; 21:517–22. 112 Delbeke D, Vitola JV, Sandler MP, et al. Staging recurrent metastatic colorectal carcinoma with PET. J Nucl Med 1997; 38:1196–201. 113 Adloff M, Arnaud JP, Schloegel M, Thibaud D. Factors influencing local recurrence after abdominoperineal resection for cancer of the rectum. Dis Colon Rectum 1985; 28:413–15. 114 Sugarbaker PH, Gianola FJ, Dwyer A, Neuman NR. A simplified plan for follow-up of patients with colon and rectal cancer supported by prospective studies of laboratory and radiologic test results. Surgery 1987; 102:79–87. 115 Valk PE, Abella-Columna E, Haseman MK, et al. Whole-body PET imaging with [18F]fluorodeoxyglucose in management of recurrent colorectal cancer. Arch Surg 1999; 134:503–11; discussion 511–13. 116 American Cancer Society. Cancer Facts and Figures 2003.: www.cancer.org/downloads/stt/caff2003. 117 Royster AP, Fenlon HM, Clarke PD, et al. CT colonoscopy of colo-rectal neoplasms: two-dimensional and threedimensional virtual-reality techniques with colonoscopic correlation. AJR 1997; 169:1237–42. 118 Littrup PJ, Bailey SE. Prostate cancer: the role of transrectal ultrasound and its impact on cancer detection and management. Radiol Clin North Am 2000; 38:87–113. 119 Zincke H, Oesterling JE, Blute ML, et al. Long-term (15 years) results after radical prostatectomy for clinically localized (stage T2C or lower) prostate cancer. J Urol 1994; 152:1850–7. 120 Gustafsson O, Carlsson P, Norming U, et al. Costeffectiveness analysis in early detection of prostate cancer: an evaluation of six screening strategies in a randomly selected population of 2,400 men. Prostate 1995; 26:299–309. 121 Smith JA Jr, Scardino PT, Resnick MI, et al. Transrectal ultrasound versus digital rectal examination for the staging of carcinoma of the prostate: results of a prospective, multiinstitutional trial. J Urol 1997; 157:902–6.
144 Tumour imaging in oncology
122 Melchior SW, Brawer MK. Role of transrectal ultrasound and prostate biopsy. J Clin Ultrasound 1996; 24:463–71. 123 Oyen RH, Van Poppel HP, Ameye FE. Lymph node staging of localized prostatic carcinoma with CT and CT-guided fineneedle aspiration biopsy: prospective study of 285 patients. Radiology 1994; 190:315–22. 124 Sonnad SS, Langlotz CP, Schwartz JS. Accuracy of MR imaging for staging prostate cancer: a meta-analysis to examine the effect of technologic change. Acad Radiol 2001; 8:149–57. 125 Jager GJ, Barentsz JO, Oosterhof GO, et al. Pelvic adenopathy in prostatic and urinary bladder carcinoma: MR imaging with a three-dimensional TI-weighted magnetization-prepared-rapid gradient-echo sequence. AJR 1996; 167:1503–7. 126 Partin AW, Pearson JD, Landis PK, et al. Evaluation of serum prostate-specific antigen velocity after radical prostatectomy to distinguish local recurrence from distant metastases. Urology 1994; 43:649–59. 127 Kaji Y, Kurhanewicz J, Hricak H, et al. Localizing prostate cancer in the presence of postbiopsy changes on MR images: role of proton MR spectroscopic imaging. Radiology 1998; 206:785–90. 128 Sharir S. Update on clinical and radiological staging and surveillance of bladder cancer. Can J Urol 2006; 13(Suppl. 1):71–6. 129 Bernhardt TM, Rapp-Bernhardt U. Virtual cystoscopy of the bladder based on CT and MRI data. Abdom Imaging 2001; 26:325–32. 130 Kim B, Semelka RC, Ascher SM, et al. Bladder tumor staging: comparison of contrast-enhanced CT, T1- and T2-weighted MR imaging, dynamic gadolinium-enhanced imaging, and late gadolinium-enhanced imaging. Radiology 1994; 193:239–45. 131 Buy JN, Moss AA, Guinet C, et al. MR staging of bladder carcinoma: correlation with pathologic findings. Radiology 1988; 169:695–700. 132 Slaton JW, Swanson DA, Grossman HB, Dinney CP. A stage specific approach to tumor surveillance after radical cystectomy for transitional cell carcinoma of the bladder. J Urol 1999; 162:710–14. 133 Montie JE. Follow-up after cystectomy for carcinoma of the bladder. Urol Clin North Am 1994; 21:639–43. 134 Kinkel K, Hricak H, Lu Y, et al. US characterization of ovarian masses: a meta-analysis. Radiology 2000; 217:803–11. 135 Buy JN, Ghossain MA, Sciot C, et al. Epithelial tumors of the ovary: CT findings and correlation with US. Radiology 1991; 178:811–18. 136 Sohaib SA, Sahdev A, Van Trappen P, et al. Characterization of adnexal mass lesions on MR imaging. AJR 2003; 180:1297–304. 137 Bagley CM Jr, Young RC, Schein PS, et al. Ovarian carcinoma metastatic to the diaphragm – frequently undiagnosed at laparotomy. A preliminary report. Am J Obstet Gynecol 1973; 116:397–400. 138 Clarke-Pearson DL, Soper JT, Berchuck A, Hunter VJ. Ovarian cancer. In: Moosa AR, Schimpff SC, Robson MC (eds),
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
Comprehensive Textbook of Oncology, 2nd edn. New York: Williams & Wilkins, 1991, 1006–20. Coakley FV, Choi PH, Gougoutas CA, et al. Peritoneal metastases: detection with spiral CT in patients with ovarian cancer. Radiology 2002; 223:495–9. Forstner R, Hricak H, Occhipinti KA, et al. Ovarian cancer: staging with CT and MR imaging. Radiology 1995; 197: 619–26. Prayer L, Kainz C, Kramer J, et al. CT and MR accuracy in the detection of tumor recurrence in patients treated for ovarian cancer. J Comput Asst Tomogr 1993; 17:626–32. Surwit EA, Childers JM, Krag DN, et al. Clinical assessment of 111In-CYT-103 immunoscintigraphy in ovarian cancer. Gynecol Oncol 1993; 48:285–92. Cho SM, Ha HK, Byun JY, et al. Usefulness of FDG PET for assessment of early recurrent epithelial ovarian cancer. AJR 2002; 179:391–5. Morrow CP, Bundy BN, Kumar RJ, et al. Relationship between surgical–pathological risk factors and outcome in clinical stages I and II carcinoma of the endometrium: a Gynaecologic Oncology Group study. Gynecol Oncol 1991; 40:55–6. Boronow RC, Morrow CP, Creasman WT, et al. Surgical staging in endometrial cancer: clinical–pathologic findings of a prospective study. Obstet Gynecol 1984; 63:825–32. Weber G, Merz E, Bahlmann F, Mitze M, Weikel W, Knapstein PG. Assessment of myometrial infiltration and preoperative staging by transvaginal ultrasound in patients with endometrial carcinoma. Ultrasound Obstet Gynecol 1995; 6:362–7. Fishman A, Altaras M, Bernheim J, Cohen I, Beyth Y, Tepper R. The value of transvaginal sonography in the preoperative assessment of myometrial invasion in high and low grade endometrial cancer and in comparison to frozen section in grade 1 disease. Eur J Gynaecol Oncol 2000; 21:128–30. Gordon AN, Fleischer AC, Reed GW. Depth of myometrial invasion in endometrial cancer: preoperative assessment by transvaginal ultrasonography. Gynecol Oncol 1990; 39:321–7. Hardesty LA, Sumkin JH, Hakim C, Johns C, Nath M. The ability of helical CT to preoperatively stage endometrial carcinoma. AJR 2001; 176(3):603–6. Kim SH, Kim HD, Song YS, Kang SB, Lee HP. Detection of deep myometrial invasion in endometrial carcinoma: comparison of transvaginal ultrasound, CT, and MRI. J Comput Assist Tomogr 1995; 19(5):766–72. Connor JP, Andrews JI, Anderson B, Buller RE. Computed tomography in endometrial carcinoma. Obstet Gynecol 2000; 95(5):692–6. Kinkel K, Kaji Y, Yu KK, et al. Radiological staging in patients with endometrial cancer: a meta-analysis. Radiology 1999; 212(3):711–18. Sironi S, Colombo E, Villa G, et al. Myometrial invasion by endometrial carcinoma: assessment with plain and gadolinium-enhanced MR imaging. Radiology 1992; 185(1):207–12.
References 145
154 Lee EJ, Byun JY, Kim B, Koong Sen, Shinn KS. Staging of early endometrial carcinoma: assessment with T2-weighted and gadolinium-enhanced T1-weighted MR imaging. Radiographics 1999; 19(4):937–45. 155 Manfredi R, Mirk P, Maresca G, et al. Local–regional staging of endometrial carcinoma: role of MR imaging in surgical planning. Radiology 2004; 231(2):372–8. 156 Rose PG. Endometrial carcinoma. N Engl J Med 1996; 335(9):640–9. 157 Creasman WT, Morrow CP, Bundy BN, Homesley HD, Graham JE, Heller PB. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group study. Cancer 1987; 60(8 Suppl.):2035–41. 158 Franchi M, La Fianza A, Babilonti L, et al. Clinical value of computerized tomography (CT) in assessment of recurrent uterine cancers. Gynecol Oncol 1989; 35(1):31–7. 159 Saga T, Higashi T, Ishimori T, et al. Clinical value of FDG-PET in the follow up of post-operative patients with endometrial cancer. Ann Nucl Med 2003; 17(3):197–203. 160 Niwa K, Tagami K, Lian Z, Onogi K, Mori H, Tamaya T. Outcome of fertility preserving treatment in young women with endometrial carcinomas. BJOG 2005; 112(3):317–20. 161 Benshushan A. Endometrial adenocarcinoma in young patients: evaluation and fertility-preserving treatment. Eur J Obstet Gynecol Reprod Biol 2004; 117(2):132–7. 162 Ben-Shacher I, Vitellas KM, Cohn DE. The role of MRI in the conservative management of endometrial cancer. Gynecol Oncol 2004; 93(1):233–7. 163 Shipp MA. Prognostic factors in aggressive non-Hodgkin’s lymphoma: who has ‘high-risk’ disease? Blood 1994; 83(5):1165–73. 164 Filly R, Bland N, Castellino RA. Radiographic distribution of intrathoracic disease in previously untreated patients with Hodgkin’s disease and non-Hodgkin’s lymphoma. Radiology 1976; 120(2):277–81. 165 Castellino RA, Hilton S, O’Brien JP, Portlock CS. NonHodgkin lymphoma: contribution of chest CT in the initial staging evaluation. Radiology 1996; 199(1):129–32. 166 Harell GS, Breiman RS, Glatstein EJ, et al Computed tomography of the abdomen in the malignant lymphomas. Radiol Clin North Am 1977; 15(3):391–400. 167 Goffinet DR, Warnke R, Dunnick NR, et al. Clinical and surgical (laparotomy) evaluation of patients with non-Hodgkin’s lymphomas. Cancer Treat Rep 1977; 61(6):981–92. 168 Lewis ER, Caskey CI, Fishman EK. Lymphoma of the lung: CT findings in 31 patients. AJR 1991; 156(4):711–14. 169 Carlsen SE, Bergin CJ, Hoppe RT. MR imaging to detect chest wall and pleural involvement in patients with lymphoma: effect on radiation therapy planning. AJR 1993; 160(6):1191–5.
170 Spiers AS, Husband JE, MacVicar AD. Treated thymic lymphoma: comparison of MR imaging with CT. Radiology 1997; 203(2):369–76. 171 Giovagnoni A, Giorgi C, Goteri G. Tumours of the spleen. Cancer Imaging 2005; 5(1):73–7. 172 Moog F, Bangerter M, Diederichs CG, et al. Extranodal malignant lymphoma: detection with FDG PET versus CT. Radiology 1998; 206(2):475–81. 173 Raderer M, Vorbeck F, Formanek M, et al. Importance of extensive staging in patients with mucosa-associated lymphoid tissue (MALT)-type lymphoma. Br J Cancer 2000; 83(4):454–7. 174 Reznek RH, Mootoosamy I, Webb JA, Richards MA. CT in renal and perirenal lymphoma: a further look. Clin Radiol 1990; 42(4):233–8. 175 Chamberlain MC, Sandy AD, Press GA. Leptomeningeal metastasis: a comparison of gadolinium-enhanced MR and contrast-enhanced CT of the brain. Neurology 1990; 40(3 Pt 1):435–8. 176 Moog F, Bangerter M, Kotzerke J, et al. 18-Ffluorodeoxyglucose-positron emission tomography as a new approach to detect lymphomatous bone marrow. J Clin Oncol 1998; 16(2):603–9. 177 Yasumoto M, Nonomura Y, Yoshimura R, et al. MR detection of iliac bone marrow involvement by malignant lymphoma with various MR sequences including diffusionweighted echo-planar imaging. Skeletal Radiol 2002; 31(5):263–9. 178 Cheson BD, Horning SJ, Coiffier B, et al. Report of an international workshop to standardize response criteria for non-Hodgkin’s lymphomas. NCI Sponsored International Working Group. J Clin Oncol 1999; 17(4):1244. 179 Raanani P, Shasha Y, Perry C, et al. Is CT scan still necessary for staging in Hodgkin and non-Hodgkin lymphoma patients in the PET/CT era? Ann Oncol 2006; 17(1):117–22. 180 Reinhardt MJ, Herkel C, Altehoefer C, et al. Computed tomography and 18F-FDG positron emission tomography for therapy control of Hodgkin’s and non-Hodgkin’s lymphoma patients: when do we really need FDG-PET? Ann Oncol 2005; 16(9):1524–9. 181 Naumann R, Vaic A, Beuthien-Baumann B, et al. Prognostic value of positron emission tomography in the evaluation of post-treatment residual mass in patients with Hodgkin’s disease and non-Hodgkin’s lymphoma. Br J Haematol 2001; 115(4):793–800. 182 WHO Handbook for Reporting Results of Cancer Treatment. Geneva: World Health Organisation, 1979, 48. 183 Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumours. J Natl Cancer Inst 2000; 92:205–16.
7 Interventional radiology TARUN SABHARWAL, ANNE P. HEMINGWAY AND ANDREAS ADAM
Introduction Biopsy procedures Percutaneous puncture and drainage procedures Venous sampling Percutaneous tumour ablation
146 147 150 150 151
INTRODUCTION Over the past three decades, a variety of invasive diagnostic and therapeutic procedures have been developed by radiologists. The term interventional radiology (IR) most appropriately refers to therapeutic procedures performed under imaging guidance.1 However, diagnostic invasive techniques are usually carried out by interventional radiologists and are included in this chapter. The emergence of IR as a specialty has been made possible by the enormous technological advances in relation to catheter and instrument design and manufacture, imaging systems and radiological expertise. Some of these procedures have largely replaced more invasive and hazardous surgical alternatives; for example, tunnelled central venous catheters are now usually inserted using imaging-guided percutaneous techniques under local anaesthesia, rather than open surgery under general anaesthesia. A summary of the main types of interventional procedures is shown in Table 7.1. The term invasive is relative; most of the procedures described in this chapter are considerably less invasive than their surgical alternative. However, they all carry some risk, which is interdependent on the underlying condition, the nature of the procedure and the experience of the radiologist. Radiological interventions in cancer patients require comprehensive assessment of the patients and careful consideration of the risks and benefits of the procedure. Factors unique to oncology patients must be evaluated, and decisions regarding therapeutic benefit must be made in light of patients’ prognoses and quality of life considerations. Effective communication of information about the nature,
Dilatation techniques Extraction techniques Infusion techniques Vascular embolization Gene therapy References
153 156 158 158 163 165
Table 7.1 Interventional radiological procedures Procedure Biopsy Drainage Dilatation/stenting
Feeding Venous access Extraction Infusion Embolization
Vertebroplasty Tumour ablation
Examples of indications Need for cytology/histology Obstruction of renal and biliary tracts, pleural effusions, ascites, abscesses Limb/organ ischaemia, gastrointestinal, biliary, ureteric and airway obstruction, superior vena cava/inferior vena cava obstruction etc. Percutaneous gastrostomy Dialysis, chemotherapy, total parental nutrition (TPN) Intravascular foreign bodies Regional, selective infusion of chemotherapy Gastrointestinal haemorrhage, posttraumatic haemorrhage, pseudoaneurysms, skeletal metastases etc. Osteoporosis, vertebral metastases Hepatic, renal, bony and soft-tissue tumours
benefits and potential complications is an integral part of the treatment.2 In this chapter, emphasis is placed on the indications, contraindications and results rather than on detailed technical descriptions.
Biopsy procedures 147
BIOPSY PROCEDURES The development of fine needles (Chiba needles) and smallgauge cutting instruments (automated biopsy devices such as the Quick-core, Temno and Biopty gun), which can be accurately directed to virtually any site within the body under imaging control and local anaesthesia, has dramatically reduced the number of patients requiring open surgical biopsy to obtain tissue for a histological diagnosis. It is possible to perform most of these procedures on an outpatient basis. Needle placement for obtaining a cytological or histological specimen may be carried out under fluoroscopic, ultrasonic (US), computed tomographic (CT) or magnetic resonance imaging (MRI) guidance.3–5 An efficient and accurate biopsy service is dependent upon the co-operation and expertise of the local departments of cytology and histology. Ideally, a cytologist should be present at the time of biopsy to determine if the specimen is adequate for examination. An initial cytological report should be available within hours; when larger specimens are sent for histological examination (as may be required with lymphoreticular malignancies), processing, sectioning and staining take longer. In certain other cases, such as primary breast cancer, histological evaluation for oestrogen and progesterone receptors and genetic analysis for the HER-2/neu or the BRCA1 or BRCA2 genes may be needed. A more recent advance is the use of coaxial introducers. The introducer stays in the lesion while the sampling needle is removed for recovery of the sample. This technique allows multiple needle passes to be performed through the introducer with only one puncture through the organ of interest. The coaxial introducer is especially useful for deep lesions or lesions that are difficult to access to minimize the risk and effort inherent in obtaining multiple samples. It is possible to perform image-guided biopsy either percutaneously (transthoracic, abdominal, musculoskeletal, lymph nodes etc.) or transluminally.
Percutaneous transthoracic biopsy Percutaneous transthoracic biopsy is a rapid, safe and effective means of establishing a diagnosis in a patient with an opacity visible on a chest radiograph.6,7 The procedure is performed under local anaesthesia using biplane fluoroscopy, US or CT for guidance (Fig. 7.1). Ultrasound guidance is useful for pleural biopsy, rib lesions, subcutaneous deposits and peripheral lung lesions reaching a pleural surface. For other more central or difficult-access lesions, either fluoroscopy or CT guidance is recommended.8 In general, it is easier, quicker and less expensive to carry out a biopsy under fluoroscopic guidance than to use CT scanning for the procedure. However, CT has several advantages over fluoroscopy. ●
It is easier to visualize small lesions on CT, especially when they are located in such a position as to be
●
●
●
projected over the mediastinum or the spine on lateral fluoroscopy. Biopsies from partially necrotic lesions should be taken from the viable parts of the wall of the mass rather than from the necrotic areas. It is easier to ensure that this is done under CT guidance. When multiple lesions are present, it is difficult to be certain that one is looking at the same lesion in both the anteroposterior and lateral projections on fluoroscopy, but much easier on CT. When a mass is close to a large vascular structure, it is better to use CT for the biopsy, as the vessel is easier to avoid because of the greater precision of the procedure.
Computed tomography guidance is also particularly useful in the performance of mediastinal biopsies. It is easier to distinguish abnormal masses from vascular structures by CT than by fluoroscopy. In addition, it is possible to avoid puncturing the lung, especially when a posterior approach is used, thus minimizing the risk of pneumothorax.9 The major drawback with CT is the lack of real-time imaging and slow image display time. To overcome this limitation, CT fluoroscopy (continuous imaging CT or real-time reconstruction CT) is now being used (Fig. 7.2).10,11 The success rate of imaging-guided biopsy is approximately 80–90 per cent. The most serious complications are pneumothorax and bleeding. On most occasions these can be managed in the interventional suite, e.g. lung aspiration or chest-drain insertion for pneumothorax, and embolization for haemoptysis. Although a small pneumothorax is seen on CT images in the majority of patients after lung biopsy, it is present on plain radiograph in only 15–20 per cent of patients; less than 2 per cent require insertion of a chest drain (cf. open lung biopsy, when all patients require a drain). Needle lung biopsy is contraindicated in patients with a bleeding diathesis, emphysema, contralateral pneumonectomy, hydatid disease or suspected arteriovenous malformation, i.e. any patients who could not withstand a pneumothorax or who might bleed uncontrollably. The procedure is of value not only in pulmonary parenchymal lesions but also in the investigation of pleural and mediastinal abnormalities. If the lesion to be biopsied shows evidence of cavitation, it is useful to obtain material from both its centre and periphery; this may require multiple passes of the biopsy needle, increasing the risk of pneumothorax. It is important to realize that with percutaneous lung biopsy, as with any other form of biopsy, a single negative result cannot be taken as proof that malignancy is not present.
Percutaneous abdominal biopsy In most cases, a percutaneous abdominal biopsy is performed under US or CT guidance. Both of these methods allow the radiologist to visualize not only the lesion to be targeted for biopsy but also the intervening organs, thus
(a)
(b)
(c)
(a)
Figure 7.1 (a) Percutaneous lung biopsy. An opacity is noted in the left upper zone (a) posteriorly (b) (The triangular metallic opacities are nipple markers). The biopsy needle has been inserted under biplane fluoroscopic control (c) Reproduced with permission from Allison (1987).
(b)
Figure 7.2 Percutaneous CT fluoroscopy-guided biopsy of retroperitoneal mass: (a) CT scan, showing soft-tissue mass surrounding the aorta and inferior vena cava; (b) with the patient in the prone position, biopsy needle positioned from left approach (photon starvation artefact confirming position of needle tip). Reproduced with permission from Dr M. Roddie.
Biopsy procedures 149
(a)
(b)
Figure 7.3 A musculoskeletal biopsy. A fine-gauge needle has been inserted into a lytic lesion at the level of C5 using biplane fluoroscopic control: (a) 5AP; (b) 5 lateral. The biopsy was performed under local anaesthesia and a diagnosis of osteoclastoma was made. Reproduced with permission from Allison (1987).
increasing the accuracy and safety of the procedure. The choice between the two techniques depends on the experience and preference of the radiologist and on the precise location of the lesion to be biopsied. For example, retroperitoneal masses are usually easier to visualize on CT scans, whereas lesions in the superior part of the right lobe of the liver immediately below the diaphragm are best biopsied under US guidance because this modality allows an oblique approach that avoids the lung. Ultrasound is usually quicker to perform, does not involve ionizing radiation and is available as a portable technique. In patients with a bleeding diathesis or massive ascites, percutaneous liver biopsy carries an increased risk of major, uncontrollable intraperitoneal haemorrhage. In such patients it is possible to obtain a liver biopsy via a transvascular route.12,13* Specially designed biopsy instruments can be passed via the right internal jugular vein into the hepatic veins. Bleeding arising from a biopsy taken from this position passes directly back into the patient’s circulation and is not therefore of haemodynamic significance. Although successful, this technique is technically demanding and not applicable to focal lesions. An alternative method of percutaneous liver biopsy, which can be used in patients with abnormal blood coagulation, is plugged liver biopsy. The needle is introduced into the liver within a guiding sheath. The biopsy is performed
and the needle is withdrawn, leaving the sheath in place. The biopsy track is then embolized with Sterispon or steel coils as the sheath is withdrawn.14,15
Musculoskeletal biopsy The presence of an abnormality within bone on an isotopic bone scan and/or plain radiography in a patient with known malignancy does not necessarily herald dissemination of the tumour, and it may prove necessary to obtain material for cytology.16,17 Solitary bony abnormalities discovered incidentally in a patient who is otherwise well also pose a difficult diagnostic problem. Lytic lesions are relatively straightforward to biopsy (Fig. 7.3), whereas sclerotic lesions give rise to more difficulties and the procedure may need to be performed under very heavy sedation or general anaesthesia. Computed tomography is extremely useful during the performance of musculoskeletal biopsy because it allows the needle to be guided precisely to the abnormal area. Quite frequently, a soft-tissue mass may be associated with lytic bone lesions, and CT allows this to be visualized and biopsied. In addition, CT demonstrates areas where the cortex is thinnest and thus easiest to penetrate with a biopsy needle. Another advantage of CT, especially when biopsying lesions in thoracic vertebrae, is that the needle
150 Interventional radiology
can be guided extrapleurally, thus avoiding the danger of pneumothorax. In a few cases, US guidance is sufficient. The great benefits of percutaneous biopsy are that the recorded incidence of complications is exceptionally low, the overall success rate is high (70–80 per cent) and the procedure is readily repeatable if the initial result is inconclusive.
Miscellaneous biopsy techniques When a mass extends into the superior or inferior vena cava, the transvascular route provides a safe and convenient way of obtaining a biopsy. The femoral vein is punctured at the groin, and a long sheath is introduced into the inferior vena cava and advanced to lie adjacent to the mass to be biopsied. A modified cardiac bioptome is then inserted via the sheath and the biopsy taken.18 Developments in endoluminal US have allowed lesions deep in the pelvis to be biopsied transrectally (prostate biopsies) or transvaginally (ovarian cysts). The advent of endoscopic and intra-operative US probes has further extended the versatility of this imaging modality (pancreatic, tracheal, oesophageal, colonic).19,20 In the management of breast cancer, non-palpable lesions are biopsied or localized using sterotactic mammography, US or MRI for guidance. In some instances with no apparent safe percutaneous approach, sufficient displacement of vital structures can be achieved by adjacent saline installation to allow safe biospy needle placement. This technique has been applied in the anterior mediastinum to displace internal mammary vessels and in the posterior mediastinum to displace lung. It has also been safely applied elsewhere in the abdomen and pelvis as needed.21
requires the percutaneous transhepatic cannulation of the portal veins. The reader is referred to more specialized texts for detailed descriptions of these techniques.
PERCUTANEOUS PUNCTURE AND DRAINAGE PROCEDURES With the use of fluoroscopy, US, CT or MRI, it is possible to image and drain obstructed renal and biliary systems, cysts, abscesses and effusions.
Renal tract Antegrade pyelography and percutaneous nephrostomy are useful in the management of benign and malignant obstruction of the urinary tract.23 The technique is also of value in a variety of other situations, e.g. in patients with malignancy undergoing chemotherapy resulting in haemorrhagic cystitis, in whom it is desirable to divert the urine to ‘rest’ the bladder. In patients with pelvic malignancy, either the disease or the treatment may result in the development of fistulas between the bladder and rectum or vagina, leading to incontinence. Diversion of urinary flow may assist in healing of the fistulas, ease nursing problems and allow patients to become ‘dry’. The procedure, usually performed under local anaesthetic and US guidance, involves an initial puncture of the pelvicalyceal system with a fine needle, followed by the instillation of radiographic contrast medium to demonstrate the anatomy and determine the level of obstruction. Urine can be aspirated for microbiological and cytological examination.
VENOUS SAMPLING The diagnosis of a hormone-secreting neoplasm is made on the basis of the clinical history and characteristic biochemical abnormalities. The localization of these tumours may, however, be quite difficult. Ultrasound and CT scanning of the common sites for tumours (e.g. adrenal glands for phaeochromocytomas, pancreas for insulinomas etc.) will reveal the majority of tumours. However, a few are either too small to be detected by these means or are found in ectopic sites. In these instances, venous sampling techniques can prove invaluable.22* Both systemic and portal blood specimens are obtained from various sites in the venous system and the areas sampled recorded on a map (Fig. 7.4). The specimens are numbered accordingly and taken immediately for appropriate analysis. Areas of high hormone concentration can then be charted on the map and searched by US, CT, angiography, MRI or surgery. Whereas systemic venous sampling is relatively safe and has a high success rate, localization of pancreatic or gastrointestinal hormoneproducing tumours is a more hazardous technique, as it
Figure 7.4 Systemic venous sampling map. A diagram of a systemic venous system indicates the main venous tributaries. Each of these vessels can be selectively catheterized and a sample of blood obtained for hormone analysis.
Percutaneous tumour ablation 151
Percutaneous nephrostomy entails the insertion of a pigtail catheter with multiple large side ports into the collecting system (Fig. 7.5). The catheter is introduced employing a needle, guide-wire and catheter exchange technique.
Biliary tract Likewise, in patients with infected obstruction to their biliary tree, external pigtail catheters or, preferably, internal–external biliary pigtail catheters can be inserted for temporary drainage.
Abscess drainage Percutaneous puncture of an abscess cavity and aspiration of contents for bacteriological analysis can be followed by insertion of a drainage catheter.24 It is possible to instill antibiotics into the cavity, and percutaneous drainage may be effective either as the definitive treatment or as a temporary measure until the appropriate surgery can be contemplated. Ultrasound, CT or MRI guidance may be used for abscess drainage. Computed tomography is particularly useful for retroperitoneal or mediastinal abscesses, for which visualization by US scanning may be inadequate. Transvaginal or transrectal drainage under US guidance is very useful for certain abscesses in the pelvis.
PERCUTANEOUS TUMOUR ABLATION
Figure 7.5 Percutaneous renal drainage. A patient with obstruction of the renal tract secondary to prostatic malignancy (note the sclerotic vertebral body due to metastatic infiltration). A pigtail catheter has been introduced under local anaesthetic into the dilated pelvicalyceal system and ureter. Urine is able to drain via the side ports in the catheter into a drainage bag on the skin surface. Reproduced with permission from Allison (1987).
(a)
In recent years, local ablative techniques have been employed successfully to treat a variety of tumours. Lesional heating techniques such as radiofrequency (RF) ablation and interstitial laser photocoagulation (ILP) achieve tumour necrosis by hyperthermia. The RF electrodes or laser fibres are inserted into the tumour under US, CT or MRI guidance (Fig. 7.6). Intralesional hyperthermia causes almost immediate coagulation necrosis. Radiofrequency waves induce ionic agitation, which results in frictional heat production within the tissue. Serious
(b)
Figure 7.6 Percutaneous liver tumour ablation: (a) axial CT scan showing radiofrequency electrode positioned within metastasis in liver before treatment; (b) after treatment, an area of necrosis is seen, indicating successful ablation.
152 Interventional radiology
complications are rare and consist mainly of intraperitoneal haemorrhage and liver abscess formation. Interstitial laser photocoagulation produces thermal coagulation by the conversion of absorbed light energy into heat. Hepatic resection is the mainstay in the curative management of primary and secondary hepatic malignancies. Surgical resection significantly improves survival, but only a minority of patients are surgical candidates because of the size or location of the lesions, extrahepatic disease or, in the case of hepatoma, limited hepatic functional reserve secondary to associated cirrhosis. Percutaneous techniques of local tumour ablation may be categorized into three major groups: injectables (ethanol, acetic acid, hot saline), heating (RF, electrocautery, interstitial laser therapy, microwave coagulation therapy and high-intensity focus US) and freezing (cryotherapy). Of these, the most widely used is RF ablation both for hepatocellular carcinoma and for hepatic metastases. Percutaneous ethanol injection therapy (PEIT) was first described in 198325 and since that time has been used extensively for the treatment of unresectable hepatocellular carcinoma.26 Surgery is the treatment of choice for most primary renal tumours. However, patients who are not good surgical candidates, have multiple recurrent tumours (as with von Hippel Lindau syndrome), borderline renal function or a solitary kidney can benefit from treatment with thermal ablation, which has a technical success rate exceeding 90 per cent in several series. Peripheral, small tumours are the easiest to treat (Fig. 7.7). Treating larger, central neoplasms is less successful and more likely to be associated with complications such as ureteric strictures. Techniques such as cooling of the ureter during treatment can help reduce the likelihood of injuring adjacent organs. Radiofrequency treatment is also being used to treat unresectable pulmonary neoplasms and for palliation of bone tumours. The follow-up of patients after all forms of percutaneous tumour ablation includes a combination of imaging, tumour marker assay and selected use of fine-needle aspiration biopsy when the serum levels of these markers are elevated prior to the initiation of therapy. The immediate goal of imaging is to assess whether complete necrosis has been achieved. Ultrasound does not usually provide useful information, as the echogenicity of fibrosis and neoplastic tissue overlaps. Contrast-enhanced MRI and contrastenhanced CT are capable of demonstrating remaining viable tumour requiring treatment. However, in difficult cases, PET scanning may provide additional information. Better methods of imaging guidance and more sophisticated equipment are likely to increase the importance of percutaneous tumour ablation in the future.
Gastrointestinal tract In the management of patients with malignant disease, nutritional support is essential, particularly in those who
(a)
(b) Figure 7.7 Patient with left renal cancer considered as unsuitable surgical candidate treated with percutaneous radiofrequency ablation.(a) contrast enhanced CT showing peripheral left renal cancer, incidental right simple renal cyst. (b) triple needle RFA probe in situ within tumour lesion.
are severely debilitated or unable to swallow. Parenteral nutrition can be provided using central venous catheters, but this is associated with significant morbidity and considerable expense. Enteric feeding is more desirable and can be provided by the insertion of gastrostomy tubes. Gastrojejunostomy tubes are preferable when there is gastric outlet obstruction or in cases of gastro-oesophageal reflux. Several different techniques have been described for percutaneous gastrostomy insertions27,28 with and without gastropexy under fluoroscopic guidance (Fig. 7.8).
Dilatation techniques 153
Figure 7.8 Fluoroscopic insertion of gastrostomy feeding tube. Gastrostomy catheter is locked with pigtail in fundus of stomach. Nasogastric tube and the two ‘stay’ sutures used for the safe insertion of gastrostomy tube are then removed.
Gastrostomy may also be useful in relieving symptoms in patients with gastric outlet obstruction due to retroperitoneal or mesenteric malignancy when insertion of a stent in the pylorus fails to relieve the obstruction. Percutaneous puncture for decompression of the caecum (in cases of distal obstruction) and for afferent loop obstruction have been described.29,30
DILATATION TECHNIQUES Dilatation procedures are most widely employed for nonmalignant conditions in the vascular tree. These techniques can also be applied to stenoses and occlusions in other systems, such as the gastrointestinal tract, renal, biliary and respiratory systems. Fluoroscopically guided oesophageal dilatation has proved to be a particularly useful technique.31* A guide-wire is manipulated through the narrowed area and a balloon catheter of suitable dimensions is then passed over the wire. Intravenous analgesia should be given immediately prior to balloon inflation, as this can cause moderate discomfort. Dilatation alone is unlikely to be effective in malignant oesophageal strictures and should be followed by some form of stenting.32 Rigid plastic tubes inserted endoscopically or under fluoroscopic guidance have been used for several years. However, recently, self-expandable metallic endoprostheses have become available.33,34* With recent advances35
and increased experience in the use of metallic stents, more than 95 per cent of patients with inoperable oesophageal strictures can be palliated successfully with these devices (Fig. 7.9). Strictures at other sites in the gastrointestinal tract can also be treated successfully by radiological dilatation techniques. The relief of obstruction at anastomotic strictures (gastroenterostomies, antral–pyloric strictures, enteroenterostomies and colorectal strictures) is particularly useful.36 In cases of acute malignant obstruction of the large bowel, the insertion of self-expanding stents can provide immediate relief (Fig. 7.10). This may be a temporizing measure, allowing stabilization of the patient prior to definitive surgery, or, alternatively, in patients who are not surgical candidates, the stent may provide adequate palliation.37 Malignant airways obstruction can cause considerable distress to patients. When surgical resection is not possible, self-expanding metallic stents can provide good palliation (Fig. 7.11).38 Plastic-covered stents are very effective in managing tracheo-oesophageal fistulas unsuitable for treatment with covered oesophageal stents.39 The stents are inserted under general anaesthesia or sedation with local anaesthetic using combined fluoroscopic and bronchoscopic guidance, following balloon dilatation of the strictures.40 Within the urinary tract, in the presence of obstruction, if drainage is to be of short duration, it is satisfactory to allow the urine to drain externally into a bag. When long-term drainage is required, a J-J stent is inserted with the proximal pigtail coiled in the renal pelvis and the distal end in the bladder,41 thus allowing the patient to be free of ‘bags’. Patients with obstructive jaundice due to unresectable malignant biliary strictures can be palliated by the insertion of endoprostheses. For surgical candidates, preoperative biliary drainage may be performed to correct metabolic disturbances and stabilize the patient. Most patients with malignant obstructive jaundice undergo endoscopic retrograde cholangiopancreaticography (ERCP) as part of the diagnostic work-up. If ERCP confirms a malignant stricture, an endoprosthesis can be inserted immediately after cholangiography. Endoscopic drainage is less invasive than percutaneous biliary drainage (PBD), is associated with fewer complications and avoids the discomfort of a percutaneous biliary catheter. The majority of strictures of the mid and lower common bile ducts, which are mainly due to carcinoma of the head of the pancreas, can be drained effectively by the endoscopic approach, the percutaneous approach being reserved for patients in whom endoscopic drainage attempts have failed. Patients with obstructive jaundice due to malignant tumours at the hilum of the liver are best palliated by percutaneous drainage methods, as the endoscopic approach in such patients often fails and is associated with a high rate of complications. In most cases, unilateral drainage is sufficient for relieving jaundice and pruritus. The procedure is usually carried out under fluoroscopic and US guidance. First a percutaneous transhepatic cholangiogram (PTC) is performed, using a 22-gauge needle. It is important to visualize the entire biliary system prior to selecting the most suitable duct for the insertion of a
154 Interventional radiology
(a)
(b)
Figure 7.9 Oesophageal stenting: (a) contrast swallow showing lower oesophageal carcinoma stricture; (b) Flamingo Wallstent (selfexpanding metallic endoprosthesis) has been placed fluoroscopically across the stricture. Contrast is now flowing freely into the stomach.
drainage catheter. Self-expandable metallic stents are now widely available and generally preferable to the conventional plastic endoprostheses. Such stents can be inserted using a relatively small introducing catheter and yet achieve a large internal diameter when released across the lesion (Fig. 7.12). The large calibre of these devices ensures that the rate of occlusion is lower than that of plastic endoprostheses. In our experience, the rate of haemorrhage and cholangitis is approximately three times lower with metallic endoprostheses than with plastic stents. Most importantly, the rate of re-intervention for the purpose of replacement of occluded stents has also been found to be three times lower with metallic stents.42 A randomized comparison of endoscopically inserted self-expanding metal stents and plastic stents has confirmed the longer patency of metallic endoprostheses and has shown that although these stents are more expensive than the conventional plastic devices, the cost per patient is lower because of the lower rate of re-intervention.43***
Venous dilatation and stenting The superior vena caval syndrome is most commonly related to mediastinal neoplasia, particularly primary and secondary lung tumours and lymphoma. The obstruction, which can be partial or complete, may be caused by caval compression or invasion by the tumour, and is sometimes complicated by venous thrombosis. Cavography delineates the site and extent of the obstruction. Percutaneous transfemoral dilatation of the narrowed cava, combined with thrombolysis where necessary, followed by the insertion of self-expandable metallic endoprostheses, restores flow and provides excellent palliation of symptoms (Fig. 7.13).44 Malignant involvement of the inferior vena cava can be managed in a similar fashion. Percutaneous insertion of an inferior vena cava filter is indicated in patients with recurrent pulmonary embolism refractory to, or unsuitable for, treatment with medical therapy. In patients with iliac or
Dilatation techniques 155
(a)
(b)
(c)
Figure 7.10 Colonic stenting. (a) Contrast study of the sigmoid colon; lateral view of the pelvis, showing carcinoma stricture of the sigmoid colon. A guide-wire and catheter have been placed across the stricture. (b) A self-expanding metallic endoprosthesis has been deployed across the stricture. (c) CT scan, showing the expanded stent within the sigmoid tumour. The patient’s obstructive symptoms were successfully palliated.
(a)
(b)
Figure 7.11 Tracheal stenting. (a) Axial CT scan of a section through the chest, showing tumour and nodal mass encasing and compressing both main bronchi. (b) Chest X-ray after deployment of two self-expanding metal endoprostheses situated in the trachea and extending into both main bronchi.
156 Interventional radiology
(b)
(a)
(c)
Figure 7.12 (a) Percutaneous transhepatic cholangiogram in a patient with hilar cholangiocarcinoma involving the left and right hepatic ducts and the upper common hepatic duct. (b) Self-expandable metallic endoprostheses have been inserted into both hepatic ducts using separate punctures. Drainage of contrast medium is taking place into the duodenum. (c) The stents are shown in position following complete drainage of the contrast medium.
iliofemoral venous thrombosis due to pelvic malignancy, a temporary filter inserted prior to surgical resection of the tumour may protect from pulmonary embolism.
EXTRACTION TECHNIQUES Developments in intravenous feeding therapy and monitoring techniques have led to a vast increase in the number of indwelling venous cannulas and catheters. Unfortunately, these occasionally break or become disconnected, resulting in loss of part or all of the catheter within the venous system.45 It is important to retrieve these intravascular foreign bodies as they can perforate vascular structures and cause
dysrhythmias, and can be a source of infection, particularly in immunosuppressed patients. Surgical retrieval of catheter fragments necessitates a thoracotomy and carries significant risk. It is almost invariably possible to retrieve these catheter fragments percutaneously under fluoroscopic guidance (Fig. 7.14).46 They usually lodge within the right side of the heart or the pulmonary arteries. A wide variety of instruments have been found to be useful in ‘catching’ these foreign bodies, including wire-loop snares, grasping forceps, steering catheters and Dormia baskets. It is usually possible to retrieve the foreign body under local anaesthesia via a percutaneous femoral venous puncture. Detailed descriptions of all the techniques available are beyond the scope of this chapter, but any interventional
Extraction techniques 157
(a)
(b)
(c)
Figure 7.13 Superior vena caval stenting. (a) Superior vena cavogram reveals severe narrowing of the superior vena cava (SVC) by tumour in the mediastinum. (b) Metal stents have been placed across the compressed area, which was initially dilated with a balloon. (c) A repeat venogram confirms patency of the SVC. The patient’s symptoms improved immediately and had resolved completely within 24 hours.
158 Interventional radiology
(b)
(a)
radiologist offering a comprehensive vascular service is well advised to acquaint himself or herself with the various methods and have the necessary equipment available.47 The ability to snare or ‘catch’ the end of a catheter can also be of value in patients receiving intravenous cytotoxic chemotherapy in whom the tip of an indwelling central venous catheter has become displaced and lodged in the jugular vein instead of the superior vena cava. It is usually possible to ‘pull’ such a catheter back to the appropriate position using a percutaneous vascular approach under local anaesthesia.
INFUSION TECHNIQUES The ability of the radiologist to site a catheter accurately into virtually any blood vessel within the body has brought into play the concepts of regional infusion of chemotherapeutic agents, monoclonal antibodies and isotopes. The principle underlying these techniques is that a high dose of the therapeutic agent is delivered to the tumour(s) with minimal systemic side effects. Recent advances include the addition of embolic materials (see ‘Vascular embolization’ section below) and the microencapsulation of cytotoxic agents to achieve gradual and sustained release. Lipiodol is now also used in combination with cytotoxic drugs for the treatment of certain hepatic tumours; it is retained in tumour vessels and also acts as a contrast agent, thus allowing monitoring of tumour response to treatment using CT guidance. These techniques have been used with varying degrees of success to treat primary and metastatic liver tumours,48 bone neoplasms, cerebral neoplasms, sarcomas, melanomas and pelvic neoplasms. It is possible to insert fine catheters that can be left in place for weeks at a time.49 A variety of long-term venous access lines (tunnelled central venous catheters, peripherally inserted central catheters (PICC) and Portacaths) have been developed for insertion under fluoroscopic and US guidance.50 Radiological techniques are best suited for their insertion.51 Tunnelled
Figure 7.14 (a) A chest radiograph in a young girl undergoing cytotoxic therapy in whom part of a central venous catheter became detached and migrated into a right lower lobe pulmonary artery (arrows). (b) The catheter fragment after extraction via the femoral vein, grasped in the catheter used for retrieval.
external catheters are most frequently used. The preferred access sites are the internal jugular and subclavian veins, but other veins can be used in cases of difficulty, including the common femoral, translumbar, inferior vena cava and hepatic veins. The PICC lines are more usually inserted for shorter duration of therapy, with the basilic vein being the preferred site of puncture.
VASCULAR EMBOLIZATION This technique involves the deliberate occlusion of arteries and/or veins by the injection of embolic agents through selectively placed catheters. It is one of the major applications of IR in patients with neoplastic disease and has been employed in the management of a wide variety of tumours. Embolization, usually performed by a percutaneous approach under local anaesthesia, offers an attractive alternative to surgery and in some situations it is the only therapeutic option available. A wide variety of embolic agents is available52 and a detailed description is beyond the scope of this chapter. The broad categories of substances used include particulate emboli (sterile sponge [Spongostan], polyvinyl alcohol [Ivalon]), mechanical emboli (balloons, steel coils) and liquids (50 per cent dextrose, absolute alcohol, lipiodol). The appropriate agent or combination of agents depends on the lesion to be treated and its site, with particular attention paid to adjacent vulnerable vascular structures. There are three ways in which embolization can assist in the management of neoplasms: definitive, preoperative or palliative. Definitive treatment can only be offered for benign lesions.
Preoperative embolization This has been widely employed in the management of renal carcinomas, although not all urologists find the technique useful. The tumour is embolized and surgery
Vascular embolization 159
(a)
(b)
(c)
(d)
(e)
Figure 7.15 Preoperative embolization of a bone neoplasm (osteoclastoma) in a 17-year-old girl which had proved resistant to radiotherapy. A plain radiograph (a) reveals a large soft-tissue mass with extensive bone destruction. Axillary arteriography shows the lesion to be highly vascular. (b) The feeding vessels selectively catheterized and embolized. The postembolization angiogram (c) shows the lesion to be completely devascularized. A plain radiograph 3 months later (d) shows considerable reduction in the size of the mass and bone regrowth. This improvement allowed reconstructive surgery to be performed. A humeral replacement (e) was successfully carried out. Reproduced with permission from Allison (1987).
undertaken within 24–48 hours, while there is maximum devascularization and minimal oedema. Preoperative embolization reduces blood loss, and may minimize dissemination of malignant cells during tumour mobilization.
Preoperative tumour embolization has proven to be of value in other situations, including nasopharyngeal tumours, paragangliomas, meningiomas and bone tumours (Fig. 7.15). In the last-mentioned example, the inclusion of
160 Interventional radiology
a cytotoxic agent within the embolization ‘cocktail’ may further reduce the risk of tumour dissemination during surgery.
Palliative embolization This technique is employed to control pain, haemorrhage and hormone production, as well as to reduce tumour bulk.
It may be used as the primary mode of treatment in inoperable malignancy. Embolization of metastatic deposits has, in some situations, been shown to extend survival times in advanced disease.53 Tumours in many organs have been treated in this fashion: liver, kidney (Fig. 7.16), bone, lung, soft tissues, nervous system and gastrointestinal tract. Hormone-secreting neoplasms, e.g. metastatic carcinoid and APUD cell tumours (Fig. 7.17), show the greatest therapeutic response to arterial embolization. Appropriate
(a)
(b)
(c)
Figure 7.16 Selective right renal arteriography, (a) arterial and (b) and (c) venous phases, reveals a highly vascular renal tumour in the lower pole of the kidney. The vessels supplying the tumour were selectively catheterized and embolized with particulate material, depriving it of its blood supply and relieving the patient of pain and haematuria. Reproduced with permission from Allison (1987).
Vascular embolization 161
(a)
(b)
(c)
Figure 7.17 Hepatic embolization for metastatic carcinoid tumour. (a) The early arterial phase shows hepatic enlargement. (b) The parenchymal phase reveals multiple tumour deposits. (c) Post-embolization arteriogram shows that the arterial supply has been obliterated. The patient’s symptoms (flushing and diarrhoea) were dramatically alleviated by this procedure. Reproduced with permission from Allison (1987).
162 Interventional radiology
pharmacological blockade is necessary during the embolization to avoid the effects of a massive outpouring of hormone as the tumour is deprived of its blood supply. The beneficial effects of embolization may become apparent within a matter of hours. In embolization procedures it is important that adequate premedication is given prior to the procedure, including broad-spectrum antibiotics. In many situations, e.g. liver and bone, it is advisable to continue antibiotics for 10 days after the procedure to prevent sepsis developing in the devascularized tissue. After embolization, patients commonly experience some discomfort and pain. They may have a pyrexia for a few days, accompanied by a feeling of malaise and an elevated white cell count. This combination of signs and symptoms is known as the post-embolization syndrome (PES), an indicator of the presence of necrotic tissue. Any sustained pyrexia should alert the clinician to the possibility of abscess formation, and blood cultures and regional US should be performed. Serum C-reactive protein (CRP) estimations can also provide a useful indication that infection may be present. Following embolization, CRP reaches a peak at about 4 days, falling off thereafter to reach normal levels at 10 days. A continued CRP rise, or failure to fall, indicates infection.54 In patients with pelvic malignancy and fistulas to the perineum, percutaneous nephrostomy is often insufficient to achieve complete diversion of the urine. It is often necessary to combine nephrostomy with ureteric embolization using steel coils and fragments of gelatin sponge, in order to ensure that no urine reaches the skin of the perineum.55
Chemoembolization Embolic agents in combination with chemotherapeutic drugs are being used in the treatment of certain malignancies.
(a)
In this technique, known as chemoembolization,56 the emboli cause ischaemia of the tumour cells and, by increasing the transit time through the tumour vascular bed, the contact time between the cytotoxic agent and the neoplastic cells is prolonged, resulting in a greater therapeutic effect.
NEUROLYSIS
Alcohol injection is useful in the palliation of the intractable pain that occasionally accompanies retroperitoneal malignancy. For example, certain patients with carcinoma of the pancreas experience severe pain due to infiltration of the coeliac ganglion by the tumour. In these patients the ganglion may be ablated by injecting alcohol in its immediate vicinity under CT guidance (Fig. 7.18).
Percutaneous vertebroplasty Metastases to the vertebrae are common. Radiation therapy is not always effective in relieving the pain. Percutaneous vertebroplasty provides immediate and sustained pain relief, and contributes to spinal stabilization.57 The procedure consists of percutaneous injection of acrylic surgical cement into a vertebra under radiological guidance (fluoroscopy and/or CT) (Fig. 7.19). This technique can also be used for treated bone metastases at other sites. Up to 80 per cent of patients with pain unresponsive to conventional treatment experience a significant degree of pain relief, and few serious side effects have been reported. Kyphoplasty is a modified technique that involves the additional step of high-pressure balloon dilatation prior to cement injection.
(b)
Figure 7.18 Coeliac axis block for intractable pain in a patient with pancreatic carcinoma. (a) Axial CT scan with a 22-gauge needle positioned with its tip at level of the coeliac ganglion. (b) Bilateral spill seen following injection of alcohol.
Gene therapy 163
AORTIC STENTING
The effectiveness of aortic abdominal aneurysm stent grafts has been recognized.58** The experience with endovascular repair in the management of descending thoracic aortic diseases such as thoracic aneurysm, perforated ulcer, intramural haematoma and type B dissection is also very encouraging.59
Amongst the advantages of endovascular techniques over open surgery are the fact that they can be performed in lessfit patients,60 a lower rate of complications, lower mortality rates and reduced hospital stay. CAROTID STENTING
Carotid artery stenting is useful in symptomatic patients, especially those unfit for surgery and particularly those who have had previous head and neck surgery or radiation (Fig. 7.20). The principal reason for intervening in the first place is for the prevention of stroke. Advantages include shorter hospital stay, avoidance of a wound and its associated complications (pain, infection, numbness, haematoma and patch infection), avoidance of cranial nerve injury, and reduced cardiovascular complications (i.e. myocardial infarction).61
GENE THERAPY
(a)
(b) Figure 7.19 Vertebroplasty for relief of pain and provision of bony stability. (a) Fluoroscopy image of needle within diseased vertebral body (b) Completion image after 3ml of cement injection.
The most rapidly evolving area in medicine is gene therapy. The underlying principle is to identify and clone a gene, and then to insert it into a vector capable of directing expression in mammalian tissues.62 The main aim at present is to treat genetic deficiencies and malignant diseases that are refractory to conventional therapies. The delivery systems involved include retroviral vectors (RNA viruses), adenoviral vectors (DNA viruses) and cationic liposomes, along with strategies that involve US-directed gene transfer, CT-guided gene transfer and transcatheter gene delivery, in particular via the hepatic artery. Examples of genes being evaluated in trials include oncogenes, tumour suppressor genes, suicide genes and anti-angiogenesis factors. The liver is an ideal therapeutic target for gene therapy. Hepatic malignancies being considered for treatment include metastatic colorectal carcinoma, hepatoma, cholangiocarcinoma, lymphoma, metastatic melanoma and haemangioma. Gene therapy strategies for managing occluded biliary stents (resulting from tumour ingrowth) and vascular transjugular intrahepatic portosystemic stents (resulting from neo-endothelialization) are also under consideration. In gene delivery, angiographic guidance will be of use in localizing tumour blood supply and directing the targeted intra-arterial delivery of genes of interest, so that vector–DNA complexes can be delivered with accuracy and specificity. Embolization techniques may also be of benefit, by prolonging vector contact with the target cells, thus delaying washout and further enhancing target-cell uptake. Radiological monitoring will be of considerable importance during gene delivery, e.g. the process of liposomal vector delivery can be monitored accurately with US because lipid vesicles are echogenic. Guided biopsy of transduced tissues for histopathologic analysis after gene delivery should also improve confidence in the evaluation of gene expression.
164 Interventional radiology
(a)
(c)
(b)
Figure 7.20 Carotid artery stent in a symptomatic patient with previous laryngeal cancer resection and radiotherapy. (a) angiogram of carotid vessels; arrow pointing to radiation stricture right common carotid artery. Note absent left carotid artery (previously occluded) (b) Selective angiogram of right carotid artery (c) Post stent angiogram demonstrating good flow within carotid artery.
References 165
◆8
KEY LEARNING POINTS ●
●
●
●
●
●
●
An interventional procedure should only be contemplated if it is likely to benefit the patient’s quality of life. Some interventional procedures are replacing surgical alternatives; others are totally new forms of treatment. Diagnostic invasive procedures, e.g. biopsy, are significantly less traumatic than their surgical alternatives. All interventional procedures require a team approach to ensure the best possible outcome for the patient. Advances in imaging technology and instrumentation are constantly occurring, leading to the emergence of new interventional techniques. Many interventional procedures, e.g. tunnelled central venous catheter insertion, SVC stenting and percutaneous drainage, are carried out in conjunction with, and enable the more effective use of, conventional forms of cancer therapy. Interventional radiology is likely to play a critical role in the exciting new field of gene therapy.
●9
10
●11
12 ◆13
●14
●15
16 ◆17
●18
19
REFERENCES 1 Adam A. (1998) The definition of interventional radiology (or, ‘When is a barium enema an interventional procedure?’). Eur Radiol 1998; 8:1014–15. 2 Bogda K. Radiological Interventions: Special considerations in cancer patients. In: Ray CE, Hicks ME, Patel NH (eds) SIR syllabus-Interventions in Oncology, North Fairfax, Virginia: Society of Interventional Radiology, 2003, 1–7. 3 Haubeck A, Gammelgaard G, Gronvall S, Holm HH. Ultrasonically guided percutaneous puncture and biopsy technique. In: Wilkins RA, Viamonte MJR (eds) Interventional Radiology. Oxford: Blackwell Scientific Publications, 1982, 373–408. 4 Husband JE, Golding SJ. Recent developments in whole body computed tomography. In: Steiner RE (ed.) Recent Advances in Radiology, Vol. 7. Edinburgh: Churchill Livingstone, 1983, 88–106. ◆5 Mueller PR, Stark DD, Simeone JF, et al. MR-guided aspiration biopsy: needle design and clinical trials. Radiology 1986; 161:605–9. ●6 Allison DJ, Hemingway AP. Percutaneous needle biopsy of the lung. Br Med J 1981; 282:875–8. 7 Greene RE. Transthoracic needle aspiration biopsy. In: Athanasoulis CA, Pfister RC, Greene RE, Robertson GH (eds) Interventional Radiology. Philadelphia: WB Saunders, 1981, 587–637.
20
21
22
23
◆24
25
◆26
●27
van Sonnenberg E, Casola G, Ho M, et al. Difficult thoracic lesions: CT-guided biopsy experience in 150 cases. Radiology 1988; 167:457–61. Adam A, McSweeney JE, Whyte MKB, et al. CT-guided extra-pleural drainage of bronchogenic cyst. J Comput Assist Tomogr 1989; 13:1065–8. Katada K, Kato R, Anno H, et al. Guidance with real-time CT fluoroscopy: early clinical experience. Radiology 1996; 200:851–6. Meyer CA, White C, Wu J, et al. Real-time CT fluoroscopy: usefulness in thoracic drainage. AJR 1998; 171:1097–101. Gilmore IT, Bradley RD, Thompson RPH. Improved method of transvenous liver biopsy. Br Med J 1978; ii:249. Gamble PM, Colapinto RF, Stronell RG, et al. Transjugular liver biopsy: a review of 461 biopsies. Radiology 1985; 157:589–93. Allison DJ, Adam A. Percutaneous liver biopsy followed by embolization of the track with steel coils. Radiology 1988; 169:261–3. Dawson P, Adam A, Edwards R. Technique for steel coilembolization of liver biopsy track for use with the ‘Biopty’ needle. Br J Radiol 1992; 65:538–40. Armstrong P, Calmers AH. Needle aspiration of the spine in suspected disc space infection. Br J Radiol 1978; 51:333–7. Laredo JD, Bellaiche L, Hamze B, et al. Current status of musculoskeletal interventional radiology. Radiol Clin North Am 1994; 32:377–98. Jackson J, Adam A. Percutaneous transcaval tumour biopsy using a ‘road map’ technique. Clin Radiol 1991; 44:195–6. Machi J, Sigel B, Kurohisi T, et al. Operative ultrasound guidance for various surgical procedures. Ultrasound Med Biol 1990; 16:37–42. Ziegler K, Sanft C, Zimmer T, et al. Comparison of computed tomography, endosonography and intraoperative assessment in TN staging of gastric carcinoma. Gut 1993; 34:604–10. Karampekios S, Hatjidakis AA, Drositis J, Gourtsoyiannis N. Artificial paravertebral widening for percutaneous CT-guided adrenal biopsy. J Comput Assist Tomogr 1998; 22(2):308–10. Allison DJ. Therapeutic embolization and venous sampling. In: Taylor S (ed.) Recent Advances in Surgery, Vol. 10. Edinburgh: Churchill Livingstone, 1980, 27–64. Papanicolaou N. Uroradiological intervention. In: Watkinson A, Adam A (eds) Interventional Radiology, a Practical Guide. Oxford: Radcliffe Medical Press, 1996, 88–118. Gerzof SG, Spira R, Robins AH. Percutaneous abscess drainage. Semin Roentgenol 1981; 16:62–71. Sugiura N, Takara K, Ohto M, et al. Percutaneous intratumoral injection of ethanol under ultrasound imaging for treatment of small hepatocellular carcinoma. Acta Hepatol Jpn 1983; 24:920–3. Livraghi T, Solbiati L. Percutaneous ethanol injection in liver cancer: methods and results. Semin Intervent Radiol 1993; 10(2):69–77. van Sonnenberg E, Wittich GR, Cabrera OA, et al. Percutaneous gastrostomy and gastroenterostomy: 2. Clinical experience. AJR 1986; 146:581–6.
166 Interventional radiology
28 Saini SJ, Mueller PR, Gaa J, et al. Percutaneous gastrostomy with gastropexy: experience in 125 patients. AJR 1990; 154:1003–6. ●29 Casola J, Withers C, van Sonnenberg E, et al. Percutaneous cecostomy for decompression of the massively distended cecum. Radiology 1986; 158:793–4. ●30 Lee LI, Teplick SK, Haskin PN, et al. Refactory afferent loop problems: percutaneous transhepatic management of two cases. Radiology 1987; 165:49–50. ◆31 Sabharwal T, Cowling M, Dussek M, Owen W, Adam A. Balloon dilatation myotomy in oesophageal achalasia; a safe and effective technique. Radiology 2002; 224:719–24. ◆32 Sabharwal T, Morales JP, Salter R, Adam A. Esophageal cancer: self-expanding metallic stents. Abdom Imaging 2004; 29:1–9. 33 Neuhaus H. Metal oesophageal stents. Semin Intervent Radiol 1991; 8:305–10. 34 Song H-Y, Do YS, Han YM, et al. Covered, expandable oesophageal metallic stent tubes: experiences in 119 patients. Radiology 1994; 193:689–95. 35 Sabharwal T, Morales JP, Irani FG, Adam A. Quality assurance guidelines for placement of oesophageal stents. Cardiovasc Intervent Radiol 2005; 28:284–8. 36 Grundy A. The radiological management of gastrointestinal strictures and other obstructive lesions. In: Adam A, Allison DJ (eds) Clinical Gastroenterology. Interventional Radiology of the Abdomen, Vol. 6, No. 2. London: Bailliere Tindall, 1992, 319–40. ●37 De Gregorio MA, Mainar A, Tejero E, et al. Acute colorectal obstruction: stent placement for palliative treatment – results of a multicenter study. Radiology 1998; 209:117–20. ●38 Tan BS, Watkinson AF, Dussek JE, Adam A. Metallic endoprosthesis for malignant tracheo-bronchial obstruction: initial experience. Cardiovasc Intervent Radiol 1996; 19:91–6. 39 Morgan R, Ellul J, Denton E, et al. Malignant esophageal fistulas and perforations: management with plastic-covered metallic endoprostheses. Radiology 1997; 204:527–32. ◆40 Hatrick AG, Sabharwal T, Adam A. Tracheobronchial stents: a review. Semin Intervent Radiol 2001; 18(3):243–50. 41 Lu DSK, Papanicolaou N, Girard M, et al. Percutaneous internal ureteral stent placement: review of technical issues and solutions in 50 consecutive cases. Clin Radiol 1994; 49:256–61. ●42 Adam A, Chetty N, Roddie M, et al. Self-expandable stainless steel endoprostheses for treatment of malignant bile duct obstruction. AJR 1991; 156:321–5. 43 Davids PHP, Groen AK, Rauws EAJ, et al. Randomized trial of self-expanding metal stents versus polyethylene stents for distal malignant biliary obstruction. Lancet 1992; 340:1488–92.
●44
45
46
47
◆48
49 ●50
51 52 53 ◆54
55
56
◆57
58
59
60
◆61
62
Irving JD, Dondelinger RF, Reidy JF, et al. Gianturco selfexpanding stents: clinical experience in the vena cava and large veins. Cardiovasc Intervent Radiol 1992; 15:351–5. Gibson RN, Hennessy OF, Collier N, Hemingway AP. Major complications of central venous catheterization. A report of five cases and brief review of the literature. Clin Radiol 1985; 36:204–8. Rossi P. Percutaneous removal of intravascular foreign bodies. In: Wilkins RA, Viamonte M (eds) Interventional Radiology. Oxford: Blackwell Scientific Publications, 1982, 359–69. Belli AM, Hemingway AP. Retrieval of intravascular foreign bodies. In: Belli AM (ed.) Interventional Radiology in the Peripheral Vascular System. London: Edward Arnold, 1994, 81–92. Balch CM, Levin B. Regional and systemic chemotherapy for colorectal metastases to the liver. World J Surg 1987; 11:521–6. Chuang VP, Wallace S. Arterial infusion and occlusion in cancer patients. Semin Roentgenol 1987; 16:13–25. Robertson LJ, Mauro MA, Jacques PF. Radiologic placement of Hickman catheters. Radiology 1989; 170:1007–9. Adam A. Insertion of long-term central venous catheter: time for a new look. Br Med J 1995; 311:341–2. Hemingway AP. Materials for embolization. Radiol Now 1986; 3:63–4. Chuang VP, Wallace S. Hepatic artery embolization in the treatment of hepatic neoplasms. Radiology 1987; 140:51–8. Hemingway AP, Allison DJ. Complications of embolization: analysis of 410 procedures. Radiology 1988; 166:669–72. Dick R, Adam A, Allison DJ. Interventional techniques in the hepatobiliary system. In: Grainger R, Allison DJ (eds) Diagnostic Radiology: an Anglo-American Textbook of Organ Imaging, 3rd edn. Edinburgh: Churchill-Livingstone, 1997, 1235–58. Kato L, Nemeto R, Mori H, et al. Arterial chemoembolization with microencapsulated anticancer drug. J Am Med Assoc 1981; 245:1123–7. Sabharwal T, Gangi A. Percutaneous vertebroplasty. CME Radiol 2004; 4(2):71–5. Prinssen M, Buskens E, Blankensteijn JD. Quality of life after open and endovascular AAA repair: results of a randomised trial. Eur J Vasc Endovasc Surg 2004; 27:121–7. Bell RE, Taylor PR, Aukett M, Sabharwal T, Reidy JF. Midterm results for second-generation thoracic stent grafts. Br J Surg 2003; 90:811–17. Morales JP, Irani FG, Jones KG, Taylor PR, Dourado R, Sabharwal T. Endovascular repair of a ruptured abdominal aortic aneurysm under local anaesthesia. BJR 2005; 78:1–3. Veith FJ, Amor M, Ohki T, et al. Current status of carotid bifurcation angioplasty and stenting based on a consensus of opinion leaders. J Vasc Surg 2001; 33:S111–16. Voss SD, Kruskal JB. Gene therapy: a primer for radiologists. Radiographics 1998; 18:1343–72.
8 Vaccination strategies for malignant diseases DEEPAK P. ASSUDANI, STEPHANIE MCARDLE, MURRIUM AHMAD, GENG LI, ROBERT C. REES AND SELMAN A. ALI
Introduction Tumour antigens Tolerance Tumour vaccines
167 168 171 171
INTRODUCTION Unlike conventional vaccines to infectious diseases, which stimulate the immune system to respond specifically to nonself, dangerous, ‘foreign’ microbial antigens, the challenge of developing effective cancer vaccines lies in the ability of the immune system to recognize and respond to self or ‘altered self’, essentially non-dangerous antigens. Apart from virally induced tumours, most tumours express antigens that fit more with a self/altered-self paradigm than with the non-self paradigm antigens of pathogens.1 To answer the most frequently asked question about whether immunity to cancer exists at all, one needs to analyse the occurrence of spontaneous regression in certain cancers and the existence of correlation between tumour prognosis and patient survival with intra-tumoral inflammatory cell infiltrations. The immunosurveillance theory, as postulated by Burnet,2 is based on the ability of the immune system effector cells, recently identified as T cells, actively to assist the body to identify and eradicate incipient tumour cells. The development of inbred strains of mice was the cornerstone in proving this theory, since it was possible for the first time to test experimentally the idea that tumours were immunologically distinguishable from normal cells. Inbred mice could be immunized and caused to reject chemically or virally induced syngeneic tumours, and this could not have happened without the existence of tumour-specific antigens. Boon and Kellermann3 demonstrated for the first time that tumour cells with new mutations induced by in-vivo treatment with a mutagen were unable to produce progressive tumours in syngeneic mice, not because of intrinsic growth defects but
Tumour escape Conclusion Acknowledgements References
178 179 180 180
changes in their antigenicity and ability to trigger an immune rejection response. The cancer immunosurveillance theory was not flawless and has clearly generated more controversial debate than any other immunological concept due to a number of factors, including the failure of immunodeficient mice (nude mice) to develop significantly more cancers than their control counterparts. It has also been postulated that under certain circumstances the immune system can aid cancer progression by selecting tumour cells with lower immunogenicity.4 However, further evidence supporting cancer immunosurveillance has recently emerged, demonstrating an increased cancer incidence in mice lacking specific cellular populations such as T cells, natural killer (NK) cells and immunostimulant molecules such as interferon-gamma (IFNγ) and interleukin-2 (IL-2) This contradiction has finally been resolved by the postulation of a broader immune process termed cancer immunoediting, in which cancer immunosurveillance represents only one of three dimensions of a complex process of interaction between the immune system and cancer, namely (a) immunosurveillance or elimination, (b) equilibrium, and (c) escape.4,5 Although the presence of immunosurveillance, as a tumour suppressor and protective mechanism against neoplasia in immunocompetent patients, is now unquestionable, failure of this process will inevitably lead to an equilibrium phase in which the immune system is unable completely to destroy all tumour cells and finally to the third phase in which tumours escape growth restraint imposed by the immune system. Cancer immunotherapy represents a promising and relatively safe approach compared to chemotherapy and
168 Vaccination strategies for malignant diseases
radiotherapy, which may be associated with severe toxicity and lack of specificity and in some cases leads to shortduration remission. Treatment involving surgery, chemotherapy and/or radiotherapy together with immunotherapy represents a more rational approach to patient management. The active immunotherapy hypothesis is based on stimulation of the immune system to recognize and destroy tumour cells. This approach to treatment can be divided into two broad areas: non-specific and antigenspecific therapies. The non-specific therapies refer to stimulation or enhancement of the immunity by substances that non-specifically activate the immune system without having pre-defined antigen specificity. This type of therapy was pioneered by a New York physician, William Coley, who more than 100 years ago used a mixture of killed bacteria, ‘Coley’s toxin’, to treat cancer patients. Coley based his novel approach on observations that tumour regression coincided with the development of streptococcal infection within ulcerated tumours. Spontaneous regression of human tumours that coincides with various types of infection is a recognized phenomenon that has been reported in the literature for the past few hundred years ago.6 Nowadays, the majority of non-specific immunostimulants, such as cytokines and other biological modifier molecules, are recognized for their supporting roles in the enhancement of interaction and communication of immune cells. These findings have now been supported by extensive research into the application of viral and bacterial vectors to deliver tumour antigens and adjuvants such as cytokines. There is compelling evidence to suggest that microbial antigens bind to ‘toll-like’ receptors expressed on dendritic cells, leading to the activation and production of T helper 1 (Th1) cytokines, including IFNγ and IL-2.7 Recruitment and local activation of dendritic and NK cells have also been associated with early phases of viral infection; both cells are capable of exchanging bidirectional activating signals that are important for the development of antigen-specific memory.8,9 In contrast, antigen-specific therapies can be accomplished by either adoptive transfer or vaccination. It is now well established that CD8 cytotoxic T lymphocytes (CTLs) can recognize and kill tumour cells displaying peptides from tumour-associated antigens presented by major histocompatibility complex (MHC) class I molecules, and optimally require participation of antigen-specific CD4 helper T cells (Fig. 8.1). The isolation, expansion and re-infusion of tumour-specific T cells derived from tumour biopsies and expanded in vitro have been shown to be feasible, and clinical responses have been demonstrated.10 To induce specific immune responses, vaccination using crude, unidentified specific tumour antigens has been attempted using a number of approaches, including the use of wholecell vaccines, cancer-cell lysates, culture-cell supernatants and delivery systems for tumour antigens, peptides and genetic sequences encoding the antigen specificities.11,12 The success in identifying a wide range of tumour antigens by a variety of techniques, including molecular cloning
Figure 8.1 Cancer and immune responses. Tumour antigens released by dying tumour cells (or administered as vaccines) are taken up by tissue-resident dendritic cells (DCs), which then migrate to the lymph nodes and present these processed antigens to CD4 T cells. Depending on the activation signals, this can lead these CD4 T cells to differentiate into Th1 (Helper T cells 1) or Th2 response. Cross-talk between dendritic cells and CD4 T cells conditions the dendritic cells to present the antigens to CD8; T cells (cytotoxic T lymphocytes, CTLs) and activate them. The CTLs migrate to the tumour cells and mediate their killing. The Th1 cells provide help for the maintenance of CTL and memory response. The Th2 cells help in the generation of the humoral (antibody) response, and these antibodies can also mediate tumour killing via antibody-dependent cellular cytotoxicity. (IFN-γ, interferon-γ, TNF-β, tumour necrosis factor-β, IL-4/5, interleukin-4/5.)
of CTL-defined tumour antigens, has led to the current clinical trials of cancer vaccines. Some of these vaccines have resulted in partial or complete tumour regression in a proportion of treated patients, and have increased the disease-free survival rate in others (see below). These outcomes are more evident in patients suffering from certain cancers, such as melanoma and renal cancer. This chapter attempts to cover aspects of recent advances in the field of tumour immunology, with emphasis on tumour antigens, various types of cancer vaccines in use in the clinic and their limitations due to immune escape of cancer cells.
TUMOUR ANTIGENS Tumour cells differ from their surrounding tissue by the expression of proteins that are either unique to them or over-expressed by them. The identification of tumour antigens capable of inducing antigen-specific anti-tumour immunity represents a challenging task confronting tumour immunologists. Since the cloning of MAGE-1,13 the first gene reported to encode a tumour antigen recognized by T cells, molecular identification and characterization of novel tumour antigens have rapidly evolved.
Tumour antigens 169
The identification of tumour antigens that elicit an immune response in the tumour-bearing host is a prerequisite for the development of immunotherapeutic approaches in cancer. Knowledge of their molecular nature provides us not only with potential targets for immunotherapeutic interventions against tumour cells, but also with new disease markers and new insights into the molecular mechanisms of malignant transformation. Various techniques have been developed over the past 15 years for the identification of these tumour antigens and these have been recently reviewed.14,15 According to their expression pattern, antigens expressed by human tumours can be loosely classified as: ●
●
●
●
onco-foetal antigens (typically only expressed in foetal tissues and in cancerous somatic cells), onco-viral antigens (encoded by tumorigenic transforming viruses), over-expressed/accumulated antigens (expressed by both normal and neoplastic tissue, with the level of expression highly elevated in neoplasia), cancer–testis antigens (expressed only by cancer cells and adult reproductive tissues such as testis and placenta),
●
●
●
●
tissue-specific differentiation antigens (expressed largely by a single cancer histotype), tumour-specific unique antigens (only expressed by cancer as a result of genetic mutation or alteration in transcription), post-translationally altered antigens (tumour-associated alterations in glycosylation etc.), idiotypic antigens (highly polymorphic genes , a specific ‘clonotype’, of which is expressed by a tumour cell i.e. as in B-cell, T-cell lymphoma/leukaemia resulting from clonal aberrancies).
Examples of tumour-associated antigens that fall into each of these categories are provided in Table 8.1. However, these categories are not mutually exclusive, and tumour antigens may fall into more than one. For example, the p53 tumour-suppressor gene is frequently mutated in cancer cells, resulting in the accumulation of p53 protein in these cells and reduced cell-cycle regulatory control by the tumour cells. Based on these parameters, p53 would be classified as both an over-expressed/accumulated tumour antigen and a mutated tumour antigen. In a similar manner, tyrosinase represents a normal melanocytic protein that can be both over-expressed and altered in its
Table 8.1 Classification of tumour antigens Group
Tumour antigens
Cancer expressing the antigen
Cancer–testis
MAGE 1–3 and 12, BAGE, GAGE, HAGE, NY-ESO-1 Head/neck, bladder, gastric and lung
Melanoma, breast
Differentiation antigens
Tyrosinase, gp-100, TRP-1, TRP-2, MART-1
Melanoma
Over-expressed/mutated antigens
HER-2/neu MUC-1 p53 (mutated) p53 (non-mutated) Ras (mutated) WT-1 Proteinase-3 PAP, PSA, PSMA
Breast, ovary, lung Breast, adenocarcinoma Colorectal, lung, bladder, head/neck Breast, colon, other cancers Pancreatic, colon, lung CML, ALL, AML CML Prostate
Viral antigens
EBV HCV HPV
Burkitt’s lymphoma Hepatocellular carcinoma Cervical and penile cancer
Onco–fetal antigens
CEA Alpha-fetoprotein 5T4 Onco-trophoblast glycoprotein
Colon, breast, pancreatic cancer Liver cancer Many carcinomas Many carcinomas
Tumour-specific antigens
Ig idiotype CDK4 Caspase-8 beta-catenin BCR/ABL
B-cell NHL, MM Melanoma Head/neck Melanoma CML
NHL, non-Hodgkin’s lymphoma; MM, multiple myeloma; CML, chronic myeloid leukaemia; ALL, acute lymphoblastic leukaemia; AML, acute myeloid leukaemia; CEA, carcinoembryonic antigen; HPV, human papillomavirus; EBV, Epstein–Barr virus; HCV, hepatitis C virus.
170 Vaccination strategies for malignant diseases
post-translational modification, leading to differential recognition of melanoma cells versus normal melanocytes by specific T lymphocytes.
Onco-foetal antigens The onco-foetal antigens are ‘self’ proteins normally expressed during foetal development and then undetectable or at low levels in normal healthy adult tissues. On the other hand, these antigens have been detected in the sera of patients with gynaecological cancer. Carcinoembryonic antigen (CEA) was one of the first-known tumour markers; many more have now been described, but CEA remains, alone or in combination with others, the one most used. It is not organ specific, and abnormal values may be found in a wide range of carcinomas.16 One of the most useful applications of this marker is as a post-surgical prognostic indicator in the treatment of neoplasms. Any elevation of CEA after the conventional treatment of neoplasms has been correlated with a recurrence of cancer.17
Onco-viral antigens Some human malignancies are associated with viral infections, e.g. Burkitt’s lymphoma and Epstein–Barr virus (EBV), hepatocellular carcinoma and hepatitis B and C virus (HBV and HCV), cervical and anal carcinoma and human papillomavirus (HPV), T-cell leukaemia and human T-lymphotropic virus (HTLV). Virally encoded tumourassociated molecules in the malignant cells offer exogenous cancer vaccine targets for which there is unlikely to be immunological tolerance at the immune repertoire level. T cells can recognize the strong antigens expressed by virally infected cells and it is appropriate to consider viral vaccines for prophylactic and therapeutic immunization against virus-associated human cancers, as in the case of HPV and cervical cancer and HBV-associated liver cancer.18
Over-expressed antigens/mutated antigens Many genetic alterations accumulate in a normal cell before it becomes fully malignant, thereby increasing its genetic instability, which will render the cell susceptible to more genetic changes. These alterations include point mutations, gene amplification, translocation and viral insertion. As a result, many genes or gene products end up being over-expressed due to an increase copy number of the gene, an increase in the stability of its protein, or a defect in the degradation pathway of its product. These antigens are therefore also expressed in many normal tissues, but at much lower levels. The epitopes processed and potentially presented by normal tissues are below the threshold level of T-cell recognition, whereas their over-expression in tumour cells may trigger an anti-tumour immune response by breaking previously established tolerance.
Cancer–testis antigens These antigens are only expressed by normal tissue in the adult testis and placenta and are termed cancer–testis antigens (CTAs). However, during carcinogenesis, these antigens are re-expressed due to a process called demethylation. The functions of most of the cancer–testis antigens are still unknown. They represent attractive targets for immunotherapy due to their widespread expression in cancer cells but localized expression in immune privileged sites such as testis and placenta. Indeed, cancer–testis antigens have been some of the most extensively investigated antigens in animal models and in clinical trials. MAGE was the first antigen to be discovered in this group, following which a large number of others such as HAGE and GAGE as well as NY–ESO 1 antigen which have been defined using SEREX.19 These antigens are frequently found in a wide range of tumour types such as melanoma, lung carcinoma, head and neck tumour, and bladder carcinoma. Recently, we have applied SEREX to identify novel CTAs from a cDNA expression library constructed from normal human testes that was screened with pooled prostate cancer patients’ sera leading to identification of several known and unknown antigens. cDNA were obtained from human testes and used to express proteins in bacteria with each clone expressing one protein. These are then screened using the antibodies obtained from cancer patients to identify novel antigens to which these patients have generated immune response, since normally testis is an immunopriviledged site and proteins expressed in them do not generate an immune response.
Differentiation antigens These antigens are generally shared between cancer cells and their normal counterparts, and are specific for a particular lineage of cells or a point of differentiation. They are also known as melanoma-associated antigens, as they are mostly found in melanomas and normal melanocytes. Some of these antigens are involved in the biosynthesis of the pigment melanin. The most well known antigens from this group are tyrosinase, TRP-1, gp100 and MART-1. Since these are expressed at low levels in normal tissues, targeting them involves the risk of autoimmune reactions such as vitiligo.
Tumour-specific unique antigens These represent the most attractive immunotherapeutic targets, as they are unique to each individual tumour or shared between specific histological tumour types. Bcr/abl is the most well known antigen in this group, which is formed due to translocation between chromosomes 9 and 22, leading to chronic myeloid leukaemia (CML). Other unique antigens are generated due to point mutations or gene fusions in some of the essential oncogenes or tumour-suppressor
Tumour vaccines 171
genes such as k-ras, β-catenin, CDK4 and p53. Some groups consider these to be the best option for therapy, as they represent the most immunogenic epitopes; however, their identification in each individual patient raises questions regarding their practical application in the wider population. Tailor-made individualized therapies are being considered seriously as an option by certain immunologists, and this group of antigens would be their priority target.
Post-translationally altered antigens Altered glycosylation appears to be a constant phenomenon associated with oncogenic transformation in all types of naturally occurring human cancers. Most of the biochemical or, more recently, immunological methods used to identify tumour-associated antigens have resulted in the isolation of glycolipids or glycoproteins (mucins) with altered glycosylation patterns. Mucins are large (200 kDa) glycoproteins with a high carbohydrate content. They are expressed by a variety of normal and malignant epithelial cells.
Idiotypic antigens B-cell lymphoma arises as a clonal event. Specific immunoglobulins are generated by a unique combination of gene segments during B-cell differentiation, and these are expressed throughout the development of the tumour. The immunoglobulin idiotypic determinants come from unique heavy (VH) and light (VL) chains and provide a unique tumour-associated antigen to target for immunotherapy against lymphoma.
T cells reach the thymus and are presented with the ‘self’ antigens by the residing antigen-presenting cell (APC), leading to either negative or positive selection. T cells having strong affinity to ‘self’ antigens are deleted, as these are potential autoimmune T cells (negative selection). On the other hand, T cells binding to MHC–peptide complex with medium affinity are positively selected to survive and form the peripheral T-cell repertoire. However, the central tolerance mechanisms are not perfect, and sometimes potentially autoreactive T cells are released into the periphery.27 These T cells are kept in check by the peripheral tolerance mechanisms, mainly under the control of regulatory T cells (Tregs). Tregs are perhaps the most important suppressive cells, and control the activation of potentially autoreactive T cells. This suggests that an effective T-cell response against cancerous cells over-expressing normal antigens will require deletion of these Tregs along with antigen-specific vaccination.
TUMOUR VACCINES Research in the field of vaccination for cancer has generated enormous interest since the identification of tumour antigens. Most of the tumour antigens are ‘self’ antigens, hence it is essential to break tolerance against them while avoiding autoimmunity. Moreover, most cancer patients are immunocompromised at later stages of the disease, which makes it difficult to mount an effective immune response against tumour antigens. Successful immunotherapy of cancer will also have to overcome the multiple tumour escape mechanisms employed by the cancer during its evolution. Attempts to overcome these challenges are illustrated by the large number of vaccine strategies that have thus far been developed and tested (Fig. 8.2).
TOLERANCE The development of cancer/malignant disease is often seen as a result of a failure in immune surveillance. Moreover, recent evidence suggests that mechanisms of tolerance that usually exist to prevent autoimmune diseases may also prevent the development of an anti-tumour response against antigens such that even tumours have the ability to preclude a response against their antigens. Immunological tolerance is the failure to mount an immune response to an antigen. This can be either natural/ ‘self’ tolerance or induced tolerance. It is now becoming apparent that there are several ‘bottlenecks’ in the immune response to cancer cells; these include: (a) the lack of highavidity tumour-specific T cells,20–22 (b) the inefficient priming of tumour-specific T cells,22–24 (c) the physical and functional deletion of primed tumour-specific T cells,24 and (d) tumour evasion and counterattack.25,26 T-cell tolerance can be either central or peripheral. Central tolerance occurs in the developing thymus. Developing
Figure 8.2 Types of cancer vaccines. Cancer vaccines can be patient-non-specific, such as allogeneic whole-cell, peptide, DNA/RNA and viral/bacterial vaccines. Personalized vaccines include autologous tumour cell or tumour-derived heat-shock-protein vaccine and dendritic cell vaccine. T cells isolated from peripheral blood mononuclear cells (PBMCs) or tumour-infiltrating lymphocytes can be expanded in vivo and injected as a form of therapy.
172 Vaccination strategies for malignant diseases
Whole-cell vaccines Whole-cell vaccines consist of irradiated autologous or allogeneic tumour cells, with or without any in-vitro modifications, and many clinical trials have explored this method of cancer vaccination. Earlier clinical trials using autologous or allogeneic tumour cells without any modification met with limited success. Galligioni et al.28** immunized 120 renal carcinoma patients with autologous irradiated tumour cells along with bacillus Calmette–Guérin (BCG) as an adjuvant. Although the vaccination was able to generate a tumour-cell-specific immune response as measured by delayed-type cutaneous hypersensitivity (DTCH), it had little effect on patient survival. Subsequently, various approaches have been used in which the tumour cells have been modified in vitro either to express co-stimulatory molecules or to secrete cytokines to enhance their immunogenicity.11,29* Granulocyte–macrophage colony-stimulating factor (GM-CSF) coding gene is the most widely used for the transfection of tumour cells leading to the recruitment and generation of dendritic cells to the site of vaccination. In a phase I clinical trial of prostate cancer patients vaccinated with irradiated autologous prostate tumour cells engineered to secrete GM-CSF, most patients generated prostate-antigen-specific T-cell and B-cell responses.29* Similarly, in a recent clinical trial of patients diagnosed with metastatic non-small-cell lung carcinoma, the patients were vaccinated with irradiated autologous tumour cells engineered to secrete GM-CSF. Immune responses were detected in most of the patients receiving the vaccine, but few clinical responses were recorded.30* In metastatic melanoma patients immunized with autologous melanoma cells modified to secrete IL-12 cytokine, thus favouring the generation of a Th1 response and activation of the innate and adaptive immune system,31*,32 two of the seven patients developed DTH reaction to the autologous tumour cells, whereas one patient had a minor clinical response and generated tumour-specific CD4 and CD8 T-cell infiltrates into the metastases.31* Targeting individual tumour antigens by vaccination limits the number of T-cell clones activated and does not take into account the variable expression pattern in different tumours. This problem could potentially be overcome by using whole tumour cells or their lysates for vaccination: they potentially contain multiple antigens, many of which may be individually specific to the tumour. Thus, although autologous tumour-cell vaccination has generated a tumour-specific immune response in most cases, clinical success has been extremely limited. This approach is restricted by the inability to generate large numbers of autologous tumour cells and is clinically impractical for larger clinical trials. Most of the recent clinical trials have been conducted using modified allogeneic tumour cell lines, as they can be genetically modified and cultured to large numbers prior to injection. Moreover, cross-presentation of antigen to T cells by the APCs does not require the tumour cells to be
MHC matched with the recipient. One concern regarding cultured cell lines in the long term relates to changes of their antigenic profile, immunogenicity and gene expression; thus quality control and consistency of the vaccine are central issues.
Heat-shock-protein-based vaccines Heat-shock proteins (HSPs) are a large family of ubiquitous and inducible proteins involved in the transport of intracellular peptides and which assist in the folding and unfolding of proteins, along with a host of other functions.33 Their role in cancer vaccination was discovered when it was observed that HSP from tumour cells but not from normal cells could be used to immunize animals against autologous tumours, although HSPs prepared from both normal and tumour cells are identical. It was later realized that this immunogenicity of the HSPs was due to bound peptides derived from tumour-associated/ tumour-specific antigens. The ability of tumour-derived HSP to generate anti-tumour responses has been demonstrated for gp96, hsp70, hsp90, hsp110, grp170 and calreticulin.34 Intracellular proteins are processed via the proteosomes in the cell; the peptides thus generated are transported to the endoplasmic reticulum by HSPs, where they bind to newly synthesized MHC molecules. Thus, immunizing with purified HSPs from autologous tumour cells can potentially generate an immune response against non-covalently bound antigenic peptides, and essentially bypasses the need to identify individual peptide epitopes presented by the cancer cells. Moreover, HSPs seem to have an intrinsic adjuvant property and can bind to the receptors on APC, resulting in their maturation and internalization followed by internal processing and presentation of HSP-associated peptides on MHC class I and II molecules.35 This property of HSPs could be extremely important, as not only are they the first and only ‘natural’ adjuvants of mammalian origin, but also they fail to elicit an immune response on their own. Moreover, a very low level of bound peptide is needed to generate a response compared to conventional peptide immunization.36 Immunizing with HSP can generate CD8 T-cell and antibody responses as well as the expansion of the NK cells,37 and they have been nicknamed the ‘Swiss Army knives’ of the immune system due to their ability to elicit diverse responses.38 In cancer patients, antigen-specific tolerance is maintained in the periphery, as T cells are not provided with co-stimulatory or cytokine signals (‘signal 2’) along with antigen presentation. Hence, it has been hypothesized that HSPs might be able to overcome tolerance to tumour cells by virtue of their ability to activate APC and generate ‘signal 2’.39 Several non-randomized clinical trials have been carried out in the last few years with the HSP–peptide complexes. In most of these trials, following surgical excision of
Tumour vaccines 173
the primary tumour, HSP–peptide complexes were purified from the tumour and the autologous complexes were injected back into the patient. The clinical response is based on the regression of metastasis and/or stabilization of tumour progression. In one such trial, out of 16 renalcell carcinoma patients immunized with 25 μg of autologous tumour-derived hsp gp96, one of the several heat shock proteins involved in intracellular protein transport, three patients demonstrated complete tumour regression, three had partial responses and three showed prolonged stabilization of the disease.40 Other clinical trials involving cancer patients have also generated some encouraging results, although large randomized controlled trials are needed to establish fully their potential.34 In a recently concluded phase III clinical trial of 300 stage IV melanoma patients immunized with autologous tumour-derived hsp (gp96)–peptide complex vaccines, increased survival period was observed in vaccinated patients compared to control patients. Although only a minority of the patients had M1a AJCC staging system for melanoma category tumour: skin and lymph-node disease only, and normal serum lactate dehydrogenase level, the medial survival period was 626 days, as compared to 383 days in the control group. Moreover, patients receiving ten or more immunizations survived for significantly longer period than those in the control arm of the trial.41
Peptide vaccines As explained previously, CD4 and CD8 T cells are considered to be the main effector cells against cancer, and functional activity by both these cells requires binding of their T-cell receptor (TCR) to the peptides presented to them by the MHC class II and class I molecules respectively. Thus, any peptide presented by the MHC molecule on the surface of tumour cells is a potential target for cancer immunotherapy. In 1992, the first study to identify a nonamer peptide recognized by cytotoxic T lymphocytes in cancer patient was published.42 Since then, several studies have identified immunogenic peptides presented by the tumour cells in vivo and recognized by the cytotoxic T lymphocytes in cancer patients.43 CD8 T cells recognizing peptides derived from tumour antigens have been shown to exist in most cancer patients, and can be activated by vaccination.44 The advantage of the peptide vaccines is that they are easily produced. Moreover, there may be differences in the processing of proteins by the tumour cells and antigenpresenting cells, which may result in the generation of different epitopes. In such instances, immunization with peptides known to be naturally processed by the tumour cells is the desired option. Several studies in animal models have shown the efficacy of peptide-specific CD8 T cells in providing protection against tumour challenge or completely eradicating established tumours.45 A number of clinical trials have
been conducted using peptide vaccines. Initial trials involved immunizing patients with MHC class I restricted peptides alone, and in some instances peptide-specific CTLs were generated, but this immunological response failed to deliver any objective clinical response.46 However, most of the other clinical trials have not produced encouraging results. In two recent independent clinical trials of stage IV melanoma and CML patients, immunizing with peptide vaccines containing epitopes from melanoma antigens (gp100, MART-1, tyrosinase and MAGE-3) and bcr-abl protein respectively produced no objective clinical response in terms of tumour regression or remission induction. However, in both these trials, some of the patients had stable disease that correlated with the generation of vaccine-induced cytotoxic response.47*,48* It has been suggested that low disease burden or combination with other therapy modalities could be more effective. Supporting this, another clinical trial showed that peptide vaccination can reduce the number of recurrences in patients who have a history of at least three resected metastases. Of 44 patients with resected cutaneous melanoma, 9 patients with more than three resected metastases were immunized with melanoma-specific peptide vaccines. Of these 9 patients, 4 had cessation of recurrences and all 4 made an immunologic response to the vaccine.49* Another school of researchers seems to favour targeting mutated antigens that might be specific to individual tumours. In one such clinical trial, patients were screened for specific mutations in p53 and K-ras antigens and then immunized with peptides containing those mutations. Although only a fraction of patients experienced a period of stable disease, median survival time for immunological responders was significantly longer than that for nonresponders.50* These results suggest that although peptide vaccines have had very little clinical benefit in terms of tumour regression, in most cases clinical outcome correlates with immunological response. Experience from animal models showed that for an effective and long-lasting response leading to tumour regression, it was essential to activate CD4 T-helper cells. Not only is T-cell help required for optimal effector and memory CTL response,11 but in certain animal models CD4 T cells have been shown to mediate tumour regression via an IFNγ-mediated mechanism.51 This has led to the intensification of efforts to identify MHC class II restricted peptides.52,53 Recent trials have used a combination of MHC class I and II peptides, but the available evidence is limited and further investigations are required. There are several possible reasons for the limited clinical efficacy of peptide vaccines. The source of the helper peptides for CD4 T-cell stimulation is an important consideration. Several groups include MHC class II epitopes from irrelevant antigens such as tetanus toxoid and hepatitis B together with CTL epitopes in peptide vaccines. The CD4 T-cell repertoire is not likely to be tolerised against these foreign antigens.54 However, there is evidence to
174 Vaccination strategies for malignant diseases
suggest that immunizing with MHC class II epitopes from antigens expressed by the tumour cells together with MHC class I epitopes is more likely to induce the regression of established tumours. Also, the adjuvant used to administer the peptide vaccines is extremely important and more likely to recruit APCs to the site of the peptide depot and thus be more effective in generating peptidespecific T cells. Currently, the adjuvants permitted for human administration are aluminium-based salts (Alum) and a squalene-oil-water emulsion (MF59).55 There is a need to identify novel and potent adjuvants, especially for peptide vaccines. Peptide vaccines delivered through liposomes or nanobeads or in combination with CpGoligonucleotides have all shown an ability to generate high-avidity CTLs.56,57 Another important factor influencing the effectiveness of vaccines is the ‘quality and quantity’ of the T-cell response. Animal models predict that not only is the number of T cells important, but also their functional status. GM-CSF and IL-2 promote the differentiation of APCs and proliferation of T cells respectively, and are instrumental in generating effector T cells.58* One of the obstacles to the success of cancer immunotherapy is the development of tolerance to ‘self’ antigens. It has been shown that during development, most of the T cells having high affinity for ‘self’ antigens have been deleted in the thymus and only T cells with low to medium binding affinity survive; their activation represents a significant challenge. However, it has been shown, both in human studies and in animal models, that altering certain amino-acid residues of the peptide may increase its binding affinity to MHC molecules and/or to the TCR. This can significantly improve the ‘quality and quantity’ of the T cells, which are able to bind to both the altered and naïve peptide and lyse tumour cells.59,60 Confirming the studies in animal models, Rosenberg et al. showed that immunizing advanced (metastatic) melanoma patients with a modified peptide vaccine and IL-2 resulted in the generation of immunologic responses in 91 per cent of patients, with 42 per cent demonstrating an objective clinical response.59* It was observed that 13 out of 31 patients immunized with modified peptide emulsified in IFA (Incomplete Freund’s Adjuvant) and concomitant IL-2 had objective clinical regressions of the metastases in brain, lung, liver, lymph nodes, skin and subcutaneous tissues. Moreover, in spite of generating an immune response against a differential antigen that is normally expressed in normal skin cells, no autoimmune response were observed in any of the patients.59* Finally, immunizing with epitopes derived from a single tumour antigen usually results in down-regulation of antigen or MHC molecule expression.61 This could be overcome by immunization against multiple epitopes.62 However, the phenomenon of epitope dominance, i.e. immune response being diverted towards one peptide, represents a possible limitation in using multi-epitope vaccines.63
Dendritic-cell vaccines Dendritic cells were first identified some 25 years ago as cells with intertwining dendritic processes that seem to be involved in antigen capture in primary lymphoid follicles.64,65 Their role in generating an immune response then became apparent,66 but it is thanks to methods developed by Steinman67 enabling the isolation and identification of dendritic cells that their extraordinary and unique ability to activate and maintain the survival of T lymphocytes was established. Nowadays it is well recognized that dendritic cells represent the most powerful antigenpresenting cell type, capable of stimulating strong immune responses after trafficking from the periphery, where they take up antigens, to lymph nodes, where they cross-present antigens to T cells through their high number of human leukocyte antigen (HLA) and co-stimulatory molecules, as well as producing important immunomodulatory molecules such as cytokines and chemokines. The origin of the cells, their maturation status and their combination with antigen in the form of peptides, protein, lysate or RNA/DNA, as well as the dose and site of injection are important considerations for optimizing vaccine protocols. Indeed, successful vaccine will have to generate an effective, long-lasting antitumour response. The choice of the antigen will direct the specificity of the immune response, and the choice of the adjuvant will affect the quality and magnitude of the anti-tumour response. Dendritic cells can be generated from various precursor cells after expansion of their progenitors or be directly isolated from various body locations. Their phenotype and function may be altered, depending on the precursor chosen, factors and agents used to mobilize, differentiate and mature them, as well as on the anatomic location from which the dendritic cells were taken. CD34 cells represent the main precursor cells and the main sources of dendritic cells, and can be isolated from blood, bone marrow, umbilical cord, lymph node, tonsil, spleen, skin or even tumours. Patients undergoing chemotherapy may receive cytokine such as granulocyte-colony-stimulating factor (G-CSF) prior to treatment in order to mobilize peripheral blood CD34 cells, which can then be positively selected and used to generate dendritic cells in vivo. More dendritic cells can be generated by this method than from other materials, and many clinical trials have adopted these cells as a source of dendritic cells.68*,69*,70* However, monocyte-derived dendritic cells remain the most widely used type in clinical trials.71 The majority of human chronic and acute myelogenous leukaemia cells can be induced to differentiate into dendritic cells under the influence of a combination of cytokines such as GM-CSF, IL-4, and/or IL-6 and tumour necrosis factor-alpha (TNFα), as estimated by the level of expression of dendritic-cell-specific markers and function.72,73 This offers two major advantages: that the cells can be grown and manipulated in vitro to achieve a maximum number but also that they act as a source of tumour antigens.
Tumour vaccines 175
Co-culture of T cells with autologous dendritic cells generated from acute myeloid leukaemia (AML) or CML patients can generate effector cells capable of lysing autologous CML targets with greater efficacy than T cells expanded with IL-2 alone.72,73 Alternatively, exosomes, which are small membrane vesicles containing both major histocompatibility complex (MHC) class-I and MHC class-II molecules secreted by dendritic cells, can be used as potent anti-tumour molecules. However, these have been shown to be efficacious only when co-injected with dendritic cells,74,75 which, following antigen up-take, migrate to the T-cell-enriched areas located in lymphoid organs.76 In order to elicit immunity in cancer patients, dendritic cells have to present the relevant tumour antigens, many of which are newly expressed, uniquely expressed or overexpressed by cancer cells (see Table 8.1). The best tumour antigen would ideally induce a broad repertoire of antigenspecific T-helper and cytotoxic T lymphocytes. The most commonly used sources of antigen in clinical trials are tumour lysates, necrotic or apoptotic tumour cells, peptides and tumour-derived DNA or RNA. Tumour lysates present two main advantages: HLA typing of the cells is not required and the antigen specificity does not need to be defined. Nestle et al. pioneered this work and immunized 16 melanoma patients with advanced disease with immature monocyte-derived dendritic cells pulsed with tumour lysate or a cocktail of peptides known to be recognized by CTLs (such as MART-1, tyrosinase, gp100 or MAGE-1 or -3), depending on the patient’s HLA haplotype. Objective responses were evident in 5 out of the 16 evaluated patients (two complete responses, three partial responses), with regression of metastases in various organs (skin, soft tissue, lung, pancreas) and one additional minor response.77* However, tumour cells will also carry ‘self’ antigen that may induce unwanted immune responses, which would divert the immune response away from cancer-associated antigens. Sufficient tumour material is also required for the preparation of the lysate, and monitoring of cancer-specific responses to undefined antigens is problematic. Some groups have fused dendritic cells directly with tumour cells; the resultant cells thus express both known and unknown tumour antigens as well as co-stimulatory and adhesion molecules necessary for the activation of T cells.78 Patients with metastatic renalcell carcinoma have been vaccinated with a hybrid cell vaccine consisting of autologous tumour cells fused to dendritic cells and, despite the poor prognosis for these patients, objective clinical responses, including four complete remissions, were seen in 7 of 17 (41 per cent) patients.79* Dendritic cells transfected with tumour-derived mRNA offer the advantage of safety over DNA, as the RNA is highly unlikely to integrate into the patient’s genome. Moreover, multiple antigens can be immunized against using this approach without any MHC restriction of the patient.
A phase I clinical trial in metastatic prostate cancer patients demonstrated the feasibility and safety of this technique.80* In this study all 13 patients receiving autologous monocyte-derived dendritic cells transfected with autologous tumour-derived mRNA demonstrated prostatespecific antigen (PSA)-specific T-cell responses without toxicity associated with significant decrease of the serum PSA in six of seven patients. No autoimmune toxicity was observed in these patients and they generated PSA specific immune response as tested in vitro. Also, serum PSA decreased in some patients. Some of these patients also showed a rapid but transient clearance of circulating tumour cells from peripheral blood. Malignancies of B-cell origin express monoclonal immunoglobulins that carry unique tumour-specific antigenic determinants in the variable region, idiotypes. These idiotypes have been isolated from B-cell malignancies and subsequently loaded onto autologous dendritic cells and used for the immunization of low-grade follicular nonHodgkin’s lymphoma (NHL). Durable tumour regression in patients with measurable disease was observed, as well as the induction of anti-idiotype T-cell and antibody response which could specifically recognize tumour cells and mediate objective clinical response.81* Combinations of selected peptides with carefully prepared dendritic cells have been successfully used to eradicate tumour in mice, and with some success against human tumours,77* the major limiting factors being that these peptides are specific for the HLA type of the patient. However, a phenomenon known as ‘peptide spreading’ can occur. This occurs after a strong immune response is generated against a particular peptide, resulting in the killing of tumour cells and leading to new antigens being released as a result of the cell destruction, which can then be taken up and processed by dendritic cells. Thus these dendritic cells present to T cells different tumourspecific peptides that were not included in the initial vaccine.82 Peptide-pulsed dendritic cells also allow the careful monitoring of the immune response in vivo. Microspheres may also be used to load dendritic cells with selected peptides and to prolong their presentation of peptides to the immune system.83 Alternatively, transfection of dendritic cells with tumour RNA or DNA using viruses such as vaccinia viruses and lentiviruses allows a broader range of peptide specificities to be expressed.84,85 Animal studies have shown that the route of immunization with dendritic cells is of prime importance, and that delivery to lymphoid organs is preferable to immunization via non-lymphoid organs such as the blood or skin. Indeed, the route of administration has been shown to determine memory cell distribution and the ability to control tumour outgrowths located at different body sites.86 The preparation of dendritic cells ex vivo to clinical grade standard is labour intensive and remains a major
176 Vaccination strategies for malignant diseases
challenge. Moreover, ex-vivo-derived dendritic cells may not offer advantages over peptide immunization.87 Although dendritic cells can be injected directly into the tumour or lymph nodes,88 techniques developed to attract and mature dendritic cells at the site of immunization or specifically target dendritic cells in vivo may be more effective in generating a strong CTL response.89
DNA vaccines DNA vaccines consist of an antigen-encoding gene on a bacterial plasmid backbone. The expression of the gene encoded by this plasmid is controlled by a strong viral promoter such as cytomegalovirus (CMV) immediate–early promoter/enhancer. They are cheap and simple to produce, and the chances of DNA integrating into the human genome are extremely low. Compared to peptide vaccines, which would be restricted to specific MHC haplotypes, DNA vaccines are unrestricted and can generate CTLs against multiple epitopes, being able to generate B-cell as well as T-cell immune responses.90* Unlike viral vaccines, DNA vaccines are not influenced by the neutralizing antibodies against the viral vector.91 DNA vaccines can be administered in a number of ways, such as intramuscularly, intranasally and intramucosally, or can be coated onto gold particles and administered into the dermis using a gene gun.92 These plasmids persist in the cell nuclei as circular, non-replicating episomes, resulting in long-term expression of the gene. Apart from when administered by gene gun immunization, the injected plasmids are taken up by somatic cells such as myocytes; these will then express the antigen encoded by the injected gene, process it and present the resulting peptides onto their surface in association with MHC class I molecules.93 The APCs play an important role in the initiation of the immune response by DNA vaccination. Interestingly, the administration of apoptosis-inducing agents along with the DNA vaccination enhances antigen acquisition and T-cell activation by the APCs. In contrast, during gene gun immunization, the plasmid DNA is directly taken up by the resident Langerhans cells in the dermis.94 In both cases, these APCs then travel to the lymph nodes and present the antigens to the T and B cells. The route of immunization with DNA vaccines seems to be important, as several studies in animal models have shown that intramuscular immunization preferentially generates a Th1 response, whereas immunizing with the gene gun generates a Th2 response, although this method requires 100 times less antigen.95 Also, DNA vaccination can be used in conjunction with other vaccination methods, i.e. transfection of dendritic cells with plasmids encoding for tumour antigen or co-stimulatory/cytokine genes prior to injecting them. An important consideration in vaccinating with tumourassociated antigen is the activation of APCs, which is intrinsic to DNA-based vaccines as they contain unmethylated CpG
(cytosine linked to guanine with a phosphate) motifs.96 The motifs are 20-fold more frequent in microbial genes and therefore the mammalian immune system (mainly the APCs) is able to recognize them through their patternrecognition receptors. As additional adjuvants, plasmids encoding cytokine genes, co-stimulatory molecules or chemokines have been shown to increase the efficacy of the vaccination.97 Recently, it has been shown that along with tumour-antigen-encoding plasmids, co-administration of plasmids encoding for dendritic cell chemotactic and growth factors enhances the antigen-specific response in mouse models. DNA vaccines encoding multiple class I and class II peptide epitopes have also been designed for human immunodeficiency virus (HIV) and malaria, targeting either the MHC class I or class II processing pathway by the addition of specific signalling sequences.98 Several clinical trials have been conducted in recent years with plasmid vaccines, and unfortunately none of the human trials has been able to match the remarkable success in the animal models. None of the 22 patients immunized with a plasmid vaccine encoding gp100 antigen was able to mount any immunological and clinical response.99 In another clinical trial, patients with stage IV melanomas were injected intranodally with the plasmid DNA vaccine; 11 out of 26 patients were able to generate immune response. Although none of the patients had any clinical regression, it was observed that more than 50 per cent had unexpectedly long survival at a median follow-up of 12 months.100* Several other phase I trials have been conducted in melanoma and prostate cancer patients, but none has reported any significant objective clinical responses, despite some patients being able to generate humoral and cellular responses.101–103 DNA vaccines by themselves are considered weakly antigenic; however, when used in combination with viral vectors in a ‘prime-boost’ strategy, they appear to be highly effective in generating antigen-specific responses. Priming with plasmid vaccine followed by boosting with recombinant virus was shown to be highly effective compared to either of them alone.104 Recently, xenogeneic immunization, i.e. immunizing with a plasmid encoding homologous antigen from another species, has emerged as a powerful method to break tolerance to several ‘self’ antigens.105 However, the efficacy of these approaches remains to be tested in randomized human clinical trials.
Recombinant viral and bacterial vaccines The requirement for potent immunogenic vectors for the delivery of tumour antigens has led researchers to investigate naturally-occurring microbes as appropriate vehicles. Retrovirus, poxvirus, alphavirus, adenovirus and herpes simplex viruses have been investigated in several animal models and shown to be effective in promoting anti-tumour immunity. Viral vectors can be injected systemically or
Tumour vaccines 177
intra-tumorally and used to modify tumour cells ex vivo prior to injection. Many viral vectors are disabled and safe to administer and ideally should infect dividing as well as non-dividing cells and have the capacity to accept large inserts of genetic information. Retroviruses have been used for direct gene delivery to tumour cells as well as for the modification of tumour cells ex vivo prior to use. Out of nine patients injected with a retrovirus encoding for the p53 antigen, three exhibited regression of their lesions and three had disease stabilization. However, the virus could integrate into the host genome and hence pose a biosafety hazard,106 and most viruses currently in use have been attenuated through genetic approaches to make them safe for clinical application. Modified vaccinia Ankara strain is one of the most promising vectors being investigated, as it has proven to be highly effective for gene delivery and safe during smallpox eradication.107 In a phase I/II clinical trial of women with cervical intraepithelial neoplasia, 34 of 36 women injected with modified vaccinia virus showed complete elimination of pre-cancerous lesions and responses correlated with generation of specific cytotoxic response.108** Thus viral vaccination might be more advantageous compared to conventional cryosurgery, as vaccination in addition to treating pre-existing lesions would also generate an immune response against future infections by human papillomavirus. Other viruses belonging to the poxvirus family, such as canarypox and fowlpox viruses, have proven their efficacy in animal models. In a recent clinical trial, vaccination of patients with canarypox virus encoding for melanoma antigen gp100 was found to be superior to peptide vaccination, as 44 per cent of patients generated antigen-specific CTL response compared to none of the peptide-immunized patients.109** In another clinical trial in advanced melanoma cases, 30 patients were immunized with recombinant canarypox virus encoding for MAGE peptides. Only one patient demonstrated partial response and two had stable disease; the response in all patients correlated significantly with antigen-specific CTL generation.110 Adenoviruses have also been shown to be extremely potent in the generation of immune response in animals; however, their efficacy in humans would be limited, as about 85 per cent of the world’s population has antibodies against common serotypes. In one of the early clinical trials, recombinant adenovirus expressing melanoma antigen MART-1 was administered to 16 patients with metastatic melanoma, and one patient experienced a complete response; however, the efficacy was possibly limited due to the high levels of neutralizing antibody generated against the virus.91 Recombinant alphaviruses are potentially very safe vectors that could be used, as they need to be co-transfected with a helper plasmid for the generation of infective viral particles. Moreover, they induce apoptosis upon infection, which would facilitate cross-priming of the antigens.
We have extensively investigated the immunotherapeutic potential of disabled infectious single cycle-herpes simplex virus (DISC-HSV) in a murine tumour model.111 Upon intra-tumour inoculation of DISC-HSV encoding for murine GM-CSF, the regression of established tumours was observed in up to 70 per cent of the mice. The induced tumour regression was concomitant with potent CTL activity that was MHC-I restricted and directed against peptides of known tumour antigens. This vector is very safe, as the essential glycoprotein H (gH) gene has been deleted and it is therefore only capable of a single round of infection.112–114 Another recent clinical trial has also demonstrated potency of viral vaccines by inducing lesion regression of high-grade cervical intra-epithelial neoplasia.115* Recombinant bacterial vaccines from strains such as Salmonella, BCG and Listeria monocytogenes are promising novel vectors for cancer immunotherapy as they can potentially be administered orally and are capable of infecting APCs directly. The dual phagolysosomal and cytoplasmic life cycle of Listeria monocytogenes allows efficient processing of both MHC class II and class I antigens respectively.116,117
Novel vaccine delivery systems Several novel vaccination strategies able to deliver tumour antigens to the APCs, which would lead to appropriate antigen processing and would be ethically safe for human administration have been investigated. Liposomes, consisting of lipid membrane and encapsulating various drug formulations, have been used for a number of years to treat Kaposi’s sarcoma, and studies are now being extended to other cancers.118 Liposomes have the ability to deliver ‘drugs’ directly to the cytosol of the cells by fusion with the cell membrane and release of the drug internally. This property has been exploited to deliver antigens to APCs, and liposomes containing tumour antigens have also been shown to generate antigen-specific responses.119 Liposomal vaccination has been shown to be clinically safe and effective in generating long-lasting CD4 and CD8 T-cell responses in animal models and in patients with advanced stage follicular lymphoma. Antigens, either coupled to beads made of iron, silica or latex, or trapped in poly-lactide-co-glycolide (PLG) microspheres, have been shown to generate CTLs and in some cases are able to provide tumour protection in animal models.120 Recently, peptides coupled to nanoparticles have also been shown to generate peptide-specific CTL responses and provide protection against tumour challenge.57 The administration of viruses encoding tumour antigens has always proved controversial, especially when given to patients with late-stage cancer who are usually immunocompromised. To overcome this concern without losing the therapeutic effect of the viruses, virus-like particles
178 Vaccination strategies for malignant diseases
(VLPs) have been designed that consist of self-assembled proteins derived from the viruses. These VLPs have been shown to prime antigen-specific responses without the need for adjuvant; however, their efficacy in humans has yet to be shown. Exosomes have recently emerged as an alternative novel method of immunization. Exosomes are small membrane vesicles that are released from various cell types during the fusion of multi-vesicular bodies with the plasma membrane121 and that carry both MHC class I and class II antigens along with co-stimulatory molecules. These exosomes can be purified from various tumour cells and loaded with peptide ex vivo before injecting them into the patients. Exosome immunization has been shown to generate immune response leading to tumour protection and rejection in animal models121 and current clinical studies will establish their utility in cancer therapy.
Cellular adoptive therapy Most cancer vaccination strategies, although proven to be highly successful in animal models, have only generated sporadic responses in clinical trials. It has been suggested that the major limiting factor in these trials is the inability to generate sufficient numbers of efficient tumour-specific T cells. This limitation has been overcome by recent clinical trials conducted using autologous T cells generated from the patients and expanded in vivo prior to infusion. Most of the tumours contain infiltrating lymphocytes that are reactive against mutated or non-mutated antigens expressed by the tumour, which can be isolated from the surgically removed tumours, expanded and activated in vitro and injected back into patients.122 Initial clinical trials with CD8 T cells isolated from autologous tumourinfiltrating lymphocytes (TILs) have provided disappointing results, possibly due to lack of the helper CD4 T cells and/or suppression by regulatory T cells.123 Regression of the tumours would therefore rely on the elimination of such suppressive T cells and the provision of helper cells for the generation of long-lasting immunity.124 Indeed, Rosenberg et al. have recently shown that 18 of 35 treated patients receiving myeloablative chemotherapy in the form of cyclophosphamide and fludarabine, followed by adoptive T-cell transfer of autologous TILs, generated objective clinical responses (50 per cent tumour reduction).10,125* Regression of metastasis in these patients was observed in sites such as lung, brain, liver, lymph nodes and subcutaneous tissues.125* In the past, adoptive cellular therapy was limited by the inability to generate sufficient numbers of tumour-specific T cells upon in-vivo stimulation. This has led to the construction of ‘artificial’ APCs, which are able to present the antigens to the T cells in vivo along with the co-stimulatory molecules and to promote T-cell expansion. An alternative novel strategy has been to clone tumour-antigen-specific T-cell receptors into autologous T cells prior to injecting
these engineered T cells back into patients. Long-lasting regression and/or remission of the tumours in patients resulting from adoptive cellular therapy would be largely dependent on the generation of memory CTLs, which in turn is dependent on the local production of cytokines such as IL-2. However, injecting IL-2 into cancer patients has been associated with several adverse effects. Hence, TILs engineered to contain the IL-2 gene are being developed, the local secretion of which would promote their own growth.10 This obviates the need for exogenous administration of IL-2 and reduces side effects in the patients.
TUMOUR ESCAPE Tumour escape is one major obstacle that has to be addressed prior to designing and delivering successful immunotherapy. There is evidence to suggest that immunogenic tumours in both pre-clinical murine models and cancer patients can be rejected by the immune system under optimum conditions. However, despite this, a large number of tumours continue to grow and evade the immune system, and the limited success of current immunotherapeutic strategies may be due to a number of reasons:126 ●
●
● ● ●
failure of effector cells to compete with the growing tumour burden; the production of humoral factors and immunosuppressive cytokines by tumours that locally block immune attack; antigen/MHC-loss tumour variants; T-cell dysfunction; the production of suppressor T cells.
When considering novel immunotherapeutic approaches for cancer, one has to consider simultaneously the correction of immune activating signals, the eradication of inhibitory factors and the evasion of newly developed immuno-resistant tumour phenotypes.
Activation versus suppression in tumour progression Tumours can grow undetected by the immune system, being seen as ‘normal tissue’ as they do not exhibit the danger signals for immune activation.127,128 It is also believed that even during progressive growth, the tumour has the ability to activate the immune system and that a fine balance between activation and suppression exists, which may determine the fate of the tumour. Tumour vaccines may induce activation and expansion of specific CD8 T cells and destruction of tumour cells in cancer patients; this has been observed in approximately 5–20 per cent of vaccinated melanoma patients. However, this activation can be diminished by the lack of appropriate
Conclusion 179
co-stimulation or the presence of immunosuppressive cytokines such as IL-10 and TGF-β. The eventual fate of the tumour is therefore decided by the net effect of immune activation and inhibition. The mechanisms of tumour escape from immune recognition/destruction are likely to be multi-factorial.
epitope.135 These mutations severely diminish T-cell recognition of the tumour antigen by a variety of mechanisms, including modulation of MHC–peptide interaction and TCR binding to MHC–peptide complex.
Immunosuppressive cytokines HLA class I down-regulation/loss Altered MHC class I antigen expression in tumours is a well-known phenomenon. It has been described in a variety of mouse models, including the T-cell leukaemic cell line, methylcholanthrene (MCA)-induced tumours and murine leukaemia-virus-induced tumours. The changes in MHC class I expression allow the tumours to avoid recognition or attack by CD8 T cells – a widespread phenomenon that has been confirmed by immunohistological studies.129 HLA class I molecule down-regulation occurs frequently in many cancers, and this abnormality might adversely affect the clinical course of disease and hence the outcome of T-cell-based immunotherapies. Changes in HLA expression can occur via a number of different mechanisms – genetic mutations, abnormalities in regulation and/or defects in HLA class-Idependent antigen processing. The various types of HLA loss have been classified into the different phenotypes that are recognized.130
Tumour antigen down-regulation, mutation or loss An alteration of tumour-associated antigen (TAA) expression is one mechanism by which tumour cells can escape CTL detection. Changes in TAA expression can range from a simple down-regulation to a complete loss and this can occur independently of the deregulation of HLA class I expression. Tumour-antigen expression is known to be heterogeneous, even within the same tumour,34 and a decrease in the expression of gp100 and MART-1 was associated with disease progression.131 The emergence of stable antigenloss variants has also been shown to be responsible for tumour escape in the murine mastocytoma P815 model.132 Antigenic drift, a mechanism used by viruses to escape immune recognition, has recently been described for tumours. Transgenic mice expressing TCR (T cell receptor) for a single antigenic epitope have been used extensively in establishing antigenic mutation(s) as a mechanism for viral escape of T-cell recognition.133,134 A transgenic mouse line expressing TCR specific for tumour antigen P1A35-43 presented by H2Ld was also developed and used to study tumour escape mechanisms. The recurrence of tumours in mice that have responded favourably to transgenic T-cell adoptive therapy was found to correlate with the presence of tumour variants with mutations within the P1A
Tumour cells produce various cytokines and chemokines that can have a suppressive effect on immune cells. In patients with non-small-cell lung cancer, the mRNA expression of IL-4, IL-10, TGF-α and TGF-β was significantly higher than that of IL-2, IL-12, IL-18 and IFN-γ as determined in pleural effusion and tumour tissue.136 Vascular endothelial growth factor (VEGF) is secreted by a number of tumours137 and is known to be a contributory factor in tumour escape. It is important for tumour vascularization and is also produced by solid tumours to inhibit immune recognition.138,139 It also prevents dendritic-cell differentiation and maturation by suppressing the transcription factor NF-κB (nuclear factor-kappa B) in haemopoietic stem cells.140 Elevated VEGF in blood was correlated with poor prognosis in human neoplasms,137 possibly as a result of its angiogenic properties and/or its ability to suppress dendritic-cell maturation. Interleukin-10 is a cytokine that is often said to be ‘suppressive’ in nature, and high levels of IL-10 have been detected in patients with melanoma141 and pancreatic cancer.25,142 It has the ability to exert its effects in many ways: it inhibits antigen presentation, IL-12 production and the induction of a Th1 response in vivo. With respect to TGF-β, high concentrations of this cytokine are found in cancer patients25,142 and are usually also associated with ‘tumour progression’.
CONCLUSION Using various vaccine-based strategies such as antigen pulsed dendritic cells or targeting dendritic cells along with appropriately delivered tumour antigens, clinical responses have been observed in patients with various types of tumour. However, these results have been rather disappointing and have failed to meet expectations. Although animal models have provided invaluable information regarding the efficacy of vaccines against tumours in preclinical studies, one has to remain cautious because most studies have used rodent tumours, the majority of which are chemically induced and grow much faster than those encountered in humans. Many human tumours evolve slowly, taking sometimes years to develop malignant potential, and undergo a series of genetic and cellular changes leading to tumour progression and avoidance of the host defence mechanisms. Also, murine tumours are usually implanted at a site other than the tissue of origin, thereby not matching the organ-specific physiological characteristics
180 Vaccination strategies for malignant diseases
of the tumour. Therefore, the immune system of patients is slowly acclimatized to tumours, whereas the immune system of animals transplanted with tumour cell lines is abruptly exposed to the cancer and its antigenic composition. Moreover, animal tumours sometimes acquire endogenous viruses and express viral antigens, rendering the tumour more immunogenic; this situation may not occur in human cancers. Cancer patients often die as a result of metastasis, which is where therapeutic vaccination should primarily focus. However, more appropriate animal models are being developed for studying and optimizing therapeutic cancer vaccines. An additional reason for the lack of progress in immunotherapy and low response rates observed in human clinical trials is that most patients involved in these trials are in stage III–IV of the disease – at a time when the immune system is already compromised. For vaccines to be effective, they will have to be used when minimal residual disease is present in order to achieve a favourable effector-to-target ratio. The mechanisms responsible for immune tolerance versus activation are being dissected, and the results of studies have provided a better understanding of host–tumour interactions. Collaboration between innate and adaptive immune responses in order to mount a strong and long-lasting immune response against tumours is considered important. Moreover, with the emergence of more powerful technologies, such as RNA amplification, proteomic technologies and DNA micro-arrays, events within the tumour microenvironment following vaccination can be studied. The identification of new biomarkers might allow the diagnosis of cancer at earlier stages of progression and at a time when immunotherapy may be of significant benefit to patients. In conclusion, immunotherapy represents an attractive modality for treating cancer but is limited by our understanding of the biological complexity of the disease and the patient. The past decade has witnessed remarkable progression in our understanding of the immune recognition of tumours, escape mechanisms and strategies for vaccination and cellular therapy. It is necessary to consider immunotherapy as an adjunct to conventional treatments and not necessarily as separate. The areas of concern relate to the selection of ‘resistant’ cells during treatment and whether combined approaches can eradicate these variants. In order to overcome tolerance and tumour escape mechanisms without side effects of autoimmunity, it will be necessary to select the most appropriate immunotherapy combined with traditional treatments. Moreover, we are faced with the challenge of understanding complex biological mechanisms, which may be cancer-type specific, and determining their relevance to ‘treatment outcome’. Finally, we do not as yet have a clear understanding of the important ‘self-renewing’ cancer stem cells and their relative susceptibility or resistance to treatments. These cells are difficult to detect and cultivate and very little information is available concerning their ‘immune phenotype’.
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
Most human tumours are immunogenic and immunosurveillance plays an important role in the clearance of many possible tumours. The immune system also helps in shaping the phenotype of cancerous cells. Hundreds of tumour antigens that can potentially be targets for the immune system have been identified. Most tumour antigens are ‘self’ antigens, hence tolerance to these antigens limits the efficacy of cancer vaccination. Various forms of vaccination, such as whole-cell, peptide, DNA, viral/bacterial vaccines, have generated limited clinical benefits in clinical trials in spite of clearly evident immunological responses. Recently, adoptive transfer of in-vitro-expanded autologous tumour-infiltrating lymphocytes has generated significant clinical responses in patients, although their widespread application is debatable. Immune escape mechanisms also limit the efficacy of tumour vaccines. Delineating the escape mechanisms and circumventing them are likely to be the key factors for successful immunotherapy. Recent progress in proteomic and genomic technologies will allow us to identify early cancer-specific markers that will enable early treatment to be given to patients and disease progression and recurrence to be monitored, and will aid decision-making concerning treatment.
ACKNOWLEDGEMENTS The authors wish to acknowledge the support of the John and Lucille van Geest Foundation, the European Commission (specifically funding received from ALLOSTEM, ENACT AND OISTER) and the National Eye Research Council for supporting research into cancer vaccines in this laboratory.
REFERENCES 1 Houghton A. Cancer antigens: immune recognition of self and altered self. J Exp Med 2004; 180:1–4. 2 Burnet F. Cancer – a biological approach. Br Med J 1957; 1:841–7. ●3 Boon T, Kellermann O. Rejection by syngeneic mice of cell variants obtained by mutagenesis of a malignant
References 181
◆4
5
6
7
8
9
●10
11
12 ●13
14
15 16
17
18
19
20 ●21
teratocarcinoma cell line. Proc Natl Acad Sci U S A 1977; 74:272–5. Dunn GP, Old LJ, Schreiber RD.The immunobiology of cancer immunosurveillance and immunoediting. Immunity 2004; 21:137–48. Ikeda H, Chamoto K, Tsuji T, et al. The critical role of type-1 innate and acquired immunity in tumor immunotherapy. Cancer Sci 2004; 95:697–703. Hoption Cann SA, van Netten JP, van Netten C, et al. Spontaneous regression: a hidden treasure buried in time. Med Hypotheses 2002; 58:115–19. Hochrein H, Schlatter B, O’Keeffe M, et al. Herpes simplex virus type-1 induces IFN-alpha production via Toll-like receptor 9-dependent and -independent pathways. Proc Natl Acad Sci U S A 2004; 101:11416–21. Gerosa F, Baldani-Guerra B, Nisii C, et al. Reciprocal activating interaction between natural killer cells and dendritic cells. J Exp Med 2002: 195:327–33. Biron CA. Activation and function of natural killer cell responses during viral infections. Curr Opin Immunol 1997; 9:24–34. Rosenberg SA, Dudley ME. Cancer regression in patients with metastatic melanoma after the transfer of autologous antitumor lymphocytes. Proc Natl Acad Sci U S A 2004; 101:14639–45. Ali SA, McLean CS, Boursnell ME, et al. Preclinical evaluation of ‘whole’ cell vaccines for prophylaxis and therapy using a disabled infectious single cycle-herpes simplex virus vector to transduce cytokine genes. Cancer Res 2000; 60:1663–70. Hockertz S. Present and future of cancer vaccines. Toxicology 2005; 214:151–61. Van der Bruggen P, Traversari C, Chomez P, et al. A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma. Science 1991; 254:1643–7. Miles AK, Matharoo-Ball B, Li G, et al. The identification of human tumour antigens: current status and future developments. Cancer Immunol Immunother 2006; 12:1–8 Li G, Ali SA, McArdle SE, et al. Immunity to tumour antigens. Curr Pharm Des 2005; 11:3501–9. Ballesta AM, Molina R, Filella X, et al. Carcinoembryonic antigen in staging and follow up of patients with solid tumours. Tumour Biol 1995; 16:32–41. Holubec L Jr, Topolcan O, Pikner R, et al. The significance of CEA, CA19-9 and CA72-4 in the detection of colorectal carcinoma recurrence. Anticancer Res 2000; 20:5237–44. Mahdavi A, Monk BJ. Vaccines against human papillomavirus and cervical cancer: promises and challenges. Oncologist 2005; 10:528–38. Simpson AJ, Caballero OL, Jungbluth A, et al. Cancer/testis antigens, gametogenesis and cancer. Nat Rev Cancer 2005; 5:615–25. Blattman JN, Greenberg PD. Cancer immunotherapy: a treatment of the masses. Science 2004; 305:200–05. Gallegos AM, Bevan MJ. Central tolerance to tissue-specific antigens mediated by direct and indirect antigen presentation. J Exp Med 2004; 200:1039–49.
●22
23 24
25
26
◆27
28
29
30
31
32
33
34
35
●36
Gronski MA, Boulter JM, Moskophidis D, et al. TCR affinity and negative regulation limit autoimmunity. Nat Med 2004; 10:1234–9. Mathis D, Benoist C. Back to central tolerance. Immunity 2004; 20:509–16. Overwijk WW, Theoret MR, Finkelstein SE, et al. Tumour regression and autoimmunity after reversal of a functionally tolerant state of self reactive CD8 T cells. J Exp Med 2003; 198:569–80. Rivoltini L, Carrabba M, Huber V, et al. Immunity to cancer: attack and escape in T lymphocyte–tumor cell interaction. Immunol Rev 2002; 188:97–113. Marincola FM, Wang E, Herlyn M, et al. Tumours as elusive targets of T cell based active immunotherapy. Trends Immunol 2003; 24:335–42. Anderton SM, Wraith DC. Selection and fine-tuning of the autoimmune T-cell repertoire. Nat Rev Immunol 2002; 2:487–98. Galligioni E, Quaia M, Merlo A, et al. Adjuvant immunotherapy treatment of renal carcinoma patients with autologous tumor cells and bacillus Calmette–Guerin: fiveyear results of a prospective randomized study. Cancer 1996; 77:2560–6. Simons JW, Mikhak B, Chang JF, et al. Induction of immunity to prostate cancer antigens: results of a clinical trial of vaccination with irradiated autologous prostate tumor cells engineered to secrete granulocyte–macrophage colony-stimulating factor using ex vivo gene transfer. Cancer Res 1999; 59:5160–8. Salgia R, Lynch T, Skarin A, et al. Vaccination with irradiated autologous tumor cells engineered to secrete granulocyte–macrophage colony-stimulating factor augments antitumor immunity in some patients with metastatic non-small-cell lung carcinoma. J Clin Oncol 2003; 21:624–30. Sun Y, Jurgovsky K, Moller P, et al. Vaccination with IL-12 gene-modified autologous melanoma cells: preclinical results and a first clinical phase I study. Gene Ther 1998; 5:481–90. Trinchieri G. Interleukin-12 and the regulation of innate resistance and adaptive immunity. Nat Rev Immunol 2003; 3:133–46. Srivastava PK, Amato RJ. Heat shock proteins: the ‘Swiss Army Knife’ vaccines against cancers and infectious agents. Vaccine 2001; 19:2590–7. Binder RJ, Vatner R, Srivastava P. The heat-shock protein receptors: some answers and more questions. Tissue Antigens 2004; 64:442–51. Singh-Jasuja H, Toes RE, Spee P, et al. Cross-presentation of glycoprotein 96-associated antigens on major histocompatibility complex class I molecules requires receptor-mediated endocytosis. J Exp Med 2000; 191:1965–74. Nieland TJ, Tan MC, Monne-van Muijen M, et al. Isolation of an immunodominant viral peptide that is endogenously bound to the stress protein GP96/GRP94. Proc Natl Acad Sci U S A 1996; 93:6135–9.
182 Vaccination strategies for malignant diseases
37 Tamura Y, Peng P, Liu K, et al. Immunotherapy of tumors with autologous tumor-derived heat shock protein preparations. Science 1997; 278:117–20. ◆38 Schild H, Rammensee HG. gp96– the immune system’s Swiss army knife. Nat Immunol 2000; 1:100–1. 39 Liu B, DeFilippo AM, Li Z. Overcoming immune tolerance to cancer by heat shock protein vaccines. Mol Cancer Ther 2002; 1:1147–51. 40 Amata RJ, Murray L, Wood LA, et al. Active specific immunotherapy in patients with renal cell carcinoma using autologous tumour derived heat shock protein–peptide complex-96 (HSPPC-96) vaccine. ASCO Abstract 1999. 41 Srivastava PK. Therapeutic cancer vaccines. Curr Opin Immunol 2006; 18:201–5. ●42 Traversari C, van der Bruggen P, Luescher IF, et al. A nonapeptide encoded by human gene MAGE-1 is recognized on HLA-A1 by cytolytic T lymphocytes directed against tumor antigen MZ2-E. J Exp Med 1992; 176:1453–7. ●43 Romero P, Cerottini JC, Speiser DE. Monitoring tumor antigen specific T-cell responses in cancer patients and phase I clinical trials of peptide-based vaccination. Cancer Immunol Immunother 2004; 53:249–55. ●44 Pittet MJ, Valmori D, Dunbar PR, et al. High frequencies of naive Melan-A/MART-1-specific CD8() T cells in a large proportion of human histocompatibility leukocyte antigen (HLA)-A2 individuals. J Exp Med 1999; 190:705–15. 45 Chen YF, Lin CW, Tsao YP, Chen SL. Cytotoxic-T-lymphocyte human papillomavirus type 16 E5 peptide with CpGoligodeoxynucleotide can eliminate tumor growth in C57BL/6 mice. J Virol 2004; 78:1333–43. 46 Lee KH, Wang E, Nielsen MB, et al. Increased vaccinespecific T cell frequency after peptide-based vaccination correlates with increased susceptibility to in vivo stimulation but does not lead to tumor regression. J Immunol 1999; 163:6292–300. 47 Cathcart K, Pinilla-Ibarz J, Korontsvit T, et al. A multivalent bcr-abl fusion peptide vaccination trial in patients with chronic myeloid leukemia. Blood 2004; 103:1037–42. 48 Hersey P, Menzies SW, Coventry B, et al. Phase I/II study of immunotherapy with T-cell peptide epitopes in patients with stage IV melanoma. Cancer Immunol Immunother 2005; 54:208–18. 49 Letsch A, Keilholz U, Fluck M, et al. Peptide vaccination after repeated resection of metastases can induce a prolonged relapse-free interval in melanoma patients. Int J Cancer 2005; 114:936–41. 50 Carbone DP, Ciernik IF, Kelley MJ, et al. Immunization with mutant p53- and K-ras-derived peptides in cancer patients: immune response and clinical outcome. J Clin Oncol 2005; 23:5099–107. ●51 Egilmez NK, Hess SD, Chen FA, et al. Human CD4 effector T cells mediate indirect interleukin-12 and interferongamma-dependent suppression of autologous HLA-negative lung tumor xenografts in severe combined immunodeficient mice. Cancer Res 2002; 62:2611–17.
52 Rojas JM, McArdle SE, Horton RB, et al. Peptide immunisation of HLA-DR-transgenic mice permits the identification of a novel HLA-DRbeta1*0101 and HLADRbeta1*0401-restricted epitope from p53. Cancer Immunol Immunother 2005; 54:243–53. 53 Zarour HM, Kirkwood JM, Kierstead LS, et al. MelanA/MART-1(51-73) represents an immunogenic HLA-DR4restricted epitope recognized by melanoma-reactive CD4() T cells. Proc Natl Acad Sci U S A 2000; 97:400–5. 54 Slingluff CL Jr, Yamshchikov G, Neese P, et al. Phase I trial of a melanoma vaccine with gp100(280–288) peptide and tetanus helper peptide in adjuvant: immunologic and clinical outcomes. Clin Cancer Res 2001; 7:3012–24. 55 Mesa C, Fernandez LE. Challenges facing adjuvants for cancer immunotherapy. Immunol Cell Biol 2004; 82:644–50. 56 Engler OB, Schwendener RA, Daia WJ, et al. A liposomal peptide vaccine inducing CD8 T cells in HLA-A2.1 transgenic mice, which recognise human cells encoding hepatitis C virus (HCV) proteins. Vaccine 2004; 23:58–68. 57 Fifis T, Mottram P, Bogdanoska V, et al. Short peptide sequences containing MHC class I and/or class II epitopes linked to nano-beads induce strong immunity and inhibition of growth of antigen-specific tumour challenge in mice. Vaccine 2004; 23:258–66. 58 Jager E, Ringhoffer M, Dienes HP, et al. Granulocyte–macrophage-colony-stimulating factor enhances immune responses to melanoma-associated peptides in vivo. Int J Cancer 1996; 67:54–62. ◆59 Rosenberg SA, Yang JC, Schwartzentruber DJ, et al. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma. Nat Med 1998; 4:321–27. 60 Bredenbeck A, Losch FO, Sharav T, et al. Identification of noncanonical melanoma-associated T cell epitopes for cancer immunotherapy. J Immunol 2005; 174:6716–24. ◆61 Khong HT, Restifo NP. Natural selection of tumor variants in the generation of ‘tumor escape’ phenotypes. Nat Immunol 2002; 3:999–1005. 62 Valmori D, Dutoit V, Ayyoub M, et al. Simultaneous CD8 T cell responses to multiple tumor antigen epitopes in a multipeptide melanoma vaccine. Cancer Immun 2003; 3:15. 63 Palmowski MJ, Choi EM, Hermans IF, et al. Competition between CTL narrows the immune response induced by prime-boost vaccination protocols. J Immunol 2002; 168:4391–98. 64 Nossal GJV, Abbot A, Mitchell J, Lummus Z. Antigens in immunity, IX: ultrastructural features of antigen capture in primary and secondary lymphoid follicles. J Exp Med 1968; 127:277–90. 65 Nossal GJV, Ada GL, Austin CM. Antigens in immunity, IV: cellular localization of 125l- and 1311-labelled flagella in lymph nodes. Aust J Exp Biol 1964; 42:311–33. 66 Tew JG, Thorbecke GJ, Steinman RM. Dendritic cells in the immune response: characteristics and recommended nomenclature. J Reticuloendothel Soc 1982; 31:371–80.
References 183
◆67
68
69
70
71
72
73
●74
75
76
77
78
●79
80
81
82
Steinman RM. The dendritic cell and its role in immunogenicity. Annu Rev Immunol 1991; 9:271–96. Banchereau J, Palucka AK, Dhodapkar M, et al. Immune and clinical responses in patients with metastatic melanoma to CD34() progenitor-derived dendritic cell vaccine. Cancer Res 2001; 61:6451–8. Palucka AK, Dhodapkar MV, Paczesny S, et al. Single injection of CD34 progenitor-derived dendritic cell vaccine can lead to induction of T-cell immunity in patients with stage IV melanoma. J Immunother 2003; 26:432–9. Paczesny S, Banchereau J, Wittkowski KM, et al. Expansion of melanoma-specific cytolytic CD8 T cell precursors in patients with metastatic melanoma vaccinated with CD34 progenitor-derived dendritic cells. J Exp Med 2004; 199:1503–11. Dietz AB, Padley DJ, Butler GW, et al. Clinical-grade manufacturing of DC from CD14 precursors: experience from phase I clinical trials in CML and malignant melanoma. Cytotherapy 2004; 6:563–70. Choudhury A, Toubert A, Sutaria S, et al. Human leukemia-derived dendritic cells: ex-vivo development of specific antileukemic cytotoxicity. Crit Rev Immunol 1998; 18:121–31. Choudhury A, Gajewski JL, Liang JC, et al. Use of leukemic dendritic cells for the generation of antileukemic cellular cytotoxicity against Philadelphia chromosome-positive chronic myelogenous leukemia. Blood 1997; 89:1133–42. Zitvogel L, Regnault A, Lozier A, et al. Eradication of established murine tumors using a novel cell-free vaccine: dendritic cell-derived exosomes. Nat Med 1998; 4:594–600. Zitvogel L, Couderc B, Mayordomo JI, et al. L-12-engineered dendritic cells serve as effective tumor vaccine adjuvants in vivo. Ann NY Acad Sci 1996; 795:284–93. Bell D, Chomarat P, Broyles D, et al. In breast carcinoma tissue, immature dendritic cells reside within the tumor, whereas mature dendritic cells are located in peritumoral areas. J Exp Med 1999; 190;1417–26. Nestle FO, Alijagic S, Gilliet M, et al. Vaccination of melanoma patients with peptide- or tumor lysate-pulsed dendritic cells. Nat Med 1998; 4:328–32. Koido S, Hara E, Torii A, et al. Induction of antigen-specific CD4- and CD8-mediated T-cell responses by fusions of autologous dendritic cells and metastatic colorectal cancer cells. Int J Cancer 2005; 117:587–95. Kugler A, Stuhler G, Walden P, et al. Regression of human metastatic renal cell carcinoma after vaccination with tumor cell–dendritic cell hybrids. Nat Med 2000; 6:332–6. Heiser A, Coleman D, Dannull J, et al. Autologous dendritic cells transfected with prostate-specific antigen RNA stimulate CTL responses against metastatic prostate tumors. J Clin Invest 2002; 109:409–17. Timmerman JM, Czerwinski DK, Davis TA, et al. Idiotypepulsed dendritic cell vaccination for B-cell lymphoma: clinical and immune responses in 35 patients. Blood 2002; 99:1517–26. Butterfield LH, Ribas A, Dissette VB, et al. Determinant spreading associated with clinical response in dendritic
83
84
85
86
●87
88
89
90
91
92
93
94
95
cell-based immunotherapy for malignant melanoma. Clin Cancer Res 2003; 9:998–1008. Waeckerle-Men Y, Scandella E, Uetz-Von Allmen E, et al. Phenotype and functional analysis of human monocytederived dendritic cells loaded with biodegradable poly(lactide-co-glycolide) microspheres for immunotherapy. J Immunol Methods 2004; 287:109–24. Heiser A, Dahm P, Yancey DR, et al. Human dendritic cells transfected with RNA encoding prostate-specific antigen stimulate prostate-specific CTL responses in vivo. J Immunol 2000; 164:5508–14. Di Nicola M, Carlo-Stella C, Mortarini R, et al. Boosting T cell-mediated immunity to tyrosinase by vaccinia virus-transduced, CD34()-derived dendritic cell vaccination: a phase I trial in metastatic melanoma. Clin Cancer Res 2004; 10:5381–90. Mullins DW, Sheasley SL, Ream RM, et al. Route of immunisation with peptide-pulsed dendritic cells controls the distribution of memory and effector T cells in lymphoid tissues and determines the pattern of regional tumor control. J Exp Med 2003; 198:1023–34. Franco A, Tilly DA, Gramaglia I, et al. Epitope affinity for MHC class I determines helper requirement for CTL priming. Nat Immunol 2000; 1:145–50. Bedrosian I, Mick R, Xu S, et al. Intranodal administration of peptide-pulsed mature dendritic cell vaccines results in superior CD8 T-cell function in melanoma patients. J Clin Oncol 2003; 21:3826–35. Bonifaz L, Bonnyay D, Mahnke K, et al. Efficient targeting of protein antigen to the dendritic cell receptor DEC-205 in the steady state leads to antigen presentation on major histocompatibility complex class I products and peripheral CD8 T cell tolerance. J Exp Med 2002; 196:1627–38. Pavlenko M, Roos AK, Lundqvist A, et al. A phase I trial of DNA vaccination with a plasmid expressing prostatespecific antigen in patients with hormone-refractory prostate cancer. Br J Cancer 2004; 91:688–94. Rosenberg SA, Zhai Y, Yang JC, et al. Immunising patients with metastatic melanoma using recombinant adenoviruses encoding MART-1 or gp100 melanoma antigens. J Natl Cancer Inst 1998; 90:1894–900. Babiuk LA, Pontarollo R, Babiuk S, et al. Induction of immune responses by DNA vaccines in large animals. Vaccine 2003; 21:649–58. Fu TM, Ulmer JB, Caulfield MJ, et al. Priming of cytotoxic T lymphocytes by DNA vaccines: requirement for professional antigen presenting cells and evidence for antigen transfer from myocytes. Mol Med 1997; 3:362–71. Porgador A, Irvine KR, Iwasaki A, et al. Predominant role for directly transfected dendritic cells in antigen presentation to CD8 T cells after gene gun immunisation. J Exp Med 1998; 188:1075–82. Weiss R, Scheiblhofer S, Freund J, et al. Gene gun bombardment with gold particles displays a particular Th2-promoting signal that over-rules the Th1-inducing effect of immunostimulatory CpG motifs in DNA vaccines. Vaccine 2002; 20:3148–54.
184 Vaccination strategies for malignant diseases
96 Sato Y, Roman M, Tighe H, et al. Immunostimulatory DNA sequences necessary for effective intradermal gene immunisation. Science 1996; 273:352–4. 97 Scheerlinck JP, Casey G, McWaters P, et al. The immune response to a DNA vaccine can be modulated by co-delivery of cytokine genes using a DNA prime-protein boost strategy. Vaccine 2001; 19:4053–60. 98 Velders MP, Weijzen S, Eiben GL, et al. Defined flanking spacers and enhanced proteolysis is essential for eradication of established tumors by an epitope string DNA vaccine. J Immunol 2001; 166:5366–73. 99 Rosenberg SA, Yang JC, Sherry RM, et al. Inability to immunize patients with metastatic melanoma using plasmid DNA encoding the gp100 melanoma–melanocyte antigen. Hum Gene Ther 2003; 14:709–14. 100 Tagawa ST, Lee P, Snively J, et al. Phase I study of intranodal delivery of a plasmid DNA vaccine for patients with Stage IV melanoma. Cancer 2003; 98:144–54. 101 Triozzi PL, Aldrich W, Allen KO, et al. Phase I study of a plasmid DNA vaccine encoding MART-1 in patients with resected melanoma at risk for relapse. J Immunother 2005; 2:382–8. 102 Miller AM, Ozenci V, Kiessling R, Pisa P. Immune monitoring in a phase 1 trial of a PSA DNA vaccine in patients with hormone-refractory prostate cancer. J Immunother 2005; 28:389–95. 103 Todorova K, Ignatova I, Tchakarov S, et al. Humoral immune response in prostate cancer patients after immunization with gene-based vaccines that encode for a protein that is proteasomally degraded. Cancer Immun 2005; 115:1. 104 Amara RR, Villinger F, Altman JD, et al. Control of a mucosal challenge and prevention of AIDS by a multiprotein DNA/MVA vaccine. Science 2001; 292:69–74. 105 Wolchok JD, Gregor PD, Nordquist LT, et al. DNA vaccines: an active immunisation strategy for prostate cancer. Semin Oncol 2003; 30:659–66. 106 Roth JA, Nguyen D, Lawrence DD, et al. Retrovirus-mediated wild-type p53 gene transfer to tumors of patients with lung cancer. Nat Med 1996; 2:985–91. 107 Perkus ME, Taylor J, Tartaglia J, et al. Live attenuated vaccinia and other poxviruses as delivery systems: public health issues. Ann N Y Acad Sci 1995; 754:222–33. 108 Corona Gutierrez CM, Tinoco A, Navarro T, et al. Therapeutic vaccination with MVA E2 can eliminate precancerous lesions (CIN 1, CIN 2, and CIN 3) associated with infection by oncogenic human papillomavirus. Hum Gene Ther 2004; 15:421–31. 109 Spaner DE, Astsaturov I, Vogel T, et al. Enhanced viral and tumor immunity with intranodal injection of canary pox viruses expressing the melanoma antigen, gp100. Cancer 2006; 106:890–9. 110 van Baren N, Bonnet MC, Dreno B, et al. Tumoral and immunologic response after vaccination of melanoma patients with an ALVAC virus encoding MAGE antigens recognized by T cells. J Clin Oncol 2005; 23:9008–21. 111 Ahmad M, Rees RC, McArdle SE, et al. Regulation of CTL responses to MHC-restricted class I peptide of the gp70
112
113
114
115
116
117
◆118
119
120
121
122
●123
●124
tumour antigen by splenic parenchymal CD4 T cells in mice failing immunotherapy with DISC-mGM-CSF. Int J Cancer 2005; 115:951–9. Rees RC, McArdle S, Mian S, et al. Disabled infectious single cycle-herpes simplex virus (DISC-HSV) as a vector for immunogene therapy of cancer. Curr Opin Mol Ther 2002; 4:49–53. Ali SA, Lynam J, McLean CS, et al. Tumor regression induced by intratumor therapy with a disabled infectious single cycle (DISC) herpes simplex virus (HSV) vector, DISC/HSV/murine granulocyte–macrophage colonystimulating factor, correlates with antigen-specific adaptive immunity. J Immunol 2002; 168:3512–19. Assudani DP, Ahmad M, Li G, et al. Immunotherapeutic potential of DISC-HSV and OX40L in cancer. Cancer Immunol Immunother 2006; 55:104–11. Garcia-Hernandez E, Gonzalez-Sanchez JL, AndradeManzano A, et al. Regression of papilloma high-grade lesions (CIN 2 and CIN 3) is stimulated by therapeutic vaccination with MVA E2 recombinant vaccine. Cancer Gene Ther 2006; 13:592–7. Gentschev I, Fensterle J, Schmidt A, et al. Use of a recombinant Salmonella enterica serovar Typhimurium strain expressing C-Raf for protection against C-Raf induced lung adenoma in mice. BMC Cancer 2005; 5:15. Weiskirch LM, Pan ZK, Paterson Y. The tumor recall response of antitumor immunity primed by a live, recombinant Listeria monocytogenes vaccine comprises multiple effector mechanisms. Clin Immunol 2001; 98:346–57. Ferrari M. Cancer nanotechnology: opportunities and challenges. Nat Rev Cancer 2005; 5:161–71. Mandal M, Lee KD. Listeriolysin O-liposome-mediated cytosolic delivery of macromolecule antigen in vivo: enhancement of antigen-specific cytotoxic T lymphocyte frequency, activity, and tumor protection. Biochim Biophys Acta 2002; 1563:7–17. Tartour E, Ciree A, Haicheur N, et al. Development of nonlive vectors and procedures (liposomes, pseudo-viral particles, toxin, beads, adjuvantsellipsis) as tools for cancer vaccines. Immunol Lett 2000; 74:45–50. Cho JA, Yeo DJ, Son HY, et al. Exosomes: a new delivery system for tumor antigens in cancer immunotherapy. Int J Cancer 2005; 114:613–22. Zhou J, Dudley ME, Rosenberg SA, Robbins PF. Persistence of multiple tumor-specific T-cell clones is associated with complete tumor regression in a melanoma patient receiving adoptive cell transfer therapy. J Immunother 2005; 28:53–62. Dudley ME, Wunderlich JR, Yang JC, et al. A phase I study of nonmyeloablative chemotherapy and adoptive transfer of autologous tumor antigen-specific T lymphocytes in patients with metastatic melanoma. J Immunother 2002; 25:243–51. Antony PA, Piccirillo CA, Akpinarli A, et al. CD8 T cell immunity against a tumor/self-antigen is augmented by CD4 T helper cells and hindered by naturally occurring T regulatory cells. J Immunol 2005; 174:2591–601.
References 185
●125
126
127
128
129
130
◆131
132
133
Dudley ME, Wunderlich JR, Yang JC, et al. Adoptive cell transfer therapy following non-myeloablative but lymphodepleting chemotherapy for the treatment of patients with refractory metastatic melanoma. J Clin Oncol 2005; 23:2346–57. Ahmad M, Rees RC, Ali SA. Escape from immunotherapy: possible mechanisms that influence tumor regression/progression. Cancer Immunol Immunother 2004; 53:844–54. Cabrera CM, Jimenez P, Cabrera T, et al.Total loss of MHC Class I in colorectal tumors can be explained by two molecular pathways: beta 2-microglobulin inactivation and MSI-positive tumors and LMP7/TAP2 downregulation in MS negative tumors. Tissue Antigens 2003; 61:211–19. Cabrera T, Pedrajas G, Cozar JM, et al. HLA class I expression in bladder carcinomas. Tissue Antigens 2003; 62:324–7. Jager E, Jager D, Knuth A. CTL-defined cancer vaccines in melanoma and other epithelial cancer. In: Stern PL, Beverely PCL, Carroll MW (eds) Cancer Vaccines and Immunotherapy. Cambridge: Cambridge University Press, 2000, 207–217. Boursnell MEG, Entwistle C, Ali SA, et al. Disabled Infectious Single Cycle (DISC) Herpes Simplex Virus as a vector for immunotherapy of cancer. Adv Exp Med Biol 1998; 451:379–84. Dunn GP, Bruce AT, Ikeda H, et al. Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol 2002; 3:991–8. Melcher A, Todryk S, Hardwick N, et al. Tumor immunogenicity is determined by the mechanism of cell death via induction of heat shock protein expression. Nat Med 1998; 4:581–7. Melero I, Singhal MC, McGowan P, et al. Immunological ignorance of an E7-encoded cytolytic T-lymphocyte
134
135
136
137
138
139
140
141 142
epitope in transgenic mice expressing the E7 and E6 oncogenes of human papillomavirus type 16. J Virol 1997; 71:3998–4004. Mellor AL, Munn DH. Tryptophan catabolism and T-cell tolerance: immunosuppression by starvation? Immunol Today 1999; 20:469–73. Mukherjee P, Ginardi AR, Madsen CS, et al. MUC1-specific CTLs are non-functional within a pancreatic tumor microenvironment. Glycoconj J 2001; 18:931–42. Plescia OJ, Grinwich K, Plescia AM. Subversive activity of syngeneic tumor cells as an escape mechanism from immune surveillance and the role of prostaglandins. Ann N Y Acad Sci 1976; 276:455–65. Ragnarsson GB, Mikaelsdottir EK, Vidarsson H, et al. Intracellular Fas ligand in normal and malignant breast epithelium does not induce apoptosis in Fas-sensitive cells. Br J Cancer 2000; 83:1715–21. Ramal LM, Feenstra M, van der Zwan AW, et al. Criteria to define HLA haplotype loss in human solid tumors. Tissue Antigens 2000; 55:443–8. Ramal LM, Maleno I, Cabrera T, et al. Molecular strategies to define HLA haplotype loss in microdissected tumor cells. Hum Immunol 2000; 61:1001–12. Gabrilovich DI, Velders MP, Sotomayor EM, Kast WM. Mechanism of immune dysfunction in cancer mediated by immature Gr-1 myeloid cells. J Immunol 2001; 166:5398–406. Restifo NP. The new vaccines: building viruses that elicit antitumor immunity. Curr Opin Immunol 1996; 8:658–63. Rohrlich PS, Cardinaud S, Firat H, et al. HLA-B*0702 transgenic, H-2KbDb double-knockout mice: phenotypical and functional characterization in response to influenza virus. Int Immunol 2003; 15:765–72.
9 Biological therapies: cytokines and adoptive cell therapy FIONA C. THISTLETHWAITE, PETER L. STERN AND ROBERT E. HAWKINS
Introduction Cytokines in cancer therapy
186 188
INTRODUCTION This chapter covers bio-immunotherapies for cancer that directly or indirectly exploit the anti-tumour activities of particular cytokines or immune effector cells. The impetus for these approaches has been reinvigorated as a result of the renaissance of the concept of immunosurveillance1 and the significant advances that have been made in our understanding of the molecular and cellular mechanisms underpinning tumour immunity.2 Faith in the concept that the immune system could control tumour development had fallen away in the 1970s when experiments of tumour induction with chemical carcinogens showed no increased frequency in animals with a severe deficiency in T-lymphocyte-mediated (T-cell-mediated) immune function – athymic nude mice. Indeed, a recent major review identifying the supposed hallmarks of cancer failed to acknowledge any significant role for the immune system in cancer.3 In part, the problem with the nude mouse experiments was the underestimation of the role of small numbers of residual T cells, as well as of natural killer (NK) cells, and the importance of cytokine effector molecules. However, there is now convincing evidence that the cellular immune system and associated cytokine networks are key players acting to control tumour emergence. This is most dramatically demonstrated by the increased susceptibility to tumour induction by methyl cholanthrene in various knock-out mouse models with defects in T-cell, B-cell and NK-cell function or in their interferon (IFN) pathways. These include mice deficient in RAG2 TCRαβ, TCRγδ, perforin, STAT1 and IFN-γ as well as the SCID model. Interestingly combinations of these defects such as RAG2 x
Adoptive cell therapy References
194 198
STAT1 or perforin x IFN-γ mice show evidence of increased spontaneous tumours and subtle difference in the prevalence of types.1 There is abundant evidence that tumours are immunogenic during cancer pathogenesis with both antibody and T-cell-recognized antigens.4,5 These tumour-associated antigens (TAAs) include the products of mutated oncogenes, other altered gene products, cancer–testis antigens, differentiation and virally encoded antigens. The tumour microenvironment, consisting of neoplastic cells, immune cells, stromal cells and extracellular matrix, is the principal combat zone. It is the result of complex interactions between all these components that ultimately determines whether tumour surveillance is successful and the tumour eradicated, or it fails and the tumour becomes established. It is likely that in the early phases of tumorigenesis a lack of danger signals is an important factor in any failure to activate the innate arm of the immune response.6 The innate response plays an important role in co-ordinating inflammatory reactions and provides the first line of defence against infection. It includes soluble factors such as complement proteins in addition to cells such as granulocytes, macrophages, NK cells and mast cells. The process of establishing an innate response critically influences the activation of antigen-presenting cells (APCs) such as dendritic cells, the sentinels that awake the adaptive response.7 Endogenous danger signals are important triggers in the initiation of the innate response. Released by damaged, stressed or abnormally dying cells these danger signals include heat shock proteins, nucleotides, reactive oxygen intermediates and cytokines. However, it is clear primary tumours often do not provide adequate danger
Introduction 187
signals, and thus avoid any early activation of innate immunity. Adaptive immunity supplies the potential for specific recognition of TAAs and subsequent expansion and differentiation of anti-tumour cytotoxic T lymphocytes (CTLs) as well as providing for memory populations. Importantly, the signals provided by the local innate immunity influence the antigen-presenting activated dendritic cells and can influence the subsequent adaptive immune response. Thus specific T-cell responses can be biased to either a T helper 1 (Th1) or T helper 2 (Th2) type, which preferentially promote the development of CTLs or B-cell antibody production and class switching respectively.8 The important role that the adaptive immune system plays is reflected in the observation that for many solid tumours, significant lymphocyte infiltration can be correlated with reduced frequency of metastasis and improved patient survival.9 The nature of tumour development per se, however, involves genomic instability and this provides the engine for generating variant tumour clones.3 Evidence exists for the selection of tumour variants that can avoid both antigen-specific T-cell and NK anti-tumour activities as a result of the immune pressure exerted over the natural history of the malignancy.10 Multiple mechanisms have been described that can account for such tumour evolution. Both T-cell and NK-cell populations are controlled by the presence or absence of the major histocompatability complex (MHC) class molecules respectively, and loss of these is a very common event during tumour evolution.1,11 The natural history of tumour development in humans is long, so sufficient time is available to allow for the emergence of tumour variants lacking the appropriate human leukocyte antigen (HLA) restricting element for TAA-specific CTL sensitivity, whereas such allele-specific loss does not lead to increased NK sensitivity.12 Chronic sub-optimal stimulation of anti-tumour T cells can lead to depletion of key populations of effector T cells. It is increasingly clear that in cancer patients there are significant increases of T regulatory (Treg) cells, which can act both specifically and non-specifically to limit anti-tumour immunity.13–15 Thus, in the presenting tumour microenvironment, there are probably strong negative regulatory influences on positive anti-tumour responses. Tumours also have endogenous mechanisms that provide for immune suppression, such as the production of immunosuppressive cytokines (TGF-β),16 the ability to induce apoptosis in infiltrating lymphocytes through Fas–Fas ligand interactions,17 and the activation of T cells without appropriate co-stimulatory molecules leading to anergy of TAA-specific T cells.18 Figure 9.1 summarizes some of the key factors that can influence the regression or progression of a tumour in the face of its recognition by cellular immunity during the natural history of a cancer. As discussed in Chapter 8, one approach to immunotherapy is to try to immunize patients using various means either to produce de-novo anti-tumour immunity (trying to engage a new anti-tumour repertoire) or to stimulate existing cellular immunity so as to overcome the various negative
Elimination
Regression
TUMOUR Escape Natural Selection / Heterogeneity
Progression
1. Poor inflammation • No Danger Signals 2. Poor immunogenicity 3. Immune suppression • Tumour (TGF β) • T regulatory cells
1. Innate immunity 2. APC processing 3. Th1 cell response 4. CTL generation 5. Long term memory
(a) TUMOUR Elimination
Regression
Escape Natural Selection / Heterogeneity
1. Innate immunity 2. APC processing 3. Th1 response 4. CTL generation 5. Long term memory
Progression
1. Blunt innate response 2. Interfere APCs 3. Skew Th1 v Th2 4. Anergize T response 5. No long term memory
(b) TUMOUR Elimination
Regression
Escape Natural Selection / Heterogeneity
1. Innate immunity 2. APC processing 3. Th1 cell response 4. CTL generation 5. Long term memory
Progression
1. Multiple tumour variants 2. MHC/Ag loss on tumour • NK and T resistance 3. T repertoire depletion 4. T regulatory cells 5. Other very bad stuff
(c)
Figure 9.1 Key factors that influence the regression or progression of a tumour. Early in the natural history, if the tumour is immunogenic, danger signals may activate an immune response resulting in cancer regression and elimination (a). With time, however, the tumour may develop escape mechanisms that lead to an equilibrium (b), or ultimately tolerance and establishment of the cancer (c).
influences discussed above. The challenge becomes increasingly daunting with more advanced cancers, especially where there is widespread metastasis of heterogeneous tumour cells. In order to treat such patients effectively, a very rapid response of the tumour target to treatment will obviously be advantageous. The basic rationale of treatments that use cytokines or adoptive cell transfer is that, as with monoclonal antibody therapy, they may exert a bolus anti-tumour activity that can override many immune modulatory mechanisms of the tumour targets. They can thus produce a shift in the equilibrium, allowing the action of other pre-existing
188 Biological therapies: cytokines and adoptive cell therapy
anti-tumour activity, thereby delivering a favorable clinical outcome with longer-term control.
CYTOKINES IN CANCER THERAPY Cytokines are secreted or membrane-bound proteins that regulate the growth, differentiation and activation of immune cells in a coordinated fashion.19 They are released in response to a wide variety of cellular stresses, including infection, inflammation and carcinogen-induced injury, and function to stimulate and coordinate a host response. A variety of cells within the tumour microenvironment produce cytokines, including host stromal cells, infiltrating immune cells and the malignant cells themselves. The biological effect of an individual cytokine can be highly pleiotrophic and determined by many factors, including the array of expressed cytokines, their relative concentrations and cytokine receptor expression patterns. In the context of immunosurveillance, the net influence of a cytokine network may be to tilt the balance towards the activation of immune effector mechanisms that limit the growth of the tumour. Alternatively, it may be tipped towards tumour growth, invasion and metastasis. Figure 9.2 highlights some of the network of important cytokine and lymphocyte interactions that occur within the tumour microenvironment.
DC
The manipulation of this cytokine balance has proved to be an attractive target for cancer therapies. Cytokines can be divided into five major families. 1. 2. 3. 4. 5.
Type I cytokines – interleukins (ILs). Type II cytokines – interferons (IFNs). Tumour necrosis factors (TNFs). Immunoglobulin supergene family. Chemokines.
The list of individual members of these families and their receptors extends to more than 100 molecules. It is clearly beyond the scope of this chapter to detail all of them individually, their biological actions and their clinical potential. However, Table 9.1 lists some of the cytokines that have had a significant impact on our understanding of the area, and highlights their principal biological and therapeutic effects. Many are currently being explored in trials and may become part of accepted clinical practice in years to come. The following is a more detailed evaluation of the cytokines used in clinical practice today.
Interferons Interferons are widely expressed cytokines with important anti-viral, anti-proliferative and immunomodulatory functions. The IFN family includes two main classes, referred to
NK IFN-α IFN-β IL-12
Tumour Kill
CD8
IL-10
IL-2 IFN-Y TNF-β
Tumour
TGF-β IL-10
T-reg
Th1
IL-10 CD4
Th2
IL-4 IL-5 IL-6 IL-10 IL-13
B-Cell
Figure 9.2 Complexity of the cytokine and lymphocyte interactions within the tumour microenvironment. CD4 cells are influenced by cytokines such as IL-12, IL-18 and IFN-α to differentiate down the Th1 pathway. Upon activation Th1 cells produce an array of cytokines including IL-2, IFNγ and TNF-β. CD8 cells are activated by these following specific recognition of peptides presented from MHC class I molecules on APCs (primary DCs) via their TCR, and co-stimulation from molecules such as CD28. Further positive signals come from IFNα, IFN-β and IL-12 which induce CD8 CTLs and NK cells to destroy target cells, by apoptosis, via multiple pathways. These pathways include secretion of granules of perforin and granzyme, or by triggering receptors for TNF-like receptors on target cells by soluble molecules such as Fas Ligand, TRAIL.
Table 9.1 Cytokines, their principal biological effects and current or potential clinical uses Cytokine
Biological effects
Therapeutic potential
IL-1α
Pro-inflammatory: increased expression of genes such as COX-2 and promotion of infiltration of leukocytes
Trials have shown some evidence of prevention of thrombocytopenia, but at the expense of significant toxicity limiting further development19
IL-1β
Role in angiogenesis: activation of vascular endothelial cells Increased invasiveness of malignant cells
Blockade of IL-1 can be helpful in inflammatory conditions, but was not found to be effective at preventing graft-versus-host disease20 in allogeneic transplantation
IL-2 (see text)
T-cell proliferation Enhanced cytotoxicity of T cells and NK cells Increased vascular permeability Co-factor in activating macrophages and B cells
Used primarily in melanoma and renal cell carcinoma Toxicity can be severe, including ‘capillary-leak’ syndrome, fever, hypotension, diarrhoea, renal and hepatic impairment
IL-3
Stimulation of proliferation and differentiation of haematopoietic stem cells Enhanced tumour-antigen presentation
Clinical trials have investigated a role in increasing haematopoiesis, for example in myelodysplastic syndrome21 Used to sustain haematopoietic stem cells ex vivo
IL-4
B-cell activation and differentiation, mediation of Ig class switching Pluripotent effects on cells, including monocytes, macrophages, mast cells, myeloid and erythroid precursors, endothelial cells and fibroblasts Directs activated T cells to Th2 pathway
Co-stimulatory immunoadjuvant Early-phase trials as direct anti-tumour drug have determined maximum tolerated dose, but not yet demonstrated efficacy22 Therapeutic target in inflammatory diseases
IL-6
Inflammatory mediator Central role in acute-phase response and fever Enhanced T-cell and B-cell function
Prognostic marker in several tumour types, including lymphoma and multiple myeloma Anti-tumour effects in clinical trials of IL-6 have been limited Promising results with anti-IL-6 antibodies in B-lymphoproliferative disorders,23 Castleman’s disease and rheumatoid arthritis Anti-IL-6 antibodies may also have a role in treating cancer-related cachexia
IL-7
Enhanced T-cell function Growth and differentiation factor for precursor B cells
Not yet tested clinically
IL-10
Inhibition of Th1 response Inhibition of macrophage-antigen presentation Inhibition of macrophage production of pro-inflammatory cytokines such as IL-1, IL-6 and TNF
Clinical trials using IL-10 in inflammatory conditions have as yet failed to define a clear role Anti-IL-10 antibodies are under investigation in systemic lupus erythematosus24
IL-11
Stimulation of haematopoiesis, particularly induction of production of megakaryocytes Spectrum of effects similar to IL-6
Administration of IL-11 has been shown to reduce the platelet transfusion requirement in breast cancer patients at high risk of thrombocytopenia25
IL-12
Stimulation of proliferation, cytokine production (especially IFN) and lytic activity of cytotoxic T cells and NK cells Promotion of maturation and activation of Th1 cells
Clinical trials have investigated subcutaneous or intra-tumoral administration as single agent or in combination, for example with IFN-α26 or monoclonal antibodies27 Clinical responses with IL-12 have been observed in lymphoid malignancies28 (non-Hodgkin’s lymphoma and cutaneous T-cell lymphoma)
IL-15
Enhanced T-cell and NK-cell proliferation and cytotoxicity
No published clinical data May have therapeutic potential similar to IL-2 May have a role in ex-vivo expansion of cells for adoptive therapy29
IL-18
Synergism with IL-12 to induce production of IFN-γ from NK and T cells Promotion of development of Th1 responses
Anti-tumour effects in vivo have been demonstrated in mouse models and clinical trials are awaited30
190 Biological therapies: cytokines and adoptive cell therapy
Table 9.1 (Continued) Cytokine
Biological effects
Therapeutic potential
IL-21
B-cell differentiation and antibody production NK-cell and T-cell differentiation, proliferation and cytotoxicity
Early-phase clinical trials currently underway
M-CSF
Proliferation and differentiation of monocyte/macrophage lineage
Lack of clinical evidence of direct anti-tumour effect in human trials Some indication that the incidence of chemotherapyrelated febrile neutropenia may be reduced when administered following chemotherapy31 Animal models show it induces thrombocytopenia, which may limit clinical applications
GM-CSF
Proliferation and activation of granulocyte and macrophage cell lineages
Phase II trials have explored its use in a number of tumour types either as a single agent or in combination, for example with chemotherapy32,33 Clinical responses noted when administered as aerosol to patients with pulmonary metastases34 Immunogenic effect noted when combined with chemotherapy and IL-2 in patients with metastatic colon cancer35 Phase III data awaited GM-CSF-based vaccines have shown promising early results36
IFN-α and IFN-β (see text)
Stimulation of proliferation and cytotoxicity of T cells and NK cells Promotion of increased Th1 responses Increased IgG secretion from B cells Regulatory and immunomodulatory effects on other cytokine and chemokine secretion Increased MHC expression and antigen presentation
The use of both has been explored, either as single agents or in combination with other molecules such as IL-2 IFN-α is widely used in many haematological and solid malignancies, particularly renal-cell carcinoma and melanoma To date, no conclusive therapeutic benefit has been demonstrated for IFN-β in malignancy (although it is extensively used to modify the course of multiple sclerosis)
IFN-γ (see text)
Augmentation of NK-cell and cytotoxic T-cell activity Induction expression of MHC molecules
Used in combination with TNF-α and melphalan for the treatment, by IPL, of in-transit metastases from melanoma and sarcoma, but activity as a single agent not proven
TNF-α (see text)
Pro-inflammatory Proliferation and activation of NK cells, T cells, B cells, macrophages and dendritic cells Induction of tumour-cell apoptosis
Dose-limiting toxicity at sub-therapeutic doses has limited systemic therapy, but, as described in the text, IPL has been shown to be clinically useful in melanoma and sarcoma Anti-TNF-α antibodies have therapeutic benefit in inflammatory conditions such as rheumatoid arthritis
FLT3 ligand
Expansion and mobilization of stem cells and progenitors Dendritic-cell and NK-cell differentiation from haematopoietic progenitor cells
Early-phase trials are examining the use of FLT3 ligand as a strategy to increase the number of dendritic cells, for example post-haematopoietic-cell transplant.ation37
Lymphotactin
Chemokine that enhances T-cell recruitment
Transgenic expression of lymphotactin (and IL-2) from an allogeneic neuroblastoma-cell vaccine has produced promising results in a phase I trial38
IL, interleukin; M-CSF, macrophage colony-stimulating factor; GM-CSF, granulocyte–macrophage colony-stimulating factor; IFN, interferon; TNF, tumour necrosis factor FLT3, fms–like tyrosine kinase 3; Th1/Th2, T helper 1 cell/T helper 2 cell; NK, natural killer; Ig, immunoglobulin; IPL, isolated limb perfusion.
Cytokines in cancer therapy 191
as type I and type II IFNs. Type I IFNs are a gene family consisting of about 20 members, including IFN-α, IFN-β, IFN-δ, IFN-ε, IFN-κ, IFN-τ and IFN-ω. IFN-α can be further subdivided into 13 functional genes in humans. The genes for the type I IFNs are located on the short arm of chromosome 9, and their proteins share significant aminoacid sequence homology. By contrast, there is only one type II IFN, IFN-γ. The gene that encodes it is located on chromosome 12 and does not share significant structural homology with the type I IFN genes. Interferon-α is the only IFN that is widely used in oncology at present, and is discussed in the paragraphs below. Therapeutic roles for IFN-β and IFN-γ have also been explored, either as single agents or in combination with other cytokines such as IFN-α or IL-2. Consistent additional therapeutic benefit has not been demonstrated for either of these agents when administered systemically.20,21 Interferon-γ, in combination with TNF-α and melphalan, has, however, been used in isolated limb perfusion (ILP) for metastatic disease for melanoma,22 which is discussed further in a later section. INTERFERON-α
Biology Interferon-α is produced primarily by leukocytes, classically when stimulated by viral infection, but other stimuli clearly exist, including bacterial endotoxins, unmethylated DNA (CpG DNA) and double-stranded RNA.23 Once induced, IFN-α must bind to its receptor to exert its biologic effect. All type I IFNs share a common receptor, which is composed of two subunits, a 100-kDa ligand-binding subunit (IFN-α R1) and a 125-kDa subunit (IFN-α R2), involved in both ligand binding and signal transduction. Binding of IFN-α to this receptor results in a cascade of events,24 including the activation of Janus tyrosine kinases JAK1 and TYK2. This enables the phosphorylation and subsequent dimerization of the signal transducers and activators of transcription 1 and 2 (STAT1 and STAT2) to form one of two distinct transcriptional activator complexes, namely IFN-α-activated factor (AAF) and IFN-stimulated gene factor 3 (ISGF3). These complexes disassociate from the IFN receptor and translocate to the nucleus, where they bind to the IFN-stimulated response element of the α-response genes, causing the induction of hundreds of target genes. The subsequent biologic effects of IFN-α on the immune system are extensive, involving both the innate and the adaptive immune branches. For example, in the innate response, NK-cell activity, trafficking, secretion of IFN-γ and cytotoxicity are increased.25 In the adaptive response, IFN-α influences CD8 T-cell and B-cell responses by up-regulating MHC class I and class II expression, increasing antigen presentation and increasing CD8 T-cell-mediated killing of tumour cells.26 Clinical roles Interferon-α is available in recombinant formulations and is licensed for the treatment of a number of malignancies
as well as chronic hepatitis B and C (usually in combination with ribavirin). Two types of recombinant IFN-α are widely used in oncology, IFN-α-2a and IFN-α-2b. They differ only slightly in their carbohydrate components, but no comparative data regarding toxicity or efficacy exist with regards to them, or to any of the other formulations available. Toxicities of IFN-α include flu-like symptoms, such as fever, chills, myalgia and fatigue, as well as weight loss, altered taste, depression, anaemia, leucopenia and elevated liver function test results. For many patients, the administration of IFN-α at bedtime can limit the impact of symptoms, and some adverse effects, especially the flu-like symptoms, diminish during long-term therapy. Interferon-α has been shown to induce responses in several haematological conditions,24 including hairy-cell leukaemia, multiple myeloma and chronic myeloid leukaemia (CML), but it is in its use in solid malignancies that there are significant areas of contention. Melanoma The use of IFN-α as adjuvant therapy in patients with resected melanoma at high risk of relapse has been a longrunning point of contention. An early report suggested that for patients randomized to receive low-dose IFN-α, administered for 18 months, there was evidence of significantly increased relapse-free survival (p 0.038) and a trend towards an overall survival benefit (p 0.059).27 Interferon-α was subsequently licensed for the adjuvant treatment of high-risk malignant melanoma in Europe. More recently, a systematic review of this trial plus four other published randomized trials comparing low-dose IFN-α with observation has failed to detect a statistically significant improvement in overall survival.28 For example, in the AIM HIGH study from the United Kingdom Coordinating Committee on Cancer Research, 674 patients with completely resected stage IIB and stage III cutaneous malignant melanoma were randomized to receive either low-dose IFN-α (3 MU three times per week for 2 years) or observation.29 At median follow-up of 3.1 years, no significant differences in overall survival (p 0.6) or relapse-free survival (p 0.3) were detected and it was concluded that low-dose IFN-α as adjuvant therapy could not be recommended in these circumstances. It is, however, the use of high-dose INF-α that has generated the greatest controversy. Whilst many of the published trials have produced improvements in diseasefree survival, this is at the expense of significant toxicity, and inconsistent results for overall survival benefits, particularly in three Eastern Cooperative Oncology Group (ECOG) studies, have generated extensive debate. In the first ECOG trial, 1684, 20 MU/m2 intravenous IFN-α was administered five times a week for 4 weeks, followed by maintenance with 10 MU/m2 three times a week for a further 11 months, and compared to an observation arm.30 A significant survival benefit was initially observed in the high-dose IFN-α arm, but this advantage was lost with longer follow-up. A confirmatory study, ECOG 1690, also failed to demonstrate
192 Biological therapies: cytokines and adoptive cell therapy
an improvement in overall survival.31 It was suggested that this might be a result of a significantly greater proportion of patients in the observation arm being salvaged by an IFN-containing regimen (31 per cent of recurrences versus 15 per cent in the IFN-α arm; p 0.003) or biochemotherapy (17 per cent of recurrences in the observation arm versus 6 per cent of recurrences in the IFN-α arm; p 0.013). An imbalance in the randomization arms of ECOG 1684, due to lack of stratification according to positive lymph-node number, was also cited as a possible contributory factor. A third ECOG trial, 1694, compared high-dose IFN-α with a vaccine, GM-2KLH (GM ganglioside-keyhole limpet hemocyanin).32 At a median follow-up of 2.1 years, highdose IFN-α demonstrated superiority to the GM2-KLH vaccine in terms of both overall survival (hazard ratio 1.33; p 0.04) and relapse-free survival (hazard ratio 1.33; p 0.006). This result has been interpreted as favoring high-dose IFN-α, but a commonly expressed concern is that without a no-treatment control arm the possibility of a detrimental effect of the vaccine cannot be ruled out. Several meta-analyses have been published looking at trials of IFN-α in the adjuvant setting.28,33 From their study, Wheatley and colleagues concluded that relapsefree survival was improved with IFN-α (hazard ratio for recurrence 0.83; 95% confidence interval (CI) 0.77–0.90; p 0.000003), corresponding to a 17 per cent reduction in the odds of recurrence. However, there was no clear benefit for survival (hazard ratio for mortality 0.93; 95% CI 0.85–1.02; p 0.1). Subgroup analysis indicated a significant trend for an increasing benefit of IFN with increasing dose in terms of relapse-free survival. In a recent publication, however, Verma and colleagues pooled the results from the three ECOG studies described. A meta-analysis of 2-year death rates yielded a risk ratio of 0.85 (95% CI 0.73–0.99; p 0.03). In light of this and the improvement in disease-free survival, they suggested that high-dose IFNα should be considered for patients at high risk of relapse following resection of their primary melanoma. The high cost, significant toxicity and small benefit do, however, still make the widespread adoption of high-dose IFN-α for these patients a contentious issue. Renal-cell carcinoma Interferon-α has undergone extensive clinical evaluation during the past two decades in metastatic renal-cell carcinoma. Most studies have shown modest anti-tumour activity, with objective response rates tending to be in the order of 10–15 per cent. Four randomized studies have been published in which IFN-α was compared to a non-IFN-α arm in the context of advanced renal-cell carcinoma.34–37 In the study from the Medical Research Council (MRC), 10 MU IFN-α administered subcutaneously three times a week for 12 weeks was compared to medroxyprogesterone acetate and was found to confer a survival benefit.34 Indeed, a recent Cochrane Collaboration Review has found that the response rate overall in these studies comparing IFN-α to a non-IFN-α arm was generally greater than for the control
arm and, overall, IFN-α was associated with a reduced 1-year mortality (OR (odds ratio) for death by 1 year 0.56; 95% CI 0.40–0.77).38 The weighted average median survival showed a modest improvement of 3.8 months. The relationship between improved survival and IFN-α dose or duration of therapy has not been established, but increased doses are certainly associated with more severe toxicity and are thus difficult to justify. Several studies have looked at the addition of various agents to IFN-α in an attempt to improve efficacy. These include chemotherapy (for example vinblastine), hormones, cytokines such as IL-2 or IFN-α, and other miscellaneous agents, including aspirin and cimetidine.38 Overall, there has been little evidence of a significant effect or survival benefit, although the combination of IFN-α and subcutaneous IL-2 with intravenous 5-fluorouracil (5FU) has been advanced in Europe as an active regimen in renal-cell carcinoma.39 It is currently being tested in metastatic disease in the MRC trial RE04. Whether or not there is any benefit of nephrectomy in patients with renal-cell cancer found to have metastatic disease at diagnosis has been the source of much debate over the years. Anecdotal reports have suggested that responses from immunotherapy are rare in patients who still have the primary tumour in situ. Two studies have addressed this question.40,41 In both trials, good performance status (ECOG performance status 0–1) patients were randomized to undergo nephrectomy or not prior to receiving IFN-α administered subcutaneously at a dose of 5 MU/m2 three times a week. Although response rates were low (6 per cent of patients in both studies combined), there was a significant survival benefit in the combined analysis (12.6 months versus 7.8 months with or without nephrectomy respectively), and this benefit was even more pronounced for the fittest patients. Other solid malignancies Some efficacy for IFN-α has been observed in carcinoid tumours, but evidence from randomized trials is often affected by the small sample size dictated by this rare tumour. This is illustrated by one study, which examined its use in patients with midgut carcinoid tumours with liver metastases who had undergone prior primary surgery and hepatic arterial embolization. Sixty-eight patients were randomized to receive octreotide alone or in addition to IFN-α. The results suggested a significantly reduced risk of tumour progression during follow-up for the patients who received the combination therapy (p 0.008), but failed to show a significant difference in survival.42 In another recent randomized trial, no difference was observed in the numbers of patients with disease progression at 1 year between the three treatment arms (lanreotide 1 mg three times a day versus IFN-α 5 MU three times a week versus a combination of both).43 Side effects leading to an interruption of therapy were, however, more frequent in the combination arm. There is some suggestion from small randomized studies that IFN-α in combination with induction chemotherapy
Cytokines in cancer therapy 193
may confer a survival advantage for limited-stage small-cell lung cancer,44 but larger studies are required to confirm this finding. In human immunodeficiency (HIV)-related Kaposi’s sarcoma, response rates to IFN-α of around 40 per cent have been demonstrated,45 but this disease is becoming rarer in the Western world, and other agents, such as liposomal doxorubicin, are now available with reduced toxicity and improved efficacy. A further indication is the use of intra-vesical IFN-α for the treatment of superficial transitional carcinoma-in-situ of the bladder. Complete response rates of 30–60 per cent with less toxicity than with the conventional intra-vesical treatment with Bacillus Calmette–Guerin (BCG) have been reported, but in a direct comparison, total number of recurrences and disease-free interval were in favour of BCG.46 Thus, whilst BCG remains the standard of care, there may be a role for intravesical IFN-α in patients who develop resistance or intolerance to BCG.
Interleukins INTERLEUKIN-2
The existence of Interleukin-2 (IL-2) was first proposed 30 years ago when it was discovered that the presence of a growth factor could induce the in-vitro proliferation of human T cells.47 The molecule, initially termed T-cell growth factor, was later isolated, cloned and subsequently designated interleukin-2.48 The biological effects of IL-2 are a result of binding to specific surface receptors. The high-affinity IL-2 receptor consists of three sub-units: α, β and γ chains. Binding of IL-2 to this receptor induces numerous downstream events, including the recruitment of kinases such as the src family member p56lck (49) and the JAK family of tyrosine kinases including JAK1, JAK2 and JAK3. In turn this leads to activation of various members of the STAT family, including STAT1, STAT3 and STAT5, in T cells.50 There are numerous consequences of this cascade of events, including proliferation, production of secondary cytokines such as IL1, IL-6 and TNF, and enhanced cytolytic activity of T cells and NK cells. Potent anti-tumour activity in vitro led to the exploration of IL-2 as an anti-cancer agent in patients with, primarily, renal-cell carcinoma and melanoma. Renal-cell carcinoma High-dose IL-2 has been used in the treatment of patients with metastatic renal-cell carcinoma for a number of years. Its approval by the Food and Drug Administration in the USA in 1992 was based on data presented on 255 patients entered into seven phase II clinical trials.51 The patients received between 600 000 and 720 000 IU/kg of recombinant human IL-2 by 15-minute infusion every 8 hours during two 5-day courses, with courses separated by 5–9 days. Treatment courses were repeated at 8–12-week intervals in responding patients. Objective responses were seen in 37 (15 per cent) of patients, with 20 partial responses (8 per cent)
and 17 complete responses (7 per cent). Whilst underwhelming in themselves, these figures belie the remarkable durability of some of the responses, particularly for those who achieved a complete response. Follow-up data published in 200252 show that although there have been some late relapses, the majority of complete responders remain in remission, and in addition four patients who achieved partial responses and subsequently underwent surgical resection of all of their residual disease remain alive and disease free. Since the response duration curve appears to have levelled off after the 30-month time point, it can be argued that these patients are unlikely to regress and may actually be cured of their metastatic disease. The widespread adoption of high-dose IL-2 as a standard therapy has been limited, not just because of the low response rates described, but also due to the significant associated toxicity. Although the side effects are limited to a few hours, due to the short half-life of IL-2, they can be severe and not infrequently require admission to an intensive care unit. They commonly include fever, chills, lethargy, diarrhoea, nausea, anaemia, thrombocytopenia, diffuse erythroderma, hepatic and renal dysfunction, and confusion. IL-2 can also cause a ‘capillary leak syndrome’, leading to fluid retention, hypotension, respiratory distress syndrome and cardiac failure. Administration of high-dose IL-2 has therefore been restricted to highly selected patients with good performance status and organ function in centres with extensive experience of this approach. In view of the significant toxicity associated with highdose IL-2, regimens using lower doses of IL-2 either alone or in combination with IFN-α have been explored. Unfortunately, these have generally produced lower response rates and reduced quality of life compared to high-dose IL-2. For example, in one study, 192 patients were randomized to receive either low-dose IL-2 (5 MU/m2 subcutaneously every 8 hours for three doses on day 1, then daily 5 days per week for 4 weeks) plus IFN-α (5 MU/m2 subcutaneously three times a week for 4 weeks) every 6 weeks or high-dose inpatient IL-2 (600 000 IU/kg per dose intravenously every 8 hours, days 1–5 and 15–19, to a maximum of 28 doses) every 12 weeks. The response rate for high-dose IL-2 was significantly higher than that for low-dose IL-2 plus IFN-α (23 per cent versus 9 per cent), and although there was no significant difference in median overall survival (17 months versus 13 months respectively; p 0.12), there appeared to be a trend towards progression-free survival at 3 years (9 patients versus 2 patients; p 0.06). Melanoma Treatment of metastatic melanoma with high-dose IL-2 can also result in durable responses. For example, in one report, a database of 270 patients treated with high-dose IL-2 between 1985 and 1993 was analysed.53 The overall objective response rate was found to be 16 per cent, with 17 patients achieving complete response (CR) (6 per cent) and 26 patients with partial response (PR) (10 per cent). Ten of the patients with CR remained disease free, and disease
194 Biological therapies: cytokines and adoptive cell therapy
did not progress in any patient responding for more than 30 months. However, unsurprisingly, severe toxicity was reported and 6 patients died (2 per cent) as a result of sepsis. This, combined with the lack of randomized controlled trials, means that, as with renal-cell carcinoma, the use of highdose IL-2 is restricted to selected centres.54
Tumour necrosis factor For many years it has been observed that patients with serious pyogenic infections could undergo spontaneous regressions of their tumours. This observation led to the hypothesis that, as a consequence of these infections, a substance might be released that has the ability to provoke tumour cell death. During the 1970s it was indeed shown that a circulating factor induced by bacterial infection had anti-tumour activity in animal experiments.55 This molecule, now known as TNF-α, was isolated and cloned in 198456 and was found to be a member of a family of molecules, known as the TNF superfamily. Unfortunately, clinical trials with systemic administration of recombinant human TNF-α failed to demonstrate significant clinical benefit due to the development of dose-limiting toxicity at levels well below those required for therapeutic effect.57 Isolated limb perfusion has, however, been successfully applied to the local administration of TNF-α and is discussed in more detail below.
rates and palliation from IPL are impressive, information regarding any survival benefit is limited, and many questions remain with regard to dosing and optimal combinations of agents.60 DIRECT TUMOUR INJECTION OF CYTOKINES
Intra-lesional therapy is another approach aimed at delivering cytokines to the local tumour environment. Various cytokines have been investigated, including IFN-α, IFN-β, granulocyte–macrophage colony-stimulating factor (GMCSF) and IL-2. IL-2 in particular has met with some success. For example, in a phase II trial of 24 patients with American Joint Committee on Cancer (AJCC) stage III or IV melanoma, Radny and colleagues injected a total of 245 metastases with IL-2 and achieved CR in 209 (85 per cent) lesions and PR in 21 (9 per cent) lesions.61 The treatment was well tolerated, and biopsies from metastases showed a dense intra-tumoral and peri-tumoral lymphocyte infiltrate surrounding and infiltrating areas of necrotic tumour cells. One potential drawback of injection with cytokines is their short half-life, and approaches that have been explored to address this include the injection of pegylated IL-262 and the intra-tumoral injection of recombinant viral vectors coding for cytokines such as IFN-γ.63
Cytokines in vaccine strategies Local administration of cytokines A number of therapeutic strategies have been developed in an attempt to reduce the systemic toxicity commonly seen with cytokine administration. ISOLATED LIMB PERFUSION
This technique allows elevated concentration of the drugs to the isolated extremity without general systemic exposure and its associated toxicities. The therapeutic mechanism of TNF-α in ILP appears to involve destruction of endothelial cells and vasculature of tumours rather than direct killing of tumour cells.58 Subsequent augmented tissue penetration of chemotherapeutic agents such as melphalan explains the synergism observed when used in combination with TNF-α. Multiple clinical trials have shown that ILP, with or without other agents, including melphalan and TNF-γ, can achieve response rates of more than 70 per cent in patients suffering from stages IIIAB and IIIB (in-transit metastases) malignant melanoma.22 In a multi-centre European study of 186 patients with soft-tissue sarcomas that would have normally required limb amputation, there was a 29 per cent (54 patients) CR rate with ILP, and 82 per cent of patients (152) underwent limb salvage surgery.59 Whilst the response
Vaccines are discussed in more detail in Chapter 8, but it is noteworthy that vaccines that incorporate the expression of cytokines are showing particular promise.64–66 For example, in one study, 21 patients with relapsed or refractory neuroblastoma received up to eight subcutaneous injections of a vaccine combining lymphotactin-secreting and IL-2-secreting allogeneic neuroblastoma cells.65 Immunological responses included significant increases of circulating CD4 T cells, NK cells, IL-5 and immunoglobulin G (IgG) antibodies that bound to the immunizing cell line. Clinical responses were also observed, including two complete responses. Systemic cytokines have also been administered in conjunction with vaccination, with some evidence of success. For example, IL-2 has been reported to increase the tumour response rate to melanoma vaccines,67 and IL-12 has also been tested in pre-clinical and clinical studies on cancer vaccination, with encouraging preliminary results.68
ADOPTIVE CELL THERAPY Adoptive cell therapy involves isolation, sensitization or activation of autologous or allogeneic cells in vitro followed by re-infusion into the patient. Host immunity is stimulated in an attempt to overcome the natural tolerance to
Adoptive cell therapy 195
the established tumour. The basic principle was developed by Borberg and co-workers in 1972, who showed that the intravenous injection of lymph-node cells from mice and sheep immunized with a chemically induced tumour induced regression or inhibition of established grafts of the same tumour in syngeneic mice.69 The power of adoptive cell therapy in the allogeneic setting has been clearly demonstrated in patients with relapsed leukaemia following bone-marrow transplantation through the use of donor lymphocyte infusions (DLIs). Lymphocytes isolated from the allogeneic donor and infused into the patient are thought to respond to tumour through MHC mismatches (either major or minor MHC mismatches) and consequently to eliminate tumour through a graft-versusleukaemic (GVL) effect. However, by the same process, they can also be destructive to healthy host tissue. This unwanted side effect, graft-versus-host disease (GVHD), is associated with high rates of morbidity and mortality. T cells are the critical component, as their depletion from the DLIs abrogates both GVL and GVHD effects. In the autologous setting, a number of effector cells have been investigated for use in adoptive cell therapy. Early strategies, for example the use of lymphokine-activated killer (LAK) cells, concentrated on non-antigen-specific cytotoxicity. In recent years, however, the resurgence in interest in adoptive cell therapy has been focused on targeting specific tumour antigens. Various strategies are under development, but common to all of them are a number of obstacles that must be overcome if adoptive cell therapy is to be accepted as part of mainstream clinical practice. These approaches, as well as some the most promising innovations in the field, are discussed below.
Lymphokine-activated killer cells Animal models have shown that if lymphocytes are expanded and activated in vitro with IL-2 and then returned to the host with intravenous IL-2, they can kill tumour cells. The cells responsible for this MHC-unrestricted killing in response to IL-2 have been termed lymphokine-activated killer cells. They display a wide spectrum of cytotoxicity and have been used clinically by adoptive transfer in conjunction with high-dose infusional IL-2. However, a randomized trial of IL-2 plus LAK infusion versus IL-2 alone failed to demonstrate either a significantly higher response rate (27 per cent versus 18 per cent; two-sided p value [p2] 0.16) or significant survival benefit (36-month actuarial survival 31 per cent versus 17 per cent; p2 0.089).70 Any trend towards response rate or survival was at the expense of unacceptably severe systemic toxicity that was attributed to the IL-2 therapy, and the treatment is also expensive in terms of infrastructure and man-hours. A number of further studies failed to capitalize, and the emphasis in this field has since shifted to targeting specific tumour antigens and defining effector-cell populations more accurately.
T cells for Epstein–Barr virus-induced malignancies Epstein–Barr virus (EBV) infection can result in malignant transformation of infected B cells leading to EBV-related lymphoproliferation and is also associated with other malignancies such as Hodgkin’s lymphoma and nasopharyngeal cancer. The EBV is highly immunogenic, and specific T cells, which can recognize EBV-infected cells but do not crossreact with normal cells, can be generated with relative ease. These features make EBV-related malignancies a suitable test for adoptive immunotherapy. Encouragingly, the EBV specific T cells can have remarkable effects against EBVrelated malignancies in the post-allogeneic transplant setting71 and can persist for some time, producing long-lasting immune control.72 Such cells can also be used in the nontransplant setting and have been shown to have beneficial effects on patients with Hodgkin’s lymphoma73 and nasopharyngeal cancer.74 The encouraging successes of EBV-specific T cells in the therapy of malignant disease have promoted the application of T-cell therapy in other diseases (see below) and have also stimulated the exploration of ways to simplify this approach. One such means of simplifying therapy is to have a ‘bank’ of specific cells that can be HLA matched and used as ‘off the shelf’ therapy for appropriate individuals. Initial evaluation of this approach is promising, and further testing is on-going.75
Tumour-infiltrating lymphocytes Tumour-infiltrating lymphocytes (TILs) are cells that infiltrate into the stroma of tumours. They provided evidence of the existence of tumour antigen-specific T cells when MHC-restricted TILs were isolated from melanoma biopsy samples.76 In an early trial using adoptively transferred TILs, cloned melanoma-reactive T cells were given to patients with metastatic melanoma.77 Despite the transferred cells being highly reactive against melanoma antigens, no objective responses were seen. In murine models, prior host immunosuppression has been shown to improve the antitumour effects of adoptively transferred cells.78 Further benefits can be gained by the systemic administration of cytokines such as IL-2. Thus, in a second trial carried out by the same group, patients were pre-treated with escalating doses of non-myeloablative, lympho-depleting chemotherapy prior to receiving cloned TILs and IL-2.79 The final cohort of patients received cyclophosphamide (60 mg/kg for 2 days) and fludarabine (25 mg/m2 for 5 days) and IL-2 (720 000 U/kg every 8 hours up to 15 doses) in conjunction with cell transfer. Disappointingly, no objective responses were found and in-vivo persistence of the TILs was so poor that transferred clones could not be detected in the peripheral blood 2 weeks after administration. It was hypothesized that the absence of CD4 helper cells within the transferred CD8 cloned populations may have limited their in-vivo activation and persistence. Thus
196 Biological therapies: cytokines and adoptive cell therapy
the protocol was modified to administer populations of TILs including both CD8 and CD4 cells that were selected for being highly reactive against melanoma antigens, but were not cloned. This also had the advantage of reducing the intensity of the ex-vivo stimulation and expansion process, with the aim of adoptively transferring cells with an improved activity and proliferation potential. The results of a study involving 35 patients with metastatic melanoma who received TILs expanded in this way in conjunction with the previously described lympho-depleting chemotherapy and IL-2 was recently published.80 The patients had bulky metastatic melanoma affecting multiple organs that was refractory to standard treatments, including chemotherapy, radiation and cytokine therapies. Eighteen out of these 35 patients (51%) had an objective response, including 4 complete responses (11%). Of note was the fact that response strongly correlated with persistence of administered anti-tumour T cells, and in some cases this translated to 70–80 per cent of peripheral blood lymphocytes for longer than 4 months. Clinical responses were also associated in some cases with the development of autoimmune phenomena such as vitiligo and uveitis. This probably reflects the shared expression by normal tissues of the relevant TAA. The autoimmunity observed in this trial was of relatively minor clinical significance, but it may be a major limitation in other trials in which other TAAs are targeted if autoimmunity is more widespread or difficult to control. This trial clearly represents a significant advance in adoptive cell therapy, although it remains to be seen whether the excellent response rate demonstrated can be maintained when this therapy is transferred to other centres around the world. Furthermore, TIL therapy has a number of limitations: TILs are only present in significant numbers in a few tumour types, primarily melanoma and renal-cell carcinoma; selection relies on reactivity to, and hence the knowledge of, specific tumour antigens; and to obtain sufficient numbers of cells there must be tumour that is amenable to surgical removal to isolate the TILs. Despite these limitations, the dramatic results from these trials clearly warrant further evaluation through multi-centre clinical trials.
Genetically engineered T cells In view of the difficulties in developing TIL therapy for a broad range of tumour types, researchers are actively investigating more generally applicable adoptive cell therapy strategies. The advent of effective methods for gene transfer into T cells provides a means for creating T cells that are specific for a defined tumour antigen.81 Retroviral vectors are the most commonly used systems for gene transfer, but other approaches, including adenoviral, lentiviral and nonviral systems, are also under development. Broadly speaking, these genetically engineered T cells can be divided into two groups: T cells expressing physiological T-cell receptors (TCRs) and T cells expressing chimeric immune receptors (CIRs).
T CELLS EXPRESSING PHYSIOLOGICAL T-CELL RECEPTORS
T-cell specificity is mediated through the TCR, a heterodimer made up of an α-chain and a β-chain. The TCR chains specific for a chosen target antigen can be molecularly cloned using polymerase chain reaction (PCR)-based methods and introduced into T cells by gene transfer. A number of tumour antigens are under investigation for targeting in this way, including the melanoma tumour antigens MART1 and MAGE-A1, p53 and MDM2.81 This approach, however, has a number of potential limitations, not least the fact that the ability of the TCR to recognize a tumour cell is HLA restricted. Thus a different TCR must be used to recognize an antigen in each HLA background. In addition, tumours can escape T-cell recognition by down-regulating expression of the HLA class I molecules on their cell surface. A further concern is the risk that TCR transduction could lead to the formation of hybrid TCRs between the transduced and endogenous αchains and β-chains, leading to novel specificity with the potential for autoimmunity. Recent work has investigated engineering the transferred TCR in such a way that it is not possible for it to pair with the endogenous chains.82 T CELLS EXPRESSING CHIMERIC IMMUNE RECEPTORS
In 1989 it was demonstrated that T cells can respond to antigen independently of the TCR–MHC interaction when transduced with a CIR.83 Many different varieties of CIRs have been developed and described,84 but the basic configuration consists of an extracellular antigen-binding motif (most commonly an scFv antibody fragment) fused to a signalling moiety (most commonly the CD3ζ chain of the TCR) by means of a transmembrane domain. Antigen recognition is thus directed to native cell-surface structures and is not MHC restricted, unlike the physiological TCR.
Future development of adoptive cell therapy Currently, adoptive cell therapy is attractive and appears clinically promising. Trials are underway involving both CIR-transduced and TCR-transduced T cells in various tumour types. Figure 9.3 summarizes the principal steps involved in adoptive cell therapy, but there are numerous aspects in the process that require optimization. CELL GROWTH
The optimal dose of cells for adoptive therapy is likely to vary according to clinical target and cell type being used. However, in many instances the numbers of cells required will necessitate a period of ex-vivo expansion. This process can be expensive, labour intensive and require specialist facilities that comply with ‘Good Manufacturing Practice’ (GMP). Where peripheral blood T cells are being used for adoptive transfer, it may also be complicated by the fact
Adoptive cell therapy 197
a
e
b
f
g
c
d
Figure 9.3 Adoptive cell therapy strategy. Suitable patient with malignancy is identified (a). Lymphocytes are isolated for example by leukapheresis or from a biopsy sample (b). The cells are modified for example by transduction with a retroviral vector that encodes a TAA specific TCR (c). The transduced cells are expanded ex vivo (d). Meanwhile remaining in vivo lymphocytes (including regulatory T cells) are eliminated from the patient for example by using non-myeloablative conditioning chemotherapy (e). The modified lymphocytes are infused back into the patient (f) and cytokine IL-2 ‘support’ therapy is given to support the further in vivo expansion of the adopted modified lymphocytes (g).
that patients with advanced malignancy are likely to have depleted T-cell pools as a result of heavy pre-treatment, making isolation and expansion more difficult. Current clinical protocols for T-cell expansion involve the initial stimulation of the cells with antibodies, often followed by transduction, then subsequent IL-2-driven expansion. While relatively simple and effective in terms of the number of cells generated, these conditions may not be optimal, for example CD4:CD8 T-cell subset ratios can be altered, with greater numbers of CD8 T cells being generated, and T-cell functionality may be adversely affected by prolonged ex-vivo expansion. Certainly it appears important not to ‘exhaust’, the proliferative cell’s potential during the expansion process, as telomere length appears to correlate with invivo persistence.85 The use of artificial APCs86 and limiting time in culture may be ways to achieve this. Other areas of active interest include optimization of the cytokine growth media, and there are already encouraging data for the use of IL-15.87–89 It is also possible to introduce cytokine genes into T cells to facilitate growth.90,91 CELL SELECTION
Greater understanding of the immune system also reveals the potential for cell selection prior to expansion or gene modification. Certainly the importance of Treg cells in cancer is being increasingly recognized,92 and using methods
to deplete these cells before use in adoptive transfer may be beneficial.93 ENGRAFTMENT AND CELL SURVIVAL
Several factors limit the survival of adoptively transferred cells in vivo, including host regulatory immune forces. Importantly, there is increasing evidence to support the use of lympho-depleting chemotherapy prior to adoptive cell therapy in enabling engraftment and rapid homeostatic expansion of adoptively transferred cells. Notable success with regard to this has been achieved by Rosenberg and colleagues in the treatment of patients with metastatic melanoma using TIL therapy as described above.79,80,94 Invivo expansion and persistence may also be aided by more stringent myelosuppression than that currently used – certainly this is the case in animal models.95 ENGINEERING CELLS FOR OPTIMAL FUNCTION
Current trials of gene-modified cells use ‘first-generation’ receptors as described above. A ‘second generation’ of receptors is now being developed which incorporate important co-stimulatory molecules. The rationale behind this is that, in addition to antigen-specific signals resulting from interaction TCR with peptide – MHC complexes (signal 1), T cells receive additional signals from cell surface molecules
198 Biological therapies: cytokines and adoptive cell therapy
known as co-stimulatory molecules (signal 2). These act to enhance and fine-tune the effector functions of the T cell as well as influencing the survival of the activated T cell. Of the pathways studied to date, the CD28-B7 pathway is the best characterized. Similar to engineering the ‘signal 1’ pathway, the CD28 pathway can also be engineered.96 A further enhancement is to include two signals (both CD3ζ and CD28) in one molecule, and certainly such receptors are more active in animal models.97 Other approaches include the manipulation of the apoptotic pathways to avoid premature apoptosis of transferred cells by introducing antiapoptotic genes such as Bcl-XL or Bcl-2.98,99 The clinical application of these possible advances remains to be established, as associated with a greater signalling power is the potential for deleterious effects such as non-specific activation and increased risk of malignancy, as was seen in gene therapy trials for SCID.100 Reassuringly, to date, gene therapy using gene transfer into differentiated T cells appears to have a better safety profile than transfer into stem cells. Several gene-marking studies have been published using retroviral vectors to transduce either TILs or mature peripheral T cells. Longterm follow-up of these patients has not shown any evidence of oncogenesis.101 However, monitoring of patients for any indication of clonal expansion is clearly appropriate for all current trials of adoptive cell therapy. In addition, researchers are actively pursuing ways to enhance the safety of vectors by engineering in ‘suicide’ genes into them. The expression of these ‘suicide’ genes can be induced if required and result either in the death of genetically modified cells,102 or control of the expression of the vector.103
as macrophages or NK cells. Certainly the Fc receptors on macrophages can be engineered in similar ways,104 and it is also possible to engineer stem cells and thus multiple effector arms.105 Overall, the possibilities for developing such cellular therapies are enormous, but they remain complex and will require careful clinical testing in expert centres if the potential is to be realized.
REFERENCES ◆1
2
3 4
5
6 7
8
ENGINEERING OTHER CELLS IN THE IMMUNE SYSTEM
To date, the major focus has been on T-cell adoptive transfer. However, the same processes can, in principle, be applied to other effector cells in the immune system such
●9
10
KEY LEARNING POINTS ●
●
●
●
●
Immunotherapies in the form of cytokines or adoptive cell therapy can overcome tolerance to tumours. Interferon-α is used particularly in melanoma and renal-cell carcinoma, but controversy still surrounds the use of high-dose IFN-α in resected melanoma at high risk of relapse. Although high-dose IL-2 is toxic, it is associated with rare, but durable, complete remissions, even in patients with metastatic disease. Many other cytokines are showing promise in their development for clinical applications. Adoptive cell therapy is a complex approach, but recent successes make it an attractive target for further development.
11
12 13 14
15
Dunn GP, Bruce AT, Ikeda H, et al. Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol 2002; 3(11):991–8. Stern PL, Beverley PLC, Carroll MW. Cancer Vaccines and Immunotherapy. Cambridge: Cambridge University Press, 2000. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100(1):57–70. Preuss KD, Zwick C, Bormann C, et al. Analysis of the B-cell repertoire against antigens expressed by human neoplasms. Immunol Rev 2002; 188:43–50. Dermime S, Armstrong A, Hawkins RE, Stern PL. Cancer vaccines and immunotherapy. Br Med Bull 2002; 62:149–62. Matzinger P. The danger model: a renewed sense of self. Science 2002; 296(5566):301–5. Kelsall BL, Biron CA, Sharma O, Kaye PM. Dendritic cells at the host–pathogen interface. Nat Immunol 2002; 3(8):699–702. Kalinski P, Hilkens CM, Wierenga EA, Kapsenberg ML. T-cell priming by type-1 and type-2 polarized dendritic cells: the concept of a third signal. Immunol Today 1999; 20(12):561–7. Zhang L, Conejo-Garcia JR, Katsaros D, et al. Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N Engl J Med 2003; 348(3):203–13. Khong HT, Restifo NP. Natural selection of tumor variants in the generation of ‘tumor escape’ phenotypes. Nat Immunol 2002; 3(11):999–1005. Cabrera T, Lopez-Nevot MA, Gaforio JJ, et al. Analysis of HLA expression in human tumor tissues. Cancer Immunol Immunother 2003; 52(1):1–9. Orange JS, Fassett MS, Koopman LA, et al. Viral evasion of natural killer cells. Nat Immunol 2002; 3(11):1006–12. Von Boehmer H. Mechanisms of suppression by suppressor T cells. Nat Immunol 2005; 6(4):338–44. Curiel TJ, Coukos G, Zou L, et al. Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. Nat Med 2004; 10(9):942–9. Ichihara F, Kono K, Takahashi A, et al. Increased populations of regulatory T cells in peripheral blood and tumor-infiltrating lymphocytes in patients with gastric and esophageal cancers. Clin Cancer Res 2003; 9(12): 4404–8.
References 199
◆16
17
18 ◆19
20
21
22
23
24
25
26
●27
◆28
29
●30
Mapara MY, Sykes M. Tolerance and cancer: mechanisms of tumor evasion and strategies for breaking tolerance. J Clin Oncol 2004; 22(6):1136–51. Takeda K, Smyth MJ, Cretney E, et al. Critical role for tumor necrosis factor-related apoptosis-inducing ligand in immune surveillance against tumor development. J Exp Med 2002; 195(2):161–9. Pardoll D. T cells and tumours. Nature 2001; 411(6841):1010–12. Dranoff G. Cytokines in cancer pathogenesis and cancer therapy. Nat Rev Cancer 2004; 4(1):11–22. Meyskens FL Jr, Kopecky KJ, Taylor CW, et al. Randomized trial of adjuvant human interferon gamma versus observation in high-risk cutaneous melanoma: a Southwest Oncology Group study. J Natl Cancer Inst 1995; 87(22):1710–13. Kleeberg UR, Suciu S, Brocker EB, et al. Final results of the EORTC 18871/DKG 80-1 randomised phase III trial. rIFNalpha2b versus rIFN-gamma versus ISCADOR M versus observation after surgery in melanoma patients with either high-risk primary (thickness 3 mm) or regional lymph node metastasis. Eur J Cancer 2004; 40(3):390–402. Lienard D, Eggermont AM, Koops HS, et al. Isolated limb perfusion with tumour necrosis factor-alpha and melphalan with or without interferon-gamma for the treatment of in-transit melanoma metastases: a multicentre randomized phase II study. Melanoma Res 1999; 9(5):491–502. Decker T, Muller M, Stockinger S. The yin and yang of type I interferon activity in bacterial infection. Nat Rev Immunol 2005; 5(9):675–87. Smyth MJ, Cretney E, Kershaw MH, Hayakawa Y. Cytokines in cancer immunity and immunotherapy. Immunol Rev 2004; 202:275–93. Biron CA, Sonnenfeld G, Welsh RM. Interferon induces natural killer cell blastogenesis in vivo. J Leukoc Biol 1984; 35(1):31–7. Steimle V, Siegrist CA, Mottet A, et al. Regulation of MHC class II expression by interferon-gamma mediated by the transactivator gene CIITA. Science 1994; 265(5168):106–9. Grob JJ, Dreno B, de la Salmoniere P, et al. Randomised trial of interferon alpha-2a as adjuvant therapy in resected primary melanoma thicker than 1.5 mm without clinically detectable node metastases. French Cooperative Group on Melanoma. Lancet 1998; 351(9120):1905–10. Verma S, Quirt I, McCready D, et al. Systematic review of systemic adjuvant therapy for patients at high risk for recurrent melanoma. Cancer 2006; 106(7):1431–42. Hancock BW, Wheatley K, Harris S, et al. Adjuvant interferon in high-risk melanoma: the AIM HIGH Study – United Kingdom Coordinating Committee on Cancer Research randomized study of adjuvant low-dose extended-duration interferon Alfa-2a in high-risk resected malignant melanoma. J Clin Oncol 2004; 22(1):53–61. Kirkwood JM, Strawderman MH, Ernstoff MS, et al. Interferon alfa-2b adjuvant therapy of high-risk resected
31
32
◆33
●34
35
36
37
38
39
40
●41
42
43
cutaneous melanoma: the Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 1996; 14(1):7–17. Kirkwood JM, Ibrahim JG, Sondak VK, et al. High- and low-dose interferon alfa-2b in high-risk melanoma: first analysis of Intergroup Trial E1690/S9111/C9190. J Clin Oncol 2000; 18(12):2444–58. Kirkwood JM, Ibrahim JG, Sosman JA, et al. High-dose interferon alfa-2b significantly prolongs relapse-free and overall survival compared with the GM2-KLH/QS-21 vaccine in patients with resected stage IIB–III melanoma: results of Intergroup Trial E1694/S9512/C509801. J Clin Oncol 2001; 19(9):2370–80. Wheatley K, Ives N, Hancock B, et al. Does adjuvant interferon-alpha for high-risk melanoma provide a worthwhile benefit? A meta-analysis of the randomised trials. Cancer Treat Rev 2003; 29(4):241–52. Interferon-alpha and survival in metastatic renal carcinoma: early results of a randomised controlled trial. Medical Research Council Renal Cancer Collaborators. Lancet 1999; 353(9146):14–17. Pyrhonen S, Salminen E, Ruutu M, et al. Prospective randomized trial of interferon alfa-2a plus vinblastine versus vinblastine alone in patients with advanced renal cell cancer. J Clin Oncol 1999; 17(9):2859–67. Kriegmair M, Oberneder R, Hofstetter A. Interferon alfa and vinblastine versus medroxyprogesterone acetate in the treatment of metastatic renal cell carcinoma. Urology 1995; 45(5):758–62. Steineck G, Strander H, Carbin BE, et al. Recombinant leukocyte interferon alpha-2a and medroxyprogesterone in advanced renal cell carcinoma. A randomized trial. Acta Oncol 1990; 29(2):155–62. Coppin C, Porzsolt F, Awa A, et al. Immunotherapy for advanced renal cell cancer. Cochrane Database Syst Rev 2005(1):CD001425. Atzpodien J, Kirchner H, Illiger HJ, et al. IL-2 in combination with IFN-alpha and 5-FU versus tamoxifen in metastatic renal cell carcinoma: long-term results of a controlled randomized clinical trial. Br J Cancer 2001; 85(8):1130–6. Flanigan RC, Salmon SE, Blumenstein BA, et al. Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N Engl J Med 2001; 345(23):1655–9. Mickisch GH, Garin A, van Poppel H, et al. Radical nephrectomy plus interferon-alfa-based immunotherapy compared with interferon alfa alone in metastatic renalcell carcinoma: a randomised trial. Lancet 2001; 358(9286):966–70. Kolby L, Persson G, Franzen S, Ahren B. Randomized clinical trial of the effect of interferon alpha on survival in patients with disseminated midgut carcinoid tumours. Br J Surg 2003; 90(6):687–93. Faiss S, Pape UF, Bohmig M, et al. Prospective, randomized, multicenter trial on the antiproliferative effect of lanreotide, interferon alfa, and their combination for therapy of metastatic neuroendocrine
200 Biological therapies: cytokines and adoptive cell therapy
44
45
46
47
48
49
50 ●51
52
53
◆54
55
56
57 58
●59
gastroenteropancreatic tumors – the International Lanreotide and Interferon Alfa Study Group. J Clin Oncol 2003; 21(14):2689–96. Prior C, Oroszy S, Oberaigner W, et al. Adjunctive interferon-alpha-2c in stage IIIB/IV small-cell lung cancer: a phase III trial. Eur Respir J 1997; 10(2):392–6. Real FX, Oettgen HF, Krown SE. Kaposi’s sarcoma and the acquired immunodeficiency syndrome: treatment with high and low doses of recombinant leukocyte A interferon. J Clin Oncol 1986; 4(4):544–51. Jimenez-Cruz JF, Vera-Donoso CD, Leiva O, et al. Intravesical immunoprophylaxis in recurrent superficial bladder cancer (Stage T1): multicenter trial comparing bacille Calmette–Guerin and interferon-alpha. Urology 1997; 50(4):529–35. Morgan DA, Ruscetti FW, Gallo R. Selective in vitro growth of T lymphocytes from normal human bone marrows. Science 1976; 193(4257):1007–8. Taniguchi T, Matsui H, Fujita T, et al. Structure and expression of a cloned cDNA for human interleukin-2. Nature 1983; 302(5906):305–10. Hatakeyama M, Kono T, Kobayashi N, et al. Interaction of the IL-2 receptor with the src-family kinase p56lck: identification of novel intermolecular association. Science 1991; 252(5012):1523–8. Leonard WJ, O’Shea JJ. Jaks and STATs: biological implications. Annu Rev Immunol. 1998; 16:293–322. Fyfe G, Fisher RI, Rosenberg SA, et al. Results of treatment of 255 patients with metastatic renal cell carcinoma who received high-dose recombinant interleukin-2 therapy. J Clin Oncol 1995; 13(3):688–96. McDermott DF, Atkins MB. Application of IL-2 and other cytokines in renal cancer. Expert Opin Biol Ther 2004; 4(4):455–68. Atkins MB, Lotze MT, Dutcher JP, et al. High-dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J Clin Oncol 1999; 17(7):2105–16. Tsao H, Atkins MB, Sober AJ. Management of cutaneous melanoma. N Engl J Med 2004; 351(10):998–1012. Carswell EA, Old LJ, Kassel RL, et al. An endotoxin-induced serum factor that causes necrosis of tumors. Proc Natl Acad Sci U S A 1975; 72(9):3666–70. Pennica D, Nedwin GE, Hayflick JS, et al. Human tumour necrosis factor: precursor structure, expression and homology to lymphotoxin. Nature 1984; 312(5996):724–9. Jones AL, Selby P. Tumour necrosis factor: clinical relevance. Cancer Surv 1989; 8(4):817–36. Renard N, Lienard D, Lespagnard L, et al. Early endothelium activation and polymorphonuclear cell invasion precede specific necrosis of human melanoma and sarcoma treated by intravascular high-dose tumour necrosis factor alpha (rTNF alpha). Int J Cancer 1994; 57(5):656–63. Eggermont AM, Schraffordt Koops H, Klausner JM, et al. Isolated limb perfusion with tumor necrosis factor and melphalan for limb salvage in 186 patients with locally
60
61
62
63
64
65
66
67
68
69
70
71
72
73
advanced soft tissue extremity sarcomas. The cumulative multicenter European experience. Ann Surg 1996; 224(6):756–64; Discussion 764–5. Bonvalot S, Laplanche A, Lejeune F et al. Limb salvage with isolated perfusion for soft tissue sarcoma: could less TNF-alpha be better? Ann Oncol 2005; 16(7):1061–8. Radny P, Caroli UM, Bauer J, et al. Phase II trial of intralesional therapy with interleukin-2 in soft-tissue melanoma metastases. Br J Cancer 2003; 89(9):1620–6. Kaplan B, Moy RL. Effect of perilesional injections of PEG-interleukin-2 on basal cell carcinoma. Dermatol Surg 2000; 26(11):1037–40. Dummer R, Hassel JC, Fellenberg F, et al. Adenovirusmediated intralesional interferon-gamma gene transfer induces tumor regressions in cutaneous lymphomas. Blood 2004; 104(6):1631–8. Osanto S, Schiphorst PP, Weijl NI, et al. Vaccination of melanoma patients with an allogeneic, genetically modified interleukin 2-producing melanoma cell line. Hum Gene Ther 2000; 11(5):739–50. Rousseau RF, Haight AE, Hirschmann-Jax C, et al. Local and systemic effects of an allogeneic tumor cell vaccine combining transgenic human lymphotactin with interleukin-2 in patients with advanced or refractory neuroblastoma. Blood 2003; 101(5):1718–26. Nemunaitis J, Sterman D, Jablons D, et al. Granulocyte–macrophage colony-stimulating factor genemodified autologous tumor vaccines in non-small-cell lung cancer. J Natl Cancer Inst 2004; 96(4):326–31. Lotem M, Shiloni E, Pappo I, et al. Interleukin-2 improves tumour response to DNP-modified autologous vaccine for the treatment of metastatic malignant melanoma. Br J Cancer 2004; 90(4):773–80. Portielje JE, Kruit WH, Eerenberg AJ, et al. Subcutaneous injection of interleukin 12 induces systemic inflammatory responses in humans: implications for the use of IL-12 as vaccine adjuvant. Cancer Immunol Immunother 2005; 54(1):37–43. Borberg H, Oettgen HF, Choudry K, Beattie EJ Jr. Inhibition of established transplants of chemically induced sarcomas in syngeneic mice by lymphocytes from immunized donors. Int J Cancer 1972; 10(3):539–47. Rosenberg SA, Lotze MT, Yang JC, et al. Prospective randomized trial of high-dose interleukin-2 alone or in conjunction with lymphokine-activated killer cells for the treatment of patients with advanced cancer. J Natl Cancer Inst 1993; 85(8):622–32. Heslop HE, Brenner MK, Rooney CM. Donor T cells to treat EBV-associated lymphoma. N Engl J Med 1994; 331(10):679–80. Heslop HE, Ng CY, Li C, et al. Long-term restoration of immunity against Epstein–Barr virus infection by adoptive transfer of gene-modified virus-specific T lymphocytes. Nat Med 1996; 2(5):551–5. Bollard CM, Aguilar L, Straathof KC, et al. Cytotoxic T lymphocyte therapy for Epstein–Barr virus Hodgkin’s disease. J Exp Med 2004; 200(12):1623–33.
References 201
74 Straathof KC, Bollard CM, Popat U, et al. Treatment of nasopharyngeal carcinoma with Epstein–Barr virus-specific T lymphocytes. Blood 2005; 105(5):1898–904. 75 Haque T, Wilkie GM, Taylor C, et al. Treatment of Epstein–Barr-virus-positive post-transplantation lymphoproliferative disease with partly HLA-matched allogeneic cytotoxic T cells. Lancet 2002; 360(9331):436–42. 76 Hom SS, Topalian SL, Simonis T, et al. Common expression of melanoma tumor-associated antigens recognized by human tumor infiltrating lymphocytes: analysis by human lymphocyte antigen restriction. J Immunother 1991; 10(3):153–64. 77 Dudley ME, Wunderlich J, Nishimura MI, et al. Adoptive transfer of cloned melanoma-reactive T lymphocytes for the treatment of patients with metastatic melanoma. J Immunother 2001; 24(4):363–73. 78 North RJ. Cyclophosphamide-facilitated adoptive immunotherapy of an established tumor depends on elimination of tumor-induced suppressor T cells. J Exp Med 1982; 155(4):1063–74. 79 Dudley ME, Wunderlich JR, Yang JC, et al. A phase I study of nonmyeloablative chemotherapy and adoptive transfer of autologous tumor antigen-specific T lymphocytes in patients with metastatic melanoma. J Immunother 2002; 25(3):243–51. ●80 Rosenberg SA, Dudley ME. Cancer regression in patients with metastatic melanoma after the transfer of autologous antitumor lymphocytes. Proc Natl Acad Sci U S A 2004; 101(Suppl. 2):14639–45. ◆81 Sadelain M, Riviere I, Brentjens R. Targeting tumours with genetically enhanced T lymphocytes. Nat Rev Cancer 2003; 3(1):35–45. 82 Willemsen RA, Weijtens ME, Ronteltap C, et al. Grafting primary human T lymphocytes with cancer-specific chimeric single chain and two chain TCR. Gene Ther 2000; 7(16):1369–77. ●83 Gross G, Waks T, Eshhar Z. Expression of immunoglobulin–T-cell receptor chimeric molecules as functional receptors with antibody-type specificity. Proc Natl Acad Sci U S A 1989; 86(24):10024–8. 84 Thistlethwaite F, Mansoor W, Gilham DE, Hawkins RE. Engineering T-cells with antibody-based chimeric receptors for effective cancer therapy. Curr Opin Mol Ther 2005; 7(1):48–55. 85 Zhou J, Shen X, Huang J, et al. Telomere length of transferred lymphocytes correlates with in vivo persistence and tumor regression in melanoma patients receiving cell transfer therapy. J Immunol 2005; 175(10):7046–52. 86 Kim JV, Latouche JB, Riviere I, Sadelain M. The ABCs of artificial antigen presentation. Nat Biotechnol 2004; 22(4):403–10. 87 Brentjens RJ, Latouche JB, Santos E, et al. Eradication of systemic B-cell tumors by genetically targeted human T lymphocytes co-stimulated by CD80 and interleukin-15. Nat Med 2003; 9(3):279–86.
88 Klebanoff CA, Finkelstein SE, Surman DR, et al. IL-15 enhances the in vivo antitumor activity of tumor-reactive CD8 T cells. Proc Natl Acad Sci U S A 2004; 101(7):1969–74. 89 Teague RM, Sather BD, Sacks JA, et al. Interleukin-15 rescues tolerant CD8() T cells for use in adoptive immunotherapy of established tumors Nat Med 2006; 12(3):335–41. 90 Hsu C, Hughes MS, Zheng Z, et al. Primary human T lymphocytes engineered with a codon-optimized IL-15 gene resist cytokine withdrawal-induced apoptosis and persist long-term in the absence of exogenous cytokine. J Immunol 2005; 175,(11):7226–34. 91 Liu K, Rosenberg SA. Transduction of an IL-2 gene into human melanoma-reactive lymphocytes results in their continued growth in the absence of exogenous IL-2 and maintenance of specific antitumor activity. J Immunol 2001; 167(11):6356–65. ◆92 Zou W. Regulatory T cells, tumour immunity and immunotherapy. Nat Rev Immunol 2006; 6(4):295–307. 93 Powell DJ Jr, Parker LL, Rosenberg SA. Large-scale depletion of CD25 regulatory T cells from patient leukapheresis samples. J Immunother 2005; 28(4):403–11. ●94 Dudley ME, Wunderlich JR, Robbins PF, et al. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science 2002; 298(5594):850–4. 95 Gattinoni L, Finkelstein SE, Klebanoff CA, et al. Removal of homeostatic cytokine sinks by lymphodepletion enhances the efficacy of adoptively transferred tumorspecific CD8 T cells. J Exp Med 2005; 202(7):907–12. 96 Alvarez-Vallina L, Hawkins RE. Antigen-specific targeting of CD28-mediated T cell co-stimulation using chimeric single-chain antibody variable fragment-CD28 receptors. Eur J Immunol 1996; 26(10):2304–9. 97 Haynes NM, Trapani JA, Teng MW, et al. Single-chain antigen recognition receptors that costimulate potent rejection of established experimental tumors. Blood 2002; 100(9):3155–63. 98 Eaton D, Gilham DE, O’Neill A, Hawkins RE. Retroviral transduction of human peripheral blood lymphocytes with Bcl-X(L) promotes in vitro lymphocyte survival in proapoptotic conditions. Gene Ther 2002; 9(8):527–35. 99 Charo J, Finkelstein SE, Grewal N, et al. Bcl-2 overexpression enhances tumor-specific T-cell survival. Cancer Res 2005; 65(5):2001–8. 100 Hacein-Bey-Abina S, von Kalle C, Schmidt M, et al. A serious adverse event after successful gene therapy for X-linked severe combined immunodeficiency. N Engl J Med 2003; 348(3):255–6. 101 Recchia A, Bonini C, Magnani Z, et al. Retroviral vector integration deregulates gene expression but has no consequence on the biology and function of transplanted T cells. Proc Natl Acad Sci U S A 2006; 103(5):1457–62. 102 Straathof KC, Pule MA, Yotnda P, et al. An inducible caspase 9 safety switch for T-cell therapy. Blood 2005; 105(11):4247–54.
202 Biological therapies: cytokines and adoptive cell therapy
103 Alvarez-Vallina L, Agha-Mohammadi S, Hawkins RE, Russell SJ. Pharmacological control of antigen responsiveness in genetically modified T lymphocytes. J Immunol 1997; 159(12):5889–95. 104 Biglari A, Southgate TD, Fairbairn LJ, Gilham DE. Human monocytes expressing a CEA-specific chimeric CD64
receptor specifically target CEA-expressing tumour cells in vitro and in vivo. Gene Ther 2006; 13(7):602–10. 105 Hege KM, Cooke KS, Finer MH, et al. Systemic T cellindependent tumor immunity after transplantation of universal receptor-modified bone marrow into SCID mice. J Exp Med 1996; 184(6):2261–9.
10 Radioimmunotherapy TIM ILLIDGE AND MIKE BAYNE
The principles of radioimmunotherapy Radioimmunotherapy for lymphoma Radioimmunotherapy in leukaemias
203 205 215
THE PRINCIPLES OF RADIOIMMUNOTHERAPY Introduction The use of monoclonal antibody (mAb) in routine clinical practice is now well established and has led to significant improvements in outcome for patients with haematological malignancies as well as in a wide range of other malignancies, including breast and bowel cancer.1,2 Although the single-agent activity of most mAb has been modest, when used in combination with other anti-tumour therapies, an additive or synergistic effect has been seen.3 The combination of mAb and multi-agent chemotherapy in a number of different histologies of non-Hodgkin’s lymphoma (NHL) has led not only to highly impressive increases in response rates, but also to improved relapsefree survival and even overall survival.3,4 Radioimmunotherapy (RIT) is the administration of therapeutic radioisotopes targeted to tumour through chemical conjugation to mAb or mAb-derived constructs. Monoclonal antibodies were initially regarded simply as direct carriers for the radioisotope that delivers systemically targeted cytotoxic radiation to areas of disease with relative sparing of normal tissue. It is, however, becoming clear that during RIT for lymphoma, the mAb effector mechanisms may also play an important additional role in killing tumour cells. The nature of RIT determines that its efficacy depends on a number of factors, including the properties of the targeted antigen (specificity, density, availability, shedding and heterogeneity of expression), the tumour (volume, degree of vascularization, blood flow and permeability), the mAb (specificity, immunoreactivity, stability and affinity) and the
Radioimmunotherapy for solid tumours Clinical perspectives for the next five years References
217 220 221
properties of the chosen radioisotope (emission characteristics, half-life and availability).5 A wide variety of different mAbs, delivery schedules, radioisotopes and doses of radioactivity have been used in RIT, and the most impressive results have been achieved in the treatment of NHL.6 Two drugs, namely yttrium-90ibritumomab tiuxetan and iodine-131-tositumomab, have been approved by the US Food and Drug Administration (FDA). Yttrium-90-ibritumomab tiuxetan is also approved for use within the European Union (EU). The use of RIT in leukaemias and solid tumours is less well developed, but the emerging data look highly encouraging, suggesting that RIT may also play a useful role in these malignancies. This chapter focuses on the current clinical indications for radioimmunoconjugates in NHL and provides an overview of the current clinical data related to the use RIT in other malignancies.
Antigen targeting The use of radiation therapy in the treatment of malignancy is well established and often highly effective for localized disease. The disseminated nature of many malignancies frequently prevents effective delivery of external-beam radiotherapy. The systemic delivery of radiation through RIT enables irradiation of tumour cells throughout the body whilst minimizing the dose to normal tissues and is a logical strategy that is showing great promise for the treatment of the exquisitely radiosensitive lymphomas and leukaemias. The effective delivery of RIT requires the selection of a suitable tumour-antigen target.
204 Radioimmunotherapy
Tumour-specific antigens (Box 10.1) would be the ideal targets for RIT, but such a degree of specificity is unusual. In practice, tumour-associated antigens, expressed abundantly on tumour cells as well as on some normal tissues, represent the majority of potential targets. As most NHLs are of B-cell origin, the pan-B-cell antigens such as human leucocyte antigen DR (HLA-DR), CD19, CD20, CD22, CD37, CD52 and MHC II have been extensively evaluated as targets for RIT.1,2,7–10 Box 10.1 shows the antigen characteristics that are considered ideal for RIT. Optimum antigen characteristics will vary depending on the radioisotope employed. For 131I, internalization can be problematic, as it results in rapid dehalogenation and loss of the 131I from the target cell. For 90Y, internalization is less of a concern, as radiometals usually remain trapped in the target cell despite internalization. When alpha-emitters and radioisotopes employing Bremsstrahlung for cytotoxicity are used, internalization is an advantage, as the very short path length of these emissions requires that the isotope comes into close contact with the nuclear material of the targeted cell in order to be effective.
Box 10.1 The characteristics considered ideal in an antigen target for radioimmunotherapy Tumour-cell specific Highly expressed on tumour cells No tendency to mutation Not secreted or shed Not rapidly modulated on antibody binding Critical for target-cell survival Not expressed on critical or non-renewable host cells.
From initial investigations, the CD20 antigen emerged as having many of the characteristics thought to be important for a good tumour target, and therefore targeting this antigen has dominated the clinical RIT of lymphoma.11 The CD20 antigen is a transmembrane phosphoprotein that is expressed on mature B lymphocytes but not on pre-B lymphocytes. The CD20 complex does not internalize or shed from the cell surface, and when bound by mAb may initiate signal transduction that triggers apoptosis through a caspase-dependent pathway.12 CD20 is highly expressed on more than 90 per cent of B-cell lymphomas but is not expressed on stem cells or plasma cells, and consequently following radiolabelled anti-CD20 mAb RIT, the depleted B-cell pool may be replenished and the transient lymphopenia has no significant clinical impact. Although most clinical RIT work has been targeted against the CD20 antigen, other B-cell antigens, such as CD22, are still being actively investigated.13 In leukaemia, CD33, CD45 and CD66 have been the most extensively investigated antigen targets for RIT. CD45 is found on nearly all leucocytes, including lymphoid and myeloid precursors. More than 90 per cent of acute myeloid
leukaemia (AML) biopsy samples and most acute lymphoblastic leukaemia (ALL) biopsy samples show CD45 antigen expression, where cell-surface antigen expression averages 200 000 copies per cell.14 Importantly in the context of RIT, the antigen does not internalize after mAb binding. The antigens targeted in RIT for solid tumours include CEA and HER/2-neu and are discussed in the solid tumour section below.
Radioisotopes used in radioimmunotherapy The physical characteristics considered important for a radioisotope in RIT include half-life, type of radioactive emissions (α, β or γ) and ionization path length. Particle energy and mean path length in tissue are important determinants of therapeutic efficacy. The emission profile of the radioisotope determines not only its suitability for therapy, but also the toxicological profile of the radiopharmaceutical. The majority of clinical trials to date have used either 131 I or 90Y because of their favourable emission characteristics, availability and radiochemistry, permitting reliable and stable attachment to mAbs. Iodine-131 has the advantages of a long history of successful use in the management of thyroid cancer and a well-documented safety profile. It is readily available, inexpensive, easily conjugated and emits both β-particles with a path length of 0.8 mm and penetrating γ-emissions. The γ-photons enable uncomplicated imaging using a gamma camera for dosimetry purposes, but also result in a significant non-targeted normal-tissue radiation dose, as well as radiation protection issues for visitors and medical/nursing staff. Yttrium-90 offers a number of theoretical advantages over 131I, although the radioisotopes have not been directly compared in clinical trials. Yttrium-90 is a pure β-emitter that produces higher-energy radiation (2.3 MeV versus 0.6 MeV) at a longer path length than 131I (5.3 mm versus 0.8 mm). Due to the path length of the β-emissions, cellular damage occurs in both the targeted lymphoma cells and neighbouring cells, as illustrated in Table 10.1 (see p. 9). The greater path length of 90Y would be expected to enhance this ‘cross-fire’ effect and could therefore be potentially advantageous in treating bulky, poorly vascularized tumours with heterogeneous antigen expression.5 It is likely, however, to increase the normal-tissue dose when targeting microscopic disease, for which the shorter β-particle path length of 131I may be preferable. The physical half-life of 90Y is 64 hours and approximates to the biological half-life of murine mAbs. The absence of penetrating γ-emissions enables delivery in the outpatient setting.15 Additionally, if a cell internalizes 90Y, it is likely to be retained within the cell.13 In contrast, once 131I-conjugates are internalized by a cell, there is rapid dehalogenation of the free iodide and subsequent excretion of the iodinated products from the cell, reducing desired tumour-absorbed radiation dose and increasing normal-tissue radiation exposure.16 The major disadvantages of 90Y relate to its
Radioimmunotherapy for lymphoma 205
data related to the treatment of leukaemia appear extremely promising.18,19 In practice, the choice of the optimal radioisotope for RIT remains controversial, with proponents advocating the relative merits of 131I, 90Y, 186Rh, 67Cu, and α-emitters such as 211 At.15 Comparative studies are difficult to conduct and scientifically robust randomized human trials have not been performed. The ideal properties of a radioisotope for RIT remain unclear and it is likely that the optimal radioisotope for a particular situation will depend upon the bulk and type of tumour being targeted. An important area of potential future research will be to define the optimal radioisotope, or cocktail of isotopes, required for different tumour sizes.
RADIOIMMUNOTHERAPY FOR LYMPHOMA Clinical experience
Figure 10.1 The cross-fire effect. Mab conjugated to a radioisotope binds to tumour-associated antigen and delivers ionizing radiation not only to the cell to which it binds but also to cells within a radius defined by the path length of the radioactive emission. For 131I this equates to 30–50 cell diameters and for 90Y this equates to 200–300 cell diameters.
greater expense, relatively limited availability and requirement for chelation radiochemistry making radiolabelling a more complex procedure. Yttrium-90 does not emit γphotons and there is therefore a need to use indium-111 as a surrogate to obtain images for biodistribution and dosimetry studies. Rhenium-186 and copper-67 are both β-emitters with physical and chemical properties that make them attractive alternatives to either 131I or 90Y. Nevertheless, their current limited availability means that these radioisotopes have had limited clinical use.17 Alpha-emitters produce a helium nucleus particle of very high energy but with a very short path length. The high linear energy transfer (LET) radiation of α-emitters may be lethal to cells with a single collision; however, the very short path length means that the isotope must be adjacent to, or internalized by, the cell to be effective and is likely to have little or no ‘cross-fire’ effect. The suitability of α-emitters therefore appears limited to readily accessible tumours such as leukaemia cells confined to the blood or bone marrow. The short half-life of α-emitters (e.g. 211At 7 hours and 213Bi, 45 minutes) complicates the radiopharmaceutical preparation, meaning that such radioisotopes are likely to require generation on the same site as delivery in the clinic. Despite this logistical hurdle, early clinical
Clinical RIT trials in NHL differ in terms of eligibility criteria, mAb and radioisotope used, dose, number of treatments, doses of unlabelled mAb pre-infused or co-infused, and the biodistribution or dosimetry estimations required for administration of a therapeutic dose of radiolabelled mAb. Nevertheless, virtually all clinical studies performed to date have shown high response rates of NHL to RIT and have been well reviewed.5,20–24 DeNardo et al. initially pioneered RIT for NHL with 131Ianti-HLA-DR mAb (Lym-1).8 Escalating fractionated doses of 131I-Lym-1 ranging from 1480 mBq/m2 to 3700 mBq/m2 (40–100 mCi/m2) resulted in an overall response rate (ORR) of 52 per cent in 21 treatment courses administered to 20 patients, with seven patients (33 per cent) achieving complete response (CR) and four patients (19 per cent) achieving partial response (PR).8 Goldenberg et al. used an 131I-LL2 (anti-CD22) mAb to treat a variety of B-cell lymphomas. In one of their trials, four out of 17 patients achieved objective remission, including one CR.25 In another trial, 90Y-LL2 was administered to seven patients with B-cell lymphomas, two of whom achieved PR.25 By far the greatest depth of experience in RIT involves the use of 90Y-ibritumomab tiuxetan and 131I-tositumomab in relapsed B-cell lymphomas, where impressive responses have been observed in all of the clinical trials reported to date. Ibritumomab and tositumomab bind to different epitopes of the same CD20 antigen.26 The radioisotopes 131I and 90 Y also have important differences in their emission characteristics. Table 10.1 compares the main characteristics of 131 I-tositumomab and 90Y-ibritumomab tiuxetan. YTTRIUM-90-IBRITUMOMAB TIUXETAN
Yttrium-90-ibritumomab tiuxetan consists of a monoclonal IgG1 kappa anti-CD20 mAb, the murine parent immunoglobulin of rituximab, covalently attached to a metal chelator molecule (tiuxetan; an isothiocyanatobenzyl derivative of
206 Radioimmunotherapy
the polyaminocarboxylic acid DTPA), which stabilizes the mAb–isotope complex for delivery to the lymphoma site.27 The biological half-life elimination of 90Y-ibritumomab tiuxetan is 30 hours. More than 90 per cent of the β-radiation is absorbed within a 5 mm proximity (corresponding to a diameter of 100–200 cells) of the radiation source. This facilitates highly targeted delivery of radiation without the need for patient isolation or shielding.15 The tiuxetan chelator molecule provides a stable link between the mAb and the radioisotope, and therefore free isotope clearance rates are minimal and predictable, with 7.3 3.2 per cent of the radiolabelled activity being excreted in the urine over 7 days.28 Consequently, 90Y-ibritumomab tiuxetan may be administered on an outpatient basis. Figure 10.2 outlines the 90 Y-ibritumomab tiuxetan therapeutic regimen. Four clinical trials, including three phase I/II and one randomized study, formed the basis of the FDA submission for 90Y-ibritumomab tiuxetan. The initial phase I/II study demonstrated that myelotoxicity was dose limiting.28 The maximum tolerated dose (MTD) was identified as 14.8 MBq/kg (0.4 mCi/kg), to a maximum of 1184 MBq (32 mCi) for patients with a baseline platelet count of 150 109/L and 11.1 MBq/kg (0.3 mCi/kg) for patients with baseline platelet counts 150 109/L but 100 109/L. In this study, a high ORR for the intent-totreat population (n 51) was seen at 67 per cent (CR 26 per cent; PR 41per cent); for low-grade disease (n 34) 82 per cent (CR 26 per cent; PR 56 per cent); for intermediategrade disease (n 14) 43 per cent. A phase II study of patients with mild thrombocytopenia (baseline platelet count 100–150 109/L) was conducted using the reduced dose of 11.1 MBq/kg (0.3 mCi/Kg). The ORR was 83 per cent (CR 37 per cent, CRu [unconfirmed] 6.7 per cent, PR 40 per cent). Kaplan–Meier estimated median time to progression (TTP) was 9.4 months (range 1.7–24.6 months). In responders, Kaplan–Meier estimated median TTP was 12.6 months (range 4.9–24.6 months). Toxicity was primarily haematologic, transient and reversible. The incidences of grade 4 neutropenia, thrombocytopenia, and anaemia were 33 per cent, 13 per cent and 3 per cent respectively. The conclusions from this study were that reduced-dose ibritumomab tiuxetan is safe and welltolerated and has significant clinical activity in patients with mild thrombocytopenia.29 A further single-arm phase II study of 90Y-ibritumomab tiuxetan was undertaken to examine its efficacy in a group with rituximab refractory disease.30 Fifty-four heavily
pre-treated patients with follicular lymphoma were recruited who were refractory to or progressed after rituximab. The trial showed an ORR of 74 per cent and a CR rate of 15 per cent, despite a median of four prior therapies and 73 per cent of patients having bulky disease (5 cm diameter). Kaplan–Meier-estimated DR (duration of remission) was 6.4 months, with a TTP of 6.8 months in all patients and 8.7 months in responders. The randomized phase III trial included 143 patients with relapsed or refractory low-grade follicular or transformed NHL and compared efficacy of a single dose of 14.8 MBq/kg (0.4 mCi/kg) 90Y-ibritumomab tiuxetan with rituximab (375 mg/m2 once weekly for 4 weeks).31 Response rates were significantly higher in the 90Y-ibritumomab tiuxetan arm, with an ORR of 80 per cent versus 56 per cent (p 0.002) and a CR rate of 30 per cent versus 16 per cent (p 0.004). Subgroup analysis revealed a superior benefit for patients with follicular histology, with an ORR of 86 per cent versus 55 per cent (p 0.001) and a significant (p 0.04) improvement in TTP for this subgroup. However, there was no difference in the overall TTP in the two treatment groups, but patients treated with 90 Y-ibritumomab tiuxetan showed a trend towards longer median DR (14.2 months versus 12.1 months) and more often achieved responses lasting longer than 6 months (64 per cent versus 47 per cent). A recent retrospective analysis suggests that treatment with 90Y-ibritumomab tiuxetan is associated with higher response rates and longer DR when used earlier in the therapy schedule.32 An integrated analysis of 211 patients treated in clinical trials compared the efficacy and safety of 90Y-ibritumomab tiuxetan in patients with one prior therapy (n 63) and in patients who had received two or more prior therapies (n 148). Patients receiving 90 Y-ibritumomab tiuxetan as second-line therapy had greater ORR (86 per cent versus 72 per cent; p 0.051) and CR/CRu rates (49 per cent versus 28 per cent; p 0.004) and a significantly longer median TTP (12.6 months versus 7.9 months; p 0.038). Amongst the CR/CRu patients, the median TTP (23.9 months versus 15.6 months; p 0.0442) and median DR (22.8 months versus 14.6 months; p 0.0429) were both significantly increased in those with only one prior therapy (n 53). A large European intergroup study of 90Y-ibritumomab tiuxetan therapy of previously untreated follicular lymphoma has now completed accrual, with more than 400 patients recruited. Patients were treated initially with Day 7
Day 0 Dosimetric dose (250 mg rituximab (1 hr), 185 MBq (5 mCi) 90Y-ibritumomab tiuxetan (1.6 mg) (10 min)
Imaging step (Required in USA) X3
Dosimetric dose used to determine individual pharmacokinetics
Figure 10.2 Treatment regimen for 90Y-ibritumomab tiuxetan.
Dosimetric dose (250 mg rituximab (1 hr), 14.8 MBq/Kg (11.1 MBq/Kg platelet counts <150 109/L) 90 Y-ibritumomab tiuxetan (3.2 mg) to deliver desired (10 min)
Radioimmunotherapy for lymphoma 207
chemotherapy (the physician’s choice) and then randomized to 90Y-ibritumomab tiuxetan or no further treatment. The data from this study may provide information for determining whether 90Y-ibritumomab tiuxetan has a potential beneficial role after primary chemotherapy. There were, however, relatively small numbers of patients treated in the latter part of the study who received rituximab in combination with chemotherapy regimens, which is now a widely adopted approach. Therefore questions regarding the additional role of 90Y-ibritumomab tiuxetan RIT after full-course rituximab–chemotherapy combinations will probably not be answered by this study. Shipley and colleagues performed an evaluation of the efficacy of short-duration R-CHOP (rituximabcyclophosphamide, doxorubicin, vincristine, prednisolone) followed by 14.8 MBq (0.4 mCi) 90Y-ibritumomab tiuxetan as a first-line treatment of patients with follicular lymphoma.33 The study was a multi-centre, phase II trial that recruited 42 patients, of which 39 completed the entire planned therapy. Of the 40 patients who received chemotherapy, 12 (30 per cent) were found to have a CR/Cru and 28 (70 per cent) showed a PR. After 90Y-ibritumomab tiuxetan RIT, 26 patients (67 per cent) showed a CR/Cru, 12 patients (31 per cent) showed a PR and 1 patient (3 per cent) showed disease progression. The group showed that treatment with 90Y-ibritumomab tiuxetan after chemotherapy/ rituximab increased the CR rate from 30 per cent to 66 per cent. Actuarial PFS (Progression Free Survival) after one year was 97 per cent (38 patients at risk) and after two years was 85 per cent (15 patients at risk). Clinical responses have also been observed for transformed follicular and relapsed diffuse large B-cell lymphoma (DLBCL) when treated with 90Y-ibritumomab tiuxetan. An initial phase I/II study reported a response rate of 58 per cent, with a 33 per cent CR rate in a group of just 12 patients who had relapsed following two previous chemotherapy regimens that included CHOP.31 A prospective, single-arm, openlabel, non-randomized, multi-centre, phase II trial was therefore undertaken to evaluate the efficacy and safety of 90 Y-ibritumomab tiuxetan in patients over the age of 60 years with relapsed or primary refractory DLBCL not suitable for autologous stem-cell transplantation (ASCT). Patients were divided into two groups of those previously treated with chemotherapy alone (Group A, n 76) and those previously treated with chemotherapy and rituximab (Group B, n 28).34 All patients received a single dose of 14.8 MBq/kg (0.4 mCi/kg) of 90Y-ibritumomab tiuxetan up to a maximum dose of 1184 MBq/kg (32 mCi). In total, 103 patients were evaluated for treatment efficacy, and 104 for safety. An ORR of 44 per cent was observed in the entire study population. In Group A, the ORR was over 50 per cent. In Group B, in which 37 per cent of patients were refractory to rituximab-CHOP, the ORR was 19 per cent. Adverse events (AEs), with the exception of haematological AEs, were generally mild (grade 1/2) and the incidence of severe infection was low, with only 7 per cent of patients hospitalized for infection during the study. The results of
this study were encouraging, and clinical trials are now underway in the USA and at an advanced stage of development in the EU to integrate 90Y-ibritumomab tiuxetan into a front-line treatment for DLBCL alongside rituximab– chemotherapy schedules. IODINE-131-TOSITUMOMAB
Tositumomab was the first mAb to be produced against a B-cell antigen35 and thus was originally called B1 and, like ibritumomab, has activity against lymphoma. Pre-clinical studies have been helpful in enhancing our understanding of the nature of the anti-CD20 mAb activity and these investigations have revealed that ibritumomab/rituximab recruits C1q and C3 complement factors well and is efficient at recruiting antibody effectors as part of ADCC (antibody dependent directed cytotoxicity), whereas in contrast, tositumomab appears more able directly to signal apoptosis.36 Unlike 90Y-ibritumomab tiuxetan, the pharmacokinetics of 131I-tositumomab vary widely between patients and so require individualized patient dosimetry. This involves delivery of a tracer dose of the radioimmunoconjugate followed by the therapeutic dose 7 to 14 days later. Whole-body gamma-camera imaging is performed three times over the week following the trace-labelled infusion to calculate the whole-body half-time and the dose required for the therapeutic infusion to deliver a 65–75 cGy whole-body dose (WBD) – usually 3700–5550 MBq (100–150 mCi).37 Dose adjustments to 65 cGy were made for a baseline platelet count of 100 000–150 000/mm3 and for obesity. The therapeutic regimen is outlined in Figure 10.3. Kaminski and colleagues initially conducted a series of trials at the University of Michigan using 131I-tositumomab, for the treatment of relapsed follicular lymphoma.38,39 In a pivotal study, 60 extensively pre-treated patients were given a single administration of 131I-tositumomab.7 Disease responses were compared to their previous responses to chemotherapy for follicular or transformed follicular lymphoma. A PR or CR was observed in 39 patients (65 per cent) after 131I-tositumomab, compared with 17 patients (28 per cent) after their last qualifying chemotherapy (LQC) (p 0.001). Iodine-131-tositumomab therapy was shown to provide greatly superior PFS compared to LQC. Since 1990, more than 800 patients with ‘low-grade’ and transformed lymphoma have been treated with 131 I-tositumomab. Long-term follow-up data were presented at the American Society of Hematology (ASH) Annual Meeting 2002; 250 of these patients had indicated response rates of 56 per cent, with 30 per cent of patients achieving a CR. Perhaps the most impressive fact is that 70 per cent of the patients who achieved a CR are alive and remain in CR at up to 7.8 years, with a median follow-up of almost 4 years.40 Impressive response rates have also been seen in patients refractory to rituximab and subsequently treated with 131I-tositumomab. Horning and colleagues used
208 Radioimmunotherapy
Thyroprotection: Day-1 continuing until day 14 post therapeutic dose
Day 0
Day 7–14
Dosimetric dose (450 mg tositumomab (1 hr), 185 MBq (5 mCi) 131 I-tositumomab (35 mg) (20 min)
Dosimetric dose (450 mg tositumomab (1 hr), 185 MBq (5 mCi) 131 I-tositumomab (35 mg) to deliver desired cGy WBD (10 min)
Imaging step (Required in USA) X3
Dosimetric dose used to determine individual pharmacokinetics
Administered activity determined by imaging
Figure 10.3 Treatment regimen for 131I-tositumomab. 131
I-tositumomab to treat 40 patients with low-grade NHL, 72 per cent of whom had received four or more previous lines of therapy and 60 per cent of whom had failed to respond to rituximab.41 An ORR of 68 per cent with a CR rate of 30 per cent was noted and a median DR of 14.7 months reported. Of the 12 complete responders, 9 remained in CR at the time of presentation, with a range of 12–26 months. More recently, an analysis including 230 patients treated with 131I-tositumomab was made. Independently assessed durable CRs were noted with similar frequency in patients with rituximab-refractory disease (28 per cent) and rituximab-naïve patients, all of whom had chemotherapy refractory disease (23 per cent). With a median followup of 4.6 years, 75 per cent of patients with durable CR continue in complete remission.40 Kaminski et al. have shown highly promising results in the front-line treatment of previously untreated low-grade follicular lymphoma using 131I-tositumomab.21 An encouraging ORR of 95 per cent was seen, with 75 per cent achieving CR. Polymerase chain reaction (PCR) was used to detect rearrangement of the BCL2 gene, which revealed molecular responses in 80 per cent of assessable patients who had a clinical CR. The most recent update included 76 patients with a median follow-up of 5.1 years. The actuarial 5-year PFS for all patients was 59 per cent, with a median PFS of 6.1 years. Haematological toxicity was moderate, with no patient requiring transfusion or granulocyte colonystimulating factor (G-CSF),21,42,43 although 48 out of 76 (63 per cent) patients developed detectable human antimouse antibody (HAMA) responses after a single course of treatment with 131I-tositumomab. More recently, Press and colleagues conducted a phase II trial in untreated follicular lymphoma that consisted of six CHOP cycles followed 4–8 weeks later by 131I-tositumomab.44 A cohort of 90 previously untreated eligible patients with advanced stage follicular lymphoma tolerated the treatment well. Reversible myelosuppression was the main AE, and was more severe during CHOP chemotherapy than after RIT. The ORR over the entire treatment regimen was 90 per cent, including 67 per cent CR and 23 per cent PR. Of the 47 fully evaluable patients who achieved less than a CR with CHOP, 27 (57 per cent) improved their remission status after pre-dosed tositumomab and 131I-tositumomab. With a median follow-up of 2.3 years, the 2-year PFS was estimated to be 81 per cent, with a 2-year overall survival of
97 per cent. Having established the feasibility and the efficacy of this approach a randomized phase III trial is currently being undertaken to compare this approach of chemotherapy followed by RIT against immuno-chemotherapy with six cycles of CHOP-R. The contribution of targeted radiation to the overall responses seen in RIT has been addressed, with two randomized studies comparing the radioimmunoconjugates 90 Y-ibritumomab tiuxetan and 131I-tositumomab with the unlabelled mAbs.20,31 Both studies have shown greatly superior clinical responses of RIT over the unlabelled mAb. The 90Y-ibritumomab tiuxetan versus rituximab is described above. The second study compared treatment outcomes for unlabelled tositumomab (pre-dose) and 131Itositumomab to an equivalent total amount of unlabelled tositumomab. This study involved 78 patients with refractory/relapsed low-grade NHL.20 The investigators reported an ORR of 55 per cent versus 19 per cent (p 0.002), with a CR of 33 per cent versus 8 per cent (p 0.012) in the 131 I-tositumomab versus unlabelled tositumomab groups respectively. The median duration of the ORR was not reached for 131I-tositumomab, compared with 28.1 months for unlabelled tositumomab. The median duration of CR was not reached in either arm, and the median TTP was 6.3 versus 5.5 months (p 0.031), respectively. Although haematological toxicity was more severe and non-haematological AEs were more frequent after 131I-tositumomab than after tositumomab alone, there were no serious infectious or bleeding complications. The frequency of developing HAMA was similar in the two arms, at 27 per cent (131I-tositumomab group) versus 19 per cent (tositumomab alone group). This study demonstrated that although unlabelled tositumomab showed single-agent activity, the conjugation of 131I to tositumomab significantly enhanced the therapeutic efficacy.20
Toxicity and safety of radioimmunotherapy ADVERSE EFFECTS OF
90
Y-IBRITUMOMAB TIUXETAN
Myelosuppression is the dose-limiting toxicity of RIT for both 90Y-ibritumomab tiuxetan and 131I-tositumomab. Due to concerns that significant involvement of the bone marrow by lymphoma will result in excessive delivery of
Radioimmunotherapy for lymphoma 209
radiation to the radiosensitive bone marrow, and so unacceptable myelotoxicity, it has become standard practice to restrict RIT to patients with less than 25 per cent bonemarrow involvement. This limitation has been applied to all clinical trials of both 90Y-ibritumomab tiuxetan and 131 I-tositumomab. An analysis of all patients treated in 90 Y-ibritumomab tiuxetan trials (n 261) indicated that 28 per cent experienced grade 4 neutropenia and 8 per cent grade 4 thrombocytopenia.45 No significant differences in the incidence of haematological and non-haematological grade 3–4 AEs were observed in patients aged 65 years or older as compared to younger patients. In a further analysis, safety data from the four clinical trials that informed the FDA application were reviewed retrospectively in an integrated analysis encompassing 349 patients, of whom 345 (99 per cent) completed treatment with 90 Y-ibritumomab.46 Although 80 per cent of patients reported non-haematological AEs, these were generally mild to moderate in severity, with asthenia, nausea and chills being the most common events that were considered to have probable or possible association with the treatment. Only 11 per cent (39 patients) of all patients experienced grade 3–4 non-haematological toxicity. For 90Y-ibritumomab tiuxetan, grade 1–2 and 3–4 thrombocytopenia occurred in 37 per cent and 63 per cent of patients, respectively. Of the patients with grade 3–4 thrombocytopenia, 87 per cent recovered to 50 109/L by week 12 following therapy. Neutropenia grade 3–4 was observed in 30 per cent of patients, with 90 per cent recovering to 100 109/L within 12 weeks post-treatment. For patients who received G-CSF support, the median duration of neutropenia was reduced from 27 to 19 days. Grade 3 and 4 anaemia developed in 13 per cent and 4 per cent of patients, respectively. Of all patients, 22 per cent required platelet transfusions and 20 per cent required red blood cell transfusions. Yttrium-90-ibritumomab tiuxetan is dosed according to the patient’s body weight and baseline platelet counts. For patients with platelet counts 150 000/mm3, 14.8 MBq/kg (0.4 mCi/kg) body weight is given, up to a maximum permissible dose of 1184 MBq (32.4 mCi). For patients with platelet counts of 100–149 109/L, 90Y-ibritumomab tiuxetan is dosed at 11 MBq/kg (0.3 mCi/kg). The incidence of severe thrombocytopenia and neutropenia correlated significantly with degree of bone-marrow involvement and platelet counts at baseline, underscoring the importance of excluding patients with 25 per cent bone-marrow infiltration and inadequate bone-marrow reserve. Patients who had more than two prior chemotherapies were twice as likely to develop grade 4 thrombocytopenia, whereas number of prior chemotherapies did not correlate with longer median duration of neutropenia, thrombocytopenia and anaemia. Total number of B cells and levels of IgM declined after treatment but recovered after 6–9 months. The median T-cell counts and levels of IgG and IgA remained stable following treatment with 90Y-ibritumomab tiuxetan. Most importantly, treatment with 90Y-ibritumomab tiuxetan was not associated with an excess rate of infections. In fact,
the incidence of infectious complications was low, with upper respiratory and urinary tract infections occurring in 5 per cent and 7 per cent of patients, respectively. Only 8 per cent of all patients received antibiotic therapy during the treatment period. Despite concerns about the potential of an increased risk of radiation-induced secondary haematological malignancies, the observed rate of secondary myelodysplastic syndromes (MDS)/AML in this same analysis was less than 1 per cent (5/348), which is comparable with a similar patient population treated with alkylating agents.46 ADVERSE EFFECTS OF 131I-TOSITUMOMAB RADIOIMMUNOTHERAPY
Short-term non-haematological AEs with 131I-tositumomab administration are generally mild and typically include fatigue, nausea, fever, vomiting, pruritus and rash, which usually respond well to anti-histamines. Follow-up examinations include weekly blood and platelet counts following RIT until haematological recovery occurs. Iodine-131-tositumomab is susceptible to dehalogenation and therefore ‘cold’ iodine is given pre-administration, starting 72 hours prior to the dosimetric dose and again at 14 days after the therapeutic dose in order to block the thyroid from radioactive iodine uptake. Despite these attempts to prevent thyroid uptake of 131I, hypothyroidism appears to be one of the most consistent long-term adverse effects after 131I-mAb treatment. However, hypothyroidism can be easily managed with thyroid hormone replacement once detected, and screening is an important component of follow-up. Zelenetz reviewed the multi-centre RIT trials using 131I-tositumomab in NHL patients and reported that elevated thyroid-stimulating hormone (TSH) was observed in 5 out of 59 patients in the phase I study.47 However, after a myeloablative dose of 131I-tositumomab, elevated TSH was observed in 59 per cent of patients.45 Human anti-mouse antibody reactions appear to be substantially lower in previously treated NHL patients compared to the rates experienced with solid-tumour RIT.5 The incidence of HAMA in a heavily pre-treated population of follicular lymphoma patients is less than 10 per cent after 131 I-tositumomab treatment. However, as noted above, HAMA response rates of 63 per cent have been reported when 131I-tositumomab is used in the first-line treatment of follicular lymphoma.21 The HAMA responses are frequently subclinical and have little immediate clinical consequence, but they may have an impact on future murine antibody delivery and are important if repeated or fractionated treatment is to be considered. The impact of administration of the 131I-tositumomab therapeutic regimen on circulating CD20-positive cells was assessed in two clinical studies,21,39 one conducted in chemotherapy-naïve patients and one in heavily pre-treated patients. The assessment of circulating lymphocytes did not distinguish normal from malignant cells. Consequently, recovery of normal B-cell function was not directly assessed. Lymphocyte recovery began at approximately 12 weeks
210 Radioimmunotherapy
following treatment. Among patients who had CD20positive cell counts recorded at baseline and at six months, eight of 58 (14 per cent) chemotherapy-naïve patients and six of 19 (32 per cent) heavily pre-treated patients had CD20positive cell counts below normal limits at 6 months. There was no consistent effect of the 131I-tositumomab therapeutic regimen on post-treatment serum IgG, IgA or IgM levels. Secondary malignancy following 131I-tositumomab has fortunately proved to be rare.21,47 This concurs with the experience described above with 90Y-ibritumomab and is supported by a study with large numbers and substantial follow-up. In this study, 1071 131I-tositumomab-treated patients were assessed for treatment-related MDS/AML.48 Among these patients, 995 with low-grade and transformed low-grade NHL had been previously treated with a median of three therapies (range 1–13 therapies) prior to RIT. As part of their initial therapy for follicular lymphoma, 76 patients received RIT. For the previously treated patients, the median follow-up from the diagnosis of NHL and RIT was 6 years and 2 years respectively; for the patients who received RIT as their initial therapy, the corresponding median follow-up times were 5.6 years and 4.6 years, respectively. Of the 995 previously treated patients, 35 (3.5 per cent) cases of treatment-related MDS/AML were reported and 13 cases were confirmed to have developed MDS/AML following RIT. This incidence was found to be consistent with that expected on the basis of patients’ prior exposure to chemotherapy. With a median follow-up approaching 5 years, no case of treatment-related MDS/AML has been reported in the 76 patients receiving 131I-tositumomab as their initial therapy.48 Data are emerging to suggest that therapy can be administered after 131I-tositumomab and 90Y-ibritumomab tiuxetan.49,50 Chemotherapy or ASCT after prior therapy with 90Y-ibritumomab tiuxetan appears feasible and reasonably well tolerated. The toxicity with subsequent therapy seems similar to that in patients not treated with 90Y-ibritumomab tiuxetan. Despite initial concerns, preliminary data suggest that patients can be safely treated after prior bone-marrow transplant or stem-cell support as long as a significant dose reduction is made. For 90Y-ibritumomab tiuxetan, a dose of 11.1 MBq/kg is suggested,51 and for 131I-tositumomab a WBD of 65 cGy has proven to be feasible.52,53 It should be emphasized that the delivery of either RIT drug post-ASCT is outside the current licensed indication. However, RIT followed by ASCT is being increasingly investigated, with the aim of total disease eradication.
Myeloablative radioimmunotherapy in non-Hodgkin’s lymphoma Myeloablative RIT has the theoretical advantages of delivering higher doses of radiation and overcoming the dose-limiting myelotoxicity with autologous stem-cell and bone-marrow rescue.54 Press and colleagues in Seattle
pioneered the myeloablative approach, demonstrating the feasibility of using high-activity doses of 131I-anti-B-cell mAbs, MB-1 (anti-CD37) or anti-B1 (anti-CD20) with either ASCT (autologous stem cell transplantation) or PBSCT (peripheral stem cell transplantation).55 Initially, the group assessed the biodistribution, toxicity and efficacy of high doses of radiolabelled anti-B1 in 43 patients with Bcell lymphoma. The dose-limiting toxicity was found to be cardiopulmonary and the lung MTD to be 27 Gy. Patients with a favourable biodistribution were eligible for RIT with 131 I-tositumomab according to a phase 1 dose-escalation protocol. Twenty-four patients had a favourable biodistribution, and 19 received therapeutic infusions of 8.7–28.7 GBq (234–777 mCi) of 131I-mAbs (58–1168 mg) followed by ASCT, resulting in CR in 16, PR in 2 and a MR (25–50 per cent tumour regression) in 1. Of these patients, 9 have remained in continuous CR for 3–53 months. Toxic effects included myelosuppression, nausea, infections and two episodes of cardiopulmonary toxicity, and were moderate in patients treated with doses of 131I- mAbs that delivered 27.25 Gy to normal organs. For patients with a favourable biodistribution (tumour burdens 500 cm3 and without massive splenomegaly), the CR rate was 84 per cent, with a PR rate of 11 per cent. These remissions appeared durable and the median DR exceeded 11 months at the time of publication. Finally, the anti-CD20 mAb (B1) was considered to be superior to the anti-CD37 (MB-1) because a favourable biodistribution was achieved with 2.5 mg/kg, as compared with 10 mg/kg for the anti-CD37. In the subsequent study, Press et al. evaluated the combination of high-dose 131I-tositumomab, etoposide and cyclophosphamide in conjunction with ASCT in 38 patients with NHL (26 low grade, 12 aggressive).55 Of the 37 evaluable patients, 33 (89 per cent) were alive and 29 (78 per cent) were progression free after a median follow-up of 1.5 years. Toxicities included grade 4 myelosuppression in all patients, grade 2/3 nausea in 26 (70 per cent), pulmonary infiltrate in four, and grade 3 veno-occlusive disease in two patients. These results confirmed the feasibility of delivering high-dose RIT in combination with high-dose chemotherapy in an ASCT setting for NHL. More recently, the long-term follow-up of the 29 patients treated in phase I and phase II trials was reported.45 Twenty-four of the 29 patients are alive and 15 are progression free after a median follow-up of 37 months. At 5 years, overall survival and PFS are projected at 68 per cent and 42 per cent respectively. None of the surviving patients has objective impairment of performance status or cardiac function. Late toxicities have been uncommon, except for elevation of the TSH levels seen in 60 per cent of the subjects. The Seattle Transplant Group performed a multivariable comparison of 125 consecutive patients with follicular lymphoma treated either with high-dose RIT using 131 I-tositumomab (n 27) or conventional high-dose therapy (C-HDT) (n 98) and ASCT.56 The groups were similar, although more patients treated with high-dose RIT had an elevated pre-transplantation level of lactate dehydrogenase
Radioimmunotherapy for lymphoma 211
(41 per cent versus 20 per cent, p 0.03) and elevated International Prognostic Score (41 per cent versus 19 per cent, p 0.02). Patients treated with high-dose RIT received individualized therapeutic doses of 131I-tositumomab (median 19.7 GBq – 531 mCi) to deliver 17–31 Gy (median 27 Gy) to critical organs. Patients treated with C-HDT received TBI (total body irradiation) plus chemotherapy (70 per cent) or chemotherapy alone (30 per cent). Patients treated with high-dose RIT experienced improved overall survival and PFS compared with patients treated with C-HDT. The estimated 5-year overall survival and PFS were 67 per cent and 48 per cent respectively for high-dose RIT, and 53 per cent and 29 per cent respectively for C-HDT. The 100-day treatment-related mortality was 3.7 per cent in the high-dose RIT group and 11 per cent in the C-HDT group. The probability of secondary MDS/AML was estimated to be 0.076 at 8 years in the high-dose RIT group and 0.086 at 7 years in the C-HDT group. These data suggest that high-dose RIT may improve outcomes when compared to C-HDT in patients with relapsed follicular lymphoma. Nademanee et al. recently studied the feasibility of including high-dose 90Y-ibritumomab with high-dose etoposide (VP-16) 40–60 mg/kg (day4) and cyclophosphamide 100 mg/kg (day2) followed by ASCT.57 This phase I/II trial of high-dose 90Y-ibritumomab tiuxetan in combination with etoposide and cyclophosphamide was conducted in 31 patients with CD20-positive NHL. Patients underwent dosimetry (day21) with (185 MBq (5 mCi0)) 111In-ibritumomab tiuxetan following 250 mg/m2 rituximab, followed a week later by 90Y-ibritumomab tiuxetan to deliver a target
dose of 1000 cGy to highest normal organ. Bone-marrow biopsy was later performed (day7) to estimate radiation dose, and stem cells were re-infused when the radiation dose was estimated to be 5 cGy. The median 90Y-ibritumomab tiuxetan dose was 2649.2 MBq (range 1354.2–3885 MBq; 71.6 mCi, range 36.6–105 mCi). At a median follow-up of 22 months, the 2-year estimated overall survival and relapse-free survival rates were 92 per cent and 78 per cent respectively. The authors concluded that high-dose 90Yibritumomab tiuxetan can be combined safely with highdose etoposide and cyclophosphamide without an increase in transplant-related toxicity or delayed engraftment.
Key issues in radioimmunotherapy of non-Hodgkin’s lymphoma IODINE-131-TOSITUMOMAB (BEXXAR®) VERSUS 90 Y-IBRITUMOMAB TIUXETAN (ZEVALIN®)
No comparative clinical trial has ever been performed for 90 Y-ibritumomab tiuxetan and 131I-tositumomab and it is now unlikely that such a trial will ever be undertaken. However, it appears reasonable to surmise from the published results that the two drugs have very similar response rates and response durations. The characteristics of the two agents are summarized in Table 10.1 and the clinical trial results are summarized in Table 10.2. The integration of 90Y-ibritumomab tiuxetan and 131Itositumomab into routine clinical practice therefore seems more likely to depend on the cost and convenience of each
Table 10.1 Characteristics of 131I-tositumomab (Bexxar) and 90Y-ibritumomab tiuxetan (Zevalin) 131
90
Bexxar Tositumomab (anti-B1)-murine Simple 131 I γ and β 0.606 MeV 0.8 mm 8 days 0.364 MeV 4–6-day inpatient stay in shielded room Renal (variable) Thyroid (pre-blocked with potassium iodide) Tositumomab (450 mg/patient)
Zevalin Ibritumomab (2B8)-murine More complex 90 Y β only 2.293 MeV 5.3 mm 2.6 days None Outpatient
75 cGy whole-body dose Dosimetric dose obligatory
14.8 MBq/kg (0.4 mCi/kg) Dosimetric dose not required Dose reduction for thrombocytopenia
I-tositumomab
US trade name Monoclonal antibody Chelation Isotope Isotope emissions β-energy β-particle path length Isotope half-life γ-energy Radiation protection measures Isotope excretion Normal tissue uptake Pre-dose (unlabelled monoclonal antibody) Dose
Y-ibritumomab tiuxetan
Limited Bone Rituximab (250 mg/m2) 2
Table 10.2 Summary of the clinical trials that led to US FDA registration of the two licensed radioimmunocojugates 90Y Ibritumomomab and 131I Tositumomab Radioimmuno-conjugate 90
90
Y ibritumomab
Y ibritumomab
Study design and eligibility
Histology
Patients (n)
ORR %
CRR/CRRu (%)
Median DR (months)*
Median TTP (months)†
Phase I–II 0.2, 0.3 or 0.4 mCi/kg LG or FL and 2 prior CT regimens or an anthracycline IG or MCL in first or subsequent relapse Platelets 150,000/mm3 No prior anti-CD20 therapy
All LG
51 34
67 82
26 26
11.7 –
–
IG§
14
43
29
–
–
FL
33
85¶
57¶
All*
73
¶
80
¶
FL*
55
Phase III versus rituximab LG, FL or T NHL Platelets 150,000/mm3 No prior rituximab
Ref
9.3
28
–
–
34
14.2
10.6
31
86¶
58¶
18.5
15.0
90
Y ibritumomab
Phase II, FL, platelets 150,000/mm3 Prior rituximab and no response or TTP 6 months
FL
54
74¶
15¶
6.4
6.8
30
90
Y ibritumomab
Phase II,LG,FL or T NHL Platelets 100,000–149,000/mm3 No prior anti-CD20 Therapy
All FL
38 25
83¶ 92¶
47¶ –
11.5 11.7
9.4 10.8
29
Phase I–II, LG, IG or T NHL Platelets 100,000/mm3 No prior anti-CD20 therapy
All
59
71
34
–
–
LG T IG
28 14 17
86 79 41
46 50 0
– – –
– – –
All
47
57
32
9.9
–
LG
37
57
27
8.2
–
T
10
60
50
12.1
–
131
I tositumomab
131
I tositumomab
Phase II, LG or T NHL Prior anthracyclinecontaining CT regimen Progression 1 year after last CT regimen No prior monoclonal antibody or RIT Platelets 100,000/mm3
131
I tositumomab
Phase III, LG or T NHL 2 prior qualifying CT regimens No prior monoclonal antibody or RIT Platelets 100,000/mm3
All
60
65
20
6.5
–
LG T
36 23
81 39
– –
– –
– –
131
Phase II, FL or T NHL Progressed after at least one course of rituximab Platelets 100,000/mm3
All
40
65
38
–
10.4**
131
Phase II, FL, SLL, LPL or T NHL First or second recurrence and subsequently progressed No prior monoclonal antibody or RIT Platelets 100,000/mm3
All LG
41 34
76 76
49 53
15 13
9.6** –
FL
29
79
59
28
–
7
71
28
41
–
42
55
33
14.7
I tositumomab
I tositumomab
131
I tositumomab
Phase III versus tositumomab LG or T NHL Prior anthracycline-containing CT regimen progression 1 year after CT regimen Platelets 100,000/mm3 No prior monoclonal antibody or RIT
T ALL
#
6.3
7
20
* Responding patients only † All patients § Including two patients with transformed NHL ¶ By International Workshop Response Criteria # Only patients that received RIT **Progression-free survival CRR: Complete response rate; CCRu: Complete response rate unconfirmed; CT: Chemotherapy; DR: Duration of response; NHL: Non-Hodgkin lymphoma; IG: Intermediate grade; LG: Low grade ; (IG and LG refer to the previously used working formulation and is as described in the original publication) FL: Follicular lymphoma LPL: Lymphoplasmacytic lymphoma; MCL: Mantle cell lymphoma; ORR: Overall response rate; RIT: Radioimmunotherapy; SLL: Small lymphocytic lymphoma; T: Transformed follicular lymphoma; TTP: Time to progression
214 Radioimmunotherapy
therapy rather than on any perceived differences in clinical efficacy. Iodine-131-tositumomab remains licensed in the USA only, and if an EU license is eventually gained, the radiation protection issues, with the necessity for 5–6 days inpatient stay for patients (within the EU) receiving 131 I-tositumomab, may influence clinicians in favour of 90 Y-ibritumomab tiuxetan on health economic grounds. The removal of the dosimetric dose, within the EU, further simplifies the delivery of 90Y-ibritumomab tiuxetan, making the whole regimen very easy for the patient. RADIOIMMUNOTHERAPY INDUCES DURABLE REMISSIONS
Perhaps the most impressive finding consistently to emerge from the maturing data using RIT in follicular lymphoma and other low-grade B-cell malignancies is the remarkable duration of response enjoyed by some patients. This durability of treatment response that was initially seen in patients treated with 131I-tositumomab40 has now also been seen with 90Y-ibritumomab tiuxetan.58 In all of the RIT studies, around 70 per cent of patients who achieve a CR remain in remission for years.32,40 Further, some patients treated in the early studies have now been in remission for more than 5 years. An analysis of long-term responders underscores this potential for 90Y-ibritumomab tiuxetan to achieve durable remissions, with an observed median DR approaching 2 years and responses greater than 6 years being observed in some patients.32 An analysis of prognostic factors has confirmed that this remarkable durability of response following RIT is unlikely to be accounted for by patient selection, as most of these durable remissions have been achieved in heavily chemotherapy pre-treated and chemo-refractory patients with validated poor prognostic factors such as extensive prior therapy (one to nine regimens), bulky disease, high LDH (lactate dehydrogenase) and extra-nodal disease.59 IS THERE A ROLE FOR DOSIMETRY IN RADIOIMMUNOTHERAPY?
Dosimetry studies are routinely carried out prior to the therapeutic administration of radioimmunoconjugates. By delivering a tracer dose of radioimmunoconjugate and undertaking a series of quantitative gamma-camera scans over the course of a week, the distribution and an estimation of the absorbed dose to the tumour and to critical normal tissues from the subsequent therapeutic dose may be calculated. Such dosimetric studies enable identification of those patients who have ‘favourable’ biodistributions, whereby the dose to tumour is predicted to be substantially greater than that to normal tissues.38 Dosimetric studies during the initial development of RIT demonstrated an advantage from delivering a dose of unlabelled mAb prior to the therapeutic RIT administration. This pre-dose is considered to block ‘non-specific’ binding sites (e.g. circulating and splenic B cells), thereby prolonging the circulating half-life of the radiolabelled mAb and resulting in
increased tumour retention of the labelled mAb. The predose is of particular importance in the presence of splenomegaly, which may act as an antigen sink preventing delivery of the radioimmunoconjugate to the tumour. The dose to the tumour and normal organs is calculated knowing the quality of radiation, the equilibrium constant and the specific absorbed fraction to the target using the Medical Internal Radiation Dose (MIRD) Committee guidelines. Due to the complexities of these calculations and to the varied and often contradictory results reported, the clinical importance of dosimetry in the RIT of NHL currently remains controversial. Whilst some investigators regard dosimetry as an essential component of RIT practice, others argue that it is unnecessary.42,49,50,52,60–62 Proponents both for and against dosimetry have valid arguments and both may be correct, with the requirement for dosimetry likely to be dependent on the radioisotopes and the mAb targeting used and the questions being asked by the dosimetry. There are at least two important issues with regard to dosimetry: namely, can dosimetry predict tumour response and can it predict normal-tissue toxicity? To date, despite the high response rates seen in lymphoma RIT, clinical dosimetry studies have failed to show a consistent dose–response relationship, although some investigators have found correlation.5 One of the factors that is very likely to complicate the analysis and may underlie the controversy is the inability of dosimetry to measure the tumoricidal capability of some mAbs. For 131I-tositumomab, gamma-camera imaging is an essential part of the regimen and critically required for patient-specific dosing. Calculated whole-body absorbed dose has been found to correlate well with the degree of subsequent myelosuppression. Due to wide variation in the pharmacokinetics of the iodinated mAb, the administered dose required to deliver the 75 cGy whole-body absorbed dose usually prescribed may vary by as much as a factor of 5 between patients. Contrasting with the experience with 131I-tositumomab, dosimetric studies have failed to demonstrate a consistent correlation between the estimated bone-marrow dose and toxicity for the Zevalin® treatment regime. Due to the residualizing properties of the radiometal 90Y, there is far less variation in the pharmacokinetics between patients, and it appears that 90 Y-ibritumomab tiuxetan can be safely prescribed according to body weight and platelet count.53 As a result, imaging studies are not required within the EU and do not form a part of licensing in the majority of EU countries.46 Despite this, two imaging scans following the rituximab pre-dose continue to form part of the approved schedule within the USA. INTEGRATION OF RADIOIMMUNOTHERAPY INTO TREATMENT ALGORITHMS FOR NON-HODGKIN’S LYMPHOMA
The high response rates and durable remissions achieved with either 90Y-ibritumomab or 131I-tositumomab make
Radioimmunotherapy in leukaemias 215
single-agent RIT an attractive treatment option for many patients with relapsed follicular lymphoma. Furthermore, the impressive duration of responses (DRs) seen after achieving a complete response (CR) are achieved with a treatment that is completed within a week, is very well tolerated, even by older patients and that has minimal non-haematological toxicity and easily manageable myelotoxicity. The introduction of RIT has some parallels with the introduction of rituximab into clinical practice in the late 1990s. There is no doubting that RIT drugs are highly active, but considerable uncertainty remains as to when and how best to integrate RIT into clinical practice, even within the licensed indication of relapsed low-grade (USA) or relapsed rituximab failure or refractory follicular lymphoma (within the EU). In the education session on follicular lymphoma at the American Society of Hematologists (2004), an algorithm was suggested for follicular lymphoma for which RIT was recommended upon first relapse after a rituximab– chemotherapy combination (Fig. 10.4).63 Currently, this seems a very reasonable strategy, as it does not exclude transplantation options at a later date, especially if, as recommended, progenitor stem-cell collection is performed at the time of the initial remission. Although data are emerging to suggest RIT can be given after transplantation (see above), this is inevitably at lower doses. Further clinical trials will need to be performed further to define the role of non-myeloablative RIT in the treatment of other NHLs.
Despite the excellent results reported with myeloablative RIT followed by stem-cell transplantation, this approach has yet to be widely adopted because of the serious radiation protection issues inherent in managing such high doses of radionuclide. In particular, for 131I with its penetrating gamma emission and long half-life, such high doses are unlikely to be delivered except in very few research institutions. With increasing experience with radioimmunoconjugates and the availability of chimeric and humanized mAbs that rarely induce HAMA(human anti-mouse antibody)/HACA (human anti-chimeric antibody) responses, dose escalation may be achieved in the future without the requirement of stem-cell transplantation, by fractionation of a course of RIT.
RADIOIMMUNOTHERAPY IN LEUKAEMIAS Radioimmunotherapy for leukaemia is less well developed than that for lymphoma, and there are no currently licensed or widely available agents. However, promising results have been seen in two settings: as part of a conditioning regimen prior to haematopoietic stem-cell transplant and in the treatment of low-volume or minimal residual disease. The majority of leukaemic cells are found in the bone marrow, and bone marrow or haematopoietic stem-cell transplantation is a component of treatment for most patients with acute leukaemia. Total body radiation (TBI) as a technique for delivering cytotoxic radiation to all the bone marrow and other sites harbouring leukaemic
Figure 10.4 A proposed treatment algorithm for follicular lymphoma patients after first relapse following a rituximab-chemotherapy combination.63
216 Radioimmunotherapy
cells is a well-established part of many conditioning regimens, but is limited by the inevitable toxicity of treating so much normal tissue. Radioimmunotherapy using beta emitters such as 131I and 90Y may be used to deliver radiotherapy selectively to the bone marrow and has enabled the bone-marrow dose to be escalated far beyond that possible with TBI. The relatively long path length of the beta emitters ensures crossfire within the bone marrow and a relatively homogeneous delivery of cytotoxic radiation to both mAb-bound and non-mAb-bound cells. By contrast, when treating minimal residual disease where leukaemic cells may be widely spaced, this crossfire effect will result in additional dose to normal tissues. Due to the very short range of alpha particles, RIT with alpha emitters results in very little crossfire but effective single cell killing, making this approach attractive for the treatment of leukaemia in the context of minimal residual disease.
Radioimmunotherapy for conditioning prior to haematopoietic stem-cell transplant In RIT for leukaemia, the CD33, CD45 and CD66 antigens are the most widely tested. CD33 is a cell-surface glycoprotein found on myeloid leukaemia cells. HuM195 is a humanized anti-CD33 mAb, which has been used, in its native form as well as in a radiolabelled form, to treat leukaemias. HuM195 as a single agent was observed to have minimal but measurable anti-leukaemic activity in patients with relapsed or refractory AML, although its activity is confined to patients with low-burden disease.64 Iodine131-M195 and 131I-HuM195 have been combined with busulfan/cyclophosphamide as conditioning for allogeneic bone-marrow transplantation.65 Radioimmunotherapy with 131I-M195 or 131I-HuM195 was performed on 31 patients with relapsed/refractory AML (n 16), accelerated/ myeloblastic chronic myeloid leukemia (CML) (n 14) or advanced MDS (n 1). Subjects received 4.5–16.2 GBq (122–437 mCi) plus busulfan (16 mg/kg) and cyclophosphamide (90–120 mg/kg) followed by infusion of relateddonor bone marrow (27 first ASCT; 4 second ASCT). Hyperbilirubinaemia was the most common extramedullary toxicity, occurring in 69 per cent of patients during the first 28 days after ASCT. Gamma-camera imaging showed targeting of the radioisotope to the bone marrow, liver and spleen, with absorbed radiation doses to the marrow of 272–1470 cGy. The median survival was 4.9 months (range 0.3–90 months). Following bone-marrow transplantation, three patients with relapsed AML remained, at the time of publication in CR for more than 59, 87 and 90 months. There have also been very encouraging results using 131 I-BC8 mAb (anti-CD45) in combination with cyclophosphamide and TBI as a marrow transplant conditioning regimen for acute leukaemia.66 Results from a phase I study of 131I anti-CD45 combined with 120 mg/kg cyclophosphamide and 12 Gy TBI in HLA-matched related
transplants for ALL in first remission were recently reported.66 A biodistribution dose of 0.5 mg/kg 131I-BC8 (murine anti-CD45) mAb was given to 44 patients with advanced MDS/ALL. Thirty-seven patients (84 per cent) had favourable mAb biodistributions, with a higher estimated absorbed radiation dose to marrow and spleen than to normal organs. Of these patients, 34 received a therapeutic dose of 131I-BC8 mAb labelled with 2.8–22.6 GBq (76–612 mCi) of 131I to deliver estimated radiation absorbed doses to liver (normal organ receiving the highest dose) of between 3.5 Gy (level 1) to the hightest level 12.25 Gy (level 6) which was given in addition to cyclophosphamide and TBI. The MTD was level 5 (delivering 10.5 Gy to liver), with grade III/IV mucositis in two patients treated at level 6. The authors concluded that 131I-anti-CD45 mAb can safely deliver substantial supplemental doses of radiation to bone marrow (approximately 24 Gy) and spleen (approximately 50 Gy) when combined with conventional cyclophosphamide/TBI. Following the success of the phase I study, two phase II studies have been performed to evaluate the short-term toxicity of the anti-CD45 radiopharmaceutical.65 In the first trial, 17 patients with advanced AML were treated with 131I-anti-CD45, TBI (12 Gy) and cyclophosphamide (120 mg/kg) to obtain a red-marrow dose of 21–22 Gy. Results of the trial showed LFS in 42 per cent of patients, 29 per cent relapsed and 29 per cent died of non-relapse disease. In the second study, 46 AML patients in CR1 with either intermediate-risk or high-risk cytogenetic features were treated with the same treatment as before but with busulfan (16 mg/kg) instead of cyclophosphamide. Red-marrow dose was 11 Gy and average LFS was 60 per cent. A further phase I trial has been initiated by the same group using only 2 Gy TBI in combination with fludarabine.67,68 Thus far, the mean red-marrow dose achieved with this treatment has been 24 Gy and of the 11 patients recruited, 8 are in remission at the time of publication. Rhenium-188 is an attractive radioisotope for RIT as it has a 16.9-hour half-life and emits a high energy (2.2 MeV β-particle and a 155 keV κ-photon) facilitating therapy and imaging. Bunjes et al. used a 188Re-anti-CD66 mAb to intensify the conditioning regimen prior to stem-cell transplantation to treat 36 patients with high-risk AML and MDS.69 As with the 131I-labelled and 90Y-labelled mAbs targeting CD45 and CD33, this strategy proved successful in delivering significant radiation to the bone marrow without excessive toxicity. Randomized trials are now required to demonstrate the benefit of this intensified conditioning with RIT.
Radioimmunotherapy for MRD (minimal residual disease) using γ-emitters The high linear energy transfer and short path length of α-particles offer the potential for selective killing of individual
Radioimmunotherapy for solid tumours 217
tumour cells. Bismuth-213 has a half-life of 45.6 minutes and emits an α-particle of 8 MeV in addition to a 440 keV photon that enables detailed dosimetry and biodistribution studies. Bismuth-213 has been conjugated with HuM195 and following promising pre-clinical studies demonstrating that 50 per cent of target cells are killed when only two bismuth atoms are bound to the cell surface,70 a phase I dose-escalation study was reported. In patients with refractory and relapsed myeloid leukaemias it was demonstrated that patient imaging of 213Bi labelled to HuM195 was possible and may be used to derive pharmacokinetics and dosimetry data.71 Bismuth-213-HuM195 was then studied in 18 refractory AML or CML patients.18 Doses of 10.36–37.0 MBq/kg of 213Bi-HuM195 were delivered. All 17 evaluable patients developed myelosuppression, with a median time to recovery of 22 days, and almost all of the activity was observed to localize rapidly and remain in areas of leukaemic involvement, including the bone marrow, liver and spleen. The reduced whole-body and increased target-organ doses (bone marrow, liver, spleen) produced absorbed dose ratios approximately 1000 times higher for 213Bi-HuM195 than those for β-emitting anti-CD33 mAb constructs. Fourteen (93 per cent) of the 15 evaluable patients had reductions in circulating blasts, and 14 (78 per cent) of 18 patients had reductions in the percentage of bone-marrow blasts. No complete remissions were observed; however, even with optimum antibody targeting it would not have been possible to deliver two 213Bi atoms to every leukaemic cell given the tumour burden of the patients involved. This study demonstrated the safety, feasibility and therapeutic effects of 213BiHuM195 in treating leukaemia and offers proof of concept for the systemic delivery of targeted α-particle therapy in humans. Clearly this approach needs to be investigated in the context of minimal residual disease, where it may offer real promise in the treatment not only of leukaemia but also of solid tumours with persisting micrometastatic disease.
RADIOIMMUNOTHERAPY FOR SOLID TUMOURS The results for RIT in solid tumours are currently some way short of the highly promising, durable response rates seen for the haematological malignancies. This is likely to be due to the limited penetration of poorly vascularized heterogeneous tumours by mAbs and the inherent radioresistance of solid tumours relative to the exquisite radiosensitivity of lymphomas. A wide range of antibodies and radionuclides has been used, delivering a spectrum of tumour doses. Most of the trials have been in extensively pre-treated patients with metastatic chemo-refractory disease, usually involving a single intravenous injection of a non-myeloablative dose of RIT. Generally, response rates have been disappointing and short lived, often with no objective responses.72
Intact IgG is a large molecule (150 kD) that is designed to remain in the bloodstream for many weeks. This is an advantage when acting as part of the immune system to remove pathogens, but when used to deliver radioactivity to tumours it may result in poor targeting and dose-limiting myelosuppression. Even when treating the usually highly vascular lymphomas, with non-myeloablative RIT the mean dose to tumour nodules is usually less than 10 Gy.42 This is a dose that may be effective against lymphoma but that is unlikely to produce a durable response in an epithelial tumour. These penetration problems may be overcome in brain tumours by intra-lesional administration and in ovarian tumours by intra-peritoneal administration. Alternatively, the dose-limiting haematological toxicity can be overcome by supporting the treatment with an ASCT. Pre-targeted RIT is a potentially more widely applicable approach that may improve the tumour:blood ratio and is showing some promise in both the laboratory and the clinic.
Pre-targeted radioimunotherapy In an attempt to improve RIT results for solid tumours, a number of pre-targeting strategies have been developed. The principle of pre-targeting is that the mAb is delivered first, allowing time for optimal penetration and tumour binding, before the radioisotope is delivered with a carrier that has a high affinity for the mAb construct and has higher permeation, diffusion and clearance rates than the mAb.73 This enables the radionuclide to localise more rapidly to the tumour, with rapid clearance of excess from the circulation improving the tumour:blood dose ratio. Pretargeting most commonly takes advantage of the highaffinity streptavidin–biotin interaction. Two-step and three-step protocols are under investigation. The threestep protocol includes a step in which excess circulating mAb construct is cleared from the circulation by the administration of a scavenging molecule prior to the administration of the radionuclide. Following administration, any unbound radionuclide is rapidly cleared through the kidneys, so minimizing normal-tissue exposure. Animal studies predicted a 25-fold advantage when using 90 Y-biotin pre-targeted with streptavidin-conjugated antiEpCAM IgG compared to directly radiolabelled mAb.74 Phase I trials in humans have repeatedly demonstrated a greater then ten-fold improvement in the tumour-tomarrow ratio using pre-targeting, suggesting that clinically relevant doses of 50 Gy to tumour should be achievable. In a recent dosimetry study of nine patients given 90 111 Y/ In-labelled biotin pre-targeted by anti-TAG-72 StAv fusion protein, an average tumour dose of 0.289 Gy/mCi was achieved, equating to a tumour dose of 50 Gy if the phase II study dose of 110 mCi could be delivered.75 At these doses, renal toxicity may become dose limiting, and further investigation of pre-targeting strategies needs to take this into account.
218 Radioimmunotherapy
Clinical experience with pre-targeting strategies is now accumulating, and promising results have been reported in the treatment of glioblastoma multiforme (GBM), with a reported survival time of 33.5 months greatly exceeding that of a matched control group (p 0.01).76 Additional promising results have also been reported using bispecific anti-carcinoembryonic antigen (CEA) for pre-targeting CEA-expressing small-cell lung cancer (SCLC)77 and medullary carcinoma of the thyroid.78 Despite enhanced tumour uptake, dosimetry studies in SCLC suggest the tumour doses in most cases will still be below 25 Gy, indicating that even with pre-targeting, RIT is unlikely to be effective in the treatment of bulky epithelial tumours but may have a role in the treatment of microscopic residual disease or when used in combination with conventional chemotherapy.77 Radioimmunotherapy has yet to enter mainstream practice in the treatment of solid tumours, and a brief review of the major tumour sites is given.
Central nervous system tumours Both primary brain tumours and leptomeningeal disease have been treated with RIT. For primary tumours, RIT has been studied as therapy for small-volume residual disease and in patients with recurrent disease after conventional treatment. Various antibodies, radionuclides and routes of administration have been utilized.79–83 Tumour response can be difficult to assess radiographically; disease-free survival, disease-free interval and quality of life assessments are commonly used as markers of disease response. Gliomas account for 70 per cent of all primary brain tumours in adults. The high-grade tumours, especially GBM, are aggressive tumours with a poor median survival of just 12 months. Standard treatment involves surgery and external beam radiotherapy and occasionally chemotherapy. The treatment of gliomas with systemic administration of RIT, although feasible, is fraught with difficulties, as described by Hopkins and colleagues.84 Radioimmunoconjugates penetrate the blood/brain barrier poorly; even in situations in which the blood/brain barrier is disrupted, the accumulation of radioisotope is low. The natural history and location of the tumours have led to the development of intrathecal administration with direct tumour injection or injection into the surgical bed. This approach has proved to be possible and has shown promising results, especially for patients with lower grade tumours and small-volume disease. Bone-marrow toxicity is not dose limiting. Cerebral oedema is seen frequently but is usually transient. Riva et al. reported the results of a phase I and a phase II trial administering 131I-labelled mAbs to the tumour bed of 111 patients, with 20 in the phase I and 91 in the phase II studies.81 Murine IgG mAbs BC-2 and BC-4 were used, which react with two separate epitopes on the intracellular and stromal glycoprotein tenascin. Tenascin is expressed at high levels in gliomas, especially GBM. Ninety-one
patients had GBM; the other pathologies were eight oligodendrogliomas (7 anaplastic oligodendrogliomas), two grade II astrocytomas and ten anaplastic grade III astrocytomas. All patients received conventional surgery and radiotherapy; 54 also received chemotherapy. Patients with both a new diagnosis and relapsed disease were recruited. Toxicity was minimal and repeated treatments were possible. The MTD was 2590 MBq, with larger doses causing marked cerebral oedema. Scintigraphy was used to demonstrate tumour uptake, which was high and sustainable. Response varied with histology and inversely with tumour grade. Tumour volume also affected response rates in the patients with GBM. Of the 74 phase II GBM patients, one patient showed a CR, nine a PR, 10 stable disease (SD) and 23 no evidence of disease (NED – patients who had RIT when the tumour mass was very small and not detectable but who remained free of disease from the start of treatment). The overall response rate (CR+PR+NED+SD) was 66.6 per cent in those with a tumour volume of 2 cm2, with a median survival of 25 months. For those with bulky lesions, the overall response rate was 17.8 per cent, with a median survival of 17 months. Quality of life was not formally assessed in this study. These results have been duplicated more recently in an additional phase II study with an alternative 131I-labelled anti-tenascin mAb 81C6, demonstrating a comparable efficacy.85 On the basis of these results with fixed-dose RIT, further studies are being performed utilizing patient-specific dosing to deliver 44 Gy to the surgical cavity. Other investigators are exploring the use of 90Y-labelled mAbs.86 Gliomas have also been treated with systemic RIT, including intravenous or intra-arterial approaches. The intra-arterial route is associated with an increased risk of complications and has not been shown consistently to give improved tumour uptake.88 An initial pilot study by Miyamoto and colleagues used 125I-anti-epidermal growth factor receptor antibody 425 in 15 patients with recurrent gliomas. An objective response was seen in 20 per cent of patients, with 40 per cent showing SD. Subsequently, the same group has given RIT as primary adjuvant therapy for GBM, and survival appears to be prolonged.83 Diffuse leptomeningeal deposits theoretically represent an ideal target for RIT, being low-volume disease and with the possibility of delivering local intrathecal treatment, thereby reducing systemic toxicity. Kemshead et al. provide a good description of the feasibility and present a review of this approach with preliminary results.84
Gastrointestinal tumours The clinical use of RIT in gastrointestinal malignancies has generally been disappointing, with poor response rates.87,88 There may be a role for RIT in the elimination of smallvolume disease, and pre-clinical studies have provided fresh encouragement that this approach may yet deliver in
Radioimmunotherapy for solid tumours 219
the clinic. Behr et al. compared responses to unlabelled mAbs (one directed against a membrane glycoprotein and the other to CEA – FO23C5), the same mAbs 131I-labelled and 5-fluorouracil (5FU)/folinic acid chemotherapy in Nude mice bearing colon cancer xenografts. There was no reduction in tumour growth with either the unlabelled mAbs or chemotherapy. Responses were seen in the RIT group, with cures reported in 35–55 per cent of mice.89 A pilot clinical study of patients with small-volume (3 cm) metastatic colon cancer who are receiving escalating doses of 131IFO23C5 is continuing, and so far ten patients have been reported with durable anti-tumour effects beyond 12 months. Bone-marrow toxicity appears to be the doselimiting effect. Anti-CEA mAbs have shown good targeting in a variety of CEA-producing tumours, including ovarian, breast, colorectal, pancreas and medullary thyroid cancer. Given the poor results with RIT for macroscopic solid tumours, the delivery of RIT as an adjuvant in the context of microscopic residual disease is being pursued. Liersch et al. have reported the safety and efficacy of delivering the 131 I-labelled anti-CEA mAb labetuzumab as an adjuvant following the resection of colorectal liver metastases. Myelosuppression was the dose-limiting toxicity, with a single administration of 55 mCi/m2 appearing safe. The median OS (overall survival) from first liver resection was 68 months (n 19). This compares favourably with the literature and with a matched contemporaneous group of patients who did not receive RIT.90 A confirmatory phase III randomized trial is planned.
Ovarian tumours The prognosis for epithelial ovarian cancers is poor, with 5-year survival rates reported of between 29 and 40 per cent for patients treated with conventional therapy consisting of cytoreductive surgery and chemotherapy. Intraperitoneal RIT has predominantly been studied for ovarian carcinoma, where peritoneal seedlings are commonly seen. Response rates depend on tumour volume and dose; there is a general trend of increasing chance of tumour control with reducing tumour volume, but this trend is not uniformly reported.91 The largest single-institution study comes from the Hammersmith Hospital, London.92,93 Yttrium-90-labelled murine mAb HMF-G1 was administered to 25 patients with stage Ic–IV epithelial ovarian cancer as a single intra-peritoneal dose following completion of conventional chemotherapy.94 No objective responses were seen in tumour nodules 2 cm. Nine out of 16 patients (56%) with 2 cm nodules responded, as did 50 per cent of patients with microscopic disease. Compared to historical controls matched for age, stage, histological type and grade, survival at 5 years was increased in the RIT group to 80 per cent, compared to 55 per cent in the controls (p 0.003). Acute and late bowel toxicity was also much lower for the RIT group compared to those who
received whole abdominal irradiation. A multi-centre phase III randomized trial is currently ongoing. Two phase I studies performed in the USA have also shown promising results, especially in the treatment of small-volume disease.95 Crippa et al. reported CRs in 5 of 16 (31%) patients with microscopic residual disease or positive washings treated with intra-peritoneal 131I-MOV18.96 Only one of these CRs remained disease free beyond 12 months (34 months), with the median disease-free survival for the other 4 CRs being just 10.5 months. Intraperitoneal RIT is currently being investigated as part of multi-modality therapy. Meredith et al. reported 27 patients who received intra-peritoneal 177Lu-labelled CC49 (anti-TAG-72) with or without interferon-alpha and with or without paclitaxel, within a phase I trial. All patients had TAG-72-expressing tumour limited to the abdomen after surgery/chemotherapy. Three of these patients with smallvolume disease achieved a CR and remain disease free at 3–5 years.97
Breast tumours Clinical trials have demonstrated responses to RIT in 30–60 per cent of heavily pre-treated patients with metastatic breast cancer.98 Target antigens include HER-2/neu (c-erb B2), a transmembrane tyrosine kinase receptor for a polypeptide growth regulatory molecule that is over-expressed in 20–30 per cent of breast adenocarcinomas. Phase III studies using unlabelled antibody to HER-2/neu (trastuzumab) have shown tumour responses in extensively treated patients.99 Radiolabelled mAbs targeting HER-2/neu have been used for the immunodetection of metastatic disease and may have a role in directing appropriate treatment with trastuzumab, but such mAbs have not been used in RIT to date. Good results have, however, been seen using radiolabelled chimeric (Ch) L6 mAb. DeNardo et al. reported a PR in four out of ten patients treated with 131I-Ch L6 and a minor response in 7 patients.100 Wilder et al. reported a 40 per cent PR rate in breast cancer patients treated with 131 I-ChL6.101 In addition to targeting tumour cells, vascular endothelium has also been found to be a target for L6, and levels of the immunomodulatory molecule IL-2R have been seen to correlate with response.102 These observations suggest that these promising results may be the result of increased delivery of RIT to tumour cells due to enhanced vascular permeability as well as synergistic effects of radiation and activated immune effector cell mechanisms. As has been observed with RIT for lymphoma, the mechanisms underlying RIT may involve more than simply targeted radiation. Myeloablative doses of RIT have been studied in breast cancer. Using PBSCT support, 2000–4000 cGy/cycle of therapy can be delivered to sites of metastatic disease.103 Cagnoni et al. reported a phase I study of high-dose RIT with the radiolabelled antibody 90Y-hu-BrE-3 (targeted against the MUC 1 antigen) and autologous PBSC support.
220 Radioimmunotherapy
Four of 9 patients showed a response, with 2 PR and 2 minor responses.104 As suggested previously, using RIT to treat microscopic residual disease and combining RIT with other therapeutic modalities are likely to be more effective than RIT alone. DeNardo et al. have observed a synergistic effect between paclitaxel chemotherapy and RIT with 90Y-ChL6 in a xenografted mouse model, which gives further credence to this approach. Further studies have shown that the optimal time of paclitaxel administration is 48 hours after the RIT, with maximal deposition of the radiolabelled mAb and paclitaxel in the tumour and no paclitaxel in the bone marrow during the time of maximum marrow irradiation.105 A clinical trial using paclitaxel and RIT in patients with metastatic breast cancer is underway.
treatment. No patient achieved a CR or PR based on PSA and/or radiological criteria. Three patients had transient subjective improvement in the symptomatology of their disease.
Renal tumours
Although the results for single-agent RIT are encouraging and make RIT an attractive treatment option for relapsed follicular lymphoma, the future is likely to involve integrating it into chemotherapy treatment protocols in an attempt to increase relapse-free and overall survival. The current challenge for clinical investigators is to determine the optimal approach of integrating RIT into chemotherapy schedules. The emerging data using both 131I-tostumomab and 90Y-ibritumomab tiuxetan as consolidation following shortened or full-course chemotherapy look extremely promising and suggest that the quality of responses can be substantially increased by this type of approach.33,44,110,111 Future randomized clinical trials will define whether this type of approach offers similar or perhaps even superior relapse-free survival over rituximab–chemotherapy regimens followed by maintenance rituximab. Future clinical studies will indicate whether the ability of RIT to improve the quality of the response from PR to CR will add to relapse-free survival and perhaps even overall survival. The highly promising phase II studies using 131I-tositumomab alone in the first-line treatment of indolent lymphoma suggest that further exploration of this approach may be worthwhile, with both 131I-tositumomab and 90Y-ibritumomab tiuxetan. Given the high single-agent activity of 90Y-ibritumomab tiuxetan in DLBCL and 131I-tositumomab in aggressive lymphoma, there are a number of clinical trials that are now underway to integrate either of these two RIT reagents into the front-line treatment of DLBCL. A number of US phase II studies and a large European inter-group study with randomization to 90Y-ibritumomab tiuxetan or no further treatment after full-course CHOP-R chemotherapy are underway. Over the next 5 years, these clinical studies are likely to define further whether RIT has a role in improving the results achieved with R-CHOP immunochemotherapy in DLBCL. In the salvage setting, it is to be expected that the results after standard BEAM (BCNU, etoposide, cytarabine, melphalan) conditioning will now result in less impressive overall survival following the introduction of rituximab
Recent results in metastatic renal-cell cancer have been promising. Divgi et al. treated 33 patients with 131I-labelled mouse mAb anti-G250 as part of a phase I/II trial.106 The G250 antigen is a transmembrane phosphoprotein expressed on the majority of renal-cell cancers. Targeting of radioactivity to all sites of disease 2 cm was demonstrated and 17 patients showed stable disease; 2 had a minimal response. The maximum tolerated 131I dose was 90 mCi/m2. Despite this promise, a phase II study with this agent has been reported more recently in which two consecutive treatments with 131I-cG250 at MTD in patients with progressive metastatic renal-cell carcinoma have been delivered. No responses were seen, although 5 of 25 patients achieved stabilization of their disease.107
Prostate tumours A phase II trial of 75 mCi/m2 131I-anti-TAG-72 high-affinity antibody CC49 has been carried out in 15 patients with hormone-resistant metastatic prostate cancer.108 No acute adverse reactions occurred, but unfortunately all patients had developed evidence of an immune response to CC49 by 4 weeks. Six of 10 symptomatic patients had bone pain relief, but no patients met the radiographic or prostatespecific antigen (PSA) criteria for objective response. Positive imaging of bone and/or soft-tissue lesions was, however, noted for 13 of the 15 patients. More recently, a phase I dose-escalation study using 90YCYT-356 mAb was performed in 12 patients with hormonerefractory prostate carcinoma by Deb et al.109 Of the 12 patients, 7 patients had at least one site of disease imaged after the administration of 111In-CYT-356. The dose range of 90Y was 1.83–12 mCi/m2. Both 111In and 90Y-CYT-356 were tolerated well, without significant non-haematological toxicity. Myelosuppression was the dose-limiting toxicity and occurred at dose levels of 4.5–12 mCi/m2. Only one patient developed a human anti-mouse antibody 4 weeks after
Many of the studies of RIT in solid tumours have produced disappointing results. These results may be improved by using strategies such as pre-targeting and autologous stem-cell support to escalate the dose to tumour. Most importantly, however, if this therapeutic modality is to be used successfully in solid tumours, it needs to be used at a time when the tumour burden is low and in combination with other treatment modalities.
CLINICAL PERSPECTIVES FOR THE NEXT FIVE YEARS
References 221
and the resulting selection of higher risk, poorer prognosis patients who fail R-CHOP regimens. There is therefore an urgent requirement to improve the results and for an intensification of the ‘conditioning regimen’. An area that is therefore currently being intensely investigated is the integration of RIT into the conditioning regimens, instead of TBI or with reduced-dose TBI, followed by ASCT rescue. The studies to date have confirmed that higher myeloablative doses can be safely delivered with ASCT support. Although Press and colleagues have demonstrated that myeloablative doses of 131I-tositumomab with and without high-dose chemotherapy result in highly impressive relapse-free survival, the difficulty has been in reproducing these excellent results given the extremely high doses of 131I that are manipulated in the conjugation process and subsequently administered. A more practical approach, which will allow the participation of many more transplant centres, is the use of ‘standard’ or escalated doses of 90Y-ibritumomab tiuxetan. The early results are encouraging and confirm the feasibility of the addition of escalated doses of 90Y-ibritumomab tiuxetan with the ‘standard’ conditioning regimen of BEAM.57 Appropriately designed randomized studies will confirm over the next few years whether the addition of RIT to high-dose chemotherapy will result in improvement in treatment outcome for patients with relapsed lymphoma. Yttrium-90-ibritumomab tiuxetan is also being investigated in the treatment of other aggressive CD20-positive lymphomas such as mantle-cell lymphoma. Preliminary data in patients with relapsed or refractory mantle-cell lymphoma suggest therapeutic efficacy of 90Y-ibritumomab tiuxetan, with disease control being achieved in around half of the patients treated. Further clinical studies are underway using RIT as consolidation after rituximab chemotherapy regimens and these studies should help to establish whether the poor results in this disease can be improved upon. The use of RIT in leukaemia is less well developed and no radioimmunoconjugate has yet been approved for routine delivery. The exquisite sensitivity of these malignancies to targeted radiation together with the highly impressive results achieved by the pioneers in this field suggest that this is very likely to be a productive area for future clinical research and to lead to tangible patient benefits. The huge progress made over the last decade with the
REFERENCES 1 Adams GP, Weiner LM. Monoclonal antibody therapy of cancer. Nat Biotechnol 2005; 23(9):1147–57. 2 Robak T. Monoclonal antibodies in the treatment of chronic lymphoid leukemias. Leuk Lymphoma 2004; 45(2):205–19. ●3 Coiffier B, Lepage E, Briere J, et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 2002; 346(4):235–42.
development of RIT in the treatment of haematological malignancies leads to distinct optimism that its further development over the next 5 years will result in significant improvements in clinical outcomes for patients.
KEY LEARNING POINTS ●
●
●
●
●
●
●
Radioimmunotherapy now has an established place in the treatment of NHL, with high response rates and durable remissions seen even in heavily pre-treated patients with multiple adverse prognostic factors. Radioimmunotherapy benefits from the crossfire effect whereby radiation is delivered not only to the targeted cell but also to many surrounding cells due to the path length of radiation emitted. The efficacy of RIT in lymphoma is the result of a combination of targeted radiotherapy and antibody effector mechanisms, which include recruitment of immune effectors and signalling by the mAb. The future of RIT in lymphoma is likely to involve integration into chemotherapy schedules, and early data using RIT as consolidation after chemotherapy look highly promising, with high conversion rates of partial responses to complete responses. Myelotoxicity is the dose-limiting toxicity for RIT and can be overcome with autologous stemcell support. Promising results have been achieved when myleoablative doses are combined with stem-cell transplantation, and further studies are required. The future of the application of RIT in other haematological malignancies such as acute leukaemias looks highly promising. Radioimmunotherapy applied to solid tumours is less well developed, with initial clinical responses currently being generally disappointing, and requires further investigation and optimization of techniques.
4 Hiddemann W, Kneba M, Dreyling M, et al. Frontline therapy with rituximab added to the combination of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) significantly improves the outcome for patients with advanced-stage follicular lymphoma compared with therapy with CHOP alone: results of a prospective randomized study of the German Low-Grade Lymphoma Study Group. Blood 2005; 106(12):3725–32. ◆5 Knox SJ, Meredith RF. Clinical radioimmunotherapy. Semin Radiat Oncol 2000; 10(2):73–93.
222 Radioimmunotherapy
6 Press OW, Leonard JP, Coiffier B, Levy R, Timmerman J. Immunotherapy of non-Hodgkin’s lymphomas. Hematology (Am Soc Hematol Educ Program) 2001; 221–40. ●7 Kaminski MS, Zelenetz AD, Press OW, et al. Pivotal study of iodine I-131-tositumomab for chemotherapy-refractory low-grade or transformed low-grade B-cell non-Hodgkin’s lymphomas. J Clin Oncol 2001; 19(19):3918–28. 8 DeNardo GL, DeNardo SJ, Goldstein DS, et al. Maximumtolerated dose, toxicity, and efficacy of (131)I-Lym-1 antibody for fractionated radioimmunotherapy of non-Hodgkin’s lymphoma. J Clin Oncol 1998; 16(10):3246–56. ●9 Illidge TM, Cragg MS, McBride HM, French RR, Glennie MJ. The importance of antibody-specificity in determining successful radioimmunotherapy of B-cell lymphoma. Blood 1999; 94(1):233–43. 10 Press OW, Farr AG, Borroz KI, Anderson SK, Martin PJ. Endocytosis and degradation of monoclonal antibodies targeting human B-cell malignancies. Cancer Res 1989; 49(17):4906–12. 11 Grossbard ML, Press OW, Appelbaum FR, Bernstein ID, Nadler LM. Monoclonal antibody-based therapies of leukemia and lymphoma. Blood 1992; 80(4):863–78. 12 Chan HT, Hughes D, French RR, et al. CD20-induced lymphoma cell death is independent of both caspases and its redistribution into triton X-100 insoluble membrane rafts. Cancer Res 2003; 63(17):5480–9. 13 Sharkey RM, Behr TM, Mattes MJ, et al. Advantage of residualizing radiolabels for an internalizing antibody against the B-cell lymphoma antigen, CD22. Cancer Immunol Immunother 1997; 44(3):179–88. 14 Andres TL, Kadin ME. Immunologic markers in the differential diagnosis of small round cell tumors from lymphocytic lymphoma and leukemia. Am J Clin Pathol 1983; 79(5):546–52. ◆15 Press OW, Rasey J. Principles of radioimmunotherapy for hematologists and oncologists. Semin Oncol 2000; 27(6 Suppl. 12):62–73. 16 Press OW, Shan D, Howell-Clark J, et al. Comparative metabolism and retention of iodine-125, yttrium-90, and indium-111 radioimmunoconjugates by cancer cells. Cancer Res 1996; 56(9):2123–9. 17 DeNardo GL, Kukis DL, Shen S, DeNardo DA, Meares CF, DeNardo SJ. 67Cu-versus 131I-labeled Lym-1 antibody: comparative pharmacokinetics and dosimetry in patients with non-Hodgkin’s lymphoma. Clin Cancer Res 1999; 5(3):533–41. 18 Jurcic JG, Larson SM, Sgouros G, et al. Targeted alpha particle immunotherapy for myeloid leukemia. Blood 2002; 100(4):1233–9. 19 McDevitt MR, Sgouros G, Finn RD, et al. Radioimmunotherapy with alpha-emitting nuclides. Eur J Nucl Med 1998; 25(9):1341–51. ●20 Davis TA, Kaminski MS, Leonard JP, et al. The radioisotope contributes significantly to the activity of radioimmunotherapy. Clin Cancer Res 2004; 10(23):7792–8.
21 Kaminski MS, Tuck M, Estes J, et al. 131I-tositumomab therapy as initial treatment for follicular lymphoma. N Engl J Med 2005; 352(5):441–9. 22 Press OW. Radioimmunotherapy for non-Hodgkin’s lymphomas: a historical perspective. Semin Oncol 2003; 30(2 Suppl. 4):10–21. ◆23 Sharkey RM, Goldenberg DM. Perspectives on cancer therapy with radiolabeled monoclonal antibodies. J Nucl Med 2005; 46(Suppl. 1):115S–27S. 24 Wilder RB, DeNardo GL, DeNardo SJ. Radioimmunotherapy: recent results and future directions. J Clin Oncol 1996; 14(4):1383–400. 25 Goldenberg DM, Horowitz JA, Sharkey RM, et al. Targeting, dosimetry, and radioimmunotherapy of B-cell lymphomas with iodine-131-labeled LL2 monoclonal antibody. J Clin Oncol 1991; 9(4):548–64. 26 Tedder TF, Engel P. CD20: a regulator of cell-cycle progression of B lymphocytes. Immunol Today 1994; 15(9):450–4. 27 Grillo-Lopez AJ. Zevalin: the first radioimmunotherapy approved for the treatment of lymphoma. Expert Rev Anticancer Ther 2002; 2(5):485–93. 28 Witzig TE, White CA, Wiseman GA, et al. Phase I/II trial of IDEC-Y2B8 radioimmunotherapy for treatment of relapsed or refractory CD20(+) B-cell non-Hodgkin’s lymphoma. J Clin Oncol 1999; 17(12):3793–803. 29 Wiseman GA, Gordon LI, Multani PS, et al. Ibritumomab tiuxetan radioimmunotherapy for patients with relapsed or refractory non-Hodgkin lymphoma and mild thrombocytopenia: a phase II multicenter trial. Blood 2002; 99(12):4336–42. 30 Witzig TE, Flinn IW, Gordon LI, et al. Treatment with ibritumomab tiuxetan radioimmunotherapy in patients with rituximab-refractory follicular non-Hodgkin’s lymphoma. J Clin Oncol 2002; 20(15):3262–9. ●31 Witzig TE, Gordon LI, Cabanillas F, et al. Randomized controlled trial of yttrium-90-labeled ibritumomab tiuxetan radioimmunotherapy versus rituximab immunotherapy for patients with relapsed or refractory low-grade, follicular, or transformed B-cell non-Hodgkin’s lymphoma. J Clin Oncol 2002; 20(10):2453–63. 32 Gordon LI, Molina A, Witzig T, et al. Durable responses after ibritumomab tiuxetan radioimmunotherapy for CD20 Bcell lymphoma: long-term follow-up of a phase 1/2 study. Blood 2004; 103(12):4429–31. 33 Shipley DL, Greco FA, Spigel DR, et al. Phase II trial of rituximab and short duration chemotherapy followed by 90Y-ibritumomab tiuxetan as first-line treatment for patients with follicular lymphoma: a Minnie Pearl Cancer Research Network phase II trial. 2004 ASCO Annual Meeting Proceedings (Post-Meeting Edition), 2004 (22, No. 14 S (July 15 Supplement) Abstract #6577). 34 Morschhauser F, Huglo D, Martinelli G, Paganelli G, Zinzani PL, Hadjiyiannakis D. Yttrium-90 ibritumomab tiuxetan (Zevalin) for patients with relapsed/refractory diffuse large B-cell lymphoma not appropriate for autologous stem cell transplantation: results of an open-label phase II trial. Blood 2004; 104:11 (Abstract #130).
References 223
35 Nadler LM, Ritz J, Hardy R, Pesando JM, Schlossman SF, Stashenko P. A unique cell surface antigen identifying lymphoid malignancies of B cell origin. J Clin Invest 1981; 67(1):134–40. 36 Cragg MS, Glennie MJ. Antibody specificity controls in vivo effector mechanisms of anti-CD20 reagents. Blood 2004; 103(7):2738–43. 37 Hohenstein MA, Augustine SC, Rutar F, Vose JM. Establishing an institutional model for the administration of tositumomab and iodine I-131-tositumomab. Semin Oncol 2003; 30(2 Suppl. 4):39–49. ●38 Kaminski MS, Zasadny KR, Francis IR, et al. Radioimmunotherapy of B-cell lymphoma with [131I]antiB1 (anti-CD20) antibody. N Engl J Med 1993; 329(7):459–65. 39 Kaminski MS, Zasadny KR, Francis IR, et al. Iodine-131-antiB1 radioimmunotherapy for B-cell lymphoma. J Clin Oncol 1996; 14(7):1974–81. ●40 Fisher RI, Kaminski MS, Wahl RL, et al. Tositumomab and iodine-131 tositumomab produces durable complete remissions in a subset of heavily pretreated patients with low-grade and transformed non-Hodgkin’s lymphomas. J Clin Oncol 2005; 23(30):7565–73. 41 Horning SJ, Younes A, Jain V, et al. Efficacy and safety of tositumomab and iodine-131 tositumomab (Bexxar) in B-cell lymphoma, progressive after rituximab. J Clin Oncol 2005; 23(4):712–19. 42 Koral KF, Dewaraja Y, Li J, et al. Update on hybrid conjugate-view SPECT tumor dosimetry and response in 131I-tositumomab therapy of previously untreated lymphoma patients. J Nucl Med 2003; 44(3):457–64. 43 Kaminski MS, Zelenetz A, Leonard J, Saleh M, Jain M. Bexxar radioimmunotherapy produces a substantial number of durable complete responses in patients with multiply relapsed or refractory low grade or transformed low grade non-Hodgkin’s lymphoma. Blood 2002; 100:11 Part 2 (Abstract 1382). 44 Press OW, Unger JM, Braziel RM, et al. A phase 2 trial of CHOP chemotherapy followed by tositumomab/iodine I131-tositumomab for previously untreated follicular nonHodgkin lymphoma: Southwest Oncology Group Protocol S9911. Blood 2003; 102(5):1606–12. 45 Liu SY, Eary JF, Petersdorf SH, et al. Follow-up of relapsed B-cell lymphoma patients treated with iodine-131-labeled anti-CD20 antibody and autologous stem-cell rescue. J Clin Oncol 1998; 16(10):3270–8. ●46 Witzig TE, White CA, Gordon LI, et al. Safety of yttrium-90 ibritumomab tiuxetan radioimmunotherapy for relapsed low-grade, follicular, or transformed non-hodgkin’s lymphoma. J Clin Oncol 2003; 21(7):1263–70. 47 Zelenetz AD. Radioimmunotherapy for lymphoma. Curr Opin Oncol 1999; 11(5):375–80. 48 Bennett JM, Kaminski MS, Leonard JP, et al. Assessment of treatment-related myelodysplastic syndromes and acute myeloid leukemia in patients with non-Hodgkin lymphoma treated with tositumomab and iodine I-131-tositumomab. Blood 2005; 105(12):4576–82.
49 Ansell SM, Ristow KM, Habermann TM, Wiseman GA, Witzig TE. Subsequent chemotherapy regimens are well tolerated after radioimmunotherapy with yttrium-90 ibritumomab tiuxetan for non-Hodgkin’s lymphoma. J Clin Oncol 2002; 20(18):3885–90. 50 Dosik AD, Coleman M, Kostakoglu L, et al. Subsequent therapy can be administered after tositumomab and iodine I-131 tositumomab for non-Hodgkin lymphoma. Cancer 2006; 106(3):616–22. 51 Joyce J, Schuster MW, McCook B, et al. Experience with yttrium 90 ibritumomab tiuxetan after autologous stem cell transplant (ASCT) in patients with non-Hodgkin’s lymphoma (NHL). J Clin Oncol (Meeting Abstracts) 2005; 23(16 Suppl.):6669. ◆52 Wahl RL. Tositumomab and 131I therapy in non-Hodgkin’s lymphoma. J Nucl Med 2005; 46(Suppl. 1):128S–40S. 53 Wagner HN Jr, Wiseman GA, Marcus CS, et al. Administration guidelines for radioimmunotherapy of nonHodgkin’s lymphoma with (90)Y-labeled anti-CD20 monoclonal antibody. J Nucl Med 2002; 43(2):267–72. 54 DeNardo GL, Mirick GR, Kroger LA, Bradt BM, Lamborn KR, DeNardo SJ. Characterization of human IgG antimouse antibody in patients with B-cell malignancies. Clin Cancer Res 2003; 9(10 Pt 2):4013S–21S. 55 Press OW, Eary JF, Gooley T, et al. A phase I/II trial of iodine131-tositumomab (anti-CD20), etoposide, cyclophosphamide, and autologous stem cell transplantation for relapsed B-cell lymphomas. Blood 2000; 96(9):2934–42. 56 Gopal AK, Gooley TA, Maloney DG, et al. High-dose radioimmunotherapy versus conventional high-dose therapy and autologous hematopoietic stem cell transplantation for relapsed follicular non-Hodgkin lymphoma: a multivariable cohort analysis. Blood 2003; 102(7):2351–7. 57 Nademanee A, Forman S, Molina A, et al. A phase 1/2 trial of high-dose yttrium-90-ibritumomab tiuxetan in combination with high-dose etoposide and cyclophosphamide followed by autologous stem cell transplantation in patients with poor-risk or relapsed nonHodgkin lymphoma. Blood 2005; 106(8):2896–902. 58 Gordon LI, Witzig T, Molina A, et al. Yttrium 90-labeled ibritumomab tiuxetan radioimmunotherapy produces high response rates and durable remissions in patients with previously treated B-cell lymphoma. Clin Lymphoma 2004; 5(2):98–101. 59 Gregory S, Kaminski M, Zelenetz A, Jain V. Characteristics of patients with relapsed and refractory low grade nonHodgkin’s lymphoma who sustained durable responses following treatment with tositumomab and iodine I-131tositumomab (Bexxar®). Blood 2002; 100(11 Part 2):Abstract 4791. 60 Postema EJ. Dosimetry and radioimmunotherapy of nonHodgkin’s lymphoma. J Nucl Med 2004; 45(12):2126–7. 61 Goldenberg DM, Sharkey RM. Radioimmunotherapy of non-Hodgkin’s lymphoma revisited. J Nucl Med 2005; 46(2):383–4. 62 Britton KE. Radioimmunotherapy of non-Hodgkin’s lymphoma. J Nucl Med 2004; 45(5):924–5.
224 Radioimmunotherapy
63 Winter JN, Gascoyne RD, Van Besien K. Low-grade lymphoma. Hematology (Am Soc Hematol Educ Program) 2004; 203–20. ●64 Feldman EJ, Brandwein J, Stone R, et al. Phase III randomized multicenter study of a humanized anti-CD33 monoclonal antibody, lintuzumab, in combination with chemotherapy, versus chemotherapy alone in patients with refractory or first-relapsed acute myeloid leukemia. J Clin Oncol 2005; 23(18):4110–16. 65 Burke JM, Caron PC, Papadopoulos EB, et al. Cytoreduction with iodine-131-anti-CD33 antibodies before bone marrow transplantation for advanced myeloid leukemias. Bone Marrow Transplant 2003; 32(6):549–56. 66 Matthews DC, Appelbaum FR, Eary JF, et al. Phase I study of (131)I-anti-CD45 antibody plus cyclophosphamide and total body irradiation for advanced acute leukemia and myelodysplastic syndrome. Blood 1999; 94(4):1237–47. 67 Niederwieser D, Maris M, Shizuru JA, et al. Low-dose total body irradiation (TBI) and fludarabine followed by hematopoietic cell transplantation (HCT) from HLA-matched or mismatched unrelated donors and postgrafting immunosuppression with cyclosporine and mycophenolate mofetil (MMF) can induce durable complete chimerism and sustained remissions in patients with hematological diseases. Blood 2003; 101(4):1620–9. 68 McSweeney PA, Niederwieser D, Shizuru JA, et al. Hematopoietic cell transplantation in older patients with hematologic malignancies: replacing high-dose cytotoxic therapy with graft-versus-tumor effects. Blood 2001; 97(11):3390–400. 69 Bunjes D, Buchmann I, Duncker C, et al. Rhenium 188labeled anti-CD66 (a, b, c, e) monoclonal antibody to intensify the conditioning regimen prior to stem cell transplantation for patients with high-risk acute myeloid leukemia or myelodysplastic syndrome: results of a phase I–II study. Blood 2001; 98(3):565–72. 70 Nikula TK, McDevitt MR, Finn RD, et al. Alpha-emitting bismuth cyclohexylbenzyl DTPA constructs of recombinant humanized anti-CD33 antibodies: pharmacokinetics, bioactivity, toxicity and chemistry. J Nucl Med 1999; 40(1):166–76. 71 Sgouros G, Ballangrud AM, Jurcic JG, et al. Pharmacokinetics and dosimetry of an alpha-particle emitter labeled antibody: 213Bi-HuM195 (anti-CD33) in patients with leukemia. J Nucl Med 1999; 40(11):1935–46. 72 Meredith RF, LoBuglio AF, Spencer EB. Recent progress in radioimmunotherapy for cancer. Oncology (Williston Park) 1997; 11(7):979–84, 87; Discussion 87–8. ◆73 Goldenberg DM, Sharkey RM, Paganelli G, Barbet J, Chatal JF. Antibody pretargeting advances cancer radioimmunodetection and radioimmunotherapy. J Clin Oncol 2006; 24(5):823–34. 74 Axworthy DB, Reno JM, Hylarides MD, et al. Cure of human carcinoma xenografts by a single dose of pretargeted yttrium-90 with negligible toxicity. Proc Natl Acad Sci U S A 2000; 97(4):1802–7.
75 Shen S, Forero A, LoBuglio AF, et al. Patient-specific dosimetry of pretargeted radioimmunotherapy using CC49 fusion protein in patients with gastrointestinal malignancies. J Nucl Med 2005; 46(4):642–51. 76 Grana C, Chinol M, Robertson C, et al. Pretargeted adjuvant radioimmunotherapy with yttrium-90-biotin in malignant glioma patients: a pilot study. Br J Cancer 2002; 86(2):207–12. 77 Vuillez JP, Kraeber-Bodere F, Moro D, et al. Radioimmunotherapy of small cell lung carcinoma with the two-step method using a bispecific anti-carcinoembryonic antigen/anti-diethylenetriaminepentaacetic acid (DTPA) antibody and iodine-131 Di-DTPA hapten: results of a phase I/II trial. Clin Cancer Res 1999; 5(10 Suppl.):3259s–67s. 78 Chatal JF, Campion L, Kraeber-Bodere F, et al. Survival improvement in patients with medullary thyroid carcinoma who undergo pretargeted anti-carcinoembryonic-antigen radioimmunotherapy: a collaborative study with the French Endocrine Tumor Group. J Clin Oncol 2006; 24(11):1705–11. 79 Papanastassiou V, Pizer BL, Coakham HB, Bullimore J, Zananiri T, Kemshead JT. Treatment of recurrent and cystic malignant gliomas by a single intracavity injection of 131I monoclonal antibody: feasibility, pharmaco-kinetics and dosimetry. Br J Cancer 1993;67(1):144–51. 80 Riva P, Arista A, Tison V, et al. Intralesional radioimmunotherapy of malignant gliomas. An effective treatment in recurrent tumors. Cancer 1994; 73(3 Suppl.):1076–82. 81 Riva P, Franceschi G, Frattarelli M, et al. 131I radioconjugated antibodies for the locoregional radioimmunotherapy of high-grade malignant glioma – phase I and II study. Acta Oncol 1999; 38(3):351–9. 82 Zalutsky MR, Moseley RP, Benjamin JC, et al. Monoclonal antibody and F(ab)2 fragment delivery to tumor in patients with glioma: comparison of intracarotid and intravenous administration. Cancer Res 1990; 50(13):4105–10. 83 Miyamoto CT, Brady LW, Rackover MA, et al. The use of epidermal growth factor receptor-425 monoclonal antibodies radiolabeled with iodine-125 in the adjuvant treatment of patients with high grade gliomas of the brain. Recent Results Cancer Res 1996; 141:183–92. 84 Hopkins KP, Chandler C, Kemshead, J. Systemic Radiotherapy with Monoclonal Antibodies. New York: Springer-Verlag, 1996. 85 Reardon DA, Akabani G, Coleman RE, et al. Salvage radioimmunotherapy with murine iodine-131-labeled antitenascin monoclonal antibody 81C6 for patients with recurrent primary and metastatic malignant brain tumors: phase II study results. J Clin Oncol 2006; 24(1):115–22. 86 Riva P, Franceschi G, Frattarelli M, et al. Loco-regional radioimmunotherapy of high-grade malignant gliomas using specific monoclonal antibodies labeled with 90Y: a phase I study. Clin Cancer Res 1999; 5(10 Suppl.):3275s–80s. 87 Tempero M, Leichner P, Dalrymple G, et al. High-dose therapy with iodine-131-labeled monoclonal antibody CC49 in patients with gastrointestinal cancers: a phase I trial. J Clin Oncol 1997; 15(4):1518–28. 88 Tempero M. Biologic therapy of gastrointestinal cancer. Cancer Treat Res 1998; 98:227–37.
References 225
89 Behr TM, Memtsoudis S, Vougioukas V, et al. Radioimmunotherapy of colorectal cancer in small volume disease and in an adjuvant setting: preclinical evaluation in comparison to equitoxic chemotherapy and initial results of an ongoing phase-I/II clinical trial. Anticancer Res 1999; 19(4A):2427–32. 90 Liersch T, Meller J, Kulle B, et al. Phase II trial of carcinoembryonic antigen radioimmunotherapy with 131 I-labetuzumab after salvage resection of colorectal metastases in the liver: five-year safety and efficacy results. J Clin Oncol 2005; 23(27):6763–70. 91 Buckman R, De Angelis C, Shaw P, et al. Intraperitoneal therapy of malignant ascites associated with carcinoma of ovary and breast using radioiodinated monoclonal antibody 2G3. Gynecol Oncol 1992; 47(1):102–9. 92 Epenetos AA, Munro AJ, Stewart S, et al. Antibody-guided irradiation of advanced ovarian cancer with intraperitoneally administered radiolabeled monoclonal antibodies. J Clin Oncol 1987; 5(12):1890–9. 93 Stewart JS, Hird V, Snook D, et al. Intraperitoneal radioimmunotherapy for ovarian cancer: pharmacokinetics, toxicity, and efficacy of I-131 labeled monoclonal antibodies. Int J Radiat Oncol Biol Phys 1989; 16(2):405–13. 94 Nicholson S, Gooden CS, Hird V, et al. Radioimmunotherapy after chemotherapy compared to chemotherapy alone in the treatment of advanced ovarian cancer: a matched analysis. Oncol Rep 1998; 5(1):223–6. 95 Jacobs AJ, Fer M, Su FM, et al. A phase I trial of a rhenium 186-labeled monoclonal antibody administered intraperitoneally in ovarian carcinoma: toxicity and clinical response. Obstet Gynecol 1993; 82(4 Pt 1): 586–93. 96 Crippa F, Bolis G, Seregni E, et al. Single-dose intraperitoneal radioimmunotherapy with the murine monoclonal antibody I-131 MOv18: clinical results in patients with minimal residual disease of ovarian cancer. Eur J Cancer 1995; 31A(5):686–90. 97 Meredith RF, Alvarez RD, Partridge EE, et al. Intraperi-toneal radioimmunochemotherapy of ovarian cancer: a phase I study. Cancer Biother Radiopharm 2001; 16(4):305–15. 98 DeNardo SJ, Kroger LA, DeNardo GL. A new era for radiolabeled antibodies in cancer? Curr Opin Immunol 1999; 11(5):563–9. 99 Baselga J, Tripathy D, Mendelsohn J, et al. Phase II study of weekly intravenous trastuzumab (Herceptin) in patients with HER2/neu-overexpressing metastatic breast cancer. Semin Oncol 1999; 26(4 Suppl. 12):78–83.
100 Denardo SJ, O’Grady LF, Richman CM, et al. Radioimmunotherapy for advanced breast cancer using I131-ChL6 antibody. Anticancer Res 1997; 17(3B): 1745–51. 101 Wilder RB, DeNardo GL, DeNardo SJ. Radioimmunotherapy: recent results and future directions. J Clin Oncol 1996; 14(4):1383–400. ◆102 DeNardo SJ. Radioimmunodetection and therapy of breast cancer. Semin Nucl Med 2005; 35(2):143–51. 103 Richman CM, Schuermann TC, Wun T, et al. Peripheral blood stem cell mobilization for hematopoietic support of radioimmunotherapy in patients with breast carcinoma. Cancer 1997; 80(12 Suppl.):2728–32. 104 Cagnoni PC, Ceriani R, Cole W. Phase 1 study of high dose radioimmunotherapy with 90-Y-Hu-BrE-3 and autologous stem cell support in patients with metastatic breast cancer. Cancer Biother Radiopharm 1998; 13:328. 105 Denardo SJ, Richman CM, Kukis DL, et al. Synergistic therapy of breast cancer with Y-90-chimeric L6 and paclitaxel in the xenografted mouse model: development of a clinical protocol. Anticancer Res 1998; 18(6A):4011–18. 106 Divgi CR, Bander NH, Scott AM, et al. Phase I/II radioimmunotherapy trial with iodine-131-labeled monoclonal antibody G250 in metastatic renal cell carcinoma. Clin Cancer Res 1998; 4(11):2729–39. 107 Brouwers AH, Mulders PF, de Mulder PH, et al. Lack of efficacy of two consecutive treatments of radioimmunotherapy with 131I-cG250 in patients with metastasized clear cell renal cell carcinoma. J Clin Oncol 2005; 23(27):6540–8. 108 Meredith RF, Bueschen AJ, Khazaeli MB, et al. Treatment of metastatic prostate carcinoma with radiolabeled antibody CC49. J Nucl Med 1994; 35(6):1017–22. 109 Deb N, Goris M, Trisler K, et al. Treatment of hormonerefractory prostate cancer with 90Y-CYT-356 monoclonal antibody. Clin Cancer Res 1996; 2(8):1289–97. 110 Link B, Kaminski MS, Coleman M, Leonard JP. Phase II study of CVP followed by tositumomab and iodine I-131tositumomab (Bexxar therapeutic regimen) in patients with untreated follicular non-Hodgkin’s lymphoma. Proc ASCO 2004; 23:560 (Abstract 6520). 111 Leonard JP, Coleman M, Kostakoglu L, et al. Abbreviated chemotherapy with fludarabine followed by tositumomab and iodine I-131-tositumomab for untreated follicular lymphoma. J Clin Oncol 2005; 23(24):5696–704.
11 Monoclonal antibodies and treatment of cancer HOSSEIN BORGHAEI, LIAT BINYAMIN, IGOR ASTSATUROV AND LOUIS M. WEINER
Introduction Mechanisms of monoclonal antibody action Antibody-based cellular cytotoxicity Complement-dependent cytotoxicity Manipulation of signal transduction Delivery of cytotoxic compounds – immunoconjugates
226 227 227 228 228 229
INTRODUCTION Antibody-based therapy has emerged as an integral part of effective therapies for a number of malignancies. Monoclonal antibodies (mAbs) have considerable utility in cancer therapy because of their specificity and ability to bind preferentially to tumour-specific antigens without causing significant side effects compared with conventional cytotoxic drugs. Antibodies raised against CD20 on the surface of B cells (e.g. rituximab) have emerged as major components of lymphoma treatment.1 Exciting and clinically important antibody therapies are widely used to treat HER2/neu over-expressing breast cancer,2 epidermal growth factor (EGF) receptor overexpressing malignancies3,4 and cancers that are driven by pro angiogenic properties of the vascular endothelial growth factor (VEGF).5 This chapter reviews some of the pertinent data and mechanisms of action for the more clinically relevant monoclonal antibodies. In the last decade, monoclonal antibodies have become major therapeutic vehicles in the treatment of malignant and non-malignant diseases. The success of immune-based treatments and an improved understanding of the role of the immune system in cancer development and progression have mapped the road for extensive development of antibody-based therapy. A century ago, Paul Ehrlich described the antibody therapy concept: to selectively target malignant cells based on the unique expressed determinants’ profile of the disease (the ‘magic bullet’ hypothesis). The advanced technology of hybridoma production provided the necessary skills to
Immunotoxins Radiolabelled monoclonal antibodies Drug-conjugated monoclonal antibodies Haematologic malignancies Solid tumours References
229 229 230 230 233 242
produce highly specific mAbs.6,7 Extensive efforts were made to apply this technique and to generate mAbs for cancer therapy. Many mAbs that are currently in use were originally from hybridoma-derived murine antibodies.8 A second generation of engineered antibodies was employed, taking into account the major issue of stimulation of human–antimouse antibody responses (HAMA) and the inefficient interaction of the murine origin constant region of the antibodies (Fc) with human immune accessory cells.9 Using an immunoglobulin G (IgG) scaffold, humanized and chimeric antibodies were generated by incorporating portions of the murine variable regions into the human IgG backbone.10 Grafting either the entire murine variable regions (chimeric mAb) or the murine complementary-determining regions (humanized mAb), mAbs are created that contain human Fc domains and preserve their target specificity.11 In addition, using molecular engineering techniques, critical human heavy-chain backbone sequences were grafted onto the xenogeneic murine antibody structure, reducing the immunogenicity and providing the important human Fc domain to the resulting antibodies.12 To improve the power of mAbs further and to design antibodies that specifically target and subsequently eliminate cancer cells, large combinatorial antibody libraries of murine, human or synthetic origin were constructed. Efficient in-vitro screening systems now allow the bypassing of immunization and the selection of recombinant antibodies of defined specificity without the need for hybridoma production.13,14 New forms of antibody modules with different size, flexibility and valency suitable for in-vivo imaging and therapy were created using genetic
Antibody-based cellular cytotoxicity 227
manipulation, recombinant production and antibody conjugation techniques.15,16 A major breakthrough in the technology of antibody engineering was the derivation of single-chain molecules (scFv). These molecules were obtained by joining the heavy and light variable domains (VH and VL) from a mAb with a flexible linker, which allowed the reconstitution of the original antigen-binding fragment association.17 A number of multi-valent scFvbased structures have been engineered, including bispecific and bivalent antibodies such as diabodies and minibodies.18–21 This has led to the development of a variety of engineered mAb antibody molecules. These mAbs are used in several formats: unconjugated, as drug–toxin conjugates or as radiolabelled antibodies, for research, diagnosis and therapy. Each format possesses advantages and disadvantages with respect to its potential mechanism of action and clinical applications. A review of ongoing research and development efforts, and of the increasing number of mAbs which have gained approval from the Food and Drug Administration (FDA) for use in cancer therapy;11 demonstrates a desire to develop more specific, less toxic and more efficient targeted therapies.22
MECHANISMS OF MONOCLONAL ANTIBODY ACTION The human backbones of IgG1 isotype have been shown effectively to mediate Fc domain-based functions such as antibody-dependent cellular cytotoxicity (ADCC) and complement fixation.23 It is therefore the most commonly used antibody format among the available mAbs for cancer therapy. The therapeutic effects of other mAbs can be achieved by binding to the specific antigen and eliminating the natural ligand from binding to its receptor, leading to signal transduction alteration.24,25 As a proof of concept, anti-idiotypic antibodies were raised and tested, taking advantage of differentially expressed tumour-associated antigen (for example cell-surface-membrane immunoglobulin present on human B-cell lymphomas).26 These antibodies were designed to bind unique idiotypes associated with cell-surface-membrane immunoglobulins. This interaction would then lead to perturbation of downstream signalling and enhanced apoptosis. Monoclonal antibodies that are currently under development incorporate additional beneficial modifications to improve the efficiency of the different potential mechanisms of action. These include alteration of Fc-domain glycosylation and sequence to enhance ADCC or modification in size and antigen-binding affinity to increase the ability of the mAb to penetrate solid tumours.27 Monoclonal antibodies directed against human tumour-associated antigens have also been used as selective vehicles for the delivery of radionuclides to malignant cells.28 In the past 40 years, although studies have confirmed this concept of using labelled antibodies for cancer diagnosis and therapy, progress in this area has been obstructed by methodological limitations. In addition to
radioisotopes, mAbs conjugated to catalytic toxins, drugs and enzymes have been used extensively in clinical trials.28 A synergistic effect can be achieved by combining mAb therapy with traditional chemotherapy agents, attacking tumours through complementary mechanisms of action.
ANTIBODY-BASED CELLULAR CYTOTOXICITY Antibody-based cellular cytotoxicity is a well-recognized immune effector mechanism in which antigen-specific antibodies direct immune effector cells of the innate immunity, such as natural killer cells (NK cells) and macrophages to the killing of antigen-expressing cancer cells.29 This property is dependent on interactions between cellular Fc receptors (FcRs) on immune accessory cells and the Fc portion of an antibody.30 The FcRs for IgG were identified more than 40 years ago with the observation that IgG antibodies could be directly cytophilic for macrophages when presented as opsonized red blood cells.31 The binding cross-links FcR on the effector cells, and as a result the effector cells become activated, e.g. natural killer (NK) cells are triggered to kill cancer cells and also release cytokines and chemokines.22 The effector cells that may mediate ADCC include NK cells, monocyte-macrophages and neutrophils. Natural killer cells comprise the principal ADCC effector cells as they express low-affinity type IIIA FcR (FcγRIIIa; CD16a) on their surface.32–34 In ADCC, NK cells generally kill their target cells by releasing cytotoxic granules (perforin) and granzymes. CD16a plays a dominant role in NK-cellmediated ADCC and is therefore is known as the ADCC receptor.35 Antibody-based cellular cytotoxicity-mediated elimination of tumour cells was demonstrated in vitro in the presence of NK cells and tumour-specific antibodies of appropriate IgG isotypes.36 Clynes and Ravetch studied the magnitude of FcR interaction in vivo by examining the anti-tumour activities of clinically effective mAbs against human tumour xenografts growing in either wild-type mice or murine FcγRII/III knock-out mice.37 Anti-tumour activity was reduced in the FcγR knock-out mice, and was conserved when only the inhibitory FcγR isoform was removed.37 The role of FcγR in the anti-tumour effects of rituximab, the first FDA-approved chimeric mAb for lymphoma treatment, was further supported with the finding that CD16 polymorphisms predict responses to rituximab in patients suffering from follicular lymphoma.38 These findings indicate that interactions between the antibody Fc domain and the FcR underlie at least some of the clinical benefit of some mAb, and imply the importance of ADCC. Indeed, ongoing research projects are focused on strategies to design and test new mAbs with an improved capacity to mediate ADCC. The approaches include manipulation of the mAb Fc region, which directly participates in activating complement via the classical pathway, and in recruiting FcγRIII on immune effector cells to mediate ADCC. Combined computational and experimental methods have identified mutations within
228 Monoclonal antibodies and treatment of cancer
the Fc domain of mAb to tune selectively the affinity for FcγRIII and other Fcγ receptors.39 An alternative strategy to enhance ADCC by mAb is to modify Fc glycosylation by the producing cell line.40 Modification of the Fc region to interact with activating or inhibitory FcR could enhance antigen presentation by dendritic cells and can be biased to promote or inhibit the generation of cytotoxic T-cell responses against the targeted antigen. Such T cell responses indicate that the induction of ADCC can lead to adaptive immune responses and finally to the elimination of tumour cells.27,41
complement.50 Down-regulating the expression of monomeric complement reactive protein (mCRP) has a potential to improve mAb-mediated complement activation.23 In this regard, the use of various cytokines for in-vivo studies has been reported.51 Furthermore, mAbs that block the function of mCRPs could enhance the complement susceptibility of tumour cells. A limitation of this approach is that mCRP is widely expressed on normal tissue. Bispecific antibodies that can recognize both a tumour-related antigen and a mCRP can selectively target tumour cells and enhance their susceptibility to complement deposition and lysis.52
COMPLEMENT-DEPENDENT CYTOTOXICITY
MANIPULATION OF SIGNAL TRANSDUCTION
Most mAbs that mediate ADCC also activate the complement system.42 The classical activation cascade of the complement system, complement-dependent cytotoxicity (CDC), involves direct killing of the tumour cells by forming membrane attack complex (MAC). Following antibody binding to antigens on the target cell, C1q binding sites on the antibodies’ Fc domain become available. Multiple C1q binding then changes the low-affinity interactions of the single C1q–IgG interaction to high-avidity interactions. This leads to the release of C3a and C5a, which, in turn, will attract and activate effector cells, resulting in an improvement of ADCC. The complement proteins will also create MAC that cause pore formation in the target cell membrane, leading to tumour-cell death. Several clinically approved mAbs activate complement on tumour cells in vitro; however, the clinical relevance of CDC has been difficult to demonstrate convincingly.43,44 Improving the ability of mAbs to activate CDC has the potential to increase clinical efficacy, and could be additive to all the existing effector mechanisms of such antibodies. For example, it is known that the high binding avidity of C1q requires the dimerization of two IgGs; however, occasionally the targeted antigens are present in low density, thus enabling the formation of IgG dimers. A recent study has suggested that multiple, different epitope-targeted mAbs directed to HER2/neu antigen could be used in order to increase the number and density of the mAbs on the target cell.45 To increase the density of bound antibodies, a secondary antibody could also be used against either the anti-tumour mAb46 or the inactivated b fragment of the C3 complement regulatory protein (iC3b) deposited on tumour cells by the primary mAb.47 Another approach to enhance complementmediated effector mechanisms is to conjugate an anti-tumour mAb to a complement activation protein, such as cobra venom factor (CVF) or C3b.48,49 Complement-receptor-3dependent cellular cytotoxicity (CR3-DCC) could be activated through the complement C3 activation product iC3b deposited on the tumour cell. This mechanism requires the presence of the cell-wall β-glucan, though associated with yeast and fungal infection. The soluble yeast β-glucan has been shown to serve as adjuvant that greatly promoted the tumour regression activity of mAbs that activated
The signalling events that are essential for the tumour cells to proliferate are mainly triggered through the interaction of extracellular ligands with cell-surface receptors. Antibodies can be directed to alter these interactions.53,54 An example of this concept may be best demonstrated in VEGF.54 The inhibition of angiogenesis and prevention of the development of the tumour neovasculature is the process that inhibits growth beyond a minimal size. The vasculature is directly accessible to antibodies, and vascular damage might affect many tumour cells depending on number of cells that each capillary supports. A mAb, bevacizumab (Avastin™, Genentech), binds to VEGF and blocks the interaction with its receptor.55 Another growth factor receptor family exploited for therapeutic purposes with respect to interference and inhibition by mAbs is the ErbB or epidermal growth factor receptor (EGFR). The activation of the EGFR is controlled by the binding of specific ligands and this induces the formation of heterodimers and activation of the intrinsic kinase domain (reviewed in reference 56). It has been shown that cancer patients with tumour cells expressing high levels of ErbB1 and ErbB2 have a more aggressive disease and an unfavourable prognosis. For this reason, ErbB receptors are attractive therapeutic targets and many different approaches have been tried to inhibit them.57 Cetuximab, a humanized mAb, recognizes the EGFR ectodomain and competes for ligand binding to the receptor, resulting in the inhibition of mitogenesis.58 Others, such as pertuzumab (2C4), allow ligand binding to occur but sterically inhibit the subsequent receptor heterodimerization required for signal transduction.59 Subsequently, in-vitro and in-vivo growth of breast and prostate tumour cells is inhibited by 2C4. The therapeutic benefit of 2C4 could complement that of other antibodies, as the prevention of receptor heterodimerization appears to be a promising novel approach that could possibly complement the use of trastuzumab and tyrosine kinase inhibitors.60 Since overexpression of growth factor receptors is essential for the maintenance of tumour cells, therapeutic agents that have the effect of reducing the density of target antigen expression are of interest. Of these, antagonistic antibodies, smallmolecular-weight kinase inhibitors, compounds causing ErbB2 degradation and scFv-mediated inactivation of ErbB2
Radiolabelled monoclonal antibodies 229
through its retention in the endoplasmic reticulum have been employed.61
DELIVERY OF CYTOTOXIC COMPOUNDS – IMMUNOCONJUGATES Immunoconjugates are molecules that consist of a ‘targeting’ domain that localizes in tumours and a therapeutic domain. Immunoconjugates, in the broadest definition, may utilize mAbs, mAb fragments, hormones, peptides or growth factors selectively to localize cytotoxic drugs, plant and bacterial toxins, enzymes, radionuclides or cytokines to antigens presented on tumour cells or on cells of the tumour neovasculature.11
IMMUNOTOXINS These are hybrid molecules derived by coupling bacterial, plant or fungal toxins to mAbs specific for molecules on the surface of tumour cells. The elucidation of the molecular structure of bacterial toxins such as Pseudomonas aeruginosa exotoxin A (ETA) and the development of recombinant antibody technologies have allowed the minimization of the size of these molecules via recombinant DNA techniques and their production as single polypeptides in large quantities and consistent quality in bacteria.62 The principal function of these immunotoxins is to inhibit protein synthesis after internalization, leading to the death of the targeted cell. Accordingly, a mAb that has no intrinsic cell-elimination function in an unconjugated format might still be useful for immunotoxin design. An advantage of immunotoxins is that an exceedingly small amount of the protein is actually required to kill both resting and dividing cells. Part of this property derives from the catalytic properties of the fusion proteins: one molecule can attack multiple intracellular targets.63, 64 Over the years, a number of clinical trails have been conducted with immunotoxins and fusion toxins.62, 63, 65, 66 These studies defined a number of pharmacologic and toxicologic barriers that needed to be overcome. The original rationale for the production and testing of these reagents was that they had a mechanism of action different from that of DNA-damaging or cell-division-damaging therapeutics and thus might be effective either alone or in combination in patients with chemotherapy-resistant malignancies.
RADIOLABELLED MONOCLONAL ANTIBODIES For radioimmunotherapy, the central issue is balancing the dose delivered to tumour against the exposure of normal organs and tissue to radiation. Two radiolabelled mAbs have been licensed to date. These are ibritumomab tiuxetan (Zevalin) and tositumomab (Bexxar), which are mAbs conjugated to the β-particle-emitting radioisotopes yttrium-90 and iodine-131, respectively.67 These radioimmunoconjugates are described in detail in Chapter 10. These agents are
used in the therapy of non-Hodgkin’s lymphoma (NHL) and are directed to the CD20 antigen.68 In patients with solid tumours, response rates to radioimmunotherapy agents are modest. In general, after intravenous injection, mAbs accumulate in solid tumours comparatively slowly, and less than 1 percent of the injected dose typically is localized per gram of tumour. Such inefficient accumulation has been attributed to a number of physiological barriers between the blood circulation and the tumour-cell surface.69 These barriers include the vascular endothelium, size-dependent diffusion properties of the antibodies, hydrostatic pressure within the tumour and long transport distances in the tumour tissue. Thus, for these less radiosensitive tumours, more effective targeting of tumour with mAb is required. Several strategies have been developed to enhance the efficacy of immunotherapy, such as the use of mAb fragments, highaffinity mAbs and labelling techniques that are stable in vivo, the active removal of the radiolabelled mAbs from the circulation, and pre-targeting strategies.68,70,71 Radioimmunotherapy has also been combined with other agents or modalities, such as cytokine administration or hyperthermia, to increase antigen expression and tumour uptake. Chemotherapy has been used to enhance radiosensitivity.72,73 Important applications of radionuclide-labelled mAb include for the imaging and localization of the tumour and the prediction of the clinical outcome of treatment. For in-vivo imaging, contrast (signal:noise ratio) is the key parameter for success. Biodistribution studies demonstrate that smaller fragments, such as diabodies and minibodies, reach their maximal tumour uptakes within 1–6 hours of administration in xenograft-bearing mice. Because of rapid blood clearance, tumour:blood ratios increase steadily over time and reach high values (20:1) by 24 hours, making these fragments prime candidates for imaging.74 Much recent work has focused on positron emission tomography (PET) imaging for evaluating, targeting and distributing drugs and tracers because of the higher sensitivity, resolution and quantification afforded by this imaging mode.75,76 Achieving high tumour-to-non-tumour ratios with the slow antibody uptake into solid tumours is a major challenge in the application of immunoconjugates.77 In pre-targeting, the radionuclide is administered separately from the tumour-targeting mAb. In the first step, the unlabelled mAb is administered and allowed to accumulate in the tumour. In the later phase, preferably when the mAb has been cleared from the circulation, the radionuclide is administered as an agent that can rapidly clear from the circulation but with high affinity for the unlabelled molecule that was injected in the first phase.68,78 Rather then radionuclides, the second-phase administration of a cytotoxic drug or enzyme specifically designed to bind the mAb could be utilized. This has been referred to as antibody-directed enzyme pro-drug therapy (ADEPT). As with radiolabelled mAbs for cancer therapy, the tumour-to-nontumour-conjugate ratio is a critical parameter in ADEPT that directly affects the amount of pro-drug that can be safely administered.79–81
230 Monoclonal antibodies and treatment of cancer
DRUG-CONJUGATED MONOCLONAL ANTIBODIES Gemtuzumab ozogamicin (Mylotarg) is FDA approved for use in relapsed acute myeloid leukaemia (AML). This is a humanized antibody directed to the MUC1 antigen, linked to calicheamicin, an antibiotic that cleaves DNA.82 The specificity of this very toxic anti-tumour compound is assured by the recognition of CD33 antigen on cell surface; through internalization of the conjugated mAb, the calicheamicin portion can selectively exert its tumour effects.67 Studies involving doxorubicin as the payload drug have also been reported.83–85 It has only in the past few years that the critical parameters for optimization have been identified and have begun to be addressed. These include the physiological barriers to mAb extravasation and intra-tumoral penetration, mAb immunogenicity, normal tissue expression of the targeted antigen, low drug potency, inefficient drug release from the mAb and difficulties in releasing drugs in their active states.28,77,86,87 The following is a detailed discussion of some of the clinically relevant mAbs and their targets.
HAEMATOLOGIC MALIGNANCIES Antibodies in bone-marrow transplantation Graft-versus-host disease (GvHD) continues to be one of the main barriers to successful allogeneic transplantation. The most recent efforts have concentrated on the depletion of T cells as a means of overcoming this problem. Muromonab (OKT3) was one of the first mAbs to be used in clinical practice. This antibody blocks all cytotoxic T-cell functions. In an early trial, 17 patients received donor marrows that were incubated with OKT3 in an attempt to deplete them of immuno-competent T lymphocytes. All donors were either matched or slightly mismatched siblings. In the first 60 days of observation, 11 of these patients showed no evidence of acute GvHD, four patients had transient limited skin rash, and three developed grade II or higher GvHD.88 In another study, the treatment of 14 patients with steroidresistant GvHD following allogeneic bone-marrow transplant with an anti-CD3/TCR antibody (OKT3/BMA031) led to a reduction or resolution of GvHD in the majority of the patients, but the infectious rates were high post-treatment.89 In this study the persistence of CD3 lymphocytopenia was associated with a complete clearance of acute GvHD. A newer antibody that targets the CD25 antigen (basiliximab) is also under investigation in this context. Ji and colleagues used this antibody in conjunction with their usual GvHD prophylaxis regimen of anti-thymocyte globulin (ATG), cyclosporine, methotrexate and mycophenolate mofetil.90 The incidence of GvHD was 11 per cent in the group of patients who were treated with basiliximab versus 33 per cent in the control group. All patients achieved engraftment with full donor chimerism, with
recovery of T, B and NK cells within 12 months posttransplant. Anti-thymocyte globulin (ATG) has also been extensively studied. When used as part of a conditioning regimen by the Gruppo Italiano Trapianti Midollo Osseo (GITMO), 15 mg/kg of this agent was shown to be effective in reducing the risk of grade III–IV acute GvHD, which reached statistical significance.91 Transplant-related mortality and overall survival did not change in this study, but extensive chronic GvHD was reduced. Both labelled and unlabelled mAbs have been used in the transplant setting as well. Rituximab has been incorporated into bone-marrow transplant in a number of different ways, with the aim of exploiting the synergism between rituximab and chemotherapy. Also, with pre-transplant administration, it is possible to obtain a lymphoma-free graft harvest. Improvements in complete response rates and relapse-free survival might be achieved with maintenance therapy.92 Rituximab was used for in-vivo purging by Magni et al. in 25 patients with relapsed or refractory follicular lymphoma or mantle-cell lymphoma.93 Fifteen patients received rituximab plus high-dose chemotherapy and ten patients received just the high-dose therapy. A total of six infusions of rituximab were given at specified intervals. Polymerase chain reaction (PCR)-negative status was achieved in 93 per cent of grafts in the chemotherapy–rituxan arm, as opposed to 40 per cent in the chemotherapy-only arm. Clinical follow-up was too short in this report for interim analysis. A similar protocol was used by Gianni94 in 28 patients with mantle-cell lymphoma. A 97 per cent complete response rate and a projected 3-year survival of 85 per cent were reported. These and other similar data are extensively reviewed by Malek and Flinn.92 Radiolabelled antibodies have also been incorporated into the transplant setting, as described in Chapter 10.
Rituximab and CD20 The first successful application of a mAb was in the treatment of low-grade NHL. The anti-CD20 mAb rituximab (Rituxan) was the first mAb to be approved by the FDA (in 1997) for use in human malignancy.95,96 Treatment with this chimeric anti-CD20 antibody led to impressive clinical responses and dramatically changed the available treatment options for patients with NHL. Rituximab is humanized and multiple doses can be safely administered. In-vitro studies have demonstrated multiple mechanisms by which anti-CD20 antibodies lead to cell death.97 CD20 is a cell-surface marker that is expressed on all normal and most malignant B cells.98 It has a low rate of internalization and therefore is an attractive target for antibody-based therapeutics.99 The CD20 protein has 297 amino acids and intracellular termini, spans the plasma membrane four times and has a single non-glycosylated extracellular loop of 43 residues.98 Its functions remain unknown, but there is evidence that it participates in calcium influx.100,101 There is also evidence that after the binding of rituximab to CD20, in addition to ADCC
Haematologic malignancies 231
and complement, apoptosis could play a role as a direct cytotoxic effect of rituximab on B cells. There is evidence of apoptosis occurring in B cells isolated from patients with chronic lymphocytic lymphoma (CLL) who were treated with rituximab, thus supporting a role for direct CD20 signalling.102 The clinical development, single-agent and combination data with chemotherapy are described in Chapter 43. Table 11.1 lists some of the early trials conducted with rituximab. Emerging data also indicate that low-grade B-cell lymphoma patients possessing the 158v/v polymorphism in FcγRIII experience superior response rates and outcomes when treated with this antibody.38,103 These findings indicate that antibody Fc domain–Fc receptor interactions underlie at least some of the clinical benefit of rituximab, and that there is a possible role for ADCC, which depends on such interactions.
Combination with cytokines The immunomodulatory effects of cytokines have been exploited in a number of early studies in order to augment the observed activity of rituximab in patients with NHL. Interleukin-2 (IL-2) is a lymphokine produced by T lymphocytes that has direct effect on B cells, NK cells and monocytes. Several studies have shown that patients treated with IL-2 can have greatly increased NK-cell concentrations,104,105 and since these cells are important effectors of ADCC, this can lead to improved cell killing in combination with rituximab. A phase I trial conducted to test this hypothesis reported an overall response rate of 55 per cent in rituximab-naive patients with follicular lymphomas.106 Patients were treated with the standard doses of rituximab on a weekly basis for four treatments in addition to daily subcutaneous injections of 1.2 million units (MIU)/m2 of IL-2 This combination seems to be feasible, and further work is needed to determine the risks and benefits of the addition of IL-2. Similarly, IL-12 has been studied in combination with rituximab.107 Escalating doses of IL-12 were used in a phase I trial in combination with standard doses of rituximab in 43 patients with NHL of various histologies. At the maximum tolerated dose of 300 ng/kg of IL-12, dose-limiting toxicities were liver function elevations and haematotoxicity. A response rate of 69 per cent was reported, with a 25 per cent complete response rate across all histologies. Type I interferon (IFN) not only has immunomodulatory effects on T cells and NK cells, but also has been shown
to up-regulate the expression of CD20 on B cells.108,109 Two trials have examined the safety and feasibility of combination therapy with IFN and rituximab.110,111 Davis and colleagues110 conducted a single-arm, multi-centre, phase II trial to assess the safety and efficacy of combination therapy with rituximab and IFN-α2a in 38 patients with relapsed or refractory, low-grade or follicular B-cell NHL. Interferonα2a, 2.5 or 5 million units (MIU) was administered subcutaneously three times weekly for 12 weeks. Starting on the fifth week of treatment, rituximab was administered by intravenous infusion (375 mg/m2) weekly for four doses. All patients received four complete infusions of rituximab and were evaluable for efficacy, but 11 patients did not receive all 36 injections of IFN. The study treatment was reasonably well tolerated, with no unexpected toxicities stemming from the combination therapy. The overall response rate was 45 per cent (17 of 38 patients); 11 per cent had a complete response, and 34 per cent had a partial response. The median response duration and the median time to progression in responders were 22.3 and 25.2 months, respectively. Further follow-up is needed to determine whether this treatment combination leads to a significantly longer time to progression than single-agent treatment with rituximab. In a study reported by Sacchi et al., 64 patients with various lymphomas were treated with standard-dose rituximab in addition to daily subcutaneous injection of IFN, starting with 1.5 MIU on week 1 and escalating to 6 MIU on weeks 4 and 5. The overall response rate was 70 per cent, with 33 per cent complete response and a median duration of remission of 19 months.111 These results need to be confirmed in subsequent trials to better define the role of combination immunotherapy in patients with NHL.
New anti-CD20 antibodies The pharmacokinetic analysis in the first pivotal trial of rituximab indicated that patients with higher and more prolonged blood levels of active drug had a better chance of responding.112 This not only led to investigations into the role of maintenance therapy with rituximab, it also sparked an interest in developing second-generation antiCD20 antibodies with more favourable pharmacokinetics and improved efficacy. Several of these antibodies are in clinical development.113,114 The HuMax-CD20 antibody is being studied in several phase I/II trials. Hagenbeek and
Table 11.1 Studies with Rituxmab
Witzig 1999 (137) Witzig 2002 (138) Witzig 2002 (139) Wiseman 2002 (140)
Phase
Patient population
CR (%)
ORR (%)
TTP (months)
Number of patients
Phase I/II Randomized Phase III Phase II Phase II
All histologies No prior rituximab Rituximab refractory Thrombocytopenia
26 34 15 37
67 80 74 83
12.9 11.2 6.8 9.4
51 143 57 30
CR, complete response; ORR, overall response rate; TTP, time to progression.
232 Monoclonal antibodies and treatment of cancer
colleagues.115 have reported the results of a study involving 40 patients with follicular lymphoma. The patients were treated with four weekly infusions of 300–1000 mg of HuMax anti body after pre-medication. Treatment led to the depletion of circulating B cells and no dose-limiting toxicities were encountered. Twelve of 24 patients in this study had an objective response, including four complete responses. Coiffier et al. recently reported the early results of a phase I/II trial with HuMax-CD20 antibody in patients with CLL.116 Thirty-three patients in three cohorts with refractory or relapsed CLL received four weekly infusions of the antibody in escalating doses up to 2000 mg. The maximum tolerated dose was not identified. Adverse events were noted mainly on the days of infusions and were related to cytokine release. Other adverse events included hepatic cytolysis, herpes zoster, neutropenia and one death from pneumonia. An overall response rate of 52 per cent was observed at week 11. Treatment also led to significant depletion of CD19CD5 cells. These results suggest that HuMax-CD20 has a favourable toxicity profile and activity in low-grade lymphomas.
Alemtuzumab Alemtuzumab is a humanized anti-CD52 mAb that has been approved by the FDA for the treatment of fludarabinerefractory CLL. CD52 is expressed on almost all human lymphocytes.117 The rat IgM and IgG antibodies were lytic with complement, but only the IgG2b had ADCC activity. The function of this antigen remains unknown. It is known, however, that CD52 is expressed on all lymphocytes at various stages of differentiation, as well as on monocytes, macrophages and eosinophils.118 The only other site of expression is the male reproductive tract.119 Haematopoietic stem cells, erythrocytes and platelets do not express this antigen and are thus spared from a direct antibody effect. The highest levels of expression is on T-prolymphocytic leukaemia (PLL) cells, followed by B-cell CLL, with the lowest levels on normal B cells.120 The first CD52 antibodies were isolated in the 1980s when researchers were seeking antibodies that would kill T cells by activating human complement. An IgM antibody was initially selected because of the efficiency with which it activated complement and because it was cytotoxic to T cells in vitro. Prior iterations included CAMPATH-1G, a murine derivative,117 which demonstrated clinical activity in refractory CLL even in patients who had experienced treatment failure with Campath-1M. Possible mechanisms of action of this agent include antibody-dependent cellular toxicity (ADCC), complement-dependent cytotoxicity and induction of apoptosis. CAMPATH has been tested as a therapeutic agent in patients with CLL121 and low-grade NHL122 and as a means to deplete T cells in bone-marrow transplantation.123,124 In one study, 33% of the patients with fludarabineresistant CLL exhibited clinical responses to CAMPATH-1.125 A larger phase II study reported a 42 per cent overall response rate in patients with relapsed or refractory CLL,
but at the cost of an increase in opportunistic infections and septicaemia.126 CAMPATH-1 has also been evaluated as first-line therapy for patients with CLL. Loss of peripheral blood malignant lymphocytes was seen in all patients treated with this antibody. However, patients with involvement of lymph nodes and/or spleen were less likely to respond completely. There was evidence of reactivation of cytomegalovirus infections. Subcutaneous administration of the antibody was found to be safe and effective.127 In a phase II multi-centre study of CAMPATH-1H in previously treated patients with low-grade NHL, 50 patients with relapsed or refractory disease were treated with 30 mg of CAMPATH-1H three times weekly for up to 12 weeks.128 Infection, anaemia and thrombocytopenia were common, and myocardial infarction occurred in one patient with a prior history of angina and congestive heart failure. The overall response rate was 20 per cent (16 per cent partial response, 4 per cent complete response). Responses were short in duration, with a median time to progression of 4 months. Patients with mycosis fungoides responded more frequently and had a longer time to progression (10 months) than patients with low-grade NHL (4 months). Treatment was associated with reactivation of herpes simplex, oral candidiasis, Pneumocystis carinii pneumonia, cytomegalovirus pneumonitis, pulmonary aspergillosis, disseminated tuberculosis, and seven cases of pneumonia and septicaemia. Alemtuzumab is also capable of inducing a minimal residual disease (MRD)-negative remission in patients with relapsed CLL refractory to fludarabine.129,130 One study has examined the relationship between the MRD-negative state and patient survival. Moreton and colleagues treated 91 patients with refractory CLL with a standard dose of alemtuzumab (84 patients received intravenous dosing and the rest subcutaneous) three times a week until a maximum response was achieved, with a goal of achieving am MRDnegative remission as determined by four-colour-flow cytometry.131 Fifty-three per cent of patients (48 patients) in this study responded to treatment and the responses were clearly correlated with the degree of adenopathy, such that patients with less adenopathy had better responses. Also, patients with minimal adenopathy had higher rates of MRD negativity. Overall, 18 patients achieved an MRD-negative state. None had bulky disease and 72 per cent had no lymphadenopathy prior to initiating treatment, although over 40 per cent of them had failed prior treatments. Two patients in this group died of opportunistic infections, but 88 per cent were alive at a median follow-up time of 36 months. The most common adverse event reported for this agent in the trial conducted by Morton was infusion-related toxicity, including rigors and fever. Other non-haematologic toxicity reported included fatigue, dyspnoea and bronchospasm. The most common haematologic toxicity reported was neutropenia.131 There were three deaths due to fungal infections and one due to cytomegalovirus reactivation. Overall, 30 patients (43 per cent) developed an infection more than 40 per cent of patients developed an infection either during therapy or within a month of finishing their treatment.
Solid tumours 233
Vascular-targeting antibodies Tumour-associated blood vessels offer numerous tumourspecific targets for therapy. Several markers have been identified by various groups and these markers provide opportunities for targeted therapy strategies. However, it is likely that great care must be taken when such markers are chosen for targeted therapy. If the target is also expressed by the endothelial cells or by other cells in a vital organ and if the targeting agent is not consumed by targeting the tumour vascular target, the results could be catastrophic. Therefore, regulation of the cytotoxic therapy to avoid excessive host toxicity is important. Vascular targeting has been validated by a number of groups as a viable and effective method of treating solid tumours in mice. One of the earlier vascular-targeting agents (although not an antibody), Combretastatin A-4 (CA4P), had shown significant activity in a number of animal studies.132,133 Combretastatin A-4 is a tubulin-binding agent that inhibits tubulin polymerisation.134 It has limited water solubility, leading to the development of a water-soluble prodrug. In experimental tumours, CA4P causes rapid and extensive vascular destruction. This effect is highly selective and leads to a haemorrhagic necrosis of the tumours and a significant reduction in tumour perfusion within 1 hour of treatment,132 which continued at 3 and 6 hours with residual areas of perfusion at the periphery of the tumours but complete vascular shut-down in the tumour centre.132 By 24 hours, perfusion in the peripheral zone had increased, but the centre of the tumour remained unperfused. Phase I studies of this agent with different treatment schedules and in combination with chemotherapy have shown minimal responses, although a reduction in tumour perfusion was indirectly shown for some of the patients in these trials.135,136 Although the toxicity profile of this agent was reasonable, responses were limited. The clinical results of this strategy have been somewhat disappointing, leading to approaches that provide more targeting specificity. For example, significant treatment efficacy has been observed in large solid tumours using either toxin-linked or tissue-factor-linked vascular targeting137,138 A single-chain antibody fragment against the ED-B domain of fibronectin (scFv L19) has shown specific tumour targeting in a murine tumour model.139 This potential target is expressed in the extracellular matrix around the newly formed blood vessels in a number of solid tumours. Fusion of this single chain with IFN-γ and IL-2 has led to impressive but inconsistent results in therapeutic studies in animal models.140,141 A mAb (3G4) directed against anionic phospholipids on the external membrane of hydrogen-peroxide-treated endothelial cells localizes to the tumour vascular endothelium in scid mice bearing MDAMB-435 breast cancer tumours.142 An average of 40 10 per cent of the vessels were stained with 3G4 in this tumour model. Interestingly, the binding of 3G4 by ezyme-linked immunosorbent assay (ELISA) to anionic phospholipids requires the presence of β2-glycoprotein I. Another intriguing finding from these studies is the demonstration that 3G4
causes monocytes to bind tumour blood vessels and macrophages to infiltrate tumours. It is possible that binding of 3G4 to exposed anionic phospholipids on tumour vessels can stimulate monocyte and macrophage binding via FcγR. A potential limitation of this antibody is that the target is only expressed under hypoxic and other oxidative distress conditions. Thus the concept of antibody-directed vascular targeting has been validated, but new targets are needed to facilitate the optimal clinical development of this strategy.
SOLID TUMOURS EGFR structure and mechanism of signalling The EGFR was initially implicated in cancer by virtue of its tyrosine kinase activity and the discovery of the v-erbB oncogene, a truncated EGFR, in avian erythroblastosis virus.143–145 Two decades of in-depth studies of EGF signalling have recently yielded a new family of antibody and small-molecule therapeutics that are rapidly changing our views of how to treat many types of cancer. The EGFR has four family members (Erb1–4 or, alternatively, Her1–4), of which the EGFR (Her1) and Her2/neu are the most fully characterized. The members of the EGFR family, their principal ligands and nomenclature are summarized in Table 11.2. All EGFR family members consist of a heavily glycosylated extracellular region containing 11 potential glycosylation sites spanning approximately 620 amino acids. A single transmembrane domain (23 residues) is flanked by a so-called juxtamembrane regulatory domain (approximately 40 amino acids) followed by a tyrosine kinase domain (approximately 260 amino acids) and a C-terminal regulatory region of 232 amino acids. Ligands for the EGFR include EGF itself, transforming growth factor-α (TGF-α), amphiregulin, epigenin, epiregulin, betacellulin and heparin-binding EGF-like growth factor (HB-EGF). The EGF and TGF-α, often co-expressed with EGFR,146 are produced by normal-tissue and tumourtissue epithelial cells, with higher levels found in neoplastic tissue and stroma than in surrounding normal mucosa. The EGFR2 (erbB2/Her2/neu) has no known ligands. Neuroregulins (NRGs) serve as ligands for ErbB3 and
Table 11.2 Erb family of receptors and their ligands Erb family receptor ErbB1/ EGFR
Ligand
EGF, TGF-α amphiregulin, epigenin, epiregulin, betacellulin, HB-EGF ErbB2/Her2 None ErbB3, ErbB4 Neuregulins 1–4, heregulin EGFRvIII, de2-7 Much lower affinity
Dimerization partner EGFR/Erb2
Erb2/Her2, ErbB3 ErbB1-4
234 Monoclonal antibodies and treatment of cancer
ErbB4; NRG1, in particular, is implicated in the pathogenesis of breast cancer, while the biological functions of the other three members of the family remain poorly understood.147 Four distinct protein domains comprise the EGFR extracellular region. Domains II and IV (also known as CR1 and CR2, respectively) are cysteine rich, with a number of disulphide bonds: eight disulphide modules in domain II and seven disulphide modules in domain IV. Domains I and III are leucine rich and are binding sites for the growth factor ligands. In studies of different mutant forms of EGFR it became evident that cooperation between domains I and III was necessary for the high-affinity binding of EGF.148 In 1997, Lemmon et al. suggested a model in which ‘one EGF monomer binds to one sEGFR monomer, and that receptor dimerization involves subsequent association of two monomeric (1:1) EGF-sEGFR complexes’.149 The dimer form of EGFR possessed biologically significant tyrosine kinase activity (via the AKT pathway) and was shown to possess transforming activity in keratinocytes.150
Anti-EGFR monoclonal antibodies At least five antibodies targeting different epitopes of EGFR have entered clinical trials. Chimeric and humanized antibodies (e.g. cetuximab or C225, and matuzumab or EMD72000) were originally raised in mice. In order to reduce the immunogenicity of these antibodies, their backbones have been exchanged with those of human immunoglobulin. In contrast, fully human IgG kappa antibodies have been derived employing XenoMouseTM technology. The latter is the result of the insertion of a large portion of the human immunoglobulin gene locus into the mouse genome. In contrast to xenogeneic (i.e. murine or chimerized) antibodies, human or humanized antibodies are less likely to elicit host immune responses that may limit therapy or cause side effects. There are several proposed mechanisms by which antiEGFR mAbs can elicit an anti-tumour response ●
●
● ●
The binding of the EGFR on the cell surface triggers receptor internalization and degradation via the lysosomal pathway, prevents nuclear translocation of DNA-PK and thereby inhibits DNA break repair induced by radiation and DNA-avid chemotherapeutic agents. Interference with EGFR signalling by preventing ligand binding and unfolding of the tethered EGFR to the extended, more active conformation. This is the likely scenario with domain-III-targeting anti-EGFR antibodies such as cetuximab, matuzumab or panitumumab. Antibodies specific to epitopes in domain II of EGFR (the dimerization arm) will cancel EGFR signalling by preventing its dimerization. ADCC, as discussed above. Another mechanism that is less well understood invokes broad interference with signalling events via antibody
disruption of lipid rafts. The EFGR is intimately associated with signalling microplatforms on the cellular membrane.151 High concentration of EGFR ligands or, perhaps, cross-linking with an antibody may shuttle EGFR internalization from the clathrin-mediated pathway of recycling to the caveolin-associated route, leading to receptor ubiquitination and degradation.152 Table 11.3 lists some of the anti-EGFR antibodies that are currently either approved or under investigation.
EGFR-targeting monoclonal antibodies in head and neck cancer CETUXIMAB
Cetuximab (ErbituxTM), a chimeric IgG1 antibody, binds domain III of EGFR, thereby interfering with ligand binding. It is the most extensively studied anti-EGFR mAb and received FDA approval in 2004 for the treatment of patients with metastatic colorectal cancer who are not suitable for or are refractory to irinotecan. Cetuximab binding to EGFR prevents the receptor from adopting an extended conformation, thereby inhibiting EGFR activation.153 Squamous cell carcinoma of head and neck (HNSCC) is a common global cancer that remains lethal for many patients despite contemporary surgery, radiation, and chemotherapy. In 2006 in the USA alone there will be approximately 11 000 deaths from HNSCC and more than 40,000 incident cases.154 Worldwide, the burden of HNSCC is substantial with over 500 000 new cases diagnosed annually.155 The incidence is rising, probably due to the role of environmental exposures, which include tobacco and alcohol; there is also emerging evidence for a role for viruses such as human papillomavirus (HPV).156 Although the mainstay of treatment for stages 1 and 2 disease remains surgical excision, the majority of patients with HNSCC unfortunately present with large tumours (greater than T2), nodal involvement or metastatic disease. With Table 11.3 Monoclonal antibodies (mAbs) targeting the EGFR receptor Antibody
Origin
C225/cetuximab EMD72000/matuzumab ABX-EGF/panitumumab ICR62 806 (EGFRvIII)
Humanized mAb Humanized mAb Human mAb Rat mAb Humanized mAb
hR3/nimotuzumab MDX-447
Humanized mAb Humanized bispecific EGFR/CD3ε
Phase of development II and III II III I Imaging pilot trials I/II I/II
Solid tumours 235
combined therapy, cure is still possible for many of these patients, but the side effects of treatment and the resultant long-term morbidity remain formidable, leaving numerous opportunities for treatment innovations. The HNSCCs are radiosensitive, and radiation therapy plays an integral role in treatment, yielding results comparable to those for surgery in early-stage disease (T1 and T2). Radiotherapy is less effective as a solo modality for intermediate-size tumours and is typically used in an adjuvant fashion following surgery or as a component of induction chemoradiotherapy for more advanced stages of disease.157,158 Patients with stages 3 and 4 disease continue to represent a significant unmet medical need. Current state-of-the-art radiation, surgery and chemotherapy cure only 40–50 per cent of stage 3 and fewer than 20 per cent of stage 4 HNSCC patients. Of note, many patients continue to succumb from cancer that persists or recurs above the clavicles; death from distant metastases occurs in less than one-fifth of patients.156 There have been significant advances in concomitant chemoradiotherapy over the past three decades,159–161 and to date this is the only approach for locally advanced disease that consistently shows improved loco-regional control and a survival advantage (approximately 4 per cent at 5 years), as shown in a recent meta-analysis.162 The improved locoregional control achieved with cisplatin-based chemoradiotherapy regimens is probably attributable to the drug’s ability to damage irreversibly DNA of resting clonogenic cells, thereby sensitizing them to the effects of radiotherapy. Despite these advances, an unacceptable proportion of patients continue to experience loco-regional disease progression or recurrence leading to death. Therefore a better understanding of the disease process and newer treatments are needed. CLINICAL RESULTS OF COMBINATION OF CETUXIMAB AND RADIOTHERAPY FOR LOCO-REGIONALLY ADVANCED HNSCC
The use of cetuximab in this clinical scenario has resulted in a survival benefit, the first time in a randomized clinical trial that a statistically significant survival benefit has been conferred by the use of any EGFR antagonist. The pivotal trial performed by Bonner et al.163 was preceded by a smaller phase I trial.164 in which the safety of co-administration of cetuximab with a course of definitive radiation delivered once or twice daily over 7 weeks was established. In this study, all patients were given a loading dose of antibody 1 week prior to the initiation of radiotherapy. The purpose of the loading dose was to saturate rapidly all EGFR-binding sites, including those in normal liver and skin. After the loading dose, patients received a weekly maintenance dose of 200–250 mg/m2 during radiation therapy. The weekly maintenance dose range was based on an earlier pharmacokinetics (PK) study showing a pattern of non-linear pharmacokinetics consistent with saturation of clearance and relatively stable antibody levels in blood with this pattern of dosing.164,165 The authors estimated the half-life of cetuximab to be around 90 hours.
The 16 patients in that study had unresectable squamous cell carcinomas of the oropharynx and oral cavity (12 patients), larynx and hypopharynx (4 patients). The three evaluated dose levels of cetuximab were 100 mg/m2, 200 mg/m2 and 250 mg/m2 weekly; three subsequent cohorts received loading doses of 400 mg/m2 or 500 mg/m2. The most common toxicities attributable to cetuximab were desquamating skin reaction in the radiation field (grade 3 in five patients) and infusion reactions in four patients (one resulting in discontinuation of protocol therapy). None of these was considered dose limiting. Mucositis and odynophagia occurred in most of the patients, as expected and at the expected intensity with radiotherapy alone. Fourteen of 15 patients evaluable for response had a partial response or complete response, many of which have been durable over a prolonged period of follow-up.164 The pivotal multi-centre trial163,166–171 enrolled 424 patients who underwent randomization to definitive radiotherapy alone (213 patients) or the combination of definitive radiotherapy and weekly cetuximab at a loading dose of 400 mg/m2 followed by 250 mg/m2 for the duration of radiotherapy. Patients were stratified by performance status, tumour stage, nodal involvement and radiotherapy fractionation regimen. The majority of patients on this study were men167 in relatively good health (68 per cent had a Karnofsky score of 90–100). Study entry did not require tumour testing for EGFR expression. Tumour sites were oropharyngeal (60 per cent), larynx (25 per cent) and hypopharynx (15 per cent). The authors reported doubling of the median duration of loco-regional control from 14.9 to 24.4 months (log-rank p 0.005). This result, in particular, validated pre-clinical models in which improvement of regional control in the radiation field was anticipated based on the radiosensitizing and apoptosis-promoting effects of cetuximab. EGFR blockade with radiotherapy reduced the risk of loco-regional failure by 32 per cent (hazard ratio 0.68) and the risk of death by 26 per cent (hazard ratio 0.74, p 0.03). There was a lack of substantial improvement in the loco-regional control of carcinomas of larynx and hypopharynx (median loco-regional control lasting in the range of 10.3–12.9 months). Similarly, overall survival of laryngeal cancer patients improved only marginally, from a median of 31.6 months to 32.8 months, and hypopharyngeal cancers uniformly showed poor outcome, with no change in short median survival (13.5 per cent versus 13.7 per cent with cetuximab). At the same time, the cancers of oropharynx showed an outstanding doubling in median duration of loco-regional control with cetuximab (23 months versus 49 months) and a remarkable improvement in overall survival in the cetuximab arm (median survival not reached at three years). The 3-year survival rates were 54 per cent and 45 per cent in favour of cetuximab and radiotherapy, largely because of the success of the investigational treatment in the oropharyngeal subgroup. Updated survival and toxicity data have been presented showing the durability of the loco-regional control and survival data.164 Of interest, there was no influence of EGFR expression on patient outcome.
236 Monoclonal antibodies and treatment of cancer
COMBINATION OF CISPLATIN, CETUXIMAB AND RADIOTHERAPY IN LOCALLY ADVANCED HNSCC
Platinum compounds are established potentiators of radiation-induced DNA damage in tumour cells. To test the clinical hypothesis that two radiosensitizing agents such as cisplatin and cetuximab would act synergistically, investigators from Memorial Sloan Kettering Cancer Center (MSKCC) reported a phase I/II clinical study exploring the feasibility of cisplatin, cetuximab and radiotherapy in locally advanced HNSCC.172 With median follow-up of 44 months, and all survivors followed for a minimum of 31 months, 3-year overall survival of 76 per cent and 3-year progression-free survival of 59 per cent were reported. This regimen was selected as the experimental arm of the recently activated follow-up study to the Bonner trial. CLINICAL RESULTS WITH CETUXIMAB IN COLORECTAL CANCER
Cetuximab has been evaluated both as a single agent and in combination with irinotecan in phase II clinical trials. In a phase II trial of patients whose tumours exhibited EGFR expression and had failed to respond to irinotecan, five of 57 patients (9 per cent) achieved a partial response, with 21 additional patients having stable disease or minor responses.173 Toxicities included rash (18 per cent with grade 3) and allergic reactions that required drug discontinuation in 3.5 per cent of patients. In a phase II trial of cetuximab versus cetuximab plus irinotecan, in patients whose disease had progressed on irinotecan alone, single-agent cetuximab produced a similar rate of partial response, 10.8 per cent, whereas the irinotecan-cetuximab combination produced a partial response rate of 22.9 per cent.4 The combination also resulted in a statistically significant increase in time to progression (4.1 months versus 1.5 months, p 0.001) and a trend towards improved median survival. Building on these results, a phase II trial has evaluated the combination of cetuximab and bevacizumab with or without irinotecan in irinotecan-refractory colon cancer.174 Cetuximab has also been tested with 5-fluorouracil (5-FU), leucovorin and irinotecan in the first-line setting in phase I/II trials, with response rates of 48–74 per cent.175–177 The combination of cetuximab and FOLFOX-4 is currently being evaluated in phase III trials in the first-line setting. Numerous clinical trials are ongoing assessing the addition of cetuximab to a variety of commonly used regimens in metastatic and stage III colorectal cancer. Each of these studies was preceded by phase I/II data demonstrating that the addition of cetuximab to various chemotherapy regimens was feasible and safe.178–184 A phase II randomized clinical trial conducted by Saltz et al.185 in irinotecan-refractory metastatic colorectal cancer patients showed similar objective response rates whether or not patients were treated with a cytotoxic drug (in that case irinotecan) added to a combination of cetuximab and bevacizumab. Interesting in this context is the report from Kerbel’s group at Sunnybrook Hospital in Toronto,186 which showed an increased level of
VEGF production by the A431 xenografts selected for resistance to EGFR-blocking antibodies. Treatment with antiVEGF antibody resulted in normalization of anomalous vessels in the tumour. Clinical exploration of bevacizumab and EGFR antagonists has been initiated in at least seven clinical trials registered with National Cancer Institute. Cetuximab received accelerated approval from the US FDA in 2004 on the basis of data showing single-agent activity173 and enhanced activity in combination with irinotecan in patients with metastatic colorectal cancer who had failed or were unsuitable for irinotecan-based therapy.4 The benefits of cetuximab in the metastatic, chemotherapy-refractory colorectal cancer setting are modest, conferring only 1.7 months of improvement in survival and a doubling of the partial response rate when given in combination with irinotecan compared to irinotecan alone.4 Given the fact that this study enrolled heavily pre-treated patients who were refractory to 5-FU and irinotecan, however, the result was considered highly encouraging, especially as the addition of cetuximab to irinotecan did not appear to intensify chemotherapy side effects. Other than acneiform rash and the occasional infusion reaction, the regimen was generally well tolerated. INDICATORS OF RESPONSE TO CETUXIMAB
An acneiform or maculopapular rash is a characteristic side effect of anti-EGFR therapies. EGFR plays a role in maintaining the integrity of the skin,187 and is expressed in epidermal and follicular keratinocytes, sebaceous and eccrine epithelia, dendritic antigen-presenting cells and connective-tissue cells. After histological analysis of rash biopsies from ten patients receiving cetuximab, Busam et al. concluded that the rash is characterized by lymphocytic perifolliculitis or suppurative superficial folliculitis, but without an infectious component. Some studies of rash associated with EGFR inhibitors state that the rash is sterile (Van Doorn189,190), whereas others report that micro-organisms are present.188 The sebaceous glands are not affected,190,191 leading to the conclusion reached at a recent EGFRinhibitor rash management forum192 that the rash is not acne vulgaris and does not appear to have an acne-like aetiology, although the exact aetiology is unclear. Skin toxicity has been shown to be significantly associated with response and overall survival in metastatic colorectal cancer patients receiving cetuximab.4 However, rash was not predictive of response among Chung’s group of patients whose tumours were EGFR negative by immunohistochemistry. Of seven patients with tumour response (partial response and stable disease), five had a grade 1–2 rash. However, three patients who progressed on cetuximab/irinotecan also developed rashes.193 At this time, we view rash as a pharmacodynamic marker indicating the presence of significant levels of antibody in the body; such levels are sufficient to mediate anti-tumour effects. However, the mere attainment of potentially inhibitory antibody levels in tumour or surrogate tissues such as skin does not assure clinical response. Clinical
Solid tumours 237
response will require the presence of the relevant target and possibly gene amplification, and a permissive tumour and host environment. Much work remains to be done to define the precise determinants of response to EGFRtargeted mAb therapy. EGFR EXPRESSION AND RESPONSE TO CETUXIMAB
Approximately 70–75 per cent of human colorectal carcinomas express EGFR when assayed by immunohistochemistry. The results of clinical studies suggest that EGFR expression is neither sufficient nor necessary for tumour response to cetuximab. Early clinical trials of cetuximab required EGFR positivity by immunohistochemistry for study entry. Within those studies, no relationship between intensity of EGFR expression and clinical activity was demonstrated.4,173 This led many oncologists to include EGFR-negative colorectal patients from standard off-protocol treatment on the basis of EGFR status alone. Reporting on the experience of 16 such EGFR-negative irinotecan-refractory patients treated with cetuximab/irinotecan at MSKCC, Chung et al.193 noted that seven achieved a degree of tumour control, with four partial responses, two minor responses, and one patient with a more than 50 per cent drop in carcinoembryonic antigen. This study provides confirmation that tumour response can be seen in EGFR-negative (by immunohistochemistry) patients. Further, the 25 per cent response rate in this small study is comparable to the 23–45 per cent response rates seen in two cetuximab-plus-irinotecan clinical trials in EGFR-positive patients.4,173 Vallbohmer et al.194 analysed mRNA levels of enzymes involved in the EGFR signalling pathway to determine whether activation of this pathway correlates with clinical response to cetuximab. Thirty-nine patients with metastatic colorectal cancer refractory to both irinotecan and oxaliplatin were treated with single-agent cetuximab, with intra-tumoral mRNA levels of CCND1, Cox-2, EGFR, IL-8 and VEGF assessed from paraffin-embedded tissue samples. All patients had immunohistochemistry evidence of EGFR expression. Only two patients had partial responses, limiting the robustness of the data, although 21 additional patients had stable disease. In this study, only low intratumoral gene expression of VEGF was associated with response to cetuximab therapy, independent of skin toxicity. Ciardello et al. demonstrated a dose-dependent inhibition of VEGF with cetuximab therapy.195 These findings suggest a possible role of intra-tumoral VEGF levels in determining the response to cetuximab. They are consistent with the findings in other studies186 of increased expression and secretion of VEGF in tumour cells with acquired resistance to cetuximab. In an important recent study, Moroni et al.196 evaluated nine colorectal cancer patients with EGFR-expressing tumours who had a response to treatment with cetuximab or panitumumab. They found the mutational status of the EGFR catalytic domain (exons 18, 19 and 21) and its immediate downstream effectors, IK3CA, KRAS and BRAF, did not correlate with disease response. However,
eight of nine patients with objective clinical responses were found to have an EGFR copy number of three or more as determined by fluorescent in-situ hybridization (FISH) performed on tumour samples. By contrast, only one of 20 non-responders had an increased EGFR copy number. These findings suggest that, as in breast cancers sensitive to trastuzumab, anti-EGFR mAbs are more likely to work against amplified rather than mutated targets. They do not explain the response of EGFR-negative colorectal tumours to anti-EGFR mAbs described by Chung, however. Further studies will be required to resolve this apparent contradiction. However, the gene amplification explanation is scientifically plausible and offers the possibility that target populations can be enriched to maximize the odds that antibody therapy will be useful. The question of why EGFR-negative tumours would respond to an anti-EGFR mAb remains unanswered for now. Some studies have reported variability in EGFR staining depending on how the tissue has been processed or stored. For example, in one study, a decreasing EGFR staining intensity was seen with increasing storage time of tissue samples.197 Thus a tumour could appear falsely negative for EGFR expression. Because DNA is a generally more stable substance than protein, EGFR copy number assays may have predictive value, even in cases in which tumour expression of EGFR was thought to be negative.
Panitumumab Panitumumab (ABX-EGF) is a fully human IgG2 mAb to EGFR that, similarly to cetuximab, blocks ligand binding by the EGFR and receptor activation. Note that the IgG2 class of antibodies lacks the ability to induce activation of the immune system cell via the FcR mechanism, which is deemed an important component in the overall effect of antibody targeting of cancer cells. In recently released, updated results from a phase I clinical trial, 96 patients, including 39 colorectal cancer patients, with at least 1 + expression of EGFR, were treated with ABX-EGF. Grade 3 skin-related toxicities occurred in 7 per cent of patients, but no maximum tolerated dose was reached, and no infusion-related reactions were observed. Partial responses were observed in 5/39 patients with colorectal cancer, and stable disease was observed in an additional 18 patients. A phase II trial is currently investigating ABX-EGF monotherapy in subjects with metastatic colorectal cancer whose tumours express low or negative EGFR levels following treatment with fluoropyrimidine, irinotecan and oxaliplatin chemotherapy.
Antibody for therapy of Her2-positive malignancies The ErbB2/Her2/neu protein is amplified in a number of malignancies, including about 30 per cent of breast cancers. It confers significantly worse prognosis and identifies
238 Monoclonal antibodies and treatment of cancer
a specific subset of the breast adenocarcinomas distinct from basal-like, normal-breast-like and luminal epithelial/ER.198 The Her2 over-expressing breast carcinomas tend to have higher proliferative markers, metastasize early and are more responsive to chemotherapy with taxanes (paclitaxel or docetaxel).199 Recent groundbreaking basic research led to better understanding of the biological function of Her2 and to the development of strategies of interference with its role in oncogenic phenotype. HER2 is a transmembrane protein of the Erb family and is active in a homodimeric or heterodimeric form, as discussed earlier. The preferred dimerization partners of Her2 are Her3/EbrB3, Her4/ErbB4 and EGFR/ErbB1 (see Table 11.2). In normal cells, few HER2 molecules exist at the cell surface, emanating only limited background ‘noise’ of growth signals. Quite strikingly, when HER2 is over-expressed, multiple HER2 heterodimers are formed and cell signalling is stronger, resulting in enhanced responsiveness to growth factors and malignant growth. The unique stimulatory properties of Her2 are conferred by its constitutive exposure of the dimerization arm of the second domain of the molecule. This fixed conformation of HER2 resembles a ligand-activated state, making HER2 poised to interact with other ErbB receptors in the absence of direct ligand binding.200 Besides breast cancer, a number of other epithelial malignancies have been shown to over-express HER2 protein.201
Trastuzumab METASTATIC BREAST CANCER
The recombinant humanized anti-HER2 mAb (rhuMAbHER2, trastuzumab, Herceptin) induces rapid removal of HER2 from the cell surface, thereby reducing its availability to heterodimers and reducing oncogenicity.202 Herceptin binds to the juxtamembrane region of HER2, identifying this site as a target for anti-cancer therapies. Its clinical use in metastatic breast cancer patients revealed modest activity as a single agent, with a response rate of 11.6 per cent reported in a pilot study of 46 heavily pretreated patients.203 In-vitro and xenograft studies of MCF-7 breast carcinoma over-expressing HER2 revealed synergistic interaction of trastuzumab with alkylating agents, platinum analogues and topoisomerase II inhibitors, and additive interaction with taxanes, anthracyclines and some anti-metabolites.204 These rational combinations laid the foundation of human clinical trials of trastuzumab.2,199 In the pivotal randomized clinical trial, patients with immunohistochemically 2+/3+ HER2-positive metastatic breast cancer were assigned to receive six cycles of chemotherapy with or without weekly trastuzumab. The choice of chemotherapy was based on prior exposure to anthracyclines, i.e. anthracycline-naive patients were treated with cyclophosphamide/doxorubicin or epirubicin combination (143 in the trastuzumab arm and 138 in the
chemotherapy-only arm), while anthracycline-exposed patients received paclitaxel chemotherapy alone (96 women) or with trastuzumab (92 women). Regardless of chemotherapy treatment, trastuzumab led to nearly doubling of time to progression (median 7.4 months versus 4.6 months; p 0.001), higher rate of objective response (50 per cent versus 36 per cent), and better median survival by almost 5 months (25.1 months versus 20.3 months; p 0.046). As expected, cardiotoxicity was a prominent side effect, with New York Heart Association class III/IV severity cardiac failure seen in 16 per cent of patients in the anthracycline/ cyclophosphamide/trastuzumab arm and in only 1 to 3% in the remaining three study groups. Treatment with trastuzumab was otherwise well tolerated. The second trastuzumab clinical trial was conducted in Europe. Patients (a total of 186) were randomly assigned to six cycles of docetaxel 100 mg/m2 every 3 weeks, with or without trastuzumab 4 mg/kg loading dose, followed by 2 mg/kg weekly until disease progression. Ninety-seven per cent of patients had HER2/neu gene amplification confirmed by FISH or 3 reactivity by immunohistochemistry. Trastuzumab plus docetaxel was significantly superior to docetaxel alone in terms of overall response rate (61 per cent versus 34 per cent; p 0.0002), overall survival (median 31.2 months versus 22.7 months; p 0.0325), time to disease progression (median 11.7 months versus 6.1 months; p 0.0001), time to treatment failure (median 9.8 months versus 5.3 months; p 0.0001) and duration of response (median 11.7 months versus 5.7 months; p 0.009). Investigators in this trial conducted close monitoring of their patients’ cardiac performance by determining ejection fraction every third cycle of treatment. With these precautions, only one patient in the combination arm experienced symptomatic heart failure (1 per cent), while two other patients died of progressive disease, although cardiac failure could not be excluded.205 Attempts to incorporate trastuzumab into chemotherapeutic regimens for the treatment of lung cancer, prostate or colorectal cancer have not yielded positive results, primarily due to the low incidence of HER2/neu gene amplification in these.206,207 Adjuvant therapy In those women who undergo curative surgery, chemotherapy is given to eradicate micro-metastatic disease that is responsible for distant recurrences of breast cancer. Several major clinical trials have reported exciting results.208,209 for the European HERA (Herceptin Adjuvant) trial and combined results of NSABP B-31 (National Surgical Adjuvant Breast and Bowel Project) and North Central Cancer Treatment Group trial N9831. All three trials included patients operated for breast cancer that tested 3 by immunohistochemistry or showed HER2 gene amplification by FISH. The North American trials included 94 per cent node-positive patients, while in the HERA trial 30 per cent of participants were node negative. After a brief observation of 1–2.5 years, the absolute benefit for disease-free
Solid tumours 239
survival was 8 per cent in the North American trial and 6 per cent in the HERA trial at 2 years. This benefit would project to 18 per cent at 4 years. Analysis of disease-free survival curves indicates an early separation in favour of trastuzumab. These findings indicate that trastuzumab therapy dramatically alters the natural history of this disease in appropriately selected patients. Docetaxel and the platinum salts are logical candidates to be combined with trastuzumab since these agents exhibit potent synergy with the antibody in pre-clinical experiments. Furthermore, the two phase II clinical trials conducted by Breast Cancer International Research Group and the University of California at Los Angeles–Oncology Research Network using the taxotere/carboplatin/herceptin (TCH) (docetaxel/platinum/trastuzumab) regimen suggest this combination has significant activity, with response rates in these two studies reported as 59 per cent and 78 per cent. The BCIRG006 trial is a three-arm adjuvant study, comparing doxorubicin/cyclophosphamide followed by docetaxel (AC-T), the same regimen with trastuzumab administered with docetaxel (TH), and TCH, in 3150 women with nodepositive or high-risk node-negative, HER2-positive breast cancer. BCIRG 007 compares TH and TCH as first-line therapy in patients with HER2-positive metastatic breast cancer. In both trials, entry is restricted to patients whose tumours are positive for HER2 gene amplification as determined by FISH. The results of the first interim analysis of the BCIRG006 adjuvant trial were presented at the 2005 San Antonio Breast Cancer Symposium. More than 1000 women enrolled in each of the three arms of this trial, with median follow-up of 23 months and 322 disease-free survival events recorded (84 deaths). Twenty-nine per cent of the patients were node negative (comparable with the HERA trial population). At 4 years, the disease-free survival was estimated to reach 73 per cent in the AC-T, 80 per cent in the TCH and 84 per cent in the AC-TH arms. The number of events in the TCH and AC-TH arms (98 and 77, respectively) did not reach a statistically significant difference (p 0.16), indicating equivalency of the two regimens. One of the hypotheses tested in this trial was the possibility of reduction of cardiac events with the non-anthracyclinecontaining TCH regimen compared to AC-TH. The total number of grade 3 and 4 cardiac events, including arrhythmias, was higher in AC-TH (n 25) than in TCH (n 14), but not statistically significant (p 0.11). This clinical trial also provided deeper insight in the genetic events associated with HER2/neu locus amplification on chromosome 17. The adjacent, locus 17q21.2 contains the gene for Topo II. Patients who did not have co-amplification of this gene with HER2/neu locus (65 per cent) fared significantly worse overall and showed poor survival when treated without trastuzumab . Future clinical research employing trastuzumab– chemotherapy combinations will probably be directed towards the identification of genetic profiles and
chromosomal rearrangements associated with poor response to HER2 blockade and to the identification of the ‘escape mechanisms’ that cause treatment resistance. The minimum necessary duration of trastuzumab treatment has not been defined. The HERA trial unambiguously showed equivalency of 1 and 2 years of therapy. The data from several neoadjuvant trials 210–212 as well as a recently presented Finnish adjuvant study 213 indicate that even a short course of trastuzumab confers a significant treatment benefit for Her2-positive breast cancer. In the latter trial,213 adjuvant 9-week trastuzumab was effective in preventing any recurrence (hazard ratio 0.46, p 0.0078). Three-year distant disease-free survival of patients who received trastuzumab was 93 per cent, and that of patients who did not receive trastuzumab was 76 per cent (p 0.0078, 11/115 versus 26/116 events, hazard ratio 0.43). There is a paucity of data to suggest any indication for trastuzumab therapy in highly curable node-negative breast cancer. The subset analysis of the HERA and BCIRG006 trials may shed some light on this issue as the follow-up data mature. PERTUZUMAB
In many human cancers, Her2 is detectable on the cell surface without gene amplification. The molecular structure of Her2 allows it to exhibit strong tyrosine kinase activity in the absence of any extracellular ligand. Its dimerization arm of domain II (also known as CR1) is constitutively exposed and ‘ready’ to dimerize with any activated isoform of an ErbB family member in physical proximity. Thus Her2 is capable of retaining EGFR on the cell surface and augmenting the signalling cascade initiated by EGFR activation.214,215 Moreover, an interaction between insulinlike growth factor type I receptor (IGF-IR) and Her2 has been described by Nahta et al.216 The investigators demonstrated the reversal of trastuzumab resistance in the SKBR3 cell line by combining trastuzumab with either antibody to IGF-IR or a novel antibody, pertuzumab, which targets the Her2 dimerization arm. Interestingly, Her2 blockade with trastuzumab in the sensitive cell lines is associated with recruitment of the inhibitory molecule PTEN, which in turn results in reduced Akt phosphorylation.215 Clearly, the mechanism of Her2-driven resistance to apoptosis and proliferation is highly complex and involves interaction with other growth factor tyrosine kinase receptors and complex trafficking of Her2 to and from the cellular membrane. Pertuzumab, a ‘dimerization inhibitor’, was demonstrated in pre-clinical studies to have a growth-inhibitory effect on breast, prostate and non-small-cell lung cancer cell lines expressing varying levels of Her2. In phase I clinical trials, pertuzumab showed activity in a number of human cancers. Agus et al.217 conducted a dose-escalation phase I clinical trial of pertuzumab in 21 patients, of whom 19 completed at least two cycles. Pertuzumab was well tolerated. The pharmacokinetics of pertuzumab
240 Monoclonal antibodies and treatment of cancer
were similar to those of other humanized IgG antibodies, supporting a 3-week dosing regimen. With good tolerance and favourable pharmacokinetics allowing 3-week dosing, pertuzumab is undergoing more advanced clinical evaluation. BEVACIZUMAB
Malignant tumour progression requires a blood supply capable of delivering essential nutrients and oxygen to the cells. Blood vessels consist of endothelial cells, smooth muscle cells and other supporting cells and are responsible for the transport of these nutrients. Numerous researchers have investigated the process of tumour angiogenesis with a goal of manipulating the host vascular response to tumours based on the pioneering work of Folkman and others.218 The blockade of tumour vessels deprives malignant cells of essential nutrients and causes tumour destruction. Since one vessel supplies a vast number of cancer cells, this approach can amplify the consequences of an initial cytotoxic insult. Moreover, this strategy does not require the penetration of other cells or molecules into the tumour parenchyma. Bevacizumab is a mAb targeting VEGF. Of the identified angiogenic factors, VEGF is the most potent and specific regulator of both normal and pathologic angiogenesis.219 It produces a number of biological effects, including endothelial cell mitogenesis and migration, induction of proteinases leading to remodelling of the extracellular matrix, increased vascular permeability, and maintenance of survival for newly formed blood vessels.219 The biological effects of VEGF are mediated through the binding and stimulation of two receptors on the surface of endothelial cells: Flt-1 (fms-like tyrosine kinase) and KDR (kinase domain region).219,220 Increased expression of VEGF has been demonstrated in most human tumours examined to date, including tumours of the lung, breast, thyroid, gastrointestinal tract, kidney, bladder, ovary and cervix, as well as angiosarcomas and glioblastomas.219 Inhibition of VEGF by using an antiVEGF mAb blocks the growth of a number of human cancer cell lines in nude mice.219 In addition, the combination of anti-VEGF antibody and chemotherapy in nude mice injected with human cancer xenografts results in an increased anti-tumour effect compared with antibody or chemotherapy treatment alone.221 Bevacizumab (previously known as rhuMAb VEGF) is a recombinant humanized version of a murine anti–human VEGF mAb.222 Approximately 93 per cent of the amino-acid sequence, including most of the antibody framework, is derived from human IgG, and 7 per cent of the sequence is derived from the murine antibody. Bevacizumab has been studied in at least 3500 patients in a number of phase I, II and III clinical trials. These clinical trials have included patients with a number of tumour types, including colorectal, breast, lung and renal carcinoma.5,223–225
Metastatic colorectal cancer In a large phase III study (AVF2107g) in patients with metastatic colorectal cancer, the addition of bevacizumab to irinotecan/5-FU/leucovorin (IFL) chemotherapy resulted in a clinically and statistically significant increase in duration of survival, with a hazard ratio of death of 0.660 (median survival 15.6 months versus 20.3 months; p 0.0001). Similar increases were seen in progression-free survival (6.2 months versus 10.6 months; p 0.0001), overall response rate (35 per cent versus 45 per cent; p 0.0029) and duration of response (7.1 months versus 10.4 months; p 0.0014) for the combination arm versus the chemotherapy-only arm.5 Based on this survival advantage, bevacizumab was designated for priority review and was approved in the USA for first-line treatment in combination with intravenous 5-FU-based chemotherapy for patients with metastatic colorectal cancer.5 One of the pivotal studies that addressed the nature of the biological effects of bevacizumab in patients with rectal cancer was reported by Willet et al.226 The authors provided evidence of rapid changes in metabolic activity and reduction of vascularization, microvascular density, interstitial fluid pressure and the number of viable circulating endothelial and progenitor cells. This study also paved the way for the rational combination of bevacizumab with chemotherapy, since the blockade of VEGF led to increases in the fraction of vessels with pericyte coverage. This ‘normalization’ of leaky vessels in the tumour bed reduces interstitial oncotic pressure and hence should improve the diffusion of the chemotherapeutic drugs within tumour masses. Preliminary results from a large, randomized clinical trial (ECOG 3200) for patients with advanced colorectal cancer who had previously received treatment were recently released. This trial offered a different choice of chemotherapy. Patients with previously treated metastatic colorectal cancer who received bevacizumab in combination with FOLFOX4 (a regimen of oxaliplatin, 5-FU and leucovorin) had a median overall survival of 12.5 months, which was significantly better than the 10.7 month median overall survival in patients treated with FOLFOX4 alone. There was a 26 per cent reduction in the risk of death for patients in this study who received bevacizumab plus FOLFOX4 compared to those who received FOLFOX4 alone.
Metastatic non-small-cell lung cancer In non-small-cell lung cancer, bevacizumab plus carboplatin and paclitaxel was tested in a phase II trial in patients with advanced or recurrent disease.227 In the three arms of this study, 99 patients were randomly assigned to bevacizumab 7.5 mg/kg (n 32) or 15 mg/kg (n 35) plus carboplatin and paclitaxel every 3 weeks or carboplatin and paclitaxel alone (n 32). Compared with the control arm, the addition of bevacizumab resulted in a dose-dependent increase in the response rate (31.3 per cent versus 40 per cent, as per
Solid tumours 241
independent review), longer median time to progression (7.4 months versus 4.2 months) and a modest increase in survival (17.7 months versus 14.9 months). Unexpectedly, patients treated with the lower dose of bevacizumab had worse response rates, time to progression and survival than those in the control arm. These differences can be attributed to the small number of patients in the study and the unusually long median survival of patients in the control arm, exceeding 1 year (14.9 months). Also, 19 of the 32 control patients crossed over to single-agent bevacizumab at 15 mg/kg at the time of disease progression and five maintained stable disease for more than 6 months. Bleeding was the most prominent adverse event and was manifested in two distinct clinical patterns: minor mucocutaneous haemorrhage and major haemoptysis. Major haemoptysis was associated with squamous cell histology, tumour necrosis and cavitation, and disease location close to major blood vessels. Patients with centrally located tumours and squamous cell histology were excluded from the phase III trial due to the tendency of these cancers to bleed. The results of a randomized phase II/III trial (E4599) of paclitaxel and carboplatin with or without bevacizumab in patients with advanced non-squamous, non-small-cell lung cancer were reported by the ECOG at the 2005 American Society of Clinical Oncology (ASCO) Annual Meeting.228 A total of 878 patients were randomized to receive paclitaxel plus carboplatin or the same chemotherapy plus bevacizumab (15 mg/kg) every 3 weeks. Chemotherapy was continued up to six cycles; patients in the experimental arm received single-agent bevacizumab after the six cycles of chemotherapy until progressive disease or intolerable toxicity. Patients in the chemotherapy-only arm were not allowed to cross over to bevacizumab. The results of the second interim analysis were reported after 469 (72.2 per cent) of the 650 deaths required for final analysis had occurred. There was a significant advantage for patients in the bevacizumab arm in terms of median survival (12.5 months versus 10.2 months; p 0.0075). In addition, patients treated with bevacizumab had a significantly higher response rate (27 per cent versus 10 per cent; p 0.0001) and a significantly longer progression-free survival time (6.4 months versus 4.5 months; p 0.0001). Both regimens were well tolerated; a higher incidence of bleeding was associated with bevacizumab administration (4.5 per cent versus 0.7 per cent). Five of ten treatment-related deaths occurred as a result of haemoptysis, all in the experimental arm.
Breast cancer The safety and efficacy of bevacizumab in patients with previously treated metastatic breast cancer were evaluated in a phase I/II trial.229 Seventy-five patients were treated with escalating doses of bevacizumab ranging from 3 mg/kg to 20 mg/kg administered intravenously every other week. Tumour response was assessed before the sixth (70 days)
and twelfth (154 days) doses. Safety was evaluated during every cycle. Eighteen patients were treated at 3 mg/kg, 41 at 10 mg/kg, and 16 at 20 mg/kg. Four patients discontinued study treatment because of an adverse event. Hypertension was reported as an adverse event in 17 patients (23 per cent). This study demonstrated that bevacizumab as a single agent has minimal activity: the overall response rate was 9.3 per cent (confirmed response rate 6.7 per cent). The optimal dose of bevacizumab in this trial was 10 mg/kg every other week and toxicity was acceptable. Miller and colleagues230 reported the results of a randomized phase III trial of capecitabine/bevacizumab versus capecitabine alone in 462 metastatic breast cancer patients pre-treated with anthracycline and taxane. The primary endpoint of this study, progression-free survival, did not show a statistically significant difference (4.86 months versus 4.17 months; hazard ratio 0.98). Combination therapy significantly increased the response rates (19.8 per cent versus 9.1 per cent; p 0.001). Overall survival (15.1 months versus 14.5 months) and time to deterioration in quality of life were comparable in both treatment groups. Further proof of efficacy of bevacizumab in the treatment of metastatic breast cancer comes form the preliminary report of the ECOG study 2100, reported at the San Antonio Breast Cancer Symposium in 2005.231 In this randomized phase III trial, 722 patients were randomly assigned to receive weekly paclitaxel either alone or with bevacizumab. The primary end point of this study was progression-free survival (PFS). Combination therapy lead to a significantly prolonged PFS (10.97 months versus 6.11 months, hazard ratio 0.498). Toxicities included hypertension and proteinuria, as expected form bevacizumab, and neuropathy. The data on overall survival is pending. TOXICITY
Four major bevacizumab-associated toxicities have been identified: hypertension, proteinuria, thromboembolic events and haemorrhage. Proteinuria, ranging from asymptomatic and transient events detected on routine dipstick urinalysis to nephrotic syndrome, has been seen in all clinical trials to date. The majority of proteinuria events have been grade 1 or 2. In the phase III pivotal trial in metastatic colorectal cancer, the rate of grade 3 or greater proteinuria by National Cancer Institute’s Common Terminology Criteria for Adverse Events, version 3.0 (NCI CTCAE Scale v3 ) was 1 per cent in both treatment arms. Venous and arterial thromboembolic events, ranging in severity from catheter-associated phlebitis to fatal, have been reported in patients treated with bevacizumab in the colorectal cancer trials and, to a lesser extent, in patients treated with bevacizumab in non-small-cell lung cancer and breast cancer trials. In the phase III pivotal trial in metastatic colorectal cancer, there was a slightly higher rate of venous thromboembolic events that was not statistically significant in the patients on the bevacizumab treatment
242 Monoclonal antibodies and treatment of cancer
arm (16 per cent versus 19 per cent). There was also a higher rate of arterial thromboembolic events (1 per cent versus 3 per cent) such as myocardial infarction, transient ischaemic attack, cerebrovascular accident/stroke and angina/unstable angina. A pooled analysis of the rate of arterial thromboembolic events from five randomized studies (1745 patients) showed that treatment with bevacizumab increased the risk of having an arterial thromboembolic event from 1.9 per cent to 4.4 per cent. Furthermore, certain baseline characteristics conferred additional risk, specifically age 65 years and a history of a prior arterial thromboembolic event. Gastrointestinal perforation was not seen in the bevacizumab phase I or II clinical trials; however, in the phase III colorectal trial, gastrointestinal perforation and wound dehiscence, complicated by intra-abdominal abscesses, occurred at an increased incidence in patients receiving bevacizumab. These events varied in type and severity, ranging from free air seen on kidney, ureter and bladder Xray, which resolved without treatment, to a colonic perforation with abdominal abscess, which was fatal. In a phase II non-small-cell lung cancer trial, six of 66 bevacizumab-treated patients experienced life-threatening haemoptysis or haematemesis. Four of these events were fatal. Centrally located lesions that were necrotic or cavitary and squamous-cell histology were identified as possible risk factors for bleeding. Bevacizumab increases the risk of congestive heart failure (CHF) in patients with a history of or concurrent anthracycline exposure. Prior radiation therapy to the chest wall may also increase the risk of CHF in these patients. There are no data to suggest that bevacizumab increases the risk of CHF in patients without exposure to anthracyclines.
KEY LEARNING POINTS ●
●
●
●
Antibodies have emerged as important therapeutic vehicles in a number of cancers. Future investigations are likely to define additional therapeutic targets, and will exploit a variety of anti-tumour mechanisms. These efforts will be abetted by advances in antibody engineering. While virtually all currently approved antibodies require the presence of the antibody target on the cell surface, in the tumour stroma or in the circulating blood, it is likely that continued progress in the area of immunotoxins will lead to important new directions for antibody therapy. Treatment with antibodies has been most successful when combined with other chemotherapeutic agents. Antibodies, however, in general have a favourable side effect profile and are tolerated well and do not significantly change the toxicities of chemotherapy regimens.
REFERENCES 1 McLaughlin P, Grillo-Lopez AJ, Link BK, et al. Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolent lymphoma: half of patients respond to a four-dose treatment program. J Clin Oncol 1998; 16(8):2825–33. 2 Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344(11):783–92. 3 Rowinsky EK, Schwartz GH, Gollob JA, et al. Safety, pharmacokinetics, and activity of ABX-EGF, a fully human anti-epidermal growth factor receptor monoclonal antibody in patients with metastatic renal cell cancer. J Clin Oncol 2004; 22(15):3003–15. 4 Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecanrefractory metastatic colorectal cancer. N Engl J Med 2004; 351(4):337–45. 5 Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004; 350(23):2335–42. 6 Kohler G, Milstein C. Derivation of specific antibodyproducing tissue culture and tumor lines by cell fusion. European Journal Of Immunology 1976; 6(7):511–9. 7 Kohler G, Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 1975; 256(5517):495–7. 8 Brekke OH, Sandlie I. Therapeutic antibodies for human diseases at the dawn of the twenty-first century. Nat Rev Drug Discov 2003; 2(1):52–62. 9 Mirick GR, Bradt BM, Denardo SJ, Denardo GL. A review of human anti-globulin antibody (HAGA, HAMA, HACA, HAHA) responses to monoclonal antibodies. Not four letter words. Q J Nucl Med Mol Imaging 2004; 48(4):251–7. 10 Kim JA. Targeted therapies for the treatment of cancer. Am J Surg 2003; 186(3):264–8. 11 Trail PA, King HD, Dubowchik GM. Monoclonal antibody drug immunoconjugates for targeted treatment of cancer. Cancer Immunol Immunother 2003; 52(5):328–37. 12 Khazaeli MB, Conry RM, LoBuglio AF. Human immune response to monoclonal antibodies. J Immunother 1994; 15(1):42–52. 13 Benhar I, Azriel R, Nahary L, et al. Highly efficient selection of phage antibodies mediated by display of antigen as LppOmpA’ fusions on live bacteria. J Mol Biol 2000; 301(4):893–904. 14 Hoogenboom HR. Overview of antibody phage-display technology and its applications. Methods Mol Biol 2002; 178:1–37. 15 Worn A, Pluckthun A. Stability engineering of antibody singlechain Fv fragments. J Mol Biol 2001; 305(5):989–1010. 16 Irving RA, Coia G, Roberts A, Nuttall SD, Hudson PJ. Ribosome display and affinity maturation: from antibodies to single V-domains and steps towards cancer therapeutics. J Immunol Methods 2001; 248(1-2):31–45. 17 Bird RE, Hardman KD, Jacobson JW, et al. Single-chain antigen-binding proteins. Science 1988; 242(4877):423–6.
References 243
18 Reiter Y, Brinkmann U, Lee B, Pastan I. Engineering antibody Fv fragments for cancer detection and therapy: disulfidestabilized Fv fragments. Nat Biotechnol 1996; 14(10):1239–45. 19 Todorovska A, Roovers RC, Dolezal O, Kortt AA, Hoogenboom HR, Hudson PJ. Design and application of diabodies, triabodies and tetrabodies for cancer targeting. Journal of Immunological Methods 2001; 248(1–2):47–66. 20 Pluckthun A, Pack P. New protein engineering approaches to multivalent and bispecific antibody fragments. Immunotechnology 1997; 3(2):83–105. 21 Le Gall F, Kipriyanov SM, Moldenhauer G, Little M. Di-, triand tetrameric single chain Fv antibody fragments against human CD19: effect of valency on cell binding. FEBS Lett 1999; 453(1–2):164–8. 22 Iannello A, Ahmad A. Role of antibody-dependent cellmediated cytotoxicity in the efficacy of therapeutic anticancer monoclonal antibodies. Cancer Metastasis Rev 2005; 24(4):487–99. 23 Gelderman KA, Tomlinson S, Ross GD, Gorter A. Complement function in mAb-mediated cancer immunotherapy. Trends Immunol 2004; 25(3):158–64. 24 Schmidt KV, Wood BA. Trends in cancer therapy: role of monoclonal antibodies. Semin Oncol Nurs 2003; 19(3):169–79. 25 Farah RA, Clinchy B, Herrera L, Vitetta ES. The development of monoclonal antibodies for the therapy of cancer. Crit Rev Eukaryot Gene Expr 1998; 8(3–4):321–56. 26 Goldenberg DM. Challenges to the therapy of cancer with monoclonal antibodies. J Natl Cancer Inst 1991; 83(2):78–9. 27 Weiner LM, Carter P. Tunable antibodies. Nat Biotechnol 2005; 23(5):556–7. 28 Wu AM, Senter PD. Arming antibodies: prospects and challenges for immunoconjugates. Nat Biotechnol 2005; 23(9):1137–46. 29 Ahmad A, Menezes J. Antibody-dependent cellular cytotoxicity in HIV infections. Faseb J 1996; 10(2):258–66. 30 Steplewski Z, Lubeck MD, Koprowski H. Human macrophages armed with murine immunoglobulin G2a antibodies to tumors destroy human cancer cells. Science 1983; 221(4613):865–7. 31 Berken A, Benacerraf B. Properties of antibodies cytophilic for macrophages. J Exp Med 1966; 123(1):119–44. 32 Ernst LK, Metes D, Herberman RB, Morel PA. Allelic polymorphisms in the FcgammaRIIC gene can influence its function on normal human natural killer cells. J Mol Med 2002; 80(4):248–57. 33 Metes D, Gambotto AA, Nellis J, et al. Identification of the CD32/FcgammaRIIc-Q13/STP13 polymorphism using an allele-specific restriction enzyme digestion assay. J Immunol Methods 2001; 258(1–2):85–95. 34 Morel PA, Ernst LK, Metes D. Functional CD32 molecules on human NK cells. Leuk Lymphoma 1999; 35(1–2):47–56. 35 O’ Hanlon LH. Natural born killers: NK cells drafted into the cancer fight. J Natl Cancer Inst 2004; 96(9):651–3. 36 Whiteside TL, Herberman RB. The role of natural killer cells in immune surveillance of cancer. Curr Opin Immunol 1995; 7(5):704–10.
37 Clynes RA, Towers TL, Presta LG, Ravetch JV. Inhibitory Fc receptors modulate in vivo cytoxicity against tumor targets. Nat Med 2000; 6(4):443–6. 38 Weng WK, Levy R. Two immunoglobulin G fragment C receptor polymorphisms independently predict response to rituximab in patients with follicular lymphoma. J Clin Oncol 2003; 21(21):3940–7. 39 Hayes RJ, Bentzien J, Ary ML, et al. Combining computational and experimental screening for rapid optimization of protein properties. Proc Natl Acad Sci U S A 2002; 99(25):15926–31. 40 Umana P, Jean-Mairet J, Moudry R, Amstutz H, Bailey JE. Engineered glycoforms of an antineuroblastoma IgG1 with optimized antibody-dependent cellular cytotoxic activity. Nat Biotechnol 1999; 17(2):176–80. 41 Rafiq K, Bergtold A, Clynes R. Immune complex-mediated antigen presentation induces tumor immunity. J Clin Invest 2002; 110(1):71–9. 42 Gorter A, Meri S. Immune evasion of tumor cells using membrane-bound complement regulatory proteins. Immunol Today 1999; 20(12):576–82. 43 Trikha M, Yan L, Nakada MT. Monoclonal antibodies as therapeutics in oncology. Curr Opin Biotechnol 2002; 13(6):609–14. 44 Velders MP, Litvinov SV, Warnaar SO, et al. New chimeric anti-pancarcinoma monoclonal antibody with superior cytotoxicity-mediating potency. Cancer Res 1994; 54(7):1753–9. 45 Spiridon CI, Ghetie MA, Uhr J, et al. Targeting multiple Her-2 epitopes with monoclonal antibodies results in improved antigrowth activity of a human breast cancer cell line in vitro and in vivo. Clin Cancer Res 2002; 8(6):1720–30. 46 Kroesen BJ, McLaughlin PM, Schuilenga-Hut PH, et al. Tumortargeted immune complex formation: effects on myeloid cell activation and tumor-directed immune cell migration. Int J Cancer 2002; 98(6):857–63. 47 Sokoloff MH, Nardin A, Solga MD, et al. Targeting of cancer cells with monoclonal antibodies specific for C3b(i). Cancer Immunol Immunother 2000; 49(10):551–62. 48 Reiter Y, Fishelson Z. Targeting of complement to tumor cells by heteroconjugates composed of antibodies and of the complement component C3b. J Immunol 1989; 142(8):2771–7. 49 Juhl H, Petrella EC, Cheung NK, Bredehorst R, Vogel CW. Complement killing of human neuroblastoma cells: a cytotoxic monoclonal antibody and its F(ab’)2–cobra venom factor conjugate are equally cytotoxic. Mol Immunol 1990; 27(10):957–64. 50 Cheung NK, Modak S, Vickers A, Knuckles B. Orally administered beta-glucans enhance anti-tumor effects of monoclonal antibodies. Cancer Immunol Immunother 2002; 51(10):557–64. 51 Blok VT, Gelderman KA, Tijsma OH, Daha MR, Gorter A. Cytokines affect resistance of human renal tumour cells to complement-mediated injury. Scand J Immunol 2003; 57(6):591–9. 52 Blok VT, Daha MR, Tijsma O, et al. A bispecific monoclonal antibody directed against both the membrane-bound complement regulator CD55 and the renal tumor-associated
244 Monoclonal antibodies and treatment of cancer
53 54
55
56 57 58
59
60
61
62 63 64 65 66 67 68
69
70 71
72
73
antigen G250 enhances C3 deposition and tumor cell lysis by complement. J Immunol 1998; 160(7):3437–43. Mendelsohn J, Baselga J. The EGF receptor family as targets for cancer therapy. Oncogene 2000; 19(56):6550–65. Olayioye MA, Neve RM, Lane HA, Hynes NE. The ErbB signaling network: receptor heterodimerization in development and cancer. Embo J 2000; 19(13):3159–67. Rosen LS. Clinical experience with angiogenesis signaling inhibitors: focus on vascular endothelial growth factor (VEGF) blockers. Cancer Control 2002; 9(2 Suppl):36–44. Holbro T, Civenni G, Hynes NE. The ErbB receptors and their role in cancer progression. Exp Cell Res 2003; 284(1):99–110. Groner B, Hartmann C, Wels W. Therapeutic antibodies. Curr Mol Med 2004; 4(5):539–47. Mendelsohn J. Antibody-mediated EGF receptor blockade as an anticancer therapy: from the laboratory to the clinic. Cancer Immunol Immunother 2003; 52(5):342–6. Franklin MC, Carey KD, Vajdos FF, Leahy DJ, de Vos AM, Sliwkowski MX. Insights into ErbB signaling from the structure of the ErbB2-pertuzumab complex. Cancer Cell 2004; 5(4):317–28. Agus DB, Akita RW, Fox WD, et al. Targeting ligand-activated ErbB2 signaling inhibits breast and prostate tumor growth. Cancer Cell 2002; 2(2):127–37. Yarden Y. The EGFR family and its ligands in human cancer. signalling mechanisms and therapeutic opportunities. Eur J Cancer 2001; 37 Suppl 4:S3–8. Kreitman RJ. Recombinant toxins for the treatment of cancer. Curr Opin Mol Ther 2003; 5(1):44–51. Pastan I, Kreitman RJ. Immunotoxins in cancer therapy. Curr Opin Investig Drugs 2002; 3(7):1089–91. Pastan I, Beers R, Bera TK. Recombinant immunotoxins in the treatment of cancer. Methods Mol Biol 2004; 248:503–18. Uckun FM. Immunotoxins for the treatment of leukaemia. Br J Haematol 1993; 85(3):435–8. Kreitman RJ. Immunotoxins in cancer therapy. Curr Opin Immunol 1999; 11(5):570–8. Gatto B. Monoclonal antibodies in cancer therapy. Curr Med Chem Anti-Canc Agents 2004; 4(5):411–4. Boerman OC, van Schaijk FG, Oyen WJ, Corstens FH. Pretargeted radioimmunotherapy of cancer: progress step by step. J Nucl Med 2003; 44(3):400–11. DeNardo SJ, DeNardo GL, Brush J, Carter P. Phage libraryderived human anti-TETA and anti-DOTA ScFv for pretargeting RIT. Hybridoma 1999; 18(1):13–21. Milenic DE, Brechbiel MW. Targeting of radio-isotopes for cancer therapy. Cancer Biol Ther 2004; 3(4):361–70. Goldenberg DM. Advancing role of radiolabeled antibodies in the therapy of cancer. Cancer Immunol Immunother 2003; 52(5):281–96. Wong JY, Shibata S, Williams LE, et al. A Phase I trial of 90Y-anti-carcinoembryonic antigen chimeric T84.66 radioimmunotherapy with 5-fluorouracil in patients with metastatic colorectal cancer. Clin Cancer Res 2003; 9(16 Pt 1): 5842–52. Sharkey RM, Hajjar G, Yeldell D, et al. A phase I trial combining high-dose 90Y-labeled humanized anti-CEA
74
75
76
77 78
79
80
81
82
83
84
85
86 87
88
89
monoclonal antibody with doxorubicin and peripheral blood stem cell rescue in advanced medullary thyroid cancer. J Nucl Med 2005; 46(4):620–33. Wu AM, Yazaki PJ. Designer genes: recombinant antibody fragments for biological imaging. Q J Nucl Med 2000; 44(3):268–83. Begent RH, Verhaar MJ, Chester KA, et al. Clinical evidence of efficient tumor targeting based on single-chain Fv antibody selected from a combinatorial library. Nat Med 1996; 2(9):979–84. Larson SM, El-Shirbiny AM, Divgi CR, et al. Single chain antigen binding protein (sFv CC49): first human studies in colorectal carcinoma metastatic to liver. Cancer 1997; 80(12 Suppl):2458–68. Jain RK. Tumor physiology and antibody delivery. Front Radiat Ther Oncol 1990; 24:32–46; discussion 64–8. Sharkey RM, Karacay H, Cardillo TM, et al. Improving the delivery of radionuclides for imaging and therapy of cancer using pretargeting methods. Clin Cancer Res 2005; 11(19 Pt 2): 7109s–21s. Senter PD, Springer CJ. Selective activation of anticancer prodrugs by monoclonal antibody-enzyme conjugates. Adv Drug Deliv Rev 2001; 53(3):247–64. Bagshawe KD, Sharma SK, Begent RH. Antibody-directed enzyme prodrug therapy (ADEPT) for cancer. Expert Opin Biol Ther 2004; 4(11):1777–89. Sharma SK, Bagshawe KD, Begent RH. Advances in antibodydirected enzyme prodrug therapy. Curr Opin Investig Drugs 2005; 6(6):611–5. Zein N, Sinha AM, McGahren WJ, Ellestad GA. Calicheamicin gamma 1I: an antitumor antibiotic that cleaves doublestranded DNA site specifically. Science 1988; 240 (4856):1198–201. Trail PA, Willner D, Knipe J, et al. Effect of linker variation on the stability, potency, and efficacy of carcinoma-reactive BR64doxorubicin immunoconjugates. Cancer Res 1997; 57(1):100–5. Trail PA, Willner D, Lasch SJ, et al. Cure of xenografted human carcinomas by BR96-doxorubicin immunoconjugates. Science 1993; 261(5118):212–5. Mosure KW, Henderson AJ, Klunk LJ, Knipe JO. Disposition of conjugate-bound and free doxorubicin in tumor-bearing mice following administration of a BR96-doxorubicin immunoconjugate (BMS 182248). Cancer Chemother Pharmacol 1997; 40(3):251–8. Payne G. Progress in immunoconjugate cancer therapeutics. Cancer Cell 2003; 3(3):207–12. Dubowchik GM, Walker MA. Receptor-mediated and enzymedependent targeting of cytotoxic anticancer drugs. Pharmacol Ther 1999; 83(2):67–123. Prentice HG, Blacklock HA, Janossy G, et al. Use of anti-T-cell monoclonal antibody OKT3 to prevent acute graft-versus-host disease in allogeneic bone-marrow transplantation for acute leukaemia. Lancet 1982; 1(8274):700–3. Hebart H, Ehninger G, Schmidt H, et al. Treatment of steroidresistant graft-versus-host disease after allogeneic bone marrow transplantation with anti-CD3/TCR monoclonal antibodies. Bone Marrow Transplant 1995; 15(6):891–4.
References 245
90 Ji SQ, Chen HR, Yan HM, et al. Anti-CD25 monoclonal antibody (basiliximab) for prevention of graft-versus-host disease after haploidentical bone marrow transplantation for hematological malignancies. Bone Marrow Transplant 2005; 36(4):349–54. 91 Bacigalupo A, Lamparelli T, Bruzzi P, et al. Antithymocyte globulin for graft-versus-host disease prophylaxis in transplants from unrelated donors: 2 randomized studies from Gruppo Italiano Trapianti Midollo Osseo (GITMO). Blood 2001; 98(10):2942–7. 92 Malek SN, Flinn IW. Incorporating monoclonal antibodies in blood and marrow transplantation. Semin Oncol 2003; 30(4):520–30. 93 Magni M, Di Nicola M, Devizzi L, et al. Successful in vivo purging of CD34-containing peripheral blood harvests in mantle cell and indolent lymphoma: evidence for a role of both chemotherapy and rituximab infusion. Blood 2000; 96(3):864–9. 94 Gianni AM, Cortelazzo S, Magni M, Martelli M. Rituximab: enhancing stem cell transplantation in mantle cell lymphoma. Bone Marrow Transplant 2002; 29 Suppl 1:S10–3. 95 Maloney DG, Grillo-Lopez AJ, Bodkin DJ, et al. IDEC-C2B8: results of a phase I multiple-dose trial in patients with relapsed non-Hodgkinís lymphoma. J Clin Oncol 1997; 15(10):3266–74. 96 Maloney DG, Grillo-Lopez AJ, White CA, et al. IDEC-C2B8 (Rituximab) anti-CD20 monoclonal antibody therapy in patients with relapsed low-grade non-Hodgkinís lymphoma. Blood 1997; 90(6):2188–95. 97 Shan D, Ledbetter JA, Press OW. Signaling events involved in anti-CD20-induced apoptosis of malignant human B cells. Cancer Immunol Immunother 2000; 48(12):673–83. 98 Deans JP, Li H, Polyak MJ. CD20-mediated apoptosis: signalling through lipid rafts. Immunology 2002; 107(2):176–82. 99 Press OW, Howell-Clark J, Anderson S, Bernstein I. Retention of B-cell-specific monoclonal antibodies by human lymphoma cells. Blood 1994; 83(5):1390–7. 100 Bubien JK, Zhou LJ, Bell PD, Frizzell RA, Tedder TF. Transfection of the CD20 cell surface molecule into ectopic cell types generates a Ca2+ conductance found constitutively in B lymphocytes. J Cell Biol 1993; 121(5):1121–32. 101 Kanzaki M, Shibata H, Mogami H, Kojima I. Expression of calcium-permeable cation channel CD20 accelerates progression through the G1 phase in Balb/c 3T3 cells. J Biol Chem 1995; 270(22):13099–104. 102 Byrd JC, Kitada S, Flinn IW, et al. The mechanism of tumor cell clearance by rituximab in vivo in patients with B-cell chronic lymphocytic leukemia: evidence of caspase activation and apoptosis induction. Blood 2002; 99(3):1038–43. 103 Cartron G, Dacheux L, Salles G, et al. Therapeutic activity of humanized anti-CD20 monoclonal antibody and polymorphism in IgG Fc receptor FcgammaRIIIa gene. Blood 2002; 99(3):754–8. 104 Meropol NJ, Porter M, Blumenson LE, et al. Daily subcutaneous injection of low-dose interleukin 2 expands
105
106
107
108
109 110
111
112
113
114
115
116
117
118
119
natural killer cells in vivo without significant toxicity. Clin Cancer Res 1996; 2(4):669–77. Caligiuri MA, Murray C, Robertson MJ, et al. Selective modulation of human natural killer cells in vivo after prolonged infusion of low dose recombinant interleukin 2. J Clin Invest 1993; 91(1):123–32. Friedberg JW, Neuberg D, Gribben JG, et al. Combination immunotherapy with rituximab and interleukin 2 in patients with relapsed or refractory follicular non-Hodgkin’s lymphoma. Br J Haematol 2002; 117(4):828–34. Ansell SM, Witzig TE, Kurtin PJ, et al. Phase 1 study of interleukin-12 in combination with rituximab in patients with B-cell non-Hodgkin lymphoma. Blood 2002; 99(1):67–74. Sivaraman S, Venugopal P, Huang X. Effects of in vitro exposure to interferon alpha on CD20 expression in chronic lymphocytic leukemia cells. Ann Oncol 1999; 10. Herberman RB. Effect of alpha-interferons on immune function. Semin Oncol 1997; 24(3 Suppl 9):S9-78–S9-80. Davis TA, Maloney DG, Grillo-Lopez AJ, et al. Combination immunotherapy of relapsed or refractory low-grade or follicular non-Hodgkinís lymphoma with rituximab and interferon-alpha-2a. Clin Cancer Res 2000; 6(7):2644–52. Sacchi S, Federico M, Vitolo U, et al. Clinical activity and safety of combination immunotherapy with IFN-alpha 2a and Rituximab in patients with relapsed low grade nonHodgkin’s lymphoma. Haematologica 2001; 86(9):951–8. Berinstein NL, Grillo-Lopez AJ, White CA, et al. Association of serum Rituximab (IDEC-C2B8) concentration and anti-tumor response in the treatment of recurrent low-grade or follicular non-Hodgkin’s lymphoma. Ann Oncol 1998; 9(9):995–1001. Teeling JL, French RR, Cragg MS, et al. Characterization of new human CD20 monoclonal antibodies with potent cytolytic activity against non-Hodgkin lymphomas. Blood 2004; 104(6):1793–800. Stein R, Qu Z, Chen S, et al. Characterization of a new humanized anti-CD20 monoclonal antibody, IMMU-106, and Its use in combination with the humanized anti-CD22 antibody, epratuzumab, for the therapy of non-Hodgkin’s lymphoma. Clin Cancer Res 2004; 10(8):2868–78. Hagenbeek A, Plesner T, Walewski A. A novel fully human anti CD20 monoclonal antibody, first clinical results from an ongoing phase I/II trial in patients with follicular nonHodgkin’s lymphoma. Abstract 114. Ann Oncol 2005; 16(S5). Coiffier B, Tilly H, Pederson H, et al. HuMax CD20 human monoclonal antibody in chronic lymphocytic leukemia. Early results from an ongoing phase I/II clinical trial. Abstract 448. Blood 2005; 106(11). Dyer MJ, Hale G, Hayhoe FG, Waldmann H. Effects of CAMPATH-1 antibodies in vivo in patients with lymphoid malignancies: influence of antibody isotype. Blood 1989; 73(6):1431–9. Hale G, Xia MQ, Tighe HP, Dyer MJ, Waldmann H. The CAMPATH-1 antigen (CDw52). Tissue Antigens 1990; 35(3):118–27. Mavromatis B, Cheson BD. Monoclonal antibody therapy of chronic lymphocytic leukemia. J Clin Oncol 2003; 21(9):1874–81.
246 Monoclonal antibodies and treatment of cancer
120 Ginaldi L, De Martinis M, Matutes E, et al. Levels of expression of CD52 in normal and leukemic B and T cells: correlation with in vivo therapeutic responses to Campath1H. Leuk Res 1998; 22(2):185–91. 121 Alinari L, Lapalombella R, Andritsos L, Baiocchi RA, Lin TS, Byrd JC. Alemtuzumab (Campath-1H) in the treatment of chronic lymphocytic leukemia. Oncogene 2007; 26(25):3644–53. 122 Faderl S, Thomas DA, OíBrien S, et al. Experience with alemtuzumab plus rituximab in patients with relapsed and refractory lymphoid malignancies. Blood 2003; 101(9):3413–5. 123 Hale G, Jacobs P, Wood L, et al. CD52 antibodies for prevention of graft-versus-host disease and graft rejection following transplantation of allogeneic peripheral blood stem cells. Bone Marrow Transplant 2000; 26(1):69–76. 124 Hale G, Cobbold S, Novitzky N, et al. CAMPATH-1 antibodies in stem-cell transplantation. Cytotherapy 2001; 3(3): 145–64. 125 Rai KR, Freter CE, Mercier RJ, et al. Alemtuzumab in previously treated chronic lymphocytic leukemia patients who also had received fludarabine. J Clin Oncol 2002; 20(18):3891–7. 126 Osterborg A, Dyer MJ, Bunjes D, et al. Phase II multicenter study of human CD52 antibody in previously treated chronic lymphocytic leukemia. European Study Group of CAMPATH-1H Treatment in Chronic Lymphocytic Leukemia. J Clin Oncol 1997; 15(4):1567–74. 127 Lundin J, Kimby E, Bjorkholm M, et al. Phase II trial of subcutaneous anti-CD52 monoclonal antibody alemtuzumab (Campath-1H) as first-line treatment for patients with Bcell chronic lymphocytic leukemia (B-CLL). Blood 2002; 100(3):768–73. 128 Lundin J, Osterborg A, Brittinger G, et al. CAMPATH-1H monoclonal antibody in therapy for previously treated lowgrade non-Hodgkin’s lymphomas: a phase II multicenter study. European Study Group of CAMPATH-1H Treatment in Low-Grade Non-Hodgkin’s Lymphoma. J Clin Oncol 1998; 16(10):3257–63. 129 Rawstron AC, Kennedy B, Evans PA, et al. Quantitation of minimal disease levels in chronic lymphocytic leukemia using a sensitive flow cytometric assay improves the prediction of outcome and can be used to optimize therapy. Blood 2001; 98(1):29–35. 130 Keating MJ, Flinn I, Jain V, et al. Therapeutic role of alemtuzumab (Campath-1H) in patients who have failed fludarabine: results of a large international study. Blood 2002; 99(10):3554–61. 131 Moreton P, Kennedy B, Lucas G, et al. Eradication of minimal residual disease in B-cell chronic lymphocytic leukemia after alemtuzumab therapy is associated with prolonged survival. J Clin Oncol 2005; 23(13):2971–9. 132 Pedley RB, Hill SA, Boxer GM, et al. Eradication of colorectal xenografts by combined radioimmunotherapy and combretastatin A-4 phosphate. Cancer Res 2001; 61(12):4716–22.
133 Dark GG, Hill SA, Prise VE, Tozer GM, Pettit GR, Chaplin DJ. Combretastatin A-4, an agent that displays potent and selective toxicity toward tumor vasculature. Cancer Res 1997; 57(10):1829–34. 134 Thorpe PE. Vascular targeting agents as cancer therapeutics. Clin Cancer Res 2004; 10(2):415–27. 135 Galbraith SM, Maxwell RJ, Lodge MA, et al. Combretastatin A4 phosphate has tumor antivascular activity in rat and man as demonstrated by dynamic magnetic resonance imaging. J Clin Oncol 2003; 21(15):2831–42. 136 Rustin GJ, Galbraith SM, Anderson H, et al. Phase I clinical trial of weekly combretastatin A4 phosphate: clinical and pharmacokinetic results. J Clin Oncol 2003; 21(15):2815–22. 137 Burrows FJ, Thorpe PE. Eradication of large solid tumors in mice with an immunotoxin directed against tumor vasculature. Proc Natl Acad Sci U S A 1993; 90(19):8996–9000. 138 Huang X, Molema G, King S, Watkins L, Edgington TS, Thorpe PE. Tumor infarction in mice by antibody-directed targeting of tissue factor to tumor vasculature. Science 1997; 275(5299):547–50. 139 Tarli L, Balza E, Viti F, et al. A high-affinity human antibody that targets tumoral blood vessels. Blood 1999; 94(1):192–8. 140 Ebbinghaus C, Ronca R, Kaspar M, et al. Engineered vascular-targeting antibody-interferon-gamma fusion protein for cancer therapy. Int J Cancer 2005. 141 Carnemolla B, Borsi L, Balza E, et al. Enhancement of the antitumor properties of interleukin-2 by its targeted delivery to the tumor blood vessel extracellular matrix. Blood 2002; 99(5):1659–65. 142 Ran S, He J, Huang X, Soares M, Scothorn D, Thorpe PE. Antitumor effects of a monoclonal antibody that binds anionic phospholipids on the surface of tumor blood vessels in mice. Clin Cancer Res 2005; 11(4):1551–62. 143 Ushiro H, Cohen S. Identification of phosphotyrosine as a product of epidermal growth factor-activated protein kinase in A-431 cell membranes. J Biol Chem 1980; 255(18):8363–5. 144 Cohen S, Carpenter G, King L, Jr. Epidermal growth factorreceptor-protein kinase interactions. Co-purification of receptor and epidermal growth factor-enhanced phosphorylation activity. J Biol Chem 1980; 255(10):4834–42. 145 Downward J, Yarden Y, Mayes E, et al. Close similarity of epidermal growth factor receptor and v-erb-B oncogene protein sequences. Nature 1984; 307(5951):521–7. 146 Messa C, Russo F, Caruso MG, Di Leo A. EGF, TGF-alpha, and EGF-R in human colorectal adenocarcinoma. Acta Oncol 1998; 37(3):285–9. 147 Falls DL. Neuregulins: functions, forms, and signaling strategies. Exp Cell Res 2003; 284(1):14–30. 148 Lax I, Fischer R, Ng C, et al. Noncontiguous regions in the extracellular domain of EGF receptor define ligand-binding specificity. Cell Regul 1991; 2(5):337–45. 149 Lemmon MA, Bu Z, Ladbury JE, et al. Two EGF molecules contribute additively to stabilization of the EGFR dimer. Embo J 1997; 16(2):281–94.
References 247
150 Sibilia M, Fleischmann A, Behrens A, et al. The EGF receptor provides an essential survival signal for SOS-dependent skin tumor development. Cell 2000; 102(2):211–20. 151 Puri C, Tosoni D, Comai R, et al. Relationships between EGFR signaling-competent and endocytosis-competent membrane microdomains. Mol Biol Cell 2005; 16(6):2704–18. 152 Sigismund S, Woelk T, Puri C, et al. Clathrin-independent endocytosis of ubiquitinated cargos. Proc Natl Acad Sci U S A 2005; 102(8):2760–5. 153 Li S, Schmitz KR, Jeffrey PD, Wiltzius JJ, Kussie P, Ferguson KM. Structural basis for inhibition of the epidermal growth factor receptor by cetuximab. Cancer Cell 2005; 7(4):301–11. 154 Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2006. CA Cancer J Clin 2006; 56(2):106–30. 155 Jemal A, Tiwari RC, Murray T, et al. Cancer statistics, 2004. CA Cancer J Clin 2004; 54(1):8–29. 156 Kufe D, Holland J, Frei E, Society AC. Cancer medicine 6. Hamilton, Ont London: BC Decker 2003;6th ed:2 v. 157 Porter EH. The statistics of dose/cure relationships for irradiated tumours. Part II. Br J Radiol 1980; 53(628):336–45. 158 Porter EH. The statistics of dose/cure relationships for irradiated tumours. Part I. Br J Radiol 1980; 53(627):210–27. 159 Ohnishi K, Ota I, Takahashi A, Yane K, Matsumoto H, Ohnishi T. Transfection of mutant p53 gene depresses X-ray- or CDDP-induced apoptosis in a human squamous cell carcinoma of the head and neck. Apoptosis 2002; 7(4):367–72. 160 Kojima H, Endo K, Moriyama H, et al. Abrogation of mitochondrial cytochrome c release and caspase-3 activation in acquired multidrug resistance. J Biol Chem 1998; 273(27):16647–50. 161 Toyozumi Y, Arima N, Izumaru S, Kato S, Morimatsu M, Nakashima T. Loss of caspase-8 activation pathway is a possible mechanism for CDDP resistance in human laryngeal squamous cell carcinoma, HEp-2 cells. Int J Oncol 2004; 25(3):721–8. 162 Pignon JP, Bourhis J, Domenge C, Designe L. Chemotherapy added to locoregional treatment for head and neck squamous-cell carcinoma: three meta-analyses of updated individual data. MACH-NC Collaborative Group. MetaAnalysis of Chemotherapy on Head and Neck Cancer. Lancet 2000; 355(9208):949–55. 163 Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006; 354(6):567–78. 164 Robert F, Ezekiel MP, Spencer SA, et al. Phase I study of anti-epidermal growth factor receptor antibody cetuximab in combination with radiation therapy in patients with advanced head and neck cancer. J Clin Oncol 2001; 19(13):3234–43. 165 Baselga J, Pfister D, Cooper MR, et al. Phase I studies of anti-epidermal growth factor receptor chimeric antibody C225 alone and in combination with cisplatin. J Clin Oncol 2000; 18(4):904–14. 166 Bonner JA, al e. Phase III evaluation of radiation with and without cetuximab for locoregionally advanced head and
167
168
169
170
171
172
173
174
175
176
177
neck cancer. International Journal of Radiation Oncology *Biology* Physics 2004; 60(Supplement 1):S147–S8. Bonner JA, et al. Cetuximab improves locoregional control and survival of locoregionally advanced head and neck cancer: independent review of mature data with a median followup of 45 months. AACR-NCI-EORTC International Conference Molecular Targets and Cancer Therapeutics Clinical Cancer Research 2005; 11:9058s. Bonner JA, Giralt J, Harari PM, et al. Phase III evaluation of radiation with and without cetuximab for locoregionally advanced head and neck cancer. International Journal of Radiation Oncology*Biology*Physics 2004; 60(Supplement 1):S147–S8. Bonner JA, Harari PM, Giralt J, et al. Cetuximab Improves Locoregional Control and Survival of Locoregionally Advanced Head and Neck Cancer: Independent Review of Mature Data with a Median followup of 45 months. AACRNCI-EORTC International Conference Molecular Targets and Cancer Therapeutics, Clinical Cancer Research 2005; 11(24, part 2):9058s. Bonner JA, Harari PM, Giralt J, et al. Improved preservation of larynx with the addition of cetuximab to radiation for cancers of the larynx and hypopharynx. Abstract No: 5533. ASCO Annual Meeting 2005. Bonner JA, Giralt J, Harari PM, et al. Cetuximab prolongs survival in patients with locoregionally advanced squamous cell carcinoma of head and neck: A phase III study of high dose radiation therapy with or without cetuximab. Abstract 5507 ASCO Annual Meeting, 2004. Su Y, et al. Concurrent cetuximab, cisplatin, and radiotherapy (RT) for locoregionally advanced squamous cell carcinoma of the head and neck (SCCHN): Updated results of a novel combined modality paradigm. ASCO Annual Meeting 2005. Saltz LB, Meropol NJ, Loehrer PJ, Sr., Needle MN, Kopit J, Mayer RJ. Phase II trial of cetuximab in patients with refractory colorectal cancer that expresses the epidermal growth factor receptor. J Clin Oncol 2004; 22(7):1201–8. Saltz LB, Lenz HJ, Wadler S, et al. Randomized phase II trial of cetuximab/bevacizumab/irinotecan (CBI) versus cetuximab/bevacizumab (CB) in irinotecan-refractory colorectal cancer. In: Abstract No: 3508, ASCO Annual Meeting, 2005. Lutz M, Schoffski P, Folprecht G, et al. A phase I/II study of cetuximab (C225) plus irinotecan (CPT-11) and 24h infusional 5FU/folinic acid (FA) in the treatment of metastatic colorectal cancer expressing the epidermal growth factor receptor. Ann Oncol 2002; 13 (Suppl 5):73. Rosenberg A, Loehrer PJ, Needle MN, et al. Erbitux (IMCc225) plus weekly iriniotecan (CPT-11), fluorouracil (5FU) and leucovorin (LV) in colorectal cancer (CRC that expresses the epidermal growth factor receptor (EGFr). Proc Am Soc Clin Oncol 2002; 21:135a. Van Laethem J-L, Raoul J-L, Mitry E, et al. Cetuximab (C225) in combination with bi-weekly irinotecan (CPT-11), infusional 5-fluorouracil (5-FU), and folinic acid (FA) in
248 Monoclonal antibodies and treatment of cancer
178
179
180
181
182
183 184
185
186
187
188
189
190
patients (pts) with metastatic colorectal cancer (CRC) expressing the epidermal growth factor receptor (EGFR). Preliminary safety and efficacy results. Proc Am Soc Clin Oncol 2003; 22:264. Burtness B. The role of cetuximab in the treatment of squamous cell cancer of the head and neck. Expert Opin Biol Ther 2005; 5(8):1085–93. Herbst RS, Arquette M, Shin DM, et al. Phase II multicenter study of the epidermal growth factor receptor antibody cetuximab and cisplatin for recurrent and refractory squamous cell carcinoma of the head and neck. J Clin Oncol 2005; 23(24):5578–87. Baselga J, Trigo JM, Bourhis J, et al. Phase II multicenter study of the antiepidermal growth factor receptor monoclonal antibody cetuximab in combination with platinum-based chemotherapy in patients with platinumrefractory metastatic and/or recurrent squamous cell carcinoma of the head and neck. J Clin Oncol 2005; 23(24):5568–77. Chan ATC, Hsu M-M, Goh BC, et al. Multicenter, Phase II study of cetuximab in combination with carboplatin in patients with recurrent or metastatic nasopharyngeal carcinoma. J Clin Oncol 2005; 23(15):3568–76. Vermorken JB, Bourhis J, Trigo J, et al. Cetuximab in recurrent/metastatic (R&M) squamous cell carcinoma of the head and neck (SCCHN) refractory to first-line platinumbased therapies. J Clin Oncol (Meeting Abstracts) 2005; 23(16 suppl):5505. Govindan R. Cetuximab in advanced non-small cell lung cancer. Clin Cancer Res 2004; 10(12 Pt 2):4241s–4s. Humblet Y, Vega-Villegas E, Mesia R, et al. Phase I study of cetuximab in combination with cisplatin or carboplatin and 5-fluorouracil (5-FU) in patients (pts) with recurrent and/or metastatic squamous cell carcinoma of the head and neck (SCCHN). J Clin Oncol (Meeting Abstracts) 2004; 22(14 suppl):5513. Saltz LB, Lenz H, Hochster H, et al. Randomized phase II trial of cetuximab/bevacizumab/irinotecan (CBI) versus cetuximab/ bevacizumab (CB) in irinotecan-refractory colorectal cancer. Abstract No: 3508, ASCO Annual Meeting, 2005. Viloria-Petit A, Crombet T, Jothy S, et al. Acquired resistance to the antitumor effect of epidermal growth factor receptor-blocking antibodies in vivo: a role for altered tumor angiogenesis. Cancer Res 2001; 61(13):5090–101. Jost M, Kari C, Rodeck U. The EGF receptor – an essential regulator of multiple epidermal functions. Eur J Dermatol 2000; 10(7):505–10. Busam KJ, Capodieci P, Motzer R, Kiehn T, Phelan D, Halpern AC. Cutaneous side-effects in cancer patients treated with the antiepidermal growth factor receptor antibody C225. Br J Dermatol 2001; 144(6):1169–76. Kimyai-Asadi A, Jih MH. Follicular toxic effects of chimeric anti-epidermal growth factor receptor antibody cetuximab used to treat human solid tumors. Arch Dermatol 2002; 138(1):129–31. Van Doorn R, Kirtschig G, Scheffer E, Stoof TJ, Giaccone G. Follicular and epidermal alterations in patients treated with
191
192
193
194
195
196
197
198
199
200
201 202 203
204
ZD1839 (Iressa), an inhibitor of the epidermal growth factor receptor. Br J Dermatol 2002; 147(3):598–601. Baselga J, Rischin D, Ranson M, et al. Phase I safety, pharmacokinetic, and pharmacodynamic trial of ZD1839, a selective oral epidermal growth factor receptor tyrosine kinase inhibitor, in patients with five selected solid tumor types. J Clin Oncol 2002; 20(21):4292–302. Perez-Soler R, Delord JP, Halpern A, et al. HER1/EGFR inhibitor-associated rash: future directions for management and investigation outcomes from the HER1/EGFR inhibitor rash management forum. Oncologist 2005; 10(5):345–56. Chung KY, Shia J, Kemeny NE, et al. Cetuximab shows activity in colorectal cancer patients with tumors that do not express the epidermal growth factor receptor by immunohistochemistry. J Clin Oncol 2005; 23(9):1803–10. Vallbohmer D, Zhang W, Gordon M, et al. Molecular determinants of cetuximab efficacy. J Clin Oncol 2005; 23(15):3536–44. Ciardiello F, Bianco R, Damiano V, et al. Antiangiogenic and antitumor activity of anti-epidermal growth factor receptor C225 monoclonal antibody in combination with vascular endothelial growth factor antisense oligonucleotide in human GEO colon cancer cells. Clin Cancer Res 2000; 6(9):3739–47. Moroni M, Veronese S, Benvenuti S, et al. Gene copy number for epidermal growth factor receptor (EGFR) and clinical response to antiEGFR treatment in colorectal cancer: a cohort study. Lancet Oncol 2005; 6(5):279–86. Atkins D, Reiffen KA, Tegtmeier CL, Winther H, Bonato MS, Storkel S. Immunohistochemical detection of EGFR in paraffin-embedded tumor tissues: variation in staining intensity due to choice of fixative and storage time of tissue sections. J Histochem Cytochem 2004; 52(7):893–901. Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature 2000; 406(6797): 747–52. Baselga J, Seidman AD, Rosen PP, Norton L. HER2 overexpression and paclitaxel sensitivity in breast cancer: therapeutic implications. Oncology (Williston Park) 1997; 11(3 Suppl 2):43–8. Cho HS, Mason K, Ramyar KX, et al. Structure of the extracellular region of HER2 alone and in complex with the Herceptin Fab. Nature 2003; 421(6924): 756–60. Scholl S, Beuzeboc P, Pouillart P. Targeting HER2 in other tumor types. Ann Oncol 2001; 12 Suppl 1:S81–7. Rubin I, Yarden Y. The basic biology of HER2. Ann Oncol 2001;12 Suppl 1:S3–8. Baselga J, Tripathy D, Mendelsohn J, et al. Phase II study of weekly intravenous recombinant humanized anti-p185HER2 monoclonal antibody in patients with HER2/neuoverexpressing metastatic breast cancer. J Clin Oncol 1996; 14(3):737–44. Pegram M, Hsu S, Lewis G, et al. Inhibitory effects of combinations of HER-2/neu antibody and chemotherapeutic
References 249
205
206
207
208
209
210
211
212
213
214
215
216
agents used for treatment of human breast cancers. Oncogene 1999; 18(13):2241–51. Marty M, Cognetti F, Maraninchi D, et al. Randomized phase II trial of the efficacy and safety of trastuzumab combined with docetaxel in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer administered as first-line treatment: the M77001 study group. J Clin Oncol 2005; 23(19):4265–74. Gatzemeier U, Groth G, Butts C, et al. Randomized phase II trial of gemcitabine-cisplatin with or without trastuzumab in HER2-positive non-small-cell lung cancer. Ann Oncol 2004; 15(1):19–27. Langer CJ, Stephenson P, Thor A, Vangel M, Johnson DH. Trastuzumab in the treatment of advanced non-small-cell lung cancer: is there a role? Focus on Eastern Cooperative Oncology Group study 2598. J Clin Oncol 2004; 22(7):1180–7. Romond EH, Perez EA, Bryant J, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 2005; 353(16):1673–84. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 2005; 353(16):1659–72. Buzdar AU, Ibrahim NK, Francis D, et al. Significantly higher pathologic complete remission rate after neoadjuvant therapy with trastuzumab, paclitaxel, and epirubicin chemotherapy: results of a randomized trial in human epidermal growth factor receptor 2-positive operable breast cancer. J Clin Oncol 2005; 23(16):3676–85. Mohsin SK, Weiss HL, Gutierrez MC, et al. Neoadjuvant trastuzumab induces apoptosis in primary breast cancers. J Clin Oncol 2005; 23(11):2460–8. Kostler WJ, Steger GG, Soleiman A, et al. Monitoring of serum Her-2/neu predicts histopathological response to neoadjuvant trastuzumab-based therapy for breast cancer. Anticancer Res 2004; 24(2C):1127–30. Joensuu H, Kellokumpu-Lehtinen P-L, Bono P, et al. Trastuzumab in combination with docetaxel or vinorelbine as adjuvant treatment of breast cancer: the FinHer Trial. In: San Antonio Breast Cancer Symposium, Abstract #2; 2005; San Antonio, TX; 2005. Haslekas C, Breen K, Pedersen KW, Johannessen LE, Stang E, Madshus IH. The inhibitory effect of ErbB2 on epidermal growth factor-induced formation of clathrin-coated pits correlates with retention of epidermal growth factor receptor-ErbB2 oligomeric complexes at the plasma membrane. Mol Biol Cell 2005; 16(12):5832–42. Longva KE, Pedersen NM, Haslekas C, Stang E, Madshus IH. Herceptin-induced inhibition of ErbB2 signaling involves reduced phosphorylation of Akt but not endocytic down-regulation of ErbB2. Int J Cancer 2005; 116(3):359–67. Nahta R, Yuan LX, Zhang B, Kobayashi R, Esteva FJ. Insulinlike growth factor-I receptor/human epidermal growth factor receptor 2 heterodimerization contributes to trastuzumab resistance of breast cancer cells. Cancer Res 2005; 65(23):11118–28.
217 Agus DB, Gordon MS, Taylor C, et al. Phase I clinical study of pertuzumab, a novel HER dimerization inhibitor, in patients with advanced cancer. J Clin Oncol 2005; 23(11):2534–43. 218 O’Reilly MS, Holmgren L, Shing Y, et al. Angiostatin: a novel angiogenesis inhibitor that mediates the suppression of metastases by a Lewis lung carcinoma. Cell 1994; 79(2):315–28. 219 Ferrara N, Davis-Smyth T. The biology of vascular endothelial growth factor. Endocr Rev 1997; 18(1):4–25. 220 Davis-Smyth T, Chen H, Park J, Presta LG, Ferrara N. The second immunoglobulin-like domain of the VEGF tyrosine kinase receptor Flt-1 determines ligand binding and may initiate a signal transduction cascade. Embo J 1996; 15(18):4919–27. 221 Borgstrom P, Gold DP, Hillan KJ, Ferrara N. Importance of VEGF for breast cancer angiogenesis in vivo: implications from intravital microscopy of combination treatments with an anti-VEGF neutralizing monoclonal antibody and doxorubicin. Anticancer Res 1999; 19(5B): 4203–14. 222 Presta LG, Chen H, O’Connor SJ, et al. Humanization of an anti-vascular endothelial growth factor monoclonal antibody for the therapy of solid tumors and other disorders. Cancer Res 1997; 57(20):4593–9. 223 Hurwitz HI, Fehrenbacher L, Hainsworth JD, et al. Bevacizumab in combination with fluorouracil and leucovorin: an active regimen for first-line metastatic colorectal cancer. J Clin Oncol 2005; 23(15):3502–8. 224 Miller KD, Chap LI, Holmes FA, et al. Randomized phase III trial of capecitabine compared with bevacizumab plus capecitabine in patients with previously treated metastatic breast cancer. J Clin Oncol 2005; 23(4):792–9. 225 D’Adamo DR, Anderson SE, Albritton K, et al. Phase II study of doxorubicin and bevacizumab for patients with metastatic soft-tissue sarcomas. J Clin Oncol 2005; 23(28):7135–42. 226 Willett CG, Boucher Y, di Tomaso E, et al. Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer. Nat Med 2004; 10(2):145–7. 227 Johnson DH, Fehrenbacher L, Novotny WF, et al. Randomized phase II trial comparing bevacizumab plus carboplatin and paclitaxel with carboplatin and paclitaxel alone in previously untreated locally advanced or metastatic non-small-cell lung cancer. J Clin Oncol 2004; 22(11):2184–91. 228 Sandler AB, Gray R, Brahmer J, Dowlati A, Schiller JH, Perry MC, Johnson DH. Randomized phase II/III Trial of paclitaxel (P) plus carboplatin (C) with or without bevacizumab (NSC # 704865) in patients with advanced non-squamous nonsmall cell lung cancer (NSCLC): An Eastern Cooperative Oncology Group (ECOG) Trial – E4599. In: ASCO Orlando; 2005. 229 Cobleigh MA, Langmuir VK, Sledge GW, et al. A phase I/II dose-escalation trial of bevacizumab in previously treated
250 Monoclonal antibodies and treatment of cancer
metastatic breast cancer. Semin Oncol 2003; 30(5 Suppl 16):117–24. 230 Miller KD, Chap LI, Holmes FA, et al. Randomized phase III trial of capecitabine compared with bevacizumab plus capecitabine in patients with previously treated metastatic breast cancer. J Clin Oncol 2005; 23(4):792–9.
231 Miller KD WM, Gralow J, Dickler M, Cobleigh MA, Perez EA, Shenkier TN, Davidson NE. A randomized phase III trial of paclitaxel versus paclitaxel plus bevacizumab as first-line therapy for locally recurrent or metastatic breast cancer: a trial coordinated by the Eastern Cooperative Oncology Group (E2100). In: Abstract #3. San Antonio Breast Cancer Symposium; 2005.
12 Angiogenesis as a target for the treatment of cancer SRINIVASAN MADHUSUDAN, DANIEL PATTERSON AND ADRIAN L. HARRIS
Introduction Tumour hypoxia Tumour angiogenesis
251 252 252
INTRODUCTION Initiation of angiogenesis is a critical step for tumour growth and metastasis.1 Whereas physiological angiogenesis is complex and tightly controlled, aberrant angiogenesis is a hallmark of cancer. Advances in our understanding of the molecular basis of angiogenesis have established tumour angiogenesis as an important prognostic, predictive and therapeutic target in cancer. The search for angiogenesis factors began more than a century ago.2 Several investigators, including Rudolf Virchow, observed that tumour growth was associated with an increased vascularity.2 In a paper published in 1927, Warren Lewis concluded that tumour micro-environment had a significant influence on the growth of blood vessels.3 Intra-vital analysis with transparent chambers to allow visualization of living tissue was established in 1928,4 and in 1939, using the same technique, Ide et al. made the seminal observations that tumour growth was accompanied by neovascularization and transplanted tumour failed to grow in the absence of new blood-vessel formation.5 In 1945, using a transparent chamber in mice, Algire et al. concluded that tumour growth was dependent on neovascularization and they proposed that the ability to initiate new vessel growth was a critical step in tumorigenesis.6 In 1968, Ian Tannock showed that a key rate-limiting step for tumour-cell growth is the diffusion of oxygen and nutrients from the vascular endothelium.7 The first direct
Tumour vasculature as a therapeutic target in cancer Conclusions References
256 262 262
experiments to demonstrate that tumour cells produced diffusible angiogenesis factors that promoted neovascularization were performed in 1968.8,9 Isolation of a tumour factor responsible for angiogenesis was first reported by Judah Folkman in 1971,10 and in the same year he proposed that angiogenesis inhibitors could be used in the treatment of cancer.1 These pioneering early studies stimulated further investigations that led to the identification of vascular endothelial growth factor (VEGF) in the 1980s2,11–13 and other angiogenesis factors.14 The next logical step was to investigate the relevance of angiogenesis in human tumours and search for angiogenesis inhibitors for cancer therapy. In the last 15 years, we have seen unprecedented research in this area, culminating in the approval by the US Food and Drug Administration (FDA) of bevacizumab (Avastin), a humanized anti-VEGF monoclonal antibody, for the first-line treatment of metastatic colorectal cancer in combination with chemotherapy. This chapter discusses how this has been achieved in recent years and future applications. It starts with an overview of the pathways of angiogenesis and discusses the prognostic and predictive significance of angiogenesis in human cancer. This is followed by a detailed review of various angiogenesis inhibitors in clinical development, with a particular focus on VEGF/VEGFR (VEGF receptor) inhibitors. Finally, there is a discussion of novel clinical trial methodologies to investigate anti-angiogenic agents, with an emphasis on biomarker analysis and angiogenesis imaging in cancer.
252 Angiogenesis as a target for the treatment of cancer
TUMOUR HYPOXIA In the complex multi-step process of carcinogenesis, genetic and epigenetic changes in cells are essential for malignant transformation. In addition, the ability to propagate and progress is critically dependent on the induction of a tumour vasculature. In fact, to maintain an adequate supply of oxygen and nutrients and to remove metabolic waste products, tumours need to initiate angiogenesis to grow beyond a size of approximately 2 mm3 (i.e. 105–106 cells).15 Cells that are located within about 100 μm from blood vessels are normoxic, whereas tumour cells located more than 100 μm away from blood vessels become hypoxic. In tumours, hypoxia sets in early on because rapidly proliferating tumour cells outgrow the capacity of the host vasculature. Hypoxic clones may survive by initiating an angiogenic pathway. Clones that are unable to initiate angiogenesis remain dormant, sometimes for months to years, before they can switch to an angiogenic phenotype.16 Tumour hypoxia is an important inducer of angiogenesis within tumours. In addition, hypoxic micro-environment selects cells that are capable of evading apoptosis and survive in the absence of normal oxygen availability.17 Direct measurement of oxygen concentrations within tumours using the ‘Eppendorf’ electrode suggests that oxygen concentrations tend to be heterogeneous, with areas of significant hypoxia (5 mmHg partial pressure of oxygen (pO2), corresponding to 0.7 per cent O2 in the gas phase).18 So, how do cells sense hypoxia and what mechanisms initiate tumour angiogenesis? Hypoxia-inducible transcription factor-1 (HIF-1)mediated pathway is critical for tumour angiogenesis.19 The HIF-1 is a heterodimer that consists of hypoxiaresponse factor HIF-1α and the constitutively expressed aryl hydrocarbon receptor nuclear translocator (ARNT, also known as HIF-1β). In the absence of oxygen, HIF-1 binds to hypoxia-responsive elements (HREs), thereby activating the expression of numerous hypoxia-responsive genes such as VEGF.19 However, in the presence of oxygen, HIF-1α is post-translationally modified and interacts with the Von Hippel-Lindau (VHL) protein, a ubiquitin ligase that ubiquitinates HIF-1α and leads to degradation of HIF-1α in proteasomes.20An enzyme family known as prolylhydroxylases (PHDs) is involved in sensing hypoxia and regulating HIF-1α expression. There are at least three PHDs, 1, 2 and 3.21,22 Under normoxic conditions, PHDs hydroxylate HIF-1α on two proline residues, which then interact with VHL, as discussed previously. Under hypoxia, hydroxylation does not occur and this leads to stabilization of HIF-1α.19 Redox active human apurinic/apyrimidinic endonuclease (APE1) also known as redox factor-1 (Ref-1) [redox faxtor-1]) may be involved in keeping HIF-1α in an active reduced state.23 In addition, the thioredoxin reductase/thioredoxin system may also be involved in the regulation of HIF-1α expression.24 Hypoxia-inducible genes regulate several biological processes critical for tumour formation, such as cell proliferation, apoptosis, immortalization, migration and
angiogenesis.19 In particular, HIF-1 activates the expression of VEGF, a critical pro-angiogenic factor. It also induces the expression of VEGFR1 on endothelial cells19 and regulates many other angiogenic factors, proteases and macrophage attractants, which could contribute to angiogenesis.25
TUMOUR ANGIOGENESIS Vasculogenesis is the process of blood-vessel formation de novo from endothelial-cell precursors (angioblasts) during embryogenesis. Angiogenesis is the formation of new blood vessels from pre-existing capillaries. In addition to the recruitment of sprouting vessels from pre-existing blood vessels, a growing body of evidence suggests that the incorporation of endothelial progenitors or stem cells into the growing vascular bed may be an essential process in tumour angiogenesis.26 As well as sprouting, vascular beds can also generate new capillary segments by a process called intussusception. In contrast to sprouting, intussusceptive angiogenesis is fast and does not need cell proliferation. The key event is the formation of transluminal tissue pillars, and the major function of the process is the expansion and increase in complexity of the capillary meshwork.27 The process of physiological angiogenesis is complex and is known to occur in normal adult tissues such as the female endometrium, organs undergoing physiological growth and during wound healing. Essential events include endothelial proliferation, migration, invasion, organization of endothelial cells into functional tubular structures, maturation of vessels and vessel regression. However, there are several important differences between tumour and physiological angiogenesis. Tumour angiogenesis is characterized by aberrant vascular structure, altered endothelial-cell–pericyte interactions, abnormal blood flow, increased permeability and delayed maturation. These phenotypic changes result from dysregulation of the angiogenesis process within tumours28 compared to physiological angiogenesis. The induction of a tumour vasculature, termed the ‘angiogenic switch’, is critically dependent on the balance that exists between endogenous pro-angiogenic and antiangiogenic factors (Table 12.1).16,29 Pro-angiogenic gene expression is strongly induced in tumours by hypoxia, oncogene activation and tumour-suppressor mutation.28,30 Switch to an angiogenic phenotype is considered a hallmark of the malignant phenotype whereby pro-angiogenic mechanisms overwhelm or circumvent negative regulators of angiogenesis in tumours.31,32 Vasculogenic mimicry is the ability of cancer cells to express endothelium-associated genes and form extracellular matrix-rich networks in three-dimensional cultures. These networks recapitulate embryonic vasculogenesis, and they have been detected in aggressive human tumours such as melanoma, breast cancer and others. The in-vivo significance of vascular mimicry in human tumours remains to be established but has been shown in breast cancer, melanoma and other human tumours.33–37
Tumour angiogenesis 253
Table 12.1 Endogenous angiogenic factors Pro-angiogenic factors Growth factors Vascular endothelial growth factor Fibroblast growth factors (acidic and basic) Hepatocyte growth factor Platelet-derived growth factor Epidermal growth factor Granulocyte colony-stimulating factor Tumour necrosis factor-α Cytokines Interleukin-1, 6 and 8 Enzymes Cathepsin Gelatinase A,B Stromelysin Small adhesion molecules αvβ3-integrin Metal ions Copper Others Angiostatin-2 Angiopoietin-1 and 2 Angiotropin
Angiogenin Adrenomedullin Erythropoietin Endothelin Hypoxia Midkine Nitric oxide synthase Prostaglandin-E Pleiotropin Platelet-activating factor Plasminogen activator inhibitor-1 Thymidine phosphorylase Thrombopoietin Urokinase tissue plasminogen activator
Anti-angiogenic factors Proteolytic fragments Angiostatin Endostatin Serpin anti-thrombin Canstatin PEX (haemopexin) Prolactin (16-kDa) Restin Tumstatin Arresten Vasostatin Kringle 1–5 Fibronectin fragments Cytokines and chemokines Interleukin-1, 4, 10, 12 and 18 Interferon-α, β and γ EMAP-II gro-β IP-10 (interferoninducible protein-10) MIG (monokine induced by interferon-γ) Platelet factor-4 Soluble receptors Soluble FGFR-1 (fibroblast growth factor receptor-1) Soluble VEGFR-1 Collagenase inhibitor TIMP-1, 2, 3 and 4 (tissue inhibitor of matrix metalloproteinase) Vitamins 1,25-(OH) Vitamin D3 Retinoic acid Tumour suppressor genes p16 p53 Other endogenous inhibitors Angiopoietin Angiotensin Angiotensin-2-receptor Caveolin Fubulin Meth-1 and 2 2-Methoxy oestradiol Osteospondin cleavage product
Table 12.1 (Continued) Pro-angiogenic factors
Anti-angiogenic factors Pigment epitheliumderived factor Prostate-specific antigen Protamine Thrombospondin-1 and 2 Transforming growth factor-β1 Troponin I Collagen fragments EFC-XV
Molecular basis of angiogenesis THE VASCULAR ENDOTHELIAL GROWTH FACTOR AND VASCULAR ENDOTHELIAL GROWTH FACTOR RECEPTOR FAMILY
The VEGF-related gene family of angiogenic and lymphangiogenic growth factors comprises six secreted glycoproteins referred to as VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGF-E and placental growth factor-1 (PLGF-1) and PLGF-2. Vascular endothelial growth factor-A, also known as VEGF, is the predominant factor that has been most extensively investigated in recent years. Of the several isoforms of VEGF, VEGF165 is predominant and is commonly over-expressed in a variety of solid tumours. VEGF plays a critical role in angiogenesis by enhancing microvascular permeability and endothelial-cell activation, and promoting endothelial-cell survival, proliferation, invasion and migration.32,38–41 In addition, VEGF may also be involved in the recruitment of endothelial progenitors at the sites of angiogenesis.42,43 More recently, VEGFRs have been shown on cancer cells, but their contribution to the anti-cancer effect of VEGF inhibitors is unknown. Vascular endothelial growth factor ligands mediate their effects by binding to at least three tyrosine kinase receptors: VEGFR-1 (also known as FLT-1), VEGFR-2 (also known as KDR) and VEGFR-3 (also known as FLT-4). VEGFR-2 is considered to be the main receptor mediating the proliferation of endothelial cells after stimulation by VEGF. VEGFR1 is also essential for physiologic and developmental angiogenesis. VEGF-A binds to VEGFR-1, VEGFR-2 and sFLT-1 (a soluble, secreted splice variant form of VEGFR-1). VEGF-C binds to VEGFR-2 and VEGFR-3 and plays a major role in lymphangiogenesis. VEGF-B, PLGF-1 and PLGF-2 bind to VEGFR-1 and sFLT-1 only. VEGF-E is a viral VEGF molecule that binds only to VEGFR-2.32,38,39,41,44 Vascular endothelial growth factor receptor belongs to the receptor tyrosine kinase (RTK) family of tyrosine kinases. The cell signalling pathway initiated by the VEGFR pathway is complex. In brief, RTKs contain an amino-terminal extracellular ligand-binding domain (usually glycosylated),
254 Angiogenesis as a target for the treatment of cancer
a hydrophobic transmembrane helix and a cytoplasmic domain, which contains a conserved protein tyrosine kinase core and additional regulatory sequences (that contain crucial C-terminal tyrosine residues and receptor regulatory motifs). Ligand binding (VEGF) to the extracellular domain (ECD) results in receptor dimerization/oligomerization, leading to activation of cytoplasmic tyrosine kinase activity and phosphorylation of tyrosine residues. Autophosphorylated tyrosine residues serve as a platform for the recognition and recruitment of a specific set of signal-transducing proteins, such as proteins containing SH2 (Src homology 2) and PTB (phosphotyrosine binding) domains, which modulate diverse cell-signalling responses.45–47 The downstream pathways include PKC (protein kinase C) and MAP (Mitogen-activated protein) kinases, so inhibitors of these pathways may also block angiogenesis.
OTHER PRO-ANGIOGENIC FACTORS
Several other components of angiogenesis pathways are also being developed for therapy. The roles of these are briefly summarized below. Angiopoietin-1 (Ang-1) and angiopoietin-2 (Ang-2) were discovered in 1996.48 Angiopoietin-1 promotes endothelial survival and is involved in the maintenance of mature vessels. It may be involved in both physiological and pathological angiogenesis. Ephrin-B2 and ephrin-B4 are involved in the initial distinction between arterial and venous vessels. The former is distributed in the endothelium of primordial arterial vessels and the latter marks the endothelium of primordial venous vessels.16 Fibroblast growth factor (FGF) consists of nine distinct family members, and the FGF receptor (FGFR) family consists of four family members. The interaction of FGF with FGFR leads to endothelial cell differentiation.49,50 The platelet-derived growth factor (PDGF) family consists of five family members (PDGF-AA, PDGF-AB, PDGFBB, PDGF-CC and PDGF-DD). These factors exert their cellular effects through two RTKs, PDGF-α and PDGF-β receptors respectively. PDGF is involved in tumour pericyte recruitment, an essential process in angiogenesis and differentiation.51,52
OTHER NOVEL ANGIOGENIC SIGNALLING PATHWAYS
Notch signalling is an evolutionarily conserved pathway that is involved in embryonic vascular development and angiogenesis. There are four notch receptors (Notch 1–4) and five ligands (Jagged-1, Jagged-2, Delta-1, Delta-3, Delta-4).44 In-vitro studies suggest that the role of notch signalling is complex and it may be involved in the inhibition of angiogenesis.53 Ephrin ligands (types A and B) and their receptors (EphA and EphB tyrosine kinase receptors) are involved in vascular development and tumour angiogenesis.54 Hedgehogs are 19-kDa proteins that are also involved
in angiogenesis.55 Sprouty (Spry), magic roundabout (Robo4) and slits are recently identified families of proteins that may be involved in tumour angiogenesis.44
Endogenous inhibitors of angiogenesis An important concept developed by Folkman has been that of endogenous inhibition of angiogenesis, suppressing angiogenesis in normal tissues and thus contributing to escape from tumour dormancy. Hypoxia and oncogenes may suppress these pathways, contributing to a pro-angiogenic micro-environment. Several endogenous inhibitors have been isolated in recent years after Folkman’s initial isolation of angiostatin and endostatin.29,56 At physiological concentrations, they seem to have an anti-angiogenic effect. A detailed description of individual inhibitors is beyond the scope of this chapter and only a brief overview is provided here. The subject has been reviewed recently.29,56 Endostatin is an endogenous collagen-XVIII-derived angiogenesis inhibitor. Recent studies suggest that endostatin interacts with VEGF-induced and FGF-2-induced signal transduction pathways. It has also been shown to inhibit cyclin D1. Endostatin also binds to α5β1-integrin and heparin. It inhibits the activation and activity of matrix metalloproteinases (MMP-2, MMP-9 and MMP13). Recently, an endostatin-like fragment from type XV collagen has been isolated, which inhibits the migration of endothelial cells. Anastellin, a fibronectin fragment, and fibulin, a basement-membrane digestion product, have also been found to have anti-angiogenic effects in pre-clinical studies. Thrombospondin-1 (TSP-1) is a large, multifunctional glycoprotein with dual activity, the individual (pro-angiogenic and anti-angiogenic) effects residing in different domains. Tumstatin is a 28-kDa fragment of the α3-chain of the NC1 domain of type IV collagen. Arresten is derived from the non-collagenous domain of the α1-chain of type IV collagen and inhibits endothelial-cell proliferation, migration and tube formation. It may act by interfering with α1β1-integrin and type I collagen. Canstatin is a 24-kDa fragment of the α2-chain of type IV collagen and endorepellin, a carboxyl terminal end of perlecan are potent inhibitors of angiogenesis. Non-matrixderived inhibitors of angiogenesis include angiostatin, cleaved antithrombin III, prothrombin kringle-2, chondromodulin-I, soluble fms-like tyrosine kinase-1, interferons (IFN-α and IFN-β), interleukins (IL-1β, IL-4, IL-12 and IL-18), pigment epithelium-derived factor (PEX), platelet factor-4, prolactin fragment, tissue inhibitors of MMPs (TIMP-2), troponin I and vasostatin.29 Since many of these are proteolytic fragments, it is likely that endogenous normal tissue proteases regulate their expression and function. This may be one reason for the lack of success, or even worsening of outcome, for firstgeneration MMP inhibitors, which were the first specific angiogenesis inhibitors developed.
Tumour angiogenesis 255
Angiogenesis as a prognostic and predictive marker in cancer Prognostic factors are measurements available at the time of diagnosis or surgery that are associated with recurrence, death or other clinical outcomes and determine how patients will fare irrespective of treatment. Predictive factors are measures that help determine which patients do well with particular types of treatment. Though only a few new prognostic or predictive factors for cancer have been validated in the past 10 years, several recent clinical studies have reported the prognostic or predictive value of angiogenesis markers. Discussion of individual studies is beyond the scope of this chapter and only a brief overview is provided. The quantification of immunohistochemically identified microvessels in vascular hot spots has been used for the assessment of tumour vascularity for several years. In fact, this method for the assessment of vascularity in tumours was first described in 1991.57 The technique involved the use of pan-endothelial markers such as factorVIII-related antigen to immunostain tumour blood microvessels.57 Pan-endothelial markers do not distinguish blood from lymphatic vessels. However, recently identified markers such as CD105 and integrin-αvβ3 are able to distinguish between proliferating and quiescent endothelium. In addition to immunohistochemistry, Chalkley eyepiece counting58,59 and multi-parametric computerized imaging analysis systems60 have been developed recently to limit subjective variability during evaluations. The evaluation of angiogenesis markers in tumour tissue and biological fluids, using colorimetric enzymatic immunoassay, immunohistochemistry or RNase protection assays, has been developed and extensively investigated.61 In addition, the evaluation of hypoxia in tumour tissue by immunohistochemistry (e.g. antibodies to HIF and carbonic anhydrase IX), by pimonidazole binding or direct measurements using the Eppendorf oxygen microelectrode has also been investigated recently.62 It is expected that gene microarray technology may predict an angiogenesis profile in tumours in the near future.61 The prognostic significance of tumour vascularity was first demonstrated in breast cancer more than 10 years ago.63,64 Since then, a number clinical studies have been performed in breast and several other tumours. In many of these studies, intra-tumoral vascularization was a significant and independent indicator of relapse-free survival and overall survival.58–60,65–70 In addition, circulating angiogenesis factors such as VEGF are used as surrogate markers of angiogenic activity and tumour progression in cancer patients. Such evaluations may provide a convenient and non-invasive prognostication of patients,71 but none of them has been validated as being reproducible to the requisite standard of a diagnostic test, or related to response to therapy, and they remain experimental. Most recently, circulating endothelial progenitor cells have emerged as having the potential for being useful markers of response to anti-angiogenic therapy, but there is no agreed standard assay.72,73
Recent studies suggest that tumour hypoxia may adversely affect prognosis in solid tumours, irrespective of the treatment modalities used.74–76 Hypoxic cancer cells are relatively radioresistant, largely due to a deficiency of molecular oxygen, which is required for free-radical formation during radiation treatment. Hypoxic cells are also relatively chemoresistant. Poor perfusion may limit drug dissemination, and alterations in pH within local tumour micro-environment may bear upon the efficacy of certain chemotherapeutic agents. In addition, hypoxia in tumour provides a selective survival advantage for biologically more aggressive cancer cells (e.g. p53 mutant phenotype, increased mutational rate, increased metastatic rate and other characteristics that confer aggressive biology), which are likely to be resistant to cytotoxic therapy. Direct measurements of intra-tumoral pO2 may predict survival in advanced cervical cancer77 and may predict radiation response in advanced squamous-cell carcinoma of the head and neck.78 Tumour hypoxia may also predict distant metastases in sarcoma.79 Immunohistochemical analyses of the expression of surrogate markers of hypoxia such as carbonic anhydrase (CA IX) may be of predictive significance in several human tumours, including cervical cancer,80 head and neck cancer,81 breast cancer,82 brain tumours,83 gastric cancer,84 lung cancer,85 sarcoma86 and renal-cell cancer.87
Imaging of angiogenesis in cancer Rapid advances in radiological techniques have enabled noninvasive imaging of angiogenesis in human tumours in recent years. Positron emission tomography (PET), multi-detector X-ray computed tomography (MD-CT), single-photon emission computed tomography (SPECT), magnetic resonance imaging (MRI), ultrasound and near-infrared optical imaging are some of the techniques that are available for such investigations.88–90 Measurements of haemodynamic parameters such as blood volume and blood flow within tumours are possible using steady-state imaging and dynamic contrast imaging techniques.88 In addition, the imaging of markers of angiogenesis, including VEGF, VEGFR, αvβ3integrins, endoglin and endosialin, has been demonstrated in pre-clinical studies.88,89 The feasibility of such an approach has been investigated in a recently published phase I study using HuMV833 anti-VEGF antibody. In that study, 20 patients with progressive solid tumours were treated with escalating doses of HuMV833. Positron emission tomography with 124I-HuMV833 was used to measure the antibody distribution in and clearance from tissues, and MRI was performed to evaluate vascular permeability.91 The use of radiolabelled 2-nitroimidazole for hypoxia imaging in tumours has been demonstrated in pre-clinical studies92,93 and its feasibility in human tumours is now under investigation. Novel imaging techniques may provide rapid and possibly effective biomarker data for quantifying angiogenesis in tumours and for the evaluation of tumour response to
256 Angiogenesis as a target for the treatment of cancer
anti-angiogenic agents. However, further studies are required to evaluate the sensitivity and reliability of these investigations. Convincing correlations between imaging data and pathological data such as micro-vessel density will have to be demonstrated before widespread clinical use.
in nature, in contrast to VDAs, which are thought to be cytotoxic (Table 12.2). However, despite the overlap between the two groups, it is easier overall to classify agents according to their primary mode of action.94
Angiogenesis inhibitors (Table 12.3) TUMOUR VASCULATURE AS A THERAPEUTIC TARGET IN CANCER During the last 10 years, significant advances have been made in the development of cancer therapies targeted at angiogenesis. Detailed description of individual approaches is beyond the scope of this chapter, but an overview of the compounds that are in an advanced stage of development and recently licensed is given below. It is important to distinguish two major therapeutic approaches to exploiting the abnormal tumour vasculature, anti-angiogenesis drugs and vascular targeting agents or vascular disrupting agents (VDAs). Anti-angiogenic agents target the formation of new blood vessels, whereas, in contrast, VDAs selectively target pre-existing tumour vasculature, leading to rapid tumour-cell ischaemia and secondary tumour-cell death. Anti-angiogenic agents are considered to be cytostatic
Table 12.2 Differences between vascular disrupting agents and anti-angiogenic agents Vascular disrupting Effects within minutes Acute dosing All intravenous Increase CEPs Increase permeability Cytotoxic
Anti-angiogenic Effects within days Chronic dosing Oral and intravenous Decrease CEPs Decrease permeability Cytostatic
CEPs, circulating endothelial progenitors.
TARGETING VASCULAR ENDOTHELIAL GROWTH FACTOR
Bevacizumab is a humanized anti-VEGF monoclonal antibody (also known as rhuMab VEGF or Avastin). It binds to and neutralizes VEGF-A isoforms with high affinity but does not neutralize VEGF-B or VEGF-C. Bevacizumab has a long half-life, of about 17–21 days.95 It inhibits tumour growth in pre-clinical tumour xenograft models.95 Bevacizumab monotherapy was found to be safe in phase I human studies and it did not potentiate the toxicity of chemotherapy in combination regimens.96,97 Phase II studies have been conducted in several tumour types.98–102 Bevacizumab showed evidence of anti-tumour activity in phase II studies in combination with chemotherapy in colorectal cancer101 and non-small-cell lung cancer,102 and showed single-agent activity in renal-cell cancer.99 A pivotal phase III trial of bevacizumab in combination with IFL (irinotecan, 5-fluorouracil (5-FU) and leucovorin) in colorectal cancer showed that median survival was increased from 15.6 months in the placebo arm to 20.3 months in the bevacizumab arm.103,104 Preliminary analysis suggests that bevacizumab also prolongs survival in combination with FOLFOX4 chemotherapy (5-FU, leucovorin and oxaliplatin).105 Phase III studies of bevacizumab are ongoing in non-small-cell lung cancer, metastatic breast cancer and renal cancer.105 Although bevacizumab was well tolerated, the reported side effects included thrombosis, bleeding, proteinuria, hypertension and gastrointestinal perforation. A pooled analysis in 1745 patients treated in randomized studies of bevacizumab has shown that thromboembolic events are more common in patients receiving bevacizumab in combination with
Table 12.3 Angiogenesis inhibitors in clinical use Inhibitor
Mode of action
Trial stage
Sponsor
Bevacizumab
Anti-VEGF monoclonal antibody
FDA approval in metastatic colorectal carcinoma
Genentech
Sorafanib
Inhibitor of Raf, VEGFR-2, VEGFR-3, PDGFR, FLT-3, c-KIT
FDA approval in metastatic renal-cell cancer
Bayer
Sunitinib
Inhibitor of VEGFR-1, VEGFR-2, FLT-3, c-KIT, PDGFR
FDA approval in metastatic renal-cell cancer
Pfizer
Erlotinib
Inhibitor of EGFR
FDA approval in metastatic NSCLC, pancreatic cancer
OSI
Thalidomide
Anti-angiogenic
FDA approval in multiple myeloma
–
Endostatin
Anti-angiogeneic
Chinese FDA approval in advanced NSCLC
–
Bortezomib
Inhibits NF-κB
FDA approval in multiple myeloma
Millennium
VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor; PDGFR, platelet-derived growth factor receptor; FLT-3 (FMS-like tyrosine kinase-3); EGFR, epidermal growth factor receptor; NF-κB (nuclear factor-kappa B); FDA, Food and Drug Administration; NSCLC, non-small-cell lung cancer.
Tumour vasculature as a therapeutic target in cancer 257
chemotherapy, and patients who were aged 65 years or older and had a previous history of atherosclerosis were particularly at risk of thromboembolic events.106 It has been proposed that bevacizumab-related adverse events may reflect an impaired response of normal endothelial cells to injury due to the inhibition of normal VEGF response. In February 2004, the US FDA approved bevacizumab for the treatment of metastatic colorectal cancer in combination with 5-FU-based chemotherapy regimes. Another effective way to inhibit VEGF-signalling pathways is to prevent VEGF from binding to its receptors by administering a decoy soluble receptor. This so-called VEGF-trap involves a fusion protein between the constant region of IgG and the extracellular domain of VEGFR.107 It is proposed that the binding affinity of the decoy receptor to VEGF is significantly higher than that of VEGF monoclonal antibody.108 Clinical trials are underway to investigate this approach in cancer. Pegaptanib sodium is an aptamer that recognizes the heparin-binding domain of VEGF and inhibits VEGF165; it is licensed for use in macular degeneration.109 TARGETING VASCULAR ENDOTHELIAL GROWTH FACTOR RECEPTOR AND OTHER RECEPTOR TYROSINE KINASES
PTK787/ZK222584 (Vatalanib) is an orally bioavailable inhibitor of VEGFR-1, VEGR-1 and PDGF-R.110 Phase I studies have confirmed its safety and bioavailability.111 Preliminary results from a large phase III study in colorectal cancer in combination with FOLFOX4 suggest that in a subset of patients with high lactate dehydrogenase, there is significant improvement in progression-free survival.112 Phase III studies in combination with FOLFORI are ongoing. ZD6474 is an orally available small-molecule inhibitor of VEGFR-2, VEGFR-3 and epidermal growth factor receptor (EGFR) with potent anti-tumour activity in pre-clinical studies.113 In phase I studies, diarrhoea, skin rash, nausea, hypertension and fatigue were the commonly reported side effects. Tumour responses were seen in lung cancer. A series of clinical trials has now been initiated in non-small-cell lung cancer. These studies include the investigation of ZD6474 versus geftinib, ZD6474 in combination with docetaxel and in combination with carboplatin and taxol.113 BAY 43-9006 (Sorafenib) is an orally available inhibitor of several kinases, including Raf, VEGFR-2, VEGFR-3, PDGFR-β, FLT-3 and c-kit.114 It inhibits angiogenesis and Raf-mediated cell proliferation. Diarrhoea, hand–foot syndrome, fatigue and hypertension were the common side effects reported in early-phase human studies.114,115 In a placebo controlled phase III randomized study in advanced renal-cell cancer, patients receiving BAY 43-9006 had an improvement in median progression-free survival.106 Sunitinib (sunitinib malate; SU11248) is a novel oral, multi-targeted tyrosine kinase inhibitor. It is a potent inhibitor of VEGFR-1, VEGFR-2, fetal liver tyrosine kinase receptor 3 (FLT-3), c-kit (stem-cell factor (SCF) receptor) and PDGFR. Pre-clinical data suggest that sunitinib has
anti-angiogenic and anti-proliferative effects. In phase I studies, sunitinib was well tolerated and showed antitumour activity in patients with metastatic renal-cell carcinoma.116 In a phase II study in metastatic renal-cell cancer, a significant proportion of patients achieved partial responses or stable disease, with a median time to progression of about 8.7 months.117 Sunitinib has also shown anticancer activity in Gleevec-resistant gastrointestinal stromal tumours and in patients who are unable to tolerate Gleevec. It significantly prolonged time to disease progression in the patient group.118 Sunitinib recently received FDA approval for clinical use in renal cancer and gastrointestinal stromal tumours. Erlotinib HCl (Tarceva) is an orally available, highly selective, reversible inhibitor of EGFR tyrosine kinase activity.119 It has been shown to have anti-proliferative and antiangiogenic effects in tumours. In pre-clinical studies, erlotinib caused significant tumour regressions both alone and in combination with chemotherapy.119 In early-phase studies erlotinib was found to be safe and well tolerated. In phase II studies anti-tumour responses were seen in several tumour types,120 and in large phase III studies in lung cancer, survival in patients receiving erlotinib was superior to that of placebo-treated patients (6.67 months compared to 4.70 months). Patients who had never smoked and those with EGFR-positive tumours were more likely to respond to treatment. Skin rash and diarrhoea were the most common side effects of treatment. In November 2004, erlotinib received FDA approval as monotherapy for the treatment of patients with locally advanced or metastatic non-smallcell lung cancer who have received at least one prior chemotherapy regimen. In the first-line treatment of nonsmall-cell lung cancer, however, two large, controlled, randomized trials failed to show any benefit from adding erlotinib to platinum-based chemotherapy.121 OTHER ANGIOGENESIS TARGETS
Thalidomide is a derivative of glutamic acid with anti-cancer properties. Its mechanism of action is complex and it has been shown to inhibit angiogenesis and tumour necrosis factor-α (TNF-α) synthesis and to block the activation of NF-κB (nuclear factor- kappa B.122 The role of thalidomide in haematological malignancies has been extensively investigated and it is currently approved for use in multiple myeloma.123 Thalidomide has been widely investigated in solid tumours such as brain tumours, renal-cell carcinoma, prostate cancer, melanoma and Kaposi’s sarcoma, with variable response rates.122 Several thalidomide analogues are currently under evaluation.124 Endostatin is a 20-kDa internal fragment of the carboxy terminus of collagen XVIII. It is an endogenous inhibitor of angiogenesis and has a broad spectrum of action. It showed anti-tumour activity in tumour xenograft studies.125 It was well tolerated in phase I clinical studies, but no significant tumour responses were seen.126–129 Clinical trials of Endostar (rh-endostatin, YH-16), a new recombinant
258 Angiogenesis as a target for the treatment of cancer
human endostatin, were conducted in China in non-smallcell lung cancer patients. Tumour responses were seen in 35.4 per cent of patients who received Endostar together with chemotherapy, compared with in 19.5 per cent of those who received chemotherapy only. The 1-year survival rate for patients on Endostar was about 60 per cent, double the rate for those who received chemotherapy alone.130 Endostatin was approved by the Chinese state FDA in September 2005 for clinical use in non-small-cell lung cancer patients. The proteasome is a ubiquitous enzyme complex involved in the degradation of proteins involved in cell-cycle regulation, apoptosis and angiogenesis. The proteasome inhibitor bortezomib (Velcade) is a potent and selective proteasome inhibitor with anti-cancer activity in several solid and haematological tumours.131,132 Bortezomib causes inhibition of the nuclear transcription factor NF-κB and consequently inhibition of tumour growth, angiogenesis and induction of apoptosis. In phase I studies, bortezomib was well tolerated, and in phase II studies, impressive antitumour responses were seen in refractory multiple myeloma patients. Bortezomib has been approved by the US FDA and the European Medicines Agency (EMEA) for progressive multiple myeloma.133 In a phase III trial comparing bortezomib with dexamethasone in refractory/relapsed multiple myeloma, survival advantage was seen in the bortezomib arm. Phase II studies with bortezomib alone or in combination with other agents are ongoing in solid tumours (lung cancer).134,135 Many other drugs previously approved by the US FDA for other indications have now been shown also to have antiangiogenic activity, including zoledronic acid,136 celecoxib137 and low-dose IFN-α.138 The anti-angiogenic effects of conventional cytotoxic chemotherapy have been well established. Chemotherapeutic agents are not only cytotoxic to proliferating tumour cells but also to proliferating endothelial cells. The cytotoxic effects on proliferating endothelial cells are significant when chemotherapy is administered ‘metronomically’, i.e. in small doses at regular intervals over a prolonged period of time. Metronomic scheduling has been shown to produce anti-tumour responses in pre-clinical tumours that are resistant to the same drugs when used at conventional dose levels. In addition, when combined with other anti-angiogenic agents, the efficacy of metronomic scheduling is increased. Clinical trials are ongoing to assess the anti-tumour efficacy of etronomic chemotherapy in combination with antiangiogenic drugs.139–143
Vascular disrupting agents The pioneer of VDA therapy was Juliana Denekamp, who demonstrated in the early 1980s that tumour regression occurred as a result of compromised tumour blood flow. He showed that tumour endothelial cells proliferate more rapidly than those in normal tissues, and by exploiting the known differences between tumour and normal blood
vessels, predicted that selective agents could be developed (Table 12.4).144,145 It was soon shown in pre-clinical studies that small-molecule VDAs induced vascular shutdown within 30 minutes of administration.146,147 The other major class of VDAs undergoing development is that of the biological or ligand-directed VDAs, which use antibodies, peptides or growth factors that selectively bind to the endothelium. Coagulation and/or endothelialcell death are then achieved by coupling the vascular targeting moiety with a toxin (e.g. ricin) or a pro-coagulant. There are a large number of potential target molecules there are expressed on the tumour endothelium and there is active research into the development of agents that can target these molecules, some of which are outlined in Table 12.5.148 The rest of this chapter focuses on small-molecule VDAs, which are at a much more advanced stage of clinical development than ligand-based therapies. SMALL-MOLECULE VASCULAR DISRUPTING AGENTS: MECHANISM OF ACTION
Flavonoids Flavone acetic acid (FAA) is a non-steroidal anti-inflammatory compound that was found in pre-clinical studies to cause haemorrhagic necrosis of murine tumours,149 but was unfortunately inactive in human trials. DMXAA (5,6-dimethylxanthenone-4-acetic acid), also known as AS1404, is a more potent FAA analogue that has since been developed. The exact molecular target is unknown, but DMXAA causes partial dissolution of the actin cytoskeleton, resulting in DNA strand breaks and the induction of endothelial-cell apoptosis.150 DMXAA also works indirectly by the activation of macrophages to release cytokines such as TNF, which helps mediates the anti-tumour effect. The anti-vascular effect is also helped by the activation of platelets and serotonin. Tubulin-binding agents Early examples of tubulin-binding agents include the spindle poisons colchicine, vinblastine and vincristine, but only at doses near the maximum tolerated dose was there evidence of anti-vascular activity.151,152 Newer agents with a much larger therapeutic window have since been developed, which have reversible effects on tubulin. Tubulin-binding agents work by disrupting the endothelial cells’ reliance on the tubulin cytoskeleton to maintain their shape. They act by binding to a different site on the tubulin molecule from that to which colchicine binds, causing subsequent tubulin depolymerization and disorganization of actin and tubulin. The subsequent change in endothelial shape leads to vessel blockage, reduced blood flow and disruption of the endothelial-cell layer, resulting in exposure of the basement membrane and increased vessel permeability. Interstitial pressures rise as a result, leading to further vessel congestion and loss of blood flow. Furthermore, exposure of the basement membrane can initiate the
Tumour vasculature as a therapeutic target in cancer 259
Table 12.4
Small-molecule vascular disrupting agents in development
Drug
Developmental stage
Comments
Phase I/II Phase I/II Phase I/II
Prodrug of CA4 Prodrug of CA4 Prodrug of N-acetylcolchinol
Phase II
Flavonoid
TZT-1027
Phase I
Synthetic derivative of dolastatin-10 (isolated from a marine mollusc)
OXi4503
Phase I
Prodrug of CA1
NPI-2358
Phase I
Synthetic diketopierazine
Various
Main use for promyelocytic leukaemia
Tubulin-binding agents CA4P (combretastatin A4 phosphate) AC7700 (AVE8062) ZD6126 Small-molecule cytokine inducers DMXAA Combined cytotoxic and vascular-targeting agents
Others Arsenic trioxide
Table 12.5 Ligand-directed vascular disrupting agents in development Ligand and effector combination Antibody – tissue factor Anti-VCAM-1 – tissue factor L19 scFv – tissue factor VEGF – gelonin Anti-endoglin – ricin A Anti-TES-23 – neocarzinostatin L19 scFv – IL-12 L19 scFv – TNF-α Anti-phosphatidylserine Targeted ATPμ–Raf gene Flk-1 fused to Fas
Comments Induces coagulation VCAM-1 is a cell adhesion molecule; tissue factor induces coagulation Targets fibronectin, an extracellular matrix marker of angiogenic vessels, leading to thrombosis Gelonin is a plant toxin Ricin is a toxin Targets CD44-related tumour endothelial cell marker; neocarzinostatin is a cytotoxic agent Cytokine effects Cytokine effects Naked antibody targeting tumour endothelium surface phospholipid leading to thrombosis Gene therapy – blocks signalling Gene therapy – induces apoptosis
VCAM (vascular cell adhesion molecule), scFV (short chain variable fragment), VEGF, vascular endothelial growth factor; TES (tumour tissue endothelium specific), IL, interleukin; TNF, tumour necrosis factor; ATP (adenosine triphosphate); Flk (fms-like tyrosine kinase).
coagulation cascade.153 Unlike tumour blood vessels, normal vessels have a much more developed cellular cytoskeleton that will maintain their shape despite the occurrence of tubulin depolymerization, and hence, in theory, vascular shutdown should not occur in normal vessels. There are some tubulin-binding agents that may have additional cytotoxic effects. SMALL-MOLECULE VASCULAR DISRUPTING AGENTS: CLINICAL STUDIES
DMXAA Two phase I studies have been conducted in parallel, with a total of 109 patients.154,155 DMXAA was given weekly in the UK trial and 3-weekly in the New Zealand trial at doses
between 6 mg/m2 and 4900 mg/m2. The drug was well tolerated, with common side effects of tumour pain, nausea and malaise. Dose-limiting toxicities were neurological, including visual disturbances, tremor and urinary incontinence. The maximum tolerated dose was 3700 mg/m2. Partial responses were seen in one melanoma patient at 1300 mg/m2 and in one cervical cancer patient at 1100 mg/m2. Dynamic contrast-enhanced MRI (DCE-MRI) detected evidence of reduced tumour blood flow at doses 500 mg/m2.156 Elevated TNF was seen in tumour biopsies, but not in the plasma, suggesting that the effects of DMXAA were local. The drug has now entered phase II trials in combination with paclitaxel and carboplatin in non-small-cell lung cancer and ovarian cancer patients, with results expected in late 2006.
260 Angiogenesis as a target for the treatment of cancer
Table 12.6 Phase I trials of single-agent combretastatin A4 phosphate (CA4P)
Schedule Infusion time Number of patients CA4P dose range DCE-MRI PET MTD DLT RP2D Number of doses of CA4P given Terminal half-life CA4P (hours) Terminal half-life CA4 (hours)
UK CRC PH1/066154,174
PENN CA4P-102159
CWRU CA4P-101158
Weekly 3 every 28 days 10 minutes 34 5–114 mg/m2 Yes Yes 68 mg/m2 Ataxia, motor neuropathy, syncope, diplopia, tumour pain, dyspnoea 52–68 mg/m2 167 0.489 2.23
Daily 5 every 21 days 10 minutes 37 6–75 mg/m2 Yes No 75 mg/m2 Tumour pain, sensorimotor neuropathy, syncope, dyspnoea 52 mg/m2 700 0.36 3.3
3 weekly 10 minutes 25 18–90 mg/m2 Yes No 60 mg/m2 Cardiac ischaemia, dyspnoea 60 mg/m2 104 0.47 4.26
DCE-MRI, dynamic contrast-enhanced magnetic resonance imaging; DLT, dose-limiting toxicity; MTD, maximum tolerated dose; PET, positron emission tomography; RP2D, recommended phase II dose.
Combretastatin A4 Combretastatin A4 (CA4) was originally isolated from the bark of the African Bush Willow tree Combretum caffrum. Combretastatin A4 phosphate (CA4P) is a more watersoluble prodrug of CA4. Pre-clinical studies have shown that CA4P causes vascular shutdown within 1 hour of administration, leading to haemorrhagic necrosis at the core of the tumour but leaving viable tumour cells at the periphery, which can rapidly re-grow.146,157 Therefore in order for CA4P to achieve its full potential, it needs to be combined with conventional treatments such as chemotherapy or radiotherapy. The three published phase I trials of single-agent CA4P are summarized in Table 12.6. All three studies used DCEMRI and one also used PET to demonstrate significant reductions in tumour blood flow. The most common side effects of CA4P are cardiovascular (hypertension, hypotension and tachycardia), tumour pain and lymphopenia, with no evidence of the traditional cytotoxic side effects of alopecia, myelosuppression and mucositis. The potential for cardiovascular side effects is a concern with CA4P, as it can cause hypertension and would potentially damage new blood vessels that developed after previous cardiac ischaemia. However dose-limiting cardiotoxicity was only seen in one of the three phase I trials, in two patients with pre-existing cardiovascular disease for whom blood pressure control was not ideal.158–160 The presence of a viable rim of tumour cells at the periphery after CA4P treatment explains the modest tumour control seen in the phase I studies. These viable cells are likely to be rapidly proliferating with a good blood supply, and therefore conventional anti-cancer therapies may help to kill them. Pre-clinical studies have shown additional responses when CA4P is combined with chemotherapy, radiation or radioimmunotherapy, which have led to a number of on-going
CA4P combination phase Ib trials, that are summarized in Table 12.7. Thus far in the phase Ib/II CA4P carboplatin/paclitaxel study, there has been minimal myelosuppression. At 72 mg/m2 CA4P, dose-limiting toxicities of ataxia and hypertension were seen. Ataxia may be prevented by the use of prophylactic steroids, and hypertension has been easily controlled with sublingual glyceryl trinitrate. In the phase I portion of the study, partial responses were seen in relapsed ovarian cancer, small-cell lung cancer and oesophageal cancer. Out of the 15 ovarian cancer patients assessable by radiological or CA125 response criteria, there were 10 (67 per cent) responses overall. The study has now entered the phase II part in platinum-resistant ovarian cancer. ZD6126 ZD6126 causes effects similar to those of CA4P on tumour endothelial cells. It is a prodrug of N-acetylcolchinol, a colchicine analogue that inhibits tubulin polymerization and causes microtubule destabilization. Pre-clinical studies have shown that ZD6126 has significant anti-tumour activity, with decreases in tumour blood flow.161 Interestingly, in murine tumours, ZD6126 enhanced the tumour response to radiation.162 Four phase I clinical trials (Table 12.8) have been published using DCE-MRI to show significant reductions in tumour blood flow, with stable disease lasting for at least four cycles occurring in three patients and a minor response in a patient who received 19 cycles.163–166 AVE8062A (AC7700) This synthetic combretastatin analogue was found in preclinical studies to be more likely to achieve vascular shutdown by inducing a contractile response of arterioles rather than a direct response on tumour vasculature.167 Phase I studies were temporarily stopped because of four vascular events occurring in daily and weekly schedules. However,
Tumour vasculature as a therapeutic target in cancer 261
Table 12.7 Combretastatin A4 phosphate (CA4P) combination trials Combination
Comments
CA4P and carboplatin184
Carboplatin AUC 4–5 followed 1 hour later by CA4P 27–36 mg/m2 over 10 minutes every 21 days for maximum of 6 cycles Grade 3/4 thrombocytopenia in 7/16 patients secondary to pharmacokinetic interaction elevating carboplatin AUC.
CA4P with carboplatin and/or taxol185
CA4P 36–54 mg/m2 and carboplatin AUC 4–5 or CA4P 27–54 mg/m2 and paclitaxel 135–175 mg/m2 doublets, progressing to triplet study: CA4P 54–72 mg/m2 with carboplatin AUC 5 and paclitaxel 175 mg/m2 CA4P given over 10 minutes, 18–22 hours prior to carboplatin and/or paclitaxel every 21 days Phase II triplet study: CA4P 63 mg/m2 with carboplatin AUC 5 and paclitaxel 175 mg/m2 triplet for platinum-resistant ovarian cancer patients
CA4P and radiotherapy186
CA4P 50 mg/m2 given with 2 different radiotherapy regimens NSCLC: radiotherapy 27 Gy per fraction, 6 fractions in 3 weeks with single dose CA4P after 2nd fraction (cohort 1), weekly CA4P (cohort 2), and twice-weekly CA4P (cohort 5) Prostate cancer: radiotherapy 55 Gy/20#/4weeks with single dose CA4P after 5th# (cohort 3), weekly CA4P (cohort 4), and weekly higher dose CA4P 63 mg/m2 (cohort 6)
CA4P and A5B7 (radiolabelled anti-CEA monoclonal antibody)
CA4P 45–50 mg/m2 followed by 1800–2400 MBq/m2 131I-A5B7 1–2 weeks later and further CA4P on days 3, 4, 11, 18, 25, 32, 39, 46 and 53 and then every 8 weeks Study open to gastrointestinal cancer patients with CEA 10–1000 g/L or positive CEA immunohistochemistry
CEA, carcinoembryonic antigen; AUC (area under the curve); NSCLC, non-small-cell lung cancer.
Table 12.8 Phase I trials of ZD6126 LoRusso et al.163
Gadgeel et al.164
Radema et al.165
DelProposto et al.166
Single dose every 3 weeks Infusion time (minutes) 10 Number of patients 19 ZD6126 dose range (mg/m2) 5–40 Elimination t1/2 2–3 (hours) Dose-limiting toxicities anaemia, cardiac ischaemia
Single dose every 3 weeks 10 27 5–112 2
Weekly
Single dose every 3 weeks
10 12 5–7 2–3
Anorexia, constipation, dyspnoea, fatigue, headache, pain, nausea
Hypokalaemia, reduced LVEF, increased ICP with brain metastases
Schedule
6 56–112
LVEF (left ventricular ejection fraction); ICP (intracranial pressure).
the 3-weekly schedule had no cardiovascular events, and thus the trial was re-started, with no subsequent dose limiting toxicities (DLTs) seen in 23 patients.168 Combination therapy with cisplatin is now being assessed.
which may have additional cytotoxic effects.170 A Cancer Research UK (CRUK) phase I trial is currently in progress.
OXi4503
PHARMACODYNAMIC ASSESSMENT OF VASCULAR DISRUPTING AGENT
This is the prodrug of combretastatin A1. Pre-clinical studies have shown OXi4503 to be at least ten times more potent than CA4P in terms of tumour vascular shutdown. In mice, the maximum tolerated doses of the two agents are similar, and therefore OXi4503 has potentially a much larger therapeutic window.169 OXi4503 also has quinone metabolites,
This is a very important component of phase I trials of VDAs. Most experience has been with the use of DCE-MRI, which can provide information on changes in dynamic parameters, with a number of phase I trials showing reduced tumour blood flow within hours of administering a VDA. It is beyond the scope of this chapter to describe DCE-MRI in more
262 Angiogenesis as a target for the treatment of cancer
detail, but interested readers can refer to excellent reviews on this topic.171,172 Alternative MRI techniques that are currently being studied include BOLD (blood oxygen level dependent) MRI, which measures deoxyhaemoglobin,171 and diffusion-weighted MRI, which can give a measure of apoptosis.173 Positron emission tomography is an alternative modality that can also measure tumour perfusion, as shown in one of the CA4P trials.174
Targeting tumour hypoxia Hypoxic cancer cells are resistant to radiotherapy and chemotherapy. However, hypoxia provides an opportunity to target cancer cells. Prodrugs activated by hypoxia, hypoxia-specific gene therapy and targeting HIF-1 transcription factor are some of the strategies that have been investigated in recent years.18 Tirapazamine is a hypoxia-activated prodrug in clinical trials in solid tumours. It is converted into a potent DNA-damaging species under hypoxia and has been shown to potentiate the cytotoxicity of ionizing radiation and chemotherapy.175,176 Clinical trials in combination with cisplatin in lung cancer177–179 and in head and neck cancer180 have shown improvements in survival. AQ4N is another prodrug that is converted to an active species, which is a DNA intercalator and topoisomerase poison. AQ4N is currently in early-phase clinical trial.18,181,182 Inhibitors of HIF-1 transcription factor are under pre-clinical development.183
KEY LEARNING POINTS ●
●
●
●
●
●
CONCLUSIONS ●
The initiation of tumour angiogenesis is a critical factor that drives the malignant phenotype and is an important prognostic factor in solid tumours as well as in haematological cancers. Tumour hypoxia is an important inducer of angiogenesis in human cancer and provides a selective micro-environment for the selection and proliferation of biologically more malignant cancer cells. Hypoxia may predict resistance to chemotherapy and radiotherapy. Inhibition of angiogenesis has now been established as a viable anti-cancer treatment strategy. Bevacizumab, sorafenib, sunitinib, erlotinib and thalidomide have now been approved for clinical use in human cancer. Vascular disrupting agents are a very promising new class of anti-cancer agents. Their true potential will become apparent from the results of phase I/II combination studies that have already started, although some of the newer VDAs may be active as single agents. There is also the exciting prospect of combining VDAs with antiangiogenic agents, as one would expect synergistic effects. More work is required on understanding the exact mechanisms of action and the pharmacodynamic assessment of early effects of VDAs and selecting patients for therapy with specific inhibitors. It is possible that only a subgroup of patients benefit from each drug, and profiling the drug to the tumour will be critical to optimize expensive therapies and maximize their effectiveness.
Angiogenesis is the development of new vessels from pre-existing vasculature, and is critical for tumour growth, invasion and metastasis. New mechanisms of angiogenesis include the recruitment of circulating endothelial-cell precursors released from the bone marrow, vascular co-option and vascular mimicry. Many oncogenes synergize with the HIF-1 pathway to induce a key angiogenic factor, VEGF. Drugs targeting VEGF or its receptor, as well as other angiogenic pathways, have been approved and are in clinical use, with improvements in response rate and survival for several common cancers, including breast cancer, colon cancer and renal cancer. Vascular targeting is a different approach to eliminating aberrant tumour vasculature, which results in acute shutdown of tumour vessels causing necrosis, which also shows synergy with conventional therapy. The approaches are complementary. The selection of appropriate patients and targeting other angiogenic pathways remain critical areas of research, as do combined approaches with different anti-angiogenic agents and conventional therapy. Anti-angiogenesis therapy has now reached maturity in clinical trials and clinical application, and has become an additional tool for use in current therapy.
REFERENCES 1 Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med 1971; 285:1182–6. 2 Ferrara N. VEGF and the quest for tumour angiogenesis factors. Nat Rev Cancer 2002; 2:795–803. 3 Lewis WH. The vascular pattern of tumors. Johns Hopkins Hosp Bull 1927; 41:156–62. 4 Sandison JC. Observations on growth of blood vessels as seen in transparent chamber introduced into rabbit’s ear. Am J Anat 1928; 41:475–96. 5 Ide AG, Baker NH, Warren SL, et al. Vascularization of the Brown Pearce rabbit epithelioma transplant as seen in the transparent ear chamber. Am J Roentgenol 1939; 42:891–9. 6 Algire GH, Chalkey, HW, Leqallis FY, et al. Vascular reactions of normal and malignant tissues in vivo. I. Vascular reactions of mice to wounds and to normal and neoplastic transplants. J Natl Cancer Inst 1945; 6:73–85.
References 263
7 Tannock IF. The relation between cell proliferation and the vascular system in a transplanted mouse mammary tumour. Br J Cancer 1968; 22:258–73. 8 Ehrmann RL, Knoth M. Choriocarcinoma. Transfilter stimulation of vasoproliferation in the hamster cheek pouch. Studied by light and electron microscopy. J Natl Cancer Inst 1968; 41:1329–41. 9 Greenblatt M, Shubi P. Tumor angiogenesis: transfilter diffusion studies in the hamster by the transparent chamber technique. J Natl Cancer Inst 1968; 41:111–24. 10 Folkman J, Merler E, Abernathy C, et al. Isolation of a tumor factor responsible for angiogenesis. J Exp Med 1971; 133:275–88. 11 Senger DR, Galli SJ, Dvorak AM, et al. Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid. Science 1983; 219:983–5. 12 Leung DW, Cachianes G, Kuang WJ, et al. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989; 246:1306–9. 13 Plouet J, Schilling J, Gospodarowicz D, et al. Isolation and characterization of a newly identified endothelial cell mitogen produced by AtT-20 cells. Embo J 1989; 8:3801–6. 14 Folkman J, Klagsbrun M. Angiogenic factors. Science 1987; 235:442–7. 15 Carmeliet P. Mechanisms of angiogenesis and arteriogenesis. Nat Med 2000; 6:389–95. 16 Tonini T, Rossi F, Claudio PP, et al. Molecular basis of angiogenesis and cancer. Oncogene 2003; 22:6549–56. 17 Hammond EM, Giaccia AJ. The role of p53 in hypoxia-induced apoptosis. Biochem Biophys Res Commun 2005; 331:718–25. 18 Brown JM, Wilson WR. Exploiting tumour hypoxia in cancer treatment. Nat Rev Cancer 2004; 4:437–47. 19 Harris AL. Hypoxia – a key regulatory factor in tumour growth. Nat Rev Cancer 2002; 2:38–47. 20 Tanimoto K, Makino Y, Pereira T, et al. Mechanism of regulation of the hypoxia-inducible factor-1-alpha by the von Hippel-Lindau tumor suppressor protein. Embo J 2000; 19:4298–309. 21 Tian YM, Mole DR, Ratcliffe PJ, et al. Characterization of different isoforms of the HIF prolyl hydroxylase PHD1 generated by alternative initiation. Biochem J 2006; 397 179–86. 22 Willam C, Nicholls LG, Ratcliffe PJ, et al. The prolyl hydroxylase enzymes that act as oxygen sensors regulating destruction of hypoxia-inducible factor alpha. Adv Enzyme Regul 2004; 44:75–92. 23 Lando D, Pongratz I, Poellinger L, et al. A redox mechanism controls differential DNA binding activities of hypoxiainducible factor (HIF) 1alpha and the HIF-like factor. J Biol Chem 2000; 275:4618–27. 24 Welsh SJ, Bellamy WT, Briehl MM, et al. The redox protein thioredoxin-1 (Trx-1) increases hypoxia-inducible factor 1alpha protein expression: Trx-1 overexpression results in increased vascular endothelial growth factor production and enhanced tumor angiogenesis. Cancer Res 2002; 62:5089–95. 25 Semenza GL. Targeting HIF-1 for cancer therapy. Nat Rev Cancer 2003; 3:721–32.
26 Rafii S, Lyden D, Benezia R, et al. Vascular and haematopoietic stem cells: novel targets for antiangiogenesis therapy? Nat Rev Cancer 2002; 2:826–35. 27 Burri PH, Hlushchuk R, Djonov V, et al. Intussusceptive angiogenesis: its emergence, its characteristics, and its significance. Dev Dyn 2004; 231:474–88. 28 Bergers G, Benjamin LE. Tumorigenesis and the angiogenic switch. Nat Rev Cancer 2003; 3:401–10. 29 Nyberg P, Xie L, Kalluri R, et al. Endogenous inhibitors of angiogenesis. Cancer Res 2005; 65:3967–79. 30 Kerbel R, Folkman J. Clinical translation of angiogenesis inhibitors. Nat Rev Cancer 2002; 2:727–39. 31 Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100:57–70. 32 Hicklin DJ, Ellis LM. Role of the vascular endothelial growth factor pathway in tumor growth and angiogenesis. J Clin Oncol 2005; 23:1011–27. 33 Folberg R, Maniotis AJ. Vasculogenic mimicry. Apmis 2004; 112:508–25. 34 Cai XS, Jia YW, Mei J, et al. Tumor blood vessel formation in osteosarcoma: vasculogenesis mimicry. Chin Med J (Engl) 2004; 117:94–8. 35 Sun BC, Zhang SW, Zhao XL, et al. (Study on vasculogenic mimicry in malignant melanoma.) Zhonghua Bing Li Xue Za Zhi 2003; 32:539–43. 36 Auguste P, Lemiere S, Larrien-Lahargne F, et al. Molecular mechanisms of tumor vascularization. Crit Rev Oncol Hematol 2005; 54:53–61. 37 Yue WY, Chen ZP. Does vasculogenic mimicry exist in astrocytoma? J Histochem Cytochem 2005; 53:997–1002. 38 Ferrara N. VEGF as a therapeutic target in cancer. Oncology 2005; 69(Suppl. 3):11–16. 39 Carmeliet P. VEGF as a key mediator of angiogenesis in cancer. Oncology 2005; 69(Suppl. 3):4–10. 40 Zachary I. VEGF signalling: integration and multi-tasking in endothelial cell biology. Biochem Soc Trans 2003; 31:1171–7. 41 Ribatti D. The crucial role of vascular permeability factor/vascular endothelial growth factor in angiogenesis: a historical review. Br J Haematol 2005; 128:303–9. 42 Lyden D, Hattori K, Dias S, et al. Impaired recruitment of bone-marrow-derived endothelial and hematopoietic precursor cells blocks tumor angiogenesis and growth. Nat Med 2001; 7:1194–201. 43 Asahara T, Murohara T, Sullivan A, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science 1997; 275:964–7. 44 Bicknell R, Harris AL. Novel angiogenic signaling pathways and vascular targets. Annu Rev Pharmacol Toxicol 2004; 44:219–38. 45 Barbieri MA, Ramkumar TP, Fernandez-Pol S, et al. Receptor tyrosine kinase signaling and trafficking – paradigms revisited. Curr Top Microbiol Immunol 2004; 286:1–20. 46 Waters C, Pyne S, Pyne NJ, et al. The role of G-protein coupled receptors and associated proteins in receptor tyrosine kinase signal transduction. Semin Cell Dev Biol 2004; 15:309–23.
264 Angiogenesis as a target for the treatment of cancer
47 Madhusudan S, Ganesan TS. Tyrosine kinase inhibitors in cancer therapy. Clin Biochem 2004; 37:618–35. 48 Davis S, Aldrich TH, Jones PF, et al. Isolation of angiopoietin1, a ligand for the TIE2 receptor, by secretion-trap expression cloning. Cell 1996; 87:1161–9. 49 Eswarakumar VP, Lax I, Schlessinger J, et al. Cellular signaling by fibroblast growth factor receptors. Cytokine Growth Factor Rev 2005; 16:139–49. 50 Grose R, Dickson C. Fibroblast growth factor signaling in tumorigenesis. Cytokine Growth Factor Rev 2005; 16:179–86. 51 Toi M, Atiqur Rahman M, Bando H, et al. Thymidine phosphorylase (platelet-derived endothelial-cell growth factor) in cancer biology and treatment. Lancet Oncol 2005; 6:158–66. 52 Ostman A. PDGF receptors – mediators of autocrine tumor growth and regulators of tumor vasculature and stroma. Cytokine Growth Factor Rev 2004; 15:275–86. 53 Li JL, Harris AL. Notch signaling from tumor cells: a new mechanism of angiogenesis. Cancer Cell 2005; 8:1–3. 54 Surawska H, Ma PC, Salgia R, et al. The role of ephrins and Eph receptors in cancer. Cytokine Growth Factor Rev 2004; 15:419–33. 55 Hooper JE, Scott MP. Communicating with Hedgehogs. Nat Rev Mol Cell Biol 2005; 6:306–17. 56 Dvorak HF. Angiogenesis: update 2005. J Thromb Haemost 2005; 3:1835–42. 57 Weidner N, Semple JP, Welch WR, et al. Tumor angiogenesis and metastasis – correlation in invasive breast carcinoma. N Engl J Med 1991; 324:1–8. 58 Fox SB, Harris AL. Markers of tumor angiogenesis: clinical applications in prognosis and anti-angiogenic therapy. Invest New Drugs 1997; 15:15–28. 59 Fox SB. Tumour angiogenesis and prognosis. Histopathology 1997; 30:294–301. 60 Fox SB, Gaspinini G, Harris AL, et al. Angiogenesis: pathological, prognostic, and growth-factor pathways and their link to trial design and anticancer drugs. Lancet Oncol 2001; 2:278–89. 61 Fox SB, Harris AL. Histological quantitation of tumour angiogenesis. Apmis 2004; 112:413–30. 62 Olive PL, Barath JP, Aquino-Parsons C, et al. Measuring hypoxia in solid tumours – is there a gold standard? Acta Oncol 2001; 40:917–23. 63 Horak ER, Leek R, Klenk N, et al. Angiogenesis, assessed by platelet/endothelial cell adhesion molecule antibodies, as indicator of node metastases and survival in breast cancer. Lancet 1992; 340:1120–4. 64 Weidner N, Folkman J, Pozza F, et al. Tumor angiogenesis: a new significant and independent prognostic indicator in early-stage breast carcinoma. J Natl Cancer Inst 1992; 84:1875–7. 65 Gasparini G. Prognostic value of vascular endothelial growth factor in breast cancer. Oncologist 2000; 5(Suppl. 1):37–44. 66 Canoglu A, Gogus C, Beduk Y, et al. Microvessel density as a prognostic marker in bladder carcinoma: correlation with tumor grade, stage and prognosis. Int Urol Nephrol 2004; 36:401–5.
67 Saito H, Tsujitani S. Angiogenesis, angiogenic factor expression and prognosis of gastric carcinoma. Anticancer Res 2001; 21:4365–72. 68 Fontanini G, Fariana P, Lacchi M, et al. A high vascular count and overexpression of vascular endothelial growth factor are associated with unfavourable prognosis in operated small cell lung carcinoma. Br J Cancer 2002; 86:558–63. 69 Shimanuki Y, Takahashi K, Cui R, et al. Role of serum vascular endothelial growth factor in the prediction of angiogenesis and prognosis for non-small cell lung cancer. Lung 2005; 183:29–42. 70 Herbst RS, Onn A, Sandler A, et al. Angiogenesis and lung cancer: prognostic and therapeutic implications. J Clin Oncol 2005; 23:3243–56. 71 Poon RT, Fan SJ, Wong J, et al. Clinical implications of circulating angiogenic factors in cancer patients. J Clin Oncol 2001; 19:1207–1225. 72 Bertolini F, Mancuso P, Kerbel RS, et al. Circulating endothelial progenitor cells. N Engl J Med 2005; 353: 26132616; author reply 2613–2616. 73 Mancuso P, Colleoni N, Calleri A, et al. Circulating endothelial cell kinetics and viability predict survival in breast cancer patients receiving metronomic chemotherapy. Blood 2006; 108(2):452–9. 74 Weinmann M, Belka C, Plasswilm L, et al. Tumour hypoxia: impact on biology, prognosis and treatment of solid malignant tumours. Onkologie 2004; 27:83–90. 75 Theodoropoulos VE, Lazaris A, Sofras F, et al. Hypoxiainducible factor 1 alpha expression correlates with angiogenesis and unfavorable prognosis in bladder cancer. Eur Urol 2004; 46:200–208. 76 Korkolopoulou P, Patsouris E, Konstantinidou AE, et al. Hypoxia-inducible factor 1alpha/vascular endothelial growth factor axis in astrocytomas. Associations with microvessel morphometry, proliferation and prognosis. Neuropathol Appl Neurobiol 2004; 30:267–278. 77 Hockel M, Vordran B, Schlenger K, et al. Tumor oxygenation: a new predictive parameter in locally advanced cancer of the uterine cervix. Gynecol Oncol 1993; 51:141–149. 78 Nordsmark M, Overgaard M, Overgaard J, et al. Pretreatment oxygenation predicts radiation response in advanced squamous cell carcinoma of the head and neck. Radiother Oncol 1996; 41:31–39. 79 Brizel DM, Scully SP, Harrelson JM, et al. Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res 1996; 56:941–943. 80 Loncaster JA, Harris AL, Davidson SE, et al. Carbonic anhydrase (CA IX) expression, a potential new intrinsic marker of hypoxia: correlations with tumor oxygen measurements and prognosis in locally advanced carcinoma of the cervix. Cancer Res 2001; 61:6394–6399. 81 Koukourakis MI, Giatromanolaki A, Sirridis E, et al. Hypoxia-regulated carbonic anhydrase-9 (CA9) relates to poor vascularization and resistance of squamous cell head and neck cancer to chemoradiotherapy. Clin Cancer Res 2001; 7:3399–3403.
References 265
82 Vleugel MM, Griejer AE, Shrarts A, et al. Differential prognostic impact of hypoxia induced and diffuse HIF-1alpha expression in invasive breast cancer. J Clin Pathol 2005; 58:172–177. 83 Haapasalo JA, Nordfors KM, Hilvo M, et al. Expression of carbonic anhydrase IX in astrocytic tumors predicts poor prognosis. Clin Cancer Res 2006; 12:473–477. 84 Chen J, Rocken C, Hoffmann J, et al. Expression of carbonic anhydrase 9 at the invasion front of gastric cancers. Gut 2005; 54:920–927. 85 Kim SJ, Rubbani ZN, Vollmer RT, et al. Carbonic anhydrase IX in early-stage non-small cell lung cancer. Clin Cancer Res 2004; 10:7925–7933. 86 Maseide K, Kandel RA, Bell RS, et al. Carbonic anhydrase IX as a marker for poor prognosis in soft tissue sarcoma. Clin Cancer Res 2004; 10:4464–4471. 87 Bui MH, Seligson D, Han KR, et al. Carbonic anhydrase IX is an independent predictor of survival in advanced renal clear cell carcinoma: implications for prognosis and therapy. Clin Cancer Res 2003; 9:802–811. 88 Miller JC, Pien HA, Sahani D, et al. Imaging angiogenesis: applications and potential for drug development. J Natl Cancer Inst 2005; 97:172–187. 89 McDonald DM, Choyke PL. Imaging of angiogenesis: from microscope to clinic. Nat Med 2003; 9:713–725. 90 Anderson, H. and Price, P. Clinical measurement of blood flow in tumours using positron emission tomography: a review. Nucl Med Commun 2002; 23:131–138. 91 Jayson GC, Zweit J, Jackson A, et al. Molecular imaging and biological evaluation of HuMV833 anti-VEGF antibody: implications for trial design of antiangiogenic antibodies. J Natl Cancer Inst 2002; 94:1484–1493. 92 Piert M, Machulla HJ, Picchio M, et al. Hypoxia-specific tumor imaging with 18F-fluoroazomycin arabinoside. J Nucl Med 2005; 46:106–113. 93 Bennewith KL, Raleigh JA, Durrant RE, et al. Orally administered pimonidazole to label hypoxic tumor cells. Cancer Res 2002; 62:6827–6830. 94 Siemann DW, Bibby MC, Dark GG, et al. Differentiation and definition of vascular-targeted therapies. Clin Cancer Res 2005; 11:416–420. 95 Ferrara N, Hillan KJ, Geber HP, et al. Discovery and development of bevacizumab, an anti-VEGF antibody for treating cancer. Nat Rev Drug Discov 2004; 3:391–400. 96 Margolin K, Gordon MS, Holmgren E, et al. Phase Ib trial of intravenous recombinant humanized monoclonal antibody to vascular endothelial growth factor in combination with chemotherapy in patients with advanced cancer: pharmacologic and long-term safety data. J Clin Oncol 2001; 19:851–856. 97 Gordon MS, Margolin K, Talpaz S, et al. Phase I safety and pharmacokinetic study of recombinant human anti-vascular endothelial growth factor in patients with advanced cancer. J Clin Oncol 2001; 19:843–850. 98 Cobleigh MA, Langmuir VK, Sledge GW, et al. A phase I/II dose-escalation trial of bevacizumab in previously treated metastatic breast cancer. Semin Oncol 2003; 30:117–124.
99 Yang JC, Haworth L, Sherry RM, et al. A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody, for metastatic renal cancer. N Engl J Med 2003; 349:427–434. 100 Yang JC. Bevacizumab for patients with metastatic renal cancer: an update. Clin Cancer Res 2004; 10:6367S–6370S. 101 Kabbinavar F, Hurwitz HI, Fehrenbacher L, et al. Phase II, randomized trial comparing bevacizumab plus fluorouracil (FU)/leucovorin (LV) with FU/LV alone in patients with metastatic colorectal cancer. J Clin Oncol 2003; 21:60–65. 102 Johnson DH, Fehrenbacher L, Novotny WF, et al. Randomized phase II trial comparing bevacizumab plus carboplatin and paclitaxel with carboplatin and paclitaxel alone in previously untreated locally advanced or metastatic non-small-cell lung cancer. J Clin Oncol 2004; 22:2184–2191. 103 Hurwitz H, Ferhrenbacher L, Novotny WF, et a.l Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004; 350:2335–2342. 104 Hurwitz HI, Ferhrenbacher L, Hainsworth JD, et al. Bevacizumab in combination with fluorouracil and leucovorin: an active regimen for first-line metastatic colorectal cancer. J Clin Oncol 2005; 23:3502–3508. 105 Ferrara N, Kerbel RS. Angiogenesis as a therapeutic target. Nature 2005; 438:967–974. 106 Zakarija A, and Soff G. Update on angiogenesis inhibitors. Curr Opin Oncol 2005; 17:578–583. 107 Holash J, Davis S, Papadopoulos N, et al. VEGF-Trap: a VEGF blocker with potent antitumor effects. Proc Natl Acad Sci USA 2002; 99:11393–11398. 108 Lau SC, Rosa DD, Jayson G. Technology evaluation: VEGF Trap (cancer), Regeneron/sanofi-aventis. Curr Opin Mol Ther 2005; 7:493–501. 109 Ng EW, Shima DT, Calias P, et al. Pegaptanib, a targeted anti-VEGF aptamer for ocular vascular disease. Nat Rev Drug Discov 2006; 5:123–132. 110 Wood JM, Bold G, Buchdunger E, et al. PTK787/ZK 222584, a novel and potent inhibitor of vascular endothelial growth factor receptor tyrosine kinases, impairs vascular endothelial growth factor-induced responses and tumor growth after oral administration. Cancer Res 2000; 60:2178–2189. 111 Thomas AL, Morgan B, Horsfield MA, et al. Phase I study of the safety, tolerability, pharmacokinetics, and pharmacodynamics of PTK787/ZK 222584 administered twice daily in patients with advanced cancer. J Clin Oncol 2005; 23:4162–4171. 112 Tyagi P. Vatalanib (PTK787/ZK 222584) in combination with FOLFOX4 versus FOLFOX4 alone as firstline treatment for colorectal cancer: preliminary results from the CONFIRM-1 trial. Clin Colorectal Cancer 2005; 5:24–26. 113 Ryan A.J. and Wedge, S.R. ZD6474—a novel inhibitor of VEGFR and EGFR tyrosine kinase activity. Br J Cancer 2005; 92:Suppl 1, S6–13. 114 Rini BI. Sorafenib. Expert Opin Pharmacother 2006; 7:453–461. 115 Strumberg D, Awada A, Hirte H, et al. Pooled safety analysis of BAY 43-9006 (sorafenib) monotherapy in patients with
266 Angiogenesis as a target for the treatment of cancer
116
117
118
119 120
121
122
123
124 125
126
127
128
129
130
131
advanced solid tumours: Is rash associated with treatment outcome? Eur J Cancer 2006; 42:548–556. Faivre S, Delbaldo C, Vera K, et al. Safety, pharmacokinetic, and antitumor activity of SU11248, a novel oral multitarget tyrosine kinase inhibitor, in patients with cancer. J Clin Oncol 2006; 24:25–35. Motzer RJ, Michaelson MD, Redman BG, et al. Activity of SU11248, a multitargeted inhibitor of vascular endothelial growth factor receptor and platelet-derived growth factor receptor, in patients with metastatic renal cell carcinoma. J Clin Oncol 2006; 24:16–24. Prenen H, Cools J, Mentens N, et al. Efficacy of the kinase inhibitor SU11248 against gastrointestinal stromal tumor mutants refractory to imatinib mesylate. Clin Cancer Res 2006; 12:2622–2627. Akita RW and Sliwkowski MX. Preclinical studies with Erlotinib (Tarceva). Semin Oncol 2003; 30:15–24. Grunwald V and Hidalgo M. Development of the epidermal growth factor receptor inhibitor Tarceva (OSI-774). Adv Exp Med Biol 2003; 532:235–246. Cohen MH, Johnson JR, Chen YF, et al. FDA drug approval summary: erlotinib (Tarceva) tablets. Oncologist 2005; 10:461–466. Eleutherakis-Papaiakovou V, Bamias A, Dimopoulos MA, et al. Thalidomide in cancer medicine. Ann Oncol 2004; 15:1151–1160. Kumar S. and Anderson KC. Drug insight: thalidomide as a treatment for multiple myeloma. Nat Clin Pract Oncol 2005; 2:262–270. Teo SK. Properties of thalidomide and its analogues: implications for anticancer therapy. Aaps J 2005; 7:E14–19. Folkman J. Antiangiogenesis in cancer therapy—endostatin and its mechanisms of action. Exp Cell Res 2006; 312:594–607. Eder JP, Supko JG, Clark JW, et al. Phase I clinical trial of recombinant human endostatin administered as a short intravenous infusion repeated daily. J Clin Oncol 2002; 20:3772–3784. Herbst RS, Hess KR, Trans HT, et al. Phase I study of recombinant human endostatin in patients with advanced solid tumors. J Clin Oncol 2002; 20:3792–3803. Thomas JP, Arzoomanian RZ. Phase I pharmacokinetic and pharmacodynamic study of recombinant human endostatin in patients with advanced solid tumors. J Clin Oncol 2003; 21:223–231. Hansma AH, Broxterman HJ, Van der Horst I, et al. Recombinant human endostatin administered as a 28-day continuous intravenous infusion, followed by daily subcutaneous injections: a phase I and pharmacokinetic study in patients with advanced cancer. Ann Oncol 2005; 16:1695–1701. Sun J, Wang Y, Liu X, et al. 2005; Results of Phase III trial of EndostarTM (rh-endostatin, YH-16) in advanced non-small cell lung cancer (NSCLC) patient. Proceeding of the American Society of Clinical Oncology. Abstract no. 7138. Montagut C, Rovira A, Mellado B, et al. Preclinical and clinical development of the proteasome inhibitor bortezomib in cancer treatment. Drugs Today (Barc) 2005; 41:299–315.
132 Spano JP, Bay JO, Blay JY, et al. Proteasome inhibition: a new approach for the treatment of malignancies. Bull Cancer 2005; 92:E61-66, 945–952. 133 Kane RC, Bross PF, Farrell AT, et al. Velcade: U.S. FDA approval for the treatment of multiple myeloma progressing on prior therapy. Oncologist 2003; 8:508–513. 134 Park DJ and Lenz HJ. The role of proteasome inhibitors in solid tumors. Ann Med 2004; 36:296–303. 135 Mack PC, Davies AM, Lara PN, et al. Integration of the proteasome inhibitor PS-341 (Velcade) into the therapeutic approach to lung cancer. Lung Cancer 2003; 41 Suppl 1:S89–96. 136 Wood J, Bonjean K, Ruetz S, et al. Novel antiangiogenic effects of the bisphosphonate compound zoledronic acid. J Pharmacol Exp Ther 2002; 302:1055–1061. 137 Masferrer JL, Leahy KM, Koki AT, et al. Antiangiogenic and antitumor activities of cyclooxygenase-2 inhibitors. Cancer Res 2000; 60:1306–1311. 138 Bukowski RM. New antiangiogenic agents for renal cell carcinoma: interferon alfa and thalidomide. Curr Oncol Rep 2004; 6:86–87. 139 Gasparini G. Metronomic scheduling: the future of chemotherapy? Lancet Oncol 2001; 2:733–740. 140 Vogelzang N. Metronomic chemotherapy: teaching old drugs new tricks? Clin Adv Hematol Oncol 2004; 2:432–433. 141 Munoz R, Shaked Y, Bertolini F, et al. Anti-angiogenic treatment of breast cancer using metronomic low-dose chemotherapy. Breast 2005; 14:466–479. 142 Bocci G, Tuccori M, Emmenegger U, et al. Cyclophosphamide-methotrexate ‘metronomic’ chemotherapy for the palliative treatment of metastatic breast cancer. A comparative pharmacoeconomic evaluation. Ann Oncol 2005; 16:1243–1252. 143 Kerbel RS and Kamen BA. The anti-angiogenic basis of metronomic chemotherapy. Nat Rev Cancer 2004; 4:423–436. 144 Denekamp J. Vascular attack as a therapeutic strategy for cancer. Cancer Metastasis Rev 1990; 9:267–282. 145 Denekamp J. Review article: angiogenesis, neovascular proliferation and vascular pathophysiology as targets for cancer therapy. Br J Radiol 1993; 66:181–196. 146 Tozer GM, Prise VE, Wilson J, et al. Mechanisms associated with tumor vascular shut-down induced by combretastatin A-4 phosphate: intravital microscopy and measurement of vascular permeability. Cancer Res 2001; 61:6413–6422. 147 Zwi LJ, Baguley BC, Gavin JB, et al. Correlation between immune and vascular activities of xanthenone acetic acid antitumor agents. Oncol Res 1994; 6:79–85. 148 Thorpe PE. Vascular targeting agents as cancer therapeutics. Clin Cancer Res 2004; 10:415–427. 149 Corbett TH, Bissery MC, Woziak A, et al. Activity of flavone acetic acid (NSC-347512) against solid tumors of mice. Invest New Drugs 1986; 4:207–220. 150 Ching LM, Cao Z, Kieda C, et al. Induction of endothelial cell apoptosis by the antivascular agent 5,6Dimethylxanthenone-4-acetic acid. Br J Cancer 2002; 86:1937–1942.
References 267
151 Hill SA, Lonergan SJ, Denekamp J, et al. Vinca alkaloids: anti-vascular effects in a murine tumour. Eur J Cancer 1993; 29A:1320–1324. 152 Baguley BC, Holdaway KM, Thomsen LL, et al. Inhibition of growth of colon 38 adenocarcinoma by vinblastine and colchicine: evidence for a vascular mechanism. Eur J Cancer 1991; 27:482–487. 153 Galbraith SM, Chaplin DJ, Lee F, et al. Effects of combretastatin A4 phosphate on endothelial cell morphology in vitro and relationship to tumour vascular targeting activity in vivo. Anticancer Res 2001; 21:93–102. 154 Galbraith SM, Maxwell RJ, Lodge MA, et al. Combretastatin A4 Phosphate has tumour anti-vascular activity in rat and man as demonstrated by dynamic magnetic resonance imaging. J Clin Oncol 2003; 21:2831–2842. 155 Jameson MB, Thompson PI, Baguley B, et al. Clinical aspects of a phase I trial of 5,6-dimethylxanthenone-4-acetic acid (DMXAA), a novel antivascular agent. Br J Cancer 2003; 88:1844–1850. 156 Galbraith SM, Rustin GJ, Lodge MA, et al. Effects of 5,6dimethylxanthenone-4-acetic acid on human tumor microcirculation assessed by dynamic contrast-enhanced magnetic resonance imaging. J Clin Oncol 2002; 20:3826–3840. 157 Tozer GM, Prise VE, Wilson J, et al. Combretastatin A-4 phosphate as a tumor vascular-targeting agent: early effects in tumors and normal tissues. Cancer Res 1999; 59:1626–1634. 158 Dowlati A, Robertson K, Cooney M, et al. A phase I pharmacokinetic and translational study of the novel vascular targeting agent combretastatin a-4 phosphate on a single-dose intravenous schedule in patients with advanced cancer. Cancer Res 2002; 62:3408–3416. 159 Stevenson JP, Rosen M, Sun W, et al. Phase I trial of the antivascular agent combretastatin A4 phosphate on a 5-day schedule to patients with cancer: magnetic resonance imaging evidence for altered tumor blood flow. J Clin Oncol 2003; 21:4428–4438. 160 Rustin GJ, Galbraith SM, Anderson H, et al. Phase I clinical trial of weekly combretastatin A4 phosphate: clinical and pharmacokinetic results. J Clin Oncol 2003; 21:2815–2822. 161 Blakey DC, Westwood FR, Walker M, et al. Antitumor activity of the novel vascular targeting agent ZD6126 in a panel of tumor models. Clin Cancer Res 2002; 8:1974–1983. 162 Horsman MR and Murata R. Vascular targeting effects of ZD6126 in a C3H mouse mammary carcinoma and the enhancement of radiation response. Int J Radiat Oncol Biol Phys 2003; 57:1047–1055. 163 LoRusso S, Gadgeel S, Wozniak A, et al. A Phase I dose escalation trial of ZD6126, a novel vascular targeting agent, in patients with cancer refractory to other treatments. Proc AACR-NCI-EORTC. 2001; Abstr 36. 164 Gadgeel S, LoRusso PM, Wozniak AJ, et al. A dose escalation study of the novel vascular targeting agent ZD6126 in patients with solid tumours. Proc Am Soc Clin Oncol 2002; 21:Abstr 438.
165 Radema S, Beerepoot LV, Witteveen PO, et al. Clinical evaluation of the novel vascular targeting agent ZD6126: assessment of toxicity and surrogate markers of vascular damage. Proc Am Soc Clin Oncol 2002; 21:Abstr 439. 166 DelProposto Z, LoRusso PM, Latif Z, et al. MRI evaluation of the effects of the vascular targeting agent ZD6126 on tumor vasculature. Proc Am Soc Clin Oncol 2002; 21:Abstr 440. 167 Hori, K. and Saito, S. Microvascular mechanisms by which the combretastatin A-4 derivative AC7700 (AVE8062) induces tumour blood flow stasis. Br J Cancer 2003; 89:1334–1344. 168 Sessa C, LoRusso P, Tolcher AW, et al. A Pharmacokinetic and DCE-MRI-dynamic Phase I study of the antivascular combretastatin analogue AVE8062A administered every 3 weeks. Proc Amer Assoc Cancer Res 2005; 46:Abstract 5827. 169 Hill SA, Tozer GM, Pettit GR, et al. Preclinical evaluation of the antitumour activity of the novel vascular targeting agent Oxi 4503. Anticancer Res 2002; 22:1453–1458. 170 Kirwan IG, Loadman PM, Swaine DJ, et al. Comparative preclinical pharmacokinetic and metabolic studies of the combretastatin prodrugs combretastatin A4 phosphate and A1 phosphate. Clin Cancer Res 2004; 10:1446–1453. 171 Padhani AR. MRI for assessing antivascular cancer treatments. Br J Radiol 2003; 76 Spec No 1:S60–80. 172 Leach MO, Brindle KM, Evelhoch JC, et al. The assessment of antiangiogenic and antivascular therapies in early-stage clinical trials using magnetic resonance imaging: issues and recommendations. Br J Cancer 2005; 92:1599–1610. 173 Thoeny HC, De Keyzer F, Chen F, et al. Diffusion-weighted MR imaging in monitoring the effect of a vascular targeting agent on rhabdomyosarcoma in rats. Radiology 2005; 234:756–764. 174 Anderson HL, Yap JT, Miller MP, et al. Assessment of pharmacodynamic vascular response in a phase I trial of combretastatin A4 phosphate. J Clin Oncol 2003; 21:2823–2830. 175 Denny WA and Wilson WR. Tirapazamine: a bioreductive anticancer drug that exploits tumour hypoxia. Expert Opin Investig Drugs 2000; 9:2889–2901. 176 Gandara DR, Lara PN, Goldberg Z, et al. Tirapazamine: prototype for a novel class of therapeutic agents targeting tumor hypoxia. Semin Oncol 2002; 29:102–109. 177 Reck M, Von Dawel J, Nimmermann C, et al. (Phase II-trial of tirapazamine in combination with cisplatin and gemcitabine in patients with advanced non-small-cell-lung-cancer (NSCLC)). Pneumologie 2004; 58:845–849. 178 Gatineau M, Rixe O, Chevalier TL, et al. Tirapazamine with cisplatin and vinorelbine in patients with advanced nonsmall-cell lung cancer: a phase I/II study. Clin Lung Cancer 2005; 6:293–298. 179 Williamson SK, Crowley JJ, Lara PN, et al. Phase III trial of paclitaxel plus carboplatin with or without tirapazamine in advanced non-small-cell lung cancer: Southwest Oncology Group Trial S0003. J Clin Oncol 2005; 23:9097–9104. 180 Rischin D, Peter L, Fisher R, et al. Tirapazamine, Cisplatin, and Radiation versus Fluorouracil, Cisplatin, and Radiation in patients with locally advanced head and neck cancer: a
268 Angiogenesis as a target for the treatment of cancer
randomized phase II trial of the Trans-Tasman Radiation Oncology Group (TROG 98.02). J Clin Oncol 2005; 23:79–87. 181 Patterson LH. Bioreductively activated antitumor N-oxides: the case of AQ4N, a unique approach to hypoxia-activated cancer chemotherapy. Drug Metab Rev 2002; 34:581–592. 182 Patterson LH and McKeown SR. AQ4N: a new approach to hypoxia-activated cancer chemotherapy. Br J Cancer 2000; 83:1589–1593. 183 Semenza GL. Development of novel therapeutic strategies that target HIF-1. Expert Opin Ther Targets 2006; 10:267–280.
184 Bilenker JH, Flaherty KT, Rosen M, et al. Phase I trial of combretastatin a-4 phosphate with carboplatin. Clin Cancer Res 2005; 11:1527–1533. 185 Rustin GJS, Nathan PD, Boxall J, et al. A dose escalation study combining combretastatin A-4 phosphate (CA4P) with carboplatin or paclitaxel in patients with advanced cancer. Clin Cancer Res 2005; 11:8968. 186 Ng QS, Carnell D, Milner J, et al. Phase Ib trial of Combretastatin A4 Phosphate (CA4P) in combination with radiotherapy (RT): Initial clinical results. J Clin Oncol (Meeting Abstracts). 2005; 23:3117
13 Gene therapy KATE RELPH, KEVIN HARRINGTON AND HARDEV PANDHA
Introduction Vectors for cancer gene therapy Regulation of expression
269 269 275
INTRODUCTION Over the past few decades, our knowledge of the molecular and genetic basis of disease has dramatically increased, which, combined with advances in technology, has resulted in novel molecular therapies for disease. In particular, gene therapy, which involves the transfer of genetic material to cells in order to produce a therapeutic effect, has become a promising approach, with 1145 clinical trials involving more than 3500 individuals currently being carried out worldwide (Journal of Gene Medicine, www.wiley.co.uk/wileychi/genmed). Initially, gene therapy was considered for monogenic disorders such as cystic fibrosis, in which one defective gene could be replaced by a functional gene. Indeed, the first gene therapy clinical trial was for adenosine deaminase deficiency,1 and we have almost reached a point at which single-gene disorders have
Table 13.1 Indications addressed by gene therapy clinical trials Disease Cancer Monogenic diseases Vascular diseases Infectious disease Other diseases Gene marking Healthy volunteers
Number of trials 762 100 100 75 37 52 19
Taken from Journal of Gene Medicine (http://www.wiley.co.uk/ genmed/clinical/).
Strategies for therapy Summary References
276 280 280
been cured. However, it is now recognized that gene therapy represents an effective approach for more complex diseases such as cancer, either as single agents or in conjunction with chemotherapy or radiotherapy.2 Cancer gene therapy represents one of the fastest growing areas of pre-clinical and clinical cancer research. Out of the 1145 gene therapy clinical trials being undertaken, 762 are in cancer gene therapy (Table 13.1). This chapter discusses the current strategies for cancer gene therapy in both preclinical and clinical settings, as well as the vectors used to deliver the genes to the target cells.
VECTORS FOR CANCER GENE THERAPY Although great progress has been made over the past few years, scientists are still trying to optimize vectors for cancer gene therapy to give them maximal efficiency and minimal toxicity. An ideal vector for the gene therapy of cancer would be highly targeted, resistant to immune degradation and safe for the recipient. The therapeutic effect of the gene should be targeted and regulated for as long as therapy is required.
Viral vectors Viruses have complex structures and life cycles, but analysis of individual viral genomes and life cycles has produced highly efficient gene delivery vehicles.3 The usual approach to constructing viral vectors is to remove the unneeded or pathogenic features of the virus, such as the virulence genes or packaging signals, whilst retaining the efficiency
270 Gene therapy
of gene delivery and expression. Genetic analysis allows determination of viral elements that are essential in cis or that can be supplied in trans. In these cases, the design of complementation systems in the form of packaging cells that supply the trans complementing functions have been developed.4 In turn, packaging cells can be used as producer cells for the generation of certain vectors. The two main obstacles to the generation of safe and efficient viral vectors are generating vectors at sufficient titre to target all the required cells, and avoiding the generation of replication competent or wild-type virus.5 However, with regard to cancer gene therapy, in addition to developing viral vectors that deliver therapeutic genes to the target cells, viral lysis of the cells is now an important arm of attack, and thus the replicative potential of certain viruses is also of importance. This chapter discusses both non-replicating and replicating viral vectors and how they are being used in the treatment of cancer. ADENOVIRAL VECTORS
Adenoviruses are one of the most popular types of vector currently used in cancer gene therapy. Adenoviral vectors are attractive reagents for gene therapy because of their ability to transduce genes into a broad range of cells and to infect both dividing and non-dividing cells.6 In addition, adenoviral vectors can accommodate large segments of DNA (up to 7.5 kilobases), and the viral genome rarely undergoes rearrangement, ensuring that therapeutic genes are maintained without change during virus replication (Table 13.2). Moreover, most adults have been exposed to the adenoviral serotypes used in gene therapy; in fact, more than 50 per cent of the worldwide population show antibodies to adenovirus serotype 5, suggesting that it is particularly safe. Furthermore, adenoviruses replicate
episomally and do not insert their genome into that of the host cell, ensuring less disruption of vital cellular genes and processes and reduced risk of insertional mutagenesis. This can be a limitation, in that transient expression of the therapeutic gene may be inadequate to treat chronic conditions such as cystic fibrosis. However, for situations in which short-term activity of the gene is needed, such as in the expression of suicide genes selectively in tumour cells, these viruses are suitable vectors. Adenoviruses have linear, double-stranded DNA genomes of approximately 38 kilobases. The capsid is nonenveloped and is comprised of the following structural proteins: the hexon, the fibre knob (which binds to the coxsackie and adenovirus receptor (CAR) on the surface of target cells) and the penton (which binds to αvβ3,5-integrins on the target cell surface and allows internalization).7 The structure of the adenovirus genome is shown in Figure 13.1. There are four early transcriptional units (E1, E2, E3 and E4), which have regulatory functions, and a late transcript, which codes for structural proteins. These genes can be separated into two groups; the cis- genes, such as those responsible for the packaging signal, which must be carried by the virus itself, and the trans- genes which can generally be complemented and therefore replaced with ‘foreign’ DNA. The first generation of adenoviral vector was used for the delivery of genes in monogenic disorders. In these vectors, the E1 region was removed to make way for the therapeutic gene and to inhibit viral replication. However, despite the removal of this region, there was still low-level transcription of viral genes which led to a host cellular immune response and a reduction in the period of gene expression due to cell-mediated destruction of the transduced cells.8 In addition, these types of vectors allowed the generation of E1-containing replication-competent
Table 13.2 Viral vectors for use in gene therapy applications Packaging capacity (kb)
Host range
Features
Retrovirus
8
Dividing cells only
Genome integration, long-term expression
Adenovirus
7.5
Most cells – low transduction of neurons
Transient expression, strong immunogenicity
Adeno-associated virus (AAV)
4
Infects both dividing and non-dividing cells
Slow initial expression, genome integration, long-term expression, inefficient large-scale production
Alphaviruses
7.5
Most cells – neuron and glial cell-specific strains
Transient, but extreme, expression levels, low immunogenicity
Herpes simplex virus
30
Most cells
Latent infection, long-term expression, low toxicity
Lentivirus
8
Dividing and non-dividing cells
Genome integration, long-term expression, safety concerns, low titres, production inefficient
Vector
Vectors for cancer gene therapy 271
adenovirus (RCA) due to homologous recombination in 293 cells, which further enhanced the adverse effects.9 In order to address these problems homologies between the vectors and the complementing cell lines have been reduced. Second-generation adenoviral vectors have further deletions in E2a, E2b or E4 and have reduced immunogenicity and RCA generation (Fig. 13.1). Despite these improvements, the complementing cell lines are difficult to engineer, can be difficult to grow and can lead to poor viral titres. As a result, a third generation of adenoviral or gutless vectors has been created.10 These have all of the viral genes deleted (except for the packaging signal) and replaced with the therapeutic gene of interest. They are therefore free from problems associated with immunogenicity and demonstrate long-term transgene expression. They are generated with a helper virus, which contains all of the genes necessary for viral replication but which contains a deletion in the packaging signal to ensure that it is not incorporated into the final vector. These vectors are still undergoing development in order to improve their purity and large-scale manufacture.11 RETROVIRAL VECTORS
Retroviral vectors are the second most popular vectors currently in clinical trials for cancer after adenoviruses, and were among the first constructed vectors for gene therapy. Retroviruses are small RNA viruses that enter the cell via an interaction between the viral envelope protein and a cellular
E1B 55Kd, 19Kd
10
L2 III, pVII, V
E3 L3 12.5Kd, 6.7Kd, gp19Kd, ADP, pVI, II, Pr RIDab, 14.7Kd
L1 IIIa
E1A 243R, 289R
20
L4 100Kd, 33Kd, pVII L5 IV
30
40
50
60
70
80
90
100
ITR E2B pTP, Pol
E2A DBP
receptor. After entry into the cell, the viral RNA is converted to double-stranded DNA (Fig. 13.2), which integrates into the host cell genome. The integrated retrovirus then produces new RNA genomes, which are packaged up into core particles that become encased in the lipoprotein envelope as they bud out of the cell membrane. The retroviral genome consists of genes essential for replication, including GAG (capsid), POL (reverse transcriptase) and ENV (envelope), together with a packaging signal that allows the genome to be packaged into virions. The construction of retroviral vectors is based on the removal of GAG, POL and ENV genes and their replacement with the desired therapeutic gene (up to 10 kb). The deletion of these essential genes makes the vector replication defective unless complemented in a packaging cell line. This is an important safety feature ensuring that the viral vectors cannot replicate and produce viral particles after transduction of the target cells. Several packaging cell lines have been developed to prevent recombination between the vector and the packaging line, which may produce replication-competent retroviruses.12 The packaging signal remains in the vector to ensure that the therapeutic gene is packaged into virions, which then bud out through the membrane of the cell. The long terminal repeats (LTRs) at either end of the genome contain regulatory sequences such as the promoter and enhancer that control gene expression. The main advantage of retroviral vectors is that they are able to integrate stably into the genome of the target cell. This ensures long-term expression of the therapeutic gene. However, retroviral vectors have several limitations including: (i) low titre, (ii) low transduction efficiency, (iii) particle instability and difficult to concentrate, and (iv) inability to transduce non-dividing cells. Despite these obstacles, retroviral vectors are currently in use in 70 clinical trials for cancer gene therapy (Journal of Gene Medicine website, 2006). Initial phase I and II clinical studies in patients with recurrent malignant glioma have shown
ITR
E4 Orfs 1-6/7
– Packaging signal
(a) E1( E3)-deleted vector (first generation vector)
DNA U3 R U5
Transgene
U3 R U5
Transcription
(b) E1( E3) E2/4 -deleted vector (second generation vector)
ΔE2
Transgene
ΔE3
ΔE4
RNA
U3 R
ΔE1
Reverse transcription
(c) High capacity vector (Ⳋgutless vector)
Transgene I
AAA
R U5
Transgene II
Figure 13.1 Schematic of adenoviral genome and adenoviral vectors. E1A must be removed to prevent recombinant virus from replicating. ITR, inverted terminal repeats.
DNA
Figure 13.2 Replication of retroviral genome. U3 region contains the viral promoter and enhancer. After transcription and reverse transcription the 3 U3 region is transferred to the 5 position.
272 Gene therapy
favourable safety profiles and some efficacy with retroviralmediated gene therapy. On the other hand, a prospective randomized phase III clinical study of retroviral gene therapy in primary malignant glioma failed to demonstrate significant extension of the progression-free or overall survival times in retroviral-treated patients. The failure of this retroviral gene therapy study may be due to the low tumour-cell transduction rate observed in vivo.13 In an attempt to avoid the problems associated with retroviral vectors, lentiviral vectors have been constructed. Lentiviruses are a special group of retroviruses – which includes human immuodeficiency virus (HIV) – that have the ability to transduce non-dividing cells whilst ensuring long-term expression and minimal inflammatory response (Table 13.2).
vectors, the risk of insertional mutagenesis is reduced, as integration occurs preferentially at a site-dependent locus on chromosome.19 However, in AAV vectors for cancer gene therapy, this characteristic integration is lost due to deletion of rep proteins (in an attempt to reduce the risk of the emergence of replication-competent AAV). Other potential drawbacks for the use of AAV vectors for cancer gene therapy are that they have limited packaging capacity (approximately 5 kb), and gene expression may be slow to reach its peak. In addition, production requires the use of helper viruses, which means that preparations for preclinical and clinical use may be contaminated with these entities.
HERPES SIMPLEX VIRAL VECTORS
Alphaviruses are members of the Togaviridae and are icosahedral enveloped viruses. There are three different alphaviruses that have been developed as vectors for gene therapy: Semliki forest virus (SFV), Sinbis virus (SIN) and Venezuelan equine encephalitis virus (VEE).18 The replication-competent alphaviruses are able to maintain all of their wild-type genes and incorporate a therapeutic gene of interest. In addition, due to their broad host range, the alphavirus vectors are able to transduce a variety of tumour cells in which they have been shown to induce apoptosis .19 Other useful features for their use in cancer gene therapy are their immunogenicity and ability to stimulate cytotoxic T-cell responses, making them suitable for vaccine production against tumours.20
Herpes simplex virus (HSV) is a highly cytotoxic human pathogen with a large and complex genome encoding at least 80 gene products. Despite its cytotoxicity, not all its natural infections lead to cell death: in neurons, the virus establishes a latent infection in which the viral genome remains episomal in the nucleus throughout the lifetime of the organism. Given its natural tropism, herpes simplex vectors have primarily been developed for delivery to neurons.14 There are several features of HSV biology that make it attractive as a vector: the virus grows well in culture to give high-titre stocks; genetic manipulation of the genome is straightforward and large amounts of DNA can be accommodated; and the latent episomal genome means integration into the host genome is avoided. Two types of vector are available: (i) replication-defective vectors, in which the cytotoxicity has been abolished by the removal of viral genes, such as deletion of the four immediate early genes ICP0, ICP4, ICP22 and ICP47,15 and (ii) amplicon vectors, which are plasmids packaged into HSV particles with the aid of a helper virus.16 Such vectors retain their ability to infect cells, are non-replicative and have enormous packaging capacity. Their use potentially circumvents the cytotoxicity and immune activation associated with viral gene expression. However, in the context of cancer gene therapy, such considerations have limited significance and these vectors are more likely to be of use in the treatment of non-malignant conditions. ADENO-ASSOCIATED VIRAL VECTORS
Adeno-associated viruses (AAVs) belong to the family of parvoviruses. They are a single-stranded, non-enveloped DNA viruses of 4.5 kb in size that establish latent infections in cells.17 They are single–stranded viruses capable of infecting non-dividing cells and are stably integrated/maintained in the host genome, although this issue is of lesser importance in cancer gene therapy, where transient expression of cytotoxic genes should be sufficient to achieve a therapeutic effect. An additional benefit is that, in contrast to retroviral
ALPHAVIRUS VECTORS
ONCOLYTIC VIRAL VECTORS
Replication-selective oncolytic viruses (virotherapy) represent a novel and unique approach to the treatment of cancer. Lytic viruses have evolved to infect cells, replicate, induce cell death, release viral particles and spread to surrounding tissue. Selective replication of the viruses within tumour tissue could increase the therapeutic index of these agents dramatically. In addition, the fact that oncolytic viruses do not always induce cell death via classical apoptotic pathways makes the likelihood of cross-resistance with standard regimens such as chemoradiotherapy or radiotherapy much less likely. The oncolytic viruses currently considered for cancer gene therapy include oncolytic adenovirus, HSV, reovirus, parvovirus, vesicular stomatitis virus and Newcastle virus,21 although most of the work has been based on the HSV and adenovirus. Over the past decade, advances in molecular biology have enabled these viruses to be engineered to enhance their safety and antitumour potency. Oncolytic adenoviruses Adenoviruses mediate cell death via several mechanisms. Viral proteins expressed late in the course of the viral life cycle are directly cytotoxic. These include the E3 11.6-kDa adenovirus death protein and E4ORF4.22 Deletion of these
Vectors for cancer gene therapy 273
Table 13.3 Selection of oncolytic viral vectors in clinical trials Virus name
Virus type
Mechanism
ONYX-015(100) CV787 HSV1716 NV1020
Adenovirus 5 Adenovirus 5 HSV HSV
Reovirus NDV PV 701 Mumps (Urabe) MV-CEA
Reovirus Newcastle disease Mumps Measles virus
E1B interference by partial deletion E1B driven by PSA promoter ICP34.5 deleted 700 bp tk deletion 15 kb deletion across the joint region, which contains an exogenous copy of tk gene under control of HSV-1α4 promoter and 5.2 kb fragment of HSV-2 DNA Natural strains of reovirus Attenuated strain of NDV Attenuated strain of mumps virus Modified MV; expresses soluble carcinoembryonic antigen (CEA)
HSV, herpes simplex virus; CEA, carcinoembryonic antigen; MV, measles virus; NDV, newcastle disease virus; PSA, prostate specific antigen
gene products results in a significant delay in cell death. Expression of the E1A protein early in the adenovirus life cycle makes the cells more refractory to killing via tumour necrosis factor (TNF); this effect is inhibited by the E3 proteins 10.4, 14.5 and 14.7. Deletion of these three E3 proteins leads to an increased TNF expression in vivo and enhanced cell sensitivity to TNF.23 There are currently two main approaches to achieving tumour-selective adenoviral replication. The first is via the use of tumour-specific promoters, which are used to drive the expression of the E1A gene in tumour cells alone. E1A functions to stimulate S phase and to stimulate both viral and cellular genes that are critical for efficient viral replication.24 This approach has been studied in a phase I clinical trial, which used the prostate-specific antigen (PSA) promoter to drive the expression of the E1A gene in patients with locally recurrent prostate carcinoma. This virus was termed CN706 and was injected directly into the tumour. Similar approaches have been used by other groups to achieve selective replication in other tumour types, including using the promoters from alpha-fetoprotein (AFP), carcinoembryonic antigen and MUC-1 (mucin-1).25 One of the first clinical trials demonstrating antitumour efficacy in a specific cancer used a replicationconditional adenovirus. This virus, dl1520, also known as ONYX-015, is defective in the early regulator protein E1B, which binds to and inactivates p53 to promote its own activation.26 In normal cells, p53 inactivates adenoviral replication, but the exact mechanism by which it does so is still not clear. This mutated virus can infect and replicate in cells defective in p53 as well as in cells with loss of p14ARF function (a protein that can mediate apoptosis by activation of p53). However, it cannot replicate in normal cells carrying wild-type p53 and an intact p53 pathway.27 To date, phase I and II trials have been conducted with virus alone or in combination with chemotherapy. dl1520 has been well tolerated at the highest practical doses that
could be administered (2 1012–2 1013 particles) by intra-tumoral, intra-peritoneal, intra-arterial and intravenous routes. Flu-like symptoms were the most common toxicities and were increased in patients receiving intravascular treatment. Two phase II trials enrolled a total of 40 patients with head and neck cancer.28 Despite a fairly aggressive injection regimen of six to eight daily needle passes for 5 consecutive days, no objective responses were documented. Similarly, no objective responses were noted in phase I or I/II trials in patients with pancreatic, colorectal or ovarian carcinomas.29 As a result, combinations with chemotherapy were explored. Evidence for a potentially synergistic interaction between oncolytic adenoviral therapy and chemotherapy has been obtained in multiple trials. Encouraging clinical data have been achieved in patients with recurrent head and neck cancer treated with intratumoral ONYX-015 in combination with cisplatin and 5-fluoruracil.30 Out of 30 patients treated, an objective response (at least 50 per cent reduction in tumour size) was observed in 19 patients and a complete response was seen in 8 patients. Tumours as large as 10 cm regressed completely, and none of the tumours that responded had progressed after a mean follow-up of 5 months. One reason why the ONYX-015 vector has limited efficacy in some studies could be the lack of the CAR on the surface of target tumour cells preventing intra-tumoral spread.31 Oncolytic herpes simplex virus In the setting of cancer gene therapy thus far, most attention has focused on the creation of oncolytic HSV vectors. A variety of vectors have been created with mutations in their genomes to ensure that only tumour cells are lysed. One strategy for targeting cancer cells involved manipulation of the gene encoding for ribonucleotide reductase (RR). As the name implies, this enzyme reduces ribonucleotides to deoxyribonucleotides and is necessary for DNA synthesis.32 The mutation inactivates the viral RR
274 Gene therapy
and renders the virus tropic for cells such as tumour cells with an active RR. Another, more widely used, strategy employed to attenuate the toxicity of HSV was deletion of one or both of the γ134.5 genes that encode the protein ICP34.5 (HSV-1716). When the host cells become infected with virus it phosphorylates the transcription factor eIF-2α, which leads to shutting off of host cell protein synthesis. The γ134.5 genes are thus important for viral synthesis and therefore neurovirulence. Cancer cells, however, continue to support replication of the viruses when infected with these mutant viruses.33 Strains G207 and MGH-1 contain both the RR and γ134.5 mutations, greatly reducing the risk that viruses will revert to wild type.34 Initially, studies with HSV mutants were undertaken in central nervous system tumours,35 but since then a range of tumours have shown susceptibility to HSV-mediated oncolysis. Efficient dissemination of viral vectors within the tumour mass has been identified as being crucial for achieving clinical efficacy.36 In this regard, oncolytic replicationcompetent viral vectors might be the best choice for cancer gene therapy applications.
Table 13.4 Non-viral vectors for cancer gene therapy Vector
Characteristics
Cationic polymers, e.g. poly(ethylenimine) (PEI), poly(L-lysine) (PLL)
Synthetic cationic polymer that can be combined with DNA to form a particulate complex
Cationic peptides
Amphiphilic peptides that can undergo conformational changes in acidic environments escaping the endosomal lysosomal pathways They contain positively charged amino acids so that they can effectively condense DNA
Cationic lipids (liposomes)
Liposomes are well-studied non-viral vectors Used mainly in the form of liposomes Lipid emulsions have been more recently evaluated
Non-viral delivery systems Although viral vector systems have so far been the most popular choice for cancer gene therapy, non-viral systems are also being developed. The advantages of non-viral vectors are a virtually unlimited cloning capacity, ease and quantity of production, low host toxicity and immunogenicity and the ability to repeat application. NAKED DNA INJECTION
The simplest form of non-viral gene therapy is the transfer of naked plasmid DNA containing the therapeutic gene of choice driven by a promoter. The DNA is taken into the cells after injection by an undefined mechanism and the gene is expressed episomally or becomes integrated into the host cell genome. In cancer gene therapy, naked DNA has been used in several pre-clinical and clinical trials, including cytokine gene therapy. The most effective route for naked DNA appears to be intra-muscular, but intra-tumoral injection has been shown to be possible in some tumour types,37 although it does result in rapid clearance from the tumour.38 Despite the advantages of naked DNA, the disadvantages still remain: low transfection efficiency and short-term expression.39 Several methods and systems have been developed in order to try to overcome these problems. One such method has been to try to improve DNA penetration of the cell membrane using electroporation.40 LIPOPLEXES AND POLYPLEXES
In addition to the transfer of naked DNA into cells, synthetic vectors have been developed to aid the transfer of DNA into cells and also to protect it from degradation.
The most common polymers in use are derived from lipids. The plasmid DNA is encapsulated in a lipid layer to form micelles or liposomes known as a lipoplexes. Anionic and neutral liposomes were initially considered for gene transfer due to their safety and compatibility with body fluids. However, these complexes were quite hard to construct and the level of gene transfer was relatively low. Recently, novel anionic and neutral complexes have been constructed for use in in-vivo gene therapy. Despite this, the vast majority of work has concentrated on cationic liposomes, which naturally create a complex with DNA and, by their positive nature, are attracted to negatively charged cell membranes and subsequently endocytosed (Table 13.4). Liposome delivery of DNA has been most extensively used in cancer gene therapy to deliver genes to stimulate anti-tumour responses or anti-oncogenes.41 However, despite their advantages, they are still associated with toxicity problems when used in vivo.42
BACTERIA
In addition to viral vectors, bacteria are now being considered as vehicles for gene delivery. This approach exploits the unique necrotic and hypoxic environment that exists in most solid tumours and the ability of some anaerobic bacteria to colonize and replicate in these areas.43 The currently known and tested bacterial vectors for use in gene therapy fall into two categories: (i) strictly anaerobic bacteria such as Clostridium and Bifidobacterium,44 and (ii) Salmonella typhimurium, which require tumour-specific nutrition factors to replicate and in doing so deplete the supply of factors preventing tumour-cell growth.45
Regulation of expression 275
REGULATION OF EXPRESSION A central problem of conventional cancer therapies is their lack of specificity for tumour tissue, which results in severe side effects and dose limitations below efficacy of the drug. Similarly, gene therapy is also faced with a specificity problem and in order for cancer gene therapy to be effective and efficient, it is essential that a high therapeutic index is achieved in order to prevent damage to surrounding tissue and to reduce the risk of germ-line transduction. Targeting can be achieved on two main levels: transcriptional targeting, which uses tissue/tumour-specific elements to drive the transcription of the gene to the specified tissue, and transductional targeting, which modifies the tropism of the vector for the target cell.
Transcriptional targeting In many gene therapy vectors, the standard promoter in use is that from cytomegalovirus (CMV). This drives a high level of expression of the gene in a wide variety of tissues but lacks expression selectivity. In addition, it has the added disadvantage of being down-regulated in vivo.46 Many laboratories have focused on the design of new promoters for cancer gene therapy (Table 13.5).
TISSUE-SPECIFIC AND TUMOUR-SPECIFIC PROMOTERS
Many tissue specific promoters have been identified (see Table 13.5). These include the PSA promoter47 and the tyrosinase promoter.48 However, the main drawback of these promoters is that, as their name implies, they are also active in the normal tissue from which the tumour has originated. Therefore unless the loss of normal cells is acceptable to the therapy, another level of targeting needs to be incorporated into the vector, especially if it is to be administered systemically. A variety of genes have been shown to be exclusively expressed in tumour tissue. These include the gene encoding oncofetal AFP, which is active in foetal liver but becomes reactivated in hepatoma cells.49 Disease-specific conditions associated with the tumour include deregulation of signalling pathways. Up-regulation of the ErbB-2 promoter and Mucin-1 gene (MUC-1) have been seen in a range of adenocarcinomas and have therefore been exploited by a range of groups targeting expression to tumours.50 In addition to tumour-specific and tissue-specific promoters, several other inducible promoters have been considered. One such system is the use of promoters inducible by certain conventional cancer therapies such as the early growth response 1 (Egr-1) promoter, which is inducible by therapeutic doses of ionizing radiation51 or by chemotherapeutic agents such as the promoter of the multi-drug resistance gene (MDR-1).52 Several groups have also used cell-cycle-specific regulatory elements
Table 13.5 Tissue-specific promoters Promoter Tissue specific Tyrosinase Prostate-specific antigen (PSA) Albumin Muscle creatine kinase (MCK) Glial fibrillary acidic protein (GFAP) Neuron specific enolase (NSE)
Target
Melanocytes Prostate cancer Liver Muscle Glial cells Neurons
Tumour endothelium Kinase insert domain-containing receptor (KDR) E-selectin Endoglin Tumour selective Grp78 Carcinoembryonic antigen (CEA) Erythroblastic leukemia viral oncogene homolog 2 (erbB2) L-plastin 5-Aminolaevulinic acid (ALA) Beta lactoglobulin gene (BLG) Osteocalcin Secretory leukoprotease inhibitor (SLP1) Hypoxia response element (HRE) Alpha-fetoprotein (AFP) Muc1(DF3) Mucin-1 Hexokinase II Cell-cycle regulated E2F transcription factor 1 (E2F-1) Cyclin A Treatment responsive Early growth response protein 1 (Egr-1) Heat Shock Protein 70 (Hsp70) Mdr-1
Solid tumours Adenocarcinoma Breast and pancreatic cancer Cancer cells Breast cancer Breast cancer Osteosarcoma, prostate cancer Ovarian, cervical cancer Solid tumours Liver tumour Breast cancer Cancer cells Proliferating cells Proliferating cells Radiation induced Heat induced Chemotherapy induced
incorporated into vectors to target the highly proliferating malignant cells.53
Transductional targeting In addition to targeting at the level of transcription, targeted gene transfer can be achieved in both viral and non-viral vectors by the incorporation of specific ligands into the vectors. TRANSDUCTIONAL TARGETING OF VIRAL VECTORS
The identification of the route by which human cells uptake virus is an important step towards re-targeting viral
276 Gene therapy
vectors to different cell types. For example, the adenovirus fibre knob anchors onto the surface of the target cell by means of the Coxsackie Adenovirus Receptor (CAR) and interaction of the capsid penton protein with integrins αvβ3 and αvβ5 on the surface of target cells allowing internalization.54 Most immortalized tumour cell lines express CAR and are therefore easily transduced by adenoviral vectors. However, certain studies have demonstrated that 50 per cent of primary epithelial cancers do not express CAR.55 This may account for some of the limited success with past clinical trials using adenoviral vectors. Transductional targeting may improve the transfer of genes to particular cancer types and, in addition, re-targeting adenoviral vectors will permit the treatment of haematologic malignancies because haematopoietic stem cells are known to lack CAR.56 There are many reports of the re-targeting of adenoviral vectors to tumour cells via the use of antibodies directed towards specific antigens on the surface of a particular tumour type.57 One group used a neutralizing anti-fibre antibody conjugated to an antibody directed against the epithelial cell adhesion molecule (EGP-2), which is highly expressed on the surface of a range of adenocarcinomas from the stomach, oesophagus, breast, ovary, colon and lung and its expression is limited in normal tissue. In this study, the adenovirus specifically infected cancer cell lines expressing EGP-2, whilst gene transfer was dramatically reduced in EGP-2-negative cell lines.58
STRATEGIES FOR THERAPY It is now widely accepted that the pathogenesis of cancer is a multi-step process that involves the sequential accumulation of separate genetic events. There are many different genes involved in the development, progression, spread and refractoriness to treatment of the common cancers. Each of these abnormalities distinguishes a tumour cell from a normal cell and represents a possible target for therapy (Fig. 13.3). Within this chapter, cancer gene therapy strategies are grouped as follows: (i) corrective gene therapy that aims to restore the normal function of a deleted or mutated gene CANCER GENE THERAPY
Corrective gene therapy
Immunomodulatory gene therapy
Immune stimulant genes
Antioncogenes
Vaccines
Gene vaccines
Cell vaccines
Antisense RNAi oligonucleotides
Cytoreductive gene therapy
GDEPT
Antiangiogenesis
Induction of apoptosis
Figure 13.3 Strategies in cancer gene therapy.
Tumour suppressor genes
(usually a tumour suppressor gene) or to counteract the effect of a mutated oncogene; (ii) cytoreductive gene therapy in which delivery of an exogenous gene results in cancer-cell death by metabolism of a prodrug to a toxic agent (suicide gene therapy), induction of apoptosis, antiangiogenic activity or enhanced localization of therapeutic radioisotopes; and (iii) immunomodulatory gene therapy in which gene delivery to tumour or normal tissue enhances the ability of the immune system to mount an effective cytotoxic response.
Corrective gene therapy The genetic mutations associated with the development of cancer are many and varied, making it an extremely difficult process to understand. However, the genes that are directly implicated in transformation can be classified into two types: oncogenes and tumour suppressor genes. Corrective gene therapy involves either the abrogation of oncogene function or the restoration of tumour suppressor gene function. ONCOGENES
During development, the expression of proto-oncogenes is vital to provide the organism with all the necessary cells to grow. After fulfilling their requirements, these genes are then switched off to prevent abnormal growth. However, during the transformation process, oncogenes are switched back on, resulting in aberrant cell proliferation and therefore tumour growth. Examples of oncogenes that are frequently mutated in cancers are Ras, Myc, ErbB2, and Bcl-2.59 Cancer gene therapy strategies can be designed to negate the activating function of these mutated genes by targeting the transcriptional/translational machinery of the cell with the aim of preventing the cell from making functional oncogenic proteins. Therapeutic strategies can be designed to interfere with transcription or prevent the transcribed mRNA being translated into protein. This can be done in several ways. The first way is by inhibition of oncogene transcription into mRNA. For example, transcription of the oncogene can be inhibited by using short, single-stranded DNA oligonucleotides that bind to specific oncogene promoter regions and prevent transcription.60 A second way to reduce oncogene translation is by the use of antisense oligonucleotides that bind in a complementary manner to the mRNA. Antisense oligonucleotides have been used to reduce the translation of Bcl2 mRNA in prostate and breast cancer cells.61 Table 13.6 indicates some clinical trials that have been carried out with antisense oncogenes. Following antisense strategies, there are further possibilities of abrogating oncogene activity through the use of small inhibitory RNAs (siRNA). In RNAi-based (RNA interference) therapy, a doublestranded, short, interfering RNA (siRNA) molecule is engineered to match precisely the protein-encoding nucleotide
Strategies for therapy 277
sequence of the target mRNA to be ‘silenced’. Following administration, the siRNA molecule associates with a group of proteins termed the RNA-induced silencing complex (RISC), and directs the RISC to the target mRNA. The siRNA-associated RISC binds to the target mRNA through a base-pairing interaction and degrades it. The RISC is then capable of degrading additional copies of the targeted mRNA (Fig. 13.4). Short interfering RNA is currently the fastest growing sector of this anti-gene field for target validation and therapeutic applications. It has been shown to inhibit the growth of several types of tumour cells in vitro, including cervical and breast tumour cells.62 Although, in
Table 13.6 A selection of clinical trials involving antisense oligonucleotides Drug/oligo MG98 Oblimerson (Genasense G3139) OGX-011 ISIS3521 ISIS2503 ISIS5132 OL(1)p53 Bcl-2 GEM231
Protein target
Reference
DNA methyltransferase 1 (DNMT-1) BCL-2
101*
Clusterin PKCα H-Ras c-Raf P53 Bcl-2 PKA-1
103 104*** – 105* 106* 102* 107*
102*
theory, the development of genomics-based agents to inhibit gene expression is simple and straightforward, the fundamental concern centres on the capacity of the oligonucleotide to gain access to the target RNA.63 TUMOUR SUPPRESSOR GENES
Tumour suppressor genes encode a range of proteins that are concerned with regulating the cell cycle and DNA repair pathways. In general, when DNA becomes damaged, the cell cycle is arrested and the DNA repaired. If this is not possible, the cell will apoptose. By inhibiting the expression of tumour suppressor genes, cancer cells containing DNA mutations are able to continue proliferating and a tumour forms. Transfection of tumour suppressor genes into cells has proved a successful route to arresting growth and inducing apoptosis. Mutations of p53 have been reported in a wide variety of tumour types, including the common epithelial tumours such as lung, breast, colorectal, prostate, pancreatic, bladder, oesophageal and head and neck.64 Furthermore, there is abundant evidence that expression of mutant p53 is associated with a poor prognosis for most tumour types that have been reported.65 Pre-clinical studies have demonstrated decreased growth of cancer cells in vitro and in vivo after delivery of wild-type p53 by adenoviral,66 retroviral67 and cationic liposomal vectors.68 In addition, and perhaps most relevantly for clinical applications, liposome-mediated and viral-mediated delivery of p53 have been shown to sensitize cancer cells to the effects of genotoxic agents such as radiotherapy69 and chemotherapy70 in vitro and in vivo. As a result of these positive pre-clinical data, p53 has been the subject of a number of preliminary phase I clinical trials.71–74
siRNA 5-P 3-HO
OH-3
ADPPi
miRNA function
P-5
ATP
Cytoreductive gene therapy
RISC
siRNA function
RISC
RISC
RISC incorporates partially complementary mRNA
Specific or non-specific gene silencing depending on the degree of homology
Translational repression
Gene silencing
Figure 13.4 RNAi-mediated silencing. The siRNA associates with cellular proteins to produce an RNA-induced silencing complex (RISC) that contains a helicase that unwinds the siRNA. The antisense strand can guide RISC to the target mRNA for endonucleolytic cleavage. Each of the siRNA strands has the potential to be incorporated into RISC and direct RNA interference. A totally complementary RNA has the ability to degrade its target mRNA. However, an RNA with a partial match functions as an miRNA and causes translational repression.
The term cytoreductive gene therapy embraces approaches that aim to deliver genes that kill cells either directly or indirectly (by sensitizing them to the effects of drugs or radiation or by depriving them of their blood supply). A range of strategies can be considered under this heading, including gene-directed enzyme prodrug therapy (GDEPT) or ‘suicide’ gene therapy, genetic induction of apoptosis, anti-angiogenic gene therapy, and gene-directed radioisotopic therapy GENE-DIRECTED ENZYME PRODRUG THERAPY
This involves the delivery of a gene that encodes an enzyme that is able to convert a relatively innocuous prodrug to a toxic active agent within tumour cells (Fig. 13.5). The strength of this approach lies in its potential to avoid the systemic toxicity and lack of tumour specificity associated with existing cytotoxic agents by ensuring that they are generated in high concentrations only at the tumour site,
278 Gene therapy
Nontoxic prodrug
Suicide enzyme
Expression of suicide gene
Active toxic drug
Virus/vector encoding suicide enzyme
Nucleus
Tumour cell
Death of transduced tumour cell
Death of neighbouring cells through bystander effect
Figure 13.5 Gene-directed enzyme prodrug therapy (GDEPT). Gene therapy vector carries gene encoding suicide enzyme into the cell. The inactive prodrug is administered and when it hits the cells containing the transgene the suicide enzyme converts the non-toxic prodrug into an active toxic drug, which then moves to adjacent cells via bystander effect.
thus increasing their therapeutic index.75 The components of an idealized GDEPT system are as follows: ●
●
●
●
●
●
a gene that encodes an enzyme that catalyses the conversion of a prodrug to a cytotoxic metabolite; a mechanism of restricting gene expression to tumour tissue; an enzyme that elicits only a minimal host immune response (although in the case of immunomodulatory gene therapy, such an immune response may be beneficial); a prodrug with minimal intrinsic cytotoxicity and wellestablished pharmacokinetics; a toxic product that kills both cycling and non-cycling cells; a toxic agent that is able to diffuse to adjacent tumour cells to cause a bystander effect.
A number of candidate suicide genes have undergone preliminary clinical evaluation. The two most widely studied examples of GDEPT are (i) the HSV thymidine kinase (HSVtk) and ganciclovir (GCV) system;76 and (ii) the Escherichia coli bacterial cytosine deaminase (CD) and 5fluorocytosine (5-FC) system.77 The first gene therapy trial involving HSVtk/GCV was to treat ovarian cancer and
took place in 1991.78* Since then, numerous clinical trials for a range of cancers have taken place. A large number of these have involved the use of adenoviral79 and retroviral vectors.80 In general, in the phase I trials only moderately toxic responses have been reported, and in the phase II trials moderate therapeutic outcomes have been observed in some patients. For example, in a phase I/II study for recurrent glioma retroviral producer cells expressing HSVtk were injected into the cavity after tumour debulking followed by intravenous GCV. This resulted in no tumour recurrence in 4 out of 12 patients by 4 months and in 1 patient the tumour had not recurred at 2.8 years after treatment.81* Despite encouraging pre-clinical results, the CD system has not proved as popular for use in clinical trials. One phase I trial involving local injection of the CD gene regulated by the erbB-2 promoter and systemic 5-FC demonstrated the safety of this approach for the treatment of breast cancer.82 Table 13.7 lists a selection of the current trials that are being undertaken. As yet, no phase II trials are underway. GENETIC INDUCTION OF APOPTOSIS
Derangement of normal apoptotic signalling pathways is a central component of malignant transformation. Therefore, dissection of the underlying molecular biology of apoptosis has opened up new prospects for gene therapy. Signalling to tumour cells via specific apoptosisinducing ligands has been reported for TNF,83 TRAIL (TNF-related apoptosis inducing ligand)84 and FasL.85 Griffith et al.84 demonstrated that an adenoviral vector directing expression of TRAIL was able to induce apoptosis through caspase activation in a variety of cancer cell lines in vitro. Much work has been done with FasL, which has been shown to be effective after direct viral delivery to cancer cells86 or after delivery by FasL-secreting myoblast cells.87 Significantly, this approach has been shown to be capable of inducing apoptosis in cell lines that were resistant to the induction of apoptosis by conventional Fas-mediated pathways (agonist anti-Fas antibody or membrane-bound recombinant FasL).87 Alternative approaches have involved the delivery of pro-apoptotic members of the Bcl-2 family, such as Bak88 or Bax,89 or active caspase molecules.90 Such approaches have been shown to be effective both in vitro and in vivo. As yet, no clinical studies of cancer gene therapy using pro-apoptotic molecules have been reported. Clearly, as with all other approaches, tight control of the expression of these genes will be necessary to guard against the risk of toxicity. ANTI-ANGIOGENESIS
The ability of a tumour to grow depends on it having an adequate blood supply to provide it with all the necessary nutrients for growth (angiogenesis). Scientists have therefore sought to interfere with the process of angiogenesis via gene therapy. The main targets for anti-angiogenesis therapy
Strategies for therapy 279
Table 13.7 Cancer gene therapy phase III open trials (2006) NS not stated Investigator
Cancer
Vector
Gene type
Date approved
Delivery
UK
Prostate
Adeno-associated virus
2005-9
NS
UK
Prostate
Adeno-associated virus
Granulocyte–macrophage colony-stimulating factor (GM-CSF) GM-CSF
2005-9
NS
USA: Walter J. Urba
Prostate
Adeno-associated virus
GM-CSF
2004
In vitro
USA: Charles G. Drake
Prostate
Adeno-associated virus
GM-CSF
2005
In vitro
USA: John Marshall
Pancreas
Poxvirus vaccinia virus
Carcinoembryonic antigen (CEA) B7.1 (CD80) Intercellular adhesion molecule-1 (ICAM-1) Lymphocyte function-related antigen-3 (LFA-3) Mucin-1 (MUC-1)
2004
In vitro
Germany: Herr PD Dr. med. Andreas Dietz
Recurrent squamous-cell carcinoma of the head and neck
Adenovirus
p53
NS
In vivo
USA: John T. Hamm
Squamous-cell carcinoma of the head and neck
Adenovirus
p53
1999
In vivo
USA: John Nemunaitis
Squamous cell carcinoma of the head and neck
Adenovirus
p53
2000
NS
Rene Gonzalez
Melanoma
Lipofection
DMRIE-DOPE Allovectin-7 (Vical VCL-1005) HLA-B7/beta-2-microglobulin
2005
In vivo
include inhibiting angiogenic inducers such as vascular endothelial growth factor (VEGF)91 and using angiogenesis inhibitors such as angiostatin and endostatin. Recent work by Lee et al. introduced Salmonella choleraesuis carrying a eukaryotic expression plasmid expressing endostatin into mice bearing melanomas or bladder tumours and found inhibition of tumour growth by 40–70 per cent.92 IMMUNOTHERAPY
Many cancer cells display on their surfaces so-called tumour-associated antigens (TAAs), that can be recognized by the humoral and cellular limbs of the immune system. Despite this fact, the immune system is rarely seen to mount a clinically meaningful anti-tumour response. Indeed, there is extensive evidence that the tumour is able
to evade immune surveillance both by reducing its own immunogenicity and by inhibiting the ability of the immune system to mount an effective response against it.93 A number of mechanisms that appear to contribute to this phenomenon have been identified. Total or partial loss of expression of the major histocompatibility complex (MHC) class I and co-stimulatory B7.1/B7.2 molecules has been documented in many cancer types and serves to reduce the efficiency of presentation of TAA to cytotoxic T lymphocytes (CTLs). Mutations in the pathways regulating the transport and presentation of peptides on the surface of tumour cells may also shield them from detection by CTLs. In addition, those CTLs that infiltrate cancers may be killed by means of release of soluble Fas ligand (FasL) from the tumour cells, which themselves appear to be partially resistant to this important pathway of cell killing.
280 Gene therapy
The identification of these aberrations in the immune system has fuelled interest in approaches aimed at delivering genes to enhance the immunogenicity of tumours and the immune responses mounted against them. Enlisting the services of the immune system may have a number of benefits. ●
●
●
●
The inherent specificity of the immune response should limit the occurrence of normal tissue toxicity. An immune response generated at one site should prime the immune system to react to disease deposits at other local and distant sites, yielding a potent bystander effect. The system involves significant signal amplification such that a small immunogenic stimulus has the capacity to trigger a large response. Once established, anti-tumour immunity should persist through the generation of memory cells, which should act to prevent disease recurrence.
In the arena of the clinical development of immunomodulatory cancer gene therapy, studies have focused on two particular approaches. The first involves attempts to deliver genes encoding MHC molecules or cytokines directly into tumour cells in vivo. The delivery of cytokine genes – granulocyte–macrophage colony-stimulating factor (GM-CSF), interferon-α (IFN-α) and interleukin-2 (IL-2) – has been accomplished using replication-deficient vaccinia, retroviral and adenoviral vectors in patients with melanoma, breast, colonic and pancreatic cancers.94 Once again, these approaches have been shown to have good safety profiles and to be capable of transducing tumour cells in vivo. However, as with the studies involving cationic liposomal vehicles, there have been few objective responses, although a number of patients have experienced disease stabilization. The second approach to clinical immunomodulatory gene therapy has involved using genetically modified normal or malignant cells as cancer vaccines. In most studies, the patient’s own (autologous) tumour cells have been harvested, transduced with a therapeutic cytokine gene (GMCSF, IFN-α and IL-2 and IL-12) and re-injected (after irradiation to stop them forming ‘metastatic’ deposits). This strategy has been seen as most likely to yield a meaningful clinical response because the tumour cells are likely to display the relevant antigens necessary for inducing anti-tumour immunity. Studies of this type have been reported in patients with prostate cancer, melanoma and glioma using ex-vivo transduction of tumour cells with plasmid cDNA or retroviral vectors.95,96 In general, treatment has been very well tolerated, with side effects limited to local inflammatory changes and fever. Surrogate measures of anti-tumour immunity have included the generation of delayed-type hypersensitivity reactions, CTL and intra-tumoral immune infiltrates.96 Responses in these studies have largely been limited to disease stabilization, although occasional objective responses have been reported.
SUMMARY Cancer gene therapy is the most promising and active field in gene therapy treatment. Although previous experimental and clinical trials have identified some exciting cases, in general the clinical benefits have been limited. At present, most gene therapy trials are in the phase I stage and preliminary results indicate mild benefits in tumour response.97–99 Only a few have progressed to stage II/III (see Table 13.7). Despite much progress since the approval of the first clinical trial 13 years ago, more work is needed before it is considered a viable treatment for this multifaceted disease.
KEY LEARNING POINTS ●
●
●
●
●
Gene therapy involves the transfer of genetic material to cells in order to produce a therapeutic effect. Out of 1145 gene therapy clinical trials currently (2006) being undertaken, 762 are in cancer gene therapy (see Table 13.1). Viral vectors include adenoviral, retroviral, adeno-associated, poxviruses, herpes simplex viruses and oncolytic viruses. Non-viral vectors include naked DNA, lipoplexes and bacteria. Regulation of expression is vital and can be achieved at the point of transcription or transduction, or both. Strategies for cancer gene therapy: corrective – knock out oncogenes or replace tumour suppressor genes; cytoreductive – gene-directed enzyme prodrug therapy, anti-angiogenesis, induction apoptosis; immunotherapy – cytokine genes, cancer vaccines.
REFERENCES 1 Culver KW, Osborne WR, Miller AD, et al. Correction of ADA deficiency in human T lymphocytes using retroviralmediated gene transfer. Transplant Proc 1991; 23(1 Pt 1):170–1. 2 Wolf JK, Jenkins AD. Gene therapy for ovarian cancer (review). Int J Oncol 2002; 21(3):461–8. ◆3 Lundstrom K. Latest development in viral vectors for gene therapy. Trends Biotechnol 2003; 21(3):117–22. ◆4 Thomas CE, Ehrhardt A, Kay MA. Progress and problems with the use of viral vectors for gene therapy. Nat Rev Genet 2003; 4(5):346–58. ◆5 Kappes JC, Wu X. Safety considerations in vector development. Somat Cell Mol Genet 2001; 26(1–6):147–58.
References 281
◆6
7 8
9
10
◆11
12 13
◆14
15
16
◆17
◆18
19 20
21
22
23
24
Relph KL, Harrington KJ, Pandha H. Adenoviral strategies for the gene therapy of cancer. Semin Oncol 2005; 32(6):573–82. Shenk T. Adenoviridae: the Viruses and their Replication. Philadelphia: Lippincott-Raven, 1996. Kay MA, Holterman AX, Meuse L, et al. Long-term hepatic adenovirus-mediated gene expression in mice following CTLA4Ig administration. Nat Genet 1995; 11(2):191–7. Lochmuller H, Jani A, Huard J, et al. Emergence of early region 1-containing replication-competent adenovirus in stocks of replication-defective adenovirus recombinants (delta E1 ; delta E3) during multiple passages in 293 cells. Hum Gene Ther 1994; 5(12):1485–91. Parks RJ, Chen L, Anton M, Sankar U, Rudnicki MA, Graham FL. A helper-dependent adenovirus vector system: removal of helper virus by Cre-mediated excision of the viral packaging signal. Proc Natl Acad Sci U S A 1996; 93(24):13565–70. Wu N, Ataai MM. Production of viral vectors for gene therapy applications. Curr Opin Biotechnol 2000; 11(2):205–8. Miller AD. Retrovirus packaging cells. Hum Gene Ther 1990; 1(1):5–14. Rainov NG, Ren H. Clinical trials with retrovirus mediated gene therapy – what have we learned? J Neurooncol 2003; 65(3):227–36. Lachmann R. Herpes simplex virus-based vectors. Int J Exp Pathol 2004; 85(4):177–90. Samaniego LA, Neiderhiser L, DeLuca NA. Persistence and expression of the herpes simplex virus genome in the absence of immediate-early proteins. J Virol 1998; 72(4):3307–20. Fraefel C, Jacoby DR, Breakefield XO. Herpes simplex virus type 1-based amplicon vector systems. Adv Virus Res 2000; 55:425–51. Snoeck HW, Tao W, Klotman ME. Adeno-associated viral vectors: background and technical aspects. Exp Nephrol 1997; 5(6):514–20. Yamanaka R. Alphavirus vectors for cancer gene therapy (review). Int J Oncol 2004; 24(4):919–23. Lundstrom K. Alphavirus vectors for vaccine production and gene therapy. Expert Rev Vaccines 2003; 2(3):447–59. Lundstrom K. Alphavirus vectors for vaccine production and gene therapy. Expert Rev Vaccines 2003; 2(3):447–59. Stanziale SF, Fong Y. Novel approaches to cancer therapy using oncolytic viruses. Curr Mol Med 2003; 3(1):61–71. Branton PE, Roopchand DE. The role of adenovirus E4orf4 protein in viral replication and cell killing. Oncogene 2001; 20(54):7855–65. Sparer TE, Tripp RA, Dillehay DL, Hermiston TW, Wold WS, Gooding LR. The role of human adenovirus early region 3 proteins (gp19K, 10.4K, 14.5K, and 14.7K) in a murine pneumonia model. J Virol 1996;70(4):2431–9. Whyte P, Ruley HE, Harlow E. Two regions of the adenovirus early region 1A proteins are required for transformation. J Virol 1988; 62(1):257–65.
25 Hallenbeck PL, Chang YN, Hay C, et al. A novel tumorspecific replication-restricted adenoviral vector for gene therapy of hepatocellular carcinoma. Hum Gene Ther 1999; 10(10):1721–33. 26 Barker DD, Berk AJ. Adenovirus proteins from both E1B reading frames are required for transformation of rodent cells by viral infection and DNA transfection. Virology 1987; 156(1):107–21. 27 Vollmer CM, Ribas A, Butterfield LH, et al. p53 selective and nonselective replication of an E1B-deleted adenovirus in hepatocellular carcinoma. Cancer Res 1999; 59(17):4369–74. 28 Nemunaitis J, Ganly I, Khuri F, et al. Selective replication and oncolysis in p53 mutant tumors with ONYX-015, an E1B-55kD gene-deleted adenovirus, in patients with advanced head and neck cancer: a phase II trial. Cancer Res 2000; 60(22):6359–66. 29 Mulvihill S, Warren R, Venook A, et al. Safety and feasibility of injection with an E1B-55 kDa gene-deleted, replicationselective adenovirus (ONYX-015) into primary carcinomas of the pancreas: a phase I trial. Gene Ther 2001; 8(4):308–15. 30 Khuri FR, Nemunaitis J, Ganly I, et al. A controlled trial of intratumoral ONYX-015, a selectively replicating adenovirus, in combination with cisplatin and 5-fluorouracil in patients with recurrent head and neck cancer. Nat Med 2000; 6(8):879–85. 31 Douglas JT, Kim M, Sumerel LA, Carey DE, Curiel DT. Efficient oncolysis by a replicating adenovirus (ad) in vivo is critically dependent on tumor expression of primary ad receptors. Cancer Res 2001; 61(3):813–17. 32 Coen DM, Goldstein DJ, Weller SK. Herpes simplex virus ribonucleotide reductase mutants are hypersensitive to acyclovir. Antimicrob Agents Chemother 1989; 33(8):1395–9. 33 Chou J, Roizman B. The gamma 1(34.5) gene of herpes simplex virus 1 precludes neuroblastoma cells from triggering total shutoff of protein synthesis characteristic of programmed cell death in neuronal cells. Proc Natl Acad Sci U S A 1992; 89(8):3266–70. 34 Kramm CM, Chase M, Herrlinger U, et al. Therapeutic efficiency and safety of a second-generation replicationconditional HSV1 vector for brain tumor gene therapy. Hum Gene Ther 1997; 8(17):2057–68. 35 Yazaki T, Manz HJ, Rabkin SD, Martuza RL. Treatment of human malignant meningiomas by G207, a replicationcompetent multimutated herpes simplex virus 1. Cancer Res 1995; 55(21):4752–6. ◆36 Jia W, Zhou Q. Viral vectors for cancer gene therapy: viral dissemination and tumor targeting. Curr Gene Ther 2005; 5(1):133–42. 37 Yang JP, Huang L. Direct gene transfer to mouse melanoma by intratumor injection of free DNA. Gene Ther 1996; 3(6):542–8. 38 Nomura T, Nakajima S, Kawabata K, Yamashita F, Takakura Y, Hashida M. Intratumoral pharmacokinetics and in vivo gene expression of naked plasmid DNA and its cationic liposome
282 Gene therapy
◆39
40
41
42
43
44
45
46 47
48
49
50
51
52 53
54
complexes after direct gene transfer. Cancer Res 1997; 57(13):2681–6. Ogris M. Non-viral cancer gene therapy – what is best? Drug Discov Today 2003; 8(2):63. Gehl J. Electroporation: theory and methods, perspectives for drug delivery, gene therapy and research. Acta Physiol Scand 2003; 177(4):437–47. Dash PR, Read ML, Barrett LB, Wolfert MA, Seymour LW. Factors affecting blood clearance and in vivo distribution of polyelectrolyte complexes for gene delivery. Gene Ther 1999; 6(4):643–50. Dass CR. Cytotoxicity issues pertinent to lipoplex-mediated gene therapy in-vivo. J Pharm Pharmacol 2002; 54(5):593–601. Liu SC, Minton NP, Giaccia AJ, Brown JM. Anticancer efficacy of systemically delivered anaerobic bacteria as gene therapy vectors targeting tumor hypoxia/necrosis. Gene Ther 2002; 9(4):291–6. Fujimori M, Amano J, Taniguchi S. The genus Bifidobacterium for cancer gene therapy. Curr Opin Drug Discov Devel 2002; 5(2):200–3. Li YH, Xie YM, Guo KY, et al. Treatment of tumor in mice by oral administration of cytosine deaminase gene carried in live attenuated Salmonella. Sheng Wu Hua Xue Yu Sheng Wu Wu Li Xue Bao (Shanghai) 2001; 33(2):233–7. Verma IM, Somia N. Gene therapy – promises, problems and prospects. Nature 1997; 389(6648):239–42. Pang S, Taneja S, Dardashti K, et al. Prostate tissue specificity of the prostate-specific antigen promoter isolated from a patient with prostate cancer. Hum Gene Ther 1995; 6(11):1417–26. Lillehammer T, Tveito S, Engesaeter BO, Fodstad O, Maelandsmo GM, Engebraaten O. Melanoma-specific expression in first-generation adenoviral vectors in vitro and in vivo – use of the human tyrosinase promoter with human enhancers. Cancer Gene Ther 2005; 12(11):864–72. Ido A, Nakata K, Kato Y, et al. Gene therapy for hepatoma cells using a retrovirus vector carrying herpes simplex virus thymidine kinase gene under the control of human alphafetoprotein gene promoter. Cancer Res 1995; 55(14):3105–9. Gupta VK, Park JO, Kurihara T, et al. Selective gene expression using a DF3/MUC1 promoter in a human esophageal adenocarcinoma model. Gene Ther 2003; 10(3):206–12. Greco O, Joiner MC, Doleh A, Powell AD, Hillman GG, Scott SD. Hypoxia- and radiation-activated Cre/loxP ‘molecular switch’ vectors for gene therapy of cancer. Gene Ther 2006; 13(3):206–15. Mezhir JJ, Schmidt H, Yamini B, et al. Chemo-inducible gene therapy. Anticancer Drugs 2005; 16(10):1053–8. Ho IA, Hui KM, Lam PY. Targeting proliferating tumor cells via the transcriptional control of therapeutic genes. Cancer Gene Ther 2006; 13(1):44–52. Bergelson JM, Cunningham JA, Droguett G, et al. Isolation of a common receptor for Coxsackie B viruses and adenoviruses 2 and 5. Science 1997; 275(5304):1320–3.
55 Kasono K, Blackwell JL, Douglas JT, et al. Selective gene delivery to head and neck cancer cells via an integrin targeted adenoviral vector. Clin Cancer Res 1999; 5(9):2571–9. 56 Huang S, Kamata T, Takada Y, Ruggeri ZM, Nemerow GR. Adenovirus interaction with distinct integrins mediates separate events in cell entry and gene delivery to hematopoietic cells. J Virol 1996; 70(7):4502–8. ◆57 Barnett BG, Crews CJ, Douglas JT. Targeted adenoviral vectors. Biochim Biophys Acta 2002; 1575(1–3):1–14. 58 Heideman DA, van Beusechem VW, Offerhaus GJ, et al. Selective gene transfer into primary human gastric tumors using epithelial cell adhesion molecule-targeted adenoviral vectors with ablated native tropism. Hum Gene Ther 2002; 13(14):1677–85. 59 Teich NM,. Oncogenes and Cancer, 2nd edn. Oxford: Oxford Medical Publications; 1991. 60 Kim HG, Miller DM. A novel triplex-forming oligonucleotide targeted to human cyclin D1 (bcl-1, proto-oncogene) promoter inhibits transcription in HeLa cells. Biochemistry 1998; 37(8):2666–72. 61 Tolcher AW. Preliminary phase I results of G3139 (bcl-2 antisense oligonucleotide) therapy in combination with docetaxel in hormone-refractory prostate cancer. Semin Oncol 2001; 28(4 Suppl. 15):67–70. 62 Yamato K, Fen J, Kobuchi H, et al. Induction of cell death in human papillomavirus 18-positive cervical cancer cells by E6 siRNA. Cancer Gene Ther 2006; 13(3):234–41. ◆63 Devi GR. siRNA-based approaches in cancer therapy. Cancer Gene Ther 2006. 13 (9):819– 29 64 Takahashi T, Nau MM, Chiba I, et al. p53: a frequent target for genetic abnormalities in lung cancer. Science 1989; 246(4929):491–4. 65 Remvikos Y, Tominaga O, Hammel P, et al. Increased p53 protein content of colorectal tumours correlates with poor survival. Br J Cancer 1992; 66(4):758–64. 66 Zhang WW, Fang X, Mazur W, French BA, Georges RN, Roth JA. High-efficiency gene transfer and high-level expression of wild-type p53 in human lung cancer cells mediated by recombinant adenovirus. Cancer Gene Ther 1994; 1(1):5–13. 67 Fujiwara T, Cai DW, Georges RN, Mukhopadhyay T, Grimm EA, Roth JA. Therapeutic effect of a retroviral wild-type p53 expression vector in an orthotopic lung cancer model. J Natl Cancer Inst 1994; 86(19):1458–62. 68 Zou Y, Zong G, Ling YH, et al. Effective treatment of early endobronchial cancer with regional administration of liposome–p53 complexes. J Natl Cancer Inst 1998; 90(15):1130–7. 69 Spitz FR, Nguyen D, Skibber JM, Meyn RE, Cristiano RJ, Roth JA. Adenoviral-mediated wild-type p53 gene expression sensitizes colorectal cancer cells to ionizing radiation. Clin Cancer Res 1996; 2(10):1665–71. 70 Ogawa N, Fujiwara T, Kagawa S, et al. Novel combination therapy for human colon cancer with adenovirus-mediated wild-type p53 gene transfer and DNA-damaging chemotherapeutic agent. Int J Cancer 1997; 73(3):367–70.
References 283
71 Roth JA, Nguyen D, Lawrence DD, et al. Retrovirus-mediated wild-type p53 gene transfer to tumors of patients with lung cancer. Nat Med 1996; 2(9):985–91. 72 Tursz T, Cesne AL, Baldeyrou P, et al. Phase I study of a recombinant adenovirus-mediated gene transfer in lung cancer patients. J Natl Cancer Inst 1996; 88(24):1857–63. 73 Schuler M, Rochlitz C, Horowitz JA, et al. A phase I study of adenovirus-mediated wild-type p53 gene transfer in patients with advanced non-small cell lung cancer. Hum Gene Ther 1998; 9(14):2075–82. 74 Swisher SG, Roth JA, Nemunaitis J, et al. Adenovirusmediated p53 gene transfer in advanced non-small-cell lung cancer. J Natl Cancer Inst 1999; 91(9):763–71. ◆75 Springer CJ, Niculescu-Duvaz I. Prodrug-activating systems in suicide gene therapy. J Clin Invest 2000; 105(9):1161–7. 76 Fillat C, Carrio M, Cascante A, Sangro B. Suicide gene therapy mediated by the Herpes Simplex virus thymidine kinase gene/ganciclovir system: fifteen years of application. Curr Gene Ther 2003; 3(1):13–26. ◆77 Greco O, Dachs GU. Gene directed enzyme/prodrug therapy of cancer: historical appraisal and future prospectives. J Cell Physiol 2001; 187(1):22–36. 78 Freeman SM, McCune C, Robinson W, et al. The treatment of ovarian cancer with a gene modified cancer vaccine: a phase I study. Hum Gene Ther 1995; 6(7):927–39. 79 Shalev M, Kadmon D, Teh BS, et al. Suicide gene therapy toxicity after multiple and repeat injections in patients with localized prostate cancer. J Urol 2000; 163(6):1747–50. 80 Sandmair AM, Loimas S, Puranen P, et al. Thymidine kinase gene therapy for human malignant glioma, using replication-deficient retroviruses or adenoviruses. Hum Gene Ther 2000; 11(16):2197–205. 81 Klatzmann D, Valery CA, Bensimon G, et al. A phase I/II study of herpes simplex virus type 1 thymidine kinase ‘suicide’ gene therapy for recurrent glioblastoma. Study Group on Gene Therapy for Glioblastoma. Hum Gene Ther 1998; 9(17):2595–604. 82 Pandha HS, Martin LA, Rigg A, et al. Genetic prodrug activation therapy for breast cancer: a phase I clinical trial of erbB-2-directed suicide gene expression. J Clin Oncol 1999; 17(7):2180–9. 83 Sato T, Yamauchi N, Sasaki H, et al. An apoptosis-inducing gene therapy for pancreatic cancer with a combination of 55-kDa tumor necrosis factor (TNF) receptor gene transfection and mutein TNF administration. Cancer Res 1998; 58(8):1677–83. 84 Griffith TS, Anderson RD, Davidson BL, Williams RD, Ratliff TL. Adenoviral-mediated transfer of the TNF-related apoptosis-inducing ligand/Apo-2 ligand gene induces tumor cell apoptosis. J Immunol 2000; 165(5):2886–94. 85 Aoki K, Akyurek LM, San H, et al. Restricted expression of an adenoviral vector encoding Fas ligand (CD95L) enhances safety for cancer gene therapy. Mol Ther 2000; 1(6):555–65. 86 Hedlund TE, Meech SJ, Srikanth S, et al. Adenovirusmediated expression of Fas ligand induces apoptosis of
87
88
89
90
91
92
93
◆94
95
96
97
98
99
human prostate cancer cells. Cell Death Differ 1999; 6(2):175–82. Hofmann A, Blau HM. Death of solid tumor cells induced by Fas ligand expressing primary myoblasts. Somat Cell Mol Genet 1997; 23(4):249–57. Pataer A, Fang B, Yu R, et al. Adenoviral Bak overexpression mediates caspase-dependent tumor killing. Cancer Res 2000; 60(4):788–92. Shinoura N, Saito K, Yoshida Y, et al. Adenovirus-mediated transfer of bax with caspase-8 controlled by myelin basic protein promoter exerts an enhanced cytotoxic effect in gliomas. Cancer Gene Ther 2000; 7(5):739–48. Shariat SF, Desai S, Song W, et al. Adenovirus-mediated transfer of inducible caspases: a novel ‘death switch’ gene therapeutic approach to prostate cancer. Cancer Res 2001; 61(6):2562–71. Cichon T, Jamrozy L, Glogowska J, Missol-Kolka E, Szala S. Electrotransfer of gene encoding endostatin into normal and neoplastic mouse tissues: inhibition of primary tumor growth and metastatic spread. Cancer Gene Ther 2002; 9(9):771–7. Lee CH, Wu CL, Shiau AL. Endostatin gene therapy delivered by Salmonella choleraesuis in murine tumor models. J Gene Med 2004; 6(12):1382–93. Melcher A, Gough M, Todryk S, Vile R. Apoptosis or necrosis for tumor immunotherapy: what’s in a name? J Mol Med 1999; 77(12):824–33. Wysocki PJ, Karczewska-Dzionk A, Mackiewicz-Wysocka M, Mackiewicz A. Human cancer gene therapy with cytokine gene-modified cells. Expert Opin Biol Ther 2004; 4(10):1595–607. Simons JW, Mikhak B, Chang JF, et al. Induction of immunity to prostate cancer antigens: results of a clinical trial of vaccination with irradiated autologous prostate tumor cells engineered to secrete granulocyte–macrophage colony-stimulating factor using ex vivo gene transfer. Cancer Res 1999; 59(20):5160–8. Sun Y, Jurgovsky K, Moller P, et al. Vaccination with IL-12 gene-modified autologous melanoma cells: preclinical results and a first clinical phase I study. Gene Ther 1998; 5(4):481–90. Kubo H, Gardner TA, Wada Y, et al. Phase I dose escalation clinical trial of adenovirus vector carrying osteocalcin promoter-driven herpes simplex virus thymidine kinase in localized and metastatic hormonerefractory prostate cancer. Hum Gene Ther 2003; 14(3):227–41. Trudel S, Trachtenberg J, Toi A, et al. A phase I trial of adenovector-mediated delivery of interleukin-2 (AdIL-2) in high-risk localized prostate cancer. Cancer Gene Ther 2003; 10(10):755–63. Vasey PA, Shulman LN, Campos S, et al. Phase I trial of intraperitoneal injection of the E1B-55-kd-gene-deleted adenovirus ONYX-015 (dl1520) given on days 1 through 5 every 3 weeks in patients with recurrent/ refractory epithelial ovarian cancer. J Clin Oncol 2002; 20(6):1562–9.
284 Gene therapy
◆100
McCormick F. Cancer-specific viruses and the development of ONYX-015. Cancer Biol Ther 2003; 2(4 Suppl. 1):S157–60. 101 Davis AJ, Gelmon KA, Siu LL, et al. Phase I and pharmacologic study of the human DNA methyltransferase antisense oligodeoxynucleotide MG98 given as a 21-day continuous infusion every 4 weeks. Invest New Drugs 2003; 21(1):85–97. 102 Webb A, Cunningham D, Cotter F, et al. BCL-2 antisense therapy in patients with non-Hodgkin lymphoma. Lancet 1997; 349(9059):1137–41. 103 Zellweger T, Miyake H, Cooper S, et al. Antitumor activity of antisense clusterin oligonucleotides is improved in vitro and in vivo by incorporation of 2’-O-(2methoxy)ethyl chemistry. J Pharmacol Exp Ther 2001; 298(3):934–40.
◆104
Holmlund JT, Monia BP, Kwoh TJ, Dorr FA. Toward antisense oligonucleotide therapy for cancer: ISIS compounds in clinical development. Curr Opin Mol Ther 1999; 1(3):372–85. 105 Stevenson JP, Yao KS, Gallagher M, et al. Phase I clinical/pharmacokinetic and pharmacodynamic trial of the c-raf-1 antisense oligonucleotide ISIS 5132 (CGP 69846A). J Clin Oncol 1999; 17(7):2227–36. 106 Bishop MR, Iversen PL, Bayever E, et al. Phase I trial of an antisense oligonucleotide OL(1)p53 in hematologic malignancies. J Clin Oncol 1996; 14(4):1320–6. 107 Chen HX, Marshall JL, Ness E, et al. A safety and pharmacokinetic study of a mixed-backbone oligonucleotide (GEM231) targeting the type I protein kinase A by two-hour infusions in patients with refractory solid tumors. Clin Cancer Res 2000; 6(4):1259–66.
PART
2
PRACTICE
14 Central nervous system Roy Rampling 15 Ocular and adnexal tumours John L Hungerford, P Nicholas Plowman 16 Head and neck cancer Alastair J Munro, Nicholas D Stafford 17 Thyroid Masud Haq, Clive Harmer 18 Endocrine cancer Maria Gueorguiev, Ashley B Grossman, P Nicholas Plowman 19 Breast cancer Navita Somaiah, John Yarnold 20 Bronchus Vanessa A Potter, Nicholas Thatcher, Penella J Woll 21 Oesophageal cancer David B Smith, Brian J Haylock 22 Liver Daniel H Palmer, Philip J Johnson, 23 Pancreas Patrick G Johnston, Martin M Eatock 24 Biliary tract Hemant M Kocher, Ajit T Abraham, Satya Bhattacharya 25 Stomach Marianne C Nicolson, Kenneth G M Park, Leslie M Samuel 26 Bladder cancer Robert Huddart, Paula Wells, Alan Horwich 27 Prostate cancer Malcolm Mason 28 Colorectal cancer Richard H J Begent, Astrid Mayer, Christopher H Collis, Russell N Moule 29 Anus Bernard J Cummings 30 Germ-cell cancers of the testis and related neoplasms Graham M Mead 31 Renal cell cancer Tom R Geldart
287 320 343 408 438 468 502 539 559 573 585 602 614 645 666 695 706 729
286 Part 2: Practice
32 Ovary and fallopian tubes Daniela D Rosa, Andrew R Clamp, Gordon C Jayson 33 Uterus Bleddyn Jones 34 Cervix Peter Blake 35 Carcinoma of the vagina and vulva Sadaf Ghaem-Maghami, W Pat Soutter 36 Gestational trophoblastic tumours Peter Schmid, Michael J Seckl 37 Non-melanoma skin cancer Stephen L Morris, Sean Whittaker, Margaret Spittle 38 Malignant melanoma Susana Banerjee, Martin Gore 39 Bone Jeremy S Whelan, Rob C Pollock, Anna M Cassoni 40 Soft tissue sarcomas Thomas F DeLaney, Andrew E Rosenberg, David C Harmon, Frances Hornicek, Sam Yoon, David G Kirsch, Henry J Mankin, Daniel Rosenthal 41 Leukaemias Tariq I Mughal, John M Goldman 42 Hodgkin’s lymphoma Beate Klimm, Andreas Engert 43 Non-Hodgkin’s lymphoma Adrian J C Bloor, David C Linch 44 Multiple myeloma Jennifer M Bird, Cathy Williams 45 Introduction to haematopoietic stem cell transplantation Effie Liakopoulou, David I Marks 46 Paediatric oncology Stephen Lowis, Eddy Estlin, Keith Sibson 47 AIDS-related cancer Mark Bower, Tom Newsom-Davis
748 764 777 796 809 835 879 905 924
981 1027 1045 1073 1098 1118 1167
14 Central nervous system ROY RAMPLING
Pathology Presentation Imaging
287 292 293
The central nervous system (CNS) is host to a remarkable variety of primary tumours that demonstrate an equal diversity of clinical behaviour, response to treatment and prognosis. Whilst most malignant tumours still carry a bleak prognosis, worthwhile extension of life can be achieved in many patients. For those with more responsive tumours, adequate management can provide prolonged survival or cure. An accurate diagnosis is required in almost all cases and this is now facilitated by advances in neuro-imaging, neurosurgical technique and neuropathology. Molecular analysis is already having an impact on the management of some brain tumours. The practising neuro-oncologist therefore needs a detailed knowledge of a modern pathological classification system and how this relates to tumour management.
PATHOLOGY Incidence The overall incidence of primary CNS tumours in the UK is around 16 per 100 000 person years at risk, of which around half are malignant.1* Incidence estimates are complicated by methodological problems, but geographical and racial variations do seem to occur.2 High rates are reported in Israeli Jews, whether born in Israel or elsewhere. White people in Los Angeles have a higher incidence than those in England and Wales. The incidence in Japan is low, but is doubled in Japanese living in Los Angeles. The distribution of histological types is fairly uniform worldwide (Table
Treatment modalities Individual tumours References
295 302 317
14.1). Exceptions to this include high rates of pineal germcell tumours in Asia, medulloblastomas in New Zealand Maoris, and primary CNS lymphomas in regions with a high acquired immunodeficiency syndrome (AIDS) incidence. There is no regional variation of brain tumours in the UK. Brain tumours account for approximately 1.6 per cent of all primary tumours,3 but nearly 7 per cent of the number of years of life lost from cancer before age 70. The rate of brain tumour registration in the UK increased by 17 per cent in the decade 1991–2000.1 Whilst some of this increase is due to improved diagnosis, there is evidence of
Table 14.1 Approximate incidence rates (worldwide) for brain tumour types (per 100 000/year) Tumour type Astrocytoma Anaplastic astrocytoma Glioblastoma Meningioma Primary central nervous system lymphoma: immune competent Overall Medulloblastoma Germ-cell tumours Pinealoma/pineoblastoma Metastases
Incidence 1.5 1.0 3 3 0.3 0.8–6.8 0.5 0.2 0.1 8
288 Central nervous system
an underlying increase in incidence, particularly among the elderly.4
astrocytomas and oligodendrogliomas. Some, like ependymomas and nerve sheath tumours, are equally distributed, and meningiomas are more common in females.
Age Brain tumours occur at any age, from neonates to old age. The age- specific incidence shows a small peak in early childhood, a poorly defined minimum in teenage years, rising at an increasing rate to a second major peak at around 75.1 After age 80, most series report a rapid decline into old age. This decline is possibly an artefact of data collection, and a Swedish series, based on high routine autopsy rates, shows the high incidence continuing into old age.5 It is important to recognize that the brain is the commonest site for solid tumours in childhood. The tumour spectrum also varies with age. The majority of brain tumours in children (70–80 per cent) arise infratentorially (glial tumours, medulloblastoma) or in the midline (germ-cell tumours, craniopharyngioma). Low-grade glial tumours are common, particularly pilocytic astrocytoma. In adults, most brain tumours are supratentorial. Gliomas, particularly glioblastomas, and meningiomas predominate.
Sex Most brain tumours occur more commonly in males than in females, particularly medulloblastomas, germ-cell tumours,
Aetiology Ionizing radiation is the only environmental factor that is clearly associated with an increased risk of developing a brain tumour. This has been shown in adults and children receiving radiation for therapeutic purposes and following occupational exposure. Radiation-induced tumours include astrocytomas of all grades, benign and malignant meningiomas, sarcomas and nerve sheath tumours. Whilst brain tumours may be induced in animals by a variety of viruses, none is known to do so in humans, although the Epstein–Barr virus (EBV) genome is present in most immunosuppressed patients with lymphoma. Likewise, chemical agents can reliably induce tumours in some animal species (e.g. nitrosourea in rats), but neither industrial nor agricultural exposure in humans has been definitely associated with an increased risk. Other putative agents have included bacteria, head injury, exposure to non-ionising radiation (e.g. power lines and mobile phones), diet and tobacco. In spite of extensive epidemiological studies, none of these has a proven relationship with brain tumours.6**
Table 14.2 Genetic syndromes associated with an increased risk of brain tumour Syndrome
Brain tumour
Other associations
Genetics
Pigmentation Peripheral neurofibromas Osseous and vascular lesions Cerebral calcification Lens opacities
NF1 on 17q11 Autosomal dominant
Von Hippel-Lindau
Neurofibromas Gliomas Sarcomas Schwannomas (acoustic neuromas) Meningiomas Gliomas (especially spinal) Haemangioblastoma
VHL on 3p25–26 Autosomal dominant
Cowden’s
Dysplastic gangliocytoma of cerebellum
Turcot’s
Glioblastomas Medulloblastomas Subependymal giant-cell astrocytoma Hamartomas Gliomas PNETs Medulloblastomas
Retinal haemangioblastoma Renal carcinoma Phaeochromocytoma Visceral cysts Peripheral hamartomas Breast cancer Thyroid neoplasia Colorectal tumours
Neurofibromatosis I
Neurofibromatosis II
Tuberose sclerosis Li Fraumeni Basal naevus
PNETs, primitive neuro-ectodermal tumours; PTCH, ‘patched’ gene
Angiofibromas Hypomelanotic patches Sarcomas Breast cancer Basal-cell carcinomas Bone abnormalities Palmer pits
NF2 on 22q12 Autosomal dominant
PTEN/MMAC1 on 10q23 Autosomal dominant 5q21 Autosomal dominant TSC1 on 9q TSC2 on16p TP53 on 17p13 Autosomal dominant PTCH on 9q22 Autosomal dominant
Pathology 289
There are a number of genetic syndromes that are associated with an increased risk of brain tumour.7 These are described in Table 14.2. A few other host-related factors are associated with an increased risk of acquiring a brain tumour. Immunosuppression, both environmental (e.g. AIDS) and iatrogenic, as in transplant recipients, predisposes to CNS lymphoma. The EBV genome is present in 95 per cent of these tumours and probably plays a major role in their development. Breast cancer may be associated with a higher incidence of meningioma.8
Tumour types The value of a pathological classification system to a clinician is that it can be used to direct management and predict behaviour. The World Health Organisation system (WHO 2000) is now almost universally accepted for this purpose.7 It has well over 100 entries. An abridged version is shown in Box 14.1. The scheme is based on a separation of tumour types according to their (presumed) tissue of origin (neuro-epithelium, meninges, etc.) and then according to the derivative cell type (astrocyte, oligodendrocyte etc.). Further classification recognizes features of the tumour cells and the associated microenvironment that grade the tumour according to its degree of malignancy.
Box 14.1 World Health Organisation brain tumour classification system (WHO 2000) Tumours of neuroepithelial tissue Astrocytic tumours 1. Astrocytoma (variants: fibrillary, protoplasmic, gemistocytic, mixed) 2. Anaplastic (malignant) astrocytoma 3. Glioblastoma (variants: giant cell, gliosarcoma) 4. Pilocytic astrocytoma 5. Pleomorphic xanthoastrocytoma 6. Subependymal giant-cell astrocytoma Oligodendroglial tumours 1. Oligodendroglioma 2. Anaplastic oligodendroglioma Ependymal tumours 1. Ependymoma 2. Anaplastic ependymoma 3. Myxopapillary ependymoma 4. Subependymoma Mixed gliomas Choroid plexus tumours 1. Choroid plexus papilloma 2. Choroid plexus carcinoma Neuroepithelial tumours of uncertain origin 1. Astroblastoma 2. Gliomatosis cerebri
Neuronal and mixed neuronal–glial tumours 1. Gangliocytoma 2. Ganglioglioma 3. Anaplastic ganglioglioma 4. Dysembryoplastic neuroepithelial tumour (of childhood) Pineal tumours 1. Pineocytoma 2. Pineoblastoma Embryonal tumours 1. Medulloepithelioma 2. Atypical teratoid/rhabdoid tumour 3. Primitive neuroectodermal tumours (PNET) 4. Medulloblastoma 5. Neuroblastoma Tumours of peripheral nerves Schwannoma (neurilemmoma, neurinoma) Neurofibroma Malignant peripheral nerve sheath tumour Tumours of the meninges 1. Meningiomas Meningioma (and variants) Atypical meningioma Anaplastic (malignant) meningioma 2. Mesenchymal, non-meningothelial tumours (haemangiopericytoma, chondrosarcoma) Tumours of uncertain histiogenesis 1. Haemangioblastoma Haematopoeitic neoplasms 1. Primary malignant lymphomas 2. Histiocytic tumours Germ-cell tumours 1. Germinoma 2. Teratoma (immature, mature) 3. Choriocarcinoma 4. Mixed Tumours of the sellar region 1. Craniopharyngioma Metastatic tumours Tumours no longer included Cysts and tumour-like lesions 1. Rathkes cleft cyst 2. Dermoid cyst (and others) Tumours of the anterior pituitary 1. Pituitary adenoma 2. Pituitary carcinoma Others 1. Paraganglioma 2. Chordoma 3. Chondrosarcoma Abridged and adapted (Kleihues and Cavanee 2000).
290 Central nervous system
These features (for example cellular pleomorphism and necrosis in glioblastoma) relate to the clinical behaviour of the tumour. Although WHO 2000 has replaced older classification systems, e.g.,9 great care must be taken when comparing modern studies with older clinical series that used different classification schemes. Whereas conventional light microscopy still provides the backbone of pathological analysis, it is often insufficient to produce a complete diagnosis. Immunohistochemistry is included in the WHO 2000 classification system and is frequently an indispensable discriminant in making a diagnosis. In particular, glial fibrillary acidic protein (GFAP) is valuable in identifying normal astrocytes and tumour cells of astrocytic origin. However, non-astrocytic and even some non-glial tumours may be positive. S-100 may also be useful, but is not specific to the nervous system. Lymphomas, germ-cell tumours, sarcomas and metastases immunostain, as do their systemic counterparts (see under relevant sections elsewhere in this book). Electron microscopy can also provide useful information. Molecular biology is playing an increasingly important role in identifying and classifying tumours as well as providing prognostic information and predicting response to treatment. Chromosomal losses (LOH 1p, 19q in oligodendroglioma), gene silencing (methyl guanine methyl transferase [MGMT-gene] in glioblastoma) and gene profiling (as in primitive neuro-ectodermal tumours) are all examples where assessment using molecular techniques can enhance management in patients with brain tumours. These are discussed in more detail in their separate sections. At the moment, molecular changes are not included in the WHO 2000 classification. They are, however, expected in the next WHO classification, due in the very near future.
The molecular biology of brain tumours The molecular pathogenesis of brain tumours is becoming increasingly clear.10 Early changes comprise loss of tumour suppressor function, whilst gene amplification is almost exclusively seen in high-grade tumours and represents late change. GENETIC ALTERATIONS IN BRAIN TUMOURS
Loss of heterozygosity (LOH) is common in brain tumours. LOH 17p is usually associated with p53 mutations and is associated with loss of tumour suppressor function. This is not the case in medulloblastoma, where LOH 17p is not associated with p53 mutation. Homozygous deletion of 9p21 is associated with inactivation of p16/CDKN2A, whilst loss of 13q inhibits RB1. Each inactivates the RB1 pathway, as does CDK4 amplification. The two abnormalities are almost never seen together. LOH 10q is particularly common in all types of high-grade glioma, but the specific LOH 10q23 (the PTEN gene) is largely confined to the subgroup of primary glioblastoma. Oncogene amplification is a late
Growth Factor (Tyrosine Kinase) receptors EGFR PDGFR
Ras
Pi3K
ERK
AKT
PTEN P16/CDKN2 A
P27 Cyclin D - CDK4
RB
Cycline/CDK2
P21 Promotes DNA Synthesis
E2F P53
P14/ARF
MDM2
Figure 14.1 The major pathways involved in the development of gliomas. Genes that are frequently altered (mutated, lost or amplified) are shown in lighter shading (oncogenes) or darker shading (tumour suppressor genes). [See text for details.]
event. Epidermal growth factor receptor (EGFR) amplification in glioblastoma is most common, especially in the primary subtype and is frequently aberrant. Platelet derived growth factor (PDGFR), CDK4 and mouse double minute 2 (MDM2) are amplified less frequently in gliomas. The above abnormalities produce disruption of three principal cellular pathways (RB1, p53 and tyrosine receptor signalling) leading to glioma development (Fig. 14.1). By facilitating repair or inducing apoptosis, the p53 system acts as a tumour suppressor gene. Excess MDM2 facilitates degradation of p53. MDM2 in turn is inhibited by p14/ARF, loss of which also leads to reduced p53. Naturally, mutated non-functioning p53 will inhibit this pathway. Phosphorylated RB1 inhibits binding to the E2F transcription factor, which indirectly promotes DNA synthesis. RB1 is regulated by the cyclin–cyclin dependent kinase (CDK) complex, which is inhibited by p16. Hence loss of p16/CDKN2A or Rb leads to loss of RB1 function and increased DNA synthesis. The phosphatidylinositol 3-kinase (PI3K) AkT pathway promotes DNA synthesis. It may be stimulated by activated EGFR and PDGFR. It is inhibited by phosphatase and tensin homologue (PTEN). Excess EGFR or the presence of the mutated form, EGFRv, which is constitutively active, will drive cell growth. Each of these pathways may be inhibited or activated in a variety of ways. The final product of this disruption, no matter how it is achieved, is the formation of similar tumours. LOW-GRADE GLIOMAS
Loss of heterozygosity for 17p occurs commonly (30 per cent). This loss is usually associated with mutations of p53,
Pathology 291
Table 14.3 Molecular abnormalities in brain tumours Tumour type Astrocytoma (low grade) Primary glioblastoma Secondary glioblastoma Oligodendroglioma Anaplastic oligodendroglioma Medulloblastoma Meningioma
High frequency
Low frequency
P53 mutations, PDGFR amplifications LOH 10q, EGFR amplification, PTEN alterations, CDKN2A deletions, p16 deletion LOH 10q, p53 mutations, PDGFR amplifications RB1 inhibition LOH 1p, LOH 19q LOH 1p, LOH 19q, CDKN2A deletions LOH 17p, PAX over-expression Deletions of Chr 22 (esp. 22q12) LOH 10q (atypical only)
P53 mutations PTEN alterations, EGFR amplifications LOH 10q, PTEN alterations LOH 10q, PTEN alterations
LOH, loss of heterozygosity; PDGFR, platelet derived growth factor; EGFR, epidermal growth factor receptor; PTEN, phosphatase and TENsin homolog
which do not occur in pilocytic astrocytomas but are very common in gemistocytic astrocytomas (80 per cent). GLIOBLASTOMA
There appear to be at least two types of glioblastoma based on their molecular abnormalities (Table 14.3). The denovo or primary glioblastomas tend to occur in older patients with no prior history of tumour and have a particularly poor prognosis. They are characterized by oncogene amplification (EGFR), CDKN2A deletions and PTEN mutations. Secondary or progressive glioblastomas arise in younger patients, frequently with preceding astrocytoma, and have a slightly better prognosis. Common abnormalities are p53 mutations and PDGFR amplification. LOH 10 is common in both types of glioblastoma. Over-expression of MDM2, without gene amplification, occurs in the majority of glioblastomas. OLIGODENDROGLIOMA
Both oligodendroglioma and anaplastic oligodendroglioma exhibit LOH 1p and 19q (80 per cent of patients). LOH 4, 6 and 11 occur much less frequently. There is a mutually exclusive pattern of LOH 1p and 19q and p53 gene mutation in oligoastrocytomas, suggesting that the former are of oligodendrocyte origin and the latter are astrocytic. LOH for 1p is associated with a good response tochemotherapy.11 The minority of oligodendrogliomas, without LOH 1p, appear to be chemo-resistant. MEDULLOBLASTOMAS
Embryonal tumours of the CNS represent a heterogeneous group of tumours with similar appearances on conventional microscopy. Pomeroy et al.12 developed a classification system based on DNA micro-array gene expression data derived from samples in 99 patients with embryonal tumours. They show that medulloblastomas are molecularly distinct from other brain tumours, including primitive
neuro-ectodermal tumours (PNETs), atypical teratoid/rhabdoid tumours (AT/RTs) and malignant gliomas. Molecular analysis suggests that medulloblastomas derive from cerebellar granule cells through activation of the Sonic Hedgehog (SHH) pathway. LOH 17p occurs in 30–40 per cent of medulloblastomas but is associated with mutated p53 in only 5–10 per cent. It is not considered the target of 17p loss. MENINGIOMAS
Deletions on chromosome 22 occur in the majority of meningiomas. Half of these incur loss at 22q12. Atypical meningiomas show further loss on 1p, 9q and 10q, suggesting that progression may be associated with genes at these loci.
Local environment The tumour extracellular matrix (ECM) comprises proteoglycans, glycoproteins and collagens. Fibronectin and laminin are present and may assist invasion. Gliomas modify the ECM by enzymatic digestion of the ECM proteins. There are changes in cell–cell and cell–matrix interactions and abnormalities in water and electrolyte composition. A characteristic of many tumours, particularly glioblastomas, is heterogeneity. The intracellular and extracellular pH varies from a borderline alkalosis in regions of high perfusion to acidosis in regions of insufficiency. Likewise the PO2 varies both in and around the tumour. It is clear, however, that large regions of the tumour are markedly hypoxic.13 The region adjacent to many tumours is characterized by oedema and inflammatory cell infiltrate.
Immunology The brain does not have a lymphatic system. There are, however, phagocytic microglial cells, which probably derive from monocytes in the developing brain. They express macrophage markers and act as macrophages do
292 Central nervous system
elsewhere in the body. Abnormal vessels in and around tumours allow the transfer of proteins and cells that would normally be prevented by an intact blood–brain barrier. Brain tumour cells exhibit a number of tumourassociated antigens, such as melanoma antigen (MAGE) and tenacsin, which may also be found on other neuroectodermal tumours and fetal cells. Astrocytic tumours express CD10 and EGFR. In vitro, human glioma cells secrete a number of inhibitory factors such as transforming growth factor (TGF)-β2, prostaglandin E2 (PGE2) and interleukin-10 (IL-10), which act to reduce the local immune response to the tumour. Other cytokines are secreted, including IL-2, IL-6, IL-8 and granulocyte–macrophage colony-stimulating factor (GM-CSF). A variable, but often marked, cellular immune response occurs at the site of a brain tumour. The infiltrate is predominantly of CD8 T lymphocytes and macrophages, with a minor component of B cells and natural killer (NK) cells. The systemic cellular immune response in patients with malignant brain tumours is often markedly impaired.14
Tumour spread By definition, benign tumours in the brain are non-invasive and non-metastatic. Their clinical effects arise from compression of adjacent structures and the associated functional detriment. Primary malignant tumours, on the other hand, are usually highly infiltrative. Cellular motility occurs early in their development. These cells typically disseminate along white-matter tracts, following the pattern set by the tumorigenic oedema, and may be found many centimetres from the apparent ‘edge’ of the tumour.15 Spread in this way does not respect boundaries between the lobes of the brain; spread across the corpus callosum is common. This pattern is followed by many malignant tumours but is characteristic for malignant gliomas (Fig. 14.2). The term ‘butterfly tumour’ is frequently used to describe a lesion that involves both frontal lobes. Local spread will also occur along the meningeal surface in superficially placed tumours. Spread via the CSF is common in a few tumour types – medulloblastoma, pineoblastoma, germ-cell tumours and lymphoma – but is clinically less apparent in gliomas. However, if a post-mortem search is made in late survivors with glioma, this route of dissemination appears more common than previously thought. Systemic spread for primary CNS tumours is uncommon. Exceptions to this are medulloblastoma, mesenchymal lesions (e.g. haemangiopericytoma and meningeal sarcoma) and, possibly, lymphomas and germ-cell tumours. In the last of these, however, it may be difficult to know whether the brain lesion was part of a systemic process from the outset. The pattern of spread of brain tumours is quite different from that of cancers at the more common sites. Hence the TMN staging system, which works well for systemic cancer, has little relevance in brain tumours.
Figure 14.2 ‘Butterfly glioma’. This CT scan shows characteristic features of a glioblastoma, including contrast enhancement, multilobar involvement, transcallosal spread, and oedema extending along the white matter tracts.
PRESENTATION The functions of the brain are so numerous and diverse that it is not surprising that brain tumour presentation is also highly varied. A detailed description of the many syndromes is beyond the scope of this chapter and the reader is referred to a standard neurology text.16 Some general principles can be applied. The effects of brain tumours can be categorized as local or global (Table 14.4). Symptoms due to pressure may be completely or partially reversed following tumour decompression, shunting or treatment with steroids. Damage to brain tissue due to infiltration or haemorrhage is frequently irreversible, though delayed recovery or compensation can occur after successful treatment. The brain’s inflammatory response to a tumour, particularly the development of oedema, may contribute to symptoms. Clinical response to steroids in these circumstances is often dramatic but may not be associated with corresponding changes on brain imaging. The mechanism of steroid action is far from clear. Likewise the cause of seizure in patients with brain tumours is not understood. Seizures frequently persist, even after anti-tumour treatment and appropriate anticonvulsants. Table 14.5 shows the first presenting symptom in 120 consecutive patients with glioblastoma. The individual symptom incidence, however, is much higher, with headache a feature in more than 70 per cent at presentation. Whilst both headache and seizure are common symptoms in patients with brain tumours, in the community as a whole brain tumours are a rare cause of these symptoms. A general practitioner will, on average, see five new patients with
Imaging 293
Table 14.4 Categorization of the influences of brain tumours Site Local
Global
Effect
Symptom
Outcome following treatment
Infiltration Haemorrhage Electrical Pressure/distortion Metabolic Pressure due to space occupation or obstructive hydrocephalus Electrical
Local syndromes
Usually irreversible
Seizure Local syndromes
Usually reversible
Headache, nausea, tiredness, cognitive change, brain-stem syndrome Seizure
Usually reversible
Table 14.5 Initial symptom in 120 consecutive patients with newly diagnosed glioblastoma presenting to the Neuro-Oncology Unit at Institute of Neurological Sciences in Glasgow Symptom Headache Seizure Limb weakness Cognitive change Visual disturbance Altered consciousness Incoordination Sensory disturbance Speech disturbance Other
% 34 23 12 8 5 4 4 3 3 4
primary brain tumours in their career but thousands with headache. Suspicion should be aroused when the headache fits the pattern for raised intracranial pressure, is unremitting, arises in unexpected circumstances or is accompanied by vomiting or neurological deficit.17 The great majority of patients presenting with seizure require a brain scan, but this is not feasible in all patients with headache. Evidence of recent neurological, psychological or behavioural change should be actively sought and examination of the fundi performed. If an abnormality is found or if the headache intensifies in the face of conservative management, specialist referral and imaging are indicated. Less usual presentations include the effects of hormonal disturbance, intellectual decline, developmental failure and disturbance of special senses.
IMAGING The mainstay of brain tumour diagnosis is structural imaging, using either computed tomography (CT) or magnetic resonance imaging (MRI). The use of contrast enhancement in both techniques is indispensable. Computed tomography is quick, has scalar integrity, is relatively easy to interpret
and is inexpensive. However, tumours less than 0.5 cm and those adjacent to bone may be missed. The basis of CT is differential absorption of X-rays. Since this is related to electron density, the derived Hounsfield numbers can be used as input to CT planning systems for radiotherapy. Magnetic resonance imaging relies on radiofrequency emission from perturbed atoms in a strong magnetic field. It is more expensive, prone to distortions and is more difficult to interpret. However, it is the more sensitive investigation and does not suffer from bone artefact. It is the diagnostic modality of choice for brain tumours in most circumstances (see Figs 14.3, 14.14 and 14.18). A few patients with metal in their bodies, from either previous surgery or accidents, may not be eligible for MRI scan. Diffusion tensor imaging is a new development in brain MRI. By showing the direction of flow of water, it allows the demonstration of the white-matter tracts. Water flow may be disrupted by the presence of tumour cells, leading to more isotropy of movement of the water molecules. Isotropy maps can, in principle, be used to delineate the extent of tumour infiltration and may be used to aid planning of radiotherapy or to determine response to treatment.18 The very small changes in blood flow that occur when the brain initiates a particular function (speech, hand movement etc.) can also be imaged using MRI. By making ‘functional maps’ of the brain in this way, a surgeon can plan an operation to avoid damaging important functional areas. Brain tumour images appear as infiltrative or spaceoccupying lesions. The position and distribution of the tumour together with the density variation, particularly after contrast injection, provides important diagnostic information. Tumours enhance with contrast (on CT and MRI) if they are hypervascular or if the vessels are abnormal and leak excessive contrast material. Areas of enhancement often indicate actively growing tumour. High-grade gliomas enhance in their actively growing rim (see Fig. 14.2) but not in their necrotic centres or in the associated oedema. Lymphomas (see Fig. 14.15), meningiomas (see Fig. 14.14), and many metastases (see Fig. 14.18) are usually uniformly enhancing. Slowly growing tumours such as
294 Central nervous system
Figure 14.4 Single photon emission computerized tomographic (SPECT) scan showing both thallium uptake and Tc-hexamethylpropyleneamine oxime (HMPAO) images in a glioblastoma (See Plate Section.). (a)
(b)
Figure 14.3 (a) T1 MR scan with gadolinium showing lack of uptake in a low-grade glioma. Also shown is a lack of brain distortion and pressure effects in this diffuse infiltrating tumour. (b) Equivalent T2 image clearly showing the extent of the high signal, which probably corresponds with the extent of tumour infiltration.
low-grade gliomas often fail to enhance at all (Fig. 14.3). Some tumours may show calcification. Whilst this is particularly common in oligodendrogliomas (see Fig. 14.12) and craniopharyngiomas, it may be seen in almost any slow-growing tumour. It is important to realize that whilst imaging appearances can give a good indication of the nature of a tumour, they are not totally diagnostic. For example a solitary, mixed density,
enhancing, space-occupying lesion may suggest an intrinsic malignant tumour. However, these appearances are also common in metastases and, less commonly, can be seen in meningioma, lymphoma or non-malignant lesions such as abscess or radiation necrosis. The magnetic resonance properties of tumour-infiltrated brain can also be used to produce magnetic resonance spectra, which may be typical of a particular tumour type and help identification. However, histological examination is the only sure way to categorize a tumour. Prior surgical intervention can also produce diagnostic problems. Postoperative imaging for residual tumour assessment should be done within 72 hours of the operation. After this time, appearances can be misleading, particularly due to the presence of altered blood. So-called functional imaging provides complementary information. Positron emission tomography (PET) scanning relies on the use of positron-emitting isotopes integrated into metabolically active molecules (e.g.18F-fluorodeoxyglucose, FDG). The oppositely directed, co-linear γ-rays can be detected by scintillation counters and turned into a three-dimensional image that represents the metabolic activity associated with the imaging agent. For example, 18 FDG uptake into glioma recurrence may distinguish it from radionecrosis in a patient previously treated with radiation. Single-photon emission computerized tomography (SPECT) similarly uses a radiolabelled, metabolically active molecule. In this technique, single γ-ray photons are imaged by a rotating gamma-camera and three-dimensional images are created using standard tomography algorithms. An example is 201thallium (chloride), which is an analogue of potassium and is taken up preferentially into rapidly proliferating tumours. It may differentiate lowgrade and high-grade tumours (Fig. 14.4). Other uses of functional imaging include assessing tumour response to treatment and measuring blood flow.
Treatment modalities 295
TREATMENT MODALITIES Surgery Surgery has three roles in brain tumour management: to obtain a diagnosis, to contribute to survival and to relieve symptoms. Biopsy alone may be attempted, either as an open procedure or by using stereotactic techniques. Stereotactic biopsy relies on relating a CT-acquired (or MRI-acquired) brain image to an external coordinate set. The image can then be used to direct a biopsy needle to the point of interest with an accuracy of 1–2 mm through a burr-hole. The diagnostic yield is very high (90 per cent) and the morbidity/mortality correspondingly low.19 There are situations in which even stereotaxy may be considered hazardous. In these circumstances, biopsy under direct vision may be required using craniotomy or ventriculoscopy to visualize the tumour. Biopsies (multiple where possible) should be taken from the most active looking part of the tumour. Functional imaging (e.g. SPECT) may be helpful in achieving this. When interpreting the biopsy, it is important to provide the pathologist with full clinical information such as the patient’s demography, presentation, previous treatment and any previous insult to that area of brain (e.g. prior irradiation). Full radiological information is also invaluable. It is important to remember that sampling error may lead to under-diagnosis in a heterogeneous tumour. Thus adequate weight must be given to the clinical situation and the imaging appearances in making treatment decisions and prognostic forecasts. This is most appropriately done in the context of a multidisciplinary team meeting.1 The morbidity associated with neurosurgery for tumour has been greatly reduced by the routine use of corticosteroids. These are prescribed before operating and for some days thereafter. In many cases they can be reduced steadily in the days following surgery as the reaction settles, provided the source of pressure has been removed. Drugs that modify platelet function, and other potential anticoagulating agents, should be stopped preoperatively. Anticonvulsants are prescribed routinely in some countries, but in the UK are normally given only to patients who are known to suffer from seizures or are at high risk because of the site of the intended operation. For the resection of a tumour, access is gained by craniotomy. A scalp incision is made and a flap of skin and peri-cranial tissue turned. The site of the craniotomy will depend on the position of the tumour. Most surgeons prefer a free bone flap, which is created first by making burrholes, freeing the dura and then cutting the flap using a high-speed craniotome. The dura is then opened only by an amount that will give adequate exposure of the tumourbearing brain. The operating microscope is routinely used to give good visualization. Minimal retraction of the brain is performed to permit access to the tumour. It is common practice to confirm the diagnosis with an intra-operative frozen section or cytological smear, prior to performing
the full resection. The resection itself is normally performed by internal decompression using a mixture of bipolar coagulation, sharp dissection, cautery, ultrasonic aspiration and, if appropriate, laser vaporization. The aim is always to remove as much of the tumour as is safely possible. The operation is completed with dural closure, replacement of the bone flap and skin closure. The devitalized free bone flap is a potential site for infection that will be resistant to antibiotics. Infection in the flap following surgery will necessitate its removal. The possibility of tumour removal has been improved by image-guided surgery. In this technique a CT image of the tumour is displayed into the operating field via the operating microscope, allowing the surgeon to visualize the tumour more clearly. This technique has been enhanced by the ability to fuse multi-modality images (CT, MRI, SPECT) prior to projection. In this way the advantages of each imaging modality can be used to improve the accuracy of localization and completeness of resection. A further refinement is to use intra-operative MRI. This is a complex technique that requires all the operative equipment to be MRI compatible. Few centres have this available, but early results suggest improved tumour-removal capability. Another technique to enhance tumour removal is to use a photosensitive dye such as 5-aminolevulinic acid (ALA), which is preferentially taken up into the tumour. When a laser is directed onto the operative field, the ALA in the tumour cells fluoresce, allowing its identification and removal.20 The technique of awake craniotomy involves ‘cortical mapping’ of neurological function. When performed during the operative procedure, it can be used to avoid resection of eloquent brain tissue.
Radiotherapy The value of radiotherapy in the treatment of brain tumours depends largely on their intrinsic radiosensitivity. In some it may be curative (medulloblastomas, germ-cell tumours), whereas in others it will only contribute to a modest prolongation of survival. Most radiotherapy is delivered using postoperative external beam X-rays. However, various other techniques are possible, including intra-operative radiotherapy, particle radiotherapy and brachytherapy. In all situations the aim is to deliver a maximal (or curative) dose to the entire tumour, whilst minimizing the normaltissue dose and volume. The particular applications are discussed under the individual tumour types. RADIOTHERAPY TECHNIQUE
Partial brain irradiation In the majority of cases, the patient’s head should be immobilized in preparation for treatment planning. Mouldable plastics such as Aquaplast or Orfit have the advantage of speed and simplicity but are perforated and less rigid and
296 Central nervous system
do not carry marking information well (Fig. 14.5b). Beamdirection shells (BDSs) can reproduce a set-up position to an accuracy of 3–5 mm. They are rigid and easy to mark (Fig. 14.5a). However, they are cumbersome and time consuming to prepare. Patients are most commonly positioned supine (Fig. 14.5b), but it may be necessary to plan the patient in the prone position (Fig. 14.5a) or even lying on the side in order to facilitate beam placement. Good-quality images (CT or MRI) must be available for the planning of radiotherapy for brain tumours. A combination of preoperative and postoperative imaging is optimal for this. It must be borne in mind that postoperative scans may be complicated by the difficulty of differentiating residual tumour from postoperative change. Simple planning in palliative situations can be done by screening at the simulator or by using a manual method of image
transfer onto orthogonal radiographs of the skull. If information from MR images is used, care must be taken to avoid errors due to distortion. Patients proceeding to radical radiotherapy for brain tumour should have access to a CT planning system for optimal use of imaging, beam placement and beam shaping. Contrast should be given to enhancing tumours. For more complex cases, where interpretation of the postoperative CT scan alone may be difficult, fusion of preoperative and postoperative CT/MRI or functional images in the planning system is an important tool (Fig. 14.6). It is best practice to use multiple fields to deliver the radiation to the tumour and minimize dose to nontumour-bearing brain. Whole-brain radiotherapy is rarely used for gliomas. Opposed pair arrangements can be used for tumours affecting both sides of the brain or in very palliative set-ups. Otherwise the choice is usually three fields with appropriate wedging (Fig. 14.7). Non-coplanar beam arrangements and the use of conformal shaping with blocks or Multi Leaf Collimator (MLC) can further reduce the dose to normal brain. It is usual to treat patients with brain tumours using a 4–6 MV linear accelerator, which has good dose characteristics over the required depth range, typically 2–12 cm. The build-up depth of around 1 cm gives good skin sparing. High dose rates lead to short treatment times and the high energy means that bone doses are not excessive. Cobalt-60 does not match these characteristics, and orthovoltage should not be used except in the most palliative of situations.
(a)
(b)
Figure 14.5 Immobilization devices for brain tumour therapy. (a) Thermoplastic beam direction shell (BDS) in the prone position. (Notice the cut-out region to preserve skin sparing.) (b) An Orfit shell.
Figure 14.6 A mosaic of fused CT and MRI images used in the radiotherapy planning process.
Treatment modalities 297
In radical treatments for brain tumours, fractionation is generally kept at or below 2 Gy per fraction (see below). Treatments are given daily, 5 days per week. No radiobiological advantage has been shown for accelerated or hyperfractionated treatments. It is customary to maintain patients on corticosteroids during their radiotherapy. However, the need for high doses is probably over-emphasized, particularly if a good decompression has been performed. Decompressed patients may require only low doses of dexamethasone (2–4 mg per day), or none at all. Hair loss is universal in the irradiated site, but regrowth may occur if the dose at the hair follicle is kept below about 45 Gy. Other side effects from radiotherapy include tiredness, which is common and may persist for weeks after completion of treatment, and skin erythema. By using megavoltage radiation and cutting out the BDS (see Fig. 14.5b), skin sparing and the possibility of hair regrowth may be maximized. Nausea is rare with modern radiotherapy, and vomiting is exceptional if the correct dose of steroids is used. If these do occur, other possible causes of gastrointestinal upset must be actively excluded. Whole neuraxis radiotherapy This technique is used in situations in which the entire meningeal content is at risk from tumour dissemination in the cerebrospinal fluid (CSF), e.g. medulloblastoma. The
problem is one of irradiating, in continuity, the brain, which is best approached with an opposed pair of photon beams, and the spine, which requires a direct posterior field of photons or electrons. The standard technique uses a 4–6 MV linear accelerator. The patient is immobilized in a prone cast extended over the shoulders to reduce movements of the trunk. Planning may be done by imaging the head and spine using plain X-rays and employing appropriate skin and magnification markers. However, it is now standard practice to plan with CT. The whole of the meningeal surface and content is outlined, with particular care being paid to the cribriform plate and the sacral sac. Further volumes, including the tumour bed, are outlined for planning the boost treatments. Fusion with MRI can be particularly valuable to clarify structures in the posterior fossa. A pair of lateral fields is used to irradiate the head. These are shaped using individually cast blocks (or ‘mini MLC’) to include the entire meningeal surface (Fig. 14.8). Particular attention must be paid to the region of the cribriform plate and the temporal lobes, which may extend lower than often appreciated. Centring the field on the outer canthus reduces divergence dose to the opposite lens. Compensators may be used to account for varying head width. The spine is irradiated using a direct posterior field (Fig. 14.8). A collimator rotation on the head fields matches the divergence of the spine field (typically 7°). In children, care must be taken to irradiate the whole of the growing vertebra. Two spinal fields, also matched, may be needed in adults. A compensator is used to produce a uniform dose to the cord (prescribed at the anterior cord surface). The lower border is defined to cover the apex of the sacral sac (normally at least to the bottom of S2). There is no evidence that a sacral ‘spade’ with extension to cover the sacral roots is required. Because of the uncertainties involved in establishing the junctions at each treatment, it is conventional in many departments to move them by 1 cm every seven treatments. Similar treatments to the spine have been described using electrons. Compensator
Temporal pole Cribriform plate
Thecal sac Spinal cord Lower border atleast at S2, 3 junction
Figure 14.7 Three-field plan created to treat a right frontal glioblastoma on a 6 MV Linac. Also shown are DRRs of the three conformed fields.
Shielding block Match posterior field and head fields at C3, 4 junction
Figure 14.8 Schematic of whole neuraxis treatment. The head fields are opposed lateral megavoltage X-ray beams. Note the shielding shaped to include the entire meningeal surface and the field tilt to match the divergence of the direct posterior spinal field.
298 Central nervous system
Special techniques Conformal therapy implies shaping each radiation field to fit an irregularly shaped tumour. This is now easily achieved with mini MLC, fitted as standard on many modern linacs, or cast blocks. Also, by increasing the number of beams and varying the intensity of radiation across them (intensity modulated radiotherapy (IMRT)) it is possible to conform the irradiated volume more closely to the tumour shape, including concavities. In principle, application of these techniques should lead to a reduction in toxicity, by reducing the volume of irradiated normal brain, and offer the possibility of dose escalation. Clinical studies to confirm the value of this approach are in progress. Stereotaxy simply relates to a method of localization in which the target is defined by an external coordinate system. This has the potential for great accuracy and implies the need for rigid fixation. This fixation may be ‘one off’, using pins screwed into the skull, or may use a re-locatable system based on a dental mouth bite or accurate BDS.21 Accuracies of 1–2 mm are standard. Treatment may be delivered using multiple sources, collimated to the same spherical volume (Gamma Knife), or using multiple coplanar arc therapy on a linac. Although the isodose volumes will be spherical in these cases, more complex shapes can be constructed from multiple, superimposed spheres. Multiple fixed conformed beams can also be delivered stereotactically. If a single high-dose treatment is given, this is known as stereotactic radiosurgery (SRS). This technique has an established role in the treatment of arteriovenous malformations and small brain metastases. Use of a re-locatable frame allows multiple stereotactic doses of radiation to be given. This is fractionated stereotactic radiotherapy (FSRT), which is increasingly used to treat lesions such as acoustic neuromas, skull-base and paediatric tumours. Whether stereotactic localization has a role in the management of gliomas has yet to be established. It is important to note that the radiobiologies of SRS and SRT are quite different. Protons travelling in a medium deposit a large amount of energy into the final few millimetres of their path – the Bragg peak. This property can be used to deliver highly localized, three-dimensional conformed dose distributions, even to volumes with concavities. This has led to the use of protons in the treatment of tumours requiring high doses but that lie immediately adjacent to vital structures. The most common application in the CNS has been to chordomas of the clivus, but other skull-base tumours would be appropriate targets. The production of high-energy proton beams is expensive and the technique is available in very few centres. At the time of writing, no beam suitable for the treatment of brain tumours exists in the UK. Brachytherapy, usually using temporary, high-energy 125 I seed implants, has been used to treat high-grade gliomas at relapse or as a boost at first presentation.22 The early results of randomized trials appeared encouraging, but mature results have failed to confirm this early promise and the technique is now little used. In Europe,
low-energy implants have been used to treat low-grade gliomas,23 and iridium wire afterloading has been used for palliation in glioblastoma. Boron neutron capture therapy (BNCT) relies on the capture interaction of very low-energy thermal or epithermal neutrons with the stable isotope boron-10 (10B). Capture causes the 10B to split, releasing an α and a lithium particle: N 10B : 4He 7Li The requirement for successful BNCT is selective accumulation of 10B into brain tumours using boron-labelled compounds such as 10B-bovine serum albumin. Irradiation with low-energy neutrons causes the release of shortrange, high-LET (linear energy transfer) particles in the vicinity of the tumour. The technique is under study in a few centres worldwide.
Normal tissue reactions to radiotherapy As in other sites, radiation injury in the CNS depends on a number of factors that may be patient related (age, vasculopathy, infection) or treatment related (total dose, dose per fraction, volume irradiated). In general, the CNS is a late-responding tissue; however, both early and intermediate effects can also occur. ACUTE EFFECTS
Acute effects of radiation begin within days or even hours and are probably an oedematous response, though this has never been clearly demonstrated. The acute tolerance of the brain is higher, both in terms of total dose (up to 80 Gy in 2-Gy fractions) and single doses (6–8 Gy), than is acceptable for late effects. Sheline reported four acute deaths in 54 patients when using single doses of 10 Gy.24 The symptoms are generally those of acute raised intracranial pressure or a worsening of neurological symptoms caused by the lesion itself. Therefore they are more commonly seen in patients who are symptomatic at the start of treatment. Using conventional dosing, acute effects are rarely troublesome if steroids are appropriately given. EARLY DELAYED (INTERMEDIATE) EFFECTS
An intermediate radiation reaction can begin within weeks of completing radiation therapy to the brain and may continue for 6–10 weeks thereafter. It comprises somnolence, lethargy and, frequently, recurrence of the original presenting symptoms and signs. It is usually self-limiting but will respond to steroids. Recovery is the rule. The condition was originally described in children as ‘somnolence syndrome’ and later in adults treated to moderate brain doses (55 Gy in 27 fractions).25 The pathogenesis is not known but is believed to correlate with interruption of myelin synthesis secondary to damage to the oligodendroglial
Treatment modalities 299
cells. In the rare cases that fail to recover, autopsy has shown extensive demyelination, central necrosis, diminished numbers of oligodendroglial cells and gliosis, but without evidence of vascular, neuronal or axonal injury. A corresponding condition occurs after spinal irradiation and presents with Lhermitte’s sign. This too is self-limiting. DELAYED RADIATION DAMAGE
This is the most sinister form of radiation damage and is uniformly irreversible. The onset may be from around 3–4 months up to many years after the exposure. Injury is predominantly to the white matter and is dose and volume dependent. The clinical manifestations are diverse and may vary from subtle deterioration in higher cognitive function and behavioural changes to gross neurological deficit associated with a space-occupying lesion. Pathologically the lesion may also vary in severity. Milder forms demonstrate gliosis, demyelination, vascular proliferation and thickening and neuron fallout. Radiation necrosis is the most severe form of damage. It appears as an expansive lesion within the gyrus, which, on section, typically comprises haemorrhagic coagulation necrosis, restricted largely to the white matter. A fibrinous exudate typically accumulates in a lamina of hypocellularity along the grey–white junction. Fibrinoid necrosis is very common. Vascular proliferation can occur in the early phase of late damage and can resemble tumour. A further characteristic of irradiated brain is a large, bizarre, widely disseminated cell, the nature of which is not clear. In the more chronic phase, some of the necrotic foci are resorbed and can evolve into multiple cysts. Calcification is characteristic of the later stages, along with encephalomalacia, telangiectasia and vascular thickening. The underlying cause of radiation necrosis is not known, but there is accumulating evidence that fibrinolytic inhibitors are involved. Late radiation damage can be difficult to differentiate clinically and radiologically from tumour recurrence. Furthermore the two can co-exist. Functional rather than structural imaging plays an increasing role in differentiating the processes. Both FDG-PET and thallium-SPECT can show differentiation. The rate of onset and the severity of the late radiation response in brain depend strongly on the fraction size and total dose. However, the dependence on overall treatment time is weak for treatments more than 12 hours apart. Accepted levels of tolerance vary according to the situation: 60 Gy in 30 fractions are commonly given to highgrade tumours and may cause around 5 per cent symptomatic late damage. In less demanding situations, 50–54 Gy in 2-Gy fractions is considered the upper limit. To minimize late damage, fraction sizes for radical brain treatments should not exceed 2 Gy. This has been emphasized in a carefully conducted series of studies of the late cognitive consequences of partial brain irradiation for tumour.26** These studies demonstrated that the tumour and other factors such as anticonvulsant medication had a
greater effect on cognition than did the radiation, provided the dose per fraction was kept below 2 Gy. Other late consequences of brain irradiation include hormone (pituitary/hypothalamic) failure, damage to optic tracts and second malignancy. Late radiation damage in the cord may be sudden or insidious in onset, with sensory and motor abnormalities (paraplegia or quadriplegia), bowel and bladder sphincter disturbance and diaphragm dysfunction in high lesions. The most serious consequence is complete transection of the cord at the irradiated level. The pathology is similar to that in brain. A combination of vascular lesions with demyelination and malacia is characteristic of radiation myelopathy. The pathogenesis remains obscure, with both the vasculature and oligodendrocytes identified as principal targets.27 Imaging may aid diagnosis. Magnetic resonance imaging performed within 8 months of the onset of symptoms shows low signal intensity on the T1-weighted image and high signal intensity on T2-weighted images, often with cord swelling. Gadolinium enhancement is common. Late scans show an atrophic cord with normal signal intensity. There is no evidence that either the volume irradiated or the anatomical level of irradiation materially affects the incidence of myelitis in patients receiving standard doses of radiotherapy. Accepted wisdom has been that spinal cord tolerance at conventional fractionation is between 45 Gy and 50 Gy, depending on the clinical situation. Others have argued that the spinal cord tolerance should be revised upwards to 50 Gy or even higher,28,29 according to the clinical circumstance. A dose of 57–60 Gy carries a 5 per cent risk of myelitis.27 Furthermore, there is evidence that reirradiation of CNS tissue is possible. Some tolerance develops with increasing time from the initial radiation and is virtually complete (50–70 per cent) by 2 years. However, full tolerance is never regained. Some chemotherapeutic drugs can enhance radiation damage in CNS tissue. These include methotrexate, cytosine arabinoside and the nitrosoureas. These are all drugs that have good CNS penetration and can produce toxic damage in their own right. Others, such as adriamycin, become toxic when given following disruption of the blood–brain barrier. The toxic pathological changes are similar to those produced by radiation, with both fibrinoid and coagulation necrosis but with much greater involvement of the grey matter and much more generalized brain changes.
Chemotherapy For a chemotherapeutic drug to be effective it must be active against the tumour and have access to it. In the brain, the intact blood–brain barrier generally inhibits the passage of molecules with molecular weights greater than 200 Daltons. Drugs with a high partition coefficient (e.g. nitrosoureas) or that are small (temozolomide) can circumvent this barrier. Although in the vicinity of tumour the barrier is partially defective, as is demonstrated by the
300 Central nervous system
penetration of contrast scanning agents, large hydrophilic molecules remain largely excluded from the brain/tumour structures and are not useful for therapy. However, some agents thought to be only modestly penetrating can be very effective, for example platinum compounds in germ-cell tumours and medulloblastoma.
Procarbazine
THE ROLE OF CHEMOTHERAPY IN BRAIN TUMOURS
Temozolomide
Chemotherapy can be given in a neoadjuvant, adjuvant, concomitant, consolidation or palliative setting. In the past, gliomas have been considered relatively chemo-resistant. However, recent molecular analysis has identified subtypes of glioma that are much more responsive to treatment with alkylating agents (e.g. some oligodendrogliomas). For the rare chemosensitive tumours (teratoma and lymphoma), neoadjuvant treatment is well established, whereas in others, such as medulloblastoma, the role of adjuvant or neoadjuvant chemotherapy remains under development. This is discussed further in the relevant sections. ACTIVE AGENTS
Nitrosoureas The chloroethyl nitrosoureas were held to be the most effective group of drugs for the treatment of brain tumours, although very few comparative studies against other agents have been reported. They are highly lipid-soluble, non-ionized drugs that rapidly cross the blood–brain barrier. They degrade into two reactive compounds, one with a carbamylation activity and the other an alkylating agent. Carmustine (BCNU), lomustine (CCNU), semustine (MeCCNU), tauromustine (TCNU), (1-(2-chloroethyl)3-(2,6-dioxo-3-piperidyl)-1-nitrosourea)), PCNU, ACNU (1-(4-amino-2-methylpyrimidin-5-yl)methyl-3-(2chloroethyl)-3- nitrosourea) and streptozotocin have different pharmacokinetic properties whilst retaining the same basic chemical activity and toxicity problems. None has proved more effective than any other, although BCNU is regarded as the benchmark drug. Given intravenously, the single-agent symptomatic response rate for glioma is 20–40 per cent. Radiological partial response is less, however. CCNU is reported to have a similar response rate and is administered orally. Adverse effects include delayed myelosuppression and lung fibrosis.30 Gliadel Gliadel is system comprising a biodegradable polymer in wafer form that is impregnated with BCNU. A number of these wafers are used to line the cavity left after resecting a brain tumour. After wound closure, the polymer slowly breaks down, delivering the BCNU in a more concentrated and protracted fashion than is possible to achieve by systemic delivery. Although Gliadel has been shown in prospective randomized clinical trials to improve survival in patients undergoing surgery for newly diagnosed and recurrent high-grade glioma, it is little used in the UK.31**
Procarbazine is activated in the liver to an alkylating agent. The single-agent symptomatic response rate for glioma is believed to be around 20 per cent. It causes nausea, vomiting and myelosuppression and interacts adversely with alcohol and some smoked and preserved foods.
Temozolomide is a small molecule that acts as an alkylating agent by adding methyl groups to the O6 position of guanine in DNA. It has high oral bioavailability and is converted spontaneously into the active compound at physiological pH on entering the bloodstream. It penetrates readily into brain tumours. It acts by methylating DNA and inducing futile cycling and apoptosis. It is opposed by MGMT, a suicide enzyme that acts to remove the methyl adducts. For this reason, protracted, fractionated delivery of temozolomide, which depletes the cell of MGMT, is likely to be more effective than short courses of drug. Temozolomide is active against astrocytic and oligodendroglial tumours and has a predictable and modest toxicity, principally myelosuppression. Temozolomide has been recommended as second-line treatment (after nitrosourea) in patients with relapsed malignant glioma and as concomitant treatment (with radiotherapy) in selected patients with newly diagnosed glioblastoma. Epidophyllotoxins and platinum compounds Drugs such as VP-16, cisplatinum and carboplatin are valuable for treating non-glial brain tumours such as medulloblastoma and germ-cell tumours. They have only minor activity against gliomas and are typically used as third-line agents. Topoisomerase inhibitors Topoisomerase inhibitors such as irinotecan have shown activity against glial tumours, including glioblastoma, both as single agents and in combination.32 However, whether they have a role in routine management has yet to be determined. COMBINATIONS
Few drugs are effective as single agents in glioma therapy and therefore combinations have been little studied. A combination of procarbazine, CCNU and vincristine (PCV) had been considered by many to be standard therapy as first-line treatment in relapsed glioma. However, the evidence that it is superior to single-agent nitrosourea is very sparse,33 and it is equally acceptable to use single-agent nitrosourea in relapsed glioblastoma. Combinations may be more successful for some less common tumours. Where appropriate, their use is discussed further in the relevant section. NOVEL APPLICATIONS OF CHEMOTHERAPY
Because of the difficulty of access of many agents to the brain, alternative strategies have been considered to try to
Treatment modalities 301
increase the concentration of drug in the tumour. Arterial catheterization has been used to deliver agents such as the nitrosoureas to the tumour. Results have been disappointing and the complication rates high. This technique remains confined to a few centres. Global blood–brain barrier disruption has been achieved with high-dose mannitol prior to infusion of hydrophilic drugs. Again, this approach has proved unacceptably toxic, as the blood–brain barrier disruption is non-selective. Convection-enhanced delivery (CED) is a surgical technique in which a thin catheter is placed into the brain and connected to an extracranial, pump-driven syringe containing a compatible medium. By selecting an appropriate delivery rate, the medium can be made to ‘flow’ into the brain or tumour. The medium can be used to carry a variety of cytotoxic agents, including large molecules or even small viruses, to regions in the brain remote from the catheter tip. This technique is currently under investigation in large randomized clinical trials.
New modalities GENE THERAPY
There are two major issues in gene therapy: what killing (or growth-limiting) strategy and what delivery mechanism should be adopted. There are many possibilities for killing, such as genes that sensitize the transfected cells to drugs that would otherwise be harmless – so-called suicide gene therapy (e.g. HSV-tk gene sensitizes to gancyclovir) – or genes whose products can switch off activated oncogenes (e.g. antisense compounds) or simply using viruses that have been engineered to kill malignant cells selectively (oncolytic viruses). Such strategies often work well in vitro. However, a more formidable problem is delivery of the gene therapy. Vehicles that have been proposed include naked DNA and liposomes, but most work has examined the potential role of viruses. Retroviruses, adenoviruses and herpes virus have all been proposed and a randomized trial using retroviral producer cells and an HSV-tk/gancyclovir strategy has been performed, though without success.34 As yet, there is no proven role for gene therapy in the treatment of brain tumours. TARGETING ANGIOGENESIS
Abnormal angiogenesis is fundamental to the growth of most brain tumours. It is controlled by a variety of growth factors, particularly vascular endothelial growth factor (VEGF) and platelet-derived growth factor (PDGF). The VEGF has been shown to be significantly up-regulated in glioblastoma. Angiogenesis would seem to be a logical target in these tumours. Although thalidomide was investigated, its toxicity did not merit the modest benefit it seemed to produce. Of the new generation of agents, enzastaurin and PTK 787 are
showing considerable promise and are in phase 3 clinical trials internationally. CELL-CYCLE MODULATORS
As discussed earlier on page 290 (under the heading ‘molecular biology of brain tumours), changes in the molecules that regulate the cell cycle (EGFR, PDGF, PTEN etc.) lead to loss in control of cell growth and division. It follows that if these molecular changes could be reversed, control might be re-established. This is the rationale behind the design of molecules whose function is specifically to block these aberrant molecules or their effects. There is no doubt that single-agent targeted therapy of this type can produce responses in small numbers of patients (Tarceva, Glivec), and some of these can be marked and durable, as shown in Figure 14.9. However, the level of response is too low to be useful and the future lies in combinations of these agents or their combination with conventional therapy. IMMUNOTHERAPY
The immunobiology of brain tumours is still poorly understood. In spite of this, some clinical attempts at immunotherapy have been made. Examples are the exploration of adoptive immunotherapy by injecting LAK cells concomitantly with IL-2 in patients with recurrent glioma or the use of vaccines produced from tumours. These approaches have major technical difficulties and are as yet unproven.35
Adjunctive treatments It has long been recognized that corticosteroids produce symptomatic benefit in patients with brain tumours. The physiological basis for this is not clear. The explanation that the steroid acts by reversing cerebral oedema is not born out fully by imaging studies and is almost certainly not wholly correct.36 Dexamethasone is most frequently used, in doses up to 64 mg daily. It is unusual for patients to be maintained on doses greater than 16 mg for more than a few days, and any dose should be titrated down to the minimum needed to control symptoms. This will minimize the impact of severe side effects, which in the elderly are particularly proximal myopathy, diabetes and osteoporosis, and in the young are acne and appetite stimulation. All patients may suffer weight gain, sleep disturbance and disorders of mood and perception. Epilepsy is common in patients with brain tumours. Anticonvulsants should be given in doses determined by the efficacy and toxicity of a drug in a particular patient and not by the measured plasma level. The established agents, carbamazepine, phenytoin and sodium valproate, are giving way to a newer generation of drugs including lamotrigine and levitracitam. The latter have the added advantage
302 Central nervous system
Figure 14.9 Response to treatment with Glivec. This 56-year-old woman with glioblastoma recurrent after surgery and radiotherapy shows an unequivocal response to daily Glivec. The images shown here are taken at 3-monthly intervals, but she has continued to respond for a further 12 months.
that they do not induce hepatic enzymes to the extent of some of the older drugs, and thus do not interfere with the metabolism of other therapeutic agents, particularly chemotherapy. Probably half of all patients with tumourassociated epilepsy will not have their fits completely controlled, even with ‘optimal’ doses of anticonvulsants. Although headache is a common presenting symptom, severe pain is fortunately unusual in patients with treated brain tumours, even following recurrence. When pain is a problem, the same analgesic ladder should be used as for other malignant conditions. The liberal use of morphine in the later stages of the disease is entirely justified. Nausea can be a particularly troublesome symptom and may have a variety of causes. It may arise as part of posterior fossa syndrome and secondary to raised intracranial pressure, but it is also associated with seizure, drug toxicity (particularly anticonvulsants) and peptic pathology. Limited-field brain irradiation uncommonly causes nausea, and this should be a diagnosis of exclusion. Clearly the cause of the toxicity should be sought and, where possible, treated. When the usual anti-emetics fail to control nausea of intracranial origin, it may be useful to try the continuous subcutaneous delivery of agents such as the major tranquillizers or antihistamines.
Additional support Brain tumours are rare but their effects are devastating. Cognitive decline and major physical disability often accompany the late stages. Because of their rarity, family doctors will have managed very few cases. Patients may require input from physiotherapists, speech therapists, social workers, palliative care workers, the GP and the district nurse. The specialist ‘brain tumour support nurse’,
who has knowledge of the patient’s current condition, needs and likely prognosis as well as of the facilities available in a particular area, can provide invaluable support to all workers involved in care, particularly the GP. Even if a brain tumour is incurable, the quality of the patient’s life can be greatly enhanced by efforts to minimize the impact of their deficit. It is not sufficient to prolong the life of patients without maximizing its quality.
INDIVIDUAL TUMOURS Astrocytomas Diffuse astrocytomas are a group of tumours, more common in adults, which can arise anywhere in the CNS but are most common in the cerebral hemispheres. Irrespective of their histological grade, they infiltrate diffusely into adjacent and distant brain. Tumour grade is based on cell density, cytological features (nuclear atypia, nucleus:cytoplasmic ratio), mitotic activity and features of the microenvironment (vascular proliferation and necrosis). The grade allocated to the tumour is the highest grade seen in any part of the specimen. The prognosis is strongly dependent on the grade. Other important prognostic factors are the patient’s age, performance status and mental status. The Radiation Therapy Oncology Group (RTOG) has used recursive-partitioning analysis to produce a classification system of six (RTOG-RPA) outcome classes. These are useful as a basis of predicting the likely outcome for particular patients or for developing historical comparisons in clinical trial work.37 Lower grades of astrocytoma progress to more malignant phenotypes. The rate of progression is very variable.
Individual tumours 303
Transformation, to glioblastoma, may be complete within a year or may not begin for 10 or even 20 years. When transformation occurs, it is accompanied by a cumulative acquisition of genetic alterations (on page 290 under the heading ‘the Molecular biology of brain tumours’). Pilocytic astrocytomas rarely transform. The aetiology of the majority of astrocytomas is obscure. The only known associations are with some inherited conditions (see Table 14.2) and exposure to ionizing radiation. GRADE I: PILOCYTIC ASTROCYTOMA
Pilocytic astrocytomas are tumours of children and young adults. They most commonly arise in the posterior fossa but are also found in the cerebral hemispheres, the optic nerve and chiasm, the thalamus and basal ganglia, the brain stem and the spinal cord. On imaging, they appear as hyperdense, well delineated, solid or solid/cystic tumours. They break the rule for low-grade tumours in that they frequently enhance with contrast. They are characterized by slow growth and may, without intervention, stop growing or even regress. Rarely, they may seed in the CNS. Unlike other low-grade astrocytomas, they very rarely progress to a more malignant phenotype. However, a rare, intrinsically aggressive form exists (which is possibly more common in adults), which progresses relentlessly without change in the low-grade histology. The histology is characterized by the presence of bipolar, GFAP-positive ‘pilocytes’ and eosinophilic hyaline masses known as Rosenthal fibres. They are often highly vascularized and, in contrast to other low-grade astrocytomas, may show endothelial hyperplasia identical to glioblastoma. They do not demonstrate inactivation of P53. The treatment of pilocytic astrocytoma is surgical. Maximal safe resection should be performed. Even if this is incomplete, further tumour progression may not occur. There is no evidence that postoperative radiation improves on surgery alone. If re-growth occurs, the treatment is again surgical. Should the second resection be incomplete, adjuvant radiotherapy can be tried (45–50 Gy), but there is no clear documentation that this is beneficial. Overall, the prognosis for pilocytic astrocytoma is very good, with long-term control or cure rates of 80–90 per cent. The achievement of a complete resection is an important prognostic factor.38 GRADE II: (DIFFUSE) ASTROCYTOMA
Low-grade diffuse astrocytomas (LGAs) are well-differentiated, slow-growing, diffuse tumours more commonly seen in young adults, although they can occur at any age. They usually arise in the cerebral hemispheres, but also occur in the brain stem and spinal cord. They are uncommon in the cerebellum. On CT imaging, they typically appear as ill-defined, hypodense, infiltrative lesions, which
may or may not displace other structures (see Fig. 14.3). They are usually non-contrast enhancing. They most commonly present with seizure. High signal intensity on T2weighted MRI or FLAIR (Fluid Attenuation Inversion Recovery) gives the best delineation of the extent of the tumour. Histologically they are recognized to contain abnormal numbers of astrocytes, which show monotonous, minor degrees of anaplasia (nuclear pleomorphism and cytoplasmic changes). Mitoses are uncommon. The background may contain increased numbers of cellular processes and microcysts, which aid diagnosis. GFAP positivity is common. These tumours are sometimes sub-grouped into fibrillary, protoplasmic and gemistocytic forms. This has prognostic significance only for the gemistocytic variant, which is generally regarded as more aggressive in nature and is treated in many departments as an anaplastic astrocytoma. The management of low-grade astrocytoma is one of the most controversial areas in neuro-oncology. We know that these tumours will progress and that the majority will transform to higher-grade malignancy. However, patients presenting with controllable seizures as their only symptom and who have non-distorting, low-density lesions on scan may remain well for many years. The prognostic factors for patients with these tumours have been defined.39** These include tumour size (6 cm), lesions crossing the midline, neurological symptoms and age above 40 years. There is no evidence that early radiotherapy intervention in good-prognosis patients improves survival when compared to treatment delayed until required;40** however, time to progression is improved. It is perfectly acceptable to observe these patients in the first instance, thus sparing them the morbidity of treatment. However, this policy may disadvantage a minority of patients whose tumours are misdiagnosed on scan (e.g. who have a more malignant lesion) and who may benefit from early intervention. A policy of biopsy ‘up front’ and then observation if appropriate is adopted in some units. Patients who find a policy of observation unacceptable after full consultation can be treated as outlined below. Patients who require treatment are those with poor prognostic features or whose tumours are clearly progressing, as witnessed by demonstrable growth on scan or the development of neurological deficit or possibly loss of seizure control. If most of the tumour mass is accessible, maximal safe resection is usually recommended, although the data to support this concept are sparse. That radiotherapy causes tumour regression and symptom relief is a matter of common experience. Furthermore, two randomized trials inform us that doses beyond 45–50 Gy are not necessary.41,42** Our recommendation is to offer radiotherapy within 4–8 weeks of surgery. Patients are planned using CT and CT/MRI fusion. The tumour is defined as the high signal region on the T2 or FLAIR sequences. This is expanded by 1–2 cm to create the clinical target volume (CTV). The additional margin required to create the planning target volume (PTV) is determined by local facilities and
304 Central nervous system
techniques. Two to four fields may be required to produce an adequate plan. These should be conformed to minimize the volume of non-tumour-bearing brain irradiated to high dose. A total of 45–50 Gy is delivered in 1.8-Gy fractions, depending on the size of the irradiated volume. The outcome after this form of management is variable. The majority of patients will relapse, with ultimate tumour progression and death. Overall, the median survival is about 5–8 years. However, some patients continue well for many years, showing no sign of progression even 20 years after treatment. The role of chemotherapy is even less clear than that of surgery or radiotherapy. Symptomatic and radiologic responses are undoubtedly observed in newly diagnosed and relapsed disease. Active agents include nitrosoureas and temozolomide. A new European Organisation for Research and Treatment of Cancer (EORTC) prospective randomized trial is examining the relative worth of chemotherapy with temozolomide and radiotherapy. For patients who relapse following treatment with radiotherapy, an attempt at salvage with chemotherapy is undoubtedly worthwhile. GRADE III: ANAPLASTIC ASTROCYTOMA
Anaplastic astrocytoma is also a diffusely infiltrating, astrocytic tumour that is typified by increasing cellularity, increasing nuclear size and variation and the presence of mitoses (growth fraction around 5–10 per cent). In the WHO classification, tumour necrosis must not be present. It is primarily a tumour of middle age. It appears with variable density or signal on CT or MRI respectively. There is usually enhancement with contrast, but this may be diffuse or patchy and is rarely of the classical ‘ring’ type seen in glioblastoma. Unlike LGA, rapid growth in anaplastic astrocytoma is the rule. It is treated as glioblastoma, but the prognosis is considerably better, with a median survival of 2–4 years. There is also evidence of increased chemosensitivity as compared to glioblastoma, and the two groups of patients should be distinguished in clinical trial work. The disease is almost uniformly fatal, and transformation to glioblastoma usually occurs in the later stages. Note that at the time of writing the role of concomitant chemo-radiotherapy for anaplastic astrocytoma has not been established. GRADE IV: GLIOBLASTOMA
Glioblastoma is a poorly differentiated, extensively invasive, highly mitotic astrocytic tumour. It is characterized by areas of necrosis and the presence of microvascular proliferation (previously known as capillary endothelial hyperplasia). It can arise at any age but is most common in the sixth and seventh decades. It occurs preferentially in the cerebral hemispheres, from where it extends into adjacent structures such as basal ganglia and across the corpus callosum to form the classical ‘butterfly tumour’. It can arise, rarely, in the cerebellum, brain stem and spinal cord. It is universally fatal.
The tumour was previously called glioblastoma multiforme, because of the marked pathological heterogeneity within and between tumours. Macroscopically, the tumours appear as grey masses with areas of haemorrhage, yellow necrosis and sometimes cysts. They may appear to have a capsule, but this is always an artefact of rapidly growing tumour and compressed brain. Adjacent brain is commonly swollen with peri-tumoral oedema, which tends to spread along the white-matter tracts and can form a conduit for the migratory tumour cells, which can be found many centimetres from the main tumour mass. These migratory cells may form secondary masses. Therefore, although glioblastomas may appear to be multi-focal, this usually represents spread from a single tumour. The tumours are also highly heterogeneous at a cellular level, with almost any size and shape of cell being seen, including multi-nucleated giant cells, which are a hallmark of glioblastoma. Secondary glioblastoma is more likely to have areas comprising lower-grade astrocytoma. Immunohistochemical staining with GFAP, S-100 and vimentin is also heterogeneous. An inflammatory infiltrate is usually seen. This comprises predominantly CD8 T-lymphocytes; CD4 and B-lymphocytes are much less common. Recognized glioblastoma variants are giant-cell glioblastoma and gliosarcoma, the natural histories of which are similar to that of classical glioblastoma. Microvascular proliferation and tumour necrosis are the histological hallmarks of glioblastoma. Angiogenic tyrosine receptors (e.g. VEGFr-1 and 2) are up-regulated in proliferating tumour vessels, and the ligands (e.g. VEGF) are found in the glioblastoma cells themselves, leading to paracrine stimulation of angiogenesis. There are two types of tumour necrosis: large-scale coagulation necrosis, which is usually visible on imaging studies, and microscopic serpiginous foci of necrosis, which form the pseudo-pallisading pattern typical of glioblastoma. The molecular biological changes are discussed earlier in this chapter. The typical imaging appearance of glioblastoma is a rim-enhancing, irregular tumour with a low-density (necrotic/cystic) centre and surrounding low-density oedematous brain. They tend to centre in the deep white matter. The brain is often distorted, with evidence of raised intracranial pressure (see Fig. 14.2). Metastases may produce a similar appearance, but tend to lie preferentially at the grey/white junction, are often smaller and less heterogeneous and are commonly multiple. In the 1980s the Mayo Clinic group performed important studies. By taking serial stereotactic biopsies through glioblastomas and peritumoral brain, they were able to correlate the pathology with the imaging appearances.15 They showed that while the main growing tumour mass corresponded to the enhancing rim of tumour, isolated tumour cells could be found throughout the peri-tumoral oedema. The overall strategy of management will depend on the likely prognosis. A diagnosis of glioblastoma will be suspected from the imaging in the majority of cases. The very elderly, or patients with severe, steroid-resistant symptoms, may need only symptomatic care, as their outlook
Individual tumours 305
with glioblastoma. Since the 1970s, numerous trials of adjuvant (nitrosourea-based) chemotherapy with radiotherapy failed to show any worthwhile advantage.48*** However, in 2005, the EORTC/NCIC (National Cancer Institute of Canada) published the results of a randomized trial in which a regime of standard radiotherapy with concomitant daily temozolomide (75 mg/m2) followed by monthly temozolomide (200 mg/m2 5) produced a survival advantage of 16 per cent (10–26 per cent) at 2 years49** (Fig. 14.11). Furthermore, the quality of life was maintained and the toxicity was minimal. Benefit was greatest in well-resected patients with WHO performance status 0 or 1. Poor performance status patients did not appear to benefit. This scheme has been adopted as standard treatment in selected patients and as the appropriate comparator arm in future studies. Another important feature of this study was the observation that if the promoter of the MGMT gene was methylated (and the gene silenced), these patients responded particularly well to temozolomide50 (see page 300 under heading ‘Active Agents, Nitrosoureas). If this observation
Definition of prognostic index Prognostic factor
Category
Score
Age (years)
44 45–59 60
0 6 12
WHO performance status
0–1 2 3–4
0 4 8
Extent of neurosurgery
Complete resection Partial resection Biopsy
0 4 8
History of fits (months)
3 3 None
0 5 10
Prognostic index sum of scores for each factor, a low score indicating a better prognosis 1 0.900 0.800 0.700 Survival rate %
will be little changed by provision of the precise diagnosis and treatment. All other patients need a minimum of biopsy proof of diagnosis. Where possible, it would appear good practice to resect as much of the tumour as is safely possible. The surgical ambitions are to provide a complete diagnosis, to relieve symptoms and to delay tumour progression. Although there is no adequate randomized study to show that maximal resection improves quality of life or longevity when compared to lesser forms of surgery, there is compelling non-randomized evidence associating complete macroscopic removal with improved outcome.43 Furthermore, it is a matter of common clinical experience that patients tolerate subsequent forms of treatment, particularly radiotherapy, much better when their tumours have been decompressed. Surgery to provide adequate symptomatic relief is to be encouraged. Between 1978 and 1981, four seminal studies showed beyond any doubt that treatment with radiation improves survival in patients with malignant glioma when compared to best supportive care alone.44 Surgery followed by limited-volume radiation therapy is now the standard treatment in this disease. Whole-brain radiotherapy has been abandoned. There are two approaches to choosing the radiotherapy target volume. In the first, we rely on the Mayo studies,15 which show that tumour cells extend to, at least, the low-density oedema seen on CT (or the region of abnormality seen on T2weighted MRI). This region then defines the phase 1 volume to be irradiated (usually to two-thirds the dose). In a second phase the full radical dose is applied the region of the enhancing tumour with a modest margin of 1-2 cm (gross tumour volume 2 cm). This technique is prevalent in the USA. The second approach is more pragmatic. It recognizes that when glioblastomas relapse they do so within 2 cm of the original enhancing rim in 85 per cent of cases.45 If this problem is not solved, what happens outside this region of relapse is of little consequence. In this one-phase approach, the CTV is therefore defined throughout as the enhancing tumour rim (GTV) plus 2–3 cm. Radiotherapy is usually delivered using a 4–6 MV linear accelerator. The standard dose is 60 Gy in 30 fractions over 6 weeks to the enhancing tumour plus 1–2 cm. If an initial phase is used, this is usually restricted to 40 Gy. Alternative regimens are 45 Gy in 20 treatments (which can be delivered on a twice-daily basis) for patients in whom the more protracted regimen cannot be justified, and 30 Gy in six fractions on alternate days as a very palliative approach.46 The outcome following radiotherapy alone remains poor. Figure 14.10 shows the survival in patients treated with radical radiotherapy according to the known prognostic factors.47 Overall median survival is around 1 year, with less than 10 per cent surviving 2 years and almost no one alive at 5 years. Good prognostic indicators are youth and good performance status, with presentation with seizure and completeness of resection having a lesser impact. Primary chemo-radiotherapy for patients with good performance status is now the standard of care for patients
0.600 0.500 0.400 0.900 0.200 0.100 0 0
3
6
9
12
15
18
21
24
27
30
33
36
Months from randomisation Prognostic groups. Index score: 0–10
16–20
26–33
11–15
21–25
34–38
Figure 14.10 Survival according to a prognostic index following treatment with radiotherapy in patients with glioblastoma. (Reprinted from British Journal of Cancer, Volume 64, pp. 769–74, Bleehan et al., 1991, by permission of Churchill Livingstone.)
306 Central nervous system
100 90
Median OS, months 2-year survival
80
HR (95% CI)
RT TMZ 14.6 26%
RT 12.1 10%
0.63 (0.52 – 0.75) P .0001
Patients, %
70 60 50
RT RT TMZ
40 30
RT TMZ
20 10
RT
0 0
6
12
18
24
30
36
42
Time, months N 286 287
Patients at risk, n 240 144 246 174
59 109
23 57
2 27
0 4
Figure 14.11 Outcome of EORTC Trial 26981 comparing survival in patients with glioblastoma treated with surgery and radiotherapy with or without concomitant and adjuvant temozolomide. (See text for details.)
is confirmed, it may provide a mechanism for selecting patients for combined chemo-radiotherapy. When patients relapse, around 30 per cent of those with good prognostic factors may enjoy a further brief improvement with chemotherapy. Single-agent CCNU (or other nitrosourea), PCV and temozolomide can all be given as outpatient treatment and produce about the same level of response with only modest acute toxicity. The results of a trial (BR12) in patients at first relapse, comparing PCV with temozolomide (in those not previously exposed), are awaited.
Oligodendrogliomas Oligodendrogliomas are diffusely infiltrating glial tumours that are believed to arise from oligodendrocytes or their progenitor cells. They account for around 5–10 per cent of gliomas and have a peak age incidence between 30 and 50 years. They arise in white matter, with a predilection for the frontal and parietal lobes. Low-grade oligodendrogliomas (WHO Grade II) are slow-growing tumours and the commonest presentation is with seizure. The imaging characteristics are similar to those of low-grade astrocytoma, but calcification is very common (approximately 50 per cent) (Fig. 14.12). Longstanding tumours in the frontal brain may extend across the corpus callosum to affect the contralateral lobe, even when the tumour is low grade. Macroscopically, the tumours usually appear as grey/pink masses, often well demarcated and with calcium often evident to the naked eye. Microscopically, the tumour comprises abundant uniform, small cells with round nuclei and a fine chromatin pattern – often referred to as ‘chicken wire’. Although the majority of oligodendrogliomas are low grade, a substantial minority are more aggressive. These are anaplastic oligodendrogliomas (WHO Grade III). Although the usual features of increased cellularity, mitotic
Figure 14.12 A CT scan without contrast showing extensive calcification in a large right frontal oligodendroglioma in a 30year-old male presenting with seizure.
activity, pleomorphism, vascular proliferation and necrosis are identified, it remains unclear just which features determine prognosis. Unlike glioblastomas, some patients with aggressive-appearing oligodendrogliomas may have prolonged survival after treatment. Grade for grade, the conventional management of these tumours is much as for astrocytomas, with surgery and radiotherapy as described above. However, it has been conclusively shown that oligodendrogliomas have a better response to chemotherapy51 and that LOH 1p and LOH 19q act as markers for chemo-responsiveness, with the combined loss being most prognostic and predictive of a good outcome11 (see also page 300 under heading ‘the Molecular biology of brain tumours’) . Tumours appear to respond equally well to PCV and to temozolomide, although no formal comparisons have been done. Two studies have conclusively shown that adjuvant chemotherapy does not improve survival in anaplastic oligodendroglioma treated with radiotherapy.52,53** What these studies tell us is that treatment with chemotherapy at relapse is as good as treatment at primary diagnosis. We have no information concerning concomitant treatment. Until further information is available, therefore, oligodendroglioma should be treated with surgery and radiotherapy initially, but most patients should be offered chemotherapy at relapse when response rates are high in 1p19q-positive patients and prolonged survival may be achieved.
Ependymoma Ependymoma is a glioma arising from the ependymal cells that normally line the cerebral ventricles and the central
Individual tumours 307
canal of the spinal cord. The incidence of intracranial tumours is greatest in young children and decreases steadily thereafter. These tumours can arise at any site, usually in association with the ventricular system. They are most common in the posterior fossa, where they present with obstruction or with posterior fossa syndrome. Imaging usually shows a well-circumscribed tumour with variable contrast enhancement in the characteristic location. Differentiation from medulloblastoma may be difficult. Ependymomas are soft, grey/pink tumours that often show their ependymal origin. Histologically, the monotonous cellular background is interrupted by perivascular pseudo-rosettes and ependymal rosettes. Anaplastic tumours may, in addition, show pleomorphism, disorganized cytoarchitecture, increased mitosis and necrosis. Ependymomas spread via the CSF, but the influence of spinal seeding on outcome has probably been over-estimated. Although post-mortem series have shown up to 30 per cent spinal seeding, in life it is detected in around only 10 per cent and is symptomatic in less than 5 per cent.54 Spread is more likely with infratentorial high-grade tumours. All patients should be staged with neuraxis imaging, as for medulloblastoma. Maximal tumour resection should be attempted regardless of site. The role of radiotherapy is more controversial and cannot clearly be determined from published studies. A reasonable policy is to deliver postoperative irradiation (50 Gy in 1.8-Gy fractions) to the tumour site for all incompletely resected low-grade tumours. Where the surgeon believes a macroscopic complete removal of a low-grade tumour has been achieved and this is confirmed on MRI, there is a case for a careful observation policy. Supratentorial high-grade tumours should be treated as glioblastoma and carry a similar prognosis. Whole neuraxis radiotherapy is then reserved for high-grade infratentorial tumours and those for which seeding is already apparent. The 5-year survival for patients with low-grade tumours is greater than 50 per cent, although it is worse for younger children. The 5-year survival for high-grade tumours is almost zero. Treatment failure is most commonly due to local recurrence. Chemotherapy is of minimal value for low-grade or high-grade tumours, both in the newly diagnosed and relapsed settings. Myxopapillary ependymoma is a variant most commonly found in the sacral region. It is of low grade and may often be ‘watched’, even after incomplete excision. The rare subependymoma is a very low-grade lesion that should be treated with surgery alone.
Rare neuro-epithelial tumours
cells that form pseudo-rosettes around the blood vessels. The tumours are often superficial, well circumscribed and amenable to surgical resection. The use of radiation and chemotherapy should follow the guidelines for astrocytomas. PLEOMORPHIC XANTHOASTROCYTOMA
This is usually a cystic and peripherally located tumour of children and young adults. Histologically, it is characterized by a mixture of spindle-like cells and mono-nucleated or multi-nucleated giant cells. There is often intracellular lipid accumulation and the marked presence of reticulin fibres. Although normally a low-grade to very low-grade tumour, best treated with complete excision alone, radiation may be of value for the treatment of more aggressive tumours or those that recur after surgery. SUBEPENDYMAL GIANT-CELL ASTROCYTOMA
This very slow-growing tumour is frequently associated with tuberose sclerosis. Treatment is with surgery or observation only. There is no role for either radiotherapy or chemotherapy in this condition. BRAIN-STEM GLIOMA
Gliomas of all types may arise in the brain stem, most commonly in the pons. Biopsy can be very dangerous and it is acceptable in these cases to treat on the basis of the imaging appearances and clinical features alone. Treatment is usually with radiotherapy (54 Gy for high-grade and 45–50 Gy for low-grade tumours). Studies of hyperfractionation and adjuvant chemotherapy have not improved outcome. The majority of patients have high-grade tumours and their outlook is very poor. Patients with lower-grade gliomas may survive for years after treatment. Chemotherapy can produce a further response in patients with low-grade tumours who relapse after radiotherapy. As well as the familiar prognostic factors, patients with exophytic tumours appear to do better. OPTIC-NERVE GLIOMA
This tumour, which is most common in children and young adults, is usually low grade and very slow growing. Because of its position, it causes devastating symptoms of visual and hormonal disturbance. Management is controversial. Surgery (for unilateral disease), radiotherapy and sometimes chemotherapy have been advocated, but the timing is crucial and some tumours will spontaneously cease to progress. Simple guidelines to management cannot be offered and the reader is referred to more detailed text.55
ASTROBLASTOMA
This unusual, probably astrocytic, tumour is regarded by some as a growth pattern rather than a separate pathological entity. The lesion comprises prominent elongated tumour
GLIOMATOSIS CEREBRI
This is a diffuse glial tumour with varied histological appearances. It diffusely affects the brain, involving multiple lobes
308 Central nervous system
and sometimes extends infratentorially, even into the spine. It may affect people of any age. Imaging appearances, though abnormal, are frequently non-specific. Biopsy may often be non-diagnostic and the diagnosis is often only made at post mortem. Surgery has almost no therapeutic role in this condition. Whole-brain radiotherapy may delay the disease process and is the treatment of choice. However, the outlook is dismal; many patients die within months of presentation. DYSEMBRYOPLASTIC NEURO-EPITHELIAL TUMOUR
This is a benign tumour that arises predominantly in the temporal lobes of children and young adults. Histopathologically, these tumours are characterized by a specific glioneuronal element, often with a nodular component and associated cortical dysplasia. Treatment is by surgical excision; there is no evidence for the value of either radiotherapy or chemotherapy. GANGLIOCYTOMA AND GANGLIOGLIOMA
These tumours comprise neoplastic ganglion cells alone (gangliocytoma) or in association with neoplastic glial cells (ganglioglioma). Most patients are under 30 years of age and the great majority present with seizure. The primary treatment of both tumours is with surgery. Radiotherapy is now usually withheld from these lesions, which are regarded as WHO class I, even if resection is incomplete. However, anaplastic transformation may occur in the glial component, when the outcome becomes much more sinister and treatment more aggressive.
neuronal, astrocytic, ependymal (muscular and melanotic) pathways. They are undifferentiated round-cell tumours, which include medulloblastoma, pineoblastoma and supratentorial PNET. They occur predominantly in children and young adults and have similar morphological and behavioural characteristics. However, recent evidence suggests there are distinct genetic differences that also correlate with prognosis.12 Medulloblastoma has an annual incidence of 0.5/100 000. Seventy per cent of medulloblastomas occur in children under the age of 16 years (peak 7 years) and they are rare after the age of 40. The male preponderance is slightly less than 2:1. By definition, medulloblastomas arise in the posterior fossa, usually in the midline (vermis). They spread by local invasion into the cerebellar hemispheres and rostrally into the fourth ventricle and project into the brain stem. They readily spread via CSF to produce metastases on the leptomeninges of the brain and cord and to involve the ventricular surfaces. Clinically, medulloblastoma presents with a cerebellar syndrome or with raised intracranial pressure due to CSF outflow obstruction. The characteristic imaging appearance is of a solid, often uniformly enhancing mass with a discrete edge (Fig. 14.13). Sometimes the tumour is more irregular, inhomogeneous and occasionally cystic. The tumours may be soft or firm and are frequently haemorrhagic. The typical histological appearance comprises densely packed, small, hyperchromatic cells that have a high nuclear:cytoplasmic ratio and a high mitotic rate. Rosettes are seen in less than 50 per cent of cases. Apoptosis is more commonly seen than necrosis. Cellular
CENTRAL NEUROCYTOMA
This is a rare tumour that arises within the ventricles of young adults. Histological features are of small round cells with neuronal differentiation. Treatment is mainly surgical and the outcome, even after incomplete resection, can be good. However, troublesome recurrence can occur, at which time radiotherapy and even chemotherapy may be of value. CHOROID-PLEXUS TUMOURS
Choroid-plexus papillomas are rare intraventricular tumours derived from the choroid-plexus epithelium. They are predominantly found in children. The malignant choroidplexus carcinoma is even less common. Treatment is surgical and the outcome is highly dependent on the completeness of resection. Radiotherapy is reserved for incompletely removed tumours and malignant lesions. Cerebrospinal fluid spread can occur, even in low-grade lesions.
Medulloblastoma and other PNETs Primitive neuro-ectodermal tumours are a group of highly malignant embryonal tumours that can differentiate along
Figure 14.13 A sagittal MRI of a patient with medulloblastoma. The lesion appears as two masses with discrete edges. One encroaches the fourth ventricle, distorting the brainstem and causing obstruction.
Individual tumours 309
differentiation, usually glial or neuronal, is often a feature. The prognostic significance of this is not established. Immunostaining with synaptophysin and vimentin is typical in medulloblastoma, and staining with GFAP and rhodopsin is also common. The desmoplastic variant is more common in older patients and is generally regarded as carrying a better prognosis. The diagnosis will usually be suspected from the clinical presentation and the imaging appearances. If hydrocephalus is present, this may constitute an emergency and require prompt ventricular drainage and steroids prior to a definitive operative procedure, although some surgeons undertake both the drainage and tumour decompression simultaneously. Full neuraxis imaging (Gd-MRI) should be done preoperatively to look for metastatic spread before blood is introduced into the CSF at operation. If this is not possible, the investigation should be done within 48 hours following surgery. The CSF should be sampled and examined for malignant cells before manipulation of the tumour. The tumour is removed through a posterior fossa craniectomy using either suction or the ultrasound aspirator. Complete tumour removal is associated with a better prognosis. Whether this is due to the resection or to the property of resectability is not clear but, wherever possible, a complete removal should be attempted. In some situations the tumour involves the brain stem and total excision is not possible. A postoperative CT (or MRI) should be done within 48 hours of surgery to assess the amount of residual disease. The prognosis for this disease improved dramatically in the late 1960s with the introduction of whole neuraxis radiotherapy.56 Current practice involves the irradiation of the entire brain and spinal cord with the meningeal coverings. Without adjuvant chemotherapy the standard dose is 35 Gy (in 20 fractions), with a boost of 19.8 Gy (in 11 fractions) being delivered to the posterior fossa. The technique is described on page 295 under ‘Radiotherapy technique’. This remains standard treatment in adults. In children, it has recently been claimed that superior results are obtained in medium-risk patients using platinum-based chemotherapy and reduced-dose radiotherapy (23.4 Gy whole neuraxis plus boost to 55 Gy).57 Although this has not been subjected to formal phase 3 evaluation against ‘standard radiotherapy’ as described above, the regime has been adopted as standard of care for children in many units. A European randomized trial will compare this protocol with a similar regime in which the radiotherapy is hyperfractionated. Patients are treated daily (or twice daily in the hyperfractionated regime). Adequacy of shielding should be checked with portal imaging and in-vivo dose measurements to the eyes. Weekly vincristine is often given as part of ‘standard’ chemotherapy.57 Acute toxicity includes complete alopecia and frequently nausea (and vomiting), which requires anti-emetic. The exit dose from the spinal field may produce a radiation oesophagitis. Leucopenia and thrombocytopenia may occur, but rarely require support or interruption of treatment (even if prior chemotherapy has been used). Later consequences are ongoing nausea
(especially in teenagers), loss of IQ, hormonal deficits, loss of height due to direct and indirect (growth hormone) effects on bone, and cataracts from scattered radiation dose. The ‘Packer’ regime demands further chemotherapy after completion of the radiotherapy comprising cisplatin, vincristine and CCNU. This can prove highly toxic in some children and most adults. The overall 5-year survival is around 60–80 per cent, depending on the prognostic mix, although patients may still relapse even many years after primary treatment. Good prognostic factors are age (infants fare worse), completeness of resection, lack of metastatic disease and desmoplastic histology. Patients who relapse should be treated aggressively with a mixture of chemotherapy, further surgery, focal radiotherapy and high-dose therapy. Durable remissions can be induced using a variety of chemotherapy regimes including VAC, platinum-based combinations, e.g. VCEP,58 and high-dose cyclophosphamide.
Meningiomas A variety of tumours develop in the meninges but the most common by far are meningiomas, which arise from the meningothelial cells themselves. They are usually benign. They occur most commonly on the convexity or falcine brain regions, but may arise anywhere that meninges are present and cause particular difficulties in sites around the base of the skull. Spinal meningiomas are most common in the thoracic region. Meningiomas present most commonly in middle age and occur more often in women (F:M 3:2). They occur rarely in children, when they tend to be more aggressive. They can be induced by radiation in high or low doses, when again they may be more aggressive in character. Some genetic syndromes, particularly neurofibromatosis-2, are associated with the development of meningiomas. A majority of tumours show the presence of progesterone receptor and a minority are oestrogen receptor positive. Whether this is important in the aetiology is not established. Loss of chromosome 22 is a consistent finding in meningioma and is particularly prominent in the atypical form. Most meningiomas are round or lobulated, smooth, firm, well-delineated tumours, often indenting and compressing brain but rarely attached to or invading it. More commonly, they invade into or through the dura and induce hyperostosis in the overlying skull. In the base of skull they may grow as plaque-like tumours. Meningiomas derive their blood supply from the adjacent meningeal artery and are highly vascular. Identification and embolization of this artery can be therapeutic in its own right and can aid surgical removal. A variety of histological variants are identified. These include the commonly recognized meningothelial, fibrous and transitional subtypes, as well as the rarer psammomatous, angiomatous, microcystic forms. In spite of this variety,
310 Central nervous system
their clinical behaviour is similar. All of these are graded as WHO grade I; however, aggressive behaviour may develop in any of them. The WHO identifies an atypical meningioma (WHO Grade II) when areas are present that show increased cellularity with ‘sheet-like’ growth (loss of growth pattern), increasing numbers of mitoses, high nucleus:cytoplasm ratio and geographic necrosis. Malignant meningioma (WHO Grade III) is said to be present when these features become increasingly prominent (very high mitotic rates) and particularly if invasion of the brain or metastasis is identified. It is clear that these definitions are not precise, and they are subject to variation from pathologist to pathologist. The rare chordoid subtype is always regarded as WHO Grade II and the rhabdoid and papillary forms are always graded as Grade III and are associated with a high rate of invasion and recurrence. Presentation in patients with meningioma is diverse. Although headache is most common, seizure and functional deficit are also frequent. Computed tomography imaging usually reveals an isodense to hyperdense lesion, with a meningeal base, which enhances strongly with contrast (Fig. 14.14), although 15 per cent or more may have atypical appearances. Gadolinium-MRI is the most sensitive detection method for meningioma. Angiography may add further diagnostic information and also provides the basis for embolization. Surgery dominates the treatment of meningioma.59 The aim is to remove the entire tumour and any involved adjacent structure (dura, soft tissue and bone) to maximize the prospect for enduring local control. Preoperative embolization of the main feeder vessel is often performed as an aid to surgical removal. The site of the tumour determines the sur-
gical procedure. Convexity meningiomas have the best chance of total removal. Removal of parasagittal tumours risks damage to the sagittal sinus and its draining veins. Tumours of the base of skull are particularly problematic because of the difficult access and the proximity of sensitive structures. A newly diagnosed meningioma is often separated from the brain by an intact layer of arachnoid, which can define a ‘plane of safety’ during the removal procedure. The outcome following surgery depends strongly on the extent of resection. Re-growth varies from less than 10 per cent for patients with a ‘complete’ excision to over 40 per cent for patients undergoing partial resection. It follows that patients with tumours in the most accessible regions (convexity) have better prospects than patients with tumours in difficult areas such as the sphenoid ridge. Second operation for recurrence is possible in the majority of patients. Radiotherapy is increasingly used in the management of meningioma, usually in the adjuvant setting following incomplete resection, or when the histology is unfavourable.60 The GTV will include any imageable residual or recurrent tumour. The CTV will include a margin for spread into adjacent structures, particularly the dura. For malignant tumours, the brain and overlying bone must also be considered at risk. Typically, doses between 50 Gy and 55 Gy in 1.8-Gy fractions are used for grade I lesions, whereas 60 Gy in 30 fractions is recommended where possible for higher-grade tumours. Sophisticated immobilizing, localization and beam-conforming techniques should be used to minimize the irradiated normal tissue, especially in sensitive areas around the skull base. Stereotactic radiotherapy can be used to advantage.60 Stereotactic radiosurgery has also been used to treat meningiomas, but the tendency to dural spread and the damaging effect of large single doses may limit the success of this approach. Chemotherapy is of no proven value in benign meningioma, although sarcoma regimes may be tried for palliation in the malignant form. The identification of hormone receptors in the majority of meningiomas has led to the use of anti-androgen hormone therapy, but with little success. The use of interferon has been advocated by some, but any benefit is minor and short lived. Not all patients with meningioma require immediate surgery. Small lesions may be found serendipitously. Particularly in the elderly, these tumours may be so slow growing as to pose no threat during the patient’s lifetime. They can be managed with follow-up and observation, intervening only if the clinical or imaging situation deteriorates.
Haemangiopericytoma Figure 14.14 A sagittal MRI showing a patient with two meningiomas exhibiting uniform enhancement with gadolinium. The large tumour arising from the skull base and the smaller parasagittal tumour both have discrete edges that distort rather than invade the adjacent brain.
Previously thought of as a variant of meningioma, this tumour is now believed to be indistinguishable from haemangiopericytoma in other sites. It is probably derived from the meningeal capillary pericyte. In comparison to meningiomas, these tumours tend to arise in a younger age group and more commonly in men. They arise in the dura
Individual tumours 311
as highly vascular, lobulated masses. They are densely cellular and often highly mitotic. Their behaviour is sarcomatous, with a marked tendency to recur after surgery alone and to metastasize within and outside the CNS, particularly to bone. Surgery alone is rarely curative and they are best treated with radical excision followed by high-dose radiotherapy (55–60 Gy). Median survival is around 5 years.
Chordoma Chordomas are malignant, embryonal tumours that arise from the notochordal remnant. They arise predominantly in the region of the clivus and at the sacrococcygeous. Rarely, other sites are affected. They normally arise in the extradural space, but grow slowly and may invade the dura as they do so. Histologically, the identifying features are ‘physaliferous’ cells, which contain large mucus-filled vacuoles. They are arranged in lobules and are usually surrounded by extracellular mucus. The management of these tumours is complex. The best possibility for cure probably results from aggressive surgical resection. However, even after apparent complete resection, local recurrence can be a problem. Conventional high-dose radiotherapy is often given in the postoperative setting, but the value of this is not proven. Dose escalation to highly conformed target volumes using proton beam facilities appears to produce better results and, where possible, referral of these rare cases to specialist centres is appropriate. Stereotactically localized, intensity-modulated radiotherapy offers the prospect of similar, linacbased dose escalation.
appear as isodense or hyperdense lesions that enhance uniformly (Fig. 14.15). They are often multiple at presentation (20–60 per cent). Characteristically they respond rapidly to treatment with steroids and may ‘disappear’ within 48 hours of starting treatment, leading to them being described as ‘ghost tumours’. Ocular disease is present in 15–20 per cent of cases. Clinically they usually present as mass lesions, much as high-grade gliomas. Cerebrospinal fluid cytology is positive in about10 per cent. Primary spinal lymphoma is extremely rare. The pathology is much as for systemic extra-nodal lymphomas. They can be firm or friable, well delineated or diffuse. They may be necrotic or haemorrhagic, yellow, grey or indistinguishable from adjacent brain. The great majority of PCNSLs are B-cell and express the usual pan B markers (e.g. CD20). Only 2 per cent are T-cell. The literature is confused with regard to the incidences of the various subtypes, and the systematic classification schemes do not apply well. Large cell types (diffuse and immunoblastic) are more common than small-cell (cleaved and non-cleaved) tumours. Highgrade tumours are more common than low grade. In most series a percentage of tumours cannot be classified. When primary Hodgkin’s disease and plasmacytoma occur in the CNS, they are usually dural based. Both are extremely rare. Whilst imaging may suggest PCNSL, histological proof is essential for adequate management. Response to nonsurgical therapy, including steroids, is usually rapid, and a (stereotactic) biopsy alone is needed. Whether these patients need full lymphoma staging including systemic
Primary CNS lymphoma Primary CNS lymphoma (PCNSL) is defined as lymphoma arising in the CNS in the absence of obvious lymphoma elsewhere at the time of diagnosis. Previous names have included microglioma and reticulum cell sarcoma of the CNS. Its development is strongly related to immunosuppression due to either disease or therapy. Thus there is a high incidence in patients with AIDS, following organ transplantation and possibly in rheumatoid disease. The incidence in both immune-competent and immune-compromised patients appears to be rising. Primary CNS lymphoma may occur at any age but in immune-competent patients is most common in middle age. It is slightly more common in men. Why PCNSL occurs at all is not clear, since the brain lacks a defined lymphatic system. It has been suggested that malignant lymphoma cells, which develop outside the brain, are imported and develop in this relatively immuneprotected site. An alternative idea is that a polyclonal inflammatory lesion in the brain may expand clonally to a neoplastic state. Primary CNS lymphomas arise preferentially in periventricular regions of the brain. On imaging, they usually
Figure 14.15 A contrast-enhanced CT scan showing a patient with a primary central nervous system lymphoma. Note the characteristic features of periventricular location and uniform enhancement. Primary central nervous system lymphoma (PCNSL) is often multifocal and can appear to resolve rapidly with steroids (ghost tumour).
312 Central nervous system
imaging and bone-marrow examination is controversial, as the pick-up for systemic disease is so low. However, there is general agreement that a full ophthalmologic examination, a lumbar puncture (where possible), whole neuraxis MRI and an immunological screen, including an human immunodeficiency virus (HIV) test, are important. In the past, patients with confirmed PCNSL were treated with radiotherapy alone. Survivals were dismal – median 18 months, 5-year disease-free survival 10–20 per cent.61 Modern management comprises primary chemotherapy for most patients, although no randomized trial has shown the supremacy of this approach. The role of additional radiotherapy is unclear. This and many other questions remain unanswered. The chemotherapeutic treatment of PCNSL is dominated by high-dose methotrexate. A dose of at least 3 g/m2 is considered necessary to achieve adequate levels in the CNS, and doses up to 8 g/m2 have been used. Other agents that can produce a response include cytosine arabinoside, procarbazine, temozolomide, the epidophyllotoxins and the nitrosoureas. Various combinations of these agents have been reported, but none has proved consistently superior.62 Complete response rates up to 80 per cent can be achieved. However, relapse is the rule and median survival is only between 2 and 4 years. The value of intrathecal drugs is unproven and their use is diminishing. High-dose chemotherapy with stem-cell rescue has some advocates, but again the procedure has not been shown to improve the results obtained with a more conventional approach. The role of post-chemotherapy irradiation treatment is controversial. Since lymphoma is a radiosensitive tumour and since relapse is predominantly local, it would seem a logical approach. Indeed, combined modality treatment does produce long-term survivors. However, the toxicity of the combination is high, with cognitive decline being very common, and almost universal after age 60, when it is frequently very severe.63 For this reason, patients over the age of 60 are now treated with chemotherapy or radiotherapy alone, with the unused modality reserved as salvage therapy at relapse. For younger patients the decision concerning irradiation is made on an individual basis or within the context of a clinical trial. If radiation is given after chemotherapy, doses are usually kept to 36 Gy in 20 fractions to the whole brain. A boost to the primary site is no longer considered beneficial. Whole neuraxis radiotherapy is of unproven value. The outcome from this approach is unclear. Whereas median survival overall remains only 2–4 years, a substantial minority (up to 30 per cent) may enjoy prolonged survival or even cure. Prognosis is better in younger, fitter patients and in those who obtain an early and complete response to chemotherapy. In 10 per cent of cases relapse outside the CNS will occur. The treatment of HIV-positive patients is even less well defined. If their condition allows it, an aggressive approach, as for the immune-competent patient, can be followed. For less well patients, radiotherapy alone can often be given. It is important to continue antiviral therapy.
Tumours of the pineal region Whereas pineal-cell and germ-cell tumours are particularly associated with this region of the brain, other tumour types, including gliomas, meningiomas, benign tumours and metastases, also occur. Presentation is most commonly with hydrocephalus and is often accompanied by complete or partial Parinaud’s syndrome. TUMOURS OF THE PINEAL PARENCHYMA
These are uncommon tumours of the pineocytes. They always arise in the pineal gland but may disseminate in the CSF. They occur in childhood or, more commonly, in mid-adult life and show no gender preference. On imaging they most commonly show as uniform isodense or hypodense lesions that enhance uniformly and often appear to have a well-defined margin. Cysts or calcifications may be present but are not as prominent as in germ-cell tumours. Two types of pineal tumours are recognized: the well-differentiated pinealoma and the highly malignant pineoblastoma. Pinealomas comprise cells identical to normal pineocytes and contain few mitotic figures. Pineoblastomas contain undifferentiated cells with high mitotic activity and are often indistinguishable from other PNETs. These represent two ends of a continuum and, not infrequently, tumours may have features intermediate between them; precise classification is then difficult. The management of pineal tumours includes imaging of the entire neuraxis, immediate decompression of hydrocephalus and establishing a tissue diagnosis. Pineocytoma is a surgical disease and may be cured by total resection. Pineoblastoma is unlikely to be cured by surgery. It requires whole neuraxis radiotherapy as for medulloblastoma but with the boost directed at the tumour (plus ventricular system, if appropriate). The difficulty lies with ‘intermediate’ tumours. Our general attitude has been to treat these aggressively on the assumption of pineoblastoma. The outcome in these rare tumours is difficult to estimate. Pineocytoma is likely to be cured by surgery. Pineoblastoma has a sinister reputation, although failure to recognize and treat it adequately has probably contributed to this. The outcome if treated aggressively is likely to be similar to that for medulloblastoma. GERM-CELL TUMOURS
These comprise a group of tumours that present a spectrum of pathology indistinguishable histologically from their gonadal counterparts. It includes pure germinoma (seminoma), teratoma, choriocarcinoma, embryonal and yolksac tumours. They are tumours of the young; approximately 90 per cent occur before the age of 20. The male:female ratio is 2:1. They are predominantly midline tumours, most common around the pineal and third ventricle (80 per cent), although suprasellar presentation is not uncommon. Tumours arising elsewhere tend to be non-germinomatous.
Individual tumours 313
There is a marked geographical variation, with a higher incidence in the Far East, particularly Japan. Typically, pure germinomas comprise sheets or lobules of large cells with abundant glycogen-rich cytoplasm with round, centrally placed nuclei. They immunostain with placental alkaline phosphatase (PLAP), usually on the surface membrane. Human chorionic gonadotrophin (hCG)positive syncytiotrophoblastic giant cells may be present and do not represent a major adverse prognostic factor. The presence of alpha-fetoprotein (AFP) staining in a specimen indicates the presence of non-germinomatous tumour and implies treatment as for immature teratoma. Teratomas may be mature or immature. Mature teratomas contain only fully differentiated elements from any or all of the three germinal layers. The presence of incompletely differentiated elements automatically classifies the tumour as immature teratoma. Production of AFP is common. Teratoma with malignant transformation indicates a teratoma in which a malignancy of conventional somatic type (e.g. sarcoma, carcinoma) has arisen. Yolk-sac tumours are characterized by primitive, AFPpositive epithelial cells in a loose myxoid matrix. Embryonal carcinoma may appear similar to germinoma but is distinguished by staining for cytokeratin. Choriocarcinoma is characterized by the presence of cytotrophoblastic elements and syncytiotrophoblastic giant cells. Staining for hCG is a regular feature; PLAP is more variable. Germ-cell tumours may grow locally, producing potentially reversible effects of pressure and obstruction. Local infiltration of adjacent brain with irreversible destruction may also occur. They tend to spread along the ventricular linings and have a marked tendency to disseminate via the CSF to affect other areas of the brain and spinal cord. The incidence of this is variably reported, but is probably around 15 per cent. Systemic spread, either blood borne or via ventriculo-peritoneal shunts, can also occur. Presentation is as for other mass lesions in the pineal region (or other site). The testicular tumour markers, hCG and AFP can often be detected in the serum and CSF of these patients and can be a useful guide to response to treatment. Decay of the marker may be delayed by the presence of cysts, which act as reservoirs. The CSF-to-serum-marker ratio is normally 1. If the ratio is reversed, systemic disease is likely. The detection of serum AFP is diagnostic of teratoma and may obviate the need for biopsy. Gadolinium-enhanced MRI is the imaging modality of choice and is essential for planning surgery. Germinomas tend to be homogeneous and enhance uniformly with contrast. Calcification is common and cystic areas may be present. Teratomas, benign and malignant, are notably heterogeneous with variable signal characteristics and irregular contrast enhancement. The whole neuraxis should be imaged. All patients presenting with suspicion of a germ-cell tumour require serum estimation for AFP and hCG. The CSF levels should also be evaluated. This may be done
following the placement of a shunt, which is commonly required for the relief of hydrocephalus. The management of germ-cell tumours is complex and highly dependent on the precise histology of the lesion. Tissue (or marker) diagnosis is essential. However, the extent of surgery required is a matter of debate. Biopsy of this region using stereotaxy may be possible but is often considered hazardous because of the location of the tumour and the risk of bleeding. An open procedure may be considered safer and also provides the opportunity for therapeutic resection. Ventriculoscopy and biopsy can also be valuable in difficult cases. Surgical excision is usually curative for differentiated teratoma but offers no advantage over biopsy alone in germinoma.64 Surgical resection of non-germinomatous germ-cell tumours may be valuable,65 but because of the increasing success of non-surgical treatment, the risk of complications should be considered in each case. It is difficult to give simple guidelines for the treatment of these diseases. In the past, the standard treatment for localized germinoma has been whole neuraxis radiotherapy (30 Gy) with a boost to the primary lesion (including the adjacent ventricular system) of 15–20 Gy.66 The toxicity from this treatment is similar to that for medulloblastoma. The use of chemotherapy with radiation confined to the tumour only has recently been explored,67 as has radiation alone to the whole brain alone with a tumour boost, relying on salvage treatment (chemotherapy and radiotherapy) for the small minority that relapse. Early results suggest that comparable control can be obtained, possibly with less toxicity. For patients with metastatic disease at the outset, a combined approach with chemotherapy and radiotherapy is required. Patients with malignant non-germinomatous germ-cell tumours require treatment with initial chemotherapy (two to four cycles). Various platinum-based regimes have been used, e.g. BEP (bleomycin, etoposide, platinum), BEC (bleomycin, etoposide, carboplatin).68,69 This should be followed by radiotherapy, at least to the primary site (45–50 Gy) and possibly to the entire neuraxis. The outcome following these approaches is good. Fiveyear, disease-free survivals of 85–100 per cent can be expected in pure germinoma.70 Radiotherapy alone for non-germinomatous germ-cell tumours produces only around 20 per cent long-term survivors, but the introduction of platinum-based chemotherapy has improved this to around 70 per cent or better.68,69 As in gonadal germ-cell tumours, it is not uncommon to image residual abnormality in the tumour site following treatment. In the absence of evidence of progression, this needs observation only. If there is doubt concerning the completeness of treatment, however, surgery may be indicated. The functional outcome following treatment is often good, although the same complications of treatment experienced by medulloblastoma survivors occur. Patients presenting with severe Parinaud’s syndrome often suffer persistence of deficit and require the input of a neuro-ophthalmologist. Patients with suprasellar tumours may suffer long-term endocrine problems.
314 Central nervous system
Craniopharyngioma Craniopharyngioma is a benign epithelial tumour of the sellar region, thought to arise from a remnant of Rathke’s pouch. It is most common in children and young adults. It grows slowly and presents with pressure symptoms, endocrine and visual disturbances. Hydrocephalus may be present. The typical imaging properties of an enhancing solid and cystic lesion, often with calcification in a sellar site, strongly suggest the diagnosis. Two forms are recognized histologically, papillary and adamantinomatous. Both are slow growing, and excite intense gliosis in neighbouring brain. Tongues of tumour also extend into adjacent brain tissue and may be very difficult to remove surgically. Single or multiple cysts are very common and contain a thick, cholesterol-rich fluid. The fluid in the adamantinomatous tumour is often oilier than the papillary type and calcification is much more common. Optimal management of these tumours is controversial. A common approach in the UK is to attempt maximal surgical removal. If this is shown to be complete, routine followup with interval imaging is sufficient. Because of the growth pattern, however, total excision is often not possible and the recurrence rate then is higher.71 In these cases radiotherapy can be given. The volume should be minimized using accurate CT localization and conforming techniques, as a dose of 50–54 Gy in 1.8-Gy fractions is needed. This treatment restores the level of tumour control to that of complete excision.72,73 In very young children, in whom radiotherapy might produce high levels of morbidity, observation may be appropriate, even after partial resection. Follow-up in all cases includes endocrine assessment, neuro-ophthalmology and neuro-psychology as well as tumour surveillance. Treatment of recurrence is difficult, as re-operation is hazardous. Radiotherapy-naive patients can be irradiated and cystic tumours may be treated with radioactive intracavitary yttrium-90, even after prior external beam treatment. Although the 10-year survival is around 90 per cent, fewer patients are free of disease. There is no clear difference in outcome between the histological subtypes.
spinal tissue and growth along nerve roots. Almost any pattern of motor deficit may occur. Combinations of complete and partial, upper and lower motor neuron lesions arise at and below the level of the lesion as the tumour develops. Likewise, sensory deficits may be complex. However, localization of the upper vertebral limit of the symptoms and signs can be a useful guide to identifying the position of the tumour within the spine. It is worth remembering that, in the adult, the spinal cord segmental level is approximately two above the bony vertebral level and that the cord ends in the conus at the level of L1–2. Lesions below this level can only produce lower motor neuron symptoms and signs. Plain spinal X-rays still have an important role in the early evaluation of the patient suspected of having metastatic disease, but the widespread availability of MRI scanning has revolutionized the investigation of spinal tumours (Fig. 14.16). The clear identification of the tumour and its boundaries allows surgeons to plan their operations accurately and radiotherapists to make rational judgements about field size and placement. Surgery is the mainstay of treatment for many low-grade spinal tumours. Good results can be expected from the complete excision of meningiomas, neurofibromas, pilocytic astrocytomas and ependymomas. However, surgical decompression of spinal metastases, fibrillary and anaplastic astrocytomas and high-grade ependymomas should usually be followed by treatment with postoperative radiotherapy. Spinal cord compression due to an intramedullary tumour nearly always requires urgent decompression. When loss of function develops rapidly (hours to days), the prospect for recovery of neurological function is likely to be worse than when cord compression develops slowly (weeks to months).
Spinal tumours A similar spectrum of tumours arises in the spine as in the brain, although the frequency of occurrence is different and some types are absent. Tumours may be grouped according to their site of occurrence into extradural (metastasis, chordoma, sarcoma), intradural–extramedullary (meningioma, neurofibroma) and intramedullary (astrocytomas, ependymomas, benign tumours). For all sites, the commonest presentation is with pain and/or loss of function at and below the level of the lesion. The pain may be exacerbated by straining or coughing and may be felt in the spine itself or in the appropriate root distribution. Tumours cause functional loss by direct pressure, generation of spinal oedema, spinal infarction, invasion into
Figure 14.16 A sagittal MRI showing a patient with an ependymoma of the filum. Note the well-delineated lesion lying between lumbar vertebrae 3 and 4.
Individual tumours 315
Radiotherapy may be given with palliative or radical intent. For palliation, either direct posterior or opposed fields are used to deliver short courses of treatment. Doses of 30 Gy in 10 fractions and 20 Gy in 5 fractions are common in metastatic disease. Higher doses for radical treatments are often delivered using a technique based on a wedged pair of fields (Fig. 14.17). Optimal planning will involve the use of CT. For treatments to low-grade tumours, doses of 45–50 Gy in 1.8-Gy fractions are usual, but for highly malignant tumours, doses up to 60 Gy may be appropriate. Although the risk to the spinal cord from the radiation is increased at these doses, it may not be as high as previously thought, provided megavoltage X-rays are used and the dose per fraction is kept below 2 Gy.29 This risk must be balanced against the risk of under-treating the tumour and the consequences of early re-growth.
EPENDYMOMAS
Ependymomas may arise anywhere in the spine but have a predilection for the lower end around the conus. Ectopic extra-spinal tumours may develop in the pre-sacral region. Most are low grade. A clear operative plane is often found and it may be possible for the surgeon to excise the tumours completely. In this situation follow-up only is required. If complete excision is not obtained, adjuvant radiotherapy is often given, but clear evidence of its value is lacking. For low-grade tumours this would comprise local radiotherapy only. High-grade tumours can rarely be excised completely, and postoperative whole-spine radiotherapy with a local tumour boost should be considered. The variant myxopapillary ependymoma is found almost exclusively in the conus/cauda equina region. It is treated surgically and has a particularly favourable prognosis.
ASTROCYTOMAS
The majority of spinal astrocytomas are low grade, but an important distinction exists here based on the precise histology. If the tumour is a pilocytic astrocytoma, the outcome following surgery is usually good. The value of adjuvant radiotherapy is not clear, although it is frequently given. Diffuse astrocytoma, usually fibrillary, has a less good prospect. It is difficult for the surgeon to find any plane between tumour and normal tissue and there is less argument about the need for additional radiotherapy. Most patients are dead within 5 years.74 Anaplastic astrocytomas and glioblastomas occur rarely in the spine but when they do should be treated as their cerebral counterparts with surgery and radiotherapy. The prognosis is very poor.
Gantry 315 45 wedge 9 15 cm
Metastases The brain is a common site of metastasis for systemic malignancy, which may be found in up to 25 per cent of patients at autopsy.75 However, the incidence varies with tumour type. Metastases from carcinomas of the bronchus, breast, bowel and from melanoma are very common, whereas spread from prostate is almost never seen. They arise most commonly at the junction of cortex and white matter (Fig. 14.18) but cerebellar, dural and lepto-meningeal
Gantry 45 45 wedge 9 15 cm Post
100 L 50
Spinal cord
Lung
R 50
Lung
30
30
ANT
Figure 14.17 A stylized diagram of a wedged pair treatment planned on CT for a spinal tumour. Treatment to intramedullary tumours would use smaller field widths (approximately 4 cm). If doses to vital structures (here lungs) are excessive, the plan can be modified by using a posterior field together with the wedged pair.
Figure 14.18 Gadolinium-enhanced MRI of a patient with multiple small brain metastases. Note their spherical shape, their location at the junction of the grey and white matter and their uniform enhancement.
316 Central nervous system
Table 14.6 Factors affecting outcome in patients with brain metastases Factor Histology Age Performance status Disease-free interval Number of metastases Systemic disease
Good prognosis
Poor prognosis
Adenocarcinoma (especially slow growing, breast, bowel, kidney) Younger Good Long (or initial presentation) Solitary Controlled
Squamous carcinoma, poorly differentiated tumours Older Poor Short Multiple Uncontrolled
metastases are also common. Less commonly, metastases are found in sites such as the pituitary and choroid plexus. Patients with suspected intracranial metastases should undergo a minimum of a CT scan, with and without intravenous contrast. However, MRI will show additional metastases in around 15 per cent of cases and, where available, is the imaging modality of choice. Prognosis in this condition is very variable and depends on a number of factors (Table 14.6), which in turn determine the management approach.76
SOLITARY METASTASIS
Patients with a single metastatic lesion and additional good prognostic factors, particularly controlled systemic disease, should be considered for either excision surgery or stereotactic radiosurgery.77,78** The role of additional, conventional radiation therapy is not clear. Although wholebrain radiation does appear to reduce intracranial relapse and neurological death, it has not been shown to improve overall survival. For patients with good performance status and controlled systemic disease there is an increasing move to offer whole-brain radiation following focal treatment. Doses of 40 Gy in 20 fractions, 30 Gy in 10 fractions and 20 Gy in 5 fractions are used, but for patients with better outlook, the higher dose options may have some advantage. Boost treatments to the tumour bed have gone out of fashion, but the evidence for and against these is lacking. In older patients it may be appropriate to omit the wholebrain radiation or restrict postoperative radiotherapy to the tumour bed only.
MULTIPLE METASTASES
Patients with multiple metastases should be carefully assessed after treatment of their symptoms with steroids. Patients who then have better prognostic features should be offered radiotherapy. For these patients, whole-brain radiotherapy is appropriate. RTOG studies have shown that the regimes 40 Gy in 20 fractions, 30 Gy in 10 fractions
and 20 Gy in 5 fractions are equivalent in both efficacy and morbidity.79 For patients who fall into the poorest prognostic categories, the survival is brief irrespective of therapy (median 6–10 weeks) and it is doubtful that any radiotherapy is justified.76 Chemotherapy may be used to treat cerebral metastases in those tumours known to be chemo-responsive. Therefore this may be first-line treatment in patients with germ-cell tumours, small-cell carcinoma and lymphoma. It can also be considered, usually as second-line treatment, in less sensitive conditions such as breast cancer. The regimes used are those appropriate to the tumour concerned, but where a choice is possible, blood–brain barrier-penetrating agents are selected.
KEY LEARNING POINTS ●
●
●
●
●
●
●
Tumours of the brain are characterized by their histological variety. Accurate histological diagnosis (WHO 2000) (7) and molecular analysis are essential for appropriate management. Surgery and radiotherapy remain the predominant treatments for the majority of tumour types. Surgical and radiotherapeutic techniques must be individualized to the specific tumour. Conventional chemotherapy has an increasing role in the primary treatment of CNS tumours and at relapse. Increased understanding of biological and molecular processes in the development and maintenance of brain tumours is leading to the development of new therapies. Cure can be obtained for some patients with brain tumours (particularly children) and useful extension of life in the majority.
References 317
REFERENCES 1 NICE BoCGDG. Improving Outcomes for People with Brain and Other CNS Tumours. London: National Institute for Health and Clinical Excellence http://guidance.nice.org.uk/ csgbraincns; 2006. 2 Giles GG, Gonzales MF. Epidemiology of brain tumours and factors in prognosis. In: Kay AH, Laws ER (eds) Brain Tumours: an Encyclopaedic Approach. Edinburgh: Churchill Livingstone, 1995, 349–60. 3 Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J (eds). Cancer Incidence in Five Continents. Lyons: IACR Scientific Publications, 1997. 4 Fleury A, Menegoz F, Groscalaude P, et al. Descriptive epidemiology of cerebral gliomas in France. Cancer 1997; 79(6):1195–202. 5 Muir C, Waterhouse J, Mack T, Powell J, Whelan S. Cancer Incidence in Five Continents, Vol 5. Lyons: International Agency for Research on Cancer, 1987. 6 Hepworth S, Schoemaker M, Muir K, Swerdlow A, van Tongeren M, McKinney PA. Mobile phone use and risk of glioma in adults: case-control study. [Miscellaneous]. BMJ 2006; 332(7546):883–7. 7 Kleihues P, Cavenee WK. Pathology and Genetics of the Central Nervous System. Lyon: International Agency for Reasearch on Cancer, 2000. 8 Markopoulos C, Sampalis F, Givalos N, Gogas H. Association of breast cancer with meningioma. Eur J Surg Oncol 1998; 24(4):332–4. 9 Kernohan J, Mabon RF, Svien HJ, Adson AW. Simplified classification of gliomas. Proc Staff Meet Mayo Clin 1949; 24:71–5. 10 Sanson M, Thillet J, Hoang-Xuan K. Molecular changes in gliomas. Curr Opin Oncol 2004; 16(6):607–13. ●11 Cairncross JG, Ueki K, Zlatescu MC, et al. Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 1998; 90(19):1473–9. 12 Pomeroy SL, Tamayo P, Gaasenbeek M, et al. Prediction of central nervous system embryonal tumour outcome based on gene expression. [Letter]. Nature 2002; 415(6870):436–42. 13 Rampling R, Cruickshank G, Lewis AD, Fitzsimmons SA, Workman P. Direct measurement of pO2 distribution and bioreductive enzymes in human malignant brain tumors. Int J Radiat Oncol Biol Phys 1994; 29(3):427–31. 14 Berger MS, Pardos M. Immunotherapy. In: Textbook of Neuro-Oncology. Philadelphia: Elsevier, 2005, 87–92. 15 Kelly PJMD, Daumas-Duport C, David B, et al. Imaging based stereotaxic serial biopsies in untreated intracranial glial neoplasms. J Neurosurg 1987; 66:865–74. 16 Lindsay K, Bone I. Neurology and Neurosurgery Illustrated, 4th edn. Edinburgh: Churchill Livingstone, 2004. 17 Scottish Referral Guidelines for Suspected Cancer. Scottish Office, 2002. http://www.scotland.gov.uk/ Publications/2002/05/14862/5417
18 Jena R, Price SJ, Baker C, et al. Diffusion tensor imaging: possible implications for radiotherapy treatment planning of patients with high-grade glioma. Clin Oncol 2005; 17(8):581–90. 19 Krieger MD, Chandrasoma PT, Zee C-S, Apuzzo MLJ. Role of stereotactic biopsy in the diagnosis and management of brain tumors. Semin Surg Oncol 1997; 14(1):13–25. 20 Stummer W, Pichlmeier U, Meinel T, Wiestler OD, Zanella F, Reulen HJ. Fluorescence-guided surgery with 5-aminolevulinic acid for resection of malignant glioma: a randomised controlled multicentre phase III trial. Lancet Oncol 2006; 7(5):392–401. 21 Gill S, Thomas DG, Warrington AP, Brada M. Relocatable frame for stereotactic external beam radiotherapy. Int J Radiat Oncol Biol Phys 1991; 20:599–603. 22 Gutin PH, Prados MD, Phillips TL, et al. External irradiation followed by an interstitial high activity iodine-125 implant ‘boost’ in the initial treatment of malignant gliomas: NCOG study 6G-82-2. Int J Radiat Oncol Biol Phys 1991; 21:601–6. 23 Kreth FW, Faist M, Warnke PC, Robner R, Volk B, Ostertag CB. Interstitial radiosurgery of low grade gliomas. J Neurosurg 1995; 82:418–29. 24 Sheline GE, Wara WM, Smith V. Therapeutic irradiation and brain injury. Int J Radiat Oncol Biol Phys 1980; 6:1215–28. 25 Rider WD. Radiation damage to the brain: a new syndrome. J Can Assoc Radiol 1963; 14:67–9. 26 Klein M, Heimans JJ, Aaronson NK, et al. Effect of radiotherapy and other treatment-related factors on midterm to long-term cognitive sequelae in low-grade gliomas: a comparative study. Lancet 2002; 360(9343):1361–8. 27 Schultheiss TE, Kun LE, Ang KK, Stephens DV. Radiation response of the central nervous system. Int J Radiat Oncol Biol Phys 1995; 31(5):1093–112. ●28 Marcus RB, Million RR. The incidence of myelitis after irradiation of the cervical spinal cord. Int J Radiat Oncol Biol Phys 1990; 19:3–8. 29 Rampling RP, Symonds RP. Radiation Myelopathy. 1998. Curr Opinion Neurology 11:672–632. 30 Rampling R. Chemotherapy: determining the appropriate treatment. In: Davies EH, Hopkins A., (eds) Improving Care for Patients with Malignant Glioma. London: Royal College of Physicians, 1997, 63–74. *31 Westphal M, Hilt DC, Bortey E, et al. A phase 3 trial of local chemotherapy with biodegradable carmustine (BCNU) wafers (Gliadel wafers) in patients with primary malignant glioma. Neuro-Oncology 2003; 5(2):79–88. 32 Gruber MLMD, Buster WPDO. Temozolomide in combination with irinotecan for treatment of recurrent malignant glioma. Am J Clin Oncol 2004; 27(1):33–8. 33 Levin VA, Wilson CB, Rubenstein L, et al. Adjuvant chemotherapy with BCNU or the combination of CCNU, procarbazine and vincristine following irradiation for glioblastoma multiforme. Proc Annual Meeting Am Assoc Cancer Res 1980; 21:474.
318 Central nervous system
34 Rainov NG, A phase III clinical evaluation of herpes simplex virus type 1 thymidine kinase and ganciclovir gene therapy as an adjuvant to surgical resection and radiation in adults with previously untreated glioblastoma multiforme. Hum Gene Ther 2000; 11:2389–2401. 35 Parney IF, Chang L-J, Farr-Jones MA, Hao C, Smylie M, Petruk KC. Technical hurdles in a pilot clinical trial of combined B7-2 and GM-CSF immunogene therapy for glioblastomas and melanomas. Journal of Neuro Oncology 2006;78(1):71–80. 36 Chumas P, Condon B, Olouch-Olunya D, Griffiths S, Hadley D, Teasdale G. Early changes in peritumorous oedema and contralateral white matter after dexamethasone : a study using proton magnetic resonance spectroscopy. J Neurol Neurosurg Psychiatry 1997; 62(6):590–5. ●37 Scott CB, Scarantino C, Urtasun R, et al. Validation and predictive power of Radiation Therapy Oncology Group (RTOG) recursive partitioning analysis classes for malignant glioma patients: a report using RTOG 90-06. Int J Radiat Oncol Biol Phys 1998; 40(1):51–5. 38 Dirven C, Mooij J, Molenaar W. Cerebellar pilocytic astrocytoma: a treatment protocol based on the analysis of 73 cases and a review of the literature. Childs Nerv Sys 1997; 13(1):17–23. ●39 Pignatti F, Van den Bent M, Curran D, et al. Prognostic factors for survival in adult patients with cerebral lowgrade glioma. J Clin Oncol 2002; 20(8):2076–84. ●40 Karim ABMF, Afra D, Cornu P, et al. Randomized trial on the efficacy of radiotherapy for cerebral low-grade glioma in the adult: European Organization for Research and Treatment of Cancer Study 22845 with the Medical Research Council Study BRO4: an interim analysis. 2002. Int J Radiat Oncol Biol Phys 52:316–324. 41 Karim AB, Maat B, Hatlevoll R, et al. A randomised trial on dose response in radiation therapy of low grade cerebral glioma: European Organisation for Research and Treatment of Cancer (EORTC) Study 22844. Int J Radiat Oncol Biol Phys 1996; 36(3):549–56. 42 Shaw E, Arusell R, Scheithauer B, et al. Prospective randomized trial of low- versus high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report of a North Central Cancer Treatment Group/ Radiation Therapy Oncology Group/Eastern Cooperative Oncology Group study [Report]. J Clin Oncol 2002; 20(9):2267–76. 43 Barker F, Aghi M, Curry W, Carter B. Survival after biopsy or resection of supratentorial lobar glioblastoma multiforme: a population-based study. [Abstract]. Neurosurgery 2004; 55(2):485–6. 44 Walker MD, Alexander E, Hunt WE, et al. Evaluation of BCNU and/or radiotherapy in the treatment of anaplastic gliomas. J Neurosurg 1978; 49:333–43. 45 Wallner KE, Galicich JH, Krol G, Arbit E, Malkim M. Patterns of failure following treatment for glioblastoma multiforme and anaplastic astrocytoma. Int J Radiat Oncol Biol Phys 1989; 16:1405–1409. 46 McAleese J, Stenning S, Ashley S, et al. Hypofractionated radiotherapy for poor prognosis malignant glioma: matched
47
◆48
●49
●50
51
52
53
54
◆55
56
●57
58
59
60
pair survival analysis with MRC controls. Radiother Oncol 2003; 67:177–82. MRC Brain Tumour Working Party. Prognostic factors for high grade malignant gliomas: development of a new prognostic index. J NeuroOncol 1990; 9:47–55. Stewart LA. Chemotherapy in adult high-grade glioma: a systematic review and meta-analysis of individual patient data from 12 randomised trials. Lancet 2002; 359(9311):1011–18. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005; 352(10):987–96. Hegi ME, Diserens A-C, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 2005; 352(10):997–1003. Van den Bent M, Kros J, Heimans J, et al. Response rate and prognostic factors of recurrent oligodendroglioma treated with procarbazine, CCNU, and vincristine chemotherapy. Neurology 1998; 51(4):1140–5. Cairncross G, Berkey B, Shaw E, et al. Phase III trial of chemotherapy plus radiotherapy compared with radiotherapy alone for pure and mixed anaplastic oligodendroglioma: Intergroup Radiation Therapy Oncology Group Trial 9402. J Clin Oncol 2006; 24(18):2707–14. van den Bent MJ, Carpentier AF, Brandes AA, et al. Adjuvant procarbazine, lomustine, and vincristine improves progression-free survival but not overall survival in newly diagnosed anaplastic oligodendrogliomas and oligoastrocytomas: a randomized European Organisation for Research and Treatment of Cancer Phase III Trial. J Clin Oncol 2006; 24(18):2715–22. Nazar GB, Hoffman HJ, Becker LE, et al. Infratentorial ependymomas in childhood: prognostic factors and treatment. J Neurosurg 1990; 72:408–17. Walker D. Recent advances in optic nerve glioma with a focus on the young patient. Curr Opin Neurol 2003; 16(6):657–64. Bloom HJ, Wallace EN, Henk JM. The treatment and prognosis of medulloblastoma in children. Am J Roentgenol 1969; 105:43–62. Packer RJ, Goldwein J, Nicholson HS, et al. Treatment of children with medulloblastomas with reduced-dose craniospinal radiation therapy and adjuvant chemotherapy: a Children’s Cancer Group study. J Clin Oncol 1999; 17(7):2127–36. Gaze MN, Smith DB, Rampling RP, Simpson E, Barrett A. Combination chemotherapy for primitive neuroectodermal and other malignant brain tumours. Clin Oncol 1994; 6(2):110–15. DeMonte F, Ossama A. Meningiomas. In: Kay AH, Laws ER (eds), Brain Tumours: an Encyclopaedic Approach. Edinburgh: Churchill Livingstone, 1995, 675–704. Milker-Zabel S, Zabel A, Schulz-Ertner D, Schlegel W, Wannenmacher M, Debus J. Fractionated stereotactic radiotherapy in patients with benign or atypical intracranial meningioma: long-term experience and prognostic factors. Int J Radiat Oncol Biol Phys 2005; 61(3):809–16.
References 319
61 Hayabuchi N, Shibamoto Y, Onizuka Y, et al. Primary central nervous system lymphomas in Japan: a nationwide survey. Int J Radiat Oncol Biol Phys 1999; 44(2):265–72. 62 Hoang-Xuan KA, Camilleri-Broet SB, Soussain CC. Recent advances in primary CNS lymphoma. Curr Opin Oncol 2004; 16(6):601–6. 63 Abrey L, DeAngelis L, Yahalom J. Long term survival in primary CNS lymphoma. J Clin Oncol 1998; 16(3):859–63. 64 Sawamura Y, De Tribolet N, Ishii N, Abe H. Management of primary intracranial germinomas: diagnostic surgery or radical resection? J Neurosurg 1997; 87(2):262–6. 65 Bruce JN, Connolly S, Stein B. Pineal and germ cell tumours. In: Kaye AH, Laws ER (eds), Brain Tumours: An Encyclopaedic Approach. Edinburgh: Churchill Livingstone, 1995, 725–58. 66 Dearnley DP, A’Hearn RP, Whittaker S, Bloom HCG. Pineal and CNS germ cell tumours: Royal Marsden Hospital experience 1962–87. Int J Radiat Oncol Biol Phys 1990; 18:773–81. 67 Baranzelli MC, Patte C, Bouffet E, et al. Nonmetastatic intracranial germinoma: the experience of the French Society of Pediatric Oncology. Cancer 1997; 80(9):1792–7. 68 Robertson PL, DaRosso RC, Allen JC. Improved prognosis for intracranial nongerminoma germ cell tumors with multimodality therapy. J Neuro-Oncol 1997; 32(1): 71–80. 69 Senan S, Rampling R, Kaye SB. Malignant pineal teratomas: a report on three patients and the case for craniospinal irradiation following chemotherapy. Radiother Oncol 1991; 22(3):209–12. 70 Huh SJ, Shin KH, Kim IL, Ahn YC, Ha SW, Park CI. Radiotherapy for intracranial germinomas. Radiother Oncol 1996; 38(1):19–23.
71 Fahlbusch R, Honegger J, Paulus W, Huk W, Buchfelder M. Surgical treatment of craniopharyngioma: experience with 168 patients. J Neurosurg 1999; 90(2):237–50. 72 Ullrich N, Scott R, Pomeroy S. Craniopharyngioma therapy: long-term effects on hypothalamic function. Neurologist 2005; 11(1):55–60. 73 Habrand J-L, Ganry O, Couanet D, et al. The role of radiation therapy in the management of craniopharyngioma: a 25year experience and review of the literature. Int J Radiat Oncol Biol Phys 1999; 44(2):255–63. 74 Minehan KJ, Shaw EG, Scheithauer BW, Davis DL, Onofrio BM. Spinal cord astrocytoma: pathological and treatment considerations. J Neurosurg 1995; 83(4):590–5. 75 Posner JB, Chernick NL. Intracranial metastases from systemic cancer. Adv Neurol 1978; 19:579–92. ●76 Gaspar LE, Scott C, Murray K, Curran W. Validation of the RTOG recursive partitioning analysis (RPA) classification for brain metastases. 2000. Int J Radiat Oncol Biol Phys 47:1001–1006. 77 Patchell R, Tibbs PA, Walsh J, et al. A randomised trial of surgery in the treatment of single metastases to brain. N Eng J Med 1990; 322:494–500. ●78 Andrews DW, Scott CB, Sperduto PW, et al. Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: Phase III results of the RTOG 9508 randomised trial. Lancet 2005 363(9422): 1665– 72. 79 Borgelt B, Gelber G, Kramer S, et al. The palliation of brain metastases: final results of the first two studies by the Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 1980; 6:1–9.
15 Ocular and adnexal tumours JOHN L. HUNGERFORD AND P. NICHOLAS PLOWMAN
Paediatric tumours Other paediatric tumours Adult tumours
320 327 329
PAEDIATRIC TUMOURS Although rare, malignant tumours arising in the eye and orbit in childhood are of great importance because of their curability. Benign tumours of the retina are of vascular or glial origin. Choroidal haemangiomas may occur as part of the Sturge–Weber syndrome, and retinal haemangiomas as part of the von Hippel–Lindau syndrome. Uncomplicated angiomas require no treatment, but when larger lesions threaten vision, by producing a retinal detachment, circumscribed tumours benefit from radioactive plaque therapy, and diffuse haemangiomas require lens-sparing externalbeam radiotherapy as described for retinoblastoma but at lower doses. In tuberous sclerosis and neurofibromatosis, benign astrocytic tumours may occur in the retina and optic nerve. The retinal lesions rarely require treatment, although large optic-nerve gliomas may require radiotherapy or resection to prevent proptosis or local invasion into the central nervous system (CNS). Very occasionally, juvenile xanthogranuloma involves ocular structures, especially the iris, producing hyphaema and secondary glaucoma. Where topical or systemic steroids fail to control these complications, external-beam radiotherapy to doses as low as 300 cGy may be very effective.
Retinoblastoma Although a rare form of childhood cancer, retinoblastoma (RB) assumes a very important position in clinical oncology because of its high curability without significant loss of vision with modern treatment in early-stage disease, its
Key references References
338 338
high lethality in late-stage disease, its autosomal dominant inheritance (with the lessons that it has taught the subject of genetic oncogenesis), and its tendency to affect both eyes.1,2 Whereas approximately one-third of all cases of RB are genetically determined and have multifocal or bilateral disease, only 10–20 per cent of affected children have a positive family history, implying a high new mutation rate. Survivors of genetically determined RB are at increased risk of developing second non-ocular cancers. Because of its familial tendency, the genetics of predisposition to RB and second tumours, genetic counselling and screening programmes have been the subject of much recent research. Knudson3 proposed that at least two gene changes are required for the development of retinoblastoma. In genetically determined cases, one gene change is already present in the genotype and only one further change needs to occur in the somatic cell to produce RB. In non-genetic cases, two changes must occur in a single somatic cell for RB formation. It is implicit in the Knudson hypothesis that the normal genotype has a suppressive influence on RB induction. The RB gene has been localized to band 13q14. Cavenee et al.4 used recombinant DNA probes that recognize individual chromosome-specific markers, termed restriction fragment length polymorphisms (RFLPs), to probe this chromosomal region. Comparisons between the chromosome-13-associated RFLPs from RB cells and those from the leucocytes of patients revealed that loci that were heterozygous in the normal white cells were often homozygous in the tumour. Cavenee concluded that homozygosity for a mutant allele on chromosome 13 is a likely prerequisite for RB. Furthermore, Cavenee extended his studies to in- utero prediction of RB risk, by the use of chromosome-13-specific RFLPs and
Paediatric tumours 321
esterase D isozymes – the esterase D gene being known to be close to the RB gene at 13q14. By comparing RFLP and isozyme patterns of individual family members over two or more generations, Cavenee and co-workers determined which haplotype pattern was associated with the maternally or paternally derived RB allele. They analysed amniotic fluid cells from second-trimester pregnancies and successfully predicted in four out of five cases whether a fetus at risk from RB would later develop the tumour.4 These preliminary genetic studies are of much wider significance in the field of genetic oncology. Friend et al.5 identified the gene that predisposes to retinoblastoma (RB1) and it was subsequently cloned by Lee et al.6 RB1 was the first human recessive cancer gene ever cloned. It appears to occupy a key position in the cell cycle, possibly regulating the choice between proliferation and differentiation. The gene product, pRB, may exercise control on the proliferation of cells by regulating the transcription of messenger RNA from the DNA of genes that promote growth.7 In London it is now possible to carry out prenatal or perinatal diagnosis of RB1 carrier status through linkage analysis in over 95 per cent of RB families. It is planned to extend this service to sporadic bilateral cases with no previous family history, employing mutation analysis. Meanwhile, non-informative family members at risk of inheriting RB1 are examined regularly under anaesthesia. The first examination takes place at age 2 weeks and the second 2 months later. Thereafter, examinations proceed 3-monthly during the remainder of the first year of life, 4monthly during the second year, and 6-monthly during the third and fourth years. Two further examinations are made, with the child awake, in the fifth year, before examinations can cease. It has long been recognized that survivors of genetic RB are at increased risk of developing second, non-ocular cancers. Estimates of the risk have ranged between 8.4 per cent at 18 years8 and 90 per cent at 30 years.9 It is now clear that inactivation of both copies of RB1 may be responsible for many of these histologically distinct second neoplasms.8 The affected child usually presents under 3 years of age. The mean age of onset of the genetically determined, multifocal type is 1 year and that of the non-genetic, unifocal variant, 2 years. In developed countries, most children present with the tumour confined within the eye and with a white pupil, a squint or, occasionally, an expanded glaucomatous eye. In advanced cases the tumour may spread via the choroid to the bone marrow, or via the optic nerve to the CNS. In developing countries, RB tends to present late, with local extra-ocular extension leading to proptosis and spread to regional lymph nodes. The initial assessment of the child with suspected RB should be carried out by an experienced ophthalmologist, employing binocular indirect ophthalmoscopy with scleral indentation, preferably under general anaesthesia. The diagnosis is a clinical one and RB must be distinguished from simulating lesions that include nematode endophthalmitis
and various forms of retinal detachment. Active RB lesions must be distinguished from non-progressive or spontaneously regressed tumours that are quite common and require no treatment. Intra-ocular procedures to sample a suspected tumour are contraindicated because of the very real risk of disseminating it locally within the orbit, but non-invasive ancillary tests, including ultrasound and computed tomography (CT) scanning, may be helpful by demonstrating a characteristic pattern of calcification.10 The histogenesis of RB has been a matter of dispute11 and the tumour has been variously considered to arise from glial cells and from photoreceptors. Recent immunohistochemical reactivity to a panel of monoclonal antibodies has suggested that RB may arise from an early multipotential cell, so that the resulting tumour cell population is heterogeneous.12 Whatever the cell of origin, there is clinical evidence that the progenitor cells lose the capability of malignant change as they mature and that this process occurs centrifugally in the retina. Thus, new tumours developing later during the risk period arise more peripherally than earlier lesions and the technique described is needed to see these neoplasms arising at the very edge of the retina. Once the diagnosis has been made, systemic staging investigations are probably unnecessary in children with small intra-ocular tumours that can be managed without enucleation.13 Computed tomography prior to surgery is usually recommended when an eye is to be enucleated, mainly to confirm the intra-ocular diagnosis. However, CT cannot detect minor optic nerve invasion. Only very rarely seen on routine CT are ectopic intra-cranial RBs, which arise as a component of multifocal disease in cells within the pineal and suprasellar region, which have a similar embryonic origin to that of the eye.14 These tumours affect approximately 4 per cent of children with genetically determined RB and may be better seen in their early stages by magnetic resonance imaging (MRI).15 They have a poor prognosis, probably due to late detection, but screening for them is impractical because the risk period during which they may develop after the diagnosis of RB extends for some 6 years. Bone marrow and cerebrospinal fluid (CSF) cytology can be deferred until after enucleation, and is only required when there is histological evidence of extensive optic-nerve or choroidal invasion. Technetium bone scans are needed only if there is clinical evidence of bony involvement. There is no satisfactory staging system for the overall management of retinoblastoma. The classification proposed by Reese and Ellsworth16 was designed to predict the prognosis for retaining an eye with RB irrespective of the visual outcome (Box 15.1). It has been used widely for 30 years to compare outcomes of treatments for intra-ocular RB. Pratt17 proposed a classification based on the clinical features at diagnosis and on histology, as a guide to treatment and prognosis. A proposed new international classification is currently under evaluation. The aim of the treatment of RB is to preserve vision where possible, with minimal morbidity and without
322 Ocular and adnexal tumours
Box 15.1 Reese–Ellsworth classification of retinoblastoma16 Group I
Single or multiple tumours less than 4 disc diameters at or behind the equator (1 optic disc diameter 1.5 mm)
Group IIA
Solitary lesion 4–10 disc diameters at or behind the equator
Group IIB
Solitary lesion larger than 10 disc diameters
Group III
Lesions anterior to the equator
Group IVA
Multiple tumours, some larger than 10 disc diameters
Group IVB
Any lesion extending anteriorly beyond the limits of ophthalmoscopy
Group VA
Massive tumours
Group VB
Vitreous seedlings
compromising survival. The choice of treatment depends on whether or not RB is confined to the eye, on the size, number and location of the intra-ocular tumours, and on whether one or both eyes are involved. Most children present with RB confined to the eye and there are now several conservative treatment options. External-beam radiotherapy has been used extensively for large and multifocal tumours, and for tumours of any size near the optic disc and macula, but has been largely supplanted by chemotherapy as the firstline management of such lesions. The best outcomes from the treatment of RB stem from combinations of treatments. Focal treatment is still preferable wherever possible for small tumours located away from the optic disc and macula. When choosing a focal treatment, methods such as cryotherapy and laser, which do not involve radiation, are to be preferred over plaque radiotherapy. Salvage treatment by cryotherapy or laser is unlikely to lead to loss of vision by potentiating the effects of earlier radiotherapy or chemotherapy. By contrast, salvage methods that involve radiation or chemotherapy must be planned very carefully to take account of earlier treatments by these methods if consequent visual loss is to be minimized. When focal treatment fails, many eyes can be salvaged by prompt use of chemotherapy or externalbeam radiotherapy. Conversely, focal treatments are useful in salvaging eyes with new or persistent tumour activity after primary chemotherapy or external radiation. The eye should be removed when RB is too advanced for conservative treatment or when conservation fails. The decision to enucleate depends to some extent on the status of the contralateral eye. Eyes with extra-scleral extension, glaucoma and tumour occupying more than half of the globe are best removed whatever the condition of the fellow eye. Eyes with tumour occupying half the volume of the globe are unlikely to be retained with useful vision and are best enucleated if the contralateral eye is healthy or has minor disease. It is justified, however, to attempt conservative
treatment if the contralateral eye has been lost or has equally serious involvement. In infants, growth of the face is so retarded by whole-eye radiotherapy that it is rarely justified to employ this method when the other eye is reasonably healthy. The survival prospects of these young patients can be maximized by subjecting the enucleated eye to a rigorous histopathological examination, on which is based a decision whether or not to give adjuvant chemotherapy. The histological risk factors predictive of metastatic death following enucleation have been well defined.18 The presence of choroidal invasion significantly increases the risk of metastasis, but only when there is concurrent optic nerve invasion.19 In London, prior to 1985, 83 per cent of 74 children with concurrent major choroidal invasion and retrolaminar optic-nerve invasion died from metastatic disease or direct intra-cranial extension. Subsequently, all children with these risk factors have received systemic adjuvant chemotherapy after enucleation, together with orbital radiotherapy when the optic nerve resection margin was involved. On this regimen there were no deaths in the first 11 children18 and only one death from CNS relapse subsequently. Orbital recurrence may follow extra-ocular extension, optic-nerve involvement to the resection margin, or an injudicious biopsy,20 although orbital relapse was occasionally seen in children with none of these features. Prompt orbital radiotherapy was formerly advised when there was an obvious risk factor, and usually prevented relapse in the orbit. Lately, adjuvant treatment for high-risk histology has dramatically reduced the incidence of orbital recurrence, and radiotherapy is now recommended only for tumour extension to the optic-nerve resection margin. Most orbital recurrences occur within a year of enucleation, and were formerly associated with a poor survival rate even if local tumour control was established by orbital radiotherapy. Chemotherapy has greatly reduced the mortality rate in the few cases of orbital recurrence that still occur. In a prior series of 16 children not receiving chemotherapy, 15 died.21 The one long-term survivor had been treated with simple excision of the recurrent tumour, adjuvant chemotherapy and orbital radiotherapy. In a subsequent report from the same group, all five children with an isolated orbital recurrence were cured with a similar regimen.22 With the advent of adjuvant chemotherapy where appropriate after enucleation, metastatic RB has become a rarity in the UK. Metastases from RB are usually apparent within a year of diagnosis of the ocular primary. Direct spread may take place through the sclera into the orbit or along the optic nerve into the CNS. Haematogenous dissemination may result in spread to bones, bone marrow and brain. Local extension into the lymphatic territory of the eyelids may lead to preauricular and submandibular lymph-node deposits and to subcutaneous deposits around the orbit and in the scalp.23 Initial responses to chemotherapy with vincristine, VM26 and platinum analogues for metastatic RB are usually good but are rarely
Paediatric tumours 323
sustained, and early fatal relapse is common.24 Recently, however, reports have been appearing in the literature of long-term survival following intensive chemotherapy.25,26 Ectopic intra-cranial RB was formerly seen in around 1 in 20 children with the genetic form of the disease.14 Some children with genetic RB develop other midline primitive neuroectodermal tumours, for example in the region of the ethmoid sinuses, and these may be forme fruste of ectopic intra-cranial RB. Like orbital recurrence, however, the incidence of ectopic intra-cranial RB seems to have reduced significantly following the use of chemotherapy, either as a primary or as an adjuvant treatment. This entity presents with symptoms and signs of secondary hydrocephalus or, when arising in the suprasellar region, of diabetes insipidus. Treatment by radiotherapy alone or in combination with chemotherapy may prolong survival, but there have been very few durable remissions reported, with most children dying from spinal metastases. CHEMOTHERAPY
The indications for chemotherapy currently in use at St Bartholomew’s Hospital, London, are summarized in Box 15.2. The JOE regimen of vincristine, etoposide and carboplatin is the most widely used for the primary treatment of intra-ocular disease and for adjuvant treatment for heavy choroidal invasion or retrolaminar optic-nerve invasion after enucleation. It is well tolerated. Six cycles are normally given for the treatment of overt disease and four for adjuvant therapy. In London, when treating intra-ocular RB, the ocular status is reassesed after every second cycle of chemotherapy. Second-line treatment with ifosphamide may be given for orbital and metastatic disease, but is rarely
Box 15.2 Indications for chemotherapy for retinoblastoma Extra-ocular disease Orbital relapse Metastatic disease Ectopic intracranial retinoblastoma Adjuvant therapy following enucleation with adverse histology Retrolaminar invasion of optic nerve with or without extension to the resection margin Extensive choroidal invasion Scleral invasion Anterior chamber involvement Intra-ocular disease Primary treatment of tumour(s) too large, too numerous or too close to the optic disc or macula for focal therapy Salvage therapy for recurrent disease bilaterally or in an only eye
recommended for intra-ocular relapse, which is better managed by enucleation or radiotherapy. Intrathecal methotrexate should be considered for intra-cranial disease and should be combined with whole-neuraxis radiation. Local intra-ocular thiotepa and intra-orbital carboplatin have been tried for local intra-ocular relapse, but sustained local tumour control is unusual after this approach. With the advent of primary chemotherapy for intraocular disease, the concept of ‘chemoreduction’ has been popularized. In this approach, chemotherapy is followed by a second treatment on the assumption that it will be insufficient by itself to destroy the tumour(s) in the eye. Chemotherapy is first given to reduce the size of the tumour. The aim is that a large lesion may then respond better to radiotherapy, or a smaller tumour may be treatable with a lower dose of external radiation, by focal radiation treatment with a radioactive plaque or, preferably, by a non-radiational focal treatment such as laser hyperthermia. This approach is severely limited by the poor visual outcome that results from the additive effects of chemotherapy and radiotherapy on the retina.27 It has been suggested that chemotherapy drug resistance in RB may be due to a chemical pump in the tumour cell wall that can be blocked by the drug cyclosporin. This principle has been applied clinically by combining cyclosporin with chemotherapy. Early results have been encouraging,28 but the method is not yet fully tested. EXTERNAL-BEAM RADIOTHERAPY
This is indicated for RBs that are (post-chemotherapy) too large, too numerous or too close to the optic disc or macula for focal treatment, when vitreous seeding is present, and for tumours that have failed to respond to focal therapy or chemotherapy. Any tumour larger than 13 mm in diameter is best treated by external-beam radiotherapy. Foote et al.29 analysed the dose–response rates and patterns of failure of external-beam radiotherapy for RB. Using 4 MV or 6 MV photons, they recommended a dose of 45 Gy in 1.8-Gy fractions and reported no eyes lost from radiation side effects when treatment was given over 4 weeks in at least 20 fractions. External-beam radiotherapy can be given using an anterior whole-eye field when one eye requires treatment, and by lateral whole-eye fields when both need radiotherapy. Using such techniques, radiation-induced cataract is inevitable and the results of cataract surgery are limited by corneal opacification resulting from eyelid shrinkage, dryness and lagophthalmos, which also result from whole-eye radiotherapy and which are resistant to treatment. The result is usually a poorly sighted, photophobic, painful and unsightly eye. A sophisticated and reproducible radiotherapy technique, developed by Schipper30 at Utrecht, has made it possible to treat the whole retina accurately up to, but not including, the lens. This technique has been developed for routine service at St Bartholomew’s Hospital.31 A contact lens on the cornea of the anaesthetized infant provides a fixed reference point
324 Ocular and adnexal tumours
from which all measurements are made. This point is connected accurately to a linear accelerator, the beam of which has been split to minimize divergence and to create a very sharp cut-off point anteriorly. The distance from the front of the cornea to the back of the lens is determined before treatment by A-scan ultrasound, and the beam edge is placed at this depth with 0.5 mm accuracy. A lateral beam encompasses the retina up to the back of the lens, and may be angled at 40° to avoid the contralateral eye when only one globe is to be treated. The approach is recommended for posteriorly placed tumours without vitreous seeding and without retinal detachment extending to the ora serrata. Tumours not fulfilling these criteria and requiring externalbeam radiotherapy are best treated by the whole-eye approach. Lens-sparing external-beam radiotherapy is particularly suited to smaller lesions located critically on the temporal side of the optic disc (Fig. 15.1). Small tumours located very anteriorly at the edge of the beam may receive a suboptimal dose and are best pretreated by cryotherapy. Response to external-beam radiotherapy is assessed by examination under anaesthesia 4–6 weeks after completion of treatment, whichever approach is utilized. Several patterns
of regression are recognized and experience is necessary for correct interpretation of the results. The relationship of regression pattern after external-beam radiotherapy to recurrence has been analysed.32 Complications of radiotherapy are now less than those that followed poorly fractionated wholeeye treatments (Fig. 15.2). Cataract and dry eye have been successfully eliminated by the lens-sparing technique, but some degree of growth retardation of the middle face is inevitable, although less marked than with the whole-eye method. RADIOACTIVE SCLERAL PLAQUE THERAPY
This is particularly useful for tumours up to 10 mm in diameter located at least 5 mm away from the optic disc and macula. The plaque is temporarily sutured to the sclera precisely over the base of the tumour and removed as soon as it has delivered 40 Gy to the apex of the lesion. Although most plaque therapy has been carried out with high-energy cobalt-60 applicators, these sources cannot be shielded on their external surface and have now been largely replaced by low-energy plaques employing either low-energy X-rays from iodine-12533 or β-rays from ruthenium-106.34 Plaque
Not to scale mm scale Vacuum Isocentre
Adjustable scatterer Collimator
Central axis of linear accelerator beam
(a)
(b)
Figure 15.1 (a) Schematic illustration of lateral beam technique used for lens-sparing external-beam radiotherapy. (b) Child in set-up position in treatment.
Figure 15.2 Patient surviving from poorly fractionated radiotherapy for retinoblastoma. The punched-out square skin portal is most prominent. The other ocular complications (listed in text) are more serious than in the modern series of patients.
Paediatric tumours 325
therapy may be repeated at least once and two plaques may be applied for separate lesions. Radiation doses from plaques are cumulative and multiples of three or more tumours requiring radiation are best managed by an external-beam approach. PHOTOCOAGULATION
Photocoagulation can destroy small RBs confined to the retina and at least 3 mm from the optic disc and macula. It is most applicable to post-equatorial tumours. Most centres have employed the indirect technique described by Höpping35 to destroy the tumour by divorcing it from its blood supply (i.e. infarcting it). Shields et al.36 reported a local tumour control rate of 76 per cent in 45 photocoagulated eyes, and best cure rates were achieved when treating tumours 1.5 mm in diameter or less. HYPERTHERMIA
Photocoagulation has been substantially superseded by hyperthermia applied directly to the tumour using infrared radiation generated by a diode laser. Some clinics have employed argon green laser to achieve the same effect. This technique is now commonly applied as an adjunct to primary chemotherapy for intra-ocular disease under the name chemothermotherapy. The method has not been fully evaluated. It is as yet uncertain whether it increases local tumour control rate compared with chemotherapy alone. Furthermore, it has not been investigated for adverse effects such as macular scarring and vitreous relapse due to direct treatment of the tumour. Both of these side effects have been seen in the clinical application of this technique at St Bartholomew’s Hospital. CRYOTHERAPY
Cryotherapy is useful for the treatment of small tumours37 anterior to the equator. The treatment is usually applied transconjunctivally, using a triple freeze–thaw technique, but posterior lesions may be treated through a small conjunctival incision provided that they are located at least 5 mm from the optic disc and macula. Abramson and Ellsworth37 and Shields et al.38 have reported local tumour control rates of 70 per cent and 79 per cent, respectively. Cryotherapy is not recommended for tumours more than 3.5 mm in diameter.38
Orbital rhabdomyosarcoma Rhabdomyosarcomas (RMSs) are the most common softtissue sarcomas of childhood and are of unknown cause. Over one-third of all new cases present with head and neck primaries. Orbital RMS accounts for one-quarter of all head and neck primary sites39 and this particular primary
RMS has long been recognized as having a later pattern of dissemination than tumours arising at other sites. Early surgical series demonstrated that, following orbital exenteration, up to 18 per cent of patients could remain longterm survivors. Similarly, Sagerman et al.40 found an excellent local control rate of 91 per cent following radical radiotherapy. However, in the late 1960s it was found that combination cytotoxic chemotherapy was extremely powerful against this condition. Following the success of adjunctive chemotherapy in Wilms’ tumour, cytotoxic drugs quickly achieved a place in the standard therapy of all patients with RMS. Furthermore, it became conventional practice to deliver one or two pulses of chemotherapy prior to radiotherapy to the orbit, to achieve cytoreduction and perhaps improve vascularity of the tumour at the time of irradiation. Together, chemotherapy and radiotherapy have improved the prognosis of RMS at all sites and, for patients with primary orbital RMS, orbital exenteration is now reserved as a salvage procedure. Typically, the patient with an orbital RMS is a child aged between 1 and 15 years who presents with a painless orbital mass, a squint or proptosis. It is essential to confirm the diagnosis histopathologically by orbital biopsy. The histological subtype is usually embryonal RMS, although older children may be found to have alveolar RMS, which has an earlier pattern of dissemination. Staging of the disease includes a high-definition orbital CT scan to supplement plain skull X-rays, chest X-ray, bone scan and bone-marrow examination. The majority of patients will have localized disease. Bad prognostic features at presentation include large tumour bulk, bone erosion, alveolar or poorly differentiated histology and metastases. Treatment of orbital rhabdomyosarcoma has evolved over time but there is a present day controversy that is germane to other oncological situations. Whilst both American and European co-operative groups are pursuing “risk adapted” protocols, the European approach is currently to use primary chemotherapy, without local therapy (for those patients achieving a complete clinical response: CR) whereas the American group had believed in routine local therapy, together with (for lower stage patients) less chemotherapy (particularly less alkylating agent chemotherapy). The potential advantages of the European approach are that they will avoid the late effects of orbital radiotherapy in many children who achieve CR; however, the diseasefree survival results are inferior to the American data and the “price” of additional therapy for those who relapse is high, in terms of intensity of salvage therapy. To exemplify: in the American IRS IV study, a subset of patients with embryonal orbital tumours, which had been resected down to microscopic levels, were only treated with actinomycin and vincristine (AV) chemotherapy alone (if microscopically clear margins) or AV plus radiotherapy to the orbit. By contrast, in the third SIOP trial, such patients would have received AV chemotherapy (if completely excised with clear margins) or IVA (i.e. phosphamide added to the AV to the dosage of 9 g/m2 per course – after 6 courses a
326 Ocular and adnexal tumours
F1
COBALT 60 75 cm SSD. 60 Wedges
0
CL
F1 5.5 6.0 cm Wt. 82% L F2 5.5 5.0 cm Wt. 91%
CL F1
F2 47
0
17 MeV e–s 5 cm dia.
L R 106
100
106 50
95
95
25
R
100 100
75
90 50 75
75 90
PIT.
50
25
25 10
B.S. 25
(a)
Transverse section
(b)
Transverse section
Figure 15.3 (a) Wedge-pair photon technique for orbital radiotherapy. (b) Anterior electron portal for orbital radiotherapy.
cumulative dose of 45 g/m2) with no local treatment for complete responders.40a,40b Radiotherapy is delivered by either megavoltage photons or an electron portal. Using a 60° wedge-pair technique (Fig. 15.3a), a dose distribution that well fits the orbit is achieved, but there is forward projection of the isodoses towards the pituitary and hypothalamus and late neuroendocrine sequelae may result. The alternative is a single anterior electron portal (Fig. 15.3b). At first, this second technique is attractive because it is theoretically possible to choose the correct energy to draw up the beam at the apex of the orbit. However, the air spaces in the ethmoid region make dosimetry less certain in the midline. Thus, the large penumbra of an electron beam and the straight antero-posterior alignment of the nasal boundary of the orbit make it essential that the medial border of an anterior electron portal extends to the midline of the patient. Nevertheless, and particularly for anteriorly placed orbital RMS, the anterior electron portal is the favoured technique, whereas wedged photon portals are preferred for extensive orbital disease. If the photon technique is chosen, an ‘eyes open’ treatment position may reduce the keratoconjunctivitis risk by minimizing the bolus effect of the eyelids. The radiation dose prescription varies with bulk of the tumour, age of the patient, histology and presence or absence of bone erosion. For younger children and smallbulk disease, doses of the order of 40 Gy are prescribed in
daily fractions of 1.75–1.8 Gy. For older children and largebulk disease, alveolar histology or bone erosion, a higher dose of 50 Gy or more is chosen, and given in similar daily dose fractions. Ophthalmic follow-up is advisable during and after irradiation, as radiation conjunctivitis and xerophthalmia are, respectively, acute and late complications of high-dose treatments. Acute radiation conjunctivitis is usually inflammatory and treated by topical steroids. Secondary infection will require antibiotic drops. Dry eyes require long-term lubrication with artificial tears during the day and with ointment at night if there is significant nocturnal lagophthalmos. Careful follow-up by CT scanning allows early detection of relapse. Almost without exception, relapse will be manifest within 2 years of treatment, and radical orbital exenteration is the treatment of choice. Following exenteration, we have successfully employed a brachytherapy technique where the margins of the resection were microscopically involved (Fig. 15.4). Similarly, and particularly in patients relapsing off therapy, a very short period of some other chemotherapy, such as doxorubicin, etoposide and cisplatin, may precede surgery in order to test the responsiveness of the tumour for postoperative chemotherapy. However, radical surgery must not be long delayed. Following such a treatment scheme, 16 patients treated at St Bartholomew’s Hospital and the Royal Marsden Hospital,
Other paediatric tumours 327
There is continuing controversy as to whether, for good histology tumours, it may be optimal to omit the alkylating agent (with its fertility and second cancer potential) or the radiation.
OTHER PAEDIATRIC TUMOURS Uveal malignant melanoma Uveal malignant melanoma may occur in the choroid, ciliary body or iris. It is exceedingly uncommon in childhood, although benign naevi may arise in the choroid or iris and increase in size without malignant change. Continued growth and a raised lesion should arouse suspicion of malignancy, and management is as for adult disease.
Medulloepithelioma
Figure 15.4 Lateral skull X-ray to show the position of a six gold-grain implant technique performed by the authors. This girl relapsed locally following chemoradiotherapy for orbital rhabdomyosarcoma. The child underwent orbital exenteration but there was residual disease in the orbital apex despite orbital wall and ethmoid resection. She remains disease-free at 14 years.
London,41 have a 5-year survival of 94 per cent, and the American intergroup RMS study reported 21 of 23 cases relapse free at 2 years.39 With such excellent results, several groups have attempted to reduce the morbidity of therapy. The intergroup RMS studies have suggested that, in goodprognosis cases, two-drug chemotherapy with vincristine and actinomycin D is not significantly inferior to VAC, and the subtraction of cyclophosphamide may reduce the incidence of second tumours and infertility. Similarly, the trend downwards in total radiation dose below 50 Gy in selected cases is also expected to reduce late effects. The late morbidity of patients with orbital RMS treated by chemoradiotherapy has been well studied by Heyn et al.42 The incidence of acute non-lymphocytic leukaemia following the alkylating agent containing VAC chemotherapy is approximately 4 per cent43 and there will be a small incidence of other second tumours following chemoradiation. Following radiotherapy to the orbit there is always some retardation of growth, which is inversely related to age, cataract and a substantial incidence of dry-eye syndrome with photophobia and keratoconjunctivitis. Where the hypothalamus has received the majority of the radiation dose, clinically important deficits in growth hormone secretion have been demonstrated. Particularly in boys, the alkylating agent in VAC may lead to infertility.
This is a rare tumour of the non-pigmented ciliary epithelium, presenting in children and young adults. The tumour is usually benign or of low-grade malignancy, with between one-quarter and two-thirds of medulloepitheliomas containing malignant cells.44,45 Medulloepithelioma has been reported outside the uveal tract in the retina46 and optic nerve,47 but usually arises in the iris and ciliary body. The clinical presentation is very variable. It usually presents as a grey mass arising from the ciliary body and displacing the lens to produce astigmatism, amblyopia and squint. There may be an associated pupillary abnormality, cataract or a painful glaucoma. Rare posterior lesions may produce a white pupil reflex similar to that seen in RB. Extra-scleral extension may lead to proptosis and ultimately a tumour may fungate between the eyelids (Fig. 15.5). Noninvasive investigations, including ultrasound and CT, cannot distinguish between medulloepithelioma and simulating lesions such as amelanotic melanoma, Toxocara granuloma, and leiomyoma. The diagnosis is best established by biopsy, but this approach is contraindicated for a posterior tumour in case it may prove to be an RB. Medulloepitheliomas are friable lesions that are difficult to excise. Small iris tumours may be resectable,48 but most cannot be removed completely.49 The eye should be removed when excision is incomplete and especially when malignant cells are seen. Primary enucleation is probably best for all except the most localized medulloepitheliomas.49 Malignant medulloepitheliomas show a tendency to invade the orbit. The overall mortality rate is only 10 per cent, and the presence of extra-scleral extension is the most important adverse prognostic indicator.45 In fatal cases, death commonly results from invasive local recurrence with intracranial extension. Metastases to regional lymph nodes and occasionally to distant sites may sometimes be seen late in the course of the disease in neglected cases.
328 Ocular and adnexal tumours
(a)
(b)
Figure 15.5 Child presenting with fungating medulloepithelioma: (a) before and (b) immediately after radiotherapy.
Optic-nerve glioma Glioma of the optic nerve is a rare tumour, now classified as a juvenile pilocytic astrocytoma, which may occur anywhere in the optic nerve or chiasm. Most examples are histologically benign, but inexorable growth of some intra-cranial lesions may lead to death. Other astrocytomas grow episodically or cease to grow without treatment. Nine out of 10 of these tumours present in the first two decades50 and there is evidence of neurofibromatosis in nearly 40 per cent of affected individuals.51 The clinical presentation depends on whether the tumour is mainly intra-orbital or intra-cranial.52 The earliest sign is often loss of vision, called to attention by a squint, nystagmus or a dilated pupil. Involvement of the intra-orbital optic nerve results in painless axial proptosis, and optic disc swelling or atrophy may be visible ophthalmoscopically. Chiasmal gliomas present with hydrocephalus, diabetes insipidus, somnolence, seizures, obesity or hypogonadism. The tumour is best seen on CT scanning, although intra-cranial extension may lead to enlargement of the optic foramen on plain X-ray and chiasmal involvement may be detected when visual field testing shows a quadrantic or hemianopic defect in the contralateral eye.
Treatment is controversial. The tendency to slow or episodic growth and to spontaneous remission has made it difficult to assess the results of policies of no treatment, surgical excision or radiotherapy. Intra-orbital lesions with good vision and not involving the chiasm may be observed for evidence of growth before considering excision, although gross proptosis in a blind eye is an indication for surgical removal of the tumour via a lateral orbitotomy. Some have advocated a combined orbital and neurosurgical approach if excision is incomplete or if the optic foramen is enlarged, and the weight of evidence from very long-term follow-up studies suggests that the death rate is high if an attempt is not made to remove these tumours surgically.53 Nevertheless, there is now a body of opinion that only tumours not involving the chiasm are resectable. In young children (particularly under the age of three years) it is now appreciated that chemotherapy with (usually) vincristine and carboplatin can often cause good regressions of this disease and spare the child from radiotherapy to the brain at a vulnerable young age. Chemotherapy has therefore moved up to front line therapy in this situation. Carefully fractionated external-beam radiotherapy to a dose of 40–50 Gy may be considered for progressive chiasmal lesions. Although some have claimed that improvement after radiotherapy is
Adult tumours 329
temporary,54 there is evidence55,56 that good regression can be achieved and that treatment should be advised as soon as the chiasm is involved and not delayed until there are symptoms of this involvement. The overall mortality rate ranges up to 56 per cent for tumours with chiasmal involvement.57
Histiocytosis X Histiocytosis X usually has its onset in childhood, and histiocytes accumulate throughout the body in soft tissues and bone. Bony involvement is seen more often in Hand–Schüller–Christian than in Letterer–Siwe disease. The membrane bones of the skull and sella turcica are preferentially affected. Orbital bone involvement is relatively infrequent58 and leads to non-axial proptosis, whereas a soft-tissue deposit within the optic nerve sheath may lead to axial forward displacement of the eye. Proptosis, bone deposits and diabetes insipidus constitute a triad of signs that suggest the diagnosis of histiocytosis X, which can be confirmed by radiological examination and biopsy. Before treatment is commenced the patient should be staged to assess the full extent of the disease. Multifocal involvement requires systemic therapy, which may begin with prednisolone alone. Cytotoxic drugs such as vincristine and vinblastine may be added and etoposide may be considered for resistant disease. Bony deposits respond well to intra-lesional injection of methyl sodium succinate, which is useful for accessible orbital deposits. The injection may need to be repeated several times before regression occurs.58 Lesions resistant to this treatment and inaccessible deposits causing opticnerve compression or severe proptosis may be treated with low-dose radiotherapy. Carefully planned external-beam radiotherapy to a dose of 12 Gy in eight fractions is followed by good regression with little morbidity.59
Secondary deposits Secondary deposits may be seen in the eye and orbit in children. Choroidal involvement is fairly common in uncontrolled acute lymphoblastic leukaemia (ALL) but, with the adequacy of modern systemic treatments and cranial radiation prophylaxis, the ocular relapse rate and pattern have both changed. We have encountered isolated anterior chamber relapses in children in bone-marrow remission, usually soon after stopping maintenance chemo-therapy, and it may be speculated that the radiation dose received by the back of the globe during cranial radiation prophylaxis is cytocidal and that chemotherapy is suppressive only with regard to the poor drug penetration to the anterior chamber sanctuary. Leukaemic infiltration of the bones and soft tissue of the orbit by green-coloured granulocytic tumour is seen in acute myeloblastic leukaemia (AML) and has led to the term chloroma. Occasionally, intra-ocular relapse is seen in AML. Both orbital and ocular ALL and AML respond well to radiotherapy to a dose of 24 Gy given in 12 fractions over 15 days. Unlike adult non-Hodgkin’s lymphoma, involvement of the
Figure 15.6 The ‘black eyes’ of orbital involvement by metastatic neuroblastoma.
orbit by lymphoma in childhood is usually associated with disseminated disease, and the emphasis of management is on chemotherapy. The orbit is involved in 50 per cent of cases of the African form of Burkitt’s lymphoma, usually by secondary extension from a maxillary primary. Neuroblastoma preferentially metastasizes to the orbits, often bilaterally (Fig. 15.6), and very occasionally a primary orbital neuroblastoma may arise in the ciliary ganglion. Local radiotherapy to a dose of 25 Gy in 13 fractions must be employed promptly if there is any threat to vision, as the tumour is reliably radiosensitive and unreliably chemosensitive.
ADULT TUMOURS Tumours of the eye and of the ocular adnexa are much more common in adults than in children. Apart from choroidal naevi and benign papillomas of the eyelids, the mostfrequently encountered neoplasms are malignant. In contrast to those of children, many adult intra-ocular malignant neoplasms are metastatic, and for most affected individuals, treatment can only be palliative. For adult primary malignant intra-ocular tumours the overall cure rates remain relatively low, although a trend towards earlier detection during routine eye testing has produced some improvement in outlook. Unlike intra-ocular malignancies, the majority of adult adnexal neoplasms are primary tumours. This contrasts once again with the situation in children, where adnexal malignancies are commonly metastatic or related to a blood dyscrasia. Orbital malignant tumours now have relatively good cure rates in the adult because most are only locally invasive. Death from such a neoplasm usually results from direct local extension rather than from metastatic spread. The outlook is similarly good for most malignant tumours of the eyelids and conjunctiva, because they seem to have a lower propensity to regional and distant spread than their equivalents in other structures. Benign tumours are common in the eye and its adnexa. The importance of these tumours is that many are associated with widespread diseases, some of which are heritable,
330 Ocular and adnexal tumours
and that, because of their location, complex treatment may be required to preserve vision or to reduce cosmetic deficit.
Benign tumours of the eye The more common benign intra-ocular tumours in adults are of melanocytic, smooth-muscle or vascular origin. MELANOCYTIC TUMOURS
Melanocytic tumours are overwhelmingly the most numerous. Naevi arise from uveal melanocytes in the iris, ciliary body and choroid. There is no doubt that naevi may occasionally undergo malignant change and that most uveal-tract malignant melanomas arise in this way. Choroidal naevi are extremely common, but the chance of malignant change has been estimated to be less than 1 in 500 during a 5-year period.60 It is impractical to propose regular follow-up for all innocent, flat naevi. Nevertheless, repeated examinations are advised for all iris naevi and for raised choroidal lesions, those developing orange lipofuscin pigment, those with an associated serous retinal detachment or visual field defect, and multiple naevi. Patients with ocular or oculodermal melanocytosis also have an increased risk of uveal melanoma61 and should undergo regular examination of the ocular fundi. SMOOTH-MUSCLE TUMOURS
Smooth-muscle tumours are represented by leiomyomas, which are slow-growing, pale, benign tumours. Leiomyomas are rare in the eye and have been described most frequently in the iris, with a few cases reported in the ciliary body.62 Re-evaluation of archival histological material suggests, however, that most so-called iris leiomyomas are really amelanotic benign melanocytic lesions.63 The only true ocular leiomyomas arise in the ciliary body, where they may grow quite large and be difficult to distinguish clinically from amelanotic melanomas. Ciliary-body leiomyomas are very circumscribed tumours that are easily excised from within the eye, the main indication for this procedure being exclusion of the alternative diagnosis of malignant melanoma. VASCULAR TUMOURS
Vascular tumours may occur in the retina but are more common in the choroid. Most retinal lesions are capillary haemangiomas, which may be solitary or multiple. New retinal haemangiomas may arise throughout life in the von Hippel–Lindau syndrome. They tend to grow and may ultimately lead to bilateral blindness from retinal detachment. Large capillary haemangiomas may respond to cryotherapy,64 but the best response rates are obtained by treating smaller lesions by cryotherapy or laser photocoagulation.65 For this reason, a policy of annual ocular assessment of von
Hippel–Lindau patients from infancy is wise, with early treatment of all new tumours. Cavernous haemangiomas are less common in the retina than tumours of the capillary type. The lesions are usually solitary, although the ocular tumour may be associated with skin or CNS vascular lesions as part of an inherited neurooculocutaneous syndrome, of which one bilateral case has been reported.66 The cavernous type of haemangioma rarely grows and is unlikely to lead to loss of vision unless it involves the macular area or unless it bleeds. Haemorrhage from one of these tumours may be treated by photocoagulation or by cryotherapy. Haemangiomas affecting the choroid are of cavernous type, though their histology and clinical behaviour are different from those of cavernous haemangiomas occurring at other sites. They are of two types. Most are circumscribed and first detected in adult life. Circumscribed choroidal haemangiomas are not associated with systemic abnormalities. A minority of haemangiomas are diffuse. Diffuse haemangiomas commonly present in children and young adults and are usually associated with vascular abnormalities of the skin and CNS, often as part of the Sturge–Weber syndrome. Choroidal haemangiomas grow slowly, if at all. Nevertheless, they may threaten vision by producing a localized or extensive serous retinal detachment. Serous exudation may be reduced by argon laser grid photocoagulation with retinal reattachment, but usually without destruction of the tumour. Should detachment persist or recur, sustained retinal reattachment and good tumour regression may be achieved by irradiating a circumscribed choroidal haemangioma to a dose of only 40 Gy to the apex of the lesion, using a radioactive scleral plaque. Diffuse choroidal haemangiomas commonly produce extensive, blinding serous retinal detachment. These tumours are usually too extensive to respond to photocoagulation or to focal radiotherapy using a scleral plaque. Sustained retinal reattachment has been achieved in diffuse haemangioma by using fractionated external-beam radiotherapy to a dose of 12 Gy, using the lens-sparing approach advocated for retinoblastoma,67 although usually without significant regression of the tumour. Recently, photodynamic therapy using a photosensitizing drug, verteporfin, in combination with focal laser application has been used to treat both circumscribed68 and diffuse69 choroidal haemangiomas with encouraging short-term results.
Benign tumours of the eyelids, conjunctiva and orbit Benign tumours of the ocular adnexa are common and are of epithelial, vascular, neurogenic, melanocytic or lacrimalgland origin. Benign orbital tumours must be distinguished from idiopathic inflammatory pseudotumour.70 EPITHELIAL TUMOURS
Epithelial tumours mainly comprise benign papillomas of the eyelid margin, which are extremely common and of viral origin. They are easily treated by shaving and light
Adult tumours 331
cautery to the base, but tend to recur. Recurrent tumours may be managed by wedge excision. Seborrhoeic keratosis is particularly common in the elderly and may be treated by curettage. Cryotherapy may be performed for tumours of this type in pale-skinned individuals, but should be avoided in those with darker colouring because it is likely to be followed by noticeable depigmentation. VASCULAR TUMOURS
Vascular tumours predominantly affect the orbit. In the adult, cavernous haemangioma is one of the most common primary orbital tumours. It usually presents in middle-aged women and is best managed by surgical resection via a lateral orbitotomy. PERIPHERAL NERVE TUMOURS
Peripheral nerve tumours are mostly neurofibromas. Most orbital neurofibromas in adults are solitary and not associated with neurofibromatosis. Recurrence after surgical excision is unusual. Schwannomas are much less common and are encapsulated tumours that rarely recur after surgical excision. MELANOCYTIC TUMOURS
Melanocytic tumours are mainly naevocellular naevi, which are common in the skin of the eyelids and must be distinguished from pigmented basal-cell and squamouscell lesions. Occasionally, a divided naevus is encountered that occupies the adjacent margins of both the upper and lower eyelids and that has arisen from a single naevus by separation of the eyelids in embryonic life.71 Naevocellular naevi are best managed by infrequent observation. In oculodermal melanocytosis (naevus of Ota) there is widespread infiltration of the eyelid skin and of all ocular structures with plump melanocytes. The condition is most common in individuals of Asian and African extraction. Malignant change does not occur in the cutaneous element of naevus of Ota, but as well as being associated with an increased incidence of intra-ocular melanoma, it may very occasionally be linked with the development of a primary orbital melanoma. Benign conjunctival naevi are frequently cystic. They are easily excised and, bearing in mind that malignant change can occur, this option should be considered as an alternative to long-term follow-up. LACRIMAL-GLAND TUMOURS
These are usually of benign mixed-cell histology and are encapsulated. The whole affected lacrimal gland should be excised with the capsule intact via a lateral orbitotomy.72 Incisional biopsy is contraindicated because recurrence is common if the capsule is ruptured and there is a significant chance of malignant change in the recurrent tumour.
Primary malignant tumours of the eye By contrast with those of children, the vast majority of adult intra-ocular malignancies arise in the uveal tract, and primary malignancies in the retina are very rare indeed. MALIGNANT TUMOURS OF THE UVEAL TRACT
Malignant tumours of the uveal tract are mainly malignant melanomas. Uveal melanoma is the most common of all primary intra-ocular tumours. The neoplasm is most commonly encountered in the choroid but also arises in the ciliary body and occasionally in the iris. This tumour was formerly managed by enucleation of the eye. The survival rate following enucleation has been shown to be dependent on the size of the tumour, on the histological cell type, on its position within the eye, on the presence or absence of extra-scleral extension and on the age of the patient. Large melanomas, ciliary-body melanomas and melanomas containing epithelioid cells or extending extra-sclerally have a relatively bad prognosis, particularly in the elderly. For large tumours, the overall mortality rate approaches 50 per cent73 and the mean interval reported from treatment to the development of metastases is 43 months.74 Metastases are rarely detectable at the time of diagnosis of the ocular primary, and it has been suggested that enucleation of the eye may in some way potentiate the development of widespread melanoma.75 This view is no longer widely held, but it has served to foster conservative management of ocular melanoma, and the majority of eyes are no longer removed as the primary treatment of this tumour. The treatment of choice depends mainly on the location of the tumour within the eye and on its size.76,77 The diagnosis of an intra-ocular tumour is usually based entirely on clinical criteria, with an accuracy of almost 98 per cent,78 but fine-needle aspiration and cytology or formal biopsy through a scleral trapdoor incision may be considered in cases of doubt that cannot be resolved by a period of observation with serial photographs and ultrasound measurements of tumour size, looking for evidence of growth. Iris melanomas are generally relatively benign spindlecell lesions and overall mortality rates appear to be as low as 3 per cent. Most iris tumours can be managed by observation alone, with treatment reserved for those that are documented to grow or that are large at presentation. By contrast with their counterparts in the iris, melanomas of the choroid, and particularly those that develop in the ciliary body, are associated with particularly poor survival rates, of around 50 per cent at 5 years for larger lesions. Three main methods of conservative treatment are available: photocoagulation, radiation and resection. Photocoagulation is applicable only to very small melanomas situated in the posterior choroid outside the macular arcades of vessels and not directly abutting the optic disc. Most of the reported experience has been obtained using xenon arc photocoagulation,79 but lately argon and krypton laser have
332 Ocular and adnexal tumours
been employed and the role of photosensitization with haematoporphyrin derivatives is being explored.80 More recently still, hyperthermia using diode laser-generated infra-red radiation, so-called transpupillary thermotherapy, has been employed with good short-term results.81,82 Photocoagulation and hyperthermia are not applicable to melanomas anteriorly located in the iris and ciliary body. By contrast, local surgical resection is technically possible for some intra-ocular melanomas. Although easier to perform on small lesions arising anteriorly in the iris and ciliary body, the complication rate is higher than for tumours that develop posteriorly in the choroid and to which the method is most applicable.83 The technique is limited to young, otherwise fit patients by the need for anaesthesia, with profound hypotension while cutting the highly vascular choroid. Tumours up to 15 mm in diameter may be excised and overlying retinal detachment facilitates this operation. Local resection alone has a lower tumour control rate than plaque radiotherapy alone, so it is now advised that this operation should always be combined with adjuvant plaque therapy.84 Radiotherapy has been by far the most widely used treatment modality for the conservative therapy of intra-ocular melanomas and may be applicable to tumours in all locations within the uveal tract.85 Melanoma is a radioresistant neoplasm, whereas the lens of the eye and the choroid, retina and optic nerve are relatively radiosensitive structures. A radiation cataract can be safely removed, but radiation choroidoretinopathy and optic neuropathy are not amenable to treatment. The risk of these side effects increases at doses to the whole eye in excess of 50 Gy, whereas a tumour dose of between 80 Gy and 120 Gy is required to regress uveal melanoma. The treatment methods employed have depended on the difficulty of confining this high dose to the vicinity of the ocular tumour and thereby of limiting the dose to the adjacent retina, choroid and optic nerve. Until recently, it has only been possible to preserve the healthy structures of the eye by brachytherapy techniques, employing radioactive applicators temporarily sutured to the surface of the sclera over the base of the tumour within the eye. After initial successes with radon,86 the most widely used source was the cobalt-60 applicator, because the conveniently long half-life of this isotope allows the source to be re-used many times and because it is simple to reactivate.87–89 The gamma-rays emitted by cobalt-60 are of high energy and the external surface of the applicator cannot be effectively shielded. More recently, the low-energy beta-ray-emitting ruthenium-106 and X-ray-emitting iodine-125 plaques used for RB treatments (Fig. 15.7) have been employed to treat ocular melanomas up to 5mm or 8mm thick, respectively.90,91 These plaques can be shielded, which allows them to be used to treat anterior melanomas without damage to the eyelids and lacrimal apparatus, as well as making them safer to handle. A dose of 80–120Gy is prescribed at the apex of the tumour at a dose rate of at least 0.4Gy per hour. Radiation-induced maculopathy and optic neuropathy limit
Figure 15.7 Iodine-125 ophthalmic applicator currently in use at St Bartholomew’s Hospital. Iodine-125 seeds are embedded in epoxy resin in a gold carrier.
Tumour 15 000 10 000 8000 6000 4000 2000
20 000 0 Design 5 mm depth 10 mm Retina
0
5 mm
15 mm 10 mm Sclera
Dose distribution for CKA.3 (in cGy/6 days)
Figure 15.8 applicator.
Dose distribution plot of 10 mm cobalt-60
the visual results of plaque therapy for tumours near the macula or optic disc. Because of the effects of the inverse square law, the dose received by the base of a neoplasm escalates substantially with increasing tumour thickness. Furthermore, the higher the dose at the base of the tumour, the larger is the area of surrounding choroid and retina that receives a dose in excess of the 50Gy above which there is a significant risk of ischaemic damage to normal ocular structures (Fig. 15.8). Over the last two decades, it has become possible to avoid this important limitation of radioactive scleral plaques. The special properties of positively charged particles give them an advantageous dose distribution. Not only is the dose uniform from base to apex of the tumour, but the Bragg peak effect means that the entry dose is reduced and that the beam is extinguished behind the treated volume (Fig. 15.9). Employing protons92,93 or helium ions94 generated by a cyclotron, these properties have allowed a high dose prescription of 60–70 Gy in four or five fractions to be delivered to ocular melanomas over 4–5 days, with the 50 per cent isodose occurring within 2 mm of the tumour edge and with successful tumour regression.
Adult tumours 333
8-SEP-84 8411210 RIGHT EYE PLANE WITHIN THE EYE FIXATION LIGHT: POLAR 45 AZIMUTHAL 45 TWIST 0 EYE CENTER IS AT 0.50 0.40 0.00 DOSE 90 50 20
Figure 15.10 Enucleated eye, showing an encapsulated modular extrascleral extension of malignant melanoma.
APERTURE IS: AP2 RANGE: 1.60 RANGE MODULATION: 1.60 PLANE POSITION PARALLEL TO BEAM X- 0.7 Y-0.0 Z-0.0
Figure 15.9 Computer-constructed plan for treatment by proton beam of an eye containing a malignant melanoma; 100%, 90% and 50% isodose plots are shown and demonstrate the Bragg peak effect.
Survival rates have been comparable with those following enucleation.95 Although charged particles can be used to treat melanomas that are too close to the optic disc to apply a radioactive scleral plaque or too thick for plaque brachytherapy, the surface-sparing advantages of the Bragg peak are progressively lost with increasing modulation of beam to treat thicker and thicker tumours and with anterior location. Consequently, eyelid damage limits the benefits of charged particle therapy for larger and more anteriorly located melanomas. Moreover, the treatment of larger tumours is not uncommonly followed by painful neovascular glaucoma, sometimes requiring enucleation.96 Gamma-knife stereotactic radiosurgery can also be used for the treatment of posterior uveal-tract melanomas.97 The treatment is delivered in one or two fractions and this plainly contributes to the high rate of side effects, which are not dissimilar to those of charged-particle therapy. Nevertheless, easier access to this method compared with charged particles has lent it a certain popularity. When appropriately selected for radiotherapy, more than 90 per cent of eyes can be retained.98 The criterion of treatment success is shrinkage of the tumour or cessation of growth. Regression is slow and can continue episodically for 2 years or more. Melanomas less than 5 mm thick will
often regress to a flat scar, but larger tumours will usually have some residual thickness when regression ceases. Enucleation is advised for large melanomas measuring more than 1.5 cm3 in volume. Ciliary-body melanomas occupying more than one-third of the circumference of the globe are too large to manage by resection or plaque therapy and are also best treated by enucleation. Circumscribed nodular extra-scleral extension may be managed by plaque therapy or by resection with a scleral graft, but extensive extra-scleral spread is better treated by removal of the eye. The presence of extra-scleral extension (Fig. 15.10) reduces the survival rate and significantly increases the risk of orbital recurrence.99 Orbital exenteration following enucleation for extra-scleral extension confers no survival advantage, and the risk of orbital recurrence can be substantially reduced without mutilation by postoperative orbital radiotherapy.100
MALIGNANT TUMOURS OF THE RETINA
These are exceedingly uncommon. Cytologically malignant adenocarcinomas of the retinal pigment epithelium have been reported,70 although these lesions rarely, if ever, metastasize. They are pigmented and are usually a chance finding when an eye is removed for what is thought to be a malignant melanoma.101 For this reason there are no data related to whether or not they can be managed conservatively.
OCULAR LYMPHOMA
Primary ocular lymphoma is a completely different disease from orbital lymphoma. Ocular lymphoma is a high-grade, usually B-cell lymphoma with a strong predisposition to occur synchronously or metachronously in the brain and both eyes. The brain involvement is as solid mass disease (exactly as primary cerebral lymphoma – same disease) and
334 Ocular and adnexal tumours
not leptomeningeal disease. This is the primary “oculocerebral” lymphoma; it usually presents to the ophthalmologist with involvement of the retina and vitreous and the symptoms and signs can easily be confused with those of a posterior uveitis or vitritis, often delaying the diagnosis. As there is no lymphatic connection from one orbit to the other, the bilateral ocular predisposition is of interest. It is worth noting the (typically) different clinical picture when systemic lymphoma involves the eye. Here, the choroid is mainly involved – the lesions appearing as solid or well-defined creamy yellow choroidal lesions with a clear vitreous. With disease progression, both presentations merge and a systemic work-up is always required. Staging of primary ocular lymphoma requires expert binocular ophthalmoscopy and high Tesla MR imaging of the brain. Treatment requires CNS and ocular penetrating drugs; the Barts: Ara-CHOMP regimen,102a which is now: standard CHOP, with day 8 methotrexate (at 4 g/m2) and cytosine arabinoside (2 g/m2 twelve hourly X2) upon count recovery – repeated, has proved effective for both primary ocular lymphoma and primary cerebral lymphoma and is used before radiotherapy, which is brought in after two full chemotherapy cycles. The order of high dose methotrexate (and cytosine arabinoside) before brain radiotherapy is of importance with regard to CNS tolerance. Where the patient has presented with mono-ocular disease, the eye receives 40-44 Gy conventionally fractionated, and it is advised to give prophylactic brain radiotherapy at this time (24–30 Gy, conventionally fractionated); at least one American centre advises prophylactic radiotherapy to the other eye – a controversy that has been discussed in the literature.102b,102c,102d The long term prognosis of this rare lymphoma remains very guarded. SECONDARY DEPOSITS
Secondary deposits are usually seen in the uveal tract and only rarely in the retina, although choroidal metastases often present with a serous retinal detachment. The most common source of a choroidal metastasis is breast cancer. The ocular secondaries rarely present before the breast primary, and the deposits are commonly multiple and bilateral. The second most common source is lung cancer. An ocular deposit is, more often than not, detected before the lung primary and is usually solitary and unilateral. The treatment of ocular metastases is palliative and mainly by external-beam radiotherapy, although solitary metastases may be treated by plaque. It is legitimate to postpone irradiation and to watch for evidence of regression if the patient is to start chemotherapy, unless there is extensive retinal detachment. Photodynamic therapy may be employed to treat small, solitary choroidal metastases, metastases not controlled by radiotherapy, and rare new lesions arising in the previously irradiated choroid.
Primary malignant tumours of the eyelids, conjunctiva and orbit Like their benign counterparts, primary malignancies of the ocular adnexa are of epithelial, vascular, neurogenic, melanocytic or lacrimal-gland origin. MALIGNANT EPITHELIAL TUMOURS
These are relatively common. Basal-cell carcinoma is by far the most frequently encountered eyelid tumour, and the inner canthus is one of the most common sites for this neoplasm. The treatment of choice for eyelid basal-cell carcinomas depends on whether the tumour is of the nodular or morphoea type and on its size, location and the presence or absence of infiltration of underlying periosteum and bone. Small lesions may be treated effectively by liquid nitrogen cryotherapy. Cryotherapy or surgical excision is to be preferred for small basal-cell carcinomas on the upper eyelid, as radiotherapy may lead to keratinization of the eyelid margin and thereby to ocular discomfort from corneal epithelial abrasion. Very high cure rates can be obtained from surgical resection provided that complete excision is confirmed by frozen section at the time of surgery, especially for morphoea-type tumours. Extensive inner canthal lesions may be treated either by excision or by radiotherapy, the radiotherapy being a well/fully fractionated regime to reduce risks (e.g. lacrimal duct stenosis, ectropion), although it is difficult to avoid epiphora with either approach. Radiotherapy is particularly suitable for tumours that have recurred after excision and for those tethered to the periosteum, once again, a fully fractionated regime is advised. We utilise an orthovoltage technique with an ‘internal’ eye shield (i.e. contact lens of lead with smoothed internal wall) placed (after amethocaine desensitisation of the cornea) between the lids, to protect the eye from the radiation beam. Occasionally, orbital exenteration will be required for persistent or very extensive tumours. Adenocarcinoma is the second most common primary eyelid malignancy and most examples are sebaceous carcinomas. The eyelid is copiously supplied with glandular structures and is a site of predilection for this neoplasm, which may also arise in sebaceous glands in the caruncle or eyebrow. The tumour tends to be multifocal, and pagetoid spread from the lid on to the bulbar conjunctiva is characteristic. Diagnosis is difficult and often delayed because the tumour is mistaken for a simple meibomian cyst or for chronic conjunctivitis or blepharitis. Persistent or recurrent meibomian cysts and chronic, indeterminate ‘infections’ of the conjunctiva and eyelid should therefore be subjected to biopsy. After melanoma, it has the highest mortality rate of all eyelid cancers103 and is best managed by wide excision. Where the orbit is invaded, orbital exenteration is recommended, and although radiotherapy may be considered when an advanced sebaceous carcinoma involves the only good eye, this neoplasm is not very radiosensitive.
Adult tumours 335
Invasive squamous carcinoma is rare in the eyelid,104 often arising in a focus of actinic keratosis. It may also arise in the conjunctiva, either de novo or in an area of preexisting intra-epithelial squamous carcinoma.105 Surgical excision of eyelid lesions is to be preferred, and block dissection may be added in the rare instances in which regional lymph-node spread occurs. Tethered lymph glands may be better treated by radiotherapy. Metastasis from an eyelid primary is uncommon and death very rare.106 The mortality rate is similarly low for conjunctival primary epithelial tumours. Most are intra-epithelial, carcinomain-situ lesions that show a low propensity to develop into invasive carcinomas. The majority of these lesions are classified as ‘conjunctival (or corneal) intra-epithelial neoplasia’ (CIN) by ophthalmologists. They have indistinct margins and may be multifocal. Consequently, they show a strong tendency to recur after simple excision. The risk of recurrence may be reduced by adjuvant cryotherapy, although very persistent lesions are best managed by radiotherapy, particularly if there is any histopathological evidence of an invasive element. Radiotherapy for bulbar conjunctival lesions is best administered using a strontium-90 beta-ray source. Recently, topical mitomycin C therapy has been shown to be effective in the treatment of CIN that has recurred after surgery and for the primary treatment of some extensive, and therefore irresectable, in-situ lesions. Mitomycin C therapy alone is not recommended for thick tumours or for invasive conjunctival carcinoma.107 Thick in situ lesions should first be debulked and mitomycin C used in an adjuvant setting. Invasive tumours require wide excision.
MALIGNANT VASCULAR TUMOURS
Malignant vascular tumours may occur in the orbit but are uncommon. Haemangiopericytoma is a slow growing, encapsulated tumour.108 If the capsule is not breached during surgical removal, a total cure is usually effected. Postoperative radiotherapy may be indicated. Radiotherapy alone has not been shown to achieve sustained control of recurrent lesions109 and recurrence should be managed by orbital exenteration. Malignant haemangioendothelioma is much less common. It is not encapsulated and local recurrence is likely after attempted surgical excision unless a wide margin of normal tissue is excised. Recurrence should be managed by orbital exenteration. There is a high mortality rate from distant spread.110
recurrence and orbital exenteration may be required. Persistent local recurrence may be managed by radiotherapy. Death may occur from direct intracranial extension or from distant metastases. Most of the meningiomas seen in the orbit represent secondary orbital invasion by an intracranial tumour,112 but very occasionally a primary orbital meningiomamay arise in the optic nerve sheath. Primary meningiomas of the orbit may be excised via a lateral orbitotomy and the local recurrence rate after this operation is much less than that following removal of an intracranial meningioma. MALIGNANT MELANOCYTIC TUMOURS
Malignant melanocytic tumours are mostly melanomas. Their management is largely surgical.113 Malignant melanoma of the eyelid skin is extremely rare and is best managed by wide surgical excision.114,115 Malignant melanoma is more common in the conjunctiva (Fig. 15.11), but nevertheless still rare. The majority of conjunctival melanomas arise in a pre-existing conjunctival lesion.116 Approximately 18 per cent of tumours arise in a pre-existing naevus and 57 per cent in primary acquired melanosis (PAM). Primary acquired melanosis was formerly termed pre-cancerous melanosis and may be a manifestation of the atypical mole syndrome.117 Atypical melanocytes spread to involve much of the conjunctiva and on to the cornea118 and eyelid.119 Ultimately, melanoma develops in most cases of PAM in which atypia are found, as indicated by biopsy or by impression cytology,120 and in most cases multifocal tumours develop sequentially. Patients with PAM should have regular re-evaluation, looking for signs of a developing invasive melanoma. Cryotherapy is recommended to try to delay or prevent the onset of malignant change in PAM.121 This treatment should be directed towards the areas of acquired conjunctival pigmentation. Unfortunately, the success of this approach is limited by the facts that most patients with PAM do not present until the first melanoma has developed
MALIGNANT NEUROGENIC TUMOURS
Malignant neurogenic tumours of neural crest origin may arise de novo or may develop in a pre-existing benign tumour, particularly in association with neurofibromatosis. Malignant schwannoma111 is an infiltrative tumour which tends to grow along neural channels and to present with pain. Local surgical excision is frequently followed by
Figure 15.11 Conjunctival malignant melanoma arising in primary acquired melanosis.
336 Ocular and adnexal tumours
and that in some instances the melanosis is not pigmented and is therefore invisible.122 Eyelid skin melanomas are also associated with PAM,123 and the conjunctival involvement may not be apparent or may be unsuspected at the time the lid tumour is diagnosed.124 Survival rates are excellent following surgical excision of solitary bulbar conjunctival tumours not associated with PAM. The local recurrence rate is low after simple excision of melanomas that do not involve the corneoscleral limbus and after en-bloc lamellar corneoscleral dissection of those that do. Recurrence after incomplete excision may be reduced by adjuvant cryotherapy115 or by beta-ray radiotherapy.125 The effectiveness of adjunctive treatments with topical cytotoxics such as mitomycin C and 5-fluorouracil in reducing relapse after surgical excision of conjunctival melanoma or as a primary treatment is not yet fully evaluated. The value of these drugs in preventing the evolution of conjunctival melanomas from PAM is also not yet known. It is more difficult to eradicate a melanoma arising in the palpebral conjunctiva, partly because adjunctive treatments are more difficult to apply in this location. Accordingly, survival rates are poorer for such tumours. Additionally, melanomas arising in this location are more often than not associated with PAM, and other interrelated prognostic factors apply.116 Tumours arising in unfavourable locations have 2.2 times the mortality of epibulbar melanomas. Unfavourable locations include the palpebral conjunctiva, the conjunctival fornices and the plica, caruncle and lid margin. Histology is relevant and mixed-cell tumours have three times the mortality of spindle-cell lesions. Access to lymphatics may be the underlying reason why some locations are less favourable prognostically than others, and lymphatic invasion carries a fourfold increase in mortality rate. Multifocal and thick tumours attract a worse prognosis than solitary, thin ones. The various risk factors conspire to make conjunctival melanoma arising in PAM a much more serious disease than melanoma arising de novo or in a naevus. Because of their poor outlook, melanomas arising in the eyelid, the conjunctival fornix and, particularly, in the caruncle have been managed by orbital exenteration. In a recent study the role of exenteration was compared retrospectively in two groups of patients, one of which underwent the operation as the primary treatment of their disease, whereas the other first underwent conservative treatment, with exenteration reserved for palliative treatment.126 This second group of patients was not seen to be disadvantaged by prior conservative surgery in terms of survival. The key to survival appeared to be the thickness of the largest tumour rather than the treatment method, with overall melanoma-related mortality ranging between zero in tumours with a maximum thickness of 1 mm and 50 per cent in those in excess of 2 mm thick. A particularly poor outcome was noted for caruncular melanoma, with six out of seven patients dying despite primary exenteration. These results suggest that, where possible, a conservative approach should be tried first. Following exenteration, some patients
have developed recurrences in the orbit, nasal passages and paranasal sinuses.123,127 Orbital recurrence is seen in large, neglected tumours and nasal recurrence in inner canthal lesions, probably because of implantation of tumour cells shed down the nasal passages. At St Bartholomew’s Hospital, patients exenterated for large melanomas now receive adjunctive orbital radiotherapy, and those undergoing exenteration for inner canthal melanomas receive radiotherapy postoperatively to the ipsilateral nasal passages, in the hope of eliminating local recurrence. MALIGNANT LACRIMAL-GLAND TUMOURS
The most common malignant tumour of the lacrimal gland is adenocystic carcinoma. This tumour is not encapsulated and is locally invasive. There is a high mortality rate, with death usually resulting from direct intra-cranial extension. The tumour grows rapidly and the history of painful proptosis is short. The pain results from infiltration of neural channels. Radical resection may be performed if adenocystic carcinoma does not extend to the orbital apex and appears to be confined within the periosteum.128 Radiotherapy should be considered post-operatively and we usually use a wedged pair of megavoiltage portals to treat the whole orbit to a dose of (50)–55 Gy in 6 weeks. This is high dose orbital radiotherapy and there is ocular morbidity (dry eye, cataractogenesis and some risk to the retina). MALIGNANT LYMPHOMAS OF THE ORBIT
The orbit is one of the more common sites of extranodal origin of lymphoma, and is a disease that must be distinguished from the rare ocular lymphoma (vide supra). Almost all orbital lymphomas are of B-cell origin; Hodgkin’s disease rarely affects the orbit. Primary conjunctival lymphomas are now recognised as mucosal associated lymphoid tissue (MALT) tumours – marginal zone lymphoma (MZL) – and these comprise 44% of all orbital cases. Chronic antigenic stimulation (e.g. chlamydial infection or the autoimmune thyroid eye disease) may predispose to the development of these lymphomas. The recent discovery of regression of conjunctival MALT lymphoma following treatment of the chlamydia infection with doxycycline129 is a fascinating observation as it is entirely analogous to the discovery of gastric MALT lymphoma responding to anti-Heliobacter therapy. Clearly this needs to be further researched. Other histologies found are: lymphoplasmacytic/lymphoplasmacytoid lymphoma (LPL) (24%), follicle centre cell lymphoma (FCL) (12%), diffuse large cell lymphoma (12%) and rare types (e.g. mantle cell lymphoma), and, of course, orbital deposits from systemic lymphoma occur. The risk of development of extra-orbital disease and lymphoma-related death has been assessed in a recent UK analysis of almost 200 patients presenting to Moorfield’s
Adult tumours 337
Eye and St. Bartholomew’s Hospitals, and with long follow-up. The risk of relapse by five years, in patients presenting with localised disease is related to pathology, being 47% for MZL, 48% for LPL, 64% for FCL, 81% for DCL and 95% for other histologies, with five year lymphomarelated mortality rates of 12%, 19%, 22%, 48% and 53% respectively.130,131 Prognosis is also related to the stage at presentation. Orbital lymphomas may be bilateral at presentation (16% in our analysis of 326 cases of primary orbital lymphoma), which is surprising on the one hand given the absence of lymphatic connections between the two orbits, but unsurprising when one considers the aetiological factors contributing to the development of MALT lymphoma. Interestingly in our series, was the fact that bilaterality at presentation doubled the risk of extra-orbital disease/ relapse – is it just doubling the chance because there are two orbits from which to spread? – we are unsure. Our analysis also examined other predicting clinical features for late relapse and death, the incidence of these two events being sequentially greater for patients with predominantly deep orbital presentations (and painful optic neuropathy was significantly prognostic as it marks for large orbital masses), lacrimal gland and eyelid (i.e.(sub)cutaneous eyelid mass) presentations. These clinical features also marked for a greater likelihood of discovering systemic disease at presentation, as compared to a patient with a lengthy history of a unilateral conjunctival (MALT) lymphoma mass.132 Diagnosis of orbital lymphoma is made by orbital scanning (preferably MR) followed by biopsy for histology. Staging follows – the authors do not perform bone marrow testing routinely on (otherwise) stage IE conjunctival MALT lymphoma. Where the disease is localised to the conjunctiva, (no disease behind the equator of the globe on good quality MR of the orbit), the optimal treatment has always been radiotherapeutic; this might change if the doxycycline therapy approach is validated. When radiotherapy is used, our current approach is with an anterior orthovoltage
(a)
Figure 15.12
(250–300 KV) portal covering the orbit and with midline ocular shielding; this last is achieved with a plastic contact lens with a centrally mounted lead cylinder – placed carefully between the lids before each radiation fraction (after numbing the conjunctiva with amethocaine eye drops) such that the entire cornea (and behind it the lens) is shielded from the primary beam. Our research demonstrated that an orthovoltage beam was superior to an electron beam, where there was more scatter inwards under the lead shield. Where the disease is behind the equator of the globe on MR or not solely conjunctival, such a technique (utilised to prevent cataractogenesis) is inappropriate and a wedged pair of megavoltage portals is used to treat the entire orbit, mindful of the possible need to bolus anteriorly where the disease comes very superficially, anteriorly. The standard radiotherapy prescription is 30 Gy in 15 fractions over 21 days, although a higher dose – to 35 Gy – may be appropriate for bigger tumour masses. For localised high grade lymphomas, four cycles of CHOP (Cyclophosphamide, adriamycin, vincristine and prednisolone) precedes radiotherapy to the orbit. Where the tumour expresses the surface CD20 antigen, the addition of the humanised monoclonal antibody directed at this determinant: rituximab, is made. Such chemotherapy is clearly appropriate for high grade-extra-orbital disease, whilst simpler chemotherapy (e.g. single agent chlorambucil) would be first employed for extra-orbital low-grade disease – indeed treatment would be along standard NHL lines. SECONDARY DEPOSITS
Secondary deposits are uncommon in the ocular adnexa in adults, although they may occasionally be seen in the orbit. Breast carcinoma is the most common primary source. They are treated by radiotherapy. Intraocular metastases also occur – presenting with deteriorating vision in the eye – and, once again, breast cancer is the most common source. There is a predisposition for
(b)
Subconjunctival deposit of lymphoma (a) before and (b) after treatment by fractionated radiotherapy.
338 Ocular and adnexal tumours
brain and the other eye to be involved and a brain MR scan and binocular ophthalmic examination is always indicated once this diagnosis has been made. Radiotherapy is the therapy of choice with doses in the range of 40 Gy in four weeks or shorter fractionated courses being delivered.
KEY LEARNING POINTS ●
●
●
●
●
The modern treatment of tumours of the eye and ocular adnexa is still primarily surgical, but a multidisciplinary approach with radiotherapy or chemotherapy is increasingly employed, particularly for malignancies in childhood. Retinoblastoma has one of the highest survival rates of all childhood cancers, achieved mainly by adjuvant chemotherapy after enucleation for tumours with high-risk histology. In the conservative treatment of retinoblastoma, focal methods and chemotherapy are currently preferred to external-beam radiotherapy in order to reduce local morbidity and to minimize the chance of inducing other cancers in genetically susceptible children. Chemoreduction is still recommended for orbital rhabdomyosarcoma with high-risk histology, but it remains controversial whether it is permissible to eliminate either the alkylating agent or radiotherapy in tumours of favourable cell type. Both the choice of treatment and the life prognosis for uveal tract melanoma depend on the size and location of the tumour within the eye. Particularly if there is extensive retinal detachment or neovascular glaucoma, globes with large melanomas are still best enucleated, although mortality rates are relatively undiminished by aggressive treatment. Employing a variety of techniques, almost all eyes with smaller tumours can be preserved with excellent cosmesis, often with good visual function, and with a fairly optimistic survival prospect. Most ocular adnexal melanomas arise in the conjunctiva. Almost all are associated with unilateral, premalignant, atypical melanosis, in which case the tumours are usually sequentially multifocal. Survival rates in conjunctival melanoma are generally better than those in the uveal tract variant of this tumour and can be maximized by a rigid policy of lifelong observation with a high index of suspicion and early treatment of potential new lesions.
KEY REFERENCES Hungerford JL. Uveal melanoma. Eur J Cancer 1993; 29:1365–8. Hungerford JL. Surgical treatment of ocular melanoma. Melanoma Res 1993; 3:305–12. Hungerford JL, Plowman PN, Kingston JE. Tumours of the eye and orbit. In: Voute PA, Barrett A, Lemerie J (eds) Cancer in Children. Berlin: Springer Verlag, 1992, 207–25. Kingston JE, Hungerford JL. Retinoblastoma. In: Plowman PN, Pinkerton CR (eds) Paediatric Oncology. Clinical Practice and Controversies. London: Chapman & Hall, 1992, 268–90. Hungerford JL. Current trends in the treatment of ocular melanoma by radiotherapy. Clin Experiment Ophthalmol 2003; 31:8–13.
REFERENCES 1 Hungerford JL, Plowman PN, Kingston JE. Tumours of the eye and orbit. In: Voute PA, Barrett A, Lemerie J (eds) Cancer in Children. Berlin: Springer Verlag, 1992, 207–25. 2 Kingston JE, Hungerford JL. Retinoblastoma. In: Plowman PN, Pinkerton CR (eds) Paediatric Oncology. Clinical Practice and Controversies. London: Chapman & Hall, 1992, 268–90. 3 Knudson AG. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci 1971; 68:820–3. 4 Cavenee WK, Murphree AL, Shull MM, et al. Prediction of familial predisposition to retinoblastoma. N Engl J Med 1986; 314:1201–7. 5 Friend SH, Bernard R, Rogel S, et al. A human DNA segment with properties of the gene that predisposes to retinoblastoma and osteosarcoma. Nature 1986; 323:643–6. 6 Lee W-H, Bookstein R, Hong FD, et al. Human retinoblastoma susceptibility gene; cloning, identification and sequence. Science 1987; 235:1394–9. 7 Nevins JR. Transcriptional activation by viral regulatory proteins. Trends Biol Sci 1991; 16:435–9. 8 Draper GJ, Sanders BM, Kingston JE. Second primary neoplasms in patients with retinoblastoma. Br J Cancer 1986; 53:661–71. 9 Abramson DH, Ellsworth RM, Kitchin D, Tung G. Second monocular tumors in retinoblastoma survivors. Are they radiation induced? Ophthalmology 1984; 91:1351–5. 10 Arrigg PG, Hedges TR, Char DH. Computed tomography in the diagnosis of retinoblastoma. Br J Ophthalmol 1983; 67:588–91. 11 Hungerford JL. Histogenesis of retinoblastoma. Br J Ophthalmol 1990; 74:131–2. 12 Tarlton JF, Easty DL. The immunohistological characterisation of retinoblastoma related ocular tissue. Br J Ophthalmol 1990; 74:144–9. 13 Pratt CB, Meyer D, Chenaille P, Crom DB. The use of bone marrow aspirations and lumbar punctures at the time of diagnosis of retinoblastoma. J Clin Oncol 1989; 7:140–3. 14 Kingston JE, Plowman PN, Hungerford JL. Ectopic intracranial retinoblastoma in childhood. Br J Ophthalmol 1985; 69:742–8.
References 339
15 Schulman JA, Peyman GA, Mafee MF, et al. The use of magnetic resonance imaging in the evaluation of retinoblastoma. J Pediatr Ophthalmol Strabismus 1986; 3:144–7. 16 Reese AB, Ellsworth RM. The evaluation and current concept of retinoblastoma therapy. Trans Am Acad Ophthalmol Otolaryngol 1963; 67:164–72. 17 Pratt CB. Management of malignant solid tumors in children. Pediatr Clin North Am 1972; 19:1141–55. 18 Hungerford JL. Factors influencing metastasis in retinoblastoma. Br J Ophthalmol 1993; 77:541. 19 Olver JM, McCartney ACE, Kingston J, Hungerford J. Histological indicators of the prognosis for survival following enucleation for retinoblastoma. In: Bornfeld N, Gragoudas ES, Happing W, et al. (eds) Tumors of the Eye. Amsterdam: Kugler, 1991, 59–67. 20 Stevenson KE, Hungerford J, Garner A. Local extraocular extension of retinoblastoma following intraocular surgery. Br J Ophthalmol 1989; 73:739–42. 21 Hungerford JL, Kingston JE, Plowman PN. Orbital recurrence of retinoblastoma. Ophthal Paediatr Genet 1987; 8:63–8. 22 Goble RR, McKenzie J, Kingston JE, Plowman PN. Orbital recurrence of retinoblastoma successfully treated by combined therapy. Br J Ophthalmol 1990; 74:97–8. 23 MacKay CJ, Abramson DH, Ellsworth RM. Metastatic patterns of retinoblastoma. Arch Ophthalmol 1984; 102:391–6. 24 Kingston JE, Hungerford JL, Plowman PN. Chemotherapy in metastatic retinoblastoma. Ophthal Paediatr Genet 1987; 8:69–72. 25 Petersen RA, Friend SH, Albert DM. Prolonged survival of a child with metastatic retinoblastoma. J Pediatr Ophthalmol Strabismus 1987; 24:247–8. 26 Saleh RA, Gross S, Cassano W, Gee A. Metastatic retinoblastoma successfully treated with immunomagnetic purged autologous bone marrow transplantation. Cancer 1988; 62:2301–3. 27 Kingston JE, Hungerford JL, Madreperla SA, Plowman PN. Results of combined chemotherapy and radiotherapy for advanced intraocular retinoblastoma. Arch Ophthalmol 1996; 114:1339–43. 28 Chan HSL, DeBoer G, Thiessen JJ, et al. Combining cyclosporin with chemotherapy controls intraocular retinoblastoma without requiring radiation. Clin Cancer Res 1996; 2:1499–508. 29 Foote RL, Garretson BR, Schomberg PJ, et al. External beam irradiation for retinoblastoma: patterns of failure and doseresponse analysis. Int J Radiat Oncol Biol Phys 1989; 16:823–30. 30 Schipper J. Retinoblastoma: a Medical and Experimental Study. Thesis, University of Utrecht, 1980. 31 Harnett AN, Hungerford J, Lambert G, et al. Modern lateral external beam (lens sparing) radiotherapy for retinoblastoma. Ophthal Paediatr Genet 1987; 8:53–61. 32 Singh AD, Garway-Heath D, Love S, et al. Relationship of regression pattern to recurrence in retinoblastoma. Br J Ophthalmol 1993; 77:12–16.
33 Harnett AN, Thomson E. An iodine-125 plaque for radiotherapy of the eye: manufacture and dosimetric considerations. Br J Radiol 1988; 61:835–8. 34 Lommatzsch P. Beta-irradiation of retinoblastoma with 106Ru/106Rh applicators. Mod Probl Ophthalmol 1977; 18:128–36. 35 Höpping W, Meyer-Schwickerath G. Light coagulation treatment in retinoblastoma. In: Boniuk M (ed) Ocular and Adnexal Tumors: New and Controversial Aspects. St Louis: Mosby, 1964, 192–6. 36 Shields JA, Shields CL, Parsons H, Giblin ME. The role of photocoagulation in the management of retinoblastoma. Arch Ophthalmol 1990; 108:205–8. 37 Abramson DH, Ellsworth RM. Cryotherapy for retinoblastoma. Arch Ophthalmol 1982; 100:1253–6. 38 Shields JA, Parsons H, Shields CL, Giblin ME. The role of cryotherapy in the management of retinoblastoma. Am J Ophthalmol 1989; 108:260–4. 39 Sutow WW, Lindberg RD, Geban EA, et al. Three year relapse free survival rates in childhood rhabdomyosarcoma of the head and neck. Cancer 1982; 49:2217–21. 40 Sagerman RH, Treffer P, Ellsworth RM. The treatment of orbital rhabdomyosarcoma of children with primary radiation therapy. Am J Roentgenol Radium Ther Nucl Med 1972; 114: 31–4. 40a Crist WM, Anderson JR, Meza JL, et al. Intergroup rhabdomyosarcoma study IV: Results for patients with nonmetastatic disease. J Clin Oncol 2001; 19:3091–3102. 40b Stephens MGC, Rey A, Bouvet N, et al. Treatment of nonmetastatic rhabdomyosarcoma in childhood and adolescence: Third study of the International Society for Paediatric Oncology – SIOP Malignant Mesenchymal Tumour 89. J Clin Oncol 2005; 23:2618–2628 41 Kingston JE, McElwain TJ, Malpas JS. Childhood rhabdomyosarcoma: experience of the Children’s Solid Tumour Group. Br J Cancer 1984; 48:195–207. 42 Heyn R, Ragab A, Raney RB Jr, et al. Late effects of therapy in orbital rhabdomyosarcoma in children. Cancer 1986; 57:1738–43. 43 Meyers PA, Ghavimi F. Secondary acute non-lymphocytic leukaemia (ANLL) following treatment of childhood rhabdomyosarcoma. Abstracts of the Proceedings of the American Society of Clinical Oncology 1983; 2:77. 44 Anderson SR. Medulloepitheliomas of the retina. Int Ophthalmol Clin 1962;2:483–506. 45 Broughton WL, Zimmerman LE. A clinicopathologic study of 56 cases of intraocular medulloepitheliomas. Am J Ophthalmol 1978; 85:407–18. 46 Mullaney J. Primary malignant medulloepithelioma of the retinal stalk. Am J Ophthalmol 1974; 77:499–504. 47 Green WR, Iliff WJ, Trotter RR. Malignant teratoid medulloepithelioma of the optic nerve. Arch Ophthalmol 1974; 91:451–4. 48 Morris AT, Garner A. Medulloepithelioma involving the iris. Br J Ophthalmol 1975; 59:276–8. 49 Canning CR, McCartney ACE, Hungerford J. Medullo epithelioma (Diktyoma). Br J Ophthalmol 1988; 72:764–7.
340 Ocular and adnexal tumours
50 Chutorian AM, Schwartz JF, Evans RA, et al. Optic gliomas in children. Neurology 1964; 14:83–95. 51 Manschot WA. Primary tumours of the optic nerve in von Recklinghausen’s disease. Br J. Ophthalmol 1954; 38:285–9. 52 Crowe FW, Schull WJ. The diagnostic importance of café au lait spot in neurofibromatosis. Arch Intern Med 1953; 91:758–66. 53 Innes RK, Hoyt WF. Childhood chiasmal gliomas: update on the fate of the patients in the 1969 San Francisco study. Br J Ophthalmol 1986; 70:179–82. 54 Reese AB. Glioma of the optic nerve, retina and orbit. In: Reese AB (ed) Tumors of the Eye. New York: Hoeber, 1963, 162–79. 55 Taveras JM, Mount LA, Wood EH. The value of radiation therapy in the management of glioma of the optic nerves and chiasm. Radiology 1956; 66:518–28. 56 Lloyd L. Gliomas of the optic nerve and chiasm in childhood. Trans Am Ophthalmol Soc 1973; 72:488–535. 57 Rush JA, Younge BR, Campbell RJ, et al. Optic glioma: long term follow up of 85 histologically verified cases. Ophthalmology 1982; 89:1213–19. 58 Moore AT, Pritchard J, Taylor DSI. Histiocytosis X: an ophthalmological review. Br J Ophthalmol 1985; 69:7–14. 59 Harnett AN, Doughty D, Hirst A, et al. Radiotherapy in benign orbital disease, II. Ophthalmic Graves’ disease and histiocytosis X. Br J Ophthalmol 1988; 72:289–92. 60 Ganley JP, Comstock GW. Benign nevi and malignant melanomas of the choroid. Am J Ophthalmol 1973; 76:19–25. 61 Gonder JR, Shields JA, Albert DM, et al. Uveal malignant melanoma associated with ocular and oculodermal melanocytosis. Ophthalmologym 1982; 89:953–60. 62 White V, Stevenson K, Garner A, Hungerford J. Mesectodermal leiomyoma of the ciliary body: case report. Br J Ophthalmol 1989; 73:12–18. 63 Foss AJE, Pecorella I, Alexander RA, et al. Are most intraocular ‘leiomyomas’ really melanocytic lesions? Ophthalmology 1994; 101:919–24. 64 Annesley WH, Leonard BC, Shields JA, Tasman WS. Fifteen-year review of treated cases of retinal angiomatosis. Trans Am Acad Ophthalmol Otolaryngol 1977; 83:446–53. 65 Goldberg MF, Koenig S. Argon laser treatment of von Hippel–Lindau retinal angiomas. I. Clinical and angiographic findings. Arch Ophthalmol 1971; 92:12–15. 66 Goldberg RE, Pheasant TR, Shields JA. Cavernous hemangioma of the retina: a four generation pedigree with neurooculocutaneous involvement and an example of bilateral retinal involvement. Arch Ophthalmol 1979; 97:2321–4. 67 Plowman PN, Harnett AN. Radiotherapy in benign orbital disease, I. Complicated ocular and orbital angiomas. Br J Ophthalmol 1986; 72:286–8. 68 Jurklies B, Bornfeld N. The role of photodynamic therapy in the treatment of symptomatic choroidal hemangioma. Graefes Arch Clin Exp Ophthalmol 2005; 243:393–6. 69 Huiskamp EA, Muskens RP, Ballast A, Hooymans JM. Diffuse choroidal haemangioma in Sturge-Weber syndrome treated with photodynamic therapy under general anaesthesia. Graefes Arch Clin Exp Ophthalmol 2005; 243:727–30.
70 Garner A. Pathology of ‘pseudotumour’ of the orbit: a review. J Clin Pathol 1973; 26:639–48. 71 Hamming N. Anatomy and embryology of the eyelid. Review with special reference to the development of divided nevi. J Pediatr Dermatol 1983; 1:518. 72 Wright JE, Stewart WB, Krohel GB. Clinical presentation and management of lacrimal gland tumours. Br J Ophthalmol 1979; 63:600–6. 73 Shields JA. Current approaches to the diagnosis and management of choroidal melanomas. Surv Ophthalmol 1977; 21:443–63. 74 Einhorn LH, Burgess MA, Gottlieb JA. Metastatic patterns of choroidal melanoma. Cancer 1974; 34:1001–4. 75 Zimmerman LE, McLean IW, Foster WD. Does enucleation of the eye containing a malignant melanoma prevent or accelerate the dissemination of tumour cells? Br J Ophthalmol 1978; 62:420–5. 76 Hungerford JL. Uveal melanoma. Eur J Cancer 1993; 29:1365–8. 77 Hungerford JL. Surgical treatment of ocular melanoma. Melanoma Res 1993; 3:305–12. 78 Char DH, Store RD, Irvine IR, et al. Diagnostic modalities in choroidal melanoma: sensitivity, specificity and reproducibility. Am J Ophthalmol 1980; 89:223–30. 79 François J. Treatment of malignant choroidal melanomas by photocoagulation. Ophthalmologica 1982; 184:121–30. 80 Tse DT, Dutton JJ, Weingeist TA, et al. Hematoporphyrin photoradiation therapy for intraocular and orbital malignant melanoma. Arch Ophthalmol 1984; 102:833–8. 81 Shields CL, Shields JA, Cater J, Lois N, Edelstein C, Gündüz K, Mercado G. Transpupillary thermotherapy for choroidal melanoma: tumour control and visual results in 100 consecutive cases. Ophthalmology 1998; 105:581–90. 82 Win PH, Robertson DM, Buettner H, McCannel CA, Bennett SRW. Extended follow-up of small melanocytic choroidal tumors treated with transpupillary thermotherapy. Arch Ophthalmol 2006; 124:503–6. 83 Foulds WS. Current options in the management of choroidal melanoma. Trans Ophthalmol Soc UK 1983; 103:28–34. 84 Damato BE, Paul J, Foulds WS. Risk factors for residual and recurrent melanoma after trans-scleral local resection. Br J Ophthalmol 1996; 80:102–8. 85 Hungerford JL. Current trends in the treatment of ocular melanoma by radiotherapy. Clin Experiment Ophthalmol 2003; 31:8–13. 86 Stallard HB. A case of malignant melanoma of the choroid successfully treated by radon seeds. Trans Ophthalmol Soc UK 1949; 69:293–7. 87 Stallard HB. Radiotherapy for malignant melanoma of the choroid. Br J Ophthalmol 1966; 50:147–55. 88 Bedford MA. The use and abuse of cobalt plaques in the treatment of choroidal malignant melanomata. Trans Ophthalmol Soc UK 1973; 93:139–43. 89 MacFaul PA. Local radiotherapy in the treatment of malignant melanoma of the choroid. Trans Ophthalmol Soc UK 1977; 97:421–7.
References 341
90 Lommatzsch PK. P-irradiation of choroidal melanoma with 106Ru/106Rh applicators, 16 years’ experience. Arch Ophthalmol 1983; 101:713–17. 91 Packer S. Iodine-125 radiation of posterior uveal melanoma. Ophthalmology 1987; 94:1621–5. 92 Gragoudas ES, Goitein M, Koehler AM, et al. Proton irradiation of small choroidal malignant melanomas. Am J Ophthalmol 1977; 83:665–73. 93 Munzenrider JE, Gragoudas ES, Seddon JM, et al. Conservative treatment of uveal melanoma: probability of eye retention after proton treatment. Int J Radiat Oncol Biol Phys 1988; 15:553–8. 94 Char DH, Castro JR. Helium ion therapy for choroidal melanoma. Arch Ophthalmol 1982; 100:935–8. 95 Seddon JM, Gragoudas ES, Egan KM, et al. Relative survival rates after alternative therapies for uveal, melanoma. Ophthalmology 1990; 97:769–77. 96 Foss AJ, Whelahan I, Hungerford JL, et al. Predictive factors for the development of rubeosis following proton beam radiotherapy for uveal melanoma. Br J Ophthalmol 1997; 81:748–54. 97 Zehetmayer M, Menapace R, Kitz K, Ertl A. Stereotatic radiosurgery for uveal melanoma. In: Kogelnik HD (ed) Progress in Radio-oncology. Bologna: Monduzzi, 1995, 451–4. 98 Wilson MW, Hungerford JL. Comparison of episcleral plaque and proton beam radiation therapy for the treatment of choroidal melanoma. Ophthalmology 1999; 106:1579–87. 99 Starr HJ, Zimmerman LE. Extrascleral extension and orbital recurrence of malignant melanoma of the choroid and ciliary body. Int Ophthalmol Clin 1962; 2:369–84. 100 Hykin PG, McCartney ACE, Plowman PN, Hungerford JL. Postenucleation orbital radiotherapy for the treatment of malignant melanoma of the choroid with extrascleral extension. Br J Ophthalmol 1990; 74:36–9. 101 Shields JA, Zimmerman LE. Lesions simulating malignant melanoma of the posterior uvea. Arch Ophthalmol 1973; 89:466–71. 102a Plowman PN, Montefiore DSM, Lightman S. Multiagent chemotherapy in the salvage cure of ocular lymphoma relapsing after radiotherapy. Clin Oncol 1993; 5:315–6. 102b Valluri S, Moorthy RS, Khan A, Rao NA. Combination treatment of intraocular lymphoma. Retina 1995; 15:125–9. 102c Plowman PN, Taylor A, Jackson ASN, Lightman S, Pavesio C. Optimal therapy of primary ocular lymphoma. Clin Oncol 2002; 14:521. 102d HormigoA, Abrey LE, De Angelis LM. Treatment of ocular lymphoma (Abstr.) Proc Am Soc Clin Oncol 2002 103 Rao NA, Hidayat AA, McLean IW, Zimmerman LE Sebaceous carcinomas of the ocular adnexa: a clinicopathologic study of 104 cases with five-year follow-up data. Hum Pathol 1982; 13:113–22. 104 Aurora A, Blodi F. Lesions of the eyelids: a clinicopathologic study. Surv Ophthalmol 1970; 15:94–104.
105 Zimmerman LE. The cancerous, precancerous, and pseudocancerous lesions of the cornea and conjunctiva. The Pocklington Memorial Lecture. In: Ryeroft PV (ed.) Corneoplastic Surgery. New York: Pergamon Press, 1969, 547–55. 106 Jakobiec FA, Rootman J, Jones IS. Secondary and metastatic tumors of the orbit. In: Jones IS, Jakobiec FA (eds) Diseases of the Orbit. Hagerstown: Harper and Row, 1979, 503–6. 107 Wilson MW, Hungerford JL, George SM, Madreperla SA. Topical mitomycin C for the treatment of conjunctival and corneal epithelial dysplasia and neoplasia. Am J Ophthalmol 1997; 124:303–11. 108 Jakobiec FA, Howard GM, Jones IS, et al. Hemangiopericytoma of the orbit. Am J Ophthalmol 1974; 78:816–34. 109 Jakobiec FA, Jones IS. Vascular tumors, malformations and degenerations. In: Jones IS, Jakobiec FA (eds) Diseases of the Orbit. Hagerstown: Harper and Row, 1979, 269–308. 110 Stout AP. Hemangio-endothelioma: a tumor of blood vessels featuring vascular endothelial cells. Ann Surg 1943; 118:445–64. 111 Grinberg M, Levy NS. Malignant neurilemmoma of the supraorbital nerve. Am J Ophthalmol 1974; 78:489–92. 112 Stern WE. Meningiomas in the cranio-orbital junction. J Neurosurg 1973; 38:428–37. 113 Hungerford J, Collin JRO. Mélanomes orbito-palpébraux. In: Adenis J-P, Morax S (eds) Pathologie Orbitopalpébrale. Paris: Masson, 1998, 355–60. 114 Garner A, Koornneef L, Levane A, Collin JR. Malignant melanoma of the eyelid skin: histopathology and behaviour. Br J Ophthalmol 1985; 69:180–6. 115 Collin JRO, Allen LH, Garner A, Hungerford JL. Malignant melanoma of the eyelid and conjunctiva. Aust N Z J Ophthalmol 1986; 14:29–34. 116 Paridaens ADA, Minassian DC, McCartney ACE, Hungerford JL. Prognostic factors in primary malignant melanoma of the conjunctiva: a clinicopathological study of 256 cases. Br J Ophthalmol 1994a; 78:252–9. 117 Bataille V, Boyle J, Hungerford JL, Newton JA. Three cases of primary acquired melanosis of the conjunctiva as a manifestation of the atypical mole syndrome. Br J Dermatol 1993; 128:86–90. 118 Paridaens ADA, Kirkness CM, Garner A, Hungerford JL. Recurrent malignant melanoma of the corneal stroma: a case of ‘black cornea’. Br J Ophthalmol 1992; 76:444–6. 119 Robertson DM, Hungerford JL, McCartney ACE. Pigmentation of the eyelid margin accompanying conjunctival melanoma. Am J Ophthalmol 1989; 108:435–9. 120 Paridaens ADA, McCartney ACE, Curling OM, et al. Impression cytology of conjunctival melanosis and melanoma. Br J Ophthalmol 1992; 76:198–201. 121 Jakobiec FA, Rini FJ, Fraunfelder FT, Brownstein S. Cryotherapy for conjunctival primary acquired melanosis and malignant melanoma. Ophthalmology 1988; 95:1058–70.
342 Ocular and adnexal tumours
122 Paridaens ADA, McCartney ACE, Hungerford JL. Multifocal amelanotic conjunctival melanoma and acquired melanosis sine pigmento. Br J Ophthalmol 1992; 76:163–5. 123 Robertson DM, Hungerford JL, McCartney ACE Malignant melanomas of the conjunctiva, nasal cavity, and paranasal sinuses. Am J Ophthalmol 1989; 108:440–2. 124 Hicks C, Liu C, Hiranandani M, et al. Conjunctival melanoma after excision of a lentigo maligna melanoma in the ipsilateral eyelid skin. Br J Ophthalmol 1994; 78:317–18. 125 Lederman M, Wybar K, Busby E. Malignant epibulbar melanoma: natural history and treatment by radiotherapy. Br J Ophthalmol 1984; 68:605–17. 126 Paridaens ADA, McCartney ACE, Minassian DC, Hungerford JL. Orbital exenteration in 95 cases of conjunctival malignant melanoma. Br J Ophthalmol 1994; 78:520–8. 127 Paridaens ADA, McCartney ACE, Lavelle RJ, Hungerford JL. Nasal and orbital recurrence of conjunctival melanoma 21 years after exenteration. Br J Ophthalmol 1992; 76:369–71.
128 Wright JE. Management of malignant orbital tumours. In: Oosterhuis JA (ed.) Ophthalmic Tumours. Dordrecht: W. Junk, 1985, 229–39. 129 Ferreri AJM, Ponzini M, Guidoboni M, et al. Regression of ocularadnexal lymphoma after chlamydia psittacieradicatiing antibiotic therapy. J Clin Oncol 2005; 23: 5067–73. 130 Hardman-Lea S, Kerr-Muir M, Wotherspoon AC, et al. Mucosal-associated lymphoid tissue lymphoma of the conjunctiva. Arch Opohthalmol 1994; 112:1207–12. 131 Jenkins C, Rose GE, Bunce C, et al. Histological features of ocular adnexal lymphoma (REAL) classification and their association with patient morbidity and survival. Brit J Opohthalmol 2000; 84:907–13. 132 Jenkins C, Rose GE, Bunce C, et al. Clinical features associated with survival of patients with lymphoma of the ocular adnexa. Eye 2003; 17:809–20.
16 Head and neck cancer ALASTAIR J. MUNRO AND NICHOLAS D. STAFFORD
Introduction Organization of services for patients Incidence Head and neck cancer is a global problem Outcome Aetiology The problem of second primary tumours Strategies for prevention General principles of surgery General principles of radiotherapy
343 344 344 345 345 346 347 347 349 352
INTRODUCTION The management of cancer of the head and neck has a reputation for being difficult, for the reasons summarized in Box 16.1. Many of the apparent difficulties can be resolved by realizing that the management of head and neck cancer follows the same basic principles that apply to the management of any tumour. Good communication between specialists is a fundamental prerequisite; no individual specialist is omniscient. Surgical and radiotherapeutic techniques are evolving rapidly, new drugs are being introduced for the prevention and treatment of the disease, and the limitations of the past are no guide to the possibilities of the present. Decision-making often involves careful appraisal of competing options: the welfare of the patient is the paramount concern. The defence of turf or the maintenance of reputation should be of no consequence. The key decision-maker should be the patient; it is, after all, the patients who, once they have been properly informed of the risks and advantages of the various options, are in the best position to make decisions that are right for them, as individuals. In order for patients to be adequately assessed and informed it is essential that there be a multidisciplinary team characterized not just by technical competence, but also by open communication. The easiest way to achieve the necessary co-operation and communication between the various specialists involved in
The role of chemotherapy Unresolved issues and future developments in chemotherapy and radiotherapy Quality of life Palliative care Novel approaches to management Staging and prognosis Host-related prognostic factors Assessment of patients References
364 366 367 367 369 370 372 372 400
Box 16.1 Factors contributing to the difficulties and complexities encountered in the management of head and neck cancer ●
●
●
●
●
●
●
The anatomy of the head and neck region is complicated This is reflected in the number of primary tumour sites specified by the TNM system for staging head and neck cancer There have been several revisions of the TNM system over the past 15 years Tumours of apparently identical histological appearance may, according to their primary site of origin, behave very differently The role of chemotherapy in head and neck cancer is poorly defined For many tumours, the decision between surgery and radiotherapy as definitive treatment is not clear cut The close proximity of vulnerable normal structures to head and neck tumours means that both radiotherapy and surgery are technically demanding. A wide repertoire of techniques is required to treat these tumours effectively
344 Head and neck cancer
●
●
The side-effects of the treatments available for head and neck cancer mean that effective palliative therapy is often difficult Patients are often elderly, in poor physical condition, or have limited social support
Box 16.2 Members of the multidisciplinary team required for the effective management of head and neck cancer ● ● ● ● ● ● ● ● ● ● ●
● ● ● ●
Otorhinolaryngologist/head and neck surgeon Oral and maxillofacial surgeon Plastic and reconstructive surgeon Oral surgeon Clinical oncologist/radiotherapist Diagnostic radiologist Pathologist Speech therapist Prosthetist Prosthodontist Clinical nurse specialist (care of stoma, rehabilitation, supportive care) Macmillan nurse (symptom control, palliative care) Social worker/counsellor Medical secretary/administrator Audit/Information co-ordinator
managing head and neck cancer is through combined clinics and multidisciplinary team meetings . Such clinics and meetings require commitment and flexibility from those who participate – the surrender of autocracy involved in the process may not always suit the individual clinician but is undoubtedly in the patient’s best interests. The recognition that the term ‘clinician’ embraces disciplines other than surgery and medicine, such as nursing, radiography and speech therapy, is an open acknowledgement of the importance of the team approach to managing cancer (Box 16.2). The patient is always the most important member of the team.
ORGANIZATION OF SERVICES FOR PATIENTS By their very nature and constitution, the multidisciplinary team and the combined clinic have to be based at a specialist centre. This immediately creates a problem: patients with head and neck cancer will not always present to large centres; they will present to local hospitals and will not arrive neatly diagnosed and staged. In the UK each year otorhinolaryngologists see nearly 1 million patients, 50 000 of whom have change in their voice as a symptom. Fewer than 5 per cent of these 50 000 patients will have cancer; 48 000 patients with voice changes will not have cancer and they cannot all be referred to combined clinics for specialist assessment and opinion. Even a comparatively common head and neck
cancer, laryngeal cancer, is rarely encountered in general practice. The average British GP will see one new patient with carcinoma of the larynx every 15 years or so. The current model for delivering cancer care in the UK assumes that diagnosis will be made at a hospital near the patient’s home but that patients will then need to travel to specialist centres for assessment and treatment. It is the ease with which this transfer can take place that is crucial to the success of this model of care. One method is to have managed clinical networks, in which there are protocols and guidelines in place at district, regional and supraregional levels. These are intended to ensure that, wherever possible, patients with a specific problem are managed uniformly and that a high standard of care is not dependent on a combination of serendipity and postcode. Data related to the interval between first symptom and specialist referral are, in the UK at least, encouraging: patients delayed a median of 4 weeks before consulting their family doctor, and the majority of patients had been seen by a specialist within 12 weeks. There was, however, a significant delay between first being seen at a specialist ENT unit and starting definitive treatment: the mean interval for surgery was 5 weeks and for primary radiotherapy 10 weeks.1 Despite the theoretical advantages associated with management at a specialist centre, there is little actual proof that referral to a specialist centre influences outcome. Given the technical demands of surgery and radiotherapy for potentially curable patients with advanced disease, it would be surprising if experience and technical aptitude counted for nought. However, these patients are a small subset of all patients with head and neck cancer and so it would be difficult to design a study with sufficient statistical power to show an overall survival difference. Average improvements from specialist referral will be hard to prove and are largely irrelevant: it is failures at an individual level that are important (the ‘if one sparrow falls’ argument).
INCIDENCE World-wide, nearly 650 000 people develop head and neck cancer each year and there are 350 000 deaths from the disease. The ratio of males to females is 3:1 (Table 16.1). Taking Scotland as a typical developed country, the pattern of disease is somewhat different. The ratio of males to females is 2:1. Head and neck cancer is the fourth most common cancer in men and the ninth most common in women. There were just under 1000 new patients diagnosed in Scotland in 2002 (which would correspond to a figure of 11 000 for the UK as a whole). The 5-year relative survival is around 55 per cent. The incidence is rising, by 6.6 per cent for men and by 15 per cent for women between 1992 and 2002. Although incidence is rising, mortality is decreasing, particularly for women: between 1994 and 2004 the mortality decreased by 22 per cent for women and by 9 per cent for men. At any one time there are more than
Outcome 345
Table 16.1 Head and neck cancer: world annual incidence and mortality data (Globocan 2002 http://www-dep.iarc.fr/ globocan/database.htm) Male
Female
Incidence Oral cavity Nasopharynx Other pharynx Larynx Total
175 916 55 796 106 219 139 230 477 161
98 373 24 247 24 077 20 011 166 708
643 869
Deaths Oral cavity Nasopharynx Other pharynx Larynx Total
80 736 39 913 67 964 78 628 262 241
46 723 34 913 67 964 11 327 89 498
351 739
45 000 people in the UK who are living with head and neck cancer or the consequences of its treatment. The standardized incidence rates suggest that the increase in numbers is mainly due to changes in the size and age structure of the population. The incidence of head and neck cancer is related to social deprivation, and so is survival. The disease is twice as common when the most deprived are compared with the most affluent (European Age Standardisation Rate 21.7 versus 10.3) and the 5-year survival is 52 per cent amongst the most deprived and 60 per cent amongst the most affluent,2 and there has been a rise in the incidence of head and neck cancer in the most deprived sections of the community. These data clearly point to the social and environmental origins of the disease: as the rich become richer and the poor become poorer, these social changes have a measurable impact on those cancers whose incidence is related to personal behaviours associated with social deprivation. This has important implications for any programmes aimed at cancer prevention. Someone whose existence is fairly miserable, whose quality of life is eroded by social deprivation, and with no prospect of improving their lot, is unlikely to be swayed by arguments involving the foreswearing of pleasures that are immediate, and actual, for gains that are both deferred and theoretical.
HEAD AND NECK CANCER IS A GLOBAL PROBLEM Head and neck cancer is an important problem in the developing world. As tobacco companies move into what is, for them, the lucrative market of the developing nations, the harvest of their salesmanship will be an ever-increasing number of head and neck tumours in those nations who can least afford to treat them. A third of Chinese males currently under 29 years old will die from smoking-related illness: this corresponds to a total of 100 million premature deaths.
There are wide geographical variations in the standardized incidence rates for specific types of head and neck cancer. One of the poorest countries in the world, Bangladesh, has some of the highest incidence rates of head and neck tumours. These wide differences provide important clues concerning the causes of the disease and the role of genetic susceptibility to the development of cancer. The distribution of nasopharyngeal cancer is a case in point. The highest incidence is in Hong Kong and South-East Asia, with world age-standardized rates per 100 000 men of 25.2 (Hong Kong), 15.8 (Singapore) and 13.4 (Malaysia). The incidence is also high in parts of North Africa – 10.9 in Tunisia – and this is, both clinically and epidemiologically, a distinct form of the disease. Because of the colonial links with France, this in turn affects the clinical features of nasopharyngeal cancer as it presents to some French hospitals. The practical point is that experience obtained in Hong Kong does not necessarily translate into appropriate guidelines for patients being treated in Paris. The reverse is also true. Another feature of the global epidemiology of head and neck cancer is the emerging problem of laryngeal cancer in the countries of the former Eastern bloc. Standardized rates per 100 000 men are approximately triple those in the UK (4.8): Ukraine 12.5; Russian Federation 12.5; Croatia 13.2; Poland 12.1; Bosnia-Herzegovina 12.6. Worldwide, these rates are only exceeded by the figures for France (15.3), Spain (14.2) and Bangladesh (16.0). Even within Europe and the developed world, there are, on a regional basis, wide variations in the incidence of particular types of head and neck cancer. These differences are particularly marked for men, suggesting that many of the tumours in men might, potentially at least, be preventable. The astonishingly high incidence in some regions of France is usually attributed to excessive consumption of heated apple brandy (Calvados), which is a digestif popular in some areas of the country. The combination with heavy smoking appears to be a particularly potent carcinogenic stimulus to the hypopharynx and oesophagus.
OUTCOME The statement is often made, particularly in the US literature, and particularly when defending the role of a new and expensive intervention, that survival has changed little over the past 30 years. Two Eurocare studies show that, in Europe, the 5-year survival for men with laryngeal cancer improved from 56 per cent for patients diagnosed between 1978 and 1985 to 63 per cent for men diagnosed between 1985 and 1989 – an absolute improvement of 7 per cent and a relative improvement of 13 per cent. The corresponding survival figures for patients with tumours of the oral cavity and pharynx were 29 per cent (1978–1985) and 34 per cent (1985–1989) – an absolute improvement of 5 per cent and a relative improvement of 16 per cent. These data also show considerable survival differences between countries.3–5 It is often assumed that countries with poorer
346 Head and neck cancer
survival rates provide a lower standard of care than countries with higher survival rates, and although this may occasionally be the case, other factors will also be important, particularly for head and neck cancer. Social and behavioural factors will not only affect incidence, but they will affect survival.6 These factors are often beyond the control of even the most technically efficient services providing care for patients with cancer. The solutions to many of these disparities in outcome will be economic and social rather than medical and technological.
may be due to tobacco – chewed, sniffed or smoked. Since heavy drinkers usually smoke, and vice versa, it is difficult to assess the precise contribution of each to the aetiology of head and neck cancer. However, even if adjustment is made for smoking habits, excessive use of alcohol is still an independent risk factor, particularly for pharyngeal tumours. Tobacco and alcohol function as co-carcinogens, and so the effect of both smoking and drinking is approximately 2.5 times greater than simply adding the risk from each.
AETIOLOGY
Tobacco
The current belief is that cancers arise as a result of the interaction between the outside world and an inherent, genetically based, susceptibility. The fact that genetic factors may be important is suggested by the finding that first-degree relatives of patients with head and neck cancer have a 3.5 times higher risk of developing the disease than the general population. Environmental factors are, however, particularly important in the aetiology of cancers of the head and neck (Table 16.2). This is hardly surprising, since our interactions with the environment largely occur by way of the structures within this region: our noses sniff the air we breathe, our mouths taste the food we eat and the liquids we swallow. The two most important carcinogens for cancers of the head and neck are alcohol and tobacco. In India, 70–80 per cent of cancers of the oral cavity, oropharynx and larynx in men
It has been estimated that, in the European Community (EC), between 33 per cent and 44 per cent of deaths from cancer in men are due to smoking and/or alcohol. The higher figure is the UK estimate, the lower figure is the Danish estimate; the estimate for the EC as a whole is 39 per cent. Data from the Yorkshire Cancer Registry show that the absolute number of head and neck cancers that can be attributed to smoking is increasing: in men, from 190 per year in the 1960s to 260 per year in the 1990s; in women, from 65 per year in the 1960s to 100 per year in the 1990s. The current data suggest that 48 per cent of the tumours in men and 35 per cent of the tumours in women are related to smoking. These figures are based on the Yorkshire population of 3.6 million. Extrapolating to the UK as a whole, these data suggest that, each year, approximately 5500 people develop cancers of the head and neck that can be attributed to smoking, and that since the 1960s this figure has risen from 3900.
Table 16.2 Environmental factors potentially implicated in the cause of head and neck cancer Environmental factor Alcohol
Tobacco
Betel nut/Pan masala (a complex mixture of areca nut, tobacco, lime, cardamom, etc.) Thorium dioxide Radiotherapy for ringworm Chromium dust/fumes Leather working Nickel dust/fumes Wood dust (e.g., beech) Iron deficiency Salt fish HIV infection
Dietary factors
Cancer location Oral cavity Pharynx Oesophagus Larynx Oral cavity Pharynx Oesophagus Larynx Oral cavity
Paranasal sinuses Parotid Skin Nasal cavity and sinuses Nasal cavity and sinuses Nasal cavity and sinuses Nasal cavity and sinuses Post-cricoid carcinoma Nasopharynx Oral cavity Oropharynx
The role of dietary factors in the causation of head and neck cancer is uncertain. There are reasonable theoretical grounds for thinking that vitamin A, retinoids and betacarotene might be important in protecting against carcinogenesis in the epithelium of the upper aerodigestive tract. Dietary investigations in epidemiology, particularly those that are retrospective, are difficult: our eating habits change and recall is imperfect. There are nevertheless some data to suggest that people whose diets contain low levels of betacarotene, vitamin A or vitamin C might be at increased risk of developing cancer of the head and neck. Other dietary factors that may help prevent the disease include genistein (from soya beans) and curcumin (in curry). Dietary nitrosamines may also be important in the epidemiology of cancer of the head and neck. For example, in parts of South-East Asia, the consumption of smoked and salted fish, which is rich in carcinogenic nitrosamines, may be associated with a high incidence of the disease.
Occupational factors Occupational exposure is an important aetiological factor. The relationship between adenocarcinoma of the nose and
Strategies for prevention 347
para-nasal sinuses and employment in the furnituremaking industry is well known. However, the associations between asbestos exposure and carcinoma of the larynx, between nickel refining and squamous carcinoma of the nose and para-nasal sinuses, and between work with vulcanization processes and carcinoma of the larynx are less well appreciated.
Viruses Three main viruses have been associated with head and neck cancer: human papilloma virus (HPV); herpes simplex virus (HSV); Epstein–Barr virus (EBV). Human papilloma virus may be associated with tumours of the oral cavity, oropharynx and larynx. A systematic review of the literature7 shows that, overall, about 26 per cent of patients with squamous cancers of the head and neck have HPV DNA present within tumour biopsies. Type 16 is the most prevalent type, particularly in oropharyngeal tumours; type 18 is also important, particularly in tumours of the oral cavity and larynx. Other types of HPV are infrequently associated with squamous cancers of the head and neck. Patients with HPV-related tumours are a distinct subset of those with head and neck cancer. They are less likely to smoke or drink, are younger and more likely to have indulged in high-risk sexual behaviour (oral sex with multiple partners). The tumours tend to be basaloid or poorly differentiated. Survival is better in patients with HPV-positive tumours.8 In the future, vaccination programmes for cervical cancer may modify the incidence of HPV-related head and neck cancer. In the meantime, we are left with evidence suggesting that HPV-related squamous cancer of the head and neck is a sexually transmitted disease.9 There is a clear association between EBV and nasopharyngeal cancer, but the mechanism, if any, whereby the virus causes the tumour is unclear. The infection may be associated with loss of function of tumour suppressor genes. Herpes simplex virus, by increasing the expression and activation of pre-existing oncogenes, may act as a co-factor in the development of squamous carcinomas.
Field cancerization, a concept introduced by Slaughter in 1953,10 imposes a major constraint on the successful treatment of head and neck cancer. Slaughter’s idea was that patients who developed this cancer did so because the mucosal lining of the upper aerodigestive tract had become unstable, a hypothesis that resonates interestingly with current concepts concerning genomic instability. This instability represents the outcome, for that particular individual, of the interplay between environment and heredity. Comparison of p53 mutations between the primary tumours and the second tumours suggests that the second tumours are genetically distinct from the original tumours. This supports the concept of environmental carcinogens, such as tobacco, acting on a vulnerable target – the mucosa of the aerodigestive tract in susceptible individuals. Following successful treatment of a primary cancer of the head and neck, the risk of developing a second malignancy is relatively constant, at between 4 per cent and 6 per cent per annum. Overall, 10–30 per cent of patients will develop second malignancy. The majority of these second tumours occur in the head and neck, lung or oesophagus and, once a second tumour has developed, long-term survival is unlikely. Only 10 per cent of such patients survive for 5 years. Screening routinely for new primary tumours at each follow-up visit may detect second primaries early, but it will not prevent them. A policy of annual panendoscopy under general anaesthetic might detect second tumours at an even earlier stage, but compliance is likely to be poor. Annual chest X-ray may be worthwhile, but has not been rigorously evaluated in terms of its cost-effectiveness.
STRATEGIES FOR PREVENTION These can be divided into primary prevention, in which the aim is to prevent individuals developing head and neck cancer in the first place, and secondary prevention, in which the goal is to prevent further primary tumours in patients who have already had head and neck cancer.
Primary prevention THE PROBLEM OF SECOND PRIMARY TUMOURS There are defined criteria that need to be fulfilled before a cancer of the head and neck can be classified as a second primary, rather than simply a recurrence of the original tumour. ●
●
It must be more than 3 years since eradication of the original tumour, or the new tumour must be separated from the original tumour by at least 2 cm of normal epithelium If the second tumour is in the lung, it should be solitary and, if less than 3 years from the time of diagnosis of the original tumour, the histological appearance should be different.
Given that in the UK almost half of the head and neck cancers in men and over one-third of those in women are caused by tobacco, there is one obvious strategy: encourage people not to smoke. However, as with many cancer prevention programmes, the most vulnerable population is that which is hardest to reach, particularly if the persuasive strategies are based on cosy middle-class values as opposed to any attempt to understand what it might actually feel like to be poor, unemployed and without any hope of things getting better. There is an enormous amount literature on interventions designed to encourage people to give up smoking: there are several Cochrane Reviews of the various methods that have been employed. The results of some of these interventions
348 Head and neck cancer
Table 16.3 Summary of results from Cochrane Reviews of smoking cessation studies183–197 Intervention
Reference
Odds ratio
95% confidence interval
Nortriptylene SSRIs Bupropion Anxiolytic (buspirone) vs. placebo Nicotine replacement Nurse counselling Intensive physician advice (vs. minimal advice) Nicobrevin
Hughes, 2004 Hughes, 2004 Hughes, 2004 Hughes, 2004 Silagy, 2004 Rice, 2004 Lancaster, 2004
2.14 0.90 1.99 0.71 1.77 1.47 1.44
1.49–3.06 0.68–1.18 1.73–2.30 0.34–1.48 1.66–1.88 1.29–1.67 1.24–1.67
Stead, 2006
Exercise Aversive smoking Group behaviour therapy (vs. self-help) Individual counselling (vs. minimal contact) Competitions and incentives Telephone counselling Self-help Acupuncture
Usher, 2005 Hajek, 2004 Stead, 2005
Insufficient evidence for any conclusion 1.21 1.98 2.04
0.45–3.20 1.36–2.90 1.60–2.60
Lancaster, 2005
1.56
1.32–1.84
Hey, 2005 Stead, 2003 Lancaster, 2005 White, 2006
1.07 1.56 1.33 1.91
0.78–1.45 1.38–1.77 1.18–1.51 0.98–3.70
SSRIs, selective serotonin reuptake inhibitors.
are summarized in (Table 16.3). Even simply posting a selfhelp manual appears to be of some slight, but measurable, benefit. In its first year of operation, the Scottish telephone helpline Smokeline helped an estimated 1.5 per cent of Scottish smokers to quit – a total of between 16 700 and 22 350 adults. Recidivism is an issue: how many of those not smoking at 6 months are still abstaining 2 years or more beyond the intervention? Nevertheless, interventions designed to stop people smoking are amongst the most cost-effective measures in the whole of health care.
Secondary prevention Given the known carcinogenic effects of tobacco and alcohol, and given the concept of field cancerization, it would seem sensible to advise patients treated radically for early tumours of the head and neck to stop smoking and to cut down on their alcohol intake. Unfortunately, there is no direct proof that these measures have any immediate beneficial effect. Chemoprevention is a potentially useful strategy for preventing further tumours in patients with head and neck cancer. Chemopreventative agents are summarized in Table 16.4. Retinoids interact with specific nuclear retinoic acid receptors (RARs), which function as ligand-dependent transcription factors. There are several subtypes of receptor, the most important being designated alpha, beta and gamma. The retinoid X-receptor (RXR), again with alpha, beta and gamma subtypes, is another crucial constituent of this system. Clinical and laboratory evidence suggests that
it is the interaction with the beta-receptor that is particularly important in growth inhibition. If insufficient levels of beta-RAR are expressed, retinoic acid is unlikely to have much preventative effect on malignant change. The retinoids that have been most widely used in clinical practice are 13-cis-retinoic acid and all-trans-retinoic acid. Since each RAR subtype may be associated with specific cellular effects, agents with selectivity for each type of receptor are now being developed and tested. The use of a combined approach to chemoprevention is now being explored. Interferon and retinoids may act synergistically, possibly by inhibiting the neovascularization that is essential for tumours to grow and invade. The addition of N-acetylcysteine (NAC) is also logical, since its mechanisms of action are rather different from those of the retinoids. NAcetylcysteine acts primarily against the environmental stimuli causing malignant transformation, whereas the retinoids can affect the regulation of those intracellular processes controlling growth and differentiation. Another approach is to combine an epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (gefitinib) with a cyclooxygenase inhibitor (sulindac or celecoxib).11 Several randomized trials of chemoprevention have now been published. The evidence demonstrates clearly that approaches using retinoids alone are ineffective.12 In interpreting chemoprevention trials the important denominator is not the number of patients completing the protocol as planned, or even the number of patients entering the study; the truly important denominator is the total number of eligible patients.
General principles of surgery 349
Table 16.4 Summary of chemopreventative agents for preventing further tumours in patients with head and neck cancer Class Retinoids
Action
Drugs
Acute toxicity
Chronic toxicity
Bind to nuclear receptors and, by transactivation, modulate gene expression; promote differentiation; anti-angiogenic
Vitamin A retinal Tretinoin Etretinate (Tigason) 13-cis-retinoic acid (Roaccutane) 4-Fenretinide (HPR) 9-cis-retinoic acid: binds only to RXR All-trans-retinoic acid: binds to RAR and RXR Bexarotene
Dry skin Desquamation Dry eye Dry mucosa
Liver damage Bone remodelling Skin erythema Teratogenic
Hyper-triglyceridaemia Arthralgia Myalgia Bone tenderness Raised ICP Hypertriglyceridaemia Pancreatitis
Retinoid-x-receptor binding
Bind to RXR and inhibits tumour growth and encourages apoptosis
Carotenoids
20% converted to vitamin A; ? independent protection via the remaining 80% through inactivation of free radicals
Beta-carotene
Interferon
Anti-angiogenic; immune stimulation
Interferon-α
Tocopherols
Inactivation of nitrate carcinogens (e.g. nitrosodimethylamine); inactivation of free radicals (anti-oxidant effect)
Vitamin E
N-Acetylcysteine
Scavenges free radicals (intracellular glutathione precursor); anti-oxidant: detoxifies carcinogens extracellularly and intracellularly; inhibits formation of DNA adducts by carcinogenic metabolites; lowers the rate of mutation – both acquired and spontaneous
N-Acetylcysteine
Non-toxic
Cyclo-oxygenase inhibitors
Blocks signals involved in malignant transformation, may abrogate increase in cyclo-oxygenase activity that is associated with smoking
Sulindac Celecoxib
Diarrhoea Gastrointestinal upset
Hypothyroidism
Yellow skin
Flu-like syndrome Abnormal liver function tests Myalgia Antagonizes vitamin K, causing prolonged PT
Cardiovascular risks
ICP, intracranial pressure; PT, prothrombin time; RAR, retinoic acid receptor; RXR, retinoid X-receptor.
GENERAL PRINCIPLES OF SURGERY Full discussion of the principles and practice of head and neck surgery is beyond the scope of this chapter and is covered in other specialist texts. For the general oncologist there are, however, a few principles and practical points
to be considered. Surgery can be associated with significant long-term morbidity and should, therefore, only be undertaken in those patients who are fit enough to withstand its effects and who are aware of the likely functional outcome. Apart from diversions of the airway or digestive tract, aimed at relieving symptoms, there is little
350 Head and neck cancer
role for palliative surgery. Surgery should only be undertaken when there is a realistic chance of curing the patient. Patients undergoing major oral cavity or oropharyngeal tumour resections are usually given a temporary tracheostomy to safeguard the airway. This should be undertaken at the start or end of the main procedure. Similarly, in patients undergoing major resections in whom the likelihood of an early return to normal swallowing is small, thought should be given to undertaking a percutaneous or, if needs be, open gastrostomy. This will safeguard the airway from overspill and allow good and safe nutritional support during the early postoperative period. Having decided on major surgery, the following issues need to be considered. ●
●
●
●
●
Accurate preoperative assessment of the size of the tumour and involvement of adjacent structures (e.g. mandible). This can usually be provided by competent endoscopy and the use of appropriate computed tomography (CT) and/or magnetic resonance imaging (MRI) scanning. The extent of the excision that will be required to remove the tumour adequately. Standard texts talk of allowing a 2-cm rim of macroscopically normal tissue around the tumour. However, such a margin is not always feasible in the head and neck, and recourse to frozen-section analysis of the excision margins is often necessary at the time of surgery. Not surprisingly, positive histological margins are associated with a worse prognosis. The reliance on frozen section for histological diagnosis of the primary tumour. Although frozen-section analysis is reliable for squamous-cell carcinoma, it is notoriously unreliable for salivary-gland malignancies. Whenever possible, the histological diagnosis should be sought at the time of the initial endoscopy. Management of nodal disease. Certain primary tumour sites in the head and neck are associated with a high incidence of metastatic nodal disease, which may be too early to be clinically palpable. Such sites include the nasopharynx, tongue base, tonsil, supraglottis and pyriform fossa. It is therefore logical to consider a prophylactic ipsilateral neck dissection when undertaking excision of a primary tumour at one of these sites. The majority of surgeons dissect the neck if the primary tumour is stage T2 or above. In the absence of palpable disease, it would be appropriate to undertake a selective neck dissection and send off any suspicious-looking nodes for frozen-section analysis. A full modified radical neck dissection for clinical N0 disease is only justified if intra-operative frozen section proves positive. A combined approach, in which the primary tumour is treated radically with radiation (with or without chemotherapy) and the neck is dealt with surgically, may be useful for patients with advanced tumours of the larynx and pharynx. Reconstruction of the defect. Primary closure of a defect is sometimes possible where there is adequate mobile surrounding epithelium. However, this is not often the
case in the head and neck region. Attempts to close a defect primarily can result in tethering of nearby structures (e.g. the tongue) or subsequent wound breakdown. If the defect cannot be closed primarily, consideration must be given to the use of transposed tissue. The types of flaps or grafts available and the situations where they are most commonly applied can be summarized as follows. ●
●
●
●
Free grafts. 1. Skin: split skin grafts can be used on a wellvascularized bed, e.g. post-hemiglossectomy or post-buccal mucosa excision. 2. Bone: free bone has been employed to reconstruct a bony defect, e.g. after segmental mandibular resection. However, the bone graft is unlikely to become fully revascularized and free bone grafts are now rarely used. Local mucosal or skin flaps. 1. Random: these flaps do not have a defined vessel supplying them. They are therefore relatively small and the length of the flap must never exceed the width of its base. They can be used for small defects in the oral cavity or on the facial skin. 2. Axial: these flaps have a named blood supply. They can be cutaneous, e.g. the nasolabial flap, which is supplied by the superior labial branch of the facial artery and which can be used to reconstruct the anterior floor of mouth, or mucosal, e.g. a tongue flap, which is based on a branch of the lingual artery. Local axial flaps should be used with care in instances where the surgical field has previously been irradiated. Distant axial flaps. These are distant in that the base of the flap is not immediately adjacent to the area of excision. 1. Cutaneous: the best example of a distant cutaneous axial flap is the delto-pectoral flap. Although now rarely employed, this flap is still useful for providing neck skin cover. It is based on the perforating branches of the internal mammary artery. 2. Myocutaneous: these flaps rely on the principle that the muscle involved carries with it a specific named blood supply, which is preserved. Perforating branches can spread out into the skin overlying the muscle and allow a small paddle of skin to be preserved on the distal portion of the muscle. Such a muscle/skin flap can then be tunnelled up into the neck and the skin paddle on the muscle used to re-line the mucosal surface appropriately. The two most popular myocutaneous flaps are the pectoralis major flap and the latissimus dorsi flap. 3. Osseomyocutaneous flaps are really a variation of myocutaenous flaps and incorporate a portion of rib to provide reconstruction of the jaw. Free flaps. These are portions of tissue that carry an identifiable arterial supply and venous drainage, allowing
General principles of surgery 351
●
the tissue to be raised, the vessels divided and the flap inset into the surgical defect with anastomosis of the vessels to appropriate vessels in the neck. The need to provide a vascular supply for a free flap may compromise the extent of radical neck dissection, but there is no evidence that this has any impact on cure rate.13 The functional results following free-flap reconstructions may be excellent,14 but there are no randomized studies to prove their superiority over more traditional methods of reconstruction. Free flaps can be constructed from: 1. Skin, e.g. radial forearm flap. This flap has the advantage that it can be harvested with a vascularized piece of radius, which can be used to reconstruct a composite resection of the jaw and oral cavity. The groin flap (antero-lateral thigh flap) provides a similar type of reconstruction to the radial forearm flap.15 2. Bone, e.g. free fibular flap.16 This can be useful for mandibular reconstruction. A vascularized piece of fibula is used to bridge the defect in mandibular bone and its viability is protected by microvascular anastomosis. 3. Muscle – rectus abdominis. This flap can also be harvested with overlying skin. It provides a good bulky flap, e.g. for defects after glossectomy. 4. Jejunum. A segment of jejunum can be harvested along with its vascular arcade and can be inset to provide a neopharynx, without disturbing the lumen of the small bowel. Alternatively, the jejunum can be opened along its anti-mesenteric border and used as a ‘patch’ type of reconstruction in the pharynx. Pedicled viscera. Following a pharyngolaryngooesophagectomy (e.g. for a post-cricoid or cervical oesophageal carcinoma), the best way to reconstruct the upper digestive tract may be to mobilize and transpose the stomach into the neck. Colon can be used in a similar fashion but is often less reliable in view of its relatively poor vascular supply.
One of the major surgical developments over the past two decades has been post-laryngectomy speech restoration. There are a number of systems available, e.g. Blom Singer and Provox. Whichever device is used, the outlook for the patient after laryngectomy can be improved dramatically. The essence of speech restoration is to create a small fistula between the posterior wall of the trachea and the reconstructed pharynx. A catheter is kept in this tract to allow it to epithelialize and is then replaced by a one-way indwelling valve with flanges to keep it in place. By occluding the stoma and exhaling, the patient can divert air through the valve and into the pharynx. The slit-type valve at the pharyngeal end of the prosthesis will set the air into vibration and the patient can then articulate in the normal way and produce excellent speech. As with any medical treatment, head and neck surgery carries with it a host of potential complications, of which the following are the most relevant.
●
Fistula formation. The development of an oro-cutaneous or pharyngo-cutaneous salivary fistula can occur after any major excision involving resection of digestive tract mucosa and synchronous dissection of the neck. It is more commonly seen in patients who have previously undergone radiotherapy. The management is initially conservative and, with modern antibiotics, the risk of a subsequent carotid artery haemorrhage is very low. However, a persistent fistula should be treated with suspicion: not only is a carotid bleed possible, but also the presence of the fistula may indicate a recurrence of the original tumour.
A chylous fistula is most likely to occur following a modified radical neck dissection. Although these fistulae usually settle spontaneously, re-exploration and surgical closure are occasionally necessary. Unlike a salivary fistula, they rarely prolong the patient’s stay in hospital by many days. ●
●
●
●
Aspiration. After any major resection of the oral cavity or oropharynx, a reconstructive flap will need to be employed. This flap will be anaesthetic and immobile and, not surprisingly, swallowing will suffer as a consequence. Depending on factors such as the site of the flap, the patient’s age, previous surgery/radiotherapy, the patient may quickly learn to compensate. However, in certain instances, long-term overspill becomes a serious problem. Management can be very difficult and the patient may benefit from a permanent tracheostomy to safeguard the airway. Following total glossectomy a minority of patients will require salvage laryngectomy purely in order to prevent chronic aspiration. Wound breakdown. This is rare unless there is an associated fistula. Facial oedema. Significant facial oedema is unusual unless synchronous neck dissections are undertaken with removal of both internal jugular veins. Because of this and the associated problem of suddenly raised intra-cranial pressure, every attempt should be made to preserve at least one vein. Frozen shoulder. This is often seen following the conventional (Crile) radical neck dissection, in which the accessory nerve is divided. Postoperative radiotherapy makes the problem even more likely to occur. The ‘solution’ is to preserve the axillary nerve whenever possible or to provide physiotherapy in the postoperative period.
The development of techniques in tissue engineering17 and transplantation will revolutionize the approach to surgical reconstruction in patients with head and neck cancer. By using bio-compatible matrices and growth factors, it is possible to encourage, either in vitro or in vivo, normal tissues to reconstitute the anatomy of tissue that has been destroyed or removed. The recombinant form of bone morphogenetic protein (rhBMP) is commercially available and, once a suitable scaffold has been provided, may be used to help repair bone removed during head and neck surgery.
352 Head and neck cancer
The recent development of facial transplants has attracted wide publicity but, quite apart from ethical and psychological dilemmas,18,19 because of problems with immunosuppression and concerns about increasing the risks of cancer recurrence, it will probably have little role to play in the management of head and neck cancer.
neck are summarized in Table 16.5. An important feature of these data, obtained from 11 separate centres, is that significant differences emerged when the results were stratified by centre. Only two of the 11 centres reported Tpot values of over 45 days; at five of the centres the maximum reported Tpot value was less than 20 days. This clearly demonstrates that, even in expert hands, there is considerable variation in the estimation of these parameters. There was no clear or consistent relationship between any of the three parameters and T stage, site or histological grade. The median Tpot was 6.4 days for well-differentiated tumours and 7.8 days for undifferentiated tumours. These Tpot values are much shorter than the volume doubling times. The discrepancy between the two parameters can be explained by cell loss:
GENERAL PRINCIPLES OF RADIOTHERAPY Kinetics of squamous-cell carcinoma of the head and neck The volume doubling time of squamous-cell carcinomas of the head and neck is between 40 and 80 days. The potential doubling time can be calculated from
θ 1 (Tpot /Td)
Tpot λ (Ts/LI)
where θ is the cell loss factor and Td is the cell-population doubling time, which will usually be longer than the volume doubling time. Cell loss by desquamation is part of the repertoire of normal squamous epithelium. Cell loss factors of around 95 per cent are characteristic of squamous-cell carcinomas of the head and neck. These high cell loss factors have important implications for the response to therapy. A l cm diameter tumour contains approximately 109 cells; without cell loss, this size would be achieved within 30–35 generations. With cell loss, however, it could take up to 1000 generations or more to reach this size. A small tumour of the head and neck is therefore genetically old (Fig. 16.1). Mutations conferring resistance to radiation and/or drugs will occur randomly and may precede exposure to the toxic agent. The chance of such a mutation occurring depends on the number of cellular divisions that it has taken a tumour to reach a given size, rather than the absolute number of cells
where LI is the labelling index, λ is a constant describing the age distribution of the population (values are typically in the range 0.75–1.0) and Ts is the duration of DNA synthesis. The results of a comprehensive review of kinetic parameters in 476 squamous carcinomas of the head and
Table 16.5 Data on tumour kinetic parameters from 476 squamous carcinomas of the head and neck.46 Parameter Ts (hours) LI (%) Tpot (days)
Median
Range
10.7 8.85 5.1
4.4–45.7 0.6–47.7 0.8–72.9
LI, labelling index; Ts, duration of S phase; Tpot, potential doubling time.
Number of generations
1200 1000 800 600 400 200 0 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Cell loss (phi) NO LAG
10 LAG
20 LAG
Figure 16.1 The relationship between cell loss factor (phi) and the number of generations it would take tumour to reach 109 cells (corresponding to a diameter of about 1 cm). Three different conditions are modelled: NO LAG, cell loss is present from the initiation of the tumour; LAG 10, cell loss only kicks in after 10 generations; LAG 20, cell loss kicks in after 20 generations. For clinically relevant values for cell loss factor (phi in the range 0.8–0.99) it could take up to 1000 generations for a tumour to reach a size of 1 cm, ample opportunity for mutations to accumulate.
General principles of radiotherapy 353
within the tumour. Cell loss thereby increases the probability of a mutation conferring resistance having occurred before diagnosis. Successful treatment may be jeopardized by events that have taken place during the pre-clinical phase of tumour development. Cell loss will exaggerate the response to treatment observed clinically. Modest reductions in the number of clonogenic cells present will, through the amplificatory effect of cell loss, produce rapid shrinkage of tumour. The underlying problem is the presence of resistant cells and these will grow through treatment, imperceptibly at first, and ultimately dominate. The failure of chemotherapy, in spite of impressive initial responses, significantly to improve outcome in head and neck cancer may, in part, be related to the kinetic behaviour of squamous-cell carcinomas and its genetic consequences (Fig. 16.2).
into the problems posed by the oxygen effect will be more broadly based than in the past. Oxygen levels within tumours of the head and neck Several studies have now reported on the direct measurement, using polarographic electrodes, of oxygen levels in head and neck tumours. A significant proportion of the cells within squamous cancers of the head and neck are hypoxic. The distribution of median PO2 values in tumours of the head and neck regions is completely differ-
Box 16.3 Methods that could be used clinically to circumvent the problems caused by the relative radioresistance of hypoxic cells within tumours
Radiobiology of head and neck cancer
Increase oxygen delivery to tumour ● Hyperbaric oxygen ● Normobaric oxygen ● Nicotinamide ● Modify blood viscosity ● Increase haemoglobin ● Lower interstitial fluid pressure within tumour
Clinical research into the radiotherapy of head and neck cancer has, historically, been dominated by the oxygen effect, and it is only recently that the importance of other radiobiological factors has been rediscovered. THE OXYGEN EFFECT
Increase killing of hypoxic cells Hypoxic cell sensitizers (e.g., nimorazole) ● Bioreductive drugs (e.g., mitomycin C, tirapazamine) ● High linear energy transfer (LET) radiation (e.g., neutrons, light ions) ●
Hypoxic but viable cells may exist within tumours. These cells are relatively resistant to radiation-induced cell killing and may also be relatively resistant to cytotoxic chemotherapy. The survival of such cells after a course of treatment might limit cure. Three main methods have been used to attempt to overcome the potential problem of hypoxic cells limiting cure in the radiation treatment of head and neck cancer: hyperbaric oxygen, neutron therapy and hypoxic cell sensitizers. There are other approaches to the problem (summarized in Box 16.3) and future research
Decrease oxygen demand within tumour Hypothermia ● Calcium channel blockers ● Meta-iodobenzylguanidine (MIGB) scintiscan ● Crabtree effect (acute hyperglycaemia) ●
1000 000 000 100 000 000 Sensitive clonogens
Cell number
10 000 000
Resistant clonogens
1 000 000 100 000 10 000 1000 100 10 1 0
50
100
150
200 Time
250
300
350
400
Figure 16.2 When cell loss is high (in this example phi 0.9) treatment will produce a rapid response as the sensitive clonogens are killed and the effect is exaggerated by the high cell loss. However, the resistant clonogens proliferate unperturbed and the tumour regrows – a pattern characteristic of squamous carcinoma of the head and neck treated with chemotherapy.
354 Head and neck cancer
ent from that of the adjacent normal tissues: the median PO2 in normal tissues follows a Gaussian (normal) distribution, with no values below 20 mmHg. The distribution of the median PO2 values from tumours is skewed to the left, with the majority of patients having tumour median PO2 values of less than 20 mmHg. It is probably the absolute number of hypoxic cells within a tumour, rather than the proportion, that is important. The concept of the ‘hypoxic subvolume’ has been introduced to describe this. The hypoxic subvolume is simply the product of the tumour volume the hypoxic fraction (percentage of cells with PO2 values below threshold, e.g. 5 mmHg). In a multivariate analysis in patients with squamous-cell carcinoma of the head and neck, the hypoxic subvolume was an important and independent predictor of outcome.20 Large tumours may have a small hypoxic subvolume, small tumours may have a large hypoxic subvolume – it is the size of the hypoxic subvolume, rather than the size of the tumour itself, that appears to be important. Both the absolute level and the duration of hypoxia are important: blood ebbs and flows through the abnormal vasculature of a tumour and it is difficult to separate out the confounding effects of chronic hypoxia, acute hypoxia and vascular anatomy. A crucial question is whether or not, during a course of fractionated radiotherapy, hypoxic cells are able to re-oxygenate. Hypoxic cells will be relatively radioresistant and any such cells persisting after radiotherapy will compromise cure. If, however, fractionation permits all cells to regain normal oxygenation, then all cells, ultimately, will become vulnerable to radiation and it should be possible to eliminate every last clonogenic cell. Re-oxygenation may occur during fractionated radiotherapy and this may be associated with improved response.21
Hypoxic cell sensitizers are still under active investigation in head and neck cancer. Initial studies used metronidazole and misonadazole, but cumulative neurotoxicity caused by the sensitizer proved limiting. These early studies showed no convincing benefit from the addition of the radiosensitizer. This failure to show benefit may have been due in part to the use of reduced doses of sensitizer dictated by the need to avoid excessive toxicity. Two large-scale, randomized studies have assessed the effectiveness of etanidazole (SR 2508) as a radiosensitizer. Although the sensitizer was well tolerated, it appeared ineffective in improving either local control or survival. The DAHANCA-5 study (Danish Head and Neck Cancer Study,22 using nimorazole as the radiosensitizer, showed a significant improvement in cause-specific survival, with nimorazole radiation compared to placebo radiation 52 per cent versus 41 per cent at 5 years (p 0.01; odds ratio 1.92, with 95% CI 1.30–2.84). For the moment at least, the most clinically useful radiosensitizer appears to be nimorazole; unfortunately, it is not widely available. Combined approaches have been used clinically in an attempt to circumvent the oxygen effect. The ARCON schedule uses accelerated radiotherapy (AR) carbogen (CO) and nicotinamide (N). Carbogen is a mixture comprising 95 per cent oxygen and 5 per cent carbon dioxide and is a means of increasing oxygen delivery to tissues whilst minimizing the vasoconstrictive effects of high concentrations of oxygen. Nicotinamide acts as a vasodilator, and so the ARCON schedule aims to give a rapid course of radiotherapy to a tumour that has an excellent supply of oxygen-rich blood. A large Phase II trial has shown that the technique is both feasible and effective, particularly for locally advanced laryngeal tumours.23
Clinical studies
HAEMOGLOBIN AND OXYGENATION
Pooling the results from ten randomized trials of hyperbaric oxygen therapy in head and neck cancer shows an absolute difference in survival of 8.6 per cent (95% confidence interval (CI), 1.3–15.9 per cent) in favour of hyperbaric oxygen treatment. This difference is small, hyperbaric treatment is both cumbersome and potentially dangerous, and the technique is no longer used in the UK. Fast neutrons kill hypoxic cells as efficiently as they kill well-oxygenated cells, and neutron therapy appeared to offer a logical means of circumventing the problems posed by the oxygen effect. The pooled results from seven randomized trials of neutron therapy in head and neck cancer show no survival benefit: the relative risk of dying was 0.98 (95% confidence interval, 0.88–1.09) with neutron therapy, and there was no significant benefit in local control. The relative risk of local failure was 0.87 (95% CI, 0.66–1.13) with neutrons. Neutron therapy was associated with a statistically significant increase in serious complications: relative risk 1.73 (95% CI, 1.08–2.78). The early promise of this form of treatment has not been fulfilled. The late damage to normal tissues caused by neutrons is excessive and abrogates the benefit from any increase in local control that might have been achieved.
There is some evidence that haemoglobin level before, and during, radiotherapy may influence therapeutic outcome in patients with head and neck cancer. The problem with interpreting the extensive literature on this topic is that different studies have used different populations of patients, different endpoints, different definitions of anaemia and different time-points for defining the presence of anaemia.24 One important finding is that simply increasing haemoglobin level may be counterproductive; there is an optimum haemoglobin range for the oxygenation of squamous carcinomas – 13–15 g/dL in men, 12–14 g/dL in women. In the belief that anaemia was an independently adverse prognostic factor in patients with head and neck cancer, a series of clinical trials investigated the question of whether using erythropoietin to correct anaemia would improve outcome. The evidence suggests that the reverse is the case. A trial published in 200325 reported increased loco-regional progression and lower survival in patients randomized to erythropoietin radiotherapy as opposed to radiotherapy alone. It is possible that tumour cells express erythropoietin receptor and the erythropoietin acted as a growth factor for the tumour as well as for the red-cell precursors.26
General principles of radiotherapy 355
Anaemia and oxygenation are related. An international review of experience using polarographic micro-electrodes to measure tumour oxygen levels in patients with head and neck cancer showed that tumour hypoxia had a major influence on survival.27 As the percentage of PO2 values 2.5 mmHg (HP25) rose above 20, survival steeply decreased – from a baseline of around 40 per cent to around 10 per cent for a patient with an HP25 of 75 per cent. BIOREDUCTIVE DRUGS
Bioreductive drugs, unlike hypoxic cell sensitizers, are cytotoxic in their own right. They are, however, selectively toxic to poorly oxygenated cells. Mitomycin C is a bioreductive drug that has been used in cancer treatment, and particularly in combination with radiotherapy, for many years. Tirapazamine, porfiromycin, AQ4N and BMS-181174 are more recently introduced bioreductive agents. The rationale behind the use of bioreductive drugs synchronously with radiotherapy is that there may be true synergy between the two treatments: there may be a supra-additive effect. Tirapazamine has been used synchronously with radiotherapy for the treatment of head and neck cancer in two Phase II studies. Pooling the results of randomized trials comparing radiotherapy alone with radiotherapy mitomycin C shows that the addition of the bioreductive drug produces a small but significant improvement in survival: median survival is 4 months longer (95% CI 1.6–6.9 months) with mitomycin C.28 A randomized trial performed at Yale compared porfiromycin (the methylated derivative of mitomycin C) radiotherapy with mitomycin C radiotherapy. The porfiromycin was significantly inferior in every respect.29
the decrease in local control: 3–7-day gap, local control 37.5 per cent; more than 15-day gap, local control 29.5 per cent.33 A comprehensive analysis, using pooled data from Manchester (3-week schedule) and Toronto (5-week schedule), of patients treated radically for T2 or T3 carcinomas of the larynx again demonstrated the deleterious effect of unplanned gaps during treatment. Each extra day added to the overall treatment time corresponded to a decrease in effective dose of between 0.6 and 0.8 Gy.34,35 It is possible to compensate for unplanned gaps in treatment by keeping the overall treatment time within the desired schedule, by treating either at weekends or twice per day. If this is logistically impossible, a less satisfactory alternative is to use a radiobiological model, such as the L-Q formula, as a guide to an upward adjustment of total dose.36 Various fractionation strategies have been devised to deal with the problem of accelerated repopulation. If accelerated repopulation is important, you must either keep treatment as short as possible or, if using longer treatments, increase the total dose to compensate for proliferation during treatment. You cannot simply increase the dose per fraction, because the combination of increased dose per fraction and the increased total dose will produce unacceptably severe late effects. One solution to this dilemma is to use multiple daily fractions. Another strategy is to shorten treatment and use multiple daily fractions. Consequently, three main strategies for unconventional fractionation have emerged over the past decade. Unfortunately, different authors have used different criteria for each of the main categories. ●
FRACTIONATION
Radiotherapists, and particularly head and neck radiotherapists, have recently been re-exploring fractionation as a means of improving the therapeutic ratio for radiation therapy. Much of the impetus has come from the realization that tumour kinetics may be perturbed by treatment and, in particular, that tumour cells may proliferate more rapidly as a response to treatment.30 Whether this represents a true acceleration or simply the unmasking of the latent Tpot (of 5–10 days) is debated.31 Other mechanisms, such as a decrease in cell-cycle time and a shift towards more symmetrical division of stem cells,32 may also be involved. There is no doubt that the phenomenon is clinically important. Even short unanticipated gaps in clinically effective schedules can significantly reduce local control and/or survival. A series of 971 patients with supraglottic laryngeal tumours treated in Gliwice, Poland, using a standard 6–7-week schedule showed that if there were no unplanned interruptions to treatment, local control was 52 per cent. In patients with any gap, defined as an unplanned interruption of treatment, the local control rate was 40 per cent (p 0.014). The longer the gap, the greater
●
●
Pure hyperfractionation: overall treatment time is neither shortened nor prolonged and total dose is increased by using multiple fractions per day. The interval between fractions should be at least 6 hours to allow time for repair in damaged, but viable, cells of the normal tissues. If the interval between fractions is any shorter than this, repair is incomplete and the damage to normal tissues may be unacceptable. Repair half-times, particularly in the spinal cord, may be longer than originally thought and some repair processes may take more than 6 hours. Pure acceleration: overall treatment time is shortened, either by using multiple fractions per day or by treating at weekends. The size of each individual fraction is usually 1.8 Gy or greater. Accelerated hyperfractionation: overall treatment time is shortened and multiple fractions per day are used. The fraction size is usually less than 1.8 Gy. The reduction in treatment time achieved with this approach may, at least in part, be offset by the need to introduce a gap in treatment to permit recovery of the normal mucosal cells. The use of the concomitant boost is a variation on this theme: a large tumour volume is treated with a conventional schedule but, during the latter phase of treatment, a supplementary treatment is given each day to a smaller boost volume. This boost is given at least 6 hours after the treatment to the large volume.
356 Head and neck cancer
The schedules employed in those randomized trials of fractionation for which data are currently available are summarized in Table 16.6. The MARCH group, using meta-analysis of individual patient data, has shown that unconventional fractionation, particularly with regimens using pure hyperfractionation, provides a modest benefit in survival and local control.37 Regimens that combine hyperfractionation with acceleration appear to offer no gain in therapeutic ratio; in fact, the reverse may be true: an increase in acute effects without any increase in locoregional control or survival (Table 16.7; Box 16.4). The definition of a ‘conventional’ fractionation schedule for treating cancers of the head and neck is by no means straightforward. The phase III trials have usually adopted schedules lasting 6 weeks or more as the ‘conventional’ arm. However, this choice is not in accordance with the actual routine clinical practice of many radiotherapists in the UK,38,39 as many clinical oncologists use radical schedules of 25 fractions or fewer for treating head and neck cancer.40 The available randomized studies have therefore failed to address questions of importance to most British radiotherapists (and health economists everywhere). If accelerated proliferation of tumour clonogens does not begin until the third week of radiotherapy, simply using a treatment schedule no longer than 4 weeks might be sufficient to prevent the problem of accelerated repopulation. It is perfectly feasible to treat in 4 weeks or less using single daily fractions; typical regimens use 50–55 Gy in 20 fractions over 4 weeks at 2.5–2.75 Gy per fraction. Because there was concern that shorter schedules might be associated, through the use of relatively high doses per fraction, with unacceptable late morbidity, the British Institute of Radiology performed a randomized comparison of short ( 4 weeks) versus long (4 weeks) treatment times in cancer of the laryngopharynx. The results of this large study show that tumour control was the same in the two arms but the longer treatment schedules actually produced a higher incidence of late reactions. For patients treated with five fractions per week, only 22 6 per cent of patients treated on the longer schedules were free of late reaction for 10 years. The corresponding figure for the shorter schedules was 49 4 per cent (p 0.008).41 This result is not as surprising as it at first seems. Total dose is just as important as dose per fraction in determining late effects. If, in order to counteract tumour proliferation during treatment, longer regimens require higher total doses, then keeping dose per fraction to 2 Gy or less will not necessarily protect against the development of late effects (see Table 16.8).
maximal in the evening; cells are at their most radiosensitive at the G2/M boundary. The proportion of normal mucosal cells at the G2/M boundary is maximal between 6 p.m. and 8 p.m. This suggests that mucositis may be less severe for patients treated in the morning. If tumours lose their circadian rhythms, this would suggest that treatment in the morning might improve the therapeutic ratio for patients irradiated for cancers of the head and neck. Some of the increase in mucositis that is observed with schedules using multiple fractions per day may be explained on the basis of circadian variations in the radiosensitivity of the normal mucosa. The results of the Canadian trial have only been reported in abstract form,42 but show no compelling advantage of morning treatment. The challenge for the future in fractionated radiotherapy is to make the punishment fit the crime: to identify before treatment those factors, such as differentiation and EGFR expression43 or molecular profiling,44 that influence the outcome of treatment and to treat with appropriately designed fractionation regimens. PREDICTIVE ASSAYS AND RADIOTHERAPY
Rational radiotherapeutic treatment for head and neck cancer is dictated by the underlying radiobiology. Ideally, for each tumour we treat, we would like to be able to estimate the classic radiobiological parameters45 not just for the tumour but also for the normal tissues: what is the intrinsic radiosensitivity, the ability to repopulate, the ability to re-oxygenate, the ability to repair and the ability to redistribute around the cell cycle? The intrinsic radiosensitivity of tumours in patients with cancers of the head and neck has been assessed using SF2 (surviving fraction after 2 Gy). The technique takes about 4–5 weeks to produce a result and therefore to use SF2 as a guide to choosing treatment for an individual patient would involve delaying the start of treatment by a month or so. This is rarely acceptable, either to patients or to clinicians. Most of the investigation of this approach to choosing therapy has been confined to modelling, in which the theoretical benefits to a population of patients – were there feasible assays of intrinsic radiosensitivity for both tumours and normal tissues? – are explored. Even the modellers disagree, some suggesting that there is no overall gain from predictive testing for individual patients, others believing that there are genuine gains to be made. Unfortunately, the attempts to use Tpot to predict response to radiotherapy have, so far, failed to deliver a clinically useful strategy.46
The proliferation of tumours and normal tissues follows circadian rhythms
Complications of radiotherapy
The National Cancer Institute of Canada coordinated a trial (NCIC HN3) in which the randomization was simple: single daily treatments given between 8 a.m. and 10 a.m. versus single daily treatments given between 4 p.m. and 6 p.m. The rate of division of normal mucosal cells is
The acceptance of a degree of damage to normal tissues is implicit within the basic principles that underlie the use of radiotherapy to treat malignant disease. This problem is brought sharply into focus in the treatment of cancers of
Table 16.6 A summary of fractionation schemes used in randomized trials of fractionation in head and neck cancer.198–219
Author Sanchiz Fu Teo Horiot Cummings Datta Pinto Fu Fu Awwad Awwad Vab den Bogaer Dische Horiot Dobrowsky Olmi Maciejewski Jackson Overgaard Poulsen Suwinski Sanguineti Ezzat Hliniak
Conventional regimen
Experimental regimen
Place/ organization
Year
Type
Size
Dose
#
dpf
Time
EQ2(3)
EQ2(10)
Dose
#
dpf
Time
EQ2(3)
EQ2(10)
Barcelona RTOG 9003 hfx Hong Kong EORTC 22791 PMH Toronto India Rio de Janeiro RTOG 9003 afx s RTOG 9003 afx c Cairo Cairo EORTC 22811 MRC chart EORTC 22851 Vienna Italian multicentre Gliwice Vancouver Dahanca 6&7 Trans tasman Gliwice Genoa Cairo Warsaw
1990 1999 1996 1992 1996 1989 1991 1999 1999 1992 1998 1995 1997 1997 1998 2003 1996/8 1997 1998 1999 2005 2005 2005 2002
hf hf hf hf hf hf hf ha ha ha ha ha ha ha ha ha a a a a a a a a
559 556 100 325 336 176 98 556 556 56 72 348 918 512 127 192 100 82 540 191 175 226 60 395
60 70 60 70 51 66 66 70 70 50 60 75 66 70 70 70 72 66 68 70 63 60 68 66
30 35 24 35 20 33 33 35 35 25 30 44 33 35 35 35 36 33 34 35 35 30 34 33
2 2 2.5 2 2.55 2 2 2 2 2 1.7 2 2 2 2 2 2 2 2 2 1.8 2 2 2
42 49 30 49 26 45 45 49 49 35 42 63 45 49 49 49 55 48 45 47 42 42 45 45
60.0 70.0 66.0 70.0 56.0 66.0 66.0 70.0 70.0 50.0 60.0 70.5 66.0 70.0 70.0 70.0 72.0 66.0 68.0 70.0 60.5 60.0 68.0 66.0
60.0 70.0 62.5 70.0 53.3 66.0 66.0 70.0 70.0 50.0 60.0 73.1 66.0 70.0 70.0 70.0 72.0 66.0 68.0 70.0 62.0 60.0 68.0 66.0
70.4 81.6 71.2 80.5 58 79.2 70.4 67.2 72 42 46.2 72 54 72 55.3 67.2 72 66 68 59.4 63 64 68 66
64 68 45 70 40 66 64 42 40 30 33 45 36 45 35 42 36 33 34 33 35 35 34 33
1.1 1.2 1.6 1.15 1.45 1.2 1.1 1.6 1.8 1.4 1.4 1.6 1.5 1.6 1.6 1.6 2 2 2 1.8 1.8 1.82 2 2
42 49 30 49 26 45 45 42 42 11 14 49 12 35 17 39 36 25 39 24 35 35 38 38
57.7 68.5 35.5 66.8 51.6 66.5 57.7 61.8 69.1 37.0 40.7 66.2 48.6 66.2 50.9 61.8 72.0 66.0 68.0 57.0 60.5 61.7 68.0 66.0
65.1 76.2 68.8 74.8 55.3 73.9 65.1 65.0 70.8 39.9 43.9 69.6 51.8 69.6 53.5 65.0 72.0 66.0 68.0 58.4 62.0 63.0 68.0 66.0
a, pure acceleration; dpf, dose per fraction; EQ2(3), equivalent dose in 2 Gy fractions assuming alpha/beta 3 Gy; EQ2(10), equivalent dose in 2 Gy fractions assuming alpha/beta 10 Gy; ha, hyperfractionation with acceleration; hf, hyperfractionation; time, treatment time in days; #, fraction number.
Table 16.7 Summary of results of meta-analyses of studies of chemotherapy and/or radiotherapy fractionation in head and neck cancer Group
Comparison
Endpoint
Any altered fractionation vs. conventional n 7073
Death
Statistical test
% absolute benefit
0.87–0.97, 95% C.I.
3
0.87–0.97, 95% C.I.
7
p 0.0001 p 0.0001
5 8
Fractionation trials (MARCH) [Pignon 2005]220
Death LRF Death LRF Death LRF
HR 0.92 RR 0.91 HR 0.82 RR 0.82 RR 0.78 RR 0.76 RR 0.96 RR 0.80 RR 0.93 RR 0.91
Any chemotherapy vs. none n 16 000 Concomitant RT-CT Concomitant postoperative RT-CT Concomitant postoperative RT-CT Concomitant altered fractionation RT-CT Concomitant with platinum vs. concomitant without platinum
Death Death Death Death Death Death
HR 0.88 HR 0.81 HR 0.80 HR 0.83 HR 0.73 Advantage to platinum
p 0.01
Concomitant RT-CT vs. RT alone (RT schedules identical in both arms) n 2197 Concomitant altered fractionation RT-CT 5-FU-containing regimen vs. RT alone Cisplatin containing regimens vs. RT alone Carboplatin containing regimens vs. RT alone MMC containing regimens vs. RT alone Accelerated RT vs. conventional RT n 4702 Hyperfractionated RT vs. conventional RT n 1523
Survival at 2 years
13.3% higher with RT-CT
9.8–16.3%, 95% C.I., p 0.0001
Survival at 2 years Median survival gain Median survival gain Median survival gain
14.7% higher with RT-CT 24 months 16 months 7 months
8.9–20.3%, 95% C.I., p 0.0001 18–31 months, 95% C.I. 12–21 months, 95% C.I. 4–10 months
Median survival gain Survival at 2 years Survival at 2 years
4 months 2 to 7 months, 95% C.I. 1.3% greater with accelerated 1.5–4.0%, 95% C.I., p 0.05 12% greater with 7.7–15.9%, p 0.0001 hyperfractionated
LRF Hyperfractionated vs. conventional n 1350 Accelerated vs. conventional Very accelerated vs. conventional Chemotherapy Trials (MACH-NC) [Bourhis jco 2004]221
Modern era drugs and radiation protocols [Budach 2006]222 Chemoradiation vs. RT alone
Altered fractionation
C.I., confidence interval; CT, chemotherapy; 5-FU, fluorouracil; HR, hazard risk; LRF, loco-regional failure; MMC, mitomycin C; RR, relative risk; RT, radiotherapy.
General principles of radiotherapy 359
the head and neck. A painful, unpleasant, acute mucosal reaction almost invariably accompanies the use of radiation to treat these tumours. Because of the differing kinetics of the cells concerned, this mucosal reaction precedes any cutaneous reaction. ACUTE REACTIONS AND THEIR MANAGEMENT
The sequential stages in the development of a mucosal reaction are hyperaemia with erythema and oedema, and the formation of an exudate (membrane), which is at first patchy but becomes confluent; with a severe reaction, the
Box 16.4 Conclusions of the MACH-NC individual patient-based meta-analysis of chemotherapy studies in head and neck cancer* ●
●
●
●
●
● ● ●
Concomitant radiotherapy/chemotherapy (RT-CT) is more effective than neoadjuvant or adjuvant chemotherapy Overall survival at 5 years is better with platinumbased regimens Multiple drug regimens are no better than single agent regimens The benefit from concomitant RT-CT is independent of fractionation scheme (conventional, or altered) The benefit from concomitant RT-CT applied to postoperative irradiation The benefit from concomitant RT-CT declines with age The benefit from altered fractionation declines with age Concomitant RT-CT increases both acute and late toxicity
*Summary of conclusions from MACH-NC and MARCH meta-analyses223
mucosa may ulcerate, and in extreme cases never heals. When treatment is fractionated over 4 weeks, reaction usually starts during the third week, is at its peak during the fifth week and has abated by the sixth week. With longer courses of treatment, the reaction arises later and lasts longer. The severity and duration of the acute reaction are dictated by the rate of dose accumulation: if too much dose is accumulated too fast, the reaction will not heal. Evidence for this comes from a randomized trial carried out in Gliwice, Poland.47 Patients treated with 2 Gy per fraction, 7 days per week, to a total dose of 72 Gy over 36 days had a 21 per cent incidence of mucosal ulceration 3 months after treatment. No such reaction would be expected were the same dose to be given in the same number of fractions over 49 days. Patients should be warned of the acute mucosal reaction that they should expect, otherwise they may assume that the symptoms are due to the progression of their tumour. Patients should be seen and examined at least once a week during treatment. Careful supervision of oral hygiene is important, as is prompt identification and management of any secondary infection, the most common of which is oral candidiasis. Patients can take a considerable share of the responsibility for their own care during a course of radiotherapy, provided they are given adequate instruction. Simple measures will often reduce the duration and severity of the acute mucosal reaction; the most important steps are to stop smoking and to avoid alcohol. Advice is best given as a set of written instructions, which are carefully explained to patients at the start of treatment and reinforced during treatment. The major points are summarized in Box 16.5. In addition to these general measures, several specific interventions are now being investigated. A variety of measures have been used as prophylaxis against oral mucositis. There is evidence from randomized trials suggesting benefit from amifostine,48 zinc sulphate,49 and recombinant human keratinocyte stimulating factor (palifermin.50 There is evidence suggesting that the following
Table 16.8 Structures in the head and neck that are vulnerable to the adverse late effects of radiotherapy BED (α/β 3 Gy)
Total dose at 2 Gy pf
Total dose at 2.75 Gy pf
Sequel
Lens Retina Optic nerve Anterior eye VIIIn Cranial nerves Temporal lobe Brain stem
11 88 99 117 108 108 92 88
6 52 59 70 65 65 55 52
5.5 46 52 61 57 57 48 46
Spinal cord Temporal bone
75 107
45 64
39 56
Cataract Loss of vision Loss of vision Dry eye Hearing loss CN palsy Epilepsy dementia Dysarthria, nystagmus, disturbed consciousness Myelopathy Osteonecrosis
Tissue
BED, biologically equivalent dose; CN, cranial nerve; pf, per fraction. The doses given indicate a level at which approximately 5 per cent of patients might be expected to experience the complication.
360 Head and neck cancer
Box 16.5 Instructions for patients’ self-care during radiotherapy to the head and neck Skin care ● Avoid aftershave or astringent cosmetics ● Wash gently, pat the skin dry ● Avoid wet shaving ● Avoid collars or other clothes that chafe the treated area ● Try not to scratch the skin Mucosal care Stop smoking ● Do not drink alcohol ● Avoid hot or spicy foods ● Use regular mouthwashes ● Carry out careful dental hygiene ● Chew sugar-free gum ● Drink plenty of clear fluids ●
Diet and nutrition Follow a bland high-protein diet ● Use dietary supplements ● Take supplementary vitamins ● Put food through blender if necessary ● Avoid constipation ●
Care of the voice Don’t talk too much ● Rest voice as much as possible ● Don’t try to force the voice ● Speak slowly and quietly ● Avoid dry or smoky atmospheres ● Reduce background noise ● Avoid excessive coughing and hawking ●
interventions may not be of benefit: antibiotic lozenges,51 aloe vera,52 and isegenan hydrochloride.53 There is some evidence of benefit from low-level laser therapy,54 N-acetylcysteine in a polymer matrix,55 and tetracaine gel.56 All these results have to be interpreted with caution: the negative trials may have been too small to show benefit and publication bias may affect the positive trials. A wide variety of topical preparations are available for the treatment of the established mucosal reaction. Systemic analgesia, up to and including opiates, will be required for severe reactions. Patients will normally lose some weight during treatment. If this becomes excessive (greater than 15 per cent of initial body weight), admission to hospital and feeding via a gastrostomy or fine-bore nasogastric tube may be necessary. The timing of percutaneous gastrostomy is important. Not all patients will develop nutritional problems, and therefore a policy of routine gastrostomy is not appropriate. Early, rather than delayed, feeding is better at maintaining nutritional state.57 However, an analysis of the results from the RTOG (Radiation Therapy Oncology
Group) 2003 fractionation trial shows that survival and local control were worse in patients who had been given nutritional support before treatment. This effect was independent of T stage, N stage or performance status.58 Evidence from other studies suggests that an aggressive approach to nutritional support may feed the tumour at the expense of the patient.59,60 When the larynx is within the treated area, speech therapy, for expert advice on the care and preservation of the voice, is advisable. The acute effects of radiation treatment, although unpleasant and distressing, are usually self-limiting. Only rarely are serious problems encountered. The dose that can be given to treat a tumour is limited by the late effects on normal tissues rather than by the acute reactions. The distinction between early and late effects of radiation is arbitrary but useful – arbitrary in that the temporal distinction cannot be universal for all tissues, and useful in that it draws attention to the possible differences in pathogenesis for the two types of response. The shape of the initial portion of the cell-survival curves for acutely responding normal tissues differs from that for the cells of late-responding tissues. The difference in shape implies that increasing the number of fractions will selectively spare the late-responding tissues. Another implication is that the intensity of the acute reaction cannot be used to predict the severity of the late effects. LATE EFFECTS OF RADIATION
The late effects of a course of radiotherapy will depend on which vulnerable structures are within the radiated volume. The structures of relevance to the radiation treatment of carcinoma of the head and neck, the clinical consequences of radiation damage, and an indication of the limiting dose are summarized in Table 16.8. In general, the late complications of radiotherapy are more severe in patients who have had surgical procedures in addition to radical radiotherapy. The salivary glands The effects of radiation on salivary glands are complex and straddle the conventional division between early and late effects. The incidence and severity of the effects of radiation on the function of the salivary glands depend on the amount of glandular tissue within the radiation field as well as on time, dose and fractionation. Shielding even a proportion of one parotid gland may preserve a considerable proportion of salivary flow. Since a chronically dry mouth is unpleasant for the patient, and can be associated with rapidly progressive dental caries, it is important to exclude as much of the salivary glands as possible from the irradiated volume. Radiation-induced parotitis, manifest as acute swelling and inflammation of the salivary glands, can occur during the first few days of treatment. The syndrome itself is self-limiting. No specific management, other than reassurance, is required.
General principles of radiotherapy 361
A decrease in the rate of flow of saliva occurs within 10–14 days of the start of treatment, a timing that suggests that cell death by apoptosis may be an important mechanism whereby radiation damages the salivary glands. The effects of radiation on salivary function are at their maximum 4–6 weeks after the start of treatment. The serous acini are more sensitive to radiation than the mucinous acini, which means that the saliva changes and becomes thick and sticky. The return of salivary function is slow; it can take several years and is almost invariably incomplete. Chronic severe xerostomia can be one of the most distressing long-term side effects of treatment for head and neck cancer. Recent randomized studies using Pilocarpine have shown that it may improve salivary flow rates but that this has only limited effect on comfort and quality of life.61–64 Disturbances of taste Radiation treatment to the mouth produces disturbances in the sensation of taste; these changes occur early in the course of treatment and persist for many months. Failure to appreciate the texture of food is a major feature. Many patients complain that food tastes like blotting paper or cotton wool. Cancer itself, or radiotherapy to regions of the body other than the head and neck, can also produce changes in the perception of taste.65,66 These disturbances are not easily managed and will interfere with eating for patients who may already be in difficulty. Radionecrosis Necrosis of soft tissue is an uncommon late complication of radiotherapy. It is usually precipitated by trauma. The late effects of radiation, manifest as depletion of parenchymal stem cells and obliterative changes in blood vessels, will impair healing. Trivial trauma, for example injury by fish bones or excessively hot fluids, can occasionally produce extensive necrosis. Patients should be advised to chew their food carefully and to avoid very hot drinks. Cartilage necrosis is a rare complication of megavoltage radiotherapy for head and neck cancer; it was far commoner in the days of orthovoltage treatment for carcinomas of the larynx. The care of the teeth The care of the teeth is one of the more contentious issues in the management of cancer of the head and neck. Routine dental clearance has been proposed for all patients with head and neck cancer. This view is too extreme and, given modern conservative dental technique, a less radical approach is justified. However, for patients who are unlikely to participate in programmes of dental care and surveillance, dental clearance may still be necessary. A basic conservative approach to the care of the teeth is summarized in Box 16.6. With a conservative approach, the majority of patients can keep their teeth and the incidence of osteoradionecrosis can be kept acceptably low. When it does occur, it can usually be managed without surgery using systemic antibiotics and vigorous oral hygiene. Small bony sequestra may be extruded spontaneously from time to time, but this often causes only
Box 16.6 Dental assessment and care for patients being treated with radiotherapy for head and neck cancer Oral assessments ● Plaque score ● Restoration of carious teeth ● Panoramic X-ray ● Checking fit of dentures ● Inspection of oral mucosa ● Surveillance culture of oral flora ● Measurement of mouth opening Prophylactic care Professional tooth cleansing ● Root planning and curettage ● Oral hygiene instructions: – tooth brushing – interdental cleansing – use of disclosing agents – cleansing of mucosal surfaces – massage of oral mucosa – discouraging denture wearing during radiotherapy – instruction in fluoride usage – consultation with dietician ●
minor inconvenience. Less than 20 per cent of patients require surgery, which can vary from simple removal of sequestrum to major procedures involving complete reconstruction of the mandible. The wearing of dentures after radiotherapy can produce problems. Ill-fitting dentures can damage the mucosa and initiate necrosis of the underlying bone. Since some remodelling of the mouth will occur after radiotherapy, patients should be discouraged from wearing dentures for 3–6 months following treatment. Muscles and soft tissue The main effect of radiation on muscles and other soft tissues is to produce fibrosis. This develops slowly, usually several years after treatment. It may not be noticed by the patient but can confuse doctors. Diffuse fibrosis of the sternocleidomastoid muscle may be mistaken for cervical lymphadenopathy. Conversely, infiltrative nodal disease can mimic post-radiation fibrosis. Severe fibrosis of the muscles and tissues around the mandible can produce trismus. This may be severe and can affect both eating and speaking. Regular jaw-opening exercises, using rubber wedges of gradually increasing thickness, can prevent or ameliorate trismus. Such treatment must be started early. If trismus is allowed to become severe, the patient’s life becomes miserable. Endocrine deficiency When the pituitary is within the treatment volume, as it may well be for tumours of the nasopharynx or ethmoid
362 Head and neck cancer
sinuses, pituitary failure may be a late complication of treatment. The onset is usually insidious; lethargy and lack of libido may go unnoticed. The diagnosis should be made as early as possible, as hypopituitarism can so easily be treated by appropriate replacement therapy. If the thyroid gland is within the radiation field, up to 50–60 per cent of patients irradiated for head and neck tumours may develop biochemical hypothyroidism, manifest as a raised thyroid-stimulating hormone (TSH) level. Only about a third of these patients will develop clinical hypothyroidism. Risks of hypothyroidism developing are increased if there has been surgery to the neck – either laryngectomy or neck dissection. The incidence also depends on total radiation dose and whether or not midline structures were shielded. Awareness of the potential problem of hypothyroidism after radiotherapy is the best aid to prompt diagnosis and treatment.
resection of the necrotic area. The only treatment is the empirical use of steroids. The overall rate of temporal lobe necrosis following treatment for nasopharyngeal carcinoma is between 0 and 3 per cent. The latent interval is between 18 months and 13 years after treatment, with a median of 5 years. Once necrosis has developed, 30 per cent of patients are severely incapacitated by a median of 3 years.
Spinal cord
The potential severity of the direct effects of radiation upon the eighth nerve is now well appreciated.67 The precise contribution of radiotherapy to sensorineural hearing loss is difficult to evaluate. Hearing declines with age (presbyacusis) and, since there may be several years’ latency between the treatment and the demonstration of impaired eighth nerve function, which is to blame – time or radiation? There is evidence of a dose–response relationship for highfrequency hearing loss after irradiation. Older patients are more vulnerable. Typically, there would be a 90 per cent probability of detectable high-tone hearing loss after a dose of 50 Gy in 2-Gy fractions to the inner ear.68
The spinal cord is vulnerable to the late effects of radiotherapy for head and neck cancer. The spinal cord will often be within the treatment volume, particularly in patients with primary tumours of the nasopharynx, maxillary antrum or posterior pharyngeal wall as well as in patients who have extensive nodal disease in the neck. Lhermitte’s sign is a transitory complication of radiotherapy to the cord and its occurrence does not imply any increased risk of serious damage to the spinal cord. It typically occurs 3–15 months following the completion of treatment, is self-limiting and requires no specific treatment. Radiation-induced myelopathy is an avoidable catastrophe – avoidable in that limiting the dose to the cord will usually prevent its occurrence, and catastrophic in that, since the damage is inflicted to the cervical cord, quadriplegia is the usual consequence. The figures for cord tolerance given in Table 16.8 should be regarded as indicating the lower limit of cord tolerance: they are appropriate for patients with a high probability of cure, in whom a 5 per cent risk of cord damage would be the maximum acceptable risk. For patients with a low probability of cure, a 10 per cent risk of severe late toxicity may be acceptable and the tolerance dose for the spinal cord could be revised upwards as follows: at 2 Gy per fraction, 47 Gy; at 2.75 Gy per fraction, 41 Gy. The provenance of these values is ultimately dubious: in order to experience a complication you have to survive, and survivors may not be typical. For experimental fractionation schemes, such as CHART, particular attention needs to be paid to cord doses, particularly because the phenomenon of slow repair may limit recovery in the spinal cord when multiple fractions per day are used. The cord dose for CHART should never exceed 44 Gy and, ideally, should be less than 40 Gy. Temporal lobe necrosis Temporal lobe necrosis is a potentially devastating consequence of radiotherapy for nasopharyngeal carcinoma. The necrosis is usually bilateral and this precludes surgical
Cranial nerves Peripheral nerves are traditionally regarded as being relatively radioresistant. However, there is evidence from clinical studies that cranial nerves and sympathetic nerves in the head and neck are less radioresistant. This is particularly important in the management of tumours of the nasopharynx (see page 388). Sensorineural hearing loss
The eye The eye is vulnerable when radiotherapy is used to treat tumours of the sinuses or nasopharynx. The eye is a complex structure and several forms of late radiation damage may occur. Lens Radiation-induced cataract is a classical example of a deterministic effect of radiation: the higher the dose, the more severe the cataract. Cataracts arise as a result of radiation effects on the germinative zone of the lens epithelium. This zone is located peripherally about 3–4 mm from the centre of the lens and about l mm in front of the equator of the lens. Changes in the lens are not usually detectable until 3 years or more after radiotherapy, and there is a lag of approximately 2 years between the first changes appearing and any detectable change in visual acuity. A single fraction of 5 Gy is sufficient to produce cataract, whereas the same dose given in 10–20 fractions is unlikely to do so. The incidence of cataract formation 8 years after 15 Gy in 15 fractions to the germinative zone of the lens is around 60 per cent and the probability of decreased visual acuity is approximately 40 per cent. Radiation-induced cataracts can be treated surgically and the radiosensitivity of the lens should not necessarily be the prime concern in planning radiotherapy to the head and neck: it is better to sacrifice
General principles of radiotherapy 363
the lens than the retina, and cure should not be put in jeopardy simply to avoid a later cataract extraction. Anterior structures The lacrimal apparatus and cornea can be affected by radiation therapy. Severe damage will cause a painful dry eye and sometimes enucleation is required. Fortunately, the tolerance dose of the anterior structures is high (equivalent dose in 2-Gy fractions 70 Gy). Retina Doses of radiation that exceed the tolerance of the retina will cause an exudative retinopathy. This will cause impaired vision and, although steroids and laser coagulation may buy time, vision may ultimately be lost. Optic nerve The tolerance to radiation of the optic nerve is higher than that of the retina, but if damage does occur, blindness is inevitable and there is no effective treatment.
neck are similar to those used at other sites: parallel opposed; parallel opposed with unequal weighting; three fields with wedges lateral fields; orthogonal wedged pair; ipsilateral wedged pair. Electron fields can be used to boost the dose superficially whilst sparing structures at depth. There is no particular magic to radiotherapy planning for head and neck tumours; it is simply that the anatomy is more complicated and more compact than at other sites. Recent innovations in radiotherapeutic techniques are proving useful in the treatment of these cancers. These include multi-leaf collimators, stereotactic techniques71 and intensitymodulated radiotherapy (IMRT).72–75 The optimal energy for treating head and neck tumours with external-beam treatment is in the range 4–6 MV. The optimal source of radiation is a linear accelerator. Some advantage has been claimed in the past for treating certain tumours on 60Co units, the argument being that too focused a beam might miss the tumour. Relying on the penumbra of a poorly collimated beam to treat a tumour is not, however, best practice.
Practical aspects of radiotherapy Interstitial radiation therapy (brachytherapy) A detailed discussion of techniques and practices for the irradiation of head and neck tumours is beyond the scope of this chapter and is covered in other texts.69 The principles involved are no different from those that apply elsewhere in the body: to devise a treatment technique that homogeneously irradiates the clinical target volume (CTV) whilst minimizing the dose to the adjacent normal structures. The definition of the CTV is a key issue in the radiotherapy of head and neck cancer. One approach is to keep the CTV as small as possible and irradiate with doses in the range of 2.75–3.5 Gy per fraction, using an overall treatment time of 4 weeks or less. Another approach is to use more generous fields and treat with lower doses per fraction (1.8–2 Gy) over a longer time, 6–7 weeks. Whichever approach is used, the CTV must be defined accurately and the patient must be precisely positioned each day for treatment. This is vital for treatment in the head and neck region because of the close proximity of vulnerable normal structures, such as the eye and spinal cord, to the CTV. An immobilization shell made individually for each patient and used throughout the processes of planning and treatment is the best way of achieving this positional accuracy. The CTV should be defined using all available information: clinical examination, under anaesthesia if necessary, and imaging. If there is difficulty defining the tumour site, marking the margins with radio-opaque seeds can be useful. Barium paste can be used to mark tumours within the oral cavity.70 Computer-assisted treatment planning techniques should be standard. The CTV is marked on outlines taken at several levels through the intended treatment fields. The optimal combination of field sizes, beam arrangements, beam weights and wedges can be defined. The beam arrangements useful in treating cancers of the head and
The use of flexible radioactive sources (such as 192Ir wire) as interstitial implants enables a high dose of radiation to be precisely localized to the site of a tumour. This can be used in the treatment of carcinomas of the lip, oral cavity, oropharynx and nasopharynx. Implantation can also be used in the treatment of lymph nodes. For tumours within the mouth and pharynx, either looped implants or 192Ir hairpins can be used. Simple planar implants can be used to treat the lip or nodes in the neck. Special applicators are used to treat tumours in the nasopharynx. Implants can be used as part of the initial treatment, as a means of boosting the dose to the main bulk of the tumour, or to treat recurrence.76 The American Brachytherapy Society has published guidelines for the use of high-dose-rate brachytherapy in the treatment of head and neck cancer.77
Postoperative radiotherapy The role of postoperative radiotherapy in the management of head and neck cancer has become more prominent in recent years. In part, this reflects the emergence of combined clinics and a multi-disciplinary approach to treatment, as well as the ability to exploit advances in both radiotherapy and surgery. There is no evidence to suggest that routine postoperative treatment is of benefit to all patients. There are, however, several factors that aid in the selection of patients who might benefit from postoperative radiation. These include positive resection margins, extracapsular involvement of lymph nodes, bulky nodes (3 cm), perineural invasion, multi-focal origin, multiple nodes positive (particularly if more than one group of nodes is involved), and
364 Head and neck cancer
extensive carcinoma-in-situ. The tolerance of the normal tissues is reduced following radical surgery and doses should be adjusted accordingly: 47.5–50 Gy in 20 fractions or 55–60 Gy in 30–35 fractions.
Re-treatment with radiotherapy This is a feasible,71,78,79 potentially risky, but occasionally necessary approach to the problem of recurrent disease, or a second primary tumour, within an area that has already been treated radically with radiotherapy.80 By definition, radical treatment is close to the tolerance of normal tissues and therefore a second course of treatment is likely to exceed tolerance. The tolerance to re-treatment increases with lapsed time following initial treatment. In highly selected patients, who have had the risks explained to them and who accept those risks, radical re-treatment may offer useful prolongation of life. Intensity-modulated radiotherapy may be particularly useful for re-treating patients with head and neck cancer. It may be possible to exclude from the treatment volume organs, such as the carotid artery and jugular vein, damage to which would otherwise be associated with an unacceptable risk of catastrophic complication. For many patients the choice may lie between chemotherapy alone and re-treatment with radiation (or chemo-radiation); the risks of the latter are greater but the chances of success are less with the former.81
Rehabilitation The main issues in the rehabilitation of the patient with head and neck cancer are summarized in Box 16.7. Sensitive and careful counselling may be required for patients who have problems with body image following surgical resections. Many of these patients have difficulties with relationships and there is a surprisingly high incidence of sexual dysfunction. Communication between patients and their partners is often disrupted – each having a completely different understanding of what is going on. Prompt and sensitive intervention can prevent unnecessary psychological morbidity. The commonest physical problems after treatment for head and neck cancer involve mastication and swallowing. Close co-operation between surgeon, prosthodontist, speech therapist and dietician is required. Videofluoroscopy can be extremely helpful in defining the functional anatomy of the problem and can indicate possible solutions. The loss of muscle and sacrifice of the spinal accessory nerve during radical neck dissection can cause problems with movements of the shoulder and neck. Appropriate physiotherapy can minimize these difficulties. Rehabilitation after laryngectomy is complex and has both physical and psychological aspects. Before consenting to laryngectomy, patients should have the opportunity to meet, and to converse with, a patient who has achieved
Box 16.7 Summary of issues of concern in the rehabilitation of patients with cancers of the head and neck Prevention of second primary tumours ● Change in smoking and drinking habits ● Chemoprevention ● Regular ENT assessment Physiotherapy After radical neck dissection or other major surgery
●
Speech therapy After laryngectomy ● Care of the voice during and after radiotherapy ● Swallowing and mastication problems (videofluoroscopy) ●
Prosthetics/prosthodontics After major head and neck surgery
●
Dental care After radiotherapy (xerostomia, caries)
●
Psychological support Accepting and dealing with diagnosis ● Specific issues related to perceived damage from treatment ● Relationships and sexuality ● Problems with body image ●
Social support Financial and social worries ● Loss of earnings ● Homelessness ● Alcoholism ●
reasonable speech following this procedure. Speech therapy is essential postoperatively; many patients develop satisfactory oesophageal speech. Other options are also available, including the electrolarynx, either as an external device or via an intraoral tube, and the Blom–Singer valve, inserted into a surgically created fistula between the oesophagus and the trachea. The ability to use the telephone after treatment is a reasonable measure of functional rehabilitation – approximately 40 per cent of patients are able to communicate by telephone after laryngectomy.
THE ROLE OF CHEMOTHERAPY Chemotherapy can play several roles in the management of head and neck cancer; these are defined and summarized in Box 16.8. Squamous carcinomas of the head and neck respond extremely well to chemotherapy. The problem is that these responses are short lived and contribute little to the cure of the disease. The explanation for the disparity between
The role of chemotherapy 365
Box 16.8 The potential roles for chemotherapy in head and neck cancer As sole treatment ● For cure ● For palliation Given before definitive treatment (neo-adjuvant) To improve survival ● To improve patients’ comfort and compliance ● To permit less radical (‘organ conserving’) surgery to be performed without compromising cure ● To predict response to radiotherapy ●
Given synchronously with radiotherapy As an attempt to improve the therapeutic ratio: increased survival without increased toxicity ● As a boost during the latter phases of XRT (‘chemoboost’) ●
Given after definitive treatment (post-adjuvant) As an attempt to lower the risks of recurrence
than 15 000 patients have been included in relevant published studies, we have no idea whether or not the side effects of treatment outweigh any reduction in tumourrelated symptoms that might have been achieved. The question of quality of life is being belatedly addressed, but there is, as yet, no information upon which to base an informed decision as to the palliative benefit, or lack of benefit, of chemotherapy for patients with head and neck cancer. This is an important issue: we fail to cure more than 60 per cent of patients who present with this disease.
Chemotherapy in the curative treatment of loco-regional disease The conclusions of a meta-analysis using individual patient data37 of 65 randomized comparisons between standard loco-regional treatment alone and similar treatment plus chemotherapy in the curative treatment of head and neck cancer can be summarized as follows.
●
●
response rates and survival benefit may lie in the kinetic behaviour of squamous carcinomas of the head and neck: the high cell-loss factor allows ample opportunity for resistant cells to evolve during the pre-clinical phases of tumour development. The rapid clinical response reflects high cell loss at a time of impaired production; the persistence and re-growth of the tumour reflects the presence of a small proportion of cells that are able to persist in spite of treatment and, ultimately, to dominate. The rapid response will often produce equally rapid relief of symptoms and this can be useful both in its own right and in convincing those patients who might be inclined to refuse, or default from, active treatment that perseverance might be worthwhile. Cis-platinum 5-fluorouracil (5-FU) has emerged as the standard combination of drugs for treating head and neck cancer. The addition of taxanes to platinum/5-FU regimens appears to achieve further benefit. One interpretation of the currently available data is that taxanes may be more effective than the platinum/5-FU combination in dealing with occult metastatic disease; because they act as radiosensitizers, the benefit from platinum and 5-FU is primarily an improvement in local control.
Metastatic disease and palliative treatment The response of metastatic head and neck cancer to chemotherapy is disappointing. The overall response rate is around 30–40 per cent, but fewer than 10 per cent of patients achieve a complete response. There is very little evidence, other than the evanescent responses observed with chemotherapy for head and neck cancer, on which to base recommendations for chemotherapy as a palliative treatment in this disease. Although more
●
●
Neoadjuvant chemotherapy has no significant impact on survival: relative risk of death 0.95 (95% CI 0.88–1.01; p 0.10). Adjuvant chemotherapy given after definitive locoregional treatment is ineffective: relative risk 0.98 (95% CI 0.85–1.19; p 0.74). When chemotherapy is given synchronously with radiotherapy, the survival rate is higher than when radiotherapy is given alone: relative risk 0.81 (95% CI 0.76–0.88; p 0.0001; difference in 5-year survival equivalent to 8 per cent).
The question of neoadjuvant, or induction, treatment is more complex than the picture that emerges from the overview quoted above. Recent trials82–85 suggest that induction regimens containing taxanes may be more effective than the platinum/5-FU regimens with which the overview was mainly concerned. Similarly, the question of whether to add chemotherapy to postoperative radiotherapy is far from settled. Two recent trials show that postoperative chemoradiation is superior to postoperative radiation alone.86–88 Overviews, by their very nature, deal with trials of a bygone era, and the relevance of such trials to modern practice is questionable. The role of overviews is, by and large, to help us to ask the right questions, rather than to provide guidance for daily practice. Another issue that is of relevance in this context is that of ‘organ conservation’. If induction chemotherapy enables the larynx safely to be conserved, when the alternative loco-regional treatment would involve laryngectomy, the failure to improve survival with neoadjuvant chemotherapy is not a reason to dismiss it. If chemotherapy were to improve the laryngectomy-free survival but have no effect on overall survival, it would still be worthwhile. The available data from three randomized trials89–91 suggest that this indeed may be the case and that, in terms of preserving the
366 Head and neck cancer
Table 16.9 The adverse consequences of chemo-radiation for head and neck cancer
Mucositis (synchronous) Rx related death (synchronous)
No. of trials
No. of patients
Rate difference (%)
Control (%)
Chemo Rx (%)
24 22
3453 2963
16 (10–21) 0.06% (4 to 5)
15 0.1
28 0.7
Data from a systematic review of randomized trials of chemotherapy in head and neck cancer. The rate difference (with 95% confidence) is simply rate in experimental arm rate in control arm. Data were pooled using the random effects method of DerSimonian and Laird.224
larynx, induction chemotherapy has an important role to play in the management of head and neck cancer. Response to a single course of induction chemotherapy provides a useful marker of response to chemo-radiation. This can be used to select non-responders for surgery and responders for conservative treatment.92 For patients with advanced disease, or where surgery would inevitably involve laryngectomy, taxane-based neoadjuvant therapy followed by chemo-radiation is a reasonable therapeutic approach. For less advanced disease, chemoradiation is usually more effective than radiation alone, but at the cost of increased toxicity (Table 16.9).
Inhibitors of signal transduction Until recently, the drugs used in treating head and neck cancer were classical cytotoxics, chosen for their ability to cause DNA damage and inhibit the replication of malignant cells. We now have drugs capable of inhibiting the abnormally active signalling pathways that contribute to the malignant phenotype.93 The pathway of particular relevance to head and neck cancer is that controlled by the EGFR. Two main drugs have been used clinically to target the EGFR pathway: cetuximab, a monoclonal antibody that targets the receptor itself, and gefitinib, a receptor tyrosine kinase inhibitor that blocks the intracellular transmission of the EGFR signal. The most robust clinical data are available from a randomized trial comparing radiotherapy alone with radiotherapy plus cetuximab in patients with AJCC stage III or IV cancer of the head and neck.94 The addition of cetuximab improved local control at 2 years by 9 per cent (from 41 per cent to 50 per cent). There was also an improvement in median survival. There was no difference between the two arms in the rate of metastatic disease. This trial provides proof of principle for the contention that EGFR blockade will improve the local efficacy of protracted radiotherapy in head and neck cancer. It does not demonstrate that cetuximab is more effective than cisplatinum/5-FU as an agent to be given synchronously with radiotherapy. It is possible that the benefit of cetuximab is mediated by an ability to counteract accelerated repopulation during treatment, and if this were the case, it might be less effective with shorter schedules of radiation treatment. There is some evidence that shorter fractionation schemes may be more effective against tumours with high EGFR expression,43 but genetic abnormalities cannot be
considered in isolation and it is the pattern of abnormalities, rather than a change in a single component, that will ultimately influence outcome.44,95 Other factors, such as differentiation and proliferation, are also important in assessing the likely benefit of accelerated fractionation.96
UNRESOLVED ISSUES AND FUTURE DEVELOPMENTS IN CHEMOTHERAPY AND RADIOTHERAPY The next generation of trials will investigate whether adding cetuximab to chemo-radiation improves results and whether adding chemotherapy, with or without cetuximab, to shorter fractionation schemes is an effective strategy. The only trial so far published that addresses the latter question is the trial from the Christie Hospital, Manchester, which showed that adding methotrexate to a 3-week course of radical radiotherapy improved local control,97,98 but not survival. We have waited more than 20 years for a trial directly comparing a short schedule (of 20–25 fractions in 4–5 weeks) with a longer schedule (of 6–7 weeks’ treatment). The experience with the CHART trial in the UK shows that such trials are feasible; the regret is that there was no third arm in the CHART study. How might 55 Gy in 20 fractions in 4 weeks have measured up against 66 Gy in 33 fractions or 54 Gy in 36 fractions in 12 days? Fowler has recently analysed the data from trials of altered fractionation in head and neck cancer [data presented RCR/NCRI meeting 6th July. His conclusion: a dose of 55 Gy given in 28 days is probably the optimal schedule]. We need to investigate safer and less unpleasant ways of giving conventional drugs. Both cis-platinum and doxorubicin can be encapsulated within liposomes, and this can increase their specificity so that the effect on the tumour is maintained but toxicity is less. Much of the toxicity of taxanes is associated with the cremophors used as diluents to prepare the drugs for intravenous use. This problem can now be avoided by using albumin-bound nanoparticles (Abraxane),99 and this is of direct relevance to the neoadjuvant therapy of head and neck cancer. The integration of surgery, chemotherapy and radiotherapy into the management of head and neck cancer will remain a challenge. We should be well beyond the days of radiotherapy and chemotherapy being requested as a means of tidying up after ill-advised or cavalier surgery; unfortunately, this is not always the case.
Palliative care 367
QUALITY OF LIFE Quality of life is, like beauty, a concept that everyone understands and that no one can put adequately into words. There are particular aspects of quality of life that are important to patients with head and neck cancer – appearance, eating and swallowing, speech and communication – but these domains are not adequately covered by many of the generic instruments used to measure whatever happens to be called ‘quality of life’. There are now several quality-of-life instruments available specifically for patients with head and neck cancer, including the University of Washington QOL (UWQOL), the Performance Status Scale – Head and Neck (PSS-HN) the EORTC (The European Organisation for Research and Treatment of Cancer) head and neck module (EORTC QLQ-H&N35). Studies of patients’ expectations and desires regarding the treatment of head and neck cancer show that patients usually rank cure as their highest priority.100,101 However, keeping a natural voice is also important and, as was shown by Pauker and McNeil102 many years ago, some individuals are prepared to compromise their chances of cure in order to retain their natural voice. There is a tendency to regard patients with head and neck cancer as non-sexual beings. This is totally inappropriate. The data from both the CHART trial and the validation studies on the EORTC QLQ-H&N35 show that sexuality is important to patients with head and neck cancer and that worries and concerns about sexuality are a cause of considerable distress. The data from the CHART study show that more than 20 per cent of patients rated their distress concerning loss of interest in sex as moderate or severe. These were patients who had not had major surgery. Imagine what it is like to kiss if you have had a tracheostomy, or to be kissed by someone who has had their larynx removed. These issues and concerns have to be addressed before, during and after treatment. There is more to treating cancer than extirpating a tumour, although, admittedly, that is an excellent start.
PALLIATIVE CARE Tumours of the head and neck have the potential to destroy people in ways that do not apply to more deeply seated tumours. Head and neck cancers are often highly visible, and even small tumours can produce severe symptoms. Treatment often adds to the mutilation and deformity and exacerbates the stigma that patients already feel. Isolation, pain, cachexia and fear can all combine to produce a miserable existence. The challenge in the palliative management of head and neck cancer is to prevent such problems developing. This challenge is not trivial: 50 per cent of patients with head and neck cancer die from uncontrolled loco-regional disease; only about 10 per cent of patients die from metastatic disease with the primary tumour controlled. A dignified, painless death from liver metastases or renal failure is unusual in patients with head and neck cancer.
This raises an important question: is it justifiable to put a patient through a protracted, painful, debilitating and potentially mutilating sequence of treatments for what may turn out to be a very low chance of cure? The only sensible answer to such a question is that such a course of action may be justifiable provided that the individual patient, fully appraised of the benefits and harms associated with radical treatment, chooses to be so treated. Given the heterogeneity that is intrinsic to cancers of the head and neck, and to the patients who are afflicted by them, there can be no stock solutions to clinical problems. Each patient, with their hopes, fears and own unique predicament, has to be managed and treated as an individual. Guidelines can provide advice but should not dictate management. There can be no useful protocol to dictate how Bach’s Goldberg Variations should be played, nor can the rigid implementation of protocols play a major role in the provision of care for patients with cancers of the head and neck. The specific symptoms and problems associated with head and neck cancer are outlined in Table 16.10. Management depends first on the recognition of the symptom, then on the appreciation of the mechanism by which it is occurring, and finally on applying appropriate treatment. Psychological support, explanation, and reassurance should be an integral part of this process. The most logical approach is often to treat the tumour itself and combine treatment with whatever adjunctive treatments (analgesia, dietary supplements etc.) are necessary. The problem is that treatment itself often produces symptoms and thus violates a basic principle of palliative treatment – first do no harm. The acute mucosal reaction caused by radiotherapy may prove more troublesome than the symptoms produced by the tumour. By allowing time for repopulation of normal mucosal cells, split courses of treatment offer a way round the problem: 14.8 Gy in four fractions over 2–4 days repeated two to three times at 3-week intervals is a useful schedule. Xerostomia may, however, still be troublesome. Palliative surgical resections of tumour are not usually helpful, but laser therapy, including photodynamic therapy, may provide usual palliation for lesions that are superficial and accessible. Cranial nerve sections may be useful for intractable pain.
Pain Pain is a common and distressing symptom in patients with head and neck cancer. Nearly 50 per cent of patients have pain at and around the time of diagnosis; 8 per cent of patients rated their pain as severe.103 Successful treatment provides effective pain control: by 2 years after successful treatment, the proportion of patients complaining of pain was 26 per cent, and 4 per cent had severe pain. In order to manage pain rationally, it is essential to discriminate between nociceptive pain caused by the stimulation of free nerve endings and non-nociceptive,
368 Head and neck cancer
Table 16.10
The symptoms that commonly require palliation in patients with head and neck cancer
Symptom Obstruction-related Dyspnoea; fatigue; stridor Dysphagia Anosmia; headache; sinusitis; cacosmia Facial congestion/swelling; headache; CNS symptoms (confusion, blurred vision) Otitis media, deafness Pain Nociceptive Non-nociceptive
Cause
Obstruction of upper airway Obstruction of pharynx Obstruction of nose Obstruction of blood vessels
Obstruction of Eustachian Spread of tumour; infection Nerve involvement Neuropathy following radical neck dissection
Aspiration
Destruction of upper airway sphincters by tumour Fibrosis following XRT Distorted anatomy following surgery
Bleeding
From local tumour internally externally From eroded vein or artery
Dysphagia
Obstruction by tumour intrinsic extrinsic Fibrosis after surgery and/or XRT Reduced tongue mobility Secondary to involvement or destruction of lower cranial nerves Secondary to aspiration Secondary to dry mouth
Dry mouth
Secondary to XRT
Special senses Diplopia Anosmia
Deafness Balance Communication
Invasion of orbit Cranial nerve involvement Ethmoid tumours After craniofacial resection After XRT Naso-pharyngeal carcinoma via VIIIth nerve rarely directly due to tumour post radiotherapy Usually secondary to other factors (anaemia, vascular insufficiency, hypotension) Receptive difficulties Expressive difficulties Psychological difficulties
Syncope
IXth nerve syndrome (pain, hypotension, bradycardia) Carotid sinus hypersensitivity (bradycardia)
Weight loss
Lack of appetite Inability to eat (tumour, side effects of treatment) Cancer-induced cachexia
Foetor
Infected tumour Mucositis Dental problems
XRT, radiotherapy
Novel approaches to management 369
neuropathic pain related to abnormal excitability of nerve fibres. Nociceptive pain is managed primarily by prostaglandin antagonists and opiates, whereas membrane stabilizers and nerve blocks are more appropriate for nonnociceptive pain. The palliative treatment of head and neck cancer is not simply about attempting to shrink tumours. Broader existential issues have to be considered: fear (of pain, of disfigurement, of dying), emotional isolation, and problems of body image (wasting, weakness), social function, nurture, nourishing and nutrition. If these, and many other, issues are to be properly dealt with, there has to be adequate and open communication between the patient, the patient’s family, the hospital-based services and the services based in the community. Primary care and community support services are essential parts of this liaison. However, the comparative rarity of these tumours means that individual practitioners may have limited experience of dealing with the problems involved.
NOVEL APPROACHES TO MANAGEMENT Photodynamic therapy Photodynamic therapy (PDT) is based on the principle that some tissues, including tumours, will specifically take up light-sensitizing chemicals. If light of a wavelength that will specifically excite the photo-chemical is shone on the tumour and its immediate environment, only the illuminated cells will be exposed to the toxic effects of singlet oxygen produced by the photo-chemical reaction. The biological effects of photodynamic therapy are complex: cell death, by apoptosis or necrosis, occurs within 4 hours of treatment. Vascular effects are also important: obliterative changes cause ischaemia and impaired perfusion. Unlike ionizing radiation, for which the dependence on oxygen is only relative, the biological effects of PDT have an absolute dependence on oxygen: PDT has no effect when oxygen concentration falls below 0.5 per cent. Oxygen is rapidly depleted within tissues treated with PDT, which imposes a biological limitation on the duration of effect. The selectivity of uptake with currently available photosensitizers is not particularly impressive: the tumourto-normal-tissue ratios are typically around 2 to 3. For true therapeutic selectivity, ratios of at least 10, and preferably 100, would be required. Any benefits from PDT at present are more to do with the localization of light penetration, and subsequent sensitization, than to any selectivity of the photochemical for tumour cells. In practice, PDT works as follows. Before exposing the tumour to light, the patient is treated with a photosensitizing chemical. Typical compounds are derived from porphyrins and include: haematoporphyrin derivative (HpD), Photofrin,
5-aminolaevulinic acid (5-ALA), meta-tetrahydroxyphenychlorin (mTHPC, Foscan), benzoporphyrin derivative (BPD), mono-aspartyl chlorin e6 (Npe6), Lutetium texaphyrin (Lu-Tex). The drug is concentrated within the tumour and eliminated from the normal tissues. The tumour is treated using a coherent source of laser light of appropriate wavelength. The approach is only feasible for tumours that are endoscopically accessible and relatively superficial. As technology develops, more tumours will be accessible using fibre-optic sources, and higher light energy will enable the light to penetrate more deeply into tumours. Currently, the limit of penetration is between 5 mm and 10 mm. Those sensitizers with long total body clearance times will cause prolonged photo-sensitization of the skin. Patients should be warned to avoid exposure to sunlight, or fluorescent light, until the drug has been eliminated – for Photofrin this could take as long as 6–12 weeks. The main role of PDT in the management of head and neck cancer has been in the curative treatment of premalignant conditions, such as erythroplasia and leukoplakia in the oral cavity. It is also establishing a niche in the palliative treatment of superficial recurrent disease. In the future, flexible light sources implanted into tumours may allow more homogeneous PDT for larger and deeper tumours. The integration of PDT with more conventional treatments as part of the primary management of head and neck cancer may be a promising area for future research.104–107
Hyperthermia As a means of specifically killing cancer cells, heat has many biological advantages over ionizing radiation. Unfortunately, there are major physical disadvantages in terms of therapeutic equipment and dosimetry. Interstitial hyperthermia can be used, with or without radiation, in the palliative treatment of recurrent disease. There is, however, little indication that hyperthermia will have a major role to play in the primary management of head and neck cancer. It may have a useful contribution to make, in combination with radiotherapy, in the re-treatment of disease that has recurred locally following previous treatment with radiation.108
Biological therapies The finding that many subtle abnormalities can be identified in squamous carcinomas has prompted the development of treatments aimed at specifically targeting those abnormalities that may be unique to the malignant cells. Abnormal p53 is present in about 50 per cent of patients with head and neck cancer and thus presents an obvious target for therapeutic manipulation.109 Restoration of normal (wild-type) p53 has been attempted using a construct
370 Head and neck cancer
(Ad-p53) based on an adenovirus as vector, which uses the wild-type p53 gene inserted into an adenoviral backbone. An alternative approach has been to use ONYX-015. This is an adenovirus that has had the E1B gene deleted and that will therefore selectively replicate within, and destroy, cells that lack wild-type p53. Phase I/II studies have shown that about 30 per cent of advanced cancers of the head and neck will respond to injection of ONYX-015 directly into the tumour. Practical difficulties with this approach led to the abandonment of ONYX-15 studies in the USA in 2003. However, ONYX-15 has been taken up by a Chinese biotechnology company (Shanghai Sunway Biotech) and more than 1000 patients have already been treated in the Far East. As our knowledge of the molecular biology of head and neck cancer increases, it is likely that many new therapeutic agents will be developed. It is equally likely that the initial clinical impact of molecular biology in head and neck cancer will be on the assessment of prognosis and the definition of patients at particularly high risk of local recurrence or distant spread.44 Staining for p53 has already demonstrated that cancer cells may be present at resection margins defined as clear by conventional histology. The presence of such cells may be of independent prognostic significance,110 and the presence of other molecular markers at resection margins, such as MMP-9 and 4E, may also be independently predictive of recurrence.111
STAGING AND PROGNOSIS The factors that influence the prognosis of patients with head and neck cancer can be grouped into three main categories: factors related to the tumour, factors related to the patient, and factors related to treatment. The traditional clinical staging systems have concentrated primarily on those factors related to the tumour. Box 16.9 summarizes patient-related factors of potential relevance to prognosis. The primary purpose of staging head and neck cancer is to classify tumours reproducibly so that valid comparisons can be made between groups of patients. Classification should also provide an indication of the prognosis for an individual patient. The UICC TNM system for cancer has undergone several revisions, most recently in 2002.112,113 The system has to cope with the heterogeneity of head and neck cancer, the wide variety of sites and sub-sites, as well as the conflicting demands of surgeons and radiotherapists. It therefore represents a pragmatic compromise that is broadly useful, but is not entirely perfect. The most recent version of the system subdivides T4 tumours into T4a (advanced but potentially operable) and T4b (advanced and inoperable). There is a uniform N staging for nodal disease in the neck (Box 16.10), but an exception is made for nasopharyngeal cancer, which has its own staging system, based primarily on the extensive experience of radiotherapists in the Far East. There are seven anatomical levels, corresponding
Box 16.9 Prognostic factors in head and neck cancer Host related Demographic/behavioural Gender (males worse) Age (under 40 worse) Ethnic group Tobacco use ● previous (worse) ● continuing (worse) Performance status (poorer performance worse) Nutritional status and diet (poorer worse) HIV status (positive worse) Laboratory Haemoglobin (low worse) Albumin (low worse) Immune competence (impaired worse) Treatment related Surgery (adverse factors) Positive margins Wound infection Transfusion requirement Tracheostomy requirement Radiotherapy Dose (lower worse) Overall treatment time (longer worse) Tumour related Primary Macroscopic ● Site ● Size ● T stage ● Appearance ● infiltrating (worse) ● exophytic (better) ● Local invasion ● Associated field change (worse) Microscopic ● Depth of invasion ● Differentiation (poor worse) ● Mitotic rate ● Pleomorphism ● Vascular invasion Molecular biology ● Ploidy (aneuploid worse) ● Tumour angiogenesis ● increased nodal metastases ● ?improved control of primary ● Syndecan-1, cell adhesion molecule (higher levels better) ● p53 mutations (?better ?worse) ● EGFR (high levels worse)
Staging and prognosis 371
p27 (low levels worse) ● c-erb B-2 overexpression (worse) ● c-myc expression (worse) ● loss of heterozygosity at 2q or 10q (worse) Nodes ● Macroscopic ● size ● N stage ● fixation ● density on CT scanning ● microscopic ● extracapsular spread ● lymphocyte depletion ● plasmacytic response ●
Box 16.10 UICC (2002) N-staging for carcinomas of the head and neck NX: Regional lymph nodes cannot be assessed N0: No regional lymph node metastasis N1: Metastasis in a single ipsilateral lymph node, 3 cm or less in greatest dimension N2: Metastasis in a single ipsilateral lymph node, more than 3 cm but not more than 6 cm in greatest dimension, or in multiple ipsilateral lymph nodes, none more than 6 cm in greatest dimension, or in bilateral or contralateral lymph nodes, none more than 6 cm in greatest dimension N2a: Metastasis in a single ipsilateral lymph node more than 3 cm but not more than 6 cm in greatest dimension N2b: Metastasis in multiple ipsilateral lymph nodes, none more than 6 cm in greatest dimension N2c: Metastasis in bilateral or contralateral lymph nodes, none more than 6 cm in greatest dimension N3: Metastasis in lymph node more than 6 cm in greatest dimension The anatomical descriptor ‘U’ or ‘L’ may be used to designate nodes as above (‘U’) or below (‘L’) the level of the cricoid
to anatomical landmarks evident at the time of surgical dissection, used to describe neck nodes: Level I: submental and submandibular triangle – bounded by bellies of digastric, mandible and hyoid. Level II: upper jugular nodes – from skull base to hyoid. Level III: middle jugular nodes – from hyoid to cricoid. Level IV: lower jugular nodes – from cricoid to clavicle. Level V: posterior triangle nodes – between anterior border of trapezius, posterior border of sterno-cleido-mastoid and clavicle. Can be subdivided into upper, middle and lower using the same criteria as levels II, III and IV.
Level VI: anterior central nodes – from hyoid to suprasternal notch and bounded on each side by the carotid sheath. Level VII: nodes in the superior mediastinum inferior to the suprasternal notch. The M staging of head and neck cancer is straightforward. MX: metastatic state unknown. M0: no metastases. M1: metastases. There are four histological classifications used for squamous cancers of the head and neck. GX: differentiation unknown. G1: well differentiated. G2: moderately differentiated. G4: poorly differentiated. There is also an R classification for indicating the presence and extent of residual tumour after primary therapy. R0: no residual tumour. R1: microscopic residual tumour. R2: macroscopic residual tumour. RX: residual tumour cannot be assessed. The primary anatomical sites in the head and neck recognized by the TNM system are the lip, oral cavity, pharynx, larynx, nasal cavity and para-nasal sinuses, major salivary glands and thyroid. The anatomical sub-sites and details of T staging for the other sites are considered in the appropriate later sections of this chapter. In general, Tis indicates carcinoma-in-situ, Tx indicates that the primary tumour cannot be assessed, and T0 indicates that there is no evidence of primary tumour.
Stage grouping The recognition that the TNM system can produce a daunting array of permutations in head and neck cancer has led to the grouping of combinations of TNM categories into five larger groups designated stages 0 to IV (Table 16.11). The main problem with stage grouping is that inherent to any process of aggregation: loss of fine discrimination and the collapse of disparate entities into biologically implausible categories. This is particularly true of stage IV disease, which encompasses a spectrum extending from patients with locally advanced disease and minimal nodal involvement all the way through to patients with widespread metastatic disease. In recognition of this problem, the recent stage groupings subdivide stage IV. Other stage groupings have been proposed114 and there are empirically defined schemes that are of more prognostic value,115 but, in terms of an international standard, the AJCC stage grouping provides the benchmark.
372 Head and neck cancer
Table 16.11 American Joint Committee for Cancer (AJCC) stage grouping for head and neck cancer (2002) Stage group 0 I II III
IVA IVB IVC
T
N
M
Tis 1 2 3 1 2 3 4a 1,2,3 4b Any Any
0 0 0 0 1 1 1 0,1,2 2 Any 3 Any
0 0 0 0 0 0 0 0 0 0 0 1
Stage groupings, based on TNM, for cancer of the Head and Neck.225
Stage migration Stage migration116 – that process whereby an artefactual improvement in stage-specific results can be produced simply by changing staging investigations or classification criteria – can cause problems in the interpretation of clinical results in head and neck cancer. Patients staged in one era, or at one centre, will not necessarily resemble patients of apparently identical stage assessed at another time or in another place. It is obvious that patients with pathologically negative necks diagnosed as N0 by radical neck dissection are not directly comparable to patients staged clinically as N0. It is less immediately obvious that patients staged as II or III after FDG-PET scan may have a better prognosis than patients staged conventionally as stage II or III.
HOST-RELATED PROGNOSTIC FACTORS The most immediately relevant host-related factors are those that, unlike age, gender or previous social habits, are capable of favourable manipulation. Low pre-treatment haemoglobin is an adverse prognostic factor for survival in patients with head and neck cancer treated primarily with radiotherapy.24,117–119 The effect can be demonstrated even within a range that would be defined as physiologically normal. Transfusion would be beneficial if the low haemoglobin directly caused impaired survival, but would not necessarily improve matters were the lower haemoglobin simply a surrogate marker for some unidentified underlying factor in tumour or host, for example impaired nutritional status. Erythropoietin will increase haemoglobin levels and tumour oxygenation but, as mentioned previously, may have an adverse effect on local control and survival.25 There is evidence that persistent smoking can adversely affect prognosis in patients treated with radiotherapy for
head and neck cancer. A Canadian study showed that the 2-year survival was 30 per cent greater in patients who did not smoke after diagnosis.120 This fact should be used to try to persuade patients to stop smoking.
ASSESSMENT OF PATIENTS A full clinical history and appropriate physical examination are essential. The history should pay particular attention to the duration of symptoms: possible aetiological factors and symptoms suggestive of metastatic disease or a second (synchronous) primary tumour should be actively sought. General physical examination should again be slanted towards the detection of systemic disease or second primary tumour. An adequate social history is also important; social and psychological problems should be anticipated and, wherever possible, dealt with. The primary tumour should be assessed clinically. Site of origin, size, appearance and involvement of adjacent structures should all be formally recorded. Standardized diagrams are helpful for both initial assessment and subsequent follow-up. Wherever possible, patients should be assessed by more than one experienced clinician. The clinicians should make their assessments independently, record their findings, and only after so doing should they discuss and assign the TNM stage. The presence or absence of premalignant change should also be formally recorded. Endoscopy performed in the outpatient clinic may be sufficient, but most patients will require formal assessment under anaesthesia. Adequate biopsy is mandatory: although the vast majority of head and neck tumours are squamous carcinomas, other histologies are encountered. It is a therapeutic disaster to mistake a lymphoma or a syphilitic gumma for a squamous carcinoma. The following investigations should be performed in all patients: full blood count, liver function tests, urea and electrolytes, and chest X-ray. Computed tomography scanning and MRI can provide important supplementary information, particularly concerning direct extension of tumour into bone or extent of nodal disease. Computed tomography is particularly useful in evaluating invasion of the laryngeal cartilage or spread into pre-epiglottic space or the paraglottic space. Computed tomography imaging is not only useful diagnostically but is also used for radiotherapy planning – both for external-beam treatment and for brachytherapy. Magnetic resonance image scanning is particularly useful for assessing tongue tumours and in the assessment of the patient with amalgam-filled teeth. Scanning with FDG-PET can provide useful information about the extent of nodal involvement, but there is considerable controversy about the role of PET and PET–CT fusion in the management of patients with head and neck cancer.74,121–127 Fine-needle aspiration cytology can be extremely useful if there is doubt about whether enlarged nodes are reactive or involved by tumour. A positive result means something,
Assessment of patients 373
but a negative result does not definitively exclude nodal involvement. There is no evidence that fine-needle aspirate performed before definitive treatment adversely effects prognosis. Although all these technological advances have improved the assessment of patients with head and neck cancer, there is still no substitute for careful clinical examination.
Table 16.12 Investigations that may be useful in identifying potentially treatable primary tumours in patients who present with carcinoma of unknown primary Primary tumour Breast Ovary
Neck nodes There are three main areas of controversy concerning the management of cervical nodes in patients with head and neck cancer: 1. the management of the patient who presents with a malignant neck node in the absence of a clinically detectable primary tumour; 2. the role of prophylactic treatment to the neck in patients with primary tumours of the head and neck who have no evidence of cervical lymphadenopathy; 3. the management of potentially operable cervical lymphadenopathy in patients with primary tumours in the head and neck. PRESENTATION WITH MALIGNANT NODE IN THE NECK
Between 5 per cent and 10 per cent of patients with malignant disease seen in ENT departments have, as their presenting symptom, a lump in the neck. The majority of tumours are squamous-cell carcinomas; the rest are adenocarcinomas, lymphomas or undifferentiated tumours. Prognosis depends critically on histology: lymphomas have a good prognosis, adenocarcinomas have an extremely poor prognosis – few patients survive for more than 3 years. The prognosis for patients with squamous carcinoma presenting in this fashion is intermediate (30–40 per cent 5-year survival). The location of the lymphadenopathy, together with histology, may give an indication of whether the occult primary tumour lies above or below the clavicles. Nodes in the supra-clavicular fossa and/or adenocarcinoma usually indicate that the primary tumour is below the clavicles. The prognosis of adenocarcinoma presenting as a neck node is so poor that aggressive treatment cannot be justified for patients in whom a primary tumour cannot be found. Only treatable primary tumours should be sought. There is no point in performing extensive gastrointestinal endoscopies or barium studies. The potentially treatable primaries and the relevant investigations are outlined in Table 16.12. SQUAMOUS CARCINOMA PRESENTING AS A CERVICAL LYMPH NODE
The traditional approach to treatment when the primary site cannot be defined has been to use extended-field radiotherapy, with a treatment volume to include not just the neck nodes bilaterally but also the potential sites for the primary tumour. These would include the nasopharynx, the piriform
Prostate Neuro-endocrine tumour Germ cell tumour (gonadal or extragonadal Thyroid
Investigations Mammography CA 125 Pelvic ultrasound Serum PSA Serum chromogranin Urinary catecholamines, 5HIAA Serum AFP and βHCG Pelvic ultrasound (female) Testicular ultrasound (male) Serum thyroglobulin Thyroid ultrasound
AFP, α-fetoprotein; βhCG, human chorionic gonadotrophin; 5HIAA, 5-hydroxyindoleacetic, PSA, prostate-specific antigen.
fossa and the whole of the larynx. The Mayo Clinic has reviewed, from its usual surgical perspective, its experience with neck nodes and unknown primary tumours.128 Of 117 patients seen with this presentation between 1965 and 1987, 24 were treated by neck dissection without radiotherapy or any other treatment directed at possible primary sites. The 10-year survival in this highly selected group was 40 per cent. Experience from the radiotherapy department at the Middlesex Hospital, London, also suggests that extended treatment may not always be necessary.129 The 5-year survival in 58 patients treated with radical radiotherapy to the neck only was 40 per cent – morbidity was minimal, mainly because the bulk of both parotid glands could be spared. MANAGEMENT
The flow chart in Figure 16.3, based on the available clinical evidence, illustrates an approach to the management of patients who present with neck nodes and an undetectable primary tumour. It assumes that all patients have been adequately investigated, which, nowadays, should include a CT of the thorax and abdomen as well as a CT of the head and neck. For patients treated radically with radiotherapy for squamous or undifferentiated carcinomas, the dose should be 47.5–52.5 Gy in 20 fractions. Palliative local radiotherapy may have a role in patients with adenocarcinomas in order to alleviate local discomfort or to prevent fungation of the nodes. It is reasonable, on the principle of Pascal’s wager,130,131 to treat women presenting with adenocarcinomas with tamoxifen. Men presenting in this fashion, in whom prostate cancer has been excluded and who seek active treatment, can be treated with systemic chemotherapy. For younger patients, it is reasonable to use bleomycin, etoposide and carboplatin empirically. For older patients, less toxic regimens such as single-agent capecitabine can be used as empirical treatment.
374 Head and neck cancer
Histology
Squamous
Adenocarcinoma
Chinese origin or positive EBV serology
Search for treatable primary
N1 Disease Found
Undifferentiated
Disease > N1 Wide-field XRT as for NPC
Not found
RND
Appropriate Rx
ECE
Younger No ECE
Older
Tamoxifen or capecitabine (females)
Capecitabine (males)
BEP-type chemotherapy
Bilateral neck irradiation Clinical follow-up
Figure 16.3 A suggested scheme for the management of a patient who presents with a node in the neck in the absence of an easily identifiable primary tumour.
THE ROLE OF PROPHYLACTIC TREATMENT TO THE NECK IN PATIENTS WITH CLINICAL NEGATIVE NECK AND PRIMARY TUMOURS OF THE HEAD AND NECK
This issue has generated controversy for many years, and no randomized surgical trial has addressed the problem. In some clinical circumstances the argument is, in any event, superfluous. A tonsil ‘commando’ operation is often combined with radical neck dissection because a pectoralis major flap can more easily repair the deficit if a block dissection has been performed. The advantages and disadvantages of elective block dissection are summarized in Box 16.11. The moves towards a more conservative approach to the management of the primary tumour (see the section on chemotherapy and organ conservation) provoke a re-consideration of the role of elective neck dissection. A reasonable approach is to treat the primary and first echelon nodes with radiotherapy (or chemo-radiation). The neck nodes in patients considered to be at high risk of occult nodal disease can be treated by elective neck dissection after completion of treatment to the primary. Prophylactic radiation of the entire neck requires large radiation fields, and total doses of no more than 47.5–50 Gy can be given if treatment is given in 20 fractions over 4 weeks. These doses may be too low to control the primary tumour in patients treated solely by radiotherapy, and boost doses to the primary may be required. Radiotherapists divide into two main schools: those who believe in high-dose, small-volume treatments and
Box 16.11 A summary of the advantages and disadvantages of radical neck dissection in patients with head and neck cancer Advantages of elective neck dissection for N0 neck ● Provides definitive treatment ● Provides prognostic information: ● whether or not nodes are involved ● if so, how many ● presence or absence of extracapsular spread Disadvantages of elective neck dissection ● Prolongs operating time with attendant anaesthetic risks ● Immediate surgical complications: ● haematoma ● lymphocoele ● wound dehiscence ● necrosis of flaps ● wound infection ● carotid rupture ● chylous fistula ● salivary fistula ● Later complications: ● facial oedema ● shoulder pain ● deafferentation pain ● XIth nerve damage
Assessment of patients 375
who therefore do not irradiate the neck prophylactically, and those who believe in using larger volumes, often encompassing all the potentially involved lymph nodes, but lower doses. The main side effect of elective wide-field radiation to the upper cervical nodes is xerostomia, because much of the parotid gland is inevitably included within the fields. A randomized trial of prophylactic neck irradiation in patients with carcinomas of the oral cavity was performed at the Christie Hospital in the 1970s. This trial was briefly reported in an earlier edition of this textbook: no statistically significant benefit could be demonstrated, in terms of survival, for prophylactic radiotherapy to the ipsilateral neck when a dose of 50 Gy was given in 15 fractions over 21 days. Prophylactic radiotherapy to the neck did, however, significantly reduce the rate of recurrence in the neck: 79 per cent of 100 irradiated patients remained free of neck disease compared with 64 per cent of 105 patients who received radiotherapy only to the primary tumour (p by λ2 0.025). If isolated nodal relapse is to occur in patients not treated prophylactically, it usually does so within 2 years of first treatment. Careful follow-up, with patients being seen every 2–3 months, is obviously essential, so that if relapse does occur it can be diagnosed and treated promptly. Patients who are unlikely to co-operate with strict follow-up should have their necks treated prophylactically. THE MANAGEMENT OF POTENTIALLY OPERABLE CERVICAL LYMPHADENOPATHY
Traditionally, operable nodal disease was managed by block dissection in conjunction with removal of the primary tumour. Radiotherapy as primary treatment to operable neck nodes was used only in patients with inoperable primary tumours. Increasing unease with the morbidity associated with radical neck dissection, and an increased awareness of the effectiveness of radiotherapy in controlling both the primary tumour and nodal disease, have led to a reappraisal of the traditional position. Bataini, at the Institute Curie, has clearly shown that even bulky nodal disease can be controlled by radiotherapy provided the dose is sufficiently high.132 Factors that indicate that nodes are unlikely to be controlled by radiotherapy alone include nodes 3 cm, fixed nodes, primary T4 or T3, and prolonged overall treatment time. It is reasonable to recommend radiotherapy, with surgery held in reserve for failure, for patients with N1 disease (3 cm). Patients with more advanced but operable neck disease and operable primary tumours should be considered for radical surgery. Patients with inoperable primary tumours and N2 or N3 disease require high doses of radiation, preferably using concomitant boost techniques, if their neck disease is to be controlled adequately. Radiotherapy need not be given routinely after adequate radical neck dissection. If, however, there is extensive evidence of extra-capsular spread, or resection margins are positive, postoperative radiotherapy can reduce the incidence of failure.
TREATMENT OF INOPERABLE NODAL DISEASE AFTER PREVIOUS RADIOTHERAPY
Large, painful, fungating nodes in the neck are extremely distressing to patients. This problem can occur with the primary tumour controlled in a patient who is otherwise generally well and is, therefore, condemned to a miserable existence with no immediate prospect of release. A combination of surgery and interstitial implantation can be used in an attempt to deal with this difficult problem. The bulk of the disease is removed surgically, in the full knowledge that disease is left behind at the base of the dissection. Afterloading tubes, for later 192Ir interstitial implantation, are placed directly across the tumour bed at the time of the open operation. The tissue deficit is then closed using a myocutaneous flap. The radioactive sources are loaded several days later. This gives a high dose of radiation to the area, at depth, of residual disease. The skin and superficial tissues, having been brought in from elsewhere, have excellent tolerance to radiation. The surgical procedure deals directly with disease bulk and fungation and the radiation delays the re-growth of the tumour, thereby providing a significant period of relief from symptoms.
Carcinoma of the lip Nearly all carcinomas of the lip are squamous carcinomas. Basal-cell carcinomas can involve the lip, but actually originate from the surrounding skin and cannot, by UICC criteria, be regarded as true tumours of the lip. Three sites are recognized: upper lip, lower lip and commissures. More than 90 per cent of squamous carcinomas of the lip arise on the lower lip, presumably because the lower lip is more exposed to the most important aetiological factors – tobacco use, particularly pipe smoking, and sun exposure. Tumours of the upper lip are uncommon and tend to be less well differentiated than tumours of the lower lip. Tumours of the upper lip and commissures, particularly those that are poorly differentiated, are more likely to spread to regional nodes than tumours of the lower lip. The classical description of lymph-node spread is from the upper lip to the deep jugular chain, the centre of the lower lip to the submental nodes, and the more lateral parts of the lower lip to the submandibular nodes. In practice, spread is variable. Overall, less than 10 per cent of welldifferentiated tumours spread to nodes, whereas 30–40 per cent of poorly differentiated tumours are associated with nodal disease. Clinically, carcinomas of the lip present as an indolent ulcer. A history of temporary regression followed by re-growth is characteristic. The ulcer often has a rolled margin, and induration may be palpable well beyond the margins of the visible lesion. There may be evidence of associated leucoplakia. The differential diagnosis can be extensive; syphilis, chronic candidiasis and other infective causes must be excluded. Biopsy is essential.
376 Head and neck cancer
The choice of treatment lies between surgery and radiotherapy, and the final decision often rests on the local availability of specialist skills and techniques. RADIOTHERAPY
Radiotherapy can be administered either as external-beam treatment (electrons or orthovoltage) or as an interstitial implant. Electron-beam treatment has the advantage that the depth of penetration in tissue can be governed by the particular beam energy that is selected. The gums and teeth can easily be shielded using 3 mm or so of lead (backed with wax) as a gum shield. The technical set-up is easy; the field size should be generous, as electron isodoses constrict at depth. Afterloaded iridium wire implants are straightforward and can easily be performed under local or general anaesthesia (17.6 old numbering). Two to three wires are placed in parallel 0.5–1.0 cm apart. SURGERY
The main indication for surgery for the treatment of carcinoma of the lip is the presence of leucoplakia adjacent to the invasive tumour. The tumour and the abnormal surrounding epithelium can be excised together. A W or V incision with primary closure is usually adequate when the extent of the excision is less than one-third of the lower lip. For superficial lesions, lip shave and vermilion advancement will produce an excellent cosmetic result. Larger lesions require excision and reconstruction using a flap. There is no indication for routine treatment to the nodes in patients with carcinoma of the lip. However, patients with larger undifferentiated tumours or who are unlikely to attend for regular follow-up should be offered prophylactic treatment to the nodes, either block dissection or radiotherapy. Patients with clinically involved nodes should be managed along the lines described previously (see the section on nodes). PROGNOSIS
The overall prognosis for carcinoma of the lip is excellent: radiotherapy will control 90 per cent of cases. The status of the neck nodes is important: less than 10 per cent of N0 patients fail, but radiotherapy fails to control disease in nearly 50 per cent of patients with clinically positive neck nodes.
Tumours of the oral cavity Although relatively uncommon in Western societies, oralcavity tumours have a high incidence in the developing world – particularly in India, where the incidence is 21/100 000 a year. This problem reflects social habits – the chewing of tobacco and betel nut. Snuff dipping (taking snuff orally – ‘saffa’ in the Sudan, ‘nasswar’ in Pakistan)
and other forms of use of smokeless tobacco are causally associated with oral-cavity carcinomas. This is scarcely surprising, since commercial snuff products contain extremely high levels of carcinogenic nitrosamines – more than 2000 parts per billion. A worrying development is the increased use of smokeless tobacco products by socially disadvantaged young people in the developed world, including native people of Alaska and the North West territories of Canada, and Blacks in the rural south of the USA. Other important aetiological factors in oral-cavity cancer are cigarette smoking, alcohol and chronic sepsis and, mainly of historical interest nowadays, syphilis. The T staging of tumours of the oral cavity is shown in Box 16.12. PRE-MALIGNANT CONDITIONS OF THE ORAL CAVITY
Leucoplakia Leucoplakia appears as a white patch on the mucous membrane. It cannot be removed by scraping and has a characteristic histological appearance – cellular atypia with acanthosis, hyperkeratosis and parakeratosis. The presence of leucoplakia indicates a high risk of developing carcinoma of the oral cavity – approximately 50–100 times compared to normal. The overall cumulative risk of developing invasive cancer for patients with leucoplakia is about 5 per cent at 20 years. Oral hairy leucoplakia Oral hairy leucoplakia is increasingly recognized as one of the complications of infection with the human immunodeficiency virus (HIV). Epstein–Barr virus is probably the direct cause. Characteristically there is an exuberant
Box 16.12 UICC (2002) T staging for carcinomas of the lip and oral cavity226 TX: T0: Tis: T1: T2:
Primary tumour cannot be assessed No evidence of primary tumour Carcinoma in situ Tumour 2 cm or less in greatest dimension Tumour more than 2 cm but not more than 4 cm in greatest dimension T3: Tumour more than 4 cm in greatest dimension T4: (lip) Tumour invades adjacent structures (e.g., through cortical bone, inferior alveolar nerve, floor of mouth, skin of face) T4a: (oral cavity) Tumour invades adjacent structures (e.g., through cortical bone, into deep (extrinsic) muscles of tongue, maxillary sinus, skin) T4b: Tumour invades masticator space, pterygoid plates or skull base or encases carotid artery (Superficial erosion alone of bone/tooth socket by gingival primary is not sufficient to classify as T4)
Assessment of patients 377
leucoplakia affecting the dorsum of the tongue. In a minority of patients with oral hairy leucoplakia and HIV infection, there can be rapid progression to invasive cancer. Lesions in patients with positive tests for hepatitis B or syphilis are particularly likely to progress. Erythroplasia Erythoplasia appears as a velvety red patch on the mucous membrane. It is associated with a high rate of malignant transformation and histologically there is marked cellular atypia. Submucous fibrosis Submucous fibrosis presents as oral discomfort, often exacerbated by spicy foods. The buccal mucosa is predominantly affected; the mucous membrane is thickened and pale. Histologically, the changes are those of chronic inflammation with accumulation of collagen at the dermoepidermo junction. Submucous fibrosis is found predominantly in people from the Indian subcontinent; the incidence may be as high as l per cent in certain areas. There is an eight-fold increase in the incidence of oral cancer in patients with submucous fibrosis.
Carcinoma of the mobile tongue The mobile tongue (synonyms: oral tongue; anterior twothirds of tongue) extends forwards from the circumvallate papillae. About 40 per cent of all oral cancers arise in the mobile tongue, and there has recently been a change in the sex incidence of these tumours: formerly about 75 per cent of patients were male, but the sex incidence is now approximately equal. Most tumours of the mobile tongue arise on the lateral borders, 15 per cent arise on the inferior surface and 10 per cent on the dorsum or the tip of the tongue. The presenting symptom is usually an ulcer, either painless or uncomfortable, which does not heal. More advanced lesions cause disturbances in speech; severe pain, often referred to the ear, is a late symptom. Secondary infections produce foetor. Occasionally, the primary lesion is unnoticed by the patient, who may present with a lump in the neck secondary to involved lymph nodes. The role of the general dental practitioner in early diagnosis is important. The tip of the tongue drains to the submental lymph nodes; the rest of the mobile tongue drains to the submandibular, subdigastric and middle deep cervical nodes. Neck nodes are frequently involved – 30–40 per cent of patients have palpable lymphadenopathy at presentation and a further 30 per cent have occult disease. Contralateral nodal involvement may occur, even with well-lateralized lesions. ASSESSMENT
Adequate biopsy, usually possible under local anaesthesia, and clinical examination of the primary and neck are
routine. Scanning with MRI is superior to CT scanning in the assessment of invasion and local spread. TREATMENT
There is little to choose, in terms of cure rate, between radiotherapy and surgery for early (stage I or II) tumours of the mobile tongue. In North America, surgery is often preferred, whereas European centres have tended to treat with radiotherapy. Overall, the morbidity rate with surgery is less, but when it does occur after surgery, morbidity may be severe. Speech and swallowing are better preserved after radiotherapy, but often at the expense of xerostomia and long-term dental problems. Surgery for larger lesions often has to be extensive, and it may be preferable to compromise surgical margins slightly and treat postoperatively with radiotherapy rather than create a tissue deficit that cannot be adequately reconstructed. Postoperative radiotherapy can be used to prevent local recurrence in patients with positive or ‘close’ (0.5 cm) margin. SURGERY
Cryosurgery has been used successfully for superficial (0.5 cm depth of invasion) tumours of the tongue. Simple wedge excision is equally effective for small superficial lesions. Larger tumours that do not extend across the midline can be treated by hemiglossectomy. Primary closure may be sensible, but excessive suturing should be avoided and granulation repair is rapid and preferable to avoid impaired function. Lesions that extend across the midline require a more aggressive approach. Exposure is best achieved by splitting the mandible anteriorly; the mandible and skin are then opened like the leaves of a book, providing excellent exposure. Sub-total excisions can be repaired using a radial forearm free flap. When the whole of the anterior two-thirds of the tongue has been removed, bulkier flaps, such as the pectoralis major, latissimus dorsi or rectus abdominis flap, are required. The base of the tongue may then act as a piston and the flap is passively pushed forward during swallowing and speaking. RADIOTHERAPY
Small (1 cm) lesions can be treated entirely by interstitial implant. Alternatively, external-beam techniques using an ipsilateral wedged pair of fields or IMRT may be employed. Such a field arrangement minimizes the volume of salivary tissue within the field and thereby may prevent troublesome xerostomia. The results from either surgery or radiotherapy alone for more advanced tumours (T3 and T4) are poor. The common approach is to combine surgery with postoperative radiotherapy, but the functional result may be poor. Treating the primary with radiotherapy and concomitant chemotherapy and then treating the neck nodes surgically is an alternative, and entirely reasonable, strategy.
378 Head and neck cancer
RESULTS OF TREATMENT, PROGNOSTIC FACTORS AND CAUSES OF FAILURE
The following factors adversely affect the prognosis in patients with carcinoma of the mobile tongue: perineural invasion, intra-lymphatic tumour emboli, more advanced T stage, and tumour thickness. Patients with tumours greater than 2–3 mm thick have reduced survival. Tongue carcinomas developing under the age of 40 appear to be biologically more aggressive than those developing in later years. However, patients over the age of 70 have a poorer prognosis compared with patients 10–20 years younger; this may reflect the better prognosis in tumours associated with HPV infection.8 Between 20 per cent and 30 per cent of patients with clinically negative necks, and who do not receive prophylactic radiotherapy to the neck, will relapse in cervical nodes. Five-year survival in such patients is approximately 30 per cent. There is, however, no clear evidence that routinely treating the neck in all patients improves overall survival. Patients with tumours more than 2–3 mm thick or with T2 or T3 tumours are at sufficiently high risk of recurrence in the neck to justify routine treatment, at least to the first echelon nodes. Second malignancies and intercurrent deaths are a major problem in patients whose primary tumours are controlled: only about 20 per cent of patients actually die from tongue cancer.
Floor of mouth The boundaries of the floor of the mouth are the inner surface of the mandibular arch, the undersurface of the anterior part of the tongue and the anterior pillar of the tonsil. Tumours often arise adjacent to, and may spread along, Warthin’s duct. Comparison between series of patients treated for carcinoma of the floor of the mouth is not straightforward. Some authors include tumours of the floor of the mouth; tumours which involve the undersurface of the tongue along with tumours of the mobile tongue. Other authors simply recognize such tumours as a subgroup, with less favourable prognosis, of floor of the mouth tumours. Cancer of the floor of the mouth is commoner in males – 9:1 male:female ratio. These tumours may not be noticed by patients, but eventually difficulties with speech and excessive salivation may lead them to seek medical advice. Pain usually indicates invasion of the mandible; involvement of the inferior dental nerve will cause anaesthesia of the lower lip on the affected side. The management of carcinoma of the floor of the mouth is, in principle, very similar to that of carcinoma of the mobile tongue. For early lesions, radiotherapy and surgery are equally effective. Where function is not compromised, surgery in experienced hands offers a rapid and straightforward approach. Invasion of the mandible is a
relative contraindication to radiotherapy. Larger lesions (T3, T4) may require a combined approach. SURGERY
The tumour and the immediate lymphatic drainage should be removed en bloc. A neck dissection – either a radical or suprahyoid block dissection – should be performed in continuity with the primary procedure. If the tumour is large, or adherent to the mandible, a marginal mandibulectomy is often required. If there is evidence of erosion of the mandible, resection of the affected segment with suitable replacement is required. The radial forearm free flap is probably best for this purpose, although there are alternatives: a groin flap with iliac bone, a pectoralis major flap with rib, and plating the mandible and packing the cavity with bone chips. RADIOTHERAPY
Radiotherapy alone will cure more than 90 per cent of T1N0 tumours. The control rate drops to below 70 per cent for T2N1 tumours. Tumours larger than 3 cm, or those that extend to the gum, are probably best treated surgically. External-beam radiotherapy can be given using either a parallel opposed pair, with or without unequal loading, or an ipsilateral wedged pair of fields. A dose of 52–55 Gy in 20 fractions in 4 weeks or its equivalent is required. The prognostic factors for tumours of the floor of the mouth are the same as those for tumours of the mobile tongue.
Carcinoma of the buccal mucosa The buccal mucosa forms the lining of the cheeks; the lymphatic drainage is to the submandibular and subdigastric nodes. Tumours often arise in relation to pre-malignant conditions such as submucous fibrosis or leucoplakia. The buccal mucosa is a common site of origin for verrucous carcinomas, the management of which is controversial. Radiotherapy allegedly can cause these well-differentiated tumours to become less differentiated and thereby adversely affect prognosis. There is no evidence to support this contention, and yet the view is often held that radiotherapy is contraindicated for verrucous carcinomas. These tumours should not be singled out for special treatment; their management should simply be as for any other squamous carcinoma.133 Small tumours can be successfully treated with either surgery or radiotherapy. Larger tumours require a combined approach, but severe fibrosis may cause trismus and impair the functional and cosmetic results. ASSESSMENT
Assessment should include adequate biopsy, clinical examination and CT scanning of the primary and neck.
Assessment of patients 379
SURGERY
Local excision for small lesions will leave a defect that can easily be closed, either with a locally rotated mucosal flap or by a split-thickness skin graft. More extensive resections require major reconstruction with flaps: pectoralis major; latissimus dorsi; cervico-facial; radial forearm. RADIOTHERAPY
Small tumours can be treated entirely by implant; larger lesions require external-beam treatment. The dose to the local tumour can be boosted with either implant or electron beam. Electron-beam treatment can be administered externally or using an intra-oral cone. Fibrosis and trismus may cause problems after radiotherapy, but can usually be minimized, provided careful attention is paid to technique and dose and jaw-opening exercises are performed regularly after treatment. Five-year survival for carcinoma of the buccal mucosa is poor: only 30–50 per cent. These figures reflect the poor general condition of many of these patients and the high incidence of second primaries.
Tumours of the hard palate The curious habit of reverse smoking, in which the lighted end of the cigarette is placed in the mouth, is an interesting aetiological factor for these rare tumours.134 SURGERY
When bone is involved, surgery is the preferred treatment. Lesions confined to the mucosa can be excised and the wound left to granulate. Attempts to repair the deficit are both unnecessary and futile. Resection of bone is often required, and careful preoperative planning is necessary because some form of obturator will be required to close the defect. Teeth should be preserved if at all possible, as they provide useful anchorage points for a wire-mounted prosthesis. Temporary obturators can be held in place by circumzygomatic wires. RADIOTHERAPY
Small superficial tumours may be managed by radiotherapy. Electron-beam therapy, using an intra-oral cone, or gold-grain implant can be used, either alone or in conjunction with external-beam treatment. The overall 5-year survival for patients with primary tumours of the hard palate is between 30 per cent and 50 per cent.
Tumours of the gum These tumours usually arise in relation to coexistent leucoplakia on the posterior part of the mandible. There is rich lymphatic drainage and 20–30 per cent of patients have positive nodes at presentation. The lingual surface of the mandible drains to subdigastric, upper deep cervical and retropharyngeal nodes. The buccal surface drains to the submandibular, submental and subdigastric nodes. Invasion of bone is common, and surgery is the treatment of choice. Only exuberant tumours with no evidence of bony erosion or involvement should be considered for radiotherapy. Overall 5-year survival is approximately 50 per cent.
Retromolar trigone According to the UICC, tumours of the retromolar trigone are classified with tumours of the buccal mucosa. The retromolar trigone is the area of mucosa that overlies the ascending ramus of the mandible between and behind the lower third molar and the upper third molar. These tumours behave similarly to tumours of the anterior faucial pillar and should be managed similarly. There is a high risk of bone involvement and, as for gingival tumours, management should be primarily surgical in these cases. External-beam radiotherapy, using a wedged pair technique, is adequate for the earlier tumours.
Carcinoma of the oropharynx The oropharynx is roughly cuboidal in shape and extends from the junction of the hard and soft palates to the level of the floor of the vallecula. The sub-sites are: the base of the tongue, including the pharyngoepiglottic folds and the glossoepiglottic folds; the tonsillar region, which includes the fossa and the anterior and posterior pillars; the soft palate, including the uvula; and the posterior and lateral pharyngeal walls. The T staging for oropharyngeal tumours is summarized in Box 16.13. When assessing the literature, it is important to remember that the epidemiology of oropharyngeal tumours varies considerably. For example, the ratio of male to female
Box 16.13 UICC (2002) T staging for carcinomas of the oropharynx T1: T2: T3: T4a:
Tumour less than 2 cm Tumour greater than 2 cm but less than 4 cm Tumour greater than 4 cm Tumour invades larynx, deep/extrinsic muscles of tongue, medical pterygoid muscle, hard palate or mandible T4b: Tumour invades lateral pterygoid muscles, pterygoid plates, lateral nasopharynx, skull base or encases carotid artery
380 Head and neck cancer
patients is typically between 2:1 and 3:1 in series from Denmark, Canada and the USA, whereas in French series, the ratio is between 12:1 and 15:1. This may, in part, reflect social habits. Hot Calvados, as drunk in Normandy, appears to be a potent carcinogen. Most of these tumours are squamous carcinomas, although at this site non-Hodgkin’s lymphoma is more common than elsewhere in the head and neck. Tumours characteristically present late, sometimes as a node in the neck from an occult primary. Difficulty swallowing or pain in the throat or pain radiating to the ear are the usual symptoms from the primary itself. Spread to retropharyngeal nodes, which can occur late in the course of the disease, may cause cranial nerve palsies (IX, X, XI, XII). Infiltrating tonsillar tumours may produce trismus. On examination, there is usually an ulcerating lesion at the primary site. Induration is caused by local infiltrative spread and is often more extensive than the visible lesion. Careful palpation, more kindly performed under general anaesthetic, is an essential part of the assessment of oropharyngeal tumours, together with adequate biopsy. Computed tomography scanning of the primary site and neck is essential.
Tumours of the posterior third of the tongue About 70 per cent of patients with tumours of the posterior third of the tongue have neck nodes at presentation. The majority of patients present with T3 or T4 tumours; fewer than 15 per cent of tumours are T1 or T2 at presentation. T1 and T2N0 tumours of the posterior tongue should be treated with radical radiotherapy or chemo-radiotherapy to both the primary tumour and the neck nodes. The primary tumour and first echelon nodes can be treated by opposed lateral fields. The low neck can be treated by an anterior split cervical field. The dose to the primary should be 52–55 Gy in 20 fractions over 4 weeks (or its equivalent); the neck nodes should receive 50 Gy in 20 fractions in 4 weeks. This approach produces 5-year disease-free survival rates of between 30 per cent and 40 per cent. For the majority of patients – those with a T3 or T4 tumour – the results of treatment are poorer, with 5-year survival rates of less than 30 per cent. Lymph-node spread is common. The surgical options are limited. A total glossectomy and laryngectomy, with all that that implies in terms of loss of function and quality of life, is unacceptable to many patients. Less aggressive surgery, combined with postoperative radiotherapy, can achieve local control in carefully selected patients. The ability to deliver effective brachytherapy to the posterior tongue may further improve results. Iridium-192-afterloaded wires can be placed either percutaneously or at open operation. Using external-beam radiotherapy and percutaneous iridium implants to both primary tumour and any involved cervical nodes, Puthawala et al. have reported local control in 30/40 (75 per cent) of
patients with T3 tumours and 8/12 (67 per cent) of T4 tumours.135 Treatment of the primary tumour with radiotherapy and of lymph nodes with bilateral neck dissection is an acceptable combination, preserving function with reasonable local control. Concomitant chemotherapy with radiotherapy may improve local control rates by about 10 per cent at the expense of increased acute morbidity.
Tumours of the tonsillar area The tonsillar area can be divided anatomically into the following sites: the tonsillar fossa, the anterior and posterior faucial pillars, and the glosso-tonsillar sulcus. Because the anterior faucial pillar marks the point of embryological fusion between the oral cavity and the oropharynx, tumours arising from it have more in common with oral-cavity tumours than they do with carcinomas of the oropharynx. The tonsil itself is a lymphatic organ and, together with lymphatic tissue in the posterior tongue, around the lower end of the Eustachian tube, in the nasopharynx and in the soft palate, forms Waldeyer’s ring. Extra-nodal nonHodgkin’s lymphomas of the head and neck often arise from Waldeyer’s ring. There is a clear association between nonHodgkin’s lymphoma arising extra-nodally in the head and neck and lymphomatous involvement of the stomach and small bowel. The investigation and treatment of nonHodgkin’s lymphoma are described in detail in Chapter 43. Clinically, is it often possible to distinguish carcinomas of the tonsil from lymphomas. Patients with tonsillar lymphomas rarely complain of pain; obstructive symptoms or the sensation of a foreign body in the throat predominate. The tumour itself is often smooth and purplish, and ulceration of the overlying mucosa is uncommon. Cervical lymphadenopathy is often bilateral and bulky; the nodes have a characteristic elastic texture, in contrast to the hard, unyielding nodes associated with squamous-cell carcinomas. Squamous carcinoma of the tonsil usually causes pain, often referred to the ear, and painful dysphagia. Typically there is an infiltrating, ulcerated mucosal lesion, although occasionally tonsillar carcinoma can have an exophytic appearance with minimal ulceration or infiltration. Exophytic tumours may have a better prognosis than infiltrative tumours; this clinical impression is not, however, supported by evidence from formal multivariate analysis. The lymphatic drainage of the tonsillar area is to the subdigastric and jugulo-digastric nodes initially and then to the middle deep cervical nodes. Between 60 per cent and 75 per cent of patients have palpable cervical nodes at presentation, and the majority (around 70 per cent) have T3 or T4 primaries. Assessment requires examination under local or general anaesthesia, with biopsy and MRI, with or without CT. Surgery alone or radiotherapy alone is probably equally effective for T1 or T2 tumours of the tonsillar area. Of the two, radiotherapy gives the better functional result, and
Assessment of patients 381
therefore better quality of life. For T3 and T4 tumours, a combined approach is usually required; radical surgery with postoperative radiotherapy is commonly used. A policy of radical radiotherapy with surgery reserved for operable failures is also reasonable and may give a better functional result. SURGERY
Radical surgery for tonsillar tumours was not really feasible until the development of the so-called commando operation in the 1940s: a neck dissection is performed initially and then the mandible is split behind or between the molar teeth. The tumour can then be visualized by rotating the posterior part of the mandible laterally and backwards and is approached via the neck incision; the removal of the tumour and nodes can then be carried out en bloc. Closure involves tissue replacement using, for example, a pectoralis major, latissimus dorsi flap, or free fibula or when bone is resected; iliac crest. Overall 5-year survival after surgery for carcinoma of the tonsil is around 50 per cent.
main cause of treatment failure is recurrent or persistent loco-regional disease.
Hypopharynx The hypopharynx extends from the level of the tip of the epiglottis to the lower border of the thyroid cartilage. There are three main sites: the pyriform sinus, the posterior pharyngeal wall and the post cricoid area. This classification of hypopharyngeal tumours can be criticized. Tumours arising on the pharyngeal surface of the aryepiglottic folds might be more appropriately considered with tumours of the supraglottic larynx. Post-cricoid tumours are more closely related to tumours of the upper third of the oesophagus than they are to tumours of the pyriform sinus. The T staging of hypopharyngeal tumours is shown in Box 16.14. The lymphatic drainage of the hypopharynx is to retropharyngeal nodes and to the subdigastric, upper and middle deep cervical nodes. Post-cricoid tumours may spread to nodes in the upper mediastinum.
RADIOTHERAPY
Early tumours of the tonsillar fossa, with no palpable cervical lymphadenopathy, can be treated using lateralized techniques:136 ipsilateral wedged pair or parallel opposed fields with unequal weighting or IMRT. These techniques may also treat the first echelon nodes. Electron beams have appealing depth–dose characteristics but cannot be relied upon: the dose actually delivered to tissues behind the mandible may be less than expected because of the unpredictable absorption of electrons by bone. For larger tumours, parallel opposed field arrangements, with treatment to the low neck using an anterior split cervical field, will be required. The primary tumour requires a dose of 50–55 Gy in 20 fractions in 4 weeks, or its equivalent. A similar dose is required for palpable neck disease. The clinically negative neck can be effectively treated with 50 Gy in 20 fractions over 4 weeks, or its equivalent. Concomitant chemotherapy offers a small but significant increase in local control. The main cause of treatment failure in patients with carcinoma of the tonsil treated with radiotherapy is failure to control loco-regional disease. Failure to control nodal disease is related to N stage: for N0 disease, the failure rate with prophylactic radiotherapy to the neck is less than 5 per cent; for N3 disease, the failure rate is 70 per cent. It is possible to implant the tonsillar fossa with afterloaded 192Ir wires. The results from series of patients who had implants as part of their treatment are excellent. Possibly this has as much to do with the careful selection of patients as it does with the effectiveness of the treatment. Overall treatment time is an important prognostic factor for local control in carcinoma of the tonsil treated by radiotherapy: by allowing time for tumour-cell repopulation, gaps and delays compromise local control.137,138 The
Pyriform fossa The pyriform (pear-shaped) fossae lie on each side of the laryngeal orifice. They are bounded medially by the aryepiglottic fold, and laterally by the thyroid cartilage and thyrohyoid membrane. The apex usually extends inferiorly as far as the lower border of the thyroid cartilage. Between 60 per cent and 70 per cent of all hypopharyngeal tumours arise in the pyriform fossa. More than 75 per cent of patients are men. The primary tumour only rarely produces symptoms; the usual presentation is with a node in the neck. Eventually the primary tumour will produce pain, hoarseness and dysphagia. These tumours are usually advanced at presentation; T1 tumours are rare (around 5 per cent). Even small primary tumours may be associated with advanced neck disease. Between 40 per cent and
Box 16.14 UICC (2002) T staging for carcinomas of the hypopharynx T1: Tumour less than 2 cm and limited to one subsite T2: Tumour greater than 2 cm but less than 4 cm without fixation of hemilarynx or involves more than one subsite T3: Tumour greater than 4 cm or with fixation of hemilarynx T4a: Tumour invades thyroid/cricoid cartilage, hyoid bone, thyroid gland, oesophagus or central compartment soft tissue T4b: Tumour invades prevertebral fascia, encases carotid artery or extends into mediastinum
382 Head and neck cancer
50 per cent of patients with clinically negative necks have occult nodal disease at presentation. Prophylactic treatment of the neck is therefore indicated in all N0 patients accepted for radical treatment. TREATMENT
Combined treatment is required for nearly all tumours of the pyriform fossa. The exception would be a small tumour arising from the hypopharyngeal surface of the aryepiglottic fold; these tumours can often be adequately controlled by primary radiotherapy. Radical surgery for tumours of the pyriform sinus involves removal of the larynx as well as the hypopharynx. Postoperative radiotherapy is required for the majority of patients. An EORTC randomized trial90 showed that induction chemotherapy is a safe option for patients with advanced hypopharyngeal cancer: patients randomized to initial chemotherapy followed by radiotherapy or surgery (according to response) had equivalent survival to patients treated with surgery and postoperative radiotherapy. The laryngectomy-free survival rate at 5 years was 35 per cent in patients treated with induction chemotherapy (and, obviously, zero in patients treated surgically). SURGERY
The ablative surgical procedure required for patients with hypopharyngeal tumours is a pharyngo-laryngectomy and block dissection of the ipsilateral neck. A permanent tracheostomy is inevitable. Several reconstructive methods have been used to restore continuity of the food passage. Up to 40 per cent of patients reconstructed with myocutaneous flaps or colon transpositions may fail to achieve normal swallowing and require long-term tube feeding. The failure rate with gastric transpositions or free jejunal grafts is lower. The longterm survival rates for patients treated with primary surgery for hypopharyngeal tumours are typically between 25 per cent and 35 per cent – 33 per cent 5-year survival in a recent Australian series.139 The discrepancy between diseasespecific survival (52 per cent) and overall survival (33 per cent) bears witness to the importance of intercurrent deaths and new primaries in this group of patients. RADIOTHERAPY
The standard technique for treating these tumours is to use parallel opposed lateral fields to the primary tumour and the upper neck nodes, with an anterior split cervical field to treat the low neck. If there are concerns about the position of the junction between fields, this can be moved halfway through treatment, for example by adding 2 cm to the lower border of the lateral fields and subtracting 2 cm from the upper border of the anterior field. For patients with short necks or where other technical or anatomical factors create problems, more complex techniques, similar to
those described for tumours of the supraglottic larynx, may be required. The results of primary radiotherapy for cancer of the pyriform fossa are poor. Long-term survival is only 5–25 per cent. Inability to control loco-regional disease is the usual cause of failure. Control of bulky nodal disease is particularly difficult. The macroscopic extent of the primary tumour may be misleading: skip lesions, occult cartilage involvement and clinically inapparent circumferential spread may all occur and will lead to underestimation of the extent of the tumour. Coping with this uncertainty by simply increasing field size may prohibitively increase morbidity. This high rate reflects both field size and the local tissue destruction caused by these advanced tumours. The early results of the FNCLCC-GORTEC trial140 suggest that chemotherapy given synchronously with hyperfractionated radiotherapy is better than hyperfractionated radiotherapy alone for advanced head and neck cancer. Forty of the 163 patients in this trial had hypopharyngeal primaries: there was evidence of benefit from chemotherapy in this subgroup but, presumably because of small sample size, the difference was not statistically significant. Given that the results with radiation alone are so poor, there is an argument for using neoadjuvant chemotherapy to assess the responsiveness of the tumour (and to produce some symptomatic relief). Patients who fail to respond should be treated surgically if operable, and palliatively if inoperable. Patients who respond should be treated with chemoradiotherapy, with salvage surgery held in reserve for failure. The palliative management of patients with pyriform fossa tumours is difficult. Chemotherapy with carboplatin and 5-FU may produce rapid relief of symptoms but, because response is usually short, may only tantalize. Splitcourse radiotherapy 14.8 Gy in four fractions over 1 week followed by a 3-week gap, followed by 14.8 Gy in four fractions over 1 week may provide useful palliation without causing undue mucosal toxicity. For patients whose tumours and general condition improve with treatment, a third course of four fractions may be given after a further 3 weeks: total dose 44 Gy in 12 fractions over 12 weeks.
Post-cricoid tumours These tumours are similar in their behaviour to tumours of the upper third of the oesophagus and should be treated as such (see Chapter 21).
Posterior pharyngeal wall These tumours are uncommon and in most series account for 20 per cent or less of tumours of the hypopharynx. Clinically, they present with dysphagia and the sensation of a foreign body at the back of the throat. As with carcinomas of the oesophagus, these tumours commonly spread
Assessment of patients 383
inferiorly and superiorly; posterior spread of the tumour is limited by the prevertebral fascia. Surgical treatment for these tumours is extremely difficult and they are usually managed using radical radiotherapy. They can be treated using wedged lateral fields to a dose of 50–55 Gy in 20 fractions over 4 weeks. The spinal cord should be shielded at tolerance. The results of treatment are poor, and only about 30 per cent of patients survive free of disease at 3 years. The usual cause of treatment failure is failure to control loco-regional disease.
Larynx Approximately one-third of all cancers of the head and neck arise in the larynx. The majority of patients are middleaged males (male:female ratio 8:1). More than 90 per cent of laryngeal tumours are invasive squamous carcinomas; between 2 per cent and 10 per cent are in-situ carcinomas. Other tumours that may arise primarily in the larynx include oat-cell carcinomas, lymphomas, tumours arising in ectopic salivary tissue and carcinoid tumours. There are three anatomical sub-sites of the larynx: supraglottis, glottis and subglottis. The supraglottis is divided into five sub-sites: suprahyoid epiglottis, laryngeal aspect of the ary-epiglottic fold, arytenoids, infrahyoid epiglottis, and false cords (ventricular bands). There are three glottic sub-sites: cords, anterior commissure, and posterior commissure. The T staging is shown in Boxes 16.15–16.17. For supraglottic tumours, the most important distinction is between tumours of the epilarynx (suprahyoid epiglottis, ary-epiglottic fold and arytenoid) and the rest. Tumours of the epilarynx tend to be more advanced than tumours of the lower supraglottis; over half of the tumours are T4. Nodal spread is also common. The common defining feature of T3 laryngeal tumours is cord fixation. However, cord fixation can arise through a variety of mechanisms, which may alter the prognosis: the sheer size of the tumour may physically trap the cord; the rotation of the arytenoid cartilage may be prevented by tumour involving the cricoarytenoid joint; the terminal branches of the recurrent laryngeal nerve may be affected by tumour, with consequent paralysis of the intrinsic muscles; and the intrinsic laryngeal muscles may be directly infiltrated by tumour. Transglottic spread, a distinction not made by the UICC TNM system, is an adverse prognostic factor within the T3 category of glottic tumours. The UICC TMN system also fails to discriminate between types of T2 glottic tumour. In addition to any consideration of anatomical extent, impaired cord mobility may be an independently adverse prognostic factor. The supraglottis has a rich lymphatic plexus draining to the subdigastric and mid-cervical nodes. The subglottis drains to the lower cervical, paratracheal, and mediastinal nodes. The glottis itself is not well supplied with lymph vessels, and tumours confined to the vocal cords rarely spread to nodes.
Box 16.15 UICC (2002) T staging for carcinomas of the supra-glottic larynx T1: T2:
Tumour limited to one subsite, cords normal mobility Tumour involves more than one subsite or extends into adjacent soft tissues; cord mobility normal T3: Limited to larynx, cord fixation and/or invasion of post-cricoid space, pre-epiglottic tissues, paraglottic space or minor erosion of thyroid cartilage T4a: Tumour invades through thyroid cartilage and/or invades tissues beyond the larynx (trachea, soft tissues of neck, tongue, strap muscles, oesophagus, trachea) T4b: Tumour invades pre-vertebral space, encases the carotid artery or extends into mediastinum
Box 16.16 UICC (2002) T staging for carcinomas of the glottic larynx T1:
Tumour limited to vocal cords, normal mobility T1a: one cord T1b: both cords T2: Tumour extends to subglottis or supraglottis and/or cord mobility impaired T3: Limited to larynx, cord fixation and/or invasion of para-glottic space or minor erosion of thyroid cartilage T4a: Tumour invades through thyroid cartilage and/or invades tissues beyond the larynx (trachea, soft tissues of neck, tongue, strap muscles, oesophagus, trachea) T4b: Tumour invades pre-vertebral space, encases the carotid artery or extends into mediastinum
Box 16.17 UICC (2002) T staging for carcinomas of sub-glottic larynx T1: T2:
Tumour limited to subglottis Tumour extends to cords with normal or impaired mobility T3: Limited to larynx with cord fixation T4a: Tumour invades cricoid or thyroid cartilage and/or invades tissues beyond the larynx (trachea, soft tissues of neck, tongue, strap muscles, oesophagus, trachea) T4b: Tumour invades pre-vertebral space, encases the carotid artery or extends into mediastinum
CLINICAL FEATURES
The cardinal symptom of laryngeal tumours is change in the quality of the voice. This can vary from mild hoarseness, often intermittent, to severe hoarseness with the voice reduced to the merest whisper. Hoarseness is a symptom that is usually taken seriously by both patients and their doctors, and this, combined with the ease with which the
384 Head and neck cancer
larynx can be assessed by indirect laryngoscopy, means that patients with glottic tumours usually present with early disease. Hoarseness is a late symptom in patients with tumours of the epilarynx; mild dysphagia may be the only initial symptom from these tumours. This may go unnoticed, and patients with high supraglottic tumours may eventually present with a lump in the neck, from nodal disease. This delay in presentation may in part explain the poorer overall prognosis for tumours of the epilarynx. ASSESSMENT
All patients with laryngeal cancer require direct laryngoscopy and biopsy as part of their initial assessment. This is usually best carried out under general anaesthesia, although fibreoptic nasendoscopy offers a possible alternative for patients in whom anaesthesia would be hazardous. Adequate biopsies are essential. Cord mobility can usually be assessed on indirect laryngoscopy and cannot be adequately assessed in the anaesthetized, paralysed patient. Computed tomograph scanning is now routinely used in staging laryngeal tumours. By demonstrating early cartilage invasion, evidence of local spread beyond the larynx and nodal disease, the main effect of CT will be to classify as T4 tumours that, on clinical grounds alone, might have been staged as T1, T2 or T3. Stage migration should be remembered when comparing series in which CT has been used for staging and those using clinical staging only. The patient who presents with stridor due to a laryngeal tumour usually has advanced disease, but careful assessment should take place if possible. To avoid tracheostomy, endoscopic debulking may be possible to secure the airway. Although the adverse effect of tracheostomy followed by delayed primary treatment is known, emergency laryngectomy is rarely practicable.
Verrucous carcinomas of the larynx The question of the radiosensitivity (or radioresistance) of verrucous carcinomas of the larynx is controversial. There is no evidence that radiotherapy in any way increases the aggressiveness of these tumours. On the other hand, there is some evidence that radiotherapy may be less effective for verrucous carcinomas than it is for other squamous carcinomas of the larynx. If conservative endoscopic surgery is feasible, this is the best option; if not, patients should be treated with radical radiotherapy rather than laryngectomy.
Supraglottic tumours Tumours of the supraglottis account for 25–40 per cent of all laryngeal tumours in most series, although a much
higher rate (60 per cent) was reported from Finland. Approximately half of the tumours arise from the lower supraglottis and half from the epilarynx. Nearly 50 per cent of patients with epilaryngeal tumours have palpable nodes at presentation, compared with only 23 per cent of patients with tumours of the lower supraglottis. Tumours of the epilarynx behave more like pharyngeal tumours; tumours in the lower supraglottis behave more like tumours of the larynx proper. The epiglottis has a pitted surface and is perforated by laryngeal nerves. This may facilitate direct extension of supraglottic tumour into the pre-epiglottic space. The T staging of supraglottic tumours is summarized in Box 16.15. The choice of treatment for early (T1, T2) supraglottic tumours lies between voice-conserving surgery and radiotherapy. Supraglottic laryngectomy will permit retention of the voice but is not always feasible – there may not be sufficient clearance between the lower end of the tumour and the vocal cords. Many patients are not fit for anaesthesia; involvement of both lingual arteries is also a contraindication to supraglottic laryngectomy. About 50 per cent of all patients are anatomically unsuitable for supraglottic laryngectomy. Of the patients who are anatomically suitable, many will have significant co-morbidity and be inoperable on medical grounds. Overall, therefore, only about one-third of all patients are suitable for supraglottic laryngectomy, and surgeons vary in their enthusiasm for the procedure. For T3 tumours, a policy of radical radiotherapy, with salvage laryngectomy reserved for failure, is appropriate for those tumours with favourable features: exophytic, limited local extension. Local control will be obtained with radiotherapy in 60–70 per cent of such patients. Infiltrating tumours or those with extension or local invasion into the pyriform fossa or glottis are best treated by laryngectomy. Postoperative radiotherapy should not be given routinely. Only patients with positive or minimal (5 mm) surgical margins or with nodal disease with extracapsular spread should be selected for postoperative radiotherapy. RADIOTHERAPY TECHNIQUE
For early (T1, T2N0) supraglottic tumours, it is not necessary to treat the whole neck prophylactically, but fields need to take into account the high incidence of occult nodal disease. Parallel opposed fields extended up to cover potential lymph-node spread at the angle of the mandible with wedges to compensate for the contour of the neck can be used. The dose should be 52–55 Gy in 20 fractions over 4 weeks, or its equivalent. The lower dose should be used for field sizes greater than 50 cm2, otherwise late complications can be severe. Patients with more advanced primary tumours, or who have nodal disease at presentation, require treatment to the whole neck. This can pose a challenge, particularly in patients with short necks. A variety of techniques have been used in order to avoid areas of under-dosage within
Assessment of patients 385
what is often an extensive treatment volume. The angleddown wedged pair offers the most elegant solution: the whole volume is treated en bloc and gaps are avoided. The downward angle is achieved by swinging the foot of the couch 10–15° to the left for the right field and by a similar amount to the right for the left field. The change in neck contour can be compensated for by using wedges with the thick end anterior. The remaining problem is what to do about the cranio-caudal changes in separation, as these will produce inhomogeneities and there is a risk of overdosing the spinal cord. The best solution is probably to use a compensator, but if this resource is unavailable, wedging in both directions may achieve a similar effect: three-dimensional conformal planning increases the potential accuracy of such treatments. In terms of dose distribution, IMRT has advantages over more conventional techniques;141 however, to date, clinical experience with IMRT for supraglottic cancers is limited. The results of radiotherapy alone for supraglottic carcinoma depend critically on stage; the local control rates vary from more than 60 per cent for T1 to less than 40 per cent for T4. The main cause of failure of radiotherapy for supraglottic carcinoma is persistence or recurrence of the primary tumour; only about 10 per cent of patients fail solely in the neck nodes.
CARCINOMA-IN-SITU
This condition may be much more common than we suppose. In one autopsy series, 15 per cent of smokers had carcinoma-in-situ of the larynx. This contrasts with the 2–10 per cent incidence of carcinoma-in-situ observed in large series of patients with laryngeal carcinoma. For carcinoma-in-situ of the vocal cords, simply stripping the cord can provide both material for diagnosis and, in the short term at least, effective treatment. Laser excision is also now commonly used for in-situ carcinoma of the larynx. Experience is preliminary, but for selected patients it may provide a reasonably expedient alternative to radiotherapy. For more extensive lesions, the only surgical approach that is feasible may be cordectomy, which will invariably affect the quality of the voice. Radiotherapy is undoubtedly effective in controlling in-situ carcinoma of the larynx, and there is no evidence whatsoever that irradiation of these lesions will cause them to become invasive. The local control for this carcinoma is between 85 per cent and 90 per cent. The voice quality after radiotherapy is usually excellent: more than 75 per cent of patients have a normal voice after treatment. A radical dose is required: 50–52 Gy in 20 fractions over 4 weeks. Fields can be kept small – 4 4 cm or 5 5 cm – but care must be taken to include the whole of the anterior commissure in the high-dose volume.
Glottic carcinoma The majority (60–70 per cent) of laryngeal carcinomas arise from the vocal cords. The T staging is summarized in Box 16.16. The site distribution by T stage is shown in Figure 16.4. Nodal involvement at presentation is rare (less than 5 per cent), reflecting the poor lymphatic supply to the vocal cords.
70% 60% 50% 40% 30% 20% 10% 0% Tis
T1
T2
T3
T4
Figure 16.4 The distribution of glottic cancers according to T-stage: pooled data from the literature.
EARLY (T1, T2) INVASIVE CARCINOMAS OF THE GLOTTIS
The results of radiotherapy for early glottic cancers are so good – local control of more than 90 per cent for T1 tumours and more than 75 per cent for T2 tumours – that it is ethically difficult to introduce alternative treatments. Nevertheless, for highly selected patients, conservative surgical techniques may offer local control rates as good as those obtained by radiotherapy. Over the last decade, laser resection of early (T1, T2) glottic carcinomas has become established as a viable alternative treatment to radiotherapy, with which it provides comparable results.142 However, in the UK, where old habits die hard, it has yet to gain general acceptance. The criteria for suitability for laser excision of glottic cancer include anterior commissure not involved, patient fit for anaesthesia, and videostroboscopy only minimally abnormal. A few European centres also use the technique for some advanced disease, and although they report good results, such treatment will have to survive the rigours of randomized controlled trials before it becomes a standard therapeutic option. Other conservative surgical approaches to early glottic tumours are possible: fronto-lateral partial laryngectomy, laryngofissure with cordectomy, hemilaryngectomy, and anterior commissure resection. The main problem is that voice quality is not as reliable after conservative surgery as it is after radiotherapy. In the Mayo Clinic series of 159 patients treated with open conservative surgery,143 14 per cent had ‘poor’ voice and, disconcertingly, 11 patients died
386 Head and neck cancer
from laryngeal cancer. Within North America there are significant differences in treatment philosophy, the USA tending towards a more surgically orientated approach, with the Canadians favouring radiotherapy. There is no evidence that the Canadian approach is associated, on a population-wide basis, with inferior outcomes.144 In the UK, the vast majority of patients with T1 and T2 tumours of the glottis are treated with a policy of radical radiotherapy, with laryngectomy held in reserve for patients whose tumours persist or recur after primary radiotherapy. With this policy, about 90 per cent of patients with T1 tumours and 70–75 per cent of patients with T2 tumours are cured without requiring laryngectomy. Long-term causespecific survival is around 90 per cent for T1 tumours and 80–85 per cent for T2 tumours. Crude survival is much lower, typically around 50–60 per cent at 10 years. This difference reflects the high rate of intercurrent deaths and second malignancies in these patients. Intercurrent deaths can cause difficulties with attribution in the long-term follow-up of patients with tumours of the upper aerodigestive tracts. If treatment produces distortion and fibrosis of the structures around the epiglottis and larynx, this may interfere directly with the protection of the airway during swallowing, and aspiration may occur. If significant aspiration goes unnoticed, and deaths are simply attributed to bronchopneumonia, this will underestimate the number of deaths occurring as a result of treatment-related complications. Any disparity in rates of death from intercurrent causes in comparisons of treatment for head and neck cancer should always be looked at carefully: this may be a clue to a significantly higher rate of complications in one treatment arm. This problem was particularly apparent in the Edinburgh neutron/photon trials in head and neck cancer, in which there were many more intercurrent deaths in patients treated with neutrons.145 T3 TUMOURS OF GLOTTIS
Surgery for laryngeal cancer is evolving. Trans-oral laser excision has been used in carefully selected patients with T3 tumours without compromising local control.146 The operation of supra-cricoid partial laryngectomy can, again in carefully selected patients, preserve voice and swallowing without compromising cure.147 Nevertheless, fewer than 10 per cent of patients with T3 glottic tumours are suitable for conservative surgery and so the initial choice is almost always between total laryngectomy and radiotherapy. The choice is often difficult. Infiltrative ulcerative tumours are best treated surgically, whereas the more exophytic tumours often respond well to radiotherapy. The general condition of the patient is also important, as many of these patients are not medically fit for anaesthesia. The patient should be involved in the discussions and decisionmaking from the start. Patients’ attitudes to possible compromises between survival and voice preservation differ. Some patients value survival at all costs and will pay virtually any price to increase their certainty of cure.
Other patients value their voices highly and would be prepared to compromise their chances of survival in order to retain normal speech.102 A detailed analysis of specialists’ recommendations for the management of laryngeal cancer showed that them to be heavily influenced by specialty (otolaryngologists as opposed to radiotherapists) and by geography (the USA and Australasia as opposed to Canada, the UK and Scandinavia).148 Surgeons tended to recommend surgery; radiotherapists were less likely to do so. Surgery was more likely to be recommended by clinicians in the USA and Australasia. Primary radiotherapy will control around 40 per cent of T3 glottic cancers. Case selection, with more unfavourable tumours being referred for surgery, invalidates any direct comparison of survival data between surgery and radiotherapy for T3 tumours of the vocal cord. No randomized trial has addressed this question. The influential Veterans Administration Study91 on neoadjuvant chemotherapy for laryngeal cancer has shed no real light on the issue. It has convinced many ENT surgeons that laryngectomy is not always necessary but, because there was no randomization to radiation alone, the relative contributions to tumour control of the radiotherapy and chemotherapy in this study cannot be assessed. The RTOG 91-11 trial89 set out to resolve this issue. Patients with advanced laryngeal cancer were randomized to radiotherapy alone or to platinum/5-FU induction chemotherapy followed by radiotherapy, or to synchronous chemo-radiation with platinum and radiotherapy. Local control at 2 years was significantly better in patients randomized to chemo-radiation (78 per cent) compared with those treated with radiation alone (56 per cent) or with induction chemotherapy followed by radiotherapy (61 per cent). The addition of any chemotherapy halved the rate of metastatic disease – from 16 per cent with radiotherapy alone to 8 per cent or 9 per cent. Treatment allocation had no effect on 5-year survival, which was between 54 per cent and 56 per cent in the three arms. T4 TUMOURS OF GLOTTIS
The majority of patients with T4 glottic cancers who have operable disease and are fit for surgery should probably be treated by laryngectomy or pharyngolaryngectomy. This is particularly true for tumours involving the pyriform fossa, for which the results from radiotherapy are particularly poor – less than 20 per cent local control at 5 years. Early cartilage invasion is not, of itself, a contraindication to radical irradiation. Salvage laryngectomy is, however, rarely possible in patients with T4 glottic tumours who fail radiotherapy. This is a reflection of their poor general condition and the extensive local destruction that can be caused by both the primary tumour and post-irradiation perichondritis. Perichondritis of the thyroid cartilage is a potentially fatal complication of these tumours. Careful supervision after radiotherapy and prompt treatment with antibiotics are essential for its prevention. Surgical series on T4 glottic
Assessment of patients 387
tumours report long-term local control rates of between 30 per cent and 55 per cent.
RADIOTHERAPY TREATMENT TECHNIQUES FOR GLOTTIC CARCINOMAS
Most early glottic tumours can be treated using parallel opposed lateral fields with wedges used to compensate for the decreased separation of the neck anteriorly. In patients with short, fat necks, it may be difficult physically to accommodate lateral fields, and oblique anterior fields with appropriate wedges may be used. Relatively small fields can be used for early tumours provided that careful attention is paid to the anterior margin. The anterior commissure may lie as little as 2–3 mm below the skin surface, and the anterior field margin (defined at 50 per cent) should therefore extend beyond the skin surface. In patients with thin necks, bolus may be required to ensure adequate dose to the anterior commissure. All patients should be treated using an immobilization shell and a well-collimated beam from a 4–6 MV linear accelerator. All fields should be treated daily. It is no longer acceptable to use cobalt units for such treatment. The field should extend from the level of the hyoid to the lower border of the cricoid cartilage, and the posterior border should overlie the vertebral bodies. A randomized study from Japan looked at the influence of field size on local control and complications in patients with T1N0M0 glottic carcinoma. There was no difference in relapse-free survival between patients randomised to 6 6 cm fields compared with patients randomised to 5 5 cm fields.149 However, patients treated with the larger fields had a significantly higher incidence of chronic arytenoid oedema. It is important to remember that the physical size of the larynx varies on an individual basis, and field sizes should be individually appropriate rather than standardized. For T1 tumours, a 3-week schedule of treatment is effective: 50–52.5 Gy in 16 fractions in 21 days.40 In patients with more advanced primary tumours, larger fields may be necessary, but this will increase the risk of significant late morbidity. A dose of 52–55 Gy in 20 fractions over 4 weeks, or equivalent, is adequate for most glottic tumours. If field sizes exceed 50 cm2, it may be necessary to limit the dose to 52 Gy in order to avoid unacceptable late complications.
Subglottic tumours These tumours are very uncommon – less than 5 per cent of all laryngeal tumours. The T staging is shown in Box 16.17. Unlike glottic tumours, they often spread to nodes; the nodal drainage is to the low neck, supraclavicular fossa and upper mediastinum. The clinical presentation is usually with wheeze or stridor, occasionally as a lump in the neck. Because tumours regress towards their site of origin,
the distinction between a true subglottic tumour and a glottic tumour with subglottic extension can sometimes only be made by observing the regression of tumour during treatment. For tumours arising at the level of the thyroid ring, surgery offers the best treatment. Tumours arising from the conus elasticus are more likely to respond to radiotherapy. Even advanced (T3, T4) subglottic tumours may be controlled with radiotherapy: overall local control rates for radiotherapy in subglottic tumours range from 30 per cent to 70 per cent. Long-term survival is 30–50 per cent with radiotherapy and 40–50 per cent with surgery. The radiation fields need to include the lymph nodes of the low neck, the upper mediastinum and supraclavicular fossa as well as the primary tumour. Cruciate anterior and posterior opposed fields are usually required, with a posterior cord block to keep the dose to the spinal cord within tolerable limits. The field size is usually such that a dose of 50 Gy in 20 fractions over 4 weeks, or its equivalent, cannot be exceeded.
Laryngectomy after radiotherapy The indications for laryngectomy after radiation therapy provide interesting clues as to an institution’s radiotherapeutic philosophy. Those centres with an aggressive approach will perform relatively more laryngectomies for necrosis and fewer for recurrence. The pattern will be reversed in centres with a less vigorous approach. In a series of 376 patients with laryngeal cancer reported from Edinburgh, 56/376 underwent laryngectomy. Of the 56 larynxes removed, 43 (77 per cent) contained tumour and 13 (23 per cent) showed necrosis only. There was no difference in symptoms (pain, hoarseness, stridor) between patients with residual tumour and those with necrosis.150 The overall rate for laryngeal necrosis in this series was 13/376 (3.4 per cent). The patients who required laryngectomy for necrosis enjoyed excellent survival – 92 per cent at 5 years. This compares with 59 per cent 5year survival in the patients requiring laryngectomy for recurrence. These data emphasize the importance of careful follow-up, a high degree of suspicion for recurrence, and a low threshold for performing microlaryngoscopy under anaesthetic in patients irradiated for laryngeal cancer. Previous radiotherapy undoubtedly increases the complication rate for laryngectomy. In the Edinburgh series of previously irradiated patients, the complication rate was high – 30/56 (54 per cent) serious complications. Fistula was the commonest complication, affecting 15/56 (27 per cent) of patients. Other complications included infection, wound dehiscence and arterial rupture.150 Follow-up should be every 4–8 weeks for the first year after treatment. In patients who are difficult to examine by indirect laryngoscopy, direct laryngoscopies should be performed at 3 and 6 months after treatment. Follow-up during the second year should be every 2 months. Thereafter, the intensity of supervision can be decreased.
388 Head and neck cancer
Although occasional problems arise after 5 years, it is reasonable to discharge patients from routine follow-up at that time. Using historical data, it is possible to indicate outcome in an imaginary cohort of 100 patients with advanced laryngeal cancer treated with radical radiotherapy. Radiotherapy alone will produce long-term local control in between 50 and 56 patients. Salvage surgery will be attempted in 28–34 patients and this will be successful in between 15 and 20 patients. There will be between 10 and 22 patients with persistent or recurrent disease who are, for one reason or another, unsuitable for salvage surgery. Ultimately, between 65 and 76 patients will obtain local control of their disease. The important feature of this analysis is that it confirms the viability of a policy of radical radiotherapy (or chemoradiation), with salvage surgery for failure, in the treatment of laryngeal cancer. Around one-third of patients will die from causes other than laryngeal cancer. Most of these intercurrent deaths are due to smoking-related diseases – 50 per cent of the deaths in the Edinburgh series of patients requiring salvage laryngectomy. Encouraging patients treated for laryngeal cancer to stop smoking should improve their immediate prognosis, but unfortunately can do little to prevent the long-term harmful effects of the cigarettes they have already smoked. Unless smoking habits change or strategies for secondary prevention prove effective, long-term survival rates for patients with laryngeal cancer will always be eroded by a high rate of intercurrent death – despite curing the majority of patients.
The nasopharynx Tumours of the nasopharynx are biologically distinct from other tumours of the head and neck. These tumours are common in the Far East and North Africa. Three main factors are important for the development of classical nasopharyngeal carcinoma: genetic susceptibility, EBV infection and dietary factors. The importance of genetic factors in the origin of these tumours is illustrated by their familial tendency. There is often serological evidence of EBV infection. The main oncogenic protein associated with EBV infection is LMP-1 (latent membrane protein 1). EBVencoded RNAs (EBER) may interfere with apoptosis151 and thus permit expansion of abnormal clones of EBV-infected cells: in other words, EBV infection may cause a failure of tumour suppression. There may also be abnormal expression of c-kit after EBV infection,152 leading to the tantalizing possibility of using imatinib to prevent, or treat, the disease. The dietary factor that is particularly associated with nasopharyngeal carcinoma is the consumption of salted fish.153 Surveys in China have shown that fish from areas with a high incidence of nasopharyngeal carcinoma contained more carcinogenic nitrosamines than samples of fish from areas with a lower incidence of nasopharyngeal
carcinoma. Thus it is not simply the consumption of salt fish in general that is important: the type of salt fish is also relevant. It is difficult, from a Western perspective, to appreciate just how common nasopharyngeal carcinoma is in Southern China, Hong Kong and Taiwan. In some areas of Guangdong Province the incidence is as high as 40 per 100 000 per annum. If applied to the UK, this rate would produce 21 000 patients per year with carcinoma of the nasopharynx. In Taiwan, 60–70 per cent of all patients treated with radiotherapy are treated for nasopharyngeal carcinoma. Analysis of trends over time suggests that, as China has developed economically, dietary and socio-economic changes have led to a fall in the incidence of this disease.154 Nasopharyngeal carcinoma in China and the Far East may be a somewhat different disease from that encountered in Western Europe. In Hong Kong, less than 10 per cent of conventionally staged patients have metastatic disease at presentation. This increases to around 15 per cent when whole-body 18FDG-PET scanning is used.155 This contrasts with the disease in patients of North African origin, in whom bone metastases are more common, at both presentation and follow-up.156,157 The age–incidence peak for nasopharyngeal carcinoma in China is between 40 and 60 years; males outnumber females by 3:1. In Caucasian patients, there is a suggestion of a secondary peak in incidence around the age of 20 years. In the West, there is also a significant incidence of nasopharyngeal carcinoma in childhood. There are several pathological classifications for nasopharyngeal carcinoma. The World Health Organisation (WHO) classification is the most widely used; three categories are recognized. Type 1: keratinizing squamous, found in 25 per cent of Caucasians with nasopharyngeal carcinoma, not related to EBV infections, behaves more like a typical squamous-cell carcinoma of the head and neck, has the worst prognosis. Type 2: non-keratinizing carcinoma. Type 3: undifferentiated tumours (UCNT). ANATOMY
The roof of the nasopharynx is formed by the base of the skull, which slopes downwards and backwards to become continuous with the posterior pharyngeal wall. The anterior boundary is the posterior choanae and the free posterior edge of the nasal septum. Its lower limit is defined as the level of the uvula, opposite the second cervical vertebra. The lateral wall contains the fossa of Rosenmuller and the Eustachian tubes. The floor is the superior surface of the soft palate. The nasopharynx is difficult to assess clinically. Surgery is rarely used in the management of nasopharyngeal carcinoma and so it is only comparatively recently (with the advent of CT scanning and MRI) that it has been possible to obtain accurate anatomical information about its origins
Assessment of patients 389
and local spread. Most tumours originate in the fossa of Rosenmuller, and the earliest radiological sign is blunting of the angle at the Eustachian cushion. Computed tomography scanning has highlighted the importance of spread into the parapharyngeal space. This is defined as abnormal soft tissue lying in the space between the pharyngeal constrictors and the pterygoid plates. Another feature is the ability of the primary tumour to invade directly into the carotid sheath; previously, involvement of the carotid sheath was assumed to be due to nodal spread. Retropharyngeal nodes are involved early in the disease. They cannot be detected by clinical examination, but can readily be demonstrated on imaging. Around 30 per cent of patients with clinically negative necks will have occult nodal involvement on imaging. Imaging is also essential for assessing the involvement of the skull base: in the pre-CT era, between 10 per cent and 15 per cent of patients had skull-base involvement at presentation; nowadays the figure is nearer 35 per cent. This has obvious implications for comparing treatments over different time periods.116 Intracranial extension and extension into other sinuses is also more common than was previously detected. CLINICAL PRESENTATION
The nasopharynx, together with the pyriform sinus and the base of tongue, is a classic site for a silent primary tumour in the head and neck. About 75 per cent of patients have palpable neck nodes at presentation and the neck mass is the presenting symptom in from 40 per cent to 50 per cent of patients. The remaining patients usually present with nasal symptoms (obstruction, epistaxis) or with symptoms from the ear (deafness, tinnitus). Headache, often severe, central and unresponsive to standard analgesics, usually indicates a locally advanced tumour. Between 20 per cent and 25 per cent of patients have cranial nerve palsies at presentation. This indicates either erosion of the base of skull or spread via the various exit foramina of the cranial nerves. Cranial nerves VI and V are the most commonly involved (50–70 per cent of patients with cranial nerve palsies at presentation). A syndrome of syncope has been described in nasopharyngeal carcinoma and similar problems may occur in other patients with head and neck cancer. The mechanisms are complex but predominantly involve pressure on the carotid sinus, causing reflex bradycardia, and stimulation of the IXth nerve, again causing a fall in heart rate as well as hypotension due to direct vasodilatation. A variety of para-neoplastic syndromes has been described in association with nasopharyngeal carcinoma: hypertrophic osteoarthropathy and the syndrome of inappropriate ADH secretion occur rarely. Younger patients, particularly those from North Africa, may present with a syndrome of nasopharyngeal carcinoma, leukaemoid changes in the peripheral blood, and pyrexia of unknown origin.
CLINICAL ASSESSMENT
Examination under anaesthesia, with adequate visualization of the primary tumour and biopsy, is essential both for confirming the diagnosis and for clinical staging. Adequate imaging (CT and/or MRI) is essential and should include the skull base and lower parts of the anterior and middle cranial fossae and should extend down to the level of the suprasternal notch. Contrast enhancement allows discrimination between vascular structures and soft tissue. Magnetic resonance imaging is extremely useful in assessing nasopharyngeal carcinoma and has the potential to detect early evidence of spread beyond the primary tumour, for example infiltration of the tensor veli palatini. Other routine investigations should include chest X-ray, routine haematology, liver function tests (including LDH) and EBV serology. The issue of routine bone scanning and liver ultrasound is contentious: young North African patients are at particular risk of having disseminated disease at presentation and should be routinely imaged; other patients should only be investigated if they have suggestive symptoms or signs. If there is any doubt about the histological diagnosis, expert pathological review, with re-biopsy if necessary, is essential, particularly in non-endemic areas. The differential diagnosis of tumours at this site is extensive and includes lymphoma, rhabdomyosarcoma, extramedullary plasmacytoma, amelanotic melanoma, polymorphic reticulosis, Wegener’s granulomatosis, sarcoidosis and pharyngeal tuberculosis. STAGING
The clinical staging of nasopharyngeal tumours reflects the fact that, other than biopsy, surgery has little role to play in this disease. The staging for nasopharyngeal tumours is therefore distinct from that of other head and neck cancers (Boxes 16.18 and 16.19; Table 16.13). In the Far East, the Ho system, rather than TNM, was used for many years. This makes comparisons between the West and the East difficult. The N staging in the Ho system is entirely dependent on the level of nodal involvement; size, laterality and fixation are not considered. The UICC system uses size and bilaterality as its criteria. The Ho system T stage depends on extent of the primary tumour, bone erosion and cranial nerve involvement. T stage according to the UICC depends mainly on anatomical extent and does not acknowledge the prognostic importance of cranial-nerve involvement. MANAGEMENT
Radiotherapy is the mainstay of treatment. The role of chemotherapy is slowly being clarified. There is good evidence from randomized controlled trials that chemo-radiation is superior to radiation alone158–164 (Table 16.7). The roles of induction chemotherapy and adjuvant therapy are less clear. A recent overview165 suggested that adjuvant
390 Head and neck cancer
chemotherapy was of little benefit, that there was clear evidence of benefit from concomitant chemo-radiation, and that there was slight benefit from neoadjuvant chemotherapy. For stage III and IV disease it appears that platinum/5FU given synchronously with radiotherapy confers a survival advantage. The acute reactions to concomitant chemotherapy and radiotherapy are undoubtedly more severe and prolonged. The potential late complications of the combination are not fully documented, but it is noteworthy that in the NPC-9901 trial161 the benefit from
Box 16.18 UICC (2002) T staging for carcinomas of the nasopharynx T1: Tumour confined to nasopharynx T2: Tumour extends to soft tissues T2a: to oropharynx and/or nasal cavity without parapharyngeal extension T2b: any tumour with parapharyngeal extension T3: Tumour involves bony structures and/or paranasal sinuses T4: Tumour with intracranial extension and/or involvement of cranial nerves, infra-temporal fossa, hypopharynx, orbit or masticator space
Table 16.13 American Joint Committee on Cancer (AJCC) stage grouping (2002) for nasopharygeal cancer Stage group
Box 16.19 UICC (2002) N-staging for carcinomas of the nasopharynx
0 I IIA IIB
NX: Regional nodes cannot be assessed N0: No regional node metastases N1: Unilateral nodal metastasis above the supraclavicular fossa and less than 6 cm in diameter N2: Bilateral nodal metastases above the supraclavicular fossa and less than 6 cm in diameter N3: Metastasis greater than 6 cm and/or to supraclavicular fossa N3a: greater than 6 cm N3b: to supraclavicular fossa
III IVA IVB IVC
T
N
M
Tis 1 2a 1,2,2a 2b 1,2a,2b 3 4 Any Any
0 0 0 1 0,1 2 0,1,2 0,1,2 N3 Any
0 0 0 0 0 0 0 0 0 1
Stage groupings, based on TNM, for cancers of the nasopharynx.225
Al-Sarraf (1998) Lin (2003) Kwong (2004) Chan (2005) Zhang (2005) Lee (2005) Wee (2005)
.25
.5
.75
Chemort better
1
1.25 XRT alone better
Risk ratio
Figure 16.5 Relative risk, with 3-year overall survival as the end point, in randomized trials in which synchronous chemoradiation was compared with radiotherapy alone for carcinoma of the nasopharynx. For references to trials, please see text.
Assessment of patients 391
adding chemotherapy, in terms of local control, was offset by an increase in the number of intercurrent deaths. The net result was that survival at 3 years was no different in the two arms of the study. RADIOTHERAPY TECHNIQUE
A superb account of radiotherapeutic techniques was given by Ho in the previous editions of this textbook. The essential principle of his technique is to use small lateral fields and a supplementary anterior field to treat the nasopharynx. The neck is treated using an anterior field with a midline shield. An alternative technique is to use larger lateral fields to treat the nasopharynx, the nasal cavity and the upper neck, and to treat the low neck using an anterior split cervical field. A two-phase technique is required in order to shield the spinal cord at tolerance and also to change the level of the junction between the anterior and the lateral fields. The dose required to treat carcinoma of the nasopharynx is 66–70 Gy in 33–35 fractions. A significant incidence of late complications is virtually inevitable. In a series from Queen Elizabeth Hospital, Hong Kong, of 4527 patients assessed for late complications, 1395 (31 per cent) had significant problems: 322/4527 (7 per cent) had severe late morbidity and 62/4527 (1.4 per cent) died as a result of treatment. Most of the fatalities were due to CNS damage, temporal lobe necrosis or damage to the brain stem.166 The commonest form of severe late damage is hearing loss. This can arise from several treatment-related causes: serous otitis media, direct damage to the cochlea by radiation, and damage to the auditory pathways in the mid-brain. Adding cis-platinum to radiotherapy significantly increases the risk of sensori-neural hearing loss.167 Soft-tissue fibrosis is the commonest cause of minor morbidity; when the temporo-mandibular joint is involved, trismus results. Dry mouth and dental problems are common. The incidence of hypothalamo-pituitary dysfunction depends on how hard it is looked for: clinical problems were recognized in only 4 per cent of patients in the retrospective review from Queen Elizabeth Hospital, but up to 60 per cent of patients followed prospectively have demonstrable endocrine dysfunction within 5 years of irradiation. The actuarial incidence of hormone deficiencies at 5 years may be much higher, with rates of deficiency of 60 per cent for growth hormone, 30 per cent for gonadotrophins, 25 per cent for adrenocorticotrophic hormone and 15 per cent for thyroid-stimulating hormone. A significant proportion (10–20 per cent) of patients with nasopharyngeal carcinoma have clinically evident residual tumour in the nasopharynx at the completion of radiotherapy. A variety of techniques for boosting the dose to the primary tumour have been proposed: small external-beam fields; 198Au grain implant; intracavitary therapy with 137Cs or 192Ir; 125I seed implant; and boost with IMRT. Conformal boost techniques have been disappointing,168 but techniques using IMRT appear promising.75,169,170
MANAGEMENT OF RECURRENCE
Patients in whom nodal disease is the sole recurrence or failure should be treated surgically; re-irradiation of the neck nodes is of little benefit. Patients in whom local recurrence is the only problem can be treated by re-irradiation. The long natural history of nasopharyngeal carcinoma means that although long-term survival after re-treatment may only be about 15 per cent, local control, and subsequent relief of symptoms, can be achieved in 30–35 per cent of re-irradiated patients. Late morbidity is a problem if re-irradiation is given solely by external beam. The best approach is probably to give 20–30 Gy by external beam and then 40–50 Gy by intra-cavitary treatment or implant.76 Patients with metastatic disease should be considered for chemotherapy. The bleomycin, epirubicin and platinum regimen used at the Institute Gustave Roussy has yielded the best results so far. PROGNOSTIC FACTORS
The main adverse prognostic factors for survival are presence of metastatic disease, higher nodal stage (when the Ho system is used), cranial-nerve paresis and base of skull involvement. Male patients and patients with parapharyngeal disease may also have a worse prognosis. Levels of circulating EBV DNA in the blood may correlate with prognosis: post-treatment levels can predict both overall survival and freedom from relapse.171 In a multivariate analysis including age, gender, performance status, histological type, T-stage, N-stage, AJCC stage and whether or not plasma EBV DNA was detectable after treatment, the only independently significant factor was the post-treatment EBV DNA status. Patients with detectable EBV DNA had a hazard ratio for death of 22.9 (95% CI 3.0–173.5) compared with patients whose EBV DNA levels were undetectable.172 Prognosis has improved steadily over the past 20 years. This reflects general technical improvements in radiotherapy technique and better pre-treatment evaluation rather than any specific contribution from altered fractionation or the addition of chemotherapy. It also illustrates the danger of using historical controls when evaluating treatment for nasopharyngeal carcinoma. It is likely that this trend will continue as treatment planning improves, IMRT becomes more widely available and synchronous chemotherapy is more often used. CAUSES OF TREATMENT FAILURE
The causes of failure in the series from Queen Elizabeth Hospital, Hong Kong,166 are summarized in Table 16.14. Half the patients died from nasopharyngeal carcinoma, 35 per cent survived, 7 per cent died from intercurrent or unknown causes, 1 per cent died from treatment-related complications, 1 per cent died from second malignancies and 7 per cent were lost to follow-up.
392 Head and neck cancer
Table 16.14 Causes of failure in patients treated for carcinoma of the nasopharynx in Hong Kong
Metastatic at presentation Unfit for radical Rx Metastases as first site of failure LR failure then metastases Regional failure alone Failure of initial treatment to achieve LC Local failure after initial LC
N
(%)
292 258 924 521 177 564 891
6 5 19 11 4 12 20
Abstracted from Lee et al. 1992.166 LC, local control; LR, loco-regional.
Tumours of the nose and para-nasal sinuses Tumours of the nose and para-nasal sinuses are characterized by histological and anatomical heterogeneity. The diversity of the population of normal cells found in the lining of the sinuses is paralleled by the diversity of the histological types of tumour that are encountered in this region: squamous carcinoma, salivary-gland tumours, adenocarcinoma, inverting papillomas, melanoma, esthesioneuroblastoma, lymphoma and sarcoma. The nasal vestibule is lined with squamous epithelium but the nasal cavity and sinuses are lined with columnar ciliated epithelium. Goblet or mucous cells are interspersed amongst the columnar cells, and beneath the basement membrane there is a virtually continuous layer of mucous and serous glands. There is diffuse lymphoid tissue throughout the region. In the olfactory region, the lining comprises three types of cell: bipolar olfactory nerve cells, basal cells and sustentacular cells. ANATOMY
The nasal vestibule is defined as that small area of the nasal fossa just proximal to the nares. The nasal cavity begins at the squamo-columnar junction and extends backwards to the posterior choanae. The nasal septum divides it into left and right, and laterally it is bounded by the ala nasi. The nasal septum ends inferiorly as the columella, the structure that separates the two nostrils. Tumours of the columella should be classified with tumours of the nasal vestibule. The olfactory portion of the nasal cavity comprises the roof, the superior concha and the adjacent upper part of the nasal septum. The maxillary sinuses are pyramidal in shape. The lateral wall of the nasal cavity lies medially, the roof is formed by the orbital floor, the alveolar process forms the floor and the apex extends into the zygomatic process of the maxilla. The cheek lies anteriorly; the pterygoid plates and pterygo-palatine fossa lie posteriorly. The maxillary sinus can be divided into a superior portion (the suprastructure) and an inferior portion (infrastructure)
by an imaginary line, Ohngren’s line, drawn between the medial canthus and the angle of the jaw. This division has prognostic significance. Tumours of the suprastructure have a worse prognosis, present later and may invade locally into the orbit and adjacent structures. The ethmoidal sinuses lie within the ethmoid bone between the medial plate of the ethmoid (part of the lateral wall of the nasal cavity) and the lateral plate of the ethmoid (part of the medial wall of the orbital cavity). The top of the ethmoid bone articulates with the frontal and splenoid bones; its floor articulates with the vomer and the septal cartilage of the nose. The frontal sinuses lie within the frontal bone, immediately above the orbits. Tumours of the frontal sinuses can easily spread into the orbits or to the anterior cranial fossa. The sphenoid sinuses lie within the body of the sphenoid bone. This sinus varies considerably in size and may extend into the occipital bone almost as far back as the foramen magnum. Tumours of the sphenoid sinus may extend directly into the middle cranial fossa. The lymphatic drainage of the nasal cavity and paranasal sinuses is poor. Tumours arising in this region rarely spread to lymph nodes. The main lymphatic drainage is to the submandibular, subdigastric and retropharyngeal nodes. Clinically enlarged nodes associated with these tumours should not be assumed to be metastatic: reactive changes, secondary to chronic infection within poorly draining sinuses, are common. The nasal vestibule and columella drain to the submandibular and subdigastric nodes. The posterior parts of the ethmoid sinus and the whole of the sphenoid sinus drain to the retropharynygeal nodes. THE NOSE
The distribution by site of 1676 tumours of the skin of the nose treated with radiotherapy173 is shown in Figure 16.6. The majority were basal-cell carcinomas, but there was a high incidence of squamous carcinomas in the vestibule (75 per cent) and columella (48 per cent). There was no overall difference in control for squamous carcinomas compared with basal-cell carcinomas. However, site was important. The control rate for carcinomas of the columella was 78 per cent (95% CI 60–95 per cent); the rate was 75 per cent (64–86 per cent) for tumours of the vestibule; and 95 per cent (93–96 per cent) for tumours at other sites. Given these differences, tumours of the nasal vestibule and columella should be considered separately from tumours at other sites on the nose. These latter tumours can be managed similarly to other tumours of the skin. The TNM staging system for skin tumours can also be used for staging carcinomas of the nasal vestibule and columella. Tumours of the vestibule will invade bone cartilage or the skin of the upper lip early, hence the high proportion of T4 tumours. This local invasion may occur surreptitiously, for example along the floor of the nasal fossa. This may, in part, explain the poorer prognosis of tumours at this site compared to other tumours of the nose. Surgery has only a limited role to play in the primary management of tumours of the nasal vestibule or columella.
Assessment of patients 393
n-1 fold Tip Bridge Vestibule Ala Columella Upper lateral Root 0%
5%
10%
15%
20%
25%
30%
Figure 16.6 Relative incidence, by site, of carcinomas of the nasal skin in a series of patients reported by Mazeron et al. in 1988.
An operation extensive enough to remove the tumour with adequate margins will be cosmetically unacceptable. A cosmetically acceptable operation will almost certainly leave positive histological margins and the subsequent requirement for radiotherapy. The appropriate radiotherapeutic technique for managing tumours of the nasal vestibule or columella is controversial. The debate concerns the relative roles of interstitial implantation and external-beam therapy. These tumours can be easily implanted using afterloaded 192Ir wires. The procedure can be carried out under local anaesthetic: 60 Gy at 0.5 Gy/h produces excellent tumour control and good cosmetic results. External-beam treatment should either be with electrons or with an anterior wedged pair of fields using megavoltage X-rays. The vexed question of the relationship between cartilage necrosis and low-energy X-rays can be avoided entirely if orthovoltage therapy is not used. The dose for megavoltage therapy should be 50–55 Gy in 20 fractions, or its equivalent. Only 5–15 per cent of patients have involvement of lymph nodes at presentation. Nodal relapse in the untreated neck is an uncommon course of treatment failure that occurs in only about 10 per cent of patients. There is therefore little indication for treating the nodes prophylactically in patients with carcinoma of the nasal vestibule. Surgical salvage is sometimes possible after failure of radiotherapy. Overall actuarial 5-year survival is between 75 per cent and 80 per cent; patients with T1 or T2 tumours have cause-specific survival of nearly 95 per cent at 5 years. NASAL CAVITY AND PARA-NASAL SINUSES
These tumours are uncommon and account for less than 10 per cent of all tumours of the head and neck. A proportion,
particularly the adenocarcinomas, may be related to occupational factors – wood dust, chromium or nickel fumes, leather dust. The distribution of histological types of tumour depends not only on the site of origin of the tumour but also, presumably as a reflection of occupational factors, on the centre reporting the results. Nevertheless, these differences will influence reported results. Doubts about the precise site of origin of the tumour will provide another source of potential confusion. In some series, it is not possible to define a site of origin in nearly 50 per cent of patients. This uncertainty will produce an effect analogous to stage shift on site-specific prognosis. NASAL CAVITY
Tumours of the nasal cavity characteristically present with nasal obstruction, discharge or epistaxis. Clinically, an ulcerating or infiltrating lesion is usually visible on nasoscopy. The T staging is summarized in Box 16.20. Nodal involvement at presentation is uncommon, between 10 per cent and 15 per cent in most series. Radiotherapy is the primary method of treatment. Anterior wedged oblique fields with 6-MV X-rays can be used, or, alternatively, a weighted anterior field plus wedged lateral fields. Iridium-192 implants may be particularly useful for treating tumours of the nasal septum. Overall survival after radiotherapy is between 50 per cent and 75 per cent. Local recurrence is the predominant course of treatment failure. Isolated nodal relapse is uncommon (less than 20 per cent of patients). Local control is a better yardstick than overall survival for assessing therapy, because approximately 50 per cent of deaths are unrelated to the original tumour.
394 Head and neck cancer
Box 16.20 UICC (2002) T staging for carcinomas of the nasal cavity and ethmoid sinus
Box 16.21 UICC (2002) T staging for carcinomas of the maxillary sinus T1:
T1:
Tumour limited to one subsite with or without erosion of bone T2: Tumour involving two subsites with or without bone invasion T3: Tumour extends to: floor of orbit or medial orbital wall; maxillary sinus; palate; cribriform plate T4a: Tumour invades anterior orbit, skin of nose or cheek, pterygoid plates, anterior cranial fossa, cribriform plate, sphenoid or frontal sinus T4b: Tumour invades any of: orbital apex; dura; brain; middle cranial fossa; cranial nerves (other than maxillary division of V); nasopharynx; clivus
The most important complication after radiotherapy for nasal cavity tumours is blindness. When tumour extends into or close to the orbit, the ipsilateral eye will be within the high-dose volume. The incidence of visual loss in patients treated with radiotherapy for nasal cavity carcinoma is between 5 per cent and 10 per cent and this must be accepted if tumour control is to be achieved. INVERTING PAPILLOMA (RINGERTZ TUMOUR)
This is a rare tumour that arises primarily in the nasal cavity. Although histologically benign, it has a conspicuous tendency to recur after surgical resection. If left untreated, it can also undergo malignant transformation, to either squamous carcinoma or adenocarcinoma. Radiotherapy has an important potential role in its management. Histologically, these tumours are characterized by hyperplasia and metaplasia of the surface epithelium with infolding into the supporting stroma. Papillary projections are often prominent, but unless there has been malignant change, the basement membrane remains intact. Careful histological review suggests that the carcinomas arise directly from the inverting papilloma rather than as coincidental separate neoplasms. Analysis with in-situ hybridization and the polymerase chain reaction suggests an association between HPV and inverting papilloma. The overall rate of malignancy associated with inverting papilloma is between 8 per cent and 10 per cent. Clinically, these tumours present as nasal obstruction and may be mistaken for polyps. The initial treatment is surgical; lateral rhinotomy and medial maxilloectomy or cranio-facial resection may be required. Although the surgical procedures sound drastic, the cosmetic and functional results are often extremely good. For small tumours, complete resection with good long-term control may be possible with per-nasal endoscopic resection. Recurrence after previous radical surgery or the presence of malignant change is an indication for radiotherapy.
Tumour limited to mucosa of maxillary sinus with no erosion or destruction of bone T2: Tumour with bone erosion or destruction extending into hard palate and/or middle nasal meatus, except extension to posterior wall of maxillary sinus and pterygoid plates T3: Tumour invades any of: bone of posterior wall of sinus; subcutaneous tissues; floor of orbit or medial orbital wall; pterygoid fossa; ethmoid sinus T4a: Tumour invades anterior orbit, skin of cheek, pterygoid plates, infra-temporal fossa, cribriform plate, sphenoid or frontal sinus T4b: Tumour invades any of: orbital apex; dura; brain; middle cranial fossa; cranial nerves (other than maxillary division of V); nasopharynx; clivus
A radical dose of radiotherapy is required: 50–55 Gy in 20 fractions, or its equivalent. The ipsilateral eye may have to be included within the treatment volume. A three-field technique with a weighted anterior and two wedged lateral fields produces a reasonable dose distribution. The longterm prognosis is good: 5-year survival is greater than 90 per cent. MAXILLARY SINUS
Early tumours of the maxillary sinus cause few symptoms: epistaxis and nasal obstruction. Tumours of the infrastructure may present as a palatal or upper alveolar swelling and can be mistaken for a primary tumour of the gum or palate. All patients with tumours of the gum or palate should have radiological investigation of the maxillary sinus to exclude the possibility that the tumour has arisen primarily at that site. Tumours that have spread beyond the sinus will cause swelling of the cheek, diplopia, proptosis, numbness and paraesthesia of the cheek or upper lip. Palpable lymphadenopathy at presentation is uncommon. The majority of tumours are squamous carcinomas. The UICC (2002) staging of maxillary sinus carcinoma is summarized in Box 16.21. It is obvious from the criteria for T staging that clinical examination alone is insufficient for staging maxillary sinus carcinomas. Adequate imaging is essential for the assessment of these tumours. The images will indicate the extent of the tumour and aid in the evaluation of operability and the planning of radiotherapy. The majority of tumours are advanced at presentation: 70–90 per cent are T3 or T4. Some early tumours can be successfully treated by surgery alone or by radiotherapy alone. However, combined treatment is required for the majority of patients. The optimal sequence for radiotherapy and
Assessment of patients 395
surgery has, in the past, been controversial. Preoperative radiotherapy may facilitate the subsequent surgery, but important prognostic information from the surgical pathology may be lost and post-surgical complications may be increased. Postoperative radiotherapy is now the preferred sequence in most centres, but extensive surgical procedures may be required for tumours of the maxillary sinus. A radical maxillectomy is usually the minimum procedure. If there has been orbital invasion, more extended surgery, possibly including orbital exenteration, may be required. Reconstruction will usually involve a myocutaneous flap and, when the orbital contents have been removed, a suitable prosthesis. The removal of the hard palate provides good access to the sinus cavity and facilitates drainage during postoperative radiotherapy. In the longer term, an obturator will be required to permit eating and speaking. The management of these patients is complicated and involves many disciplines; rehabilitation is critical to successful functional results. Radiotherapy can be given using an anterior and wedged lateral fields. When the orbit has been invaded, the eye shielding on the anterior field cannot be used and the dose to the eye may well exceed radiation tolerance. When megavoltage beams are used to treat tumours invading the cheek, bolus will be required to eliminate skin sparing and bring the high-dose volume anteriorly. Tumours that extend across midline will require both right and left wedged lateral fields. It is sometimes possible to spare some of the eye by angling these fields posteriorly, but the problem with doing this is that the dose to the spinal cord will be increased. The dose required postoperatively is 50 Gy in 20 fractions, or its equivalent in 30 fractions. When radiotherapy is used alone, the doses need to be a little higher: 55 Gy or 65 Gy. The probability of 5-year survival varies according to T stage: from 60 per cent for early tumours to less than 40 per cent for T4 tumours. Results with radiotherapy alone may be as good as those achieved with combined surgery and radiotherapy. This calls into question the absolute need for surgery in all patients with tumours of the maxillary sinus. Combined treatment is possibly unnecessary for T1 and T2 tumours and is often impossible for T4 tumours. It is probably only in the T3 tumours, therefore, that combined treatment will be required. The main cause of treatment failure is failure of local control – more than 80 per cent of all failures. Only about 5–10 per cent of patients relapse in the regional nodes. About 5 per cent of patients develop distant metastases and less than 5 per cent develop second primaries. The main late complications of radiotherapy are ocular and neurological. The temporal lobes lie closer to the antrum than is often appreciated and may, for advanced tumours, be within the high-dose volume. Temporal lobe necrosis, often manifest as temporal lobe epilepsy, may occur after radiotherapy. The brain stem and cord are also vulnerable and only careful treatment planning will avoid overdosage. If the eye is within the high-dose volume,
eventual blindness – because of severe dry eye, cataract or retinal damage – is virtually inevitable. ETHMOID SINUS
Tumours of the ethmoid sinus present with headache, visual disturbances, nasal obstruction and, in the later stages, lateral displacement of the globe of the eye. Orbital involvement is common and the ipsilateral eye may have to be sacrificed; this applies whether treatment is by surgery or by radiotherapy. These tumours are extremely rare and a variety of histological types may be encountered: squamous carcinoma, adenocarcinoma, adenoid cystic carcinoma, sarcoma, mucoepidermoid carcinoma and esthesio-neuroblastoma. This makes comparison of various approaches to treatment difficult. The 5-year survival rates reported in the literature vary from 35 per cent to 50 per cent. Local recurrence is the main cause of treatment failure; distant relapse is uncommon. Radical surgery for these tumours may involve craniofacial resection and orbital exenteration. Radical radiotherapy requires a dose of 50–55 Gy in 20 fractions, or its equivalent. The optimal field arrangement is to use a heavily weighted anterior field, with bolus to the medial canthus if necessary, and wedged lateral fields. Computed tomography planning is essential to avoid overdosing the contralateral eye, and also to ensure that any intracranial extension is adequately encompassed. SPHENOID SINUS
The main symptom of carcinoma of the sphenoid sinus is headache, which is persistent, severe and central. Cranialnerve palsies occur later. These tumours are usually advanced at presentation. Radical treatment with radiotherapy may be attempted, but the chance of cure is low. Surgery has little to offer, either as primary or as secondary therapy. For the majority of patients, palliative treatment is all that can be offered.
Tumours of the ear The ear can be divided anatomically into the pinna, the external auditory canal, the middle ear and the mastoid antrum. SQUAMOUS CARCINOMA OF THE MIDDLE EAR
These tumours are extremely rare, with an incidence in the UK of less than 1/1 000 000 per annum. There is an undoubted association with chronic suppurative ear disease: 65–70 per cent of patients with tumours of the middle ear have a history of chronic ear infection. The incidence of malignancy in patients with chronic ear disease has been estimated at between 1 in 4000 and 1 in 20 000. A crude calculation would suggest that chronic ear infection of
396 Head and neck cancer
10 years’ duration would be associated with a five-fold to twenty-five-fold risk of developing a carcinoma of the middle ear. Tumours of the middle ear can be classified by site of origin into petromastoid tumours and tympanotubal tumours. The petromastoid tumours present with pain and discharge and VIIth nerve palsy. The tympanotubal tumours can spread along the Eustachian tube and, with consequent multiple palsies of cranial nerves, imitate nasopharyngeal tumours. Nodal spread is uncommon in petromastoid tumours because the otic capsule provides a barrier to tumour spread. Tumours of the middle ear are often diagnosed late; the symptoms of the associated ear infection mask those of the tumour. Only about 20 per cent of patients have T1 tumours at presentation according to the staging system proposed by Stell.174 T1: Tumour limited to the site of origin, with no facial-nerve paralysis and no bone destruction. T2: Tumour extends beyond the site of origin, indicated by facial paralysis or radiological evidence of bone destruction, but no extension beyond the organ of origin. T3: Clinical or radiological evidence of extension to surrounding structures, dura, base of skull, parotid gland, temporo-mandibular joint etc. TX: Patients with insufficient data for classification. About 30 per cent of patients present with disease that is too far advanced to permit radical therapy. Clinically, there is destruction of the middle ear and a visible tumour obscures any proper assessment of the extent of local invasion into the structures of the middle ear and adjacent bone. Both CT and MRI are therefore essential in the assessment of patients before treatment. These tumours are best managed by a combination of conservative surgery and postoperative radiotherapy. The advantages of using surgery first are that it removes the bulk of the tumour, any infection in bone is physically removed, and the resulting cavity can be easily inspected and monitored during treatment. The surgical specimen also provides information which, taken in conjunction with the radiological investigations, enables the postoperative radiotherapy to be planned accurately and rationally. Most tumours of the middle ear can be treated using an ipsilateral wedged pair of fields. Occasionally, for tumours extending deep into the temporal bone, a contralateral field may also be required. Additional fields using electrons may be useful when there is extensive involvement of the skin and subcutaneous tissues. The absorption of electrons by bone is unpredictable and it is unwise to rely on electrons to treat disease that is unquestionably invading bone. The postoperative radiation dose should be 50 Gy in 20 fractions over 4 weeks. Some very early tumours may be satisfactorily treated solely by radiotherapy; the dose should be 52–55 Gy in 20 fractions. The main risk from radiotherapy to the temporal bone is osteoradionecrosis. Provided
overdosage is avoided – and this will require careful attention to any hot spots seen on the treatment plan – and provided that any infection is aggressively treated, this complication is unlikely. The dose to the brain stem should be limited to 40 Gy in 20 fractions, or its equivalent. Surgery and radiotherapy for carcinoma of the middle ear will produce a 5-year survival rate of around 34 per cent. Failure to control local disease is the usual cause of treatment failure; distant metastases or nodal relapse are uncommon. Effective palliative treatment is extremely difficult for tumours of the middle ear. Extensive surgical resections produce morbidity that is often worse than the disease itself, and any relief of symptoms is often brief. Conventional radiotherapy is not very effective in controlling pain. Unconventional palliative schedules, such as 14.8 Gy in four fractions in 2 days repeated two to three times at 3weekly intervals, may provide some measure of control of the tumour with minimal morbidity or upset. Palliative chemotherapy using cisplatin and 5-FU may have a shortterm benefit. TUMOURS OF THE EXTERNAL AUDITORY CANAL
The majority of malignant tumours of the external auditory canal are squamous carcinomas, but other histologies may also be encountered: adenocarcinoma, adenoid cystic carcinoma and basal-cell carcinoma. Pain and discharge are the usual symptoms, although deafness and a mass around the ear may also occur. Mastoid tenderness may be present on physical examination, but the diagnosis is usually made on the presence of obvious tumour, often polypoidal, in the canal. Facial palsy is unusual (less than 20 per cent of patients). The external auditory canal can be divided into two portions: the osseous and the cartilaginous. There are no defining differences in presentation between the two sites. Tumours of the bony canal may spread to the middle ear and it is sometimes difficult to decide whether a tumour has arisen primarily within the middle ear or has invaded secondarily from the osseous canal. Computed tomography and MRI are required to stage patients adequately. The staging system proposed by Stell174 can also be applied to tumours of the external auditory canal. Conservative surgical excision and postoperative radiotherapy is the treatment of choice. Radiation dosage and technique are similar to those described for tumours of the middle ear. The prognosis for tumours of the external auditory canal is better than that for tumours of the middle ear: long-term survival is typically in the range of 30–50 per cent. TUMOURS OF THE PINNA
Tumours of the pinna usually arise from the skin. The majority are squamous carcinomas; a minority are basalcell carcinomas. Between 5 per cent and 10 per cent of all skin tumours arise on the pinna, a disproportionately high
Assessment of patients 397
incidence. Most of these tumours can be satisfactorily treated with radiotherapy. There are tumours arising at two sites, however, for which surgery may be the preferred treatment: those arising around the origin of the external auditory canal and those arising at the retroauricular sulcus. Electron-beam therapy produces excellent results for carcinoma of the pinna. The beam energy should be chosen according to the thickness of the tumour; build-up will be required for the lower energies. For tumours on the helix, it is usually possible to shield the skin behind the ear using thin lead backed with wax, to absorb any knock-on photons. It may be necessary to use a wax earplug as a compensator for tumours arising around the concha. A dose of 45 Gy in ten fractions in 2 weeks is adequate for small (2 cm) tumours; for larger tumours, a dose of 50–55 Gy in 20 fractions in 4 weeks may be required. Perichondritis may complicate the treatment of these tumours, particularly when the cartilage has been directly invaded. The pinna becomes swollen, reddened and exquisitely tender. Fever may be present. The condition usually responds to prompt therapy with a broad-spectrum antibiotic. Late necrosis of cartilage is an accepted complication of treatment and usually occurs after trauma, particularly from spectacles; treatment with systemic antibiotics and topical steroids is usually effective. The prognosis for carcinomas of the pinna is excellent. Local control with radiotherapy as sole treatment is more than 90 per cent and survival is close to 100 per cent.
Rare and unusual tumours of the head and neck PARAGANGLIOMAS
The nomenclature and classification of these rare tumours are complex, inconsistent and confusing. They can arise at a variety of sites in the head and neck, and are described by a variety of synonyms: chemodectomas, glomus tumours etc. The literature is a poor guide, different authors applying different criteria for inclusion in ostensibly comprehensive series that often turn out to be fairly restricted. Management is controversial: extreme views are often taken with only slender, and highly selected, evidence to support them. Paragangliomas are benign tumours arising from the APUD cells of the chemoreceptor bodies. Histologically, they are highly vascular with cell nests (zellballen) separated by a dense capillary network. The cells are pale and uniform and on electron microscopy contain neurosecretory granules. Only 5–10 per cent of paragangliomas metastasize. They grow slowly, and respond slowly to non-surgical treatment. There is an association between chronic hypoxaemia (for example living at high altitude) and paragangliomas. Familial forms also occur, and there is some overlap with the MEN syndromes. Germline mutations in succinic dehydrogenase genes are associated with familial
paragangliomas.175 A significant proportion (around 20 per cent) of patients will have multiple tumours. This is particularly true of familial cases. In keeping with their neuroendocrine origin, these tumours may secrete enzymes or amines into the bloodstream: NSE, VMA and 5-HIAA. The simplest classification of paragangliomas is by site: ● ● ● ●
middle ear tympanicum jugular bulb middle ear jugulare hypopharynx vagale carotid bifurcation carotid body tumour.
The incidence is low (2/100 000 per annum) and females outnumber males by approximately 3 to 1. The symptoms and clinical findings depend on the site of origin. Tumours of the middle ear classically present with pulsatile tinnitus (75 per cent of patients), deafness (52 per cent), fullness in the ear (15 per cent) or dizziness or vertigo (10 per cent). Glomus vagale or carotid-body tumours usually present as a mass in the neck. Indentation of the pharyngeal wall is visible in approximately a third of patients with glomus vagale tumours, and there is usually a corresponding external mass just behind the angle of the mandible. A visible mass is usually present on examination of patients with tumours of the middle ear. About 30 per cent may have a visible polyp in the external auditory canal. Cranial-nerve palsies are frequent: VIII, IX, X for tympanojugulare tumours X and XII for vagale tumours. The investigation of patients with paragangliomas has been revolutionized by CT scanning and MRI. Magnetic resonance imaging angiography is an extremely useful investigation for assessing these tumours.176 Assessment of operability and the planning of surgery or radiotherapy are now far more precise than when angiography was the only reliable investigation. Detection of multiple tumours is also important, and radioisotope imaging, with either 131 I-MIBG or 111In-octreotide, can be extremely useful. The choice of treatment lies between surgical resection and radiotherapy. Surgical techniques have improved considerably over the past 10–20 years and much of the older data on surgically treated patients are now only of historical interest. Microsurgical techniques and advances in anaesthesia, with hypothermia and cardiopulmonary bypass, have improved surgeons’ ability to resect tympanicum and jugulare tumours whilst sparing blood vessels and nerves. Advances in diagnostic imaging mean that these tumours are now being diagnosed at an earlier, more easily resectable, stage. Radiotherapy alone offers more than palliation: local control can be achieved in around 75–90 per cent of carefully selected patients. Where radical surgery is feasible, the results are equally good. For younger, fitter patients with resectable tumours, radical surgery is the treatment of choice. Older patients, or those in whom surgical resection would produce unacceptable morbidity, can be treated with radiotherapy. Despite their neuroendocrine background, these tumours do not respond to treatment with somatostatin analogues.177
398 Head and neck cancer
Computed tomography planning is essential for tumours of the middle ear: an ipsilateral wedged pair of fields will usually produce a satisfactory dose distribution. Tumours extending deeply towards the midline may require a contralateral top-up field. The dose should be 50–55 Gy in 25 fractions over 5 weeks. This produces adequate control of disease without excessive morbidity. Response to treatment will depend upon how local control is defined. Paragangliomas are, by virtue of their natural history, tumours for which stable disease is an acceptable and useful therapeutic end-point. Roughly speaking, about 15 per cent of tumours regress completely after radiotherapy, 60 per cent regress by 50 per cent and 15 per cent remain stable; 10 per cent of tumours will grow in spite of treatment. Long follow-up is needed, because even an apparently well-controlled tumour may eventually re-grow. ESTHESIO-NEUROBLASTOMA (OLFACTORY NEUROBLASTOMA)
Esthesio-neuroblastoma is a small round-celled tumour that arises from the olfactory epithelium in the roof of the nasal cavity and may spread into the ethmoid sinus. The age incidence is bimodal, with peaks in the second and sixth decades. The histological appearance varies and the differential diagnosis from other small round-celled tumours may be difficult. Two features are characteristic of olfactory neuroblastoma: an intracellular fibrillary network and Homer Wright rosettes. These latter structures comprise a ring of cells around a central mass of eosinophilic fibrils. Patients usually present with anosmia, nasal obstruction, nasal discharge or epistaxis. Invasion of the orbit will produce proptosis and diplopia. The Kadish system has been used for clinical staging: A: confined to the nasal cavity B: involvement of the paranasal sinuses C: spread beyond the nasal cavity or paranasal sinuses. Treatment is primarily surgical, with craniofacial resection the standard procedure. These tumours are radiosensitive and postoperative radiotherapy should be given to patients with high-grade tumours, or those in whom excision is incomplete. The dose should be 50 Gy in 20 fractions, or its equivalent. Inoperable tumours, or tumours in patients who are unfit for surgery, can be treated with radical radiotherapy. A dose of 55 Gy in 20 fractions, or its equivalent, is necessary. The radiation treatment technique is similar to that used for ethmoid tumours. Computed tomography planning is essential; MRI may demonstrate tumour extension beyond that which is visible on CT. Late recurrence, beyond 10 years, is typical of this tumour. Distant metastases occur in up to 20 per cent of patients. There is some experience with chemotherapy in these tumours: regimens based upon adriamycin, cyclophosphamide and vincristine have traditionally been used. Platinum and etoposide regimens may also be effective.178
Overall, the 5-year survival rate with multi-modality treatment is nowadays around 70 per cent.179 JUVENILE ANGIOFIBROMA
This tumour, originally described by Hippocrates in the fourth century BC, characteristically occurs in adolescent males. The tumour is histologically benign but may behave in a malignant fashion, with local invasion and a tendency to recur after surgical resection. It is nodular, bluish red, glistening and frequently ulcerated, and obtains its blood supply from the internal maxillary artery. The majority of tumours arise at the junction of the nasal cavity and the nasopharynx. The tumour will grow readily into soft tissues and cavities, and erosion of bone is common. Clinically, these tumours present with unilateral nasal obstruction, facial deformity and/or epistaxis. Careful evaluation is required with CT and MRI and angiography. The last mentioned may usefully be combined with preoperative embolization. However, embolization alone is not adequate treatment. The management of these tumours is not straightforward.180 Primary surgery has become the treatment of choice, although radiotherapy can also achieve local control.181 The hazards of surgery should not be underestimated, but in general the principle of avoiding radiotherapy for benign disease, particularly in the young, should be adhered to. Endoscopic surgery now offers a less risky alternative to open operation – with no apparent compromise of effectiveness.182 In cases that recur after surgery, or where there is residual tumour after surgery, radiotherapy is unquestionably of benefit. Low doses are sufficient: 30 Gy in 22 fractions over 4.5 weeks will achieve local control, with minimal morbidity. There is little to be gained from the use of higher doses. There are anecdotal accounts that these tumours respond to oestrogen therapy. Because of unacceptable side effects in the adolescent male, such an approach would be a therapy of last resort. It is suggested that these tumours may regress spontaneously on maturation, and observation is acceptable in some rare cases. MERKEL-CELL CARCINOMA
This neuroendocrine tumour of the skin was first described in 1972. It has at various times also been called primary endocrine carcinoma of the skin or trabecular skin carcinoma. Cases presenting before 1972 were often classified as dermal adult neuroblastoma. Histologically, and in tissue culture, the tumour cells resemble small-cell carcinoma of the lung. Merkel-cell tumours may express NSE and/or calcitonin. Abnormalities of chromosomes, particularly translocations involving p36, 11 and 13 have been described in up to 50 per cent of cases. More than half the cases of Merkel-cell carcinoma described in the literature have involved the skin of the head and neck. Patients are typically elderly and present with a relatively short history. The lesions arise in the dermis and therefore rarely ulcerate; typically they appear as pinkish or
Assessment of patients 399
greyish blebs. About 16 per cent of patients have nodal disease at presentation; survival is poor for patients with positive nodes. Local surgery alone was initially thought to be sufficient treatment for these tumours, but the recurrence rate with this approach is around 40 per cent. The 5-year survival in patients treated with surgery alone is around 30 per cent. Wide local excision and radical node dissection is now the recommended surgical approach, and this should be combined with postoperative radiotherapy. The radiation dose should be 45–50 Gy in 20 fractions for prophylactic radiotherapy and 50–55 Gy in 20 fractions when microscopic disease is present. Both local control and survival will be improved if radiotherapy is routinely used in the management of Merkel-cell carcinoma. Approximately 25 per cent of patients develop distant metastases. Chemotherapy, using regimens similar to those used for small-cell lung cancer, has some logic, but the value is uncertain and the toxicity for the mostly elderly patients may be excessive.
there is doubt about the adequacy of the surgical margins. For patients whose tumours are inoperable, radiotherapy alone may be effective treatment. The dose required will be 50–55 Gy in 20 fractions, or its equivalent. Unconventional fractionation has also been used for melanomas; the 0,7,21 regimen (24 Gy in three fractions over 4 weeks) is simple and effective. Care must be taken with this regimen to the head and neck to ensure that vulnerable normal structures such as brain, spinal cord and retina are shielded after two fractions, as the full three fractions would exceed the tolerance of these organs. There is, however, no evidence that this is superior to conventional fractionation in head and neck melanomas. Long-term survival for patients with mucosal melanomas is poor, with less than 25 per cent in most series surviving for 5 years. Late relapses are typical, and so, given the limited data, it is unlikely that more than 20 per cent of patients with mucosal melanomas are cured by treatment. AMELOBLASTOMA
BASALOID SQUAMOUS CARCINOMA
This aggressive tumour was first recognized as a distinct pathological entity in 1986. Since then, they have been more frequently reported. Histology characteristically shows a basaloid pattern with squamous differentiation. The tumours are cytokeratin positive and there are no neurosecretory granules on electron microscopy. This suggests that basaloid squamous carcinoma is an aggressive variant of squamous-cell carcinoma rather than of any other cellular origin. On fine-needle aspiration cytology, these tumours can be confused with small-cell undifferentiated carcinoma or adenoid cystic carcinoma. The tumours occur most commonly in the hypopharynx, larynx and floor of mouth, although other sites of involvement, such as the buccal mucosa and oesophagus, have recently been recognized. The majority of patients have involved lymph nodes at presentation; extensive local invasion and distant metastases occur early in the course of the disease. The aggressiveness of the treatment should match that of the tumour. The value of chemotherapy, however, remains to be defined. MUCOSAL MELANOMAS
Only about 0.5–4 per cent of melanomas occurring in the head and neck arise from the mucosa. The nose and para-nasal sinuses are the commonest sites, and lymphatic spread is unusual. Since the structure of mucosa is different from that of skin, Clark’s histological staging system is inappropriate for mucosal melanomas. The management of mucosal melanoma has historically been surgical. However, there is increasing appreciation that these tumours may respond well to radiotherapy. Tumours that are resectable should be treated surgically, and postoperative radiotherapy should be given to those patients in whom the tumours were large or thick or where
This rare tumour arises in the vicinity of the tooth socket. Its precise histogenesis is uncertain. The classical site is in the mandible, close to the molar teeth. Patients are usually aged between 20 and 40 years and give a history of a painless swelling, often of several years’ duration. These tumours grow slowly and rarely metastasize. Plain X-rays show non-specific cystic changes, and bone involvement is often more extensive than would be suspected radiologically. The standard treatment is surgical excision. Islands of tumour may be found within the adjacent, apparently normal, bone and the excision may need to be fairly extensive. Curettage is not adequate therapy for ameloblastoma. PLASMACYTOMA
Plasmacytomas can occur as apparently isolated tumours in the head and neck. The sites most often involved are the maxilla, the maxillary antrum and the mandible. Clinically, the tumour presents as a tender swelling. In the extraosseous form of the disease this is simply a reddish purple mass; in the intra-osseous form there is concomitant destruction of bone, and pain may be severe. The diagnosis of plasmacytoma should not be accepted without a search for evidence of multiple myeloma. All patients should have a skeletal survey, bone marrow aspirate, electrophoresis of serum proteins, and urine tested for Bence Jones protein. The presence of a paraprotein does not necessarily imply that a patient has disseminated myeloma. The abnormal protein may be produced by the plasmacytoma itself and might prove useful in monitoring response to therapy. The primary treatment of plasmacytoma of the head and neck is with radiotherapy. A dose of 40–45 Gy in 20 fractions is adequate. Supplementary chemotherapy may be used in patients with extensive tumours, although the
400 Head and neck cancer
role of adjuvant chemotherapy in this context is unproven. Patients who have been treated for plasmacytoma should be carefully followed up, because many, if not all, may progress to develop systemic (multiple) myeloma.
KEY LEARNING POINTS ●
●
●
●
●
●
The majority of cases of Head and Neck Cancer are entirely preventable. If people smoked and drank less the incidence of Head and Neck Cancer would appreciably decrease. Successful management of Head and Neck Cancer requires a multiprofessional, multidisciplinary approach with the individual patient placed firmly at the centre of the decision-making process. Although the majority of Head and Neck Cancers are squamous carcinomas these tumours are remarkably heterogeneous, both biologically and clinically. There are no universal rules for managing Head and Neck Cancer. Systemic therapies (chemotherapy, biologicallybased therapeutic agents) have a limited but increasingly important role to play in the treatment of Head and Neck Cancer. Some of their apparent benefit may be related to their local radiosensitising effects rather than independent cytotoxicity. Technical developments in surgery and radiotherapy will decrease the toxicity and longterm morbidity associated with the treatment of Head and Neck Cancer.
REFERENCES 1 Jones TM, Hargrove O, Lancaster J, Fenton J, Shenoy A, Roland NJ. Waiting times during the management of head and neck tumours. J Laryngol Otol 2002; 116(4):275–9. 2 Scottish Health Statistics: the website of ISD Scotland. Head and Neck Cancer Statistics. 22nd May 2006 [cited; Available from: http://showcc.nhsscotland.com/isd/cancer_ definition.jsp;jsessionidF8D2FE497B878860D9EE74AEF0A9 C1A4?pContentID2867&p_applicCCC&p_serviceConten t.show& 3 Berrino F, Gatta G. Variation in survival of patients with head and neck cancer in Europe by the site of origin of the tumours. Eur J Cancer 1998; 34(14):2154–61. 4 Sant M, Aareleid T, Berrino F, et al. EUROCARE-3: survival of cancer patients diagnosed 1990–94 – results and commentary. Ann Oncol 2003; 14(Suppl. 5):v61–118. 5 Coleman MP, Gatta G, Verdecchia A, et al. EUROCARE-3 summary: cancer survival in Europe at the end of the 20th century. Ann Oncol 2003; 14(Suppl. 5):v128–49.
6 Munro AJ. Keynote comment. Deprivation and survival in patients with cancer: we know so much, but do so little. Lancet Oncol 2005; 6(12):912–13. 7 Kreimer AR, Clifford GM, Boyle P, Franceschi S. Human papillomavirus types in head and neck squamous cell carcinomas worldwide: a systematic review. Cancer Epidemiol Biomarkers Prev 2005; 14(2):467–75. 8 Fakhry C, Gillison ML. Clinical implications of human papillomavirus in head and neck cancers. J Clin Oncol 2006; 24(17):2606–11. 9 Scully C. Oral cancer; the evidence for sexual transmission. Br Dent J 2005; 199(4):203–7. 10 Slaughter DP, Southwick HW, Smejkal W. Field cancerization in oral stratified squamous epithelium; clinical implications of multicentric origin. Cancer 1953; 6(5):963–8. 11 Chen Z, Zhang X, Li M, et al. Simultaneously targeting epidermal growth factor receptor tyrosine kinase and cyclooxygenase-2, an efficient approach to inhibition of squamous cell carcinoma of the head and neck. Clin Cancer Res 2004; 10(17):5930–9. 12 Khuri FR, Lee JJ, Lippman SM, et al. Randomized phase III trial of low-dose isotretinoin for prevention of second primary tumors in stage I and II head and neck cancer patients. J Natl Cancer Inst 2006; 98(7):441–50. 13 Roostaeian J, Suh JD, Sercarz JA, Abemayor E, Lee JT, Blackwell KE. Factors affecting cancer recurrence after microvascular flap reconstruction of the head and neck. Laryngoscope 2005; 115(8):1391–4. 14 Vos JD, Burkey BB. Functional outcomes after free flap reconstruction of the upper aerodigestive tract. Curr Opin Otolaryngol Head Neck Surg 2004; 12(4):305–10. 15 Lin DT, Coppit GL, Burkey BB. Use of the anterolateral thigh flap for reconstruction of the head and neck. Curr Opin Otolaryngol Head Neck Surg 2004; 12(4):300–4. 16 Peled M, El-Naaj IA, Lipin Y, Ardekian L. The use of free fibular flap for functional mandibular reconstruction. J Oral Maxillofac Surg 2005; 63(2):220–4. 17 Nussenbaum B, Teknos TN, Chepeha DB. Tissue engineering: the current status of this futuristic modality in head neck reconstruction. Curr Opin Otolaryngol Head Neck Surg 2004; 12(4):311–15. 18 Morris PJ, Bradley JA, Doyal L, et al. Facial transplantation: a working party report from the Royal College of Surgeons of England. Transplantation 2004; 77(3):330–8. 19 Wigmore SJ. Face transplantation: the view from Birmingham, England. South Med J 2006; 99(4):424–6. 20 Dunst J, Stadler P, Becker A, et al. Tumor volume and tumor hypoxia in head and neck cancers. The amount of the hypoxic volume is important. Strahlenther Onkol 2003; 179(8):521–6. 21 Dietz A, Vanselow B, Rudat V, Conradt C, Weidauer H, Kallinowski F, et al. Prognostic impact of reoxygenation in advanced cancer of the head and neck during the initial course of chemoradiation or radiotherapy alone. Head Neck. 2003 Jan;25(1):50–8. 22 Overgaard J, Hansen HS, Overgaard M, et al. A randomized double-blind phase III study of nimorazole as a hypoxic radiosensitizer of primary radiotherapy in supraglottic larynx
References 401
23
24 25
26
27
28
29
30
31 32
33
34
35
36
and pharynx carcinoma. Results of the Danish Head and Neck Cancer Study (DAHANCA) Protocol 5-85. Radiother Oncol 1998; 46(2):135–46. Hoogsteen IJ, Pop LAM, Marres HAM, et al. Oxygen-modifying treatment with ARCON reduces the prognostic significance of hemoglobin in squamous cell carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 2006; 64(1):83–9. Varlotto J, Stevenson MA. Anemia, tumor hypoxemia, and the cancer patient. Int J Radiat Oncol Biol Phys 2005; 63(1):25–36. Henke M, Laszig R, Rube C, et al. Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial. Lancet 2003; 362(9392):1255–60. Arcasoy MO, Amin K, Chou SC, Haroon ZA, Varia M, Raleigh JA. Erythropoietin and erythropoietin receptor expression in head and neck cancer: relationship to tumor hypoxia. Clin Cancer Res 2005; 11(1):20–7. Nordsmark M, Bentzen SM, Rudat V, et al. Prognostic value of tumor oxygenation in 397 head and neck tumors after primary radiation therapy. An international multi-center study. Radiother Oncol 2005; 77:18–24. Budach W, Hehr T, Budach V, Belka C, Dietz K. A metaanalysis of hyperfractionated and accelerated radiotherapy and combined chemotherapy and radiotherapy regimens in unresected locally advanced squamous cell carcinoma of the head and neck. BMC Cancer 2006; 6(1):28. Haffty BG, Wilson LD, Son YH, et al. Concurrent chemoradiotherapy with mitomycin C compared with porfiromycin in squamous cell cancer of the head and neck: final results of a randomized clinical trial. Int J Radiat Oncol Biol Phys 2005; 61(1):119–28. Withers HR, Taylor JM, Maciejewski B. The hazard of accelerated tumor clonogen repopulation during radiotherapy. Acta Oncol 1988; 27(2):131–46. Fowler JF. Rapid repopulation in radiotherapy: a debate on mechanism. Radiother Oncol 1992; 24:125. Marcu L, van Doorn T, Olver I. Modelling of post-irradiation accelerated repopulation in squamous cell carcinomas. Phys Med Biol 2004; 49(16):3767–79. Skladowski K, Law MG, Maciejewski B, Steel GG. Planned and unplanned gaps in radiotherapy: the importance of gap position and gap duration. Radiother Oncol 1994; 30(2):109–20. Groome PA, O’Sullivan B, Mackillop WJ, et al. Compromised local control due to treatment interruptions and late treatment breaks in early glottic cancer: population-based outcomes study supporting need for intensified treatment schedules. Int J Radiat Oncol Biol Phys 2006; 64(4):1002–12. Robertson C, Robertson AG, Hendry JH, et al. Similar decreases in local tumor control are calculated for treatment protraction and for interruptions in the radiotherapy of carcinoma of the larynx in four centers. Int J Radiat Oncol Biol Phys 1998; 40(2):319–29. Dale RG, Hendry JH, Jones B, Robertson AG, Deehan C, Sinclair JA. Practical methods for compensating for missed treatment days in radiotherapy, with particular reference to head and neck schedules. Clin Oncol 2002; 14(5):382–93.
37 Pignon JP, Baujat B, Bourhis J. Apport des meta-analyses sur donnees individuelles au traitement des cancers ORL. Cancer/Radiotherapie 2005; 9(1):31–6. 38 Radiotherapy Activity Planning Group. Radiotherapy Activity Planning for Scotland: 2011–2015. Edinburgh: Scottish Executive Health Department, 2005. 39 Williams MV, James ND, Summers ET, Barrett A, Ash DV. National Survey of Radiotherapy Fractionation Practice in 2003. Clin Oncol 2006; 18(1):3–14. 40 Gowda RV, Henk JM, Mais KL, Sykes AJ, Swindell R, Slevin NJ. Three weeks radiotherapy for T1 glottic cancer: the Christie and Royal Marsden Hospital experience. Radiother Oncol 2003; 68(2):105–11. 41 Wiernik G, Alcock CJ, Bates TD, et al. Final report on the second British Institute of Radiology fractionation study: short versus long overall treatment times for radiotherapy of carcinoma of the laryngo-pharynx. Br J Radiol 1991; 64(759):232–41. 42 Bjarnason GA, Mackenzie R, Hodson I, et al. A randomized prospective phase-III study comparing the acute oral mucositis of morning vs. afternoon radiotherapy (RT) in patients (pts) with squamous cell carcinoma of the head and neck (SCCHN): NCIC-CTG HN.3. J Clin Oncol (Meeting Abstracts) 2005; 23(16 Suppl.):LBA5500 43 Eriksen JG, Steiniche T, Overgaard J, on behalf of the Danish Head and Neck Cancer Study Group. The influence of epidermal growth factor receptor and tumor differentiation on the response to accelerated radiotherapy of squamous cell carcinomas of the head and neck in the randomized DAHANCA 6 and 7 Study. Radiother Oncol 2005; 74(2):93–100. 44 Buffa FM, Bentzen SM, Daley FM, et al. Molecular marker profiles predict locoregional control of head and neck squamous cell carcinoma in a randomized trial of continuous hyperfractionated accelerated radiotherapy. Clin Cancer Res 2004; 10(11):3745–54. 45 Withers HR. The 4 R’s of radiotherapy. Adv Radiat Biol 1975; 5:241–71. 46 Begg AC, Haustermans K, Hart AA, et al. The value of pretreatment cell kinetic parameters as predictors for radiotherapy outcome in head and neck cancer: a multicenter analysis. Radiother Oncol 1999; 50(1):13–23. 47 Maciejewski B, Skladowski K, Pilecki B, et al. Randomized clinical trial on accelerated 7 days per week fractionation in radiotherapy for head and neck cancer. Preliminary report on acute toxicity. Radiother Oncol 1996; 40(2):137–45. 48 Sasse AD, Clark LG, Sasse EC, Clark OA. Amifostine reduces side effects and improves complete response rate during radiotherapy: results of a meta-analysis. Int J Radiat Oncol Biol Phys 2006; 64(3):784–91. 49 Ertekin MV, Koc M, Karslioglu I, Sezen O. Zinc sulfate in the prevention of radiation-induced oropharyngeal mucositis: a prospective, placebo-controlled, randomized study. Int J Radiat Oncol Biol Phys 2004; 58(1):167–74. 50 Spielberger R, Stiff P, Bensinger W, et al. Palifermin for oral mucositis after intensive therapy for hematologic cancers. N Engl J Med 2004; 351(25):2590–8.
402 Head and neck cancer
51 Duncan GG, Epstein JB, Tu D, et al. Quality of life, mucositis, and xerostomia from radiotherapy for head and neck cancers: a report from the NCIC CTG HN2 randomized trial of an antimicrobial lozenge to prevent mucositis. Head Neck 2005; 27(5):421–8. 52 Su CK, Mehta V, Ravikumar L, et al. Phase II double-blind randomized study comparing oral aloe vera versus placebo to prevent radiation-related mucositis in patients with headand-neck neoplasms. Int J Radiat Oncol Biol Phys 2004; 60(1):171–7. 53 Trotti A, Garden A, Warde P, et al. A multinational, randomized phase III trial of iseganan HCl oral solution for reducing the severity of oral mucositis in patients receiving radiotherapy for head-and-neck malignancy. Int J Radiat Oncol Biol Phys 2004; 58(3):674–81. 54 Genot MT, Klastersky J. Low-level laser for prevention and therapy of oral mucositis induced by chemotherapy or radiotherapy. Curr Opin Oncol 2005; 17(3):236–40. 55 Chambers M, Welsh D, Scrimger R, et al. RK-0202 for radiation-induced oral mucositis. J Clin Oncol 2006; 24(18s):5523. 56 Alterio D, Jereczek-Fossa BA, Zuccotti GF, et al. Tetracaine oral gel in patients treated with radiotherapy for head-andneck cancer: final results of a phase II study. Int J Radiat Oncol Biol Phys 2006; 64(2):392–5. 57 Beer KT, Krause KB, Zuercher T, Stanga Z. Early percutaneous endoscopic gastrostomy insertion maintains nutritional state in patients with aerodigestive tract cancer. Nutr Cancer 2005; 52(1):29–34. 58 Rabinovitch R, Grant B, Berkey BA, et al. Impact of nutrition support on treatment outcome in patients with locally advanced head and neck squamous cell cancer treated with definitive radiotherapy: a secondary analysis of RTOG trial 90-03. Head Neck 2006; 28(4):287–96. 59 Jin D, Phillips M, Byles JE. Effects of parenteral nutrition support and chemotherapy on the phasic composition of tumor cells in gastrointestinal cancer. J Parenter Enteral Nutr 1999; 23(4):237–41. 60 McGeer AJ, Detsky AS, O’Rourke K. Parenteral nutrition in cancer patients undergoing chemotherapy: a meta-analysis. Nutrition 1990; 6(3):233–40. 61 Nyarady Z, Nemeth A, Ban A, et al. A randomized study to assess the effectiveness of orally administered pilocarpine during and after radiotherapy of head and neck cancer. Anticancer Res 2006; 26(2B):1557–62. 62 Ringash J, Warde P, Lockwood G, O’Sullivan B, Waldron J, Cummings B. Postradiotherapy quality of life for head-andneck cancer patients is independent of xerostomia. Int J Radiat Oncol Biol Phys 2005; 61(5):1403–7. 63 Scarantino C, LeVeque F, Swann RS, et al. Effect of pilocarpine during radiation therapy: results of RTOG 97-09, a phase III randomized study in head and neck cancer patients. J Support Oncol 2006; 4(5):252–8. 64 Warde P, O’Sullivan B, Aslanidis J, et al. A Phase III placebocontrolled trial of oral pilocarpine in patients undergoing radiotherapy for head-and-neck cancer. Int J Radiat Oncol Biol Phys 2002; 54(1):9–13.
65 Brewin TB. Can a tumour cause the same appetite perversion or taste change as a pregnancy? Lancet 1980; 2(8200):907–8. 66 Brewin TB. Appetite perversions and taste changes triggered or abolished by radiotherapy. Clin Radiol 1982; 33(4):471–5. 67 Grau C, Overgaard J. Postirradiation sensorineural hearing loss: a common but ignored late radiation complication. Int J Radiat Oncol Biol Phys 1996; 36(2):515–17. 68 Honore HB, Bentzen SM, Moller K, Grau C. Sensori-neural hearing loss after radiotherapy for nasopharyngeal carcinoma: individualized risk estimation. Radiother Oncol 2002; 65(1):9–16. 69 Barrett A, Dobbs J, Ash DV. Practical Radiotherapy Planning, 3rd edn. Oxford: Oxford University Press, 1999. 70 Yap BK, Slevin NJ, Ahamad A. Barium paste: useful for primary tumour localization in oral cancer. Br J Radiol 2004; 77(914):143–5. 71 Krempien RC, Grehn C, Haag C, et al. Feasibility report for retreatment of locally recurrent head-and-neck cancers by combined brachy-chemotherapy using frameless imageguided 3D interstitial brachytherapy. Brachytherapy 2005; 4(2):154–62. 72 Fiorino C, Dell’Oca I, Pierelli A, et al. Significant improvement in normal tissue sparing and target coverage for head and neck cancer by means of helical tomotherapy. Radiother Oncol 2006; 78(3):276–82. 73 Mendenhall WM, Amdur RJ, Palta JR. Intensity-modulated radiotherapy in the standard management of head and neck cancer: promises and pitfalls. J Clin Oncol 2006; 24(17):2618–23. 74 Wang D, Schultz CJ, Jursinic PA, et al. Initial experience of FDG-PET/CT guided IMRT of head-and-neck carcinoma. Int J Radiat Oncol Biol Phys 2006; 65(1):143–51. 75 Wolden SL, Chen WC, Pfister DG, Kraus DH, Berry SL, Zelefsky MJ. Intensity-modulated radiation therapy (IMRT) for nasopharynx cancer: Update of the Memorial Sloan-Kettering experience. Int J Radiat Oncol Biol Phys 2006; 64(1):57–62. 76 McLean M, Chow E, O’Sullivan B, et al. Re-irradiation for locally recurrent nasopharyngeal carcinoma. Radiother Oncol 1998; 48(2):209–11. 77 Nag S, Cano ER, Demanes DJ, Puthawala AA, Vikram B. The American Brachytherapy Society recommendations for highdose-rate brachytherapy for head-and-neck carcinoma. Int J Radiat Oncol Biol Phys 2001; 50(5):1190–8. 78 De Crevoisier R, Bourhis J, Domenge C, et al. Full-dose reirradiation for unresectable head and neck carcinoma: experience at the Gustave-Roussy Institute in a series of 169 patients. J Clin Oncol 1998; 16(11):3556–62. 79 De Crevoisier R, Domenge C, Wibault P, et al. Full dose reirradiation combined with chemotherapy after salvage surgery in head and neck carcinoma. Cancer 2001; 91(11):2071–6. 80 Stewart FA. Re-treatment after full-course radiotherapy: is it a viable option? Acta Oncol 1999; 38(7):855–62. 81 Wong SJ, Machtay M, Li Y. Locally recurrent, previously irradiated head and neck cancer: concurrent re-irradiation
References 403
82
83
84
85
86
87
88
89
90
91
92
and chemotherapy, or chemotherapy alone? J Clin Oncol 2006; 24(17):2653–8. Calais G, Pointreau Y, Alfonsi M, et al. Randomized phase III trial comparing induction chemotherapy using cisplatin (P) fluorouracil (F) with or without docetaxel (T) for organ preservation in hypopharynx and larynx cancer. Preliminary results of GORTEC 2000-01. J Clin Oncol 2006; 24(18s):5506. Hitt R, Grau J, Lopez-Pousa A, et al. Randomized phase II/III clinical trial of induction chemotherapy (ICT) with either cisplatin/5-fluorouracil (PF) or docetaxel/cisplatin/ 5-fluorouracil (TPF) followed by chemoradiotherapy (CRT) vs. crt alone for patients (pts) with unresectable locally advanced head and neck cancer (LAHNC). J Clin Oncol 2006; 24(18s):5515. Remenar E, Van Herpen C, Germa Lluch J, et al. A randomized phase III multicenter trial of neoadjuvant docetaxel plus cisplatin and 5-fluorouracil (TPF) versus neoadjuvant PF in patients with locally advanced unresectable squamous cell carcinoma of the head and neck (SCCHN). Final analysis of EORTC 24971. J Clin Oncol 2006; 24(No. 18S, June 20 Suppl.):5516. Vermorken JB, Remenar E, van Herpen C, et al. Standard cisplatin/infusional 5-fluorouracil (PF) vs docetaxel (T) plus PF (TPF) as neoadjuvant chemotherapy for nonresectable locally advanced squamous cell carcinoma of the head and neck (LASCCHN): a phase III trial of the EORTC Head and Neck Cancer Group (EORTC 24971). J Clin Oncol (Meeting Abstracts) 2004; 22(14 Suppl.):5508 Bernier J, Cooper JS, Pajak TF, et al. Defining risk levels in locally advanced head and neck cancers: a comparative analysis of concurrent postoperative radiation plus chemotherapy trials of the EORTC (22931) and RTOG (9501). Head Neck 2005; 27(10):843–50. Bernier J, Domenge C, Ozsahin M, et al. Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. N Engl J Med 2004; 350(19):1945–52. Cooper JS, Pajak TF, Forastiere AA, et al. Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med 2004; 350(19):1937–44. Forastiere AA, Goepfert H, Maor M, et al. Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 2003; 349(22):2091–8. Lefebvre JL, Chevalier D, Luboinski B, Kirkpatrick A, Collette L, Sahmoud T. Larynx preservation in pyriform sinus cancer: preliminary results of a European Organization for Research and Treatment of Cancer phase III trial. EORTC Head and Neck Cancer Cooperative Group. J Natl Cancer Inst 1996; 88(13):890–9. The Department of Veterans Affairs Laryngeal Cancer Study Group. Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer. N Engl J Med 1991; 324(24):1685–90. Urba S, Wolf G, Eisbruch A, et al. Single-cycle induction chemotherapy selects patients with advanced laryngeal
93 94
95
96
97
98
99
100
101
102
103
104
105
106
107
cancer for combined chemoradiation: a new treatment paradigm. J Clin Oncol 2006; 24(4):593–8. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100(1):57–70. Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006; 354(6):567–78. Eriksen JG, Buffa FM, Alsner J, Steiniche T, Bentzen SM, Overgaard J. Molecular profiles as predictive marker for the effect of overall treatment time of radiotherapy in supraglottic larynx squamous cell carcinomas. Radiother Oncol 2004; 72(3):275–82. Wilson GD, Saunders MI, Dische S, et al. Pre-treatment proliferation and the outcome of conventional and accelerated radiotherapy. Eur J Cancer 2006; 42(3):363–71. Gupta NK, Swindell R. Concomitant methotrexate and radiotherapy in advanced head and neck cancer: 15-year follow-up of a randomized clinical trial. Clin Oncol (R Coll Radiol) 2001; 13(5):339–44. Munro AJ. Anticipating the future role of radiotherapy and chemotherapy in the treatment of head and neck cancer: a lesson from Manchester. Clin Oncol (R Coll Radiol) 2001; 13(5):336–8. Desai N, Trieu V, Yao Z, et al. Increased antitumor activity, intratumor paclitaxel concentrations, and endothelial cell transport of cremophor-free, albumin-bound paclitaxel, ABI007, compared with cremophor-based paclitaxel. Clin Cancer Res 2006; 12(4):1317–24. List MA, Stracks J, Colangelo L, et al. How do head and neck cancer patients prioritize treatment outcomes before initiating treatment? J Clin Oncol 2000; 18(4):877–84. Sharp HM, List M, MacCracken E, Stenson K, Stocking C, Siegler M. Patients’ priorities among treatment effects in head and neck cancer: evaluation of a new assessment tool. Head Neck 1999; 21(6):538–46. McNeil BJ, Weichselbaum R, Pauker SG. Speech and survival: tradeoffs between quality and quantity of life in laryngeal cancer. N Engl J Med 1981; 305(17):982–7. Chaplin JM, Morton RP. A prospective, longitudinal study of pain in head and neck cancer patients. Head Neck 1999; 21(6):531–7. Allison RR, Cuenca RE, Downie GH, Camnitz P, Brodish B, Sibata CH. Clinical photodynamic therapy of head and neck cancers – a review of applications and outcomes. Photodiag Photodyn Ther 2005; 2(3):205–22. Hopper C, Niziol C, Sidhu M. The cost-effectiveness of Foscan mediated photodynamic therapy (Foscan-PDT) compared with extensive palliative surgery and palliative chemotherapy for patients with advanced head and neck cancer in the UK. Oral Oncol 2004; 40(4):372–82. Lou PJ, Jager HR, Jones L, Theodossy T, Bown SG, Hopper C. Interstitial photodynamic therapy as salvage treatment for recurrent head and neck cancer. Br J Cancer 2004; 91(3):441–6. Lou PJ, Jones L, Hopper C. Clinical outcomes of photodynamic therapy for head-and-neck cancer. Technol Cancer Res Treat 2003; 2(4):311–17.
404 Head and neck cancer
108 Dunlop PR, Dickinson RJ, Hand JW, Munro AJ, Vallis KA. Early experience with combined interstitial hyperthermia and brachytherapy. Br J Radiol 1986; 59(701):525–7. 109 Bouchet BP, de Fromentel CC, Puisieux A, Galmarini CM. p53 as a target for anti-cancer drug development. Crit Rev Oncol/Hematol 2006; 58:190–207 110 Brennan JA, Mao L, Hruban RH, et al. Molecular assessment of histopathological staging in squamous-cell carcinoma of the head and neck. N Engl J Med 1995; 332(7): 429–35. 111 Nathan CA, Amirghahri N, Rice C, Abreo FW, Shi R, Stucker FJ. Molecular analysis of surgical margins in head and neck squamous cell carcinoma patients. Laryngoscope 2002; 112(12):2129–40. 112 American Joint Committee on Cancer. Cancer Staging Manual. New York: Springer, 2002. 113 Sobin LH, Wittekind C. TNM Classification of Malignant Tumours, 6th edn. Hoboken, NJ: John Wiley, 2002. 114 Lydiatt WM, Shah JP, Hoffman HT. AJCC stage groupings for head and neck cancer: should we look at alternatives? A report of the Head and Neck Sites Task Force. Head Neck 2001; 23(8):607–12. 115 Groome PA, Schulze KM, Mackillop WJ, et al. A comparison of published head and neck stage groupings in carcinomas of the tonsillar region. Cancer 2001; 92(6):1484–94. 116 Feinstein AR, Sosin DM, Wells CK. The Will Rogers phenomenon. Stage migration and new diagnostic techniques as a source of misleading statistics for survival in cancer. N Engl J Med 1985; 312(25):1604–8. 117 Daly T, Poulsen MG, Denham JW, et al. The effect of anaemia on efficacy and normal tissue toxicity following radiotherapy for locally advanced squamous cell carcinoma of the head and neck. Radiother Oncol 2003; 68(2):113–22. 118 Hu K, Harrison LB. Impact of anemia in patients with head and neck cancer treated with radiation therapy. Curr Treat Options Oncol 2005; 6(1):31–45. 119 Johansen LV, Grau C, Overgaard J. Laryngeal carcinoma – multivariate analysis of prognostic factors in 1252 consecutive patients treated with primary radiotherapy. Acta Oncol 2003; 42(7):771–8. 120 Browman G, Wong G, Hodson I, et al. Influence of cigarette smoking on the efficacy of radiation therapy in head and neck cancer. N Engl J Med 1993; 328:159–63. 121 Dietl B, Marienhagen J, Kuhnel T, Schaefer C, Kolbl O. FDGPET in radiotherapy treatment planning of advanced head and neck cancer – a prospective clinical analysis. Auris Nasus Larynx 2006; 33:303–9. 122 Gor DM, Langer JE, Loevner LA. Imaging of cervical lymph nodes in head and neck cancer: the basics. Radiol Clin North Am 2006; 44(1):101–10, viii. 123 Hafidh MA, Lacy PD, Hughes JP, Duffy G, Timon CV. Evaluation of the impact of addition of PET to CT and MR scanning in the staging of patients with head and neck carcinomas. Eur Arch Otorhinolaryngol 2006; 263(9):853–9. 124 Rumboldt Z, Gordon L, Gordon L, Bonsall R, Ackermann S. Imaging in head and neck cancer. Curr Treat Options Oncol 2006; 7(1):23–34.
125 Schoder H, Carlson DL, Kraus DH, et al. 18F-FDG PET/CT for detecting nodal metastases in patients with oral cancer staged N0 by clinical examination and CT/MRI. J Nucl Med 2006; 47(5):755–62. 126 Vanderstraeten B, Duthoy W, Gersem WD, Neve WD, Thierens H. [18F]fluoro-deoxy-glucose positron emission tomography ([18F]FDG-PET) voxel intensity-based intensity-modulated radiation therapy (IMRT) for head and neck cancer. Radiother Oncol 2006; 79(3):249–58. 127 Veit P, Ruehm S, Kuehl H, et al. Lymph node staging with dual-modality PET/CT: enhancing the diagnostic accuracy in oncology. Eur J Radiol 2006; 58(3):383–9. 128 Coster JR, Foote RL, Olsen KD, Jack SM, Schaid DJ, DeSanto LW. Cervical nodal metastasis of squamous cell carcinoma of unknown origin: indications for withholding radiation therapy. Int J Radiat Oncol Biol Phys 1992; 23(4):743–9. 129 Glynne-Jones RG, Anand AK, Young TE, Berry RJ. Metastatic carcinoma in the cervical lymph nodes from an occult primary: a conservative approach to the role of radiotherapy. Int J Radiat Oncol Biol Phys 1990; 18(2):289–94. 130 Pascal B. Pensees. London: Dent, 1670. 131 Siegler M. Pascal’s wager and the hanging of crepe. N Engl J Med 1975; 293(17):853–7. 132 Bataini JP, Bernier J, Asselain B, et al. Primary radiotherapy of squamous cell carcinoma of the oropharynx and pharyngolarynx: tentative multivariate modelling system to predict the radiocurability of neck nodes. Int J Radiat Oncol Biol Phys 1988; 14(4):635–42. 133 Jyothirmayi R, Sankaranarayanan R, Varghese C, Jacob R, Nair MK. Radiotherapy in the treatment of verrucous carcinoma of the oral cavity. Oral Oncol 1997; 33(2):124–8. 134 Pindborg JJ, Mehta FS, Gupta PC, Daftary DK, Smith CJ. Reverse smoking in Andhra Pradesh, India: a study of palatal lesions among 10,169 villagers. Br J Cancer 1971; 25(1):10–20. 135 Puthawala AA, Syed AM, Eads DL, Gillin L, Gates TC. Limited external beam and interstitial 192iridium irradiation in the treatment of carcinoma of the base of the tongue: a ten year experience. Int J Radiat Oncol Biol Phys 1988; 14(5):839–48. 136 O’Sullivan B, Warde P, Grice B, et al. The benefits and pitfalls of ipsilateral radiotherapy in carcinoma of the tonsillar region. Int J Radiat Oncol Biol Phys 2001; 51(2):332–43. 137 Hoffstetter S, Marchal C, Peiffert D, et al. Treatment duration as a prognostic factor for local control and survival in epidermoid carcinomas of the tonsillar region treated by combined external beam irradiation and brachytherapy. Radiother Oncol 1997; 45(2):141–8. 138 Withers HR, Peters LJ, Taylor JM, et al. Local control of carcinoma of the tonsil by radiation therapy: an analysis of patterns of fractionation in nine institutions. Int J Radiat Oncol Biol Phys 1995; 33(3):549–62. 139 Bova R, Goh R, Poulson M, Coman WB. Total pharyngolaryngectomy for squamous cell carcinoma of the hypopharynx: a review. Laryngoscope 2005; 115(5):864–9. 140 Bensadoun R-J, Benezery K, Dassonville O, et al. French multicenter phase III randomized study testing concurrent twice-a-day radiotherapy and cisplatin/5-fluorouracil
References 405
141
142
143
144
145
146
147
148
149
150
151
152
153 154
chemotherapy (BiRCF) in unresectable pharyngeal carcinoma: results at 2 years (FNCLCC-GORTEC). Int J Radiat Oncol Biol Phys 2006; 64(4):983–94. Eisbruch A, Schwartz M, Rasch C, et al. Dysphagia and aspiration after chemoradiotherapy for head-and-neck cancer: which anatomic structures are affected and can they be spared by IMRT? Int J Radiat Oncol Biol Phys 2004; 60(5):1425–39. Kadish SP. Can I treat this small larynx lesion with radiation alone? Update on the radiation management of early (T1 and T2) glottic cancer. Otolaryngol Clin North Am 2005; 38(1):1–9, vii. Thomas JV, Olsen KD, Neel HB 3rd, DeSanto LW, Suman VJ. Early glottic carcinoma treated with open laryngeal procedures. Arch Otolaryngol Head Neck Surg 1994; 120(3):264–8. Groome PA, O’Sullivan B, Irish JC, et al. Glottic cancer in Ontario, Canada and the SEER areas of the United States. Do different management philosophies produce different outcome profiles? J Clin Epidemiol 2001; 54(3):301–15. MacDougall RH, Orr JA, Kerr GR, Duncan W. Fast neutron treatment for squamous cell carcinoma of the head and neck: final report of Edinburgh randomised trial. BMJ 1990; 301(6763):1241–2. Iro H, Waldfahrer F, Altendorf-Hofmann A, Weidenbecher M, Sauer R, Steiner W. Transoral laser surgery of supraglottic cancer: follow-up of 141 patients. Arch Otolaryngol Head Neck Surg 1998; 124(11):1245–50. Chevalier D, Laccourreye O, Brasnu D, Laccourreye H, Piquet JJ. Cricohyoidoepiglottopexy for glottic carcinoma with fixation or impaired motion of the true vocal cord: 5-year oncologic results with 112 patients. Ann Otol Rhinol Laryngol 1997; 106(5):364–9. O’Sullivan B, Mackillop W, Gilbert R, et al. Controversies in the management of laryngeal cancer: results of an international survey of patterns of care. Radiother Oncol 1994; 31(1):23–32. Chatani M, Matayoski Y, Masaki N, Teshima T, Inoue T. Radiation therapy for early glottic carcinoma (T1N0M0). The final results of prospective randomized study concerning radiation field. Strahlenther Onkol 1996; 172(3):169–72. Crellin RP, Gaze MN, White A, Maran AG, MacDougall RH. Salvage laryngectomy after radical radiotherapy for laryngeal carcinoma. Clin Otolaryngol Allied Sci 1992; 17(5):449–51. Wong HL, Wang X, Chang RC, et al. Stable expression of EBERs in immortalized nasopharyngeal epithelial cells confers resistance to apoptotic stress. Mol Carcinog 2005; 44(2):92–101. Sheu LF, Lee WC, Lee HS, Kao WY, Chen A. Co-expression of c-kit and stem cell factor in primary and metastatic nasopharyngeal carcinomas and nasopharyngeal epithelium. J Pathol 2005; 207(2):216–23. Yu MC. Diet and nasopharyngeal carcinoma. Prog Clin Biol Res 1990; 346:93–105. Yang L, Parkin DM, Li L, Chen Y. Time trends in cancer mortality in China: 1987–1999. Int J Cancer 2003; 106(5):771–83.
155 Liu F-Y, Chang JT, Wang H-M, et al. [18F]Fluorodeoxyglucose positron emission tomography is more sensitive than skeletal scintigraphy for detecting bone metastasis in endemic nasopharyngeal carcinoma at initial staging. J Clin Oncol 2006; 24(4):599–604. 156 Altun M, Fandi A, Dupuis O, Cvitkovic E, Krajina Z, Eschwege F. Undifferentiated nasopharyngeal cancer (UCNT): current diagnostic and therapeutic aspects. Int J Radiat Oncol Biol Phys 1995; 32(3):859–77. 157 Khanfir A, Frikha M, Ghorbel A, Karray H, Drira MM, Daoud J. Les cancers metastatiques du nasopharynx: etude clinique et resultats therapeutiques de 95 cas. Cancer/Radiotherapie 2006; 10:545–9. 158 Al-Sarraf M, LeBlanc M, Giri PG, et al. Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: phase III randomized Intergroup Study 0099. J Clin Oncol 1998; 16(4):1310–17. 159 Chan ATC, Leung SF, Ngan RKC, et al. Overall survival after concurrent cisplatin-radiotherapy compared with radiotherapy alone in locoregionally advanced nasopharyngeal carcinoma. J Natl Cancer Inst 2005; 97(7):536–9. 160 Kwong DLW, Sham JST, Au GKH, et al. Concurrent and adjuvant chemotherapy for nasopharyngeal carcinoma: a factorial study. J Clin Oncol 2004; 22(13):2643–53. 161 Lee AWM, Lau WH, Tung SY, et al. Preliminary results of a randomized study on therapeutic gain by concurrent chemotherapy for regionally advanced nasopharyngeal carcinoma: NPC-9901 Trial by the Hong Kong Nasopharyngeal Cancer Study Group. J Clin Oncol 2005; 23(28):6966–75. 162 Lin J-C, Jan J-S, Hsu C-Y, Liang W-M, Jiang R-S, Wang W-Y. Phase III study of concurrent chemoradiotherapy versus radiotherapy alone for advanced nasopharyngeal carcinoma: positive effect on overall and progression-free survival. J Clin Oncol 2003; 21(4):631–7. 163 Wee J, Tan EH, Tai BC, et al. Randomized trial of radiotherapy versus concurrent chemoradiotherapy followed by adjuvant chemotherapy in patients with American Joint Committee on Cancer/International Union Against Cancer stage III and IV nasopharyngeal cancer of the endemic variety. J Clin Oncol 2005; 23(27):6730–8. 164 Zhang L, Zhao C, Peng P-J, et al. Phase III Study Comparing Standard Radiotherapy With or Without Weekly Oxaliplatin in Treatment of Locoregionally Advanced Nasopharyngeal Carcinoma: Preliminary Results. J Clin Oncol 2005 November 20, 2005;23(33):8461–8. 165 Langendijk JA, Leemans CR, Buter J, Berkhof J, Slotman BJ. The additional value of chemotherapy to radiotherapy in locally advanced nasopharyngeal carcinoma: a meta-analysis of the published literature. J Clin Oncol 2004; 22(22):4604–12. 166 Lee AW, Law SC, Ng SH, et al. Retrospective analysis of nasopharyngeal carcinoma treated during 1976–1985: late complications following megavoltage irradiation. Br J Radiol 1992; 65(778):918–28. 167 Low WK, Toh ST, Wee J, Fook-Chong SMC, Wang DY. Sensorineural hearing loss after radiotherapy and
406 Head and neck cancer
168
169
170
171
172
173
174
175
176
177
178 179
180
181
182
chemoradiotherapy: a single, blinded, randomized study. J Clin Oncol 2006; 24(12):1904–9. Wolden SL, Zelefsky MJ, Hunt MA, et al. Failure of a 3D conformal boost to improve radiotherapy for nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2001; 49(5):1229–34. Lee N, Xia P, Quivey JM, et al. Intensity-modulated radiotherapy in the treatment of nasopharyngeal carcinoma: an update of the UCSF experience. Int J Radiat Oncol Biol Phys 2002; 53(1):12–22. McMillan AS, Pow EH, Kwong DL, et al. Preservation of quality of life after intensity-modulated radiotherapy for early-stage nasopharyngeal carcinoma: results of a prospective longitudinal study. Head Neck 2006; 28:712–22. Le Q-T, Jones CD, Yau T-K, et al. A comparison study of different PCR assays in measuring circulating plasma Epstein–Barr virus DNA levels in patients with nasopharyngeal carcinoma. Clin Cancer Res 2005; 11(16):5700–7. Lin J-C, Wang W-Y, Chen KY, et al. Quantification of plasma Epstein–Barr virus DNA in patients with advanced nasopharyngeal carcinoma. N Engl J Med 2004; 350(24):2461–70. Mazeron JJ, Chassagne D, Crook J, et al. Radiation therapy of carcinomas of the skin of nose and nasal vestibule: a report of 1676 cases by the Groupe Europeen de Curietherapie. Radiother Oncol 1988; 13(3):165–73. Stell PM, McCormick MS. Carcinoma of the external auditory meatus and middle ear. Prognostic factors and a suggested staging system. J Laryngol Otol 1985; 99(9):847–50. Bayley JP, van Minderhout I, Weiss MM, et al. Mutation analysis of SDHB and SDHC: novel germline mutations in sporadic head and neck paraganglioma and familial paraganglioma and/or pheochromocytoma. BMC Med Genet 2006; 7:1. van den Berg R, Schepers A, de Bruine FT, et al. The value of MR angiography techniques in the detection of head and neck paragangliomas. Eur J Radiol 2004; 52(3):240–5. Duet M, Guichard JP, Rizzo N, Boudiaf M, Herman P, Tran Ba Huy P. Are somatostatin analogs therapeutic alternatives in the management of head and neck paragangliomas? Laryngoscope 2005; 115(8):1381–4. Dulguerov P, Allal AS, Calcaterra TC. Esthesioneuroblastoma: a meta-analysis and review. Lancet Oncol 2001; 2(11):683–90. Loy AH, Reibel JF, Read PW, et al. Esthesioneuroblastoma: continued follow-up of a single institution’s experience. Arch Otolaryngol Head Neck Surg 2006; 132(2):134–8. Marshall AH, Bradley PJ. Management dilemmas in the treatment and follow-up of advanced juvenile nasopharyngeal angiofibroma. ORL J Otorhinolaryngol Relat Spec 2006; 68(5):211–16. McAfee WJ, Morris CG, Amdur RJ, Werning JW, Mendenhall WM. Definitive radiotherapy for juvenile nasopharyngeal angiofibroma. Am J Clin Oncol 2006; 29(2):168–70. Pryor SG, Moore EJ, Kasperbauer JL. Endoscopic versus traditional approaches for excision of juvenile nasopharyngeal angiofibroma. Laryngoscope 2005; 115(7):1201–7.
183 Hughes J, Stead L, Lancaster T. Antidepressants for smoking cessation. Cochrane Database Syst Rev 2004(4):CD000031 184 Hughes JR, Stead LF, Lancaster T. Anxiolytics for smoking cessation. Cochrane Database Syst Rev 2000(4):CD002849. 185 Silagy C, Lancaster T, Stead L, Mant D, Fowler G. Nicotine replacement therapy for smoking cessation. Cochrane Database Syst Rev 2004(3):CD000146. 186 Rice VH, Stead LF. Nursing interventions for smoking cessation. Cochrane Database Syst Rev 2004(1):CD001188. 187 Lancaster T, Stead L. Physician advice for smoking cessation. Cochrane Database Syst Rev 2004(4):CD000165. 188 Stead L, Lancaster T. Nicobrevin for smoking cessation. Cochrane Database Syst Rev 2006(2):CD005990. 189 Usher M. Exercise interventions for smoking cessation. Cochrane Database Syst Rev 2005(1):CD002295. 190 Hajek P, Stead LF. Aversive smoking for smoking cessation. Cochrane Database Syst Rev 2004(3):CD000546. 191 Stead LF, Lancaster T. Group behaviour therapy programmes for smoking cessation. Cochrane Database Syst Rev 2005(2):CD001007. 192 Lancaster T, Stead LF. Individual behavioural counselling for smoking cessation. Cochrane Database Syst Rev 2005(2):CD001292. 193 Hey K, Perera R. Competitions and incentives for smoking cessation. Cochrane Database Syst Rev 2005(2):CD004307. 194 Hey K, Perera R. Quit and Win contests for smoking cessation. Cochrane Database Syst Rev 2005(2):CD004986 195 Stead LF, Lancaster T, Perera R. Telephone counselling for smoking cessation. Cochrane Database Syst Rev 2003(1):CD002850. 196 Lancaster T, Stead LF. Self-help interventions for smoking cessation. Cochrane Database Syst Rev 2005(3):CD001118. 197 White AR, Rampes H, Campbell JL. Acupuncture and related interventions for smoking cessation. Cochrane Database Syst Rev 2006(1):CD000009 198 Sanchiz F, Milla A, Torner J, et al. Single fraction per day versus two fractions per day versus radiochemotherapy in the treatment of head and neck cancer. Int J Radiat Oncol Biol Phys 1990; 19(6):1347–50. 199 Fu KK, Pajak TF, Trotti A, et al. A Radiation Therapy Oncology Group (RTOG) phase III randomized study to compare hyperfractionation and two variants of accelerated fractionation to standard fractionation radiotherapy for head and neck squamous cell carcinomas: first report of RTOG 9003. Int J Radiat Oncol Biol Phys 2000; 48(1):7–16. 200 Teo PM, Leung SF, Chan AT, et al. Final report of a randomized trial on altered-fractionated radiotherapy in nasopharyngeal carcinoma prematurely terminated by significant increase in neurologic complications. Int J Radiat Oncol Biol Phys 2000; 48(5):1311–22. 201 Horiot JC, Le Fur R, N'Guyen T, et al. Hyperfractionation versus conventional fractionation in oropharyngeal carcinoma: final analysis of a randomized trial of the EORTC cooperative group of radiotherapy. Radiother Oncol 1992; 25(4):231–41. 202 Cummings BJ. Benefits of accelerated hyperfractionation for head and neck cancer. Acta Oncol 1999; 38(2):131-6.
References 407
203 Datta NR, Choudhry AD, Gupta S, Bose AK. Twice a day versus once a day radiation therapy in head and neck cancer. Int J Radiat Oncol Biol Phys 1989; 17 (suppl 1) Abs 35:132–133. 204 Pinto LH, Canary PC, Araujo CM, Bacelar SC, Souhami L. Prospective randomized trial comparing hyperfractionated versus conventional radiotherapy in stages III and IV oropharyngeal carcinoma. Int J Radiat Oncol Biol Phys 1991; 21(3):557–62. 205 Awwad HK, Khafagy Y, Barsoum M, et al. Accelerated versus conventional fractionation in the postoperative irradiation of locally advanced head and neck cancer: influence of tumour proliferation. Radiother Oncol 1992; 25(4):261–6. 206 Awwad HK, Lotayef M, Shouman T, et al. Accelerated hyperfractionation (AHF) compared to conventional fractionation (CF) in the postoperative radiotherapy of locally advanced head and neck cancer: influence of proliferation. Br J Cancer 2002; 86(4):517-23. 207 Van den Bogaert W, van der Schueren E, Horiot J-C, et al. The EORTC randomized trial on three fractions per day and misonidazole (trial no. 22811) in advanced head and neck cancer: long-term results and side effects. Radiotherapy and Oncology 1995; 35(2):91–99. 208 Dische S, Saunders M, Barrett A, Harvey A, Gibson D, Parmar M. A randomised multicentre trial of CHART versus conventional radiotherapy in head and neck cancer. Radiother Oncol 1997; 44(2):123–36. 209 Horiot JC, Bontemps P, van den Bogaert W, et al. Accelerated fractionation (AF) compared to conventional fractionation (CF) improves loco-regional control in the radiotherapy of advanced head and neck cancers: results of the EORTC 22851 randomized trial. Radiother Oncol 1997; 44(2):111–21. 210 Dobrowsky W, Naude J. Continuous hyperfractionated accelerated radiotherapy with/without mitomycin C in head and neck cancers. Radiother Oncol 2000; 57(2):119–124. 211 Olmi P, Crispino S, Fallai C, et al. Locoregionally advanced carcinoma of the oropharynx: conventional radiotherapy vs. accelerated hyperfractionated radiotherapy vs. concomitant radiotherapy and chemotherapy--a multicenter randomized trial. Int J Radiat Oncol Biol Phys 2003; 55(1):78–92. 212 Maciejewski B, Skladowski K, Pilecki B, et al. Randomized clinical trial on accelerated 7 days per week fractionation in radiotherapy for head and neck cancer. Preliminary report on acute toxicity. Radiother Oncol 1996; 40(2):137-145. 213 Jackson SM, Weir LM, Hay JH, Tsang VH, Durham JS. A randomised trial of accelerated versus conventional radiotherapy in head and neck cancer. Radiother Oncol 1997; 43(1):39-46. 214 Overgaard J, Hansen HS, Specht L, et al. Five compared with six fractions per week of conventional radiotherapy of
215
216
217
218
219
220
221
222
223
224 225 226
squamous-cell carcinoma of head and neck: DAHANCA 6&7 randomised controlled trial. The Lancet 2003; 362(9388):933–940. Poulsen MG, Denham JW, Peters LJ, et al. A randomised trial of accelerated and conventional radiotherapy for stage III and IV squamous carcinoma of the head and neck: a Trans-Tasman Radiation Oncology Group Study. Radiother Oncol 2001; 60(2):113–22. Suwinski R, Bankowska-Wozniak M, Majewski W, et al. Randomized clinical trial on continuous 7-days-a-week postoperative radiotherapy for high-risk squamous cell head-and-neck cancer: a report on acute normal tissue reactions. Radiother Oncol 2005; 77(1):58–64. Sanguineti G, Richetti A, Bignardi M, et al. Accelerated versus conventional fractionated postoperative radiotherapy for advanced head and neck cancer: results of a multicenter Phase III study. Int J Radiat Oncol Biol Phys 2005; 61(3):762–71 Ezzat M, Shouman T, Zaza K, et al. A randomized study of accelerated fractionation radiotherapy with and without mitomycin C in the treatment of locally advanced head and neck cancer. J Egypt Natl Canc Inst 2005; 17(2):85–92 Hliniak A, Gwiazdowska B, Szutkowski Z, et al. A multicentre randomized/controlled trial of a conventional versus modestly accelerated radiotherapy in the laryngeal cancer: influence of a 1 week shortening overall time. Radiother Oncol 2002; 62(1):1–10. Pignon JP, Baujat B, Bourhis J. Apport des meta-analyses sur donnees individuelles au traitement des cancers ORL. Cancer/Radiotherapie 2005; 9(1):31–36. Bourhis J, Audry H, Overgaard J, et al. Meta-analysis of conventional versus altered fractionated radiotherapy in head and neck squamous cell carcinoma (HNSCC): Final analysis. International Journal of Radiation Oncology Biology Physics 2004; 60(Supplement 1):S190–S191. Budach W, Hehr T, Budach V, Belka C, Dietz K. A metaanalysis of hyperfractionated and accelerated radiotherapy and combined chemotherapy and radiotherapy regimens in unresected locally advanced squamous cell carcinoma of the head and neck. BMC Cancer 2006; 6(1):28. Bourhis J, Pignon JP. Chemo-radiation vs altered fractionation in Head & Neck Cancer patients: how, when, for whom? Radiother Oncol 2006; 78(Suppl 1):s19. DerSimonian R, Laird N. Meta-analysis in clinical trials. Controlled Clin Trials 1986; 7:177–188. American Joint Committee on Cancer. Cancer Staging Manual. New York: Springer, 2002 Sobin LH, Witterkind CH, TNM classification of malignant tumors (6th ed.). John Wiley & Sons, New York: UICC 2002, 19–51.
17 Thyroid MASUD HAQ AND CLIVE HARMER
Introduction Epidemiology Aetiology Pathogenesis Pathology and natural history Diagnostic evaluation Management of differentiated thyroid cancer Follow-up
408 408 408 409 410 412 415 421
INTRODUCTION The thyroid follicular cell can give rise to a wide variety of neoplasms, ranging from incidental papillary microcarcinoma that has no effect on life expectancy despite minimal treatment, to lethal anaplastic cancer, invariably fatal despite aggressive treatment. The majority of cases occur in young adults, but thyroid cancer can affect any age group. Although the estimated incidence has increased by 14.6 per cent over the last four decades, the estimated death rate has fallen by 21 per cent. Mortality from thyroid cancer represents less than 1 per cent of all cancer deaths and only 10 per cent of patients with thyroid cancer will die from it. Because of these low rates, morbidity caused by treatment should not exceed that caused by the disease. The management of thyroid cancer demands multidisciplinary care, including consultation with the pathologist, surgeon, oncologist, endocrinologist and nuclear physician.
EPIDEMIOLOGY Thyroid cancer is the commonest endocrine malignancy but it remains rare, accounting for only 1 per cent of all cancers. Incidence rates vary widely, with the highest in Iceland and Hawaii (12–15 per 100 000) and only 2.2 per 100 000 population in the UK. These geographical differences are
Treatment of metastatic disease Dosimetry of 131I therapy Complications of radioiodine treatment External-beam radiotherapy and chemotherapy Management of anaplastic carcinoma Management of medullary thyroid cancer Management of thyroid lymphoma Future prospects References
421 422 423 424 426 428 430 431 433
probably caused by environmental and dietary factors rather than by race or heredity.1 The incidence has been increasing, partly reflecting the past use of radiotherapy for benign childhood conditions but also due to earlier detection with the wide availability of ultrasonography and fine-needle aspiration cytology (FNAC). In England and Wales, a significant increase in incidence was seen during the period 1962–1984, especially in north and mid Wales, where the highest levels of fallout radiation from nuclear reactors were documented.2 In 2001, there were 1431 new cases diagnosed in the UK, with an estimated 300 deaths. Despite these changes, the 5-year survival in the UK has improved to 77 per cent.3 Most cases occur in patients between 25 and 60 years of age, but thyroid cancer can occur in the very young and the elderly. The median age at diagnosis is earlier in females than in males, for both papillary and follicular tumours.4 For papillary carcinoma, the median age at diagnosis is 40 years for females and 44 years for males, while for follicular cancer the median age at diagnosis is 48 years for females and 53 years for males. The incidence of both is higher in women, with a female:male ratio of 3:1.
AETIOLOGY Radiation exposure is the only risk factor known definitively to increase the incidence of well-differentiated cancer,
Pathogenesis 409
although a large study from the Connecticut Tumor Registry showed only 9 per cent of thyroid cancers could be related to radiation.5* By contrast, in several American studies from the 1950s, between 32 per cent and almost 100 per cent of children with thyroid cancer had received prior irradiation for a variety of conditions, including enlarged thymus, tonsils, adenoids or acne. The recognition of the association between irradiation and thyroid cancer led to the elimination of the widespread use of radiotherapy for benign conditions in infants and children by 1960. Large series of patients who had neck irradiation during childhood show that the latent period is at least 3–5 years, with most cases occurring between 20 and 40 years after exposure; there is no apparent drop-off in the increased risk, even after 40 years, following radiation exposure.6* The age at exposure is inversely related to the risk. Analysis of the pooled individual data from seven studies showed that the probability of developing thyroid cancer is related to the radiation dose absorbed by the thyroid;7*** this is a strong indication that radiation is a cause of thyroid cancer. A radiation effect is seen at doses as small as 10 cGy. Over most of the dose range, the data fit best to an excess relative risk model, although an absolute risk model cannot be excluded. At highest doses, cell killing was thought to overtake tumour formation; with increasing dose, the slope (the excess relative risk) of the dose–response curve does not decline, indicating that the relative risk remains significant.6* Data from the acute radiation exposure among the survivors of the atomic bombs in Hiroshima and Nagasaki are similar, showing an increased risk in the younger age population, and in females in particular. The estimated dose to thyroid from these acute incidents also showed a proportional relationship to risk.8*** Improvements in multi-modality therapy incorporating radiation for neoplastic conditions in infants, children and young adults result in many patients cured of tumour who can then be followed up for long-term effects of the therapeutic radiation they received. A dramatically increased relative risk (between 132 and 310) of developing thyroid cancer among individuals treated for neuroblastoma or Wilms’ tumour indicates the significance of age at exposure.9* Patients treated for Hodgkin’s disease or non-Hodgkin’s lymphoma tend to be older and, although the radiation dose to thyroid is high, they more frequently develop hypothyroidism or thyroid nodules than thyroid cancer; the estimated relative risk of developing cancer is 16–80. In the adult population treated with therapeutic radiation, the risk drops off. A relative risk of 2.3 was estimated among 150 000 women treated with radiotherapy for cervical cancer.10* A second type of radiation exposure to the thyroid is from radioisotopes which concentrate in the gland. Large welldesigned studies have shown no increase in thyroid cancer among individuals who had diagnostic nuclide scans or were treated with radioactive iodine for thyrotoxicosis.11* However, data regarding exposure to nuclear fallout in the Marshall Islands, Nevada, and Chernobyl all show a significant increase in thyroid cancer. The importance of age at
exposure is evident in recent studies. In Belarus, the most affected country post-Chernobyl, the incidence increased to 13.5 per 100 000 children in 2000, compared to the usual incidence of less than 1 per 100 000. The majority of cases occurred in children under the age of 10 years at the time of the accident, with at least two-thirds younger than 5.12*** Thyroid tumours can be produced in animals by iodine deficiency or drugs. A common factor in these experimental conditions is prolonged stimulation by thyroid-stimulating hormone (TSH). A sequence of reversible hyperplasia followed by irreversible hyperplasia and in some cases by the subsequent development of follicular carcinoma has been noted.13 Evidence of primary TSH-related induction of thyroid tumours in humans is not convincing. However, papillary carcinoma is more common in iodine-rich areas such as islands, while a number of case-controlled studies have strongly suggested that low dietary iodine content is responsible for the increased rates of follicular and anaplastic cancer in areas of endemic goitre.14*** In addition, dietary iodine supplementation has been shown to increase the relative proportion of papillary cancer and to decrease the frequency of follicular cancer. Because of the strong female predominance, the influence of sex-hormone status has been investigated. Factors such as parity, early menopause, contraceptive use and late age at first birth have been associated with an increased risk of thyroid cancer, although these associations are weak and inconsistent. A pooled analysis of case-control studies also confirmed a weak association of menstrual and reproductive factors with thyroid cancer risk.15*** The biological basis of these epidemiological observations could be that oestrogen acts as a growth promoter on thyrocytes. Some experimental evidence suggests that thyrocytes express oestrogen receptors, and oestrogens may stimulate thyrocyte growth in cell-culture systems. It has also recently been shown that tamoxifen inhibits the growth of papillary cancer cells both in vitro and in vivo. Genetic factors may play a role in a small group of patients with differentiated thyroid tumours. Familial syndromes associated with thyroid tumours of follicular-cell origin include familial adenomatous polyposis coli, Gardner’s syndrome, Turcot’s syndrome, Cowden’s disease and Carney complex. Familial occurrence of differentiated thyroid cancer has also been recognized, although it appears to be rare, occurring in only 2.5–4 per cent of cases.16
PATHOGENESIS Although thyroid follicular cells can give rise to both benign and malignant tumours, the evidence to support an adenoma to carcinoma multi-step pathogenesis is not universally accepted. Malignant transformation is due to the activation of proto-oncogenes or the inactivation of tumoursuppressor genes in combination with environmental factors. Defects to the tyrosine kinase (trk) receptor genes are commonly found in thyroid cancer. The interaction of
410 Thyroid
these receptors with growth factors leads to the activation of the mitogen-activated protein (MAP) kinase pathway through ras and braf proteins, resulting in uncontrolled cell division. Abnormalities of the trk receptor genes have almost exclusively been associated with papillary carcinoma; these include ret/papillary thyroid cancer (PTC) rearrangements, trk rearrangements and met over-expression. The higher frequency of ret/PTC rearrangements in patients post-Chernobyl suggests that radiation exposure may be one of the mechanisms leading to their activation. Activation of the ras proto-oncogene by a point mutation is common in both papillary and follicular tumours (approximately 40 per cent). Activation of this oncogene is found at all stages from benign through well-differentiated to undifferentiated carcinoma, suggesting that it represents an early event and that this defect is not by itself sufficient for carcinogenesis. A high frequency of braf protein point mutations has also been identified in papillary thyroid cancer. This mutation is associated with tumours of aggressive behaviour and may serve as a novel independent molecular prognostic marker in risk evaluation.17* Translocation between genes encoding transcription factors Pax8 (Paired box 8) and PPARγ (peroxisome proliferator-activated receptor gamma) has been identified in both follicular adenoma and carcinoma, suggesting a role in their pathogenesis. Inactivating point mutations in the p53 tumour suppressor gene have been observed in many undifferentiated (anaplastic) thyroid cancers.
PATHOLOGY AND NATURAL HISTORY Thyroid tumours can originate from the follicular epithelium, from parafollicular or C cells, or from non-epithelial stromal elements.18*** The World Health Organisation (WHO) classifies malignant epithelial thyroid tumours as: papillary carcinoma, follicular carcinoma, medullary carcinoma and undifferentiated (anaplastic) carcinoma.19 The pTNM classification is recommended to assist in management decisions and for uniformity of case-series reporting.
Papillary carcinoma Papillary carcinoma is the most common type, comprising 80 per cent of all thyroid malignancies. These tumours are almost three times as common in women as in men, with a peak incidence in the third and fourth decades. The histological hallmarks are branching papillae arranged on a fibrovascular stalk. Tumour cells are cuboidal with homogeneous cytoplasm and characteristic hypochromatic nuclei with absent nucleoli (Orphan Annie eyes) and may contain laminated calcified psammoma bodies (Fig.17.1a). The term ‘mixed’ papillary and follicular carcinoma is no longer used, because the majority of papillary carcinomas do contain some follicular areas. Features of papillary cancer include an
infiltrating pattern of growth, multi-focality (up to 75 per cent of cases) and spread to the regional lymph nodes. Obvious cervical adenopathy is seen in 50 per cent of patients at presentation, but has been reported in as many as 90 per cent of those who underwent elective node dissection. Haematogenous metastases are uncommon and mainly involve the lungs; lung involvement at diagnosis occurs in 5–10 per cent of cases in adults but may be in up to 25 per cent of children. Certain variants of papillary cancer such as the tall cell, columnar and diffuse sclerosing variant have been shown to be more aggressive. Micropapillary or ‘occult’ carcinoma has the same histological features as papillary cancer but has been defined by the WHO to be a tumour 1 cm or less in maximum dimension. These tumours are a common incidental finding at autopsy, their incidence being dependent on overall age of the population, the ethnic group studied and the diligence with which the pathologist looks for these foci. Incidence ranges from 0.5 per cent to 14 per cent, with a greater incidence in older age groups. The discrepancy between the incidences of occult and clinically detected thyroid cancer indicates that these minimal lesions may have a different biology. Although they can metastasize to regional lymph nodes, they rarely cause significant morbidity or mortality.20***
Follicular carcinoma Follicular carcinoma accounts for 5–20 per cent of thyroid tumours, is also three times more common in females than in males, but tends to present in mid-life. It is a tumour of follicular origin but lacks the diagnostic features of papillary cancer. It may be extremely difficult to diagnose when well differentiated, as the appearance is similar to that of both normal thyroid and benign follicular adenoma. The presence of capsular or vascular invasion is often the only feature to denote malignancy (Fig. 17.1b). Depending on the degree of invasiveness, they may be described as minimally or widely invasive. This distinction defines the less favourable prognosis associated with the latter. Lymphnode metastases are less common than in papillary tumours, but haematogenous spread, mainly to bones and lungs, is present in 14 per cent of patients at diagnosis. Hürthle-cell carcinoma was previously considered a variant of follicular cancer and is composed of cells that exhibit oncocytic changes (Fig. 17.1c). However, it is now recognized as a distinct pathological entity because of its different oncogenic expression. The majority of Hürthle-cell tumours are benign, but malignancy is well documented in the form of local recurrence and distant metastases. Histopathological studies have shown that either capsular or vascular invasion is a reliable criterion of malignancy. Although they are usually well differentiated and produce thyroglobulin, these tumours rarely take up iodine. This is probably a contributory factor to their poorer prognosis than other follicular carcinomas.21,22
Pathology and natural history 411
(a)
(b)
(c)
(d)
(e) Figure 17.1 (a) Papillary cancer, tall cell variant; (b) follicular cancer, showing capsular invasion; (c) Hürthle cell carcinoma; (d) medullary carcinoma with amyloid stroma; (e) non-Hodgkin’s lymphoma of MALT type
Insular carcinoma is a form of poorly differentiated tumour arising from follicular cells with behaviour intermediate between differentiated (papillary/follicular) and undifferentiated (anaplastic) carcinoma. Microscopically, the tumour is characterized by a nested growth pattern and prominent vascularity. It invades both lymphatics and veins, resulting in nodal and distant metastases, and has a poor prognosis.
Anaplastic carcinoma Anaplastic thyroid cancer is one of the most aggressive of all malignancies and one of the most lethal. Epidemiological studies indicate that the incidence has decreased to approximately 1–3 per cent of all thyroid tumours23* in the period 1968 to 1992. This decrease has been attributed partially to dietary iodine prophylaxis and an overall decrease
412 Thyroid
Table 17.1 Medullary carcinoma of the thyroid (MTC) syndromes. Phenotype Sporadic MTC MEN IIa
Frequency (%)
Presentation
80 9
MTC MTC, phaeochromocytoma, hyperparathyroidism MTC, phaeochromocytoma, neuromas (see Fig. 17.2), Marfanoid habitus MTC
MEN IIb
3
FMTC
8
MEN, multiple endocrine neoplasia; FMTC, familial medullary thyroid cancer.
in endemic iodine-deficient goitre. It is most common in the elderly, with 75 per cent of patients being over 60 years old. Most series quote a ratio of men to women of 1:3. It may be associated with a long history of goitre, and a significant number of cases are probably causally associated with a pre-existing well-differentiated thyroid tumour.24 These tumours also arise from the follicular cell. However, the natural history, clinical presentation and outcome reflect their undifferentiated biology with rapid growth and invasive characteristics. Histological variants include small cell, giant cell and spindle cell, although their behaviour does not differ significantly. Patients present with a rapidly enlarging collar of tumour and confluent lymphadenopathy often invading the trachea, larynx or oesophagus, resulting in stridor, hoarseness or dysphagia. The majority die within 6 months of the first symptom, from aggressive loco-regional disease. At diagnosis, 25–50 per cent of patients have pulmonary metastases and at death this figure approaches 100 per cent.
intestinal peptide (VIP). It is composed of small round cells within an amyloid stroma. Immunohistochemical staining for calcitonin granules is the most accurate method to establish the diagnosis (Fig. 17.1d). Fewer than 20 per cent of sporadic tumours are bilateral, but in the familial syndromes, medullary cancer is usually bilateral and multi-centric. At presentation, involvement of cervical or mediastinal lymph nodes is seen in 11–75 per cent of patients, and distant metastases (mainly to lung, bone and liver) in 12 per cent.
Lymphoma of the thyroid Lymphoma of the thyroid is a rare disease, representing 2 per cent of thyroid malignancies and 2 per cent of extranodal lymphomas. Chronic autoimmune stimulation, as in Hashimoto’s thyroiditis, is a predisposing factor. There is a strong female predominance, ranging from 3:1 to 8:1, while the median age at diagnosis is in the seventh decade, similar to that of anaplastic cancer, from which it must be distinguished (Fig. 17.1e). Patients present with a rapidly enlarging, painless neck mass, and one-third of patients experience compressive symptoms. Symptoms also include fever, night sweats and weight loss, but these are rare.28* Almost all thyroid lymphomas are non-Hodgkin’s of B-cell origin, intermediate or high grade. Biologically, many thyroid, breast, parotid, lung and gastrointestinal lymphomas are a distinct subset of extranodal lymphomas derived from mucosa-associated lymphoid tissue (MALT). These small-cell lymphomas are characterized by a low grade of malignancy, slow growth rate and a tendency for late relapse or second lymphomas in other MALT sites.29*
Rare tumours Medullary carcinoma Medullary carcinoma of the thyroid (MTC) was first described by Hazard in 1959 as a solid neoplasm without follicular histology. It accounts for 7 per cent of all thyroid tumours but for 14 per cent of all thyroid cancer mortality. The incidence has been increasing in recent years due to screening.25*** Sporadic MTC accounts for 70–80 per cent of cases, with the remainder being familial. Hereditary MTC can occur alone – familial medullary thyroid carcinoma (FMTC)26* – or as the thyroid manifestation of multiple endocrine neoplasia type II syndromes (MEN IIa and MEN IIb), as shown in Table 17.1. These are autosomal dominant disorders caused by germline mutations in the RET protooncogene, located on the long arm of chromosome 10, band q11.2, which codes for a tyrosine kinase receptor.27 Medullary carcinoma of the thyroid arises from the parafollicular or ‘C’ cells, which are of neural crest origin and secrete calcitonin as well as other peptides: carcinoembryonic antigen (CEA), adrenocorticotrophic hormone (ACTH), serotonin, bradykinin, prostaglandin and vasoactive
While the relatively low mortality rate of thyroid cancer in general is due to the preponderance of well-differentiated carcinoma, there exists a subset of rare thyroid tumours that exhibit aggressive behaviour and have a poor prognosis. These require careful consideration and different treatment paradigms to optimize clinical outcome. Extremely rare types of thyroid cancer include thymus-like tumours, mucoepidermoid carcinoma, mixed medullary follicular cancers, teratomas and sarcomas, including angiosarcoma and liposarcoma.30* In contrast, the thyroid is more commonly affected by metastases originating from other primary tumour sites.31*
DIAGNOSTIC EVALUATION Palpable thyroid nodules are present in 4–7 per cent of all adults; age, gender, history of exposure to ionizing radiation and method by which the nodules are detected all significantly influence the findings of different retrospective
Diagnostic evaluation 413
Table 17.2 Diagnostic evaluation for thyroid tumours. Procedure History
Physical examination
Finding
Significance
Radiotherapy to head and neck Family history of MTC Family or personal history of phaeochromocytoma or hyperparathyroidism Diarrhoea, flushing Hashimoto’s thyroiditis Solitary thyroid nodule
Known aetiology of thyroid cancer Inherited in an autosomal dominant pattern Suggestive of MEN IIA or IIB syndrome
Common in MTC Known association with thyroid lymphoma Higher incidence of cancer in solitary nodule (5–15%) Lower incidence of cancer; cancer may present in dominant nodule May indicate cancer
Multiple nodules Thyroid fixation, hoarseness, Horner’s syndrome Enlarged cervical lymph nodes Fine-needle aspiration Ultrasonography X-ray Radionuclide imaging (131I, 99 mTc) 131
I mIBG, 111In octreotide
CT scan, MRI Thyroglobulin
Malignant, suspicious, benign, insufficient sample Differentiates solid from cystic nodules, assists in fine-needle biopsy Psammomatous calcification Cold, warm or hot nodule
Imaging of medullary thyroid cancer Extent of primary tumour, lymph nodes and metastases Preoperative elevated value Postoperative elevated value
Calcitonin
More consistent sign of malignancy
Normal value Preoperative elevated value Postoperative elevated value Postoperative normal value
70–97% accuracy Solid nodules more often malignant Suggests thyroid nodule is malignant 15–25% of cold nodules are malignant; lower incidence of cancer in warm and hot nodules May detect residual, recurrent or metastatic cancer Assist in treatment planning Does not distinguish between benign and malignant tumour Indicates residual, recurrent or metastatic thyroid cancer Supportive evidence of lack of disease Indicates C-cell hyperplasia or MTC Indicates residual/recurrent/metastatic MTC Indicates lack of disease
MEN, multiple endocrine neoplasia; MTC, medullary carcinoma of the thyroid; mIBG, meta-iodobenzylguanidine; CT, computed tomography; MRI, magnetic resonance imaging.
studies. In one pathological study, up to 90 per cent of women over the age of 70 years and 60 per cent of men over the age of 80 years had nodular goitre. In most series, a 5–15 per cent risk of cancer in all thyroid nodules for the total population is reported.32 Therefore, it is neither practical nor necessary to remove every nodule in order to exclude malignancy. Investigations should be directed towards selecting those with an increased risk of malignancy. However, no single clinical feature, physical finding or laboratory test is pathognomonic for the detection of thyroid cancer, except for the serum calcitonin level in medullary carcinoma and FNAC. A diagnostic algorithm is presented in Table 17.2. Information from the history and physical examination may help in assessing the risk of malignancy. Exposure to
ionizing radiation, extremes of age, family history of thyroid cancer or MEN syndromes (Fig. 17.2) and other inherited disorders, such as Gardner’s syndrome and Cowden’s disease, increase the suspicion of cancer. Although not specific for malignancy, a history of rapid growth, pain, hoarseness, or airway obstruction is of concern. On examination of the neck, attention should be paid to the size, consistency, mobility and number of nodules as well as to the presence of enlarged lymph nodes. The risk of malignancy is greater in a solitary nodule (5–15 per cent) than in multiple nodules; a dominant nodule or a nodule that changes size in a multinodular goitre requires further investigation. Cervical adenopathy is probably the most consistent feature of malignancy with a thyroid mass, but lacks specificity.
414 Thyroid
Figure 17.2 Patient with MEN IIb demonstrating neuromas of the tongue. Neuromas may also involve the buccal mucosa, eyelid, conjunctiva and glans penis
High-resolution ultrasonography is a useful adjunct to clinical examination for the assessment of nodule size, the detection of multiple nodules and for assisting in FNAC. A hyperechoic nodule with well-defined margins is more likely to be benign. In contrast, a nodule that appears hypoechoic with irregular margins, microcalcification, and increased blood flow should raise suspicion. Despite these features, ultrasound cannot reliably distinguish benign from malignant lesions; it can, however, establish whether a lesion is solid or cystic. In a review of 16 series, 21 per cent of the solid lesions, 12 per cent of the mixed and 7 per cent of the cystic lesions were cancerous.33* Therefore, a solid mass, although most often benign, has the highest chance of being malignant. The single most important investigation for evaluating thyroid nodules is FNAC. It can often be undertaken in the clinic, without the need for ultrasound guidance. The impact this procedure has had on clinical practice is reflected by a reduction in the number of thyroid operations performed, a greater proportion of malignancies removed at surgery, and an overall reduction in the cost of managing patients with nodules.34*** The diagnostic accuracy of FNAC ranges from 70 per cent to 95 per cent, but is dependent on the skills of the operator and the reporting cytopathologist. The report should be defined as inadequate (C1), benign (C2), follicular (C3), suspicious (C4) or malignant (C5). Results typically comprise benign 70 per cent, malignant 4 per cent, suspicious 9 per cent, or inadequate 17 per cent. Inadequate samples should be followed by repeat FNAC, preferably under ultrasound guidance. The malignant potential of follicular neoplasms cannot be determined cytologically, therefore surgical excision is mandatory. False-negative results are reported in 1–6 per cent and false-positive results in 3–6 per cent. Fine-needle aspiration cytology is adequate to diagnose anaplastic cancer, but Trucut biopsy is required to distinguish reliably the subtypes of primary thyroid lymphoma by immunohistochemistry and ascertain MALT status.
Radionuclide imaging with iodine (131I) or sodium pertechnetate (99 mTc) is of limited value in the initial evaluation of thyroid nodules since the advent of FNAC. Malignant thyroid tissue either does not incorporate iodine or incorporates less iodine than normal thyroid, so that a malignant lesion appears as a cold area on the scan.35*** Scanning cannot differentiate benign from malignant nodules and thus is used to assign only a probability of malignancy. Malignancy has been reported in 10–15 per cent of cold nodules, 5 per cent of warm and less than 2 per cent of hot nodules.36*** These results indicate that cold nodules are more likely to be malignant, but warm and hot nodules can also be malignant. The goals of scintigraphy in patients with established thyroid cancer are to locate metastases or residual neoplasm and to predict the efficacy of therapy with 131I. Following thyroidectomy and radioiodine ablation of normal residual thyroid tissue, diagnostic whole-body 131I scanning is highly specific; foci of uptake correspond to thyroid cancer metastases. While it is clear that higher scanning doses improve the visualization of disease, even low doses of 75 MBq diminish the uptake of subsequent therapeutic 131 I. Proposals to avoid this tumour stunning include the use of smaller diagnostic doses and the use of an alternative tracer such as 123I. Less well-differentiated tumours and some that are well differentiated (especially in older patients) may concentrate so little 131I that the diagnostic scan will prove false negative and the therapeutic dose will not treat effectively. Locating these tumours may be helpful in planning alternative treatment such as surgical resection. Thallium-201 chloride and radiolabelled somatostatin analogues concentrate in more than half of thyroid tumours not visualized by 131I.37*,38* Fluorodeoxyglucose positron emission tomography (18FDG-PET) may also be useful in suspected recurrent well-differentiated tumour, but is of greater value for poorly differentiated carcinoma and MTC. Indium-111 (111In) octreotide and 131I meta-iodobenzylguanidine (mIBG) are concentrated in 25–60 per cent of MTC and may be helpful in locating sites of disease, particularly in patients with an elevated calcitonin but no clinical or radiological evidence of tumour39*** see Table 17.2. Pentavalent dimercapto succinic acid (V-DMSA) whole-body scintigraphy is the best method for locating occult MTC, but has no therapeutic potential. Computed tomography (CT) can define the morphology of the thyroid gland, tumour extension to structures such as trachea or vessels, and nodal involvement in the neck and mediastinum. Imaging of the chest may reveal micronodular disease in the lungs not shown on chest X-ray. In addition, CT is particularly valuable in radiotherapy treatment planning and in assessing the effectiveness of therapy. Magnetic resonance imaging (MRI) is preferred in order to avoid the use of iodinated contrast, which remains in the body for a considerable time, reducing the uptake of subsequent radioiodine therapy. Magnetic resonance imaging is useful in depicting lesion margins, the invasion of adjacent
Management of differentiated thyroid cancer 415
structures and cervical lymphadenopathy. The presence of a cystic node or a nodal diameter of 15 mm suggests metastasis; using these two criteria, a specificity of 100 per cent with an 82 per cent accuracy but a sensitivity of only 60 per cent has been estimated.40* Magnetic resonance spectroscopy has yet to be evaluated.
MANAGEMENT OF DIFFERENTIATED THYROID CANCER Differentiated thyroid cancer is one of the most controversial malignancies as regards treatment. Debate involves almost every stage of management decision: the extent of initial surgery, the need for lymph-node dissection, the role of radioactive iodine ablation, the value of dosimetry in radioiodine therapy, and the role of adjuvant externalbeam radiotherapy (EBRT).41*** The major reasons for these ongoing controversies are that thyroid cancer is rare and that its behaviour is indolent (although there are groups of tumours that behave more aggressively). Due to lack of prospective studies, the potential for relapse may be either underestimated, resulting in patients receiving inadequate treatment, or, due to referral bias, overstated and patients are overtreated. Evidence-based guidelines for the management of thyroid cancer in adults were published in 2002 by the British Thyroid Association in conjunction with the Royal College of Physicians of London.42 Certain factors have been linked to the behaviour of welldifferentiated carcinoma and are used to determine prognosis. Groups such as the Mayo Clinic, the Lahey Clinic and the European Thyroid Association have published data on factors identified through retrospective analyses that correlate with survival. Because all factors are closely interrelated,
only multivariate analysis can identify their individual prognostic significance. In most studies, age is found to be the most important predictor of outcome; a significant increase in mortality is seen over the age of 40 (Fig. 17.3). Males tend to fare worse than females, but gender is of much less prognostic importance. Tumour size, extension and grade are also related to the risks of recurrence and survival (Fig. 17.4). Patients with papillary cancer fare better than those with follicular well-differentiated carcinoma; less well-differentiated follicular tumours show a significantly higher relapse rate and shorter survival. The impact of nodal status is controversial. In some series, it is found to be associated with an increased risk of loco-regional recurrence, although the effect on survival is less clear. The results of multivariate analysis for 1390 patients with differentiated thyroid cancer treated at the Royal Marsden Hospital, London, between 1929 and 1999 are shown in Table 17.3.43 Based on these prognostic factors, several scoring systems have been developed to help in assessing individual patient risk of dying from cancer and planning treatment (AGES, AMES, GAMES, MACIS). None of these systems is perfect, but all succeed in identifying at least high-risk and low-risk patients.44*,45*** The low-risk group includes young patients with small (1 cm) well-differentiated tumours confined to the gland with no evidence of nodal or distant spread. Prognosis in this group is excellent, with cancerspecific mortality less than 1 per cent at 30 years.46* The recurrence and survival rates are strikingly different in the high-risk group, which includes older patients with locally advanced or metastatic disease at presentation47* or less well-differentiated tumours. Risk-group analysis makes a selective approach to treatment possible and can spare many patients the morbidity of unnecessarily aggressive treatment, without compromising outcome.
% Probability of cause-specific survival
100
80
40 40–49
60 50–59 40 60 AGE 40 N 548 AGE 40–49 N 275 AGE 50–59 N 243 AGE 60 N 321 Chi-squared 242.2
20
O 31 O 49 O 67 O 123 df 3
E 127.1 E 58.6 E 41.9 E 42.4 p 0.000
0 0
10
20
30
Time since diagnosis (years)
40
50
Figure 17.3 Differentiated thyroid cancer: Royal Marsden Hospital experience 1929–99 (1390 patients). Cause-specific survival according to age.
416 Thyroid
Both serum Tg monitoring and iodine scanning may be difficult to interpret in the presence of a substantial thyroid remnant. Remnant ablation in this situation is associated with a low success rate, of the order of only 30 per cent, with higher administered activities or repeat doses required to achieve complete ablation. The main argument against radical surgery, i.e. morbidity, is less important now that experienced surgeons are reporting reduced complication rates. Vocal-cord paralysis occurs in only 1–3 per cent and permanent hypocalcaemia in 1–6 per cent in most specialist centres following total or near-total thyroidectomy. The role of elective level VI tracheo-oesophageal groove clearance of nodes down to the level of the thymus awaits clarification. Those in favour argue that this is the commonest site of lymph-node metastases,52* that the potential morbidity associated with re-operating at a later date is much greater, and that there is an increased risk of subsequently finding inoperable disease. However, this procedure does carry a slightly increased risk of hypoparathyroidism and recurrent laryngeal nerve damage. Our policy is to perform an elective tracheo-oesophageal groove clearance on the ipsilateral side only, proceeding to a lateral neck dissection if paratracheal disease is demonstrated clinically, radiologically or during the operation. If extensive superior mediastinal disease is detected, we favour a combined thoraco-cervical approach using a sternal split to extend dissection down to the arch of the aorta.53* The surgical management of lymph-node metastases is also controversial. The presence of metastatic disease in nodes, especially in the lower-risk young patients, is not always seen as an adverse prognostic factor.50* It does correlate with increased loco-regional recurrence in most series, although the effect on survival is not clear. However, the largest single institution series of papillary carcinom a (2192 cases) did confirm a significant adverse effect of
Surgical treatment Surgery remains the initial and potentially curative treatment for differentiated thyroid cancer. However, there is no universal agreement as to the extent of the surgical procedure, partly because many still regard thyroid cancer as a non-fatal disease. There have been no prospective randomized trials and in retrospective analyses there is always patient-selection bias as well as the confounding effect of non-surgical adjuvant therapy. The minimum requirement is complete excision of all macroscopic disease, which usually includes ipsilateral lobectomy and isthmusectomy, avoiding damage to the parathyroid glands and recurrent laryngeal nerves.48 A more radical initial approach is advocated by the majority of surgeons in view of the high incidence of bilateral multi-focality (as high as 87 per cent), especially in papillary carcinoma. Most studies have shown a significant reduction in local recurrence rates following total (or near-total) thyroidectomy, with some also reporting improved overall survival.49*,50* Even in the low-risk patients with tumours smaller than 1 cm diameter, the loco-regional recurrence rate after lobectomy significantly exceeded that seen after total thyroidectomy – 20 per cent and 5 per cent respectively at 20 years – although overall survival was similar.46* Survival was improved with bilateral resection in patients with unfavourable prognostic factors.49* This was also true for locally advanced tumours invading the aerodigestive tract: following complete resection, the 5-year survival was greater than 50 per cent, but without such intervention, 80 per cent of these patients were dead at 5 years.51* Other major advantages in favour of total thyroidectomy are that postoperative follow-up using serum thyroglobulin (Tg) and radioiodine diagnostic scanning are facilitated, as well as subsequent 131I therapy if this becomes necessary.
100
% Probability of survival
80
T1
60
40
T2
T3 T1 N 206 T2 N 684 T3 N 199 T4 N 254
20
Chi-squared 140
0 0
10
O 27 O 183 O 80 O 116
E 65.6 E 238.5 E 55.2 E 46.7
df 1
p 0.0001
20
T4 30
Time since diagnosis (years)
40
50
Figure 17.4 Differentiated thyroid cancer: Royal Marsden Hospital experience 1929–99 (1390 patients). Survival according to tumour stage
Management of differentiated thyroid cancer 417
Table 17.3 Prognostic factors for survival, local recurrence and distant recurrence for 1390 patients with differentiated thyroid cancer treated at RMH between 1929 and 1999 (Cox proportional hazards regression model for multivariate analysis). HR (95% CI) Factor Age (years) 40 40–49 50–59 60 T stage T1 T2 T3 T4 M stage M0 M1 Surgery NT/TT ST/HT/L B/E Iodine ablation No Yes Grade I II III
Survival
Local recurrence
Distant recurrence
1.0 2.5 (2.3–2.7) 6.2 (5.6–6.8) 15.3 (13.9–16.8) P 0.001
1.0 1.4 (1.3–1.6) 2.1 (1.9–2.3) 2.9 (2.7–3.2) P 0.001
1.0 1.8 (1.6–2.1) 3.3 (2.9–3.7) 6.0 (5.3–6.8) P 0.001
1.0 1.9 (1.7–2.1) 3.5 (3.1–3.9) 6.5 (5.8–7.2) P 0.001
1.0 1.7 (1.5–1.9) 2.8 (2.5–3.2) 4.8 (4.2–5.4) P 0.001
1.0 1.8 (1.6–2.1) 3.4 (2.9–3.9) 6.2 (5.4–7.2) P 0.001
1.0 6.9 (5.4–8.7) P 0.001
1.0 0.92 (0.51–1.6) NS
1.0 3.9 (2.6–6.0) P 0.001
1.0 1.89 (1.49–2.4) 4.66 (3.56–6.1) P 0.001
1.0 1.8 (1.38–2.36) 3.48 (2.49–4.85) P 0.001
1.0 1.87 (1.3–2.5) 3.72 (2.5–5.5) P 0.001
1.0 0.67 (0.55–0.82) P 0.001
1.0 0.39 (0.31–0.50) P 0.001
1.0 0.4 (0.3–0.58) P 0.001
1.0 2.6 (2.3–2.9) 6.6 (5.9–7.5) P 0.001
1.0 2.0 (1.7–2.3) 3.8 (3.3–4.5) P 0.001
1.0 2.6 (2.1–3.1) 6.7 (5.6–8.0) P 0.001
RMH, Royal Marsden Hospital; NS, not significant; NT, near total thyroidectomy; TT, total thyroidectomy; ST, sub-total thyroidectomy; HT, hemi-thyroidectomy; L, lobectomy; B, biopsy; E, enucleation; HR, hazards regression.
clinical nodal status on survival.54* Papillary carcinoma can be found in the regional lymph nodes of 35–75 per cent of patients when dissections are performed and carefully examined. However, recurrence is not seen in 75 per cent of patients with clinically uninvolved nodes not undergoing routine neck dissection. This argues against the need for elective lateral deep cervical-node dissection of the clinically uninvolved neck. If clinically apparent nodal disease is present, many surgeons recommend a modified neck dissection preserving the sternomastoid, spinal accessory nerve and internal jugular vein.55*** However, some still perform simple node excision, previously known as ‘berry picking’; although regional recurrence rates are high, a survival disadvantage has not been proven. If nodes are clinically involved,
we favour a selective node dissection of levels II, III, IV and VI; thyroid cancer rarely spreads to submandibular or posterior cervical lymph nodes (levels I and V).
Endocrine treatment Thyroid stimulating hormone is the main regulator of thyroid function, differentiation and proliferation. Binding of TSH to its receptor on thyroid cells primarily activates a cyclic adenosine monophosphate (cAMP) cascade, leading to thyroid hormone synthesis and release, as well as to expression of thyroid-specific genes, including those encoding Tg and thyroperoxidase. Differentiated thyroid
418 Thyroid
100
% Probability of survival
80
60
40 Iodine ablation Iodine ablation N 923 None N 467 Chi-squared 14.61
20
O 192 O 233 df 1
E 231.2 E 193.8 p 0.001
None
0 0
10
20
30
Time since diagnosis (years)
carcinomas retain some degree of thyroid-specific gene expression and function similar to normal thyroid cells; therefore they are responsive to stimulation by TSH. In thyroid cancer cell lines, TSH has been shown to stimulate vascular endothelial growth factor secretion and angiogenesis. Thus TSH may promote growth in some thyroid cancers.56* The beneficial effect of TSH suppression has not been assessed in prospective studies. However, available data suggest that thyroxine reduces the risk of recurrence, tumour progression and death from thyroid cancer.57*** It is generally accepted that a TSH level below 0.1 mU/L is desirable,42 but there is no evidence that undetectable TSH levels offer any advantage over low but detectable levels. Monitoring the free thyroxine (T4) level in the athyroid patient receiving thyroxine often gives a false high value, therefore we favour total TSH suppression (0.1 mU/L), maintaining the free tri-iodothyronine (T3) level in the normal range in order to avoid hyperthyroidism.58* Concern regarding permanently low TSH levels on bone metabolism have failed to show an increased fracture risk despite accelerated bone turnover59*,60* based on biochemical markers of osteoclastic activity. Nevertheless, postmenopausal women who are at risk may benefit from adequate dietary calcium intake, biphosphonates, or hormone-replacement therapy.
Radioactive iodine ablation of residual thyroid tissue The value of postoperative 131I to ablate residual normal thyroid is still debated. Arguments in favour of remnant ablation are that it permits the subsequent identification by a whole-body scan of any residual or metastatic carcinoma and increases the sensitivity of Tg measurement for follow-up.61 Most importantly, several retrospective
40
50
Figure 17.5 Differentiated thyroid cancer: Royal Marsden Hospital experience 1929–99 (1390 patients). Survival in relation to radioiodine
studies have documented that it decreases tumour recurrence and death50*,62*** (Fig. 17.5). However, the beneficial effect of 131I ablation can be seen mainly in patients who are at high risk of recurrence, such as those with larger tumours, extrathyroid extension and involved lymph nodes, as well as those with residual disease.63* In low-risk patient groups, and especially in those with microcarcinoma, prognosis is so favourable after surgery alone that little further improvement is possible with 131I ablation.46* The optimal activity of 131I required to achieve successful ablation remains controversial, with doses ranging between 1.1 GBq and 3.7 GBq. Higher initial iodine doses were thought to be more effective in achieving complete ablation with a single administration. The philosophy of a large dose ablation was based on the possibility that it ablates not only remnants but also possible micrometastatic deposits.64* The same authors also stressed the importance of delivering maximal radiation dose from the first iodine administration; due to mechanisms that are at present poorly understood, the biological half-life of subsequent administrations falls, therefore reducing the radiation dose delivered. In 1976, McCowan et al. were the first to report that iodine doses of 3–3.7 GBq were not more effective than 1.1 GBq,65* findings that were confirmed by several retrospective analyses.66* The long-term tumour recurrence rate was 7 per cent following a low dose of 1.1–1.85 GBq, compared with 9 per cent following a high dose of 1.9–7.4 GBq.67*** The advantages of administering the smallest effective dose of 131I are patient convenience and lower cost, as well as a reduced risk of treatment-related complications from lower whole-body radiation exposure. The only prospective randomized clinical trial to evaluate the optimal 131I ablation dose (involving 149 patients) showed that increasing the administered activity beyond 1.85 GBq resulted in plateauing of the dose–response curve; a radiation absorbed dose to the thyroid remnant
Management of differentiated thyroid cancer 419
New patient referred with Differentiated Thyroid Cancer History and clinical examination Review histology With total thyroidectomy
Without total thyroidectomy Completion thyroidectomy ⴙ Level VI
selective dissection of levels II, III, IV
At 4 weeks – avoid sea food, added salt, iodine-containing medicines and x-ray contrast FBC, biochemistry, TSH, Tg, CXR Ablation dose 3 GBq 131I 131
I whole body scan at 3 days, start T3 OPD appointment at 6 weeks TSH, T3, Tg
Abnormal scan or adverse feature Book 4 months after ablation 5.5 GBq 131I Stop T3 for 14 days FBC, biochemistry, TSH, Tg
Normal scan, no adverse feature 185 MBq 131I whole-body scan only if Tg-antibodies or clinical suspicion. Stop T3 for 14 days (or give rhTSH) TSH, Tg
Therapy dose 5.5 GBq 131I 131
I whole body scan at 3 days, restart T3
OPD appointment at 6--8 months stimulated-Tg neck US If tests negative, change to thyroxine 200mcg, target TSH 0.1mU/l
OPD appointment at 6 weeks TSH, T3, Tg
Known tumour Negative scan PBI 0.01%
Reconsider surgery
Positive scan Tg 3ug/L PBI 0.01%
Normal
External beam radiotherapy
Annual follow-up TSH, T3, Tg
Repeat 5.5 GBq 131 I at 6 months
Figure 17.6a New patient referred with differentiated thyroid cancer. T3, tri-iodothyronine; FBC, full blood count; TSH, thyroid-stimulating hormone; CXR, chest radiograph; US, ultrasound; Tg, thyroglobulin; PBI, protein-bound radioactive iodine concentration; rhTSH, recombinant human TSH.
greater than 300 Gy did not result in a higher ablation rate.68** Successful ablation was achieved in 77 per cent of thyroid remnants with the lower dose of 1.85 GBq. Maxon et al. used dosimetry to individualize administered activity so as to deliver a radiation dose of 300 Gy to the thyroid remnants.69* They reported an 81 per cent ablation rate, with no apparent gain from using a dose greater than 300 Gy. However, one meta-analysis found that a single administered activity of 1110 MBq was more likely to be unsuccessful in fully ablating thyroid remnants compared to higher activities of 2775–3700 MBq.70***
Four weeks after total thyroidectomy, by which time the level of TSH should be 30 mU/L,71*** we recommend an ablation dose of 3 GBq 131I to all patients with differentiated thyroid cancer, except children over the age of 10 with small node-negative tumours and patients in whom carcinoma is an incidental microscopic histological finding (Fig. 17.6a). This dose ablates 75 per cent of remnants and delivers a mean radiation dose of 410 Gy.72* Scans of the neck and whole body are obtained on the third day (Fig. 17.7), when the patient is usually discharged from the ward, subject to the total-body radioactivity having fallen below the
420 Thyroid
Follow-up algorithm: differentiated thyroid cancer low risk patients Patients clinically disease free at 6–8 months after surgery and 131I ablation
Clinical examination, US neck, stimulated-Tg* on TSH suppression
Tg < 1ug/L no other abnormality
Tg between 1 and 2ug/L no other abnormality
Tg > 2ug/L or clinical disease
MRI neck, CT chest
surgery or radioiodine therapy
repeat rhTSH-Tg* after 6 months
Tg 1ug/L
Tg stable
Tg rising
post-therapy scan
Positive
Negative
annual follow-up Tg on TSH suppression rhTSH – FDG PET/CT
18
Patients suitable for inclusion:Low risk patients for recurrence or persistent disease Papillary and minimally invasive follicular thyroid cancers Initial total or near-total thyroidectomy followed by 131I ablation No uptake outside the thyroid bed post-ablation Undetectable Tg (1ug/L) on TSH suppression Absence of anti -Tg antibodies
Patients excluded:Distant metastases at presentation Extensive disease pT4 Poorly differentiated tumours Incomplete surgery Anti -Tg antibodies
* Stimulated –Tg, rhTSH on TSH suppression Days 1 and 2 – 0.9 mgs im injections, Day 5 – Tg (72 hrs after last injection) TSH day 3 measurement
Figure 17.6b Follow-up algorithm: differentiated thyroid cancer, low-risk patients. Tg, thyroglobulin; MRI, magnetic resonance imaging; CT, computerized tomography; US, ultrasound; TSH, thyroid-stimulating hormone; 18FDG-PET/CT, 18-fluorodeoxyglucose positron emission tomography/computerized tomography
permitted level. Replacement thyroid hormone is then commenced in the form of T3 20 mcg three times a day. Blood is also taken on day 6 to measure the protein-bound 131 I level (PBI). Historically, ablation success was determined by a diagnostic scan performed with 74–185 MBq of 131 I at 6–12 months,73* and was defined by no visible uptake in the thyroid bed or uptake of less than 0.1 per cent
above background in association with an undetectable Tg. This is no longer universally practised, except in patients with anti-Tg antibodies, when serum Tg becomes invalid. The current recommended criterion for successful ablation is an undetectable serum Tg (1 μg/L) following recombinant TSH (rhTSH) stimulation (in association with a negative neck ultrasound) at 9–12 months.74 Provided these
Treatment of metastatic disease 421
disease, psychiatric disorders, post-partum, hypopituitarism, and those unable to tolerate prolonged hypothyroidism. Its expense currently limits wider use in the UK, although not in Europe or North America.
FOLLOW-UP
Figure 17.7 Whole-body scan following an ablation dose of 3 GBq 131 I, showing intense uptake in the thyroid bed
criteria are met and there are no adverse features, no further treatment is required (Fig. 17.6b). The patient is switched to lifelong T4 at an average daily dose of 200 mcg in order to suppress TSH to an undetectable level. If the results prove abnormal, further imaging and treatment are required. If therapeutic radioiodine is indicated, this should be repeated at 6–12-monthly intervals until uptake disappears and the Tg becomes undetectable. To optimize iodine uptake by both residual normal thyroid and cancer, TSH stimulation is necessary and therefore patients should be hypothyroid at the time of 131I administration. Proper preparation is achieved by a low-iodine diet, avoidance of iodine-rich contrast media, and discontinuation of T3 for 14 days (or 28 days for T4) prior to ablation. As an alternative to thyroid hormone withdrawal, patients may be prepared for ablation with rhTSH while remaining on thyroid hormone. The uncertainty over whether ablation rates differ between patients given rhTSH and those prepared with standard withdrawal has largely been resolved. A retrospective study comparing the ablation rate in 42 patients prepared with withdrawal and 45 patients with rhTSH reported comparable success rates (81 per cent versus 84 per cent).75*** The more recent prospective study following ablation with 3.7 GBq confirmed comparable ablation rates (each of 100 per cent) in patients prepared by either rhTSH or withholding thyroid hormone.76** Previously, rhTSH was available only for diagnostic 131I scans and stimulated Tg measurement. This has now changed, with rhTSH licensed in the UK for ablation with activity of 3.7 GBq, although it remains unlicensed for lowdose ablation or 131I therapy.77 It markedly improves the patients’ quality of life, which is otherwise severely impaired during prolonged periods of hypothyroidism. We recommend its routine use for diagnostic purposes and for ablation or therapy, especially in patients unable to produce TSH and those in whom thyroid-hormone withdrawal is medically contraindicated. This includes patients with cardiac
Annual follow-up comprising clinical examination and estimation of free T3, TSH and Tg is essential to ensure normal thyroid function with TSH suppression and to detect recurrent tumour. Early discovery of recurrence is of paramount prognostic significance, for both cure and survival.78* Local or regional relapse develops in 5–20 per cent of patients with differentiated thyroid cancer. Most relapses occur during the early years of follow-up, but may be detected even after 40 years; follow-up should therefore be lifelong. The risk of loco-regional failure relates partly to tumour aggressiveness, being higher with certain histological subtypes (tall cell, columnar cell and diffuse sclerosing papillary variants), poorly differentiated carcinomas, large tumours and lymph-node involvement at presentation. The risk of recurrence is also closely related to the extent of initial treatment, with limited thyroidectomy resulting in a higher recurrence rate than complete thyroidectomy.57*** Recurrence in the thyroid bed or cervical lymph nodes may be discovered by palpation. Ultrasonography or MRI is useful to delineate disease extent. Serum Tg is usually elevated, although it may be undetectable in 20 per cent of patients on T4 who have isolated lymph-node metastasis.79* Whole-body scanning following the administration of 131I, especially high activities, will reveal uptake in 60–80 per cent of patients with lymph-node disease. Surgery is the treatment for loco-regional recurrence, and complete resection should be attempted in all patients who are fit. Even if disease cannot be completely removed, surgical debulking is beneficial and facilitates the subsequent use of radioiodine. If surgical removal would result in unacceptable morbidity or has to be incomplete, both radioiodine treatment and EBRT should be used to control local disease.80*,81* The outcome for patients with locoregional recurrence is closely related to its site, initial prognostic factors and response to treatment. Mortality after local recurrence has been high in most series; a 10-year survival rate of only 60 per cent has been reported.67***
TREATMENT OF METASTATIC DISEASE Distant metastases develop in 5–23 per cent of patients with differentiated thyroid carcinoma, mainly in lung and bone, less frequently in liver and brain. In any individual patient, the long-term result of treatment is unpredictable; an interplay between patient and tumour characteristics seems to determine outcome. Both univariate and multivariate analyses have highlighted the adverse prognostic effect on survival of older age at the time of discovery of metastases.78*,82* Treatment comprises repeated doses of
422 Thyroid
(a)
(b)
Figure 17.8 (a) Whole-body scan following an ablation dose of 3 GBq 131I demonstrating diffuse lung metastases plus uptake in the right side of the neck. (b) Following repeated 131I therapy, the last post-therapy whole-body scan demonstrates complete eradication of tumour
radioiodine. Activities ranging from 3.7 GBq to 11.1 GBq at 3–9-month intervals have been employed; many centres use a dose of 5.5 GBq at 6-monthly intervals (Fig. 17.8). There is no maximum limit to the cumulative 131I dose that can be given to patients with persistent disease, provided that individual doses do not exceed 2 Gy total-body exposure, progressive improvement can be documented and each pre-treatment blood count confirms the absence of bone-marrow damage. A whole-body scan 3 days after iodine administration (by which time the blood background will be negligible) provides scintigraphic assessment of disease, and serial scanning will document response to treatment. Diagnostic scanning using a tracer dose of iodine is not necessary prior to therapy and may have an adverse effect, because tumour stunning by the diagnostic dose may reduce the uptake of therapeutic 131I.83* The value of 123I as a scanning agent to prevent stunning has been confirmed.84* In addition, a significant proportion of patients with residual tumour, as evidenced by an elevated Tg, demonstrate a negative diagnostic scan but uptake can be documented in the posttherapy scan.85*,86* The real benefit of iodine treatment has been questioned; however, at least one large study clearly demonstrated its
independent prognostic benefit on survival. Younger patients with limited-volume disease, mainly in the lungs, who achieve a complete response to radioiodine treatment have been consistently shown to have the best prognosis, with a 15-year survival of 89 per cent.78* In contrast, older patients and those with large metastases or bone involvement are less likely to respond.87* Although distant metastases, particularly in the lung, may remain stable for years, there is evidence that early treatment may affect outcome. Microscopic foci are more radiosensitive; complete response was reported in 82 per cent of patients with uptake in lung metastases not seen on chest radiography but in only 15 per cent of those with visible micronodules or macronodules.88* The radioresistance of large deposits may be due to poor vascularity, resulting in limited and inhomogeneous iodine distribution, or to the appearance of radioresistant clones. Bone lesions demonstrate a low response rate to radioiodine; surgical excision, when possible, or EBRT should be added.89*,90* Surgical resection with curative intent for patients with a solitary deposit not concentrating iodine and those with bulky disease resistant to iodine has achieved a 5-year post-metastasectomy survival of 46 per cent.91* Sometimes metastases persist despite the administration of substantial 131I therapy doses.92 This may be the consequence of rapid turnover of radioiodine in tumour (short effective half-life) with discharge before adequate energy has been deposited. The effective half-life in metastases responding to therapy has been shown to be more than twice as long as in those not responding: 5.5 days compared with 2.5 days.93* Lithium carbonate can prolong the biological halflife of 131I within thyroid tumours without increasing wholebody radiation exposure.94* However, blood levels need to be monitored closely to avoid toxicity, and the need for psychiatric expertise to be available makes its routine use impracticable. In a retrospective review of 400 patients, FDG-PET positivity was a strong adverse predictor of survival on multivariate analysis (in addition to age), suggesting that these patients should be treated more aggressively.95*
DOSIMETRY OF 131I THERAPY Historically, the use of radioiodine has been empirical. Fixed activities of 1–3.7 GBq for remnant ablation and 3.7–7.5 GBq for therapy are still administered based on experience and likely side effects. However, measurement of the absorbed dose (Gy) has several advantages.96,97 One is that patients are not over-treated and their overall radiation exposure is kept as low as possible. Second, it is the only way to determine whether further 131I therapy will be effective, so that alternative treatment can be considered in unsuccessful cases. But the most important reason for basing iodine therapy on lesion dosimetry is that optimizing the administered dose gives the highest probability that the lesion will be eradicated. Because current information suggests that a stunning effect occurs with incomplete or inadequate therapy and may be permanent, the most effective
Complications of radioiodine treatment 423
Radiation absorbed dose (Gy)
1000
100
10
1 NR
PR Patient clinical response
strategy is to attempt to eradicate a tumour with either a single 131I administration or as few treatments as possible. Over the last decade attempts have been made to calculate the radiation absorbed dose by thyroid remnants and metastatic deposits.69*,72* In order to calculate the radiation dose (D), three parameters must be determined: the initial activity in the target tissue (Ao), the effective half-life of the radioiodine (Te) and the mass of tissue (m). We use single photon emission computerized tomography (SPECT) or PET imaging to measure the volume of metabolically active thyroid tissue or tumour, and following iodine treatment, perform sequential quantitative scans from which time–activity curves can be produced. By fitting the data and extrapolating to the time of administration, the initial activity in the target tissue and the effective half-life of iodine are determined. Calculations are then performed using the Medical Internal Radiation Dosimetry (MIRD) formula: D 0.16AoTe/m. Preliminary analysis of 25 dosimetry studies in patients with metastatic lesions showed a wide variation in radiation absorbed dose (5–621 Gy) from a fixed administered 131 I activity of 5.5 GBq.98 There was evidence of a dose–response relationship clearly explaining the spectrum of clinical response (Fig. 17.9). However, MIRD dosimetry calculations are based on two major assumptions: that radioactivity is uniformly distributed throughout the tumour, and that washout of 131I is governed by a single exponential function. If either of these assumptions is inaccurate, errors will be introduced into the dosimetry estimates. In addition, errors on each parameter (percentage uptake, target activity, half-life and mass) will contribute to a combined error of absorbed dose.99 Given all the problems with dosimetry and the potential for large errors, it could be questioned whether trying to perform dose calculations is worthwhile. With current efforts to produce accurate
CR
Figure 17.9 Dose–response relationship in radioiodine therapy for patients with metastatic thyroid cancer. NR, no response; PR, partial response; CR, complete response
sequential registered three-dimensional SPECT images and dose–volume histograms of therapy distributions, a greater level of accuracy may be achieved eventually, resulting in improved effectiveness of treatment.
COMPLICATIONS OF RADIOIODINE TREATMENT Radioiodine therapy is well tolerated, with a few patients experiencing mild nausea within the first 24 hours postadministration. Radiation thyroiditis may occur in the first week following ablation and is characterized by pain, swelling and localized tenderness in the neck. Symptoms may be severe if there is a large thyroid remnant, but respond to steroid treatment. Acute sialadenitis affecting the parotid or submandibular glands occasionally occurs within 48 hours of administration and may last a few days. A liberal fluid intake and frequent use of lozenges after the first 24 hours should be routine to reduce salivary uptake and limit this reaction. Intravenous administration of amifostine has been reported to result in reduced salivary uptake.100* Sialadenitis may persist into a chronic phase, with episodes recurring over years; about 70 per cent of patients showed a significant decrease in salivary function, which can result in xerostomia and appeared to be dose related.101* Persistent painful masses may require excision. Most patients demonstrate a transient slight reduction in platelet and white cell counts (lymphopenia) after 131I therapy. It is of no clinical importance. These effects reach a nadir at 4–6 weeks after therapy, with recovery in the majority within 6 months. Myelodysplasia leading to aplastic anaemia is rare and likely to occur only in patients with extensive bone metastases who have received a high cumulative dose in excess of 2 Gy per treatment.102* Acute radiation pneumonitis and chronic pulmonary fibrosis have been reported in patients with diffuse functioning lung metastases following single therapeutic activities exceeding 9 GBq.103* A 6–12-month interval between iodine doses may reduce
424 Thyroid
the risk of this complication. If serial lung function tests indicate early damage, future doses can be fractionated. Because differentiated thyroid cancer occurs in women of childbearing age and young men, the possibility that iodine treatment may affect fertility has created concern. A temporary increase in follicular stimulating hormone (FSH) levels has been noted following 131I treatment in both male and female patients, indicating temporary gonadal dysfunction. A positive correlation between FSH levels and the cumulative activity of iodine has been also reported.104* In a small prospective study using thermoluminescent dosimetry, the radiation absorbed dose to the testes was found to be relatively low: 5.4–9.8 cGy and 12–19.2 cGy following the administration of 3 GBq and 5.5 GBq respectively.105* Regarding female patients, no significant difference was observed in fertility rates, birth rates or prematurity among women treated with radioiodine and those not treated.106*,107* Amongst 406 patients under the age of 40 years, temporary amenorrhoea and minor menstrual irregularities were seen in 20 per cent; 427 normal children were born to 276 women, with only one patient unable to conceive.108* The carcinogenic hazard of 131I in the treatment of differentiated thyroid cancer (DTC) has been the subject of several reports.109* An increased risk of acute myeloid leukaemia was seen in the past, especially in patients receiving a cumulative activity in excess of 40 GBq, although patient numbers were small.110*** A recent multi-centre study involving 6841 thyroid cancer patients has quantified the risk of subsequent second primary malignancy.111* Compared to the general population, an increased risk of 27 per cent was seen. It was estimated that a 3.7 GBq dose of 131 I would result in an excess of 53 solid malignant tumours and 3 leukaemias in 10 000 patients during 10 years of follow-up. In addition, a strong correlation existed between the cumulative activity of radioiodine and the risk of bone, soft-tissue, colorectal and salivary-gland cancers. These results highlight the need to restrict radioiodine treatment to those likely to benefit.
EXTERNAL-BEAM RADIOTHERAPY AND CHEMOTHERAPY The role of EBRT in the management of differentiated thyroid cancer remains controversial because published data are conflicting. In many reports, results are presented with no distinction between prophylactic (adjuvant) postoperative EBRT and the treatment of microscopic or macroscopic residual disease. Due to the rarity of the disease and its long natural history, there are no prospective randomized controlled trials. External-beam radiotherapy does not prevent the simultaneous administration of radioiodine, although 131I should be given first whenever possible, as uptake may be diminshed after radiotherapy and if there is good uptake by tumour, EBRT may become unnecessary. However, 20 per cent of tumours fail to concentrate iodine effectively.
Radiotherapy is not indicated in patients with favourable prognostic features, or in young patients with residual disease demonstrating avid iodine uptake.112*** Patients at higher risk of persistent disease or loco-regional recurrence comprise patients at presentation with a large number of lymph-node metastases (greater than 10), nodes demonstrating extracapsular extension (greater than 3), involvement of central lymph nodes, and tumour size greater than 4 cm.113* Indications for EBRT therefore include macroscopic unresectable residual tumour and microscopic disease or involved excision margins. Adjuvant irradiation is required only in older patients with less differentiated cancers that are unlikely to concentrate radioiodine, especially those with extensive extra-thyroidal spread, extracapsular lymph-node extension or recurrent disease. Farahati et al. suggest that adjuvant EBRT should be restricted to patients older than 40 years with locally advanced tumours (pT4) that are non-iodine avid.114* Treatment improved local control in those aged over 40 years with invasive papillary cancer and lymph-node involvement from 22 per cent to 90 per cent at 10 years (p 0.01). A similar group of patients with follicular cancer did not show any significant benefit. In patients without lymph-node disease, loco-regional recurrence is infrequent, such that EBRT is rarely necessary. In 1985, Tubiana et al. reported 97 patients treated with EBRT after incomplete surgery:81* local recurrence at 15 years was 11 per cent in the irradiated group compared to 23 per cent in those treated with surgery alone. More recently, Tsang et al. reported on 207 patients (155 papillary, 52 follicular) with postoperative residual microscopic disease.115* In papillary carcinoma, those irradiated had a 10-year cause-specific survival of 100 per cent and a local relapse-free rate of 93 per cent, compared to the non-irradiated group with a cause-specific survival of 95 per cent (p 0.038) and a local relapse-free rate of 78 per cent (p 0.01). External-beam radiotherapy did not significantly effect cause-specific survival or local relapse-free rate in follicular tumours. The most plausible explanation is that patients with follicular carcinoma have a worse survival due to haematogenous spread, and any benefit of local treatment may be obscured by this biological pattern of behaviour. The presence of gross inoperable macroscopic disease is another indication for EBRT. In our retrospective study, complete regression was achieved in 37.5 per cent and partial regression in 25 per cent.80* Similarly, Chow et al.116* reported the beneficial effects of EBRT in patients with gross macroscopic residual disease, with an improvement in local control from 24 per cent to 56 per cent at 10 years (p 0.001). Irradiation is also effective for advanced and recurrent inoperable Hürthle-cell carcinoma, claiming a relatively more important role because this tumour takes up iodine less frequently.22 Despite the small study size, the 5-year local recurrence rates from Birmingham UK indicate a possible dose response.117* These were 63 per cent following a dose of less than 50 Gy but only 15 per cent and 18 per cent for doses of
External-beam radiotherapy and chemotherapy 425
50–54 Gy and more than 54 Gy. Most patients had either macroscopic or microscopic residual disease. Our policy is to use EBRT infrequently because high dose is required and side effects, especially oesophagitis, are unavoidable. The phase I target volume comprises both sides of the neck (bilateral deep cervical plus supraclavicular nodes), thyroid bed and superior mediastinum from the level of the hyoid down to the carina, with shielding of the subapical portions of the lungs (Fig. 17.10). Anterior and undercouched fields ensure comprehensive coverage,
Figure 17.10 Radiotherapy technique for thyroid cancer: phase I volume covering the thyroid bed, neck and superior mediastinum
with the patient supine and the neck maximally extended. Lead protection of the submandibular salivary glands is required if the treatment volume needs to extend proximally to the mastoid tips. A Perspex shell is avoided initially so as to achieve skin-sparing, but is then fashioned for the phase II volume, which includes sites of microscopic or macroscopic tumour. We recommend three-dimensional planning and conformal beam shaping assisted by a multileaf collimator (Figs 17.11 and 17.12). The aim is to deliver 60 Gy in 30 daily fractions over 6 weeks using 4–6 MV photons. The phase I prescription should be a mid-plane dose of 46 Gy in 2-Gy daily fractions (maximum spinal cord tolerance), with phase II delivering 14 Gy in seven fractions. Known residual tumour in the region of the thyroid bed or neck nodes may be treated with a small phase III volume, adding 6 Gy in three fractions, provided there is no additional dose to the spinal cord. Intensity-modulated radiotherapy can improve the dose distribution by minimizing the dose to the spinal cord and thus permit dose escalation.118* A brisk cutanoeus erythema is invariable, with radiation oesophagitis requiring liquid analgesia, liberal hydration and adequate dietary intake. Symptoms resolve within 2 weeks after the completion of treatment. Acute laryngitis and dysphonia also resolve completely. Late effects include dysphagia, which may occur months or years later, caused by stricture or motility changes as a result of muscle or nerve damage. Reduction in the length of the oesophagus in the phase II volume minimizes such risks. Due to shielding of the subapical portions of the lungs, apical fibrosis may be visible on chest radiograph but is of no clinical significance. Palliative radiotherapy is indicated for fungating nodes, bleeding, stridor, dysphagia and superior vena caval obstruction due to progressive inoperable disease. Bone metastases causing pain, vertebral involvement threatening the spinal cord, long bone involvement if there is a potential
Figure 17.11 Three-dimensional plan (anterior and pair of antero-oblique fields) for phase II volume covering microscopic disease in the thyroid bed (See Plate Section.)
426 Thyroid
for fracture, and brain metastases should also be treated with palliative radiotherapy. Tumour in the lung or mediastinum can be treated if unresectable. Low dose is inadequate; 35 Gy in 15 fractions is required, or 6 Gy once weekly for up to 4 fractions when the central nervous system is not in field. Experience with chemotherapy in differentiated thyroid cancer is limited by the rarity of tumours not controlled by surgery, radioiodine and EBRT. There is, however, a minority of patients who do not respond to conventional therapy and survive for many years with minimal symptoms. Because the presently available drugs are of limited benefit and cause significant morbidity, chemotherapy is reserved for patients with inoperable progressive and symptomatic disease that fails to concentrate radioiodine.119* Of the several agents investigated, doxorubicin has been the most effective, with response rates of 30–40 per cent. Combination with cisplatin has produced similar response rates but with greater toxicity. Responses are usually partial and of short duration, although worthwhile palliation has been reported.120*** Doxorubicin in combination with EBRT appeared to be effective in some patients with large inoperable tumours.121* Chemotherapy may be beneficial in patients with advanced non-iodine-concentrating thyroid cancer by inducing uptake and allowing subsequent radioiodine therapy.122
leaving residual tumour for treatment by radiation and/or chemotherapy.123* Unfortunately, initial surgery is rarely possible. Anaplastic cancer is the least radiosensitive of all thyroid tumours. Doses of less than 50 Gy given in conventional fractionation are associated with a very low probability of local control; the evidence for control with higher doses is scant and subject to selection bias. Experience with 50–60 Gy administered over 5–6 weeks achieved local response in less than 45 per cent of patients, and 75 per cent still died from local progression.124* There is little effect on survival, and the majority of patients spend a significant period of their remaining lives undergoing treatment and recovering from its toxicity. In an attempt to improve local control, alternative approaches were investigated, including accelerated radiotherapy and chemotherapy. Clinical and some laboratory data on the proliferation rate of high-grade thyroid cancer suggest that the potential tumour doubling time is less than 5 days, making accelerated radiotherapy attractive. Several centres have shown an improvement in local tumour response. In our series of 17 patients treated twice daily, 5 days a week, to a dose of 60.8 Gy in 32 fractions over 20–24 days, significant response was achieved in 59 per cent (including three patients with a complete clinical response). Unfortunately, there was a corresponding increase in treatment toxicity that was unacceptable; grade 3–4 oesophagitis occurred in more than 90 per cent of patients and persisted for several weeks following the cessation of radiotherapy.125 Furthermore, despite success in achieving local control, survival remained poor, with almost all patients dying within 6 months of treatment. Better response rates are reported with combined chemotherapy and radiotherapy, particularly if the latter is delivered in a hyperfractionated schedule, although at the
MANAGEMENT OF ANAPLASTIC CARCINOMA Patients with anaplastic carcinoma present with rapidly progressive local and regional nodal disease. The prognosis is dismal, with a median survival of only 6 months from the original symptom. Local growth results in upper airway and oesophageal obstruction. For this reason, it was thought that maximal surgical debulking should be attempted,
(a)
(b)
Figure 17.12 Three-dimensional reconstruction showing field positions in relation to planning target volume and critical organs: (a) spinal cord; (b) lung apices
Management of anaplastic carcinoma 427
New patient referred with Anaplastic Thyroid Cancer History and clinical examination Haematology, biochemistry, CXR, review histology or FNAC CT neck, mediastinum, lungs, liver and bone scan Predominantly Loco-Regional Tumour Surgery Total resection (not usually possible) (Tracheostomy only if unavoidable)
Accelerated Radiotherapy Anterior and undercouched Phase I 1.67 Gy twice daily MPD: 40 Gy 24# 2.5 weeks outpatient or day ward Therapy CT scan for Phase II (excluding cord) 1.67 Gy twice daily TD: 10 Gy 6# 3 days In-patient for 2 weeks
Surgery Total resection (not usually possible)
Chemotherapy Doxorubicin (or equivalent)
Surgery Total resection (not usually possible)
Predominantly Metastatic Disease Progressive, symptomatic
Localised Radiotherapy Simulator planning
Lung
Bone
6 Gy MPD weekly 1– 4 (excluding cord)
Chemotherapy Doxorubicin or experimental agent to maximal response Other – mediastinum, brain, abdomen 35 Gy MPD 15# 3 weeks
Follow-up monthly ⴞ referral to continuing care
Figure 17.13 New patient referred with anaplastic thyroid cancer. CXR, chest radiograph; FNAC, fine-needle aspiration cytology; CT, computerized tomography; MPD, mid-plane dose; TD, total dose
cost of increased morbidity.126* Doxorubicin is the most effective agent and even low doses in combination with radiation appear to have a synergistic effect. There is a very small number of patients who demonstrate prolonged survival having been rendered free of disease by subsequent surgery. In a study from Sweden, 22 patients were treated with hyperfractionated accelerated radiotherapy (46 Gy in 29 fractions, each of 1.6 Gy, twice daily) concurrently with 20 mg doxorubicin weekly 3, followed by debulking surgery. Despite the patients’ advanced ages (over 60 years) and locally extensive disease, such an aggressive treatment modality was feasible. Local control
was achieved in 77 per cent, and in those undergoing surgery, none demonstrated local failure; 9 per cent of patients survived for more than 2 years.127* Control of local disease is important both for palliation and if there is to be any chance of prolonging survival. Improvements in radiotherapy fractionation schedules and conformal beam shaping could maximize the probability of local control whilst limiting toxicity. Finally, since almost all patients who achieve local control still die from metastatic disease, a more effective systemic treatment awaits discovery. No response was observed in distant metastases in either the Swedish study127* or the French
428 Thyroid
study126* employing doxorubicin (60 mg/m2), cisplatin (90 mg/m2) and local radiation. Experimental agents are under study. Our management algorithm is shown in Figure 17.13.
MANAGEMENT OF MEDULLARY THYROID CANCER Total thyroidectomy with routine dissection of lymph nodes in the central compartment of the neck and sampling of lateral jugular nodes is the optimal surgical treatment for MTC. Total thyroidectomy is indicated because in more than 90 per cent of familial and about 20 per cent of sporadic cases disease is multi-centric and bilateral. Furthermore, the incidence of local recurrence is lower in patients treated by radical surgery.128 Cervical lymph-node involvement at presentation ranges from 15 per cent to 75 per cent. A modified neck dissection with preservation of the sternomastoid muscle, the spinal accessory nerve and the internal jugular vein is indicated if metastatic nodes are found during sampling.129* Bilateral neck dissection has been recommended130* because adequacy of the initial operation is a prerequisite for cure. Excision of mediastinal lymph nodes, if involved, should be attempted. Aggressive surgery is justified by the initial loco-regional extent of disease in most cases. Ideally, calcitonin will fall to an undetectable level postoperatively. Measurement of calcitonin level should be repeated at annual follow-up, together with clinical evaluation. Lifelong thyroxine is prescribed at physiological dose; there is no advantage in TSH suppression. Unfortunately, elevated calcitonin levels often persist following initial operation and may be detected in up to 70 per cent of patients with node involvement.131* The most common sites of disease are nodes in the neck and mediastinum, and distant metastases may involve the liver, lungs and bones. Non-invasive imaging methods for detecting occult disease include ultrasonography, computed tomography (CT) scan or magnetic resonance imaging (MRI) and radionuclide scanning (99 mTc-pentavalent dimercaptosuccinic acid (DMSA), 123I-metaiodobenzylguanidine (mIBG), 111In-octreotide), although none of the latter is specific. We have previously reported a 30 per cent sensitivity of radionuclide imaging in MTC.132 Other methods of investigating recurrent MTC include selective venous catheterization to assay calcitonin levels133* and radioimmunoscintigraphy with monoclonal antibodies such as 131I-anti-carcinoembryonic antigen (CEA).134* More recently, PET has probably become the investigation of first choice.135* The frequent identification of liver metastases by angiography136* explains the low efficacy of lymph-node dissections to render a high calcitonin level undetectable postoperatively. Residual MTC is usually progressive, as reflected by a rise in calcitonin levels over time; a mean annual increase of 117 per cent of the initial value was calculated in 35 of 40 patients137* This progressive increase can continue from
the first postoperative measurement but may not appear until after a long period of stability. In a series from the Mayo Clinic, only 11 of 31 patients with raised calcitonin but negative imaging developed overt recurrent disease when followed for a mean period of 12 years. Re-operation for clinically documented local recurrence did not result in normalization in calcitonin level. However, overall survival at 5 and 10 years was 90 per cent and 86 per cent, respectively.138* Analysis of calcitonin doubling times in 65 patients with abnormal calcitonin levels after total thyroidectomy and bilateral lymph-node dissection by the French Neuroendocrine Tumour Group revealed that all 41 patients with a doubling time of more than 2 years were alive at the end of the study (3–30 years). On univariate analysis, TNM stage and doubling time were significant predictors of survival, but only the latter was significant on multivariate analysis (p 0.002).139* A more aggressive approach towards the localization of residual disease was adopted by Tisell, who performed meticulous 12-hour neck dissections, often removing 40–60 cervical lymph nodes.140* In a series of 11 patients, the calcitonin normalized in four (36 per cent) and dramatically improved in four. However, follow-up was short (2–4.5 years) and there is no evidence that these biochemical improvements translate into a survival advantage.141* This policy is not without risk, with higher complication rates than conventional surgery. In view of this and the long-term survival seen with patients on observation, many advocate close follow-up, with surgery reserved for when clinical recurrence can be documented.25*** The role of postoperative radiotherapy is controversial due to the lack of prospective studies; retrospective series comparing surgery alone with surgery plus radiotherapy are subject to selection bias. Favourable responses in terms of tumour reduction and local control have been reported.142*,143*** At the Institut Gustave-Roussy, the survival of 68 patients treated with surgery alone was similar to that of 59 who received postoperative radiotherapy. However in patients with involved lymph nodes, 5-year survival improved significantly with postoperative radiotherapy, from 36 per cent to 81 per cent.144 In contrast, an adverse effect of radiotherapy was reported from the M.D. Anderson Cancer Center, Houston, USA; survival was significantly worse for 24 patients given postoperative radiotherapy compared with 39 age-matched and disease-matched patients treated with surgery alone.145* We recommend the use of adjuvant radiotherapy in patients with locally advanced disease at presentation, and multiple involved lymph nodes in particular, who have persistently elevated calcitonin levels postoperatively, indicating microscopic residual disease, as shown in Figure 17.14.146* Radiotherapy should be also considered for patients with bulky inoperable tumours, for whom significant palliation can be achieved with doses of 60 Gy in 6 weeks; occasionally, subsequent surgery becomes possible. Palliative radiotherapy also has a role in inoperable mediastinal masses and painful bone metastases.
Management of medullary thyroid cancer 429
New patient referred with Medullary Thyroid Carcinoma Partial thyroidectomy ⴞ Lymph node biopsy History and clinical examination Review histology Calcitonin and CEA CT neck, superior mediastinum
Exclude hyperparathyroidism serum Ca, AlkP, Phosphate / PTH
Exclude phaeochromocytoma BP Urine serum catecholamines
Exclude familial MTC Family history Refer to genetics clinic
Completion thyroidectomy ⴞ Modified neck dissection ⴙ Central compartment resection Radiotherapy if extensive disease
Normal post-op calcitonin
Elevated post-op calcitonin
Annual clinical follow-up Calcitonin CEA
Determine site of tumour CT neck, chest, abdomen Bone scan,18FDG-PET 123I mIBG, 111In-octreotide Inoperable
Excision ⴞ Radiotherapy Normal post-op calcitonin Annual clinical follow-up
Positive mIBG or octreotide
Pre-treatment dosimetry Therapeutic 131I mIBG or 90Y-octreotide
Repeat after 6/12
Negative imaging Progressive ⴙ Symptomatic
Chemotherapy ⴞ Radiotherapy Somatostatin analogues (octreotide) Pain relief, continuing care
Asymptomatic
Follow-up only
Figure 17.14 New patient referred with medullary thyroid carcinoma. CEA, carcinoembryonic antigen; CT, computerized tomography; MTC, medullary thyroid cancer; Ca, serum calcium; AlkP, serum alkaline phosphatase; PTH, parathyroid hormone; 99 mTc-DMSA, 99 mTc pentavalent dimercaptosuccinic acid; 18FDG PET, 18F-fluorodeoxyglucose positron emission tomography; mIBG, meta-iodobenzylguanidine; 111 In, 111-indium; 90Y, 90-yttrium
Many patients with metastatic medullary cancer survive for years with minimal symptoms and, apart from medications to control diarrhoea, may not require any other treatment. Chemotherapy should be reserved for those with unresectable progressive and symptomatic disease. Doxorubicin produces symptomatic response in about 30 per cent of cases, but most are partial and of short duration.119* The same response rate is obtained when doxorubicin is used alone or in combination with other drugs.147* The selective uptake of 131I-mIBG and 111In-octreotide by 30–50 per cent of medullary cancers has generated interest in their potential use for targeted radiotherapy, although treatment is
relatively ineffective.148*** Treatment with unlabelled somatostatin analogues may be helpful to control severe diarrhoea from metastatic disease, although side effects can be troublesome. Finally, the use of recombinant interferon alpha-2a has not demonstrated significant benefit. The clinical course of MTC varies widely. In our series, overall survival was 72 per cent at 5 years, 56 per cent at 10 years, 40 per cent at 15 years, and 30 per cent at 20 years.149* Patients with MEN IIb have the most aggressive tumours, often with early development of metastases and death. On multivariate analysis, older age at diagnosis, lymph-node involvement, metastases at presentation and
430 Thyroid
incompleteness of initial surgical resection were significant adverse prognostic factors.149* In some retrospective series, patients with familial cancer have a significantly longer survival compared to those with sporadic cancer;145* however, when patients were matched for age, gender, extent of disease and treatment, this difference disappeared.
Management of familial medullary thyroid cancer (FMTC, MEN IIa and IIb) The RET proto-oncogene is a 21-exon gene on chromosome 10q 11-2 that encodes for a tyrosine kinase transmembrane receptor. The goal of screening for MEN II is to identify gene carriers early in an attempt to modify the outcome of the disease. The two manifestations that are life threatening are MTC and phaeochromocytoma. There is compelling evidence for both that early intervention will improve outcome.150* Genetic testing is the most cost-effective approach to detect affected individuals.151*** All techniques currently use DNA fragments generated by polymerase chain reaction (PCR) amplification of genomic DNA. Several analytic techniques have been applied to detect specific mutations, including direct DNA sequencing, denaturing gradient gel electrophoresis, restriction analysis of amplified products and allele-specific hybridization. Each of these has proved reliable for the detection of the most common mutations causing MEN II or FMTC. Genetic testing should be performed soon after birth. Family members found not to be gene carriers by RET mutation analysis do not require further genetic or biochemical testing and no tests need to be performed on their descendants. Adults who are gene carriers are at high risk of developing MTC; total thyroidectomy with central lymphnode dissection should be performed after exclusion of phaeochromocytoma. For the management of children found to be gene carriers, annual pentagastrin stimulation of calcitonin is performed.152*** The concern related to pentagastrin testing is failure to identify C-cell abnormalities at the earliest stage, since approximately 50 per cent of children had microscopic MTC rather than C-cell hyperplasia.153* Thus, in MEN IIb, total thyroidectomy should be performed as soon as possible, even under 2 years of age.154 Surgery is well tolerated, and the risks of recurrent laryngeal nerve damage or hypoparathyroidism are no greater than in older children.150* In MEN IIa, total thyroidectomy should be performed at 10 years of age or less if the pentagastrin stimulation test is positive. In FMTC, yearly follow-up should be performed, with a pentagastrin stimulation test, with surgery indicated at the first positive test. Annual measurement of urinary catecholamines and metanephrines on a 24-hour specimen provides a straightforward outpatient screening approach to detect phaeochromocytoma. Elevated epinephrine or an elevated epinephrine:norepinephrine ratio is the most commonly observed pattern. Basal or exercise-stimulated plasma catecholamines provides an alternative method of screening.
Magnetic resonance imaging is used to confirm phaeochromocytoma or an enlarged medulla. In most cases abnormalities involve both adrenals, and bilateral adrenalectomy is recommended.25*** The procedure is well tolerated but must be preceded by alpha-blockade and beta-blockade for 7–10 days. Measurement of serum calcium should be performed annually in MEN IIa gene carriers to screen for hyperparathyroidism. Once hypercalcaemia is documented, serum intact parathyroid hormone (PTH) should be measured to confirm the diagnosis. The majority of patients with hyperparathyroidism have diffuse but unequal multiglandular hyperplasia, with only a small proportion (10–15 per cent) having a single adenoma. There is controversy regarding total parathyroidectomy with immediate autotransplantation versus subtotal parathyroidectomy. The value of genetic screening is beyond doubt. However, it is important that family members be counselled regarding the impact of a positive genetic test. A long-term strategy of education and support is recommended.
MANAGEMENT OF THYROID LYMPHOMA Most patients with primary thyroid lymphoma present with confluent cervical/mediastinal lymphadenopathy (stage IIE), but in about one-third, tumour is confined to the thyroid gland (IE). Haematology, biochemistry, CT scan of the neck, thorax and abdomen, and bone-marrow aspirate plus trephine are required for staging. However, these patients are often elderly and may require urgent therapy to relieve airway obstruction, making full staging impracticable until later. Aggressive surgery to debulk the tumour is neither feasible nor necessary. For localized disease, EBRT has been the standard practice for several decades, resulting in 5-year survival rates of approximately 35 per cent. Local bulky disease and gross mediastinal involvement are significantly associated with failure distant from the irradiated volume.155* Chemotherapy for high-grade lymphomas has demonstrated better local and distant disease control, with overall long-term disease-free survival of about 50 per cent. The combination of radiotherapy preceded by chemotherapy has become the standard practice in most institutions and has resulted in 5-year survival rates of 65–90 per cent.156* Six cycles of rituximab plus cyclophosphamide, doxorubicin, vincristine and prednisolone (R-CHOP) given over 4 months are usually recommended.157 Lymphomas showing MALT characteristics usually present as localized extra-nodal tumours without adverse prognostic factors and follow a more indolent course.158* Radiotherapy as single-modality treatment resulted in a complete response rate of almost 100 per cent, a relapse rate of around 30 per cent, a salvage rate of more than 50 per cent, and an overall cause-specific survival of almost 90 per cent at 5 and 10 years.29* Our policy is to treat stage IEA MALT-positive lymphoma with radiotherapy only, but to use combination treatment for all other tumours, as
Future prospects 431
New patient referred with Primary Lymphoma of the Thyroid History and clinical examination Haematology, biochemistry, CXR, review histology CT neck, thorax, abdomen, pelvis Bone marrow aspirate and trephine
Stage IE or IIE
Stage III or IV
Surgery Open biopsy (if not already done) (Tracheostomy only if unavoidable)
MALT Positive IE
MALT Negative
Chemotherapy R - CHOP q21 6 cycles
Radiotherapy Simulate for anterior and undercouched Phase I to neck and mediastinum 2 Gy daily MPD: 40 Gy 20# 4 weeks
Partial response
Complete response
consider additional chemotherapy
Radiotherapy 35 Gy 20# 4 weeks involved field
no additional response
Follow-up clinic 3 monthly Haematology and CXR
Complete remission
Recurrence
Chemotherapy
Figure 17.15 New patient referred with primary lymphoma of the thyroid. CXR, chest radiograph; MALT, mucosa-associated lymphoid tissue; R-CHOP, rituximab, cyclophosphamide, hydroxydaunomycin (doxorubicin), vincristine (Oncovin), and prednisone; MPD, mid-plane dose. Ann Arbor Classification: Stage IE: disease in single lymph-node region; Stage IIE: disease in two or more lymph regions on same side of diaphragm; Stage III: disease in lymph-node regions on both sides of the diaphragm; Stage IV: widespread disease, including multiple involvement at one or more extra-nodal sites, such as the bone marrow
shown in Figure 17.15. Treatment volume includes the neck and superior mediastinum, irradiated by a pair of anterior and undercouched fields to 35–40 Gy in 20 fractions over 4 weeks. Primary Hodgkin’s disease of the thyroid is very rare and treated in a similar fashion to extra-nodal Hodgkin’s at any other site.
FUTURE PROSPECTS A large randomised, prospective, multi-centre trial (HiLo, Cancer Research UK) began recruiting patients in 2007 to compare the efficacy of low-dose (1.1 GBq) versus high-dose (3.7 GBq) remnant ablation prepared with either rhTSH or
432 Thyroid
thyroid hormone withdrawal. This study will serve to address the controversy remaining concerning the optimal activity to achieve successful ablation, but will also evaluate the roles of rhTSH and thyroid hormone withdrawal. Over the past decade, developments in dosimetry have permitted the calculation of absorbed tumour dose in metastatic disease with greater accuracy. This has enabled the construction of dose–response curves that explain the spectrum of clinical response from fixed activities of radioiodine.98 These developments allow for the tumoricidal dose to be calculated and may enable precise prescription of further 131I therapy so as to maximize tumour kill while minimizing toxicity, staff exposure and unnecessary expense. The fundamental problem for recurrent or metastatic iodine-negative disease is the decreased expression of the sodium iodide symporter. In these patients, the options for treatment are limited. Initial reports with redifferentiation to enhance the efficacy of iodine uptake have not been substantiated.159* However, studies using selective agonists of retinoic X receptors have been demonstrated to enhance sodium iodide symporter gene expression in vitro. An alternative approach is the administration of peptide receptor-targeted therapy with 90Y-lanreotide, based on the premise that non iodine-avid tumours may express somatostatin receptors.160* Initial results have been promising and invite further studies. In patients with advanced progressive disease, chemotherapy with doxorubicin is disappointing, yielding poor overall response rates, with success limited by cardiotoxicity. The challenge therefore remains to find an effective agent with minimal cardiac toxicity. This may be possible with the introduction of polyethylene glycolcoated (pegylated) liposomal doxorubicin (Caelyx). A multicentre prospective study is ongoing in the UK (Newcastle, U. Mallick) with liposomal doxorubicin for any advanced metastatic thyroid cancer. Chemotherapy may also be beneficial in patients with non iodine-avid disease by inducing the uptake of subsequent radioiodine therapy. A recent study has reported significant uptake in previously non iodine-avid lesions following cisplatin and doxorubicin.122 The use of monoclonal antibodies against the epidermal growth factor receptor (EGFR) has demonstrated a reduction in tumour growth in thyroid cancer cell lines.161* Phase II studies are in progress evaluating tyrosine kinase inhibitors for non iodine-avid tumours and metastatic MTC.162*** The use of gene therapy using vectors that lead to the expression of proteins in cancer cells that are directly toxic or aid the sensitivity of cancer treatments has been described as suicide gene therapy. Preliminary work has demonstrated success in follicular, anaplastic and MTC cell lines but requires assessment in patients. A further area of interest is associated with the p53 gene that encodes the p53 protein, a regulator of DNA repair. p53 mutations are more common in anaplastic thyroid cancer. However, malignant cells bearing the wild-type p53 mutation are more susceptible to chemotherapy agents compared to other p53 mutations. Re-expression of wild-type p53 in thyroid cancer cell lines
results in growth arrest, but no human studies are currently being performed. A combination of treatments targeting several steps involved in tumour growth may have a synergistic effect and improve outcome. Medullary thyroid cancer is a particularly attractive target for gene therapy. Potential gene targets include the inhibition of oncogene RET signalling, suicide gene therapy, immunotherapy and combination therapy. In animal models, RET signal transduction can be blocked, resulting in the suppression of tumour growth. Suicide gene therapy using gene transfer of herpes simplex virus thymidine kinase in combination with application of the antiviral drug ganciclovir or nitric oxide synthase has been assessed. Immunotherapy with delivery of interleukin 2 or 12 may help support the normal immune response in MTC, but combination treatment appears to be more effective than treatment with any single agent.163*** The British Society of Paediatric Endocrinology and Diabetes (www.bsped.org.uk) has recently published guidelines for the management of thyroid cancer in children; an updated version of the guidelines for adults is now available. The British Thyroid Foundation Cancer Group has also published a patient information leaflet for patients (www.btf-thyroid.org). A national thyroid dataset (UK) has received conditional approval from the NHS Information Authority Standards Board. This should improve patient records, data retrieval and analysis for the future.
KEY LEARNING POINTS ●
●
●
●
●
●
The incidence of thyroid cancer has increased significantly over the past decades. However, mortality rates have fallen, possibly reflecting earlier diagnosis and improved treatment. Fine-needle aspiration cytology represents the gold-standard investigation for evaluating thyroid nodules. Age is the single most important independent predictor of recurrence and survival in differentiated thyroid cancer. Tumour size, grade, extension and the presence of metastases also predict outcome. Surgery remains the initial and potentially curative treatment for differentiated thyroid cancer and medullary carcinoma. Remnant 131I ablation post-thyroidectomy reduces the risk of loco-regional recurrence and cause-specific mortality in differentiated thyroid cancer, particularly in patients with adverse prognostic features. Radioiodine therapy is the mainstay of treatment for metastatic differentiated thyroid cancer: a significant proportion of patients can be cured, and durable palliation can be achieved in others.
References 433
REFERENCES 1 Waterhouse JA. Epidemiology of thyroid cancer. In: Preece PE, Rosen RD, Maran AGD (eds), Head and Neck Oncology for the General Surgeon. London: WB Saunders, 1991, 1–10. 2 dos Santos Silva I, Swerdlow AJ. Thyroid cancer epidemiology in England and Wales: time trends and geographical distribution. Br J Cancer 1993; 67(2):330–40. 3 Office for National Statistics. Cancer Statistics Registrations: Registrations of Cancer Diagnosed in 2001, England. 32. London: Office of National Statistics, 2004. 4 Correa P, Chen VW. Endocrine gland cancer. Cancer 1995; 75(1 Suppl.):338–52. 5 Ron E, Kleinerman RA, Boice JD Jr, et al. A population-based case-control study of thyroid cancer. J Natl Cancer Inst 1987; 79(1):1–12. 6 Schneider AB, Ron E, Lubin J, et al. Dose–response relationships for radiation-induced thyroid cancer and thyroid nodules: evidence for the prolonged effects of radiation on the thyroid. J Clin Endocrinol Metab 1993; 77(2):362–9. 7 Ron E, Lubin JH, Shore RE, et al. Thyroid cancer after exposure to external radiation: a pooled analysis of seven studies. Radiat Res 1995; 141(3):259–77. ◆8 Takeichi N, Ezaki H, Dohi K. A review of forty-five years study of Hiroshima and Nagasaki atomic bomb survivors. Thyroid cancer: reports up to date and a review. J Radiat Res (Tokyo) 1991; 32 Suppl.:180–8. 9 Tucker MA, Jones PH, Boice JD Jr, et al. Therapeutic radiation at a young age is linked to secondary thyroid cancer. The Late Effects Study Group. Cancer Res 1991; 51(11):2885–8. 10 Boice JD Jr, Engholm G, Kleinerman RA, et al. Radiation dose and second cancer risk in patients treated for cancer of the cervix. Radiat Res 1988; 116(1):3–55. ●11 Holm LE, Hall P, Wiklund K, et al. Cancer risk after iodine131 therapy for hyperthyroidism. J Natl Cancer Inst 1991; 83(15):1072–7. ◆12 Becker DV, Robbins J, Beebe GW, et al. Childhood thyroid cancer following the Chernobyl accident: a status report. Endocrinol Metab Clin North Am 1996; 25(1):197–211. 13 Schaller RT Jr, Stevenson JK. Development of carcinoma of the thyroid in iodine-deficient mice. Cancer 1966; 19(8):1063–80. 14 Franceschi S, Levi F, Negri E, et al. Diet and thyroid cancer: a pooled analysis of four European case-control studies. Int J Cancer 1991; 48(3):395–8. 15 Negri E, Dal Maso L, Ron E, et al. A pooled analysis of case-control studies of thyroid cancer. II. Menstrual and reproductive factors. Cancer Causes Control 1999; 10(2):143–55. 16 Hawkins M, Vini L, Houlston R, Harmer C. Familial differentiated thyroid cancer. Br J Cancer 1999; 8:112. 17 Xing M, Westra WH, Tufano RP, et al. BRAF mutation predicts a poorer clinical prognosis for papillary thyroid cancer. J Clin Endocrinol Metab 2005; 90(12):6373–9.
18 Livolsi VA, Merino MJ. Histopathologic differential diagnosis of the thyroid. Pathol Annu 1981; 16(Pt 2):357–406. ●19 Hedinger C, Williams E, Sobin L. World Health Organization: histological typing of thyroid tumours. Berlin: SpringerVerlag, 1988, 9–11. ◆20 Bramley MD, Harrison BJ. Papillary microcarcinoma of the thyroid gland. Br J Surg 1996; 83(12):1674–83. 21 Bowen J, Haq M, Rhys-Evans P, Harmer C. Hürthle cell carcinoma: prognostic factors affecting survival. Clin Oncol 2004; 16(7):S29. 22 Vini L, Fisher C, A’Hern R, Harmer C. Hürthle cell cancer of the thyroid: the Royal Marsden experience. Thyroid 1998; 8(12):1228. 23 Tan RK, Finley RK III, Driscoll D, et al. Anaplastic carcinoma of the thyroid: a 24-year experience. Head Neck 1995; 17(1):41–7. ●24 Rosai J, Carcangiu M, DeLellis R. Tumours of the thyroid gland. In: Atlas of Tumour Pathology. Washington DC: Armed Forces Institute of Pathology, 1992. ◆25 Heshmati HM, Gharib H, van Heerden JA, Sizemore GW. Advances and controversies in the diagnosis and management of medullary thyroid carcinoma. Am J Med 1997; 103(1):60–9. 26 Hyer SL, Newbold K, Harmer C. Familial medullary thyroid cancer: clinical aspects and prognosis. Eur J Surg Oncol 2005; 31(4):415–19. 27 Ponder B. Multiple endocrine neoplasia type 2. BMJ 1990; 300(6723):484–5. 28 Harrington KJ, Michalaki VJ, Vini L, et al. Management of non-Hodgkin’s lymphoma of the thyroid: the Royal Marsden Hospital experience. Br J Radiol 2005; 78(929):405–10. ●29 Laing RW, Hoskin P, Hudson BV, et al. The significance of MALT histology in thyroid lymphoma: a review of patients from the BNLI and Royal Marsden Hospital. Clin Oncol (R Coll Radiol) 1994; 6(5):300–4. 30 Mitra A, Fisher C, Rhys-Evans P, Harmer C. Liposarcoma of the thyroid. Sarcoma 2004; 8:91–6. 31 Wood K, Vini L, Harmer C. Metastases to the thyroid gland: the Royal Marsden experience. Eur J Surg Oncol 2004; 30(6):583–8. 32 Mazzaferri EL. Thyroid cancer in thyroid nodules: finding a needle in the haystack. Am J Med 1992; 93(4):359–62. 33 Shimamoto K, Endo T, Ishigaki T, et al. Thyroid nodules: evaluation with color Doppler ultrasonography. J Ultrasound Med 1993; 12(11):673–8. 34 Hamburger JI. Diagnosis of thyroid nodules by fine needle biopsy: use and abuse. J Clin Endocrinol Metab 1994; 79(2):335–9. 35 Sisson JC. Selection of the optimal scanning agent for thyroid cancer. Thyroid 1997; 7(2):295–302. 36 Mazzaferri EL. Management of a solitary thyroid nodule. N Engl J Med 1993; 328(8):553–9. 37 Ramanna L, Waxman A, Braunstein G. Thallium-201 scintigraphy in differentiated thyroid cancer: comparison with radioiodine scintigraphy and serum thyroglobulin determinations. J Nucl Med 1991; 32(3):441–6.
434 Thyroid
38 Tenenbaum F, Lumbroso J, Schlumberger M, et al. Radiolabeled somatostatin analog scintigraphy in differentiated thyroid carcinoma. J Nucl Med 1995; 36(5):807–10. 39 Krenning EP, Kwekkeboom DJ, Bakker WH, et al. Somatostatin receptor scintigraphy with [111In-DTPA-D-Phe1]- and [123I-Tyr3]-octreotide: the Rotterdam experience with more than 1000 patients. Eur J Nucl Med 1993; 20(8):716–31. 40 Takashima S, Sone S, Takayama F, et al. Papillary thyroid carcinoma: MR diagnosis of lymph node metastasis. Am J Neuroradiol 1998; 19(3):509–13. ◆41 Vini L, Harmer C. Management of thyroid cancer. Lancet Oncol 2002; 3(7):407–14. ●42 British Thyroid Association (ed.). Guidelines for the Management of Thyroid Cancer in Adults. London: Royal College of Physicians, 2002. 43 Vini L, A’Hern R, Fisher C, Harmer C. Differentiated thyroid cancer: the Royal Marsden experience. Br J Cancer 1999; 8(Suppl. 2):112. 44 Cady B, Rossi R. An expanded view of risk-group definition in differentiated thyroid carcinoma. Surgery 1988; 104(6):947–53. 45 Loree TR. Therapeutic implications of prognostic factors in differentiated carcinoma of the thyroid gland. Semin Surg Oncol 1995; 11(3):246–55. ●46 Hay ID, Grant CS, van Heerden JA, et al. Papillary thyroid microcarcinoma: a study of 535 cases observed in a 50-year period. Surgery 1992; 112(6):1139–46. ●47 Vini L, Hyer SL, Marshall J, et al. Long-term results in elderly patients with differentiated thyroid carcinoma. Cancer 2003; 97(11):2736–42. 48 Rhys-Evans P, See A, Harmer C. Cancer of the thyroid gland. In: Rhys-Evans P, Montgomery P, Gullane P (eds), Principles and Practice of Head and Neck Oncology. London: Martin Dunitz, 2003, 405–30. 49 Hay ID, Grant CS, Taylor WF, McConahey WM. Ipsilateral lobectomy versus bilateral lobar resection in papillary thyroid carcinoma: a retrospective analysis of surgical outcome using a novel prognostic scoring system. Surgery 1987; 102(6):1088–95. ●50 Mazzaferri EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer. Am J Med 1994; 97(5):418–28. 51 Ballantyne AJ. Resections of the upper aerodigestive tract for locally invasive thyroid cancer. Am J Surg 1994; 168(6):636–9. 52 Pereira JA, Jimeno J, Miquel J, et al. Nodal yield, morbidity, and recurrence after central neck dissection for papillary thyroid carcinoma. Surgery 2005; 138(6):1095–101. 53 Ladas G, Rhys-Evans PH, Goldstraw P. Anterior cervicaltranssternal approach for resection of benign tumors at the thoracic inlet. Ann Thorac Surg 1999; 67(3): 785–9. 54 Noguchi S, Murakami N, Kawamoto H. Classification of papillary cancer of the thyroid based on prognosis. World J Surg 1994; 18(4):552–7.
55 Shaha AR. Management of the neck in thyroid cancer. Otolaryngol Clin North Am 1998; 31(5):823–31. 56 Soh EY, Sobhi SA, Wong MG, et al. Thyroid-stimulating hormone promotes the secretion of vascular endothelial growth factor in thyroid cancer cell lines. Surgery 1996; 120(6):944–7. 57 Mazzaferri EL. Papillary thyroid carcinoma: factors influencing prognosis and current therapy. Semin Oncol 1987; 14(3):315–32. 58 Liewendahl K, Helenius T, Lamberg BA, et al. Free thyroxine, free triiodothyronine, and thyrotropin concentrations in hypothyroid and thyroid carcinoma patients receiving thyroxine therapy. Acta Endocrinol 1987; 116(3):418–24. 59 Giannini S, Nobile M, Sartori L, et al. Bone density and mineral metabolism in thyroidectomized patients treated with long-term L-thyroxine. Clin Sci 1994; 87(5):593–7. 60 Leese GP, Jung RT, Guthrie C, et al. Morbidity in patients on L-thyroxine: a comparison of those with a normal TSH to those with a suppressed TSH. Clin Endocrinol 1992; 37(6):500–3. 61 Schlumberger M, Fragu P, Parmentier C, Tubiana M. Thyroglobulin assay in the follow-up of patients with differentiated thyroid carcinomas: comparison of its value in patients with or without normal residual tissue. Acta Endocrinol 1981; 98(2):215–21. 62 Vini L, Harmer C. Radioiodine treatment for differentiated thyroid cancer. Clin Oncol (R Coll Radiol) 2000; 12(6):365–72. 63 Taylor T, Specker B, Robbins J, et al. Outcome after treatment of high-risk papillary and non-Hurthle-cell follicular thyroid carcinoma. Ann Intern Med 1998; 129(8):622–7. 64 Beierwaltes WH, Rabbani R, Dmuchowski C, et al. An analysis of ‘ablation of thyroid remnants’ with I-131 in 511 patients from 1947–1984: experience at University of Michigan. J Nucl Med 1984; 25(12):1287–93. ●65 McCowen KD, Adler RA, Ghaed N, et al. Low dose radioiodide thyroid ablation in postsurgical patients with thyroid cancer. Am J Med 1976; 61(1):52–8. 66 Hackshaw A, Mallick U, Harmer C, Haq M, Franklyn J. 131I activity for remnant ablation in patients with differentiated thyroid cancer: a systematic review. J. Clin. End. Metab. 2007:92(1);28–38. 67 Mazzaferri EL. An overview of the management of papillary and follicular thyroid carcinoma. Thyroid 1999; 9(5):421–7. 68 Bal C, Padhy AK, Jana S, et al. Prospective randomized clinical trial to evaluate the optimal dose of 131-I for remnant ablation in patients with differentiated thyroid carcinoma. Cancer 1996; 77(12):2574–80. 69 Maxon HR III, Englaro EE, Thomas SR, et al. Radioiodine-131 therapy for well-differentiated thyroid cancer – a quantitative radiation dosimetric approach: outcome and validation in 85 patients. J Nucl Med 1992; 33(6):1132–6. 70 Doi SA, Woodhouse NJ. Ablation of the thyroid remnant and 131 I dose in differentiated thyroid cancer. Clin Endocrinol 2000; 52(6):765–73.
References 435
71 Harmer CL, McCready VR. Thyroid cancer: differentiated carcinoma. Cancer Treat Rev 1996; 22(3):161–77. ●72 O’Connell ME, Flower MA, Hinton PJ, et al. Radiation dose assessment in radioiodine therapy. Dose–response relationships in differentiated thyroid carcinoma using quantitative scanning and PET. Radiother Oncol 1993; 28(1):16–26. 73 Taylor H, Hyer S, Vini L, et al. Diagnostic 131I whole body scanning after thyroidectomy and ablation for differentiated thyroid cancer. Eur J Endocrinol 2004; 150(5):649–53. 74 Pacini F, Schlumberger M, Harmer C, et al. Post-surgical use of radioiodine (131I) in patients with papillary and follicular thyroid cancer and the issue of remnant ablation: a consensus report. Eur J Endocrinol 2005; 153(5):651–9. ◆75 Robbins RJ, Larson SM, Sinha N, et al. A retrospective review of the effectiveness of recombinant human TSH as a preparation for radioiodine thyroid remnant ablation. J Nucl Med 2002; 43(11):1482–8. 76 Pacini F, Ladenson PW, Schlumberger M, et al. Radioiodine ablation of thyroid remnants after preparation with recombinant human thyrotropin in differentiated thyroid carcinoma: results of an international, randomized, controlled study. J Clin Endocrinol Metab 2006; 91 (3);926–32. 77 Haq M, Harmer C. Advances in follow-up and radioiodine preparation for patients with differentiated thyroid cancer. British Thyroid Foundation Newsletter 2004; 50:4. ●78 Schlumberger M, Challeton C, de Vathaire F, et al. Radioactive iodine treatment and external radiotherapy for lung and bone metastases from thyroid carcinoma. J Nucl Med 1996; 37(4):598–605. 79 Schlumberger M, Baudin E. Serum thyroglobulin determination in the follow-up of patients with differentiated thyroid carcinoma. Eur J Endocrinol 1998; 138(3):249–52. 80 O’Connell ME, A’Hern RP, Harmer CL. Results of external beam radiotherapy in differentiated thyroid carcinoma: a retrospective study from the Royal Marsden Hospital. Eur J Cancer 1994; 30A(6):733–9. 81 Haq M, Harmer C. Non-surgical management of thyroid cancer. In: Mazzaferri E, Harmer C, Mallick U, Kendall-Taylor P (eds), Practical Management of Thyroid Cancer. London: Springer-Verlag, 2006;171–191. ●82 Haq M, Harmer C. Differentiated thyroid carcinoma with distant metastases at presentation: prognostic factors and outcome. Clin Endocrinol 2005; 63(1):87–93. ●83 Park HM, Perkins OW, Edmondson JW, et al. Influence of diagnostic radioiodines on the uptake of ablative dose of iodine-131. Thyroid 1994; 4(1):49–54. 84 Siddiqi A, Foley RR, Britton KE, et al. The role of 123 I-diagnostic imaging in the follow-up of patients with differentiated thyroid carcinoma as compared to 131 I-scanning: avoidance of negative therapeutic uptake due to stunning. Clin Endocrinol 2001; 55(4):515–21. 85 Fatourechi V, Hay ID, Mullan BP, et al. Are post-therapy radioiodine scans informative and do they influence subsequent therapy of patients with differentiated thyroid cancer? Thyroid 2000; 10(7):573–7.
86 Pineda JD, Lee T, Ain K, et al. Iodine-131 therapy for thyroid cancer patients with elevated thyroglobulin and negative diagnostic scan. J Clin Endocrinol Metab 1995; 80(5):1488–92. 87 Haq M, McCready R, Harmer C. Treatment of advanced differentiated thyroid carcinoma with high activity radioiodine therapy. Nucl Med Commun 2004; 25(8):799–805. 88 Maxon HR, Thomas SR, Samaratunga RC. Dosimetric considerations in the radioiodine treatment of macrometastases and micrometastases from differentiated thyroid cancer. Thyroid 1997; 7(2):183–7. 89 Marcocci C, Pacini F, Elisei R, et al. Clinical and biologic behavior of bone metastases from differentiated thyroid carcinoma. Surgery 1989; 106(6):960–6. 90 Niederle B, Roka R, Schemper M, et al. Surgical treatment of distant metastases in differentiated thyroid cancer: indication and results. Surgery 1986; 100(6):1088–97. ●91 Vini L, Harmer C, Goldstraw P. The role of metastasectomy in differentiated thyroid cancer. Eur J Surg Oncol 1998; 24:348. 92 Wechalekar K, Haq M, Harmer C, Cook G. Metastatic thyroid carcinoma causing superior vena caval obstruction diagnosed on I-131 scan. Clin Nucl Med 2005; 30(8):548–9. ●93 Maxon HR, Thomas SR, Hertzberg VS, et al. Relation between effective radiation dose and outcome of radioiodine therapy for thyroid cancer. N Engl J Med 1983; 309(16):937–41. 94 Koong SS, Reynolds JC, Movius EG, et al. Lithium as a potential adjuvant to 131I therapy of metastatic, well differentiated thyroid carcinoma. J Clin Endocrinol Metab 1999; 84(3):912–16. 95 Robbins RJ, Wan Q, Grewal RK, et al. Real-time prognosis for metastatic thyroid carcinoma based on FDG-PET scanning. J Clin Endocrinol Metab 2006; 91(2): 498–505. 96 Haq M, Pratt B, Harmer C, Flux G. Whole-body, blood and patient-specific dosimetry of radioiodine in the treatment of differentiated thyroid cancer. Eur J Nuc Med 2004; 31(2):P877. 97 Haq M, Gear J, Flux G, Harmer C. Whole-body and blood dosimetry for radioiodine treatment of differentiated thyroid cancer. Nucl Med Comm 2004; 25(3):314. 98 Vini L, Chittenden S, Pratt B, Harmer C. In vivo dosimetry of radioiodine in patients with metastatic differentiated thyroid cancer. J Nucl Med 1998; 25:904. 99 Flower MA, McCready VR. Radionuclide therapy dose calculations: what accuracy can be achieved? Eur J Nucl Med 1997; 24(12):1462–4. 100 Bohuslavizki KH, Klutmann S, Brenner W, et al. Salivary gland protection by amifostine in high-dose radioiodine treatment: results of a double-blind placebo-controlled study. J Clin Oncol 1998; 16(11):3542–9. 101 Hyer S, Kong A, Pratt B, Harmer C. Salivary gland toxicity after radioiodine therapy for thyroid cancer. Clin Oncol. 2007; 19(1):83–6.
436 Thyroid
●102
103
104
105
106
107
●108
109
110
●111
◆112
113
●114
●115
116
Benua R, Cicale NR, Sonenberg M, Rawson RW. The relation of radioiodine dosimetry to results and complications in the treatment of metastatic thyroid cancer. Am J Roentgenol Radium Ther Nucl Med 1962; 87:171–82. Rall JE, Alpers JB, Lewallen CG, et al. Radiation pneumonitis and fibrosis: a complication of radioiodine treatment of pulmonary metastases from cancer of the thyroid. J Clin Endocrinol Metab 1957; 17(11): 1263–76. Pacini F, Gasperi M, Fugazzola L, et al. Testicular function in patients with differentiated thyroid carcinoma treated with radioiodine. J Nucl Med 1994; 35(9):1418–22. Hyer S, Vini L, O’Connell M, et al. Testicular dose and fertility in men following I(131) therapy for thyroid cancer. Clin Endocrinol 2002; 56(6):755–8. Dottorini ME, Lomuscio G, Mazzucchelli L, et al. Assessment of female fertility and carcinogenesis after iodine-131 therapy for differentiated thyroid carcinoma. J Nucl Med 1995; 36(1):21–7. Schlumberger M, de Vathaire F, Ceccarelli C, et al. Exposure to radioactive iodine-131 for scintigraphy or therapy does not preclude pregnancy in thyroid cancer patients. J Nucl Med 1996; 37(4):606–12. Vini L, Hyer S, Al Saadi A, et al. Prognosis for fertility and ovarian function after treatment with radioiodine for thyroid cancer. Postgrad Med J 2002; 78(916):92–3. Edmonds CJ, Smith T. The long-term hazards of the treatment of thyroid cancer with radioiodine. Br J Radiol 1986; 59(697):45–51. Maxon HR III, Smith HS. Radioiodine-131 in the diagnosis and treatment of metastatic well differentiated thyroid cancer. Endocrinol Metab Clin North Am 1990; 19(3):685–718. Rubino C, de Vathaire F, Dottorini ME, et al. Second primary malignancies in thyroid cancer patients. Br J Cancer 2003; 89(9):1638–44. Harmer C, Bidmead M, Shepherd S, et al. Radiotherapy planning techniques for thyroid cancer. Br J Radiol 1998; 71(850):1069–75. Leboulleux S, Rubino C, Baudin E, et al. Prognostic factors for persistent or recurrent disease of papillary thyroid carcinoma with neck lymph node metastases and/or tumor extension beyond the thyroid capsule at initial diagnosis. J Clin Endocrinol Metab 2005; 90(10):5723–9. Farahati J, Reiners C, Stuschke M, et al. Differentiated thyroid cancer. Impact of adjuvant external radiotherapy in patients with peri-thyroidal tumor infiltration (stage pT4). Cancer 1996; 77(1):172–80. Tsang RW, Brierley JD, Simpson WJ, et al. The effects of surgery, radioiodine, and external radiation therapy on the clinical outcome of patients with differentiated thyroid carcinoma. Cancer 1998; 82(2):375–88. Chow SM, Law SC, Mendenhall WM, et al. Papillary thyroid carcinoma: prognostic factors and the role of radioiodine and external radiotherapy. Int J Radiat Oncol Biol Phys 2002; 52(3):784–95.
117 Ford D, Giridharan S, McConkey C, et al. External beam radiotherapy in the management of differentiated thyroid cancer. Clin Oncol (R Coll Radiol) 2003; 15(6):337–41. ●118 Nutting CM, Convery DJ, Cosgrove VP, et al. Improvements in target coverage and reduced spinal cord irradiation using intensity-modulated radiotherapy (IMRT) in patients with carcinoma of the thyroid gland. Radiother Oncol 2001; 60(2):173–80. 119 Hoskin PJ, Harmer C. Chemotherapy for thyroid cancer. Radiother Oncol 1987; 10(3):187–94. 120 Ahuja S, Ernst H. Chemotherapy of thyroid carcinoma. J Endocrinol Invest 1987; 10(3):303–10. 121 Kim JH, Leeper RD. Treatment of locally advanced thyroid carcinoma with combination doxorubicin and radiation therapy. Cancer 1987; 60(10):2372–5. 122 Morris JC, Kim CK, Padilla ML, Mechanick JI. Conversion of non-iodine-concentrating differentiated thyroid carcinoma metastases into iodine-concentrating foci after anticancer chemotherapy. Thyroid 1997; 7(1):63–6. 123 McIver B, Hay ID, Giuffrida DF, et al. Anaplastic thyroid carcinoma: a 50-year experience at a single institution. Surgery 2001; 130(6):1028–34. 124 Levendag PC, De Porre PM, van Putten WL. Anaplastic carcinoma of the thyroid gland treated by radiation therapy. Int J Radiat Oncol Biol Phys 1993; 26(1): 125–8. 125 Mitchell G, Huddart R, Harmer C. Phase II evaluation of high dose accelerated radiotherapy for anaplastic thyroid carcinoma. Radiother Oncol 1999; 50(1):33–8. 126 Schlumberger M, Parmentier C, Delisle MJ, et al. Combination therapy for anaplastic giant cell thyroid carcinoma. Cancer 1991; 67(3):564–6. ●127 Tennvall J, Lundell G, Wahlberg P, et al. Anaplastic thyroid carcinoma: three protocols combining doxorubicin, hyperfractionated radiotherapy and surgery. Br J Cancer 2002; 86(12):1848–53. 128 Modigliani E, Franc B, Niccoli-sire P. Diagnosis and treatment of medullary thyroid cancer. Baillieres Best Pract Res Clin Endocrinol Metab 2000; 14(4):631–49. ●129 Dralle H, Scheumann GF, Proye C, et al. The value of lymph node dissection in hereditary medullary thyroid carcinoma: a retrospective, European, multicentre study. J Intern Med 1995; 238(4):357–61. 130 Scollo C, Baudin E, Travagli JP, et al. Rationale for central and bilateral lymph node dissection in sporadic and hereditary medullary thyroid cancer. J Clin Endocrinol Metab 2003; 88(5):2070–5. 131 Block MA, Jackson CE, Tashjian AH Jr. Management of occult medullary thyroid carcinoma: evidenced only by serum calcitonin level elevations after apparently adequate neck operations. Arch Surg 1978; 113(4):368–72. 132 Vini L, Al-Saadi A, Pratt B, Harmer C. The role of radionuclide imaging (V-DMSA, 131I-mIBG, 111In-Octreotide) in medullary thyroid cancer. Nucl Med Commun 1998; 19:384. 133 Medina-Franco H, Herrera MF, Lopez G, et al. Persistent hypercalcitoninemia in patients with medullary thyroid cancer: a therapeutic approach based on selective venous
References 437
134
●135
136
137
138
139
140
141
142
◆143
144
●145
146
147
sampling for calcitonin. Rev Invest Clin 2001; 53(3):212–17. Juweid M, Sharkey RM, Swayne LC, Goldenberg DM. Improved selection of patients for reoperation for medullary thyroid cancer by imaging with radiolabeled anticarcinoembryonic antigen antibodies. Surgery 1997; 122(6):1156–65. Diehl M, Risse JH, Brandt-Mainz K, et al. Fluorine-18 fluorodeoxyglucose positron emission tomography in medullary thyroid cancer: results of a multicentre study. Eur J Nucl Med 2001; 28(11):1671–6. Szavcsur P, Godeny M, Bajzik G, et al. Angiography-proven liver metastases explain low efficacy of lymph node dissections in medullary thyroid cancer patients. Eur J Surg Oncol 2005; 31(2):183–90. Tisell LE, Dilley WG, Wells SA Jr. Progression of postoperative residual medullary thyroid carcinoma as monitored by plasma calcitonin levels. Surgery 1996; 119(1):34–9. van Heerden JA, Grant CS, Gharib H, et al. Long-term course of patients with persistent hypercalcitoninemia after apparent curative primary surgery for medullary thyroid carcinoma. Ann Surg 1990; 212(4):395–400. Barbet J, Campion L, Kraeber-Bodere F, Chatal JF. Prognostic impact of serum calcitonin and carcinoembryonic antigen doubling-times in patients with medullary thyroid carcinoma. J Clin Endocrinol Metab 2005; 90(11):6077–84. Tisell LE, Hansson G, Jansson S, Salander H. Reoperation in the treatment of asymptomatic metastasizing medullary thyroid carcinoma. Surgery 1986; 99(1):60–6. Moley JF, Wells SA, Dilley WG, Tisell LE. Reoperation for recurrent or persistent medullary thyroid cancer. Surgery 1993; 114(6):1090–5. Fife KM, Bower M, Harmer CL. Medullary thyroid cancer: the role of radiotherapy in local control. Eur J Surg Oncol 1996; 22(6):588–91. Simpson WJ. Radioiodine and radiotherapy in the management of thyroid cancers. Otolaryngol Clin North Am 1990; 23(3):509–21. Schlumberger M, Gardet P, de Vathaire F. External radiotherapy and chemotherapy in MTC patients. In: Calmettes C, Guliana J (eds), Medullary Thyroid Carcinoma. Paris: Inserm/John Libbey, 1991, 213–20. Samaan NA, Schultz PN, Hickey RC. Medullary thyroid carcinoma: prognosis of familial versus sporadic disease and the role of radiotherapy. J Clin Endocrinol Metab 1988; 67(4):801–5. Fersht N, Vini L, A’Hern R, Harmer C. The role of radiotherapy in the management of elevated calcitonin after surgery for medullary thyroid cancer. Thyroid 2001; 11(12): 1161–8. Nocera M, Baudin E, Pellegriti G, et al. Treatment of advanced medullary thyroid cancer with an alternating combination of doxorubicin–streptozocin and 5 FU–dacarbazine. Groupe d’Etude des Tumeurs a Calcitonine (GETC). Br J Cancer 2000; 83(6):715–18.
148 Kaltsas G, Rockall A, Papadogias D, et al. Recent advances in radiological and radionuclide imaging and therapy of neuroendocrine tumours. Eur J Endocrinol 2004; 151(1):15–27. ●149 Hyer SL, Vini L, A’Hern R, Harmer C. Medullary thyroid cancer: multivariate analysis of prognostic factors influencing survival. Eur J Surg Oncol 2000; 26(7):686–90. 150 Wells SA Jr, Chi DD, Toshima K, et al. Predictive DNA testing and prophylactic thyroidectomy in patients at risk for multiple endocrine neoplasia type 2A. Ann Surg 1994; 220(3):237–47. 151 Ponder BA. Genetic screening for multiple endocrine neoplasia type 2. Exp Clin Endocrinol 1993; 101(1):53–6. ◆152 Brandi ML, Gagel RF, Angeli, A et al. Guidelines for diagnosis and therapy of MEN type 1 and type 2. J Clin Endocrinol Metab 2001; 86(12):5658–71. 153 Gagel RF, Tashjian AH Jr, Cummings T, et al. The clinical outcome of prospective screening for multiple endocrine neoplasia type 2a. An 18-year experience. N Engl J Med 1988; 318(8):478–84. 154 Pinchera A, Elisei R. Medullary thyroid cancer: diagnosis and management. In: Mazzaferri E, Harmer C, Mallick U, Kendall-Taylor P (eds), Practical Management of Thyroid Cancer – a Multidisciplinary Approach. London: SpringerVerlag, 2006; 255–279. 155 Tupchong L, Hughes F, Harmer CL. Primary lymphoma of the thyroid: clinical features, prognostic factors, and results of treatment. Int J Radiat Oncol Biol Phys 1986; 12(10):1813–21. ●156 Matsuzuka F, Miyauchi A, Katayama S, et al. Clinical aspects of primary thyroid lymphoma: diagnosis and treatment based on our experience of 119 cases. Thyroid 1993; 3(2):93–9. 157 Haq M, Harmer C. Rare thyroid cancers. In: Mazzaferri E, Harmer C, Mallick U, Kendall-Taylor P (eds), Practical Management of Thyroid Cancer. London: Springer-Verlag, 2006 393–402. 158 Thieblemont C, Mayer A, Dumontet C, et al. Primary thyroid lymphoma is a heterogeneous disease. J Clin Endocrinol Metab 2002; 87(1):105–11. 159 Short SC, Suovuori A, Cook G, et al. A phase II study using retinoids as redifferentiation agents to increase iodine uptake in metastatic thyroid cancer. Clin Oncol (R Coll Radiol) 2004; 16(8):569–74. ●160 Christian JA, Cook GJ, Harmer C. Indium-111-labelled octreotide scintigraphy in the diagnosis and management of non-iodine avid metastatic carcinoma of the thyroid. Br J Cancer 2003; 89(2):258–61. 161 Gabler B, Aicher T, Heiss P, Senekowitsch-Schmidtke R. Growth inhibition of human papillary thyroid carcinoma cells and multicellular spheroids by anti-EGF-receptor antibody. Anticancer Res 1997; 17(4B):3157–9. 162 Kodama Y, Asai N, Kawai K, et al. The RET proto-oncogene: a molecular therapeutic target in thyroid cancer. Cancer Sci 2005; 96(3):143–8. 163 Herbst RS. Role of novel targeted therapies in the clinic. Br J Cancer 2005; 92(Suppl. 1):S21–7.
18 Endocrine cancer MARIA GUEORGUIEV, ASHLEY B. GROSSMAN AND P. NICHOLAS PLOWMAN
The APUD concept and APUDomas Pituitary tumours Parathyroid adenoma and carcinoma Medullary thyroid carcinoma Gastro-entero-pancreatic endocrine tumours Carcinoid tumours
438 438 449 450 451 452
THE APUD CONCEPT AND APUDOMAS In 1966 Pearse first described cytochemical and ultrastructural properties which were shared by several apparently disparate cell series in the body – initially adrenomedullary chromaffin cells, enterochromaffin cells, the corticotroph, the melanotroph, the pancreatic islet B cell and the thyroid C cell. Pearse later proposed the generic name APUD for these cells from the initial letters of their common cytochemical characteristics, which include amine precursor uptake and decarboxylase activity within the cells.1 Since that time, the list of APUD cells has expanded enormously. The structural and chemical similarity of APUD cells to neurons suggested a neural crest origin. Indeed, APUD cells of the adrenal medulla, melanocytes, thyroid, gastrointestinal tract and carotid body are of principally neuroectodermal lineage, and the ultrastructural similarity is true for all APUD cells. Pearse considered these cells as ‘neuroendocrine’ programmed cells derived from determined precursors arising in the embryonic epiblast, or in one of its principal early descendants. They are conceived as constituting a diffuse neuroendocrine system (DNES), which may be regarded as a third division of the nervous system, products of which suppress, amplify or modulate the activities of the other two divisions. 2 The DNES is divided into central and peripheral divisions, the first of which contains the cells of the hypothalamo-pituitary axis and the pineal gland, while the cells of the second division are primarily located in the gastrointestinal tract and pancreas, where they comprise the gastro-entero-pancreatic endocrine cells. However, APUD
Phaeochromocytoma Multiple endocrine neoplasia Adrenocortical tumours Key references References
454 456 457 459 459
cells are actually distributed throughout the body, where they are all prone to both hyperplasia and neoplasia, and more recent concepts have tended to decrease the emphasis on a truly discrete and distinct neuroendocrine ‘network’, and also on their presumed embryological commonality. This chapter deals with many neuroendocrine tumours, although pineal tumours and medullary carcinoma of the thyroid are covered in more detail in their respective chapters. The management of pituitary tumours is discussed first, followed by some of the more important neoplastic conditions in other organs. Finally, the topic of adrenocortical tumours is covered.
PITUITARY TUMOURS Tumours of the pituitary gland are not uncommon and represent approximately 10 per cent of all intracranial tumours. However, this figure is usually based on mass lesions of the pituitary, which present with visual-field defects or local destructive changes, and it is now realized that small, hormonally active pituitary tumours are considerably less rare. These smaller tumours present due to the consequent endocrinopathy, most commonly sexual or reproductive dysfunction. The management of such tumours differs considerably from conventional oncological treatment programmes. Nevertheless, tumours of the pituitary form a continuum from the relatively insignificant minor aberration requiring no treatment to the lethal massive tumour resistant to all modalities of therapy.
Pituitary tumours 439
Classification Pituitary tumours were originally classified in terms of their staining characteristics with conventional histological techniques, and have been principally divided into eosinophilic, basophilic and chromophobe adenomas. In such a classification the majority of tumours are either eosinophilic or chromophobe, with 10–15 per cent being basophilic. However, the staining characteristics are principally a reflection of the nature of the secretory product of the cell, and these can now be visualized directly using immunohistochemistry or immunofluorescence in most instances. The hormone product of the cell is also more closely related to the function of the tumour, its clinical presentation and its biological behaviour, and is thus a more logical way to classify histological types. On this basis, pituitary tumours may be classified as either secretory or non-secretory, and the former may be subdivided on the basis of their principal hormone product. The majority of secretory pituitary tumours are undoubtedly prolactin secreting or prolactinomas. Prolactincontaining vesicles are usually seen throughout the cytoplasm, and may be released aberrantly from non-apical surfaces of the adenomatous lactotroph. Occasionally, tumours of the somatotrophs (growth hormone (GH)secreting cells) have lactotrophs (prolactin-secreting cells) scattered throughout their substance, appearing as though such cells had become isolated in the tumour matrix during its growth. However, there are also true mixed tumours consisting of adenomatous elements of both lactotrophs and somatotrophs, and certain tumours appear to secrete both prolactin and GH from the same cytoplasmic granules. It has been speculated that the latter tumours have arisen from a common prolactin–GH cell, the ‘somato-mammotroph’, which may be present in normal pituitary tissue or might represent a progenitor of both lactotrophs and somatotrophs. Basophil tumours contain one of the glycoprotein hormones and most often consist of adenomatous corticotrophs. Adrenocorticotrophin (ACTH) is not itself glycosylated, but originates from a 31-kDa precursor, pro-opiomelanocortin, which has sugar moieties attached at several sites upstream and downstream to ACTH. Approximately 10–15 per cent of all pituitary tumours are classified conventionally as ACTH-secreting basophil adenomas. Tumours of the thyrotroph, secreting thyroid-stimulating hormone (TSH), are rare. However, approximately 30 per cent of all large pituitary tumours are said to be non-secretory or functionless, and are usually chromophobe adenomas. Many of these may be shown by electron microscopy to contain secretory vesicles. It appears increasingly likely that these tumours are related to, or originate from, gonadotrophinsecreting cells. They often secrete alpha-subunit, the common subunit of luteinizing hormone (LH), folliclestimulating hormone (FSH) and TSH, both in vivo and in vitro; in addition, secretion of LH and FSH may frequently be demonstrated in vitro. However, they rarely present
with clinical symptoms and signs of gonadotrophin excess; true gonadotrophinomas are extremely rare. Tumours of the posterior pituitary are occasionally seen, but generally arise from the glial or non-endocrine elements of the gland. Secretory tumours of the neurohypophysis have not been described.
Treatment objectives It is particularly important in the management of pituitary disorders that the objectives of treatment are clarified. In the case of pituitary tumours, the principal problems are due to the local mass effects of the lesion, especially visual impairment, partial or complete hypopituitarism, and the distant (target tissue) effects of any hormonal hypersecretion. Treatment thus needs to be directed towards reversing the neurological impairment and avoiding its recurrence, replacing any endocrine deficits and normalizing the levels of any elevated hormones. The disruption of the patient’s lifestyle should be minimized; this implies careful consideration of the necessity for long-term medication, with its possible side effects, and the frequency of outpatient visits and inpatient reassessments. It is difficult to optimize all these objectives, as it is usually the case that the more radical therapy with the highest probability of tumour sterilization will be most likely to induce long-term endocrine sequelae. Furthermore, individual patients may differ in their requirements for a normalization of their endocrine status and their desire to avoid medication. Not all neuroendocrine changes defined by subtle alterations during complex test procedures necessitate therapeutic intervention. It is, evident, therefore, that a range of treatment options can often be made available, and a therapeutic plan optimized according to the needs and wishes of the individual patient.
Prolactinomas Prolactin-secreting pituitary adenomas (prolactinomas) are the most common functional pituitary tumours. Although these adenomas are classified with regard to their size, they differ also with regard to their biological behaviour and clinical features. Microadenomas or microprolactinomas measure less than 1 cm in maximum diameter and are entirely contained within the pituitary fossa. Macroadenomas or macroprolactinomas are larger than 1 cm and usually expand beyond the pituitary fossa and cause local compression and displacement of adjacent structures. MACROPROLACTINOMAS
Clinical symptoms are the consequence of direct effects of hyperprolactinaemia, or of tumour expansion either due to local compression of anterior pituitary (hormonal deficits) or of adjacent structures. Usually, serum prolactin levels are markedly increased in prolactinomas, and correlate with
440 Endocrine cancer
tumour size, but mild hyperprolactinaemia can be present in patients with hypothalamic or pituitary lesions of other aetiology. Serum prolactin usually exceeds 4000–6000 mIU/L in patients with a macroprolactinoma, but such levels can occasionally be found with non-functioning pituitary adenomas (normal level, 450 mIU/L).3 Indeed, prolactin release is predominantly regulated by the inhibitory action of hypothalamic dopamine, and any disruption of the portal vasculature causing a decrease in dopamine delivery to the lactotrophs would result in an increase of prolactin levels. A skull X-ray will usually show an enlargement of the pituitary fossa. Improvement in radiodiagnostic techniques such as computed tomography (CT) and, particularly, magnetic resonance imaging (MRI) with gadolinium allows the visualization of the pituitary adenoma and its extra-sellar extension, and the exclusion of a possible hypothalamic lesion. The conventional surgical approach for macroprolactinomas was originally transcranial craniotomy with extirpation of the tumour and fossa contents. This technique is highly effective in the rapid restoration of visual deficits and debulking tumoral mass, but is associated with high morbidity and often with hypopituitarism.4 Trans-ethmoidal surgery has been used, but a high rate of orbital cellulitis has led to its abandonment. Originally introduced by Harvey Cushing, and then substantially improved by Guiot and Hardy, trans-sphenoidal resection, (currently with or without endoscopic assistance) with an operating microscope and intra-operative screening and, most recently, with computer-assisted neuronavigation, was then widely used for pituitary adenomas of all types. Rare cases of tumours with significant temporal or anterior cranial extension may still necessitate a trans-cranial approach or a combined trans-sphenoidal and trans-cranial approach. Morbidity is low in experienced hands, with a rate of less than 1–2 per cent for major complications (cerebrospinal fluid (CSF) rhinorrhoea and meningitis, ischaemic stroke or intracranial haemorrhage and visual loss) and of approximately 5 per cent for minor complications; surgically induced diabetes insipidus is usually transient. Transsphenoidal surgery for macroprolactinomas is rarely curative, as small residual tumoral rests remain. In a large series, trans-sphenoidal surgery induced remission in 56 per cent of patients with macroprolactinomas (normalization of postoperative prolactin levels), with a recurrence rate of 13 per cent (including microadenomas and macroadenomas) at 10 years.5*** The addition of postoperative standard external-beam radiotherapy to this regimen decreased the risk of recurrence from 50 per cent to 3 per cent or less at 10 years.6* Following the introduction in 1972 of the ergot alkaloid bromocriptine for the treatment of galactorrhoea and hyperprolactinaemia, it was rapidly introduced into tumour practice, where it was shown to normalize prolactin levels in patients with prolactinomas when used in addition to surgery and radiotherapy. Wass et al.7* reported a reduction in tumour size with bromocriptine in 9 out of 11 hyperprolactinaemic patients with presumed prolactinomas, and
normalization of visual fields in 5 patients. It appears that the majority of patients (75–80 per cent) with large prolactinomas, even with a significant extra-sellar extension, will show tumour reduction. These changes often occur very rapidly, and clinical symptoms may improve within hours of initiation of bromocriptine therapy, although there may still be ongoing shrinkage even after many months. Marked tumour shrinkage of large prolactinomas with large extra-sellar extensions may lead to a partially empty fossa, and occasionally CSF rhinorrhoea. Although significant tumour reduction is expected to occur within 6 weeks,8 our experience showed that these tumours may continue to decrease in size more slowly over many months of treatment. Other dopamine agonists have proven to be as effective: lisuride, pergolide, mesulergine, the longacting cabergoline9–12 and the non-ergot quinagolide.13,14 Re-expansion of tumour after the cessation of therapy may occur relatively rapidly, or more slowly in about 10 per cent cases.15–17 For example, Figure 18.1 demonstrates the serial CT scans of one of our patients treated for 2 years with bromocriptine who demonstrated marked tumour shrinkage; on cessation of therapy (due to patient noncompliance) for 5 weeks, there was marked and rapid re-growth of the tumour to its former size. However, there are also reports of patients in whom withdrawal of dopamine agonist after several years of therapy did not induce tumour re-growth, at least in the short term. Progressively, bromocriptine and, more recently, cabergoline therapy has become the primary therapy for all prolactinomas. Clinical improvement follows tumour shrinkage and includes early resolution of visual defects if their duration has not been too long.18 Evaluation of the therapeutic response to cabergoline and related drugs should be made within 4–6 weeks after the initiation of therapy. A long-term management decision should then be taken, but this may need to be earlier if there is no response or only a very limited one. If the prolactinoma has considerably decreased in size and the optic chiasm is free of tumour, we previously advised that external-beam radiotherapy be given. Tsagarakis et al.19* reported that radiotherapy induced normalization of prolactin levels in substantial numbers of patients 5–10 years postradiotherapy, although there is a long-term risk of hypopituitarism due to hypothalamic damage, most importantly the early loss of GH reserve.20–22 However, most would now simply maintain the dopamine agonist at the lowest effective dose, with the possibility of long-term ‘cure’ on removal of the drug in a significant minority of patients.23** The outcome of trans-sphenoidal surgery may be improved with prior bromocriptine; however, complete cure is unlikely and postoperative radiotherapy or dopamine agonist is usually necessary. It should be noted that in a woman of childbearing age there remains a 25 per cent probability of symptomatic tumour expansion during pregnancy. Bromocriptine withdrawal is also associated with this risk and may be required for life even if dose requirements decrease, although recent studies have suggested
Pituitary tumours 441
(a)
(b)
(c)
(d)
Figure 18.1 CT scans of a patient with a large prolactinoma: (a) at presentation (prolactin 206 000 mU/L) and (b) after 2 years bromocriptine treatment (prolactin 170 mU/L); the patient then stopped bromocriptine treatment for 5 weeks and his tumour enlarged rapidly, as shown in (c), when his prolactin had risen to 50 000 mU/L. He was restarted on bromocriptine, and was re-scanned 3 months later (d), by which time his serum prolactin had fallen to 400 mU/L.
that after long-term treatment re-expansion of the tumour following treatment cessation may be relatively uncommon.23** In the case of patients in whom a tumour with large supra-sellar extension with associated visual impairment persists after 4 weeks bromocriptine or cabergoline treatment, decompressive trans-sphenoidal surgery must be undertaken. Indeed, the decision for medical primary therapy in a patient with visual deficit can be taken only if experienced physicians work closely with neurosurgeons, and also have the facilities serially to monitor visual fields and acuity, as well as tumour volume with MRI scanning. Very rarely, ‘escape’ from therapy may occur,24 or an important supra-sellar extension may remain, which may represent fibrous change within the tumour, making trans-sphenoidal extirpation difficult.25 In the case of very large prolactinomas with invasion to adjacent structures, complete or partial surgical resection may be difficult, and cabergoline treatment may eventually be completed with radiotherapy unless there is evidence for drug-induced tumour necrosis.4 In summary, dopamine agonist therapy has transformed the management of macroprolactinomas, and surgery has been limited to those patients demonstrating residual chiasmal compression after attempted tumour shrinkage. Such
surgery is rarely curative and may occasionally have to be complemented by radiotherapy. There remains uncertainty concerning the optimal long-term therapy; some centres advise tumour control with dopamine agonist therapy alone, whereas others would suggest that definitive treatment with radiotherapy will lead to gradual sterilization of the tumour, with a low medium-term risk of hypopituitarism. However, the high incidence of GH deficiency, and the increasing acceptance of the morbidity associated with this state, has led to a decrease in enthusiasm for radiotherapy in this situation. With either approach, longterm close surveillance is necessary. MICROPROLACTINOMAS
Microprolactinomas have a prevalence of up to 10 per cent in adults, as revealed by some autopsy series, although hyperprolactinaemia is produced by these tumours in only 0.1 per cent of female patients.26 The prevalence in men is much lower (0.005 per cent), presumably due to their different hormonal milieu. There is, however, no evidence that oestrogen-containing oral contraceptives are oncogenic.27 The biochemical diagnosis of microprolactinoma
442 Endocrine cancer
may be difficult, as similar serum prolactin levels are found in patients with ‘functional’ or ‘idiopathic’ hyperprolactinaemia (up to 4000–6000 mIU/L).28 High-resolution MRI scans reveal abnormalities of the pituitary fossa in the majority of these patients. In this context, most women with persistent serum prolactin levels above 1000 mIU/L are likely to have microprolactinomas. Most clinical centres would recommend primary therapy with dopamine agonist as treatment for secondary amenorrhoea in such patients; resistance to this drug is rare and there is little cross-intolerance between different agonists.29 Previous estimates of pregnancy-induced tumour expansion were almost certainly exaggerated, and it seems unlikely that such problems will occur in more than 1 per cent of pregnancies, although authorities differ as to the exact incidence. When this does occur, it may be treated rapidly by the re-institution of dopamine agonist therapy. However, treatment may need to be continued for many years, as spontaneous resolution is only infrequently seen. Occasionally after treatment with a dopamine agonist, serum prolactin may return to a level much lower than previously, such that therapy may be discontinued;30 this tends to occur particularly following a pregnancy.31 Very rarely, there is progression from a microadenoma to a macroadenoma, with a gradual increase in tumour size and prolactin level and the onset of local compressive symptoms. In such a case the treatment should be adapted to that appropriate for a macroprolactinoma. We now generally initiate treatment with cabergoline, starting with 0.25 mg once a week and increasing over 2–3 weeks to 0.5 mg once or twice a week. True resistance may occasionally be responsive to an alternative dopamine agonist such as bromocriptine or quinagolide, but this is rarely so in our experience. Most data on safety in inducing conception have been obtained with bromocriptine, with more than 20 years’ evidence of a lack of teratogenicity or problems in pregnancy. To date, cabergoline and quinagolide appear to be equally safe, but the relative long-term experience is much more limited. In a recent Italian study with 2–5 years’ follow-up,23** withdrawal of long-term cabergoline used as primary therapy resulted in recurrent hyperprolactinaemia at rates of 31 per cent and 36 per cent in patients with microprolactinomas or macroprolactinomas respectively (although levels remained lower than those before treatment), but in the absence of tumour re-growth. The estimated rate of recurrence at 5 years was higher if a residual tumour was visible on MRI at the time of cabergoline withdrawal. These results suggest that it is safe to interrupt long-term cabergoline treatment under careful monitoring during follow-up, as there probably exists a persistent anti-tumoral effect of cabergoline that is additive to the natural evolution of some microprolactinomas. In a situation in which there is drug resistance or intolerance, surgical intervention may be considered. In large centres with an experienced neurosurgeon, trans-sphenoidal removal is curative in 70–90 per cent of patients, with an inverse correlation between cure rate and initial serum
prolactin levels.32,33 In recent series,5*** the recurrence rate for prolactinomas following trans-sphenoidal surgery is estimated to be about 13 per cent at 10 years, with remission assessed by normalization of prolactin levels in 87 per cent of patients with microprolactinomas. The risk of hypopituitarism is very low in the case of microprolactinomas. In our opinion, trans-sphenoidal surgery is, on current evidence, an acceptable alternative to dopamine agonist therapy only for those patients unable or unwilling to take long-term medication; 10-year and 15-year followup studies of these patients are awaited with interest. If a patient’s symptoms are minimal, it may be prudent in some cases to avoid treatment altogether. The natural history of the microprolactinoma is generally benign, and many patients only require reassurance. There are, however, three provisos. First, since hyperprolactinaemia may cause subtle changes in sexual function and libido, every patient should at least be offered a trial of a dopamine agonist to assess their clinical response. Second, hyperprolactinaemia is associated with oestrogen deficiency and long-term osteoporosis and therefore should necessarily be treated when it causes either complete amenorrhoea or low circulating oestradiol levels.34–36 (As a general rule, we consider that menses should not be less frequent than every other month.) Finally, all patients should be followed up long term to monitor the possible progression to a macroprolactinoma. An alternative approach is simply to treat the patient with some form of oestrogen replacement therapy, to induce regular withdrawal bleeds and minimize the risk of osteoporosis. Although such treatment has previously been contraindicated owing to the assumed risk of tumour-growth stimulation, several studies have demonstrated that such a risk may be relatively small.37 Pregnancy in a patient with a microprolactinoma has a low but positive risk of oestrogen-induced tumour expansion. It is probably wise, therefore, to monitor such patients during pregnancy for clinical symptoms and visual-field defects. Postpartum, it is unlikely that the tumour will enlarge in size if it has not done so previously, and thus lactation should proceed normally. Microprolactinomas appear to be extremely uncommon in men. It has been suggested that this is because the principal clinical symptom of impotence is not brought early to medical attention38 and the patients are only seen if they progress to a macroprolactinoma some 10 years later. Whatever the case, the treatment of a male microprolactinoma follows the same guidelines as in the female. Finally, a number of patients with apparent hyperprolactinaemia and a relative absence of symptoms may have ‘macroprolactinaemia’, where biochemical assays show a spuriously elevated level due to prolactin’s association with an immunoglobulin. This can be established by various techniques, but it is a vital test in all patients with hyperprolactinaemia in whom the clinical and hormonal correlates are discordant. One additional caveat is that in some prolactin assays the presence of very high prolactin levels will saturate the assay antibody and may show a spuriously low or even normal level of prolactin – the ‘hook effect’. Most
Pituitary tumours 443
laboratories are aware of this and will dilute out samples where this may be a confounder, and in an case modern assays are usually robust up to 1 000 000 mIU/L. In summary, microprolactinomas generally have a benign natural history, and therapy must take this into account. The mainstay of treatment is dopamine agonist therapy, supplemented where necessary by trans-sphenoidal surgery. Radiotherapy should only be considered where the tumour gives evidence of high growth characteristics such as local invasiveness or a very high serum prolactin, or there is resistance to dopamine agonist therapy.
Functionless tumours A significant proportion (30–40 per cent) of large pituitary tumours are apparently functionless, as they are unassociated with a hypersecretory syndrome, although a number may be capable of secreting a hormonal product in very low quantities, and about 80 per cent exhibit positive immunostaining for one glycoprotein hormone, mainly gonadotrophins. In 10 per cent of these tumours, proopiomelanocortin-derived products can be detected, and about 2 per cent immunostain for GH. These tumours represent the so-called silent corticotroph adenomas, gonadotrophinomas or somatotropinomas. The rest are chromophobe adenomas, although sparse secretory granules and some endoplasmic reticulum may be evident ultrastructurally. These tumours usually present due to mass effects consequent on their size and with visual defects and headache, but partial or complete hypopituitarism is often present on dynamic testing. The principal treatment for functionless tumours is surgical.39 Surgery is used to decompress the visual pathways, and nowadays this is almost always trans-sphenoidal in all but the most massive tumours.40 As functionless pituitary tumours express dopamine and somatostatin receptors, early work suggested that pre-treatment with dopamine agonists or somatostatin derivatives might cause tumour shrinkage, similar to that seen with prolactinomas. However, prospective studies of patients carefully followed up for prolonged periods have shown that any shrinkage that does occur is relatively minor or even if more important is observed in a small number of patients, and does not obviate the need for surgery.18,29,41,42 It is conceivable that the slight but definite evidence of a decrease in apparent tumour size is due to a diminution in size of the normal lactotrophs or somatotrophs, but that once this has occurred, the inevitable tumour progression is seen. Whatever the mechanism, dopamine agonists or somatostatin analogues only very rarely induce a clinically useful improvement in symptoms or signs in patients with functionless tumours, and we would advise urgent definitive surgical treatment in patients with large pituitary tumours and a normal or minimally elevated serum prolactin (1000 mU/L). Newer somatostatin analogues with selectivity for somatostatin receptor subtypes more characteristic of the non-functioning tumours may alter this approach in the future.
If the surgeon believes that he or she has achieved complete clearance, and if the tumour shows no obvious evidence of invasiveness, many centres would follow this with serial MRI scans (the first at 3–6 months) to assess recurrence. However, a number of surveys in the 1990s demonstrated significant recurrence rates, even in those tumours thought to have been removed completely, of the order of 50 per cent at 10 years.43,44* More recently, Laws et al.5*** reported better results, with 16 per cent recurrence at 10 years (but only 6 per cent requiring re-operation), and 83 per cent of patients alive and without evidence of disease. Postoperative improvement of visual deficits occurred in 87 per cent and normalization of hormone secretion in 27 per cent of patients. It is therefore our own policy to consider external-beam radiotherapy in the majority of patients following surgery, accepting that there will be a gradual increase in hypopituitarism in those patients with postoperative preserved pituitary function. Of course, other factors need to be taken into account, such as the age of the patient and the presence of pre-existing hormone defects, but generally, adhering to this policy, the rate of recurrence has been much less than 5 per cent at 10 years. We have not seen clear evidence of second tumours or neurocognitive defects using our current prescription, although a second-tumour rate, particularly meningioma, of up to 3 per cent at 10 years has been reported. There remains the group of patients whose serum prolactin varies between 1000 mU/L and 6000 mU/L, who may have either a prolactin-secreting tumour or a functionless tumour causing stalk compression hyperprolactinaemia. Both categories respond to dopamine agonists with a normalization of serum prolactin, but only the former are likely to show tumour regression. A short trial of cabergoline may be appropriate to test the response of the tumour, but this must be monitored extremely carefully by serial visual-field testing and MRI scanning, as recourse to surgery is important in non-responders. Silent corticotroph tumours, gonadotrophin-secreting and TSH-secreting tumours should be treated as functionless tumours. Some silent corticotroph tumours may progress to clinical Cushing’s disease and should be treated particularly vigorously. Thyroid-hormone-secreting tumours may be sensitive to dopamine agonists or to the somatostatin analogue octreotide, which may be given a therapeutic trial, while gonadotrophinomas may occasionally respond to gonadotrophin-releasing hormone (GnRH) antagonists (and, infrequently, to GnRH agonists). In summary, the primary approach to the functionless tumour is surgical, trans-sphenoidal in the great majority, followed either by radiotherapy or a policy of regular imaging and close monitoring.
Acromegaly and gigantism Acromegaly is generally considered a rare condition, with an incidence of about three to four new cases per million of population per year.45,46 Its prevalence in the UK is
444 Endocrine cancer
approximately 40 per million population. In more than 99 per cent of cases acromegaly is due to a benign pituitary GH- secreting adenoma, the majority being macroadenomas. The large majority of somatotrophinomas are sporadic, although they could be part of a familial syndrome. As part of multiple endocrine neoplasia type I (MEN-I), they represent about 10 per cent of the associated pituitary tumours and occur, as in sporadic cases, during the fourth and fifth decades of life. These tumours can also appear in the context of Carney complex, where they are the exclusive pituitary tumours found in about 21 per cent of patients. Carney complex has been linked to loss of heterozygosity (LOH) at chromosome 17q or 2p, and to germ-line mutations in the protein kinase A (PKA) regulatory subunit 1 (PRKAR1A) gene;47–49 its loss of function results in constitutive activation of PKA and its signalling, and consequently increased somatotroph proliferation. This signalling pathway is also activated by somatic mutations of GSα gene, detected in up to 40 per cent of sporadic somatotrophinomas,50,51 but also by mutations in patients with McCune–Albright syndrome, characterized by frequently present somatotroph hyperplasia and adenoma.52 Isolated familial somatotrophinomas cluster in some families, with earlier onset (in two-thirds of subjects before the age of 30 years) in the absence of other endocrinopathies; they are associated with LOH at chromosome 11q13, and this is suggestive of a tumour suppressor gene not yet identified – no mutations detected have been in the menin gene.53,54 However, as pituitary imaging and biochemical assays for GH have become more widespread, the spectrum of the disease has widened, with increasing numbers of patients having purely intra-sellar tumours or microadenomas. This has, in turn, modified the balance of the principal therapeutic approaches. It is now clearly established that, when untreated, acromegaly is associated with an increased
mortality. In addition, acromegalic patients have an increased risk (3-fold to 14-fold) of developing colorectal carcinoma or tubular adenomas.55 The association of acromegaly with breast and prostate cancer remains uncertain, but, if present, is a very low risk.56,57 Definitive treatment of acromegaly aimed at lowering serum GH and IGF-1 is indicated in almost all cases. Diagnosis is based on an elevated serum IGF-I and a serum GH that fails to suppress to 0.2 μg/L during a standard glucose tolerance test.58–60 Trans-sphenoidal surgery is highly effective in considerably reducing tumour mass and substantially lowering mean GH levels in the majority of patients, although a surgical cure is rare. Previous considerations for ‘cure’ of acromegaly (mean serum GH, basally or after glucose, 5 mU/L, 10 mU/L or even 20 mU/L) have become more strict in the last years, establishing a therapeutic aim for a serum GH below 5 mU/L (2.5 μg/mL, as residual mean GH levels above 5 mU/L were still associated with elevated mortality.61 Most series report a fall in mean serum GH to less than 5 mU/L in approximately 40–60 per cent of all acromegalic patients treated surgically, but this ‘cure’ rate rises to 80–90 per cent for microadenomas. More recent surgical series (using strict criteria for cure: normalization of IGF-1 levels, random GH 2.5 μg/L and 1 μg/L during the glucose tolerance test) report improved results, with a cure rate of 65 per cent for macroadenomas and 88 per cent for microadenomas, and a 10-year overall recurrence rate of 1.3 per cent.5***,62 Hypopituitarism occurs in approximately 25 per cent of such patients. External-beam radiotherapy has also been used extensively in the treatment of acromegaly, and although it induces a fall in mean serum GH, this may take several years to become fully effective (Fig. 18.2). In our own series 250
200
3000
Mean serum GH (mU/L)
Serum prolactin (mU/L)
4000
2000
1000
150
100
50
<2
0 0
(a)
3
6
9
Pre 2
12
Years since radiotherapy
(b)
4
6
8
10
15
20
Years since radiotherapy
Figure 18.2 Fall in serum prolactin (a) and serum growth hormone (GH) (b) in patients with prolactinomas and acromegaly treated by our standard technique and prescription of external-beam radiotherapy (see text). In each case there was an initial rapid fall in the pathologically elevated hormone levels, followed by a more gradual decrease over succeeding years. Each data point was obtained with the patient at least 1 month off dopamine agonist therapy
Pituitary tumours 445
of 80 patients treated with 45 Gy via a three-field technique, the ‘cure’ rate at 10 years was approximately 70–80 per cent, with levels below 10 mU/L in 90 per cent of patients at this time.63,64 It has been argued that this fall is not paralleled by serum IGF-I levels,65 but most studies have confirmed its efficacy. In addition, normal pituitary function may be compromised. At 6 years we found that 25 per cent of patients who did not require replacement therapy beforehand now needed some form of therapy, including gonadal steroids (11 per cent), thyroxine (8 per cent) or hydrocortisone (16 per cent). Bromocriptine will reduce GH levels in about 60–70 per cent of patients, but only a minority (10–15 per cent) will have a mean serum level of less than 2 μg/L.66 Cabergoline is more effective, with control in around 30–35 per cent of patients.12 Somatostatin analogues have emerged as a more efficient option in the control of tumour growth and serum GH levels. The native short-acting octreotide (Sandostatin®), administered subcutaneously at 100 μg three times daily, has been replaced with long-acting formulations achieving significant reduction of GH levels in the great majority of patients.66 The inhibition of gut and pancreatic peptides may lead to mild diarrhoea/steatorrhoea and cholestasis, but this is rarely problematic. The inhibition of gastric-acid output by octreotide has also been associated with chronic gastritis, but the clinical relevance of this has been disputed. More importantly, the inhibition of gall-bladder motility in conjunction with a change in bile composition, induced by octreotide, gives rise to a steady increase in the incidence of gallstones. Long-acting octreotide is given as a depot injection of 20–30 mg every 4–6 weeks and, after a gradual rise in drug level, it produces a prolonged inhibition of GH release in more than 80 per cent of patients, often to within the ‘safe’ range.67** Longacting lanreotide (Somatuline LA®) is more rapid in onset but only lasts for 1–2 weeks, although a longer-acting gel formulation can be given subcutaneously and lasts at least 4 weeks. Unfortunately, all of the long-acting preparations are expensive. In the last few years, medical therapy with somatostatin analogues appears to be the first-line therapy in acromegaly for selected patients when surgery is deemed inappropriate. Two recent reviews68***,69*** support this new therapeutic direction: treatment with somatostatin analogues (either short-acting or long-acting formulations of octreotide or lanreotide) produces significant tumour shrinkage in up to 50 per cent of patients; further, this therapeutic effect may be more evident if used as primary therapy than when used as adjuvant therapy following incomplete surgical resection. The new somatostatin analogue SOM230 (Novartis, Basel, Switzerland) is also of major therapeutic interest, as it has a broader receptoraffinity spectrum and may control GH levels in patients resistant to octreotide and lanreotide. In occasional cases, the association of somatostatin analogue and dopamine agonist may be effective. For cases resistant to the classical, previously described, agents, there is the option of using
pegvisomant, a novel GH receptor antagonist. Pegvisomant is a genetically engineered GH analogue that is ‘pegylated’ (to increase its half-life and reduce its immunogenicity) and that carries eight amino-acid substitutions at one binding site and one amino-acid substitution at the second binding site. This results in interference with GH receptor dimerization and its conformational modification, both necessary for signal transduction and the production of IGF-1 at peripheral sites.70,71 This drug normalizes IGF-1 levels in 90–97 per cent of cases if administered subcutaneously daily at doses up to 40 mg for 12 months or longer; GH levels initially increase and then stabilize.72,73 It is well tolerated, but remains expensive. Improvement in insulin sensitivity and glucose metabolism was also reported in acromegalic patients treated with pegvisomant.72,74 Pegvisomant is the therapeutic option of choice for patients with acromegaly who are resistant or intolerant to somatostatin analogue therapy.75,76 It may well be the first-line treatment for some patients with impaired glucose metabolism or overt diabetes mellitus, but currently its expense limits its use, and there is concern regarding the effects on tumour growth. A recent report showed an additive benefit of combining long-acting somatostatin analogues with weekly pegvisomant, achieving good control of biochemical parameters and no tumour re-growth in 26 acromegalic patients resistant to somatostatin analogues (with or without previous neurosurgery or radiotherapy), and a consequent reduction in therapeutic costs.77 There is some concern about the consequently increased circulating GH levels, with loss of negative feedback on the tumour. A few cases of tumour re-growth under pegvisomant have been described,72,73,78 with an estimated rate of tumour progression of 1.6–2.9 per cent.79 However, this tumoral escape could be explained by the cessation of intercurrent octreotide therapy and/or the absence of prior radiotherapy. Magnetic resonance imaging surveillance of the pituitary (every 6–12 months) is recommended.80 In summary, both surgery and radiotherapy can substantially lower GH levels in about 75 per cent of patients, but whereas the former is effective immediately, the latter may take up to 10 years. Both are more efficacious with lower GH levels and may cause partial or complete hypopituitarism. For rapidity of results, therefore, primary therapy should usually be trans-sphenoidal adenomectomy, followed by external-beam radiotherapy for residual tumour (as revealed by elevated non-suppressible serum GH levels). The radiotherapy not only leads to a further fall in GH levels but also, as discussed above, greatly decreases the probability of re-growth or recurrence. Focused radiotherapy, or radiosurgery, may allow for higher local doses of irradiation with less risk of adverse effects. Medical therapy is most useful as interim treatment, although it can occasionally be used as sole treatment in patients with very mild disease who are unenthusiastic about surgery, or in those deemed unfit for surgery. In all others, definitive treatment aimed at ablating the tumour and lowering serum GH and IGF-I levels into a ‘safe’ range is mandatory.
446 Endocrine cancer
Cushing’s disease Cushing’s disease, or pituitary-dependent Cushing’s syndrome, is seen in 10–15 per cent of patients with pituitary tumours. Almost all cases are secondary to an ACTHsecreting basophil adenoma of the pituitary, although very occasionally patients with hyperplastic nests of basophilic cells have been reported. The tumours are usually small (microadenomas) and may often lie at or near the midline. Therapy is directed towards normalization of the excess secretion of corticosteroids and either removal of the tumour or at least avoidance of its re-growth. In earlier years, bilateral adrenalectomy was certainly able to treat the cortisol hypersecretion, but the pituitary tumour remained, and in some patients enlarged further in size, with a concomitant increase in circulating ACTH and hyperpigmentation – the development of Nelson’s syndrome. Localizing the source of ACTH excess is important and its eradication is most effectively obtained by trans-sphenoidal microadenomectomy, aiming at normalizing/reducing the high levels of circulating corticosteroids. Preoperative localization of these corticotroph microadenomas is first attempted with conventional techniques such as MRI scanning of the pituitary. The diagnostic sensitivity remains low (40–60 per cent) due to their small size,81**,82*,83* although some studies have found higher positive predictive values of 72–93 per cent.84*,85* Bilateral inferior petrosal sinus sampling (BIPSS) – an invasive technique available in specialized centres – allows for the confirmation of the pituitary source and can help to localize the corticotroph (micro)adenoma within the pituitary, which is of crucial importance for successful surgical resection. Its diagnostic sensitivity in confirming a pituitary source is around 81–89.5 per cent (85–97 per cent post-corticotropin-releasing hormone (CRH) stimulation), with a specificity of 90–100 per cent.84*,86*,87*,88* The sensitivity for correct lateralization of pituitary adenoma is 83 per cent,84* although others have produced lower rates. The tumours are usually small, and in the best centres immediate ‘cure’ rates of 75–80 per cent after trans-sphenoidal surgery are reported, 89,90 although a true cure rate of around 50 per cent is produced when strict criteria are used. In a larger US series including 640 ACTH-secreting pituitary adenomas, Laws achieved remission in 91 per cent of microadenomas but only in 65 per cent of macroadenomas, with a recurrence rate of 12 per cent at 10 years in adults (and higher, 42 per cent, in children).5*** As the normal corticotrophs undergo long-term suppression (histologically demonstrable as Crooke’s cell changes), serum cortisol postoperatively should be extremely low and patients will require long-term corticosteroid cover until their own pituitary–adrenal axis recovers. Recurrence appears to be uncommon (3–5 per cent) when patients are truly cured (i.e. many apparent recurrences simply reflect inadequate criteria for primary cure), but long-term follow-up is essential. Where attempted microadenomectomy has not led to a cure, the alternatives are either to re-explore the fossa and
perform a total hypophysectomy or to irradiate the pituitary tumour. Radiotherapy of ACTH-secreting tumours, as in the case of other hormonally active pituitary tumours, leads to a gradual fall in ACTH levels over several years and a decrease in requirements for medical therapy.91 In our series of 22 patients treated with radiotherapy and medical therapy alone, 60 per cent were cured 1–12 years after irradiation, insofar as they have been able to stop all medical treatment.92 It is particularly efficacious in childhood,93* where surgery may be hazardous or very difficult, and in our series of paediatric Cushing’s disease, normalization of cortisol was usually seen within 1 year.94* Stereotactic radiosurgery was reported to achieve 17–83 per cent endocrinological ‘cure’ in 314 patients with Cushing’s disease from different series, with a mean follow-up of 2 years (review 95***). This percentage was lower (0–36 per cent) in patients with Nelson’s syndrome, and tumour-growth-control rate was 82–100 per cent at a mean follow-up of 3 years. Metyrapone is an 11-hydroxylase steroid synthesis inhibitor that blocks the final stage in cortisol biosynthesis and may be used to normalize cortisol levels while awaiting improvement following radiotherapy, or as preoperative preparation. The principal side effects of metyrapone are associated with the increased shuttling of steroid precursors into adrenal androgen biosynthesis, causing menstrual problems and hirsutism in women; metyrapone may also be teratogenic. Mitotane (1,1-dichloro-2-(o-chlorophenyl)-1-(p-chlorophenyl)ethane; o⬘p⬘-DDD) is adrenolytic and causes a slow-onset fall in all adrenal products. Impurities may render the drug toxic in high doses, but when pure it may be remarkably effective in lowering cortisol levels in the long term. Unfortunately, recent evidence suggests that o⬘p⬘DDD causes severe hypercholesterolaemia; if essential, the drug may be combined with drugs that lower cholesterol such as lovastatin or simvastatin, but in general this finding has severely limited its long-term use, as has its interference with cortisol estimation as it elevates CBG levels. The antifungal agent ketoconazole is also adrenolytic and is finding increasing favour as first-line medical therapy for hypercortisolaemia, especially in young women. It should be used with care as it is potentially hepatotoxic. However, these drugs are only useful when the primary pituitary pathology has been tackled by other means. Intravenous etomidate is available to lower cortisol levels where oral therapy is not possible.96,97 More recent agents such as retinoic acid and rosiglitazone have been shown to inhibit tumour growth in vitro, but in our experience the latter seems to be ineffective in vivo. Bilateral adrenalectomy had been of mainly historical interest, but with the advent of laparoscopic adrenalectomy its place is again being recognized. Rather than persevere with long-term control using adrenostatic agents, both adrenals can be removed at a single procedure, or in staged unilateral procedures. Indeed, it has been claimed recently that if control is achieved after pituitary radiotherapy and unilateral adrenalectomy, the long-term results are equivalent
Pituitary tumours 447
to those of successful trans-sphenoidal surgery. Bilateral adrenalectomy may also be considered in the treatment of patients with the ectopic ACTH syndrome, where the source of ACTH has defied localization. It is uncertain as to whether pituitary irradiation following adrenalectomy for Cushing’s disease will necessarily prevent the onset of Nelson’s syndrome, but we consider that it is best given prophylactically, especially in the presence of an obvious pituitary tumour. Our data suggest that the onset of Nelson’s syndrome may be reduced, or at least delayed, by such treatment.98* In Nelson’s syndrome, or where the ACTH-secreting tumour presents as a macroadenoma – in approximately 10–15 per cent of cases99* – surgical cure is rare and radiotherapy usually essential. When the tumour recurs following conventional radiotherapy, ‘radiosurgery’ may still be applicable (see below). Such tumours may be extremely invasive and prone to recurrence, and often prove to be amongst the most difficult to treat in neuroendocrinology. Certain drugs, such as bromocriptine, cyproheptadine, sodium valproate and octreotide, may be tumoristatic in some patients, and can certainly be tried in difficult cases. Unfortunately, these tumours may prove resistant to all modalities of treatment. In summary, ACTH-secreting pituitary tumours are usually small and best treated surgically, with radiotherapy reserved for postoperative residual tumour or recurrence. Of the drugs currently available for blocking cortisol production while definitive treatment is awaited, none is perfect, but metyrapone, o⬘p⬘-DDD and ketoconazole may play a useful role. Drugs which may interact with central neurotransmitters are rarely necessary or effective, except where other therapeutic approaches have failed.
Pituitary carcinoma About 45–55 per cent of pituitary adenomas, depending on the neuroradiological, neurosurgical or histological criteria, behave in a more invasive manner without being classified as malignant, infiltrating dura (95 per cent at histology), bone and/or surrounding tissue.100–102,103***,104,105 The diagnosis of pituitary carcinoma is considered only when metastases are found in the central nervous system, or more distantly,100,102,106 and its prevalence is low – approximately 0.2–0.5 per cent101,107–109 Like adenomas, pituitary carcinomas originate from cells of the adenohypophysis.100,106 A number of factors and events have been implicated in the genesis and progression of pituitary tumours towards a selective growth potential and a more invasive phenotype, and acquisition of the ability to metastasize, explaining the monoclonal nature of these neoplasms.110*** Among these are the accumulation of genetic aberrations,111–113 deletion of the retinoblastoma gene Rb, increased or reduced expression of mutated tumour suppressor genes or oncogenes, or of growth factors and their receptors – p53, H-ras, nm23 protein, c-myc, cyclin-dependent kinases and inhibitors (p27), galectin-3, HER-2/neu proto-oncogene,
activated EGF, cyclooxygenase-2 and telomerase, while alterations in the menin and PRKAR1 genes are more common in familial syndromes (MEN-1, Carney’s complex) than in sporadic pituitary neoplasms (for review see110***,103***). Classical histologic/morphologic criteria are not specific for malignancy. Additional criteria predictive of malignancy are required: these are estimation of the cell-cycle-specific antigen Ki-67 labelling indices (LI) of proliferation (MIB-1 antibody immunostaining), estimation of the proliferating cell nuclear antigen (PCNA) proliferation index, and expression of p53 proto-oncogene.114–116 According to the recent World Health Organization (WHO) classification of endocrine neoplasms, primary neuroendocrine tumours that show a high mitotic activity, associated with Ki-67% LI 3 per cent, and/or p53 expression should be included in the category of atypical pituitary adenomas.117 The majority of pituitary carcinomas secrete hormones. The most common are ACTH-secreting and prolactinsecreting, less common are GH-secreting, and rare are TSH-secreting, gonadotrophin-secreting carcinomas and non-functioning tumours, many of which immunostain for a hormone type even if not released in circulation,103***,104,118 and the plasma levels are indistinguishable from adenomas unless metastases are present. In keeping with this observation is the higher Ki-67% LI in hormonesecreting adenomas, which possibly explains their more aggressive behaviour. Reduction of hormonogenesis has been reported and attributed to tumour dedifferentiation.102,119 The detection of metastases or recurrence of pituitary carcinomas can be improved by using scintigraphy with 111In-labelled octreotide or positron emission tomography (PET) scanning with 18F-labelled deoxyglucose.102,109,120–125 Treatment for pituitary carcinomas usually comprises surgical resection (more often by the trans-sphenoidal approach, at least initially) followed by external-beam radiotherapy and chemotherapy. Surgical resection will be palliative if the tumour invades adjacent structures, or if there is recurrence or metastases. In some cases with distant metastases, resection could lead to clinical and/or biochemical remission, at least temporarily.102 A few series have suggested a better survival rate after aggressive therapy.102,126 External-beam radiotherapy was shown to produce local control of the primary tumour, particularly if used early following surgery for expansive lesions, before the diagnosis of carcinoma is obvious; the results are less good for delayed radiotherapy (established diagnosis of carcinoma).102,127–129 Radiotherapy has also been used to prevent re-growth or recurrence following incomplete resection.102,109,127,130 We have administered radiosurgery (SMART or gamma-knife – see below) after prior conventional radiotherapy for recurrence or metastases.102,131,132 Our group has reported prolonged survival in patients with extra-nervous system metastasis who had received chemotherapy (see below) in the course of their treatment, although the results are highly variable.102,109 Somatostatin
448 Endocrine cancer
analogues have failed to control tumour growth in a number of hormone-secreting pituitary carcinomas.102,109,120,125 Bromocriptine and dopamine agonists should be used early, but often there is resistance or escape in the response of hormonal secretion by pituitary carcinomas (prolactin, ACTH or TSH secreting) (review 103***). Somatostatin analogues for GH-secreting and TSH-secreting carcinomas have shown various responses (review 103***). Interferon-α did not improve patient outcome.109,125
Pituitary radiation technique and dose prescription The conventional external radiotherapy technique for pituitary tumours requires a head-fixation device – usually a plastic head-shell with the patient supine. Modern radiotherapy simulator facilities, together with modern-generation MRI scanning (in both transaxial and coronal planes), allow the field sizes to be minimized. The volume for irradiation comprises the boundaries of the tumour on the imaging procedures plus 0.5 cm in all planes. If a tumour has shrunk following medical therapy, the post-drug imaging tumour size is used. If the tumour has been resected surgically, the preoperative tumour size is used for planning. We have documented evidence of supra-sellar recurrences in such cases referred from other units where the postoperative radiation volume was confined to the fossa. By such planning techniques, a day-to-day set-up reproducible within 2 mm is achieved routinely, and a three-field technique using fixed portals has been adopted – two laterals and an antero-oblique or direct superior field (6–8 MeV X-rays are used at St Bartholomew’s Hospital, London). It is worth stressing that with the advent of computerized planning of radiotherapy for pituitary tumours, it is essential that cognisance of the preoperative MR-defined extent of tumour is made. The dose distribution is shaped ‘bespoke’ for the patient’s tumour using the multi-leaf collimator system with which all modern linear accelerators are equipped – this is conformal radiotherapy. There have been differences in the dose prescriptions, but those publications employing 50 Gy or more have not produced control rates superior to those employing 44–45 Gy. Our recommended prescription to the tumour volume is 45 Gy in 25 fractions over 35 days. The prescription utilizes daily dose fractions of 1.8 Gy and therefore respects the well-documented association between radiation damage to the nervous system and high fraction size. Utilizing this technique and dose prescription in the treatment of more than 700 patients with pituitary adenoma presenting to St Bartholomew’s Hospital, we have not encountered late optic chiasmal damage, but late defects in hypothalamo-anterior pituitary function may occur. Radiation-induced second tumours may also occur, but in our experience are extremely rare. Stereotactically delivered radiotherapy has many attractions. Pituitary adenomas usually represent discrete and
well-demarcated targets that would seem ideal for highly ‘focused’ radiation therapy, and the rapid ‘fall-off’ in the dose at the edges of the treatment volume in these techniques spares adjacent nervous system structures. In early work in the 1970s, Kjellberg in Boston successfully treated many pituitary adenomas with proton-beam therapy, but his work was never published in peerreviewed journals and proton-beam therapy is rarely available; this focal radiation therapy has lapsed. Of the two proton methods, gamma-knife has dosimetric advantages over linear accelerator methods, although newer techniques such as the Cyberknife may soon compete. In the gamma-knife technique, a pinned stereotactic frame and MR mapping define the exact extent of the pituitary adenoma in terms of its geometry, and 201 well-collimated cobalt beams converge exactly on that target to deliver a high single obliterative dose of ionizing radiation within this volume, and very little to surrounding structures. Early stereotactic radiation treatment data were not persuasive, but in recent years good results have been reported. Landolt et al. compared the endocrine results of treating acromegaly by gamma-knife radiosurgery (the term given to ‘one-shot’, high-dose focal therapy by stereotaxy) with conventionally fractionated radiotherapy.133 These workers concluded that there is equivalence with regard to efficacy (albeit with shorter follow-up in the radiosurgery patients), and made the interesting observation of faster normalization of the GH levels in the radiosurgery patients, although others would deny this accelerated response. With good and safe long-term data on the high efficacy of conventionally fractionated radiotherapy in the treatment of pituitary adenomas, our own group has been slower to embrace radiosurgery than some others, and has argued that ‘partial pituitary fossa’ radiation therapy may be too focal in some instances. However, with accumulating and improved data on durable control of pituitary adenomas coming from the focal radiation literature, we are selecting (carefully) more patients for this primary radiation therapy option. Stereotactic radiosurgical techniques are contraindicated where the adenoma abuts the chiasm, as single, high-dose therapy cannot spare this adjacent, radiovulnerable structure. Similarly, radiosurgery is rarely useful for craniopharyngiomas, as the optic chiasm is usually within the target volume. Our own research interest in the use of stereotactic radiosurgery for pituitary adenoma lies in the use of this technique in patients relapsing in the fossa or, especially, the surgically inaccessible cavernous sinuses. Swords et al. reported a series of acromegalics failing due to tumour re-growth after conventional radiotherapy and treated by our linear accelerator radiosurgical technique, with good endocrine and tumour control results and no serious morbidity.131 Of course, our prescribed radiation doses were more conservative in this retreatment situation, but the data were compelling. Our research in this area of salvage
Parathyroid adenoma and carcinoma 449
therapy continues, utilizing the gamma-knife, which gives a dosimetrically superior radiation plan.
(DDAVP®, Ferring, Malmo) subcutaneously (1–2 μg, once or twice daily), intranasally (10–20 μg daily), or orally (100–400 μg twice or three times daily).
Chemotherapy PARATHYROID ADENOMA AND CARCINOMA Although there is no well established role for chemotherapy in the treatment of pituitary adenoma, nevertheless, there is anecdotal evidence of its usefulness in advanced and relapsed cases.102,109 Recently we have been impressed by the usefulness of capecitabine and lomustine (CCNU) in combination and very recently, with the realisation of temozolomide’s efficacy in melanoma and carcinoids, has come anecdotal data demonstrating good responses of recurrent and heavily pre-treated cases of pituitary tumours so responding; we have observed such cases ourselves. This is an area that needs further observation, here and throughout the field of carcinoids/apudoma.
Replacement therapy All patients with pituitary tumours require endocrine evaluation, often with dynamic function tests, to determine their requirements for replacement therapy. In steroid insufficiency, hydrocortisone should be given in a dose necessary to mimic the normal circadian rhythm of cortisol. This is usually 10 mg on rising, 5 mg at around lunchtime and then a further 5 mg in the early evening; this should be doubled during any febrile illness. Parenteral hydrocortisone may be required during surgery and in the presence of vomiting or diarrhoea. Thyroid replacement consists of a dose of thyroxine, usually 0.1–0.2 mg once daily, to normalize circulating thyroxine levels. For gonadal replacement, women should be given cyclical oestrogen and progesterone, or some form of contraceptive pill; men will require injectable depot testosterone every 3–4 weeks or every 3 months, testosterone undecanoate orally, testosterone implants (where available), or one of the newer testosterone patches or gels. Our current preference is for a testosterone gel applied daily or 3-monthly injection of testosterone undecanoat. For fertility, regular injections of FSH and LH, or their analogues, are required, but these should be administered in specialist units. The treatment of GH deficiency in adults remains somewhat controversial. There is good evidence that such deficiency is associated with decreased lean body mass and increased adipose tissue, and undoubtedly with psychological dysfunction.134 The latter usually takes the form of increased lethargy and diminished vitality. There are also data suggesting that mortality, principally associated with premature atherosclerosis, is increased.135 We have generally found a marked improvement in well-being and the quality of life in patients with severe GH deficiency treated with replacement GH in a gradually incremental dose-titration regimen.136 Growth-hormone deficiency should be treated vigorously with biosynthetic GH injections in children. Finally, diabetes insipidus responds to therapy with desmopressin
The normal adult parathyroid glands vary from 3 mm to 6 mm in diameter, and weigh approximately 50 mg. The upper glands arise from the fourth pharyngeal pouch and are fairly constantly located by the upper poles of the thyroid. The lower glands arise, in conjunction with the thymus, from the third pharyngeal pouch, and this thymic association accounts for the not infrequent ectopic location of these glands in the mediastinum. In approximately 80–85 per cent of cases with primary hyperparathyroidism (PHPT; incidence 1:1000 cases), a single benign adenoma is the cause of parathyroid dysfunction, while hyperplasia of all four parathyroid glands occurs in about 15–20 per cent of cases, and carcinoma in 1 per cent.137–139 Parathyroid tumours are commonly sporadic, but in less than 5 per cent of cases they are part of hereditary syndromes such as MEN-1 and MEN-2A, familial isolated hyperparathyroidism (FIHP; the majority with as yet unrecognized genes) or hyperparathyroidism– jaw tumour syndrome (HPT-JT).140 The prevalence of parathyroid carcinoma is higher (around 15 per cent) in patients with HPT-JT.140,141*,142 Several genetic abnormalities have been detected among patients with parathyroid tumours. Cyclin D1/PRAD-1 oncogene, due to a translocation, leads to clonal expansion,143–145 present in 20–40 per cent of parathyroid adenomas.146–149 Bi-allelic abnormalities of the product of MEN-1 gene, a tumour suppressor, were found in 12–20 per cent of patients with sporadic parathyroid tumours.150–153 Other genetic defects affect the calcium-sensing receptor (CaSR) or vitamin D receptor (VDR).154–157 Recently, inactivating germline mutations in the tumour-suppressor gene HRPT2, which encodes for parafibromin, have been associated with parathyroid carcinoma in HPT-JT syndrome or in sporadic cases.141*,158*,159 Interestingly, parafibromin was shown to inhibit cell proliferation and to block cyclin D1 expression.160 In about 10 per cent of carriers the mutation HRPT2 gene remain silent through adulthood.161,162* Very rare (1.8 per cent) somatic mutations of HRPT2 have been detected in sporadic adenomas, in patients without mutations of the MEN-1 gene or in unselected adenomas (0.8 per cent), whereas this number increases to 77 per cent in parathyroid carcinomas.141*,158*,159,163*,164* In addition, Carpten identified germline mutations in 14/24 cases of HPT-JT, with 6/14 families (43%) having parathyroid cancer.141* Although not diagnostic, screening for HRPT2 mutations could be useful in patients with clinical suspicion of malignancy, and may indicate the need for a more aggressive surgical approach and family screening.165*** Histopathological features such as fibrous bands, mitoses and invasion are frequently described in carcinoma, but
450 Endocrine cancer
they can be found in adenomas and are non-specific. Clinical and biochemical features that could suggest malignancy are: mass in the anterior neck; renal (60 per cent) and skeletal (73 per cent) metastases, which often present simultaneously; earlier age of onset (mid-40s versus mid50s in benign adenomas); severe PHPT with a serum calcium often exceeding 3.8 mmol/L and a plasma parathyroid hormone (PTH) of up to or above 1000 pg/mL (immunoradiometric assay). The discovery of a biochemically raised serum calcium, a lowered serum phosphate and an elevated PTH level is diagnostic of the condition. Treatment is by surgical excision, but the critical question concerns which patients should be referred for surgery. Most authorities agree that a patient who has nephrolithiasis, renal dysfunction, nephrogenic diabetes insipidus or a serum calcium 3 mmol/L requires early surgery. Asymptomatic patients may be regularly observed, although recent data suggest that even these patients run a significant risk of osteoporosis, and some 10 per cent will require operation within 10 years.166* Furthermore, many apparently ‘asymptomatic’ patients show improved mood and/or cognition following parathyroidectomy, although this has been difficult to confirm in formal studies. It is therefore a matter of fine clinical judgement if, and when, to offer surgery to such patients. Localization is not usually difficult in the hands of an experienced surgeon, but may be aided by preoperative localization techniques such as subtractive isotope scanning (sestamibi), ultrasound, CT and/or MRI. Probably the optimal current modalities are ultrasound and sestamibi subtractive scanning. Sestamibi scanning is particularly helpful, and indeed should always be used before re-operation following failed surgical exploration.167 Technetium-99 m sestamibi scintigraphy combined with single photon emission computed tomography (SPECT) was shown to have high accuracy and positive predictive value in patients with primary hyperparathyroidism.168 Selective venous sampling for PTH in the draining thyroid veins (also combined with arteriography) is very useful if the parathyroid tumour has not been located by other imaging studies, but this invasive technique remains expensive and needs to be performed by an experienced radiologist.169 The management of parathyroid carcinomas is surgical resection, and its completeness (excision en bloc and involved structures) assessed by postoperative serum PTH levels at the initial intervention determines the outcome. The operation may need to comprise a radical thyroidectomy and formal dissection of neck nodes, but this is not de rigueur, and individual assessment as to the correct operative procedure is necessary. In a pooled series, Schantz and Castleman found a 30 per cent recurrence rate, with less than one-half of patients dying within 5 years of the disease, indicating the very slow growth pattern of this tumour in most cases.170* Recurrence of the tumour is associated with refractory hypercalcaemia, which is often a considerable problem in these patients but may respond to
oestrogen therapy. More recently, mithramycin and certain diphosphonate derivatives such as pamidronate (APD) have also been used to control persistent hypercalcaemia. While pamidronate is less effective in hyperparathyroidism than in malignant hypercalcaemia, either it, or one of its congeners (e.g. clodronate, zoledronate), should always be considered in such patients. Chemotherapy and radiotherapy have little efficacy, although the rarity of this cancer has prevented large trials; dacarbazine has some activity.171*,172* Cinecalcet, an inhibitor of the calcium feedback receptor, may also be useful in controlling hypercalcaemia in patients when surgery has been unsuccessful or is contraindicated.
MEDULLARY THYROID CARCINOMA Discussion of this subject is limited here, as it is treated more extensively in Chapter 17. Medullary thyroid carcinoma (MTC) is a rare tumour, originating from parafollicular C cells (which are considered as part of the amine precursor uptake and decarboxylation (APUD) system) and represents about 3–9 per cent of all thyroid cancers.173 It can occur sporadically (75 per cent) or as part of a familial syndrome (25 per cent) with an autosomal dominant pattern of inheritance, i.e. part of MEN-2A or MEN-2B, or as familial medullary thyroid carcinoma (FMTC) without MEN association. These hereditary malignancies are caused by mutations in the RET proto-oncogene, a tyrosine kinase receptor, located at chromosome 10q11-2. Somatic RET mutations have been also detected in some sporadic MTC. The distinction between a hereditary syndrome and a sporadic manifestation is important, as the clinical presentation, prognosis and therapeutic approach differ. Sporadic MTC is usually a slow-growing indolent tumour, with early metastases to the regional lymph nodes, which are often present at the time of diagnosis; haematogenous dissemination occurs variably. As opposed to these features, hereditary MTC, with nearly 100 per cent penetrance, affects virtually all patients with these syndromes; the tumours are multi-focal, and are more aggressive. Medullary thyroid carcinoma in the setting of MEN-2B is the most aggressive, with invasive carcinoma and lymphnode metastases often present in the first years of life.174*,175* Medullary thyroid carcinoma as part of MEN-2A is the next most aggressive, with hyperplasia or invasive carcinoma with lymph-node metastases detected slightly later but still in the first decade.176*,177* Medullary thyroid carcinoma in the setting of FMTC is less aggressive, classically occurring in the second or even third decade of life. As phaeochromocytomas in the context of MEN have a peak incidence in the fourth to fifth decade of life, some of these MTCs are misclassified as FMTC. The high probability of the hereditary form and the possibility of early effective therapy justify genetic screening, which has tended to replace the calcitonin stimulation tests with calcium gluconate or pentagastrin.
Gastro-entero-pancreatic endocrine tumours 451
Studies have shown that early prophylactic thyroidectomy improves outcomes in patients with MEN-2. Disease-free interval of 4 years was reported at 91.5 per cent out of 71 patients treated with early thyroidectomy.178* Skinner et al. reported that 88 per cent of young patients (carrying genetic mutations for MEN-2A) who underwent total thyroidectomy at a younger age (100 per cent of children operated on before the age of 8 years) were free of persistent or recurrent disease after a mean follow-up of 7 years.179* Recent consensus recommendations advocate total thyroidectomy alone or with central node dissection and parathyroidectomy with auto-transplantation to be done at the earliest age at which the surgical team can perform it safely.173,180,181 Further, the EUROMEN (European Multiple Endocrine Neoplasia) Study Group established an age-dependent progression of early MTC specific to the RET codon.182 Neck re-operation for persistent or recurrent MTC resulted in significant decrease or normalization of calcitonin levels.173 Scintigraphy with 111In-octreotide has better sensitivity than 123I-meta-iodobenzylguanidine (MIBG) (50–75 per cent versus 35–50 per cent) or MRI in detecting occult recurrent MTC.183–186 In terms of screening, it is probably wise to assess any patient with a MTC for hyperparathyroidism or a phaeochromocytoma, and probably to repeat these estimations at intervals unless there is clear evidence that the tumour is indeed sporadic. Certainly, any form of therapeutic intervention without exclusion of a phaeochromocytoma could be extremely hazardous. Systemic chemotherapy with doxorubicin alone or with cisplatin and vindesine has resulted in partial responses in a limited number of patients, or in long-term stabilization following the combination dacarbazine and 5-fluorouracil (5-FU),187* although our own recommendations are those for all apudoma lineage tumours (see below). Symptomatic improvement was observed with interferon-α, immunotherapy or a somatostatin analogue (lanreotide).188* The experience with 131I-MIBG therapy remains limited due to the small number of patients, with partial response or complete remission in isolated cases.184,189*–192* There are ongoing clinical trials with tyrosine kinase inhibitors. The role of external-beam radiotherapy to the neck in controlling cervical disease is unclear as results are incomplete; its palliative benefit in metastatic bone disease is more certain.173
GASTRO-ENTERO-PANCREATIC ENDOCRINE TUMOURS Another major group of apudomas is the gastro-enteropancreatic (GEP) neuroendocrine tumours. Insulinoma is the most common (incidence 1/1000000 per year) and generally presents with ‘Whipple’s triad’ of hypoglycaemic symptoms (headache, slurred speech, pallor, palpitations, sweating/fainting and impaired consciousness) associated with a demonstrable low blood glucose and alleviated by
glucose administration; plasma insulin and/or C-peptide levels remain unsuppressed in the presence of the hypoglycaemia. Gastrinomas have an incidence of approximately 1/million per year, and present with intractable peptic ulceration (the Zollinger–Ellison syndrome). Other functioning tumours include VIPomas (watery diarrhoea syndrome), glucagonomas (rash, weight loss, diabetes), somatostatinomas (diabetes, steatorrhoea, gallstones, weight loss, hypochlorhydria), PPomas (watery diarrhoea syndrome or absence of clinical symptoms), GRFoma (acromegaly due to ectopic secretion of GH-releasing factor) and ACTHoma (ectopic ACTH, Cushing’s syndrome). These are all relatively rare, and overlap with non-pancreatic gut-derived neuroendocrine tumours with and without the carcinoid syndrome. Approximately 90 per cent of insulinomas are benign, but the percentage is considerably less for gastrinomas; between 8 per cent (insulinomas) and 20–25 per cent (glucagonomas, gastrinomas) are associated with MEN-1, where they are multi-centric, and all have a capacity for metastatic spread, although this is rare for insulinomas and, even when present, may be compatible with long survival. The prognosis and survival of gastrinomas of the stomach are according to their WHO classification in three main groups, based on on Ki-67 index, tumour size, invasion and the presence of metastases and tumour differentiation.193,194 Diagnosis is based on clinical suspicion, serum tests and imaging. Serum tests involve a gut hormone profile and other relevant endocrine testing, but in particular plasma chromogranin A has been found to be a useful marker in all these tumours. Conventional CT or MRI (sensitivity of 40–85 per cent) and endoscopic/peri-operative ultrasound have an important contribution in locating the primary tumour and metastases or detecting small pancreatic lesions, respectively. Angiography with calcium gluconate intra-arterial stimulation195 remains indicated in selected cases for the preoperative detection of functional lesion(s), particularly in the context of multi-focal tumours in MEN1 syndrome. Single photon emission computerised tomography seems to provide more accurate localization and detection of lesions.196 Scintigraphy with the somatostatin analogue 111In-labelled octreotide has a superior sensitivity, of 82–95 per cent, for identifying islet-cell tumours and their metastases (80 per cent of gastrinomas, glucagonomas and VIPomas, and 60 per cent of insulinomas express the somatostatin receptor subtypes 2 and 5, and, to some extent, 3).111,122,197,198 111In-pentreotide and 111In-DOTA lanreotide are now widely used. Scintigraphy with 123I-MIBG radionuclide may also allow imaging of GEP lesions preparatory to therapy with 131I-MIBG, and a hand-held detection probe may be used intra-operatively to demonstrate disease. The initial treatment for GEP tumours is surgical removal of all resectable primary and secondary lesions.193 Gut neuroendocrine tumours often pursue an indolent course over many years,199 and where systemic therapy is
452 Endocrine cancer
ineffective or delayed in efficacy, surgical debulking or hepatic arterial embolization of large hepatic metastases may well be indicated. Hepatic thermo-ablation or radiofrequency ablation, or even transplantation, may be proposed for metastases confined to the liver. With regard to the therapy of inoperable or metastatic disease, recent data for 131I-MIBG as therapy suggest that 6-monthly treatment with around 200-mCi (approximately 7-GBq) doses is safe and well tolerated, and may improve survival, although bulk tumour regression is unusual.192* 111In-octreotide and 111In-pentreotide were reported to induce variable responses, but are unlikely to have any major impact as 111In is a very poor therapy isotope (its low energy photon emissions lending themselves better to imaging functions). Partial shrinkage of the tumour, stabilization of disease, clinical and hormonal response, prolonged survival and no change have all been reported (review 103). Yttrium-90(β-emitter)-labelled octreotide, octreotate or lanreotide, perceived to be a much superior therapy isotope due to the pure β-emissions of 90 Y, used alone or in combination with 111In-octreotide, showed partial responses in controlling tumour growth (one-third of patients) and disease (remission for isolated cases, but also progression of disease); the response, while superior compared to 111In-octreotide alone, remains limited. 177Lu-labelled octreotate, a β-emitter and γ-emitter, used alone, in combination or sequentially with 90Y analogues, seems to offer major therapeutic advantages. The few data currently available indicate: 2–3 per cent complete remission, 26–35 per cent partial remission, 35–41 per cent stabilization of disease, 18–21 per cent progressive disease and 10–30 per cent reduction of tumour size.103,200,201 Systemic chemotherapy is recommended for progressive metastatic disease that is negative on radionuclide scanning or after radio-isotopic therapy has failed. The combination of 5-FU and lomustine (CCNU), as demonstrated by work from the Mayo Clinic, or streptozotocin and/or doxorubicin is recommended in less aggressive tumours,193 with a 30–50 per cent symptomatic and/or biochemical response rate and a 10–20 per cent tumour response rate; this is not dissimilar to results from our group treated with 5-FU and lomustine.202 As these tumours regress very slowly, a tumour response is observed after a longer interval than for most other tumours, and any significant volume reduction or similar marker response should not be expected in the first three to six cycles. We have recently further modified this basic ‘Mayo Clinic’ regime to a completely oral regime of lomustine and capecitabine (a 5-FU pro-drug), and we continue to see good responses. For poorly differentiated malignant neuroendocrine tumours, cisplatin (carboplatin) and etoposide chemotherapy, as for small-cell lung cancer, is recommended193 and is currently our optimal therapy for high growth fraction/labelling index tumours. This regimen has resulted in a high response of 69 per cent overall regression rate, with 17 per cent complete regression rate, but still with an overall 2-year survival reported below 20 per cent.203–205
The treatment of patients with malignant pancreatic endocrine tumours with a combination of 5-FU and interferon showed hormonal responses in 73 per cent, reduction of tumour mass (by 50 per cent) in 27 per cent, and a complete response in 9 per cent.206 The extended followup (mean 20 months) showed biochemical and radiological responses in 47 per cent and 12 per cent of cases, respectively.207 In another large review, interferon treatment induced an objective response rate of 44 per cent.208 Our own experience with interferon is less successful than that of Oberg and colleagues. In summary, our systemic recommendations for GEP, as for all malignant apudomas are: if the tumour is positive on radio-isotope scanning in all sites, 131I-MIBG (or 90 Y-octreotide) therapy is given first, unless the tumour demonstrates a very aggressive growth pattern; for such high labelling index tumours we recommend primary treatment with cisplatin and etoposide chemotherapy. For slower growing tumours for which the radio-isotope option is not useful or is exhausted, lomustine/capecitabine oral chemotherapy is recommended. Interferon is kept in reserve in our unit. As noted above, many of these tumours are very slow growing, and symptomatic palliation of the sequelae of their endocrine products is an extremely important part of their management. In particular, diazoxide will often (but not invariably) inhibit neoplastic insulin secretion, while omeprazole (usually in high doses) may protect from the consequences of hypergastrinaemia. In addition, somatostatin derivatives such as octreotide or lanreotide can be very useful in inhibiting hormonal hypersecretion; both are available as long-acting formulations, with lanreotide having a more rapid onset of action. Although some authorities claim tumour shrinkage with somatostatin analogues, this is controversial, and their main role is to control endocrine secretion.
CARCINOID TUMOURS Carcinoid tumours most commonly occur in the gastrointestinal tract and arise from the argentaffin cells, essentially a pathological staining technique that identifies certain types of neuroendocrine cells. The most common sites are the appendix, the small bowel and the rectum. The overall incidence of appendiceal carcinoids is 1 in 150 to 1 in 1000, according to appendix histology reports. The histopathological reporting of appendiceal carcinoids is relatively more common in the appendix specimens of young adults than in the elderly, which mirrors the incidence of argentaffin cells in the body, the total number of which rises and then falls in later life. This observation also implies an involution of some benign carcinoid tumours. Appendiceal carcinoid tumours are almost invariably benign; those rare examples that have metastasized have all been more than 2 cm diameter (itself a remarkable rarity for a carcinoid at this level).
Carcinoid tumours 453
In the Mayo Clinic experience, small-bowel carcinoids occur with the following distribution: duodenum 2 per cent, jejunum 7 per cent, ileum 89 per cent (with the distal ileum being much more commonly the site of origin than the proximal ileum); 2 per cent occur in a Meckel’s diverticulum. They are often multicentric. Most small-bowel carcinoids are small and arise deep in the crypts. Tumours with metastatic potential are almost always more than 2 cm in diameter, but this malignant potential is otherwise difficult to assess by the morphology of the cells. Once the tumour reaches the bowel mesentery, it engenders a massive fibrous reaction. This contracts the mesentery and this ‘encasement phenomenon’ frequently leads to bowel obstruction, the most common presenting feature of extramural small-bowel carcinoid tumours. Metastases almost invariably spread to the liver, to which they may be confined for long periods; metastatic carcinoids have one of the longest doubling times of any malignant human tumour. Despite enormous hepatomegaly, patients may remain well for long periods. Indeed, the mean natural history from operation on a malignant small-bowel carcinoid to death is between 7 and 10 years. The ‘carcinoid syndrome’, manifest in about 5–15 per cent of patients, is a clinical syndrome due to the secretion by carcinoid tumour cells of vasoactive amines, including 5-hydroxytryptamine (5-HT), prostaglandins and tachykinins. The syndrome is virtually confined to patients with bulky liver metastases, when the urine contains abnormally high quantities of the 5-HT catabolite 5-hydroxyindole acetic acid (5-HIAA). There is an approximate correlation between disease severity and the level of urinary 5-HIAA, although there is little doubt that many of the symptoms and signs are not a direct consequence of 5-HT excess. The dominant clinical features of the carcinoid syndrome are the flushing attacks, which may be precipitated by stress and alcohol, as well as calcium and pentagastrin infusions. It is difficult to demonstrate a rise in the blood 5-HT levels during a flushing attack, but recent studies have demonstrated a close association with circulating tachykinins. Diarrhoea, often with colic, is the second major clinical feature of the carcinoid syndrome, and this also does not seem to relate to 5-HT secretion. Other components of the carcinoid syndrome include bronchial asthma and congestive or right-sided cardiac failure due to tricuspid or pulmonary stenosis. Operable disease is resected where possible, e.g. appendicectomy or right hemicolectomy if the appendiceal base nodes and mesentery are involved. Indeed, excision of discrete hepatic metastases of this slowly growing tumour is also perceived to be advantageous. Nevertheless, the tumour usually exhibits a strong desmoplastic response, whereby surrounding tissues readily become bound down to overlying gut, and surgical excision can be very difficult. Similar to for other GEP tumours noted above, radionuclide therapy is of increasing importance, and if the relevant tracer scanning is positive,185 this – notably 131 I-MIBG – may replace chemotherapy as far as initial
systemic therapy is concerned192* for all but the most aggressive tumours. For faster-growing disease, aggressive and atypical carcinoids, the combination of cis-platinum and etoposide is favoured, as for the more malignant isletcell tumours. Atypical carcinoids are almost oat-cell tumours. With regard to other cytotoxic drug therapy, the oral lomustine/capecitabine regime is, as for GEP, our first-choice regime, although the evidence suggests that carcinoids are less sensitive to chemotherapy than islet-cell tumours. Somatostatin analogues effectively control symptoms in patients with the carcinoid syndrome (40–80 per cent of patients), with a reduction of biochemical markers in up to 40 per cent. Stabilization of tumour growth is observed in 24–57 per cent of patients selected with documented tumour progression prior to somatostatin therapy. Partial and complete tumour response has been observed in less than 10 per cent of patients, with partial tumour regression reported in overall 30 patients (reviews in 193,209). The problem of tachyphylaxis to somatostatin analogues occurs in the great majority of patients and contributes to the reduction of the duration of therapeutic response (median 12 months). In 1983, Oberg et al.210 reported that leucocyte interferon therapy ameliorated the carcinoid syndrome due to metastatic midgut carcinoids, and led to prompt decreases in urinary 5-HIAA levels, although they failed to demonstrate shrinkage of tumour masses. The same Swedish group of workers later reported the treatment of 22 patients with advanced, malignant pancreatic apudomas with leucocyte interferon and demonstrated both endocrine and oncological remissions.211 These authors raise the fascinating, yet speculative, hypothesis that for a carcinoid tumour to grow, tumour-derived hormones are required as autocrine growth factors; interferon may control cell growth via inhibition of such factors. Unfortunately, this method has not been substantiated in recent years and few now use interferon in the treatment of this disease. In a review of the literature, the median symptomatic and biochemical response rates were evaluated at 40–70 per cent and 44 per cent, respectively; the median tumour response was only 11 per cent, and disease stabilization was observed in 35 per cent of patients and tumour progression in 15–20 per cent.209 The association of somatostatin analogues and interferon for the treatment of midgut and progressive pancreatic carcinoids was hypothesized to be more effective in patients resistant to either treatment alone. Indeed, the combination was shown to be effective for symptom relief and on biochemical responses, but its effects on tumour regression were limited, although possibly prolonging survival in the responder group.212*,213* A phase II study with docetaxel therapy in patients with metastatic carcinoid led to biochemical responses (in 31 per cent) but no radiological response (radiologically stable disease in 81 per cent) over a median overall survival interval of 24 months.214* More recent phase II trials with sunitinib malate, a multi-targeted tyrosine kinase inhibitor – which
454 Endocrine cancer
inhibits vascular endothelial growth factor (VEGFR)– PDGFR, KIT, RET and FLT3, have shown promising clinical results as a single-agent therapy for advanced neuroendocrine tumours. A partial response was observed in 13 per cent and 2 per cent respectively for islet-cell tumours and carcinoids, stable disease in 75 per cent and 93 per cent, and progressive disease in 7 per cent and 10 per cent, with a median time to tumour progression of 33 and 42 weeks respectively (prior cytotoxic chemotherapy accepted or continuation on octreotide).215* Gefitinib, another receptor tyrosine kinase inhibitor of the EGFR, used in a phase II trial, produced stabilization of disease in 64 per cent and 13 per cent of patients with carcinoids and islet-cell progressive metastatic tumours, with progression-free survival of 6 months (no objective response observed).216* Temsirulimus, a novel inhibitor of mTOR, is currently being used in phase II trials for metastatic neuroendocrine carcinomas: preliminary results show prolonged disease stabilization, and in a few patients tumour shrinkage or clinical benefit.217* Another mTOR inhibitor, everolimus, is also under clinical trial. In refractory cases with massive and painful hepatomegaly or carcinoid syndrome, hepatic artery embolization with gelatin sponge has satisfactorily palliated the disease for some time. Although there were some early reports that low-dose, whole-liver radiotherapy similarly palliated the disease, this optimism may not be well founded. As for appendiceal carcinoids, small rectal carcinoids are usually benign. Rectal carcinoids occur above the dentate line, almost invariably on the anterior or lateral walls, and almost never posteriorly. Larger rectal carcinoids, on the contrary, are usually extremely malignant. Above 13 cm, carcinoids are extremely rare in the sigmoid, descending, transverse and ascending colons, until the caecum and appendix are reached. When colonic carcinoids do occur, they are usually malignant. Tumours more than 2 cm diameter should always be suspected of malignancy, and abdoperineal resections may well be appropriate for the rare, large, rectal carcinoid tumour. Carcinoid tumours may arise at other sites, such as the lung, thymus, ovary, etc. Bronchial carcinoids (discussed more extensively in Chapter 20 are particularly well known and are the most common ‘benign’ bronchial tumour. Bronchial carcinoids arise endobronchially and frequently in the major bronchi, more commonly in the right lung. They usually present clinically due to haemoptysis or bronchial obstruction, and although resection (lobectomy, pneumonectomy) is usually curative, histopathologically these tumours usually show some evidence of local invasion. Indeed, approximately 5 per cent of bronchial carcinoids show nodal metastases and, in a considerably higher proportion, the histopathologist reports the primary tumour as containing ‘cellular atypia’. Atypical carcinoids show more aggressive behaviour and metastasize to mediastinal lymph nodes in 30–50 per cent of cases, with a 5-year survival rate of 40–60 per cent (review218). Somatostatin analogues or interferon used alone or in
combination for treating bronchial carcinoids seem to be ineffective.219* Although gefitinib, an EGFR receptor tyrosine kinase inhibitor, was shown to exert activity in patients with non-small-cell lung carcinoma, it failed to induce a positive response in patients with small-cell lung carcinoma (chemoresistant and chemosensitive NETs: 10.5 per cent stable disease and 89.5 per cent progressive disease with 1-year overall survival rate of 21 per cent have been described.220* In another ongoing trial, patients with advanced carcinoids benefited from prolonged progression-free survival with bevacizumab (monoclonal antibody anti-VEGF), alone or in association with PEG interferon α-2b.216* Temsirolimus, an mTOR inhibitor, significantly extended survival in patients with extensive-stage small-cell lung carcinoma in remission after induction chemotherapy, as shown in a phase II study (median survival 19.8 months, median progression-free survival 5.5 months).221* In conclusion, neuroendocrine neoplasms represent a heterogeneous group in respect to their response to cytotoxic agents, which is site specific. Targeted radio-isotopic therapy is selectively employed in many centres as first therapy for scanning positive and slower growing disease. Chemotherapy is always first employed for fast-growing disease, and we utilize different chemotherapy regimes dependent on the Ki-67 index. Ongoing clinical trials with interferon or novel receptor tyrosine kinase or mTOR inhibitors are being conducted for advanced stages of neuroendocrine carcinomas: the preliminary promising results need to be confirmed.
PHAEOCHROMOCYTOMA Phaeochromocytomas are catecholamine-secreting tumours deriving in 90 per cent of cases from adrenomedullary chromaffin cells. The remaining 10 per cent, known as paragangliomas, arise from extra-adrenal chromaffin tissue located next to sympathetic ganglia or around the origin of the inferior mesenteric artery; other primary sites are the bladder, thorax and carotid arch. Phaeochromocytomas are rare, with a prevalence about 1:6500–1:10 000, and are reported in less than 1 per cent of hypertensive patients. Sporadic phaeochromocytomas are often single and unilateral, whereas familial phaeochromocytomas (about 10 per cent) are often multi-centric, bilateral, have an earlier onset and a lower risk of being malignant. About 10 per cent of phaeochromocytomas are malignant, metastasizing to bone, lung, brain or liver, this proportion being more significant in families with mutations of SDHB (see below). Hereditary phaeochromocytomas (autosomal dominant inheritance) can be part of MEN-2A or MEN-2B (discussed later in this chapter), syndromes associated with mutations in the RET protooncogene. About 10–20 per cent of patients with von Hippel–Lindau disease develop phaeochromocytoma, with an early age of onset. In neurofibromatosis type 1, the
Phaeochromocytoma 455
risk of phaeochromocytomas is about 1 per cent and it occurs in the fifth decade, but because this is a common genetic disease, they represent an important proportion of hereditary tumours. Mutations in the succinate dehydrogenase (SDH) gene were identified in familial paraganglioma syndromes.222 Allele-specific LOH has been detected at the chromosome 2 locus in phaeochromocytomas.223 However, as phaeochromocytomas are identified in families with as yet unidentified genetic mutations, it is likely that other genes are involved in the aetiology of the disease.224,225 Clinical manifestations of phaeochromocytomas result from catecholamine release by the tumour, which may occur at rest or be induced by physical activity, emotion, tumour stimulation, tyramine-rich food or certain drugs. Hypertension, either paroxysmal or sustained (in 50 per cent of patients), is the most common feature, found in more than 90 per cent. Episodes of hypotension, either postural or alternating with hypertension, may also occur. Headache is also common, in about 80–90 per cent of cases, and sweating, mainly truncal, is seen in 60–70 per cent. The classical ‘intermittent attacks’ comprise headache, sweating and palpitations, and often pallor and anxiety in a hypertensive patient, and are highly suggestive of phaeochromocytoma. The duration of these paroxysmal episodes is variable. Biochemical diagnosis of phaeochromocytoma is based on the demonstration of increased levels of catecholamine or their metabolites. Plasma catecholamine levels are of lesser sensitivity than total urinary catecholamine levels over a 24-hour period but are useful if measured during a suspect event. Because of the low sensitivity of catecholamine metabolites such as VMAs, direct catecholamine assay (HPLC/ECD or GCMS) is used in most major centres. However, there is increasing evidence that plasma or urinary metanephrine, where available, has the highest diagnostic sensitivity. Provocative tests (glucagon) or suppression tests (clonidine, pentolinium) may be useful for differentiating elevated circulating catecholamines from phaeochromocytoma, but none is universally favoured. Usually adrenal phaeochromocytomas secrete high levels of epinephrine, most extra-adrenal tumours or paragangliomas produce high levels of norepinephrine, and metastatic phaeochromocytomas often also secrete high levels of dopamine. Conventional CT and MRI scanning makes an important contribution to localizing small adrenal extra-adrenal or metastatic lesions, but despite a high sensitivity (95 per cent and more), the specificity remains about 65–75 per cent. Venous catheterization with sampling at different levels in the inferior vena cava or elsewhere allows confirmation and localization of tumour origin or multiple lesions. It is safer than arteriography as it is unlikely to provoke a hypertensive crisis, but patients should still be prepared with full adrenoceptor blockade. Octreotide scintigraphy has shown only limited applicability, whereas 123 I-MIBG scintigraphy offers better specificity (95–100 per cent) and reasonable sensitivity (78–83 per cent) for
tumours with catecholamine uptake.226 More recently, the use of PET coupled with 6-18F-fluorodopamine seems to offer superior results for the detection and localization of primary adrenal or extra-adrenal tumours and metastatic lesions.227 The treatment of phaeochromocytoma requires surgical resection, which usually implies adrenalectomy and exploration of the contralateral adrenal. Complete surgical resection is usually employed in localized malignant phaeochromocytomas, but debulking in metastatic tumours may still be helpful. Surgical intervention necessitates preoperative preparation of the patient with α-adrenoceptor (phenoxybenzamine) and β-adrenoceptorblocking agents and restoration of normal blood volume, and expert anaesthetic management is crucial during the procedure. Radiation therapy has not been proven to be effective but may help in controlling symptomatic bone disease. Chemotherapy with cyclophosphamide, vincristine and dacarbazine is partially successful, with 57 per cent tumour reduction and 79 per cent biochemical response.228 Based on the β-emitting 131I isotope coupled to MIBG, this therapy is used in several centres but remains expensive. In advanced disease, 131I-MIBG produces a partial therapeutic response in about one-third of patients and stabilization of disease in 82 per cent over an extended period with hormonal control (Fig. 18.3).192*,229,230 It is at least equally important to produce adequate adrenergic blockade to avoid pressor crises. Where phenoxybenzamine is poorly tolerated, newer alpha-blockers such as doxazosin may also be tried. Radiolabelled octreotide therapy is also available, but experience is very limited and the therapeutic response is variable.231
Figure 18.3 123I-MIBG scan of a patient with a metastatic paraganglioma. Metastatic deposits are seen in the mediastinum and in the skull. Uptake in the nasopharynx and salivary glands is not pathological.
456 Endocrine cancer
The adrenal medullary neuroblastomas are discussed in Chapter 40.
MULTIPLE ENDOCRINE NEOPLASIA This title describes a group of familial syndromes in which more than one endocrine gland in an individual undergoes hyperplasia or tumour formation. In 1954, Wermer described the association of tumours of the parathyroid, pituitary and pancreas within individuals and their families – noting an autosomal dominant inheritance.232 This syndrome is referred to as Wermer’s syndrome or MEN-1. In 1961, Sipple reported a patient with two phaeochromocytomas and a malignant thyroid tumour;233 Sipple’s review of the literature demonstrated that there was, indeed, an association between phaeochromocytoma and thyroid cancer (now known to be MTC). This association is now referred to as Sipple’s syndrome or MEN-2A, and is also inherited as an autosomal dominant condition, often with parathyroid adenomas. A variant syndrome (MEN-2B) has been described in which, in addition, the patient manifests mucosal neuromas and other phenotypic features, but without parathyroid disease. The gene for both MEN syndromes has now been identified, although its precise function is still unclear.
MEN-1 (Wermer’s syndrome) Ballard et al. reviewed 85 patients with this syndrome.234 In order of frequency, the involved endocrine glands were: parathyroids (88 per cent), pancreatic islets (80 per cent), anterior pituitary (65 per cent), adrenal cortex (38 per cent) and thyroid parafollicular cells (19 per cent). Pairs of glands were involved in 60 per cent of cases and all of the first three glands in 40 per cent. These early figures have now been reviewed in several large series from the USA and Europe, and they appear to be broadly compatible. There is also an association with lipomas and carcinoids, at least in certain families. The parathyroid glands usually undergo hyperplasia or adenoma formation, and are usually the first abnormality to be noted. The islet cells of the pancreas develop either an adenoma or a carcinoma or, more rarely, diffuse hyperplasia, and often the tumours are multi-focal. The pituitary gland develops adenomas, usually but not always secretory; there is a particular predilection for prolactinoma formation. The adrenal cortex develops either hyperplasia or adenomas. Thyroid adenomas are also well documented. Adenomas in all these sites may be functioning or non-functioning. The gene mutated has been located at chromosome 11q13 and identified as producing a peptide product, menin, which binds to and may repress the transcription factor JunD. Unfortunately, there is no common area of mutation throughout this large gene, and little genotype–phenotype correlation. Genetic screening is
therefore not usually possible outside of a research setting.235 Indeed, about 20 per cent of MEN-1 kindreds do not carry a mutation in the menin gene. Loss of heterozygosity, which affects the non-mutated allele from the normal chromosome, has been detected in endocrine tumours from patients with MEN-1 syndrome. Thus, patients presenting with or suspected of having MEN-1 syndrome should be screened carefully: serum calcium and intact PTH for hyperparathyroidism; measurement of serum gastrin for gastrinoma; blood glucose, insulin and pro-insulin serum levels for insulinoma; glucagon, pancreatic polypeptide, GIP and VIP for other gastro-pancreatic endocrine tumours; 24-hour urinary 5HIAA for carcinoid; serum prolactin levels and, less frequently, IGF-1 levels or ACTH for other secretory pituitary adenomas as well as MRI of the pituitary. As these tumours may present sequentially, the diagnostic screening tests should be performed periodically. In the absence of any features, the authors recheck serum calcium and prolactin, and fasting gut peptides and a pancreatic MRI, at 6–12month intervals.
MEN-2 (Sipple’s syndrome) Usually, but not invariably, this autosomal dominantly inherited condition presents in young adult life. In the most common form, a MTC is associated with a phaeochromocytoma, both tumours characterized by high malignant potential. This association is common enough to warrant screening for the other tumours (by calcitonin or urinary catecholamine estimations) in any patient presenting with either one. Occasionally, patients present with parathyroid hyperplasia or adenoma (5–20 per cent). In other patients, the MTC and phaeochromocytoma occur with multiple small, subcutaneous and/or submucosal neuromas of the oral cavity and lips, plus autonomic ganglioneuromatosis and a Marfanoid habitus. Patients with genetic forms of MTC have diffuse bilateral C-cell hyperplasia of the thyroid, or frank MTC. In the setting of MEN2A or MEN-2B, patients may also have adrenal hyperplasia preceding the development of phaeochromocytoma. These syndromes appear to result from a neural-crest dysplasia during embryonic development. The gene involved is the ret oncogene, encoding a membrane-spanning peptide with multiple cysteines, which acts as the receptor for glialcell-derived neurotropic factor (GDNF). Mutations are clustered around a small number of ‘hot spots’, which renders screening relatively straightforward and which has greatly assisted in patient counselling.236,237 More recently, it has been recognized that these syndromes are not mutually exclusive, and that occasionally patients may be seen with features of both MEN-1 and MEN-2 syndromes – so-called ‘overlap MEN’. With regard to management, each endocrine disease should be treated on its own merits, but the clinician must
Adrenocortical tumours 457
remain wary to the possible development of other aspects of disease. Thus, it is reasonable to screen all phaeochromocytoma patients for the presence of a MTC (serum calcitonin levels), and vice versa (plasma and 24hour urinary catecholamines), while patients with pancreatic endocrine tumours require, at the very least, assessment of serum calcium and radiology of the pituitary fossa. Repeat screening is indicated, as these tumours may present sequentially. Where the gene mutation has been identified, it is now common to offer total thyroidectomy to all affected individuals with MEN-2 at, or near, 5 years of age. For MEN-2B, prophylactic thyroidectomy may be necessary even earlier (within the first 2 years of life). The cloning of the genes for these conditions has considerably altered screening policies and follow-up plans for these families.
Other genetic disorders associated with phaeochromocytomas The von Hippel–Lindau syndrome (mutations in VHL tumour suppressor gene) is an autosomal dominant condition in which tissues appear to sense hypoxia inappropriately due to mutations of the gene product elongin: this leads to a failure to break down VEGF, which in turn causes the development of vascular malformations in the eye (retinal angiomas) and central nervous system, especially cerebellar and spinal haemangioblastomas.238 In addition, there is a high incidence of phaeochromocytomas, especially bilateral. However, for reasons that are unclear, there is also a very high incidence of renal-cell carcinomas; these are prone to be multiple and recurrent, and patients may require sequential partial nephrectomies followed at some stage by bilateral nephrectomy and longterm dialysis. The optimal screening strategy is not well defined, but any renal lesion on imaging screening must be treated as possibly malignant until proven otherwise. The gene is very large and there are no ‘hot spots’, so family screening is only possible if the index mutation has been identified. In neurofibromatosis type 1 (mutations in tumor suppressor NF1 gene), phaeochromocytomas occur with a frequency of around 1 per cent. The gene is extremely large and screening is a research procedure. They should be treated as for the sporadic disease. More recently, mutations in the SDH gene subunits B, C and D were identified in familial paraganglioma syndromes type 4 (PGL4), type 3 (PGL3) and type 1 (PGL1).222
ADRENOCORTICAL TUMOURS Although benign adrenal cortical adenomas are not infrequently found at autopsy, most are non-functioning and of no clinical importance. The majority of functioning cortical
adenomas are found in female patients. Those occurring in pre-pubertal patients tend to present with virilization, whereas those in post-pubertal patients present with Cushing’s syndrome. An adrenal cortical adenoma producing aldosterone (Conn’s syndrome) is a very rare cause of hypertension. Diagnostic tests include the endocrine demonstration of non-suppressible and excessive levels of circulating adrenal steroids, and appropriate diagnostic imaging tests (e.g. MR/CT scanning). Treatment is by surgical excision (adrenalectomy). Increasingly, small adrenal tumours are being detected on imaging of the abdomen by CT or MRI for an ostensibly unrelated reason – the so-called ‘incidentaloma’. Some radiologists have reported specific features at MRI allowing accurate distinction of the type of lesion: an intermediate to high signal intensity on T2-weighted images for malignant adrenocortical tumours; low signal intensity for nonfunctional adenomas; and extremely high signal intensity for phaeochromocytomas.239 Positron emission tomography with 18F-fluorodeoxyglucose (FDG-PET) has been shown to detect metastases and recurrence of adrenocortical tumours with sensitivity of 96 per cent, specificity of 99 per cent and accuracy of 98 per cent.240,241 Various algorithms have been proposed to differentiate tumours that require surgical removal from those that can be simply monitored, most attempting to exclude a hypersecretory state biochemically and assess the probability of adrenal malignancy on imaging criteria.242 In general, small (3 cm) lesions with ‘benign’ imaging characteristics can be simply observed and rescanned at intervals, although the recent finding of ‘subclinical Cushing’s syndrome’ in a significant minority, if not the majority, of such patients renders the production of clear clinical guidelines extremely difficult. Adrenal carcinoma is a rare malignancy, accounting for 0.2–0.5 per cent of all cancers, with an annual incidence of 0.5–2 cases per million population, and it accounts for 0.2 per cent of all cancer deaths.243,244*,245 It affects women slightly more often than men,246 and the mean age of presentation in women may be lower than in men.247* Overall, the disease tends to afflict a younger age group than most carcinomas (median age at presentation 37–55 years) but is also found in children less than 10 years of age. The relative incidence of functioning to non-functioning carcinomas is equal; it seems to be unaffected by the age of presentation and may not be different between the sexes, although in the MD Anderson Hospital series, more women had functioning tumours.247* Interestingly, the left adrenal has been documented as the more common site of primary disease;246,247* rarely, the disease is bilateral. Advances in the genetics of adrenocortical cancer have identified a number of genes involved in these malignant tumours and some associated syndromes (reviews 248–250). Inactivating mutations in the TP53 tumour suppressor gene cause the Li-Fraumeni syndrome, which is characterized by adrenocortical cancer and other
458 Endocrine cancer
malignancies, and is detected in unrelated patients in Southern Brazil, where a high incidence of adrenocortical carcinoma is observed. Alterations at chromosome 11p15 locus, where the IGF2, H19 and CDKN1C (p57Kip2) genes are located, are associated with Beckwith–Wiedemann syndrome, in which adrenocortical carcinoma is present. Loss of heterozygosity of APC tumour suppressor gene – mutations associated with adenomatosis polyposis coli (Gardner syndrome) – has been found in adrenal tumours from patients affected by this condition. More recently, it was reported that adrenocortical carcinoma may be present in patients with MEN-1 syndrome (mutations in the menin gene at locus 11q13). Also, patients with Carney complex or McCune–Albright syndrome, associated with mutations of PRKAR1A (locus 17q22-24)49 or GNAS genes, respectively, can present with functional or nonfunctional adrenal lesions, although malignancy has not been reported. Loss of chromosome 17p (containing the TP53 gene) and structural rearrangement of chromosome 11p15 (with consequent over-expression of IGF2) seem to correlate with malignant phenotype in these tumours. Gain of chromosomes 4 and 5 and loss of chromosomes 2, 11p and 17p have also been detected in adrenocortical carcinomas. The most common clinical presenting feature of this disease is abdominal symptomatology (fullness, indigestion, nausea, vomiting, pain or the patient finding an abdominal lump). However, other prominent symptoms include weight loss, weakness, fever, features of tumour function and symptoms due to metastases. Between onequarter and one-third of patients with primary adrenal carcinoma have clinical evidence of endocrine dysfunction at presentation, most commonly Cushing’s syndrome, often supplemented by virilism. A slightly higher fraction of the patients have chemical endocrine evidence of abnormal hormone secretion, i.e. subclinical dysfunction. Obviously, the weight gain of Cushing’s syndrome may be absent from patients with an aggressive malignant tumour and the virilism may be more marked. Feminizing tumours due to over-secretion of oestrogens are rare (10 per cent), as are aldosterone-secreting tumours (2 per cent).251 As for suspected cortical adenomas, diagnostic tests include the demonstration of excessive and non-suppressible levels of adrenal steroids and relevant abdominal diagnostic imaging. At least half of all patients presenting with adrenal cortical carcinoma will have metastatic disease at the time of diagnosis, the most common sites of spread being the lung, liver, peritoneum and abdominal nodes. For patients with localized disease or disease apparently confined to this region, radical surgery is the recommended definitive treatment, with an appreciable cure rate for early disease ‘completely’ resected. Radiotherapy to the tumour bed is recommended if there is disease at resection margins or in regional nodes, and we use a parallel opposed pair of megavoltage portals to 40 Gy mid-plane in 20 fractions, considering the position of the adjacent kidney carefully.
For patients with more advanced adrenal carcinomas, there is no curative treatment, and orthodox cytotoxic chemotherapy has yet to make any impact on this disease. Nevertheless, there is one specific therapy: o⬘p⬘-DDD (mitotane).252* This drug causes necrosis and atrophy of normal adrenal tissue and also of differentiated adrenal carcinoma cells. In the original study, a steroid response rate of 72 per cent was recorded, with an objective regression of tumour in 34 per cent. Other workers have subsequently confirmed this drug’s usefulness in adrenal carcinoma, but responses may be slow (months) and there may be associated gastrointestinal (vomiting and diarrhoea) and neuromuscular (lethargy and weakness) side effects in many patients so treated. However, recent data suggest that doses in the region of 1–3 g per day are as effective as much higher doses, and are better tolerated. The dose is slowly accumulative, and measurement of the drug level 2–4 months after treatment will allow precise control of therapeutic dosimetry. The drug is not myelosuppressive. Glucocorticoid cover is essential for patients receiving o⬘p⬘-DDD therapy. Alternative drugs are suramin and gossypol, but neither has been shown to have a major impact on disease progression. Cushing’s syndrome due to adrenal carcinoma may also be palliated by metyrapone therapy (250 mg to 1 g four times daily, commencing at the lower dose). Metyrapone inhibits 11αhydroxylase (the enzyme converting the metabolically inactive 11-deoxycortisol to cortisol), as previously discussed under Cushing’s disease. When used as sole therapy, metyrapone may be highly effective, although the dose may require careful adjustment so as not to cause an Addisonian crisis. Some patients experience pronounced gastrointestinal upset on metyrapone, and occasional allergic reactions have been documented. Metyrapone also shunts steroid precursors into androgen precursors and may cause marked virilism. The antifungal drug ketoconazole is also showing increasing promise. It should be noted that in any patient with Cushing’s syndrome secondary to an adrenal tumour, removal or ablation of the tumour will need to be followed by long-term steroid-replacement therapy until the suppressed hypothalamo-pituitary-adrenal axis has recovered function; this may take several years. Similarly, removal of a Conn’s tumour is often followed by transient hyperkalaemia. Chemotherapy has a poor track record and our single-agent data using taxol, carboplatin and didadriamycin have not shown any major impact on disease progression. Novel therapeutic agents are currently under trial (review 250). Competitive inhibitors of the MDR1-mediated drug transport (multi-drug resistance protein, which acts as an adenosine phosphate (ATP)-dependent drug efflux pump, transporting out of the cell hydrophobic cytotoxic agents) have shown low rates of response, but a more potent compound, the non-competitive inhibitor Tariquidar, is in phase II study. Gefitinib, an EGFR inhibitor, is already used in clinical trials and results are awaited.
References 459
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
●
Pituitary tumours represent around 10 per cent of intra-cranial tumours and may cause problems due to local compression, hormone hypersecretion or hypopituitarism. For the majority of tumours, primary therapy is trans-sphenoidal surgery, followed where appropriate by external-beam radiotherapy to diminish the chance of recurrence. Prolactin-secreting tumours can often be shrunk with dopamine agonist therapy, thereby avoiding pituitary surgery. Growth-hormone hypersecretion by growthhormone-secreting tumours may be well controlled medically with somatostatin analogues, while awaiting the effect of more definitive treatment. Parathyroid adenomas are common and best treated surgically. The outcome of familial medullary thyroid carcinoma is improved if early preventive surgical resection is performed. Modern chemotherapy regimens can often provide useful life extension in neuroendocrine tumours of the gut, while hormone oversecretion will usually respond to somatostatin analogue therapy. Radionuclide therapy with labelled MIBG or a somatostatin derivative should be considered in most cases. Phaeochromocytomas are an uncommon cause of hypertension. They are often not diagnosed in life and may be associated with a variety of genetic disorders. Adrenal ‘incidentalomas’ are being found increasingly. Treatment protocols are designed to exclude hypersecretory states and adrenal carcinoma. Adrenal carcinomas are often highly malignant and refractory to therapy, but the adrenolytic drug mitotane is usually worth a therapeutic trial.
KEY REFERENCES Bevan JS, Webster J, Burke CW, Scanlon MF. Dopamine agonists and pituitary tumor shrinkage. Endocr Rev 1992; 13:220–40. Chew SL, Eng C. Multiple endocrine neoplasia type 2 and related genetic conditions. Curr Opin Endocrinol Dis 1995; 2:121–6. Drake WM, Perry LA, Hinds CJ, et al. Emergency and prolonged use of intravenous etomidate to control hypercortisolemia in a patient with Cushing’s syndrome and peritonitis. J Clin Endocrinol Metab 1998; 83:3542–4.
Newell-Price J, Trainer P, Besser M, Grossman AB. The diagnosis and differential diagnosis of Cushing’s syndrome and pseudo-Cushing’s states. Endocr Rev 1998; 19: 647–72. Silverberg SJ, Shan E, Jacobs TP, et al. A 10-year prospective study of hyperparathyroidism with or without parathyroid surgery. N Engl J Med 1999; 341:1249–55. Turner HE, Stratton IM, Byrne JV, et al. Audit of selected patients with nonfunctioning pituitary adenomas treated without irradiation – a follow-up study. Clin Endocrinol 1999; 51:281–4.
REFERENCES 1 Pearse AGE. Common cytochemical and ultrastructural characteristics of cells producing polypeptide hormones (the APUD series) and their relevance to the thyroid and ultimobronchial C cells and calcitonin. Proc R Soc Lond 1968; 170:71–80. 2 Pearse AGE Embryology of the diffuse neuroendocrine system and its relationship to the common peptides. Fed Proc 1979; 38:2288–91. 3 Ross RJM, Grossman A, Bouloux P, Rees LH, Doniach I and Besser G.M. The relationship between serum prolactin and immunocytochemical staining for prolactin in patients with pituitary macroadenomas. Clin Endocrinol 1985; 22:227–36. 4 Symon L and Jakubowski J Transcranial management of pituitary tumours with suprasellar extension. J Neurol Neurosurg. Psychiatry 1979; 42:123–33. 5 Law, ER and Jane Jr, JA. Neurosurgical approach to treating pituitary adenomas. Growth Horm IGF Res 2005; 15: S36–41. 6 Brada M, Rajan B, Traish D et al. The long-term efficacy of conservative surgery and radiotherapy in the control of pituitary adenomas. Clin Endocrinol 1993; 38:571–5. 7 Wass JA, C Williams J, Charlesworth M et al. Bromocriptine in management of large pituitary tumours. Br Med J 1982; 284:1908–11. 8 Bassetti M, Spada A, Pezzo G and Giannattasio G Bromocriptine treatment reduces the cell size in human macroprolactinomas: a morphometric study. J Clin Endocrinol Metab 1984; 58:268–73. 9 Webster J, Piscitelli G, Polli A et al. Dose-dependent suppression of serum prolactin by cabergoline in hyperprolactinaemia: a placebo controlled, double blind, multicentre study. European Multicentre Cabergoline Dosefinding Study Group. Clin Endocrinol 1992; 37:534–41. 10 Webster J, Piscitelli G, Polli A, Ferrari CI, Ismail I and Scanlon MF. A comparison of cabergoline and bromocriptine in the treatment of hyperprolactinemic amenorrhea. N Engl J Med 1994; 331:904–9. 11 Biller BMK, Molitch ME, Vance ML et al. Treatment of prolactin-secreting macroadenoma with the once-weekly dopamine agonist cabergoline. J Clin Endocrinol Metab 1996; 81:2338–43.
460 Endocrine cancer
12 Verhelst J, Abs R, Maiter D et al. Cabergoline in the treatment of hyperprolactinemia: a study of 455 patients. J Clin Endocrinol Metab 1999; 84:2518–22. 13 Vance ML, Cragun JR, Reimnitz C et al. CV 205–502 treatment of hyperprolactinemia. J Clin Endocrinol Metab 1989; 68:336–9. 14 Newman CB, Hurley AM and Kleinberg DL. Effect of CV 205–502 in hyperprolactinaemic patients intolerant to bromocriptine. Clin Endocrinol 1989; 31:391–400. 15 Johnston DG, Hall K, Kendall-Taylor P et al. Effect of dopamine agonist withdrawal after long-term therapy in prolactinomas: studies with high-definition computerised tomography. Lancet 1984; 2:187–92. 16 Thorner MO, Perryman RL, Rogol AD et al. Rapid changes of prolactinoma volume after withdrawal and reinstitution of bromocriptine. J Clin Endocrinol Metab 1981; 53:180–4. 17 van t’Verlaat JW and Croughs RJ. Withdrawal of bromocriptine after long-term therapy for macroprolactinomas: effect an plasma prolactin and tumour size. Clin Endocrinol 1991; 34:175–8. 18 Bevan JS, Webster J, Burke CW and Scanlon MF. Dopamine agonists and pituitary tumor shrinkage. Endocr Rev 1992; 13:220–40. 19 Tsagarakis S, Grossman A, Plowman PN et al. Megavoltage pituitary irradiation in the management of prolactinomas: long-term follow-up. Clin Endocrinol 1991; 34:399–40. 20 Grossman A, Lytras N, Savage MO et al. Growth hormonereleasing factor: comparison of two analogues and demonstration of hypothalamic defect in growth hormone release after radiotherapy. BMJ 1984; 288:1785–7. 21 Blacklay A, Grossman A, Ross RJM et al. Cranial irradiation for cerebral and nasopharyngeal tumors in children – evidence for the production of a hypothalamic defect in growth hormone release. J Endocrinol 1986; 108:25–9. 22 Lam KSL, Wang C, Yeung TT et al. Hypothalamic hypopituitarism following cranial irradiation for nasopharyngeal carcinoma. Clin Endocrinol 1986; 24:643–51. 23 Colao A, Di Sarno A, Cappabianca P et al. Withdrawal of long-term cabergoline therapy for tumoral and nontumoral hyperprolactinemia. N Engl J Med 2003; 349:2023–33. 24 Breidahl HD, Topliss DJ and Pike JW Failure of bromocriptine to maintain reduction in size of a macroprolactinoma. Br Med J 1983; 287:451–2. 25 Esiri M.M, Bevan JS, Burke CW et al. Effect of bromocriptine treatment on the fibrous tissue content of prolactinsecreting and nonfunctioning macroadenomas of the pituitary gland. J Clin Endocrinol Metab 1986; 63:383–8. 26 Miyai K, lchihara K, Kondo K and Mori S. Asymptomatic hyperprolactinaemia and prolactinoma in the general population – mass screening by paired assays of serum prolactin. Clin Endocrinol 1986; 25:549– 54. 27 Davis JRE, Selby C and Jeffcoate WJ. Oral contraceptive agents do not affect serum prolactin in normal women. Clin Endocrinol 1984; 20:427–34. ●28 Prescott RWG, Johnston DG, Taylor PK et al. The inability of dynamic tests of prolactin and TSH secretion to
29
30
31
32
33
34
35
36
37
38
39 40
41
42
43
44
differentiate between tumorous and non-tumorous hyperprolactinaemia. J Endocrinol Invest 1985; 8: 49–54. Grossman A, Ross R, Charlesworth M et al. The effect of dopamine-agonist therapy on large functionless pituitary tumours. Clin Endocrinol 1985; 22:679–86. Schlechte JA, Dolan K, Sherman B et al. The natural history of untreated hyperprolactinemia: a prospective analysis. J Clin Endocrinol Metab 1989; 68:412–18. Rjosk H-K, Fahlbusch R and von Werder K. Spontaneous development of hyperprolactinaemia. Acta Endocrinol 1982; 100:333–6. Randall RV, Laws ER, Abboud CF et al. Transsphenoidal microsurgical treatment of prolactin-secreting pituitary adenomas, results in 100 patients. Mayo Clinic Proc 1983; 58:108–21. Bevan JS, Adams CBT, Burke CN et al. Factors in the outcome of transsphenoidal surgery for prolactinomas and non-functioning pituitary tumour, including pre-operative bromocriptine therapy. Clin Endocrinol 1987; 26: 541–56. Koppelman MCS, Kurtz DW, Morrish KA et al. Vertebral body bone mineral content in hyperprolactinemic women. J Clin Endocrinol Metab 1984; 59:1050–4. Ciccarelli E, Savino L, Carlevatto V et al. Vertebral bone density in non-amenorrheic hyperprolactinaemic women. Clin Endocrinol 1988; 28:1–6. Klibanski A, Biller BMK, Rosenthal DI et al. Effects of prolactin and estrogen deficiency in amenorrheic bone loss. J Clin Endocrinol Metab 1988; 67:124–30. Corenblum B and Donovan L. The safety of physiological estrogen plus progestin replacement therapy and with oral contraceptive therapy in women with pathological hyperprolactinemia. Fertil Steril 1993; 59:671–3. Spark RF, Wills CA, O’Reilly G et al. Hyperprolactinaemia in males with and without pituitary macroadenomas. Lancet 1982; ii, 129–32. Kreutzer J and Fahlbusch R. Diagnosis and treatment of pituitary tumors. Curr Opin Neurol 2004; 17:693–703. Harris PE, Afshar F, Coates P et al. The effects of the transsphenoidal surgery on endocrine function and visual fields in patients with functionless pituitary tumours. QJM 1989; 265:417–27. Lohmann T, Trantakis C, Biesold M et al. Minor tumor shrinkage in non functioning pituitary adenomas by longterm treatment with the dopamine agonist cabergoline. Pituitary 2001; 4:173–8. Pivonello R, Matrone C, Filippella M et al. Dopamine receptor expression and function in clinically nonfunctioning pituitary tumors: comparison with the effectiveness of cabergoline treatment. J Clin Endocrinol Metab 2004; 89:1674–83. Gittoes NJL, Bates AS, Tse W et al. Radiotherapy for nonfunctioning pituitary tumours. Clin Endocrinol 1998; 48:331–7. Turner HE, Stratton IM, Byrne JV, Adams CBT and Wass JAH. Audit of selected patients with nonfunctioning pituitary
References 461
45 46
47
48
49
50
51
52
53 54
55
56
57
■58
59
adenomas treated without irradiation – a follow-up study. Clin Endocrinol 1999; 51:281–4. Nabarro JDN. Acromegaly. Clin Endocrinol 1987; 26:481–512. Bengtsson BA, Eden S, Ernest I et al. Epidemiology and longterm survival in acromegaly. A study of 166 cases diagnosed between 1955 and 1984. Acta Med Scand 2000; 233:327–35. Pack SD, Kirschner LS, Pak E et al. Genetic and histologic studies of somatomammotropic pituitary tumors in patients with the “complex of spotty skin pigmentation, myxomas, endocrine overactivity and Schwannomas” (Carney complex). J Clin Endocrinol Metab 2000; 85:3860–65. Casey M, Vaughan CJ, He J et al. Mutations in the protein kinase A R1α regulatory subunit cause familial cardiac myxomas and Carney complex. J Clin Invest 2000; 106:R31–8. Kirschner LS, Carney JA, Pack SD et al. Mutations of the gene encoding the protein kinase A type I-α regulatory subunit in patients with the Carney complex. Nat Genet 2000; 26:89–92. Landis CA, Masters SB, Spada, A et al. GTPase inhibiting mutations activate the alpha chain of Gs and stimulate the adenylyl cyclase in human pituitary tumours. Nature 1989; 340:692–6. Spada A, Arosio M, Bochicchio D et al. Clinical, biochemical, and morphological correlates in patients bearing growth hormone-secreting pituitary tumors with or without constitutively active adenylyl cyclase. J Clin Endocrinol Metab 1990; 71:1421–6. Akintoye SO, Chebli C, Booher S et al Characterization of gsp-mediated growth hormone excess in the context of McCune-Albright syndrome. J Clin Endocrinol Metab 2002; 87:5104–12. Frohman LA and Eguchi K Familial acromegaly. Growth Horm. IGF Res 2004; 14:S90–6. Soares BS, Eguchi, K and Frohman LA Tumor deletion mapping on chromosome 11q13 in 8 families with isolated familial somatotropinomas and in 15 sporadic somatotropinomas. J Clin Endocrinol Metab 2005; 90:6580–7. Jenkins PJ, Besser G.M and Fairclough PD. Screening guidelines for colorectal cancer and polyps in patients with acromegaly. Gut 1999; 51: Suppl. 5, V13–4. Hankinson SE, Willett WC, Colditz GA et al. Circulating levels of insulin-like growth factor-I and risk of breast cancer. Lancet 1998; 351:1393–6. Chan, JM, Stampfer MJ, Ma J et al. Insulin-like growth factor-I (IGF-I) and IGF binding protein-3 as predictors of advanced-stage prostate cancer. J National Cancer Institute 2002; 94:1099–106. Giustina A, Barkan A, Casanueva, FF et al. Criteria for cure of acromegaly: a consensus statement. J Clin Endocrinol Metab 2000 85:526–9. Swearingen B, Barker FG, Katznelson L et al. Long-term mortality after transsphenoidal surgery and adjunctive therapy for acromegaly. J Clin Endocrinol Metab 1998; 83:3419–26.
■60
61
62
63
64
65
66
67
◆68
◆69
◆70
71
72
73
74
75
76
77
Melmed S, Casanueva F, Cavagnini F et al. Consensus statement: medical management of acromegaly. Clin Endocrinol 2005; 153:737–40. Bates AS, Van’t Hoff W, Jones JM and Clayton RN An audit of outcome of treatment in acromegaly. QJM 1993; 86:293–300. Kreutzer J, Vance ML, Lopes MBS and Laws ER. Surgical management of GH-secreting pituitary adenomas: an outcome study using modern remission criteria. J Clin Endocrinol Metab 2001; 86:4072–7. Wass JAH, Laws ER, Randall RV and Sheline GE. The treatment of acromegaly. Clin Endocrinol Metab 1986; 15:683–707. Wass JAH, Plowman PN, Jones AE and Besser G.M The treatment of acromegaly by external pituitary irradiation and drugs. In Growth hormone, growth factors and acromegaly. New York: Raven Press, 1987; 199–206. Barkan AL, Halasz I, Dornfeld KJ et al. Pituitary irradiation is ineffective in normalizing plasma insulin-like growth factor I in patients with acromegaly. J Clin Endocrinol Metab 1997; 82:3187–91. Lamberts SWJ, Uitterlinder P, Verschoor L et al. Long-term treatment of acromegaly with the somatostatin analogue, SMS 201–995. N Engl J Med 1985; 313:1576–80. Lancranjan I, Atkinson AB, and the sandostatin LAR group. Results of a European multicenter study with sandostatin LAR in acromegalic patients. Pituitary 1999; 1:105–14. Bevan JS. Clinical review: The antitumoral effects of somatostatin analog therapy in acromegaly. J Clin Endocrinol Metab 2005; 90:1856–63. Melmed S, Sternberg R, Cook D et al. A critical analysis of pituitary tumor shrinkage during primary medical therapy in acromegaly. J Clin Endocrinol Metab 2005; 90:4405–10. Kopchick JJ, Parkinson C, Stevens EC et al. Growth hormone receptor antagonists: Discovery, development, and use in patients with acromegaly. Endocr Rev 2002; 23:623–46. Rabasseda X and Leeson P. Pegvisomant: Treatment of acromegaly, growth hormone receptor antagonist. Drug Future 1999; 24:24–9. van der Lely AJ, Hutson RK, Trainer PJ et al. (a) Long-term treatment of acromegaly with pegvisomant, a growth hormone receptor antagonist. Lancet 2001; 24:1754–9. Trainer PJ, Drake WM, Katznelson L et al. Treatment of acromegaly with the growth hormone-receptor antagonist pegvisomant. N Engl J Med 2000; 342:1171–7. Drake WM, Rowles SV, Roberts ME et al. Insulin sensitivity and glucose tolerance improve in patients with acromegaly converted from depot octreotide to pegvisomant. Eur J Endocrinol 2003; 149:521–7. Vance ML and Laws ER. Role of medical therapy in the management of acromegaly. Neurosurgery 2005; 56:877–85. Paisley AN, Trainer PJ and Drake WM. The place of pegvisomant in the acromegaly treatment algorithm. Growth Horm & IGF Res 2004; 14:S101–6. Feenstra J, de Herde WW, ten Have SMTH et al. Combined therapy with somatostatin analogues and weekly
462 Endocrine cancer
78
◆79
80
81
82 83
84
85
●86
87
88
89 90
91
pegvisomant in active acromegaly. Lancet 2005; 365:1644–6. van der Lely AJ, Muller A, Janssen JA et al. Control of tumor size and disease activity during cotreatment with octreotide and the growth hormone receptor antagonist pegvisomant in an acromegalic patient. J Clin Endocrinol Metab 2001; 86:478–81. Besser G.M, Burman P and Daly AF. Predictors of rates of treatment-resistant tumor growth in acromegaly. Eur J Endocrinol 2005; 153:187–93. Strasburger CJ. Biochemical assessment and long-term monitoring in patients with acromegaly: statement from a joint consensus conference of the Growth Hormone Research Society and the Pituitary Society. J Clin Endocrinol Metab 2004; 89:3099–102. Invitti C, Giraldi FP, de Martin M and Cavagnini F. Diagnosis and management of patients with Cushing’s syndrome: results of an Italian multicentre study. Study Group of the Italian Society of Endocrinology on the Pathophysiology of the Hypothalamic-Pituitary-Adrenal Axis. J Clin Endocrinol Metab 1999; 84:440–8. Kaye TB and Crapo L. The Cushing syndrome: an update on diagnostic tests. Ann Intern Med 1990; 112:434–44. Escourolle H, Abecassis JP, Bertagna X et al. Comparison of computerized tomography and magnetic resonance imaging for the examination of the pituitary gland in patients with Cushing’s disease. Clin Endocrinol 1993; 39:307–13. Kaltsas GA, Giannulis MG, Newell-Price J et al. A critical analysis of the value of simultaneous inferior petrosal sinus sampling in Cushing’s disease and the occult ectopic adrenocorticotropin syndrome. J Clin Endocrinol Metab 1999; 84:487–92. de Herder WW, Uitterlinden P, Pieterman H et al. Pituitary tumour localization in patients with Cushing’s disease by magnetic resonance imaging. Is there a place for petrosal sinus sampling? Clin Endocrinol 1994; 40:87–92. Oldfield EH, Doppmann JL, Nieman LK et al. Petrosal sinus sampling with and without corticotropin- releasing hormone for the differential diagnosis of Cushing’s syndrome. N Engl J Med 1991; 325:897–905. Colao A, Faggiano A, Pivonell, R et al. Inferior petrosal sinus sampling in the differential diagnosis of Cushing’s syndrome: results of an Italian multicentre study. Eur J Endocrinol 2001; 144:499–507. Invitti C, Giraldi FP and Cavagnini F. Inferior petrosal sinus sampling in patients with Cushing’s syndrome and contradictory responses to dynamic testing. Clin Endocrinol 1999; 51:255–7. Fahlbusch R, Buchfelder M and Muller OA. Transsphenoidal surgery for Cushing’s disease. J R Soc Med 1986; 79:262–9. Swearingen B, Biller BMK, Barker FG et al. Long-term mortality after transsphenoidal surgery for Cushing’s disease. Ann Intern Med 1999; 130:821–4. Estrada J, Boronat M, Mielgo M et al. The long-term outcome of pituitary irradiation after unsuccessful transsphenoidal surgery for Cushing’s disease. N Engl J Med 1997; 336:172–7.
92 Howlett TA, Plowman PN, Was, JA et al. Metyrapone and pituitary irradiation as primary treatment for Coushing’s disease. Presented at 7th International Congress of Endocrinology, Quebec, 1989; July 1–7, Abstract 892. 93 Jennings AS, Liddle GW and Orth DN. Results of treating childhood Cushing’s disease with pituitary irradiation. N Engl J Med 1977; 297:957–62. 94 Storr HL, Plowman PN, Carroll PV et al. Clinical and endocrine responses to pituitary radiotherapy in pediatric Cushing’s disease: an effective second-line treatment. J Clin Endocrinol Metab 2003; 88:34–7. ◆95 Laws ER, Sheehan JP, Sheehan JM et al. Stereotactic radiosurgery for pituitary adenomas: a review of the literature. J Neuro-Oncol 2004; 69:257–72. 96 Drake WM, Perr, LA, Hinds CJ et al. Emergency and prolonged use of intravenous etomidate to control hypercortisolemia in a patient with Cushing’s syndrome and peritonitis. J Clin Endocrinol Metab 1998; 83: 3542–4. 97 Greening JE, Brain CE, Perry LA et al. Efficient short-term control of hypercortisolaemia by low-dose etomidate in severe paediatric Cushing’s disease. Horm Res 2005; 64:140–3. 98 Jenkins PJ, Trainer PJ, Plowman PN et al. The long-term outcome after adrenalectomy and prophylactic radiotherapy in Cushing’s syndrome. J Clin Endocrinol Metab 1995; 80:165–71. 99 Woo SY, Isidori AM, Wat WZ et al. Clinical and biochemical characteristics of adrenocorticotropin-secreting macroadenomas. J Clin Endocrinol Metab 2005; 90:4963–9. 100 Scheithaue BW, Kovacs KT, Laws Jr ER et al. Pathology of invasive pituitary tumors with special reference to functional classification. J Neurosurg 1996; 65: 733–44. 101 Pernicone PJ, Scheithauer BW, Sebo TJ et al. Pituitary carcinoma: a clinicopathologic study of 15 cases. Cancer 1997; 79:804–12. ◆102 Kaltsas GA and Grossman AB. Pituitary carcinoma: a review. Pituitary 1998; 1:69–81. ◆103 Kaltsas GA, Nomikos P, Kontogeorgos G et al. Clinical review: Diagnosis and Management of pituitary carcinomas. J Clin Endocrinol Metab 2005; 90:3089–99. ◆104 Ragel BT and Couldwell WT. Pituitary carcinomas: a review of the literature. Neurosurg. Focus 2004; 16:E7. 105 Ironside JW. Best practice no 172: pituitary gland pathology. J Clin Pathol 2003; 56:561–8. ●106 Doniach I. Pituitary carcinoma. Clin Endocrinol 1992; 37:194–5. 107 Beauchesne P, Trouillas J, Barral F and Brunon J Gonadotropic pituitary carcinoma: case report. Neurosurgery 1995; 37:810–5. 108 Gsponer J, de Tribolet N, Déruaz JP et al. Diagnosis, treatment, and outcome of pituitary tumours and other abnormal intrasellar masses. Medicine (Baltimore) 1999; 78:236–69.
References 463
109 Kaltsas GA, Mukherjee JJ, Plowman PN et al. The role of cytotoxic chemotherapy in the management of aggressive and malignant pituitary tumors. J Clin Endocrinol Metab 1998; 83:4233–8. ●110 Asa SL and Ezzat S. The pathogenesis of pituitary tumours Nat Rev Cancer 2002; 2:836–49. 111 Farrel WE, Talbot JA, Bicknell EJ et al. Genomic sequence analysis of a key residue (Arg183) in human Gαq in invasive nonfunctional pituitary adenomas. Clin Endocrinol 1997; 47:241–44. 112 RN and Farrel WE. Clonality of pituitary tumours: more complicated than initially envisaged? Brain Pathol 2001; 11:313–27. 113 Bates AS, Farrel WE, Bicknell EJ et al. Allelic deletion in pituitary adenomas reflects aggressive biological activity and has potential value as a prognostic marker. J Clin Endocrinol Metab 1997; 82:818–24. 114 Thapar K, Kovacs K, Scheithauer BW et al. Proliferative activity and invasiveness among pituitary adenomas and carcinomas: An analysis using the MIB-1 antibody. Neurosurgery 1996; 38:99–107. 115 Thapar K, Scheithauer BW, Kovacs K et al. p53 expression in pituitary adenomas and carcinomas: Correlation with invasiveness and tumor growth functions. Neurosurgery 1996; 38:765–71. 116 Thapar K, Yamada Y, Scheithauer BW et al. Assessment of mitotic activity in pituitary adenomas and carcinomas Endocr. Pathol 1996; 7:215–21. ■117 Lloyd RV, Kovacs K, Young Jr WF et al. Pituitary tumors. In: DeLellis R, Lloyd RV, Heitz PV, Eng C, eds. Introduction. WHO classification of tumors of the endocrine organs: pathology and genetics of endocrine organs. Lyon: IARC Press; 2004: 10–13. ◆118 Landman RE, Horwith M, Peterson RE et al. Long-term survival with ACTH-secreting carcinoma of the pituitary: a case report and review of the literature. J Clin Endocrinol Metab 2002; 87:3084–89. 119 Mixson AJ, Friedman TC, Katz DA et al. Thyrotropinsecreting pituitary carcinoma. J Clin Endocrinol Metab 1993; 76:529–533. 120 Greenman Y, Woolf P, Coniglio J et al. Remission of acromegaly caused by pituitary carcinoma after surgical excision of growth hormone-secreting metastasis detected by 111-indium pentetreotide scan. J Clin Endocrinol Metab 1996; 81:1628–33. 121 Dayan C, Guilding T, Hearing S et al. Biochemical cure of recurrent acromegaly by resection of cervical spinal canal metastases. Clin Endocrinol 1996; 44:597–602. 122 Kaltsas GA, Mukherjee JJ, and Grossman AB. The value of radiolabeled MIBG and octreotide in the diagnosis and management of neuroendocrine tumours. Ann Oncol 2001; 12:S47–S50. 123 Komori T, Martin WH, Graber DL and Delbeke D. Serendipitous detection of Cushing’s disease by FDG positron emission tomography and a review of the literature. Clin Nucl Med 2002; 27:176–8.
124 Eriksson B, Bergstrom M, Sundin A et al. The role of PET in localization of neuroendocrine and adrenocortical tumors. Ann NY Acad Sci 2002; 970:159–69. 125 Ahmed M, Kanaan I, Alarifi A et al. ACTH-producing pituitary cancer: experience at the King Faisal Specialist Hospital, Research Centre. Pituitary 2000; 3:105–12. 126 Walker JD, Grossman A, Anderson JV et al. Malignant prolactinoma with extracranial metastases: a report of three cases. Clin Endocrinol 1993; 38:411–19. 127 Wilson DF. Pituitary carcinoma occurring as middle ear tumor. Otolaryngol Head Neck Surg 1982; 90:665– 6. 128 Cartwright DM, Miller TR and Nasr AJ. Fine-needle aspiration biopsy of pituitary carcinoma with cervical lymph node metastases: a report of two cases and review of the literature. Diagn Cytopathol 1994; 11:68–73. 129 Garrao AF, Sobrino LG, Pedro O et al. ACTH-producing carcinoma of the pituitary with haematogenic metastases. Eur J Endcorinol 1997; 137:176–80. 130 Martin NA, Hale M, and Wilson CB. Cerebellar metastasis from a prolactinoma during treatment with bromocriptine. J Neurosurg 1981; 55:615–19. 131 Swords F.M, Allan CA, Plowman PN et al. Stereotactic radiosurgery. XVI. A treatment for previously irradiated pituitary adenomas. J Clin Endocrinol Metab 2003; 88:5334–40. 132 Winkelmann J, Pagotto U, Theodoropoulou M, et al. Retention of dopamine 2 receptor mRNA and absence of the protein in craniospinal and extracranial metastasis of a malignant prolactinoma: a case report. Eur J Endocrinol 2002; 146:81–8. 133 Landolt AM, Haller D, Lomax N et al. Stereotactic radiosurgery for recurrent surgically treated acromegaly: a comparison with fractionated radiotherapy. J Neurosurg 1998; 88:1002–8. 134 Cuneo RC, Salomon F, MeGauley GA and Sonksen PH. The growth hormone deficiency syndrome in adults. Clin Endocrinol 1992; 37:387–97. 135 Rosen T and Bengtsson BA. Premature mortality due to cardiovascular disease in hypopituitarism. Lancet 1990; 336:285–8. 136 Drake WM, Coyte D, Camacho-Hubner C et al. Optimising growth hormone replacement therapy by dose titration in hypopituitary adults. J Clin Endocrinol Metab 1998; 83:3913–19. 137 Grimelius L and Johansson H. Pathology of parathyroid tumors. Semin Surg Oncol 1997; 13:142–54. ◆138 Shane E. Clinical review 122: parathyroid carcinoma. J Clin Endocrinol Metab 2001; 86:485–93. 139 Marx SJ. Hyperparathyroidism and hypoparathyroid disorders. N Engl J Med 2000; 343:1863–75. 140 Marx SJ, Simonds WF, Agarwal SK et al. Hyperparathyroidism in hereditary syndromes: special expressions and special managements. J Bone Miner Res 2002; 17 (Suppl. 2):N37–N43. ●141 Carpten JD, Robbins CM, Villablanca A et al. HRPT2, encoding parafibromin, is mutated in hyperparathyroidismjaw tumor syndrome. Nat Genet 2002; 32:676–80.
464 Endocrine cancer
142 Szabo J, Health B, Hill VM et al. Hereditary hyperparathyroidism-jaw tumor syndrome: the endocrine tumor gene HRPT2 maps to chromosome 1q21–q31. Am J Hum Genet 1995; 56:944–50. 143 Arnold A, Kim HG, Ga, RD et al. Molecular cloning and chromosomal mapping of DNA rearranged with the parathyroid hormone gene in a parathyroid adenoma. J Clin Invest 1989; 83:2034–40. 144 Rosenberg CL, Kim HG, Show, TB et al. Rearrangement and overexpression of D11S287E, a candidate oncogene on chromosome 11q13 in benign parathyroid tumors. Oncogene 1991; 6:449–53. 145 Motokura T, Bloom T, Kim HG et al. A novel cyclin encoded by a bc11-linked candidate oncogene. Nature 1991; 350:512–5. 146 Hemmer S, Wasenius VM, Haglund C et al. Deletion of 11q23 and cyclin D1 overexpression are frequent aberrations in parathyroid adenomas. Am J Path 2001; 158:1355–62. 147 Hsi ED, Zukerberg LR, Yang WI et al. Cyclin D1/PRAD1 expression in parathyroid adenomas: an immunohistochemical study. J Clin Endocrinol Metab 1996; 81:1736–9. 148 Tominaga Y, Tsuzuki T, Uchinda K et al. Expression of PRAD1/cyclin D1, retinoblastoma gene products, and Ki67 in parathyroid hyperplasia caused by chronic renal failure versus primary adenoma. Kidney Int 1999; 55:1375–83. 149 Vasef MA, Byrnes RK, Sturm M et al. Expression of cyclin D1 in parathyroid carcinomas, adenomas, and hyperplasia: a paraffin immunohistochemical study. Mod Pathol 1999; 12:412–6. 150 Chandrasekharappa SC, Guru SC, Manickam P et al. Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science 1997; 276:404–7. ■151 European consortium on MEN1. Identification of the multiple endocrine neoplasia type 1 (MEN1) gene. Hum Mol Genet 1997; 6:1177–83. 152 Carling T, Correa P, Hessman O et al. Parathyroid MEN1 gene mutations in relation to clinical characteristics of nonfamilial primary hyperparathyroidism. J Clin Endocrinol Metab 1998; 83:2960–3. 153 Farnebo F, Teh BT, Hytola S et al. Alterations of the MEN1 gene in sporadic parathyroid tumors. J Clin Endocrinol Metab 1998; 83:2627–30. 154 Tahara H, Smith AP, Gas RD et al. Genomic localization of novel candidate tumor suppressor gene loci in human parathyroid adenomas. Cancer Res 1996; 56: 599–605. 155 Palanisamy N, Imanishi Y, Rao PH et al. Novel chromosomal abnormalities identified by comparative genomic hybridization in parathyroid adenomas. J Clin Endocrinol Metab 1998; 83:1766–70. 156 Farnebo F, Kytola S, Teh BT et al. Alternative genetic pathways in parathyroid tumorigenesis. J Clin Endocrinol Metab 1999; 84:3775–80. 157 Yano S, Sugimoto T, Tsukamoto T et al. Decrease in vitamin D receptor and calcium-sensing receptor in highly
158
159
160
161
162
163
164
◆165
166
167
168
169
170 171
172
173
proliferative parathyroid adenomas. Eur J Endocrinol 2003; 148:403–11. Shattuck TM, Valimaki S, Obara T et al. Somatic and germline mutations of the HRPT2 gene in sporadic parathyroid carcinoma. N Engl J Med 2003; 349:1722–9. Howell VM, Haven CJ, Kahnoski K et al. HRPT2 mutations are associated with malignancy in sporadic parathyroid tumours. J Med Genet 2003; 40:657–63. Woodard GE, Lin L, Zhang JH et al. Parafibromin, product of the hyperparathyroidism-jaw tumor syndrome gene HRPT2, regulates cyclin D1/PRAD1 expression. Oncogene 2005; 24:1272–6. Teh BT, Farnebo F, Kristofferson U et al. Autosomal dominant primary hyperparathyroidism and jaw tumor syndrome associated with renal hamartomas and cystic kidney disease: linkage to 1q21–q32 and loss of the wild type allele in renal hamartomas. J Clin Endocrinol Metab 1996; 83:4204–11. Simonds WF, James-Newton LA, Agarwal S et al. Familial isolated hyperparathyroidism (FIHP). Clinical and genetic characteristics of 36 kindreds. Medicine 2002; 81:1–26. Cetani F, Pardi F, Borsari S et al. Genetic analysis of the HRPT2 gene in primary hyperparathyroidism: germline and somatic mutations in familial and sporadic parathyroid tumors. J Clin Endocrinol Metab 2004; 89:5583–91. Krebs LJ, Shattuck TM and Arnold A. HRPT2 mutational analysis of typical sporadic parathyroid adenomas. J Clin Endocrinol Metab 2005; 90:5015–7. Rubin MR and Silverberg SJ. HRPT2 in parathyroid cancer: a piece of the puzzle. J Clin Endocrinol Metab 2005; 90:5505–7. Silverberg SJ, Shane E, Jacobs TP et al. A 10-year prospective study of hyperparathyroidism with or without parathyroid surgery. N Engl J Med 1999; 341:1249–55. Johnston LB, Carroll MC, Britton KE et al. The accuracy of parathyroid gland localisation in primary hyperparathyroidism using sestamibi radionuclide imaging. J Clin Endocrinol Metab 1996; 81:346–52. Civelek AC, Ozalp E, Donovan P et al. Prospective evaluation of delayed technetium-99m sestamibi SPECT scintigraphy for preoperative localization of primary hyperparathyroidism. Surgery 2002; 131:149–57. Doppman JL. Preoperative localization of parathyroid tissue in primary hyperparathyroidism. In Bilezikian JP, Marcus R, Levine MA. (eds) The Parathyroids, 2nd Edition, San Diego, Academic Press, 2001; 475–86. Schantz A and Castleman B. Parathyroid carcinoma: a study of 70 cases. Cancer 1973; 31:600–5. Bukowski RM, Sheclef L, Cunningham J et al. Successful combination chemotherapy for metastatic parathyroid carcinoma. Arch Intern Med 1984; 144:399–400. Calandra DB, Chejfec G, Foy BK et al. Parathyroid carcinoma: biochemical and pathologic response to DTIC. Surgery 1984; 96:1132–7. Quayle FJ and Moley JF. Medullary thyroid carcinoma: including MEN 2A and MEN 2B syndromes. J Surg Oncol 2005; 89:122–9.
References 465
174 Skinner MA, DeBenedetti MK, Moley JF et al. Medullary thyroid carcinoma in children with multiple endocrine neoplasia types 2A and 2B. J Pediatr Surg 1996; 31:177–81. 175 Saaman NA, Draznin MB, Halpin RE et al. Multiple endocrine syndrome type IIb in early childhood. Cancer 1991; 68:1832–4. 176 Modigliani E, Cohen R, Campos J.M et al. Prognostic factors for survival and for biochemical cure in medullary thyroid carcinoma: Results in 899 patients. The GETC Study Group. Groupe d’etude des tumeurs a calcitonine. Clin Endocrinol 1998; 48:265–73. 177 Gill JR, Reyes-Mugica M, Iyengar S et al. Early presentation of metastatic medullary carcinoma in multiple endocrine neoplasia, type IIA: Implications for therapy. J Pediatr 1996; 129:459–64. 178 Niccoli-Sire P, Murat A, Baudin E et al. Early or prophylactic thyroidectomy in MEN 2/FMTC gene carriers: Results in 71 thyroidectomized patients. The French Calcitonin Tumours Study Group (GETC). Eur J Endocrinol 1999; 141:468–74. 179 Skinner MA, Moley JF, Dilley WG et al. Prophylactic thyroidectomy in multiple endocrine neoplasia type 2A. N Engl J Med 2005; 353:1105–13. ■180 Brandi ML, Gagel RF, Angeli A et al. Guidelines for diagnosis and therapy of MEN type 1 and type 2. J Clin Endocrinol Metab 2001; 86:5658–71. 181 Machens A, Ukkat J, Brauckhoff M et al. Advances in the management of hereditary medullary thyroid cancer. J Intern Med 2005; 257:50–9. ■182 Machens A, Niccoli-Sire P, Hoegel J et al. Early malignant progression of hereditary medullary thyroid cancer. N Engl J Med 2003; 349:1517–25. 183 Wiseman GA and Kvols LK. Therapy of neuroendocrine tumours with radiolabelled MIBG and somatostatin analogues. Semin Nucl Med 1995; 25:272–8. 184 Hoefnagel CA. Metaiodobenzylguanidine and somatostatin in oncology: role in the management of neural crest tumours. Eur J Nucl Med 1994; 21:561–81. 185 Kaltsas G, Korbonits M, Heintz E et al. Comparison of somatostatin analog and metaiodobenzylguanidine (MIBG) radionuclides in the diagnosis and localisation of advanced neuroendocrine tumors. J Clin Endocrinol Metab 2001; 86:895–902. 186 Dorr U, Frank-Raue K, Raue F et al. The potential value of somatostatin receptor scintigraphy in medullary thyroid carcinoma. Nucl Med Commun 1993; 14:439–45. 187 Schlumberger M, Abdoulmoumene N, Delisle MJ et al. Treatment of advanced medullary thyroid cancer with an alternating combination of 5-FU-streptozotocin and 5-FUdacarbazine. Br J Cancer 1995; 71:363–5. 188 Vitale G, Tagliaferri P, Caraglia M et al. Slow release lanreotide in combination with interferon-alpha 2b in the treatment of symptomatic advanced medullary thyroid carcinoma. J Clin Endocrinol Metab 2000; 85:983–8. 189 Clarke SEM. 131I-metaiodobenzylguanidine therapy in medullary thyroid cancer. Guy’s Hospital experience. J Nucl Biol Med 1991; 35:323–6.
190 Hoefnagel CA, Delprat CC and Valdes Olmos RA. Role of 131 I-metaiodobenzylguanidine therapy in medullary thyroid carcinoma. J Nucl Biol Med 1991; 35:334–6. 191 Troncone L and Rufini V. 131I-MIBG therapy of neural crest tumours. Anticancer Res 1997; 17:1823–32. 192 Mukherjee JJ, Kaltsas GA, Islam N et al. Treatment of metastatic carcinoids, pheochromocytomas, paragangliomas and medullary carcinoma of the thyroid with 131I-metaiodobenzylguanidine. Clin Endocrinol 2001; 55:47–60. ■193 Plöckinger U, Rindi G, Arnold R et al. Guidelines for the diagnosis and treatment of neuroendocrine gastrointestinal tumours. Neuroendocrinology 2004; 80:394–424. 194 Rindi G, Bordi C, Rappel S et al. Gastric carcinoids and neuroendocrine carcinomas: Pathogenesis, pathology and behavior. World J Surg 1996; 20:173–82. 195 Doppman JL, Chang R, Fraker DL et al. Localization of insulinomas to regions of the pancreas by intra-arterial stimulation with calcium. Ann Intern Med 1995; 123:269–73. 196 Krausz Y, Keidar Z, Kogan I et al. SPECT/CT hybrid imaging with 111In-pentatreotide in assessment of neuroendocrine tumours. Clin Endocrinol 2003; 59:565–73. 197 Krenning EP, Kwekkeboom DJ, Bakker WH et al. Somatostatin receptor scintigraphy with 111In-DTPA-DPhe1]- and [123I-Tyr3]-octreotide: the Rotterdam experience with more than 1000 patients. Eur J Nucl Med 1993; 20:716–31. 198 Alexander HR, Fraker DL, Norton JA et al. Prospective study of somatostatin receptor scintigraphy and its effect on operative outcome in patients with Zollinger-Ellison syndrome. Ann Surg 1998; 228:228–38. 199 Mignon M. Natural history of neuroendocrine enteropancreatic tumors. Digestion 2000; 62 (Suppl. 1):51–58. 200 Kwekkeboom DJ, Bakker WH, Kam BL et al. Treatment of patients with gastro-entero-pancreatic (GEP) tumours with the novel radiolabelled somatostatin analogue [177Lu DOTA(0),Tyr3]octreotate. Eur J Nucl Med Molec Imaging 2003; 30:417–22. 201 Kwekkeboom DJ, Mueller-Brand J, Paganelli G et al. Overview of results of peptide receptor radionuclide therapy with 3 radiolabeled somatostatin analogs. J Nucl Med 2005; 46:62S–66S. 202 Kaltsas GA, Mukherjee JJ, Isidori A et al. Treatment of advanced neuroendocrine tumors using combination chemotherapy with lomustine and 5-fluorouracil. Clin Endocrinol 2002; 57:169–83. 203 Moertel CG, Kvols LK, O’Connell MJ et al. Treatment of neuroendocrine carcinomas with combined etoposide and cisplatin. Cancer 1991; 68:227–32. 204 Mitry E, Baudin E, Ducreux M et al. Treatment of poorly differentiated neuroendocrine tumors with etoposide and cisplatin. Br J Cancer 1999; 81:1351–55. 205 Fjällskog M-LH, Granberg DPK, Welin SLV et al. Treatment with cisplatin and etoposide in patients with neuroendocrine tumors. Cancer 2001; 92:1101–07.
466 Endocrine cancer
206 Oberg K, Norheim I and Alm G. Treatment of malignant carcinoid tumors: a randomized controlled study of streptozotocin plus 5-FU and human leucocyte interferon. Eur J Cancer Clin Oncol 1989; 25:1475–9. 207 Eriksson B, Oberg K. An update of the medical treatment of malignant endocrine pancreatic tumours. Acta Oncol 1993; 32:203–8. 208 Oberg K, Eriksson B and Janson ET. The clinical use of interferons in the management of neuroendocrine gastroenteropancreatic tumors. Ann N Y Acad Sci 1994; 733:471–8. 209 Kaltsas GA, Besser GM and Grossman AB The diagnosis and medical management of advanced neuroendocrine tumors. Endocrine Rev 2004; 25:458–511. 210 Oberg K, Funa K and Alm G. Effects of leucocyte interferon on clinical symptoms and hormone levels in patients with mid-gut carcinoid tumours and carcinoid syndrome. N Engl J Med 1983; 309:129–33. 211 Eriksson B, Oberg K, Alm G et al. Treatment of malignant endocrine pancreatic tumours with human leucocyte interferon. Lancet 1986; ii:1307–9. 212 Frank M, Klose KJ, Wied M et al. Combination therapy with octreotide and α-interferon: effect on tumor growth in metastatic endocrine gastroenteropancreatic tumors. Am J Gastroenterol 1999; 94:1381–87. 213 Faiss S, Pape UF, Böhmig M et al. Prospective, randomized, multicenter trial on the antiproliferative effect of lanreotide, interferon alfa, and their combination for therapy of metastatic neuroendocrine gastroenteropancreatic tumors The International Lanreotide and Interferon Alfa Study Group. J Clin Oncol 2003; 21:2689–96. 214 Kulke M, Kim H, Stuart K et al. A Phase II study of docetaxel in patients with metastatic carcinoid tumors. Cancer Invest 2004; 22:353–9. 215 Kulke M, Lenz H, Meropol N et al. Results of a phase II study with sunitinib malate (SU11248) in patients (pts) with advanced neuroendocrine tumours (NETs). Eur J Cancer Supplements 2005 ; 3:204 (Abstract 718). 216 Hobday TJ, Mahoney M, Erlichman R et al. Preliminary results of a phase II trial of gefitinib in progressive metastatic neuroendocrine tumours (NET): A phase II Consortium (P2C) Study. J Clin Oncol 2005; 33:16S, 328s (Abstract 4083). 217 Duran I, Le L, Saltman D et al. A phase II trial of Temsirolimus in metastatic neuroendocrine carcinomas. J Clin Oncol 2005; 33:16S, 215s (Abstract 3096). 218 Kulke MH. Neuroendocrine tumours: clinical presentation and management of localized disease. Cancer Treat Rev 2003; 29:363–70. 219 Granberg D, Eriksson B, Wilander E et al. Experience in treatment of metastatic pulmonary carcinoid tumors. Ann Oncol 2001; 12:1383–91. 220 Moore AM, Estes D, Govindan R et al. A phase II trial of gefitinib (Iressa) in patients with chemosensitive and chemorefractory relapsed neuroendocrine cancers. A Hoosier Oncology Group Trial J Clin Oncol 2005; 33:16S, 660s (Abstract 7160).
221 Pandya KJ, Levy DE, Hidalgo M et al. A randomized, phase II ECOG trial of two dose levels of temsirolimus (CCI-779) in patients with extensive stage small cell lung cancer in remission after induction chemotherapy. A preliminary report. J Clin Oncol 2005; 33:16S, 622s (Abstract 7005). 222 Bryant J, Farmer J, Kessler LJ et al. Pheochromocytoma: The expanding genetic differential diagnosis. J Natl Cancer Inst 2003; 95:1196–204. 223 Dahia PLM, Hao K, Rogus J et al. Novel pheochromocytoma susceptibility loci identified by integrative genomics. Cancer Res 2005; 65:9651–8. 224 Astiti D, Hart-Holden N, Latif F et al. Genetic analysis of mitochondrial complex II subunits SDHD, SDHB and SDHC in paraganglioma and phaeochromocytoma susceptibility. Clin Endocrinol 2003; 59:728–33. 225 Neumann HP, Bausch B, McWhinney SR et al. Germ-line mutations in nonsyndromic pheochromocytoma. N Engl J Med 2002; 346:1459–66. 226 Eisenhofer G, Pacak K, Goldstein DS et al. 123I-MIBG scintigraphy of catecholamine systems: impediments to applications in clinical medicine. Eur J Nucl Med 2000; 27:611–2. 227 Pacak K, Lineham WM, Eisenhofer G et al. Recent advances in genetics, diagnosis, localization, and treatment of pheochromocytoma. Ann Int Med 2001; 134:315–29. 228 Averbuch SD, SteakleyCS, Young RC et al. Malignant pheochromocytoma: effective treatment with a combination of cyclophosphamide, vincristine, and dacarbazine. Ann Intern Med 1988; 109:267–73. 229 Krempf M, Lumrboso J, Mornex R et al. Use of 131I-metaiodobenzylguanidine in the treatment of malignant pheochromocytoma. J Clin Endocrinol Metab 1991; 72:455–61. 230 Loh KC, Fitzgerald PA, Matthay KK et al. The treatment of malignant pheochromocytoma with 131I-metaiodobenzylguanidine (131I-MIBG): a comprehensive review of 116 reported patients. J Endocrinol Invest 1997; 20:648–58. 231 Kopf D, Bockisch A, Steinert H et al. Octreotide scintigraphy and catecholamine response to an octreotide challenge in malignant phaeochromocytoma. Clin Endocrinol 1997; 46:39–44. 232 Sipple JH. The association of phaeochromocytoma with carcinoma of the thyroid gland. Am J Med 1961; 31:163–6. 233 Wermer P. Genetic aspects of adenomatosis and endocrine glands. Am J Med 1954; 16:363–71. 234 Ballard HS, Frame B and Hartsock RJ. Familial multiple endocrine adenoma–peptic ulcer complex. Medicine 1964; 43:481–516. 235 Bassett JH, Forbes SA, Pannett AA et al. Characterization of mutations in patients with multiple endocrine neoplasia type 1. Am J Hum Genet 1998; 62:232–44. 236 Chew SL and Eng C. Multiple endocrine neoplasia type 2 and related genetic conditions. Curr Opin Endocrinol Dis 1995; 2:121–6. 237 Marsh DJ, Mulligan LM and Eng C. RET proto-oncogene mutations in multiple endocrine neoplasia type 2 and medullary thyroid carcinoma. Horm Res 1997; 47:168–78.
References 467
238 Maxwell PH, Wiesener MS, Chang GW et al. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 1999; 399:271–5. 239 Saeger W. Tumours of the adrenal gland. Recent Results Cancer Res 1990; 118:79–96. 240 Hellwig D, Ukena D, Paulsen F et al. Meta-analysis of the efficacy of positron emission tomography with 18Ffluorodeoxyglucose in lung tumors. Basis for discussion of the German Consensus Conference on PET in Oncology 2000. Pneumologie 2001; 55:367–77. 241 Becherer A, Vierhapper H, Potzi C et al. FDG-PET in adrenocortical carcinomas. Cancer Biother Radiopharm 2001; 16:289–95. 242 Peppercorn PD, Reznek RH and Grossman AB. Incidentallydiscovered adrenal masses. Clin Endocrinol 1998; 48:379–88. 243 Lipsett MB, Hertz R and Ross GT. Clinical and pathophysiologic aspects of adrenocortical carcinoma. Am J Med 1963; 35:374–83. 244 Crucitti F, Bellantone R, Ferrante A et al. The Italian Registry for adrenal cortical carcinoma: analysis of a multiinstitutional series of 129 patients. The ACC Italian Registry Study group. Surgery 1996; 119:161–70.
245 Wajchenberg BL, Albergaria Pereira MA, Medonca BB et al. Adrenocortical carcinoma: clinical and laboratory observations. Cancer 2000; 88:711–36. 246 Didolkar MS, Bescher A, Elias EG and Moore RH. Natural history of adrenal cortical carcinoma. A clinicopathologic study of 42 patients. Cancer 1981; 47:2153–61. 247 Nader S, Hickey RC, Sellin RV and Samaan NA. Adrenal cortical carcinoma. A study of 77 cases. Cancer 1983; 52:707–11. 248 Koch CA, Pacak K and Chrousos GP. The molecular pathogenesis of hereditary and sporadic adrenocortical and adrenomedullary tumors. J Clin Endocrinol Metab 2002; 87:5367–84. 249 Stratakis CA. Genetics of adrenocortical tumors: gatekeepers, landscapers and conductors in symphony. Trends Endocrinol. Metab 2003; 14:404–10. 250 Kirschner LS. Emerging treatment strategies for adrenocortical carcinoma: a new hope. J Clin Endocrinol Metab 2005; In Press. 251 Dackiw AP, Lee JE, Gagel RF. et al. Adrenal cortical carcinoma. World J Surg 2001; 25:914–26. 252 Hutter AM and Kayhoe DE. Adrenal cortical carcinoma. Results of treatment with op-DDD in 138 patients. Am J Med 1966; 41:581–91.
19 Breast cancer NAVITA SOMAIAH AND JOHN YARNOLD
Epidemiology Identification of high-risk groups Breast cancer prevention Presentation and investigation Screening Molecular genetics Pathology and prognostic indices Surgery for early breast cancer Endocrine therapies for early breast cancer Cytotoxic therapies for early breast cancer Biological therapies in early breast cancer Bisphosphonates in early breast cancer
468 468 469 469 469 469 470 471 473 477 480 481
Radiotherapy in early breast cancer Special presentations of early breast cancer Treatment of patients with locally advanced disease Management of loco-regional recurrence Treatment of patients with metastatic disease Complications of metastatic disease Psychosocial support Hormone replacement therapy Follow-up Future developments References
481 487 489 490 490 491 492 492 493 493 494
EPIDEMIOLOGY
IDENTIFICATION OF HIGH-RISK GROUPS
Breast cancer incidence rates vary widely around the world, influenced by population demographics and lifestyle. World age-adjusted incidence rates are 20 per 100 000 in China compared with 90 per 100 000 in the Netherlands and USA.1 Epidemiological models of breast cancer focus on lifetime exposure to the mitogenic effects of oestrogen, including early age at menarche, late age at first birth, parity, late age at menopause, use of oral contraceptive, hormone replacement therapy (HRT), obesity and lack of physical activity.2 In the UK, the age-specific incidence rate rises fourfold between the ages of 35 and 70 years; the risk of developing breast cancer by the age of 85 years is 1 in 10. Approximately 40 000 UK women and a few hundred men develop breast cancer every year. Thirteen thousand of these patients die of the disease annually, representing 22 per cent of UK cancer mortality. Between 1989 and 2002, there was a 28 per cent reduction in UK breast cancer mortality, despite rising incidence. The reduction is attributed to more effective treatments and the national mammographic screening programme.
Family history identifies a minority of patients at increased lifetime risk of breast cancer due to a mix of shared environmental and genetic factors. Definitions vary, but the average lifetime risk of 8 per cent in UK is increased to 12–25 per cent if the family has just one or two first-degree relatives diagnosed with breast cancer before the age of 50 years, or two first-degree or second-degree relatives on the same side of the family with breast or ovarian cancer (these scenarios apply to 4 per cent of the female population in the UK). Women with three or more first-degree or second-degree relatives with breast or ovarian cancer on the same side of the family, a history of bilateral tumours, male breast cancer or sarcoma, or women aged more than 40 years at diagnosis have a 25–50 per cent lifetime risk of breast cancer (applies to 1 per cent of the female population). Women who inherit proven BRCA1 or BRCA2 mutations have a 60–80 per cent lifetime risk of breast cancer, plus a significant risk of ovarian cancer in the case of BRCA1.3,4 Nevertheless, this leaves 95 per cent of the female population with a lifetime risk of less than 12 per cent,
Molecular genetics 469
including women with one first-degree or second-degree relative diagnosed with breast cancer at age 50 years or older, and women with one second-degree relative diagnosed with breast cancer at any age. Women at intermediate or high risk according to the factors described above are offered referral to a specialist risk clinic, comprising a clinical geneticist, breast oncologist and nurse counsellor with close links to a dedicated molecular genetics laboratory.5 There is currently little evidence on which to base reliable guidelines. Women at moderately increased risk (15–25 per cent) are advised to examine their breasts once a month and asked to attend for annual clinical breast examination and annual mammography between the ages of 35 and 50 years. The frequency of screening is commonly reduced above this age but continues until the age of 69 years. Women considered or proven to have BRCA1 mutations are referred for annual pelvic ultrasound, in view of the elevated risk of ovarian cancer. Women at high risk (25 per cent) of breast cancer may wish to consider further measures, including bilateral prophylactic mastectomies and reconstruction (and oophorectomy in the case of BRCA1 mutation carriers). These interventions are taken only after careful consideration, including psychological counselling.6 Inclusion in ongoing chemical prevention trials may also be considered.
BREAST CANCER PREVENTION An overview of trials testing anti-oestrogens in the adjuvant and preventive settings confirms that 5 years’ tamoxifen reduces oestrogen-receptor-positive breast cancer risk by about 30 per cent.7*** Adverse effects, including hot flushes and endometrial cancer, limit the role of tamoxifen as an effective preventive strategy in high-risk populations. Current research is testing the benefits of aromatase inhibitors in post-menopausal women (IBIS II (International Breast Cancer Intervention Study)trial) and LHRH (luteinizing hormone releasing hormone) agonists in young high-risk women. Non-pharmacological approaches report positive results of randomized trials evaluating physical exercise and weight loss in the obese.8
PRESENTATION AND INVESTIGATION Typical mammographic features of screen-detected cancer include spiculate densities, clusters of fine calcifications, retraction or thickening of the overlying skin or nipple, and enlarged axillary nodes. The investigation of a screendetected impalpable abnormality requires localization for core biopsy guided by stereotactic ultrasound or radiography. If negative, this is followed by diagnostic wire-guided excision biopsy (10 g) in order to establish a histological diagnosis. If core-biopsy positive, therapeutic wire-guided excision with wider margins is undertaken if the patient is suitable for breast-preserving surgery. Either way, specimen
radiography is important to confirm that the radiological abnormality has been removed in its entirety. The management of a palpable lump is fairly straightforward, although cancer can present as nipple inversion, bloodstained nipple discharge, skin tethering, peau d’orange or axillary adenopathy as well. If the core biopsy confirms malignancy (duct carcinoma in situ [DCIS] cannot be reliably distinguished from invasive cancer on core biopsy), the extent of abnormalities on mammography will influence the type and extent of primary surgery (frozen section will be needed prior to mastectomy if only cytology is available beforehand). Simple staging investigations are adequate prior to surgery, i.e. haematology, biochemistry and chest radiograph, with further tests according to symptomatology or test results.
SCREENING Meta-analysis of nine randomised trials and four observational studies has reported a statistically significant reduction in breast cancer mortality for women aged 50–74 years (Hazard Ratio (HR) 0.74, 95% confidence interval (CI) 0.66–0.83), but not for women under 50 years (HR 0.93, 95% CI 0.76–1.13).9*** Meta-analyses based on stringent trial selection have challenged the conclusions in the older age group, claiming no significant benefit. Controversy persists, although the International Agency for Research on Cancer (IARC) concludes that mammography is effective in the 50–69-year age group but is, as yet, of unproven effectiveness under 50 years.10 Two-view (medio-lateral and cranio-caudal) mammography is currently offered every 3 years to all UK women between the ages of 50 and 70 years. During the initial prevalence screen, 10 per cent of women are recalled because of some mammographic abnormality, and after additional investigation, 1.5 per cent require diagnostic biopsy and about 0.5 per cent have a cancer. Overall, approximately 70 per cent of new cancers are expected to be screen detected in a population attending for mammography at 3-yearly intervals assuming 100 per cent compliance. Assuming 70 per cent uptake, these measures should save 1250 lives per year in the UK. The recent trial suggests that magnetic resonance imagine (MRI) is a more sensitive technique than mammography for screening young women with a strong family history of breast cancer.11
MOLECULAR GENETICS Somatic mutation theory postulates the accumulation of mutations in genes that progressively deregulate normal breast-cell behaviour. Several processes predispose to mutation, including exposure to mutagens such as ionizing radiation and mitogens such as oestrogen. Somatic mutations activate or inactivate the genes responsible for regulating a wide range of critical functions, including genomic stability, cell proliferation and cell adhesion.12,13
470 Breast cancer
Genes that are activated and over-expressed in 20–30 per cent of breast cancers include the epidermal growth factor receptors (EGFR or c-erbB1), c-erbB2 (neu or HER-2), bcl-2, cyclin D1 and c-myc. Over-expression of c-erbB2 is reported in approximately 40 per cent of patients with solid, comedotype DCIS, but virtually never in lobular carcinoma. Genes inactivated somatically in varying proportions of breast cancers include the CHK2, Rb, p53 and NM23 tumour suppressor genes.14 Inheritance of highly penetrant dominant syndromes (caused by mutations that are recessive at the cellular level, requiring inactivation of both alleles), associated with a strong family history, accounts for 5 per cent of breast cancer cases. The most important examples involve the BRCA1 and BRCA2 genes, which account for about 2 per cent and 1 per cent, respectively, of breast cancer incidence. Affected individuals are born heterozygous for the gene defect, but random loss of the normal allele in a breast duct epithelial cell leads to breast cancer in about 60 per cent of individuals, often at a young age (BRCA1 mutations also predispose to ovarian cancer and BRCA2 mutations also to male breast cancer). The molecular basis of the cellular effect in both syndromes is a defect in DNA repair by homologous recombination, rendering the genome prone to mutation, a state referred to as genomic instability.15 Genomic instability is also introduced by an inherited defect in the p53 gene, a rare cause of breast cancer predisposition (Li-Fraumeni syndrome). Ataxia telangiectasia (AT) is a rare autosomal recessive disorder, which renders homozygous individuals more vulnerable to cancer. Ataxia telangiectasia heterozygotes constitute 0.5–1 per cent of the general population and are probably breast cancer prone.16 Some low penetrance mutations and common genetic polymorphisms confer a much lower increment of cancer risk than BRCA1 and BRCA2, but may contribute a more important component of genetic predisposition in the general population as a whole.17
and myoepithelial cells, are involved in the pathogenesis of different tumour types. Recognizable precursor lesions of invasive carcinoma include DCIS.18 The intra-duct components of invasive tumours share the same grade as the invasive disease and a very similar genetic profile. Similarly, the grade of pure DCIS is usually correlated with the grade of an invasive recurrence. Local recurrences after breast conservation treatment of invasive disease are usually the same grade as the primary lesion. All these features suggest a highly stable genotype that is determined at an early stage in breast cancer development. Genotyping of precursor lesions and invasive cancers is advancing rapidly, and characteristic patterns of DNA losses and gains are associated with different histological types of disease and different grades of malignancy. In some instances, these are already characterized at the level of the gene, for example loss of the adhesion molecule E-cadherin on the long arm of chromosome 16 in lobular carcinoma.19 Future classifications of breast cancer are likely to include gene-expression profiles. For example, five subgroups identified on the basis of differing expression profiles have been reported to have different prognoses.20 The prognosis of the basal-like and HER-2-positive clusters is significantly worse than that of the predominantly oestrogen-receptor-positive (ER-positive) luminal categories. HER-2 status is currently confirmed by fluorescent in situ hybridization (counting gene copy number) and, as for ER positivity, gene expression status can increasingly be established using immunohistochemical analyses of specific proteins. Thus, the basal-like tumours can be identified by positive staining for cytokeratins 5 and 14 as well as by ER-negative, progesterone-receptor-negative (PR-negative) and HER-2-negative status. Such tumours are often deficient in BRCA1 protein, and may be more sensitive to DNA cross-linking agents such as platinum compounds. Expression and protein profiling are currently a very active research area.
Traditional histological classification PATHOLOGY AND PROGNOSTIC INDICES Pathogenesis and molecular classification The breast is a modified sweat gland with 15–20 duct systems converging on the nipple. These branching structures constitute a collecting system for the glandular elements of the breast comprising clusters of 1 mm lobules connected via intralobular ductules to a terminal duct (together, they comprise the terminal duct lobular unit). Several tens of thousand lobules in each breast are suspended in subcutaneous fat supported in a fibrous stroma and by elastic ligaments. Stereomicroscopy of whole-breast sections supports the origin of most breast carcinomas from the terminal duct lobular unit. It is not yet clear whether a single stem-cell type gives rise to all forms of breast cancer, or whether partially differentiated stem-cell progenitors, for example those that give rise to the terminal duct epithelial
The large majority of breast carcinomas are adenocarcinomas, and most are currently classified morphologically as ductal or lobular. Invasive ductal carcinoma (IDC) with no special histological features accounts for 75 per cent of invasive breast cancers. It is referred to as ductal carcinoma not otherwise specified (NOS) or of no special type (NST). Invasive lobular carcinoma (ILC) accounts for 10 per cent and medullary carcinoma for 5 per cent of invasive breast cancers.21 There are also three common special types of invasive breast cancer that have a significantly better prognosis than the others, including tubular, mucinous and papillary carcinoma, which together account for 5 per cent of invasive disease. Grading of carcinomas using the Bloom and Richardson system is based on the degree of tubule formation, nuclear pleomorphism and mitotic index, each scored on a threepoint scale. Grading criteria have been difficult to standardize
Surgery for early breast cancer 471
Table 19.1 The University of Southern California (USC)/Van Nuys Prognostic Index (VNPI) scoring system Score Size (mm) Margin width (mm) Pathologic classification Age (years)
1
2
3
15 10 Non-high grade without necrosis (nuclear grades 1 or 2) 60
16–40 1–9 Non-high grade with necrosis (nuclear grades 1 or 2) 40–60
40 1 High grade with or without necrosis (nuclear grade 3) 40
One to three points are awarded for each of four different predictors of local breast recurrence (size, margin width, pathologic classification and age). Scores for each of the predictors are totalled to yield a VNPI score ranging from a low of 4 to a high of 12. Adapted from M.J. Silverstein.22
in the past, but progress in reducing observer variation has been made in recognition of its powerful prognostic significance. The proportion of tumours in different series is variable, approximately 1:2:2 for grades 1, 2 and 3 respectively. Histological grading is also applied to ILC, although tubule formation cannot be scored.
Duct carcinoma in situ Duct carcinoma in situ comprises a spectrum of lesions characterized by proliferation of malignant cells within ducts without invasion of surrounding stroma. It presents as a mass or nipple discharge, nowadays more commonly as screen-detected microcalcification. It is characterized by cells with large pleomorphic nuclei, often with multiple mitotic figures, with or without comedo-type necrotic debris in the centre of the ducts. Calcification of necrotic debris produces characteristic microcalcifications on mammography, ranging in size from a single focus, measuring a few millimetres, to extensive involvement of a whole quadrant. Duct carcinoma in situ is nearly always unilateral, in contrast to lobular carcinoma in situ (LCIS, see below). The Van Nuys Prognostic Index classifies DCIS according to the risk of local recurrence after breast-preserving surgery has been proposed (Table 19.1).22* This index distinguishes high-nuclear and non-high-nuclear grade lesions, the latter subdivided according to the presence or absence of comedotype necrosis. It is increasingly likely that atypical ductal hyperplasia is not a distinct entity, but is more accurately classified as small (2 mm) foci of DCIS.
Lobular carcinoma in situ Lobular carcinoma in situ is not clinically palpable or detectable by mammography, but is identified incidentally in about 1 per cent of benign breast biopsies. Under the microscope, LCIS appears as a solid proliferation of small cells within breast lobules with uniform small round/oval nuclei, rarely enlarging lobules to more than two to three times their normal size. Mitoses are uncommon and necrosis is not seen. It is predominantly found in pre-menopausal women, and is typically multi-centric in the breast and often
(35 per cent) bilateral. Lobular carcinoma in situ has long been regarded as an indicator of elevated cancer risk rather than as a pre-malignant condition in its own right, but new data challenge this model.18 It is associated with an approximately 30 per cent lifetime risk of developing invasive carcinoma, usually ductal and usually (60 per cent) in the ipsilateral breast. Like DCIS, LCIS is more than a risk factor; it is clearly also a precursor to invasive disease.
Prognostic indices Conventional pathological features of invasive carcinomas correlating with prognosis include histological subtype, tumour size, grade, lympho-vascular invasion and axillarynode metastases. Pathological node status is the single most powerful prognostic indicator, node-positive patients having, on average, a 10-year survival of 40 per cent, compared with 80 per cent for node-negative women after loco-regional therapy alone. Multivariate analyses generate algorithms based on pathological tumour size, grade and node status that separate patients into prognostic groups, the best known until recently being the Nottingham Index (NI): NI 0.2 pathological tumour size (cm) node stage/level (1–3) grade (1–3).23 This has now been superseded by an online index based on the (National Cancer Institute (NCI) Surveillance, Epidemiology, and End Results (SEER) Program and estimated using patient age, tumour size, grade, oestrogen-receptor status and lymph-node status (www.adjuvantonline.com). As an indication of how fast molecular analysis is progressing, a 70-gene expression profile that appears to discriminate good-risk and poorrisk node-negative tumours better than conventional indices is due for evaluation in a prospective randomized comparison against a standard St Galen Index in the proposed MINDACT (Microarray In Node negative Disease may Avoid ChemoTherapy) trial.24
SURGERY FOR EARLY BREAST CANCER Surgery remains the single most effective curative modality for women with breast cancer. The proportion of patients
472 Breast cancer
biologically cured by surgery alone (not just personally cured in the sense of dying of other causes) cannot be directly estimated, but is likely to be at least 60 per cent overall and much higher in favourable screen-detected lesions. Randomized trials testing mammographic screening do provide direct evidence for the existence of patients cured by earlier local therapies.25*** More dramatically, the significant reduction in breast cancer mortality demonstrated by postoperative radiotherapy also provides direct evidence of a subgroup of patients with disease truly confined to the breast and lymphatic pathways.26***
be accounted for by a higher probability of macroscopic residual disease in the tumour bed, which can often be dealt with by re-excision.30* If there is no extensive intra-duct component, an isolated microscopic focus of intra-duct disease at radial resection margins in women over the age of 40 years is not an absolute indication for re-excision, provided full-dose breast radiotherapy is given. There is increasing interest in marking the cavity walls with titanium clips or gold seeds in order to help localize radiotherapy more accurately. This is likely to be increasingly important when onco-plastic procedures that mobilise breast flaps to fill the excision cavity are applied more widely.31
Primary surgery Breast reconstruction Randomized trials confirm the equivalence of breast-preserving surgery plus radiotherapy and mastectomy in terms of local control and overall survival.27 Breast-conserving surgery can, therefore, be offered as an alternative to mastectomy to the majority of women with early-stage disease. Clinical or radiological multi-centricity is an indication for mastectomy, unless discrete lesions can be independently localized and completely excised. When the expected surgical defect is large in relation to breast size, or when the nipple needs to be removed, the patient may opt for mastectomy.28 If the patient finds it difficult to lie in the radiotherapy treatment position or has active collagen vascular disease (systemic sclerosis), there may be a lower threshold for recommending mastectomy. After tumour excision, the surgical specimen must be marked with surface sutures to allow orientation and the whole specimen Indian inked before fixation. There is no consensus on the optimal margins of excision. A randomized trial in Milan tested tumorectomy ( 1 cm margins of healthy tissue at surgery) against quadrantectomy, both followed by radiotherapy, and reported superior local control with the latter at the expense of breast appearance.29** Complete microscopic clearance of invasive disease based on pathological examination of multiple paraffin sections cut from blocks prepared at the narrowest macroscopic tumour margins is achieved in the majority of cases (if not, re-excision or mastectomy is needed). The depth of excision is variable, some surgeons removing a cylinder of tissue down to the pectoral fascia. Image-guided resection may become more common in future, in which case the tumour is more likely to be in the centre of the resection specimen than is currently the case. If the resection extends to the deep fascia, microscopic evidence of disease at the deep margin is disregarded unless the fascia is breached. Microscopic evidence of intra-duct disease at the radial or superficial resection margins in women under the age of 40 years is a risk factor for local recurrence if there is extensive intra-duct carcinoma. Extensive intra-duct carcinoma is defined as DCIS occupying more than 25 per cent of the tumour cross-section at high-power microscopy plus extension of DCIS beyond the peripheral margins of the invasive component. The elevated recurrence risk seems to
Breast reconstruction can be performed as a one-stage procedure at the time of mastectomy or deferred until other treatment is complete, including radiotherapy.32* Reconstruction of a breast mound can be achieved using a fixed-volume implant or an adjustable breast expander inserted beneath the pectoralis major muscle, which is inflated with saline over few weeks prior to replacement with a silicone implant. It is now common to use a Becker implant, a device that acts both as an expander and as an implant, having separate envelopes accessed via a single subcutaneous port that is removed at a later date. The alternative is a musculocutaneous flap of skin, fat and muscle, beneath which an implant is placed. A latissimus dorsi (LD) flap from the upper back or the transverse rectus abdominis myocutaneous-free (TRAM) flap from the abdominal wall may be combined with a reduction mammoplasty of the contralateral side. Breast reconstruction is consistent with subsequent local radiotherapy, although some, but not all, retrospective studies report higher rates of late adverse effects, including fat necrosis, flap necrosis, contraction, induration and infection.33
Axillary surgery Axillary sampling traditionally involves the surgeon identifying by palpation and inspection a minimum of four lymph nodes from the axillary fat pad, a diagnostic procedure defined and tested against axillary dissection in a welldesigned randomized trial.34** Sampling of the sentinel lymph node, defined as the nearest regional lymph node to take up a radiolabelled colloid injected subdermally superficial to the primary tumour, is now being widely adopted in clinically node-negative patients as a refinement of the sampling approach.35 The evidence suggests that the technique has similar accuracy and less morbidity than more extensive surgical procedures if performed by a surgeon specifically trained in the technique. Preoperative ultrasound with cytology of abnormal nodes is increasingly used to help select the most appropriate surgical approach to the axilla. The standard management of patients with
Endocrine therapies for early breast cancer 473
less than 0.2 mm focus of malignant cells, or individual cells staining positive with immunohistochemistry, in a single sentinel node should be as for node-negative disease, although a current European Organisation for Research and Treatment of Cancer (EORTC) trial is comparing observation with axillary clearance (AMAROS). Otherwise, axillary dissection or radiotherapy is indicated in nodepositive patients. Axillary dissection removes a median of ten nodes from below the lower border of pectoralis minor (level I), five nodes from beneath this muscle (level II) and a further five nodes medial to the lateral border of pectoralis minor (level III). Axillary recurrence is uncommon (1 or 2 per cent at 10 years) in node-positive women after a full dissection by an experienced surgeon. Incomplete excision margins, rather than extra-nodal extension per se, is a risk factor for axillary recurrence. Numbness, shoulder stiffness and pain are common complications of complete axillary dissection.36 The risk of arm oedema depends on the anatomical level of dissection, a 10 per cent risk being typical after a level II dissection. Arm oedema is managed with the early introduction of a well-fitting support stocking or sleeve. Breast oedema after axillary surgery is also common. Cording is a painful restriction of shoulder mobility of unknown aetiology after axillary surgery, presenting with thickened strands in the subcutaneous tissues. The syndrome usually subsides after several months.
ENDOCRINE THERAPIES FOR EARLY BREAST CANCER The Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) meta-analysis, updated with 15 years of followup, reinforces the evidence that adjuvant endocrine therapies significantly reduce disease recurrence and mortality in women with early breast cancer (Table 19.2).26***
Adjuvant ovarian suppression The 2005 overview by the EBCTCG evaluated almost 8000 women below the age of 50 years with ER-positive or ER-unknown disease randomized into trials of ovarian ablation/suppression (approximately half had chemotherapy as well).37*** In patients less than 50 years old randomized to ovarian ablation versus no other systemic treatment for ERpositive or ER-unknown disease, suppression of ovarian function achieved a 30 per cent reduction in the annual odds of breast cancer mortality. In young women who wish to preserve fertility, the use of gonadotrophin-releasing hormone (GnRH) agonists is a safe and reversible alternative to surgery or radiotherapy ablation.
Adjuvant tamoxifen The 2005 EBCTCG systematic overview confirmed that in women with ER-positive disease, 5 years of adjuvant
tamoxifen reduces the annual breast cancer death rate by 31 per cent (standard error [SE] 3),26*** with as many deaths prevented between years 5 and 14 as during the first 5 years. The magnitude of effect appears to be independent of the use of chemotherapy and of patient age below 40, 40–59, 50–69 and more than 70 years (Fig. 19.1). The benefits, and possible adverse effects, of longer duration tamoxifen are being tested in the adjuvant Tamoxifen Treatment offer more (aTTom) and adjuvant Tamoxifen Longer Against Shorter (ATLAS) trials. In ER poor disease, trials confirm that tamoxifen has no significant impact on recurrence or breast cancer mortality.38 Although hot flushes and sweats are common and unpleasant, the most serious adverse effects are thromboembolisms and endometrial carcinoma, both consequences of the weak agonist effects of the drug.
Adjuvant aromatase inhibitors Aromatase inhibitors (AIs) block the conversion of androstenedione to oestrogen, and are ineffective before the menopause. In post-menopausal women, all of the third-generation AIs suppress circulating oestrogen levels by approximately 98 per cent.39–41 At least ten adjuvant trials with AIs including over 40 000 women are currently ongoing or have just been completed, with significant implications for the management of post-menopausal women with breast cancer. So far, AIs have been explored (i) as replacement for tamoxifen,42**,43 (ii) following 2–3 years of tamoxifen,44** and (iii) following 5 years of tamoxifen45** (Table 19.3). Updated results from the Arimidex or Tamoxifen Alone or in Combination (ATAC) trial (anastrozole versus tamoxifen versus combined, n 9366) showed that anastrazole 1 mg daily improved disease-free survival (HR 0.87, p 0.001) and time to recurrence (HR 0.79, p 0.0005) compared to tamoxifen for 5 years at a median follow-up of 68 months.42,46 The IES trial (Intergroup Exemestane Study) randomized 4742 women after 2–3 years of adjuvant tamoxifen to either tamoxifen or exemestane 25 mg daily for the remainder of the 5 years. After a median follow-up of 30.6 months, the hazard ratio for breast cancer recurrence was 0.68 ( p 0.001) in favour of exemestane.44 The MA17 trial tested letrozole 2.5 mg daily after 5 years of tamoxifen and reported a significantly superior disease-free survival in favour of letrozole at a median follow-up of 30 months (94 per cent versus 90 per cent, p 0.001).45 Based on this interim analysis, the trial was terminated by the Independent Data and Safety Monitoring Committee in order to offer women taking placebo an opportunity to take letrozole. The trial has since reported a significant overall survival benefit in the nodepositive subgroup (HR 0.61, p 0.04).45 The BIG 1-98 Femara-Tamoxifen Breast International Group (FEMTA) trial is ongoing, but data on the comparison of tamoxifen versus letrozole have recently been published in favour of letrozole (Table 19.3).47,48
Table 19.2 Estimated effects of 6 months of anthracycline-based chemotherapy, 5 years of tamoxifen, or both, on 15-year breast cancer mortality (%), in the absence of other causes of death: relevance of oestrogen-receptor (ER) status, age and underlying risk (10–15%, 25% or 50%) Proportional effect on annual breast cancer mortality rate (treatment versus control)
15-year breast cancer mortality with treatment (risk and absolute gain) versus corresponding risk without treatment (M) M ⴝ 12.5 (e.g. low risk, node negative)
M ⴝ 25 (e.g. node negative)
M ⴝ 50 (e.g. node positive)
Systemic adjuvant treatment
Ratio of
Proportional reduction
and age at diagnosis (years)
rates (R)
(%)
Risk (%)
Gain
Risk (%)
Gain
Risk (%)
Gain
Chemotherapy only in ER-poor or ER-positive disease* None (any age) Anthracycline (age 50 years) Anthracycline (age 50–69 years) Anthracycline (age 70 years)
1.0 0.62 0.80 ?
– 38 20 ?
12.5 7.9 10.1 ?
– 4.6 2.4 ?
25.0 16.3 20.6 ?
–
50.0 34.9 42.6 ?
15.1 7.4 ?
Endocrine, or chemo-endocrine, therapy in ER-positive disease* None (any age) Tamoxifen (any age) Anthracycline (age 50 years) Anthracycline (age 50–69 years) Anthracycline (age 70 years)
1.0 0.69 0.62 0.69 0.80 0.69 ? 0.69
– 31 57 45 ?
12.5 8.8 5.6 7.1 ?
– 3.7 6.9 5.4 ?
25.0 18.0 11.6 14.7 ?
–
50.0 38.0 25.7 31.8 ?
– 12.0 24.3 18.2 ?
8.7 4.4 ?
7.0 13.4 10.3 ?
Anthracycline: about 6 months of anthracycline-based adjuvant chemotherapy with regimens such as FAC or FEC FAC, 5-fluorouracil, doxorubicin, cyclophosphamide; FEC, 5-fluorouracil, epirubicin, cyclophosphamide, as in the reviewed trials. Tamoxifen: about 5 years of adjuvant tamoxifen. The 15-year survival probability with treatment is calculated as (1 M/100) to the power R. The webappendix 1 to the EBCTCG article26 gives the 15-year prognosis of untreated control patients, subdivided by ER and nodal status. *For women of given nodal status, the 5-year mortality is greater for ER-poor than for ER-positive disease, but the 15-year risks may be similar, as may the 15-year benefits of anthracycline-based chemotherapy (since the age-specific breast cancer mortality ratios for anthracycline-based versus no chemotherapy do not depend significantly on ER status). Combination of the direct and indirect randomized evidence yields breast cancer death rate ratios (treatment versus control) of 0.62 standard error (SE) 0.05 at younger than 50 years and 0.80 (SE 0.04) at age 50–69 years for allocation to anthracycline and 0.69 (SE 0.03) for allocation to tamoxifen. (Allowance for any inappropriate non-compliance with the treatment allocations in these trials would, in expectation, further reduce breast cancer mortality.) Adapted from the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG).26
Endocrine therapies for early breast cancer 475
An assessment of AIs is complex and there is no consensus view as to the optimum sequencing and duration of treatment. More data will become available from previously reported trials and from ongoing and/or unreported trials. Meanwhile, a 2005 American Society of Clinical Oncology (ASCO) technological assessment statement recommended that adjuvant hormonal therapy for postmenopausal woman with hormone-receptor-positive breast cancer: ‘…should include an AI as initial therapy or after treatment with tamoxifen. AIs are appropriate as initial treatment for women with contraindications to tamoxifen.
Category
Deaths/Women Tamoxifen deaths Allocated Allocated Logrank Variance Ratio of annual death rates Tamoxifen : Control tamoxifen control O-E of O-E
Entry age (trend χ21 0.4; 2p 0.1; NS) Age 40
74/417 (17.7%)
119/398 21.9 (29.9%)
44.0
0.61 (SE 0.12)
40–49
173/1119 (15.5%)
219/1139 24.8 (19.2%)
90.3
0.76 (SE 0.09)
50–59
330/1591 (20.7%)
394/1535 45.2 (25.7%)
161.7
0.76 (SE 0.07)
60–69
379/1822 (20.8%)
527/1789 87.3 (29.5%)
200.4
0.65 (SE 0.06)
62/266 (23.3%)
89/286 13.6 (31.1%)
29.9
0.63 (SE 0.15)
543.6
0.683 (SE 0.036) 2p < 0.00001
70 Age unknown
0/10
Total
1018/ 5225 (19.5%)
99% or
0/14 1348/ 207 5161 (26.1%)
95% confidence intervals
0
0.5
Tamoxifen better
1.0
1.5
2.0
Tamoxifen worse
Treatment effect 2p < 0.00001
Figure 19.1 Breast cancer mortality in trials of 5-year Tamoxifen versus no tamoxifen (ER). (From https://www.ctsu.ox.ac.uk/⬃ebctcg/systemic2000/aldbc.htm.)
Treatment options include 5 years of AI treatment or sequential therapy consisting of tamoxifen (for either 2 to 3 years or five years) followed by AIs for 2 to 3 years or five years’.49 This statement is controversial, as the existence and magnitude of the predicted overall survival benefit are not yet confirmed. At the time of writing, it remains common UK practice to offer tamoxifen at the time of primary treatment, unless the patient has a history of thromboembolism or some form of intolerance, most commonly night sweats. Although not yet reviewed by the National Institute for Clinical Excellence (NICE), it is also increasingly common to switch poor-prognosis ER-positive patients to several years of an AI following several years of tamoxifen. The potential value of AIs in ER-positive PR-negative and HER2-positive subgroups is discussed on page 476 (under heading Molecular Basis of Endocrine Response and Resistance).
Combined adjuvant endocrine therapies The Intergroup 101 (INT101) trial reported a survival benefit (HR 0.73, 95% CI 0.59–0.90, p 0.01) when 5 years’ adjuvant tamoxifen was added to chemotherapy plus a GnRH agonist in pre-menopausal women with node-positive ER-positive disease.50** In pre-menopausal women with ER-positive disease prescribed 5 years’ tamoxifen, there is little evidence for an added benefit of ovarian suppression, a question partly addressed by the UK Adjuvant Breast Cancer (ABC) trial and to be further tested in the BIG SOFT trial. Certainly, the 2005 EBCTCG overview raised no concerns about the efficacy of 5 years’ tamoxifen in very young women. The hazard ratios for breast cancer mortality for 5 years’ tamoxifen versus no tamoxifen were 0.61 (SE 0.12) for women below the age of 40
Table 19.3 Results of trials ATAC,42 IES,44 MA1745 and BIG 1-98 (FEMTA)48 in terms of disease-free survival and and overall survival (OS) Median follow-up (months)
ATAC 68
Tamoxifen (5 years)
IES 30.6
AIs (duration of therapy)
Anastrozole (5 years)
Tamoxifen (2–3 years)
N
3092
3094
2362
MA17 30
Tamoxifen (5 years)
BIG 1-98 51
Tamoxifen (5 years)
Tamoxifen (5 years)
Letrozole : Tamoxifen
Tamoxifen : Letrozole
: Letrozole
: placebo
(total: 5 years)
(total: 5 years)
2380
2593
2594
4003
4007
: Exemestane (2–3 years)
DFS (%)
86.9
84.5
91.5
86.8
94.4
89.8
84.0
81.4
HR (95% CI)
0.86 (0.76–0.99)
p 0.03
0.68 (0.56–0.82)
p 0.001
0.58 (0.45–0.76)
p 0.001
0.81 (0.70–0.93)
p 0.03
OS (%)
–
–
–
–
95.4
95.0
–
–
HR
0.97 (0.85–1.12)
p 0.7
0.88 (0.67–1.16)
p 0.37
0.82 (0.7–1.19)
p 0.3
0.86 (0.70–1.06)
p 0.16
AIs: aromatase inhibitors, N: number of patients in trial, DFS: disease free survival, HR: hazard ratio, OS: overall survival
476 Breast cancer
years, compared to 0.76 (SE 0.09) for women aged 40–49 and 0.76 (SE 0.07) for women aged 50–59 years.26*** Even though a proportion of patients were randomized to tamoxifen against a background of chemotherapy, it is likely that only a minority of women below the age of 40 years would have been rendered post-menopausal.51 In pre-menopausal patients with ER-positive metastatic cancer, AIs appear to be effective when combined with GnRH analogues.52 The use of AIs plus ovarian suppression with or without chemotherapy or tamoxifen in pre-menopausal women is being addressed in the Suppression of Ovarian Function Trial (SOFT), Tamoxifen and Exemestane Trial (TEXT) and Premenopausal Endocrine-Responsive Chemotherapy (PERCHE) trials. In post-menopausal women, concurrent tamoxifen and anastrozole has no advantage over tamoxifen alone, and is significantly inferior to anastrozole alone (ATAC). The explanation for the inferiority of combined treatment is that tamoxifen acts as an oestrogen agonist in the oestrogen-depleted environment induced by the AI.
Neoadjuvant endocrine therapy The validity of primary tumour response as a predictor of effect against occult metastases is not established.53 Nevertheless, this is an active field of research. For example, 337 post-menopausal women with ER-positive and/or PR-positive primary untreated breast cancer were randomized to once-daily letrozole (2.5 mg) or tamoxifen (20 mg) for 4 months prior to surgery.54** At baseline, none of the patients was considered to be a candidate for breastconserving surgery. Overall objective response rate (clinical palpation) was statistically significantly superior in the letrozole group, 55 per cent versus 36 per cent (p 0.001). Secondary endpoints of ultrasound response (35 per cent versus 25 per cent, p 0.042), mammographic response (34 per cent versus 16 per cent, p 0.001) and breast-conserving surgery (45 per cent versus 35 per cent, p 0.022) all favoured the AI. Regression analysis demonstrated that patients receiving letrozole were more than twice as likely to achieve a clinical response as patients receiving tamoxifen (odds ratio 2.23; p 0.001). There is no published evidence bearing on whether neoadjuvant endocrine therapies are superior to the same drugs given in the adjuvant setting in terms of disease-free or overall survival. The main current appeal of neoadjuvant therapy is that the primary tumour represents opportunities for identifying factors predictive of response and for investigating mechanisms of endocrine resistance.
Molecular basis of endocrine response and resistance Oestrogen regulates the transcription of multiple genes involved in cancer development and progression via interaction with the ER, a nuclear protein that functions as a transcription factor. The ER has defined domains
Androstenedione
Aromatase inhibitor
Aromatase
Testosterone
Aromatase
Aromatase inhibitor
Cestradiol
Cestrone
Tamoxifien
ER ER ↑ Proliferation EREs
ER target genes
Figure 19.2 Mechanism of action of aromatase inhibitors and tamoxifen. Oestradiol binds to the oestrogen receptor (ER), leading to dimerization, conformational change and binding to oestrogen response elements (EREs) upstream of oestrogenresponsive genes, including those responsible for proliferation. Tamoxifen competes with oestradiol for ER binding, whereas aromatase inhibitors reduce the synthesis of oestrogens from their androgenic precursors. (With permission of Macmillan Publishers Ltd, Nature Review Cancer,54 copyright 2003.)
mediating oestrogen binding, ER dimerization, binding to promoter/enhancer regions of oestrogen-responsive genes and, finally, transcriptional activation (Fig. 19.2).55,56 Like oestrogen, tamoxifen induces dimerization of ER and binding to DNA, but fails to activate transcription via the AF-2 site (responsible for oestrogen antagonist effects), although it does allow some transcription via the AF-1 site in selected tissues, including endometrium (responsible for partial agonist effects).57 Progress has been made in recent years in understanding some of the molecular mechanisms involved in denovo and acquired endocrine resistance.58,59 It is clear that the molecular biology of ER in breast cancer is complex and several aspects other than mere expression of adequate levels of ER determine initial endocrine sensitivity. Experimental models60 and clinical studies61 have shown that ER-positive tumours over-expressing EGFR or HER-2 are more resistant to tamoxifen. This receptor profile also predicted for greater sensitivity to AIs in two neoadjuvant studies in postmenopausal women.62 Loss of PR in ER-positive tumours appears to indicate activation of the EGFR/HER-2 pathway and may serve as an indirect marker for resistance to tamoxifen. In this context, retrospective subgroup analysis of the ATAC adjuvant trial reported greatest clinical benefit for anastrazole in ER-positive PR-negative tumours, although initial results from the BIG-198 and MA17 studies testing letrozole have not confirmed this.47 Laboratory and clinical data support the concept that, over time, breast cancer cells utilize alternative plasma
Cytotoxic therapies for early breast cancer 477
membrane receptor signalling pathways that enhance ER activity (cytoplasmic oestrogen/ER also activates HER-2 signalling via phosphorylation, a non-genomic effect of ER), block apoptosis and allow cells to escape initial control by tamoxifen.63,64 Strategies are emerging to block these signalling pathways by co-treatment with signal transduction inhibitors, such as trastuzumab, lepatinib and gefitinib.65 Acquired resistance to AIs reflects acquired hypersensitivity to low levels of residual oestrogen via upregulation of ER transcription. This provides the theoretical rationale for testing the anti-oestrogen fulvestrant to an AI in endocrine-responsive disease at the time of progression. Fulvestrant targets the oestrogen/ ER complex for degradation, depriving the cell of functional ER and overcoming this particular source of endocrine resistance.
Polychemotherapy (trend χ2 0.9; 2p 0.1; NS) 1
Age 40
Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) overview, 2005 The 2005 EBCTCG systematic overview confirmed that six cycles of anthracycline-containing chemotherapy, typically some form of 5-fluorouracil, doxorubicin, cyclophosphamide (FAC) or 5-fluorouracil, epirubicin, cyclophosphamide (FEC), reduces the annual odds of death from breast cancer by 38 per cent in women aged less than 50 years and by 20 per cent in women aged 50–69 years (see Table 19.2).26*** These odds ratios are 16 per cent higher than an equivalent number of cyclophosphamide, methotrexate, 5-fluorouracil (CMF) cycles in each of these age groups. It is unclear why young women have the most to gain. Part of the explanation (some argue a complete explanation in ER-positive tumours) is that
99/303 (32.7%)
10.9
34.0
0.73 (SE 0.15)
21.3
60.4
0.70 (SE 0.11)
40–49
130/586 (22.2%)
162/541 (29.9%)
370/1118 (33.1%)
430/1073 (40.1%)
37.0 135.1
0.76 (SE 0.08)
60–69
358/938 (38.2%)
397/939 (42.3%)
20.1 126.6
0.85 (SE 0.08)
70
22/64 (34.4%)
35/91 (38.5%)
3.8
0.54 (SE 0.30)
1/4
0/2
958/ 3034 (31.6%)
1123/ 2949 (38.1%)
Age unknown subtotal
99% or
Entry age
CYTOTOXIC THERAPIES FOR EARLY BREAST CANCER
77/324 (23.8%)
50–59
Adverse effects of endocrine therapy Adverse effects are related to oestrogen deprivation, with the exception that tamoxifen acts as a weak oestrogen agonist in endometrium and bone. In the National Surgical Adjuvant Breast and Bowel Project P-1 (NSABP P-1) breast cancer prevention trial, there was a four-fold increase in risk of endometrial cancer, a three-fold higher risk of pulmonary embolism, a 1.8-fold increased risk of stroke, and a 1.7-fold increased risk of deep venous thrombosis for women aged 50 years and older who received tamoxifen rather than placebo.66** Tamoxifen has been shown to prevent bone loss in post-menopausal women. The long-term safety profile of AIs is still evolving, but it is clear that they are associated with a lower incidence of hot flushes, vaginal bleeding and discharge, uterine cancer and thromboembolism compared to tamoxifen. On the other hand, they are associated with an increased incidence of muscle and joint pains and osteoporotic fractures. The tendency to bone-density loss can be monitored and managed with dietary supplements or bisphosphonates.
Deaths/Women Chemotherapy deaths Allocated Adjusted Logrank Variance Ratio of annual death rates chemotherapy control O-E of O-E Chemotherapy : Control
Entry age
6.1
93.1 362.2
95% confidence intervals
0.77 (SE 0.05) 2p 0.00001
0 0.5 1.0 1.5 2.0 Chemotherapy better Chemotherapy worse
Deaths/Women Chemotherapy deaths Allocated Adjusted Logrank Variance Ratio of annual death rates chemotherapy control O-E of O-E Chemotherapy : Control
Polychemotherapy (trend χ2 7.4; 2p 0.007) 1
Age 40
77/332 (23.2%)
86/333 (25.8%)
10.6
29.3
0.70 (SE 0.16)
40–49
185/1198 (15.4%)
282/1243 (22.7%)
44.9
89.3
0.61 (SE 0.08)
50–59
725/3004 (24.1%)
868/3219 (27.0%)
39.8 294.0
0.87 (SE 0.05)
60–69
820/3222 (25.5%)
949/3361 (28.2%)
21.9 325.7
0.93 (SE 0.05)
70
115/483 (23.8%)
156/515 (30.3%)
9.6
40.8
0.79 (SE 0.14)
0/3
2/5
1922/ 8242 (23.3%)
2343/ 8676 (27.0%)
126.7 779.1
0.85 (SE 0.03) 2p 0.00001
Age unknown subtotal
99% or
95% confidence intervals
0 0.5 1.0 1.5 2.0 Chemotherapy better Chemotherapy worse
Figure 19.3 (a) Breast cancer mortality in trials of polychemotherapy versus no chemotherapy (ER). (From https://www.ctsu.ox.ac.uk/⬃ebctcg/systemic2000/aldbc.htm.) (b) Breast cancer mortality in trials of polychemotherapy versus no chemotherapy (ER). (From https://www.ctsu.ox.ac.uk/⬃ebctcg/systemic2000/aldbc.htm.)
a significant component of chemotherapy effect is related to chemical castration. The risk of chemotherapy-induced ovarian failure is related to age and type of chemotherapy, occurring in approximately 40 per cent after four cycles of doxorubicin, cyclophosphamide (AC) and in up to 70 per cent after six cycles of classical CMF.51,67 On average, less than 50 per cent of women under the age of 40 compared to more than 50 per cent of pre-menopausal women aged 40 years or older are rendered post-menopausal by chemotherapy, and this has been related to a trend for greater benefit of chemotherapy in women with ER-positive tumours in the 40–49-year age group compared to women with ER-positive tumours aged less than 40 years. However, there is growing interest in the enhanced benefits of adjuvant chemotherapy in patients with ER-negative disease, where no age effect was reported (Fig. 19.3). On average, ER-positive tumours in young women express less ER than ER-positive tumours in old women, documentation of
478 Breast cancer
which requires a semi-quantitative score based on immunohistochemistry, e.g. the Aldred or H-score. An inverse relationship between strength of ER expression and chemoresponsiveness is postulated as a plausible mechanistic explanation for the age effects of chemotherapy, independent of the effects on ovarian function described above.68 There is epidemiological evidence based on more than 10 000 Danish breast cancer patients aged less than 50 years at diagnosis suggesting that the adverse prognosis in the less than 35-year subgroup (relative risk of death 2.1 compared to women aged 40–49) is abolished by chemotherapy, even in pathological stage I disease.69* At the other end of the age spectrum, women aged 70 years or more are very poorly represented (n 1224) in randomized trials, and there is not enough power to demonstrate a statistically significant reduction in breast cancer mortality despite several months of polychemotherapy in this age group (odds ratio 0.87; SE 0.12). It is likely that the minority of women over the age of 70 years with ER-negative tumours have the most to gain, and a randomized trial (Adjuvant Cytotoxic chemotherapy In Older Women (ACTION)) is testing the benefits of four cycles of AC chemotherapy versus no chemotherapy in women with node-positive disease.70 Apart from the strong effect of age, the relative effects of chemotherapy are largely independent of node status or of 5 years’ use of tamoxifen. The implications for treatment take into account the fact that chemotherapy prevents deaths in years 10–14 as well as in years 0–9, so that the absolute mortality reduction in years 5–15 is as great as in years 0–4. The absolute gains from chemotherapy (CMF and anthracycline schedules) in different prognostic subgroups are summarized in Table 19.2.26 Where the optimal dose of anthracyclines is concerned, the Cancer and Leukaemia Group B (CALGB) 9344 Table 19.4 Results of trial CALGB 9344 in terms of disease-free survival (DFS) and overall survival (OS) at 5 years, in relation to the dose intensity of doxorubicin Dose intensity of doxorubicin (mg/m2) DFS % OS %
60
70
90
p
69 79
66 79
67 77
0.60 0.31
Adapted from M. Campone et al.72
trial tested three dose levels of doxorubicin (60, 70 and 90 mg/m2) combined with cyclophosphamide 600 mg/m2 in a factorial 2 2 design also evaluating paclitaxel.71** No differences in disease-free or overall survivals were reported over this dose range (Table 19.4).72 However, a significant threshold effect, rather than a dose response, is suggested by the French Adjuvant Study Group-05 (FASG05) trial, which reported 10-year overall survival rates of 50 per cent and 55 per cent in 565 patients randomized to six cycles of F500E50C500 or F500E100C500 (p 0.05).73** Influenced by these data, six cycles of F600E75C600 represent a common UK standard, although four cycles of Epi100 plus four cycles of classical CMF are also widely used following the National breast cancer study of Epirubicin + CMF v classical CMF Adjuvant Therapy (NEAT) trial (see below).
Adjuvant taxanes The 2005 EBCTCG overview did not address the role of adjuvant taxanes, drugs that interfere with microtubule disassembly. Randomized trials are currently testing many different aspects of taxane therapy, including issues surrounding added benefit to anthracycline regimens, scheduling, dose density etc. For example, 11 000 patients have participated in trials testing the value of adding taxanes to anthracycline-containing schedules, either sequentially (CALGB 9344, NSABP B-28, Programme Adjuvant Cancer du Sein 01 (PACS-01)) or concomitantly (MDAnderson Hospital (MDAH), Breast Cancer International Research Group (BCIRG) 001).72 Docetaxel and paclitaxel were licensed for node-positive patients on the basis of BCIRG 001 and CALGB 9344 trials, respectively. Of these five trials, two reported no benefit in overall survival, including the NSABP B-28 (four cycles of A60 mg/m2C600 mg/m2 four cycles of paclitaxel225 mg/m2) and MDAH (four cycles of F500A50C500 plus four cycles of FAC or four cycles of paclitaxel250) trials. The three other trials were positive for overall survival, including the CALGB 9344 (four cycles of A60/70/90 mg/m2 C600 mg/m2
four cycles of paclitaxel175 mg/m2), BCIRG 001 (six cycles of T(docetaxel)75A50C500 or F500A50C500) and PACS-01 (six cycles of F500E100C500 versus three cycles of FEC plus three cycles of docetaxel100). The benefits correspond to a reduction in the hazard ratio for death of 20 per cent (Table 19.5).72 Toxicities include bone-marrow suppression,
Table 19.5 Results of trial CALGB 9344, NSABP-B-28, BCIRG 001 and PACS 01 in terms of disease-free survival (DFS) and overall survival (OS) at 5 years CALGB 9344
NSABP-B-28
AC AC:P p
AC
AC:P p
FAC
TAC
p
FEC
FEC:T
p
FEC
FEC:P
p
72 85
76 85
68 81
75 87
0.001 0.008
73.2 86.7
78.3 90.7
0.0014 0.0017
79 92
85 94
0.0009 0.13
DFS (%) 65 70 OS (%) 77 80
0.0023 0.00064
BCIRG 001
0.008 0.46
PACS 01
GEICAM 9906
AC, doxorubicin, cyclophosphamide; P, paclitaxel; FAC, 5-fluorouracil, doxorubicin, cyclophosphamide; TAC, docetaxel, doxorubicin, cyclophosphamide; FEC, 5-fluorouracil, epiribicin, cyclophosphamide; T, docetaxel Adapted from M. Campone et al.72
Cytotoxic therapies for early breast cancer 479
(b) Disease progression
(a) Death Avril/Mauriac
Avril/Mauriac
Danforth
Danforth
Gazet
Makris
Makris NSABP B18
NSABP B18
Scholl
Scholl/Broet
Scholl/Broet
Semiglazov
Semiglazov
Van der Hage
Van der Hage ALL 2
ALL 4
6
8
1
2
4
2
4
6
8
1
2
4
Risk ratio (95% CI) for neo-adjuvant vs. adjuvant treatment
Risk ratio (95% CI) for neo-adjuvant vs. adjuvant treatment
(d) Loco-regional recurrence
(c) Distant recurrence
Avril/Mauriac
Avril/Mauriac
Danforth
Danforth
Gazet
Gazet
Makris
Makris
NSABP B18 Scholl
NSABP B18
Scholl/Broet
Scholl/Broet
Semiglazov Semiglazov
Van der Hage ALL
ALL 2
4
6
8
1
2
4
Risk ratio (95% CI) for neo-adjuvant vs. adjuvant treatment
2
4
6
8
1
2
4
Risk ratio (95% CI) for nee-adjuvant vs. adjuvant treatment
Figure 19.4 Meta-analysis for primary outcomes with neoadjuvant therapy compared with adjuvant therapy for breast cancer. In each panel, each study [Van der Hage et al.,180 Avril et al./Mauriac et al.,181,182 Semiglazov et al.,183 Scholl et al.,184 Scholl et al.,185 Broet et al.,186 Makris et al.,187 NSABPB-18,188,189 Gazet et al.,190 Danforth et al.,191] is shown by the point estimate of the risk ratio (square proportional to the weight of each study) and 95% confidence interval (CI) for the risk ratio (extending lines); the summary risk ratio (ALL) and 95% confidence intervals by fixed effects calculations are also shown by diamonds. For all panels, values greater than 1 indicate that neoadjuvant treatment has a worse outcome compared with adjuvant treatment. (A) Death. (B) Disease progression. (C) Distant disease recurrence. (D) Loco-regional disease recurrence. Arrow 95% confidence interval extends beyond the depicted range. (With permission of Oxford University Press and Mauri et al.77)
hypersensitivity, neuropathy, myalgia and arthralgia. Meanwhile, taxanes have become the gold standard in the USA for node-positive patients, and docetaxel has been accepted for use with doxorubicin and cyclophosphamide for node-positive patients in Scotland. Residual doubts focus on the efficacy of the control regimens used in several of the trials. For example, it is not clear that taxane regimens would be superior to four cycles of Epi100 plus four cycles of classical CMF as tested in the NEAT trial.74** At the 2005 St Galen Conference, only 44 per cent of experts recommended adjuvant taxanes for node-positive patients with ER-negative disease, 24 per cent if the tumour was ER positive. An association between ER-negative status and taxane benefit reported by an unplanned subgroup analysis by CALGB 9344 has not been reproduced by other trials. Meanwhile, in the UK, NICE is due to publish a full appraisal of adjuvant taxanes in June 2007.
Dose-dense chemotherapy The CALGB/INTC 9741 trial randomized 2005 women to 3-weekly versus 2-weekly concomitant or sequential A60C600 4, using filgrastim in the test arm to support marrow. The hazard ratio for relapse after 2-weekly versus 3-weekly cycles of AC was 0.74 (p 0.01) and for overall survival it was 0.67 (p 0.13), based on a total of 315
relapses or deaths.75 There are a number of ongoing trials testing a similar strategy that have yet to report, including the Canadian MA20 trial and US Breast Intergroup 40101. Other approaches include the evaluation of metronomic chemotherapy based on weekly or daily regimens.76
Neoadjuvant/primary cytotoxic therapy In the last two decades, the potential benefits of preoperative versus postoperative chemotherapy have been tested in terms of disease-free and overall survival. A recent metaanalysis of nine randomized trials involving a total of 3946 patients reported no differences in death, disease progression or distant recurrence between randomized arms (Fig. 19.4).77 However, there appears to be a significant excess risk of loco-regional recurrence after neoadjuvant chemotherapy, with an odds ratio of 1.22 (95% CI 1.04–1.43). The excess loco-regional recurrence risk is confined to trials with an excess of patients in the neoadjuvant arm that avoid surgery altogether, suggesting that complete clinical/radiological response is not a safe basis for proceeding directly to radiotherapy. One of the most significant finding in the recent NSABP B-27 trial is that, although pathological complete response (pCR) increased from 13.7 per cent to 26.1 per cent (p 0.01) when four
480 Breast cancer
cycles of neoadjuvant docetaxel100 were added to four cycles of AC alone (n 1609), the disease-free and overall survivals were no different in the two arms.78** Although pCR is generally regarded as correlating with prognosis, the NSABP B27 result appears to challenge the validity of primary tumour response as a predictor of anti-metastatic therapeutic effect. Currently, neoadjuvant chemotherapy is justified as part of a research protocol, or to down-stage primary tumours in patients who would otherwise need mastectomy.
Combined endocrine and cytotoxic therapies The 2005 systematic overview of the EBCTCG confirms independent and additive benefits of 5 years’ adjuvant tamoxifen and several months of polychemotherapy in women with ER-positive disease.26*** In the absence of tamoxifen, the added benefit of ovarian suppression in conjunction with chemotherapy in pre-menopausal women with ER-positive disease depends on patient age (probability of chemical castration) and on the concurrent use of tamoxifen. Gonadotrophin-releasing hormone agonists added to the disease-free survival of women below the age of 40 years (but not 40 years) after six cycles of classical CMF in the IBCSG VIII trial.79** In the presence of tamoxifen it is unclear if ovarian ablation offers an advantage, even in women below the age of 40 years, regardless of chemotherapy, a question that will be addressed by the BIG SOFT trial. The UK ABC trial failed to demonstrate benefits of ovarian ablation in addition to 5 years’ tamoxifen
chemotherapy.80 If chemotherapy and tamoxifen are to be combined, concomitant is associated with an elevated risk of thromboembolism compared to sequential therapy regimens, with several excess thromboembolic events per 100 women treated.81** In addition, a comparison of concurrent versus sequential administration of tamoxifen plus FAC in postmenopausal hormone-receptor-positive patients reported superior disease-free survival at 8 years for the sequential regimen (67 per cent versus 62 per cent; p 0.045), adding weight to a preference for the sequential use of these modalities.82
Prediction of benefit after adjuvant endocrine and cytotoxic therapies The best-studied markers are the ER and PR, nuclear proteins detectable in about 70 per cent of tumours, that predict for benefit of adjuvant endocrine therapies.83,84 Patients with ER-negative tumours do not benefit, although it is likely that the 5 per cent of patients who are ER negative PR positive are endocrine responsive. There is fairly strong evidence indicating that over-expression of EGFR (cerb-B1) and/or HER-2 (cerb-B2) in ER-positive tumours predicts for resistance to tamoxifen, but for retained sensitivity to AIs.61
Over-expression of HER-2 down-regulates PR, and offers an explanation for why ER-positive PR-negative tumours are resistant to tamoxifen but retain sensitivity to AIs.85 Recently, a 21-gene expression profile has been reported to identify the subgroup of node-negative patients with ER-positive tumours who derive survival benefit from 5 years’ adjuvant tamoxifen.86 Prediction of benefit to adjuvant cytotoxic chemotherapy is an intensive research area. As discussed above, patients with ER-negative tumours derive more benefit than those with ER-positive tumours.67 Conventional markers, including measures of apoptosis and proliferation performed on pre-treatment and post-treatment biopsies of patients treated with neoadjuvant chemotherapy, have not proved clinically useful, although biologically informative. HER-2 over-expression appears to predict for relative resistance to CMF and sensitivity to anthracycline or taxane schedules.87 Over-expression of topoisomerase II, which may be coamplified with the HER-2 gene, is also associated with sensitivity to anthracyclines.88 In addition to the use of HER-2 testing in selecting patients for trastuzumab, there is a suggestion that HER-positive patients may be more sensitive to carboplatin (see below under ‘Biological therapies in early breast cancer: Adjuvant trastuzumab); Intense interest is currently directed at expression profiles and their immunohistochemical correlates that identify subgroups of tumours sensitive to specific agents. One example is patients with triple negative tumours (ER negative, PR negative, HER-2 negative) with DNA repair defects that render tumour cells sensitive to platinum adducts.
BIOLOGICAL THERAPIES IN EARLY BREAST CANCER Adjuvant trastuzumab The outstanding event of 2005 was the early reporting of adjuvant trials testing trastuzumab (Herceptin) in conjunction with cytotoxic chemotherapy in HER-2-positive cancer (IHC or FISH). The NSABP B-31 and the North Central Cancer Treatment Group (NCCTG) trials, combined for the analysis of 3351 patients, tested four cycles of A60C600 followed by 12 weeks of once-weekly paclitaxel80 or four cycles of 3-weekly paclitaxel125 12 months of trastuzumab delivered with the taxane.89** There was an absolute 12 per cent reduction in the rate of relapse at 3 years and a 33 per cent reduction in the risk of death in women allocated trastuzumab (p 0.015) (Fig. 19.5). The Herceptin in Adjuvant Breast Cancer (HERA) trial randomized 5000 women to 0, 1 or 2 years of adjuvant trastuzumab, starting after the completion of standard adjuvant cytotoxic therapy. Reporting the comparison between 1 year of trastuzumab versus none, the hazard ratio for recurrence after drug was 0.54 (95% CI 0.43–0.67, p 0.001). So far, the odds ratio for death is not significant.90** The use of concomitant taxane–trastuzumab in
Radiotherapy in early breast cancer 481
Trastuzumab (133 events)
100
Trastuzumab (62 deaths)
100
94.3% 87.1%
90
90 85.3%
91.7% Overall survival (%)
Disease-free survival (%)
91.4%
Control (92 deaths)
80 Control (261 events) 75.4%
70
67.1% 60
80 86.6% 70
60 P=0.015 Hazard ratio, 0.67
P < 0.0001 Hazard ratio, 0.48 50
50
0
0 0
1
2
3
4
5
Years after randomization
(a)
1
0
2
3
4
5
Years after randomization
(b)
No.at Risk
3351
2379
1455
801
133
0
No.at Risk
3351
2441
1571
908
165
0
Control
1679
1162
689
374
59
0
Control
1679
1200
766
448
83
0
Trastuzumab
1672
1217
766
427
74
0
Trastuzumab 1672
1241
805
460
82
0
the North American trials was associated with a 4 per cent risk of cardiac failure, compared to less than 1 per cent in the control group, an effect not seen in the HERA trial using sequential chemotherapy and trastuzumab. Meanwhile, a BCIRG trial has confirmed the benefits of adjuvant trastuzumab concomitantly with four cycles of AC plus six cycles of docetaxel. A third arm evaluating six cycles of carboplatin and docetaxel was not significantly better or worse than AC–docetaxel in terms of metastasisfree survival, but lacked the excess cardiotoxicity.91** A small randomized Finnish trial suggested that 3 months of trastuzumab may be as effective as 12 months.92** Meanwhile, adjuvant trastuzumab in combination with adjuvant cytotoxic therapy is currently being licensed in the UK and will be implemented in routine practice before evaluation by NICE.
BISPHOSPHONATES IN EARLY BREAST CANCER In three randomized trials testing adjuvant oral clodronate in a total of 1653 women with early breast cancer, a reduction of borderline statistical significance in the incidence of bone metastases (HR 0.82, 95% CI 0.66–1.01) and a statistically significant reduction in mortality (HR 0.82, 95% CI 0.69–0.97) have been reported.93**,94**,95** They, and the more powerful bisphosphonates, pamidronate, zolendronate and ibandronate, are currently only appropriately prescribed in the context of well-designed randomized clinical trials.
Figure 19.5 Kaplan–Meier estimates of disease-free survival (Panel A) and overall survival (Panel B). The hazard ratios are for the comparison of the trastuzumab group with the control group. (Copyright (c) 2005, Massachusetts Medical Society. All rights reserved.88)
RADIOTHERAPY IN EARLY BREAST CANCER Effects of radiotherapy The 2005 overview of radiotherapy effects in 25 000 women by the EBCTCG confirmed a 67 per cent odds reduction in local recurrence and an 11 per cent odds reduction in breast cancer mortality in patients randomized to postsurgical radiotherapy96*** (Table 19.6). On average, the prevention of four local recurrences prevents one breast cancer death from metastases, a causal association. In 7300 women with predominantly node-negative disease treated by tumour excision plus axillary dissection, randomization to breast radiotherapy reduced the 5-year local recurrence risk from 26 per cent to 7 per cent (reduction 19 per cent) and the 15-year breast cancer mortality from 35.9 per cent to 30.5 per cent (reduction 5.4 per cent). In 8500 women with predominantly node-positive disease randomized after mastectomy and axillary surgery, radiotherapy (predominantly loco-regional) reduced the isolated local recurrence at 5 years from 23 per cent to 6 per cent (reduction 17 per cent) and the 15-year breast cancer mortality from 60.1 per cent to 54.7 per cent (reduction 5.4 per cent). The excess non-breast-cancer mortality (half due to heart disease and stroke; other causes including lung and oesophageal cancers) was not seen in trials started since 1975. The greatest absolute gain from improved loco-regional control is seen in axillary-node-positive patients, in whom the absolute reduction in breast cancer deaths is 8–10 per cent. These effects, which are additive with those of adjuvant systemic therapies, confirm that axillary metastases are the first source of
482 Breast cancer
Table 19.6 Effects of age and tumour characteristics on 5-year risks of local recurrence in trials of radiotherapy (RT) (a) after breast conserving surgery (BCS) in women with node-negative disease and (b) after mastectomy and axillary clearance (AC) in women with node-negative disease 5-year local recurrence risk (%) in trials of: (a) RT after BCS (node negative)
(b) RT after mastectomy and AC (node positive)
RT versus control
Absolute reduction (SE)
RT versus control
Age (years) 50 50–59 60–69 70
11 vs 33 7 vs 23 4 vs 16 3 vs 13
22 (2) 16 (2) 12 (1) 11 (2)
6 vs 23 6 vs 24 5 vs 23 –
17 (1) 18 (2) 18 (2) –
Tumour grade Well differentiated Moderately differentiated Poorly differentiated
4 vs 14 9 vs 26 12 vs 34
10 (2) 17 (2) 22 (3)
4 vs 22 4 vs 30 6 vs 40
18 (3) 26 (2) 34 (4)
5 vs 20 14 vs 35 –
15 (1) 21 (3) –
5 vs 22 6 vs 30 8 vs 36
17 (2) 24 (2) 28 (4)
12 vs 30 6 vs 25
18 (3) 19 (2)
8 vs 28 6 vs 24
20 (2) 18 (20)
– – 7 vs 23
– – 16 (1)
4 vs 16 12 vs 26 6 vs 23
Tumour size (mm) (T category) 1–20 (T1) 21–50 (T2) 50 (T3 or T4*) ER status ER poor ER positive Number of involved nodes 1–3 4 All women
Absolute reduction (SE)
12 (2) 14 (2) 17 (1)
See webfigures 6a and 6b to the EBCTCG article97 for more details on characteristics, including separate results for those in whom the relevant characteristic is not known. *T4 tumour of any size with direct extension to skin or chest wall. ER, oestrogen receptor; SE, standard error. Adapted from Early Breast Cancer Trialists’ Collaborative Group (EBCTCG).97
distant metastases in a significant minority of patients. In selecting patients for post-mastectomy chest-wall radiotherapy, the data indicate that two breast cancer deaths are avoided by irradiating 100 patients, with a residual 10 per cent risk of chest-wall recurrence after optimal surgery and systemic therapies. The EORTC randomized boost trial in more than 5000 women confirmed a 43 per cent reduction in the odds of local recurrence following 16 Gy in eight fractions to the tumour bed after 50 Gy in 25 fractions to whole breast, regardless of patient age and tumour characteristics.97** The absolute benefit was greatest in women aged 40 years or less, in whom the local recurrence risk was reduced from 19.5 per cent to 10.2 per cent (reduction 9.3 per cent) at 8 years. Women aged over 50 years gained little in absolute terms from the boost, because local recurrence risk without boost was less than 5 per cent. The high risk of local recurrence in young women appears to reflect a greater probability of bulky foci of residual DCIS after local excision than in older women.98
Axillary radiotherapy is usually given to all patients in the absence of axillary surgery, on the basis that it reduces the morbidity of uncontrolled axillary recurrence,99 and contributes to cure. After axillary clearance (level II/III), axillary radiotherapy is not usually given, even if nodes are positive or in the presence of extra-capsular spread, due to an enhanced risk of treatment-related morbidity.100,101 In the uncommon subgroup with microscopic tumour at the margins of resection, treatment is justified. If sentinel node biopsy or axillary sampling is positive on standard haematoxylin and eosin sections and level II/III axillary dissection is not performed, radiotherapy is a satisfactory alternative. The need for radiotherapy in patients with single micrometastatic (2 mm) foci of disease after sentinel node biopsy is being evaluated in the EORTC AMAROS trial. Patients with positive axillary nodes have up to 5 per cent risk of supraclavicular fossa (SCF) node involvement. In a group of heavily axillary-node-positive patients, radiotherapy to the SCF is expected to prevent more morbidity than it causes in terms of brachial plexopathy and
Radiotherapy in early breast cancer 483
Table 19.7 Randomized trial testing excision tamoxifen in pT1 ER pN-, age 50 years Treatment after tumour excision
n
Local Recurrence (%)
Follow-up (years)
NSABP B21 Fisher 2002175
Tamoxifen only Breast radiotherapy tamoxifen
334 334
16.5 28
8
Canada Fyles 2004176
Tamoxifen only Breast radiotherapy tamoxifen
383 386
7.7 0.6
5
CALGB Hughes 2004177
Tamoxifen only Breast radiotherapy tamoxifen
319 317
4.0 1.0
5
GBSG Winzer 2004178
Tamoxifen only Breast radiotherapy tamoxifen
80 94
2.8 3.8
6
ABCSG-8A Gnant 2005179
Tamoxifen only
416
3.2
4
Study
pT1, pathological tumour size 2 cm; ER , oestrogen receptor positive; pN-, pathological node status negative
lymphoedema.99,102 Against this, SCF radiotherapy contributes to vascular disease, including stroke.103 About 30 per cent of patients with positive axillary nodes have positive internal mammary chain (IMC) nodes.104 Uncertainties persist with regard to the beneficial and adverse effects of radiotherapy to the IMC. Until the EORTC 22922 and NCIC MA20 trials have clarified its role, IMC radiotherapy is not yet part of standard treatment in the UK, although practised in some other countries.
Effects of withholding radiotherapy There is, as yet, no reliably identified group of patients whose breast recurrence risk is less than 10 per cent at 10 years after local excision alone. A summary of randomized trials testing radiotherapy in patients with small node-negative ER-positive tumours treated with local excision plus tamoxifen is shown in Table 19.7. A review of 5352 women entered in International Breast Cancer Study Group (IBCSG) trials between 1978 and 1993 presented an algorithm for estimating chest-wall relapse risk after mastectomy without radiotherapy (Table 19.8).105 Patients received variable adjuvant systemic therapies as follows: 3–12 months of classical CMF if axillary-node positive and/or tamoxifen. Two-thirds of node-negative patients received a single cycle of classical CMF. Overall, one-third received no adjuvant therapy. The 2005 EBCTCG update of systemic therapy effects confirms substantial reductions in the annual hazards of local recurrence in women randomized to 5 years’ tamoxifen and/or several months of polychemotherapy.96*** The reduction in local recurrence risk after adjuvant systemic therapies is comparable to the reduction in metastasis risk, being 53 per cent after 5 years’ tamoxifen regardless of age,
47 per cent after several months of polychemotherapy in women aged less than 50 years and 30 per cent in the 50–69year age group.96 If 10 per cent chest-wall recurrence risk represents the appropriate lower threshold for considering chest-wall radiotherapy, it is not clear what combination of tumour and treatment characteristics identifies this subgroup in the current era. This uncertainty is being addressed by the Selective Use of Postoperative Radiotherapy aftEr MastectOmy (SUPREMO) trial, which tests the benefits of chest-wall radiotherapy following mastectomy and optimal axillary and systemic therapy in intermediate-risk patients, most typically those with one to three positive axillary lymph nodes.
Duct carcinoma in situ In all three prospective randomized trials testing the effects of breast radiotherapy after complete local excision of pure DCIS, the total ipsilateral tumour recurrence risk after surgery alone was in the range 20–30 per cent at 10 years, of which roughly one-half were invasive tumours.106**,107**,108** The risk of recurrence after local excision of symptomatic DCIS has been reported by Silverstein to be related to tumour size, nuclear grade, patient age and margins of excision.109 Risks of ipsilateral recurrence at 12 years after local excision alone in Silverstein’s series were as follows: score 4–6 5 per cent; score 7–9 20 per cent; score 10–12 80 per cent.22,110 The Van Nuys Prognostic scoring index system (VNPI) has not been independently validated as a guide to therapy. Retrospective evaluation of trial data by the NSABP and EORTC failed to confirm the reliability of the VPNI, although their ability to test the index properly has been challenged.107,111
484 Breast cancer
if VPNI 6 8. Patients with VPNI 9–12 are recommended mastectomy.
Table 19.8 Chest-wall recurrence risk 10 years after mastectomy and selective systemic therapy but without radiotherapy Age (years) 50
50
pT
pN
VI
1 1 1 1
– – – – 1–3 1–3 1–3 1–3 1–3 1–3 4 4 4 4 4 4
1 1 1 1 1 1 1 1 1 1 1 1
1–3 1–3 1–3 1–3 1–3 1–3 4 4 4 4 4 4
Grade
10-year risk (%)
1 1 2 2 3 3 1 1 2 2 3 3
10 15 15 15–20 10–15 10–15 10–15 15–20 10–15 25–30 15–20 25–30 25–30 35 25–30 35
1 1 2 2 3 3 1 1 2 2 3 3
10–15 10–15 10–15 15–20 15–20 20–25 20–25 20–25 20–25 30–35 30–35 30–35
PT, pathological tumour size; pN, pathological nodal status and VI, vascular invasion Adapted from Wallgren et al.106
Whole-breast radiotherapy reduces the risk of ipsilateral breast relapse after complete excision of pure DCIS. The NSABP B-17, EORTC and UK DCIS trials indicate that the risks of recurrent DCIS and invasive tumours are each reduced by a factor of approximately 2.106–108 The dose prescription in all trials was 50 Gy in 25 fractions (or similar) to the whole breast without a boost dose to the tumour bed. The NSABP recommends whole-breast radiotherapy for all DCIS patients treated by breast-conservation surgery, on the basis that there is currently no reliable way to identify subgroups of patients who might be safely spared radiotherapy. The UK DCIS II trial is testing endocrine therapy breast radiotherapy in patients with small ER-positive tumours. Despite the limited supporting evidence, some centres apply the VPNI to determine treatment allocation after complete microscopic excision, eg. no radiotherapy if VPNI 1 5; whole-breast radiotherapy
Age greater than 70 years Women over the age of 70 years have an average life expectancy of 15 years and account for one-third of breast cancer presentations.112 However, due to omission of the elderly from the majority of the clinical trials, there are few data defining the optimum treatment for breast cancer in this arbitrarily defined age group. Two recent audits of current UK practice have confirmed that the use of primary endocrine therapy is widespread, with 42 per cent of all women over the age of 70 years being treated in this way.113,114 A recent Cochrane Database review evaluated surgery versus primary endocrine therapy (tamoxifen) for operable primary breast cancer in elderly women.115 It was unable to show a statistically significant difference in favour of either in respect of overall survival, but there was a statistically significant difference in terms of progression-free survival, which favoured surgery with or without endocrine therapy. On the basis of these data, either local excision or mastectomy should be offered to most elderly women with early breast cancer except those who are unfit for or refuse it. Post-mastectomy radiotherapy reduces loco-regional recurrence (16 per cent versus 28 per cent, p 0.03) in elderly women with high-risk breast cancer (tumours 5 cm or 4 positive axillary nodes).116 Women in their seventies and eighties are usually fit enough to cope with radiotherapy if it is needed. There is no evidence that age per se influences late normal-tissue reactions.
Radiotherapy delivery The patient lies supine in a stable position that must be reproduced exactly during planning, simulation and treatment. After breast-preserving surgery, the clinical target volume includes the soft tissues of the palpable breast down to the deep fascia, but excludes underlying muscle, ribcage, overlying skin and scar. The shape and size of the clinical target volume after breast-preserving surgery correspond to the edges of the palpable breast, although this is imprecisely defined superiorly. After mastectomy, scar and skin flaps are target tissues. The planning target volume includes the clinical target volume plus a 1-cm margin on all borders. The breast/chest wall is encompassed with 6-MV X-rays (10 MV for wide separations). The maximum thickness of lung is limited to 2 cm when possible, and cardiac shielding should be considered if any myocardium is included in the treatment volume, even at the cost of shielding target tissues (there is a trade-off to be made). If lymphatic fields are used, divergence of the superior border of tangential fields must be eliminated, and posterior divergence of tangential fields into heart and lung must be avoided. Depending on the location of the primary tumour in the preserved breast,
Radiotherapy in early breast cancer 485
(a)
Alternatively, an interstitial implant can be used. The clinical target volume of lymphatic fields is defined according to fixed anatomical landmarks, including soft tissue and bony structures. The field arrangement includes an anterior field with the reference points at the build-up depth, i.e. 100 per cent. A daily posterior axillary contribution may be added in order to ensure that the midline axilla dose does not fall below 80 per cent or 85 per cent; this involves a posterior contribution of approximately 10 per cent in the current UK Standardisation of breast RadioTherapy (START) trial. Radiotherapy dose to the breast/chest wall is prescribed to a reference point at or near the centre of the target volume, typically halfway between the anterior skin surface and the lung/chest wall interface.117 Unless the dose distribution is within 95 and 107 per cent of prescribed dose at all levels, three-dimensional dose compensation should be introduced (Figs 19.7–19.9). Radiotherapy dose and fractionation for the whole breast is in the range 46–50 Gy to the International Commission on Radiation Units (ICRU) reference point in 2.0-Gy fractions. The EORTC boost dose is 16 Gy to the reference point (100 per cent if electrons are used) in 2.0-Gy fractions. A variety of alternative fractionation schedules are in common use that aim to improve treatment outcome by reducing the number of fractions, including 40 Gy in 15 fractions, 42.5 Gy in 16 fractions and 45 Gy in 20 fractions. Greater standardization is anticipated after the UK START trial reports its results. The reproducibility of field localization should be monitored by machine check films or electronic portal imaging, the latter increasingly purchased as standard accessories with new treatment machines (Fig. 19.10).
Scheduling of radiotherapy and chemotherapy
(b)
Figure 19.6 (a) The apex of the heart (green) exposed to radiotherapy in a standard open tangential field. (b) The apex of the heart (green) protected throughout every exposure using one leaf of a multi-leaf collimator that moves into the beam in the longitudinal axis (See Plate Section.).
the medial and/or lateral field borders can be adjusted to reduce the amount of lung/heart in the tangential fields. Heart shielding is easily implemented using a multi-leaf collimator or a manually placed shielding block, unless this involves shielding the tumour bed (Fig. 19.6). The diameter of an electron boost field is influenced by the size of the original tumour, the amount of surrounding normal tissue removed by the surgeon and the accuracy of localization.
Several randomized trials have now addressed the scheduling of chemotherapy and radiotherapy. A trial from Boston tested adjuvant chemotherapy before radiotherapy against the reverse sequence in 244 women after primary surgery for early breast cancer.118** Small sample size prevented firm conclusions being drawn, but no differences in local relapsefree or overall survivals were reported. A randomized trial from Paris reported significantly more frequent and severe late adverse effects after concurrent compared to sequential chemo-radiotherapy using mitoxantrone, fluorouracil and cyclophosphamide after breast-conservation surgery in 297 women recalled for assessment a median of 6.7 years (range 4.3–9 years) post-treatment.119** There is no evidence that a gain in loco-regional control justifies the added morbidity, and it is currently usual to complete chemotherapy before starting radiotherapy, especially when anthracyclines are used.
Cardiac effects of radiotherapy Analysis of the different techniques used in the Scandinavian randomized trials of post-mastectomy radiation suggests
486 Breast cancer
(a)
(b)
Figure 19.7 Poor two-dimensional dosimetry in a standard breast plan using an open tangential field: (a) transverse view; (b) sagittal view (See Plate Section.).
(a)
(b)
Figure 19.8 Improved dosimetry using IMRT avoids hotspots and ensures better coverage of chest wall: (a) transverse view; (b) sagittal view (See Plate Section.).
that the risk of ischaemic heart disease is associated with radiation dose rather than volume.120 The implication is that reducing volume does not reduce risk unless the heart is totally excluded, presumably because the anterior descending coronary artery lies within the high-dose volume whenever cardiac tissue is exposed. On the precautionary principle, and recognizing the potential for enhancement by anthracyclines in chemotherapy patients, it is usually preferable to protect the myocardium encroaching in the treatment volume. This is fairly straightforward in patients treated by breast conservation, because most tumours are
in the central or upper quadrants and more than 75 per cent of tumour recurrence risk is concentrated in the index quadrant. It is more difficult to apply after mastectomy, when cardiac protection may shield skin very close to, or including, the medial end of the scar, and other strategies are probably superior, e.g. treatment gated to respiration cycle. The increased risk of non-breast second malignancies is another important complication of treatment to consider, including contralateral breast cancer, sarcomas, oesophageal and lung cancers.
Special presentations of early breast cancer 487
these risk factors may reflect a genetic predisposition in a minority of patients. Where the management of bilateral or contralateral tumours is concerned, each cancer is treated on its own merits.
Pregnancy
(a)
(b)
Figures 19.9 This explains how the poor two-dimensional dosimetry in Figure 19.7 is improved in Figure 19.8 by correcting for dose variation using computer-controlled leaves of a multileaf collimator which move into the beam along a transverse axis at the appropriate time during each daily treatment to shield tissue that has received the prescribed dose. (a) LAO IMRT. (b) RAO IMRT.
Figure 19.10 Electronic portal imaging verification of cardiac shielding during treatment.
SPECIAL PRESENTATIONS OF EARLY BREAST CANCER Bilateral breast cancer The incidence of bilateral primary tumours at presentation is about 1 per cent. In patients presenting with unilateral disease, the annual incidence of a contralateral primary is about 0.75 per cent. Risk factors include known or suspected genetic predisposition, young age at first presentation, a positive family history, multi-centric disease in the ipsilateral breast and lobular histological subtype. Some of
It has been estimated that up to 3 per cent of breast cancers may be diagnosed in pregnant women.121 Breast cancer arising during pregnancy presents a particularly difficult set of clinical problems to resolve, where the interests of the fetus and mother may not coincide.122,123 Studies have shown that patients with pregnancy-associated breast cancer commonly present with pathological lymph-node involvement (56–67 per cent), large tumours and lymphovascular invasion.124 In case-matched control studies, ERnegative tumours have been found to be more common in pregnant patients than in age-matched controls, although HER-2 staining pattern has been inconclusive due to small numbers.125 The diagnosis may be delayed and difficult owing to the physiological changes within the breast and limitations on investigations. Haematology and a biochemical screen, including liver function tests and alkaline phosphatase, are needed to supplement the clinical assessment. Breast ultrasound should be used in preference to mammography. A chest X-ray should not be performed routinely, but may be done with appropriate shielding, when clinically indicated.126 Computed tomography (CT) scans are usually avoided, and metastases are sought using alternative imaging modalities such as ultrasonography. There are theoretical risks to the fetus from exposure to the high magnetic fields of MRI, and the UK Medical Devices Agency recommends avoiding MRI scans during the first trimester until more information becomes available.127 Management is critically influenced by the stage of the pregnancy and should be decided by a multi-disciplinary team, incorporating not only the breast surgeon, medical and clinical oncologists, but also obstetricians, neonatologists and specialist nursing staff in order to ensure the best possible outcome for both mother and infant. Small case series suggest that breast-conserving surgery, mastectomies and axillary surgery can all be performed safely with no unexpected complications,128 although anaesthetic risks will have to be considered. The safety of sentinel lymph-node biopsy in pregnant patients is not known. Irradiation in the peri-implantation and early postimplantation period (up to 8 days) may lead to embryonic death.129 During organogenesis (up to 8 weeks), the fetus is at its most sensitive to radiation-induced malformations, which may occur with exposure to more than 0.05 Gy. Other risks include intrauterine growth delay, microcephaly and mental retardation. In terms of adjuvant breast radiotherapy, fetal dose exposure may be as low as 0.036–0.038 Gy when completed by the sixth week of gestation, but exposure may increase markedly later in pregnancy as the fetus moves closer to the radiation field.130
488 Breast cancer
These exposures may still put the fetus at risk, and therefore adjuvant breast radiotherapy is usually delayed until after delivery. The physiological changes observed in pregnancy may alter the pharmacokinetics and pharmacodynamics of chemotherapy in the mother. The amniotic fluid may also act as a pharmacological third space and delay elimination of agents such as methotrexate. Chemotherapy is avoided in the first trimester, as the estimated risk of fetal malformations is up to 17 per cent.122,123 The French National Survey, the London Hospitals Series and the MD Anderson Cancer Center Prospective Study suggest that in the short term chemotherapy can be safely administered to women within the second and third trimesters of pregnancy.128,131*,132 The only prospective series treated 24 women with doxorubicin 50 mg/m2 as a continuous infusion over 72 hours, cyclophosphamide 500 mg/m2 on day one and bolus 5-fluorouracil 500 mg/m2 on days one and four, all administered every 3–4 weeks.131 However, little is known about the long-term effects on the fetus of in-utero chemotherapy exposure. Large prospective studies are needed to provide further information regarding the sequelae of treatment. Tamoxifen may potentially be teratogenic, as shown by animal studies, and therefore its use is usually delayed until the end of pregnancy.133 The use of trastuzumab during pregnancy cannot currently be recommended, as HER-2 expression is seen in embryonic tissues, suggesting a role in embryonic development.125 The long-term effects of pamidronate on bone growth and development in the neonate are not known, although there have been reports of its use during the third trimester with resulting reversible neonatal hypocalcaemia.134 It is recommended that delivery occurs approximately 3 weeks after the last dose of chemotherapy so as to minimize the risk of maternal and fetal neutropenia and subsequent infection. The platelet count should be sufficient to prevent bleeding complications. The next dose of chemotherapy should usually be given after adequate time to recover from delivery. If treatment is to continue postpartum, breast-feeding during chemotherapy and hormonal therapy is contraindicated, as most of the agents used can be excreted in breast-milk. It is still reasonable to discuss termination if the fetus is likely to be exposed to significant risk of harm by potentially curative treatment given to the mother. In addition, some women with metastatic or high-risk cancer may not want to carry on with the pregnancy. Patients and their relatives must be provided with adequate counselling in order that they can make an informed rational decision. Casematched control studies show no reduction in life expectancy in women who become pregnant 6 months or more after treatment for early breast cancer. Patients are advised to wait at least several months after the end of radiotherapy, adjuvant systemic therapy and tamoxifen before trying to have a child. Depending on the intensity of chemotherapy, a majority of women under the age of 40 years and a minority
of pre-menopausal patients over 40 years resume ovulatory cycles.
Paget’s disease Paget’s disease is a form of breast cancer presenting with eczematous changes of the nipple, which may extend onto the adjacent areola and skin. It may be associated with nipple discharge, itching, burning, inversion or erosion, and clinical or mammographic evidence of an underlying primary tumour mass. Malignant cells in the epidermis are diagnostic and have a characteristic appearance with large, pale-staining cells with clear cytoplasm staining positively for mucin. The cells represent intra-epidermal spread from an underlying intra-duct or invasive adenocarcinoma. The commonest treatment is mastectomy, but there are now several reports of successful breast conservation with radiotherapy following removal of the nipple and underlying tumour with clear surgical margins.135
Male breast cancer Breast cancer in men represents less than 1 per cent of breast cancer. Risk factors include radiation exposure, a history of unopposed oestrogens resulting from primary or secondary testicular failure, including Klinefelter’s syndrome and exogenous oestrogens. An increased risk of male breast cancer has been reported in families with BRCA2 mutation on chromosome 13q.136 Patients present with a lump, with or without nipple retraction, bleeding, ulceration, deep fixation and axillary adenopathy. The main differential diagnosis is gynaecomastia. Staging procedures are the same as for female patients, and the treatment of choice for operable disease is mastectomy. The principles of axillary surgery are the same as for women, with axillary clearance reducing the risk of regional recurrence. There are no clear guidelines for the use of post-mastectomy radiotherapy because of small patient numbers and data confined to retrospective studies. In the absence of good data, it is reasonable to apply the same criteria as in women. In men with node-positive tumours, both chemotherapy and tamoxifen have been used and can increase survival to the same extent as in women with breast cancer. Currently, no controlled studies have compared adjuvant treatment options. Approximately 85 per cent of all male breast cancers are ER positive, and 70 per cent of them are PR positive. Hormonal therapy has been recommended in all receptor-positive patients.137 Tamoxifen use, however, is associated with a high rate of treatment-limiting symptoms, such as hot flashes and impotence, in male breast cancer patients.138 Overall survival is similar to that of women with breast cancer after stratification for known factors such as TNM stage, ER status and grade. The impression that male breast cancer has a worse prognosis may stem from the tendency towards diagnosis at a later stage.
Treatment of patients with locally advanced disease 489
Axillary adenopathy A presentation with enlarged axillary nodes accounts for less than 0.5 per cent of breast cancers. The differential diagnosis includes non-malignant causes such as infection and autoimmune disorders, malignancies of the lung, gastrointestinal tract, melanoma and lymphoma. By far the likeliest primary site if the lymph node contains metastatic adenocarcinoma is the breast, confirmed if ER positive, and patients should be investigated and treated accordingly with clinical assessment, mammography and, increasingly, MRI. In the absence of symptoms or signs of a primary tumour or distant metastases, the options for operable disease are as for any other patient with invasive cancer, including mastectomy (breast conservation if a primary tumour can be localized), axillary dissection, systemic therapies and radiotherapy. An occult breast primary is found in more than half the mastectomy specimens in the larger series.
Inflammatory breast cancer Inflammatory breast carcinoma (IBC) is a rare but aggressive subtype of breast cancer, which historically was considered uniformly fatal. However, the advent of combined-modality therapy has resulted in a marked improvement in prognosis. In general, women with IBC present at a younger age, are more likely to have metastatic disease at diagnosis, and have shorter survival than women with non-IBC.139 According to the latest revision of the American Joint Committee on Cancer Staging guidelines, inflammatory carcinoma is classified at T4d, which makes all patients with IBC stage IIIB, IIIC or IV, depending on the nodal status and presence of distant metastases.140 Clinically, IBC is characterized by the rapid onset of breast warmth, erythema and oedema (peau d’orange), often without a well-defined mass. The rapidity of growth can be used to distinguish true ‘primary’ inflammatory carcinoma from neglected locally advanced breast tumours that have developed inflammatory features. Inflammatory breast cancer is not associated with any particular histologic subtype.141 The characteristic pathologic finding is dermal lymphatic invasion by carcinoma, which can lead to obstruction of the lymphatic drainage resulting in the classical clinical picture. However, the diagnosis is made on clinical grounds, and the absence of dermal lymphatic invasion does not exclude the diagnosis. The presence of lymph-node involvement, extensive erythema, and p53 mutations and the absence of ER have all been associated with poorer outcomes in patients with IBC.142 Inflammatory cancers have distinct biological characteristics, including a high S-phase fraction, aneuploidy and increased angiogenesis, and are found to over-express RhoC GTPase and lose expression of LIBC gene (lost in IBC).143,144 With current multi-modality therapy, approximately one-third of women diagnosed with IBC will become longterm survivors. Although no standard management has been
defined for this disease, neoadjuvant chemotherapy (mostly anthracycline based), with surgery and radiotherapy for loco-regional disease, has become accepted practice, with or without additional systemic treatment after completion of local therapy. In the MD Anderson series, 28 per cent of patients were alive and without evidence of disease beyond 15 years using combination therapy.145 Initial response to induction chemotherapy was an important predictor of survival; disease-free survival at 15 years was 44 per cent in patients who had a complete response to induction chemotherapy, 31 per cent in those who had a partial response, and 7 per cent in patients who did not respond to therapy. In patients who have achieved a good partial or complete clinical response to chemotherapy, one study has questioned whether mastectomy is needed in addition to breast and regional nodal irradiation.146 Other series have shown that the addition of mastectomy to chemo-radiation improves local control and, in patients with an objective response to chemotherapy, also improves disease-free survival.147 Even after induction chemotherapy and local therapy, the rates of relapse remain very high. Thus, studies have recommended further adjuvant chemotherapy (anthracycline or taxane based) and 5 years of adjuvant hormone therapy in patients with ER-positive or PR-positive tumours.148 The diversity of treatment sequencing, chemotherapeutic agents used, and total duration of systemic treatment makes the comparison of clinical trials difficult. The role of high-dose chemotherapy followed by autologous stem-cell transplantation remains experimental. Current research on the biological characteristics of inflammatory carcinoma has made significant strides in the understanding of the aggressive behaviour of this tumour, and future research will be vital in developing targeted therapies.149
TREATMENT OF PATIENTS WITH LOCALLY ADVANCED DISEASE About 20 per cent of patients present with locally advanced disease without evidence of metastases. Locally advanced disease refers to tumours over 5 cm in diameter (T3), and tumours of any size with overlying oedema, chest-wall fixation, skin infiltration or inflammatory features (T4). In terms of management, the loco-regional disease is classified as operable (stage IIIA) or inoperable (stage IIIB). Several randomized trials have compared different local and systemic treatment strategies, either alone or in combination.150 The general consensus is that a multi-modality treatment approach is required involving systemic therapy, surgery and radiotherapy.
Primary surgery for patients with operable disease Mastectomy with axillary dissection is associated with loco-regional recurrence in at least 30 per cent of patients.
490 Breast cancer
Five-year survival rates for patients with positive nodes are in the range 40–50 per cent. A small subgroup of patients with large primary tumours and negative axillary histology has a more favourable outcome. On the basis of its proven effect in early-stage disease, it is assumed that post-mastectomy radiotherapy in patients with advanced local disease will also be useful in reducing the risk of loco-regional relapse. Retrospective studies of mastectomy with radiotherapy report loco-regional relapse rates in the range 10–20 per cent.
Primary medical therapy Intensive chemo-endocrine therapy as primary treatment in women with locally advanced breast cancer is becoming standard practice as the majority of patients gain at least a partial response, with approximately 10 per cent of patients achieving a complete pathological response.151 These response data lead to encouraging expectations that this approach will offer real benefits to patients in terms of loco-regional control, relapse-free survival and, possibly, overall survival in the future. The other main appeal of this approach is to use the tumour as an in-vivo measure of response. Evidence also shows that some patients treated in this way can achieve tumour regression sufficiently to avoid a mastectomy and allow breast-conserving surgery.152 Primary chemotherapy is generally anthracycline based. Poor/non-responders are either switched to second-line chemotherapy or proceed directly to surgery if possible. However, even with complete clinical/radiological response, patients must be referred for subsequent surgery, as the risk of residual microscopic disease is high.77*** With regard to primary endocrine therapy, current evidence suggests that letrozole is more effective than tamoxifen in terms of both clinical response and avoidance of mastectomy in postmenopausal patients with large ER-positive breast cancer, otherwise requiring a mastectomy.61** Therefore, ER-positive patients may be offered a therapeutic trial of letrozole prior to surgery. Following neoadjuvant therapy and surgery, patients are offered adjuvant endocrine therapy as in early breast cancer. Postoperative radiotherapy should be considered for most of these patients regardless of their response to initial chemotherapy.153 Radiotherapy alone, or as a prelude to surgery in patients with tumours of borderline operability, is increasingly reserved for patients who are too unfit for, or refuse, chemo-endocrine therapy. Ongoing trials are investigating the role of trastuzumab and other biological agents in the neoadjuvant setting.
MANAGEMENT OF LOCO-REGIONAL RECURRENCE Loco-regional relapse after mastectomy alone is associated with distant metastases on routine restaging in approximately 50 per cent of patients. In these women, appropriate systemic measures constitute the treatment of first choice. Patients with isolated loco-regional recurrence should be
considered for surgery and high-dose radiotherapy (if not given before). Approximately half the patients will be alive 5 years later, and approximately half will achieve long-term loco-regional control.154 The treatment of inoperable chestwall recurrences depends on several factors, including age, performance status, extent of recurrence and hormonereceptor positivity. Recurrence in the breast after breast-conserving surgery and radiotherapy for ductal carcinoma usually develops close to the primary site.155 The majority of local recurrences are discrete tumours in the breast parenchyma, which are operable by mastectomy or further wide excision. If staging investigations are negative and the disease operable, the prognosis of this selected subgroup is relatively good, with 70 per cent alive at 5 years. In patients entered into EORTC randomized trials, 5-year survival rates of 58 per cent were reported after ipsilateral breast relapse, no better than that of patients developing chest-wall recurrence after randomization to mastectomy.156
TREATMENT OF PATIENTS WITH METASTATIC DISEASE Presentation and evaluation As an initial presentation, metastatic breast cancer is uncommon, occurring in only about 6 per cent of newly diagnosed cases.157 However, approximately 30 per cent of women initially diagnosed with earlier stages of breast cancer eventually develop recurrent advanced or metastatic disease. Common sites of metastatic relapse include the skeleton, lung, pleura, liver, spinal cord and brain. Clinical assessment is used to guide staging investigations. Prognostic factors include relapse-free interval and the pattern of organ involvement. If the skeleton or pleura is the dominant site, the median survival is more than 2 years. Disease in the liver or central nervous system is associated with a median survival of less than 1 year.
Principles of management There is no single standard of care for patients with metastatic breast cancer. The treatment intent is essentially palliative and therefore, in addition to efficacy, issues of tolerability, quality of life and patient preference are important considerations when evaluating available options. The choice of treatment also depends on various patient and disease characteristics, including age/menopausal status, ER/PR and HER-2 receptor status, growth rate of disease, presence of visceral metastases, as well as history of prior therapy and response. Beyond the management of co-morbid conditions, the first choice of systemic treatment is often endocrine therapy supplemented by other measures, including radiotherapy, to palliate specific symptoms. The use of a sequential cascade of non-cross-resistant endocrine therapies is common in this setting. The order in which treatments
Complications of metastatic disease 491
are given depends on which prior adjuvant or first-line treatments patients have received and their responses to such agents. A durable symptomatic remission may be associated with less than a partial response, i.e. stable disease. Patients who relapse within 2 years of presentation, have multiple organ involvement, especially liver, or who have progressive symptoms of visceral disease are usually considered for cytotoxic therapy as first-line palliative treatment. Anthracycline-based and taxane-based therapies have traditionally shown the highest degree of activity in metastatic breast cancer. However, most patients with recurrent disease will already have had exposure to anthracyclines and possibly taxanes in the adjuvant setting. Other drugs that have demonstrated substantial activity in the metastatic setting include capecitabine, gemcitabine and vinorelbine.158 The use of combination therapy versus monotherapy or sequential single agents remains a controversial issue.159 Combination therapies generally result in higher overall response rates and time to disease progression than with sequential single agents, but usually at a cost of greater toxicity. Although numerous randomized clinical trials have shown improvements in overall response rates, few have found clear survival benefits. In recent years, however, there has been a small but growing series of clinical trials demonstrating modest but meaningful survival advantages in metastatic disease. Trials with chemotherapy and trastuzumab have demonstrated impressive improvements in overall survival, ranging from 4 to 8 months, representing increases of 24 per cent and 37 per cent in survival time.160**,161** In addition, early data from the combination of paclitaxel and bevacizumab also appear to support a survival benefit.162** Thus the potential for enhanced or synergistic activity forms a compelling argument for the use of these targeted biologic agents in combination with traditional chemotherapy.
COMPLICATIONS OF METASTATIC DISEASE Skeletal secondaries Skeletal metastases occur in up to 70 per cent of patients with metastatic breast cancer, and are responsible for major morbidity, including pain, fracture and spinal cord compression. Medical management involves use of the World Health Organisation ladder for pain control. Antidepressants, steroids, anticonvulsants and muscle relaxants also have a role, depending on specific clinical features. Other approaches to pain management include bisphosphonates, radiotherapy and chemo-endocrine therapies. Bisphosphonates interfere with tumour-mediated lysis by inhibiting osteoclast recruitment and function. For patients with plain radiographic evidence of bone destruction, ASCO guidelines recommend intravenous pamidronate 90 mg delivered over 2 hours or zoledronic acid 4 mg over 15 minutes every 3–4 weeks.163 There is insufficient evidence supporting the efficacy of one bisphosphonate over
the other. A recent Cochrane Review concluded that in women with advanced breast cancer and clinically evident bone metastases (osteolytic and/or mixed osteolytic/ osteoblastic), the use of bisphosphonates (oral or intravenous) in addition to hormone therapy or chemotherapy reduces the risk of developing a skeletal event as well as increasing the time to skeletal event.164*** Some bisphosphonates may also reduce bone pain and improve global quality of life. Of the currently available bisphosphonates, 4 mg intravenous zoledronate reduces the risk of developing a skeletal event by 41 per cent (Relative Risk (RR) 0.59, 95% CI 0.42–0.82), compared with 33 per cent by 90 mg intravenous pamidronate (RR 0.77, 95% CI 0.69–0.87), 18 per cent by 6 mg intravenous ibandronate (RR 0.82, 95% CI 0.67–.00), 14 per cent by 50 mg oral ibandronate (RR 0.86, 95% CI 0.73–1.02) and 16 per cent by 1600 mg oral clodronate (RR 0.84, 95% CI 0.72–0.98). The optimal timing of the initiation of bisphosphonate therapy and the duration of treatment are uncertain. Current research is focusing on bisphosphonates as adjuvant therapy. Radiotherapy is a very effective treatment for breast cancer patients with painful skeletal metastases. Patients have at least a 65 per cent chance of worthwhile pain relief, with complete response in about 20 per cent, as judged by self-assessment questionnaires. No clear evidence for a dose response has emerged for the short-term relief of bone pain. The long-term effectiveness of a single fraction of 8 Gy compared to multi-fraction schedules has been established using patient self-assessments for 12 months in 761 patients with uncomplicated metastatic bone pain, more than half with breast cancer.165** Pathological fracture of a long bone is a complication that should be prevented as often as possible by periodic review of femoral radiographs in patients with known skeletal disease. It is difficult to predict pathological fracture, but erosion of more than half the cortical thickness is an indication for an orthopaedic opinion concerning prophylactic surgical fixation or pinning. There is no randomized evidence relating to the contribution of post-surgical radiotherapy, or the optimal dose, but 20 Gy in five fractions is commonly prescribed.
Brain secondaries Patients present with symptoms and signs of raised intracranial pressure with or without focal neurological signs. The clinical diagnosis is confirmed by CT or MRI. There is a role for surgical resection followed by wholebrain radiotherapy for solitary lesions on MRI. Standard treatment for multiple symptomatic secondaries is palliative radiotherapy to the whole brain. There is no survival advantage demonstrated for high-dose palliation, and 20 Gy in five fractions is a commonly used schedule. A UK trial comparing 30 Gy in ten fractions with 12 Gy in two fractions delivered 1 week apart reported no difference in functional status or overall survival, although it is not clear
492 Breast cancer
to what extent this trial has influenced routine practice.166** Chemotherapy can offer effective palliation of intracerebral disease and should be considered for patients with symptomatic multi-system disease, especially those needing palliative chemotherapy for other symptoms.167*
Compression of spinal cord and cauda equina Diagnostic suspicion of cord compression constitutes an oncological emergency. Local and radicular pains, sensory changes, motor deficits and bladder/bowel dysfunction are all common presenting features. Cauda equina lesions present with lower motor neurone weakness affecting the bladder and bowel, associated lax anal sphincter tone and sacral anaesthesia. Magnetic resonance imaging of the whole spine is the investigation of choice, and needs to be performed on the same day if possible. Immediate surgical referral and decompression must be achieved within 24–48 hours of symptoms for the maximum chances of functional recovery. Dexamethasone 16 mg daily in divided doses should be introduced as soon as a clinical diagnosis is made to reduce inflammation and ease pain. In non-surgical cases, urgent radiotherapy is the mainstay of palliative treatment. It is common to give a tumour dose of 20 Gy in five fractions at 5 cm depth. Chemotherapy does not have a defined role in the management of this complication. The success of treatment and rehabilitation depends strongly on the neurological deficit at presentation. Patients who can still walk have good chances of full functional recovery, whereas paraplegic patients are unlikely to be helped to a useful degree.
than patients with other metastatic sites of relapse. Patients usually complain of dyspnoea, and a chest radiograph is usually sufficient to confirm a diagnosis. Cytology or pleural biopsy helps to distinguish cancer from non-malignant causes of effusion. Management in symptomatic patients includes thoracocentesis prior to commencing endocrine or chemotherapy as appropriate. If pleura is the sole or main site of relapse, thoracoscopy plus pleurodesis is the treatment of choice. Malignant pericardial effusions are far less common. Cardiac ultrasound is the most sensitive way to confirm a clinical diagnosis. If the patient is mildly symptomatic, endocrine therapy or chemotherapy can be considered as the first option; however, in the presence of more marked symptoms or tamponade, a pericardial window will need to be considered urgently.
Hypercalcaemia Hypercalcaemia is one of the commonest complications in patients with advanced breast cancer and is nearly always associated with skeletal metastases. Parathyroid-hormonelike factors appear to be less important than in some other cancers. Endocrine therapy is sometimes a cause of transient hypercalcaemia within a few weeks of instituting treatment. The symptoms of hypercalcaemia are highly variable and reflect alterations in neurological, gastrointestinal, renal, musculoskeletal and cardiac functions. The basis of management includes intravenous rehydration with correction of electrolyte imbalances, the use of bisphosphonates and a review of systemic therapy options.
Carcinomatous meningitis
PSYCHOSOCIAL SUPPORT
Metastases to the meninges are usually established via the arachnoid vessels and affect the meninges along the whole cranio-spinal axis. Presenting features include headache, altered mental state, cranial nerve palsies, lower motor neurone signs and positive Babinski’s sign. The diagnosis can be difficult and rests on noting a variety of symptoms and signs at more than one level in the central nervous system. Cerebrospinal fluid cytology may need to be repeated to confirm a clinical diagnosis. Unfortunately, intrathecal or intraventricular chemotherapy seldom offers little more than transient improvement in symptoms and signs, and local palliative radiotherapy may be a more appropriate way of relieving local symptoms and signs in many patients. The prognosis is very poor, usually a few weeks, although some long-term responders to intrathecal methotrexate have been reported.
Shock, fear and disbelief are common initial reactions met by the clinician who imparts the diagnosis.168 Speed in reaching the diagnosis is important. The ability of patients to cope with the uncertainty improves with counselling and good professional and social support.169** Trained nurse counsellors are of great value in this respect, and breast care nurses are one of the most valuable components of a multi-disciplinary breast clinic. Anxiety is common in the initial stages and is best managed by counselling in the first instance, with short-term anxiolytics reserved for overcoming severe anxiety or panic attacks. Women find it increasingly helpful to be informed and involved in the decision-making process. Many patients have clear ideas about the level of information that they find helpful. The approach needs to be modified according to the patient’s level of specialized knowledge and understanding.
Malignant effusions
HORMONE REPLACEMENT THERAPY
Malignant pleural effusions tend to develop ipsilaterally and affected patients have a more favourable prognosis
The use of HRT poses a dilemma for the rising numbers of breast cancer survivors, many of whom enter menopause
Future developments 493
prematurely as a result of therapy. Hormone replacement therapy is generally considered a contraindication because oestrogen is a growth factor for most breast cancer cells in the laboratory; however, empirical data on the safety of HRT after breast cancer are limited.170 The only reported randomized trial comparing HRT with no hormonal supplementation included 345 evaluable breast cancer patients with menopausal symptoms and was terminated early because of an increased incidence of recurrences and new primaries in the HRT group (26 in the HRT group versus 7 in the non-HRT group).171** This study, however, was not double-blinded, and it is possible that patients on HRT were monitored more closely. Education, counselling and non-hormonal drug therapy have been shown to reduce menopausal symptoms and to improve sexual functioning in breast cancer survivors. If symptoms interfere significantly with life quality, it is reasonable to recommend HRT for at least a few months in the first instance after complementary remedies have failed, especially if this helps compliance with adjuvant tamoxifen. Oestrogen/progestogen skin patches allow greater flexibility in adjusting the minimum dose of hormone replacement to suppress symptoms. Progestogen can be omitted in patients who have had a hysterectomy. Further data are needed, especially as there are proven benefits relating to protection from osteoporosis and heart disease.
a tool for better characterizing patients, and eventually tailoring treatment to individual needs. Despite recent advances in radiation technology, most centres worldwide use basic radiotherapy techniques based on two-dimensional breast data. Incorporating new approaches to breast radiotherapy, such as Intensity Modulated RadioTherapy (IMRT) and partial breast irradiation, may result in a reduction in morbidity. These more complex radiotherapy methods will require precise localization of the tumour bed and application of appropriate margins. Ongoing and proposed randomized trials will test these concepts, and will need to demonstrate the safety, efficacy and cost-effectiveness of these techniques. The future prospects for exploiting the biology of hypofractionation in breast cancer using advanced radiotherapy technologies also look bright. Recent research suggests that breast cancer may be more sensitive to the size of individual radiotherapy fractions than formerly assumed. Current trials are testing schedules delivering 16, 15 and 5 fractions. This may lead to very short treatment schedules delivered in 5 days rather than 5 weeks, something that would transform breast radiotherapy for patients and health services.
KEY LEARNING POINTS ●
FOLLOW-UP Several randomized trials have demonstrated that frequent hospital-based follow-up with detailed investigations has little, if any, clinical benefit.172**,173** This is because patients themselves are usually the first to notice symptoms or signs of recurrence. Apart from breast examination, no routine investigations are required except for annual mammography (after mastectomy this is biennial for the opposite breast) for the first 5 years and every other year thereafter, at least up to the age of 70 years.174 However, any suspicion of relapse should be investigated appropriately.
FUTURE DEVELOPMENTS The number of new drugs under clinical testing has never been higher, and the pace is likely to quicken in the years ahead. The identification and molecular characterization of signalling pathways responsible for abnormal growth, inhibition of apoptosis, angiogenesis, cellular invasion and metastasis have generated several targets for new therapeutic agents. In addition, micro-array technology holds the promise of further increasing our understanding of the complexity and heterogeneity of breast cancer, providing new information for the prognostication and prediction of breast cancer outcome. The application of gene-expression profiling for studying breast cancer will probably become
●
●
●
●
In the UK, approximately 40 000 women and a few hundred men develop breast cancer every year. The age-specific incidence rate rises fourfold between the ages of 35 and 70 years; the risk of developing breast cancer by the age of 85 years is 1 in 10. Breast cancer diagnosis relies on a triple assessment, namely clinical examination, mammography ( ultrasound) and aspiration cytology. Two-view (medio-lateral and cranio-caudal) mammography is currently offered every 3 years to all women in the UK between the ages of 50 and 70 years. Assuming 70 per cent uptake, these measures should save 1250 lives per year in the UK. Pathological node status is the single most powerful prognostic indicator, node-positive patients having, on average, a 10-year survival of 40 per cent compared with 80 per cent for nodenegative women after loco-regional therapy alone. The NCI Surveillance, Epidemiology, and End Results (SEER) Program has produced an online prognostic index based on patient age, tumour size, grade, ER status and lymph-node status (www.adjuvantonline.com). Sampling of the sentinel lymph node is now being widely adopted in clinically node-negative patients as a refinement of the axillary sampling approach.
494 Breast cancer
●
●
●
●
●
●
The 2005 EBCTCG systematic overview confirms that in women with ER-positive disease, 5 years of adjuvant tamoxifen reduces the annual breast cancer death rate by 31 per cent (SE 3), with as many deaths prevented between years 5 and 14 as during the first 5 years. There is no consensus view as to the optimum sequencing and duration of treatment of aromatase inhibitors; however, their use is becoming more widespread in the management of post-menopausal patients, with several trials showing a disease-free survival benefit in the adjuvant setting. The systematic overview confirms that 6 cycles of adjuvant anthracycline-containing chemotherapy reduces the annual odds of death from breast cancer by 38 per cent in women below the age of 50 years and by 20 per cent in women aged 50–69 years. The adjuvant use of taxanes is evolving. In 2005, adjuvant trials testing trastuzumab (Herceptin) in conjunction with chemotherapy in HER-2-positive cancer all reported in favour of trastuzumab. As a result, adjuvant trastuzumab in combination with cytotoxic therapy has been licensed for use in the UK. The 2005 overview of radiotherapy effects by the EBCTCG confirms a 67 per cent odds reduction in local recurrence and 11 per cent odds reduction in breast cancer mortality in patients randomized to post-surgical radiotherapy. On average, the prevention of four local recurrences prevents one breast cancer death from metastases. There is, as yet, no reliably identified group of patients whose breast recurrence risk is 10 per cent at 10 years after local excision alone. Intensive chemo-endocrine therapy as primary treatment in women with locally advanced breast cancer is becoming standard practice as the majority of patients gain at least a partial response, with approximately 10 per cent of patients achieving a complete pathological response In recent years, there has been a small but growing series of clinical trials demonstrating modest, but meaningful, survival advantages in metastatic disease using targeted biological agents in combination with traditional chemotherapy.
REFERENCES 1 Toms J. Cancer Stats Monograph. London: Cancer Research UK, 2004, 21–30.
2 Key TJ, Verkasalo PK, Banks E. Epidemiology of breast cancer. Lancet Oncol 2001; 2(3):133–40. 3 Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1998; 62(3):676–89. 4 Familial breast cancer: collaborative reanalysis of individual data from 52 epidemiological studies including 58,209 women with breast cancer and 101,986 women without the disease. Lancet 2001; 358(9291):1389–99. 5 Vasen HF, Haites NE, Evans DG, et al. Current policies for surveillance and management in women at risk of breast and ovarian cancer: a survey among 16 European family cancer clinics. European Familial Breast Cancer Collaborative Group. Eur J Cancer 1998; 34(12):1922–6. 6 Hopwood P. Genetic risk counselling for breast cancer families. Eur J Cancer 1998; 34(10):1477–9. 䉬7 Cuzick J, Powles T, Veronesi U, et al. Overview of the main outcomes in breast-cancer prevention trials. Lancet 2003; 361(9354):296–300. 8 Holmes MD, Chen WY, Feskanich D, Kroenke CH, Colditz GA. Physical activity and survival after breast cancer diagnosis. JAMA 2005; 293(20):2479–86. 9 Kerlikowske K, Grady D, Ernster V. Benefit of mammography screening in women ages 40–49 years: current evidence from randomized controlled trials. Cancer 1995; 76(9):1679–81. 10 (IARC) IAfRoC. On the Efficacy of Screening for Breast Cancer. Lyon: IARC Press, 2002. 11 Leach MO, Boggis CR, Dixon AK, et al. Screening with magnetic resonance imaging and mammography of a UK population at high familial risk of breast cancer: a prospective multicentre cohort study (MARIBS). Lancet 2005; 365(9473):1769–78. 12 Osin P, Shipley J, Lu YJ, Crook T, Gusterson BA. Experimental pathology and breast cancer genetics: new technologies. Recent Results Cancer Res 1998; 152:35–48. 13 Ingvarsson S. Molecular genetics of breast cancer progression. Semin Cancer Biol 1999; 9(4):277–88. 14 Meijers-Heijboer H, van den Ouweland A, Klijn J, et al. Lowpenetrance susceptibility to breast cancer due to CHEK2(*)1100delC in noncarriers of BRCA1 or BRCA2 mutations. Nat Genet 2002; 31(1):55–9. 15 Venkitaraman AR. Cancer susceptibility and the functions of BRCA1 and BRCA2. Cell 2002; 108(2):171–82. 16 Thompson D, Duedal S, Kirner J, et al. Cancer risks and mortality in heterozygous ATM mutation carriers. J Natl Cancer Inst 2005; 97(11):813–22. 17 Balmain A, Gray J, Ponder B. The genetics and genomics of cancer. Nat Genet 2003; 33(Suppl.):238–44. 18 Lakhani SR. In-situ lobular neoplasia: time for an awakening. Lancet 2003; 361(9352):96. 19 Simpson PT, Reis-Filho JS, Gale T, Lakhani SR. Molecular evolution of breast cancer. J Pathol 2005; 205(2): 248–54. 䊉20 Sorlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses
References 495
21
22
23
䉬24
25
䉬26
27
28
29
30
31
32 33
䊉34
35
36
䉬37
with clinical implications. Proc Natl Acad Sci U S A 2001; 98(19):10869–74. Azzopardi JG, Chepick OF, Hartmann WH. The World Health Organization Histological Typing of Breast Tumors. Am Soc Clin Pathol 1982; 78(6):806–17. Silverstein MJ. The University of Southern California/Van Nuys prognostic index for ductal carcinoma in situ of the breast. Am J Surg 2003; 186(4):337–43. Todd JH, Dowle C, Williams MR, et al. Confirmation of a prognostic index in primary breast cancer. Br J Cancer 1987; 56(4):489–92. Mauriac L, Debled M, MacGrogan G. When will more useful predictive factors be ready for use? Breast 2005; 14(6):617–23. de Koning HJ. Mammographic screening: evidence from randomised controlled trials. Ann Oncol 2003; 14(8):1185–9. EBCTCG. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005; 365(9472):1687–717. Margolese RG. Mastectomy or lumpectomy? The choice of operation for clinical stages I and II breast cancer. Can Med Assoc J 1998; 158:S15–S21. Sainsbury JR, Anderson TJ, Morgan DA, Dixon JM. ABC of breast diseases. Breast cancer. BMJ 1994; 309(6962):1150–3. Mariani L, Salvadori B, Marubini E, et al. Ten year results of a randomised trial comparing two conservative treatment strategies for small size breast cancer. Eur J Cancer 1998; 34(8):1156–62. Macmillan RD, Purushotham AD, George WD. Local recurrence after breast-conserving surgery for breast cancer. Br J Surg 1996; 83(2):149–55. Anderson BO, Masetti R, Silverstein MJ. Oncoplastic approaches to partial mastectomy: an overview of volumedisplacement techniques. Lancet Oncol 2005; 6(3):145–57. Ahmed S, Snelling A, Bains M, Whitworth IH. Breast reconstruction. BMJ 2005; 330(7497):943–8. Senkus-Konefka E, Welnicka-Jaskiewicz M, Jaskiewicz J, Jassem J. Radiotherapy for breast cancer in patients undergoing breast reconstruction or augmentation. Cancer Treat Rev 2004; 30(8):671–82. Forrest AP, Everington D, McDonald CC, Steele RJ, Chetty U, Stewart HJ. The Edinburgh randomized trial of axillary sampling or clearance after mastectomy. Br J Surg 1995; 82(11):1504–8. Lyman GH, Giuliano AE, Somerfield MR, et al. American Society of Clinical Oncology guideline recommendations for sentinel lymph node biopsy in early-stage breast cancer. J Clin Oncol 2005; 23(30):7703–20. Ivens D, Hoe AL, Podd TJ, Hamilton CR, Taylor I, Royle GT. Assessment of morbidity from complete axillary dissection. Br J Cancer 1992; 66(1):136–8. EBCTCG. Ovarian ablation in early breast cancer: overview of the randomised trials. Early Breast Cancer Trialists’ Collaborative Group. Lancet 1996; 348(9036):1189–96.
䉬38
39
40
41
䊉42
43
䊉44
䊉45
46
䊉47
48
ⴱ49
䊉50
EBCTCG. Tamoxifen for early breast cancer: an overview of the randomised trials. Early Breast Cancer Trialists’ Collaborative Group. Lancet 1998; 351(9114):1451–67. Geisler J, King N, Anker G, et al. In vivo inhibition of aromatization by exemestane, a novel irreversible aromatase inhibitor, in postmenopausal breast cancer patients. Clin Cancer Res 1998; 4(9):2089–93. Geisler J, King N, Dowsett M, et al. Influence of anastrozole (Arimidex), a selective, non-steroidal aromatase inhibitor, on in vivo aromatisation and plasma oestrogen levels in postmenopausal women with breast cancer. Br J Cancer 1996; 74(8):1286–91. Demers LM. Effects of Fadrozole (CGS 16949A) and Letrozole (CGS 20267) on the inhibition of aromatase activity in breast cancer patients. Breast Cancer Res Treat 1994; 30(1):95–102. Howell A, Cuzick J, Baum M, et al. Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years’ adjuvant treatment for breast cancer. Lancet 2005; 365(9453):60–2. Thurlimann B. New Clinical Data on Response to Adjuvant endocrine Therapy in Breast Cancer. In. 28th Annual San Antonio Breast Cancer Symposium: PeerView in Session; 2005. Coombes RC, Hall E, Gibson LJ, et al. A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. N Engl J Med 2004; 350(11):1081–92. Goss PE, Ingle JN, Martino S, et al. Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: updated findings from NCIC CTG MA.17. J Natl Cancer Inst 2005; 97(17):1262–71. Baum M, Buzdar A, Cuzick J, et al. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with earlystage breast cancer: results of the ATAC (Arimidex, Tamoxifen Alone or in Combination) trial efficacy and safety update analyses. Cancer 2003; 98(9):1802–10. Thurlimann B, Keshaviah A, Coates AS, Mouridsen H, Mauriac L, Forbes JF, et al. A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. N Engl J Med 2005; 353(26):2747–57. Coates AS, Keshaviah A, Thurlimann B, et al. 2007 Update of study BIG 1-98 J Clin Oncol 25(5):486–92.Coates AS, Keshaviah A, Thurlimann B, Mouridsen H, Winer EP, Hudis C, Burstein HJ, Wolff AC, Pritchard KI, Ingle JN, et al. American Society of Clinical Oncology technology assessment on the use of aromatase inhibitors as adjuvant therapy for postmenopausal women with hormone receptor-positive breast cancer: status report 2004. J Clin Oncol 2005; 23(3):619–29. Davidson NE, O’Neill A, Vukov A, Osborne CK, Martino S, White D, et al. Chemohormonal therapy in premenopausal node-positive, receptor-positive breast cancer: an eastern cooperative oncology group phase III intergroup trial (E5188, INT-0101). In. ASCO 2001; 2001.
496 Breast cancer
51 Dellapasqua S, Colleoni M, Gelber RD, Goldhirsch A. Adjuvant endocrine therapy for premenopausal women with early breast cancer. J Clin Oncol 2005; 23(8):1736–50. 52 Forward DP, Cheung KL, Jackson L, Robertson JF. Clinical and endocrine data for goserelin plus anastrozole as second-line endocrine therapy for premenopausal advanced breast cancer. Br J Cancer 2004; 90(3):590–4. 53 Huober J, Krainick-Strobel U, Kurek R, Wallwiener D. Neoadjuvant endocrine therapy in primary breast cancer. Clin Breast Cancer 2004; 5(5):341–7. 54 Eiermann W, Paepke S, Appfelstaedt J, Llombart-Cussac A, Eremin J, Vinholes J, et al. Preoperative treatment of postmenopausal breast cancer patients with letrozole: A randomized double-blind multicenter study. Ann Oncol 2001; 12(11):1527–32. 55 Johnston SRD, Dowsett M. Aromatase Inhibitors for breast cancer: Lessons from the laboratory [AU: OK?]. Nature Reviews/Cancer 2003; 3(October):821–32. 56 Jordan VC, Lewis JS, Osipo C, Cheng D. The apoptotic action of estrogen following exhaustive antihormonal therapy: a new clinical treatment strategy. Breast 2005; 14(6):624–30. 57 Dowsett M, Ashworth A. New biology of the oestrogen receptor. Lancet 2003; 362(9380):260–2. 58 Johnston SR, Dowsett M. Aromatase inhibitors for breast cancer: lessons from the laboratory. Nat Rev Cancer 2003; 3(11):821–31. 59 Ellis M. Overcoming endocrine therapy resistance by signal transduction inhibition. Oncologist 2004; 9 Suppl 3:20–6. 60 Shou J, Massarweh S, Osborne CK, Wakeling AE, Ali S, Weiss H, et al. Mechanisms of tamoxifen resistance: increased estrogen receptor-HER2/neu cross-talk in ER/HER2-positive breast cancer. J Natl Cancer Inst 2004; 96(12):926–35. 61 Mass R. The role of HER-2 expression in predicting response to therapy in breast cancer. Semin Oncol 2000; 27(6 Suppl 11):46–52; discussion 92–100. 62 Ellis MJ, Coop A, Singh B, Mauriac L, Llombert-Cussac A, Janicke F, et al. Letrozole is more effective neoadjuvant endocrine therapy than tamoxifen for ErbB-1- and/or ErbB2-positive, estrogen receptor-positive primary breast cancer: evidence from a phase III randomized trial. J Clin Oncol 2001; 19(18):3808–16. 63 Gutierrez MC, Detre S, Johnston S, Mohsin SK, Shou J, Allred DC, et al. Molecular changes in tamoxifen-resistant breast cancer: relationship between estrogen receptor, HER2, and p38 mitogen-activated protein kinase. J Clin Oncol 2005; 23(11):2469–76. 64 Nicholson RI, McClelland RA, Robertson JF, Gee JM. Involvement of steroid hormone and growth factor crosstalk in endocrine response in breast cancer. Endocr Relat Cancer 1999; 6(3):373–87. 65 Johnston SR. Clinical trials of intracellular signal transductions inhibitors for breast cancer–-a strategy to overcome endocrine resistance. Endocr Relat Cancer 2005; 12 Suppl 1:S145–57. 66 Fisher B, Costantino JP, Wickerham DL, Cecchini RS, Cronin WM, Robidoux A, et al. Tamoxifen for the prevention of
67
68
69
70
䊉71
䉬72
䊉73
䊉74
䊉75
76 䉬77
䊉78
breast cancer: current status of the National Surgical Adjuvant Breast and Bowel Project P-1 study. J Natl Cancer Inst 2005; 97(22):1652–62. Minton SE, Munster PN. Chemotherapy-induced amenorrhea and fertility in women undergoing adjuvant treatment for breast cancer. Cancer Control 2002; 9(6):466–72. Piccart MJ, de Valeriola D, Dal Lago L, de Azambuja E, Demonty G, Lebrun F, et al. Adjuvant chemotherapy in 2005: standards and beyond. Breast 2005; 14(6):439–45. Kroman N, Jensen MB, Wohlfahrt J, Mouridsen HT, Andersen PK, Melbye M. Factors influencing the effect of age on prognosis in breast cancer: population based study. Bmj 2000; 320(7233):474–8. Leonard RC, Malinovszky KM. Chemotherapy for older women with early breast cancer. Clin Oncol (R Coll Radiol) 2005; 17(4):244–8. Henderson IC, Berry DA, Demetri GD, Cirrincione CT, Goldstein LJ, Martino S, et al. Improved outcomes from adding sequential Paclitaxel but not from escalating Doxorubicin dose in an adjuvant chemotherapy regimen for patients with node-positive primary breast cancer. J Clin Oncol 2003; 21(6):976–83. Campone M, Fumoleau P, Bourbouloux E, Kerbrat P, Roche H. Taxanes in adjuvant breast cancer setting: which standard in Europe? Crit Rev Oncol Hematol 2005; 55(3):167–75. Bonneterre J, Roche H, Kerbrat P, Bremond A, Fumoleau P, Namer M, et al. Epirubicin increases long-term survival in adjuvant chemotherapy of patients with poor-prognosis, node-positive, early breast cancer: 10-year follow-up results of the French Adjuvant Study Group 05 randomized trial. J Clin Oncol 2005; 23(12):2686–93. Poole CJ, Earl HM, Dunn JA, Hiller L, Bathers S, Spooner D, et al. NEAT (National Epirubicin Adjuvant Trial) and SCTBG BR9601 (Scottish Cancer Trials Breast Group) phase III adjuvant breast trials show a significant relapse-free and overall survival advantage for sequential ECMF. In. ASCO 2003; 2003. Citron ML, Berry DA, Cirrincione C, Hudis C, Winer EP, Gradishar WJ, et al. Randomized trial of dose-dense versus conventionally scheduled and sequential versus concurrent combination chemotherapy as postoperative adjuvant treatment of node-positive primary breast cancer: first report of Intergroup Trial C9741/Cancer and Leukemia Group B Trial 9741. J Clin Oncol 2003; 21(8):1431–9. Citron ML. Dose density in adjuvant chemotherapy for breast cancer. Cancer Invest 2004; 22(4):555–68. Mauri D, Pavlidis N, Ioannidis JP. Neoadjuvant versus adjuvant systemic treatment in breast cancer: a metaanalysis. J Natl Cancer Inst 2005; 97(3):188–94. Bear HD, Anderson S, Brown A, Smith R, Mamounas EP, Fisher B, et al. The effect on tumor response of adding sequential preoperative docetaxel to preoperative doxorubicin and cyclophosphamide: preliminary results from National Surgical Adjuvant Breast and Bowel Project Protocol B-27. J Clin Oncol 2003; 21(22):4165–74.
References 497
79 Castiglione-Gertsch M, O’Neill A, Price KN, Goldhirsch A, Coates AS, Colleoni M, et al. Adjuvant chemotherapy followed by goserelin versus either modality alone for premenopausal lymph node-negative breast cancer: a randomized trial. J Natl Cancer Inst 2003; 95(24): 1833–46. 80 Yarnold J, Bliss JM, Earl H, George D, Lawrence D, Mortazavi SH, et al. Ovarian ablation in pre-menopausal women with early breast cancer prescribed 5 years tamoxifen or tamoxifen plus chemotherapy – results from the UK NCRI Adjuvant Breast Cancer (ABC) International Trial of 2,144 patients. In. ASCO; 2004. ⴱ81 Pritchard KI, Paterson AH, Paul NA, Zee B, Fine S, Pater J. Increased thromboembolic complications with concurrent tamoxifen and chemotherapy in a randomized trial of adjuvant therapy for women with breast cancer. National Cancer Institute of Canada Clinical Trials Group Breast Cancer Site Group. J Clin Oncol 1996; 14(10):2731–7. 82 Albain KS, Green S, Ravdin PM. Adjuvant chemohormonal therapy for primary breast cancer should be sequential instead of concurrent: results from intergroup trial 0100 (SWOG-8814) – A phase III trial. In; 2002. 83 Jordan VC. Molecular biology of the estrogen receptor aids in the understanding of tamoxifen resistance and breast cancer prevention with raloxifene. Recent Results Cancer Res 1998; 152:265–76. 84 Harvey JM, Clark GM, Osborne CK, Allred DC. Estrogen receptor status by immunohistochemistry is superior to the ligand-binding assay for predicting response to adjuvant endocrine therapy in breast cancer. J Clin Oncol 1999; 17(5):1474–81. 85 Osborne CK, Schiff R, Arpino G, Lee AS, Hilsenbeck VG. Endocrine responsiveness: understanding how progesterone receptor can be used to select endocrine therapy. Breast 2005; 14(6):458–65. 䊉86 Paik S, Shak S, Tang G, Kim C, Baker J, Cronin M, et al. A multigene assay to predict recurrence of tamoxifentreated, node-negative breast cancer. N Engl J Med 2004; 351(27):2817–26. 87 Yamauchi H, Stearns V, Hayes DF. When is a tumor marker ready for prime time? A case study of c-erbB-2 as a predictive factor in breast cancer. J Clin Oncol 2001; 19(8):2334–56. 88 Di Leo A, Isola J. Topoisomerase II alpha as a marker predicting the efficacy of anthracyclines in breast cancer: are we at the end of the beginning? Clin Breast Cancer 2003; 4(3):179–86. 䊉89 Romond EH, Perez EA, Bryant J, Suman VJ, Geyer CE, Jr., Davidson NE, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 2005; 353(16):1673–84. 䊉90 Piccart-Gebhart MJ, Procter M, Leyland-Jones B, Goldhirsch A, Untch M, Smith I, et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 2005; 353(16):1659–72. 91 Slamon DJ, Eiermann W, Robert N, Pienkowski T, Martin M, Pawlicki M, et al. Phase III randomized trial comparing
92
䊉93
䊉94
䊉95
䉬96
ⴱ97
98
99
100
101
102
103
doxorubicin and cyclophosphamide followed by docetaxel with doxorubicin and cyclophosphamide followed by docetaxel and trastuzumab with docetaxel, carboplatin and trastuzumab in HER2 positive early breast cancer patients: BCIRG 006 study. In. San Antonio Breast Cancer Symposium; 2005. Joensuu H, Kellokumpu-Lehtinen PL, Bono P, Alanko T, Kataja V, Asola R, et al. Adjuvant docetaxel or vinorelbine with or without trastuzumab for breast cancer. N Engl J Med 2006; 354(8):809–20. Powles T, Paterson S, Kanis JA, McCloskey E, Ashley S, Tidy A, et al. Randomized, placebo-controlled trial of clodronate in patients with primary operable breast cancer. J Clin Oncol 2002; 20(15):3219–24. Diel IJ, Solomayer EF, Costa SD, Gollan C, Goerner R, Wallwiener D, et al. Reduction in new metastases in breast cancer with adjuvant clodronate treatment. N Engl J Med 1998; 339(6):357–63. Saarto T, Blomqvist C, Virkkunen P, Elomaa I. Adjuvant clodronate treatment does not reduce the frequency of skeletal metastases in node-positive breast cancer patients: 5-year results of a randomized controlled trial. J Clin Oncol 2001; 19(1):10–7. Clarke M, Collins R, Darby S, Davies C, Elphinstone P, Evans E, et al. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005; 366(9503):2087–106. Bartelink H, Horiot JC, Poortmans P, Struikmans H, Van den Bogaert W, Barillot I, et al. Recurrence rates after treatment of breast cancer with standard radiotherapy with or without additional radiation. N Engl J Med 2001; 345(19):1378–87. Schnitt SJ, Connolly JL, Khettry U, Mazoujian G, Brenner M, Silver B, et al. Pathologic findings on re-excision of the primary site in breast cancer patients considered for treatment by primary radiation therapy. Cancer 1987; 59(4):675–81. McKinna F, Gothard L, Ashley S, Ebbs SR, Yarnold JR. Lymphatic relapse in women with early breast cancer: a difficult management problem. Eur J Cancer 1999; 35(7):1065–9. Mignano JE, Zahurak ML, Chakravarthy A, Piantadosi S, Dooley WC, Gage I. Significance of axillary lymph node extranodal soft tissue extension and indications for postmastectomy irradiation. In: Cancer; 1999. p. 1258–62. Hetelekidis S, Schnitt SJ, Silver B, Manola J, Bornstein BA, Nixon AJ, et al. The significance of extracapsular extension of axillary lymph node metastases in early-stage breast cancer. Int J Radiat Oncol Biol Phys 2000; 46(1):31–4. McKinna F, Gothard L, Ashley S, Ebbs S, Yarnold J. Selective avoidance of lymphatic radiotherapy in the conservative management of women with early breast cancer. Radiother Oncol 1999; 52(3):219–23. EBCTCG. Favourable and unfavourable effects on long-term survival of radiotherapy for early breast cancer: an overview
498 Breast cancer
104 105
䊉106
䊉107
䊉108
109
110
111
112
113
114
115
116
117
of the randomised trials. Early Breast Cancer Trialists Collaborative Group. Lancet 2000; 355(9217):1757–70. Handley RS. Carcinoma of the breast. Ann R Coll Surg Engl 1975; 57(2):59–66. Wallgren A, Bonetti M, Gelber RD, Goldhirsch A, Castiglione-Gertsch M, Holmberg SB, et al. Risk factors for locoregional recurrence among breast cancer patients: results from International Breast Cancer Study Group Trials I through VII. J Clin Oncol 2003; 21(7):1205–13. Fisher B, Dignam J, Wolmark N, Mamounas E, Costantino J, Poller W, et al. Lumpectomy and radiation therapy for the treatment of intraductal breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-17. J Clin Oncol 1998; 16(2):441–52. Julien JP, Bijker N, Fentiman IS, Peterse JL, Delledonne V, Rouanet P, et al. Radiotherapy in breast-conserving treatment for ductal carcinoma in situ: first results of the EORTC randomised phase III trial 10853. EORTC Breast Cancer Cooperative Group and EORTC Radiotherapy Group. Lancet 2000; 355(9203):528–33. UKCCCR, Party DW. Radiotherapy and tamoxifen in women with completely excised ductal carcinoma in situ of the breast in the UK, Australia, and New Zealand: randomised controlled trial. Lancet 2003; 362(9378):95–102. Silverstein M. Biological variables and prognosis in patients with ductal carcinoma in situ of the breast. The Breast 2001; 10(Supplement 3):39–53. Silverstein MJ, Lagios MD, Craig PH, Waisman JR, Lewinsky BS, Colburn WJ, et al. A prognostic index for ductal carcinoma in situ of the breast. Cancer 1996; 77(11):2267–74. Fisher ER, Dignam J, Tan-Chiu E, Costantino J, Fisher B, Paik S, et al. Pathologic findings from the National Surgical Adjuvant Breast Project (NSABP) eight-year update of Protocol B-17: intraductal carcinoma. Cancer 1999; 86(3):429–38. Silliman RA, Balducci L, Goodwin JS, Holmes FF, Leventhal EA. Breast cancer care in old age: what we know, don’t know, and do. J Natl Cancer Inst 1993; 85(3):190–9. Wyld L, Garg DK, Kumar ID, Brown H, Reed MW. Stage and treatment variation with age in postmenopausal women with breast cancer: compliance with guidelines. Br J Cancer 2004; 90(8):1486–91. Monypenny I. UK Symptomatic breast audit 1.4.200131.3.2002. In. British Association of Surgical Oncology, London: British Association of Surgical Oncology; 2003. Hind D. Surgery versus primary endocrine therapy for operable primary breast cancer in elderly women (70 years plus): Cochrane Database Syst Rev. 2006 Jan 25;(1): CD004272; 2006 Jan 25 (1). Report No.: CD004272. Lee JC, Truong PT, Kader HA, Speers CH, Olivotto IA. Postmastectomy radiotherapy reduces locoregional recurrence in elderly women with high-risk breast cancer. Clin Oncol (R Coll Radiol) 2005; 17(8):623–9. ICRU. Prescribing, Recording and Reporting Photon Beam Therapy. Bethsada: International Commission on Radiation Units and Measurements (ICRU); 1993.
118 Bellon JR, Come SE, Gelman RS, Henderson IC, Shulman LN, Silver BJ, et al. Sequencing of chemotherapy and radiation therapy in early-stage breast cancer: updated results of a prospective randomized trial. J Clin Oncol 2005; 23(9):1934–40. 119 Toledano A, Garaud P, Serin D, Fourquet A, Bosset JF, Breteau N, et al. Concurrent administration of adjuvant chemotherapy and radiotherapy after breast-conservative surgery enhances late toxicities: Long-term results of the arcosein multicenter randomized study. Int J Radiat Oncol Biol Phys 2006. 120 Gagliardi G, Lax I, Ottolenghi A, Rutqvist LE. Long-term cardiac mortality after radiotherapy of breast cancer–application of the relative seriality model. Br J Radiol 1996; 69(825):839–46. 121 White TT. Carcinoma of the breast and pregnancy; analysis of 920 cases collected from the literature and 22 new cases. Ann Surg 1954; 139(1):9–18. 䉬122 Doll DC, Ringenberg QS, Yarbro JW. Antineoplastic agents and pregnancy. Semin Oncol 1989; 16(5):337–46. 123 Clark RM, Chua T. Breast cancer and pregnancy: the ultimate challenge. Clin Oncol (R Coll Radiol) 1989; 1(1):11–8. 124 Zemlickis D, Lishner M, Degendorfer P, Panzarella T, Burke B, Sutcliffe SB, et al. Maternal and fetal outcome after breast cancer in pregnancy. Am J Obstet Gynecol 1992; 166(3):781–7. 125 Aziz S, Pervez S, Khan S, Siddiqui T, Kayani N, Israr M, et al. Case control study of novel prognostic markers and disease outcome in pregnancy/lactation-associated breast carcinoma. Pathol Res Pract 2003; 199(1):15–21. 126 Osei EK, Faulkner K. Fetal doses from radiological examinations. Br J Radiol 1999; 72(860):773–80. 127 UK MDA. Pregnancy and MR exposure. In: Guidelines for Magnetic Resonance Equipment in Clinical Use; 2002. p. 22–23. 128 Ring AE, Smith IE, Jones A, Shannon C, Galani E, Ellis PA. Chemotherapy for breast cancer during pregnancy: an 18year experience from five London teaching hospitals. J Clin Oncol 2005; 23(18):4192–7. 129 Greskovich JF, Jr., Macklis RM. Radiation therapy in pregnancy: risk calculation and risk minimization. Semin Oncol 2000; 27(6):633–45. 130 Antypas C, Sandilos P, Kouvaris J, Balafouta E, Karinou E, Kollaros N, et al. Fetal dose evaluation during breast cancer radiotherapy. Int J Radiat Oncol Biol Phys 1998; 40(4):995–9. 䊉131 Berry DL, Theriault RL, Holmes FA, Parisi VM, Booser DJ, Singletary SE, et al. Management of breast cancer during pregnancy using a standardized protocol. J Clin Oncol 1999; 17(3):855–61. 132 Giacalone PL, Laffargue F, Benos P. Chemotherapy for breast carcinoma during pregnancy: A French national survey. Cancer 1999; 86(11):2266–72. 133 Furr BJ, Valcaccia B, Challis JR. The effects of Nolvadex (tamoxifen citrate; ICI 46,474) on pregnancy in rabbits. J Reprod Fertil 1976; 48(2):367–9.
References 499
134 Illidge TM, Hussey M, Godden CW. Malignant hypercalcaemia in pregnancy and antenatal administration of intravenous pamidronate. Clin Oncol (R Coll Radiol) 1996; 8(4):257–8. 135 Bulens P, Vanuytsel L, Rijnders A, van der Schueren E. Breast conserving treatment of Paget’s disease. Radiother Oncol 1990; 17(4):305–9. 136 Thorlacius S, Tryggvadottir L, Olafsdottir GH, Jonasson JG, Ogmundsdottir HM, Tulinius H, et al. Linkage to BRCA2 region in hereditary male breast cancer. Lancet 1995; 346(8974):544–5. 137 Giordano SH, Buzdar AU, Hortobagyi GN. Breast cancer in men. Ann Intern Med 2002; 137(8):678–87. 138 Anelli TF, Anelli A, Tran KN, Lebwohl DE, Borgen PI. Tamoxifen administration is associated with a high rate of treatment-limiting symptoms in male breast cancer patients. Cancer 1994; 74(1):74–7. 139 Levine PH, Steinhorn SC, Ries LG, Aron JL. Inflammatory breast cancer: the experience of the surveillance, epidemiology, and end results (SEER) program. J Natl Cancer Inst 1985; 74(2):291–7. 140 Singletary SE, Allred C, Ashley P, Bassett LW, Berry D, Bland KI, et al. Revision of the American Joint Committee on Cancer staging system for breast cancer. J Clin Oncol 2002; 20(17):3628–36. 141 Jaiyesimi IA, Buzdar AU, Hortobagyi G. Inflammatory breast cancer: a review. J Clin Oncol 1992; 10(6):1014–24. 142 Aziz SA, Pervez S, Khan S, Kayani N, Azam SI, Rahbar MH. Case control study of prognostic markers and disease outcome in inflammatory carcinoma breast: a unique clinical experience. Breast J 2001; 7(6):398–404. 143 Kleer CG, van Golen KL, Merajver SD. Molecular biology of breast cancer metastasis. Inflammatory breast cancer: clinical syndrome and molecular determinants. Breast Cancer Res 2000; 2(6):423–9. 144 van Golen KL, Davies S, Wu ZF, Wang Y, Bucana CD, Root H, et al. A novel putative low-affinity insulin-like growth factorbinding protein, LIBC (lost in inflammatory breast cancer), and RhoC GTPase correlate with the inflammatory breast cancer phenotype. Clin Cancer Res 1999; 5(9):2511–9. 145 Ueno NT, Buzdar AU, Singletary SE, Ames FC, McNeese MD, Holmes FA, et al. Combined-modality treatment of inflammatory breast carcinoma: twenty years of experience at M. D. Anderson Cancer Center. Cancer Chemother Pharmacol 1997; 40(4):321–9. 146 De Boer RH, Allum WH, Ebbs SR, Gui GP, Johnston SR, Sacks NP, et al. Multimodality therapy in inflammatory breast cancer: is there a place for surgery? Ann Oncol 2000; 11(9):1147–53. 147 Fleming RY, Asmar L, Buzdar AU, McNeese MD, Ames FC, Ross MI, et al. Effectiveness of mastectomy by response to induction chemotherapy for control in inflammatory breast carcinoma. Ann Surg Oncol 1997; 4(6):452–61. 148 Panades M, Olivotto IA, Speers CH, Shenkier T, Olivotto TA, Weir L, et al. Evolving treatment strategies for inflammatory breast cancer: a population-based survival analysis. J Clin Oncol 2005; 23(9):1941–50.
149 van Golen KL, Bao L, DiVito MM, Wu Z, Prendergast GC, Merajver SD. Reversion of RhoC GTPase-induced inflammatory breast cancer phenotype by treatment with a farnesyl transferase inhibitor. Mol Cancer Ther 2002; 1(8):575–83. 150 Dorr FA, Bader J, Friedman MA. Locally advanced breast cancer current status and future directions. Int J Radiat Oncol Biol Phys 1989; 16(3):775–84. 151 Kuerer HM, Newman LA, Smith TL, Ames FC, Hunt KK, Dhingra K, et al. Clinical course of breast cancer patients with complete pathologic primary tumor and axillary lymph node response to doxorubicin-based neoadjuvant chemotherapy. J Clin Oncol 1999; 17(2):460–9. 䉬152 Kaufmann M, von Minckwitz G, Smith R, Valero V, Gianni L, Eiermann W, et al. International expert panel on the use of primary (preoperative) systemic treatment of operable breast cancer: review and recommendations. J Clin Oncol 2003; 21(13):2600–8. 153 Huang EH, Tucker SL, Strom EA, McNeese MD, Kuerer HM, Buzdar AU, et al. Postmastectomy radiation improves local-regional control and survival for selected patients with locally advanced breast cancer treated with neoadjuvant chemotherapy and mastectomy. J Clin Oncol 2004; 22(23):4691–9. 154 Schwaibold F, Fowble BL, Solin LJ, Schultz DJ, Goodman RL. The results of radiation therapy for isolated local regional recurrence after mastectomy. Int J Radiat Oncol Biol Phys 1991; 21(2):299–310. 155 Kurtz JM, Spitalier JM, Amalric R, Brandone H, Ayme Y, Jacquemier J, et al. The prognostic significance of late local recurrence after breast-conserving therapy. Int J Radiat Oncol Biol Phys 1990; 18(1):87–93. 156 van Tienhoven G, Voogd AC, Peterse JL, Nielsen M, Andersen KW, Mignolet F, et al. Prognosis after treatment for loco-regional recurrence after mastectomy or breast conserving therapy in two randomised trials (EORTC 10801 and DBCG-82TM). EORTC Breast Cancer Cooperative Group and the Danish Breast Cancer Cooperative Group. Eur J Cancer 1999; 35(1):32–8. 157 NCIDCCPS. Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov). SEER Stat Database: Incidence – SEER 9 Regs Public-Use, Nov 2004 Sub (1973-2002), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2005. In; 2005. 158 Gralow JR. Optimizing the treatment of metastatic breast cancer. Breast Cancer Res Treat 2005; 89 Suppl 1:S9–S15. 159 Miles D, von Minckwitz G, Seidman AD. Combination versus sequential single-agent therapy in metastatic breast cancer. Oncologist 2002; 7 Suppl 6:13–9. 䊉160 Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344(11):783–92. 161 Marty M, Cognetti F, Maraninchi D, Snyder R, Mauriac L, Tubiana-Hulin M, et al. Randomized phase II trial of the
500 Breast cancer
䊉162
163
䉬164
165
166
167
168
169
170
171
172
173
174
efficacy and safety of trastuzumab combined with docetaxel in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer administered as first-line treatment: the M77001 study group. J Clin Oncol 2005; 23(19):4265–74. Miller KD, Wang M, Gralow J, Dickler M, Cobleigh MA, Perez EA, et al. A randomized phase III trial of paclitaxel versus paclitaxel plus bevacizumab as first-line therapy for locally recurrent or metastatic breast cancer: a trial coordinated by the Eastern Cooperative Oncology Group (E2100). In. San Antonio Breast Cancer Symposium; 2005. Hillner BE, Ingle JN, Chlebowski RT, Gralow J, Yee GC, Janjan NA, et al. American Society of Clinical Oncology 2003 update on the role of bisphosphonates and bone health issues in women with breast cancer. J Clin Oncol 2003; 21(21):4042–57. Pavlakis N. Bisphosphonates for breast cancer: Cochrane Databse Syst Rev, 2005, Jul 20;(3): CD003474; 2005 Jul 20 (3). Report No.: CD003474. 8 Gy single fraction radiotherapy for the treatment of metastatic skeletal pain: randomised comparison with a multifraction schedule over 12 months of patient followup. Bone Pain Trial Working Party. Radiother Oncol 1999; 52(2):111–21. Priestman TJ, Dunn J, Brada M, Rampling R, Baker PG. Final results of the Royal College of Radiologists’ trial comparing two different radiotherapy schedules in the treatment of cerebral metastases. Clin Oncol (R Coll Radiol) 1996; 8(5):308–15. Boogerd W, Dalesio O, Bais EM, van der Sande JJ. Response of brain metastases from breast cancer to systemic chemotherapy. Cancer 1992; 69(4):972–80. Fallowfield LJ. Psychosocial adjustment after treatment for early breast cancer. Oncology (Williston Park) 1990; 4(4):89–97; discussion 97–8, 100. McArdle JM, George WD, McArdle CS, Smith DC, Moodie AR, Hughson AV, et al. Psychological support for patients undergoing breast cancer surgery: a randomised study. Bmj 1996; 312(7034):813–6. Rostom AY. The management of menopausal sequelae in patients with breast cancer. Clin Oncol (R Coll Radiol) 2001; 13(3):174–80. Holmberg L, Anderson H. HABITS (hormonal replacement therapy after breast cancer–-is it safe?), a randomised comparison: trial stopped. Lancet 2004; 363(9407):453–5. Grunfeld E, Levine MN, Julian JA, Coyle D, Szechtman B, Mirsky D, et al. Randomized trial of long-term follow-up for early-stage breast cancer: a comparison of family physician versus specialist care. J Clin Oncol 2006; 24(6):848–55. Kokko R, Hakama M, Holli K. Follow-up cost of breast cancer patients with localized disease after primary treatment: a randomized trial. Breast Cancer Res Treat 2005; 93(3):255–60. Maher M, Campana F, Mosseri V, Dreyfus H, Vilcoq JR, Gautier C, et al. Breast cancer in elderly women: a
175
176
177
178
179
180
181
182
183
184
185
retrospective analysis of combined treatment with tamoxifen and once-weekly irradiation. Int J Radiat Oncol Biol Phys 1995; 31(4):783–9. Fisher B, Anderson S, Bryant J, Margolese RG, Deutsch M, Fisher ER, et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med 2002; 347(16):1233–41. Fyles AW, McCready DR, Manchul LA, Trudeau ME, Merante P, Pintilie M, et al. Tamoxifen with or without breast irradiation in women 50 years of age or older with early breast cancer. N Engl J Med 2004; 351(10): 963–70. Hughes KS, Schnaper LA, Berry D, Cirrincione C, McCormick B, Shank B, et al. Lumpectomy plus tamoxifen with or without irradiation in women 70 years of age or older with early breast cancer. N Engl J Med 2004; 351(10):971–7. Winzer KJ, Sauer R, Sauerbrei W, Schneller E, Jaeger W, Braun M, et al. Radiation therapy after breast-conserving surgery; first results of a randomised clinical trial in patients with low risk of recurrence. Eur J Cancer 2004; 40(7):998–1005. Gnant MFX, Poetter R, Kwasny W, Tausch C, Handl-Zeller E, Pakesch B, et al. Breast conservation without radiotherapy in low risk breast cancer patients – results of 2 prospective clinical trials of the Austrian Breast and Colorectal Cancer Study Group involving 1,518 postmenopausal patients with endocrine responsive breast cancer. In. San Antonio Breast Cancer Symposium; 2005. van der Hage JA, van de Velde CJ, Julien JP, Tubiana-Hulin M, Vandervelden C, Duchateau L. Preoperative chemotherapy in primary operable breast cancer: results from the European Organization for Research and Treatment of Cancer trial 10902. J Clin Oncol 2001; 19(22):4224–37. Avril A, Faucher A, Bussieres E, Stockle E, Durand M, Mauriac L, et al. [Results of 10 years of a randomized trial of neoadjuvant chemotherapy in breast cancers larger than 3 cm.]. Chirurgie 1998; 123(3):247–56. Mauriac L, MacGrogan G, Avril A, Durand M, Floquet A, Debled M, et al. Neoadjuvant chemotherapy for operable breast carcinoma larger than 3 cm: a unicentre randomized trial with a 124-month median follow-up. Institut Bergonie Bordeaux Groupe Sein (IBBGS). Ann Oncol 1999; 10(1):47–52. Semiglazov VF, Topuzov EE, Bavli JL, Moiseyenko VM, Ivanova OA, Seleznev IK, et al. Primary (neoadjuvant) chemotherapy and radiotherapy compared with primary radiotherapy alone in stage IIb–IIIa breast cancer. Ann Oncol 1994; 5(7):591–5. Scholl SM, Asselain B, Palangie T, Dorval T, Jouve M, Garcia Giralt E, et al. Neoadjuvant chemotherapy in operable breast cancer. Eur J Cancer 1991; 27(12):1668–71. Scholl SM, Fourquet A, Asselain B, Pierga JY, Vilcoq JR, Durand JC, et al. Neoadjuvant versus adjuvant chemotherapy in premenopausal patients with tumours considered too large for breast conserving surgery:
References 501
preliminary results of a randomised trial: S6. Eur J Cancer 1994; 30A(5):645–52. 186 Broet P, Scholl SM, de la Rochefordiere A, Fourquet A, Moreau T, De Rycke Y, et al. Short and long-term effects on survival in breast cancer patients treated by primary chemotherapy: an updated analysis of a randomized trial. Breast Cancer Res Treat 1999; 58(2):151–6. 187 Makris A, Powles TJ, Ashley SE, Chang J, Hickish T, Tidy VA, et al. A reduction in the requirements for mastectomy in a randomized trial of neoadjuvant chemoendocrine therapy in primary breast cancer. Ann Oncol 1998; 9(11):1179–84. 188 Fisher B, Bryant J, Wolmark N, Mamounas E, Brown A, Fisher ER, et al. Effect of preoperative chemotherapy on the outcome of women with operable breast cancer. J Clin Oncol 1998; 16(8):2672–85.
189 Wolmark N, Wang J, Mamounas E, Bryant J, Fisher B. Preoperative chemotherapy in patients with operable breast cancer: nine-year results from National Surgical Adjuvant Breast and Bowel Project B-18. J Natl Cancer Inst Monogr 2001; (30):96–102. 190 Gazet JC, Ford HT, Gray R, McConkey C, Sutcliffe R, Quilliam J, et al. Estrogen-receptor-directed neoadjuvant therapy for breast cancer: results of a randomised trial using formestane and methotrexate, mitozantrone and mitomycin C (MMM) chemotherapy. Ann Oncol 2001; 12(5):685–91. 191 Danforth DN, Jr., Cowan K, Altemus R, Merino M, Chow C, Berman A, et al. Preoperative FLAC/granulocyte-colonystimulating factor chemotherapy for stage II breast cancer: a prospective randomized trial. Ann Surg Oncol 2003; 10(6):635–44.
20 Bronchus VANESSA A. POTTER, NICHOLAS THATCHER AND PENELLA J. WOLL
Incidence and aetiology Prevention and screening Clinical features Diagnosis and staging investigations
502 504 505 506
INCIDENCE AND AETIOLOGY Lung cancer is the leading cause of death in the developed world, with the majority of patients (80 per cent) dying within a year of diagnosis. Lung cancer accounts for 28 per cent of male cancer deaths in the UK and 31 per cent in the USA. For women, lung cancer accounts for 20 per cent of cancer deaths in the UK and 25 per cent in the USA. In 2004, 19 490 men and 13 550 women died from lung cancer in the UK.1 Mortality figures within England and Wales show significant regional differences in incidence, with the highest incidence in the North of England. The mortality in this region is the highest in Europe; in Tyneside, the age-standardized mortality for men is 101.3 per 100 000,
Pathology Treatment of non-small-cell lung cancer Treatment of small-cell lung cancer References
509 511 521 529
compared to only 18.9 in Potenza, Italy, and 23.8 for women in Merseyside, compared to 2.4 in Potenza. In the Far East, non-smokers have a far higher incidence of lung cancer than in the West due to a high level of EGFR mutations. The incidence of lung cancer has increased dramatically in the past century, although the number of deaths amongst men is now stable in the UK and has begun to fall in the USA (Fig. 20.1). However, amongst women, the incidence of lung cancer continues to rise and it has overtaken breast cancer as the commonest cause of cancer death. The incidence of lung cancer increases with age. Less than 1 per cent of cases occur under the age of 40 years, and 56 per cent of cases occur in the over-70 age group.
80
Rate per 100 000 population
Women
Men
60 Lung 40 Breast 20
Colorectal Lung
0 1930
1950
1970 Year
1990
1930
1950
1970 Year
1990
Figure 20.1 Incidence of lung cancer in men and women during the 20th century, compared with the incidence of breast cancer in women and colorectal cancer in men. Modified from.2
Incidence and aetiology 503
Cigarette smoking is the main risk factor for lung cancer, accounting for up to 94 per cent of lung cancer deaths in men and 83 per cent in women. The relationship between cigarette smoking and lung cancer was established by the classic epidemiological studies of Doll and Hill,3* but has been confirmed by numerous prospective and retrospective studies worldwide. The rapid rise in incidence of lung cancer in women over the last few decades is due to the fact that smoking was an uncommon habit until after the Second World War, whereas it had been socially acceptable for men several decades earlier. Although the overall consumption of cigarettes has decreased since the 1970s, it is particularly worrying that smoking is most common in the 20–24-age group and that the proportion of children who smoke has actually increased, with 26 per cent of 15-yearold girls in the UK being regular smokers.4 The risk of lung cancer is directly related to both the duration of exposure and the number of cigarettes smoked and is 15-fold greater in smokers of 25 cigarettes per day than in non-smokers. The risk of developing lung cancer falls rapidly after stopping smoking, but even 15–20 years later the risk is still 1.5 to 4 times that of life-long nonsmokers.5* The risk is increased with the tar and nicotine content of the tobacco smoked, inhalation of smoke, taking more puffs from each cigarette, keeping the cigarette in the mouth between puffs and relighting half-smoked cigarettes. The lower risks of pipe and cigar smoking may be due to the social rituals associated with their use, including less inhalation. The age of onset of smoking is an important factor in defining the risk from cigarette smoking. Therefore the increased prevalence of smoking in children is alarming, as the resulting cancers will not be seen for several decades. The risk of lung cancer at the age of 60 is three times greater for those who started to smoke at age 15 than for those who started 10 years later. A recent study6 has demonstrated significantly increased levels of tobacco-induced DNA damage in patients who began smoking before the age of 25 compared to those who began in later life. In recent years there has been a slight reduction in male lung cancer deaths in Britain, possibly attributable to reductions in the tar and nicotine contents of cigarettes, as well as reduced consumption. However, since 1950 there has been a 500 per cent increase in lung cancer mortality in women. It has been proposed that women are more susceptible to tobacco carcinogens. Zang and Wynder7 examined the risk of women developing lung cancer compared to men according to exposure (cumulative tar, pack-years and current number of cigarettes smoked) and found that, for the same life-long exposure to cigarette smoke, women were at a 1.5-fold higher risk of developing lung cancer than men. These clinical findings may be due to physiological differences that result in the slowed detoxification of carcinogens. Many non-smokers are exposed to tobacco smoke at home, during work and in public places. The health risks of such indirect (passive) smoking are difficult to quantify, but the incidences of lung cancer and ischaemic heart disease are
increased. In the UK, approximately 1 in 6 of the population are exposed passively to tobacco smoke at home. It is estimated that a non-smoker in a smoking household is exposed to the equivalent of 1 per cent of the cigarettes actively smoked. In this non-smoking population, the risk of developing lung cancer is increased by 24 per cent,8 the risk increasing with the number of cigarettes smoked and the length of co-habitation. Increasing restrictions on smoking in public places, such as offices, bars and restaurants, is expected to reduce passive exposure to tobacco smoke considerably. Additional factors associated with the development of lung cancer include atmospheric pollution, ionizing radiation and various occupational hazards. Atmospheric pollution appears to increase the risk of lung cancer mainly in smokers. In some large industrial cities, there is considerable variation in lung cancer rates, which can be linked to material deprivation. Radiation exposure has been associated with increased rates of lung cancer in atomic bomb survivors and patients treated with radiotherapy for ankylosing spondylitis, breast cancer and Hodgkin’s disease. The effects of natural radiation from radon are seen in the increased lung cancer risk of miners in these areas. There is concern that radon may accumulate in houses constructed with energy-saving techniques in areas of hard-rock geology, causing a hazard to the inhabitants. Despite this, lung cancer incidence in Cornwall is well below the national average. Industrial carcinogens such as asbestos, chromium, nickel, vinyl chloride and arsenic have been well documented. Workers engaged in the coal distillation and gas industry also have an increased risk of lung cancer. Interestingly, chloromethylether (unlike many other carcinogens) is particularly associated with a specific histological subtype, small-cell lung cancer (SCLC). In many of these cases, the effects of the carcinogens are seen mainly in smokers. A number of broncho-pulmonary diseases may also predispose to the development of lung cancer, including tuberculosis, pulmonary scars from previous infection and trauma. Fibrosing alveolitis is said to be usually associated with the development of non-small-cell lung cancers (NSCLCs). There is increasing evidence that immunosuppression is associated with an increased risk of lung cancer, whether resulting from human immunodeficiency virus (HIV) infection or immunosuppressive drugs following renal, heart or lung transplantation. Although most patients with lung cancer are smokers, only a minority of smokers develop lung cancer. This suggests that certain individuals are more susceptible to the effects of smoking than others, and has led to a search for heritable predisposing factors. In lung cancer patients aged less than 50 years, an autosomal recessive cancer-prone gene could contribute to the development of as many as 70 per cent of tumours. Genetic variations in metabolism, such as debrisoquine oxidation phenotype, glutathione transferase and arylhydroxylase activity, may contribute to lung cancer susceptibility. These metabolic pathways may detoxify
504 Bronchus
Table 20.1 Genetic abnormalities detected in SCLC and NSCLC. From9 and10 Site of mutation Chromosome 3 3p14, 3p21, 3p25 Chromosome 9 9p21 Chromosome 12 12p Chromosome 13 13q Chromosome 17 17p13 Chromosome 17 17q
Name and type
Frequency
Stage of carcinogenesis
Fragile histidine triad (FHIT) Tumour suppressor p16 Oncogene K-ras Oncogene Retinoblastoma gene Tumour suppressor p53 Tumour suppressor c-erb B2 Oncogene
SCLC ⬃90% NSCLC ⬃50%
Early
SCLC 80% NSCLC 50% Adenocarcinoma ⬃30% SCLC ⬃90% NSCLC ⬃30% SCLC ⬃80% NSCLC⬃50% NSCLC ⬃25%
Early
3p14 3p21
17p13
3p25 9p21
Normal
Metaplasia
Dysplasia
inhaled or ingested carcinogens. Dietary factors such as vitamin A deficiency and exposure to nitrosoamines may also be important.
Genetic changes in the development of lung cancer In order for normal cells to evade the normal cell regulatory processes and become malignant, it has been proposed that six or seven mutations are required. No clear hereditary component has been detected in lung cancer, although mutation of the p53 gene in the Li Fraumeni cancer syndrome and loss of heterozygosity of the Rb gene are associated with increased risk of developing lung cancer. A large number of mutations of both tumour suppressor genes and oncogenes have been described (Table 20.1). The loss of tumour suppressor activity requires the loss of heterozygosity to an inactive homozygous state. Each step of bronchial carcinogenesis is a result of the accumulation of genetic damage involving several oncogenes and tumour suppressor genes (Fig. 20.2). This process of carcinogenesis appears to be increased when exposed to cigarette-smoke carcinogens.13 Mao et al. showed that loss of heterozygosity of 3p21–22, 17p13 and 9p21 was present in 82 per cent of smokers, but in only 10 per cent of non-smokers. Clonal changes do persist in ex-smokers. It also appears that the presence of certain mutations (K-ras and c-erb B2) is associated with poor prognosis.9
Intermediate Early Late
c-erb-B2 K-ras
5q21 13q
CIS
Late
Invasive carcinoma
Metastatic carcinoma
Figure 20.2 Genetic changes that occur in the progression from normal epithelium to metastatic carcinoma in SCLC. From 10,11 and 12
PREVENTION AND SCREENING Primary prevention As cigarette smoking causes the vast majority of lung cancer, most prevention strategies are aimed at reducing tobacco consumption, and several approaches have been tried. Health education programmes aim to promote the image of smoking as an abnormal, unpleasant activity, to discourage non-smokers from taking up smoking and to help smokers to give up. However, while the advertising budget of the tobacco industry far outweighs that of health education, these will have limited effect. The ban on tobacco advertising and smoking in many public places should help further. Indeed, in Ireland, where a ban on smoking in public places came into force in 2004, tobacco sales fell by 16 per cent in the first 6 months following the legislation. It has also been clearly shown that increasing the price of cigarettes results in decreased consumption (a 10 per cent increase in real price reduces consumption by 4 per cent), and therefore governments should continue to increase taxation above inflation. Finally, new measures are being taken to prevent young children taking up the habit. The results from campaigns in Scandinavia in the 1970s have been favourable and can be attributed to integrated central government and local policies. Chemo-prevention has been proposed to reduce the incidence of lung cancer. Many case-controlled and cohort
Clinical features 505
studies have found that the incidence of lung cancer is lower in people who have a high consumption of fruit and vegetables and in particular a high intake of β-carotene. In order to determine whether supplementation with β-carotene would reduce the risk of lung cancer in smokers, two large randomized studies have been performed. Surprisingly, both the Alpha-Tocopherol, Beta-Carotene (ATBC) study, involving more than 29 000 patients14** and the Beta-Carotene and Retinol Efficacy Trial (CARET)15** demonstrated an increased incidence of lung cancer (16–28 per cent) in those receiving β-carotene. In addition, this group also had higher rates of overall mortality, particularly due to ischaemic heart disease. Studies in non-smokers show no significant benefit.16 The current advice is that fruit and vegetable consumption should be increased, but that additional supplementation with β-carotene cannot be recommended.
Screening The purpose of screening is to detect early-stage tumours or pre-cancerous lesions in the hope that treatment will lead to reduced mortality. Lung cancer should be an ideal model, in that high-risk groups can easily be identified and methods for screening such as sputum cytology and chest radiography are widely available. However, four large randomized, controlled studies involving more than 37 000 patients have shown that the use of sputum cytology and/ or plain chest X-rays led to more early cases of lung cancer being detected, but none showed a significant reduction in mortality (reviewed in 17 and 18). Many new methods for the early detection of the disease are currently under investigation. Conventional sputum cytology has a poor detection rate, but the use of induced sputum and advanced molecular biology may allow the detection by polymerase chain reaction (PCR) techniques of chromosomal mutations found commonly in lung cancer, for example p53 and K-ras. Newer immunohistochemical techniques using an antibody staining the ribonucleoprotein (RNP) binding protein hnRNP A2/B1 have also shown promise. Preliminary studies have shown that this may be one of the most powerful predictors of early disease, with a detection rate of 80 per cent.19 Others have investigated automated methods of examining the nuclear changes in cells, like those used in cervical screening. Payne et al.20* used this method on both lung biopsy specimens and sputum cytology from participants in the NCI Early Lung Cancer Study and demonstrated significant changes in more than 75 per cent of normal-appearing specimens. Other novel methods of early detection have used breath tests, which rely on the fact that normal human breath contains volatile organic compounds. Several of these compounds have been identified as markers for lung cancer. Using this method, Phillips et al.21 predicted the diagnosis in 72 per cent of patients undergoing bronchoscopy.
RADIOLOGICAL SCREENING
Plain chest X-rays are poor at detecting very early-stage disease. Manser22*** reviewed seven screening trials including 245 000 patients. Frequent screening with chest X-ray resulted in an 11 per cent increase in mortality from lung cancer, while there was a non-significant trend in mortality in those screened with a combination of chest X-ray and sputum cytology. Chest X-ray has been compared with lowdose spiral computed tomography (CT) scanning in Japan and the USA. In the Japanese study,23* carried out in an area of low incidence, the rate of detection of lung cancer was 0.48 per cent, almost ten-fold that using chest X-rays, and of these patients, 84 per cent had stage I disease. The detection rate was considerably higher (2.7 per cent by CT versus 0.7 per cent by chest X-ray) in the USA study, which was performed in high-risk individuals.24* However, in both of these studies a larger number of patients with benign non-calcified nodules were detected who then underwent further investigation. While this method does seem to be particularly sensitive at detecting stage I disease in asymptomatic individuals, the mortality benefits are unproven and the health economic implications are enormous. The National Lung Screening Trial in the USA, which has randomized approximately 50 000 current or ex-smokers to screening with spiral CT or chest X-ray, should provide further information.
CLINICAL FEATURES The vast majority of patients with lung cancer present with symptomatic disease. In a study of 1277 patients, only 12 per cent were asymptomatic, cough was a presenting feature in 50 per cent, dyspnoea in 34 per cent, chest pain in 30 per cent and haemoptysis in 30 per cent, and 50 per cent of patients presented with non-specific systemic symptoms or symptoms of metastatic spread.25 Patients who present with a long history of symptoms related to the primary tumour have been found to have a better 5-year survival, suggesting that the growth of these tumours may be more indolent. The presenting features of lung cancer depend on the location of the tumour, rapidity of growth and co-morbidity of the patient. The majority of patients (70 per cent or more) have regional or metastatic disease, making resection impossible. Small-cell lung cancer has the most rapid growth rate of the four major types of lung cancer and metastasizes early. Symptoms from SCLC occur earlier and are more varied than those from NSCLC. Non-metastatic syndromes affecting the nervous and endocrine systems are much more common with the small-cell subtype. The median interval between the onset of symptoms and diagnosis is about 4 months. This delay occurs because symptoms are often of gradual onset and on a background of chronic obstructive airways disease in smokers. A summary of the initial symptoms of patients with lung carcinoma is given in Table 20.2.26
506 Bronchus
Table 20.2 Initial symptoms at presentation26 Initial symptoms Cough Weight loss Dyspnoea Chest pain Haemoptysis Bone pain Clubbing Fever Weakness SVCO Dysphagia Wheeze/stridor
Range of frequency % 8–75 0–68 3–60 20–49 6–35 6–25 0–20 0–20 0–10 0–4 0–2 0–2
Symptoms related to the primary tumour Cough is the most frequently reported symptom and is often disregarded by the patient and doctors for some time in smokers. The development of a new cough or change in its character is therefore important. In particular, the failure of a chest infection to resolve or the onset of haemoptysis should prompt further investigation. Dyspnoea may be secondary to intrinsic or extrinsic obstructive lesions or the development of pleural effusions. If tumour involves the major airways, very severe dyspnoea can rapidly develop, and the presence of stridor must always be regarded as an emergency. Chest pain can arise from the primary tumour or superimposed infection and does not necessarily imply metastatic disease. The pain is often described as heavy, is poorly localized and may be dismissed as unimportant. More severe pain can be caused by chest-wall invasion.
Symptoms related to regional spread Nerve involvement is often seen in the Pancoast syndrome, caused by superior sulcus tumours invading the brachial plexus and ribs. This is associated with severe shoulder pain radiating down the arm, classically in the T1 distribution, and weakness of handgrip. Horner’s syndrome due to involvement of the sympathetic chain can also occur. Superior vena caval obstruction (SVCO) is another dramatic and extremely unpleasant symptom complex occurring in up to 4 per cent of patients at presentation. The syndrome results from local tumour extension or nodal metastases with compression of the SVC. Venous engorgement of the face and neck is an early presenting feature, but eventually a collateral circulation with dilatation of superficial chest-wall veins develops. Other symptoms of SVCO may include headaches, drowsiness, vertigo, dyspnoea and dysphagia. Symptoms of intra-thoracic spread of tumour include dysphagia from mediastinal lymphadenopathy or direct
infiltration of the oesophagus. Pericardial infiltration can cause arrhythmias, effusions and occasionally tamponade. Both the phrenic and left recurrent laryngeal nerves may be damaged by central tumours, resulting in paralysis of the diaphragm and hoarse voice, respectively. The use of Teflon injection into the vocal cords can be helpful in improving the cough reflex to clear bronchial secretions and to strengthen the voice.
Extra-thoracic metastases Extra-thoracic lymph-node metastases may be visible or palpable in the supraclavicular, cervical and axillary regions. The presence of upper abdominal lymphadenopathy, usually arising from SCLC, can cause poorly localized upper abdominal or back pain, jaundice, nausea, anorexia or vomiting. Other common sites of metastases include liver, bone and brain. Liver metastases are seen in more than 35 per cent of patients at autopsy, and characteristic symptoms are pain and a feeling of fullness. Bone lesions occur in about 25 per cent of patients and are usually lytic and most common in the spine, pelvis and femur. Brain metastases are particularly common in SCLC and adenocarcinoma, seen in up to 80 per cent of patients at autopsy. They can present with focal neurological deficits, with features of raised intracranial pressure such as headaches and vomiting or impaired mental function or with personality changes. The adrenals and other endocrine organs are also prone to metastases in SCLC; the incidence may be as high as 30 per cent, although patients remain asymptomatic until 90 per cent of the gland has been replaced. Less common metastases include choroidal metastases, causing retinal detachment, and skin metastases (less than 5 per cent). Lung cancer is frequently associated with constitutional symptoms including fatigue and weight loss, which occur in more than 50 per cent of patients at presentation. Pyrexia of uncertain origin and normochromic, normocytic anaemia are also features of lung cancer. Anaemia, particularly in the case of SCLC, may result from marrow infiltration. A multitude of non-metastatic or paraneoplastic syndromes may occur, particularly in SCLC, some preceding the diagnosis of lung cancer by many months (Table 20.3).
DIAGNOSIS AND STAGING INVESTIGATIONS Diagnosis Patients with lung cancer usually present with the symptoms described above. Some asymptomatic patients will have abnormalities picked up incidentally on chest X-ray. Patients presenting with symptoms suggestive of lung cancer should have a chest X-ray. A contrast-enhanced CT scan should be performed in any patient with a chest X-ray abnormality; CT appearances may allow differentiation between a benign and malignant process within the lung
Diagnosis and staging investigations 507
Table 20.3 Paraneoplastic syndromes associated with lung cancer, particularly small cell lung cancer System
Syndrome
Comments
Skeletal
Clubbing Pulmonary hypertrophic osteoarthropathy
⬃30% of patients, squamous cell carcinoma common ⬃5% of patients, adenocarcinoma common, periostitis of long bones.
Cutaneous
Dermatomyositis
Weakness of pelvic girdle muscles, heliotrope rash, 15–20% of patients will have underlying malignancy. Hyperkeratosis and pigmentation of axillae, neck and flexures.
Acanthosis nigricans Neurological
Subacute cerebellar degeneration Sensorimotor peripheral neuropathy Polymyositis Lambert- Eaton syndrome Autonomic neuropathy
Endocrine
Ectopic parathyroid hormone Cushing’s syndrome Syndrome of inappropriate ADH secretion (SIADH) Hypercalcitonaemia Gynaecomastia
Subacute, symmetrical cerebellar failure. Anti-Purkinje and Yo antibodies detected. Distal motor and sensory loss. Anti-Hu antibodies are a marker. Proximal muscle weakness Fatigue of proximal muscles, poor response to edrophonium test. 1–6% SCLC. Orthostatic hypotension, neurogenic bladder, GI dysmobility. SCLC. Squamous cell carcinoma. ⬃15% of patients develop hypercalcaemia Ectopic ACTH production, usually presents with hypokalaemia, 1–5% of SCLC patients. Hyponatraemia, ⬃30% patients with SCLC. May lead to confusion and seizures. Hypocalcaemia, may present with paraesthesia, cramps and tetany Due to ectopic gonadotropin production, common in large cell and adenocarcinomas.
Renal
Nephrotic syndrome Membranous glomerulonephritis
Secondary to immune complexes, thrombus, amyloid.
GI
VIP secretion
Profuse diarrhoea, hypokalaemia
and CT also permits staging. Ideally, the CT scan should be performed before bronchoscopy, thus providing information on the position of the tumour and the presence of metastases, and allowing clinicians to determine the most appropriate method of obtaining tissue diagnosis. Histological confirmation of the diagnosis of the primary tumour should be obtained by bronchoscopic biopsy, transthoracic biopsy or, rarely, surgical biopsy. Histology can also be obtained from metastatic sites, e.g. by fine-needle aspiration of lymph nodes, liver biopsy. Computed tomography staging is limited by poor sensitivity and specificity. FDG-PET scans are of additional benefit in assessing patients who are being considered for radical treatment. However, in patients with positive mediastinal nodes on CT, there is still a high false-positive rate, even using combined imaging. In those patients who are thought to have N2/N3 disease on CT/PET, the nodes should be sampled, using transbronchial needle aspiration (Wang technique) or mediastinoscopy. FDG-PET is also useful in detecting metastatic disease, and in an observational study the results of a PET scan changed the management of
83 per cent of patients.27 Additional investigations (e.g. bone scan, brain scan) are recommended only if prompted by symptoms.
Staging Accurate staging is important, first to give patients and carers accurate prognostic information, second to help decide the most effective treatment, and finally to allow comparison between different treatments and institutes. The international standard for staging in NSCLC is the TNM system, originally proposed by Denoix in 1946. The TNM subgroups can be amalgamated into a smaller number of disease stages. The original TNM system adopted by the American Joint Committee on Cancer Staging was updated in 1986 and 1997.28 The current international TNM staging classification is given in Table 20.4 and the new stage groupings in Table 20.5.28 The prognostic significance of various stages and TNM subsets is discussed later in the treatment section, but the overall influence of stage is shown in Figure 20.3.
508 Bronchus
Table 20.4 International TNM staging for lung cancer. From Mountain28 Primary tumour (T) TX Primary tumour can not be assessed, or tumour proven by the presence of malignant cells in sputum or bronchial washings but not visualised by imaging or bronchoscopy T0 No evidence of primary tumour Tis Carcinoma in situ T1 Tumour 3 cm or less in greatest dimension, surrounded by lung or visceral pleura, without bronchoscopic evidence of invasion more proximal than the lobar bronchus. T2 Tumour with any of the following features of size or extent: • 3 cm in greatest dimension • Involves main bronchus, at least 2 cm distal to the carina • Invades the visceral pleura • Associated with atelectasis or obstructive pneumonitis that extends to the hilar region but does not involve the entire lung T3 Tumour of any size that directly invades any of the following: chest wall (including superior sulcus tumours), diaphragm, mediastinal pleura, parietal pleura or tumour in the main bronchus less than 2 cm from the carina, but without involvement of the carina; or associated atelectasis or obstructive pneumonitis of the entire lung. T4 Tumour of any size that invades any of the following: mediastinum, great vessels, trachea, oesophagus, vertebral body, carina or tumour with malignant pleural or pericardial effusion, or with satellite tumour nodule(s) within the ipsilateral primary-tumour lobe of the lung Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastases N1 Metastases to ipsilateral peribronchial and/ or ipsilateral hilar lymph nodes, and intrapulmonary nodes involved by direct extension of the primary tumour N2 Metastases to ipsilateral mediastinal and/or subcarinal lymph nodes N3 Metastases to contralateral mediastinal, contralateral hilar, ipsilateral or contralateral scalene or supraclavicular lymph nodes Distant metastases (M) MX Presence of distant metastases cannot be assessed M0 No distant metastases M1 Distant metastases present Notes T1 T4
M1
The uncommon superficial tumour of any size with its invasive component limited to the bronchial wall, which may extend proximal to the main bronchus is also classified as T1 Most pleural effusions associated with lung cancer are due to the tumour. However, there are a few patients in whom multiple cytopathological examinations of pleural fluid show no tumour. In these cases, the fluid is non-bloody and is not an exudate. When these elements and clinical judgement dictate that the effusion is not related to the tumour, the effusion should be excluded as a staging element and the patient should be staged T1, T2 or T3. Pericardial effusion is classified according to the same rules. Separate metastatic tumour nodules in the ipsilateral nonprimary-tumour lobe(s) of the lung are also classified M1
Patients with clinical stage I, II and IIIA NSCLC can be considered for curative intent surgery or radical radiotherapy; for patients with stage IIIB and IV disease, palliative radiotherapy, chemotherapy or combined modality treatment might be appropriate. Detailed preoperative staging is required for NSCLC in order to avoid unnecessary surgery (see above). Additional investigations are required to assess the suitability of patients for surgery, including cardiovascular evaluation and pulmonary function tests to assess postoperative reserve. Small-cell lung cancer requires a different staging system, as no difference in survival is found in resected patients with TNM stages I, II and III. Small-cell lung cancer has therefore
been divided into two stages: limited and extensive. About 30 per cent of patients have limited disease, which is defined as disease confined to one hemithorax with regional metastases, including hilar, ipsilateral and contralateral mediastinal, and ipsilateral and contralateral supraclavicular nodes and ipsilateral pleural effusions, whether or not the cytology is positive. Extensive stage disease is defined as disease beyond this, including distant lymph nodes, brain, liver, bone, bone marrow and intra-abdominal soft-tissue metastases. The importance of stage for survival is discussed below. Restaging investigations are necessary to monitor the progress of the disease and the response to treatment. The definition of remission of the tumour requires full
Pathology 509
Table 20.5 Stage of lung cancer as defined by the TNM classification. 5 year survival is given according clinical estimates of the extent of the disease. From Mountain28 Stage
TNM subset
0 IA IB IIA IIB
Carcinoma in situ T1 N0 M0 T2 N0 M0 T1 N1 M0 T2 N1 M0 T3 N0 M0 T3 N1 M0 T1 N2 M0 T2 N2 M0 T3 N2 M0 T4 N0 M0 T4 N1 M0 T4 N2 M0 T1 N3 M0 T2 N3 M0 T3 N3 M0 T4 N3 M0 Any T Any N M1
IIIA
IIIB
IV
5 year survival (%)
61 38 34 24 13
5
1
100
% Survival
80 60
IA
40
IB IIA IIB
20
IIIA IIIB IV
0 1
2 3 Years from treatment
4
5
Figure 20.3 Overall survival in patients with NSCLC according to stage at diagnosis.
restaging to determine whether previous abnormalities have resolved and to what extent. It is recommended that patients be followed up regularly to identify progressive disease, complications of treatment such as pulmonary fibrosis, or second malignancies.
Other prognostic factors Additional features at presentation can be useful in assessing prognosis. Performance status is one of the most important prognostic indicators. As expected, patients who are fully mobile with few or no symptoms respond better to treatment and have better survival than those who are bed-bound.
The various scales in common use – World Health Organisation (WHO), Karnofsky, ECOG and Zubrod – all predict for survival. In addition to assessing prognosis, performance status is also valuable in monitoring the effects of treatment. More complex measures of quality of life are now being incorporated in clinical studies to assess the benefit of different treatment strategies that have a similar survival. It is important to note that age has not consistently been shown to be an independent prognostic factor affecting survival for either SCLC or NSCLC. A number of multivariate analyses of adverse prognostic factors have been performed in SCLC. The parameters identified have included low performance status, low serum albumin, low sodium, high alkaline phosphatase, extensive disease, elevated lactate dehydrogenase, high aspartate aminotransferase and low bicarbonate. A systematic review29*** identified poor performance status and raised lactate dehydrogenase as the most consistent factors associated with poor prognosis. Other prognostic factors of moderate importance were extensive stage disease, male sex and the presence of multiple metastatic sites. The importance of raised alkaline phophastase, low sodium and old age was less well defined. Placing patients into a defined prognostic group, for example with no adverse features, allows treatment to be tailored. Thus, intensive treatment can be given to those patients in the better prognostic groups, while patients identified as having a poor prognosis can be offered a more palliative approach to therapy. An analysis of prognostic factors for NSCLC examined features identifying patients who survived for more than 12 months following chemotherapy. Favourable characteristics were good performance status, female, no bone, liver or subcutaneous metastases, non-large-cell histology, little weight loss and no shoulder or arm pain. Other groups reported a favourable prognosis for patients with a single site of extrathoracic metastases. The use of cisplatin-based therapy and a low lactate dehydrogenase have also been shown to be powerful independent predictors in multivariate analyses.30
PATHOLOGY It is extremely important that an accurate pathological diagnosis is made. Lung cancer may be suspected radiologically, but pathological confirmation is required and may be performed by cytology or histology, using a variety of techniques. Sputum cytology is the least invasive method, but bronchoscopy with washings or biopsy, transthoracic needle biopsy or biopsy of lymph nodes or metastatic deposits is often required. Histologically, lung cancers are classified according to their light-microscopic appearance rather than their histogenesis. They are divided pragmatically into two clinicopathological groups, small cell and non-small cell. Non-small-cell cancers include squamous cell, adenocarcinoma and largecell tumours as defined by the WHO/ IASLC in 199931 (see Table 20.6).
510 Bronchus
Table 20.6 Histological classification of lung tumours, WHO/IASLC 199931 1. Squamous cell carcinoma Variants Papillary Clear cell Small cell Basaloid 2. Small cell carcinoma Variant Combined small cell carcinoma 3. Adenocarcinoma Acinar Papillary Bronchioloalveolar carcinoma Non-mucinous (Clara cell / type II pneumocyte type) Mucinous (Goblet cell type) Mixed mucinous and non-mucinous (Clara cell/type II pneumocyte and goblet cell type) or indeterminate Solid carcinoma with mucin formation Adenocarcinoma with mixed subtypes Variants Well differentiated fetal adenocarcinoma Mucinous (‘colloid’) adenocarcinoma Mucinous cystadenocarcinoma Signet ring adenocarcinoma Clear cell adenocarcinoma 4. Large cell carcinoma Variants Large cell neuroendocrine carcinoma Combined large cell neuroendocrine carcinoma Basaloid carcinoma Lymphoepithelioma-like carcinoma Clear cell carcinoma Large cell carcinoma with rhabdoid phenotype 5. Adenosquamous carcinoma 6. Carcinomas with pleomorphic, sarcomatoid or sarcomatous elements Carcinomas with spindle and/or giant cells Pleomorphic carcinoma Spindle cell carcinoma Giant cell carcinoma Carcinosarcoma Pulmonary blastoma 7. Carcinoid tumour Typical carcinoid Atypical carcinoid 8. Carcinomas of salivary gland type Mucoepidermoid carcinoma Adenoid cystic carcinoma Others 9. Unclassified carcinoma
In the UK and many other European countries, squamouscell carcinomas account for the majority of cases (40 per cent), with adenocarcinomas and small-cell carcinomas accounting for 15–25 per cent each and large-cell carcinomas and the other rarer forms accounting for 10–20 per cent. The incidence of the different forms of cancer varies considerably between countries, with adenocarcinoma being the most common in the USA. This range of incidence may reflect the source of tissue for diagnosis. For example, as squamous-cell carcinomas are more often resected, these may be overrepresented, and small-cell carcinomas are prone to crushing, making interpretation difficult. Both intra-observer and inter-observer disagreement can occur,32 which obviously has implications for the management of patients. Therefore, in clinical trials, review mechanisms should be available. Squamous-cell carcinomas occur more often in men than in women and are most commonly related to smoking. The tumours typically occur in the major bronchi, and the surrounding epithelia commonly shows areas of metaplasia or dysplasia. Macroscopically, these tumours appear as firm lesions with a pale-grey, gritty surface. Microscopically, the hallmarks of squamous differentiation are apparent, with prominent intercellular bridges and keratinization. Immunohistochemical staining with antibodies against highmolecular-weight cytokeratins (63 kD) is helpful in the diagnosis, in particular cytokeratin 14. Other antibodies include those staining for involucrin and carcinoembryonic antigen (CEA). Adenocarcinomas occur with equal frequency in men and women and are less commonly associated with smoking. A precursor lesion, atypical alveolar hyperplasia, is now well recognized. These tumours are typically peripheral. Macroscopically, they form irregular round masses, whose cut surfaces often have a myxoid appearance. Microscopically, varying degrees of glandular differentiation are seen and the nuclei often have prominent nucleoli. Immunohistochemistry and mucin histochemistry are often important to differentiate these tumours from mesotheliomas. Antibodies staining the surfactant apoprotein and CEA may be positive, whereas BerEP4 and the mesothelial markers CK5, thrombomodulin and calretinin are negative. Differentiating between pulmonary adenocarcinomas and metastases from non-pulmonary adenocarcinomas has at times been difficult; however, the use of stains against thyroid transcription factor-1 (TTF-1) has transformed this. Thyroid transcription factor-1 is a homeodomaincontaining transcription factor that plays a role in lung development, cell growth and differentiation; TTF-1 staining has been reported to be positive in 75–100 per cent of pulmonary adenocarcinomas, 40–79 per cent of large-cell carcinomas and 86 per cent of small-cell carcinomas.33,34 Mucinous pulmonary adenocarcinomas do not stain well. Staining is also positive in thyroid carcinomas and has been reported in a very small number of non-pulmonary adenocarcinomas, such as gastrointestinal, ovary and endometrial. Carcinoid tumours account for only 2 per cent of all lung cancers and usually behave in a benign fashion. They
Treatment of non-small-cell lung cancer 511
tend to arise in the major bronchi and contain large numbers of neurosecretory granules. Small-cell lung cancers are usually found in the major bronchi and are associated with smoking. The tumours are associated with the secretion of a number of hormones, including antidiuretic hormone, calcitonin and adrenocorticotrophic hormone (ACTH). Macroscopically, the tumours are friable, with a pinky/tan cut surface. Microscopically, they are characterized by the presence of small cells, approximately three times larger than a lymphocyte, with large pleomorphic nuclei and scanty cytoplasm. Crush artefacts following bronchoscopic biopsy are common. Antibodies such as Cam 5.2, chromogranin, synaptophysin, CD56 (neural cell adhesion molecule, NCAM) and NSE (neurone-specific enolase) are the most useful, but some NSCLCs may stain with these. Therefore morphology is very important and forms the principal means of diagnosis. A problem that may arise for pathologists is that approximately 10 per cent of tumours are of mixed cell type. Mixed small-cell and large-cell tumours account for between 4 per cent and 6 per cent of tumours and are associated with a poorer prognosis and poor response to therapy. In addition, changes in cell type may occur following treatment; in one study, 11 of 40 cases revealed NSCLC histology at relapse.35 The presence of tumours of mixed cell type implies either separate primary tumours or a common stem cell undergoing different differentiation processes. The proposition of a common stem cell is supported by the observation that neuroectodermal markers such as L-dopa decarboxylase and NSE, which were formerly associated exclusively with SCLC, can also be found in some NSCLCs. Indeed, the presence of these markers is associated with a more aggressive clinical course and greater sensitivity to chemotherapy. The development of techniques for culturing lung cancer cell lines has led to improved understanding of the biology of lung cancer. The SCLCs are characterized by the presence of dense core granules on electron microscopy, associated with the ability to secrete a variety of ectopic peptides and hormones. It is now increasingly evident that many of these can act as growth factors. They may act on the secreting cell (autocrine) or nearby cells (paracrine) to stimulate tumour growth. The development of antagonists to these growth factors is being explored as a potential therapeutic intervention. Many researchers have established antibodies to lung cancer cells. These have been tested against each other on panels of different tissues to characterize the antigenic epitopes they recognize. Five such antigen clusters are now recognized for SCLC and the most commonly expressed, cluster 1, has been identified as CD56 (NCAM). The development of monoclonal antibodies specific for SCLC will permit their use in diagnosis, imaging and treatment. Specific tumour markers would be useful in screening a high-risk population for early cancers and could also be useful in diagnosis, staging and monitoring the response to therapy. Unfortunately, none of the markers examined to date, including NSE, ACTH, calcitonin and neurophysin, has
been sufficiently sensitive and specific to be clinically useful in this role.
TREATMENT OF NON-SMALL-CELL LUNG CANCER Surgery Although there is little evidence from randomized controlled studies evaluating the benefits of surgery, many observational studies show a survival advantage. Surgical resection remains the prime modality of curative treatment for NSCLC. Resectability depends on the tumour location and stage and on the patient’s ability to tolerate the procedure. British Thoracic Society guidelines define fitness for surgery, with particular reference to age, pulmonary function, cardiovascular fitness, nutrition, weight loss and performance status.36 Patients with stage I and II tumours should be offered resection if preoperative assessment suggests that such a procedure would be safe. Some patients with stage IIIA tumours also have potentially resectable disease. Survival is dependent on the extent of the disease (Table 20.7), with patients with T3 N1 tumours doing better than those with N2 disease. N2 disease has a particularly poor prognosis if associated with T3 tumours, multiple nodal station involvement and adenocarcinoma histology. Patients with stage IIIB or IV disease are considered unresectable. The most suitable type of resection is determined by tumour size, location, stage and patient fitness. In patients with stage I disease, lobectomy is the procedure of choice, but in patients with poor pulmonary function, more limited resections can be considered and include segmentectomy and wedge resection. Systematic lymph-node evaluation should be performed in all patients to provide accurate pathological staging. A systematic review including 13 observational studies comparing segmentectomy and lobectomy has shown 5-year survivals of 62 per cent versus 80 per cent respectively, with increased loco-regional recurrence in those undergoing limited resection.37*** In studies comparing wedge resection and lobectomy, 5-year survival was reduced by 12 per cent in patients undergoing wedge
Table 20.7 Survival following surgical resection according to stage at diagnosis. From Mountain28 Stage IA IB IIA IIB IIIA
TNM
5 year survival %
T1 N0 M0 T2 N0 M0 T1 N1 M0 T2 N1 M0 T3 N0 M0 T3 N1 M0 T1-3 N2 M0
67 57 55 39 38 25 23
512 Bronchus
resection. Only one randomized trial of limited resection versus lobectomy has been reported.38** In the 247 patients randomized, lobectomy had a statistically non-significant 5-year survival advantage (73 versus 56 per cent) and locoregional recurrence rate was higher in the limited resection group. Average 5-year survival following lobectomy is 69 per cent for stage IA (T1) and 52 per cent for stage IB (T2). Minimally invasive video-assisted thoracic surgery (VATS) is being increasingly used for lung resection. No randomized controlled studies have compared VATS resections with open resections, but a number of observational studies have been reported and have shown similar survival figures, although patients are likely to have been highly selected. In terms of mortality and morbidity, two randomized controlled studies have been reported. One showed fewer postoperative complications but no change in length of hospital stay,39* and the second showed no significant difference in complication rates, length of hospital stay or postoperative lung function.40* However, more recent observational studies have shown a reduction in postoperative mortality and morbidity using VATS procedures.41* In patients with N1 disease (stages IIA and IIB), lobectomy should always be considered the operation of choice, but in patients with a central tumour, more extensive surgery may be appropriate. Pneumonectomy requires good pulmonary reserve. In patients with compromised lung function, sleeve resection can be considered, although there is a higher risk of loco-regional recurrence and postoperative complication rates are similar.42 T3 tumours involving the chest wall have an overall 5-year survival of 33 per cent, and survival is dependant on nodal status and whether or not complete resection has been possible. Patients with T3 tumours invading mediastinal structures have a poorer prognosis. The survival of patients with stage IIIA (N2) tumours depends on whether the N2 disease was detected only within the pathological specimen, or whether the nodes were felt to be resectable on preoperative staging. The number of patients undergoing surgery for unresectable N2 disease should reduce with improved staging, including the addition of PET. Advances in anaesthetic and surgical techniques have reduced postoperative mortality. The risk of postoperative death is higher in patients over the age of 70 years, in those with co-morbid conditions and following more extensive resections. Mortality rates following pneumonectomy are 6–8 per cent, whereas after lobectomy this is only 3 per cent. Morbidity is high, with about 20 per cent of patients having minor complications, the commonest being supraventricular tachycardia, and 10 per cent suffering from major problems such as embolus and pulmonary insufficiency. Studies examining quality of life post-resection have shown that preoperative levels are not achieved until at least 6 months after surgery.43 Treatment failure results from lack of control of local disease in about a third of patients and from metastatic spread in the remainder, with up to 10 per cent of patients
developing brain metastases. The majority of local relapses occur within the first 2 years following surgery; therefore close follow-up is required during this period. In addition to recurrent disease, a proportion of patients develop second primary lung cancer. Resection of new primary lung cancers should be performed if possible, even in the unusual circumstance of synchronous lung cancer, although survival is poor (approximately 25 per cent) even for low-stage disease. It is clear that surgery can cure some patients; however, the problem remains that the majority of tumours are unresectable at presentation. There is increasing interest in the use of neoadjuvant chemotherapy to render the tumour resectable, especially in patients with N2 disease. This is discussed below.
Radiotherapy Radiotherapy may be used with curative intent in patients with disease that is potentially operable but who decline or are not suitable for surgery. These patients represent the minority of those who receive radiotherapy, with the majority receiving palliative treatment.
RADICAL RADIOTHERAPY
Curative intent radiotherapy should be considered as an alternative to surgery in patients with stage I and II lung tumours who decline operation and in patients whose co-morbid conditions would make surgery unsafe. It may also be considered in some patients with stage III disease. Full assessment of patients is required, including pulmonary function and performance status. Significant weight loss prior to treatment is a poor prognostic indicator. If the forced expiratory volume (FEV) is greater than 1.0, most patients can be safely treated, but there is a significant risk of radiation pneumonitis and fibrosis. Radical doses of radiotherapy increase the risk of normaltissue damage and therefore CT planning and shielding of normal tissue are essential in order to deliver effective doses to the tumour whilst limiting toxicity to the normal lung and spinal cord. Treatment schedules vary, but most studies have used the biological equivalent of 60 Gy in 30 2-Gy fractions (Table 20.8). However, in the UK most centres now use 55 Gy in 20 fractions (equivalent to 64 Gy in 32 fractions). The current recommendation is that in patients with stage I disease the tumour only is irradiated, but in stage II disease the evidence is less clear and the ipsilateral hilum is often included. In an attempt to improve outcomes, the use of hyperfractionation has been investigated. A randomized controlled trial involving 563 patients compared continuous hyperfractionated accelerated radiotherapy (CHART) (54 Gy in 1.5-Gy fractions three times a day over 12 days) with conventional radiotherapy (60 Gy in 2-Gy fractions daily for 6 weeks). Overall, there was a
Treatment of non-small-cell lung cancer 513
Table 20.8 Examples of recent studies demonstrating survival following radical radiotherapy for NSCLC (F fractions of radiotherapy) Study
No patients
Talton et al. 199044 Rosenthal et al. 199245
77 40
Dosoretz et al. 199246 Graham et al. 199547 Morita et al. 199748
152 103 149
Tumour stage T1–3 N0 T1–2 N0 T1–2 N1 T1–3 N0 T1–2 N0–1 T1–2 N0
RT 60 Gy in 30 F 18–65 Gy 45–65 Gy in 1.8–2 Gy F Median 60 Gy in 30 F 55–74 Gy in 2 Gy F
22 per cent reduction in the risk of death in those who received CHART; 2-year survival was significantly improved from 20 per cent to 29 per cent.49**,50** Toxicity was increased, with earlier and more severe dysphagia and an increased incidence of pulmonary fibrosis and myelitis. Modified hyperfractionated regimens have been investigated and appear to produce similar results.51,52 COMBINING RADIOTHERAPY AND SURGERY
The use of preoperative radiotherapy in NSCLC has shown no benefit. A systematic review analysed two trials published in the 1970s.53*** A total of 899 patients were recruited and randomized to receive either 40 Gy or 50 Gy of preoperative radiotherapy followed by surgery or surgery alone. The 5-year survival rates were not statistically different. The studies have a number of limitations due to the staging and radiotherapy techniques at the time; in addition, 10–15 per cent of patients had SCLC. In an attempt to reduce the risk of local recurrence following surgery, a number of studies have investigated the role of postoperative radiotherapy. A systematic review54*** examined the outcome of 2232 patients from ten trials (median follow-up of 4.25 years). The results showed a significant adverse effect of postoperative radiotherapy on survival, with a hazard ratio of 1:1.8 or 18 per cent relative increase in the risk of death. This is equivalent to an absolute detriment of 6 per cent at 2 years (95% confidence interval (CI), 2–9 per cent), reducing overall survival from 58 per cent to 52 per cent. Subgroup analyses suggested that this effect was most pronounced for patients with stage I/II, N0–N1 disease, whereas for stage III, N2 patients, there was no clear evidence of an adverse effect.
PALLIATIVE RADIOTHERAPY
The principal aims of palliative treatment are symptom relief and symptom prevention, rather than life extension. The vast majority (80 per cent) of patients referred for consideration of radiotherapy for lung cancer are only suitable for palliative treatment, and in about 10 per cent radiation is not considered to be worthwhile. For patients with inoperable tumours in whom there is a negligible chance of
Med survival (months)
1 year (%)
2 year (%)
5 year (%)
18 15 18 17 16 27
57 72 70 NR NR 78
36 35 33 40 35 NR
17 NR 12 10 13 22
long-term survival, there is a difference of opinion as to whether patients who are relatively asymptomatic should receive immediate radiotherapy or whether treatment should be deferred. The Medical Research Council (MRC) study, LU17, has addressed this issue. Two hundred and thirty patients with untreated NSCLC whose disease was locally too advanced for resection or radical radiotherapy and with minimal symptoms were randomized to receive supportive care plus immediate thoracic radiotherapy, or supportive care plus delayed radiotherapy to treat symptoms. The study concluded that immediate palliative radiotherapy resulted in no improvement in symptom control or survival.55** The doses of thoracic radiotherapy used are lower than in radical treatments, the duration of treatment is shorter, and an important objective is to minimize side effects. A number of randomized controlled studies have addressed the effect of radiation dose on palliation and survival. It has been demonstrated that shorter courses of treatment (10 Gy in a single fraction or 17 Gy in two fractions) are as good as, if not better than, longer courses in palliating symptoms, with fewer adverse effects (Table 20.9). However, in an MRC study,58** the use of higher doses did produce a survival advantage. Improved survival was also seen in Bezjak’s study,60** in which 230 patients were randomized to 10 Gy in a single fraction or 20 Gy in five fractions. Those patients in the fractionated arm lived 2 months longer (p 0.03) and had improved scores for pain and global quality of life. Therefore many oncologists argue that high-dose palliation should be considered in patients with good performance status. The symptoms that respond best to palliative radiotherapy include SVCO and haemoptysis, which are relieved in 80 per cent or more of patients, while cough, chest pain and breathlessness are relieved in about 60 per cent of patients. These rates for symptom control are similar to those reported for palliative chemotherapy in NSCLC. Anorexia, weight loss, hoarseness, pleural effusion and non-metastatic features of lung cancer rarely respond to palliative radiotherapy. Other methods of palliation, including steroids, appropriate analgesia, nursing care and psychological support, should always be considered for patients with lung cancer. The use of endobronchial radiotherapy (brachytherapy) has been shown to be effective in controlling cough,
514 Bronchus
Table 20.9 Examples of recent trials investigating the effect of dose in palliative radiotherapy on symptom control in NSCLC Study
RT
No patients
Comment
MRC Lung Cancer Working Party 199156 MRC Lung Cancer Working Party 199257 MRC Lung Cancer Working Party 199658
30 Gy in 10 F vs 17 Gy in 2 F
369
Results for survival and palliation equivalent.
17 Gy in 2 F vs 10 Gy in 1 F
233
39 Gy in 13 F or 36 Gy in 12F vs 17 Gy in 2 F
509
Rees et al. 199759
22.5 Gy in 5 F vs 17 Gy in 2 F
216
Bezjak et al. 200260
10 Gy in 1 F vs 20 Gy in 5 F
230
1F group better for control of haemoptysis. No difference in survival Survival advantage in 13 F group, 9 vs. 7 months median survival. Palliation at 2–3 months better with 2 F and less side effects. Tendency towards better palliation in 2 F group. Median survival increased by 2 months and pain and global QoL improved in 5 F arm.
(F fractions of radiotherapy)
Table 20.10 Efficacy of brachytherapy for palliation in NSCLC Reference Langendijk et al. 200161
Stout et al. 200062
No patients 95
99
Regimen
Outcomes
Ext beam vs Brachy Ext beam
Improved re-expansion rates (57 vs 35%) and dyspnoea scores in EBB arm
Ext beam vs EBB
Ext beam improved re-expansion rates and symptom scores
breathlessness and haemoptysis in selected tumours causing intrinsic or extrinsic bronchial compression, with improvement in symptoms in 75 per cent. The most serious complication of this treatment is major haemoptysis. There is no clear evidence that this method of delivery is better than external beam (Table 20.10), but it may be useful in retreating patients who have received maximum doses to the spinal cord, although this needs evaluation in clinical trials. Laser therapy can also be used to destroy endobronchial tumours via the bronchoscope. Unfortunately, current laser delivery systems can only treat rather small endobronchial lesions, which therefore often require multiple treatments. Radiotherapy may also be useful in the treatment of metastatic disease, with bone pain being relieved in more than 50 per cent of patients, frequently with a single fraction of irradiation. Large-field, e.g. hemi-body, irradiation can be considered for widespread bony metastases. Palliative brain irradiation for metastatic disease can result in symptomatic improvement in 80 per cent or more of patients,
particularly those who have fits and headaches. However, in patients with focal neurological signs from brain metastases and spinal-cord compression, the results of radiotherapy are much less impressive. A neurosurgical opinion should be sought in otherwise reasonably fit patients with a single brain metastasis or spinal-cord compression secondary to NSCLC. ADVERSE EFFECTS OF RADIOTHERAPY
The side effects can be divided into short-term, intermediateterm and long-term effects. The most common short-term toxicity is dysphagia secondary to oesophagitis. This can occur in up to 50 per cent of patients and typically lasts for 2 weeks after completion of therapy, although in a minority of patients (5 per cent) this may be prolonged. Radiation pneumonitis is evident in up to 100 per cent of chest X-rays, but is only clinically apparent in 5–10 per cent of patients, occurring 6–12 weeks following treatment. The long-term effects include radiation fibrosis, which again is more common radiologically than clinically, but these changes may make the detection of recurrent tumour difficult. Transient radiation-induced myelitis may also occur and give rise to Lhermitte’s sign. However, there is usually complete recovery and only a small minority of patients (1 per cent) develop progressive limb weakness and paraparesis. Again this unusual side effect (due to vascular impairment) may be difficult to distinguish from tumour involvement, although the latter often causes pain.
Chemotherapy EARLY-STAGE DISEASE
There has been increasing interest in the role of chemotherapy in stages I, II and IIIA NSCLC in combination with
Treatment of non-small-cell lung cancer 515
Table 20.11 Examples of preoperative chemotherapy studies in NSCLC Study
No patients/Stage
Randomisation
Median Survival
Rosell et al. 199963
60 Stage IIIA
Surgery vs 3 MIC surgery
Roth et al. 199864
60 Stage IIIA
Surgery vs 3 CEP surgery 3 CEP
Depierre et al. 200265
355 Stage IB-IIIA
Surgery RT vs 2 MIP surgery 2 MIP RT
10 months 22 months* 14 months 21 months* 26 months 37 months
* p 0.05
other treatment modalities. Studies have shown that the response to chemotherapy in locally advanced disease is approximately twice that seen in advanced disease. Neoadjuvant chemotherapy prior to surgery The rationale for using chemotherapy prior to surgery is based on the effect of down-staging the tumour, the eradication of micrometastases and the fact that growth factors produced by the primary tumour may be reduced, so decreasing the stimulus to any residual cancer. However, surgeons remain concerned that preoperative chemotherapy may increase the peri-operative morbidity. It is clear that response rates to therapy are greater in early-stage disease than in advanced disease. The early studies in patients with stage IIIA tumours suggested a survival benefit, but these have not yet been confirmed in larger studies. Two small, randomized studies compared surgery alone to chemotherapy followed by surgery (Table 20.11). These studies were both closed prematurely, so only 120 patients were included in total. Although they both demonstrated a survival benefit for patients receiving neoadjuvant chemotherapy, the number of patients was small and the survival for patients undergoing surgery alone in the Rosell study was very poor.66 Objective response rates to chemotherapy were 60 per cent66 and 35 per cent,67 with only one patient and four patients, respectively, having progressive disease. The surgical resection rates did not differ significantly between the treatment groups in either study, and postoperative complications were not increased in the patients receiving preoperative chemotherapy. The median survivals have been revised with longer follow-up and are not as dramatic.63,64 A subsequent study of 355 patients with stage IB–IIIA disease randomized to receive neoadjuvant chemotherapy or surgery alone demonstrated that although disease-free survival was prolonged by chemotherapy (27 versus 13 months), improvements in median survival and 3-year survival were not significant.65** At present, neoadjuvant chemotherapy cannot be recommended for routine use. However, its role has been addressed further in randomized trials in a variety of stage settings, including the MRC study LU22. This trial included 519 patients and showed that 3 cycles of pre-operative cisplatin-based chemotherapy was well tolerated. Forty-nine per cent of patients responded. However, there was no evidence of benefit in overall survival (hazard ratio 1:1.02).68
Adjuvant chemotherapy following surgery A meta-analysis of chemotherapy after surgery in early NSCLC was included in the NSCLC Collaborative Group meta-analysis published in 1995. This included data from 4357 patients in 14 randomized trials of surgery versus surgery chemotherapy published between 1965 and 1990. It showed that adjuvant chemotherapy using alkylating agents resulted in a statistically significant increase in the risk of death (15 per cent, p 0.005), but that cisplatinbased adjuvant chemotherapy resulted in a 13 per cent reduction in the risk of death, but with borderline statistical significance (p 0.08), suggesting an absolute benefit of 5 per cent at 5 years.69*** The 1995 meta-analysis created an intense interest in the question of adjuvant chemotherapy for resected NSCLC and led to a number of large randomized trials being set up. Seven have now been published, in abstract or in full (Table 20.12). Two found no significant benefit from adjuvant chemotherapy and five found a survival advantage. An updated meta-analysis (including data from five of the seven new studies) has now been undertaken.78*** This included data from 7200 patients in 19 randomized trials. The authors separated the data into two groups: uracil and tegafur (UFT)-based and platinum-based regimens. An overall estimate of 13 per cent relative reduction in mortality (95% CI, 7–19 per cent) was found. There was an 11 per cent relative reduction in mortality associated with postoperative cisplatin-based regimens (95% CI, 4–18 per cent, p 0.004) and 17 per cent associated with UFT (95% CI, 5–27 per cent, p 0.006) as compared to surgery alone. This means that there would be an additional survivor at 5 years for 25 patients treated with cisplatin or for 30 with UFT. The two studies not included in the new meta-analysis recruited a further 826 patients and each shows a significant survival benefit for platinum-based chemotherapy.75**,76** Incorporating these data would strengthen the case for adjuvant treatment. It is now recommended that adjuvant chemotherapy should be discussed with patients who have undergone curative resection.41 It should be noted that the trials were restricted to patients with a performance status of 0–1 starting chemotherapy within 6–8 weeks of surgery. The survival benefits are restricted to those patients with completely resected stage IB–IIIA disease, and who therefore represent
516 Bronchus
Table 20.12 Trials of adjuvant chemotherapy in NSCLC Study
No pts
Stage
Chemo regimen
Median survival control vs adj
HR for survival
5 year survival Control vs adj
NR
NR
45 vs 63% p 0.04
48 vs 55 months NS
0.96 [0.81–1.13] p 0.59
1% NS
1209
I: 39%
Cisplatin etoposide 6 cycles MVP 3 cycles
1867
II: 33% III: 28% I: 37%
Cisplatin based
NR
0.86 [0.76–0.98] p 0.03
40 vs 45% p 0.03
Kato et al. 200473
999
II: 24% III: 40% I
UFT po 2 years
NR
NR
BLT, Waller et al. 200474
381
I: 27%
Cisplatin based 3 cycles
33 vs 34 months
1.02 [0.77–1.35] p 0.90
85 vs 88% p 0.05 NR
JBR.10, Winton et al. 200575
482
II: 38% III: 34% IB: 45%
Cisplatin VRB 4 cycles
73 vs 94 months p 0.04
0.69 p 0.04
54 vs 69% p 0.03
CALGB, Strauss et al. 200476
344
II: 55% Stage IB
Carboplatin taxol 4 cycles
NR
0.62 [0.41–0.95] p 0.03
4 year 59 vs 71% p 0.03
ANITA, Douillard et al. 200577
840
I: 35%
Vinorelbinecisplatin
43.7 vs 65.8 months p 0.013
NR
Mineo et al. 200170 ALPI, Scagliotti et al. 200371
IALT, Arriagada et al. 200472
66
IB
39 vs 51%
II: 30% IIIA: 35% MVP, mitomycin C, vinblastine and cisplatin
a subset of patients undergoing surgery, likely to comprise about one-third of all operated patients. The choice of regimen remains controversial. Trials of UFT have been restricted to Japan and Asia, where the pattern of disease is different. In Europe and North America, platinum-based regimens are preferred. Most of the trials used secondgeneration platinum combinations (e.g. cisplatin–etoposide), but there are some data for the third-generation drugs vinorelbine and paclitaxel. In the absence of trial data supporting the use of carboplatin, and given the continuing controversy over its possible inferiority to cisplatin in advanced NSCLC, cisplatin is preferred in the curative adjuvant setting. Combination chemotherapy and radiotherapy In patients with stage IIIA disease who decline surgery or in those with stage IIIB disease, chemo-radiotherapy is superior to radiotherapy alone. Performance status is of major prognostic importance in this patient group, and all the trial data are limited to patients of good performance status. A meta-analysis of the effects of chemotherapy in NSCLC included the analysis of data from 3033 patients in 22 trials comparing radical radiotherapy to chemo-radiotherapy.
It showed a small survival benefit from chemo-radiotherapy corresponding to an absolute survival benefit of 3 per cent at 2 years. However, when the analysis was restricted to trials of platinum-based chemotherapy, including data from 1780 patients in 11 trials, combination chemo-radiotherapy was associated with a 13 per cent reduction in the risk of death (p 0.005) and an absolute benefit of 4 per cent at 2 years.69*** Several more recent randomized studies using modern drug and radiation regimens have shown greater benefits.79–82 The largest, including 490 patients, showed that chemoradiation improved 2-year survival rates from 21 per cent to 32 per cent,83** a similar survival gain to that obtained with CHART (which included Stage I and II patients). The use of hyperfractionated radiotherapy in several studies has not led to improved survival, and standard chemo-radiotherapy was superior to hyperfractionated radiotherapy in one direct comparison.83** ASCO guidelines84 state that chemotherapy in association with definitive dose radiotherapy prolongs survival in patients with locally advanced unresectable stage III NSCLC, but that the optimum sequence remains unclear. The sequential use of chemotherapy and radiotherapy has
Treatment of non-small-cell lung cancer 517
Table 20.13 Examples of randomised studies investigating the use of chemo-radiotherapy in NSCLC compared with radiotherapy alone No of patients
Dillman et al. 199085 Schaake-Koning et al. 199286
Study
Rx
Median survival (months)
2 year survival (%)
Cis/Vinblastine
CT RT
13.8
26
60 Gy in 30F Cis 30 mg/m2 weekly or 6 mg/m2 daily 30 Gy in 10F 25 y in 10F
RT CT30 RT
9.7 NR
13 19
Sequence
Chemotherapy/ Radiotherapy
155
Sequential
331
Concurrent
CT6 RT RT
26 13
Jeremic et al. 199580
169
Concurrent
Carbo etoposide 64.8 Gy in 1.2 Gy F bd
CT RT RT
8 18
25 35
Sause et al. 200083
490
Sequential
Cis/Vinblastine 60 Gy in 2 Gy F 69.6 Gy in 1.2 Gy F bd
CT RT RT RT
13.2 11.4 12.0
8 at 5 years 5 at 5 years 6 at 5 years
(F, fractions; Cis, cisplatin; NR not reported, Carbo, carboplatin)
been the most studied and appears to have less toxicity than when given concurrently (Table 20.13). A phase III study in which 320 patients were randomized to receive concurrent or sequential thoracic radiotherapy with chemotherapy (mitomycin, vindesine and cisplatin) demonstrated that median survival was significantly improved (16.5 versus 13.3 months) in the concurrent group, but that this was offset by significantly greater myelotoxicity.87** A meta-analysis of concurrent chemo-radiotherapy versus radiotherapy alone (14 trials and 2393 patients) demonstrated a 7 per cent reduction in risk of death at 2 years. In a meta-analysis of three trials comparing concurrent with sequential chemo-radiotherapy, concurrent chemo-radiotherapy was associated with a significant reduction in risk of death at 2 years (hazard ratio, 0.86; 95% CI, 0.78–0.95 per cent; p 0.003), but with increased toxicity, particularly oesophagitis and myelosuppression.88*** Further studies are required to determine the impact on quality of life. The National Institute for Clinical Excellence (NICE) guidance41 concludes that the evidence of improved survival at 2 years with concurrent treatment may be at the expense of added toxicity, so concurrent chemo-radiotherapy cannot be recommended for routine use. Patients declining or unfit for chemotherapy may be offered radical radiotherapy alone, preferably CHART. CHEMOTHERAPY FOR ADVANCED DISEASE
The response rates seen with single-agent chemotherapy as reported in phase I and II studies are shown in Table 20.14, divided into first-generation, second-generation and thirdgeneration agents. For many years there was a great deal of pessimism concerning the role of chemotherapy in advanced NSCLC. This
Table 20.14 Mean overall response rates using single chemotherapy agents in NSCLC Drug
Mean overall response %
1st generation Cyclophosphamide Doxorubicin Methotrexate
8 13 10
2nd generation Cisplatin Etoposide Ifosfamide Irinotecan Mitomycin C Topotecan Vindesine
20 11 26 27 20 13 17
3rd generation Paclitaxel Docetaxel Gemcitabine Vinorelbine
26 26 21 20
arose because a number of trials using first-generation agents resulted in worse survival in patients receiving chemotherapy than in those receiving supportive care alone. In a metaanalysis, data were evaluated from 1190 patients in 11 trials comparing chemotherapy plus best supportive care with best supportive care alone. The analysis demonstrated that while alkylating agents had an adverse effect on prognosis (hazard ratio 1:1.26), the use of platinum-containing
518 Bronchus
regimens resulted in improved survival (hazard ratio 0.73) equivalent to an increase in 1-year survival of 10 per cent, from 15 per cent to 25 per cent.69*** At least nine more recent studies comparing a cisplatin-based regimen with best supportive care in more than 2000 patients confirmed that the use of chemotherapy improved prognosis leading to an improvement in median survival of about 9 weeks.41 The survival advantage is probably limited to those patients achieving disease control i.e. objective response and stable disease. Unfortunately, few studies have formally examined the effect of chemotherapy on quality of life. In those that have, however, it is clear that the effect is beneficial89 and is not limited to young, fit patients.90 In terms of specific symptom relief, Ellis et al.91 found that following a single cycle of chemotherapy (mitomycin C, vinblastine and cisplatin, MVP), 61 per cent of patients had a symptomatic response and that this increased to 96 per cent after two cycles. Symptoms that were best controlled included pain (60 per cent), cough (66 per cent), dyspnoea (59 per cent) and general malaise (53 per cent). In Socinski’s review,92*** the rate of symptom relief appeared to be higher than the objective response rate. In general, the subjective response rates are about double the objective response rates – i.e. 40–60 per cent. This is the main advantage of chemotherapy for many patients. Patient selection is important and a systematic review of prognostic factors found that performance status was the most important indicator of response, survival and toxicity.92*** Whilst it is clear that patients with performance status 0–1 benefit from chemotherapy, there remains controversy as to whether it should be offered to patients with performance status 2. An expert review reported that chemotherapy is justified in these patients, and either singleagent third-generation agents or carboplatin-based combinations should be considered.93*** In the MILES study.94 toxicity did not appear to be dependent on performance status, but survival was significantly worse in the performance status 2 patients (Median survival 10 weeks versus 19 weeks in the gemcitabine arm). Until recently, the standard regimens used contained platinum in combination with second-generation chemotherapeutic agents such as ifosfamide, vinblastine, vindesine and mitomycin C. However, the third-generation drugs such as gemcitabine, vinorelbine, paclitaxel and docetaxel have shown significant activity as single agents and also in combination and their use is now widespread. A number of phase III studies have compared regimens containing second-generation and third-generation agents (Table 20.15) and in general the response rate and survival are increased with the third-generation combinations. The NICE review identified six comparisons of platinum combinations with second-generation or third-generation drugs. In terms of clinical effectiveness, differences in 1-year survival rate and median survival did not reach statistical significance. However, this group concluded that third-generation combinations were preferred.41 Evidence to date suggests that three-drug combinations are more
toxic and no more effective than two-drug combinations. The current standard of care is therefore a two-drug combination including a platinum and a third-generation agent. Cisplatin combinations require pre-hydration and are more emetogenic and nephrotoxic than those using the analogue carboplatin. Carboplatin is therefore easier to administer, but is associated with more myelosuppression. Studies comparing carboplatin-containing and cisplatin-containing regimens have generally shown no significant differences in response rates or survival.105–108 However, a recent metaanalysis of five trials of platinum plus third-generation regimens included a subgroup analysis of cisplatin versus carboplatin and found that patients receiving cisplatin had an 11 per cent increased survival.109*** Further studies are directly comparing cisplatin versus carboplatin and different cisplatin doses. Direct comparisons of the different third-generation agents plus platinum have generally shown no obvious superiority in terms of response between any of the combin-ations (Table 20.16). Although Fossella’s study107** demonstrated that the docetaxel/cisplatin regimen resulted in significantly higher response rates and median survival than cisplatin and vinorelbine, this was not observed with carboplatin and docetaxel. Interestingly, patients on both docetaxel-containing arms had better quality of life compared to those receiving vinorelbine. Despite concerns about the equivalence of cisplatin and carboplatin, the most commonly used regimen in the UK is gemcitabine and carboplatin, whilst in the USA a taxane-containing regimen is often favoured. The use of non-platinum-containing regimens has also been investigated (Table 20.17). Gemcitabine in combination with vinorelbine, paclitaxel and docetaxel has been compared with a platinum-containing regimen and a systematic review of 14 randomized controlled studies found only one trial in which there was a significant difference in survival.117 In general, there was less myelotoxicity with the non-platinum-containing arm, but other toxicity was not significantly different. The 2004 ASCO guidelines have suggested that non-platinum-containing regimens could be used as an alternative in first-line therapy,84 but elsewhere platinum-containing regimens remain the gold standard. The optimal duration of treatment has been investigated. Socinski et al.118 randomized 230 patients to receive carboplatin and paclitaxel for either four cycles or continuously until disease progression. In the continuous arm, patients received between 0 and 15 cycles, but the median number given was four due to cumulative toxicity or disease progression. Response rates, quality of life and survival were not significantly different. In a second study,119** 308 patients were randomized to receive three or six cycles of MVP. In the cycle 3 arm, 72% of patients completed three cycles, while in the 6 cycle arm, 73% completed three cycles and only 31% of patients received six cycles. Again, there was no difference in response rates or median survival, but quality of life was improved in those patients who only received three cycles. In both these studies the survival advantage with longer schedules failed to reach statistical significance.
Treatment of non-small-cell lung cancer 519
Table 20.15 Examples of phase III trials of new chemotherapy agents in the treatment of NSCLC Third generation agent Vinorelbine
Gemcitabine
Paclitaxel
Chemotherapy regimens
Response Rates (%)
Med survival
1 year survival (%)
Wozniak et al. 199896 n 432
NVB NVB Cis Vin Cis Cis NVB Cis
14 30* 19* 12 26
31 weeks 40 weeks 32 weeks 6 months 8 months
30 33* 27* 20 36*
Crino et al. 199897 n 307 Cardenal et al. 199998 n 135 Danson et al. 200399 n 372 Rudd et al. 2005100 n 422
Gem Cis MIC Gem Cis Cis Etop Gem Carbo MIC Gem Carbo MIC
40 28* 40.6 21.9* 30 33 42 41
35 weeks 38 weeks 8.7 months 7.2 months* 236 days 248 days 10 months 7.6 months*
NR
Bonomi et al. 1996101 n 560
Tax (250/24) cis (75 G) Tax (135/24) cis Cis 75 etop Tax (175/3) Cis (80) Cis tenip HD-Cis (100) Tax (175/3) Cis 80
32.1 26.5* 12* 44 30 17 26*
10 months 9.5 months 7.7 months 9.4 months 9.7 months 8.6 months NS 8.1 months
Cis (75)
13.7
27.7 weeks*
21
Cis (75) Tir (390)
27.5*
34.6 weeks
33*
Study Le Chevalier et al. 199495 n 612
Giaccone et al. 1997102 n 332 Gatzemeier et al. 1998103 n 414 Tirapazamine
von Pawel and von Roemling 1998104 n 446
32 26* 33.2 32.5 40 30 39 37 NS 32 NS NR NR
NVB, vinorelbine; Cis, cisplatin; Vin, Vindesine; Gem, gemcitabine; MIC, mitomycin C; ifosfamide, cisplatin; Etop, etoposide; Tax, paclitaxel; Tenip, teniposide; HD, high dose; Tir, tirapazamine; NR, not reported; NS, not significant; *, p 0.05.
Table 20.16 Phase III studies comparing different 3rd generation regimens in NSCLC Study
Regimen
RR
Schiller et al. 2002106
Cis 75 paclitaxel 135 Cis 100 gem 1000 Cis 75 docetaxel 75 Carbo AUC6 paclitaxel 225
21 22 17 17
7.8 8.1 7.4 8.1
31 36 31 34
Scagliotti et al. 2002110
Cis 100 vinorelbine 25 Cis 75 gemcitabine 1250 Carbo AUC 6 paclitaxel 225
30 30 32
9.5 9.8 10.0
37 37 37
Fossella et al. 2003107
Cis 75 docetaxel 75 Carbo AUC 6 docetaxel 75 Cis 100 vinorelbine 25
31.6* 23.9 24.5
11.3* 9.4 10.1
46 38 41
Chemotherapy in the elderly With an increasingly elderly population – currently approximately 30 per cent of new patients with NSCLC are over the age of 70 years – discussions regarding the most suitable treatment will be increasingly common. Elderly patients often have considerable co-morbidity, the biology
MS
1 year survival
of their tumours may be different and the physiological changes of ageing all need to be considered. The evidence for the treatment of this population comes from trials designed for the elderly and those in which no upper age was set and in which a subset analysis was performed – and in these latter trials it could be argued that the patients were highly selected. In phase II trials of single-agent third-generation agents,
520 Bronchus
Table 20.17 Phase III trials of non-platinum combinations in NSCLC Drug Vinorelbine
Study
Regimens
RR %
Gridelli et al. 2003111 n 501
Gem Vin Gem Cis Vin Cis Vin Carbo Gem Vin
25 30 20.8 28
8.6 11.5*
Tan et al. 2005112 n 316
MS (months) 7.4 8.8
Paclitaxel
Kosmidis et al. 2002113 n 509 Smit et al. 2003114 n 480
Taxol Carbo Taxol Gem Taxol Cis Gem Cis Taxol Gem
28 35 32 37 28
10.4 9.8 8.1 8.9 6.7
Docetaxel
Georgoulias 2001115 n 317 Pujol et al. 2005116 n 311
D 100 Cis 80 D 100 Gem 1100 Gem 1000 D 85 Cis 100 Vin 30
35 33 31 35.9
10 9.5 11.1 9.6
Gem, gemcitabine; Vin, vinorelbine; Cis, cisplatin; Taxol, paclitaxel; Carbo, carboplatin; D, docetaxel.
response rates of 5–39 per cent and median survival times of 5–10 months were reported, with tolerable toxicity. A number of phase III trials were therefore instigated. The ELVIS trial120** randomized 161 patients over 70 years old to best supportive care or single-agent vinorelbine. Those patients who received chemotherapy had significantly improved survival (median 28 versus 21 weeks, 1-year survival 32 versus 14 per cent) and quality of life, both in terms of functional scores and tumour-related symptoms. Two phase III studies have also investigated the use of nonplatinum combinations. The SICOG study of 120 patients compared single-agent vinorelbine with the combination of gemcitabine and vinorelbine; combination therapy resulted in improved survival and delay in symptomatic deterioration.121** The MILES study (698 patients) compared single-agent vinorelbine, single-agent gemcitabine and the vinorelbine/gemcitabine combination.122** Overall survival was not significantly different between the arms (median survival 36, 28 and 30 weeks respectively) and there was no difference in quality of life scores. Randomized phase III trials of platinum-based combinations in the elderly are lacking, but subgroup analyses performed on studies without an upper age limit suggest that carboplatin-based regimens may improve outcomes and are tolerable in carefully selected patients.123 Current recommendations are that single-agent therapy should be considered in this population.124*** Second-line chemotherapy The role of second-line therapy has remained unclear for many years due to the limited numbers of patients fit enough to receive such treatment. A phase III study randomized 204 patients to single-agent docetaxel or best supportive care.125** The initial dose of 100 mg/m2 was reduced to 75 mg/m2 due to a high incidence of toxicity-related
deaths. The overall response rate was 5.8 per cent; median survival and 1-year survival were improved in the chemotherapy arm (7 months versus. 4.6 months, 29 per cent versus 19 per cent, respectively). Quality of life and symptom control were signifi-cantly improved in the chemotherapy arm. This trial formed the basis of the NICE guidance on second-line therapy. Pemetrexed is a multi-targeted antifolate agent, with the primary mechanism of action being the inhibition of thymidylate synthase. The toxicity of pemetrexed depends on folate and vitamin B12 nutritional status, and safe use requires supplementation with these vitamins. Hanna et al.126** randomized 571 patients who had previously received chemotherapy for advanced NSCLC to either docetaxel 75 mg/m2 or pemetrexed 500 mg/m2. Median survival was 8.3 months in the pemetrexed arm versus 7.9 months; 1year survival was 29.7 per cent in both arms, but there was significantly less neutropenia, febrile neutropenia and hospitalization in the pemetrexed arm.
Molecular targeted therapy EPIDERMAL GROWTH FACTOR RECEPTOR INHIBITORS
Epidermal growth factor receptor (EGFR) is overexpressed in approximately 30 per cent of NSCLC and is an adverse prognostic factor. Gefitinib and erlotinib are smallmolecule tyrosine kinase inhibitors that target the EGFR and are orally administered. Gefitinib (Iressa) was initially investigated in two phase II studies (TALENT133 and TRIBUTE134) as a single agent in patients who had received at least one line of chemotherapy for advanced NSCLC. In fact, many patients had received three or more prior chemotherapy regimens, making them a
Treatment of small-cell lung cancer 521
Table 20.18 Impact of treatment on survival in SCLC according to extent of disease (modified from Minna et al.137)
Therapy Supportive care Thoracic radiotherapy Single-agent chemotherapy Combination chemotherapy Combination chemotherapy thoracic radiotherapy
Median survival (months)
2 year survival (%)
LS
ES
LS
ES
3 3–9 6 10–14 12–16
1.5 – 4 7–11 7–11
– 2–7 – 5–15 10–35
– – – 1–3 1–2
LS limited stage, ES extensive stage
very unusual group. In both studies, patients were randomized to receive either 250 mg or 500 mg daily. The 250 mg dose was as active as the higher dose but with less toxicity (acne-like rash and diarrhoea), with overall response rates of 18.4 per cent and 12 per cent, and symptom improvement rates of 40.3 per cent and 43 per cent.127,128 It was noted that never-smokers, female patients, Asians, adenocarcinomas and rash were associated with a higher chance of response. However, a recent phase III study129** that randomized 1692 patients in a 2:1 ratio to gefitinib or placebo plus best supportive care in second or third line showed no difference in median survival in the overall population or for the patients with adenocarcinoma. However, a pre-planned subgroup analysis did demonstrate significant improvement in survival in patients who had never smoked (MS 8.9 versus 6.1 months) and in patients of Oriental origin (MS 9.5 versus 5.5 months). Gefitinib has also been investigated in combination with chemotherapy in two phase III trials – INTACT 1130 and INTACT 2131 – and again no additional benefit was seen. Erlotinib, however, has been shown to produce a survival benefit in patients who have previously received chemotherapy. The BR.21 study132** randomized patients to erlotinib or placebo and median survival was prolonged in those receiving the active agent (6.7 months versus 4.7 months, p 0.001). A response rate of 8.9 per cent was reported and improvement in quality of life and symptoms was statistically significant. Subgroup analysis has again shown increased benefit in females, non-smokers, patients with adenocarcinoma and those of Oriental origin. As with gefitinib, phase III studies investigating the addition of erlotinib to chemotherapy showed no benefit in survival (TALENT133 and TRIBUTE134). VASCULAR ENDOTHELIAL GROWTH FACTOR RECEPTOR INHIBITORS
Tumours require a blood supply in order to grow. Increased blood vessel counts and over-expression of vascular endothelial growth factor receptor (VEGFR) in NSCLC are associated with adverse prognosis. Bevacizumab is a humanized monoclonal antibody and blocks the binding of all VEGF-A isoforms to the receptor. A randomized phase II study comparing doses of 7.5 mg/kg and 15 mg/kg in combination with carboplatin plus paclitaxel versus
chemotherapy alone demonstrated response rates 10 per cent higher in the bevacizumab arms, with a trend towards improved median survival.135 The risk of bleeding is increased with bevacizumab and in this study there were four cases of fatal haemoptysis, all in patients with squamous-cell carcinomas. A subsequent phase III study136 comparing carboplatin/paclitaxel alone and in combination with bevacizumab excluded patients with squamouscell carcinoma and a history of haemoptysis. Median survival (12.5 versus 10.2 months), response rates (27 per cent versus 10 per cent) and progression-free survival (6.4 versus 4.5 months) were all significantly improved in the bevacizumab arm. Toxicity was increased in this arm and there were fatalities due to haemoptysis.
TREATMENT OF SMALL-CELL LUNG CANCER Chemotherapy Without treatment, the median survival from diagnosis of SCLC is 6 weeks for patients with extensive disease and 12 weeks for limited disease. Until the 1970s, radiotherapy was the standard treatment for SCLC, but this had little effect on median survival (Table 20.18). The introduction of chemotherapy significantly improved survival and it is now the treatment of choice. The beneficial effect of chemotherapy was first reported by Green and co-workers in 1969;138 patients with advanced SCLC who received single-agent cyclophosphamide compared to placebo had significantly improved survival. Active agents include alkylating agents (e.g. cyclophosphamide, ifosfamide), anthracyclines (e.g. doxorubicin, epirubicin), platinums (e.g. cisplatin, carboplatin), topoisomerase inhibitors (e.g. topotecan, etoposide and irinotecan), tubulin-binding agents (e.g. paclitaxel), and the vinca alkaloids (e.g. vincristine, vindesine). Objective responses in the order of 15–45 per cent have been reported for single agents, but combination chemotherapy results in much higher response rates and dramatic improvements in median survival. As discussed in “Staging investigations”, the TNM classification is rarely used in SCLC; instead, prognostic indicators and extent of disease are used to determine prognosis and potential response to treatment.
522 Bronchus
Table 20.19 Outcome of the use of single-agent oral etoposide compared to IV chemotherapy for poor prognosis SCLC
MRC Lung Cancer Working Party 1996139 n 339 Souhami et al. 1997140 n 155
Overall Response Rate
1 year survival
E
IV
E
IV
Comments
EV or CAV
45%
51%
11%
13% p 0.03
Increased haematological toxicity in E arm
PE/CAV
32.9%
45.3% p 0.01
9.8%
19.3% p 0.05
Decreased QoL in E arm
Etoposide (E)
Intravenous (IV) chemotherapy
50 mg bd for 10 days
100 mg bd for 5 days
EV, etoposide/vincristine; CAV, cyclophosphamide/doxorubicin/vincristine; PE, cisplatin/etoposide; QoL, quality of life.
It is clear that in particular patients with limited-stage disease and good performance status can achieve longterm survival. Standard combination chemotherapy regimens result in response rates of 85–95 per cent in patients with limited-stage disease, with complete responses in about 50 per cent. The overall response rate in patients with extensive disease falls to 65–85 per cent, with complete responses in only 25 per cent. Few patients have disease progression on initial chemotherapy, so the vast majority achieve some symptomatic benefit. The use of prognostic factors allows the identification of patients who are likely to have a good outcome from treatment, and the choice of chemotherapy regimen can be selected to match potential gain and current performance status. Combined chemo-radiotherapy or more toxic drug combinations, such as vincristine, ifosfamide, carboplatin and etoposide (VICE) and ifosfamide, carboplatin and etoposide (ICE), have impressive response rates and in some studies a 5-year survival of 15–17 per cent, but are associated with greater toxicity. These combinations may be an appropriate choice for fit patients with limited disease and no other adverse prognostic features. In contrast, the aim of treatment in patients with extensive disease is palliation to control the disease and improve symptoms with minimal toxicity. Drugs are chosen for combination therapy because they are active as single agents, have different mechanisms of action, differing toxicities and compatible schedules. Two important studies have confirmed that combination chemotherapy is superior to single-agent oral therapy in SCLC patients of poor prognosis.139,140** Both studies compared oral etoposide (50 mg or 100 mg daily for 10 days) with intravenous chemotherapy (etoposide and vincristine (EV) or cyclophosphamide, doxorubicin and vincristine or platinum and etoposide (PE/CAV). The response rates and 1-year survival were significantly better with intravenous combination chemotherapy (Table 20.19), whereas haematological toxicity and quality of life were worse with oral etoposide. These findings are consistent with those of an earlier randomized controlled study141 in which single-agent
intravenous cyclophosphamide was inferior to combination therapy for response rates (12 per cent versus 59 per cent) and median survival (18 months versus 31 months). Although combination chemotherapy is clearly superior to single-agent oral treatment,41 there is little evidence to support the use of more than three myelosuppressive drugs concurrently, as dosage is compromised by increasing myelosuppression with additional agents. Numerous chemotherapy regimens have been used. A recent survey in the UK showed that 34 regimens (151 schedules) were in routine use for SCLC.142 Simple outpatient regimens such as CAV and cyclophosphamide, doxorubicin and etoposide (ACE) were preferred in the UK. These were thought to have similar efficacy to the more complex platinum-containing regimens. However, there is now increasing evidence that platinum-containing regimens should be preferred. Early studies comparing cisplatin and etoposide (PE) with CAV have shown that whereas overall response rates were higher in the PE arm, there was little effect on overall survival.143,144 However, a meta-analysis145*** and two randomized controlled trials146,147** have demonstrated that platinum-based regimens are more effective. The meta-analysis, which included more than 4000 patients in 19 trials, demonstrated an increased response rate and improved 6-month and 12month survival. In the Sundstrom study, both 2-year survival (14 per cent versus 6 per cent) and 5-year survival (5 per cent versus 2 per cent) were statistically significantly increased in the PE arm,146 whereas the MRC LU21 study147 found prolonged survival in patients receiving VICE compared with standard chemotherapy. In addition, platinum-based regimens are associated with less myelosuppression and mucosal toxicity and are easier to give in combination with radiotherapy. The use of cisplatin does require pre-hydration and post-hydration, often necessitating an overnight stay. Carboplatin has therefore been incorporated in a number of regimens and appears to have similar efficacy to cisplatin.29 In practice, most oncologists use cisplatin where the primary aim is survival and carboplatin where palliation is the aim. In the UK, NICE
Treatment of small-cell lung cancer 523
Table 20.20 The effect of extended courses of chemotherapy on survival in patients with SCLC Median Survival Study Bleehen, et al. 1989150 n 497 Spiro et al. 1989151 n 616 Bleehen, et al. 1993152 n 458
2 year survival %
Chemotherapy
long
short
long
short
EMCV 12 vs 6
35 weeks NS 39 weeks p 0.007 8.6 mths NS
29 weeks
6
6
32 weeks
NR
NR
7.4 mths
7
8
CVE 8 vs 4 EMCV 6 vs 3
(EMCV, etoposide/cyclophosphamide/methotrexate/vincristine; CVE, cyclophosphamide/vincristine/etoposide; NR, not reported; NS, not significant)
guidelines41 now recommend a platinum-based regimen as first-line therapy. Chemotherapy should be given as soon as possible after diagnosis, even where the treatment is palliative. Two randomized studies have shown improvement in survival and quality of life if chemotherapy is given as soon as possible after diagnosis rather than being used to treat symptoms. The MRC Lung Cancer Working Party148 compared immediate treatment with six cycles of etoposide, cyclophosphamide, methotrexate and vincristine (ECMV) with the use of single-agent chemotherapy and radiotherapy for symptom control. The median survival was 32 weeks in patients receiving immediate therapy compared to only 16 weeks in those treated symptomatically. Earl et al.149 treated 300 patients either immediately with eight cycles of etoposide, cyclophosphamide and vincristine (CVE) or delayed treatment with the same regimen until onset of symptoms. Although no difference in survival was shown, quality of life score was better in patients receiving immediate treatment. Although chemotherapy improves survival in patients with SCLC, the optimal duration is unclear. Randomized studies have not shown an improved survival with prolonged treatment (Table 20.20). In a large British study,151 616 patients were randomized to receive either four or eight cycles of induction treatment with CVE. On progression, patients were randomized to receive either second-line chemotherapy or symptomatic treatment. The response rate after four cycles was 61 per cent, compared to 63 per cent after eight cycles. Whereas the patients who received only four cycles of induction chemotherapy and no second-line chemotherapy had the poorest survival, the length of induction therapy had no effect on survival in the patients who received further treatment on relapse. However, two MRC studies showed no survival advantage in patients who received extended courses of chemotherapy; indeed, quality of life was worse in patients receiving 12 cycles of treatment.152,153 MAINTENANCE CHEMOTHERAPY
As discussed above, there appears to be little benefit in increasing the number of cycles of chemotherapy given;
indeed, prolonged treatment has been shown to impair quality of life. Maintenance therapy with recombinant interferon following response to induction chemotherapy has been investigated by several groups, without clear evidence of benefit (Table 20.21). In some of these studies, the compliance with interferon treatment was limited by toxicity.154 Schiller et al.157 reported a randomized study in which 420 patients initially received four cycles of PE; those who had stable disease or who had responded were randomized to receive a further four cycles of topotecan or no further chemotherapy. Whereas progression-free survival was significantly improved, there was no improvement in overall survival or quality of life. A similar study was reported by Hanna et al.158 Patients with extensive-stage disease were treated with four cycles of cisplatin, etoposide and ifosfamide; 144 non-progressing patients were then randomized to receive three cycles of oral etoposide or no further therapy. Progression-free survival was significantly improved and there was a trend towards improved overall survival in the etoposide arm. A meta-analysis of 14 trials including 2550 patients has demonstrated an improvement in 2-year survival of 4 per cent in those receiving maintenance chemotherapy.159*** It is unclear as to which patients may benefit most, and the routine use of maintenance chemotherapy is not currently recommended.
ALTERNATING REGIMENS
The rationale for the use of alternating regimens is to reduce the development of resistant clones within the tumour. There does appear to be at least some lack of cross-resistance between PE and CAV, with an objective response rate of about 50 per cent compared with only 10 per cent with CAV following failure with PE. Reported results are contradictory, so alternating regimens are not routinely used. No benefit in the survival of patients with limited-stage disease was demonstrated by Feld et al.,160 but it has been reported with the use of this regimen by Fukuoka et al.143 These findings were not confirmed in a later study in patients with extensive disease144 (Table 20.22).
524 Bronchus
Table 20.21 Results of maintenance therapy using recombinant interferon (IFN) compared to observation only following response to induction chemotherapy for SCLC Study
No patients
Interferon dose
Median survival
Mattson et al. 1992153
237 with OR
IFN-2α 3 106 units
Jett et al. 1994154
100 with CR
IFN-γ 4 106 units per day for 6 months
Kelly et al. 1995155
171 with OR
IFN-2α 3 106 units 3x week for 2 years
van Zandwijk et al. 1997156
117 in CR or good PR
IFN-γ 4 106 units alternate days for 4 months or observation
IFN – 11 months Obs – 10 months IFN – 13.3 months Obs – 18.8 months IFN – 13 months Obs – 16 months IFN – 8.9 months Obs – 9.9 months
There was no significant difference between the IFN and observation arms in any of the above studies. CR, complete response; OR, objective response.
Table 20.22 Examples of the effect of alternating chemotherapy regimens on response rate and survival in SCLC
Fukuoka et al. 1991143 Roth et al. 1992144
Overall Response Rate (%)
Median survival (months)
CAV
PE
CAV/PE
CAV
PE
CAV/PE
No patients
55 51
78 61
76 59
9.9 8.3
9.9 8.6
11.8 8.1
300 437
Extent of disease LS ES ES
CAV, cyclophosphamide/doxorubicin/vincristine; PE, cisplatin/etoposide
DOSE INTENSIFICATION
Dose escalation can overcome cytotoxic drug resistance and increase cures in animal cancer models. A variety of methods have been used to increase cytotoxic dose intensity, including increased doses, shorter treatment intervals, haemopoietic growth factor support and high-dose therapy with haemopoietic progenitor cell transplantation. Suboptimal chemotherapy doses result in inferior survival, but it is uncertain how much survival in SCLC can be improved by increasing cytotoxic dose intensity. Early studies were underpowered, rarely assessed the delivered dose intensity and often compared dissimilar regimens. The results were contradictory. A meta-analysis of 60 trials, using a variety of regimens, concluded that there was no statistically significant improvement in response rate or survival in the dose-intense arms.161*** However, a French study involving 105 SCLC patients with limited-stage disease reported a significant survival benefit (2-year survival 43 per cent versus 26 per cent, p 0.02) for those randomized to receive higher doses of cisplatin and cyclophosphamide for the first dose only.162** A study in elderly patients with SCLC randomized patients to receive an attenuated dose or standard dose of cisplatin/etoposide. Response rates and 1-year survival were lower in the attenuated-dose arm (response rate 39 per cent versus 69 per cent, 1-year survival 18 per cent versus 39 per cent).163
ACCELERATED CHEMOTHERAPY
Shortening the interval between cycles of chemotherapy might reduce the opportunity for tumour cells to re-grow and mutate to drug-resistant clones. Typically, myelosuppressive drugs are alternated weekly with non-myelosuppressive agents. Randomized trials, however, have shown no advantage for weekly chemotherapy over standard treatment at 3-week intervals. A randomized trial alternated cisplatin and etoposide with ifosfamide and doxorubicin for 12 consecutive weeks, which was compared to a standard 3weekly regimen of alternating CAV/PE in good performance status, limited-stage and extensive-stage patients. Response rates, median and 2-year survivals were not significantly different.164 A similar study conducted by the European Lung Cancer Working Party has shown no survival advantage in patients treated with weekly multiple-drug chemotherapy as against standard combination treatment with ACE.165 The Canadian weekly regimen, CODE (cisplatin, vincristine, doxorubicin and etoposide), was compared with CAV/PE in a randomized study in 220 patients with extensive-stage SCLC.166 The dose intensity of CODE was two-fold that of CAV/PE and led to higher response rates (87 per cent versus 70 per cent) but was associated with more toxic deaths. There was no significant difference in progression-free or overall survival.
Treatment of small-cell lung cancer 525
Table 20.23 Effect of dose intensification with or without haemopoietic growth factors in the treatment of SCLC Study
Regimen
RDI
Arriagada et al. 1993162
PCAE n 105
Woll et al. 1995170 Trillet-Lenoir et al. 1996171 Wolf et al. 1996172 Woll et al. 1996173 Steward et al. 1998174 Thatcher et al. 2000175 Sculier et al. 2001176 Ardizzoni et al. 2002177
VICE n 65 AVI n 54 AIO/PE n 330 ICE 2 vs 4 wk n 50 VICE 3 vs 4 wk n 301 ACE 2 vs 3 wk n 403 EVI 2 vs 3 wks CDE 2 vs 3 wks
P 100 vs 80 mg/m2 C 300 vs 225 mg/m2 1.34 vs 1.17
Response Rate %
Median survival
2 year survival %
87 vs 70
–
43 vs 26
93.5 vs 94.1
69 vs 65 weeks
32 vs 15
1.76 higher than 3 weekly –
77
8 months
22
–
1.8 vs 0.99
80 vs 76
17.1 vs 15.1 months –
34 both groups –
26% higher
83 vs 84
443 vs 351 days
33 vs 18*
34% higher
78 vs 79
–
13 vs 8 *
– 70% higher
76 vs 59 84 vs 79
264 vs 286 days 52 vs 54 wks
6 vs 5 18 vs 15
(RD, relative dose intensity; PCAE, cisplatin/cyclophosphamide/doxorubicin/etoposide; AVI, doxorubicin/etoposide/ifosfamide; AIO, doxorubicin/ifosfamide/vincristine; CDE, cyclophosphamide/doxorubicin/etoposide; ACE, doxorubicin/cyclophosphamide/etoposide; ICE, ifosfamide/carboplatin/etoposide; VICE, vincristine/ifosfamide/cyclophosphamide/etoposide; PE, cisplatin/etoposide; EVI, epirubicin/vindesine/ifosfamide * p 0.05).
HAEMOPOIETIC GROWTH FACTORS
Granulocyte colony-stimulating factor (G-CSF, filgrastin, lenograstin) reduces the depth and duration of neutropenia during standard chemotherapy for SCLC, leading to fewer infections and a higher proportion of patients receiving the planned dose on time.167,168 Unexpectedly, granulocyte/ macrophage colony-stimulating factor (GM-CSF) was associated with more thrombocytopenia, toxic deaths, days in hospital and use of blood products and intravenous antibiotics in a South West Oncology Group (SWOG) study.169 Several groups have tested the use of haemopoeitic growth factors to increase cytotoxic dose intensity in SCLC. Modest increases have been achieved, and two recent studies have reported improved survival in the dose-intensified arm (Table 20.23). The evidence is not sufficient for dose intensification to be routinely used, but should reinforce the importance of trying to minimize dose reductions and delays. HIGH-DOSE THERAPY
The success of high-dose therapy with haemopoietic progenitor cell support in leukaemia and lymphoma gave impetus to using a similar approach in chemo-responsive solid tumours such as SCLC. High-dose chemotherapy has been administered as primary treatment or following
tumour remission with conventional induction treatment. This requires hospitalization and the re-infusion of previously stored autologous bone marrow or peripheral blood progenitor cells to abrogate the severe and potentially lethal bone marrow suppression. The most popular regimen for late intensification has been ICE. Pasini et al.178*** reviewed the data from 505 patients in 26 studies. Eight studies used early intensification, and complete response rates of 51 per cent were seen. In the late intensification studies, only 30 per cent of the population receiving induction chemotherapy were actually suitable for the procedure because of toxicity from the induction regimen, lack of initial response or inability to withstand the late intensification chemotherapy with its severe toxicity. Within this group of patients, 39 per cent achieved complete response, 2-year survival rates were between 30 per cent and 50 per cent, and the toxic death rate was 7 per cent. An alternative approach is to use haemopoietic progenitor cells and G-CSF to support accelerated multi-cyclic chemotherapy.179 Lorigan et al.180** randomized 318 patients with good-prognosis SCLC to either standard 4-weekly ICE or 2-weekly ICE. In the 2-weekly arm, patients received G-CSF (day 4–14) and had autologous blood collected just prior to cycle 2–6, which was returned 24 hours after chemotherapy. Despite achieving almost twice the dose intensity, there was significantly less haematological toxicity in the 2-weekly arm, but response rates and overall survival were not improved.
526 Bronchus
CHEMOTHERAPY FOLLOWING RELAPSE
Despite dramatic responses to first-line therapy, the majority of SCLC patients relapse with chemo-resistant disease. In general, patients who develop disease progression more than 3 months from the completion of first-line therapy obtain the best results with second-line chemotherapy. Although second responses can be obtained by re-introducing the firstline regimen, particularly in patients with remissions of more than 6 months, the use of drugs that have not previously been used results in improved responses. The use of the PE regimen following CAV has been shown to produce response rates of 50 per cent. However, the use of CAV second-line following treatment failure with PE has produced response rates of only 13–28 per cent.29 Phase II studies of topotecan in the second-line setting have reported response rates of 11 per cent in chemo-refractory patients181 and up to 38 per cent in chemo-sensitive patients.182 Von Pawel et al.183** reported a randomized trial comparing the efficacy of topotecan and CAV in 211 patients with relapsed SCLC. No significant differences in response rates (24.3 per cent topotecan versus 18.3 per cent CAV), time to progression (13.3 weeks versus 12.3 weeks) or median survival (25.0 weeks versus 24.7 weeks) were seen between the two arms. There did, however, appear to be significant improvement in certain disease-related symptoms in the topotecan arm. Topotecan can be given orally, and a phase II study comparing oral and intravenous administration184 demonstrated improved response rates and median survival in the oral arm, with less neutropenia and increased ease of administration. O’Brien et al.185 randomized 141 patients with relapsed SCLC to receive oral topotecan or best supportive care and demonstrated that topotecan extended survival (MS 26 weeks versus 14 weeks) and improved quality of life. NEW DRUG REGIMENS AND NOVEL AGENTS
New drugs A number of newer agents have recently been investigated in the treatment of relapsed or extensive SCLC (Table 20.24). These new agents may play an important role both in first-line therapy and also in the treatment of recurrent disease, which is resistant to standard agents. A phase III study196** comparing irinotecan and cisplatin (IP) with etoposide and cisplatin (EP) in extensive-stage disease was closed early (154 patients) because interim analysis demonstrated significantly improved median survival in the IP arm (12.8 months versus 9.4 months, p 0.002). Two-year survival was also increased, from 5.2 per cent in the EP arm to 19.5 per cent in the IP arm. Toxicity was comparable, with increased diarrhoea in the IP arm and increased myelotoxocity in the EP arm. However, these results have not been supported by a confirmatory phase III study conducted in the USA (9.3 months in the IP arm and 10.2 months in the EP arm (p 0.74));197 331 patients were randomized and no significant difference was seen in
response rates (52 versus 51 per cent), median or 1-year survival (35 per cent versus 36.1 per cent). Gemcitabine in combination with other agents has been investigated as first-line therapy in patients with poorprognosis SCLC and in patients with recurrent disease. In a phase II study in 69 patients with extensive-stage SCLC, carboplatin and gemcitabine was well tolerated, with reported response rates of 42.5 per cent and a median survival of 9.2 months.198 De Marinis et al.199 compared cisplatin and gemcitabine with cisplatin, etoposide and gemcitabine. Response rates were increased using the three agents (63 versus 57 per cent), but median survival was not improved and there was greater toxicity. Two phase II studies using singleagent gemcitabine in the second-line setting have shown modest activity.195,200 Vinorelbine has also been investigated in first and second line in combination with other agents, but the regimens have been poorly tolerated.201,202 Likewise, regimens incorporating paclitaxel have shown increased toxicity.203 The non-platinum-containing regimen docetaxel/gemcitabine was well tolerated in elderly patients with extensive-stage SCLC, but response rates were disappointing (23 per cent).204 Novel agents A number of novel agents have recently been investigated in the management of SCLC and include immunotherapy, molecular targeting and anti-angiogenic agents. Bec2 is an anti-idiotypic antibody that mimics the ganglioside GD3, which is expressed on the surface of SCLC cells. In a phase III trial in 515 limited-stage SCLC patients who had a good response to primary chemotherapy, Giaccone et al.205** randomized patients to receive five vaccinations of Bec2/ Bacillus Calmette–Guérin (BCG) vaccine or observation only. No improvement in survival was seen in the vaccinated arm, although a trend to improved survival was seen in the third of patients who developed a humoral response. Other immunotherapeutic approaches have been investigated; Harper-Wynne et al.206 used the immunological adjuvant Mycobacterium vaccae in addition to chemotherapy in patients with SCLC. No significant differences in overall response rate or survival were seen, although patients receiving M. vaccae did have a significant improvement in symptomatic response. A number of potential molecular targets have been identified and include c-kit and CD56. Imatinib (Glivec) inhibits c-kit tyrosine kinase activity. Two phase II studies using imatinib in patients with recurrent SCLC that express the c-kit protein have shown no evidence of anti-tumour activity.207,208 Studies of agents targeting CD56 are in progress. Matrix metalloproteinases (MMPs) are proteins that are able to digest the extracellular matrix and basement membrane. High levels of expression are found in tumours and the surrounding stromal tissue. In SCLC, high levels of expression are associated with poor prognosis. In order to investigate whether MMP inhibitors could prolong survival
Treatment of small-cell lung cancer 527
Table 20.24 Phase II trials of a number of new agents in the treatment of SCLC. Sensitive disease relapse 3 months following first line chemotherapy, Resistant disease progression within 3 months of first line chemotherapy. Study Smit et al. 1998186 n 24 Groen et al. 1999187 n 35 Deppermann et al. 1999188 n 75 Lyss et al. 1999189 n 34 Neubauer et al. 2004190 n 77 Ardizzoni et al. 1997191 n 101 Kudoh et al. 1998192 n 75 Nakamura et al. 1999193 n 51 De Vore et al. 1998194 n 44 Masters et al. 2003195 N 46
Drug (dose mg/m2)
Patient group
ORR (%)
Median survival
Paclitaxel 175
Resistant
29
100 days
Paclitaxel 175 Carboplatin AUC7 Paclitaxel 200 Carboplatin AUC 6 Paclitaxel 230 Cisplatin 75 G-CSF Paclitaxel 80 Carboplatin AUC 2-weekly Topotecan
Resistant
73.50
31 weeks
Chemonaive extensive disease
61
359 days
Chemonaive extensive disease
71
7.6 months
38
7.2 months
Irinotecan cisplatin
Chemonaive ED, PS2, 70 years old Resistant Sensitive Chemonaive limited extensive
6.4 37.8 Overall 84
Irinotecan etoposide
Chemonaive extensive disease
66
4.7 months 6.9 months 14.3 months 13 months 12 months
Irinotecan
Sensitive Resistant Refractory Sensitive
35.3 3.7 5.6 16.7
5.9 months 2.8 months 6.9 months 7.3 months
Gemcitabine
in SCLC patients who had achieved a partial response or better following primary chemotherapy, Shepherd et al.209** randomized 532 patients to receive the oral MMP inhibitor marimastat or placebo for up to 2 years. The use of marimastat did not improve either time to progression or median survival, but did impair quality of life due to musculoskeletal toxicity.
Table 20.25 Three year survival figures following single or combined modality treatment for SCLC (from Pignon et al.210).
Combined-modality treatment for SCLC
CT, chemotherapy; RT, radiotherapy
THORACIC RADIOTHERAPY AND CHEMOTHERAPY
Despite improvements in the survival of SCLC patients with the widespread use of chemotherapy, 30–80 per cent of patients develop local recurrence of their disease. The use of radiotherapy in addition to chemotherapy has therefore been investigated. A meta-analysis of 13 randomized trials comparing chemotherapy with chemotherapy plus radiotherapy, including 2103 patients with limited disease, clearly demonstrated a significant survival advantage.210*** Patients receiving combined-modality therapy had a 14 per cent reduction in risk of death, with overall survival at 3 years being 14.3 per cent for combined modality versus 8.9 per cent with chemotherapy alone. The greatest benefits were seen in patients under the age of 55 years, and patients over the age of 70 years suffered from greater pulmonary toxicity (Table 20.25). A second meta-analysis211***
3 year survival %
55 y 70 y
CT
CT ⫹ Thoracic RT
9.2 10.2
17.4 8.7
demonstrated a 25 per cent absolute increase in intrathoracic tumour control, with a 5.4 per cent absolute improvement in 2-year survival. In a randomized controlled study in 210 patients with extensive-stage SCLC who had achieved a complete response at distant sites and at least a partial response within the chest, the addition of hyperfractionated radiotherapy improved 5-year survival from 3.7 per cent to 9.1 per cent.212** Combined-modality treatment clearly improves local control of the tumour, but the optimal timing for the radiotherapy remains unclear. Of six randomized trials investigating the effect of early or late radiotherapy, two have shown a benefit of ‘early’ radiotherapy (with first or second cycle of chemotherapy)213,214** and two have shown a trend of improvement in survival with early radiotherapy.215,216** In the Canadian study, patients who received radiotherapy with the second cycle of platinum-based
528 Bronchus
chemotherapy had a median survival of 21.2 months, compared with only 16 months for patients treated with radiotherapy with the final cycle. More importantly, the probability of long-term survival and cure was enhanced with early thoracic irradiation, with 5-year survival rates of 20 per cent compared with 11 per cent.213 Takada et al.216 randomized 231 patients to receive radiotherapy starting on day 2 of platinum-based chemotherapy or after the fourth cycle. Although survival was not significantly increased, there was a trend, with 5-year survival rates of 23.7 per cent in the concurrent arm compared with 18.3 per cent in the late arm. Two recent meta-analyses have been performed. Freid et al.217*** included seven RCTs involving 1524 patients and concluded that there was a small but significant improved in 2-year survival in patients receiving early radiotherapy. These benefits were increased in the studies using platinum-based chemotherapy and hyperfractionated radiotherapy. However, a Cochrane Review218*** found no significant improvement in 2-year survival in patients who received radiotherapy within 30 days of starting chemotherapy. It is currently recommended in the UK that patients with limited disease and good performance status who have achieved a good response to induction chemotherapy should be offered consolidation thoracic radiotherapy. The optimal dose of radiotherapy is also uncertain. A single randomized study219 has shown that higher doses (37.5 Gy versus 25 Gy) resulted in better control, and the majority of recent studies have used 45–50 Gy in 2-Gy fractions. Hyperfractionation of radiotherapy has also been investigated,220 but the incidence of severe oesophagitis was significantly increased, with no difference in survival at 18 months. However, in the Turrisi study,221** 417 patients were randomized to receive 45 Gy in twice-daily fractions or in once-daily fractions concurrently with cisplatin and etoposide. Those patients receiving hyperfractionated radiotherapy had improved median survival (23 months versus 19 months, p 0.04), although oesophagitis was more common. PROPHYLACTIC CRANIAL IRRADIATION
Up to 50 per cent of SCLC patients achieving remission develop brain metastases within 2 years. Although palliative radiotherapy may improve symptoms in a proportion of patients, the effect of metastases can be devastating for both the patients and their carers. A number of early clin-ical trials investigating the effect of prophylactic cranial irradiation (PCI) demonstrated reduction in the incidence of metastases but no survival benefit. A meta-analysis of 987 patients in seven randomized controlled trials comparing PCI with no PCI in patients with complete response following induction chemotherapy has been published.222*** Overall survival was significantly increased in the patients receiving PCI, with a 16 per cent reduction in risk of death and a 5.4 per cent increase in survival at 3 years (Table 20.26). The risk of brain metastases was reduced by 54 per cent. There was no significant difference between doses of radiation received, although
Table 20.26 Adapted from (Auperin et al.222). Three year survival and incidence of brain metastases in SCLC patients in complete remission treated with or without prophylactic cranial irradiation (PCI)
Overall 3 years survival (%) Incidence of brain metastases at 3 years (%)
PCI
No PCI
Significance
20.7
15.3
p 0.01
33.3
58.6
p 0.001
there was a trend towards better survival in patients receiving higher radiation doses. The majority of patients in this study had limited-stage disease (86 per cent). However an EORTC study222a has confirmed benefits in patients with extensive stage disease with a reduction in incidence in symptomatic brain metastases in patients receiving PCI (15% versus 40%). Overall survival was also increased form 5.4 to 6.7 months. One of the major concerns about the use of PCI is the possibility of long-term neurotoxicity. Such toxicity includes memory loss, tremors, somnolence, ataxia and cortical atrophy. Evaluation of the impact of therapy on 64 patients who were in remission for at least 2 years showed that the majority of patients had a significant degree of cognitive dysfunction, demonstrated by neuropsychometric testing, although no pre-treatment assessment was available for comparison.223 A randomized controlled study run by the EORTC and UKCCCR224** included cognitive function and quality of life as endpoints. This study showed that there was cognitive impairment before treatment and increased impairment at both 6 and 12 months after treatment, although there was no consistent difference between those who received PCI and those who did not. SURGICAL ADJUVANT THERAPY
Surgical resection is generally contraindicated in SCLC, as most patients have bulky central and metastatic disease, either evident or occult at the time of presentation. Early studies comparing surgery and radiotherapy showed no survival advantage in patients undergoing surgical resection. However, the value of surgery in SCLC has recently been re-evaluated. In one of the largest reviews from the University of Toronto Lung Oncology Group, 79 SCLC patients underwent surgery as their first treatment and most then received CAV-based postoperative chemotherapy. Overall, the median survival for the entire group was 26 months, with a 5-year survival of 39 per cent. Five-year survival in a small group of these patients with stages I and II disease was 51 per cent.225 The investigators still felt that surgery should not be employed initially other than in the rare situation of a small peripherally located pulmonary lesion or in patients in whom SCLC was found unexpectedly at thoracotomy and a complete excision was possible. In a
References 529
prospective randomized phase III study in which patients who achieved an objective response following induction chemotherapy with CAV were randomized to undergo surgical resection or observation only,226** all patients received thoracic and cranial radiotherapy. Surgical resection did not influence the pattern of relapse and there was a trend towards improved survival in the non-surgical arm. Two other observational studies have suggested a benefit in survival, but these involved small numbers of highly selected patients.227 Rarely, salvage surgery may be beneficial to patients with mixed SCLC and NSCLC histology who relapse within the chest or to patients who at relapse are found to have purely NSCLC histology following chemotherapy for SCLC.
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
Lung cancer is the leading cause of cancer deaths, accounting for 28 per cent of male and 20 per cent of female deaths in the UK. The major pathological types in the UK are squamous-cell carcinoma (40 per cent), adenocarcinoma (20 per cent) and small-cell lung cancer (20 per cent). Surgery is the treatment of choice for stages I and II and some stage IIIA NSCLC patients. Adjuvant chemotherapy using cisplatin and vinorelbine improves survival in patients with stages II and IIIA disease. Radical radiotherapy is possible in a small percentage of patients with NSCLC, and hyperfractionated regimens improve survival further. Chemotherapy for stages IIIB and IV NSCLC, using platinum agents, increases 1-year survival by 10 per cent and can improve quality of life. Epithelial growth factor receptor inhibitors play a role in second-line therapy, with response rates similar to those for docetaxel. Chemotherapy is the treatment of choice for SCLC. Response rates are high and survival is extended from a median of 2 months to 18 months. In limited-stage SCLC, the use of concurrent chemo-radiotherapy and prophylactic cranial irradiation is beneficial.
REFERENCES 1 http://info.cancerresearchuk.org/cancerstats 2 Jemal A, Murray T, Ward E, Samuels A, et al. Cancer Statistics, 2005. CA Cancer J Clin 2005; 55:10–30. ●3 Doll R, Hill AB. Mortality in relation to smoking: Ten years’ observations of British doctors. British Medical Journal 1964;1:1399–1410.
4 http://info.cancerresearchuk.org/cancerstats/types/lung/ smoking/ 5 Mao L, Lee JS, Kurie JM, et al. Clonal genetic alterations in the lungs of current and former smokers. Journal of the National Cancer Institute 1997; 89:857–862. 6 Wiencke JK, Thurston SW, Kelsey KT, et al. Early age at smoking initiation and tobacco carcinogen DNA damage in the lung. Journal of the National Cancer Institiute 1999; 91:614–619. 7 Zang EA, Wynder EL. Differences in lung cancer risk between men and women: Examination of the evidence. Journal of the National Cancer Institute 1996; 88: 183–192. 8 Hackshaw AK. Lung cancer and passive smoking. Stat Methods Med Res 1998; 7:119–36. 9 Roland M, Rudd RM. Somatic mutations in the development of lung cancer. Thorax 1998; 53:979–983. 10 Salgia R, Skarvin AT. Molecular abnormalities in lung cancer. Journal of Clinical Oncology 1998; 16:1207–1217. 11 Thiberville L, Payne P, Vielkinds J, LeRiche J, et al. Evidence of cumulative gene losses with progression of premalignant epithelial lesions to carcinoma of the bronchus. Cancer Research 1995; 55:5133–5139. 12 Petersen S, Wolf G, Bockmuhl U, Gerllert K, et al. Allelic loss on chromosome 10q in human lung cancer: association with tumour progression and metastatic phenotype. British Journal of Cancer 1998; 77:270–276. ●13 Mao L, Lee JS, Kurie JM, Fan YH, et al. Clonal genetic alterations in the lungs of current and former smokers. Journal of the National Cancer Institute 1997; 89: 857–862. 14 Alpha-Tocopherol Beta Carotene Group. The effect of Vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. New England Journal of Medicine 1994; 330:1029–1035. 15 Omenn G, Goodman G, Thornquist M. Effects of a combination of beta-carotene and vitamin A on lung cancer and cardiovascular disease. New England Journal of Medicine 1996; 334:1150–1155. 16 Albanes D. b-carotene and lung cancer: a case study. American Journal of Nutrition 1999; 69 (Suppl): 1345S–1350S. 17 Rizvi N, Hayes DF. A breathalyser for lung cancer. Lancet 1999; 353:1897–1898. 18 Smith IE. Screening for lung cancer: time to think positive. Lancet 1999; 354:86–87. 19 Tockman MS, Mulshine JL, Piantadosi S, et al. Prospective detection of preclinical lung cancer: results form two studies of hnRNP overexpression. Clinical Cancer Research 1997; 3:2237–2246. 20 Payne PW, Sebo TJ, Doudkine A, et al. Sputum screening by quantitative microscopy: A reexamination of a portion of the National Cancer Institute Cooperative Early Lung Cancer Study. Mayo Clinic Proceedings 1997; 72:697–704. 21 Phillips M, Gleeson K, Hughes JMB, et al. Volatile organic compounds in breath as markers of lung cancer: a cross-sectional study. Lancet 1999; 353:1930–1933.
530 Bronchus
◆22
●23
●24
25
●26
27
◆28
◆29
30
◆31
32
33
34
35
❉36
◆37
Manser R, Irving L, Stone C. Screening for lung cancer. The Cochrane Database of Systematic Reviews 2006 (1). Sone S, Takashima S, Li F, et al. Mass screening for lung cancer with mobile spiral computed tomography scanner. Lancet 1998; 351:1242–1245. Henschke CI, Mccauley DI, Yankelevitz DF, et al. Early Lung Cancer Project: overall design and findings from baseline screening. Lancet 1999; 354:99–105. Buccheri G, Ferrigno D. Lung cancer: clinical presentation and specialist referral time. European Respiratory Journal 2004; 24:898–904. Beckles M, Spiro S, Colice G, Rudd R. Initial evaluation of the patient with lung cancer: symptoms, signs, laboratory tests and paraneoplastic syndromes. Chest 2003; 123:97–104. McManus M, Hicks R, Fisher R, et al. FDG-PET detected extracranial metastasis in patients with non-small cell lung cancer undergoing staging for surgery or radical radiotherapy – survival correlates with metastatic burden. Acta Oncologica 2003; 42:48–54. Mountain CF. Revisions in the International system for staging lung cancer. Chest 1997; 111:1710–1717. Gillenwater H, Socinski M. Extensive stage small cell lung cancer. In: Detterbeck F, Riveria M, Socinski M, Rosenman J, editors. Diagnosis and treatment of lung cancer: An evidence based guide for the practicing clinician. Philidelphia: WB Saunders; 2001: 360–375. Albain KS, Crowley JJ, LeBlanc M, Livingston RB. Survival determinants in extensive-stage non-small cell lung cancer: The Southwest Oncology Group Experience. Journal of Clinical Oncology 1991; 9:1618–1626. Verbeken E, Brambilla E. WHO classification of lung and pleural tumours. The WHO/IASLC 1999 revision. European Respiratory Reviews 2002; 12:172–176. Mooi WJ. Common Lung Cancers. In: Hasleton PS, editor. Spencer’s pathology of the Lung. 5th ed. New York: McGraw – Hill; 1996:1009–1064. Kaufmann O, Dietel M. Thyroid transcription factor-1 is the superior immunohistochemical marker for pulmonary adenocarcinomas and large cell carcinomas compared to surfactant proteins A and B. Histopathology 2000; 36: 8–16. Srodon M, Westra W. Immunohistochemical staining for thryoid transcription factor-1: a helpful aid in discerning primary site of tumor origin in patients with brain metastases. Human Pathology 2002; 33:642–645. Abeloff MB, Eggleston JC, Mendelsohn G, et al. Changes in morphological and biochemical characteristics of SCLC. A clinicopathological study. American Journal of Medicine 1979; 66:757–764. British Thoracic Society and Society of Cardiovascular Surgeons of Great Britain and Ireland Working Party. Guidelines for the selection of patients with lung cancer for surgery. Thorax 2001; 56:89–108. Jones D, Detterbeck F. Surgery for stage I non-small cell lung cancer. In: Detterbeck F, Riveria M, Socinski M, Rosenman J, editors. Diagnosis and treatment of lung
38
39
40
❉41
42
43
44
45
46
47
48
●49
●50
51
●52
cancer: an evidence based guide for the practicing clinician. Philidelphia: WB Saunders; 2001: 177–90. Ginsberg RJ, Rubenstein LV. Randomised trial of lobectomy versus limited resection for T1 N0 NSCLC. Annals of Thoracic Surgery 1995; 60:615–623. Kirby T, Mack M, Landreneau R, Rice T. Lobectomy – video assisted thoracic surgery versus muscle-sparing thoracotomy. A randomized trial. Journal of Thoracic and Cardiovascular Surgery 1995; 109:997–1001. Guidicelli R, Thomas P, Lonjon T, et al. Video-assisted mini-thoracotomy versus muscle-sparing thoracotomy for performing lobectomy. Annals of Thoracic Surgery 1994; 58:712–717. National Institute of Clinical Excellence. The diagnosis and treatment of lung cancer; 2005. Martin-Ucar A, Chaudhuri N, Edwards J, Waller D. Can pneumonectomy for non-small cell lung cancer be avoided? An audit of parenchymal sparing lung surgery. European Journal of Cardiothoracic Surgery 2002 21:601–605. Dales RE, Belanger R, Shamji FM, et al. Quality of life following thoractomy for lung cancer. Journal of Clinical Epidemiology 1994; 47:1443–1449. Talton BM, Constable WC, Kersch CR. Curative radiotherapy in non-small cell carcinoma of the lung. Int Radiat Oncol Biol Phys 1990; 19:15–21. Rosenthal SA, Curran WJ, Herbert SH, et al. Clinical stage II non-small cell lung cancer treated with radiation therapy alone. Chest 1992; 70:2410–2417. Dosoretz DE, Katin MJ, Blitzer PH. Radiation therapy in the management of medically inoperable carcinoma of the lung: results and implications for future treatment stategies. Int J Radiat Oncol Biol Phys 1992; 24:3–9. Graham PH, Gebski VJ, Langlands AO. Radical radiotherapy for early non-small cell lung cancer. Int J Radiat Oncol Biol Phys 1995; 31:261–266. Morita K, Fuwa N, Suzuki Y, et al. Radical radiotherapy for medically inoperable non-small cell lung cancer in clinical stage I: a retrospective analysis of 149 patients. Radiother Oncol 1997; 42:31–36. Saunders M, Dische S, Barrett A, et al. Continuous hyperfractionated accelerated radiotherapy (CHART) versus conventional radiotherapy in non-small cell lung cancer: a randomised multicentre trial. Lancet 1997; 350: 161–165. Saunders M, Dische S, Barrett A, et al. Continuous, hyperfractionated, acclerated radiotherapy (CHART) versus conventional radiotherapy in non-small cell lung cancer: mature data from the randomised multicentre trial. CHART Steering committee. Radiotherapy Oncology 1999; 52:137–148. Mehta MP, Tannehill SP, Adak S, et al. Phase II trial of hyperfractionated accelerated radiation therapy for nonresectable NSCLC: results of Easterm Cooperative Oncology Group 4593. Journal of Clinical Oncology 1998; 16:3518–3523. Saunders M, Rojas A, Lynn BE, et al. Experience with dose esclation using CHARTWEL continuous hyperfractionated
References 531
◆53
◆54
55
56
57
58
59
●60
61
62
●63
●64
accelerated radiotherapy (weekend less) in NSCLC. British Journal of Cancer 1998; 78:1323–1328. Detterbeck F, Socinski M. Induction therapy and surgery for stage I-IIA, B non-small cell lung cancer. In: Detterbeck F, Riveria M, Socinski M, Rosenman J, editors. Diagnosis and treatment of lung cancer: An evidence-based guide for the practicing clinician. Philidelphia: WB Saunders; 2001: 267–82. PORT Meta-analysis Trialists Group. Postoperative radiotherapy for non-small cell lung cancer. Cochrane Database Systematic Review 2005 (2). Girling D, Falk S, White R, et al. Immediate versus delayed thoracic radiotherapy (TRT) in patients with unresectable, locally advanced non-small cell lung cancer (NSCLC) and minimal symptoms: Results of an MRC/BTS randomised trial. Lung Cancer 2000; 29 (Suppl 1): 164. Medical Research Council Lung Cancer Working Party. Inoperable non-small cell lung cancer (NSCLC): A Medical Research Council randomised trial of palliative radiotherapy with two fractions or ten fractions. British Journal of Cancer 1991; 63:265–270. Medical Research Council Lung Cancer Working Party. A Medical Research Council randomised trial of palliative radiotherapy with two fractions or a single fraction in patients with inoperable non-small cell lung cancer and poor performance status. British Journal of Cancer 1992; 65:934–941. Medical Research Council Lung Cancer Working Party. Randomised trial of palliative two fraction versus more intensive 13-fraction radiotherapy for patients with inoperable non-small cell lung cancer. Clinical Oncology 1996; 8:167–175. Rees GJG, Devrell CE, Barley VI, Newman HFV. Palliative radiotherapy for lung cancer: two versus five fractions. Clinical Oncology 1997; 9:90–95. Bezjak A, Dixon P, Brundage M, et al. Randomized phase III trial of single versus fractionated thoracic radiation in the palliation of patients with lung cancer. Int J Radiat Oncol Biol Phy 2002; 54:719–728. Langendijk H, de Jong T, Tjwa M, et al. External irradiation versus external irradiation plus endobronchial brachytherapy in inoperable non-small cell lung cancer: a prospective randomized study. Radiotherapy Oncology 2001; 58:257–268. Stout R, Barber P, Burt P, et al. Clinical and quality of life outcomes in the first United Kingdom randomized trial of endobronchial brachytherapy (intraluminal radiotherapy) vs. external beam radiotherapy in the palliative treatment of inoperable non-small cell lung cancer. Radiotherapy Oncology 2000; 56:323–327. Rosell R, Gomez-Codina J, Camps C, et al. Pre-resectional chemotherapy in stage IIIA non-small cell lung cancer: a 7year assessment of a randomized controlled trial. Lung Cancer 1999; 47:7–14. Roth J, Atkinson E, Fossella F. Long term follow-up of patients enrolled in a randomized trial comparing perioperative chemotherapy and surgery with surgery alone
●65
66
67
68
◆69
70
●71
●72
●73
74
●75
76
77
in resectable stage IIIA non-small cell lung cancer. Lung Cancer 1998; 21:1–6. Depierre A, Milleron B, Moro-Sibilot D, et al. Preoperative chemotherapy followed by surgery compared with primary surgery in resectable stage I (except T1N0), II, and IIIA nonsmall-cell lung cancer. J Clin Oncol 2002; 20:247–253. Rosell R, Gomez-Codina J, Camps C, et al. A randomized trial comparing preoperative chemotherapy plus surgery with surgery alone in patients with NSCLC. New England Journal of Medicine 1994; 330:153–158. Roth JA, Fossella F, Komaki R, et al. A randomized trial comparing perioperative chemotherapy and surgery with surgery alone in resectable stage IIIA NSCLC. Journal of the National Cancer Institute 1994; 86:673–680 Gilligan D, Nicolson M, Smith I, et al. Preoperative Chemotherapy in patients with resectable non-small cell lung cancer: results of the MRC LU22/NVALT2/EORTC 08012 multicentre randomised trial and update of systematic review. Lancet 2007; 369:1929-1937. Non-small cell lung cancer collaborative group. Chemotherapy in non-small cell lung cancer: a metaanalysis using updated data on individual patients from 52 randomised trials. British Medical Journal 1995; 311:899–909. Mineo T, Ambrogi V, Corsaro V, Roselli M. Postoperative adjuvant therapy for stage IB non-small cell lung cancer. European Journal of Cardiothoracic Surgery 2001; 20:378–384. Scagliotti G. The ALPI trial: the Italian/European experience with adjuvant chemotherapy in resectable nonsmall cell lung cancer. Clinical Cancer Research 2005; 11:5011s–5016s. Arriagada R, Bergman B, Dunant A, et al. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small cell lung cancer. New England Journal of Medicine 2004; 350:351–360. Kato H, Ichinose Y, Ohta M, et al. A randomized trial of adjuvant chemotherapy with uracil-tegafur for adenocarcinoma of the lung. New England Journal of Medicine 2004; 350:1713–1721. Waller D, Peake M, Stephens R, et al. Chemotherapy for patients with non-small cell lung cancer:the surgical setting of the Big Lung Trial. European Journal of Cardiothoracic Surgery 2004; 26:173–182. Winton T, Livingston R, Johnson D, et al. Vinorelbine plus cisplatin versus observation in resected non-small cell lung cancer. New England Journal of Medicine 2005; 352:2589–2597. Strauss G, Herndon J, Maddaus M, et al. Randomized clinical trial of adjuvant chemotherapy with paclitaxel and carboplatin following resection of stage IB non-small cell lung cancer: report of Cancer and Leukaemia Group B protocol 9633. Proc ASCO 2004: abstract 7019. Douillard JY, Rosell R, De Lena M, et al. Adjuvant vinorelbine plus cisplatin versus observation in patients with completely resected stage IB-IIIA non-small cell lung cancer (Adjuvant Navelbine International Trialist Association
532 Bronchus
◆78
●79
80
●81
82
83
❉84
●85
86
87
◆88
89
90
[ANITA]): a randomised controlled trial. Lancet Oncology 2006; 7:719–727. Sedrakyan A, Van der Meulen J, O’Byrne K, et al. Postoperative chemotherapy for non-small cell lung cancer: A systematic review and meta-analysis. Journal of Thoracic and Cardiovascular surgery 2004; 128:414–419. Dillman RO, Herndon J, Seagren SL, et al. Improved survival in stage III NSCLC: seven year follow-up of cancer and leukaemia group B (CALGB) 8433 trial. Journal of the National Cancer Institiute 1996; 88:1210–1215 Jeremic B, Shibamoto Y, Acimovic L, Djuric L. Randomised trial of hyperfractionated radiation therapy with or without concurrent chemotherapy for stage III non-small cell lung cancer. Journal of Clinical Oncology 1995; 13:452–458. Jeremic B, Shibamoto Y, Alimovic L, Milisavljevic S. Hyperfractionated radiation therapy with or without concurrent low dose daily carboplatin/etoposide for stage III non-small cell lung cancer: a randomized study. Journal of Clinical Oncology 1996; 14:1065–1070. Cullen M, Billingham L, Woodroffe C, et al. Mitomycin, ifosfamide and cisplatin in unresectable non-small cell lung cancer: effects on survival and quality of life. Journal of Clinical Oncology 1999; 17:3188–3194. Sause W, Kolesar P, Taylor S, et al. Final results of phase III trial in regionally advanced unresectable non-small cell lung cancer. Chest 2000; 117:358–364. Pfister D, Johnson D, Azzoli C, et al. American Society of Clinical Oncology Treatment of unresectable non-small cell lung cancer guideline: Update 2003. Journal of Clinical Oncology 2004; 22:330–353. Dillman RO, Seagren SL, Propert KL, et al. A randomised trial of induction chemotherapy plus high-dose radiation versus radiation alone in stage III non-small cell lung cancer. New England Journal of Medicine 1990; 323:940–945. Schaake-Koning C, van den Bogaert W, Dalesio O, et al. Effects of concomitant cisplatin and radiotherapy on inoperable non-small cell lung cancer. New England Journal of Medicine 1992; 326:524–530. Furuse K, Fukuoka M, Kawahara M, et al. Phase III study of concurrent versus sequential thoracic radiotherapy in combination with mitomycin, vindesine and cisplatin in unresectable stage III non-small cell lung cancer. Journal of Clinical Oncology 1999; 17:2692–2699. Rowell N, O’Rourke N. Concurrent chemoradiotherapy in non-small cell lung cancer. The Cochrane Database of Systematic Reviews 2006(1). Billingham LJ, Cullen MH, Woods J, et al. Mitomycin, ifosfamide and cisplatin (MIC) in non-small cell lung cancer (NSCLC): 3. Results of a randomised trial evaluating palliation and quality of life. Lung Cancer 1997; 18 (S1):9. Hickish TF, Smith IE, O’Brien MER, et al. Clinical benefit from palliative chemotherapy in non-small cell lung cancer extends to the elderly and those with poor prognostic factors. British Journal of Cancer 1998; 78:28–33.
91 Ellis PA, Smith IE, Hardy JR, et al. Symptom relief with MVP (mitomycin C, vinblastine and cisplatin) chemotherapy in advanced NSCLC. British Journal of Cancer 1995; 71:366–370. ◆92 Socinski M. Chemotherapy for stage IV non-small cell lung cancer. In: Detterbeck F, Riveria M, Socinski M, Rosenman J, editors. Diagnosis and treatment of lung cancer: An evidence-based guide for the practicing clinician. Philidelphia: WB Saunders; 2001:307–325. 93 Gridelli C, Ardizzoni A, Le Chevalier T, I. Treatment of advanced non-small cell lung cancer patients with ECOG performance status 2: results of an European Experts Panel. Annals of Oncology 2004; 15:419–426. 94 Perrone F, Di Maio M, Gallo C, Gridelli C. Outcome of patients with a performance status of 2 in the multicenter Italian lung cancer in the elderly study (MILES). Journal of Clinical Oncology 2004; 22:5018–5020. 95 Le Chevalier T, Brisgand D, Douillard J-Y, et al. Randomized study of vinorelbine and cisplatin versus vindesine and cisplatin versus vinorelbine alone in advanced non-small cell lung cancer: results of a European multicenter trial includig 612 patients. Journal of Clinical Oncology 1994; 12:360–367. 96 Wozniak A, Crowley J, Balcerzak S, et al. Randomised trial comparing cisplatin with cisplatin plus vinorelbine in the treatment of advanced non-small cell lung cancer: A Southwest Oncology Group study. Journal of Clinical Oncology 1998; 16:2459–2465. 97 Crino L, Conte P, De Marinis F, et al. A randomised study of gemcitabine cisplatin (GP) versus mitomycin, ifosfamide and cisplatin (MIC) in advanced non-small cell lung cancer (NSCLC). A multicenter phase III study. Proc ASCO. 1998; 17:455a. 98 Cardenal F, Lopez-Cabrerizo M, Anton A, et al. Randomised phase III study of gemcitabine-cisplatin versus etoposidecisplatin in the treatment of locally advanced or metastatic non-small cell lung cancer. Journal of Clinical Oncology 1999; 17:12–18. 99 Danson S, Middleton M, O’Byrne K, et al. Phase III trial of gemcitabine and carboplatin versus mitomycin, ifosfamide and cisplatin or mitomycin, vinblastine and cisplatin in patients with advanced non-small cell lung carcinoma. Cancer 2003; 98:542–553. 100 Rudd R, Gower N, Spiro S, et al. Gemcitabine plus carboplatin versus mitomycin, ifosfamide and cisplatin in patients with stage IIIB or IV non-small cell lung cancer: a phase III randomized study of the London Lung Cancer Group. Journal of Clinical Oncology 2005; 23:142–153. 101 Bonomi P, Kim K, Chang A, Johnson D. Phase III trial comparing etoposide cisplatin versus taxol with cisplatinG-CSF versus taxol- cisplatin in advanced non-small cell lung cancer. Proc ASCO1996; 15:382. 102 Giaccone G, Postmus P, Debruyne C, et al. Final results of an EORTC phase III study of paclitaxel versus tenoposide in combination with cisplatin, in advanced NSCLC. Proc ASCO 1997; 16:460a.
References 533
103 Gatzemeier U, von Pawel J, Gottfried M, et al. Phase III comparative study of high dose cisplatin versus a combination of paclitaxel and cisplatin in patients with advanced non-small cell lung cancer. Proc ASCO 1998; 17:454a. 104 von Pawel J, von Roemling R. Survival benefit from Tirazone (Tirapazamine) and cisplatin in advanced nonsmall cell lung cancer patients: Final results from the international phase III CATAPULT I trial. Proc ASCO 1998; 17:454a. 105 Rosell R, Gatzemeier U, Betticher D, et al. Phase III randomised trial comparing paclitaxel / carboplatin with paclitaxel / cisplatin in patients with advanced non-small cell lung cancer: a cooperative multinational trial. Annals of Oncology 2002; 13:1539–1549. ●106 Schiller J, Harrington D, Belani C, et al. Comparison of four chemotherapy regimens for advanced non-small cell lung cancer. New England Journal of Medicine 2002; 346:92–98. 107 Fossella F, Pereira J, von Pawel J, et al. Randomized multinational phase III study of docetaxel plus platinum combinations versus vinorelbine plus cisplatin for advanced non-small cell lung cancer: The TAX326 study group. Journal of Clinical Oncology 2003; 21:3016–3024. 108 Zatloukal P, Petruzelka L, Zemanova M, et al. Gemcitabine plus cisplatin vs. gemcitabine plus carboplatin in stage IIIB and IV non-small cell lung cancer: a phase III randomized trial. Lung Cancer 2003; 41:321–331. ◆109 Hotta K, Matsuo K, Ueoka H, et al. Role of adjuvant chemotherapy in patients with resected non-small cell lung cancer: reappraisal with meta-analysis of randomized controlled trials. Journal of Clinical Oncology 2004; 22:3860–3867. 110 Scagliotti G, De Marinis F, Rinaldi M, et al. Phase III randomized trial comparing three platinum-based doublets in advanced non-small cell lung cancer. Journal of Clinical Oncology 2002; 20:4285–4291. 111 Gridelli C, Gallo C, Shepherd F, et al. Gemcitabine plus vinorelbine compared with cisplatin plus vinorelbine or cisplatin plus gemcitabine for advanced non-small cell lung cancer: a phase III trial of the Italian GEMVIN investigators and the National Institute of Canada Clinical Trials Group. Journal of Clinical Oncology 2003; 21:3025–3034. 112 Tan E, Szczesna A, Krzakowski M, et al. Randomized trial of vinorelbine- gemcitabine versus vinorelbine – carboplatin in patients with advanced non-small cell lung cancer. Lung Cancer 2005; 49:233–240. 113 Kosmidis P, Mylonakis N, Nicolaides C, et al. Paclitaxel plus carboplatin versus gemcitabine plus paclitaxel in advanced non-small cell lung cancer: a phase III randomized trial. Journal of Clinical Oncology 2002; 20:3578–3585. 114 Smit E, Van Meerbeeck J, Lianes P, et al. Three-arm randomized study of two cisplatin-based regimens and paclitaxel plus gemcitabine in advanced non-small cell lung cancer: a phase III trial of the European Organization for Research and Treatment of Cancer Lung Cancer
115
116
117
118
●119
●120
●121
●122
123
◆124
●125
●126
group – EORTC 08975. Journal of Clinical Oncology 2003; 21:3909–3917. Georgoulias V, Samonis G, Papadakis E, et al. Comparison of docetaxel / cisplatin to docetaxel / gemcitabine ad first line treatment of advanced non-small cell lung cancer:early results of a randomised trial. Lung Cancer 2001; 34:S47–51. Pujol J, Breton J, Gervais R, et al. Gemcitabine-docetaxel versus cisplatin-vinorelbine in advanced or metastatic non-small-cell lung cancer: a phase III study addressing the case for cisplatin. Annals of Oncology 2005; 16:602–610. Barlesi F, Pujol J. Combination of chemotherapy without platinum compounds in the treatment of advanced nonsmall cell lung cancer: a systematic review of phase III trials. Lung Cancer 2005; 49:289–298. Socinski M, Schell M, Peterman A, et al. Phase III trial comparing a defined duration of therapy versus continuous therapy followed by second-line therapy in advanced stage IIIB / IV non-small cell lung cancer. Journal of Clinical Oncology 2002; 20:1335–1343. Smith I, O’Brien M, Talbot D, et al. Duration of chemotherapy in advanced non-small cell lung cancer: a randomized trial of three versus six courses of mitomycin, vinblastine and cisplatin. Journal of Clinical Oncology 2001; 19:1336–1343. The Elderly Lung Cancer Vinorelbine Study Group. Effects of vinorelbine on quality of life and survival of elderly patients with advanced non-small cell lung cancer. Journal of the National Cancer Institute 1999; 91:66–72. Frasci G, Lorusso V, Panza N, et al. Gemcitabine plus vinorelbine versus vinorelbine alone in elderly patients with advanced non-small cell lung cancer. Journal of Clinical Oncology 2000;18:2529–2536. Gridelli C, Perrone F, Gallo C, et al. Chemotherapy for elderly patients with advanced non-small cell lung cancer: the multicenter Italian lung cancer in the elderly (MILES) phase III randomized trial. Journal of the National Cancer Institute 2003; 95:362–372. Lilenbaum R, Herndon J, List M, et al. Single agent versus combination chemotherapy in advanced non-small cell lung cancer: a CALGB randomized trial of efficacy, quality of life and cost-effectiveness. Proc ASCO 2002; 21:1a. Gridelli C, Aapro M, Ardizzoni A, et al. Treatment of advanced non-small cell lung cancer in the elderly: results of international expert panel. Journal of Clinical Oncology 2005; 23:3125–3137. Shepherd F, Dancey J, Ramlau R, et al. Prospective randomized trial of docetaxel versus best supportive care in patients with non-small cell lung cancer previously treated with platinum-based chemotherapy. Journal of Clinical Oncology 2000; 18:2095–2103. Hanna N, Shepherd F, Fossella F, et al. Randomized phase III trial of pemetrexed versus docetaxel in patients with advanced non-small cell lung cancer previously treated with chemotherapy. Journal of Clinical Oncology 2004; 22:1589–1597.
534 Bronchus
127 Fukuoka M, Yano S, Giaccone G, et al. Multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small cell lung cancer. Journal of Clinical Oncology 2003; 21:2237–2246. 128 Kris M, Natale R, Herbst R, et al. Efficacy of gefitinib, an inhibitor of the epidermal growth factor recepto tyrosine kinase, in symptomatic patients with non-small cell lung cancer: a randomized trial. Journal of American Medical Association 2003; 290:2149–2158. ●129 Thatcher N, Chang A, Parikh P, et al. Gefitinib plus best supportive care in previously treated patients with refractory advanced non-small cell lung cancer: results from a randomised placebo-controlled, multicentre study (IRESSA Survival Evaluation in Lung Cancer). Lancet 2005; 366:1527–1537. 130 Giaccone G, Herbst R, Manegold C, et al. Gefitinib in combination with gemcitabine and cisplatin in advanced non-small cell lung cancer: a phase III trial – INTACT 1. Journal of Clinical Oncology 2004; 22:777–784. 131 Herbst R, Giaccone G, Schiller J, et al. Gefitinib in combination with paclitaxel and carboplatin in advanced non-small cell lung cancer: a phase III trial – INTACT 2. Journal of Clinical Oncology 2004; 22:785–794. ●132 Shepherd F, Pereira J, Ciuleanu T, et al. Erlotinib in previously treated non-small cell lung cancer. New England Journal of Medicine 2005; 353:123–132. 133 Gatzemeier U, Pluzanska A, Szczesna A, et al. Results of a phase III trial of erlotinib (OSI-774) combined with cisplatin and gemcitabine chemotherapy in advanced nonsmall cell lung cancer. Proc ASCO 2004; 23:617. 134 Herbst R, Prager D, Hermann R, et al. TRIBUTE: a phase III trial of erlotinib hydrochloride (OSI-774) combined with carboplatin and paclitaxel chemotherapy in advanced nonsmall cell lung cancer. Journal of Clinical Oncology 2005; 23:5892–5899. 135 Johnson D, Fehrenbacher L, Novotny W, et al. Randomized phase II trial comparing bevacizumab plus carboplatin and paclitaxel with carboplatin and paclitaxel alone in previously untreated locally advanced or metastatic non-small cell lung cancer. Journal of Clinical Oncology 2004; 22:2184–2191. 136 Sandler A, Gray R, Perry MC, et al. Paclitaxel-carboplatin alone or with bevacizumab for non-small cell lung cancer. New England Journal Medicine 2006; 355: 2542–2550. 137 Minna JD, Pass H, Glatstein E, Ihde DC. Cancer of the lung. In: De Vita V, Hellma S, Rosenberg SA, editors. Principle and Practice of Oncology. Philidelphia: JB Lippincott 1989: 591–705 138 Green RE, Humphrey E, Close H, Patno ME. Alkylating agents in bronchogenic carcinoma. American Journal of Cancer 1969; 46:516–525. ●139 Medical Research Council Lung Cancer Working Party. Comparison of oral etoposide and standard intravenous multidrug chemotherapy for small cell lung cancer: a stopped multicentre randomised trial. Lancet 1996; 348:563–566.
●140
141
142
143
144
◆145
●146
●147
148
149
150
151
Souhami RL, Spiro SG, Rudd RM, et al. Five day etoposide for advanced small cell lung cancer: randomized comparison with intravenous chemotherapy. Journal of the National Cancer Institute 1997; 89:577–580. Lowenbraun S, Bartolucci A, Smalley R, et al. The. superiority of combination chemotherapy over single agent chemotherapy in small cell lung cancer. Cancer 1979; 44:406–413. Sambrook R, Girling D. A national survey of the chemotherapy regimens used to treat small cell lung cancer in the United Kingdom. British Journal of Cancer 2001; 84:1447–1452. Fukuoka M, Furose K, Saijo N, et al. Randomized trial of cyclophosphamide, doxorubicin and vincristine versus cisplatin and etoposide versus alternation of these regimens in small cell lung cancer. Journal of the National Cancer Institute 1991; 83:855–861. Roth BJ, Johnson DH, Einhorn LH, et al. Randomized study of cyclophosphamide, doxorubicin and vincristine versus etoposide and cisplatin versus alternation of these two regimens in extensive small cell lung cancer: a phase III trial of the Southeastern Cancer Study Group. Journal of Clinical Oncology 1992; 10:282–291. Pujol J, Carestia L, Daures J. Is there a case for cisplatin in the treatment of small cell lung cancer? A meta-analysis of randomized trials of a cisplatin-containing regimen versus a regimen without this alkylating agent. British Journal of Cancer 2000; 83:8–15. Sundstrom S, Bremnes R, Kaasa S, et al. Cisplatin and etoposide regimen is superior to cyclophosphamide, epirubicin and vincristine regimen in small cell lung cancer: results from a randomized phase III trial with 5 years follow-up. Journal of Clinical Oncology 2002; 20:4665–4672. Thatcher N, Qian W, Clark P, et al. Ifosfamide, carboplatin and etoposide with midcycle vincristine versus standard chemotherapy in patients with small cell lung cancer and good performance status: clinical and quality of life results of the British Medical Research Council multicenter randomized LU21 trial. Journal of Clinical Oncology 2005; 23:8371–8379. Medical Research Council Lung Cancer Working Party. Survival, adverse reactions and quality of life during chemotherapy compared with selective palliative treatment for small-cell lung cancer. Respiratory Medicine 1989; 83:51–58. Earl HM, Rudd RM, Spiro SG, et al. A randomised trial of planned versus as required chemotherapy in small cell lung cancer: A Cancer Research Campaign trial. British Journal of Cancer 1991; 64:566–572 Bleehen NM, Fayers PM, Girling DJ, Stephens RJ. Controlled trial of twelve versus six courses of chemotherapy in the treatment of SCLC. British Journal of Cancer 1989; 59:584–590. Spiro SG, Souhami RL, Geddes DM, et al. Duration of chemotherapy in SCLC: a Cancer Research Campaign trial. British Journal of Cancer 1989; 59:578–583.
References 535
152 Bleehen NM, Girling DJ, Machin D, Stephens RJ. A randomised trial of three or six courses of etoposide, cyclophosphamide, methotrexate and vincristine or six courses of etoposide and ifosfamide in SCLC: survival and prognostic factors. British Journal of Cancer 1993; 68:1150–1156. 153 Mattson K, Niiranen A, Ruotsalainen T, et al. Interferon maintenance therapy for small cell lung cancer: improvement in long-term survival. Journal of Interferon and Cytokine Research 1997; 17:103–105. 154 Jett JR, Maksymiuk AW, Su JQ, et al. Phase III trial of recombinant interferon gamma in complete responders with small cell lung cancer. Journal of Clinical Oncology 1994; 12:2321–2326. 155 Kelly K, Crowley JJ, Bunn PA, et al. Role of recominant interferon alfa-2a maintenance in patients with limited stage small cell lung cancer responding to concurrent chemoradiation – A Southwest Oncology Group Study. Journal of Clinical Oncology 1995; 13:2924–2930. 156 van Zandwijk N, Greon HJM, Postmus PE, et al. Role of recombinant interferon-gamma maintenance in responding patients with small cell lung cancer. A randomised phase III study of the EORTC lung cancer cooperative group. European Journal of Cancer 1997; 33:1759–1766. 157 Schiller J, Adak S, Cella D, et al. Topotecan versus observation after cisplatin plus etoposide in extensive stage small cell lung cancer: E7593 – a phase III trial of the Eastern Cooperative Oncology Group. Journal of Clinical Oncology 2001; 19:2114–2122. 158 Hanna N, Sandler A, Loehrer PS, et al. Maintenance daily oral etoposide versus no further therapy following induction chemotherapy with etoposide plus ifosfamide plus cisplatin in extensive stage small cell lung cancer: a Hoosier Oncology Group randomized study. Annals of Oncology 2002; 13:95–102. 159 Bozcuk H, Artac M, Ozdogan M, Savas B. Does maintenance/consolidation chemotherapy have a role in the management of small cell lung cancer? Cancer 2005; 104:2650–2657. 160 Feld R, Evans W, Coy P, et al. Canadian multicenter randomized trial comparing sequential and alternating administration of two non-cross resistant chemotherapy combinations in patients with limited small cell carcinoma of the lung. Journal of Clinical Oncology 1987; 5:1401–1409. ◆161 Klasa R, Murray N, Coldman A. Dose-intensity metaanalysis of chemotherapy regimens in small cell carcinoma of the lung. Journal of Clinical Oncology 1991; 9:499–508. ●162 Arriagada R, LeChevalier Y, Pignon JP, et al. Initial chemotherapeutic doses and survival in patients with limited small cell lung cancer. New England Journal of Medicine 1993; 329:1848–1852. 163 Ardizzoni A, Favaretto A, Boni L, et al. Platinum-etoposide chemotherapy in elderly patients with small cell lung cancer: results of a randomized multicenter phase II study assessing attenuated-dose or full-dose with lenograstim prophylaxis – A Forza Operativa Nazionale Italiana
164
165
●166
167
168
169
170
171
172
173
174
●175
Carcinoma Polmonare and Gruppo Studio Tumori Polmonari Veneto (FONICAP-GSTPV) study. Journal of Clinical Oncology 2005; 23:569–575. Souhami RL, Rudd R, Ruiz de Elvira M-C, et al. Randomized trial comparing weekly versus 3-week chemotherapy in small cell lung cancer: A Cancer Research Campaign Trial. Journal of Clinical Oncology 1994; 12:1806–1813. Sculier JP, Paesmans M, Bureau G, et al. Multiple-drug weekly chemotherapy versus standard combination regimen in small cell lung cancer: A phase III randomized study conducted by the European Lung Cancer Working Party. Journal of Clinical Oncology 1993; 11:1858–1865 Murray N, Livingston RB, Shepherd FA, et al. Randomized study of CODE versus alternating CAV/EP for extensive stage small cell lung cancer: An intergroup study of the National Cancer Institute of Canada Clinical Trials Group and the Southwest Oncology Group. Journal of Clinical Oncology 1999; 17:2300–2308. Crawford J, Ozer H, Stoller R, et al. Reduction by granulocyte colony-stimulating factor of fever and neutropenia induced by chemotherapy in patients with small cell lung cancer. New England Journal of Medicine 1991; 325:164–170. Trillet-Lenoir V, Green J, Manegold C, et al. Recombinant granulocyte colony stimulating factor reduces the infectious complications of cytotoxic chemotherapy. European Journal of Cancer 1993; 29A:319–324. Bunn P, Crowley J, Kelly K, et al. Chemoradiotherapy with or without granulocyte macrophage colony stimulating factor in the treatment of limited stage small cell lung cancer. Journal of Clinical Oncology 1995; 13:1632–1641. Woll PJ, Hodgetts J, Lomax L, et al. Can cytotoxic dose intensity be increased by using granulocyte colonystimulating factor? A randomised controlled trial of lenograstim in small cell lung cancer. Journal of Clinical Oncology 1995; 13:652–659. Trillet-Lenoir V, Soler P, Arpin D, et al. The limits of chemotherapy dose intensification using granulocyte colony stimulating factor alone in extensive small cell lung cancer. Lung Cancer 1996; 14:331–341. Wolf M, Hans K, Drings P, et al. Treatment intensification with GM-CSF in patients with non-metastatic SCLC. Results of a multicenter trial including 330 patients. Proc ASCO1996;15:270. Woll PJ, Lee SM, Lomax L, et al. Randomised phase II study of standard versus dose-intensive ICE chemotherapy with reinfusion of haemopoietic progenitors in whole blood in small cell lung cancer. Proc ASCO 1996;15:333. Steward WP, von Pawel J, Gatzemeier U, et al. Effects of granulocyte-macrophage colony-stimulating factor and dose intensification of VICE chemotherapy in small cell lung cancer: A prospective randomised study of 300 patients. Journal of Clinical Oncology 1998; 16:642–650. Thatcher N, Girling DJ, Hopwood P, et al. Improving survival without reducing quality of life in small cell lung cancer patients by increasing the dose-intensity of chemotherapy with granulocyte colony-stimulating factor support: results
536 Bronchus
176
●177
◆178
179
●180
181
182
●183
184
185
186
187
188
of a British Medical Research Council Multicenter randomized trial. Journal of Clinical Oncology 2000; 18:395–404. Sculier J, Paesmans M, Lecomte J, et al. A three arm phase III randomised trial assessing in patients with extensive disease small cell lung cancer, accelerated chemotherapy with support of haematological growth factors or oral antibiotics. British Journal of Cancer 2001; 85:1444–1451. Ardizzoni A, Tjan-Heijnen V, Postmus P, et al. Standard versus intensified chemotherapy with G-CSF support in small cell lung cancer: a prospective EORTC – lung cancer group phase III trial 08923. Journal of Clinical Oncology 2002; 20:3947–3955. Pasini F, Durante E, De Manzoni D, Rosti G, Pelosi G. Highdose chemotherapy in small cell lung cancer. Anticancer Research 2002; 22:3465–3472. Pettengell R, Woll PJ, Thatcher N, et al. Multicyclic doseintensive chemotherapy supported by sequential reinfusion of haemopoietic progenitors in whole-blood. Journal of Clinical Oncology 1995; 13:148–156. Lorigan P, Woll P., O’Brien M, et al. Randomized phase III trial of dose-dense chemotherapy supported by wholeblood haemopoietic progenitors in better-prognosis small cell lung cancer. Journal of the National Cancer Institute 2005; 97:666–674. Perez-Soler R, Glisson B, Lee J, et al. Treatment of patients with small cell lung cancer refractory to etoposide and cisplatin with the topoisomerase inhibitor I poison topotecan. Journal of Clinical Oncology 1996; 14:2785–2790. Eckhardt J, Gralla R, Palmer M, et al. Topotecan as second line therapy in patients with small cell lung cancer: a phase II study. Annals of Oncology 1996; 7:107 (abstract). von Pawel J, Schiller J, Shepherd F, et al. Topotecan versus cyclophosphamide, doxorubicin and vincristine for the treatment of recurrent small cell lung cancer. Journal of Clinical Oncology 1999; 17:658–667. Von Pawel J, Gatzemeier U, Pujol J, et al. Phase II comparator study of oral versus intravenous topotecan in patients with chemosensitive small cell lung cancer. Journal of Clinical Oncology 2001; 19:1743–1749. O'Brien ME, Ciuleanu TE, Tsekov H, et al. Phase III trial comparing supportive care with supportive care with oral topotecan in patients with relapsed small-cell lung cancer. Journal of Clinical Oncology 2006; 24:5441-5447. Smit EF, Fokkema E, Biesma B, et al. A phase II study of paclitaxel in heavily pretreated patients with SCLC. British Journal of Cancer 1998; 77:347–351. Groen HJM, Fokkema E, Biesma B, et al. Paclitaxel and carboplatin in the treatment of small cell lung cancer patients resistant to cyclophosphamide, doxorubicin and etoposide: A non-cross resistant schedule. Journal of Clinical Oncology 1999; 17:927–932. Deppermann KM, Serke M, Oehm C, et al. Paclitaxel and carboplatin in advanced SCLC: A phase II study. Proc ASCO 1999; 18:482a.
189 Lyss AP, Herndon JE, Lynch TC, et al. Paclitaxel cisplatin G-CSF in patients with previously untreated extensive stage small cell lung cancer: Preliminary analysis of Cancer and Leukemia Group B (CALGB) 9430. Proc ASCO 1999; 18:468a. 190 Neubauer M, Schwartz J, Caracandas J, et al. Results of a phase II study of weekly paclitaxel plus carboplatin in patients with extensive small cell lung cancer with ECOG performance status of 2 or age 70 years. Journal of Clinical Oncology 2004; 22:1872–1877. 191 Ardizzoni A, Hansen H, Dombernowsky P, et al. Topotecan, a new active drug in the second line treatment of small cell lung cancer: a phase II study in patients with refractory and sensitive disease. Journal of Clinical Oncology 1997; 15:2090–2096. 192 Kudoh S, Fujiwara Y, Takada Y, et al. Phase II study of irinotecan combined with cisplatin in patients with previously untreated SCLC. Journal of Clinical Oncology 1998; 16:1068–1074. 193 Nakamura S, Kudoh S, Komuta K, et al. Phase II study of irinotecan (CPT-11) combined with etoposide for previously untreated extensive-disease small cell lung cancer: A study of the West Japan Lung Cancer Group. Proc ASCO 1999; 18:470a. 194 De Vore RF, Blanke CD, Denham CA. Phase II study of irinotecan in patients with previously treated small cell lung cancer. Proc ASCO 1998; 17:451a. 195 Masters G, Declerck L, Blanke C, et al. Phase II trial of gemcitabine in refractory or relapsed small cell lung cancer: Eastern Cooperative Oncology Group Trial 1597. Journal of Clinical Oncology 2003; 21:1550–1555. 196 Noda K, Nishiwaki Y, Kawahara M, et al. Irinotecan plus cisplatin compared with etoposide plus cisplatin for extensive small-cell lung cancer. New England Journal of Medicine 2002; 346:85–91. 197 Hanna N, Bunn PA Jr, Langer C, et al. Randomized phase III trial comparing irinotecan/cisplatin with etoposide/cisplatin in patients with previously untreated extensive-stage disease small cell lung cancer. Journal of Clinical Oncology 2006; 24:2038–2043. 198 Neubauer M, Heaven R, Olivares J, et al. Results of a phase II study of carboplatin plus gemcitabine in patients with untreated extensive stage small cell lung cancer. Lung Cancer 2004; 46:369–375. 199 De Marinis F, Nelli F, Lombardo M, et al. A multicenter, randomized phase II study of cisplatin, etoposide and gemcitabine or cisplatin plus gemcitabine as first-line treatment in patients with poor prognosis small cell lung carcinoma. Cancer 2005; 103:772–779. 200 Hoang T, Kim K, Jaslowski A, et al. Phase II study of second-line gemcitabine in sensitive or refractory small cell lung cancer. Lung Cancer 2003; 42:97–102. 201 Gridelli C, Rossi A, Berletta E, et al. Carboplatin plus vinorelbine plus G-CSF in elderly patients with extensive stage small cell lung cancer: a poorly tolerated regimen. Results of a multicentre phase II study. Lung Cancer 2002; 36:327–332.
References 537
202 Johnson E, Lake D, Herndon J, et al. Phase II trial of vinorelbine plus doxorubicin in relapsed small cell lung cancer: CALGB 9332. American Journal of Clinical Oncology 2004; 27:19–23. 203 Niell H, Herndon J, Miller A, et al. Randomized phase III intergroup trial of etoposide and cisplatin with or without paclitaxel and granulocyte colony-stimulating factor in patients with extensive stage small cell lung cancer: Cancer and Leukemia Group B Trial 9732. Journal of Clinical Oncology 2005; 23:3752–3759. 204 Hainsworth J, Carrell D, Drengler R, et al. Weekly combination chemotherapy with docetaxel and gemcitabine as first-line treatment for elderly patients and patients with poor performance status who have extensive stage small cell lung cancer: a Minnie Pearl Cancer Research Network phase II trial. Cancer 2004; 100:2437–2441. ●205 Giaccone G, Debruyne C, Felip E, et al. Phase III study of adjuvant vaccination with Bec2/bacille Calmette-Guerin in responding patients with limited-disease small cell lung cancer (European organisation for Research and Treatment of Cancer 08971-08791B; Silva study). Journal of Clinical Oncology 2005; 23:6854–6864. 206 Harper-Wynne C, Sumpter K, Ryan C, et al. Addition of SRL172 to standard chemotherapy in small cell lung cancer improves symptom control. Lung Cancer 2005; 47:289–290. 207 Dy G, Miller A, Mandrekar S, et al. A phase II trial of imatinib (ST1571) in patients with c-kit expressing relapsed small cell lung cancer: a CALGB and NCCTG study. Annals of Oncology 2005; 16:1811–1816. 208 Krug L, Crapanzano J, Azzoli C, et al. Imatinib mesylate lacks activity in small cell lung carcinoma expressing c-kit protein: a phase II clinical trial. Cancer 2005; 103:2128–2131. ●209 Shepherd FA, Giaccone G, Seymour L, et al. Prospective, randomized, double-blind, placebo-controlled trial of marimastat after response to first-line chemotherapy in patients with small cell lung cancer: a trial of the National Cancer Institute of Canada – Clinical Trials Group and the European Organisation for Research and Treatment of Cancer. Journal of Clinical Oncology 2002; 20:4434–4439. ◆210 Pignon J-P, Arriagada R, Ihde DC, Johnson DH, Perry MC, Souhami RL, et al. A meta-analysis of thoracic radiotherapy for small cell lung cancer. New England Journal of Medicine 1992; 327:1618–1624. ◆211 Warde P and Payne D. Does thoracic irradiation improve survival and local control in limited-stage small cell carcinoma of the lung? A meta-analysis. Journal of Clinical Oncology 1992; 10:890–895. ●212 Jeremic B, Shibamoto Y, Nikolic N, et al. Role of radiation therapy in the combined-modality treatment of patients with extensive stage small cell lung cancer. A randomized study. Journal of Clinical Oncology 1999; 17:2092–2099. ●213 Murray N, Coy P, Pater JL, et al. Importance of timing for thoracic irradiation in the combined modality treatment of
214
215
216
◆217
◆218
219
220
221
◆222
222a
223
224
limited stage lung cancer. Journal of Clinical Oncology 1993; 11:336–344. Jeremic B, Shibamoto Y, Acimovic L, Milisavljevic S. Initial versus delayed acclerated hyperfractionated radiation therapy and concurrent chemotherapy in limited stage small cell lung cancer: a randomized study. Journal of Clinical Oncology 1997; 15:893–900. Work E, Nielson O, Bentzen S, et al. Randomized study of initial versus late chest irradiation combined with chemotherapy in limited stage small cell lung cancer. Aarhus Lung Cancer Group. Journal of Clinical Oncology 1997; 15:3030–3037. Takada M, Fukuoka M, Kawahara M, et al. Phase III study of concurrent versus sequential thoracic radiotherapy in combination with cisplatin and etoposide for limited stage small cell lung cancer: results of the Japan Clinical Oncology Group Study 9104. Journal of Clinical Oncology 2002; 20:3054–2060. Fried D, Morris D, Poole C, et al. Systematic review evaluating the timing of thoracic radiation therapy in combined modality therapy for limited stage small cell lung cancer. Journal of Clinical Oncology 2004; 22:4837–4845. Pijls-Johannesma M, De Ruysscher D, Rutten I, et al. Early versus late chest radiotherapy for limited stage small cell lung cancer. Cochrane Database Systematic Review 2004(2). Coy P, Hodson I, Payne DG, et al. The effects of dose of thoracic irradiation on recurrence in patients with limited stage small cell lung cancer: initial results of a Canadian multicenter randomised trial. Int Radiat Oncol Biol Phys 1988; 14:219–226. Bonner J, Sloan J, Shanahan T, et al. Phase III comparison of twice daily split-course irradiation versus once-daily irradiation for patients with limited stage small cell lung carcinoma. Journal of Clinical Oncology 1999; 17:2681–2691. Turrisi AT, Kim K, Blum R, et al. Twice daily compared with once daily thoracic radiotherapy in limited stage small cell lung cancer treated concurrently with cisplatin and etoposide. New England Journal of Medicine 1999; 340:265–271. Auperin A, Arrigada R, Pignon J-P, et al. Prophylactic cranial irradiation for patients with small cell lung cancer in complete remission. New England Journal of Medicine 1999; 341:476–484. Slotman B, Faivre-Finn C, Kramer G et al. Prophylactic cranial irradiation in extensive small cell lung cancer. NEJM 2007; 357:664–672. Cull A, Gregor A, Hopwood P, et al. Neurological and cognitive impairment in long-term survivors of small cell lung cancer. European Journal of Cancer 1994; 30A:1067–1074. Gregor A, Cull A, Stephens R.J, et al. Prophylactic cranial irradiation is indicated following complete response to induction therapy in small cell lung cancer: results of a multicentre randomised trial. European Journal of Cancer 1997; 33:1752–1758.
538 Bronchus
225 Shepherd FA, Ginsberg RJ, Feld R, et al. Surgical treatment for limited stage small cell lung cancer: the University of Toronto Lung Oncology Group experience. Journal of Thoracic and Cardiovascular Surgery 1991; 101: 385–393. 226 Lad T, Piantadosi S, Thomas P, et al. A prospective randomized trial to determine the benefit of surgical
resection of residual disease following response of small cell lung cancer to combination chemotherapy. Chest 1994; 106:320S–323S. 227 Fujimori K, Yokoyama A, Kurita Y, Terashima M. A pilot phase II study of surgical treatment after induction chemotherapy for resectable stage I to IIIA small cell lung cancer. Chest 1997; 111:1089–1093.
21 Oesophageal cancer DAVID B. SMITH AND BRIAN J. HAYLOCK
Introduction Epidemiology and aetiological factors Clinical manifestations Management of localized (operable) oesophageal cancer
539 539 540 542
INTRODUCTION Carcinoma of the oesophagus is the eighth commonest cancer in Western countries, with an incidence of just over 8:100 0001 and accounting for approximately 5 per cent of cancer-related deaths.2 The disease is rare before the age of 40 but the incidence increases sharply thereafter, with an eightfold increase between the age ranges 45–54 years and 65–74 years. The male-to-female ratio for carcinoma of the thoracic oesophagus is about 2.5:1. This ratio is reversed for carcinoma of the cervical oesophagus, especially for post-cricoid carcinoma, which predominantly affects women. During the past 40 years there has been a significant change in the incidence of the two common histological subtypes.3 In the 1960s, 90 per cent of oesophageal cancers were squamous cell carcinomas, but with a decline in the numbers of these patients and a relentless rise in adenocarcinomas of the lower oesophagus and gastric cardia, adenocarcinoma now represents the most common histological subtype. Despite advances that have been made in surgical and radiotherapy techniques over the past 20 years and the inclusion of combination chemotherapy in treatment programmes, the outlook for oesophageal cancer remains poor. Thus only about 30 per cent of patients present with potentially curable early-stage disease and overall the 5-year survival remains less than 10 per cent.4
EPIDEMIOLOGY AND AETIOLOGICAL FACTORS Squamous cell carcinoma Oesophageal cancer shows greater geographical variation in incidence than any other cancer.5 Differences are found
Management of locally advanced oesophageal cancer Management of advanced oesophageal cancer References
550 551 555
both between local areas of the same region and between different ethnic groups within regions. Such differences suggest that environmental factors play an important part in its aetiology. Squamous carcinomas usually develop in the upper and middle third of the oesophagus and occur most commonly in China, Africa and Iran and with reduced frequency in most Western countries.5,6 Several risk factors have been identified, whose relative importance differs depending on the region.7 In Western countries, a history of smoking, excess alcohol consumption and a diet low in fruit and green vegetables account for most of the increased risk.8 For those patients exhibiting several adverse factors, including a history of chronic gastro-oesophageal reflux for more than 20 years, the relative risk may be as high as 44%.9 Moreover, most of the risk factors for oesophageal cancer are associated with lower socio-economic status10 and an increased risk of other cancers, e.g. head and neck cancer and lung cancer. Thus 10–15 per cent of patients with squamous cell carcinoma of the oesophagus will also develop a squamous cell carcinoma of the head and neck during their lifetime. Other dietary factors associated with squamous cell carcinoma of the oesophagus include foods containing N-nitroso compounds,11 betel-nut chewing in regions of Asia12 and the ingestion of high-temperature drinks in Iran, Russia and South Africa.13 Squamous cell carcinoma may be associated with other disease processes such as achalasia, caustic stricture and Plummer–Vinson’s syndrome. Pre-existing achalasia increases the risk by up to 16-fold14 and the presence of tylosis palmaris (type A) has been reported to be associated with squamous cell carcinoma of the oesophagus, possibly due to the absence of a tumour suppressor gene on 17q25.1.15 Interestingly, deletions in this gene have
540 Oesophageal cancer
also been noted in up to 70 per cent of patients developing sporadic squamous cell carcinoma. In areas of high incidence there is no difference in gender specificity, but in low-incidence countries males predominate. The majority of squamous cell carcinomas arise in the mid-oesophagus. Squamous cell carcinoma invades the submucosa at an early stage, and extends along the mucosal surface, usually in a cephalad direction. Local lymph-node invasion occurs early and quickly because the lymphatics in the oesophagus are located in the lamina propria, in contrast to the rest of the gastrointestinal tract, in which they are located beneath the muscularis mucosa. The tumour spreads to regional lymph nodes along the oesophagus, the coeliac area, and adjacent to the aorta. Invasion of local structures may result in fistula formation, e.g. to the trachea, and erosion into the aorta can cause massive upper gastrointestinal haemorrhage. Distant metastases to the liver, bone and lung are seen in nearly 30 per cent of patients at diagnosis and are more common in gastro-oesophageal adenocarcinoma. In addition, bone-marrow invasion can be detected in up to 40 per cent of cases using immunohistochemical techniques.16
or Barrett’s – should be encouraged to lose weight, stop smoking, increase their consumption of fruit and vegetables and take an aspirin each day. Patients with GERD or Barrett’s should also receive acid suppression with protonpump inhibitors. The majority of cases of oesophageal adenocarcinoma are located at the gastro-oesophageal junction and are associated with endoscopic evidence of Barrett’s oesophagus. Adenocarcinomas arising in Barrett’s oesophagus may present as an ulcer or a mass, or there may be no endoscopic abnormality.22 Early adenocarcinomas not associated with Barrett’s oesophagus usually arise from an ulcer, plaque or nodule at the gastro-oesophageal junction.23 In a manner analogous to squamous cell carcinoma, lymph-node metastases occur early to adjacent or regional lymph nodes, but involvement of coeliac and peri-hepatic nodes is more common with adenocarcinoma because of the location of the tumour at the gastro-oesophageal junction.24 The differences in risk factors, patterns of spread and changing incidence suggest that squamous cell carcinoma and adenocarcinoma of the oesophagus are different diseases. However, for the purposes of staging, management and prognosis, the principles are essentially the same.
Adenocarcinoma Rarer histological subtypes The incidence of oesophageal adenocarcinoma has increased in Western countries by around six-fold since 1975.3 In the USA these tumours have increased more rapidly than almost any other cancer, at a rate of 5–10 per cent per year during the 1980s, and in Norway the rise has approached 15 per cent per year.17 The major risk factor for adenocarcinoma is gastro-oesophageal reflux disease (GERD). A Swedish case-controlled study suggested that people who had suffered from recurrent reflux 5 years earlier were nearly eight times more likely to develop adenocarcinoma of the oesophagus than those who had not.18 In addition, more severe symptoms were associated with greater risk. Barrett’s oesophagus is also associated with an increased risk of adenocarcinoma of the oesophagus, although the exact nature of the link remains uncertain.19 In this condition, an abnormal intestinal-type epithelium replaces oesophageal squamous epithelium damaged by reflux, and the excess risk of developing an adenocarcinoma may be 30–125 times the average for age group. It has been estimated that patients with Barrett’s develop adenocarcinoma at the rate of about 0.5 per cent per year. Gastro-oesophageal reflux and Barrett’s both occur with greatest frequency in white males. Obesity is also a strong risk factor, accounting for a three-fold increase,20 but in contrast to squamous cell carcinoma, cigarette smoking is associated with quite a small increase and high alcohol consumption does not appear to influence the risk of developing adenocarcinoma. The use of aspirin and other non-steroidals has been found to protect against oesophageal cancer, as does a diet high in fruit and vegetables.21 Thus patients at increased risk of oesophageal adenocarcinoma – older white males with GERD
In addition to squamous and adenocarcinomas, a number of other cancers occur rarely in the oesophagus. These include small-cell carcinomas, melanomas, leiomyosarcomas and adenoid cystic tumours. In general, the management of these cancers depends on the histological subtype rather than the tissue of origin. Thus, for example, small-cell carcinoma of the oesophagus is treated in a manner analogous to small-cell lung cancer, with primary chemotherapy followed by consolidation radiotherapy where appropriate and with consideration given to prophylactic cranial irradiation.
CLINICAL MANIFESTATIONS Both adenocarcinomas and squamous cell carcinomas have similar clinical presentations, with the exception that adenocarcinomas arise more commonly in the distal oesophagus. Obstruction of the oesophagus by the tumour causes progressive dysphagia for solid food, usually occurring when the oesophageal lumen is less than 13 mm in diameter. Dysphagia is often accompanied by weight loss, which may be exacerbated by changes in diet and the onset of tumourrelated anorexia. Early symptoms of oesophageal cancer may be subtle and non-specific and easily dismissed as unimportant, hence the average duration of symptoms at presentation exceeds 4 months. Patients may also notice retrosternal discomfort, odynophagia or a retrosternal burning sensation. Regurgitation of saliva or food uncontaminated by gastric secretions can also occur in patients with advanced disease.
Clinical manifestations 541
Aspiration pneumonia is infrequent and hoarseness may occur if the recurrent laryngeal nerve is invaded. Chronic gastrointestinal blood loss from oesophageal cancer may occur, resulting in iron-deficiency anaemia. However, patients seldom notice blood in regurgitated food and melaena is also uncommon. Similarly, acute upper gastrointestinal bleeding is rare and may be the result of catastrophic tumour erosion into the aorta or pulmonary or bronchial arteries. Tracheo-bronchial fistulae are a late complication of oesophageal cancer. The fistulae are caused by direct invasion through the oesophageal wall and into the main stem bronchus. Such patients often present with intractable coughing or frequent pneumonias and their prognosis is usually limited to a few weeks.
●
Screening and early detection
STAGING
There is as yet no reliable, simple and cost-effective screening test for squamous cell carcinoma. For adenocarcinoma, patients known to have Barrett’s oesophagus may be offered surveillance endoscopy, usually at 2-yearly intervals, although whether or not this reduces mortality from oesophageal cancer is an area of controversy.25 Dysplasia, if found, should be ablated or resected. Photodynamic therapy (PDT) has been effective in treating in-situ disease and high-grade dysplasia. More information about this technique is given in the section of this chapter on endoscopic treatments.
Diagnosis and staging
● ● ● ● ● ● ●
The multidisciplinary team will usually meet every 1–2 weeks to discuss new patients, review staging investigations and plan management. The introduction of structured multidisciplinary working in the UK over the past 10 years has greatly improved the quality, coordination and delivery of patient care.
In common with many cancers, the most important prognostic factor for outcome is stage at presentation.4,27 Patients with disease that has spread beyond the regional lymph nodes cannot be cured by surgery or combined-modality therapy. The major goals of management in this group should be to extend survival where possible and, more importantly, to maintain quality of life and keep potentially distressing interventions to a minimum. Surgical resection carries significant morbidity and mortality and the main aim of staging investigations is to select those fit patients with early-stage disease who have the best chance of benefiting from resection. Modern staging now frequently involves: ●
DIAGNOSIS
Upper gastrointestinal endoscopy is the initial investigation of choice in patients presenting with dysphagia, and many centres now offer a rapid access service. Early oesophageal cancer may appear as a superficial plaque or ulceration and more advanced lesions as a stricture, an ulcerated mass, a circumferential mass or a large ulcer. Biopsy of the abnormal area provides the diagnosis in more than 90 per cent of cases. THE MULTIDISCIPLINARY TEAM
In 2000, the UK National Cancer Plan set out to improve efficiency and standardize cancer treatments. One key pillar of the plan is the multidisciplinary team. All newly diagnosed oesophageal cancer patients should be referred to the specialist multidisciplinary team for staging and management decisions. In the UK, the ideal composition and functioning of such teams are outlined in the Upper GI Cancer Improving Outcomes Guidance.26* Most teams will include: ● ●
surgeons gastroenterologists
radiologists clinical and medical oncologists palliative care physicians dieticians pathologists upper gastrointestinal specialist nurses intensive care specialists multidisciplinary team coordinator.
● ● ●
●
computed tomography (CT) scanning, endo-oesophageal ultrasound (EUS), positron emission tomography (PET) scanning, bronchoscopy (especially for tumours of the middle third), laparoscopy and mediastinoscopy.
Initial CT scanning is directed at the detection of distant metastatic disease, most commonly seen in the liver, lungs, peritoneum and abdominal and mediastinal lymph nodes. Care should be taken that the scan extends into the neck and is reviewed to exclude cervical adenopathy. The CT scan can also provide information about involvement of the tumour with the major blood vessels and trachea (Fig. 21.1). Patients who are free of metastatic disease should then have an EUS.28 This scan can provide detailed images of oesophageal masses and their relationship with the five-layered structure of the oesophageal wall. The accuracy of EUS staging has been investigated in a large number of patients, comparing results with operative specimens.29 It consistently provides better regional TNM tumour staging than CT scanning. Although local invasion is often well documented by CT, EUS is more accurate for detecting lymph-node metastases. The most important
542 Oesophageal cancer
Figure 21.1 CT scan showing bulky oesophageal carcinoma with no local nodal disease or aortic involvement (By kind permission of Dr David White.)
nodal group to be examined is the coeliac area. The finding of malignant lymphadenopathy in the retrogastric area remote from the primary tumour predicts the presence of distant metastatic disease and is a sign of unresectability. Endo-oesophageal ultrasound can also detect small volumes of ascites and pericardial effusion and, if necessary, suspicious areas can be biopsied by fine-needle aspiration (FNA).29 A caveat for EUS staging of oesophageal lymph nodes is that the instrument may not be able to traverse a tumour-induced stenosis (which occurs in about 30 per cent of cases). In these patients, EUS may under-stage the tumour, since the entire lesion and the coeliac axis are not seen. Attempts to overcome this problem by dilating the oesophagus may result in perforation. An alternative is to use miniature EUS probes,30 which are passed through the biopsy channel of an endoscope. However, these devices only provide a partial solution to the problem because of their limited depth of penetration (approximately 3 cm). A non-optical, wire-guided echo-endoscope is another alternative that may allow more complete EUS staging in the majority of patients.31 In addition, EUS may be less accurate for staging patients who have undergone preoperative chemo-radiotherapy. In one report, for example, EUS correctly predicted a complete response to chemo-radiotherapy in only 3 of 17 patients (18 per cent) who had no residual tumour demonstrated in resection specimens.32 Measuring the change in maximal cross-sectional area before and after chemo-radiotherapy may be more useful in assessing the response of oesophageal cancer to preoperative chemo-radiation. Positron emission tomography with 18F-fluorodeoxyglucose is a non-invasive staging modality that is complementary to CT for the detection of distant metastases. In two studies totalling 94 patients, PET prevented needless surgery in 22 who had metastatic disease undetected by CT
alone.33,34 In a study of 91 patients with oesophageal cancer, PET scans detected 51 metastases in 27 of 39 cases (69 per cent sensitivity, 93.4 per cent specificity, 84 per cent accuracy) and was more accurate than CT at detecting distant metastases.35 In contrast to the improved detection of metastatic disease, PET scans are not as useful in the evaluation of the primary site and loco-regional nodes. One study prospectively compared PET scanning to preoperative CT and EUS in 74 patients with potentially operable cancer of the oesophagus or gastro-oesophageal junction.36 At the primary site, PET scanning showed increased activity in 70 of 74 patients, but was falsely negative in four patients with T1 tumours. Thirty of the 74 patients underwent lymphadenectomy at the time of oesophageal resection. Positron emission tomography scanning was not as sensitive as EUS for the detection of local lymph nodes (sensitivity 33 versus 81 per cent) and had better specificity (89 versus 67 per cent). Nevertheless, the improved sensitivity of PET for the detection of metastatic disease makes it potentially the most cost-effective method of identifying patients with occult metastases, for whom aggressive therapy should not be pursued. In addition, it is likely that in the future co-registered PET and CT will find a place in enhancing the accuracy of radiotherapy planning. Laparoscopy is an important final investigation for patients with apparently operable gastro-oesophageal junction cancers. Previously unrecognized peritoneal and liver metastases may be seen and the coeliac nodes can be directly visualized and biopsied if necessary. Bronchoscopy is required in tumours of the middle third of the oesophagus that are shown to be lying adjacent to the trachea or main bronchi on CT scanning. Direct endotracheal or endobronchial invasion will indicate inoperability and contraindicates radiotherapy due to the risk of fistula formation. The most widely used staging system for oesophageal cancer is the Union Internationale Contre le Cancer Tumour Node Metastasis (UICC TNM) system37 (Box 21.1). The importance of staging is reflected in prognosis. Reported 5-year survival rates for stages I to IV disease are 60, 31, 20 and 4 per cent, respectively.27
MANAGEMENT OF LOCALIZED (OPERABLE) OESOPHAGEAL CANCER The optimum management for patients with localized oesophageal cancer is a controversial area. Only 30–40 per cent of patients have potentially resectable disease at presentation, and in many series only 5–20 per cent of those undergoing surgery are alive at 3–5 years.38 As a result, much research has centred on the use of combined-modality therapy incorporating both chemotherapy and radiotherapy into treatment protocols in an attempt to improve these results.
Management of localized (operable) oesophageal cancer 543
Box 21.1 Staging TNM T – Primary tumour Tx Primary tumour cannot be assessed. T0 No evidence of primary tumour. Tis Carcinoma-in-situ. T1 Tumour invades lamina propria or submucosa. T2 Tumour invades muscularis propria. T3 Tumour invades adventitia. T4 Tumour invades adjacent structures. N – Regional lymph nodes Nx Regional lymph nodes cannot be assessed. N0 No regional lymph-node metastasis. N1 Regional lymph-node metastasis. M – Distant metastasis Mx Distant metastasis cannot be assessed. M0 No distant metastasis. M1 Distant metastasis. For tumours of lower thoracic oesophagus M1a Metastasis in coeliac lymph nodes. M1b Other distant metastasis. For tumours of the upper thoracic oesophagus M1a Metastasis in cervical lymph nodes. M1b Other distant metastasis. For tumours of mid-thoracic oesophagus M1a Not applicable. M1b Non-regional lymph-node metastasis or other distant metastasis. Stage grouping Stage 0 Stage I Stage IIA Stage IIB Stage III Stage IVA Stage IVB
In the first stage of the operation, the stomach is approached through an upper abdominal incision and is mobilized; in stage II, the oesophagus is approached through a right fifth interspace thoracotomy for dissection and resection of the tumour. A margin of approximately 8 cm on either side of the tumour is desirable. Continuity of the oesophageal remnant and the mobilized stomach is restored by anastomosis, which is carried out by hand or with a staple gun, according to preference. Operative and postoperative mortality may be up to 10–15 per cent, depending on case mix and patient selection, the main problems being related to cardiopulmonary complications, sepsis and anastomotic leakage.40 The involvement of intensive care specialists in multidisciplinary team discussions and in selecting patients for surgery plays an important part in keeping these complications to a minimum. A series of 500 patients treated with a standard transthoracic oesophagectomy has been reported by Ellis.41 Approximately one-third of these patients had squamous carcinomas and two-thirds adenocarcinomas of the oesophagus or gastro-oesophageal junction. Operative mortality was 3.3 per cent and up to 21 per cent required a prolonged postoperative hospital stay due to complications. The overall 5-year survival was 24.7 per cent, with no significant difference according to histological subtype. Five-year survival was 79 per cent for patients with stage I disease, 38 per cent for those with IIA, 27 per cent for those with IIB and 13.7 per cent for those with stage III tumours.
TRANS-HIATAL OESOPHAGECTOMY Tis T1 T2 T3 T1 T2 T3 T4 Any T Any T
N0 N0 N0 N0 N1 N1 N1 Any N Any N Any N
M0 M0 M0 M0 M0 M0 M0 M0 M1a M1b
From reference 37
Surgery TRANSTHORACIC OESOPHAGECTOMY
For good-performance-status patients with tumours at or below the carina, surgical resection remains the treatment of choice. The most widely performed operation is a total oesophagectomy, usually using the stomach for reconstruction. The precise type of surgical procedure carried out varies according to surgeon and centre, but in the UK the two-stage technique of Ivor Lewis is favoured by many.39*
Trans-hiatal oesophagectomy is an alternative approach, which carries lower postoperative morbidity and mortality but is more likely to result in inadequate resection margins,42* and local failure rates approach 40 per cent.43* Trans-hiatal oesophagectomy does not require a thoracotomy incision and is often reserved for less fit patients who may not be able to withstand more aggressive surgery.
RADICAL SURGICAL RESECTION AND RECONSTRUCTION
This can be done by dissection and en-bloc removal of the posterior mediastinum, with transposed stomach or colon being used to restore continuity of the alimentary tract.44 There is a serious danger of leakage from the colonic anastomoses or of sloughing of the transposed colon owing to ischaemia. The highest mortality from oesophageal surgery is encountered in association with this kind of resection, but the operation does allow the possibility of achieving radical clearance of the tumour and associated lymph nodes. Lymphadenectomy forms part of the procedure, with two-field and three-field lymphadenectomy procedures, the latter providing superior results. It also allows the most definitive method of surgical staging.
544 Oesophageal cancer
EXTENDED LYMPHADENECTOMY
Oesophagectomy with an extended or three-field lymphadenectomy (usually including mediastinal, abdominal and cervical nodes) is commonly practised in Japan.45 In skilled hands, it can be performed with relatively low mortality and morbidity compared to other techniques, but the long-term benefit of extended lymphadenectomy is controversial.46 Proponents of extended lymphadenectomy quote impressive survival rates as evidence of its therapeutic benefit. In one American series of 80 patients undergoing this technique with en-bloc oesophagectomy, the overall 5-year survival rate was 51 per cent (88 per cent for node-negative and 33 per cent for node-positive patients).47 However, since the extent of lymph-node dissection can affect the assignment of the final stage of disease, the resulting stage migration will inevitably compromise stage-by-stage comparison between different forms of surgical resection. In addition, although unsuspected metastases in the recurrent laryngeal or cervical nodes were detected in 36 per cent of patients in the abovenoted series, others have reported a low incidence of cervical nodal recurrence following a two-field lymphadenectomy.45 The cervical oesophagus The cervical oesophagus is 6–8 cm long and extends from the cricopharyngeus to the thoracic inlet, where it is contiguous with the thoracic oesophagus. Carcinoma of the cervical oesophagus, usually squamous cell carcinoma, is relatively uncommon, accounting for less than 5 per cent of all oesophageal cancers.49 Locally advanced disease is often present at diagnosis. In one report, tracheal invasion and vocal cord paralysis were evident in 35 per cent and 24 per cent of patients, respectively.50 Carcinoma of the cervical oesophagus presents a unique management situation. Treatment choices include surgery, radiation or combined-modality treatment. Where surgical resection is attempted, a radical approach is adopted involving pharyngo-laryngo-oesophagectomy and reconstruction. This includes removal of the thyroid gland, and radical neck dissections are usually carried out.51 Thus, the management is more closely related to that of squamous cell carcinoma of the head and neck than of malignancies of the more distal portions of the oesophagus. This is very major, arduous and hazardous surgery and can be justified only for cases that have undergone the most careful selection, and where suitable surgical and supportive care expertise is available. Radiation combined with chemotherapy is generally preferred over surgery in this situation, since the opportunity for long-term survival appears to be similar, and major morbidity is avoided in most patients.52 In one series of 32 patients treated with chemo-radiotherapy and salvage surgery, the 10-year survival rate was 27 per cent, and 15 of the 32 patients successfully preserved their larynx.53 The importance of hospital volume There are certainly individual surgeons who obtain excellent results despite managing limited numbers of patients.
However, it is clear that patients who undergo oesophagectomy at hospitals where large numbers of procedures are performed have lower peri-operative mortality rates and better early clinical outcomes than those who undergo resection at lower-volume institutions.54–56 Moreover, with the recognition of the importance of multidisciplinary team working and a continuing decline in resection rates due to more sophisticated staging, there is a drive towards increasing centralization of services. In the UK, the 2001 Improving Outcomes Guidance states that each specialized surgical unit should serve a population of approximately 1 million. This will inevitably lead to the phasing-out of small-volume units. POSTOPERATIVE (ADJUVANT) THERAPY
A number of trials have investigated adjuvant therapy following potentially curative resection of oesophageal cancer but none has shown a consistent advantage of chemotherapy, radiotherapy or chemo-radiotherapy.57 Postoperative radiotherapy can be difficult to deliver safely because the large ‘at-risk’ volume, which includes the violated tissue planes, prevents the delivery of therapeutically useful doses without significant morbidity. However, in selected cases it may be used if a definite site of residual disease can be identified, for example with clips, and encompassed in an appropriately small volume. The only survival benefit for postoperative radiotherapy comes from a very small series of node-negative patients.58 More recent studies report better local control rates but no advantage in terms of overall survival.59*,60* PREOPERATIVE (NEOADJUVANT) THERAPY
Prolonged recovery times following surgery often prevent the administration of postoperative adjuvant therapy for oesophageal cancer within a reasonable timeframe. One strategy to overcome this problem is to use preoperative (neoadjuvant) treatment. This approach has a number of potential advantages, including improved drug and oxygen delivery, resulting in enhanced effectiveness of chemotherapy and radiotherapy, and the possibility of down-sizing the cancer, allowing a greater chance of complete surgical resection. Chemotherapy alone and chemo-radiotherapy have been investigated as neoadjuvant therapy for operable oesophageal cancer. Intra-operative radiotherapy has produced interesting results, principally from Japanese studies; however, the practice has never become widespread because of the high associated morbidity and complex specialized nature of this treatment. PREOPERATIVE CHEMOTHERAPY
A number of randomized trials have evaluated the benefit of chemotherapy administered prior to resection in patients with operable oesophageal cancer according to radiological
Management of localized (operable) oesophageal cancer 545
assessment. Only two of these studies demonstrate a survival benefit.61*,62 In the United States Intergroup Trial 0113, 467 patients with potentially resectable oesophageal cancer were randomly assigned to surgery alone or to three cycles of preoperative chemotherapy consisting of cisplatin 100 mg/m2 on days 1, 29 and 58, and 5-fluorouracil (5-FU) 1000 mg/m2 by continuous infusion on days 1–5 of each cycle, followed by surgery.63* Patients with chemotherapy-responsive disease (stable or objective response with preoperative chemotherapy) who underwent potentially curative resection received an additional two courses of postoperative chemotherapy with a reduced dose of cisplatin (75 mg/m2) plus 5-FU. The majority of patients had adenocarcinoma (55 per cent) and outcomes were similar for both histological subtypes. The clinical response rate to preoperative chemotherapy (as assessed by barium-contrast study) was 19 per cent. A complete pathological response (pCR) was noted at resection in only five (2.5 per cent) of 202 patients who received at least one cycle of chemotherapy. There were no differences between the groups in the rate of complete resection (65 versus 66 per cent), treatment-related mortality (6.4 versus 4.0 per cent), median survival (14.9 versus 16.1 months), or survival at 1, 2 or 3 years (59 per cent, 35 per cent and 23 per cent versus 60 per cent, 37 per cent and 26 per cent, respectively). Pre-treatment weight loss of more than 10 per cent was a significant predictor of poor outcome. In contrast, the UK MRC OE02 trial demonstrated a benefit for neoadjuvant cisplatin 5-FU. In this study, 802 patients with operable oesophageal cancer (two-thirds adenocarcinoma) were randomly assigned to resection alone or to resection preceded by two courses of cisplatin (80 mg/m2 on day 1) and 5-FU (1000 mg/m2 by continuous infusion on days 1–4) given 3 weeks apart.61* Preoperative radiation, administered at the discretion of the treating clinician, was received by 9 per cent of the patients in each group. The percentage of patients undergoing surgery was similar for the two groups, 92 per cent versus 97 per cent, as was the curative resection rate (i.e. R0 resection), 60 per cent versus 54 per cent. Preoperative chemotherapy was associated with significantly greater overall survival (hazard ratio 0.79; 95% confidence interval (CI) 0.67–0.93; 2-year survival 43 per cent versus 34 per cent; and median survival 16.8 months versus 13.3 months. The frequency of postoperative deaths and non-fatal complications were similar in the two groups. On examination of the resected specimen, tumours in the chemotherapy group were significantly smaller, extended less frequently into surrounding tissue, and showed less lymph-node involvement. The frequency of local recurrence as a component of failure was similar in patients undergoing chemotherapy compared to surgery alone (12 per cent versus 11 per cent). Additional data are available from the MRC MAGIC trial.64 This was originally a study of gastric cancer (74 per cent of patients had gastric cancer) but was later extended to include gastro-oesophageal junction tumours
(15 per cent) and distal oesophageal adenocarcinoma (11 per cent). Patients were randomly assigned to surgery with or without peri-operative chemotherapy – consisting of three preoperative plus three postoperative cycles of ECF (epirubicin 50 mg/m2 on day 1, cisplatin 60 mg/m2 on day 1, and infusional 5-FU 200 mg/m2 per day on days 1–21. The final report of this trial shows a significant improvement in 5-year survival for patients receiving chemotherapy surgery (36 per cent, 95% CI 30–43 per cent) compared to surgery alone (23 per cent, 95% CI 17–29 per cent), a magnitude of benefit in line with that seen in the OE02 study. Effective down-staging was suggested by the fact that more chemotherapy-treated patients had T1/2 tumours (52 per cent versus 37 per cent, p 0.002) and N0/N1 disease (84 per cent versus 71 per cent, p 0.01). Preoperative chemotherapy as delivered in OE02 or MAGIC is well tolerated and does not increase operative morbidity or mortality and has been accepted as standard practice in the UK. PREOPERATIVE CHEMO-RADIOTHERAPY
Combined chemotherapy and radiotherapy takes advantage of the radiosensitizing effect of cisplatin-based chemotherapy to enhance local control while at the same time attempting to eradicate distant micrometastases. Chemo-radiotherapy had been shown to be an effective treatment for locally advanced oesophageal cancer65* and this led to its investigation as neoadjuvant therapy for operable disease. At least six trials have directly compared preoperative (neoadjuvant) chemo-radiotherapy followed by surgery with surgery alone for patients with potentially resectable oesophageal carcinoma, five of which have been published;66–70,71* only one demonstrated a significant survival benefit for combined-modality therapy.71* Two general approaches have been used: concurrent and sequential chemo-radiotherapy. Two completed randomized trials have compared preoperative concurrent chemo-radiotherapy with surgery alone. In a trial conducted in Ireland, 113 patients with oesophageal adenocarcinoma were randomly assigned to surgery alone or to surgery preceded by chemoradiotherapy.71* Preoperative treatment consisted of two courses of 5-FU (15 mg/kg by bolus on days 1–5) and cisplatin (75 mg/m2 on day 7 of each cycle), both administered during weeks 1 and 6 with concurrent radiotherapy (40 Gy in 15 fractions over weeks 1–3). A pCR was noted in 25 per cent of patients treated with preoperative chemoradiotherapy, and when the surgical specimens were compared, regional nodal involvement was less common in this group (42 per cent versus 82 per cent). Combined-modality therapy was associated with significantly longer median survival (16 months versus 11 months) and 3-year survival (32 per cent versus 6 per cent). This relatively small trial is often cited but has several flaws. The major criticisms were that the results of surgery alone were inferior to those of other contemporary series, the preoperative staging was
546 Oesophageal cancer
inadequate, and there were disproportionate withdrawals between the two arms. In a second study, from Michigan, 100 patients with loco-regional oesophageal cancer (25 squamous cell carcinoma, 75 adenocarcinoma) were randomly assigned to trans-hiatal oesophagectomy with or without preoperative chemo-radiotherapy.67 Neoadjuvant treatment consisted of cisplatin (20 mg/m2 per day by continuous infusion on days 1–5 and 17–21), 5-FU (300 mg/m2 per day by continuous infusion on days 1–4 and 17–20) and vinblastine (1 mg/m2 per day by intravenous bolus on days 1–4 and 17–20) plus concurrent hyperfractionated radiotherapy (45 Gy in 1.5-Gy fractions, given twice daily for 3 weeks). Radiation was administered using a three-dimensional conformal treatment-planning technique. Surgery was performed on day 42, after a 3-week rest from completion of the chemo-radiotherapy. A pCR was observed in 28 per cent of patients receiving neoadjuvant treatment. Despite this, at a median followup of 8.2 years, the median survival was similar with both treatments (16.9 months versus 17.6 months for multimodality therapy and surgery respectively), and there was nearly a two-fold higher, but non-statistically significant, improvement in 3-year survival for the combined treatment group (30 per cent versus 16 per cent). Patients achieving a pCR had a significantly improved 3-year survival compared to those with residual tumour in the resected specimen (64 per cent versus 19 per cent, respectively). Although the combined therapy group had a significantly lower locoregional recurrence rate (19 per cent versus 42 per cent), preoperative chemo-radiotherapy had no significant effect on the distant metastatic rate (65 per cent versus 60 per cent). The study was statistically powered to detect a doubling of median survival from 12 to 26 months. Thus, although this was a negative study, it was insufficiently powered to detect a smaller survival difference. Three trials comparing surgery with or without preoperative sequential chemo-radiotherapy have failed to show an advantage for the combined-modality group in terms of overall survival. These studies have been criticized with respect to their size and adequacy of radiation dose. A meta-analysis of randomized controlled trials comparing neoadjuvant chemo-radiation followed by surgery with surgery alone included 1116 patients enrolled on nine trials.72 When compared to surgery alone, the odds ratios showed a non-significant trend towards improved survival with neoadjuvant chemo-radiotherapy (0.79, 0.77 and 0.66 for 1-year, 2-year and 3-year survival, respectively). However, the improvement in 3-year survival reached the level of statistical significance only when the analysis was restricted to those trials using concurrent chemotherapy and radiation (odds ratio 0.45, 95% CI 0.26–0.79). Although patients treated with surgery alone were significantly more likely to undergo resection, those receiving preoperative chemo-radiotherapy were more likely to undergo complete (R0) resection (odds ratio 0.53, 95% CI 0.33–0.84).
The data suggest a trend in favour of preoperative chemo-radiotherapy for oesophageal cancer but without consistent evidence of a survival benefit. Thus in the UK it has not been adopted as standard practice. IMPACT OF NEOADJUVANT THERAPY ON MORBIDITY
Whether neoadjuvant therapy affects the complication rate following oesophagectomy is unclear; the available data are conflicting. In retrospective non-randomized series, the postoperative complication rate compared to historical controls undergoing surgery alone is higher in some series73 but not in others.74 However, in randomized trials that had a surgery-only control arm, the addition of chemotherapy or chemo-radiotherapy did not appear significantly to increase the morbidity or mortality associated with surgery. NECESSITY FOR SURGERY
Contemporary series suggest that chemo-radiotherapy as a non-surgical approach may provide long-term survival in up to 27 per cent of patients with squamous cell carcinoma,65 a result that is not dissimilar to that achieved with preoperative chemo-radiotherapy and surgery,75 preoperative chemotherapy and surgery,61 and surgery alone.67 However, there is a high rate of locally persistent/recurrent disease after chemo-radiotherapy alone, and a lack of data related to the non-surgical management of patients with adenocarcinoma. This is partly because it has proven very difficult accurately to assess a complete response with historical staging tests in the non-surgical cases. At least two randomized trials directly comparing chemo-radiotherapy alone to chemo-radiotherapy followed by surgery have failed to demonstrate better survival in the surgically treated patients.75,76 Nevertheless, locoregional control is better in patients who undergo surgery as a component of multi-modality management. In the first trial, 172 patients with oesophageal squamous cell carcinoma were randomly assigned to three cycles of induction 5-FU, leucovorin, etoposide and cisplatin followed by concomitant chemo-radiotherapy (cisplatin plus etoposide on days 2 and 8 only, and 40 Gy external-beam irradiation) and then resection, or the same chemotherapy followed by concomitant chemo-radiotherapy (chemotherapy on days 2 and 8 only) with an additional 20 Gy of radiotherapy added instead of surgery.75 Treatment-related mortality was higher in the trimodality arm (12.8 per cent versus 3.5 per cent). The surgically treated patients had significantly better local control (2-year local progression-free survival 58 per cent versus 36 per cent, p 0.003), but the 3-year overall survival (31 per cent versus 24 per cent, respectively) and median survival durations (16 months versus 15 months, respectively) were not significantly different. Of note, in comparison to other trials using concomitant chemo-radiotherapy or chemo-radiotherapy followed by surgery, radiation doses and the intensity of chemotherapy were lower in this study. This may in part explain the
Management of localized (operable) oesophageal cancer 547
lower survival rate in the non-surgically treated patients when compared to the patients with locally advanced disease treated with chemo-radiotherapy in Radiation Therapy Oncology Group (RTOG) 85-01.65 In the second study, 455 patients with T3–4, N0–1, M0 oesophageal squamous cell carcinoma or adenocarcinoma received induction chemo-radiotherapy with either protracted (46 Gy in 4.5 weeks) or split-course (2 15 Gy on days 1–5 and 22–26) radiation plus two courses of 5-FU and cisplatin chemotherapy.76 Patients with at least a partial response and who lacked a contraindication to surgery (n 259) were then randomly assigned to continue chemo-radiotherapy – three more cycles of chemotherapy with either 30 Gy (protracted) or 15 Gy (split-course) radiation – or to undergo surgery. In a preliminary report, continued chemo-radiotherapy was associated with similar rates of 2-year survival (40 per cent versus 34 per cent) and median survival (19 months versus 18 months), although surgically treated patients were significantly less likely to require oesophageal stents (13 per cent versus 27 per cent, p 0.005) or repeated dilatation (22 per cent versus 32 per cent, p 0.07). The data suggest that patients who initially respond to chemotherapy should complete it rather than stop and undergo surgery. Although there is good rationale for its use, there is no evidence that the combination of surgery and chemo-radiotherapy improves the results of either treatment alone.
Alternatives to surgery ⴞ neoadjuvant therapy for localized disease A significant number of patients will either not wish to undergo surgery or will not be fit enough to withstand the operation. For these patients, a number of options are available, depending on performance status, tumour stage and patient choice. The options include: ● ● ● ●
chemo-radiotherapy radiotherapy alone endoscopic treatments symptom control.
CHEMO-RADIOTHERAPY
There are no trials comparing surgery with or without neoadjuvant treatment with chemo-radiotherapy alone in operable oesophageal cancer. The difficulties of recruitment combined with problems obtaining fully informed consent for widely differing treatments such as surgery versus chemo-radiotherapy make it unlikely that a suitably powered, randomized controlled trial will ever take place. However, data from studies using chemo-radiotherapy in locally advanced disease suggest that this treatment can achieve results at least equivalent to those of surgery in patients with localized disease. Thus for fit patients who are able to tolerate the treatment, it appears to be a viable
alternative and is the treatment of choice for patients with upper third cancers.
Radiotherapy Radiation alone can result in long-term survival in a minority of patients. It may be considered when chemotherapy is not contemplated or inappropriate due to age, pre-existing medical condition or patient choice. Historically, radical radiation treatment was associated with 5-year survival rates of 5–20 per cent, depending on tumour extent. In a large historical review of 49 early series involving more than 8400 patients treated with radiation alone, survival rates at 1, 2 and 5 years were 18 per cent, 8 per cent and 6 per cent, respectively.77* Better results are reported in later studies from single institutions in well-defined patient populations using specific radiation protocols. Three-year and 5-year survival rates of 27 per cent and 21 per cent, respectively, were observed in a series that included 101 selected patients with clinically localized oesophageal cancer treated with radiation alone (45–52.5 Gy in 15 or 16 fractions over 3 weeks).78* In a cohort of 17 patients with clinical stage I oesophageal squamous cell carcinoma (after EUS and CT) treated with radiation alone (median 60.6 Gy), the 5-year overall survival rate was 59 per cent.79 Radiation carries a lower mortality than surgery, but can nevertheless be associated with significant morbidity. As with surgery, careful selection of patients for radical radiation treatment is of great importance. Patients who are malnourished or have a poor performance status will be less able to withstand treatment and more likely to suffer toxicity than fit patients. Early and continuing fluid and nutritional support is important for all cases. Modern three-dimensional conformal radiotherapy techniques employing complex CT planning and the use of modern linear accelerators with multi-leaf collimators has improved the reliable delivery of radiotherapy to allow relatively higher doses to be delivered to the tumour with better sparing of critical structures such as the lungs and spinal cord. This benefit can also be achieved using endoluminal brachytherapy to deliver a boost to the tumour. A brachytherapy boost delivered after rather than before external-beam radiotherapy, for bulky tumours, is thought to be more effective, as this allows the tumour to ‘shrink’ into the localized brachytherapy volume.80 In this Canadian randomized study, approximately 50 per cent of patients went on to receive oesophagectomy. This revealed the sterilization rates for squamous cell carcinoma and adenocarcinoma to be 36 per cent and 21 per cent using external-beam radiotherapy with brachytherapy boost and no chemotherapy. Two randomized trials have compared external-beam radiotherapy with and without a brachytherapy boost.81,82 In one, the use of brachytherapy was associated with better local control and palliation of dysphagia, and a significant survival advantage at 1 year (78 per cent versus 44 per cent). In the second, a significant survival advantage at 5 years
548 Oesophageal cancer
(17 per cent versus 10 per cent) also favoured the brachytherapy group. Localized cancers of the upper and middle oesophagus are most suited to these techniques, although very high tumours are excluded from brachytherapy because of technical difficulties and the risk of cartilage necrosis. For tumours of the lower third, particularly those involving the gastro-oesophageal junction, and where the inferior extent of the tumour is difficult to determine accurately, the benefits of conformal treatments are less clear and these tumours may be best approached surgically. The limited tolerance of the upper portion of the stomach, together with the high probability of infra-diaphragmatic spread, make a radical, purely radiotherapeutic approach in this situation unrealistic.
Figure 21.2 Four-field conformal plan for radical oesophageal radiotherapy – coaxial view (See Plate Section.).
RADIOTHERAPY TECHNIQUES
The challenge in treating the oesophagus arises from the anatomy and nearby critical structures. As the oesophagus descends within the thorax, its position changes relative to the radiosensitive spinal cord, and it is flanked by the equally radiosensitive lungs. Below the diaphragm, the density of surrounding structures increases and the oesophagus becomes more anterior and lateral and the treatment volume increases. Planning must ensure that the tolerance of nearby sensitive structures, including the liver, kidneys and small intestine, is not exceeded. External-beam radiotherapy Patients are planned and treated supine with their arms above the head, holding a suitable device to allow immobilization. Historically, simulator planning required barium to identify the level of the tumour and to track the course of the oesophagus. Lateral images were needed to allow an appropriate plane of treatment and shielding to be added to avoid the spinal cord. This process is much more simple and efficient with modern virtual fluoroscopy and simulation. Three-dimensional conformal planning allows more reliable conformity to the determined tumour volumes.83* It is, however, important to supplement the CT anatomy with all the information gleaned from the pre-treatment staging investigations, including the barium swallow, endoscopy and EUS and PET where available. Barium may still be required after CT planning to distinguish the ‘functional’ level of the tumour, and this stored digitally or on film to better inform the planning process and, where brachytherapy boost is contemplated, to identify the region to be boosted. The gross tumour volume (GTV) may be defined by outlining the tumour on a series of coaxial CT images. The clinical treatment volume (CTV) may extend well beyond the apparent tumour in order to allow for submucosal spread and lymph-node involvement. The CTV is defined as the GTV together with margins of at least 4 cm beyond the radiographic tumour extent in the superior–inferior direction, and 2 cm beyond the lateral tumour borders. For lesions of the lower third of the oesophagus and gastrooesophageal junction, the CTV includes the coeliac, gastric, and gastrohepatic lymph nodes. For lesions involving the
Figure 21.3 Four-field conformal plan for radical oesophageal radiotherapy – sagittal view (See Plate Section.).
upper two-thirds of the thorax, the mediastinal nodes are included, and for the cervical oesophagus, both supraclavicular regions are commonly included in the CTV, along the lines of head and neck tumour protocols. Large volumes are still generally required and in practical terms this means that the treatment volume will never be less than 14 cm long, even for early lesions. In the past, a typical radical treatment plan would consist of two phases: first a parallel opposed pair of fields anterior–posterior to avoid lung tissue, followed by a threefield plan comprising an anterior and two lateral oblique fields to reduce the spinal dose to within tolerance. This was particularly important when concurrent chemo-radiotherpy was prescribed. More recently, with modern conformal planning, a four-field arrangement is employed with an anterior, posterior and two lateral oblique fields weighted and with the use of multi-leaf collimation to achieve the optimal dose distribution (Figs 21.2 and 21.3). This approach gives much better and more comprehensive dosimetry; however, when planning from CT data, it is important to stress that the true extent of the tumour cannot always be seen on the CT image. The temptation to use very small volumes with inadequate margins should be resisted. The
Management of localized (operable) oesophageal cancer 549
possible existence of extensive submucosal spread and lateral extension should always be kept in mind. In terms of dose, conformal therapy should theoretically allow much higher tumour doses to be achieved, with the aim of reducing local recurrence. In practice, it is difficult to increase doses significantly without unacceptable toxicity to, often frail, patients. However, modern radiotherapy techniques offer a more reliable reduction of dose to the normal surrounding structures. It remains to be seen whether future developments in functional tumour mapping such as CT-PET planning and image fusion or co-registration will allow us better to define areas at particularly high risk of recurrence and to deliver a higher dose where it is needed. The role of intensity-modulated radiotherapy (IMRT) is yet to be evaluated in this disease. In practice, a commonly used radical regime is 50 Gy in 2Gy fractions, ‘boosted’ with concurrent chemotherapy or intra-luminal brachytherapy. It should be noted that a brachytherapy boost should not be used in patients receiving combined chemo-radiotherapy because of the increased risk of mucosal ulceration. In our practice, many patients are considered too frail to undergo this radical approach but nevertheless are potentially curable. These patients are offered 40 Gy in 15 fractions in 3 weeks combined with cisplatinum and 5-FU during the first week, or, if unfit for chemotherapy, they are offered a brachytherapy boost to intensify the treatment. Both these ‘modified radical’ approaches have been shown to achieve respectable sterilization rates.71,84* For cervical tumours, the dose regimens are more akin to those used to treat head and neck tumours and may be in the order of 60 Gy in 2-Gy fractions or higher. Intra-luminal brachytherapy Remote afterloading intra-luminal brachytherapy has become more available with the development of the relatively portable, convenient to use, high-dose-rate (HDR) Microselectron (Fig. 21.4). This has largely replaced earlier treatments described in Exeter84 using the slower-dose-rate selectron, which required over an hour rather than minutes to treat. Brachytherapy can permit treatment of a localized area of the oesophagus to high radiation doses with relative sparing of the surrounding structures. This technique may be used alone or in combination with external-beam radiation. There is a much higher risk of mucosal ulceration if it is used concurrently with chemotherapy and radiation, and this is therefore not recommended. Modern brachytherapy treatment involves the intraluminal placement of a remote afterloading radioactive source train into the oesophagus. Specialized cuffed applicators are available in order to position the source centrally within the oesophagus. High-dose-rate doses of 15 Gy at 1 cm from the central axis may be delivered as a single dose using these applicators, which limit the mucosal dose. However, they require oral placement with sedation and an anaesthetic presence. In our centre, a simpler system employs the use of a nasogastric Ryles tube of a minimum 6 mm diameter. However, this limits the dose to 10 Gy at 1 cm from the central axis in order to avoid excessive localized
Figure 21.4
High-dose rate (HDR) Microselectron.
mucosal doses, which may result in ulceration. It is possible to fractionate brachytherapy to spare toxicity, especially where it is the sole modality of treatment, but this would rarely be the case in a radical treatment plan. COMPLICATIONS OF RADIOTHERAPY
Radiation oesophagitis Radiation oesophagitis is common, and develops in the second and third weeks of treatment. It is often perceived by patients as a sensation of obstruction, and varies from a minor symptom to a severe problem. A bismuth-chelating agent such as Sulcrafate may improve mucosal protection, but some patients prefer simple aspirin and hydrocortisone mucilage or its equivalent given orally. Radiation-induced oesophagitis may be exacerbated by concurrent chemotherapy, and nutritional and fluid support is often required to speed recovery in radically treated patients. Radiation pneumonitis This may produce a dry cough, even some months after treatment has been completed. It may be precipitated by intercurrent infection and may recur with subsequent infections. A hazy shadowing, corresponding to the boundaries of the radiation treatment fields, is sometimes seen. It responds promptly to soluble prednisolone (15 mg twice a day) and gradual tapering of the dose once symptoms have
550 Oesophageal cancer
been controlled. Appropriate gastric protection is required when steroids are used. Fistula formation Broncho-oesophageal fistulae may occur as a result of destruction of tumour that has invaded the tracheobronchial tree. Bronchoscopic examination of the airway is recommended in all patients with middle third tumours. Tumours thought to be invading in this way should only be irradiated with great caution, usually only after the placement of a covered stent. The incidence of fistula formation during chemo-radiotherapy was 6 per cent in one series, and accounted for one-half of all tracheo-oesophageal fistulae in association with oesophageal cancer; the other half presented with this complication.85 For patients with persistent fistulae, symptomatic management may include covered airway and oesophageal stents, or surgery. A fistula, when present, gives rise to explosive coughing and choking following a sip of water or other fluid. The bronchial tree may outline with contrast when this is swallowed and the site of the fistula can sometimes be seen. Gastrograffin or non-irritant contrast should be used rather than barium. In a similar manner, breakdown of tumour may occur where it is infiltrating large blood vessels, with disastrous results and catastrophic haemorrhage. Such catastrophic haemorrhage may be preceded by one or more minor haemoptyses. Acute dysphagia This may occasionally occur where the lumen of the oesophagus becomes further reduced by post-treatment radiationinduced swelling. Such cases respond to steroids. Late complications The late complications of radical radiation therapy include mediastinal fibrosis and post-radiation stricture. The latter is managed by dilatation. It is important to consider frequent endoscopic dilatation for benign fibrotic strictures during the first 6 months post-radiotherapy to prevent the development of late chronic stricturing. The prevalence of malignant and non-malignant post-radiotherapy strictures was about equal in one series.86 The majority of patients with benign strictures were successfully dilated and had a 12-month survival rate of 88 per cent, compared to 19 per cent for those with malignant strictures. Radiation damage to the spinal cord may occur where tolerance is locally exceeded, usually due to adverse patient geometry. It is very rare with modern planning and dosimetry. Patients treated palliatively with opposed portals may occasionally experience transient radiation myelopathy, with sensations like electric shocks in the arms or legs following neck flexion (L’hermitte’s sign). These symptoms are generally self-limiting without specific treatment.
Endoscopic resection Endoscopic treatment options are available for patients with early-stage tumours but who are too ill to undergo
surgery or who refuse surgery or radiotherapy. Endoscopic mucosal resection (EMR) or endoscopic mucosectomy can be curative in selected patients and provides a histological specimen that can be evaluated for adequacy of resection margins. The technique has traditionally involved lifting the mucosa by submucosal injection of saline and endoscopic suction into an overtube in which a snare is fitted. Photodynamic therapy is an alternative treatment that may be considered in these patients.87,88 In the largest series of 123 such patients who were recommended for nonoperative treatment, the complete response rate at 6 months was 87 per cent with PDT alone or as a component of multimodality therapy, and the 5-year disease-specific survival rate was 74 per cent.87 Photodynamic therapy is also a useful treatment for in-situ disease and high-grade dysplasia.
MANAGEMENT OF LOCALLY ADVANCED OESOPHAGEAL CANCER Following initial staging, many patients will be found to have locally advanced unresectable disease. For these patients, chemo-radiotherapy should be considered.
Primary chemo-radiotherapy The addition of cisplatin-based chemotherapy to radiation has been shown to provide a significant survival benefit compared with radiotherapy alone for patients with locally advanced oesophageal cancer,65 and although the available randomized trial data are almost exclusively in squamouscell carcinoma, the results have been extrapolated to adenocarcinoma. A landmark study, RTOG 85-01, compared radiation therapy alone (64 Gy in 2-Gy fractions over 6.5 weeks) with concurrent chemo-radiotherapy (four cycles of 5-FU, 1000 mg/m2 by continuous infusion for the first 4 days of weeks 1, 5, 8 and 11, plus cisplatin, 75 mg/m2 on day 1 every 4 weeks) and radiation therapy (50 Gy in 25 fractions over 5 weeks and concurrent with the first two cycles of chemotherapy) in patients with loco-regional thoracic oesophageal cancer (i.e. no evidence of spread beyond mediastinal and supraclavicular lymph nodes).58 Ninety per cent of the trial participants had squamous cell carcinoma. The trial was closed prematurely, after 121 patients had been enrolled, when an interim analysis showed a significant survival advantage in the chemo-radiotherapy arm. In 1999, Cooper et al. updated the results of this study: combined-modality treatment was associated with a significantly better median survival (14 months versus 9.3 months) and 5-year survival (26 per cent versus 0 per cent).88* Analysis of patterns of failure showed a significant reduction in both loco-regional and distant failure for the chemo-radiotherapy arm. Despite this apparent benefit, 46 per cent of patients in the experimental treatment group had recurrence or persistence of disease in the oesophagus
Management of advanced oesophageal cancer 551
at 12 months. Criticisms of this study include the unusually poor results in the radiotherapy-alone arm, which contained a higher proportion of black patients (associated with a worse prognosis), and the toxicity associated with the chemotherapy. However, this landmark study defined combined chemo-radiotherapy as the standard of care for patients with inoperable disease. The issue of the unacceptably high loco-regional failure rate was addressed in a follow-up trial, INT 0123. In this study, all 236 enrolled patients received concurrent chemotherapy with cisplatin and 5-FU (as in RTOG 85-01), but they were randomly assigned to one of two different radiation doses: 50.4 Gy (28 fractions of 1.8 Gy each, 5 fractions per week) or 64.8 Gy (36 fractions of 1.8 Gy each, 5 fractions per week).52 Patients without evidence of distant metastatic squamous cell carcinoma or adenocarcinoma of the thoracic oesophagus were eligible for the study. The use of higher radiation doses was not associated with a higher median survival (13 months versus 18 months), 2-year survival rate (31 per cent versus 40 per cent) or incidence of loco-regional persistent or recurrent disease (56 per cent versus 52 per cent for the high dose and control groups, respectively). In addition, the high-radiation-dose arm was also associated with significantly more toxicity. A number of other series have explored modifications of the chemo-radiotherapy regimen used in RTOG 85-01 in patients with unresectable or inoperable oesophageal cancer. In one report, 35 patients with non-metastatic but inoperable oesophageal cancer and dysphagia were treated with concurrent cisplatin (70 mg/m2, on days 2 and 30), 5-FU (1000 mg/m2, given by continuous infusion on days 1 through 5 and 29 through 33), and split-course radiation therapy (20 Gy in five fractions, delivered on days 1 through 5 and 29 through 33). The overall median survival was 17 months, with a 2-year survival rate of 41 per cent. At posttreatment endoscopy, no tumour was found in 25 patients (71 per cent).89 A second series included 211 patients with inoperable squamous cell oesophageal cancer who were randomly assigned to receive radiotherapy alone (20 Gy in five fractions given twice, separated by a 15-day rest period) or concurrent radiation (same dose and schedule) plus cisplatin (100 mg/m2 before each radiation cycle, and repeated every 21–28 days for a total of six cycles).90 Compared to radiation alone, combined therapy was associated with significant improvements in overall survival (9.6 months versus 7.9 months), median progression-free survival (6.9 months versus 5.0 months) and loco-regional progression-free survival (10.9 months versus 6.2 months). Paclitaxel is a potent radiation sensitizer.91 In preliminary data, the combination of radiation and concurrent paclitaxel was highly active and acceptably toxic in patients with loco-regionally advanced or metastatic oesophageal cancer. In one report, 40 patients with loco-regionally advanced disease (endoscopic T3, N1 or M1 nodal disease) received two courses of cisplatin (20 mg/m2 per day by continuous infusion for 4 days) and paclitaxel (175 mg/m2 over 24 hours
on day 1) concurrent with a split course of accelerated fractionation radiation (1.5 Gy twice daily to a total dose of 45 Gy).92 Complete surgical resection was possible following chemo-radiotherapy in 93 per cent, 81 per cent achieved long-term loco-regional control, and 30 per cent of patients were still alive at 3 years. Higher doses of external-beam radiation have been combined with concurrent chemotherapy. Intergroup Trial 0123 included 236 patients with clinical stage T1–4 Nx M0 oesophageal cancer who were randomly assigned to 50.4 Gy versus 64.8 Gy of radiation with concurrent cisplatin (75 mg/m2 on day 1) and 5-FU (1000 mg/m2 by continuous infusion on days 1 through 4 of weeks 1 and 5).93* Two additional chemotherapy courses were repeated 4 weeks after radiation. The trial was closed early because there was no demonstrated survival advantage, and there was a higher rate of toxic deaths (9 per cent versus 2 per cent) with the higher radiation dose. It therefore appears that excellent results can be achieved in single centres with careful patient selection using intensified treatment schedules. However, these treatments are attended by significantly increased morbidity, and there are no randomized data to suggest that the protocol employed in RTOG 85-01 should not continue as the standard of care.
MANAGEMENT OF ADVANCED OESOPHAGEAL CANCER Approximately two-thirds of patients with oesophageal cancer have metastatic disease at presentation, and others will have too poor a performance status to tolerate surgery or combined-modality therapy. For these patients, median survivals are short and their management centres around effective palliation. The options to consider in this situation include: ● ● ● ●
radiotherapy chemotherapy endoscopic treatments palliative care.
Radiotherapy Radiation has an important role in the management of unresectable oesophageal cancer, both for palliation of dysphagia and for maintenance of local disease control. Significant palliative benefit can be achieved in the majority of irradiated patients. In one series that examined the difference in food-passage scores before and after radiation therapy in 127 patients with oesophageal cancer, dysphagia was improved and adequately palliated until death in 71 per cent and 54 per cent of patients, respectively.94 However, this benefit may be of too short a duration in patients who are expected to survive longer than 3–6 months. Good palliation should aim to deliver effective treatment in as short a space of time as possible, with minimal
552 Oesophageal cancer
morbidity and maximum improvement in quality of life. Radiotherapy can be administered as external-beam treatment or via intra-luminal brachytherapy, as discussed earlier. For palliative external-beam treatment, principally designed to relieve dysphagia, simulator planning is commonly used. This often employs a parallel opposed anterior–posterior pair of fields to cover the tumour seen on a barium swallow image and complimented by other staging images. This rapid simulation often enables the treatment to commence the same day. The dose and fractionation vary from 30 Gy in ten daily fractions, or 20 Gy in five daily fractions to 18 Gy in three daily fractions, all to mid-plane, depending on the overall performance status and estimated survival of the patient. Radiation is also commonly used for the palliation of local tumour masses, e.g. bone and brain metastases, and can provide excellent symptomatic relief.
Brachytherapy Modern HDR equipment, e.g. HDR Microselectron (see Fig. 21.4), delivers radiation much more quickly than older systems, thus permitting the delivery of a planned dose within minutes. Unless simply using HDR brachytherapy to boost an external-beam treatment or in patients with a limited life expectancy, fractionation is necessary to avoid fibrotic stricturing. Typically, two to four fractions are administered when this is the sole treatment. Brachytherapy alone, however, can provide successful long-term palliation of dysphagia in patients with unresectable and/or advanced oesophageal cancer. In a trial in which 209 patients with obstruction from oesophageal or gastro-oesophageal junction tumours were randomly assigned to brachytherapy alone (12 Gy) or to endoscopic placement of a metal stent, the stented group had more rapid improvement within 30 days of the procedure.95 However, at later time points, brachytherapy was associated with significantly lower dysphagia severity scores and more days with almost no dysphagia (115 days versus 82 days). The brachytherapy group also had a significantly lower complication rate, as well as better quality-of-life scores, and they were no more likely to require re-treatment for recurrent or persistent dysphagia than the stented group. The optimal dose and timing of brachytherapy remain uncertain. In one prospective trial, 232 patients with unresectable thoracic oesophageal cancer were randomly assigned to HDR brachytherapy in one of two schedules: 16 Gy in two fractions during 3 days, or 18 Gy in three fractions over 5 days.96 The median dysphagia-free survival was 214 days, and was similar in both groups. Treatment was complicated by tracheo-oesophageal fibrous stricture in 10 per cent and 11 per cent of patients, respectively, and the rate of stricture did not differ between the two groups. The precise role of brachytherapy in the treatment of unresectable oesophageal cancer still remains to be defined. The use of brachytherapy alone should be restricted to patients with a short life expectancy. When used as a boost
following concomitant chemotherapy and external-beam radiation, brachytherapy may be associated with better disease control. However, it should not be administered concurrently with chemotherapy when using these regimens. Useful guidelines for the use of brachytherapy have been produced by the Clinical Research Committee of the American Brachytherapy Society.
Chemotherapy Many patients with oesophageal cancer have metastatic disease at the time of presentation or develop metastases following treatment for apparently localized disease. The goals of chemotherapy in patients with advanced oesophageal cancer are to palliate symptoms, especially malignant dysphagia, and to improve survival. When assessing the results of trials in this area it is important to remember that objective responses are often difficult to assess, particularly in the primary tumour. Several agents, including bleomycin, mitomycin-C, methotrexate and etoposide, have shown activity in advanced oesophageal cancer, but the most widely used remain cisplatin and 5-FU. All these drugs have modest single-agent response rates of 15–20 per cent, with the benefit of treatment usually lasting 2–3 months. More recently, a range of newer agents has undergone testing in advanced oesophageal cancer, including carboplatin, taxotere, taxol, irinotecan and vinorelbine. In general, none of these drugs has shown consistent improvements over cisplatin 5-FU. In particular, carboplatin appears to be significantly less active than the parent drug,97 and should not replace cisplatin in standard regimens. However, the more recently introduced platinum analogue oxaliplatin shows more promise. Irinotecan and vinorelbine both appear to have activity in the 15–20 per cent range, but the taxanes, taxol and taxotere, have been studied in a number of different schedules and have shown limited benefit and significant toxicity. Although it is likely that squamous cell carcinoma and adenocarcinoma have somewhat different sensitivities to chemotherapy, histological subtype probably does not have a major role in either the response rate or duration of survival in patients with metastatic oesophageal cancer. In addition, it is possible to extrapolate data from randomized studies performed for gastric cancer, which has twice the incidence of oesophageal cancer. Many of these trials have included adenocarcinoma of the gastro-oesophageal junction and thus there is considerable overlap with oesophageal cancer trials. It should be noted, however, that distal gastric cancer has a different natural history and prognosis from proximal gastric tumours and should probably be evaluated separately in chemotherapeutic trials. COMBINATION CHEMOTHERAPY
Most reported series of combination chemotherapy in oesophageal cancer have included small numbers of patients
Management of advanced oesophageal cancer 553
with either advanced loco-regional or metastatic disease. The majority of combination regimens have been cisplatin based. The combination of cisplatin and 5-FU is one of the most commonly used regimens in both metastatic and localized oesophageal cancer due to its activity and well-established toxicity profile. In a randomized phase II study, 88 patients with locally advanced or metastatic oesophageal squamous cell carcinoma were randomly assigned to either single-agent cisplatin (100 mg/m2 every 21 days) or the same dose of cisplatin combined with 5-FU (1000 mg/m2 per day by continuous infusion on days 1–5).98 Although the response rate was higher for the combination (35 per cent versus 19 per cent), the median survival rate (33 weeks versus 28 weeks) and 1-year survival rate (34 per cent versus 27 per cent) were not significantly different. Furthermore, a 17 per cent treatmentrelated mortality rate (primarily due to sepsis and cerebrovascular episodes) was observed in the cisplatin and 5-FU arm. A similar degree of activity (response rate of 27 per cent) was reported with much less toxicity (mortality rate of 3 per cent) in a second study that used split-dose cisplatin (20 mg/m2 per day on days 1–5) in combination with leucovorin and bolus 5-FU (300 mg/m2 per day for 5 days).99 The combination of cisplatin and etoposide has also been studied. In a trial sponsored by the Rotterdam Oesophageal Tumour Study Group, 73 patients with unresectable or metastatic oesophageal squamous cell carcinoma received cisplatin (80 mg/m2 over 4 hours on day 1) plus etoposide (100 mg/m2 intravenously on days 1 and 2, followed by etoposide 200 mg/m2 orally on day 3), with courses repeated every 4 weeks.100 Five complete and 26 partial responses were reported in 65 evaluable patients (response rate 48 per cent) and the median time to progression was 7 months. There were two toxic deaths due to neutropenic sepsis during treatment. Many phase II trials have been undertaken using combinations involving newer agents. These studies frequently report relatively high response rates, in excess of 40 per cent, but median survivals remain short and no regimen has been shown to be superior to cisplatin 5-FU in phase III randomized testing. In one trial, cisplatin 60 mg/m2 and taxol 180 mg/m2 were administered every 2 weeks. The overall response rate was 43 per cent, median survival was 9 months and 1-year survival was 43 per cent.101 One area of promise is the use of oral 5-FU prodrugs. In the treatment of oesophageal cancer, 5-FU is generally administered by continuous infusion over 4–21 days, depending on the schedule employed. Such treatment requires either inpatient admission or the placement of a central line, with its attendant risks and morbidity. Thus replacement of 5-FU by capecitabine or UFT would allow patients easier access to more conveniently administered outpatient treatment. A large randomized trial in patients with advanced gastric and gastro-oesophageal junction adenocarcinomas comparing ECF with the same regimen but replacing infusional 5-FU with capecitabine and cisplatin with oxaliplatin has recently completed recruitment and the results are awaited with interest.
For good-performance-status patients with advanced squamous cell carcinoma of the oesophagus, cisplatin 5-FU remains the treatment of choice. In our experience, it is well tolerated, with low rates of neutropenic sepsis and other toxicities. Four cycles usually provide the maximum therapeutic benefit. For patients with advanced adenocarcinoma of the oesophagus or gastro-oesophageal junction, cisplatin 5-FU is a reasonable option. The addition of epirubicin in the ECF schedule may provide a somewhat higher response rate but is attended by more toxicity, including alopecia. Unfortunately, all of the currently available treatments are sub-optimal, and wherever possible patients should be considered for inclusion in clinical trials. It is likely that conventional surgery, radiotherapy and chemotherapy have reached their limits in terms of enhancing survival in this disease, and current research is centring on the place of antiangiogenic agents such as bevacizumab and drugs designed to inhibit various targets involved in the signal transduction pathway. For example, a significant proportion of patients with adenocarcinoma of the oesophagus and gastrooesophageal junction over-express epidermal growth factor receptors, and agents such as trastuzumab that block this receptor are under investigation in this disease.
Endoscopic treatment Oesophageal cancer is often diagnosed at an advanced, incurable stage and a significant proportion of patients are not well enough to tolerate systemic chemotherapy or radiotherapy. For these patients, local treatment using endoscopic manoeuvres may offer the best palliation for improving swallowing and nutritional status. A variety of endoscopic methods are available, including dilatation, stenting, laser therapy and PDT. Oesophageal dilatation with either through-the-scope balloon or wire-guided polyvinyl bougies can provide temporary relief of dysphagia until more definitive treatment can be accomplished. Most malignant strictures can be safely dilated to 16 mm or 17 mm in several sessions. However, repeat dilatation is usually required every 3–4 weeks, and oesophageal dilatation carries a small risk of perforation.102 Laser therapy with neodymium-yttrium-aluminiumgarnet (Nd:YAG) has been the traditional form of palliative treatment for oesophageal cancer, although other forms of laser, e.g. the argon plasma coagulator, are also now available. Treatment with the Nd:YAG laser typically requires several sessions. The goal is to ablate sufficient malignant tissue in order to restore luminal patency. Exophytic masses in the mid or distal oesophagus are easily treated, but masses at the gastro-oesophageal junction or near the cricopharyngeus are difficult to access. Treatments are performed every other day and are usually completed in three to four sessions.103 Luminal patency can be achieved in more than 90 per cent of cases with the Nd:YAG laser, and functional success
554 Oesophageal cancer
occurs in 70–80 per cent. Relief may last for one to several months and treatments may have to be repeated. Perforation, the most serious complication, occurs in less than 5 per cent of patients.103 The disadvantages of laser therapy include difficulty in the management of long or tortuous segments, frequency of treatment sessions, and cost. There is some evidence that the immediate benefits of laser treatments can be made more durable if supplemented by simple external-beam radiotherapy in suitably fit individuals. Photodynamic therapy uses a photosensitizing agent, porfimer sodium, in combination with endoscopic lowpower laser exposure. Porfimer sodium is a haematoporphyrin derivative that accumulates in malignant tissue after intravenous injection, and the area is then exposed to an endoscopically placed low-power laser diffuser with monochrome light (630 nm), which initiates a photochemical reaction that results in tumour necrosis. The malignant tissue can be repeatedly treated in order to provide optimal tissue ablation. It is also effective for the treatment of highgrade oesophageal dysplasia and Barrett’s oesophagus. Initial trials in oesophageal cancer have demonstrated that PDT provides significant palliation from dysphagia in patients with advanced disease or loco-regional failure after chemo-radiotherapy. In a preliminary report of ten patients with residual or locally recurrent disease following definitive chemo-radiotherapy, seven had a complete clinical response from PDT, three of whom remained progression free at 6 months.104 Photodynamic therapy is technically easier than Nd:YAG laser therapy and is better tolerated by patients. The expense of PDT is similar to that of traditional laser therapy, due to the cost of the Photofrin, and requires sophisticated equipment. The most common side effects of PDT include chest pain and worsening dysphagia. These complications gradually resolve over several weeks, except for a small percentage of patients with stricture development. Although porfimer sodium is cleared from a variety of tissues within 40–72 hours after injection, tumours, the skin and some organs (i.e. liver and spleen) retain the drug for a longer period. Skin photosensitivity may persist for 4–6 weeks after the treatment; sunscreens are ineffective, as they do not block visible light. Photodynamic therapy can also be used to treat tissue ingrowth or overgrowth in patients who have already undergone oesophageal stenting.
in the days following placement, which may require opiate analgesia. Other problems include stent migration, creation of fistulae, airway obstruction and haemorrhage. Moreover, proximal stents may cause a foreign-body sensation or respiratory compromise, and distal stents that traverse the gastro-oesophageal junction can result in intractable reflux requiring proton-pump inhibitors. The value of the first-generation metal stents was reduced by tumour ingrowth, often within 3–6 months of placement and resulting in recurrence of dysphagia. This led to the investigation of radiotherapy as an adjunct to stenting, but there is little evidence that it helps to prolong stent life and it does lead to increased toxicity. However, the introduction of stents covered with semi-permeable membranes has decreased the frequency of tumour ingrowth, although obstruction due to benign causes, including granulation tissue and reactive hyperplasia, remains a problem in some patients. In addition, the problem of gastro-oesophageal reflux in stents crossing the gastro-oesophageal junction has been reduced by newer stents that include an anti-reflux mechanism. The complications of self-expanding metal stents are more common in patients who have been treated with radiation and/or chemotherapy.
Choice of modality for palliation of dysphagia There are now a number of interventions available to palliate dysphagia, and the choice in an individual patient is not always easy. It is important to appreciate that stents do not restore swallowing to normal and should not be used too early in the disease process, when radiotherapy or chemotherapy may be a better option. As suggested previously, the location of the tumour will have a bearing on the
Endoscopic stenting Patients with advanced oesophageal cancer, those who are poor surgical candidates and those whose disease has recurred105* can be offered stenting as a palliative treatment for dysphagia. The most commonly used stents are selfexpanding metal stents (Fig. 21.5). Stents are expensive and can fail as a result of tissue ingrowth. In addition, a significant proportion of patients experience mediastinal pain
Figure 21.5 CT scan showing expandable metal stent in place (By kind permission of Dr David White.)
References 555
most appropriate treatment, with stenting being contraindicated in high tumours and more problematic in junctional cancers. Radiotherapy, including brachytherapy, chemotherapy, laser treatment and PDT should all be available in centres managing oesophageal cancer, and patients should be able to access them without undue delay. In practice, the timing of all procedures is best discussed on an individual basis with the multidisciplinary team, with input from clinical oncology, the endoscopists, dieticians and palliative care team.
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
●
●
●
Carcinoma of the oesophagus remains a highly lethal disease, with an overall 5-year survival of less than 10 per cent. Patients with suspicious symptoms, usually dysphagia, should be offered rapid-access endoscopy, with biopsy where appropriate. Following diagnosis, all patients should be referred to the specialist multidisciplinary team for investigation and management. Patients who are fit enough to undergo potentially curative surgery or chemoradiotherapy should then have full staging, including CT, EUS and PET scanning. Following staging, patients with localized disease of the middle and lower third of the oesophagus should be offered surgery or chemoradiotherapy. There is evidence that patients opting for surgery benefit from two cycles of neoadjuvant cisplatin 5-FU. The evidence for preoperative chemoradiotherapy is less secure, and there is also little evidence that surgery is of benefit in patients responding to chemo-radiotherapy. For patients with localized cancers of the cervical oesophagus, chemo-radiotherapy is the preferred treatment. Overall, the 5-year survival for patients with operable disease is of the order of 20–25 per cent. For fit patients with locally advanced disease, chemo-radiotherapy should be considered. Patients with metastatic disease have a median survival of 4–8 months. Palliative chemotherapy can offer those with a good performance status enhanced symptom-free survival and a small increase in median survival, of the order of 4–6 months. Local therapy with stenting, laser therapy, PDT and radiotherapy can offer significant symptomatic benefit for selected patients. All patients should have access to specialist palliative care when necessary.
REFERENCES 1 Office for National Statistics, registrations of cancer diagnosed in 1993–1996, England and Wales. Health Statistics Quarterly 1999; 4:59–70. 2 Office for National Statistics, Mortality Statistics, Cause. London: ONS, 1999. 3 Devesa S, Blot W, Fraumeni J. Changing patterns in the incidence of esophageal and gastric cancer in the United States. Cancer 1998; 83:2049–53. 4 Faivre J, Forman D, Esteve J, et al. The Eurocare Working Group. Survival of patients with oesophageal and gastric cancers in Europe. Eur J Cancer 1998; 34:2167–75. 5 Parkin DM, Laara E, Muir CS. Estimates of the world-wide frequency of sixteen major cancers in 1980. Int J Cancer 1988; 41:184–97. 6 Yang CS. Research on esophageal cancer in China: a review. Cancer Res 1980; 40:2633–44. 7 Blot WJ. Esophageal cancer trends and risk factors. Semin Oncol 1994; 21:403–10. 8 Engel LS, Chow WH, Vaughan TL, et al. Population attributable risks of esophageal and gastric cancers. J Natl Cancer Inst 2003; 95:1404–13. 9 Lagergren J, Bergstrom R, Lindgren A, Nyren O. Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. N Engl J Med 1999; 340:825–31. 10 Gammon MD, Schoenberg JB, Ahsan H, et al. Tobacco, alcohol, and socioeconomic status and adenocarcinomas of the esophagus and gastric cardia. J Natl Cancer Inst 1997; 89:1277–84. 11 Siddiqi M, Tricker AR, Preussmann R. The occurrence of preformed N-nitroso compounds in food samples from a high risk area of esophageal cancer in Kashmir, India. Cancer Lett 1988; 39:37–43. 12 Pickwell SM, Schimelpfening S, Palinkas LA. ‘Betelmania’. Betel quid chewing by Cambodian women in the United States and its potential health effects. West J Med 1994; 160:326–30. 13 Ghadirian P. Thermal irritation and esophageal cancer in northern Iran. Cancer 1987; 60:1909–14. 14 Sandler RS, Nyren O, Ekbom A, et al. The risk of esophageal cancer in patients with achalasia. A population-based study. JAMA 1995; 274:1359–62. 15 Iwaya T, Maesawa C, Ogasawara S, Tamura G. Tylosis esophageal cancer locus on chromosome 17q25.1 is commonly deleted in sporadic esophageal cancer. Gastroenterology 1998; 114:1206–10. 16 Thorban S, Roder JD, Nekarda H, et al. Immunocytochemical detection of disseminated tumor cells in the bone marrow of patients with esophageal carcinoma. J Natl Cancer Inst 1996; 88:1222–7. 17 Hansen S, Wiig J, Giercksky K, et al. Esophageal and gastric cancer in Norway 1958–1992: incidence time trend variability according to morphological subsites and organ subsites. Int J Cancer 1997; 71:340–4. 18 Lagergren J. Oesophageal cancer and gastro-oesophageal reflux: what is the relationship? Gut 2004; 53(8):1064–5.
556 Oesophageal cancer
19 Cameron AJ, Lomboy CT, Pera M, Carpenter HA. Adenocarcinoma of the esophagogastric junction and Barrett’s esophagus. Gastroenterology 1995; 109:1541–6. 20 Lagergren J, Bergstrom R, Nyren O, et al. Association between body mass and adenocarcinoma of the esophagus and gastric cardia. Ann Intern Med 1999; 130:883–90. 21 Speckler SJ. Strategies in prevention screening and diagnosis of high grade dysplasia in Barrett’s esophagus. ASCO Gastrointesinal Cancers Symposium 2000; 35 22 Paraf F, Flejou JF, Pignon JP, et al. Surgical pathology of adenocarcinoma arising in Barrett’s esophagus. Am J Surg Pathol 1995; 19:183–91. 23 Johansson J, Johnsson F, Walther B, et al. Adenocarcinoma in the distal esophagus with and without Barrett’s esophagus. Differences in symptoms and survival rates. Arch Surg 1996; 131:708–13. 24 Lieberman MD, Shriver CD, Bleckner S, Burt M. Carcinoma of the esophagus: prognostic significance of histologic type. J Thorac Cardiovasc Surg 1995; 109:130–8. 25 Sharma P, McQuaid K, Dent J, et al. A critical review of the diagnosis and management of Barrett’s esophagus: the AGA Chicago Workshop. Gastroenterology 2004; 127:310. ✽26 Upper GI Cancer Improving Outcomes Guidance. London: Department of Health, 2001. 27 Rationale of staging in cancer of the esophagus. In: Delarue N, Wilkins EW, Wong J (eds), International Trends in General Thoracic Surgery: Esophageal Cancer, Vol. 4. St Louis, MI: CV Mosby, 1988. 28 McLoughlin RF, Cooperberg PL, Mathieson JR, et al. High resolution endoluminal ultrasonography in the staging of esophageal carcinoma. J Ultrasound Med 1995; 14:725–30. 29 Romagnuolo J, Scott J, Hawes RH, et al. Helical CT versus EUS with fine needle aspiration for celiac nodal assessment in patients with esophageal cancer. Gastrointest Endosc 2002; 55:648–54. 30 Hunerbein M, Ghadimi BM, Haensch W, Schlag PM. Transendoscopic ultrasound of esophageal and gastric cancer using miniaturized ultrasound catheter probes. Gastrointest Endosc 1998; 48:371–5. 31 Mallery S, Van Dam J. Increased rate of complete EUS staging of patients with esophageal cancer using the nonoptical, wire-guided echoendoscope. Gastrointest Endosc 1999; 50:53–7. 32 Zuccaro G, Rice TW, Goldblum J, et al. Endoscopic ultrasound cannot determine suitability for esophagectomy after aggressive chemoradiotherapy for esophageal cancer. Am J Gastroenterol 1999; 94:906–12. 33 Flanagan FL, Dehdasht F, Siegel BA, et al. Staging of esophageal cancer with 18F-fluorodeoxyglucose positron emission tomography. Am J Roentgenol 1997; 168:417–24. 34 Block MI, Patterso GA, Sundaresan, RS, et al. Improvement in staging of esophageal cancer with the addition of positron emission tomography. Ann Thorac Surg 1997; 64:770. 35 Luketich JD, Friedman DM, Weigel TL, et al. Evaluation of distant metastases in esophageal cancer: 100 consecutive
36
37 38
●39
40
41 ●42
●43
44 45
46 47
48
49
50 51
52
53
positron emission tomography scans. Ann Thorac Surg 1999; 68:1133–6. Flamen P, Lerut A, Van Cutsem E, et al. Utility of positron emission tomography for the staging of patients with potentially operable esophageal carcinoma. J Clin Oncol 2000; 18:3202–10. Sobin LH, Wittekind C (eds). TNM Classification of Malignant Tumours, 6th edn. New York: Wiley, 2002. Hofstetter W, Swisher SG, Correa AM, et al. Treatment outcomes of resected esophageal cancer. Ann Surg 2002; 236:376–84. Lewis I. The surgical treatment of carcinoma of the esophagus: special reference to a new operation for growth of the middle third. Br J Surg 1946; 33:19. Griffin SM, Shaw IH, Dresner SM. Early complications after Ivor Lewis subtotal esophagectomy with two-field lymphadenectomy: risk factors and management. J Am Coll Surg 2002; 194:285–97. Ellis FH Jr. Standard resection for cancer of the esophagus and cardia. Surg Oncol Clin North Am 1999; 8:279. Orringer MB, Marshall B, Iannettoni MD. Transhiatal esophagectomy: clinical experience and refinements. Ann Surg 1999; 230:392–400. Urba S, Orringer M, Turrisi A, et al. A randomised trial comparing surgery to pre-operative concomitant chemoradiation plus surgery in patients with respectable esophageal cancer. Update analysis. Proc ASCO 1997; 6:227. Collard JM, Otte JB, Fiasse R, et al. Skeletonizing en bloc esophagectomy for cancer. Ann Surg 2001; 234:25–32. Igaki H, Kato H, Tachimori Y, et al. Prognostic evaluation for squamous cell carcinomas of the lower thoracic esophagus treated with three-field lymph node dissection. Eur J Cardiothorac Surg 2001; 19:88. Law S, Wong J. Two-field dissection is enough for esophageal cancer. Dis Esophagus 2001; 14:98–103. Altorki N, Kent M, Ferrara C, Port J. Three-field lymph node dissection for squamous cell and adenocarcinoma of the esophagus. Ann Surg 2002; 236:177–83. Dresner SM, Griffin SM. Pattern of recurrence following radical oesophagectomy with two-field lymphadenectomy. Br J Surg 2000; 87:1426–33. Mendenhall WM, Sombeck MD, Parsons JT, et al. Management of cervical esophageal carcinoma. Semin Radiat Oncol 1994; 4:179–91. Collin CF, Spiro RH. Carcinoma of the cervical esophagus: changing therapeutic trends. Am J Surg 1984; 148:460–6. Triboulet JP, Mariette C, Chevalier D, Amrouni H. Surgical management of carcinoma of the hypopharynx and cervical esophagus: analysis of 209 cases. Arch Surg 2001; 136:1164–1170. Stuschke M, Stahl M, Wilke H, et al. Induction chemotherapy followed by concurrent chemotherapy and high-dose radiotherapy for locally advanced squamous cell carcinoma of the cervical oesophagus. Oncology 1999; 57:99–105. Bidoli P, Bajetta E, Stani SC, et al. Ten-year survival with chemotherapy and radiotherapy in patients with squamous cell carcinoma of the esophagus. Cancer 2002; 94:352–61.
References 557
54 Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002; 346:1128–37. 55 Bachman M, Alderson D, Peters T, et al. Survival, clinical practice and costs in patients with pancreatic, oesophageal and gastric cancers: influence of doctor and hospital volumes. Repost to NHS National Cancer Research and Development Programme, 1999. 56 Gillison EW, Powell J, McConkey CC, Spychal RT. Surgical workload and outcome after resection for carcinoma of the oesophagus and cardia. Br J Surg 2002; 89:344–8. 57 Minsky BD. Carcinoma of the esophagus. Part 2: Adjuvant therapy. Oncology 1999; 13:1415–27. 58 Kasai M, Mon S, Watanabe T. Follow-up results after resection of oesophageal cancer. World J Surg 1978; 2:543–51. ●59 Fok M, Law S, Stipa F, et al. A comparison of transhiatal and transthoracic resection for oesophageal carcinoma. Endoscopy 1993; 25:660–3. ●60 Teniere P, Hay JM, Fingerhut A, et al. Postoperative radiation therapy does not increase survival after curative resection for squamous cell carcinoma of the middle and lower third esophagus as shown by a multicenter controlled trial. French University Association for Surgical Research. Surg Gynaecol Obstet 1991; 173:123–30. ●61 Bancewicz J, Clark PI, Smith DB, et al. Surgical resection with or without preoperative chemotherapy in oesophageal cancer: a randomised controlled trial. Lancet 2002; 359:1727–33. 62 Kok TC, von Lanschot J, Siersema PD, et al. Neoadjuvant chemotherapy in operable esophageal squamous cell cancer: final report of a phase III multicenter randomized controlled trial (abstract). Proc Am Soc Clin Oncol 1997; 16:277. ●63 Kelsen DP, Ginsberg R, Pajak TF, et al. Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med 1998; 339:1979–84. ●64 Cunningham D, Allum WH, Stenning SP, et al.. Perioperative chemotherapy in operable gastric and lower esophageal cancer: final results of a randomised, controlled trial (the MAGIC trial, ISRCTN 93793971) (abstract). J Clin Oncol 2005; 23:308s. ●65 Herskovic A, Martz K, al-Sarraf M, et al. Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 1992; 326:1593–8. ●66 Bosset JF, Gignoux M, Triboulet JP, et al. Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N Engl J Med 1997; 337:161–7. ●67 Urba SG, Orringer MB, Turrisi A, et al. Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J Clin Oncol 2001; 19:305–13. 68 Nygaard K, Hagen S, Hansen HS, et al. Preoperative radiotherapy prolongs survival in operable esophageal carcinoma: a randomized, multicenter study of preoperative
69
●70
●71
72
73
74
75
76
●77
●78
79
80
81
82
●83
radiotherapy and chemotherapy – the second Scandinavian trial in esophagus cancer. World J Surg 1992; 16:1104–9. Le Prise E, Etienne PL, Meunier B, et al. A randomized study of chemotherapy, radiation therapy, and surgery versus surgery for localized squamous cell carcinoma of the esophagus. Cancer 1994; 73:1779–84. Burmeister BH, Smithers BM, Gebski V, et al. Surgery alone versus chemoradiotherapy followed by surgery for resectable cancer of the oesophagus: a randomised controlled phase III trial. Lancet Oncol 2005; 6:659–68. Walsh TN, Noonan N, Hollywood D, et al. A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N Engl J Med 1996; 335:462–7. Urschel JD, Vasan H. A meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for resectable esophageal cancer. Am J Surg 2003; 185:538–43. Liedman B, Johnsson E, Merke C, et al. Preoperative adjuvant radiochemotherapy may increase the risk in patients undergoing thoracoabdominal esophageal resections. Dig Surg 2001; 18:169–75. Lin FC, Durkin AE, Ferguson MK. Induction therapy does not increase surgical morbidity after esophagectomy for cancer. Ann Thorac Surg 2005; 78:1783–9. Stahl M, Stuschke M, Lehmann N, et al. Chemoradiation with and without surgery in patients with locally advanced squamous cell carcinoma of the esophagus. J Clin Oncol 2005; 23:2310–17. Bedenne L, Michel P, Bouche O, et al. Randomized phase III trial in locally advanced esophageal cancer: radiochemotherapy followed by surgery versus radiochemotherapy alone (abstract). Proc Am Soc Clin Oncol 2002; 21:130a. Earlam R, Cunha-Melo JR. Oesophogeal squamous cell carcinoma: II. A critical view of radiotherapy. Br J Surg 1980; 67:457–61. Sykes AJ, Burt PA, Slevin NJ, et al. Radical radiotherapy for carcinoma of the esophagus: an effective alternative to surgery. Radiother Oncol 1998; 48:15–21. Shioyama Y, Nakamura K, Sasaki T, et al. Clinical results of radiation therapy for stage I esophageal cancer: a single institutional experience. Am J Clin Oncol 2005; 28:75–80. Hay JH, Flores AD, Haylock BJ, et al. A randomised trial comparing brachytherapy given before and after external beam irradiation for carcinomas of the esophagus and cardia. Proc of the Am Brachy Soc 1998. Yin WB. Brachytherapy of carcinoma of the esophagus in China, 1970–1974 and 1982–1984. In: Martinez AA, Orton CG, Mould RF (eds), Brachytherapy: HDR and LDR. Columbia, MD: Nucleotron Corp., 1990, 52. Sur RK, Singh DP, Sharma SC, et al. Radiation therapy of esophageal cancer: role of high dose rate brachytherapy. Int J Radiat Oncol Biol Phys 1992; 22:1043–6. Tate T, Brace JA, Morgan H, Skeggs DBL. Conformation therapy in the treatment of carcinoma of the oesophagus. Clin Radiol 1986; 37:267–71.
558 Oesophageal cancer
●84
85
86
87
●88
89
90
91 92
●93
94
95
Rowlands CG, Pagliero KM. Intracavitary irradiation in the palliation of carcinoma of the oesophagus and cardia. Lancet 1985; 2:981–3. Muto M, Ohtsu A, Miyamoto S, et al. Concurrent chemoradiotherapy for esophageal carcinoma patients with malignant fistulae. Cancer 1999; 86:1406–13. O’Rourke IC, Tiver K, Bull C, et al. Swallowing performance after radiation therapy for carcinoma of the esophagus. Cancer 1988; 61:2022–6. Corti L, Skarlatos J, Boso C, et al. Outcome of patients receiving photodynamic therapy for early esophageal cancer. Int J Radiat Oncol Biol Phys 2000; 47:419–24. Cooper JS, Guo MD, Herskovic A, et al. Chemoradiotherapy of locally advanced esophageal cancer: long-term follow-up of a prospective randomized trial (RTOG 85-01). Radiation Therapy Oncology Group. JAMA 1999 May 5; 281(17):1623–7. Seitz JF, Giovannini M, Padaut-Cesana J, et al. Inoperable nonmetastatic squamous cell carcinoma of the esophagus managed by concomitant chemotherapy (5-fluorouracil and cisplatin) and radiation therapy. Cancer 1990; 66:214–19. Wobbes T, Baron B, Paillot B, et al. Prospective randomised study of split-course radiotherapy versus cisplatin plus splitcourse radiotherapy in inoperable squamous cell carcinoma of the oesophagus. Eur J Cancer 2001; 37:470–7. Choy H. Combining taxanes with radiation for solid tumors. Int J Cancer 2000; 90:113–27. Adelstein DJ, Rice TW, Rybicki LA, et al. Does paclitaxel improve the chemoradiotherapy of locoregionally advanced esophageal cancer? A nonrandomized comparison with fluorouracil-based therapy. J Clin Oncol 2000; 18:2032–9. Minsky BD, Berkey B, Kelsen DK, et al. Preliminary results of Intergroup INT 0123 randomized trial of combined modality therapy (CMT) for esophageal cancer: standard vs high dose radiation therapy. Proc Am Soc Clin Oncol 2000; 19:239a. Caspers RJ, Welvaart K, Verkes RJ, et al. The effect of radiotherapy on dysphagia and survival in patients with esophageal cancer. Radiother Oncol 1988; 12:15–23. Homs MY, Steyerberg EW, Eijkenboom WM, et al. Singledose brachytherapy versus metal stent placement for the
96
97
98
99
100
101
102
103 104
●105
palliation of dysphagia from oesophageal cancer: multicentre randomised trial. Lancet 2004; 364:1497–504. Sur RK, Levin CV, Donde B, et al. Prospective randomized trial of HDR brachytherapy as a sole modality in palliation of advanced esophageal carcinoma – an International Atomic Energy Agency study. Int J Radiat Oncol Biol Phys 2002; 53:127–33. Lovett D, Kelsen D, Eisenberger M, Houston C. A phase II trial of carboplatin and vinblastine in the treatment of advanced squamous cell carcinoma of the esophagus. Cancer 1991; 67:354–6. Bleiberg H, Conroy T, Paillot B, et al. Randomised phase II study of cisplatin and 5-fluorouracil (5FU) versus cisplatin alone in advanced squamous cell oesophageal cancer. Eur J Cancer 1997; 33:1216–20. Warner E, Jensen JL, Cripps C, et al. Outpatient 5-fluorouracil, folinic acid and cisplatin in patients with advanced esophageal carcinoma. Acta Oncol 1999; 38:255–9. Kok TC, Van der Gaast A, Dees J, et al. Cisplatin and etoposide in oesophageal cancer: a phase II study. Rotterdam Oesophageal Tumour Study Group. Br J Cancer 1996; 74:980–4. Polee MB, Eskens FA, van der Burg ME, et al. Phase II study of bi-weekly administration of paclitaxel and cisplatin in patients with advanced oesophageal cancer. Br J Cancer 2002; 86:669–73. Heit HA, Johnson LF, Siegel SR, Boyce HW Jr. Palliative dilation for dysphagia in esophageal carcinoma. Ann Intern Med 1978; 89:629–31. Haddad NG, Fleischer DE. Endoscopic laser therapy for esophageal cancer. Gastrointest Endosc 1994; 4:863–74. Yano T, Muto M, Minashi K, et al. Photodynamic therapy as salvage treatment for local failures after definitive chemoradiotherapy for esophageal cancer. Gastrointest Endosc 2005; 62:31–6. Siersema PD, Schrauwen SL, van Blankenstein M, et al. Self-expanding metal stents for complicated and recurrent esophagogastric cancer. Gastrointest Endosc 2001; 54:579–86.
22 Liver DANIEL H. PALMER AND PHILIP J. JOHNSON
Introduction Incidence Hepatocellular carcinoma
559 559 560
INTRODUCTION In many parts of the world, particularly sub-Saharan Africa and the Far East, hepatocellular carcinoma (HCC) is a major public health problem. Although progress towards effective therapy has been slow, low-cost vaccines against the hepatitis B virus and anti-viral treatment may now be starting to reduce the incidence of this malignancy. Worldwide, the liver is the most common, and often the clinically predominant, site of metastatic cancer. After considering briefly the aetiological factors and the clinicopathological features of HCC, this chapter goes on to describe the possible modes of therapy. These have been grouped broadly into surgery, locoregional treatments and cytotoxic, and other systemic, chemotherapy. Hepatocellular carcinoma will be considered first, then selected aspects of metastatic liver disease. Since management of hepatic metastases from primary carcinoid tumours may pose additional problems, because of the associated carcinoid syndrome, the possible therapeutic approaches are discussed separately.
Metastatic liver disease Neuroendocrine tumours metastatic to the liver References
567 567 569
Table 22.1 Typical age-adjusted incidence rates per 100 000 of population in various countries Country Low-incidence areas
UK USA (white) Australia Germany Denmark
Intermediate-incidence areas
High-incidence areas
Male
Female
1.6 2 1.1 4 3.6
0.8 1 0.5 1.2 2.3
Italy Spain Romania Argentina
7.5 7.5 11.8 8
3.5 4 7.9 5
Japan Hong Kong Zimbabwe Senegal Taiwan
20 32 65 25 85
5 7 25 9 –
Many registries do not distinguish between hepatocellular carcinoma and other primary liver tumours.
INCIDENCE Hepatocellular carcinoma is one of the most common malignant tumours in the world today. The highest annual incidence rates, of around 100 per 100 000, occur in parts of southern Africa and the Far East (Table 22.1).1 Some estimates suggest that male Chinese carriers of the hepatitis B virus (HBV), who may comprise up to 15 per cent of males in certain populations, carry a lifetime risk of developing the tumour of over 20 per cent. In contrast, HCC is
much less common in northern Europe, the USA and Australia, although there is some evidence that the frequency may be increasing, perhaps in relation to the spread of chronic hepatitis C virus (HCV) infection.2 In most series, between 10 per cent and 20 per cent of primary liver tumours are accounted for by cholangiocarcinoma (see Chapter 24) and about 1 per cent by primary hepatic sarcomas. Liver metastases are found at approximately 1 per
560 Liver
cent of all autopsies and the liver is involved in up to 40 per cent of adult patients with primary extra-hepatic malignancies who come to autopsy. Up to 75 per cent of primary tumours drained by the portal venous system will have spread to involve the liver before death occurs.3 As such, the liver is the most common organ to be involved.
HEPATOCELLULAR CARCINOMA Aetiology, risk factors and prevention In view of the limited therapeutic options that are available once HCC has developed, the identification of risk factors (Box 22.1), which may in turn lead to interventional strategies that reduce its incidence and form the basis of a screening programme, is of particular importance. The wide geographic variation in the incidence of HCC, as outlined above, has pointed to the likely importance of environmental factors. Prime among these have been the HBV and HCV and exposure to aflatoxin. Beasley4 followed up more than 22 000 Chinese males, 15 per cent of whom were HBV carriers, for up to 9 years. The development of HCC was almost exclusively confined to those who had serum markers of HBV infection at presentation, and the relative risk for HCC development was almost 100. An HBV vaccination programme was initiated in Taiwan in the early 1980s, first inoculating neonates born to carrier mothers, and then extended to all neonates and children. The incidence (per 100 000 population) of HCC in children aged 6–9 years declined from 0.52 for those born between 1974 and 1984 to 0.13 for those born between 1984 and 1986.5 Preliminary evidence also suggests that anti-viral therapy in patients with chronic hepatitis B may also decrease the incidence of subsequent HCC development.6
Box 22.1 Major risk factors for hepatocellular carcinoma
chronic liver disease that chronic HCV infection may cause (see below). Hepatocellular carcinoma usually arises in a cirrhotic liver, although, particularly in high-incidence areas, the cirrhosis is not always symptomatic and, indeed, the development of HCC may be the first indication of the underlying cirrhosis. Males with cirrhosis are affected significantly more often than females.9 Aflatoxin, formed by the fungus Aspergillus flavus, which grows on cereals stored under damp conditions, is one of the most potent hepatic carcinogens known. In several high-HCC-incidence areas of the world, a clear relationship between intake and the incidence of HCC has been established, both by conventional dietary assessment10 and, more recently, by the use of biomarkers.11,12 As well as the obvious approach of improving grain storage to overcome the problem of aflatoxin exposure, the possibility of chemoprevention is an area of active research.13,14
Anatomy of the liver Recognition of the segmental anatomy of the liver has been important in allowing more refined operative techniques for surgical resection.15 The portal vascular supply and the bile ducts define two functional hepatic lobes, the line of demarcation running from the gall-bladder bed to the inferior vena cava. Each of these two lobes is split into four segments, none of which has any surface markings (Fig. 22.1). The standard surgical resections are based on this description (Fig. 22.2). The left, right and middle hepatic veins drain the conventional ‘anatomical’ left and right lobes, as defined by the falciform ligament, respectively. It has been recognized for more than half a century that whereas the normal liver receives about 70 per cent of its blood supply from the portal vein, the dominant blood supply to liver tumours is from the hepatic artery. This forms the rationale for several therapeutic approaches, including hepatic artery occlusion and regional chemotherapy (see ‘Chemoembolization’ below).
Chronic liver disease (usually at the stage of cirrhosis) Chronic viral hepatitis Type B Chronic hepatitis Type C Dietary exposure to aflatoxin Increasing age Male gender
VIII
II
VII III
The epidemiological evidence linking chronic HCV infection and HCC is similar to that for HBV. Indeed, in Europe, Japan and the USA, the attributable risk may be even higher.7,8 However, since HCV has no reverse transcriptase activity and is not a retrovirus, it should, in theory, have no direct oncogenic potential. Most authors have therefore attributed the association with HCC to the
VI
V
IV
Gall bladder Principal plane
Figure 22.1
Ligamentum teres
Portal Umbilical traid fissure
Segmental anatomy of the liver.
Hepatocellular carcinoma 561
Symptoms and presentation The most common mode of presentation for primary and secondary liver tumours is in the triad of abdominal pain, weight loss and the presence of hepatomegaly, often massive. In addition, patients with HCC often present with signs of hepatic decompensation, such as ascites or variceal haemorrhage.16 A particularly dramatic presentation is spontaneous rupture of the tumour: there is a sudden onset of severe abdominal pain with shock, and paracentesis reveals bloodstained ascitic fluid.17 Rarer presentations of HCC include hypoglycaemia, hypercalcaemia and polycythaemia. Intra-hepatic (‘peripheral type’) cholangiocarcinomas have no specific clinical features that distinguish them from HCC, but those arising at the bifurcation (‘hilar type’), and below, present with obstructive jaundice, weight loss and abdominal pain.18 Malignant liver tumours are increasingly diagnosed pre-symptomatically. In the case of HCC, this follows screening patients with cirrhosis by serial estimations of alpha-fetoprotein (AFP) and/or ultrasound examination. Certain metastatic tumours, particularly neuroendocrine tumours (see ‘Neuroendocrine tumours’ below), involve the liver with relatively little disruption of hepatic function. Indeed, patients with large tumours may have normal liver function tests and are often surprisingly well despite extensive tumour deposits.
Diagnosis Either dynamic triphasic computed tomography (CT) or gadolinium-enhanced magnetic resonance imaging (MRI)
will classically show marked enhancement in the arterial phase with relative hypovascularity (‘wash-out’) in the portal or late phases. In a patient known to have cirrhosis and in the presence of a mass greater than 2 cm in diameter, this radiological appearance (confirmed by two imaging modalities) is now regarded as being diagnostic of HCC without the need for histological confirmation.19 The presence of an enhancing intra-hepatic mass and an AFP level of greater than 400 ng/mL are also diagnostic of HCC if only one imaging modality is characteristic. Cross-sectional imaging is also required to assess the extent of disease so that treatment can be planned. In particular, size, number and distribution of tumours can be established, as well as the presence of macrovascular (portal vein) invasion and extra-hepatic disease. Although contrast-enhanced CT and MRI are the best current imaging modalities, both techniques may miss up to 30 per cent of lesions (as detected in the explanted liver following liver transplantation), especially those less than 1 cm.20 The first serologic assay for the detection and clinical follow-up of patients with HCC was AFP. Serum AFP is elevated in 50–70 per cent of HCCs.21 It is of value in the diagnosis of HCC in patients with cirrhosis and has been used as a screening tool in high-risk populations (see ‘Screening’ below). However, levels of up to 500 ng/mL can occur in other, benign, liver diseases, especially chronic active hepatitis and fulminant hepatic failure. Nevertheless, a rising AFP is strongly suggestive of HCC. Alpha-fetoprotein is also useful in monitoring the effects of treatment and for the detection of disease recurrence or progression following treatment. It may be of prognostic value, higher values being associated with a worse prognosis (see ‘Staging and prognosis’).
Non-anatomical resection
Right trisegmentectomy
Left trisegmentectomy
Right lobectomy
Left lobectomy
Left lateral segmentectomy
Figure 22.2 Standard surgical resections.
562 Liver
Screening In the light of the limited therapeutic options for HCC once the lesion is greater than 4–5 cm in diameter and the fact that high-risk groups can now be identified, screening (or, more strictly, surveillance) seems a logical approach and there is some limited evidence that it does indeed decrease disease-specific mortality.23 Current guidelines suggest that those in a high-risk group undergo 6-monthly ultrasound examinations. Serial AFP estimation has also been used as a screening test, but current evidence suggests that it is insufficiently sensitive to be used as a screening procedure.19,24
Histology For patients with a liver mass not fulfilling the European Association for Study of the Liver (EASL) criteria (as defined above), the diagnosis of HCC must be made histologically. Fine-needle biopsy may be limited by sampling error/missing small lesions, and by difficulty in distinguishing well-differentiated HCC from dysplasia or from adenoma or even normal liver. Some groups believe that biopsy may risk tumour seeding along the needle track and recommend avoidance of the procedure in candidates for surgical resection or transplantation. The presence of bile in tumour cells is diagnostic of HCC. There are several histological subtypes. The most typical pattern is microtrabecular, which differs from normal liver by the absence of portal tracts or bile ducts. Other patterns include acinar, pseudoglandular and scirrhous, but histological type is not of prognostic significance, with the exception of the fibrolamellar variant characterized by tumour cords with collagen strands. Fibrolamellar HCC typically occurs in younger patients (mean 26 years) without underlying chronic liver disease and AFP is not elevated.25,26 In this setting, resection rates are higher and prognosis is better (median survival 5 years), although this may reflect the younger age, absence of cirrhosis and low AFP, all of which are associated with a better prognosis, rather than being related to the histological subtype.27
Natural history Symptomatic HCC is usually rapidly fatal and most untreated patients die within 12 months of the onset of symptoms. In Africa and the Far East, the tumour appears to behave in a particularly aggressive manner. Geddes and
Falkson28 reported a mean duration of symptoms of 5 months before death in South African Bantu mineworkers; in China, the mean survival time from symptoms/diagnosis is less than 3 months. In the West, patients without underlying cirrhosis survive longer16 (Fig. 22.3) and survival up to 3 years may occur, particularly in patients who present without symptoms, vascular invasion or extrahepatic metastases.29 The prognosis of patients with other primary liver tumours is equally poor. The only exceptions are the rare ‘epitheloid haemangioendothelioma’ and the equally rare variant of HCC fibrolamellar carcinoma (see above). Both have a rather better prognosis, with a median survival of around 5 years. As local treatment becomes more effective, symptomatic lung, bone and lymph-node metastases are more frequently detected. Untreated, the overall median survival for patients with secondary liver tumours is in the range of 6–9 months, with less than 1 per cent of patients surviving for more than 5 years. Metastases from primary carcinoid tumours are the only secondary tumours to exhibit a significantly better prognosis, survival periods of up to 10 years not being uncommon. With the wide use of more sophisticated radiological techniques, the diagnosis of liver metastases is being established earlier, often while the patient is asymptomatic, and this makes for an apparent increase in survival. As a rule, the major factor influencing survival is the percentage of liver involved by the tumour.
Staging and prognosis For most cancers, prognosis is predominantly related to tumour stage. However, most HCCs occur on a background of cirrhosis that independently contributes to prognosis.30 For the majority of patients, the prognosis of HCC remains poor, with treatments other than surgery having little impact on survival. TNM staging for HCC is
100
75 Survival (%)
The other commonly used assay is for des-gamma-carboxy prothrombin protein (PIVKA-II). Levels of this protein are increased in up to 90 per cent of patients with HCC, but may also be elevated in patients with vitamin K deficiency, chronic active hepatitis or metastatic carcinoma.22
50
Cirrhosis
No cirrhosis
25
0 6
12
18
24
30
Months
Figure 22.3 Cumulative survival curves for patients with hepatocellular carcinoma with and without cirrhosis. (Reproduced with permission from Melia et al., 1984.)
Hepatocellular carcinoma 563
summarized in Table 22.2. Tumour size is a key prognostic factor, with survival approaching 3 years for tumours less than 3 cm but only 3 months for those larger than 8 cm.31 Vascular invasion increases with tumour size, but is an independent prognostic factor; even large tumours can have a good prognosis following surgical resection in the absence of vascular invasion.32,33 Underlying cirrhosis may limit prognosis independently of tumour-related factors, and the degree of liver dysfunction influences treatment options such as surgical resection and chemoembolization, which require sufficient hepatic reserve to be performed
Table 22.2 TNM staging T1 T2 T3 T4 N1 M1
Single tumour, no vascular invasion (VI) Single tumour VI, or multiple tumour none 5 cm Multiple tumours any 5 cm, involvement of major branch of PV or HV Direct invasion of adjacent organs (except GB) or perforation of visceral peritoneum Regional lymph-node involvement Distant metastases
Stage I II IIIA IIIB IIIC IV
T
N
M
T1 T2 T3 T4 Any T Any T
N0 N0 N0 N0 N1 Any N
M0 M0 M0 M0 M0 M1
From reference 36. PV, portal vein; HV, hepatic vein; GB, gallbladder.
safely. Prognostic models for HCC are therefore complex and should take into account tumour stage, degree of liver impairment, patient fitness and treatment efficacy. Systems omitting any of these are likely to be of limited value (e.g. the Child–Pugh score takes into account only the underlying liver function, and TNM only tumourrelated factors). Attempts have been made to refine this system by incorporating more detailed tumour information and liver function (e.g. the Cancer of the Liver Italian Programme [CLIP] and Chinese University Prognostic Index (CUPI)), but they are still not widely applied (Table 22.3).34,35 Following surgical resection, a key determinant of prognosis remains, as for other tumours, the TNM tumour stage.36 Nevertheless, this can still be refined by the inclusion of other tumour-related and patient-related factors and may be further improved by emerging biological markers. Further, TNM can only be applied once a patient has been deemed resectable, which has already taken into account liver function and performance status, themselves markers of prognosis.
Treatment Resection, transplantation and local ablation may be considered potentially curative treatments, but none of these has been compared to an untreated group in large randomized controlled trials and evidence for benefit is based on comparison with historical controls. Thus, the best 5-year survival rate in untreated series is less than 20 per cent, whereas with transplantation, resection or ablations the comparable figures range from 40 per cent to 70 per cent.
Table 22.3 Prognostic scoring systems Tumour factors
Host factors
Environmental factors
TNM Child–Pugh
See Table 22.1 Nil
Nil Bilirubin, albumin, INR, ascites, encephalopathy
Nil Nil
Okuda
Tumour size
Bilirubin, albumin, ascites
Nil
CLIP
Morphology (uni/multi-nodular AFP, portal vein thrombosis
Child–Pugh score
Nil
CUPI
TNM, AFP
Bilirubin, ascites, alkaline phosphate, symptoms
Nil
BCLC
Tumour size, symptoms morphology, extra-hepatic disease
Portal hypertension, Child–Pugh score, bilirubin, performance status, co-morbidity
Surgery, ablation, TACE, trials, supportive care
CLIP, Cancer of the Liver Italian Programme; CUPI, Chinese University Prognostic Index; BCLC, Barcelona Clinic Liver Cancer; AFP, alpha-fetoprotein; INR, International Normalisation Ratio; TACE, transarterial chemoembolization.
564 Liver
LIVER RESECTION
The ability to resect a tumour depends on its size, location, relation to blood vessels and underlying liver function. In patients without cirrhosis, up to 75 per cent of the liver can be removed safely and hepatic resection is clearly the treatment of choice in this situation. However, resection in patients with cirrhosis is still associated with significant morbidity and mortality, although this has fallen to less than 5 per cent with improvements in surgical technique, postoperative management and anatomical resections combined with better patient selection. Nonetheless, only 10–20 per cent of patients are suitable for resection, although this may increase as screening programmes detect tumour at an earlier stage. The use of preoperative portal vein embolization of the branches supplying the potential resected liver, thereby expanding the volume of the remnant, is now widely used and may extend the limits of surgical resection. Survival is better for small (5 cm), solitary tumours with negative resection margins and an absence of vascular invasion or lymph-node involvement, with 5-year survival rates up to 70 per cent.37,38 TRANSPLANTATION
Liver transplantation has the potential to treat both tumour and, where present, the underlying cirrhosis. However, for patients with viral hepatitis there is a risk of re-infection of the new liver. In particular for HCV, response to anti-viral drugs is lower than in untransplanted patients, and if infection persists, liver damage and cirrhosis tend to occur more rapidly than in a native liver. Accurate preoperative assessment of tumour size and number, vascular involvement and extra-hepatic disease is essential to identify those patients who are most likely to benefit from long-term survival without tumour recurrence. In patients with no more than three small (5 cm) tumours, survival was similar to that of patients with benign end-stage liver disease, and transplantation has become the treatment of choice for HCC in a cirrhotic liver. This experience has underpinned the development of the Milan criteria (Table 22.4) to guide the selection of patients for transplantation, which can lead to 5-year survival in excess of 70 per cent.39 Recent reports have indicated that these criteria may be extended whilst retaining good outcomes (University California San Francisco (UCSF) criteria, Table 22.4).40 A major limitation to transplantation is the supply of donor organs. This results in a period, of uncertain duration, between listing and transplantation, with the risk that the tumour will grow beyond the indications for transplant.41 For this reason, in North America, the scoring systems used to allocate donor organs equitably are weighted to prioritize patients with HCC for transplant.42 Drop-out (i.e. disease progression to a level at which transplantation is no longer appropriate whilst awaiting a donor liver) may be best reduced by increasing the number/availability of donor organs, and this may be helped by the use of living
Table 22.4 Eligibility criteria for liver transplantation in patients with hepatocellular carcinoma Milan criteria One tumour 5 cm or Up to 3 nodules, each 3 cm
UCSF criteria One tumour 6.5 cm or Up to 3 nodules, each 4.5 cm and total tumour diameter 8 cm
UCSF, University California San Francisco.
donors. Early data indicate similar results to those achieved with cadaveric organs if the Milan criteria are followed, with low mortality amongst donors.43 PERCUTANEOUS ETHANOL INJECTION AND RADIOFREQUENCY ABLATION
Image-guided ablation is currently regarded as the best therapeutic option for patients with small HCC not suitable for surgery. Treatment is usually performed percutaneously under ultrasound or CT guidance. Several methods for tumour destruction have been used, the most widely studied being percutaneous ethanol injection (PEI) and radiofrequency ablation (RFA). The injection of 90 per cent ethanol under ultrasound guidance is technically straightforward, inexpensive, safe and results in 5-year survival of 50 per cent in Child A, 30 per cent in Child B, but less than 10 per cent in Child C cirrhotics.44 Complete tumour necrosis is achieved in 70 per cent of tumours less than 3 cm in diameter, but this falls with increasing size (50 per cent in lesions 3–5 cm), probably due to the inability of the injected volume to disperse evenly throughout larger tumours that may contain fibrous septae. The procedure also becomes more tedious above 3 cm, requiring multiple sessions. Radiofrequency ablation is a localized thermal treatment producing tumour destruction by heating a probe inserted into the tumour to temperatures exceeding 50 °C, which can be performed percutaneously under image guidance, laparoscopically or at laparotomy. Several randomized studies have compared RFA with PEI. A study in patients with tumours up to 4 cm demonstrated that RFA was superior in terms of tumour necrosis and survival, with 3-year survival of 74 per cent versus 51 per cent.45 Further studies have demonstrated similar advantages for RFA in treating smaller tumours (3 cm).46,47 In general, RFA was associated with fewer sessions to achieve complete tumour necrosis and with no significant differences in morbidity and a likely improvement in survival. It is generally considered that RFA is superior to PEI for larger lesions. CHEMOEMBOLIZATION
Hepatocellular carcinomas greater than 2 cm are highly vascularized, mostly via the hepatic artery. In contrast, 75 per cent of normal liver parenchyma derives its blood
Hepatocellular carcinoma 565
R R (a)
(c)
(b)
(d)
supply from the portal vein. Thus the hepatic artery provides the potential for tumour-selective targeting of drugs and a rationale for arterial obstruction as a therapeutic approach. Transarterial chemoembolization (TACE) utilizes super-selective catheterization of the hepatic artery to deliver regional chemotherapy and embolize tumour-feeding arteries (Fig. 22.4). Chemotherapy is first injected, mixed with lipiodol, an oily compound that tends to accumulate in tumours, probably via enhanced permeability of leaky tumour vasculature and retention due to impaired lymphatic drainage, with the aim of also retaining the chemotherapy to increase tumour concentration and reduce systemic exposure. For chemoembolization to be performed safely, there must be adequate blood supply to the non-tumorous liver via the portal vein, and therefore it is contraindicated in the presence of main portal vein thrombosis. Other contraindications include extra-hepatic disease, advanced cirrhosis and poor performance status. Embolization frequently causes a characteristic syndrome of abdominal pain, fever and nausea, which is normally self-limiting within 2–4 days, although occasionally patients go on to develop liver abscess.48 Unlike earlier studies, recent randomized trials have shown survival benefit from chemoembolization49,50 (Fig. 22.5) and a meta-analysis of seven trials, incorporating 545 patients, confirmed improved survival compared to supportive care or systemic therapy.51 The general applicability of this meta-analysis is, however, limited in view of its
Figure 22.4 Hepatic arterial angiogram (a and b) and contrast-enhanced CT scan (c and d) before and after chemoembolization (non-lipiodol) in a patient with multi-focal hepatocellular carcinoma. (a) and (c) Tumours are characterized by hypervascular lesions in arterial phase. (b) and (d) Post-chemoembolization. (b) Angiogram demonstrating occlusion of right hepatic artery. (d) Hypodense lesions on CT.
relatively small size, the heterogeneity of the patient populations and of the techniques used, and differences in the choice of chemotherapeutic agent, embolic agent and the use of lipiodol. The key to successful chemoembolization is patient selection, and both the positive trials suggest the ideal candidate has very well preserved liver function and asymptomatic disease. Irrespective of the approach used, the major limitation of all these loco-regional approaches is disease recurrence. The risk of recurrence is conventionally related to tumour size, the rate increasing in direct relation to tumour size. To this extent, the Milan criteria do, in effect, define an arbitrary acceptable recurrence rate. In fact, it is probably the presence or absence of vascular invasion (micro- or macro-) that really defines the risk of recurrence, but since the former cannot readily be determined prior to treatment, and correlates well with tumour size, tumour size is the more convenient parameter. There are two types of recurrence. The first, and by far the most common, represents the growth of pre-existing micrometastases from the primary tumour. These usually occur within the immediate vicinity of the tumour. To this extent, tumour recurrence is sometimes classified as treatment failure, i.e. the ablative therapy has not taken a sufficient rim of surrounding tissue to remove the micrometastases. Whilst it is not yet certain that RFA is superior to PEI in terms of increased survival, there is definitely a decrease in local recurrence, presumably
566 Liver
CHEMOTHERAPY
Probability of survival (%)
100 80
Chemoembolization (n 40)
60 40 Log-rank p < 0.009 Control (n 35)
20 0 0
12 24 36 48 Time since randomization (months)
Patients at risk Chemoembolization Control
40 35
29 19
14 7
60
4 3
2 0
(a)
Probability of survival
1 0.8 0.6
Chemoembolization group
0.4 Control group
0.2 0
0
No. at risk Chemoembolization Control
6
40 39
12
29 17
18 24 Time (months) 22 12
16 7
30
36
42
12 4
10 3
10 1
(b)
Figure 22.5 Survival curves comparing trans-arterial chemoembolization against best supportive therapy. (a, see reference 51; b, see reference 50.)
reflecting the ability of the former procedure to take an appropriate margin of surrounding liver tissue (usually 1 cm), whereas this is not possible with PEI. The second type of recurrence is the development of a ‘new’ tumour. This is usually considered to have occurred where tumour develops more than 3 years after treatment. As yet, there are no conclusive data indicating that any adjuvant treatment can decrease the risk of tumour recurrence.
Systemic therapies The majority of patients with HCC have multi-focal disease, bilobar disease, extra-hepatic disease and/or underlying cirrhosis, such that surgery, embolization or ablation is not indicated. For these patients, systemic therapy is required.
Response rates for single-agent chemotherapy are low and significant durable remission is rare. The most widely used single cytotoxic agent has been doxorubicin.52,53 However, in systematic reviews of randomized trials of doxorubicin therapy, no significant survival effect was discernable.54 No other systemic therapies have fared significantly better and systemic therapy is now largely confined to clinical trials. Combination chemotherapy appears to give a higher response rate, although again the duration of remission is usually short. In general, even for well-selected patients, the expected objective response rate is only around 20–30 per cent and, as such, seems unlikely to have a significant impact on survival.55 A phase II study of a four-drug systemic combination regimen – cisplatin, recombinant interferon alpha-2b, doxorubicin and 5-fluorouracil (PIAF) – was encouraging, showing that although the response rate was not high (25 per cent), 9 of the 13 partial responders had their disease rendered resectable.56 However, a prospective randomized study comparing PIAF to conventional systemic doxorubicin suggested that any benefit in terms of increased survival was counteracted by increased toxicity.57 An alternative systemic approach has been endocrine manipulation based on reports of oestrogen receptor expression in some HCCs. Early small studies with antioestrogenic and anti-androgenic agents showed some promise.58 However, recent large-scale prospective controlled studies have refuted any role for hormonal agents, including tamoxifen.59,60 More promising data are emerging for molecularly targeted systemic therapies. For example, sorafenib, a multikinase inhibitor of tumour growth (via inhibition of the raf/mek/erk signalling pathway) and of angiogenesis (via inhibition of Vascular Endothelial Growth Factor Receptor, VEGFR, signalling) has been shown in a large, placebocontrolled, randomised trial to significantly prolong survival of patients with advanced HCC.61 Other agents have demonstrated encouraging activity in phase II trials and a new era of targeted therapies for HCC is anticipated. RADIATION THERAPY
The use of external-beam radiotherapy is significantly limited by the radiosensitivity of normal hepatocytes, with normal tissue tolerance 25–30 Gy.62 Higher doses are associated with increasing risk of radiation hepatitis. Further, in patients with cirrhosis, the liver may be even less tolerant of radiotherapy. The use of conformal techniques allows tumour doses up to 72 Gy without severe hepatitis and with evidence of objective responses.63 Radiation techniques that allow tumour-directed delivery appear promising, with evidence of anti-tumour activity, although to date the impact on survival has not been defined.64 Radio-isotopes can be delivered via the hepatic artery using yttrium-90-labelled microspheres or iodine-131containing lipiodol. Iodine-131 lipiodol, a gamma emitter,
Neuroendocrine tumors metastatic to the liver 567
has been reported to induce response rates of 40 per cent, whilst maintaining a radiation dose to normal liver of less than 20 Gy.65 Yttrium-90 is a beta emitter with shorter tissue penetration, which may further improve normal liver tolerance and reduce the risk of radiation hepatitis. In patients receiving doses in excess of 120 Gy, median survival was 15 months, with no evidence of radiation hepatitis, even in cirrhotic patients.66 These procedures are associated with a risk of lung toxicity if there is extensive arteriovenous shunting in patients with cirrhosis, and an assessment of the degree of shunting should be made prior to treatment.
METASTATIC LIVER DISEASE The management of metastatic liver disease is usually an integral part of the overall management of the primary tumour. However, in the case of colorectal cancer, the liver may be the sole or predominant site of metastatic disease, and in carefully selected patients local treatment is potentially curative. The recent advances of cytotoxic chemotherapy for colorectal cancer (see Chapter 28) also raise the possibility of the down-staging of liver metastases, thereby extending the limits of resectability.
Surgical resection of liver metastases
leads to 5-year survival in the order of 35–50 per cent. Although this has never been shown to be better than no active treatment or chemotherapy in a controlled trial, the evidence is convincing that it is only surgical resection that can produce such figures. Since combination chemotherapy now consistently achieves responses in up to 50 per cent of cases,72,73 the possibility of rendering initially unresectable disease resectable arises, and there are now several series in which this has been achieved in up to 40 per cent of cases.74,75 However, the approach is still in the early stages of development. It is clear that recurrence is still the rule,76 and several other problems need to be addressed. Criteria for just which tumours are ‘resectable’ are still evolving; deciding whether or not residual tumour after chemotherapy is actually viable remains difficult, as does the management of the patient who undergoes a complete response. The role of hepatic-artery infusion of cytotoxic agents for down-staging and systemic administration prior to resection of tumours for which there is no requirement for down-staging (i.e. in the neoadjuvant setting,77 with a view to limiting subsequent recurrence and extra-hepatic disease) are both areas of active research.
NEUROENDOCRINE TUMOURS METASTATIC TO THE LIVER
In approximately 20 per cent of patients with colorectal cancer, liver metastases will be detected at the same time as the primary lesion presents (synchronous), and a further 20 per cent will eventually develop metastases, typically within 3 or 4 years of resection of the primary tumour (metachronous). Of these, complete surgical resection is feasible in 20 per cent.67,68 There are no randomized clinical trials to confirm the efficacy of surgical resection, but extensive studies of the natural history of isolated colorectal metastases suggests 5-year and 10-year survival rates are less than 5 per cent and 1 per cent respectively, whereas most surgical series report figures of about 35 per cent and 25 per cent respectively.69–71 Factors suggesting successful outcome are complete resection (so-called R0; if margins are involved, the survival is little better than in patients not undergoing surgical resection) and low tumour volume, whereas the number of tumours and whether or not they are synchronous or metachronous are less important. All patients with predominant liver disease should be referred to a multidisciplinary team for consideration of resection. Guidelines for resection in the UK have now been published. Extra-hepatic disease is usually a contraindication, but occasionally surgery can be successful. The integration of ablative therapies may extend the limits of surgical resection and is an active area of research.
Neuroendocrine tumours are derived from Kulchitsky (enterochromaffin) cells, which are widely distributed in the body. Typically tumours are described as originating from the foregut, mid-gut or hindgut. A small proportion of tumours, predominantly those arising from the mid-gut (small bowel, appendix and caecum), are ‘functional’, with secretion of hormones, which can manifest as clinical symptoms. The carcinoid syndrome of episodic flushing, increased stool frequency, abdominal pain, bronchial constriction and right-sided heart disease, caused by systemic action of vasoactive peptides (5-hydroxytryptamine [serotonin], kinins and prostaglandins) secreted by the tumour, invariably indicates the presence of hepatic metastases, usually originating from a primary tumour of the small bowel. The extent of liver involvement is closely related to prognosis, but even with liver metastases, prolonged survival without active anti-cancer therapy remains a possibility, with 5-year survival rates of around 38 per cent.78 A minority of patients may present with solitary liver metastases or limited liver involvement, and in this subgroup of patients, surgical resection should be considered.79 For the majority of patients with diffuse, unresectable disease, other modalities of treatment such as embolization/chemoembolization, systemic chemotherapy and biological or hormonal agents need to be considered to palliate symptomatic hepatomegaly and hormonal symptoms.
Down-staging of liver metastases
Tumour embolization/chemoembolization
As noted above, about 15 per cent of cases of colorectal metastases appear confined to the liver and surgical resection
Selective embolization of hypervascular carcinoid liver metastases is attractive as a palliative treatment option as it
568 Liver
allows tumour debulking as well as reduction in the tumour’s capacity for hormone secretion. Objective response rates of between 48 per cent and 81 per cent have recently been reported, with symptomatic benefit in as many as 80 per cent of patients.80,81 It is common practice to repeat a series of embolizations over a period of time following an initial response. A retrospective series from the MD Anderson Cancer Center reported a deleterious effect on response rates of chemoembolization versus embolization alone, with a trend towards greater toxicity.82 The efficacy of therapy can be effectively monitored using estimation of 24-hour 5-hydroxyindolacetic acid (5HIAA) and chromogranin A.
Cytotoxic chemotherapy Single-agent activity is particularly poor in patients with mid-gut tumours, with response rates of less than 10 per cent.83 Combination studies have produced higher response rates and have centred on the use of streptozocin-based combinations, but convincing evidence for improved survival from chemotherapy compared to supportive care is lacking. However, a study randomizing patients with pancreatic islet-cell tumours between the addition of 5-fluorouracil or doxorubicin to streptozocin-based therapy indicated a significant increase in survival (2.2 years versus 1.5 years) in favour of the anthracycline-containing regimen.84 The applicability of these results to other types of neuroendocrine tumour is not clear. Newer agents have failed to demonstrate superiority. There is a limited evidence base to support the treatment of metastatic carcinoid tumours with high proliferative indices (the poorly differentiated, ‘small cell type’) with cisplatin and etoposide chemotherapy. One study reported response rates of 67 per cent amongst a group of patients with (prospectively identified) such tumours, although this effect remains to be validated by larger controlled studies.85
Hormonal treatment Hormone secretion from functional neuroendocrine tumours is mediated via stimulation of somatostatin receptors on the tumour-cell membrane. The advent of synthetic somatostatin analogues, which can down-regulate this process, has revolutionized the management of the carcinoid syndrome. Dramatic improvement or complete disappearance of symptoms (flushing more so than diarrhoea) is experienced by the great majority of patients, with minimal toxicity. The first analogue (octreotide, Sandostatin) was administered subcutaneously three times per day, but slow-release formulations are now available with similar activity and toxicity profile but with more convenience. These have now become the gold-standard therapy.86 Sandostatin LAR (20–30 mg every 4 weeks by deep intramuscular injection) or lanreotide (Somatuline
LA, 30 mg every 2 weeks by deep intramuscular injection) achieves similar response rates, but with increased compliance and patient satisfaction. The extent to which these analogues exert an anti-proliferative effect remains controversial. Disease ‘stabilization’ has been frequently reported and there are occasional reports of tumour shrinkage.87 Thus, in the German Sandostatin Study, 52 patients with progressive disease were evaluated following treatment with short-acting octreotide, with tumour stabilization for at least 3 months seen in around one-third of patients,88 and Eriksson et al. demonstrated an 81 per cent rate of disease stability amongst a cohort of 55 patients with advanced progressive neuroendocrine tumours.89 However, the indolent course of these tumours makes disease stabilization difficult to quantify as a true anti-proliferative effect of the drug. Nevertheless, there is a retrospective cohort study suggesting that since the introduction of octreotide there has been a striking increase in the survival of patients with carcinoid tumours, but confirmation must await properly controlled clinical trials.90
Interferon Interferon produces biochemical responses in up to 40 per cent of patients with functional neuroendocrine tumours and radiological tumour shrinkage in up to 10 per cent.91 The mechanism of action is unclear but may be a direct inhibition of cell proliferation or immune-mediated cytotoxicity. Faiss et al. randomized treatment-naïve patients to a somatostatin analogue alone, interferon alone or the combination. No significant differences in the rates of partial remission, stable disease or tumour progression were observed amongst the three groups, although a significant reduction in symptoms was seen in the combination group.92 Interferon-alpha thus remains a treatment option to consider for the palliation of symptomatic hepatic metastatic carcinoid disease, but the evidence base fails to support a survival advantage and its toxicity can substantially impact on quality of life.
KEY LEARNING POINTS ●
●
Identification of the major risk factors (chronic liver disease, chronic HBV and/or HCV infection and aflatoxin exposure) for HCC has laid the basis for screening and prevention. Hepatocellular carcinoma presents with the triad of pain, hepatomegaly and weight loss, or decompensation of pre-existing chronic liver disease, although with screening, presymptomatic diagnosis is becoming more common.
References 569
●
●
●
●
●
The diagnosis of HCC can often be established without recourse to biopsy by characteristic features on CT and MRI scanning, serum AFP estimation and the presence of underlying chronic liver disease. Surgical resection, liver transplantation and ablative techniques (such as PEI and RFA) offer the possibility of ‘cure’ for smaller tumours, but as tumour size increases, so does the rate of recurrence. Chemoembolization may offer useful palliation in a small subset of patients with excellent liver function and minimal symptoms. Patients with liver-predominant metastases from colorectal cancer should be considered at multidisciplinary meetings for surgical resection. Where this is possible, 30–50 per cent 5-year survival can be expected. A small percentage of such patients may benefit from neoadjuvant therapy to downsize tumour to a size at which resection is feasible. Symptomatic control of the carcinoid syndrome is achievable, at least in the short term, for the great majority of patients with the use of somatostatin analogues and embolization of the causative liver metastases.
11
12 ◆13
14
15 16
17
18
19
REFERENCES ●1
2
3 ●4
●5
●6
7 8
◆9
10
Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002 CA. Cancer J Clin 2005; 55:74–108. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 1999; 340:745–50. Willis RA. Secondary tumours of the liver. In: The Spread of Tumours in the Human Body. London: Butterworths, 1973. Beasley RP. Hepatitis B virus. The major etiology of hepatocellular carcinoma. Cancer 1988; 61(10):1942–56. Chang MH, Chang MH, Chen CJ, et al. Universal hepatitis B vaccination in Taiwan and the incidence of hepatocellular carcinoma in children. N Engl J Med 1997; 336:1855–9. Liaw YF, Sung JJ, Chow WC, et al. Lamivudine for patients with chronic hepatitis B and advanced liver disease. N Engl J Med 2004; 351(15):1521–31. Colombo M. Hepatitis C virus and hepatocellular carcinoma. Semin Liver Dis 1999; 19(3):263–9. Davila JA, Morgan RO, Shaib Y, et al. Hepatitis C infection and the increasing incidence of hepatocellular carcinoma: a population-based study. Gastroenterology 2004; 127:1372–80. Johnson PJ, Williams R. Cirrhosis and the aetiology of hepatocellular carcinoma. J Hepatol 1987; 4(1):140–7. Van Rensburg SJ, Cook-Mozaffari P, Van Schalkwyk DJ, et al. Hepatocellular carcinoma and dietary aflatoxin in
20
◆21
22
23
◆24
25
26
Mozambique and Transkei. Br J Cancer 1985; 51(5): 713–26. Groopman JD, Wild CP, Hasler J, et al. Molecular epidemiology of aflatoxin exposures: validation of aflatoxinN7-guanine levels in urine as a biomarker in experimental rat models and humans. Environ Health Perspect 1993; 99:107–13. Ross RK, Yu MC, Henderson BE, et al. Aflatoxin biomarkers. Lancet 1992; 340(8811):119. Kensler TW, Egner PA, Wang JB, et al. Chemoprevention of hepatocellular carcinoma in aflatoxin endemic areas. Gastroenterology 2004; 127(5 Suppl. 1):S310–18. Turner PC, Sylla A, Gong YY, et al. Reduction in exposure to carcinogenic aflatoxins by postharvest intervention measures in West Africa: a community-based intervention study. Lancet 2005; 365(9475):1950–6. Bismuth H. Surgical anatomy and anatomical surgery of the liver. World J Surg 1982; 6(1) 3–9. [volume number??] Melia WM, Wilkinson ML, Portmann BC, et al. Hepatocellular carcinoma in the non-cirrhotic liver: a comparison with that complicating cirrhosis. QJM (New Series LIII) 1984; 211:391–400. Lai EC, Lau WY. Spontaneous rupture of hepatocellular carcinoma: a systematic review. Arch Surg 2006; 141(2):191–8. Altaee MY, Johnson PJ, Farrant JM, Williams, R. Etiological and clinical characteristics of peripheral and hilar cholangiocarcinomas. Cancer 1991; 68:2501–5. Bruix J, Sherman M. Practice Guidelines Committee, American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma. Hepatology 2005; 42(5):1208–36. Bhattacharjya S, Bhattacharjya T, Quaglia A, et al. Liver transplantation in cirrhotic patients with small hepatocellular carcinoma: an analysis of pre-operative imaging, explant histology and prognostic histologic indicators. Dig Surg 2004; 21(2):152–9; discussion 159–60. Johnson PJ. The role of serum alpha-fetoprotein estimation in the diagnosis and management of hepatocellular carcinoma. Clin Liver Dis 2001; 5(1):145–59. Sakon M, Monden M, Gotoh M, et al. The effects of vitamin K on the generation of des-gamma-carboxy-prothrombin (PIVKA-II) in patients with hepatocellular carcinoma. Am J Gastroenterol 1991; 86:339–43. Chen JG, Parkin DM, Chen QG, et al. Screening for liver cancer: results of a randomised controlled trial in Qidong, China. J Med Screen 2003; 10(4):204–9. Sherman M. Hepatocellular carcinoma: epidemiology, risk factors, and screening. Semin Liver Dis 2005; 25(2):143–54. Craig JR, Peters RL, Edmondson HA, Omata M. Fibrolamellar carcinoma of the liver: a tumor of adolescents and young adults with distinctive clinico-pathologic features. Cancer 1980; 46(2):372–9. Katzenstein HM, Krailo MD, Malogolowkin MH, et al. Fibrolamellar hepatocellular carcinoma in children and adolescents. Cancer 2003; 97(8):2006–12.
570 Liver
27 Okuda K. Natural history of hepatocellular carcinoma including fibrolamellar and hepato-cholangiocarcinoma variants. J Gastroenterol Hepatol 2002; 17(4):401–5. 28 Geddes EW, Falkson G. Malignant hepatoma in the Bantu. Cancer 1970; 25(6):1271–8. 29 Llovet JM, Bustamante J, Castells A, et al. Natural history of untreated nonsurgical hepatocellular carcinoma: rationale for the design and evaluation of therapeutic trials. Hepatology 1999; 29(1):62–7. 30 Palmer DH, Johnson PJ. Hepatocellular carcinoma. In: Gospodarowicz MK, O’Sullivan B, Sobin LH (eds), Prognostic Factors in Cancer, 3rd edn. Wiley-Liss, 2006, 143–146 31 Ebara M, Ohto M, Shinagawa T, et al. Natural history of minute hepatocellular carcinoma smaller than three centimeters complicating cirrhosis. A study in 22 patients. Gastroenterology 1986; 90(2):289–98. 32 Wayne JD, Lauwers GY, Ikai I, et al. Preoperative predictors of survival after resection of small hepatocellular carcinomas. Ann Surg 2002; 235(5):722–30; discussion 730–1. 33 Poon RT, Ng IO, Fan ST, et al. Clinicopathologic features of long-term survivors and disease-free survivors after resection of hepatocellular carcinoma: a study of a prospective cohort. J Clin Oncol 2001; 19(12):3037–44. 34 Prospective validation of the CLIP score: a new prognostic system for patients with cirrhosis and hepatocellular carcinoma. The Cancer of the Liver Italian Program (CLIP) Investigators. Hepatology 2000; 31(4):840–5. 35 Leung TW, Tang AM, Zee B, et al. Construction of the Chinese University Prognostic Index for hepatocellular carcinoma and comparison with the TNM staging system, the Okuda staging system, and the Cancer of the Liver Italian Program staging system: a study based on 926 patients. Cancer 2002; 94(6):1760–9. 36 Sobin LH, Wittekind CH (eds). UICC: TNM Classification of Malignant Tumors, 6th edn. New York: Wiley-Liss, 2002. 37 Farmer DG, Rosove MH, Shaket A, Busuttil RW. Current treatment modalities for hepatocellular carcinoma. Ann Surg 1994; 219:236–47. 38 Vauthey JN, Klimstra D, Franceschi D, et al. Factors affecting long term outcome after hepatic resection for hepatocellular carcinoma. Am J Surg 1995; 169:28–34. ●39 Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996; 334(11):693–9. 40 Yao FY, Ferrell L, Bass NM, et al. Liver transplantation for hepatocellular carcinoma: expansion of the tumor size limits does not adversely impact survival. Hepatology 2001; 33(6):1394–403. 41 Yao FY, Bass NM, Nikolai B, et al. A follow-up analysis of the pattern and predictors of dropout from the waiting list for liver transplantation in patients with hepatocellular carcinoma: implications for the current organ allocation policy. Liver Transpl 2003; 9(7):684–92. 42 United Network for Organ Sharing. www.unos.org. 43 Gondolesi GE, Roayaie S, Munoz L, et al. Adult living donor liver transplantation for patients with hepatocellular
44
45
46
47
48
49
50
51
52
53
54
54 56
57
58
carcinoma: extending UNOS priority criteria. Ann Surg 2004; 239(2):142–9. Livraghi T, Giorgio A, Marin G, et al. Hepatocellular carcinoma and cirrhosis in 746 patients: long-term results of percutaneous ethanol injection. Radiology 1995; 197(1):101–8 Shiina S, Teratani T, Obi S, et al. A randomized controlled trial of radiofrequency ablation with ethanol injection for small hepatocellular carcinoma. Gastroenterology 2005; 129(1):122–30. Lencioni R, Cioni D, Crocetti L, et al. Early-stage hepatocellular carcinoma in patients with cirrhosis: longterm results of percutaneous image-guided radiofrequency ablation. Radiology 2005; 234(3):961–7. Livraghi T, Goldberg SN, Lazzaroni S, et al. Small hepatocellular carcinoma: treatment with radio-frequency ablation versus ethanol injection. Radiology 1999; 210(3):655–61. Brown DB, Geschwind JF, Soulen MC, et al. Society of Interventional Radiology position statement on chemoembolisation for hepatic malignancies. J Vasc Interv Radiol 2006; 17:217–23. Llovet JM, Bruix J. Systematic review of randomized trials for unresectable hepatocellular carcinoma: chemoembolization improves survival. Hepatology 2003; 37(2):429–42. Lo CM, Ngan H, Tso WK, et al. Randomized controlled trial of transarterial lipiodol chemoembolization for unresectable hepatocellular carcinoma. Hepatology 2002; 35(5):1164–71. Llovet JM, Real MI, Montana X, et al. Barcelona Liver Cancer Group. Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomised controlled trial. Lancet 2002; 359(9319):1734–9. Olweny CL, Toya T, Katongole-Mbidde E, et al. Treatment of hepatocellular carcinoma with adriamycin. Preliminary communication. Cancer 1975; 36(4):1250–7. Johnson PJ, Williams R, Thomas H, et al. Induction of remission in hepatocellular carcinoma with doxorubicin. Lancet 1978; 1(8072):1006–9. Simonetti RG, Leberati A, Angiolini C, et al. Treatment of hepatocellular carcinoma: a systematic review of randomized controlled trials. Ann Oncol 1997; 8:117–36. Leung TW, Johnson PJ. Systemic therapy for hepatocellular carcinoma. Semin Oncol 2001; 28:514–20. Leung TW, Patt YZ, Lau WY, et al. Complete pathological remission is possible with systemic combination chemotherapy for inoperable hepatocellular carcinoma. Clin Cancer Res 1999; 5(7):1676–81. Yeo W, Mok TS, Zee B, et al. A randomized phase III study of doxorubicin versus cisplatin/interferon alpha2b/doxorubicin/fluorouracil (PIAF) combination chemotherapy for unresectable hepatocellular carcinoma. J Natl Cancer Inst 2005; 97(20):1532–8. Farinati F, De Maria N, Fornasiero A, et al. Prospective controlled trial with antiestrogen drug tamoxifen in
References 571
59
60
◆61
62
63
64
65
66
67
68
◆69
70
71
72
73
patients with unresectable hepatocellular carcinoma. Dig Dis Sci 1992; 37(5):659–62. CLIP Group (Cancer of the Liver Italian Programme). Tamoxifen in treatment of hepatocellular carcinoma: a randomised controlled trial. Lancet 1998; 352(9121):17–20. Chow PK, Tai BC, Tan CK, et al. Asian–Pacific Hepatocellular Carcinoma Trials Group. High-dose tamoxifen in the treatment of inoperable hepatocellular carcinoma: a multicenter randomized controlled trial. Hepatology 2002; 36(5):1221–6. Llovet J, Ricci S, Mazzaferro V, et al. Sorafenib improves survival in advanced Hepatocellular Carcinoma (HCC): Results of a Phase III randomized placebo-controlled trial (SHARP trial). J Clin Oncol 2007; 25(18S): LBA1. Hawkins MA, Dawson LA. Radiation therapy for hepatocellular carcinoma: from palliation to cure. Cancer 2006; 106(8):1653–63. Robertson JM, Lawrence TS, Dworzanin LM, et al. Treatment of primary hepatobiliary cancers with conformal radiation therapy and regional chemotherapy. J Clin Oncol 1993; 1(7):1286–93. Raoul JI, Bretagne JF, Caucanas JP, et al. Internal radiation therapy for hepatocellular carcinoma. Results of a French multicenter phase II trial of transarterial injection of iodine 131-labeled lipiodol. Cancer 1992; 69(2):346–52. Lau WY, Ho S, Leung TW, et al. Selective internal radiation therapy for nonresectable hepatocellular carcinoma with intraarterial infusion of 90-yttrium microspheres. Int J Radiat Oncol Biol Phys 1998; 40(3):583–92. Salem R, Lewandowski RJ, Atassi B, et al. Treatment of unresectable hepatocellular carcinoma with use of 90Y microspheres (TheraSphere): safety, tumor response, and survival. J Vasc Interv Radiol 2005; 16(12):1627–39. Wagner JS, Adson MA, Van Heerden JA, et al. The natural history of hepatic metastases from colorectal cancer. A comparison with resective treatment. Ann Surg 1984; 199(5):502–8. Biasco G, Derenzini E, Grazi G, et al. Treatment of hepatic metastases from colorectal cancer: many doubts, some certainties. Cancer Treat Rev 2006; 32(3):214–28. Simmonds PC, Primrose JN, Colquitt JL, et al. Surgical resection of hepatic metastases from colorectal cancer: a systematic review of published studies. Br J Cancer 2006; 94(7):982–99. Wei AC, Greig PD, Grant D, et al. Survival after hepatic resection for colorectal metastases: a 10-year experience. Ann Surg Oncol 2006; 13(5):668–76. Fong Y, Cohen AM, Fortner JG, et al. Liver resection for colorectal metastases. J Clin Oncol 1997; 15(3):938–46. de Gramont A, Figer A, Seymour M, et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000; 18(16):2938–47. Goldberg RM, Sargent DJ, Morton RF, et al. A randomized controlled trial of fluorouracil plus leucovorin, irinotecan, and oxaliplatin combinations in patients with previously
74
75
76
77
78 79
80
◆81
82
83
84
85
◆86
87
88
untreated metastatic colorectal cancer. J Clin Oncol 2004; 22(1):23–30. Vibert E, Canedo L, Adam R. Strategies to treat primary unresectable colorectal liver metastases. Semin Oncol 2005; 32(Suppl. 8):33–9. National Institute for Health and Clinical Excellence. Irinotecan, oxaliplatin and raltitrexed for the treatment of advanced colorectal cancer. Review of Technology Appraisal 33, August 2005 (www.nice.org.uk). Leonard GD, Brenner B, Kemeny NE. Neoadjuvant chemotherapy before liver resection for patients with unresectable liver metastases from colorectal carcinoma. J Clin Oncol 2005; 23(9):2038–48. Kemeny NE, Niedzwiecki D, Hollis DR, et al. Hepatic arterial infusion versus systemic therapy for hepatic metastases from colorectal cancer: a randomized trial of efficacy, quality of life, and molecular markers (CALGB 9481). J Clin Oncol 2006; 24(9):1395–403. Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715 carcinoid tumors. Cancer 2003; 97(4):934–59. Norton JA. Endocrine tumours of the gastrointestinal tract. Surgical treatment of neuroendocrine metastases. Best Pract Res Clin Gastroenterol 2005; 19(4):577–83. Strosberg JR, Choi J, Cantor AB, Kvols LK. Selective hepatic artery embolization for treatment of patients with metastatic carcinoid and pancreatic endocrine tumors. Cancer Control 2006; 13(1):72–8. O’Toole D, Ruszniewski P. Chemoembolization and other ablative therapies for liver metastases of gastrointestinal endocrine tumours. Best Pract Res Clin Gastroenterol 2005; 19(4):585–94. Gupta S, Johnson MM, Murthy R, et al. Hepatic arterial embolization and chemoembolization for the treatment of patients with metastatic neuroendocrine tumors: variables affecting response rates and survival. Cancer 2005; 104(8):1590–602. Moertel CG. Treatment of the carcinoid tumor and the malignant carcinoid syndrome. J Clin Oncol 1983; 1(11):727–40. Moertel CG, Lefkopoulo M, Lipsitz S, et al. Streptozocin–doxorubicin, streptozocin–fluorouracil or chlorozotocin in the treatment of advanced islet-cell carcinoma. N Engl J Med 1992; 326(8):519–23. Moertel CG, Kvols LK, O’Connell MJ, Rubin J. Treatment of neuroendocrine carcinomas with combined etoposide and cisplatin. Evidence of major therapeutic activity in the anaplastic variants of these neoplasms. Cancer 1991; 68(2):227–32. Akerstrom G, Hellman P, Hessman O, Osmak L. Management of midgut carcinoids. J Surg Oncol 2005; 89(3):161–9. Leong WL, Pasieka JL. Regression of metastatic carcinoid tumors with octreotide therapy: two case reports and a review of the literature. J Surg Oncol 2002; 79(3):180–7. Arnold R, Trautmann ME, Creutzfeldt W, et al. Somatostatin analogue octreotide and inhibition of tumour growth in metastatic endocrine gastroenteropancreatic tumours. Gut 1996; 38(3):430–8.
572 Liver
89 Eriksson B, Renstrup J, Imam H, Oberg K. High-dose treatment with lanreotide of patients with advanced neuroendocrine gastrointestinal tumors: clinical and biological effects. Ann Oncol 1997; 8(10):1041–4. 90 Quaedvlieg PF, Visser O, Lamers CB, et al. Epidemiology and survival in patients with carcinoid disease in The Netherlands. An epidemiological study with 2391 patients. Ann Oncol 2001; 12(9):1295–300.
91 Schnirer II, Yao JC, Ajani JA. Carcinoid – a comprehensive review. Acta Oncol 2003; 42(7):672–92. 92 Faiss S, Pape UF, Bohmig M, et al. Prospective, randomized, multicenter trial on the antiproliferative effect of lanreotide, interferon alfa, and their combination for therapy of metastatic neuroendocrine gastroenteropancreatic tumors – the International Lanreotide and Interferon Alfa Study Group. J Clin Oncol 2003; 21(14):2689–96.
23 Pancreas PATRICK G. JOHNSTON AND MARTIN M. EATOCK
Incidence Aetiology Genetics Anatomy Pathology and staging
573 573 574 574 575
Pancreatic cancer represents a major public health problem and accounts for around 5 per cent of all cancer deaths.
INCIDENCE The incidence of cancer of the pancreas in the UK has been unchanged over the last 15 years, with around 7000 new cases per year diagnosed. The age-standardized incidence of pancreatic cancer is 10.2 cases per 100 000 in men and 7.7 per 100 000 in women, suggesting a slight male predominance (National Statistics Online Series MB1 No. 33 and 25). 130 120 110
Rate per 100,000
100
Symptoms and signs Natural history Investigation Treatment References
575 575 576 576 581
In the USA, the age-adjusted incidence rate is 12.5 per 100 000 men and 9.8 per 100 000 women. Mortality rates from pancreatic cancer are almost identical to the incidence rates, with an age-adjusted mortality rate of 12.2 per 100 000 men and 9.2 per 100 000 women. Pancreatic cancer is the fourth leading cause of cancer death in men and the fifth in women. In the USA, there is evidence of racial differences in incidence, with a lower incidence in Native Americans and a higher incidence in African Americans. Furthermore, there is significant variation in incidence throughout the world, with a very low incidence observed in India (0.5–2.4 per 100 000 in men and 0.5–1.8 per 100 000 in women).1 As can be seen in Figure 23.1, pancreatic cancer is rare in those aged less than 40. The peak incidence is in the seventh and eighth decades of life.
90 80
AETIOLOGY
70 60 50 40 30 20 10 85
80–84
75–79
70–74
65–69
60–64
55–59
50–54
45–49
40–44
35–39
30–34
25–29
20–24
15–19
10–14
05–09
00
01–04
0
Age at diagnosis White
Black
Asian or Pacific islander
American Indian/Alaska native
Hispanic
Figure 23.1 Incidence of pancreatic cancer by age and racial origin in USA (Surveillance, Epidemiology, and End Results (SEER) Program).
Smoking and dietary factors remain the strongest associations with the aetiology of pancreatic cancer. It has been recognized for almost 30 years that tobacco smoking is associated with an increased risk of exocrine pancreatic cancer.2 Smokers have a risk of developing pancreatic cancer that is more than twice that of non-smokers3 and that is unrelated to the type of tobacco used.4 Recent evidence has also suggested a weak but significant association between exposure to environmental tobacco smoke and pancreatic cancer in non-smokers.5 Diets consisting of a high consumption of red meat or processed meat and that are high in simple carbohydrates appear to be associated with an increased risk of pancreatic
574 Pancreas
cancer. The association of high cholesterol intake with pancreatic cancer is controversial and the published studies are contradictory. It is clear, however, that diets with a high intake of saturated fats do increase the risk of pancreatic cancer. In contrast, dietary fibre, in particular from vegetables and cereals, appears to be protective.6–9 A number of studies in the late 1990s have suggested that obesity is strongly associated with pancreatic cancer; however, meta-analysis shows significant heterogeneity between these studies and, whilst there is a positive association, it is weak and may be explained by confounding variables.10 Type II diabetes has also been suggested as a risk factor. A recent meta-analysis indicates a 50 per cent increase in risk for those who have had diabetes for less than 4 years compared to those who have had the condition for 5 years or more.11 However, it is unclear whether diabetes is an independent risk factor for this disease, and this association is confounded by common aetiological factors such as diet and obesity or whether diabetes is itself a consequence of an undiagnosed pancreatic cancer. For example, chronic pancreatitis, which approximately doubles an individual’s risk of developing pancreatic malignancy,12–15 is also associated with late-onset diabetes mellitus. Although initial studies in the UK and Italy in the 1980s suggested an association between coffee drinking and pancreatic cancer,16,17 recent large cohort studies have failed to confirm this.7,18 In contrast to these dietary studies, pancreatic cancer has a clear familial association. First-degree relatives have almost double the population risk of developing the disease. Furthermore, if the proband is aged less than 60 years, this risk increases to three times the population risk.19 This translates to an overall 4.7 per cent lifetime risk of pancreatic cancer for first-degree relatives of a patient with pancreatic cancer, and a 7.2 per cent lifetime risk for first-degree relatives of a proband diagnosed before the age of 60.20 There are a number of inherited familial cancer syndromes that include pancreatic cancer (see below). Understanding of the genetic abnormalities associated with these has led to a greater understanding of the biology of pancreatic cancer, as discussed in the next section.
GENETICS Nearly 10 per cent of pancreatic cancers are hereditary in origin. There are a number of associated familial syndromes, including hereditary non-polyposis colon cancer (HNPCC), the familial atypical mole/malignant melanoma syndrome (FAMMM), hereditary pancreatitis, familial adenomatous polyposis (FAP) and familial breast/ovarian cancer related to BRCA2 mutations.21 However, the majority of familial pancreatic cancers are not associated with one of these syndromes. Recent evidence suggests that there is a specific autosomal dominant genetic abnormality on chromosome 4 that may be implicated in familial pancreatic cancer.
Hereditary pancreatitis is an autosomal dominant disease. The cumulative risk of developing pancreatic cancer is 44 per cent by age 70. In 80 per cent of cases, hereditary pancreatitis is due to mutations in the cationic trypsinogen gene, a number of which have been described. However, there is no clear evidence that the risk of developing pancreatic cancer is related to any particular genotype. Familial atypical mole/malignant melanoma syndrome is the result of an inactivating germline mutation in the CDKN2 (p16/INK4a) gene. This tumour suppressor gene is an important regulator of the cell cycle and mutations in this gene are frequently observed in sporadic pancreatic cancer.22 Mutations in BRCA2 are particularly interesting. This gene plays an important role in the repair of DNA damage, particularly double-strand breaks. Pancreatic cancer due to germline BRCA2 mutation is described in association with breast cancer, male breast cancer and prostate cancer, but not with ovarian cancer. It is now evident that the K3326X BRCA2 polymorphism, initially though to be non-deleterious and not associated with other cancers, does appear to confer an increased risk of pancreatic cancer.23 A number of genetic abnormalities have been described in sporadic pancreatic cancer, the most commonly occurring of which include mutations in k-RAS, p53, p21 (WAF1/CIP1) and MTS-1. A range of other abnormalities, including over-expresson of NMYC and reduced expression of CDKN1, have also been recently described. Studies of genetic abnormalities in pancreatic intra-epithelial neoplasia (PanIN) suggest that k-RAS and p21 mutations occur early in the development of pancreatic cancer, followed by changes in the expression of p53 and cyclin D1. Loss of expression of SMAD4 is described and is a late change in the evolution of this disease.24 p53 mutation occurs in approximately 50 per cent of pancreatic cancers and is an independent prognostic factor.22
ANATOMY Sixty-six per cent of pancreatic cancers arise in the head of the pancreas and a further 20 per cent arise in the body or tail. The remainder occur as multiple lesions throughout the pancreas. Tumours arising in the body or tail of the pancreas tend to present at a later stage than tumours of the pancreatic head. A heterogeneous group of tumours arise in the region of the ampulla of Vater and may consist of tumours arising from the ampulla itself, pancreatic tumours, cholangiocarcinomas or duodenal carcinomas. In practice, it is often difficult to distinguish these on clinical or radiological grounds; however, tumours arising in the ampulla or from duodenal mucosa have a significantly better prognosis than those arising in the pancreas. Lymphatic spread from the head of the pancreas is to the nodes around the inferior pancreatico-duodenal artery, the superior mesenteric artery and the gastro-duodenal artery leading to the coeliac axis. Tumours in the distal pancreas may also drain towards the splenic hilum.
Natural history 575
PATHOLOGY AND STAGING Ninety per cent of tumours arising in the pancreas are pancreatic ductal adenocarcinomas. The other less common ones are pancreatic neuroendocrine tumours and mucinous cystic neoplasms of the pancreas (cystadenoma or cystadenocarcinoma) (Table 23.1). Rarely, other tumours may arise in the pancreas, including mesenchymal tumours and lymphomas. The predominant histological feature in pancreatic ductal adenocarcinoma is a dense collagenous stroma with an increase in the number of ducts, which are atypical in appearance. The diagnosis is dependent on the presence of cellular atypia, nuclear pleomorphism, the identification of mitoses and evidence of perineural, vascular or lymphatic invasion. There is no universally accepted staging system for carcinoma of the pancreas, and a number of systems are used, the most common being the TNM system (Box 23.1).
Box 23.1 TNM/AJCC staging of exocrine pancreatic cancer Primary tumour (T) TX Primary tumour cannot be assessed T0 No evidence of primary tumour Tis In-situ carcinoma (including pancreatic intraepithelial neoplasia III) T1 Tumour limited to pancreas, 2 cm or less in greatest dimension T2 Tumour limited to pancreas, more than 2 cm in greatest dimension T3 Tumour extends beyond the pancreas but without involvement of the coeliac axis or the superior mesenteric artery T4 Tumour involves coeliac axis, or the superior mesenteric artery Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph-node metastases N1 Regional lymph-node metastases Metastases (M) MX Distant metastases cannot be assessed M0 No distant metastases M1 Distant metastases Stage grouping Stage 0 Tis N0 M0 Stage IA T1 N0 M0 Stage IB T2 N0 M0 Stage IIA T3 N0 M0 Stage IIB T1–3 N1 M0 Stage III T4 N0 M0 Stage IV Any T Any N M1
Table 23.1 World Health Organization classification of pancreatic exocrine tumours Malignant
Uncertain malignant potential
Benign
Ductal adenocarcinoma Mucinous cystadenocarcinoma Acinar carcinoma Unclassified large-cell carcinoma Small-cell carcinoma Pancreatoblastoma Intraductal papillary mucinous tumour Mucinous cystadenoma Papillary cystic neoplasm Serous cystadenoma
SYMPTOMS AND SIGNS The presenting features of pancreatic cancer depend on the situation of the tumour within the pancreas. Classically, tumours in the head of the pancreas present with obstructive jaundice and are smaller than tumours in the body or tail of the gland. Abdominal pain and weight loss are present in 72 per cent and 58 per cent of patients respectively25 and tend to be associated with more advanced disease. Weight loss in pancreatic cancer is often profound and is usually due to one or a combination of anorexia, early satiety, diarrhoea, malabsorption and gastric outlet obstruction. A recent diagnosis of maturity-onset diabetes mellitus in a patient who is losing weight is highly suspicious of an underlying diagnosis of pancreatic cancer. Pancreatic cancer is associated with a number of eponymous signs. Courvoisier’s sign (a non-tender, palpable gallbladder) is present in approximately 20 per cent of patients presenting with obstructive jaundice. Trousseau’s sign (migratory thrombophlebitis), Troisier’s sign (palpable left supraclavicular lymphadenopathy), and a Sister Mary Joseph nodule (metastasis to the umbilicus) are all signs of advanced disease and late-stage presentation. In summary, a diagnosis of pancreatic cancer should be considered early and investigated appropriately in those with any of the following. ● ●
● ● ●
Jaundice due to extra-hepatic biliary obstruction. Recent-onset, unexplained, upper abdominal or thoracolumbar back pain. Unexplained weight loss 10 per cent. Unexplained acute pancreatitis in those aged 50 years. Unexplained recent onset of diabetes mellitus.
NATURAL HISTORY Approximately 20 per cent of patients have disease confined to the head of the pancreas, which may be potentially resectable. Of the remainder, 40 per cent have locally advanced disease unsuitable for surgical resection and
576 Pancreas
40 per cent have either distant or peritoneal metastases. Overall, fewer than 5 per cent of patients with a diagnosis of pancreatic cancer are alive at 5 years. Even for those with resectable disease, the prognosis remains relatively poor, with 5-year survival rates between 10 per cent and 20 per cent in those undergoing surgical resection.26–28 Despite improved treatment options for those with unresectable disease, fewer than 20 per cent of such patients are alive 1 year following diagnosis.
INVESTIGATION The aims of investigation in those with a suspected diagnosis of pancreatic cancer are to establish the diagnosis and to determine the extent and resectability of the disease. The diagnosis of pancreatic cancer is usually based on the results of abdominal ultrasound scan, computed tomography (CT) scanning, endoscopic retrograde cholangio-pancreatography (ERCP) or magnetic resonance cholangio-pancreatography (MRCP). Pancreatic cancer may only be definitively diagnosed pathologically, and ultimately a tissue biopsy or cytology should be obtained. There are no specific laboratory tests for the diagnosis of pancreatic cancer. Serum tumour markers such as carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9) are the most extensively evaluated. CA19-9 is the most specific to pancreatic cancer;29,30 however, it may also be elevated in a number of other malignancies. Nonetheless, it has been demonstrated to be useful in the follow-up and monitoring of treatment in patients with pancreatic cancer.31–34 Other serum tumour markers, such as CA242, have undergone clinical evaluation but have shown little advantage over CA19-9 in the diagnosis and assessment of pancreatic cancer.35–37 Ultrasound is usually the initial investigation in those presenting with obstructive jaundice. The usual findings are dilatation of the extra-hepatic bile ducts and/or a mass in the head of the pancreas. A recent meta-analysis has demonstrated the superiority of helical or multi-array detector CT scanning to either magnetic resonance imaging (MRI) or ultrasound in the diagnosis of pancreatic cancer, with a sensitivity and specificity of 91 per cent and 85 per cent respectively for helical CT, compared to 76 per cent and 75 per cent for ultrasound.38 Although MRCP and ERCP are of equivalent sensitivity in the diagnosis of pancreatic cancer,39 MRCP avoids the risks of acute pancreatitis associated with ERCP. On the other hand, biliary obstruction may be effectively managed by ERCP and bile sampling and brushings of the lower biliary tree may be taken to establish the diagnosis of adenocarcinoma. 18-F-fluoro-2-deoxyglucose positron emission tomography (FDG-PET) in addition to CT may be superior to CT alone in the diagnosis of pancreatic cancer, increasing sensitivity by 9 per cent and specificity by 2 per cent;40 PET CT may be particularly useful in the diagnosis of cystic
pancreatic masses.41 Although the data at present are insufficient to recommend the routine use of PET combined with CT in diagnosis, it is likely to become increasingly employed as this technology becomes more widely available. Both CT and PET are of use in determining the presence of distant metastatic disease. Endoscopic ultrasound (EUS) is superior to conventional CT in the diagnosis of small pancreatic tumours.42 Recent advances in CT technology, with the advent of spiral and multi-array detector CT scanning, have greatly improved the sensitivity of this modality. The use of EUS in the diagnosis of pancreatic cancer is restricted to those cases in which no mass is seen on multi-detector CT scanning.43 Once a diagnosis has been established, EUS is a cost-effective intervention when used to define the relationship of a pancreatic mass to vascular anatomy prior to surgical resection.44–46 Furthermore, EUS may facilitate fine-needle-guided aspirate samples for cytology, which is safer and more accurate than percutaneous biopsy techniques.47–49 Laparoscopy can detect small peritoneal deposits that are not visualized using conventional imaging and may avoid an unnecessary laparotomy in up to 29 per cent of patients. Therefore it is still recommended in the routine preoperative work-up of patients with potentially resectable pancreatic tumours.50–52
TREATMENT Surgical management GENERAL PRINCIPLES
The preoperative assessment of patients with pancreatic cancer should include careful assessment of their nutritional status and cardiac, pulmonary and renal function. There is considerable debate regarding the management of obstructive jaundice prior to resection in those with pancreatic head tumours. No improvement in postoperative morbidity and mortality has been demonstrated with the routine use of biliary stenting prior to pancreatic resection.53 Indeed, some studies suggest that preoperative biliary drainage should be avoided wherever possible, as it is associated with increased morbidity (particularly sepsis) and mortality.54,55 Therefore the use of biliary stenting in patients with resectable pancreatic cancer should be restricted to those for whom surgical resection is likely to be delayed by several weeks or where jaundice is associated with coagulopathy, sepsis or renal dysfunction. Where possible, the resectability of the tumour should be determined prior to laparotomy. A small proportion of patients proceed to surgery without a preoperative tissue diagnosis. This is appropriate as the preoperative clinical and radiological features suggestive of pancreatic cancer are correct in approximately 90 per cent of cases. The most common differential in these patients is of chronic pancreatitis, and resection may also be appropriate for them.
Treatment 577
SURGICAL RESECTION
Whipple first performed a one-stage pancreatico-duodenectomy, the operation that now bears his name, in 1942. The classical Whipple’s procedure involves a distal gastrectomy, cholecystectomy, and removal of the distal common bile duct, head of the pancreas, duodenum, proximal jejunum and regional lymph nodes. Thereafter gastrojejunostomy, pancreatico-jejunostomy and choledochojejunostomy are fashioned. A modified procedure with preservation of the distal stomach and pylorus is performed by some surgeons, particularly if the pancreatic tumour is small and confined to the head of the pancreas. Japanese surgeons have advocated a more extensive resection including an extended lymphadenectomy, often including resection of sections of the portal and superior mesenteric veins. Only one randomized clinical trial comparing a standard pancreatico-duodenectomy with resection including extended lymphadenectomy has been published. This trial shows no significant difference in 1-year, 3-year and 5-year survival rates and although operative mortality and morbidity in the two arms of the trial were comparable, those undergoing extended lymphadenectomy had a poorer quality of life 4 months following surgery than those undergoing the standard procedure.56 In view of this, radical resection with extended lymphadenectomy should still be considered experimental. Previously, the involvement of the portal vein or superior mesenteric vein by the tumour was considered a contraindication to surgical resection. Several surgeons have now published results of the surgical resection of tumours involving these veins followed by venous reconstruction. Their results suggest that 5-year survival rates are similar to those achieved in patients without portal or superior mesenteric venous invasion,7,57–62 assuming a complete resection is achieved. It is clear that hospital volume influences the outcome in patients undergoing pancreatic resection.28,63–65 This is in part related to surgical specialization, but also to expertise in the supportive care of patients following pancreatic resection and improvements in postoperative management. In high-volume centres, operative mortality for Whipple’s resection is now less than 2 per cent. SURGERY
Surgical resection offers the only chance of cure from pancreatic cancer. Despite this, the median survival following surgery alone for this disease is poor, ranging from 13 months to 18 months. The 5-year survival rate following surgery alone is between 10 per cent and 20 per cent.66–68 Tumour size has been established as an independent prognostic factor following surgery, with tumours less than 3 cm in diameter associated with a greater 5-year survival rate than larger tumours.67 The presence of regional lymphnode metastases or intra-pancreatic perineural invasion of tumour is also associated with a poorer outcome
following surgery.69–71 Other variables considered to be independent prognostic factors are tumour grade, DNA ploidy and microscopic involvement of tumour resection margins.
Post-surgical adjuvant therapy CHEMOTHERAPY AND CHEMO-RADIOTHERAPY
The poor survival of those who have undergone resection for pancreatic cancer has led to the investigation of postoperative adjuvant treatment. Over recent years there has been considerable progress in the understanding of the role of both chemotherapy and chemo-radiotherapy following surgery. Two trials have compared chemo-radiotherapy to observation following resection of pancreatic cancer. The first of these, conducted by the Gastrointestinal Tumour Study Group (GITSG), randomized 43 patients recruited over an 8-year period to treatment with 40 Gy in 20 fractions in a split course along with 5-fluorouracil -FU). The 5-FU was then continued as maintenance therapy until relapse or for 2 years. Despite the small numbers, this trial showed a significant improvement in median survival, from 11 months to 20 months, with chemo-radiotherapy.72 The European Organisation for the Research and Treatment of Cancer (EORTC) subsequently tried to reproduce these results, randomizing 114 patients to receive postoperative treatment with a continuous infusion of 5-FU along with radiotherapy administered in the same schedule as in the GITSG trial. There was no significant difference in either overall 2-year or 5-year survival in this trial.73 However, in contrast to the GITSG trial, patients randomized to chemo-radiotherapy in the EORTC trial did not receive maintenance chemotherapy. The ESPAC 1 trial represents the largest trial of postoperative adjuvant therapy in patients following resection of pancreatic cancer published to date. This trial employed a 2 2 factorial design, randomizing patients to observation, chemotherapy, chemo-radiotherapy alone or chemoradiotherapy followed by chemotherapy (Fig. 23.2). The chemotherapy used in this trial was six cycles of 5FU and folinic acid administered in a modified Mayo Clinic schedule (5-FU 425 mg/m2 and folinic acid 20 mg/m2 daily on days 1–5, repeated every 28 days). This trial demonstrated a significant benefit in favour of the use of postoperative chemotherapy, with a 29 per cent reduction in the risk of death. The median, 2-year and 5-year survival figures were 21 months (40 per cent and 21 per cent respectively) in those receiving chemotherapy compared to 15.5 months (30 per cent and 8 per cent) in those not receiving chemotherapy. The trial did not show any benefit for the use of chemo-radiotherapy following surgery.74 In addition, a recent meta-analysis has examined individual patient data from five randomized clinical trials in the adjuvant treatment of pancreatic cancer. Again there
578 Pancreas
289 patients with histologically proven adenocarcinoma of the pancreas who had undergone potentially curative resection
69 assigned to observation
73 assigned to chemoradiotherapy
72 assigned to chemoradiotherapy and chemotherapy
75 assigned to chemotherapy
Preoperative (neoadjuvant) treatment in potentially resectable pancreatic cancer
Treatment comparison
No chemoradiotheraphy vs. chemoradiotheraphy (144 vs. 145)
No chemotheraphy vs. chemotheraphy (142 vs. 147)
Figure 23.2 Design of the ESPAC 1 Trial. (Copyright Neoptolemos JP, et al, N Engl J Med, March 18 2004; 350:12.)
In view of the low resectability rates in those with localized pancreatic cancer, a number of investigators have conducted studies of preoperative treatment with chemoradiotherapy in an attempt to improve resectability. The use of chemo-radiotherapy may have three advantages. 1. The use of preoperative rather than postoperative treatment allows a greater proportion of patients to receive all components of the planned treatment. 2. Preoperative treatment allows a period of observation, which will avoid patients with rapidly progressive disease undergoing surgery. 3. Preoperative treatment may allow the down-staging of localized tumours, increasing the chance of obtaining an R0 resection.
100 75 % Survival
It is clear from these trials that adjuvant chemotherapy may be considered in the routine care of patients following pancreatic resection. 5-Fluorouracil-based treatment is considered the standard of care. However, gemcitabine appears promising, and clinical trials comparing adjuvant treatment with 5-FU and gemcitabine are ongoing.
50 25 0 0
12
24
36
48
60
29 49
16 33
Months No. at risk No CT 338 CT 348
184 230
88 125 No CT
44 75 CT
Figure 23.3 Pooled survival data from randomized trials of adjuvant chemotherapy in pancreatic cancer. (Adapted from Stocken et al., Br J Cancer 2005, 928)
appears to be no benefit from the use of adjuvant chemoradiotherapy. However, there is a significant benefit from the use of postoperative adjuvant chemotherapy, with a 25 per cent reduction in the risk of death (Fig. 23.3). Subgroup analysis in this meta-analysis suggests that there may be some benefit from postoperative chemo-radiotherapy in those with positive resection margins, and the use of adjuvant chemo-radiotherapy in this indication should be the subject of further randomized clinical trials.68 The nucleoside analogue gemcitabine is the current standard of care in the chemotherapeutic management of advanced pancreatic cancer. Recently, an interim report of a randomized trial of gemcitabine following surgery for pancreatic cancer has been reported. The trial randomized 378 patients who had undergone complete resection of their primary pancreatic cancer to observation or treatment with gemcitabine. The use of gemcitabine is associated with a significant improvement in disease-free survival, although overall survival data are still awaited.75
To date, the published data regarding this approach comprise small phase II feasibility trials or reports of institutional experience. Although numerous series suggest that there may be a survival advantage,23,76–79 these must be interpreted with caution due to the highly selected patient populations studied. No randomized controlled trial comparing this approach to surgery alone has yet been conducted and, in the absence of randomized trial data, this approach should be considered experimental. Despite this, the National Comprehensive Cancer Network in the USA does recommend the use of preoperative chemo-radiotherapy for selected patients. For patients with localized, but unresectable, pancreatic cancer, the use of chemo-radiotherapy as the main treatment modality may be justified (see later) and some authors suggest that resection should be considered if assessment following treatment indicates that the disease has become resectable. This, however, represents a minority (approximately 10–25 per cent) of patients undergoing such treatment.80–82
Treatment of locally advanced unresectable pancreatic cancer The use of combined-modality treatment with chemoradiotherapy for localized unresectable pancreatic cancer is commonplace. However, this practice is based on largely circumstantial evidence of improvements in overall survival. Although there are many phase II trials of this
Treatment 579
Table 23.2 Randomized trials in locally advanced unresectable pancreatic cancer Reference
Gy
Randomized (evaluable)
Chemo-radiotherapy versus radiotherapy alone Moertel et al. 1969 (86) 35–40 5-FU 35–40 Placebo GITSG 9273 1981 (87) 60 5-FU 40 5-FU 60 – Chemo-radiotherapy versus chemotherapy alone ECOG 1985 (85) 40 5-FU – 5-FU GITSG 9283 1988 (84) 54 5-FU SMF – SMF
NR (32) NR (32) 111 (86) 117 (83) 25 (25) NR (47) NR (44) 24 (22) 24 (21)
Chemo-radiotherapy with one drug/combination versus another SWOG 1989 (88) 60 mCCNU 5-FU NR (33) 60 mCCNU 5-FU NR (29) testolactone GITSG 9277 1985 (89) 60 5-FU 79 (73) 40 Doxorubicin 78 (70) Earle et al. 1994 (90) 50–60 5-FU 44 (44) 50–60 Hycanthone 43 (43) Li et al. 2003 (7) 50.4–61.2 Gemcitabine 18 50.4–61.2 5-FU 16
Median survival (months)
1-year survival (estimate)
10.4 6.3 (mean) 11.4 8.4 5.3
NR (25) NR (6) 44 39 14
8.3 8.2 9.7 7.4
NR (28) NR (31) 41 19
8.8 6.9
NR (40) NR (27)
8.5 7.6 7.8 7.8 14.5 6.7
NR (33) NR (26) NR (34) NR (26) NR NR
5-FU, 5-fluorouracil; SMF, streptozocin, mitomycin, and 5-fluorouracil; mCCNU, methyl CCNU; NR, not reported; GITSG, Gastrointestinal Tumour Study Group; ECOG, Eastern Co-operative Oncology Group; SWOG, South Western Oncology Group.
combined-modality treatment in the medical literature, the data they provide are difficult to interpret due to variations in the chemotherapy schedules employed and also in radiotherapy dosage and fractionation. The randomized trials of chemo-radiotherapy for localized pancreatic cancer are shown in Table 23.2. Only one trial has compared chemo-radiotherapy to observation, randomizing 31 patients and demonstrating a doubling of median survival associated with the use of chemo-radiotherapy compared to observation.83 Due to the small size of this trial, no firm conclusions can be drawn. However, the 13.2-month median survival for those treated with chemo-radiotherapy is similar to that other series. Two randomized trials comparing radiotherapy alone to chemo-radiotherapy have been conducted by the GITSG. Although these trials are small, the results are consistent and suggest that the addition of fluoropyrimidinebased chemotherapy to radiation is associated with a significant survival benefit. The use of chemo-radiotherapy compared to chemotherapy alone has been examined in two randomized trials. The results of these trials are conflicting, with one suggesting a 2-month improvement in median survival and doubling of 1-year survival,84 whereas the other does not suggest any advantage.85 This has led to a divergence in opinion regarding the use of chemo-radiotherapy in localized pancreatic cancer and greater use of this modality in North America
compared to Europe. The data that do exist suggest that, for selected individuals, there may be benefit associated with the use of combined-modality treatment over chemotherapy alone; however, this benefit is small. Recent trials have compared fluoropyrimidine-based chemo-radiotherapy to radiotherapy in combination with other agents such as gemcitabine, cisplatin or the taxanes. Again, these trials are small and do not have sufficient power to give definitive evidence for the benefit of one combination over another. However, the use of gemcitabine in combination with radiotherapy does appear worthy of further study.7 Although the subject of treatment with chemoradiotherapy remains controversial, there is clearly a need for large, well-designed and appropriately powered randomized clinical trials.
Palliative chemotherapy for advanced pancreatic cancer The benefit of palliative chemotherapy for advanced pancreatic cancer has been consistently demonstrated in clinical trials. A single-agent phase II trial had suggested that the objective radiological response rate to gemcitabine was disappointing, at 10 per cent. However, a proportion of patients seemed to gain clinical benefit from treatment assessed by improvements in pain control, reduction in
580 Pancreas
Analgesic consumption
Pain intensity
Performance status
Pain
Responder Improvement in both parameters Stable in one parameter, improvement in the other parameter
Stable in both parameters
Nonresponder Worsening in either parameter
Weight
Responder ≥7% increase in body weight
Nonresponder Stable or decreased weight
Figure 23.4 Definition of clinical benefit response
analgesic consumption, performance status and weight (Fig. 23.4).91 The subsequent randomized trial was relatively small, with 126 patients randomized to receive either gemcitabine (1000 mg/m2 per week for 7 weeks followed by 1 week’s rest, then for 3 weeks out of every 4) or weekly treatment with 5-FU (600 mg/m2). This demonstrated a modest but statistically significant improvement in median overall survival from 4.41 months to 5.65 months and in 1year survival rate from 2 per cent to 18 per cent. More importantly, the use of gemcitabine resulted in a clinical benefit response in 24 per cent of patients, compared to 4 per cent receiving 5-FU.92 Although this trial is flawed due to the choice of comparator arm, and the method of assessment of clinical benefit remains controversial, these results led to the licensing of gemcitabine for the treatment of advanced pancreatic cancer by regulatory bodies in Europe and the USA. Furthermore, based on this trial, gemcitabine is accepted as the current standard of care for the management of this disease and the comparator arm for subsequent randomized clinical trials. A number of cytotoxic and molecularly targeted agents have been assessed in combination with gemcitabine. The cytotoxic agents assessed include taxanes, platinum analogues and fluoropyrimidines; the results of phase II and phase III trials of these agents in addition to gemcitabine are shown in Table 23.2. Combinations of gemcitabine and other cytotoxic agents have not resulted in statistically significant improvements in overall survival.7,93–95 Despite this, there are indications of improved activity, with improvements in disease response rates and clinical benefit response observed in some studies. Both oxaliplatin and cisplatin have been compared to gemcitabine alone. Despite initially promising results with both combinations, neither resulted in a statistically significant overall survival benefit over gemcitabine alone.94,95 Although the combination of oxaliplatin and gemcitabine does not result in an overall survival advantage, there is an
increase in the proportion of patients experiencing a clinical benefit response. Furthermore, in the subgroup of patients with metastatic disease, the combination was associated with an improvement in median survival of 1.8 months. Recently, the UK National Cancer Research Institute completed a study comparing the combination of gemcitabine and capecitabine with gemcitabine alone. This trial showed a significant improvement in overall survival, from 6 months with gemcitabine to 7.4 months for the combination. There was also an improvement in 1-year survival from 19 per cent to 26 per cent.96 The litany of failed agents continues when molecular therapies are considered. Matrix metalloprotease inhibitors and farnesyl transferase inhibitors have failed to show any benefit over, or in combination with, gemcitabine.97–99 One strategy that has shown promise is the inhibition of epidermal growth factor receptor (EGFR) signalling. Two agents that affect this have been examined in pancreatic cancer. The first is erlotinib, an orally administered type 1 tyrosine kinase inhibitor. In a randomized placebo-controlled trial, erlotinib combined with gemcitabine resulted in a small improvement in overall survival, of approximately 2 weeks, but a 7 per cent improvement in the 1-year survival rate. This improvement is at the cost of increased toxicity. Although the improvement in survival is statistically significant, it is not clear that it is clinically significant. The second anti-EGF receptor (EGFR) strategy is the use of anti-EGFR monoclonal antibodies. The best known of these are cetuximab and panitumumab, and phase II trials of these agents, combined with gemcitabine, suggest that this strategy may be worthy of further study in randomized trials.
Special considerations in the management of pancreatic cancer Patients with pancreatic cancer are often debilitated by a number of problems indirectly related to their disease. Recurrent biliary obstruction and biliary sepsis relating to biliary stent dysfunction are common and lead to significant delays in the administration of chemotherapy, even in those with good performance status. Diabetes is often very difficult to manage in these patients, as blood glucose levels are often very labile and they have a degree of insulin resistance. Whilst it is not necessary to achieve optimal diabetic control, it is necessary to maintain reasonable control to prevent problems such as hypoglycaemia and acute complications of hyperglycaemia such as hyperosmolar non-ketotic coma or (less commonly) diabetic ketoacidosis. These patients are also often malnourished, which may be due to pancreatic duct obstruction and a failure of pancreatic exocrine function. It should be noted that approximately 90 per cent of patients with advanced pancreatic cancer develop malabsorption during their illness, and careful note should be taken of the presence or absence of steatorrhoea. Patients describing these symptoms should
References 581
be routinely prescribed pancreatic enzyme supplements to be taken before meals. In addition, careful dietetic assessment and advice are often required. As discussed previously, duodenal obstruction occurs in up to 40 per cent of patients with tumours in the head of pancreas. This is increasingly being managed by duodenal stenting. Pain from pancreatic cancer is usually described as upper lumbar back pain or epigastric pain and often has neuropathic features. The use of opiate analgesia and adjuvant analgesia such as tricyclic antidepressants or anticonvulsants is often helpful, but for some patients pain may be intractable. Options such as coeliac plexus nerve block or ablation may need to be considered. Coeliac plexus block may be performed using a number of different techniques, the most reliable of which is done under direct visualization at the time of laparotomy for those patients undergoing biliary or gastric bypass surgery. However, this is not applicable to the majority of patients. Other less invasive techniques include percutaneous ablation under radiological guidance or ablation of the coeliac plexus under EUS guidance, which may be more accurate. For those undergoing chemical ablation of the coeliac plexus, the benefit is often short lived, lasting from 6 to 12 weeks. For many patients this is acceptable due to their short life expectancy; however, for those with a better performance status and a longer life expectancy, thoracoscopic ablation of the thoracic splanchnic nerves may be considered and may provide a longer duration of pain control than chemical ablation.100
REFERENCES ●1
●2
●3
●4
◆5
●6
●7
●8
KEY LEARNING POINTS ●
●
●
●
●
Aetiology: the major aetiological factors implicated in the development of pancreatic adenocarcinoma. Genetics: the genetic basis of familial pancreatic malignancy and the common genetic changes described in sporadic pancreatic cancer. Clinical features: the common presenting features of pancreatic cancer, its differential diagnosis and investigation. The role of surgery in management: the staging of pancreatic cancer, the indications for curative and palliative surgery and the outcomes thereof. The role of chemotherapy and radiotherapy in treatment: the role of palliative treatment with chemotherapy in advanced pancreatic cancer and the potential role and controversies surrounding radiotherapy in the management of localized pancreatic cancer. The controversies surrounding the use of adjuvant therapy in pancreatic cancer and the differences in approach between Europe and North America.
●9
●10
●11
●12
●13
●14
●15
Dhir V, Mohandas KM. Epidemiology of digestive tract cancers in India. IV. Gall bladder and pancreas. Indian J Gastroenterol 1999; 18(1):24–8. Williams RR, Horm JW. Association of cancer sites with tobacco and alcohol consumption and socioeconomic status of patients: interview study from the Third National Cancer Survey. J Natl Cancer Inst 1977; 58(3):525–47. Bonelli L, Aste H, Bovo P, et al. Exocrine pancreatic cancer, cigarette smoking, and diabetes mellitus: a case-control study in northern Italy. Pancreas 2003; 27(2):143–9. Alguacil J, Silverman DT. Smokeless and other noncigarette tobacco use and pancreatic cancer: a case-control study based on direct interviews. Cancer Epidemiol Biomarkers Prev 2004; 13(1):55–8. Villeneuve PJ, Johnson KC, Mao Y, Hanley AJ. Environmental tobacco smoke and the risk of pancreatic cancer: findings from a Canadian population-based case-control study. Can J Public Health 2004; 95(1):32–7. Bueno de Mesquita HB, Maisonneuve P, Runia S, Moerman CJ. Intake of foods and nutrients and cancer of the exocrine pancreas: a population-based case-control study in The Netherlands. Int J Cancer 1991; 48(4):540–9. Lutz MP, Cutsem EV, Wagener T, et al. Docetaxel plus gemcitabine or docetaxel plus cisplatin in advanced pancreatic carcinoma: randomized phase II study 40984 of the European Organisation for Research and Treatment of Cancer Gastrointestinal Group. J Clin Oncol 2005; 23(36):9250–6. Nothlings U, Wilkens LR, Murphy SP, Hankin JH, Henderson BE, Kolonel LN. Meat and fat intake as risk factors for pancreatic cancer: the multiethnic cohort study. J Natl Cancer Inst 2005; 97(19):1458–65. Nkondjock A, Krewski D, Johnson KC, Ghadirian P. Specific fatty acid intake and the risk of pancreatic cancer in Canada. Br J Cancer 2005; 92(5):971–7. Berrington de Gonzalez A, Sweetland S, Spencer E. A metaanalysis of obesity and the risk of pancreatic cancer. Br J Cancer 2003; 89(3):519–23. Huxley R, Ansary-Moghaddam A, Berrington de Gonzalez A, Barzi F, Woodward M. Type-II diabetes and pancreatic cancer: a meta-analysis of 36 studies. Br J Cancer 2005; 92(11):2076–83. Andren-Sandberg A, Dervenis C, Lowenfels B. Etiologic links between chronic pancreatitis and pancreatic cancer. Scand J Gastroenterol 1997; 32(2):97–103. Bansal P, Sonnenberg A. Pancreatitis is a risk factor for pancreatic cancer. Gastroenterology 1995; 109(1):247–51. Ekbom A, McLaughlin JK, Karlsson BM, et al. Pancreatitis and pancreatic cancer: a population-based study. J Natl Cancer Inst 1994; 86(8):625–7. Lowenfels AB, Maisonneuve P, Cavallini G, et al. Pancreatitis and the risk of pancreatic cancer. International Pancreatitis Study Group. N Engl J Med 1993; 328(20):1433–7.
582 Pancreas
●16
●17
●18
●19
●20
●21
◆22
●23
●24
●25
●26
●27
●28
◆29
●30
La Vecchia C, Liati P, Decarli A, Negri E, Franceschi S. Coffee consumption and risk of pancreatic cancer. Int J Cancer 1987; 40(3):309–13. MacMahon B, Yen S, Trichopoulos D, Warren K, Nardi G. Coffee and cancer of the pancreas. N Engl J Med 1981; 304(11):630–3. Michaud DS, Giovannucci E, Willett WC, Colditz GA, Fuchs CS. Coffee and alcohol consumption and the risk of pancreatic cancer in two prospective United States cohorts. Cancer Epidemiol Biomarkers Prev 2001; 10(5):429–37. McWilliams RR, Rabe KG, Olswold C, De Andrade M, Petersen GM. Risk of malignancy in first-degree relatives of patients with pancreatic carcinoma. Cancer 2005; 104(2):388–94. Ghadirian P, Liu G, Gallinger S, et al. Risk of pancreatic cancer among individuals with a family history of cancer of the pancreas. Int J Cancer 2002; 97(6):807–10. Murphy KM, Brune KA, Griffin C, et al. Evaluation of candidate genes MAP2K4, MADH4, ACVR1B, and BRCA2 in familial pancreatic cancer: deleterious BRCA2 mutations in 17%. Cancer Res 2002; 62(13):3789–93. Howe JR, Conlon KC. The molecular genetics of pancreatic cancer. Surg Oncol 1997;6(1):1–18. Aristu J, Canon R, Pardo F, et al. Surgical resection after preoperative chemoradiotherapy benefits selected patients with unresectable pancreatic cancer. Am J Clin Oncol 2003; 26(1):30–6. Biankin AV, Kench JG, Biankin SA, et al. Pancreatic intraepithelial neoplasia in association with intraductal papillary mucinous neoplasms of the pancreas: implications for disease progression and recurrence. Am J Surg Pathol 2004; 28(9):1184–92. Bakkevold KE, Arnesjo B, Kambestad B. Carcinoma of the pancreas and papilla of Vater: presenting symptoms, signs, and diagnosis related to stage and tumour site. A prospective multicentre trial in 472 patients. Norwegian Pancreatic Cancer Trial. Scand J Gastroenterol 1992; 27(4):317–25. Bramhall SR, Allum WH, Jones AG, Allwood A, Cummins C, Neoptolemos JP. Treatment and survival in 13,560 patients with pancreatic cancer, and incidence of the disease, in the West Midlands: an epidemiological study. Br J Surg 1995; 82(1):111–15. Gudjonsson B. Carcinoma of the pancreas: critical analysis of costs, results of resections, and the need for standardized reporting. J Am Coll Surg 1995; 181(6):483–503. Parks RW, Bettschart V, Frame S, Stockton DL, Brewster DH, Garden OJ. Benefits of specialisation in the management of pancreatic cancer: results of a Scottish population-based study. Br J Cancer 2004; 91(3):459–65. Lucarotti ME, Habib NA, Kelly SB, et al. Clinical evaluation of combined use of CEA, CA19-9 and CA50 in the serum of patients with pancreatic carcinoma. Eur J Surg Oncol 1991; 17(1):51–3. Nazli O, Bozdag AD, Tansug T, Kir R, Kaymak E. The diagnostic importance of CEA and CA 19-9 for the early diagnosis of pancreatic carcinoma. Hepatogastroenterology 2000; 47(36):1750–2.
●31
●32
●33
●34
●35
●36
●37
◆38
●39
●40
●41
●42
●43
●44
Stemmler J, Stieber P, Szymala AM, et al. Are serial CA 19-9 kinetics helpful in predicting survival in patients with advanced or metastatic pancreatic cancer treated with gemcitabine and cisplatin? Onkologie 2003; 26(5):462–7. Ziske C, Schlie C, Gorschluter M, et al. Prognostic value of CA 19-9 levels in patients with inoperable adenocarcinoma of the pancreas treated with gemcitabine. Br J Cancer 2003; 89(8):1413–17. Micke O, Bruns F, Kurowski R, et al. Predictive value of carbohydrate antigen 19-9 in pancreatic cancer treated with radiochemotherapy. Int J Radiat Oncol Biol Phys 2003; 57(1):90–7. Heinemann V, Schermuly MM, Stieber P, et al. CA19-9: a predictor of response in pancreatic cancer treated with gemcitabine and cisplatin. Anticancer Res 1999; 19(4A):2433–5. Ozkan H, Kaya M, Cengiz A. Comparison of tumor marker CA 242 with CA 19-9 and carcinoembryonic antigen (CEA) in pancreatic cancer. Hepatogastroenterology 2003; 50(53):1669–74. Pezzilli R, Billi P, Plate L, Laudadio MA, Sprovieri G. Serum CA 242 in pancreatic cancer. Comparison with CA 19-9 and CEA. Ital J Gastroenterol 1995; 27(6):296–9. Rothlin MA, Joller H, Largiader F. CA 242 is a new tumor marker for pancreatic cancer. Cancer 1993; 71(3):701–7. Bipat S, Phoa SS, van Delden OM, et al. Ultrasonography, computed tomography and magnetic resonance imaging for diagnosis and determining resectability of pancreatic adenocarcinoma: a meta-analysis. J Comput Assist Tomogr 2005; 29(4):438–45. Adamek HE, Albert J, Breer H, Weitz M, Schilling D, Riemann JF. Pancreatic cancer detection with magnetic resonance cholangiopancreatography and endoscopic retrograde cholangiopancreatography: a prospective controlled study. Lancet 2000; 356(9225):190–3. Orlando LA, Kulasingam SL, Matchar DB. Meta-analysis: the detection of pancreatic malignancy with positron emission tomography. Aliment Pharmacol Ther 2004; 20(10):1063–70. Sperti C, Pasquali C, Chierichetti F, Liessi G, Ferlin G, Pedrazzoli S. Value of 18-fluorodeoxyglucose positron emission tomography in the management of patients with cystic tumors of the pancreas. Ann Surg 2001; 234(5):675–80. Rosch T, Lorenz R, Braig C, Classen M. Endoscopic ultrasonography in the diagnosis and staging of pancreatic and biliary tumours. Endoscopy 1992; 24(Suppl. 1):304–8. Aghrawal B, Abu Hamda E, Molke KL, Correa AM, Ho L. Endoscopic ultrasound-guided fine needle aspirate and multidetector spiral CT in the diagnosis of pancreatic cancer. Am J Gastroenterol 2004; 99(5):844–50. Tierney WM, Fendrick AM, Hirth RA, Scheiman JM. The clinical and economic impact of alternative staging strategies for adenocarcinoma of the pancreas. Am J Gastroenterol 2000; 95(7):1708–13.
References 583
◆45
●46
●47
●48
●49
●50
●51
●52
●53
●54
●55
●56
●57
◆58
●59
Mertz HR, Sechopoulos P, Delbeke D, Leach SD. EUS, PET, and CT scanning for evaluation of pancreatic adenocarcinoma. Gastrointest Endosc 2000; 52(3):367–71. Delbeke D, Pinson CW. Pancreatic tumors: role of imaging in the diagnosis, staging, and treatment. J Hepatobiliary Pancreat Surg 2004; 11(1):4–10. Afify AM, al-Khafaji BM, Kim B, Scheiman JM. Endoscopic ultrasound-guided fine needle aspiration of the pancreas. Diagnostic utility and accuracy. Acta Cytol 2003; 47(3):341–8. Binmoeller KF, Thul R, Rathod V, et al. Endoscopic ultrasound-guided, 18-gauge, fine needle aspiration biopsy of the pancreas using a 2.8 mm channel convex array echoendoscope. Gastrointest Endosc 1998; 47(2):121–7. Faigel DO, Ginsberg GG, Bentz JS, Gupta PK, Smith DB, Kochman ML. Endoscopic ultrasound-guided real-time fineneedle aspiration biopsy of the pancreas in cancer patients with pancreatic lesions. J Clin Oncol 1997; 15(4):1439–43. Menack MJ, Spitz JD, Arregui ME. Staging of pancreatic and ampullary cancers for resectability using laparoscopy with laparoscopic ultrasound. Surg Endosc 2001; 15(10):1129–34. Liu RC, Traverso LW. Diagnostic laparoscopy improves staging of pancreatic cancer deemed locally unresectable by computed tomography. Surg Endosc 2005; 19(5):638–42. Reddy KR, Levi J, Livingstone A, et al. Experience with staging laparoscopy in pancreatic malignancy. Gastrointest Endosc, 1999; 49(4 Pt 1):498–503. Pitt HA, Gomes AS, Lois JF, Mann LL, Deutsch LS, Longmire WP Jr. Does preoperative percutaneous biliary drainage reduce operative risk or increase hospital cost? Ann Surg 1985; 201(5):545–53. Heslin MJ, Brooks AD, Hochwald SN, Harrison LE, Blumgart LH, Brennan MF. A preoperative biliary stent is associated with increased complications after pancreatoduodenectomy. Arch Surg 1998; 133(2):149–54. Povoski SP, Karpeh MS Jr, Conlon KC, Blumgart LH, Brennan MF. Association of preoperative biliary drainage with postoperative outcome following pancreaticoduodenectomy. Ann Surg 1999; 230(2):131–42. Farnell MB, Pearson RK, Sarr MG, et al. A prospective randomized trial comparing standard pancreatoduodenectomy with pancreatoduodenectomy with extended lymphadenectomy in resectable pancreatic head adenocarcinoma. Surgery 2005; 138(4):618–28; discussion 28–30. Jain S, Sacchi M, Vrachnos P, Lygidakis NJ. Carcinoma of the pancreas with portal vein involvement – our experience with a modified technique of resection. Hepatogastroenterology 2005; 52(65):1596–600. Hartel M, Niedergethmann M, Farag-Soliman M, et al. Benefit of venous resection for ductal adenocarcinoma of the pancreatic head. Eur J Surg 2002; 168(12):707–12. Lygidakis NJ, Singh G, Bardaxoglou E, et al. Mono-bloc total spleno-pancreaticoduodenectomy for pancreatic head carcinoma with portal-mesenteric venous invasion. A prospective randomized study. Hepatogastroenterology 2004; 51(56):427–33.
●60
●61
●62
●63
●64
●65
◆66
●67
●68
◆69
◆70
●71
●72
●73
●74
●75
Nakao A, Takeda S, Sakai M, et al. Extended radical resection versus standard resection for pancreatic cancer: the rationale for extended radical resection. Pancreas 2004; 28(3):289–92. Howard TJ, Villanustre N, Moore SA, et al. Efficacy of venous reconstruction in patients with adenocarcinoma of the pancreatic head. J Gastrointest Surg 2003; 7(8):1089–95. Capussotti L, Massucco P, Ribero D, Vigano L, Muratore A, Calgaro M. Extended lymphadenectomy and vein resection for pancreatic head cancer: outcomes and implications for therapy. Arch Surg 2003; 138(12):1316–22. Birkmeyer JD, Siewers AE, Finlayson EVA, et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002; 346(15):1128–37. Glasgow RE, Mulvihill SJ. Hospital volume influences outcome in patients undergoing pancreatic resection for cancer. West J Med 1996; 165(5):294–300. Janes RH Jr, Niederhuber JE, Chmiel JS, et al. National patterns of care for pancreatic cancer. Results of a survey by the Commission on Cancer. Ann Surg 1996; 223(3):261–72. Boadas J, Balart J, Capella G, Lluis F, Farre A. Survival of cancer of the pancreas. Bases for new strategies in diagnosis and therapy. Rev Esp Enferm Dig 2000; 92(5):316–25. Yeo CJ, Cameron JL, Lillemoe KD, et al. Pancreaticoduodenectomy for cancer of the head of the pancreas. 201 patients. Ann Surg 1995; 221(6):721–31; discussion 31–3. Stocken DD, Buchler MW, Dervenis C, et al. Meta-analysis of randomised adjuvant therapy trials for pancreatic cancer. Br J Cancer 2005; 92(8):1372–81. Mu DQ, Peng SY, Wang GF. Risk factors influencing recurrence following resection of pancreatic head cancer. World J Gastroenterol 2004; 10(6):906–9. Meyer W, Jurowich C, Reichel M, Steinhauser B, Wunsch PH, Gebhardt C. Pathomorphological and histological prognostic factors in curatively resected ductal adenocarcinoma of the pancreas. Surg Today 2000; 30(7):582–7. Ozaki H, Hiraoka T, Mizumoto R, et al. The prognostic significance of lymph node metastasis and intrapancreatic perineural invasion in pancreatic cancer after curative resection. Surg Today 1999; 29(1):16–22. Kalser MH, Ellenberg SS. Pancreatic cancer. Adjuvant combined radiation and chemotherapy following curative resection. Arch Surg 1985; 120(8):899–903. Klinkenbijl JH, Jeekel J, Sahmoud T, et al. Adjuvant radiotherapy and 5-fluorouracil after curative resection of cancer of the pancreas and periampullary region: phase III trial of the EORTC Gastrointestinal Tract Cancer Cooperative Group. Ann Surg 1999; 230(6):776–82; discussion 82–4. Neoptolemos JP, Stocken DD, Friess H, et al. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med 2004; 350(12):1200–10. Neuhaus P, Oettle H, Post S, et al. A randomised, prospective, multicenter, phase III trial of adjuvant chemotherapy with gemcitabine vs. observation in patients with resected pancreatic cancer. J Clin Oncol (Meeting Abstracts) 2005; 23(16 Suppl.):LBA4013
584 Pancreas
●76
●77
●78
●79
●80
●81
●82
●83
●84
●85
●86
●87
●88
Breslin TM, Hess KR, Harbison DB, et al. Neoadjuvant chemoradiotherapy for adenocarcinoma of the pancreas: treatment variables and survival duration. Ann Surg Oncol 2001; 8(2):123–32. Sasson AR, Wetherington RW, Hoffman JP, et al. Neoadjuvant chemoradiotherapy for adenocarcinoma of the pancreas: analysis of histopathology and outcome. Int J Gastrointest Cancer 2003; 34(2–3):121–8. Snady H, Bruckner H, Cooperman A, Paradiso J, Kiefer L. Survival advantage of combined chemoradiotherapy compared with resection as the initial treatment of patients with regional pancreatic carcinoma. An outcomes trial. Cancer 2000; 89(2):314–27. White RR, Hurwitz HI, Morse MA, et al. Neoadjuvant chemoradiation for localized adenocarcinoma of the pancreas. Ann Surg Oncol 2001; 8(10):758–65. Magnino A, Gatti M, Massucco P, et al. Phase II trial of primary radiation therapy and concurrent chemotherapy for patients with locally advanced pancreatic cancer. Oncology 2005; 68(4–6):493–9. Sa Cunha A, Rault A, Laurent C, et al. Surgical resection after radiochemotherapy in patients with unresectable adenocarcinoma of the pancreas. J Am Coll Surg 2005; 201(3):359–65. Smeenk HG, de Castro SM, Jeekel JJ, et al. Locally advanced pancreatic cancer treated with radiation and 5-fluorouracil: a first step to neoadjuvant treatment? Dig Surg 2005; 22(3):191–7. Shinchi H, Takao S, Noma H, et al. Length and quality of survival after external-beam radiotherapy with concurrent continuous 5-fluorouracil infusion for locally unresectable pancreatic cancer. Int J Radiat Oncol Biol Phys 2002; 53(1):146–50. GITSG. Treatment of locally unresectable carcinoma of the pancreas: comparison of combined-modality therapy (chemotherapy plus radiotherapy) to chemotherapy alone. Gastrointestinal Tumor Study Group. J Natl Cancer Inst 1988; 80(10):751–5. Klaassen DJ, MacIntyre JM, Catton GE, Engstrom PF, Moertel CG. Treatment of locally unresectable cancer of the stomach and pancreas: a randomized comparison of 5fluorouracil alone with radiation plus concurrent and maintenance 5-fluorouracil – an Eastern Cooperative Oncology Group study. J Clin Oncol 1985; 3(3):373–8. Moertel CG, Childs DS Jr, Reitemeier RJ, Colby MY Jr, Holbrook MA. Combined 5-fluorouracil and supervoltage radiation therapy of locally unresectable gastrointestinal cancer. Lancet 1969; 2(7626):865–7. Moertel CG, Frytak S, Hahn RG, et al. Therapy of locally unresectable pancreatic carcinoma: a randomized comparison of high dose (6000 rads) radiation alone, moderate dose radiation (4000 rads 5-fluorouracil), and high dose radiation 5-fluorouracil: the Gastrointestinal Tumor Study Group. Cancer 1981; 48(8):1705–10. McCracken JD, Ray P, Heilbrun LK, et al. 5-Fluorouracil, methyl-CCNU, and radiotherapy with or without testolactone for localized adenocarcinoma of the exocrine
●89
●90
●91
●92
●93
●94
●95
●96
●97
●98
●99
●100
pancreas: a Southwest Oncology Group study. Cancer 1980; 46(7):1518–22. GITSG. Radiation therapy combined with adriamycin or 5fluorouracil for the treatment of locally unresectable pancreatic carcinoma. Gastrointestinal Tumor Study Group. Cancer 1985; 56(11):2563–8. Earle JD, Foley JF, Wieand HS, et al. Evaluation of externalbeam radiation therapy plus 5-fluorouracil (5-FU) versus external-beam radiation therapy plus hycanthone (HYC) in confined, unresectable pancreatic cancer. Int J Radiat Oncol Biol Phys 1994; 28(1):207–11. Rothenberg ML, Moore MJ, Cripps MC, et al. A phase II trial of gemcitabine in patients with 5-FU-refractory pancreas cancer. Ann Oncol 1996; 7(4):347–53. Burris HA 3rd, Moore MJ, Andersen J, et al. Improvements in survival and clinical benefit with gemcitabine as firstline therapy for patients with advanced pancreas cancer: a randomized trial. J Clin Oncol 1997; 15(6):2403–13. Berlin JD, Catalano P, Thomas JP, Kugler JW, Haller DG, Benson AB 3rd. Phase III study of gemcitabine in combination with fluorouracil versus gemcitabine alone in patients with advanced pancreatic carcinoma: Eastern Cooperative Oncology Group Trial E2297. J Clin Oncol 2002; 20(15):3270–5. Colucci G, Giuliani F, Gebbia V, et al. Gemcitabine alone or with cisplatin for the treatment of patients with locally advanced and/or metastatic pancreatic carcinoma: a prospective, randomized phase III study of the Gruppo Oncologia dell’Italia Meridionale. Cancer 2002; 94(4):902–10. Louvet C, Labianca R, Hammel P, et al. Gemcitabine in combination with oxaliplatin compared with gemcitabine alone in locally advanced or metastatic pancreatic cancer: results of a GERCOR and GISCAD phase III trial. J Clin Oncol 2005; 23(15):3509–16. Cunningham D, Chau I, Stocken DD, et al. Phase III randomised comparison of gemcitabine (GEM) versus gemcitabine plus capecitabine (GEM-CAP) in patients with advanced pancreatic cancer. Eur J Cancer Suppl 2005; 3(4):4. Bramhall SR, Schulz J, Nemunaitis J, Brown PD, Baillet M, Buckels JA. A double-blind placebo-controlled, randomised study comparing gemcitabine and marimastat with gemcitabine and placebo as first line therapy in patients with advanced pancreatic cancer. Br J Cancer 2002; 87(2):161–7. Moore MJ, Hamm J, Dancey J, et al. Comparison of gemcitabine versus the matrix metalloproteinase inhibitor BAY 12-9566 in patients with advanced or metastatic adenocarcinoma of the pancreas: a phase III trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2003; 21(17):3296–302. Van Cutsem E, van de Velde H, Karasek P, et al. Phase III trial of gemcitabine plus tipifarnib compared with gemcitabine plus placebo in advanced pancreatic cancer. J Clin Oncol 2004; 22(8):1430–8. Imrie CW, Menezes N, Carter CR. Diagnosis of chronic pancreatitis and newer aspects of pain control. Digestion 1999; 60(Suppl. 1):111–13.
24 Biliary tract HEMANT M. KOCHER, AJIT T. ABRAHAM AND SATYA BHATTACHARYA
Tumours of the bile duct (cholangiocarcinoma) Gall-bladder cancer
585 592
TUMOURS OF THE BILE DUCT (CHOLANGIOCARCINOMA)
Future trends References
595 596
in men than women (1.5:1). These differences must carry important messages about aetiology, and may relate to parasitic infestations (see below).
Incidence Tumours of the biliary tract can be divided into those of the intra-hepatic bile ducts, extra-hepatic bile ducts and gall bladder. Intra-hepatic bile duct tumours have been described in Chapter 22, but a brief review is included here. For most epidemiological studies, the hilar cholangiocarcinomas (which are anatomically extra-hepatic) are considered, perhaps mistakenly, as intra-hepatic. Biliary tract tumours account for 10–15 per cent of all primary hepato-biliary cancers and 3 per cent of all gastrointestinal cancers worldwide.1 There is a suggestion that the intra-hepatic cholangiocarcinomas may be rising in incidence, whilst the extra-hepatic cholangiocarcinomas show an opposite trend.1–5 The 2004 data from the UK Office of National Statistics showed that for England and Wales there were 2321 (951 men) new cases of cancers of the liver and intra-hepatic bile-ducts, while for extra-hepatic bile ducts and gall bladder there were 525 (194 men) new cases.6 Age-standardized rates in England and Wales vary from 1.5 to 3.4 per 100 000 population. These figures are in remarkable contrast to those reported from some other areas of the world, notably north-east Thailand, where agestandardized annual incidence rates of 135.4 per 100 000 men and 43.0 per 100 000 women have been reported.7 Other high-risk areas include Japan, Korea and Eastern Europe (European Russia, Czech Republic, Poland), and American Indians have an age-adjusted incidence rate of 4–12 per 100 000 women. In most areas, biliary tumours (except gall-bladder cancers) have a slightly higher incidence
Aetiology Environmental, genetic and anatomic predisposing factors have been associated with the development of cholangiocarcinomas. Anatomic factors include the presence of fibrocystic disease of the liver (Caroli’s disease),8,9 choledochal cysts10 and abnormal pancreatico-biliary duct junction.11 Environmental factors include exposure to toxins such as Thorotrast (a radiologic material used in the 1950s)12 and industrial toxins such as dioxins and polyvinyl chloride.13,14 Infestations of liver-flukes such as Opisthorchis viverrini in Thailand, Laos and Cambodia7,15,16 and Clonorchis sinensis in China and Korea17 are associated with a high incidence of intra-hepatic cholangiocarcinomas in those areas. Smoking and alcohol consumption further increase this risk in these ethnic groups.15,16 Hepatitis C seems to have a causative epidemiologic link (and may explain the rise in cholangiocarcinomas in the West), but a link with hepatitis B remains to be proven.18,19 Underlying cirrhosis increases the risk of intra-hepatic cholangiocarcinomas.20 In the West, the best known predisposing factor is primary sclerosing cholangitis (PSC), which carries a lifetime risk of 9–23 per cent.21–26 However, screening and surveillance in this group, with the current tools, have proven to be less than successful.27 In the Far East, hepatolithiasis is a known predisposing factor, and in fact a stepwise progression from hyperplasia, dysplasia, carcinoma in-situ and adenocacinoma has been shown in patients with hepatolithiasis.27–29
586 Biliary tract
Biliary–enteric anastomosis (done mostly for choledochal cysts) is also a known aetiological factor. Previous cholecystectomy,30 thalassaemia31 and prior external-beam radiotherapy32 may increase the risk of cholangiocarcinoma, but so far the evidence is anecdotal.
Molecular biology and genetics A multi-hit hypothesis has been suggested for carcinogenesis in the bile ducts.33,34 It has been suggested that this occurs in four stages: 1. predisposition and risk factors of biliary cancer, 2. genotoxic events and alterations leading to specific DNA damage and mutation patterns, 3. dysregulation of DNA-repair mechanisms and apoptosis, permitting survival of mutated cells, 4. morphological evolution from pre-malignant biliary lesions to cholangiocarcinoma. Mutations in the p53 and K-ras oncogenes have been the two most commonly studied genetic defects for cholangiocarcinomas, although K-ras mutation has not been found with uniform high frequency.35–37 Other genes implicated in carcinogenesis include c-erbB-2, c-Met, Bcl-2 and p16.38–41
The link between liver-fluke infestation and carcinogenesis is provided by increased in-vivo production of N-nitrosamines and DNA alkylation damage due to the induction of cytochromal enzymes in macrophages.42 Enzymes such as cytochrome P450 (CYP) and N-acetyl transferase (NAT) may have single nucleotide polymorphisms, explaining their differential capacity for degradation of xenobiotics and thus predisposition to biliary-tract cancers.42,43 In addition, there may be similar changes in the various bile-salt exporter proteins that allow changes in levels of exposure to biliary toxins.
Anatomy and lymphatic drainage The difficult tumours at the confluence of the right and left ducts are sometimes eponymously referred to as Klatskin tumours, after Klatskin’s report of 13 cases in 1965.44 Intra-hepatic cholangiocarcinomas could be classified as mass-forming, periductal-infiltrating, intraductal or mixed. The gross appearance of the tumour may be related to its prognosis.45 Formal pathological staging is by the TNM system46 (Table 24.1). Full staging can only be made following surgery and pathological examination of the resected specimen. Bile-duct tumours spread to the lymph-node groups along the proper and common hepatic arteries, the
Table 24.1 Tumour node metastasis (TNM) staging for extra-hepatic bile-duct cancer Primary tumour (T) Tx T0 Tis T1 T2 T3 T4
Primary tumour cannot be assessed No evidence of primary tumour Carcinoma-in-situ Tumour confined to the bile duct Invades beyond the wall of the bile duct Invades adjacent structures: liver, gall bladder and/or unilateral tributaries of portal vein or hepatic artery Invades main portal vein or hepatic artery, or other adjacent structures: duodenum, colon, stomach, abdominal wall
Lymph nodes (N) Nx N0 N1
Regional lymph nodes cannot be assessed No regional lymph-node metastasis Regional lymph-node metastasis
Distant metastasis (M) Mx M0 M1
Distant metastasis cannot be assessed No distant metastasis Distant metastasis
Stage grouping Stage 0 Stage IA Stage IB Stage IIA Stage IIB Stage III Stage IV
Tis TI T2 T3 T1, T2, T3 T4 Any T
N0 N0 N0 N0 N1 Any N Any N
M0 M0 M0 M0 M0 M0 M1
Tumours of the bile duct (cholangiocarcinoma) 587
coeliac nodes and, for distally placed lesions, the retropancreatic and superior mesenteric nodes. It is important to sample lymph nodes when considering curative resection and when resecting tumours to skeletonize the hepatic artery and remove all lymphatic tissues and associated neural tissue. These tumours have a propensity for spread along the subepithelial planes and for longitudinal perineural invasion, both proximally and distally. Such invasion has a negative impact on survival.47 It also produces difficulty in operative diagnosis, as choledochoscopic biopsies may underestimate the extent of tumour. Bileduct cancer may give rise to very well differentiated nests of biliary epithelial cells within lymph nodes. The authors would regard such deposits as a contraindication to major liver resection for attempted cure, although their presence should not preclude local resection for palliation. In the Mayo Clinic regimen of extended neoadjuvant chemo-radiotherapy prior to liver transplantation (see below), a laparotomy is performed to rule out lymph-node metastasis. Lymph-node metastases have been seen to the left supraclavicular lymph nodes (Virchow’s sign).
Natural history Untreated, most patients with bile-duct cancer die within 6 months to 1 year of diagnosis from a combination of local tumour spread and cholangitis. Direct spread to the liver is the rule, but distant involvement of intra-hepatic bile radicles may also be found and may represent a field change. In rare cases, distant metastases occur and the authors have patients with metastatic deposits in bone and distant lymph nodes.
Pathology This has been described well by Weinbren and Mutum.48 The tumours are generally well-differentiated adenocarcinomas (cholangiocarcinomas).49–51 Grossly, papillary (intraductal) lesions predominate in the distal segment and sclerotic stenosing (periductal) lesions in the upper third, while those in the central portion of the ducts tend to be nodular (mass forming). The importance of the location of these tumours lies in their propensity to invade deeply into liver substance by direct extension proximally and also to involve the portal vein and hepatic artery in the region of the hilus – features that may preclude resection. They form acini in which mucin secretion is almost invariable and mucin lakes occur in 50 per cent of cases; this feature may help to distinguish cholangiocarcinomas from hepatocellular carcinoma at the hilus. A dense fibrous reaction surrounds the hilar tumours, which may be indistinguishable from tumour radiologically, or even at operation. Even histological differentiation of this desmoplastic reaction from tumour requires careful and experienced examination of multiple sections. Invasion of nerve trunks occurs in 80 per
cent. The macroscopically papillary variety of tumour may arise in choledochal cysts, or from malignant change in cases of multiple biliary papillomatosis.52 Immunocytochemical staining is positive for carcinoembryonic antigen (CEA) in 50 per cent and for cytokeratin in 80 per cent but is usually negative for alpha-fetoprotein. Proximal to the obstructed bile duct there are changes of cholestasis and acute inflammatory cell infiltrate, with marked peri-lobular fibrosis in longstanding cases. Longstanding obstruction to one or more segmental hepatic ducts produces such intense ductal dilatation and fibrosis that segments or a lobe of the liver may become shrunken and functionally ineffective.48 This phenomenon may be compounded by hepatocyte atrophy if the portal vein to that segment of liver is also involved directly by tumour encasement. This atrophy–hypertrophy phenomenon has to be taken into account when considering surgical resections and/or non-operative decompression.
Symptoms and presentation Cholangiocarcinoma rarely occurs in individuals under the age of 40. Presenting features are determined by the tumour location. Patients with extra-hepatic bile-duct or hilar cholangiocarcinomas usually present with unremitting painless jaundice and pruritus, dark urine and pale stools. Prior abdominal pain consistent with a biliary origin may be present in up to 40 per cent of patients.53 Cholangitis may occur and is suggested by pain in the right hypochondrium, fever and rigors. Weight loss and anorexia are late symptoms. Intra-hepatic cholangiocarcinoma needs to be considered as a possibility in patients presenting with an unknown primary and suspected liver metastasis. Peripheral cholangiocarcinomas may present with pain, malaise and weight loss because of sectoral or lobar biliary obstruction prior to actual involvement of the confluence by the tumour causing jaundice. Patients known to have primary sclerosing cholangitis who develop cholangiocarcinoma may present with a rapid deterioration in their performance status and worsening cholestasis. Occasionally, asymptomatic patients with the disease may be identified during investigation for unexplained increases in cholestatic enzymes. The differential diagnosis for cholangiocarcinoma includes pancreatic, biliary and other pathologies such as malignancy of pancreatic head, peri-ampullary area and duodenum, chronic pancreatitis, stones in the common bile duct, primary sclerosing cholangitis, recurrent pyogenic cholangitis, iatrogenic injuries to the bile duct, parasitic infestations and porta hepatis lymph-node metastases. Rarely, hepatitis can masquerade as obstructive jaundice. Hepatomegaly may be present in up to 75 per cent of patients54 and ascites may be seen in advanced tumours. Virchow’s node (metastasis to left supraclavicular lymph node) should be looked for, as its confirmation with fine-needle aspiration or biopsy indicates advanced metastatic disease.
588 Biliary tract
Diagnosis Most patients with cholangiocarcinoma present with hilar strictures and cholestatic jaundice.55 However, confirmation of the diagnosis can prove difficult. A high index of suspicion and a multidisciplinary investigative approach are required,56 especially in patients with PSC who already have dominant biliary strictures and deranged liver function with secondary biliary cirrhosis. These patients do not always develop significant intra-hepatic biliary dilatation due to pre-existing biliary fibrosis.57 Obstructive jaundice may manifest with an elevation in the serum bilirubin and derangement of the liver enzymes alkaline phosphatase and gamma-glutamyl transpeptidase. Hepatic transaminases may also be elevated. Tumour markers have a limited value, as none is specific for cholangiocarcinoma. Carbohydrate antigen (CA) 19-9 may be elevated in cholangiocarcinoma, but is also raised in other cancers, including pancreatic, colorectal, gastric and gynaecological malignancies,58 cholangitis,59 benign cholestasis and following liver injury. The estimated sensitivity for CA19-9 over 100 U/mL in predicting cholangiocarcinoma in patients without PSC is 53 per cent.60 The specificity of CA19-9 is 76 per cent and 92 per cent in comparison with non-malignant liver disease and benign biliary strictures respectively.60 In the context of PSC, the estimated sensitivity of CA19-9 in predicting cholangiocarcinoma is 38–89 per cent and its specificity is 50–98 per cent.61 Carcinoembryonic antigen is primarily a marker for colorectal cancer, but its combination with CA19-9 may increase diagnostic sensitivity for cholangiocarcinoma, particularly in patients with PSC.61,62 Other new and potentially useful tumour markers that have been studied include Mac-2 binding protein, CA125, interleukin-6, DUPAN-2 and CA50 and biliary fibronectin, although clear data on clinical relevance are awaited.63–65 Accurate imaging is essential for the diagnosis, staging and planning of treatment. Ultrasound is the initial diagnostic investigation of choice. It is highly sensitive for biliaryduct dilatation and may localize the site of obstruction, showing intra-hepatic dilatation only in cases of hilar obstruction and both intra-hepatic and extra-hepatic dilatation in more distal lesions.66 Ultrasound may also suggest alternative diagnoses such as gallstones. Extensive ductal dilatation may not be a feature in PSC because of ductal fibrosis. Doppler ultrasound may suggest vascular encasement or thrombosis.67 Contrast-enhanced, triplephase helical computed tomography (ceCT) can detect ductal dilatation and intra-hepatic mass lesions greater than 1 cm68 as well as the level of biliary obstruction, presence of lymphadenopathy and vascular involvement. The hepatic atrophy–hypertrophy complex may be seen in patients with cholangiocarcinoma where tumours obstruct a single hepatic lobe, with ipsilateral portal vein invasion resulting in ipsilateral hepatic lobar atrophy and contralateral hypertrophy.69 Computed tomography may only help determine resectability in 60 per cent of cases.70
Figure 24.1 A magnetic resonance cholangio-pancreatogram (MRCP) demonstrating a hilar cholangiocarcinoma with marked dilatation of the right and left intra-hepatic bile ducts
Magnetic resonance imaging (MRI) is perhaps the most useful single investigation for cholangiocarcinoma and is rapidly becoming the imaging modality of choice.71,72 It can delineate the extent of tumour and invasion of adjacent hepatic parenchyma by hilar cholangiocarcinoma, and can also detect intra-hepatic cholangiocarcinoma, local lymphadenopathy, vascular involvement and distant metastases.72,73 Magnetic resonance cholangiopancreatography (MRCP) using non-invasive three-dimensional biliary reconstruction provides a roadmap for the biliary anatomy above and below the stricture (Fig. 24.1). Magnetic resonance angiography (MRA) helps to clarify vascular anatomy, which has many normal variants. With the advent of MRA, the role for traditional invasive angiography is limited, although it is occasionally still indicated in selected cases being considered for radical resection. Magnetic resonance cholangiopancreatography is superior to endoscopic retrograde cholangiopancreatography (ERCP) in defining biliary anatomy and resectability as it allows visualization of undrained bile ducts. Being noninvasive, it also avoids the risk of cholangitis.72 However, correlation with the extent of resected cholangiocarcinoma suggests that MRCP underestimates the extent of disease in up to 20 per cent of malignant hilar strictures.74 Endoscopic retrograde cholangiopancreatography and percutaneous transhepatic cholangiography (PTC) are invasive modalities that allow therapeutic drainage and washings, brushings and even intra-ductal biopsies to be obtained for cytopathological analysis (Fig. 24.2). The diagnostic yield is low,71 although sensitivity may be increased by techniques such as fluorescent in-situ hybridisation.75 The choice between PTC and ERCP is usually determined by local expertise, availability and failure of one or the other technique, usually ERCP.71,76 Percutaneous transhepatic cholangiography is preferable for more proximal strictures,69 and often a combination of
Tumours of the bile duct (cholangiocarcinoma) 589
Figure 24.2 An endoscopic retrograde cholangiopancreatogram (ERCP) demonstrating a hilar cholangiocarcinoma. The common bile duct has been occluded with a balloon catheter to fill the proximal biliary system with contrast at high pressure
lymph-node involvement and extent of biliary spread. The combination of laparoscopy with laparoscopic ultrasound may prevent unnecessary laparotomy in up to a third of patients with hilar or gall-bladder cancer deemed resectable radiologically, with an accuracy of 48 per cent for hilar cholangiocarcinomas.77 Histological and cytological confirmation is necessary for the definitive diagnosis of cholangiocarcinoma. Brush cytology is positive only in about 30 per cent of cases, but combining cytology with endoscopic biopsy may enable correct diagnosis in 40–70 per cent of cases of cholangiocarcinoma.78 Immunohistochemical markers such as CA19-9 and CA50 may distinguish cholangiocarcinoma from hepatocellular carcinoma, and staining with anti cytokeratin type-1 (monoclonal antibody AE1) could distinguish a biliary tract rather than hepatocyte origin.69 Histological diagnosis is not mandatory prior to surgical exploration. Positron emission tomography (PET) using the nucleotide tracer 18-F-fluorodeoxyglucose (FDG) can detect mass-like nodular cholangiocarcinoma lesions79 as well as the presence of distant metastases. This ability of PET to detect distant metastases may alter surgical management in up to 30 per cent of cases.80,81 Positron emission tomography is less useful for patients with infiltrating lesions and those with stents in situ or background primary sclerosing cholangitis.82 Endoscopic ultrasound (EUS) increasingly plays a role in the diagnosis of cholangiocarcinoma and can detect duct dilatation and lymphadenopathy, and also guide fineneedle aspiration for cytological diagnosis.83 Endoscopic ultrasound-guided fine-needle aspiration is reported to have a greater sensitivity than ERCP brushings for detecting malignancy. Other less frequently used modalities include intra-ductal ultrasound84 percutaneous flexible cholangioscopy85,86 and radiolabelled antibody87 and ligand imaging.88,89 Further evaluation of the clinical value of these modalities is required.
Staging
Figure 24.3 Hilar cholangiocarcinoma involving right and left hepatic ducts. A percutaneous transhepatic cholangiogram (PTC) demonstrating cannulation of the left hepatic duct and placement of an external drain. The right hepatic duct system had already been stented at ERCP.
both procedures is required (Fig. 24.3). These invasive techniques carry the risk of procedural complications (up to 9 per cent), including bleeding, biliary leakage, pancreatitis, cholangitis and duodenal perforation. Laparoscopy can detect small-volume peritoneal disease or liver secondaries missed on cross-sectional imaging, although it is less accurate for vascular invasion,
The Bismuth classification is a practical and commonly used surgical classification90 of hilar tumours based on the extent and location of ductal involvement (Fig. 24.4). However, this does not take into account radial tumour growth, vascular involvement and lobar atrophy and so is not truly indicative of tumour staging or resectability. Generally accepted criteria for unresectability of hilar cholangiocarcinomas include vascular involvement in terms of hepatic artery involvement, complete occlusion or encasement of the main portal vein proximal to the bifurcation, bilateral involvement of the hepatic arteries; distant spread with extra-hepatic invasion, nodal metastases beyond the hepatoduodenal ligament, or disseminated disease; extensive local disease with bilateral involvement of hepatic ducts to the level of the secondary biliary radicles, atrophy of one liver lobe with encasement of the contralateral portal vein branch, or atrophy of one liver lobe with contralateral
590 Biliary tract
Figure 24.4 The Bismuth Classification of biliary strictures. Type I tumours are entirely below the confluence of the right and left ducts; Type II tumours reach the confluence but do not involve either right or left duct; Type III tumours occlude the common hepatic duct and either the right (IIIa) or left (IIIb) hepatic duct; and Type IV indicates tumours that are multi-centric or involve both right and left hepatic ducts
secondary biliary radicle involvement.69 Formal pathological staging for cholangiocarcinoma is based on the TNM system (see Table 24.1), which is of limited clinical value in the preoperative situation. A modified version of preoperative T staging for tumours of the biliary confluence has been proposed by Jarnagin et al.91 which takes into account the degree of tumour extension to second-order biliary radicles, and ipsilateral or contralateral portal venous involvement as well as hepatic atrophy. In an analysis of 225 patients with cholangiocarcinoma using this system, resectability was 59 per cent for T1, decreasing to 0 per cent for T3.
Treatment The treatment options for cholangiocarcinoma are determined by the stage of the disease in terms of local extent, vascular involvement and presence or absence of metastases. SURGERY
Complete surgical resection with histologically negative (R0) resection margins offers the only potentially curative treatment for cholangiocarcinoma. Unfortunately, most patients (80 per cent) present with unresectable disease at diagnosis. The few patients considered suitable for surgery must be medically fit, with no metastases, and have local
disease that can be resected with clear margins.55,91,92 Patients with poor performance status, major cardiovascular disease or Child–Pugh–Turcotte stage B or C liver cirrhosis are not candidates for surgery. Sepsis, severe jaundice and malnutrition also predict poor outcomes but may be corrected to an extent prior to surgery. Preoperative biliary drainage has not been shown significantly to decrease morbidity or mortality. It increases the risk of cholangitis and increases hospital stay, but may be considered in patients with severe cholestasis who run the risk of liver failure. Selective ipsilateral portal vein embolization (PVE) can induce compensatory hypertrophy of the future liver remnant (FLR) and reduce the risk of hepatic dysfunction following liver resections in patients with a predicted FLR less than 25 per cent.93–95 Preoperative biliary drainage of the obstructed FLR prior to contralateral PVE may aid hypertrophy after PVE. Further controlled studies are required to determine the role of PVE and preoperative biliary drainage.76 The type and extent of surgery depend on the location of the cancer within the liver or the biliary tract. Extrahepatic cholangiocarcinomas of the common bile duct (Bismuth Type I and II) require excision of the extra-hepatic bile duct and regional hepatoduodenal lymphadenectomy with Roux-en-Y biliary reconstruction. Tumours involving the confluence (Bismuth Type III) require additional partial hepatectomy, and Type IV tumours need an
Tumours of the bile duct (cholangiocarcinoma) 591
extended right or left hepatectomy to achieve negative margins. Caudate lobe (segment 1) excision should be performed routinely for Type II, III and IV tumours as part of the hepatectomy, as these tumours often involve the ducts of this lobe. In one series, 98 per cent of resected specimens for hilar cholangiocarcinoma showed caudate lobe involvement.96 Distal cholangiocarcinomas should be managed with pancreato-duodenectomy (pylorus-preserving or standard Kausch–Whipple pancreatico-duodenectomy) just as for ampullary or pancreatic head cancers, with the aim once again being an R0 resection. The majority of Western series report a resectability rate of 15–35 per cent for cholangiocarcinoma,97–100 although adoption of a more aggressive policy in terms of en-bloc resection with vascular reconstruction leads to higher resectability rates of 35–66 per cent.98,101 The Japanese series report even higher resectability rates (52–92 per cent) than in the West.96,102–104 The adoption of an increasingly aggressive resection policy and achievement of higher R0-negative resection margin rates correlate with a significant trend towards prolonged survival, with 40–60 per cent 3-year survival rates in most series. However, the more extensive resections are associated with a higher mortality of around 10 per cent and significant postoperative morbidity up to 40 per cent93 Reported overall 5-year survival rates for hilar cholangiocarcinoma range from 11 per cent to 40 per cent.55,91,93 ADJUVANT THERAPY
Neoadjuvant chemotherapy has been shown to increase resectability in a small series, but is limited by the presence of obstructive jaundice and poor performance status in many patients. Adjuvant chemotherapy using 5-fluorouracil (5-FU) alone or in combination with agents such as methotrexate, cisplatin, mitomycin C, leucovorin and interferon alpha, and adjuvant radiotherapy in the form of external-beam irradiation with or without iridium-192 brachytherapy have not been convincingly shown to prolong survival, reduce recurrence or improve quality of life in marginnegative resections.56,105,106 There may possibly be a benefit for adjuvant radiotherapy alone or with chemotherapy for patients with residual disease.107 Most studies are small, retrospective, single-centre trials with poor-quality data. Adjuvant therapy cannot be routinely recommended outside the setting of clinical trials. LIVER TRANSPLANTATION
Earlier results of orthotopic liver transplantation (OLT) for cholangiocarcinoma were disappointing. Pichlmayr et al.108 found no difference in 1-year and 5-year survival for patients with proximal bile-duct cancer undergoing either surgical resection or OLT. Meyer et al. found a tumour recurrence rate of 51 per cent for patients undergoing OLT for hilar and intra-hepatic cholangiocarcinoma, with survival after recurrence rarely exceeding 1 year.109 Liver transplantation is currently contraindicated for unresectable hilar
cholangiocarcinoma because of the unacceptably high recurrence rate of up to 90 per cent within 2 years.109,110 However, a more recent study from the Mayo Clinic treated 28 highly selected patients who had unresectable disease and negative staging laparotomy with a new neoadjuvant regime combining external-beam irradiation, iridium-192 brachytherapy, 5-FU and oral capecitabine followed by liver transplantation. Overall 5-year survival was 82 per cent, which is comparable to current liver transplantation survival rates across the USA and better than present surgical resection survival rates.111 These data support a possible role for liver transplantation in cholangiocarcinoma and highlight the need for further trials to optimize chemoradiotherapy and patient selection.
Palliation ENDOSCOPIC AND RADIOLOGICAL BILIARY STENTING
Patients with unresectable cholangiocarcinomas have a median survival of 3 months without, and 6 months with, biliary drainage.112 Endoscopic or percutaneous palliative stenting for biliary drainage seeks to relieve jaundice and pruritus, prevent cholangitis, avoid liver failure and improve quality of life.113 Endoscopic stenting is suitable for distal cholangiocarcinomas, whereas percutaneous transhepatic insertion of stents may be preferable for hilar tumours.113 Both plastic and self-expanding mesh metal stents are available for biliary drainage. The choice of stent must be tailored to the individual patient. Overall, self-expanding metal stents offer the more cost-effective treatment for patients likely to survive longer than 3–6 months. Although more expensive, they have a larger diameter and longer patency rates (6–12 months) compared to plastic stents (3–6 months) 114 and are associated with shorter hospital stays, fewer repeat procedures and reduced duration of antibiotic treatment. Covered metal stents may last even longer than bare metal stents by preventing tumour ingrowth and occlusion.115 A study comparing unilateral versus bilateral stenting for hilar cholangiocarcinomas showed that bilateral stent insertion increased survival (176 versus 119 days) and reduced the risk of cholangitis.116 However, a prospective randomized controlled trial by DePalma et al.117 showed a higher success rate for unilateral stent insertion and drainage with reduced early complication rates, suggesting unilateral drainage may suffice for palliation in most cases.118 Tissue confirmation of cancer is essential prior to the insertion of metal stents, which tend to become embedded in the bileduct wall and can only be removed at operation.
Surgical palliation Surgical palliation with biliary-enteric bypass (usually segment 3 hepatico-jejunostomy) is now rarely done, following the introduction of percutaneous and endoscopic stenting.
592 Biliary tract
However, surgical bypass can still provide effective palliation and similar survival to stenting 119 in fit patients found to have unresectable disease on exploratory laparotomy. Morbidity is greater with surgery, but fewer repeat interventions are required as compared to stenting. Neurolytic coeliac plexus blockade can be considered for unremitting pain.55
Palliative chemotherapy and radiotherapy Response rates to palliative chemotherapy with 5-FU alone are poor. However, early studies using gemcitabine as a single agent show promise. Gemcitabine in combination therapy with agents such as cisplatin, docetaxel and irinotecan is associated with higher toxicities, without clear benefits in terms of treatment responses.120 Newer agents under evaluation include erlotinib, rebeccamycin analogue and capecitabine. Survival benefit from conventional radiotherapy for palliation is unproven, though some small series have suggested a possible role for external-beam radiotherapy or intraluminal brachytherapy. Currently the available data do not support the routine use of either, and clinical trials to evaluate their roles further are warranted.121,122
Photodynamic therapy Photodynamic therapy (PDT) involves the intravenous administration of a photo-sensitizer such as sodium porfimer, which localizes preferentially in tumour tissue over 24–48 hours. Subsequent endoscopic laser illumination of the tumour bed with a specific wavelength of light activates the porfimer, generating oxygen free radicals and causing cancer-cell death. Survival benefit for cholangiocarcinoma has been reported in some small series. One prospective randomized trial comparing stenting plus PDT versus stenting alone showed increased median survival (493 versus 98 days), improved stabilization of Karnofsky performance status, less cholestasis and better quality of life scores.123 This treatment modality could offer useful palliation, and further studies of PDT in combination with chemotherapy or radiotherapy or both are required. Other local ablation strategies such as radiofrequency ablation and ethanol injection are unproven. AUTHORS’ VIEW ON SURGERY
In summary, it is the authors’ policy to carry out aggressive surgical resection for cholangiocarcinomas whenever possible, as surgery offers the only potentially curative option. All patients undergo intensive investigation including CT, MRI, PET, EUS and laparoscopy as required for accurate staging and assessment of resectability, with preoperative biliary drainage and PVE when indicated. Surgery may include extended resection of the liver, biliary tree, portal
vein and lymph nodes to achieve negative resection margins and is warranted despite high morbidity and not insignificant mortality. For young and fit patients with unresectable lesions or those patients found at exploration to be unresectable, a biliary-enteric bypass is the treatment of choice. For most others, percutaneous transhepatic or endoscopic stent placement is preferred. The choice of plastic stents or self-expanding metal stents is based on estimated survival. Photodynamic therapy offers promise as a palliative treatment. Photodynamic therapy together with single-agent gemcitabine or combination chemotherapy and radiotherapy may be considered in selected patients, ideally in the context of clinical trials.
GALL-BLADDER CANCER Epidemiology Gall-bladder cancer is the most common biliary malignancy in Central and South America (Native American Indians, Hispanic), Central and Eastern Europe (Poland, Czech Republic, Slovakia), Japan and northern parts of India.124–132 Underlying genetic predisposition or dietary factors may explain this segregation of cases. American Indians, especially women in Southern Chile, have a very high incidence of gall-bladder cancer, reaching 35 per 100 000.130,133 On the other hand, in the whole of USA, only 2000 gall-bladder cancers are diagnosed per annum. In England and Wales, a total of 324 (101 men) patients’ mortality could be attributed to gall-bladder cancer per year.6
Aetiology Gallstones are perhaps the single most important risk factor for gall-bladder cancer, although the aetiologic link is unclear.134,135 Usually patients with large stones of long duration are at increased risk for cancer. All the other risk factors associated with gallstone disease are also associated with gall-bladder cancer, such as elevated body mass index, high caloric and carbohydrate intake, female sex, high parity and young age at first childbirth, oestrogen exposure.134–136 Calcification in the wall of gall bladder, such as porcelain gall bladder, also predisposes to gall-bladder cancer risk.137 Chronic Salmonella infection, which predisposes to gallstone formation, also predisposes to gall-bladder cancer risk, perhaps by the mechanism of chronic inflammation.138–140 Similarly, Helicobacter colonization has been suggested to predispose to gall-bladder cancer.141 Patients with anomalous pancreatico-biliary ductal union are also known to be at increased risk of gall-bladder cancer.142 Single, large (1 cm), sessile polyps are more likely to be malignant.143 Environmental and dietary factors may also predispose to the disease; however, the causal evidence is weak.144,145
Gall-bladder cancer 593
Molecular biology and genetics The most commonly involved genetic mutations involve the proto-oncogenes K-Ras and p53, and others, such as cell-cycle checkpoint genes, e.g. CDKN2 and MTS2, have also been involved.146–148 The molecular epidemiology of the nature of p53 gene mutations suggests that the aetiologic pathways for cancer in the Chilean and Japanese populations may be different.146,149 For example, it has been proposed that the anatomic abnormality of anomalous duct union may predispose to different Ras and p53 mutations (and these anatomic anomalies appear to be more common in the Japanese population).146,149–151 Allelic loss, particularly in chromosomes 16q, 3p, 8p, 9q and 22q, is common in gall-bladder cancer,152–154 and mitochondrial DNA mutations are relatively common and start occurring in chronic cholecystitis, giving credence to a chronic inflammation–cancer sequence.155
Anatomy and lymphatic drainage The lymphatic drainage follows the arterial supply and is commonly to the cystic lymph node, pancreatico-duodenal
lymph nodes (superior and posterior) and coeliac lymph nodes. Para-aortic and aorto-caval lymph nodes are also involved in advanced cases.
Metastasis and natural history The commonest mode of spread for gall-bladder cancer is contiguous spread to the liver, bile ducts, stomach, colon and duodenum, in that order. Distant metastasis occurs to lymph nodes, liver, peritoneum and lungs. Usually patients succumb to the locally advanced disease rather than to distant metastasis.
Pathology Most cancers are adenocarcinomas, but adenosquamous and squamous-cell carcinomas are also seen. Most tumours are of the flat, infiltrating variety, but polypoid and nodular tumours are also seen.156 Lymphatic spread depends on the tumour stage, with T1 tumours having no lymphatic spread, and T3 and T4 tumours having up to 75 per cent lymphatic spread.
Table 24.2 Tumour node metastasis (TNM) staging for gall-bladder cancer Primary tumour (T) Tx T0 Tis T1 T1a T1b T2 T3 T4
Primary tumour cannot be assessed No evidence of primary tumour Carcinoma-in-situ Tumour invades lamina propria or muscle layer: (a) invades lamina propria (b) invades muscle layer Invades perimuscular connective tissue, but no extension beyond serosa or into liver Breaches serosa (visceral peritoneum) and/or liver and/or one adjacent organ Invades portal vein or hepatic artery, or two or more extrahepatic organs
Lymph nodes (N) Nx N0 N1
Regional lymph nodes cannot be assessed No regional lymph-node metastasis Regional lymph-node metastasis
Distant metastasis (M) Mx M0 M1
Distant metastasis cannot be assessed No distant metastasis Distant metastasis
Stage grouping Stage IA Stage IB Stage IIA Stage IIB Stage III Stage IV
T1 T2 T3 T1, T2, T3 T4 Any T
N0 N0 N0 N1 Any N Any N
M0 M0 M0 MO M0 M1
594 Biliary tract
Symptoms and presentation The commonest presentation for gall-bladder cancers is pain, anorexia and weight loss and obstructive jaundice.157 A gall-bladder mass may be palpable in half the patients on presentation.157
Staging The TNM classification46 is the most commonly used staging tool for gall-bladder cancers (Table 24.2).
be helpful. Computed tomography and MR usually give adequate information about the local vascular anatomy, and hepatic angiography is rarely indicated. Endoscopic ultrasound is used only occasionally. Fine needle aspiration cytology (FNAC) is not recommended as routine because of the risk of peritoneal or needle-tract seeding in curable cases (although the instances are anecdotal), but is perfectly acceptable in irresectable cases being considered for palliative therapy. The differential diagnosis on imaging is essentially stone disease and its sequelae. Xanthogranulamatous cholecystitis, adenomyomatosis of the gall bladder and stone disease causing Mirizzi’s syndrome can be quite difficult to distinguish from gall-bladder carcinoma.
Investigations Treatment Apart from baseline blood investigations including liver function tests, serum levels of the tumour markers CEA and CA19-9 should be measured. They are elevated in 18 per cent and 30 per cent of patients respectively,158 and may be useful in monitoring progress or identifying recurrence after therapy. The first imaging investigation is ultrasound of the liver and biliary tract. Ultrasound detects thickening of the wall of the gall bladder (the differential diagnosis being inflammation), but polypoid lesions in the gall bladder may also be detected. Ultrasound also gives an estimate of the degree of the bile-duct and liver involvement and with duplex scanning, it may give an estimate of vascular involvement. Spiral CT scan probably gives the best estimate of the loco-regional disease (Fig. 24.5). However, MR gives equally high-resolution images, with the additional benefit of providing an MRCP.159 Nevertheless, most patients undergo an ERCP after ultrasound to relieve the obstructive jaundice, and therefore ERCP provides information on ductal involvement. In those cases in which it is difficult to cannulate the bile duct from below, PTC may
Figure 24.5 A computed tomography (CT) scan showing a gallbladder cancer. A stone is visible within the gallbladder. The tumour has breached the gallbladder wall and extends towards the structures at the liver hilus
SURGERY
There are three groups of patients warranting discussion: 1. Patients with the incidental finding of a gall-bladder polyp on imaging; 2. Patients in whom a hitherto unsuspected gall-bladder carcinoma is found at the time of cholecystectomy for gallstone disease, or on subsequent histological assessment of the gall bladder after cholecystectomy; 3. Symptomatic patients in whom gall-bladder carcinoma is suspected or diagnosed on imaging. Gall-bladder polyps are often discovered incidentally on imaging, and not all of them warrant a cholecystectomy. Polyps that are sessile, solitary and 1 cm or more in diameter should be deemed highly suspicious. Also, polyps that have developed in older patients (over the age of 50 years), developed in association with stones or are associated with symptoms are at a higher risk of being malignant or subsequently turning malignant.160–162 Laparoscopic cholecystectomy is the treatment of choice in these patients (unless the suspicion of malignancy is high, in which case open exploration and intra-operative frozen section are preferable, with preparation for extended resection if necessary). For polyps that are deemed low risk, the authors’ practice is to recommend 6-monthly ultrasonographic follow-up at first, which can stop after 2 years if they remain unchanged. If a surgeon suspects gall-bladder carcinoma at the time of a laparoscopic or open cholecystectomy for stone disease, it is reasonable to take tissue samples for histology and then terminate the operation, with the intention of subsequently referring the patient to a specialist centre. If expert hepato-biliary surgical help is at hand, it may be reasonable to proceed immediately to a major operation, with histological confirmation of malignancy on frozen section as the first step. But the surgeon will then have to very carefully evaluate the extent of disease (without the benefit of cross-sectional imaging) and determine if it is appropriate
Future trends 595
to proceed to a major resection, which the patient has not consented to or prepared for. If gall-bladder carcinoma is found on histological assessment after cholecystectomy, the depth of tumour invasion through the gall-bladder wall (T stage) and the involvement of the surgical margins should be carefully assessed, as these are the two major determinants of prognosis, although the presence of lymph-node metastasis and perineural invasion is also highly relevant. Gall-bladder perforation at the time of the cholecystectomy greatly increases the chances of recurrence.163 Simple cholecystectomy is adequate for mucosal or T1a lesions (where only the lamina propria has been invaded), and further surgery is not necessary. For T1b tumours (where the muscle layer has been invaded but not breached), there is evidence to suggest that a simple cholecystectomy is adequate, especially if the surgical margins are clear, there is no evidence of lymph-node spread and the gall bladder was not perforated at the time of surgery.164,165 For T2 tumours (which have invaded the perimuscular connective tissue but there is no extension beyond the serosa or into the liver) and tumours where the surgical resection margins are involved or the cystic lymph node is involved, a second radical operation should be performed. This should involve a hepatic resection in the form of an extended cholecystectomy (excision of the gall-bladder bed) or a formal excision of liver segments 4b and 5, with regional lymphadenectomy and excision of the extra-hepatic bile duct with biliary reconstruction. The laparoscopic port sites should also be excised.166–168 Gall-bladder cancers that have caused symptoms and then been identified on imaging are already quite advanced. For T3 and T4 lesions, the prognosis is poor and the majority of patients have unresectable disease. But radical resection offers the only chance of cure or prolongation of survival time, and a specialist surgical opinion should be sought in all patients who do not have distant metastases. A diagnostic laparoscopy prior to laparotomy may be useful in further staging the extent of disease. If the possibility of benign disease is being considered, a laparotomy and frozen-section confirmation of the nature of the disease should be the first step. Radical excision involves a liver resection involving at the very least segments 4b and 5, but a wide central resection of segments 4, 5 and 8 or, in some instances, even a trisegmentectomy may be necessary, depending on the extent of liver involvement. This is combined with regional lymphadenectomy (removal of nodes in the hepatoduodenal ligament, posterosuperior pancreato-duodenal nodes and nodes along the hepatic artery), and excision of the extra-hepatic bile duct with creation of a Roux-loop hepatico-jejunostomy. A formal pancreatoduodenectomy and/or portal vein resection may be necessary in some instances to achieve complete excision. If the para-aortic or retropancreatic nodes are involved, radical excision and regional lymphadenectomy provide no survival benefit, and frozen-section assessment of these nodes may be useful before commencing a radical resection.
To summarize, T3 and N1 disease should be seriously assessed with a view to resection; T4 disease may still be occasionally resectable; but para-aortic or retropancreatic node involvement or M1 disease should be considered unresectable.169–173 Cytoreductive surgery does not increase survival.156 Following surgical resection with clear margins, 5-year survivals of 62–100 per cent have been reported in patients with T2 lesions, and 20–50 per cent in patients with more advanced lesions. Adjuvant treatments have been tried, in the form of chemotherapy (5-FU with mitomycin C), chemotherapy (5-FU) combined with external-beam radiation, and intra-operative radiotherapy. They seem to confer modest benefit.174–176
RADIOTHERAPY
Gall-bladder cancer is generally considered relatively radioresistant. Patients with advanced disease do have some survival benefit from radiotherapy, but the studies are too few and too small to allow any effective judgement of the value of various types of radiotherapy, e.g. intraoperative or brachytherapy and conformal postoperative external-beam radiotherapy.177,178
CHEMOTHERAPY
Various drugs have been tried in single-agent or combination chemotherapy. These include gemcitabine, paclitaxel, 5-FU and mitomycin C (all given as single agents), 5-FU and folinic acid with hydroxyurea, and 5-FU with carboplatin. Thus far, the benefits have been found to be very modest (e.g. one study with 5-FU with folinic acid yielded 30 per cent partial response and 27 per cent disease stabilization, with median overall survival of 8 months). At present, the authors’ patients receive combination chemotherapy as for hilar cholangiocarcinomas, with a palliative intent.179–183
FUTURE TRENDS Cancers of the bile duct and gall bladder remain difficult tumours to diagnose and to treat effectively. Improvements in the outcome of surgical resection have caused this approach to be re-evaluated. The use of interventional radiology and endoscopy has considerably improved our ability to obtain successful biliary drainage in these patients. The addition to this of internal radiotherapy and PDT may produce valuable increases in survival over the next decade, and the value of chemotherapy has yet to be fully investigated. Although they still pose formidable surgical problems, much of the nihilistic thinking about these cancers has been replaced by a more aggressive and somewhat optimistic outlook.
596 Biliary tract
KEY LEARNING POINTS ●
●
●
Recent surgical results for cancers of the biliary tree and gall bladder provide some optimism for the possibility of cure, although the surgery involved is technically demanding and should be performed in specialist centres. Unresectable tumours can usually be effectively palliated by percutaneous or endoscopic stenting. Modern chemotherapy regimens may have an increasing role, and need to be tested in clinical trial settings. The value of radiotherapy remains unproven.
REFERENCES 1 Patel T. Worldwide trends in mortality from biliary tract malignancies. BMC Cancer 2002; 2:10. 2 Shaib Y, El Serag HB. The epidemiology of cholangiocarcinoma. Semin Liver Dis 2004; 24(2):115–25. 3 Patel T. Increasing incidence and mortality of primary intrahepatic cholangiocarcinoma in the United States. Hepatology 2001; 33(6):1353–7. 4 Taylor-Robinson SD, Toledano MB, Arora S, et al. Increase in mortality rates from intrahepatic cholangiocarcinoma in England and Wales 1968–1998. Gut 2001; 48(6):816–20. 5 Khan SA, Taylor-Robinson SD, Toledano MB, Beck A, Elliott P, Thomas HC. Changing international trends in mortality rates for liver, biliary and pancreatic tumours. J Hepatol 2002; 37(6):806–13. 6 Great Britain Office for National Statistics. Cancer Statistics Registration: Registrations of Cancer Diagnosed in 1991, England and Wales. London. HMSO 7 Watanapa P, Watanapa WB. Liver fluke-associated cholangiocarcinoma. Br J Surg 2002; 89(8):962–70. 8 Dayton MT, Longmire WP Jr, Tompkins RK. Caroli’s disease: a premalignant condition? Am J Surg 1983; 145(1):41–8. 9 Fozard JB, Wyatt JI, Hall RI. Epithelial dysplasia in Caroli’s disease. Gut 1989; 30(8):1150–3. 10 Voyles CR, Smadja C, Shands WC, Blumgart LH. Carcinoma in choledochal cysts. Age-related incidence. Arch Surg 1983; 118(8):986–8. 11 Ohta T, Nagakawa T, Ueno K, et al. Clinical experience of biliary tract carcinoma associated with anomalous union of the pancreaticobiliary ductal system. Jpn J Surg 1990; 20(1):36–43. 12 Sharp GB. The relationship between internally deposited alpha-particle radiation and subsite-specific liver cancer and liver cirrhosis: an analysis of published data. J Radiat Res (Tokyo) 2002; 43(4):371–80. 13 Walker NJ, Crockett PW, Nyska A, et al. Dose-additive carcinogenicity of a defined mixture of ‘dioxin-like compounds’. Environ Health Perspect 2005; 113(1):43–8.
14 Bond GG, McLaren EA, Sabel FL, Bodner KM, Lipps TE, Cook RR. Liver and biliary tract cancer among chemical workers. Am J Ind Med 1990; 18(1):19–24. 15 Haswell-Elkins MR, Mairiang E, Mairiang P, et al. Crosssectional study of Opisthorchis viverrini infection and cholangiocarcinoma in communities within a high-risk area in northeast Thailand. Int J Cancer 1994; 59(4):505–9. 16 Mitacek EJ, Brunnemann KD, Hoffmann D, et al. Volatile nitrosamines and tobacco-specific nitrosamines in the smoke of Thai cigarettes: a risk factor for lung cancer and a suspected risk factor for liver cancer in Thailand. Carcinogenesis 1999; 20(1):133–7. 17 Shin HR, Lee CU, Park HJ, et al. Hepatitis B and C virus, Clonorchis sinensis for the risk of liver cancer: a casecontrol study in Pusan, Korea. Int J Epidemiol 1996; 25(5):933–940. 18 Shaib YH, El Serag HB, Davila JA, Morgan R, McGlynn KA. Risk factors of intrahepatic cholangiocarcinoma in the United States: a case-control study. Gastroenterology 2005; 128(3):620–6. 19 Kobayashi M, Ikeda K, Saitoh S, et al. Incidence of primary cholangiocellular carcinoma of the liver in Japanese patients with hepatitis C virus-related cirrhosis. Cancer 2000; 88(11):2471–7. 20 Sorensen HT, Friis S, Olsen JH, et al. Risk of liver and other types of cancer in patients with cirrhosis: a nationwide cohort study in Denmark. Hepatology 1998; 28(4):921–5. 21 Roberts-Thomson IC, Strickland RG, Mackay IR. Bile duct carcinoma in chronic ulcerative colitis. Aust N Z J Med 1973; 3(3):264–7. 22 Kornfeld D, Ekbom A, Ihre T. Survival and risk of cholangiocarcinoma in patients with primary sclerosing cholangitis. A population-based study. Scand J Gastroenterol 1997; 32(10):1042–5. 23 Knechtle SJ, D’Alessandro AM, Harms BA, Pirsch JD, Belzer FO, Kalayoglu M. Relationships between sclerosing cholangitis, inflammatory bowel disease, and cancer in patients undergoing liver transplantation. Surgery 1995; 118(4):615–19. 24 Farrant JM, Hayllar KM, Wilkinson ML, et al. Natural history and prognostic variables in primary sclerosing cholangitis. Gastroenterology 1991; 100(6):1710–17. 25 Broome U, Olsson R, Loof L, et al. Natural history and prognostic factors in 305 Swedish patients with primary sclerosing cholangitis. Gut 1996; 38(4):610–15. 26 Bergquist A, Glaumann H, Persson B, Broome U. Risk factors and clinical presentation of hepatobiliary carcinoma in patients with primary sclerosing cholangitis: a case-control study. Hepatology 1998; 27(2):311–16. 27 Kubo S, Kinoshita H, Hirohashi K, Hamba H. Hepatolithiasis associated with cholangiocarcinoma. World J Surg 1995; 19(4):637–41. 28 Lee KT, Sheen PC. Lectin histochemical study of cholangiocarcinoma arising from stone-bearing intrahepatic bile duct. J Surg Oncol 1995; 59(2):131–5. 29 Terada T, Nakanuma Y, Ohta T, Nagakawa T. Histological features and interphase nucleolar organizer regions in
References 597
30
31
32
33
34 35
36
37
38
39
40
41
42
43
hyperplastic, dysplastic and neoplastic epithelium of intrahepatic bile ducts in hepatolithiasis. Histopathology 1992; 21(3):233–40. Ekbom A, Hsieh CC, Yuen J, et al. Risk of extrahepatic bile duct cancer after cholecystectomy. Lancet 1993; 342(8882):1262–5. Insiripong S, Thaisamakr S, Amatachaya C. Hemoglobin typing in cholangiocarcinoma. Southeast Asian J Trop Med Public Health 1997; 28(2):424–7. Burmeister BH, Turner SL. External beam radiation therapy as an agent in the aetiology of carcinoma of the bile duct: a report on two patients. Clin Oncol (R Coll Radiol) 1995; 7(1):48–9. Holzinger F, Z’graggen K, Buchler MW. Mechanisms of biliary carcinogenesis: a pathogenetic multi-stage cascade towards cholangiocarcinoma. Ann Oncol 1999; 10 Suppl. 4:122–6. Berthiaume EP, Wands J. The molecular pathogenesis of cholangiocarcinoma. Semin Liver Dis 2004; 24(2):127–37. Sturm PD, Baas IO, Clement MJ, et al. Alterations of the p53 tumor-suppressor gene and K-ras oncogene in perihilar cholangiocarcinomas from a high-incidence area. Int J Cancer 1998; 78(6):695–8. Isa T, Tomita S, Nakachi A, et al. Analysis of microsatellite instability, K-ras gene mutation and p53 protein overexpression in intrahepatic cholangiocarcinoma. Hepatogastroenterology 2002; 49(45):604–8. Wang Y, Yamaguchi Y, Watanabe H, Ohtsubo K, Wakabayashi T, Sawabu N. Usefulness of p53 gene mutations in the supernatant of bile for diagnosis of biliary tract carcinoma: comparison with K-ras mutation. J Gastroenterol 2002; 37M(10):831–9. Aishima SI, Taguchi KI, Sugimachi K, Shimada M, Sugimachi K, Tsuneyoshi M. c-erbB-2 and c-Met expression relates to cholangiocarcinogenesis and progression of intrahepatic cholangiocarcinoma. Histopathology 2002; 40(3):269–78. Harnois DM, Que FG, Celli A, LaRusso NF, Gores GJ. Bcl-2 is overexpressed and alters the threshold for apoptosis in a cholangiocarcinoma cell line. Hepatology 1997; 26(4):884–90. Okaro AC, Deery AR, Hutchins RR, Davidson BR. The expression of antiapoptotic proteins Bcl-2, Bcl-X(L), and Mcl-1 in benign, dysplastic, and malignant biliary epithelium. J Clin Pathol 2001; 54(12):927–32. Taniai M, Higuchi H, Burgart LJ, Gores GJ. p16INK4a promoter mutations are frequent in primary sclerosing cholangitis (PSC) and PSC-associated cholangiocarcinoma. Gastroenterology 2002; 123(4):1090–8. Satarug S, Lang MA, Yongvanit P, et al. Induction of cytochrome P450 2A6 expression in humans by the carcinogenic parasite infection, Opisthorchis viverrini. Cancer Epidemiol Biomarkers Prev 1996; 5(10):795–800. Prawan A, Kukongviriyapan V, Tassaneeyakul W, Pairojkul C, Bhudhisawasdi V. Association between genetic polymorphisms of CYP1A2, arylamine N-acetyltransferase 1 and 2 and susceptibility to cholangiocarcinoma. Eur J Cancer Prev 2005; 14(3):245–50.
●44
45
46 47
48 49
50
51 52
53
54
55 56
57
58 59
60
61
62
Klatskin G. Adenocarcinoma of the hepatic duct at its bifurcation within the porta hepatis. An unusual tumor with distinctive clinical and pathological features. Am J Med 1965; 38:241–56. Yamamoto M, Takasaki K, Yoshikawa T, Ueno K, Nakano M. Does gross appearance indicate prognosis in intrahepatic cholangiocarcinoma? J Surg Oncol 1998; 69(3):162–7. Sobin LH, Wittekind C. (eds). TNM Classification of Malignant Tumours, 5th edn. New York; Chichester: Wiley, 1997. Bhuiya MR, Nimura Y, Kamiya J, et al. Clinicopathologic studies on perineural invasion of bile duct carcinoma. Ann Surg 1992; 215(4):344–9. Weinbren K, Mutum SS. Pathological aspects of cholangiocarcinoma. J Pathol 1983; 139(2):217–38. Liver Cancer Study Group of Japan. The General Rules for the Clinical and Pathological Study of Primary Liver Cancer, 2nd edn. Tokyo: Kanehara, 2003. Yamasaki S. Intrahepatic cholangiocarcinoma: macroscopic type and stage classification. J Hepatobiliary Pancreat Surg 2003; 10(4):288–91. Lim JH, Park CK. Pathology of cholangiocarcinoma. Abdom Imaging 2004; 29(5):540–7. Gouma DJ, Mutum SS, Benjamin IS, Blumgart LH. Intrahepatic biliary papillomatosis. Br J Surg 1984; 71(1):72–74. Beazley RM, Hadjis N, Benjamin IS, Blumgart LH. Clinicopathological aspects of high bile duct cancer. Experience with resection and bypass surgical treatments. Ann Surg 1984; 199(6):623–636. Chu KM, Lai EC, Al Hadeedi S, Arcilla CE, Jr., Lo CM, Liu CL et al. Intrahepatic cholangiocarcinoma. World J Surg 1997; 21(3):301–305. Jarnagin WR, Shoup M. Surgical management of cholangiocarcinoma. Semin Liver Dis 2004; 24(2):189–199. Anderson CD, Pinson CW, Berlin J, Chari RS. Diagnosis and treatment of cholangiocarcinoma. Oncologist 2004; 9(1):43–57. van Leeuwen DJ, Reeders JW. Primary sclerosing cholangitis and cholangiocarcinoma as a diagnostic and therapeutic dilemma. Ann Oncol 1999; 10 Suppl 4:89–93. Lamerz R. Role of tumour markers, cytogenetics. Ann Oncol 1999; 10 Suppl 4:145–149. Albert MB, Steinberg WM, Henry JP. Elevated serum levels of tumor marker CA19-9 in acute cholangitis. Dig Dis Sci 1988; 33(10):1223–1225. Patel AH, Harnois DM, Klee GG, LaRusso NF, Gores GJ. The utility of CA 19-9 in the diagnoses of cholangiocarcinoma in patients without primary sclerosing cholangitis. Am J Gastroenterol 2000; 95(1):204–207. Ramage JK, Donaghy A, Farrant JM, Iorns R, Williams R. Serum tumor markers for the diagnosis of cholangiocarcinoma in primary sclerosing cholangitis. Gastroenterology 1995; 108(3):865–869. Nakeeb A, Lipsett PA, Lillemoe KD, Fox-Talbot MK, Coleman J, Cameron JL et al. Biliary carcinoembryonic antigen levels are a marker for cholangiocarcinoma. Am J Surg 1996; 171(1):147–152.
598 Biliary tract
63 Nehls O, Gregor M, Klump B. Serum and bile markers for cholangiocarcinoma. Semin Liver Dis 2004; 24(2):139–154. 64 Goydos JS, Brumfield AM, Frezza E, Booth A, Lotze MT, Carty SE. Marked elevation of serum interleukin-6 in patients with cholangiocarcinoma: validation of utility as a clinical marker. Ann Surg 1998; 227(3):398–404. 65 Chen CY, Lin XZ, Tsao HC, Shiesh SC. The value of biliary fibronectin for diagnosis of cholangiocarcinoma. Hepatogastroenterology 2003; 50(52):924–927. 66 Saini S. Imaging of the hepatobiliary tract. N Engl J Med 1997; 336(26):1889–1894. 67 Triller J, Losser C, Baer H. Hilar cholangiocarcinoma: radiological assessment of resectability. European Journal of Radiology 1994; 4:9–17. 68 Valls C, Guma A, Puig I, Sanchez A, Andia E, Serrano T et al. Intrahepatic peripheral cholangiocarcinoma: CT evaluation. Abdom Imaging 2000; 25(5):490–496. 69 Patel T. Cholangiocarcinoma. Nat Clin Pract Gastroenterol Hepatol 2006; 3(1):33–42. 70 Zhang Y, Uchida M, Abe T, Nishimura H, Hayabuchi N, Nakashima Y. Intrahepatic peripheral cholangiocarcinoma: comparison of dynamic CT and dynamic MRI. J Comput Assist Tomogr 1999; 23(5):670–677. 71 Khan SA, Davidson BR, Goldin R, Pereira SP, Rosenberg WM, Taylor-Robinson SD et al. Guidelines for the diagnosis and treatment of cholangiocarcinoma: consensus document. Gut 2002; 51 Suppl 6:VI1–VI9. 72 Manfredi R, Barbaro B, Masselli G, Vecchioli A, Marano P. Magnetic resonance imaging of cholangiocarcinoma. Semin Liver Dis 2004; 24(2):155–164. 73 Yeh TS, Jan YY, Tseng JH, Chiu CT, Chen TC, Hwang TL et al. Malignant perihilar biliary obstruction: magnetic resonance cholangiopancreatographic findings. Am J Gastroenterol 2000; 95(2):432–440. 74 Zidi SH, Prat F, Le Guen O, Rondeau Y, Pelletier G. Performance characteristics of magnetic resonance cholangiography in the staging of malignant hilar strictures. Gut 2000; 46(1):103–106. 75 Baron TH, Harewood GC, Rumalla A, Pochron NL, Stadheim LM, Gores GJ et al. A prospective comparison of digital image analysis and routine cytology for the identification of malignancy in biliary tract strictures. Clin Gastroenterol Hepatol 2004; 2(3):214–219. 76 Khan SA, Thomas HC, Davidson BR, Taylor-Robinson SD. Cholangiocarcinoma. Lancet 2005; 366(9493):1303–1314. 77 Weber SM, DeMatteo RP, Fong Y, Blumgart LH, Jarnagin WR. Staging laparoscopy in patients with extrahepatic biliary carcinoma. Analysis of 100 patients. Ann Surg 2002; 235(3):392–399. 78 Sugiyama M, Atomi Y, Wada N, Kuroda A, Muto T. Endoscopic transpapillary bile duct biopsy without sphincterotomy for diagnosing biliary strictures: a prospective comparative study with bile and brush cytology. Am J Gastroenterol 1996; 91(3):465–467. 79 Kim YJ, Yun M, Lee WJ, Kim KS, Lee JD. Usefulness of 18F-FDG PET in intrahepatic cholangiocarcinoma. Eur J Nucl Med Mol Imaging 2003; 30(11):1467–1472.
80 Uttaravichien T, Bhudhisawasdi V, Pairojkul C, Pugkhem A. Intrahepatic cholangiocarcinoma in Thailand. J Hepatobiliary Pancreat Surg 1999; 6(2):128–135. 81 Fritscher-Ravens A, Bohuslavizki KH, Broering DC, Jenicke L, Schafer H, Buchert R et al. FDG PET in the diagnosis of hilar cholangiocarcinoma. Nucl Med Commun 2001; 22(12):1277–1285. 82 Anderson CD, Rice MH, Pinson CW, Chapman WC, Chari RS, Delbeke D. Fluorodeoxyglucose PET imaging in the evaluation of gallbladder carcinoma and cholangiocarcinoma. J Gastrointest Surg 2004; 8(1):90–97. 83 Fritscher-Ravens A, Broering DC, Knoefel WT, Rogiers X, Swain P, Thonke F et al. EUS-guided fine-needle aspiration of suspected hilar cholangiocarcinoma in potentially operable patients with negative brush cytology. Am J Gastroenterol 2004; 99(1):45–51. 84 Farrell RJ, Agarwal B, Brandwein SL, Underhill J, Chuttani R, Pleskow DK. Intraductal US is a useful adjunct to ERCP for distinguishing malignant from benign biliary strictures. Gastrointest Endosc 2002; 56(5):681–687. 85 Nimura Y, Shionoya S, Hayakawa N, Kamiya J, Kondo S, Yasui A. Value of percutaneous transhepatic cholangioscopy (PTCS). Surg Endosc 1988; 2(4):213–219. 86 Nimura Y. Staging of biliary carcinoma: cholangiography and cholangioscopy. Endoscopy 1993; 25(1):76–80. 87 Jewkes AJ, Macdonald F, Downing R, Drolc Z, Allum WH. Labelled antibody imaging in pancreatic cancer, cholangiocarcinoma, chronic pancreatitis and sclerosing cholangitis. Eur J Surg Oncol 1991; 17(4):354–357. 88 Tan CK, Podila PV, Taylor JE, Nagorney DM, Wiseman GA, Gores GJ et al. Human cholangiocarcinomas express somatostatin receptors and respond to somatostatin with growth inhibition. Gastroenterology 1995; 108(6):1908–1916. 89 Murata T, Nagasaka T, Kamiya J, Nimura Y, Wakai K, Yoshida K et al. P53 labeling index in cholangioscopic biopsies is useful for determining spread of bile duct carcinomas. Gastrointest Endosc 2002; 56(5):688–695. 90 Bismuth H, Majno PE. Biliary strictures: classification based on the principles of surgical treatment. World J Surg 2001; 25(10):1241–1244. 91 Jarnagin WR, Fong Y, DeMatteo RP, Gonen M, Burke EC, Bodniewicz BJ et al. Staging, resectability, and outcome in 225 patients with hilar cholangiocarcinoma. Ann Surg 2001; 234(4):507–517. 92 Burke EC, Jarnagin WR, Hochwald SN, Pisters PW, Fong Y, Blumgart LH. Hilar Cholangiocarcinoma: patterns of spread, the importance of hepatic resection for curative operation, and a presurgical clinical staging system. Ann Surg 1998; 228(3):385–394. 93 Nimura Y, Kamiya J, Kondo S, Nagino M, Uesaka K, Oda K et al. Aggressive preoperative management and extended surgery for hilar cholangiocarcinoma: Nagoya experience. J Hepatobiliary Pancreat Surg 2000; 7(2):155–162. 94 Hemming AW, Reed AI, Howard RJ, Fujita S, Hochwald SN, Caridi JG et al. Preoperative portal vein embolization for extended hepatectomy. Ann Surg 2003; 237(5):686–691.
References 599
95 Abdalla EK, Barnett CC, Doherty D, Curley SA, Vauthey JN. Extended hepatectomy in patients with hepatobiliary malignancies with and without preoperative portal vein embolization. Arch Surg 2002; 137(6):675–680. 96 Nimura Y, Hayakawa N, Kamiya J, Kondo S, Shionoya S. Hepatic segmentectomy with caudate lobe resection for bile duct carcinoma of the hepatic hilus. World J Surg 1990; 14(4):535–543. 97 Blumgart LH, Hadjis NS, Benjamin IS, Beazley R. Surgical approaches to cholangiocarcinoma at confluence of hepatic ducts. Lancet 1984; 1(8368):66–70. 98 De Groen PC, Gores GJ, LaRusso NF, Gunderson LL, Nagorney DM. Biliary tract cancers. N Engl J Med 1999; 341(18):1368–1378. 99 Pinson CW, Rossi RL. Extended right hepatic lobectomy, left hepatic lobectomy, and skeletonization resection for proximal bile duct cancer. World J Surg 1988; 12(1):52–59. 100 Longmire WP, McArthur MS, Bastounis EA, Hiatt J. Carcinoma of the extrahepatic biliary tract. Ann Surg 1973; 178(3):333–345. 101 Launois B, Campion JP, Brissot P, Gosselin M. Carcinoma of the hepatic hilus. Surgical management and the case for resection. Ann Surg 1979; 190(2):151–157. 102 Tsuzuki T, Ogata Y, Iida S, Nakanishi I, Takenaka Y, Yoshii H. Carcinoma of the bifurcation of the hepatic ducts. Arch Surg 1983; 118(10):1147–1151. 103 Mizumoto R, Kawarada Y, Suzuki H. Surgical treatment of hilar carcinoma of the bile duct. Surg Gynecol Obstet 1986; 162(2):153–158. 104 Nagino M, Nimura Y, Kamiya J, Kanai M, Uesaka K, Hayakawa N et al. Segmental liver resections for hilar cholangiocarcinoma. Hepatogastroenterology 1998; 45(19):7–13. 105 Deziel DJ, Kiel KD, Kramer TS, Doolas A, Roseman DL. Intraoperative radiation therapy in biliary tract cancer. Am Surg 1988; 54(7):402–407. 106 Todoroki T, Ohara K, Kawamoto T, Koike N, Yoshida S, Kashiwagi H et al. Benefits of adjuvant radiotherapy after radical resection of locally advanced main hepatic duct carcinoma. Int J Radiat Oncol Biol Phys 2000; 46(3):581–587. 107 Nakeeb A, Tran KQ, Black MJ, Erickson BA, Ritch PS, Quebbeman EJ et al. Improved survival in resected biliary malignancies. Surgery 2002; 132(4):555–563. 108 Pichlmayr R, Weimann A, Klempnauer J, Oldhafer KJ, Maschek H, Tusch G et al. Surgical treatment in proximal bile duct cancer. A single-center experience. Ann Surg 1996; 224(5):628–638. 109 Meyer CG, Penn I, James L. Liver transplantation for cholangiocarcinoma: results in 207 patients. Transplantation 2000; 69(8):1633–1637. 110 Jeyarajah DR, Klintmalm GB. Is liver transplantation indicated for cholangiocarcinoma? J Hepatobiliary Pancreat Surg 1998; 5(1):48–51. 111 Heimbach JK, Gores GJ, Haddock MG, Alberts SR, Nyberg SL, Ishitani MB et al. Liver transplantation for unresectable perihilar cholangiocarcinoma. Semin Liver Dis 2004; 24(2):201–207.
112 Farley DR, Weaver AL, Nagorney DM. “Natural history” of unresected cholangiocarcinoma: patient outcome after noncurative intervention. Mayo Clin Proc 1995; 70(5):425–429. 113 Abu-Hamda EM, Baron TH. Endoscopic management of cholangiocarcinoma. Semin Liver Dis 2004; 24(2):165–175. 114 Kaassis M, Boyer J, Dumas R, Ponchon T, Coumaros D, Delcenserie R et al. Plastic or metal stents for malignant stricture of the common bile duct? Results of a randomized prospective study. Gastrointest Endosc 2003; 57(2):178–182. 115 Isayama H, Komatsu Y, Tsujino T, Sasahira N, Hirano K, Toda N et al. A prospective randomised study of “covered” versus “uncovered” diamond stents for the management of distal malignant biliary obstruction. Gut 2004; 53(5):729–734. 116 Chang WH, Kortan P, Haber GB. Outcome in patients with bifurcation tumors who undergo unilateral versus bilateral hepatic duct drainage. Gastrointest Endosc 1998; 47(5):354–362. 117 De Palma GD, Galloro G, Siciliano S, Iovino P, Catanzano C. Unilateral versus bilateral endoscopic hepatic duct drainage in patients with malignant hilar biliary obstruction: results of a prospective, randomized, and controlled study. Gastrointest Endosc 2001; 53(6):547–553. 118 De Palma GD, Pezzullo A, Rega M, Persico M, Patrone F, Mastantuono L et al. Unilateral placement of metallic stents for malignant hilar obstruction: a prospective study. Gastrointest Endosc 2003; 58(1):50–53. 119 Prat F, Chapat O, Ducot B, Ponchon T, Fritsch J, Choury AD et al. Predictive factors for survival of patients with inoperable malignant distal biliary strictures: a practical management guideline. Gut 1998; 42(1):76–80. 120 Daines WP, Rajagopalan V, Grossbard ML, Kozuch P. Gallbladder and biliary tract carcinoma: A comprehensive update, Part 2. Oncology (Williston Park) 2004; 18(8):1049–1059. 121 Lazaridis KN, Gores GJ. Cholangiocarcinoma. Gastroenterology 2005; 128(6):1655–1667. 122 Olnes MJ, Erlich R. A review and update on cholangiocarcinoma. Oncology 2004; 66(3):167–179. 123 Ortner ME, Caca K, Berr F, Liebetruth J, Mansmann U, Huster D et al. Successful photodynamic therapy for nonresectable cholangiocarcinoma: a randomized prospective study. Gastroenterology 2003; 125(5):1355–1363. 124 Black WC, Key CR, Carmany TB, Herman D. Carcinoma of the gallbladder in a population of Southwestern American Indians. Cancer 1977; 39(3):1267–1279. 125 Lazcano-Ponce EC, Miquel JF, Munoz N, Herrero R, Ferrecio C, Wistuba II et al. Epidemiology and molecular pathology of gallbladder cancer. CA Cancer J Clin 2001; 51(6):349–364. 126 O’Brien K, Cokkinides V, Jemal A, Cardinez CJ, Murray T, Samuels A et al. Cancer statistics for Hispanics, 2003. CA Cancer J Clin 2003; 53(4):208–226.
600 Biliary tract
127 Wagener DK, McDonald M. Increased gallbladder-related mortality among Hispanics: does education play a role? Ethn Health 1996; 1(3):197–205. 128 Boss LP, Lanier AP, Dohan PH, Bender TR. Cancers of the gallbladder and biliary tract in Alaskan natives: 1970–79. J Natl Cancer Inst 1982; 69(5):1005–1007. 129 Dhir V, Mohandas KM. Epidemiology of digestive tract cancers in India IV. Gall bladder and pancreas. Indian J Gastroenterol 1999; 18(1):24–28. 130 Miquel JF, Covarrubias C, Villaroel L, Mingrone G, Greco AV, Puglielli L et al. Genetic epidemiology of cholesterol cholelithiasis among Chilean Hispanics, Amerindians, and Maoris. Gastroenterology 1998; 115(4):937–946. 131 Strom BL, Soloway RD, Rios-Dalenz JL, Rodriguez-Martinez HA, West SL, Kinman JL et al. Risk factors for gallbladder cancer. An international collaborative case-control study. Cancer 1995; 76(10):1747–1756. 132 Jemal A, Tiwari RC, Murray T, Ghafoor A, Samuels A, Ward E et al. Cancer statistics, 2004. CA Cancer J Clin 2004; 54(1):8–29. 133 Nervi F, Duarte I, Gomez G, Rodriguez G, Del Pino G, Ferrerio O et al. Frequency of gallbladder cancer in Chile, a high-risk area. Int J Cancer 1988; 41(5):657–660. 134 Tazuma S, Kajiyama G. Carcinogenesis of malignant lesions of the gall bladder. The impact of chronic inflammation and gallstones. Langenbecks Arch Surg 2001; 386(3):224–229. 135 Diehl AK. Gallstone size and the risk of gallbladder cancer. JAMA 1983; 250(17):2323–2326. 136 Lambe M, Trichopoulos D, Hsieh CC, Ekbom A, Adami HO, Pavia M. Parity and cancers of the gall bladder and the extrahepatic bile ducts. Int J Cancer 1993; 54(6):941–944. 137 Stephen AE, Berger DL. Carcinoma in the porcelain gallbladder: a relationship revisited. Surgery 2001; 129(6):699–703. 138 Caygill CP, Hill MJ, Braddick M, Sharp JC. Cancer mortality in chronic typhoid and paratyphoid carriers. Lancet 1994; 343(8889):83–84. 139 Singh H, Pandey M, Shukla VK. Salmonella carrier state, chronic bacterial infection and gallbladder carcinogenesis. Eur J Cancer Prev 1996; 5(2):144. 140 Nath G, Singh H, Shukla VK. Chronic typhoid carriage and carcinoma of the gallbladder. Eur J Cancer Prev 1997; 6(6):557–559. 141 Leong RW, Sung JJ. Review article: Helicobacter species and hepatobiliary diseases. Aliment Pharmacol Ther 2002; 16(6):1037–1045. 142 Chao TC, Jan YY, Chen MF. Primary carcinoma of the gallbladder associated with anomalous pancreaticobiliary ductal junction. J Clin Gastroenterol 1995; 21(4):306–308. 143 Aldridge MC, Bismuth H. Gallbladder cancer: the polypcancer sequence. Br J Surg 1990; 77(4):363–364. 144 Darby SC, Whitley E, Howe GR, Hutchings SJ, Kusiak RA, Lubin JH et al. Radon and cancers other than lung cancer in underground miners: a collaborative analysis of 11 studies. J Natl Cancer Inst 1995; 87(5):378–384. 145 Serra I, Yamamoto M, Calvo A, Cavada G, Baez S, Endoh K et al. Association of chili pepper consumption, low
146
147
148
149
150
151
152
153
154
155
156 157
158
159
160
socioeconomic status and longstanding gallstones with gallbladder cancer in a Chilean population. Int J Cancer 2002; 102(4):407–411. Wistuba II, Albores-Saavedra J. Genetic abnormalities involved in the pathogenesis of gallbladder carcinoma. J Hepatobiliary Pancreat Surg 1999; 6(3):237–244. Yoshida S, Todoroki T, Ichikawa Y, Hanai S, Suzuki H, Hori M et al. Mutations of p16Ink4/CDKN2 and p15Ink4B/ MTS2 genes in biliary tract cancers. Cancer Res 1995; 55(13):2756–2760. Ajiki T, Fujimori T, Onoyama H, Yamamoto M, Kitazawa S, Maeda S et al. K-ras gene mutation in gall bladder carcinomas and dysplasia. Gut 1996; 38(3):426–429. Yokoyama N, Hitomi J, Watanabe H, Ajioka Y, Pruyas M, Serra I et al. Mutations of p53 in gallbladder carcinomas in high-incidence areas of Japan and Chile. Cancer Epidemiol Biomarkers Prev 1998; 7(4):297–301. Hanada K, Itoh M, Fujii K, Tsuchida A, Ooishi H, Kajiyama G. K-ras and p53 mutations in stage I gallbladder carcinoma with an anomalous junction of the pancreaticobiliary duct. Cancer 1996; 77(3):452–458. Iwase T, Nakazawa S, Yamao K, Yoshino J, Inui K, Yamachika H et al. Ras gene point mutations in gallbladder lesions associated with anomalous connection of pancreatobiliary ducts. Hepatogastroenterology 1997; 44(17):1457–1462. Wistuba II, Tang M, Maitra A, Alvarez H, Troncoso P, Pimentel F et al. Genome-wide allelotyping analysis reveals multiple sites of allelic loss in gallbladder carcinoma. Cancer Res 2001; 61(9):3795–3800. Matsuo K, Kuroki T, Kitaoka F, Tajima Y, Kanematsu T. Loss of heterozygosity of chromosome 16q in gallbladder carcinoma. J Surg Res 2002; 102(2):133–136. Wistuba II, Maitra A, Carrasco R, Tang M, Troncoso P, Minna JD et al. High resolution chromosome 3p, 8p, 9q and 22q allelotyping analysis in the pathogenesis of gallbladder carcinoma. Br J Cancer 2002; 87(4):432–440. Tang M, Baez S, Pruyas M, Diaz A, Calvo A, Riquelme E et al. Mitochondrial DNA mutation at the D310 (displacement loop) mononucleotide sequence in the pathogenesis of gallbladder carcinoma. Clin Cancer Res 2004; 10(3):1041–1046. Misra S, Chaturvedi A, Misra NC, Sharma ID. Carcinoma of the gallbladder. Lancet Oncol 2003; 4(3):167–176. Kapoor VK, Benjamin IS. Biliary malignancies. Bailliere’s Clinical Gastroenterology 11[4], 801–836. 1999. London, Balliere Tindall. Biliary tract. Noshiro H, Chijiiwa K, Yamaguchi K, Shimizu S, Sugitani A, Tanaka M. Factors affecting surgical outcome for gallbladder carcinoma. Hepatogastroenterology 2003; 50(52):939–944. Schwartz LH, Black J, Fong Y, Jarnagin W, Blumgart L, Gruen D et al. Gallbladder carcinoma: findings at MR imaging with MR cholangiopancreatography. J Comput Assist Tomogr 2002; 26(3):405–410. Chattopadhyay D, Lochan R, Balupuri S, Gopinath BR, Wynne KS. Outcome of gall bladder polypoidal lesions detected by
References 601
161 162
163
164
165
166
167
168
169
170
171
172
173
transabdominal ultrasound scanning: a nine year experience. World J Gastroenterol 2005; 11(14):2171–2173. Lee KF, Wong J, Li JC, Lai PB. Polypoid lesions of the gallbladder. Am J Surg 2004; 188(2):186–190. Sun XJ, Shi JS, Han Y, Wang JS, Ren H. Diagnosis and treatment of polypoid lesions of the gallbladder: report of 194 cases. Hepatobiliary Pancreat Dis Int 2004; 3(4):591–594. Z’graggen K, Birrer S, Maurer CA, Wehrli H, Klaiber C, Baer HU. Incidence of port site recurrence after laparoscopic cholecystectomy for preoperatively unsuspected gallbladder carcinoma. Surgery 1998; 124(5):831–838. Yamaguchi K, Chijiiwa K, Saiki S, Nishihara K, Takashima M, Kawakami K et al. Retrospective analysis of 70 operations for gallbladder carcinoma. Br J Surg 1997; 84(2):200–204. Wakai T, Shirai Y, Yokoyama N, Nagakura S, Watanabe H, Hatakeyama K. Early gallbladder carcinoma does not warrant radical resection. Br J Surg 2001; 88(5):675–678. Chijiiwa K, Nakano K, Ueda J, Noshiro H, Nagai E, Yamaguchi K et al. Surgical treatment of patients with T2 gallbladder carcinoma invading the subserosal layer. J Am Coll Surg 2001; 192(5):600–607. Fong Y, Heffernan N, Blumgart LH. Gallbladder carcinoma discovered during laparoscopic cholecystectomy: aggressive reresection is beneficial. Cancer 1998; 83(3):423–427. Wakai T, Shirai Y, Hatakeyama K. Radical second resection provides survival benefit for patients with T2 gallbladder carcinoma first discovered after laparoscopic cholecystectomy. World J Surg 2002; 26(7):867–871. Kondo S, Nimura Y, Hayakawa N, Kamiya J, Nagino M, Uesaka K. Extensive surgery for carcinoma of the gallbladder. Br J Surg 2002; 89(2):179–184. Kondo S, Nimura Y, Hayakawa N, Kamiya J, Nagino M, Uesaka K. Regional and para-aortic lymphadenectomy in radical surgery for advanced gallbladder carcinoma. Br J Surg 2000; 87(4):418–422. Bartlett DL, Fong Y, Fortner JG, Brennan MF, Blumgart LH. Long-term results after resection for gallbladder cancer. Implications for staging and management. Ann Surg 1996; 224(5):639–646. Todoroki T, Takahashi H, Koike N, Kawamoto T, Kondo T, Yoshida S et al. Outcomes of aggressive treatment of stage IV gallbladder cancer and predictors of survival. Hepatogastroenterology 1999; 46(28):2114–2121. Chijiiwa K, Noshiro H, Nakano K, Okido M, Sugitani A, Yamaguchi K et al. Role of surgery for gallbladder carcinoma with special reference to lymph node metastasis and stage
174
175
176
177
178
179
180
181
182
183
using western and Japanese classification systems. World J Surg 2000; 24(10):1271–1276. Takada T, Amano H, Yasuda H, Nimura Y, Matsushiro T, Kato H et al. Is postoperative adjuvant chemotherapy useful for gallbladder carcinoma? A phase III multicenter prospective randomized controlled trial in patients with resected pancreaticobiliary carcinoma. Cancer 2002; 95(8):1685–1695. Kresl JJ, Schild SE, Henning GT, Gunderson LL, Donohue J, Pitot H et al. Adjuvant external beam radiation therapy with concurrent chemotherapy in the management of gallbladder carcinoma. Int J Radiat Oncol Biol Phys 2002; 52(1):167–175. Todoroki T, Iwasaki Y, Orii K, Otsuka M, Ohara K, Kawamoto T et al. Resection combined with intraoperative radiation therapy (IORT) for stage IV (TNM) gallbladder carcinoma. World J Surg 1991; 15(3):357–366. Mehta A, Bahadur AK, Aranya RC, Jain AK. Role of radiation therapy in carcinoma of the gall bladder–a preliminary indian experience. Trop Gastroenterol 1996; 17(1):22–25. Houry S, Haccart V, Huguier M, Schlienger M. Gallbladder cancer: role of radiation therapy. Hepatogastroenterology 1999; 46(27):1578–1584. Gebbia V, Majello E, Testa A, Pezzella G, Giuseppe S, Giotta F et al. Treatment of advanced adenocarcinomas of the exocrine pancreas and the gallbladder with 5-fluorouracil, high dose levofolinic acid and oral hydroxyurea on a weekly schedule. Results of a multicenter study of the Southern Italy Oncology Group (G.O.I.M.). Cancer 1996; 78(6):1300–1307. Taal BG, Audisio RA, Bleiberg H, Blijham GH, Neijt JP, Veenhof CH et al. Phase II trial of mitomycin C (MMC) in advanced gallbladder and biliary tree carcinoma. An EORTC Gastrointestinal Tract Cancer Cooperative Group Study. Ann Oncol 1993; 4(7):607–609. Jones DV, Jr., Lozano R, Hoque A, Markowitz A, Patt YZ. Phase II study of paclitaxel therapy for unresectable biliary tree carcinomas. J Clin Oncol 1996; 14(8):2306–2310. Gallardo JO, Rubio B, Fodor M, Orlandi L, Yanez M, Gamargo C et al. A phase II study of gemcitabine in gallbladder carcinoma. Ann Oncol 2001; 12(10):1403–1406. Sanz-Altamira PM, Ferrante K, Jenkins RL, Lewis WD, Huberman MS, Stuart KE. A phase II trial of 5-fluorouracil, leucovorin, and carboplatin in patients with unresectable biliary tree carcinoma. Cancer 1998; 82(12):2321–2325.
25 Stomach MARIANNE C. NICOLSON, KENNETH G.M. PARK AND LESLIE M. SAMUEL
Incidence Aetiology Diagnosis Pathology Molecular biology
Staging Treatment Adjuvant chemotherapy Future directions References
602 603 603 604 604
INCIDENCE
605 606 611 612 612
dramatic geographical variations in the incidence of gastric cancer characterized by reducing rates in Westernized countries such as the UK (current incidence 8/100 000) while countries such as Japan, Brazil, Portugal and Lithuania continue to have a high incidence of the disease (50–100/100 000). The pattern of the disease also differs between countries; in the UK and USA (Fig. 25.1) tumours of the proximal stomach and gastro-oesophageal junction
Gastric cancer accounts for approximately 6 per cent of all cancer deaths in the UK, a percentage which comprises almost 10 000 people. Within Europe there are approximately 207 000 new cases each year and although the mortality from gastric cancer is reducing worldwide it remains the second commonest cause of cancer death. There are 130 120 110 100
80 70 60 50 40 30 20 10
Age at diagnosis White Black Asian or Pacific Islander
American Indian/Alaska Native Hispanic
85+
80–84
75–79
70–74
65–69
60–64
55–59
50–54
45–49
40–44
35–39
30–34
25–29
20–24
15–19
10–14
05–09
01–04
0 00
Rate per 100,000
90
Figure 25.1 Incidence of gastric cancer by age and racial origin in USA (Surveillance, Epidemiology, and End Results (SEER) Program)
Diagnosis 603
now outnumber antral cancers. This change in position of the tumours is not observed to the same extent in high incidence regions.
AETIOLOGY Geographical factors do not in themselves provide an adequate explanation of the rates of gastric cancer. In a study of immigrant Japanese, the incidence of gastric cancer did not fall in the first generation whereas in their offspring there was a decreased incidence in developing the disease. It is likely that there is an interaction between genetic and environmental factors in the development of most gastric cancers. Data from Scotland and the USA indicate a familial clustering of gastric cancer. Most familial gastric cancers are histologically diffuse tumours and a high proportion have mutations in the E-cadherin gene (CDH-1) resulting in a disruption of the normal E-cadherin-catenin complex required to maintain intercellular adhesion. In most patients this is a sporadic mutation but in a small proportion of cases there is an inherited germ-line inactivating mutation in the CDH-1 gene resulting in an autosomal dominant inheritance pattern. Although familial diffuse gastric cancers account for only a small proportion of gastric tumours it is important to determine the role of other germ-line mutations in patients with a weaker family history. In regions with a high incidence of gastric cancer it is the intestinal pattern of the tumour that predominates, frequently associated with a high incidence of Helicobacter pylori infection. Epidemiological studies indicate that H. pylori infection is associated with a 3–8 times increased incidence of gastric cancer compared with non-infected individuals. Furthermore a meta-analysis indicated that there was a close link between H. pylori seropositivity and gastric cancer. However even in high incidence regions of the world the majority of patients infected with H. pylori do not develop gastric cancer and the interaction between different strains of H. pylori and the individual may be more complex than was first envisaged. H. pylori strains are divided according to their ability to express CagA (cytotoxin-associated gene) and to secrete VacA (vacuolating cytotoxin) and this may determine their pathogenicity. Gastric atrophy and intestinal metaplasia, both regarded as pre-malignant conditions, are more likely in patients infected with CagA positive strains of H. pylori which incite intense inflammatory response. The role of VacA in gastric carcinogenesis appears to be more complex but strains of H. pylori expressing this protein do appear to be associated with severe gastritis. Individual responses to infecting agents may also vary in that patients with gastritis limited to the gastric antrum appear to produce excess acid and are at risk of duodenal ulceration whereas those with an extensive response in the body of the stomach develop gastric atrophy, achlorhydria, intestinal metaplasia and are at the highest risk of gastric cancer. Genetic heterogeneity in the cytokine response to infection may determine the pattern of disease observed in individuals.
Other infective agents have also been implicated in the carcinogenesis of gastric cancer including the Epstein–Barr virus (EBV). A study of EBV sequences detected by polymerase chain reaction revealed that 21 per cent of gastric adenocarcinomas in men from Los Angeles and 14 per cent in Hawaiian men were EBV associated. The figures were lower in the female population. EBV termini analysis indicated that the viral infection occurred early in transformation and was associated with an inflammatory component in the tumour. H. pylori eradication has been demonstrated to reverse many of the findings regarded as premalignant in the stomachs of infected individuals. There is, however, only indirect evidence that eradication may prevent gastric cancer. Other interventional studies have looked at increasing dietary antioxidants and COX2 inhibitors in at-risk populations but such studies have had inconclusive results to date. These were low intake of animal fat and protein, high intake of complex carbohydrates, nitrate or salt and a low intake of salads, fresh greens and fruit. When the pH of the stomach is higher than usual, bacteria can survive and may reduce dietary nitrate to nitrite to form N-nitroso compounds through nitrosation of dietary amines. N-nitroso compounds are known to be carcinogenic in animals. The source of dietary nitrites is preservatives and colouring agents, especially in home cured meats, dried fish and sausages. Nitrates are found in crop fertilizers and recycled sewage and they can be converted to nitrites by bacterial action in the food or in the stomach. The protective effect of greens may be explained by the possibility that ascorbic acid may increase gastric acidity and block the bacterial conversion of ingested nitrate to nitrite thereby decreasing the concentration of nitrosanes. This may also explain the role of diffuse H. pylori infections associated with achlohydria. Other conditions associated with reduced gastric acid production such as pernicious anaemia and previous gastric surgery have also been linked with a higher incidence of gastric cancer.
DIAGNOSIS In Britain 1 in 50 patients presenting to their general practitioner for the first time with dyspepsia will have gastric cancer. Up until 1980 only 1 per cent of cases in this country were diagnosed at an early stage – carcinoma confirmed within the submucosa. Conversely, the higher incidence of the disease in Japan has encouraged screening of the population, which resulted in an increase in the proportion of early diagnoses from 2 per cent in 1955 to 30 per cent in 1978. The Japanese mass screening was by indirect radiology. In Britain an improvement in the proportion of early tumours was seen when British dyspeptic patients over the age of 40 were referred for endoscopy following the first consultation with their general practitioner; 26 per cent of the gastric cancers found were early and 63 per cent of the cases were operable. The importance of detecting the disease at an early stage is emphasized in the 5-year survival figures for British
604 Stomach
patients presenting with advanced disease – 5 per cent1 compared with 97 per cent for the Japanese who are diagnosed and treated for early disease. Similar good results for the treatment of early gastric cancer have been reported from Western series in specialist centres. The importance of detecting gastric cancer at an early stage is undeniable but there is controversy on the merits of open access endoscopy in low incidence countries such as the UK. There is also debate regarding the routine endoscopy of patients with dyspeptic symptoms. It is clear that the detection rate of early gastric cancer in patients with uncomplicated dyspepsia under the age of 50 years is very low. This has led many to advocate screening for H. pylori in such patients and treating accordingly, with endoscopy reserved only for those whose symptoms fail to settle. Although this may be a cost-effective policy it ignores the potential impact of the occasional early tumour detected in a young patient which may not be cured with such a conservative policy.
PATHOLOGY Of stomach cancers, 97 per cent are adenocarcinoma. The other 3 per cent – sarcomas and lymphomas – will not be discussed in this chapter. Several different histology subclassifications of adenocarcinoma are recognized. The intestinal and diffuse varieties were originally described by Lauren and are widely accepted with good concordance in reports among pathologists – up to 80 per cent. In the infiltrative type of tumour neoplastic cells adhere to each other to form glandlike structures. It is more common in elderly men in populations who demonstrate an increased incidence of gastric cancer. The pre-cancerous period is longer with this histological type, which occurs mainly in the body and fundus along the greater curvature. Survival figures are slightly better for the intestinal than the ‘diffuse’ tumour. The diffuse variety of gastric cancer shows less cohesion and cause thickening of the stomach wall by infiltration. It is found mainly in areas where gastric cancer is endemic and is associated with an overrepresentation of blood group A. Lauren’s differentiations had earlier been described by Borman who described polypoid (I), ulcerating (II), combined (III) and infiltrating (IV) growth patterns and these are easily recognized on endoscopy. The Japanese added early gastric cancer, which is also an endoscopic diagnosis, made on the discovery of discrete single or multiple lesions combined to the mucosa. The site of most primary gastric cancers has changed in the recent past with an increase in the incidence of proximal lesions and a smaller number of distal and antral tumours. It is postulated that the increase in fruit and vegetable intake with consequent reduction in exposure of the distal stomach to carcinogenic nitrates may have led to the increase in the proportion of adenocarcinomas of the gastric cardia. The relevance of the change in anatomical site of stomach cancer is related to the associated poorer prognosis, possibly because disease at this proximal site
presents at a later stage or because surgery carries a higher morbidity due to the necessity for a thoracic approach. Although it is important to establish the histological type of gastric cancer, there has been poor correlation between histology and prognosis in this tumour and biological features are under investigation.
MOLECULAR BIOLOGY Biological characteristics of gastric carcinoma have been studied with the aim of identifying new prognostic indicators and possible future targets for treatment. The c-erb-2 oncogene is a member of the tyrosine kinase oncogene family. It codes for a 185 kD transmembrane glycoprotein and application has been reported in gastric carcinomas. In one large study of 189 gastric carcinomas 12.2 per cent of tumour specimens had evidence of c-erbB-2 overexpression localized to the cell membrane. These tumours were more advanced and more commonly associated with the serosal invasion, nodal involvement and peritoneal metastases, a finding confirmed by other authors. There was a significant difference in 5-year survival between the two groups (11 per cent for c-erbB-2positive vs. 50 per cent for c-erb-2-negative patients) indicating that this protein expression correlates with poor prognosis as it has been shown to do in breast cancer. Overexpression of the epidermal growth factor receptor (EGFR), which is incoded by the erb-1 oncogene, was found in 18 per cent of tumour samples in one study. There was an increased frequency in intestinal tumours (27 per cent) compared with diffuse (12 per cent). Unlike the c-erb-2 there was no association with overexpression in stage of disease in this study but others have reported a range of overexpression from 4 per cent (early disease) to 35 per cent (advanced disease). There are some data suggesting the EGFR expression is associated with a worse prognosis in gastric cancer which may be predicted from the data linking poor prognosis and presence of the antigen in patients with lung and head and neck cancers. The H-ras oncogene has been found to be overexpressed both in primary stomach cancers and also in areas of gastric dysplasia. Enhanced expression of C H-ras B21 in human stomach adenocarcinomas was defined by immunoassays using monoclonal antibodies and in situ hybridization. There was no correlation between expression and tumour depth or invasiveness. The 5-T4 antigen is the cell membrane antigen found in few normal tissues but in the wide variety of transform cell line. In a study incorporating 199 gastrointestinal cancer specimens (27 gastric and 72 colorectal) a positive correlation was found between 5-T4 and stage of disease but not cancer grade. The possibility is raised that with a monoclonal antibody raised to the 5-T4, immunotherapy could be effective in gastrointestinal cancer. Investigation and mutation of the p53 gene in gastric cancer has found involvement mainly of exons 5 and 7 (codons
Staging 605
173 and 251), which occur most frequently in advanced stage of disease. Exon 6 and 9 mutations of p53 have also been recognized, but the relationship to pathological staging and prognosis has not been fully elucidated. It has been postulated that there is an increased likelihood of lymph node metastases in tumours which overexpress P53 (85 per cent vs. 64 per cent). It has been estimated that overexpression of P53 may be present in up to 30 per cent of gastric cancers when immunohistochemical staining is used. In addition to investigating molecular biology to predict prognosis in gastric cancer, there has been investigation of the role of the multidrug-resistant gene (MDR-1) in the tumours resistance to chemotherapy. In a series of 22 patients MDR-1 was detected in 41 per cent with high levels in 18 per cent. Although the study was small the indication was clear that multidrug resistance may be a feature in the lack of responsiveness to chemotherapy in gastric cancer.
STAGING As in other tumours the stage of disease at diagnosis is relevant both to the plan of management and to the prognosis. Patients with early gastric cancer may be expected to be cured by appropriate treatment whereas patients with more advanced tumours will have a limited life expectancy and treatment must be aimed at providing good quality symptom relief. Although surgical resection remains the only proven method of providing long-term survival it is only likely to be effective if a complete macro- and microscopic resection is possible (R0). It is the ability of imaging techniques to determine whether a R0 resection is possible or not that is likely to have the largest impact on management; further refinement of pre-operative staging techniques to give more accurate anatomical staging of the disease is likely to be of less significance. For patients with early gastric cancers (limited to the mucosa and submucosa) the gross appearance, described by experienced endoscopists, may allow analysis of the extent of disease. Subdivisions are: I II III
Protruding type Superficial, A elevated, B flat, C depressed Excavated
Types IIB and IIC predominate and are associated with a 95 per cent 5-year survival rate, possibly related with the fact the lymphatic channels are very rarely found in the gastric mucosa and superficial submucosa. Computed tomography (CT) permits direct visualization of gastric lumen, wall and adjacent structures to provide a comprehensive picture of the disease including the detection of distant metastases. For localized tumours a comparison of preoperative CT scanning and operative staging in 75 patients was disappointing in that it demonstrated understaging in 31 per cent and overstaging in 16 per cent
of cases. The main failing seemed to be that CT scan was unable to detect metastases in normal-sized lymph nodes and failed to detect peritoneal metastases in 30 per cent of patients. Since this study was performed finer cross-sections are now available and water contrast has improved definition of the primary lesion. However CT scanning is limited in its ability to detect peritoneal disease. Endoscopic ultrasonography (EUS) has been found to have an overall accuracy in measuring the depth of the tumour penetration of around 92 per cent. However conventional EUS does not differentiate between mucosal and submucosal disease, which may be important in determining between surgery and endoscopic treatment; neither does it improve on CT in determining invasion of surrounding organs. Higher frequency ultrasound probes may be important in improving the value of this technique but it is still being actively evaluated. Laparoscopy of the peritoneal cavity may be extremely useful in the detection of peritoneal disease and allows for direct assessment of fixation to adjacent organs. When combined with ultrasound evaluation of lymph nodes, laparoscopic ultrasound identified pathology that mitigated against resection in approximately 20 per cent of cases. Positron emission tomography (PET) provides a metabolic image of tumours by determining the uptake of radiolabelled substrates into the tumour tissue. Most studies have used the obligate glucose metabolism of gastric tumours and the glucose analogue fluorodeoxyglucose (FDG) to visualize upper gastrointestinal tumours. Initial results with FDG-PET imaging have demonstrated the ability to detect metastatic disease but had a limited role in refining local staging. Further evaluation of combining anatomical images with CT and metabolic PET images is likely to provide better accuracy. In more advanced cases the preferred staging in the UK is the TNM system (Table 25.1). For operative purposes the stomach is divided anatomically into three sections – upper third (C), middle third (M), and lower third (A). The extent of stomach wall involvement is documented as SO (no serosal involvement), SI (suspected serosal involvement), SII (definite serosal involvement) or SIII (invasion of contiguous structures). Lymph nodes are categorized as NI (perigastric nodes along the lesser and greater curvature), NII (paragastric nodes along the coeliac access and its trifurcation) or NIII (involvement of other intraabdominal lymph nodes), which is now considered to signify distant metastases. The curability of the tumour is directly related to its operability and the exact nodal groups that must be removed to attempt cure depend on the location of the primary tumour. Following resection, which may involve total or subtotal gastrectomy, accurate histological evaluation of tumour subtype involvement of tumour margins complete the staging assessment. It is only when these procedures have been completed that a statement can be made on the curative or non-curative role of surgery in each particular case. Curative resection is possible when there are neither peritoneal nor hepatic deposits
606 Stomach
and a tier of non-involved lymph nodes has been removed beyond those containing secondary tumour. Relating the staging to prognosis, serosal involvement predicts a 5-year survival of approximately 15 per cent compared with 50 per cent when the serosa is free of tumour. Nodal involvement reduces the 5-year survival from 31 per cent to 17 per cent. The roles of surgery, radiotherapy and chemo-therapy in stages I–IV will be discussed later. The World Health Organization (WHO) classify pancreatic exocrine tumours according to the categories given in Table 25.2.
Table 25.1 TNM/AJCC staging of gastric cancer Primary tumour (T) TX Primary tumour cannot be assessed T0 No evidence of primary tumour Tis In situ carcinoma (including PanIN III) T1 Tumour limited to gastric, 2 cm or less in greatest dimension T2 Tumour limited to gastric, 2 cm in greatest dimension T3 Tumour extends beyond the gastric but without involvement of the celiac axis or the superior mesenteric artery T4 Tumour involves celiac axis, or the superior mesenteric artery Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastases N1 Regional lymph node metastases Metastases (M) MX Distant mesastases cannot be assessed M0 No distant metastases M1 Distant metastases Stage grouping Stage 0 Stage IA Stage IB Stage IIA Stage IIB Stage III Stage IV
Tis N0 M0 T1 N0 M0 T2 N0 M0 T3 N0 M0 T1-3 N1 M0 T4 N0 M0 AnyT AnyN M1
Table 25.2 WHO classification of gastric tumours • • • • •
Papillary adenocarcinoma Tubular adenocarcinoma Mucinous adenocarcinoma Signet ring adenocarcinoma Adenosquamous carcinoma
TREATMENT Surgery In patients with tumours limited to the gastric mucosa, and perhaps superficial submucosa, endoscopic mucosal resection (EMR) has been demonstrated to be effective treatment. Clearly such treatment avoids both the immediate and long-term morbidity of surgery. It is important to select patients carefully as there is an increasing incidence of lymph node involvement when tumours invade into the submucosa and beyond. Most authorities would limit EMR to endoscopic type I and II lesions (see earlier) and in whom there was pathological confirmation of tumour clearance with no invasion beyond the superficial submucosa. The cornerstone of treatment for gastric cancer remains surgical resection but meaningful long-term survival is only achieved in patients in whom it is possible to achieve complete macroscopic and microscopic clearance of their tumour (R0 resection). The poor prognosis of patients with residual microscopic (R1) or macroscopic tumour (R2) after resection is highlighted by many authors. As a result there is a trend towards a more conservative approach to offering patients surgery. A review of papers dating from 1938 to 1982 incorporating more than 7000 patients concluded that 80 per cent of patients underwent surgery. Whereas Akoh and McIntyre in a review of 100 English language papers on the treatment of gastric reported that the number of patients operated fell from 92 per cent before 1970 to 71 per cent by 1990 but the proportion of operated patients undergoing resection increased from 37 per cent to 48 per cent in the same period, suggesting an improvement in patient selection. Better staging and selection procedures perhaps in combination with improved operative techniques resulted in a larger proportion of operated patients being thought suitable for curative or radical resection (9 per cent before 1970 vs. 31 per cent before 1990). Historically there have been several problems associated with the surgery for gastric cancer in the Western hemisphere. Accurate surgical staging was often not carried out, resections were not standardized and there was frequently a lack of pathological data in the reporting of the surgical specimens. Perioperative mortality was frequently excessive and even after surgical procedures that were thought to be curative the local regional relapse rate was around 85 per cent. This is in contrast to series from Japan and East Asia, which have consistently reported operative mortality rates below 5 per cent and disease-free survival rates above 60 per cent at 5 years. Although there are potentially many reasons for these differences it is clear that a structured approach to surgery, as laid down by the Japanese Society for Gastric Cancer (JRSGC), does lead to a more uniform approach. Perhaps the area of greatest controversy in the surgical management of patients with potentially curative gastric cancer is the extent of surgical resection required to achieve this goal. Most authorities would now agree that total gastrectomy is not always required if sufficient tumour clearance is
Treatment 607
possible without removal of the whole stomach. Simlarly routine removal of adjacent organs is only indicated where there is direct tumour extension into these, and even then the increased surgical morbidity must be balanced against the limited benefits in patients with such extensive disease. The upper GI surgical community is however divided in the attitude towards the extent of lymphadenectomy required in curative resections of the stomach. Meticulous mapping of lymph node involvement for tumours in different anatomical locations within the stomach led to the concept of systemic lymph node dissection in Japan. In non-randomized series both in Japan and Europe it has been shown that extended lymphadenectomy (D2 resection) improves disease-free survival in selected patients compared with historical controls. Two large European randomized, controlled trials failed to show any benefit to extended lymph node dissection compared with limited dissection (D1) and indeed both demonstrated an excess morbidity and mortality with the more extensive dissection. Both trials have been heavily criticized because much of the morbidity in the D2 arms of the trials was due to the routine removal of the spleen and pancreatic tail. As discussed earlier the routine removal of adjacent organs is no longer recommended for tumour clearance. Furthermore in both trials surgeons were being asked to perform procedures with which they were unfamiliar and this may have been a cause for the increased morbidity in the D2 arms. This calls into question the whole philosophy of randomization in surgical trials and whether it is possible now to perform a randomized trial of D2 vs. D1 gastrectomy. The protagonists for D2 dissections argue that they can achieve this with low morbidity and mortality and that this will give more accurate staging of the tumours. They
would also argue that to perform a D1 lymphadenectomy would be a procedure with which they are unfamiliar and may not be as safe. The converse argument is that those familiar with the D1 dissection may not be comfortable with the D2 approach. The lack of standardization in the surgical approach does pose problems when considering the results of neoadjuvant and adjuvant trials (Figs 25.2 and 25.3). There is further controversy regarding laparoscopy and intra-operative lymph node mapping. If lessons are to be learned from the controversy of lymphadenectomy in gastric cancer it is clear that these techniques should be subjected to rigorous scientific scrutiny from the outset. In the surgical management of locally advanced disease where by definition there are involved regional lymph nodes,
% Survival
100
No CT CT
75 50 25 0 0
No. at risk No CT CT
12
24
36
48
60
29 49
16 33
Months 338 348
184 230
88 125
44 75
Figure 25.3 Pooled survival data from randomised trials of adjuvant chemotherapy in gastric cancer. (Adapted from Stocken et al. Br J Cancer 2005, 92(8))
289 patients with histologically proven adenocarcinoma of the gastric who had undergone potentially curative resection
69 assigned to observation
73 assigned to chemoradiotherapy
75 assigned to chemotherapy
72 assigned to chemoradiotherapy and chemotherapy
Treatment comparison
No chemoradiotherapy vs. chemoradiotherapy (144 vs. 145)
No chemotherapy vs. chemotherapy (142 vs. 147)
Figure 25.2 Design of the ESPAC 1 trial (Copyright: Neoptolemos JP, et al., NEJM, March 18 2004, 350:12)
608 Stomach
adjacent tissues or organs which preclude resection en bloc. The median survival is in the region of 5 months with all patients dead by two years. The poor figures belie the symptomatic benefit enjoyed by many patients who obtain relief from obstruction bleeding or pain associated with their tumour preoperatively (Fig. 25.4). In fitter patients downstaging with chemotherapy may be appropriate as discussed later in the chapter. For those unsuitable for chemotherapy, resection should be attempted wherever possible because this confers both a greater palliative relief from gastrointestinal symptoms and an improved survival rate. Total gastrectomy is not recommended as a palliative procedure because of the significant morbidity and high mortality from the procedure in this setting. For tumours of the gastro-oesophageal junction however oesophagogastrectomy is the treatment of choice for fit patients. Although palliative gastrostomy and jejunostomy are practised in some centres the poor response and high complication rates make these unattractive options in the majority of cases. The high morbidity and mortality associated with endoscopic insertion of a prosthetic stent make it a procedure suitable only for those with very poor prognosis who are unfit for surgery but who require palliation of salivary aspiration. Endoscopic light amplification by stimulative emission of radiation (LASER) may provide a means of haemostasis and relief of obstruction, particularly in polypoidal tumours. The laser light, converted into heat, coagulates bleeding points or vaporizes neoplastic tissue. Bleeding lesions in the fundus and cardia are very difficult to treat but effective palliation can be achieved in 80 per cent of patients. Laser palliation of cardiac stenosis was successful
Analgesic consumption
in 65 per cent of 11 patients but less encouraging in distal (antral lesions) because of associated poor motility resulting in prevention of gastric emptying despite formation of an adequate lumen. The conclusion on surgical management of gastric cancer is that only complete resection can currently offer a cure for the disease and the trend towards more careful selection of patients suitable for curative resection must be welcomed. Meticulous documentation of the extent of disease by surgeons and pathologists will aid in determining what constitutes a curative procedure and specialization of surgeons will also help to reduce operative mortality. As always, entering patients into trials for gastric cancer will expedite the accumulation of knowledge of the optimum treatment for different disease stages. In the palliative setting development of non-invasive techniques and improvement of those which are currently available will ensure maximum benefit with minimum morbidity for people with unresectable tumours. In addition the link between surgeons and oncologists should be encouraged, enabling all groups of patients with gastric cancer to have the opportunity of participation in an appropriate clinical trial.
Radiotherapy With local recurrence rates of 30–70 per cent after potentially curative surgery, radiation therapy has been used in an attempt to reduce locoregional recurrence of gastric cancer. As the location and stage of the primary gastric cancer1–4 and the extent of the surgical resection influence
Pain intensity
Performance status
Pain
Responder Improvement in both parameters Stable in one parameter, improvement in the other parameter
Stable
Nonresponder
In both parameters
Worsening in either parameter
Weight
Responder
Nonresponder
7% increase in body weight
Stable or decreased weight
Figure 25.4 Definition of Clinical Benefit Response
Treatment 609
the incidence and pattern of locoregional recurrence, the size and radiation field arrangement portals also vary. These potential confounding factors as well as differences in dose and fraction size together with the frequent the use of concurrent chemotherapy have to be remembered when comparing the results from studies. Radiation therapy has been used in the pre-, intra- and postoperative setting, often combined with chemotherapy as a radiosensitizer. Zhang et al5 randomized 370 patients with cancers of the gastric cardia to preoperative radiation therapy (40 Gy in 20 fractions) or surgery alone. They reported improved local control (61 per cent vs. 48 per cent) and survival (20 per cent vs. 13 per cent) for the group receiving preoperative radiation therapy. However the group undergoing surgery alone had significantly more lymph node involvement. Small phase I and II studies adding chemotherapy to preoperative radiotherapy have shown R0 resection rates of about 70 per cent and pathological complete response (pCR) rates of up to 30 per cent.6–8 Neoadjuvant combinations of chemotherapy and radiotherapy have been investigated. Ajani et al.6 used a combination of neoadjuvant chemotherapy with paclitaxel, cisplatin and 5-fluorouracil (5-FU) for two cycles followed by radiation therapy (45 Gy in 25 fractions) with concurrent infusional 5-FU and weekly paclitaxel over 5 weeks. The results from 42 patients were reported, with 78 per cent having a R0 resection and 20 per cent a pCR. The degree of pathological response as well as having a R0 resection correlated with overall survival. This strategy is being explored by large cooperative groups on both sides of the Atlantic. Close attention to perioperative care and surgical technique as well as radiotherapy planning is important given the relatively poor radiation tolerance of the organs within the upper abdomen. Intraoperative radiotherapy (IORT) has an advantage over external beam radiation (EBRT) in that the surgeon can move the radiosensitive organs out of the radiotherapy portal, facilitating the delivery of a higher local dose. However as IORT cannot be fractionated, and practical positioning of the cone for electron therapy in theatre can be difficult, there will inevitably be scatter of dose onto radiosensitive tissue. Japanese investigators, using doses of 28–35 Gy in a single fraction, have reported an improved local control and a survival benefit of about 20 per cent in stage II and III patients in non-randomized studies.9,10 A phase III trial was started in the USA but stopped before sufficient accrual of patients was reached, and reported a small benefit in terms of local control with no impact on survival.11 A French study used both IORT (doses about 15 Gy) and postoperative EBRT, but limited lymph node dissection at surgery.12 After 61 months of follow-up, local failures were seen in only 25 per cent of the patients, with no survival benefit apparent. Owing to the requirement for special equipment, and the prolonged operation time, IORT has not become standard therapy despite enthusiastic advocates. Postoperative radiation therapy, combined with concurrent bolus 5-FU-based chemotherapy, has a long history with early results showing a survival advantage for the combined
radiation and chemotherapy group.13 Following publication of the Intergroup 0116 study in 2001,14 it is now the standard of care in the USA and in some other parts of the world. A total of 556 patients were randomized to either surgery alone or postoperative EBRT with concurrent bolus 5-FU-based therapy. There was a survival advantage in the group receiving postoperative therapy, with a 3-year overall survival rate of 50 per cent vs. 41 per cent for the group receiving surgery alone. In the study, 54 per cent of patients had less than a D1 dissection, and less than 10 per cent had a D2 dissection, indicating that extensive lymph node dissection was not routine practice in the USA at the time of the study.15 In present practice in Western countries a dissection between D1 and D2, a D1 operation is regarded as standard practice.16 A substantial number (35 per cent) of the patients had to have alteration in position of their radiation portals due to protocol deviations. Although most of these were corrected prior to the start of therapy there remained major deviations from the protocol in 6.5 per cent of treatment plans in a posttreatment audit.14 Owing to these concerns advice was published on field arrangements.17 In INT0116 only 64 per cent of patients completed the treatment as planned, with 17 per cent stopping due to toxic side effects. Bolus 5-FU regimens are associated with higher rates of haematological toxicity and this was seen in the Intergroup Study (54 per cent grade 3/4). Rich et al.18 have summarized the basic science behind the move to both infusional and oral 5-FU regimes with concurrent radiation therapy and the outcome in a number of tumour sites, including gastric cancer. A thorough knowledge of the anatomy of the upper abdomen and patterns of spread of gastric cancer is required when planning the radiation therapy.17 A typical target volume includes the gastric bed, anastomosis, stump, and regional node areas, including perigastric nodes and nodes along the left gastric common hepatic, splenic, and coeliac arteries. Patients are usually treated in the supine position, using 6–15 MV photons, and either a 4-field or 3-field technique. Anterior-posterior/posterior-anterior (APPA) portals typically extend from the hemidiaphragm to the L3 body. They take in the external border of the postoperative gastric remnant to include the whole gastric area, duodenus, and porta hepatis (identified from preoperative imaging and surgical clips). Relevant parts of the liver and both kidneys, identified on simulation films after IV contrast, are shielded using a multileaf collimator. Lateral fields cover from the most anterior extent of the preoperative gastric wall to the anterior level of the spinal cord, with shielding for the portion of both kidneys outside the target volume. There is usually less weighting on the lateral fields compared to the APPA fields, to further reduce unwanted dose to the kidneys and liver. As has been the case in other tumour sites the technique of intensity modulated radiation therapy (IMRT) may offer the opportunity to optimize dose to the target areas whilst reducing further unwanted dose to critical organs such as kidneys and liver. With the remaining risks of locoregional recurrence after optimal gastric cancer surgery and technical improvements
610 Stomach
in the planning and delivery of radiation therapy, such as IMRT, it is likely that radiation therapy will continue to be incorporated in clinical trials looking at both preoperative ‘neoadjuvant’ regimens as well as postoperative adjuvant.
Chemotherapy In the Western world, most patients with gastric cancer are diagnosed with advanced disease. Stomach cancer is a chemosensitive tumour with single-agent drug activity achieving objective response in approximately 36 per cent of patients. Combination chemotherapy is more effective than single-agent treatment, although there is, as expected, increased toxicity with two or more drugs. There is debate regarding the ‘optimal’ treatment. Most regimens are based on 5-FU with anthracycline (doxorubicin or epirubicin). CHEMOTHERAPY IN ADVANCED DISEASE
The ideal regimen in patients with advanced disease is that with maximal efficacy and minimal toxicity, since quality of life as well as survival duration have to be balanced. A recent systematic review and meta-analysis concluded that ECF (epirubicin, cisplatin and continuous infusion 5-FU) has become the standard chemotherapy regimen.19 This
regimen was first developed in the late 1980s, reporting response rates in 128 patients of up to 70 per cent (CR 15 per cent) in advanced disease with median survival 8 months. ECF was developed to exploit both the increase in activity of 5-FU seen in infusional rather than bolus delivery and the synergy of cisplatin and 5-FU. The 5-FU cisplatin synergy is believed to be mediated through the depletion of intracellular methiamine by cisplatin. This in turn leads to an increase in the reduced metabolite concentration enhancing the binding of fluorodeoxidine monophosphate to thimidylate synthase. There may also be a reduction in cisplatin damage DNA repair due to interruption of cellular thimadine utilization by 5-FU. The response rate with ECF reduced as larger numbers of patients were treated, but when compared with one of the previous standards FAMTX (5-FU, doxorubicin and methotrexate), it demonstrated superior response rate (45 per cent vs. 21 per cent), median survival (8.9 vs. 5.7 months) and 1-year survival (36 per cent vs. 21 per cent along with a 4-month improvement in failure-free survival duration.20 Mitomycin-C has single agent activity in gastric cancer and was combined with cisplatin and infusional 5-FU (MCF) for a phase III randomized trial compared with ECF. There was no significant difference in response or median survival but quality of life was better in the ECF arm.21
Table 25.3 Randomized trials in locally advanced unresectable gastric cancer Reference
Gy
Randomized (evaluable)
Median survival (months)
1 year survival (estimate)
a. Chemotherapy vs. radiotherapy alone Moertel et al. 196913 35–40 35–40 GITSG 9273, 1981 60 40 60
5-FU Placebo 5-FU 5-FU –
NR (32) NR (32) 111 (86) 117 (83) 25 (25)
10.4 6.3 (mean) 11.4 8.4 5.3
NR (25) NR (6) 44 39 14
b. Chemoradiotherapy vs. chemotherapy alone ECOG, 1985 40 – GITSG 9283, 1988 54 –
5-FU 5-FU 5-FU SMF SMF
NR (47) NR (44) 24 (22) 24 (21)
8.3 8.2 9.7 7.4
NR (28) NR (31) 41 19
NR (33) NR (29)
8.8 6.9
NR (40) NR (27)
79 (73) 78 (70) 44 (44) 43 (43) 18
8.5 7.6 7.8 7.8 14.5
NR (33) NR (26) NR (34) NR (26) NR
16
6.7
NR
c. Chemoradiotherapy with one drug/combination vs. another SWOG, 1980 60 mCCNU 5-FU 60 mCCNU 5-FU testolactone GITSG 9277, 1985 60 5-FU 40 doxorubicin Earle et al. 1994 50–60 5-FU 50–60 hycanthone Li et al. 2003 50.4– Gemcitabine 61.2 50.4– 5-FU 61.2
Adjuvant chemotherapy 611
Some studies have compared combination chemotherapy against best supportive care and in every case there was improved survival in the treated arm (Table 25.3). Prior to the results obtained with ECF, commonly used regimens in gastric cancer included FAM, FAMTX and EAP. FAM (5-FU, Adriamycin and Mitomycin C) and FAMTX were compared in a randomized trial and response rates were 41 per cent (FAMTX) vs. 9 per cent (FAM) with median survival 42 vs. 29 weeks (P 0.004). At 1 year 41 per cent of FAMTX vs. 22 per cent of FAM patients were alive, but toxicity limits use of FAMTX. Epirubicin, Adriamycin and cisplatin (EAP) produced an overall response rate of 73 per cent (complete response 16 per cent) in a phase II trial of advanced metastatic gastric cancer. EAP vs. FAMTX showed the latter to be less toxic although equally effective (response rates 20 per cent vs. 33 per cent). There were four treatment-related deaths in the EAP group and none in the patients treated with FAMTX. EAP and FAM were then compared in a population of 90 patients with advanced gastric cancer who were treated until disease progression. The response rates were 55.6 per cent (EAP) vs. 16.7 per cent (FAM) with the former regimen producing less toxicity. The median diseasefree survival of complete responders was longer with EAP treatment (12 months vs. 8 months) but no difference was seen in overall survival (7 months vs. 5 months). The studies detailed earlier along with the FAMTX vs. ECF give a clear picture of how sequential phase II trials can result in selection of the optimal chemotherapy regimen. It may be appropriate to substitute carboplatin for cisplatin if the latter is contraindicated for renal or ototoxicity reasons since in a population of 20 evaluable patients, carboplatin and combination with infusional 5-FU resulted in a response rate of 45 per cent. Toxicity was mild and there was no measured deterioration in renal function. The median survival time was 9 months. Substitution of doxorubicin for epirubicin (ACF) in a phase II trial of 32 patients resulted in similar efficacy (MS 9.3 months in previously untreated patients) but increased toxicity.22 Biological response modifiers were investigated in conjunction with chemotherapy in advanced gastric cancer in an attempt to emulate the successful biochemical modulation of 5-FU seen in colorectal tumours. A combination of 5-FU, folinic acid and alpha interferon 2B resulted in an objective response in 17 of 36 (47 per cent) patients with a significant reduction in tumour-related pain in 22 of 26 (85 per cent). Toxicity was acceptable and the median duration of response was 5.5 months. Docetaxel has single agent activity and although myelosuppressive is also a useful option in combination treatment for advanced stomach cancer. The REAL2 trial compared four regimens in advanced stomach cancer – standard ECF, EC with capecitabine (ECX), E oxaliplatin F (EOF) and E oxaliplatin capecitabine (EOX). There was equivalence for infusional 5-FU with capecitabine, thus removing the requirement for Hickman
line insertion and replacing it with oral therapy. The study was not powered to determine whether oxaliplatin was superior to cisplatin. It is possible that chemotherapy results have plateaued and that little improvement will be seen with changing drug combinations. The results of chemotherapy in combination with targeted agents – for example the antiangiogenic bevacizumab – are awaited but the promise shown in other solid tumours makes it likely that this and other targeted drugs will be of added benefit in therapy of gastric cancer.
ADJUVANT CHEMOTHERAPY Early 5-FU-based trials of adjuvant chemotherapy suggested that survival could be improved. Studies of adjuvant Mitomycin C started within 6 weeks after surgery resulted in improved survival in the treated group (P 0.025) with a reduction in liver metastases and less likelihood of dying from recurrence (74 per cent control vs. 59 per cent treatment). The FAM regimen was randomized against no postoperative treatment in a population of 315 patients. No significant difference in survival was seen in comparison with 281 assessable patients. The South West Oncology Group studied the same regimen in stage IB IC II and III tumours and reported no significant difference in 5-year survival in their population of 176. Other studies have compared against surgery only the combinations 5-FU plus methylCCNU for 2 years, 5-FU plus methyl CCNA plus Lothamasol, 5-FU plus Adriamycin. The conclusion of a meta-analysis of adjuvant chemotherapy trials involving 2096 patients was that there was no survival benefit. INTRAPERITONEAL CHEMOTHERAPY
Cytological examination of perioperative peritoneal washings in patients with gastric cancer reveals malignant cells in 20–30 per cent of patients considered to have a poor prognosis. Because of data in ovarian cancer where intraperitoneal therapy can salvage some patients with small volume residual disease, single agent cisplatin 60 mg/m2 was given as an intra-peritoneal treatment via a Tenckhoff catheter to 18 patients within 8 weeks of complete resection of the gastric cancer tumour where there was serosal involvement, positive regional nodes or cytologically positive peritoneal washings. Of the treated population 22 per cent had cytologically positive peritoneal washings before or during chemotherapy. The median survival of the group was 17 months and the pattern of relapse was similar to that of untreated historical controls. Six patients relapsed with local/peritoneal disease and four with liver metastases suggestive either of cisplatin resistance or of poor drug penetration into peritoneal seedlings. Postoperative intraperitoneal 5-FU and cisplatin following three cycles of neo-adjuvant FAMTX has been studied in high-risk gastric cancer (defined by findings of endoscopic
612 Stomach
ultrasonography). A total of 23 patients have completed treatment to date and FAMTX was found to down-stage the tumours, resulting in 70 per cent resectability. The results of impact on disease-free survival and overall survival are pending. The final study is probably indicated by the future direction of intra-peritoneal therapy. Intra-peritoneal mitomycin with IV poly chemotherapy to cover both local and systemic relapse would be a feasible and interesting approach to the management of high-risk disease. PERI-OPERATIVE/NEOADJUVANT CHEMOTHERAPY
The main difficulty with adjuvant therapy in gastric cancer is the patients’ fitness post extensive surgery. Given preoperatively, this problem is avoided and it is argued that the tumour bulk may be reduced and ‘sterilized’ to prevent spread of the tumour cells at the time of operation. The practical problems with neoadjuvant treatment in gastric cancer include the relatively poor efficacy of established regimens and the difficulty in running studies with adequate controls. Etoposide/Adriamycin/cisplatin (EAP) was used to treat 34 patients with gastric cancer who had failed laparotomy. A total of 23 patients achieved a major response and 20 then progressed to definitive surgery with resection possible in 15. Of these patients, 5 had a pathological complete response, 2 were unresectable and overall the relapse rate of 20 months was 60 per cent. Median survival was 18 months overall and 24 months in the disease-free group. In the EORTC study of FAM vs. FAMTX in unresectable patients 19 received FAMTX and 7 proceeded to second laparotomy; 3 of these patients had complete resection of the residual disease. Intra-arterial chemotherapy with 5-FU and Adriamycin vs. IV therapy with the same agents vs. surgery alone was investigated in a randomized study incorporating 207 patients preoperatively; 62 per cent of patients had no residual tumour in the resected stomach following intraarterial treatment. A further 19 per cent of patients had small residual foci tumour cells. There was no significant response to IV preoperative chemotherapy. There was no improvement in 2-year survival between IV vs. surgery contrasting with an improvement of 29 per cent in the intraarterial vs. IV group and 45 per cent in the intra-arterial group vs. surgery. Experience with ECF initially was that 28 of 35 unresectable patients achieved a response and 19 of these went on to have surgery. A total of 10 complete resections were possible and of these 4 patients had a pathological complete response. A total of 6 patients had unresectable disease. Following on from these findings, the MAGIC trial compared standard surgery alone with three cycles of ECF pre- and postoperatively.23 There was a statistically significant improvement in completeness of resection progression-free survival (Hazard ration 0.66) and survival (hazard ration for death 0.75; 5-year survival 36 per cent vs. 23 per cent).23
Since only 40% of patients completed the postoperative chemotherapy, it is likely that the preoperative treatment had most bearing on the survival benefit. In most UK centres, neoadjuvant chemotherapy is yet to be accepted as standard treatment for resectable gastric cancers.
FUTURE DIRECTIONS Facilitation of the early diagnosis of gastric cancer is an obvious ambition to those whose aim it is to cure more people of the disease but the incidence in this country is not high enough to warrant random screening. Since it is unlikely that there will be significant progress with currently available cytotoxics alone or in combination, incorporation of novel ‘targeted’ agents needs to be fully evaluated in randomized trials. Pharmacogenetics may also contribute knowledge about patients’ likelihood of responding to particular pharmaceutical regimens. There are few clinical trials for this patient population, and more are required to discover better treatment for advanced or relapsed gastric cancer.
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
●
Incidence of gastric cancer is reducing in the West. Proximal gastric cancers now outnumber those of the antrum. Genetic response to infection with H. pylori may be involved in the likelihood of an individual developing gastric cancer. Surgical removal of organs adjacent to the stomach is not recommended. Surgical standardisation is necessary in trials of operative and other therapies in stomach cancer. Intraoperative radiotherapy (IORT) is not standard treatment. Intensity modulated radiation therapy (IMRT) maximises therapeutic dose to the tumour while protecting critical organs. The standard chemotherapy for advanced gastric cancer is epirubicin, cisplatin and infusional 5-fluorouracil (ECF). While adjuvant chemotherapy may be of benefit in high risk subgroups, post-operative morbidity makes it difficult to deliver. Median survival for advanced disease remains around 9 months.
REFERENCES 1 Allum WH, Hallissey MT, Kelly KA. Adjuvant chemotherapy in operable gastric cancer. Five-year follow-up of first British Stomach Cancer Group trial. Lancet 1989; 1(8638):571–4.
References 613
2 Gunderson LL, Sosin H. Adenocarcinoma of the stomach: areas of failure in a re-operation series (second or symptomatic look) clinicopathologic correlation and implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 1982; 8:1. 3 Landry J, Tepper JE, Wood WC, et al. Patterns of failure following curative resection of gastric carcinoma. Int J Radiat Oncol Biol Phys 1990; 19:1357–62. 4 Wisbeck WM, Becher EM, Russell AH. Adenocarcinoma of the stomach: autopsy observations with therapeutic implications for the radiation oncologist. Radiother Oncol 1986; 7:13. 5 Zhang ZX, Gu XZ, Yin WB, et al. Randomized clinical trial on the combination of preoperative irradiation and surgery in the treatment of adenocarcinoma of gastric cardia (AGC) – report on 370 patients. Int J Radiat Oncol Biol Phys 1998; 42:929. 6 Ajani JA, Mansfield PF, Janjan N, et al. Multi-institutional trial of preoperative chemoradiotherapy in patients with potentially resectable gastric carcinoma. J Clin Oncol 2004; 22:2774–2780. 7 Roth AD, Allal AS, Brundler MA, et al. Neoadjuvant radiochemotherapy for locally advanced gastric cancer: A phase l-ll study. Ann Oncol 2003; 14:110–115. 8 Ajani JA, Mansfield PF, Crane CH, et al. Paclitaxel-based chemoradiotherapy in localized gastric carcinoma: Degree of pathologic response and not clinical parameters dictated patient outcome. J Clin Oncol 2005; 23:1237–1244. 9 Takahashi M, Abe M. Intraoperative radiotherapy for carcinoma of the stomach. Eur J Surg Oncol 1986; 12:247–50. 10 Ogata T, Araki K, Matsuura K, et al. A 10-year experience of intraoperative radiotherapy for gastric carcinoma and a new surgical method of creating a wider irradiation field for cases of total gastrectomy patients. Int J Radiat Oncol Biol Phys 1995; 32:341. 11 Sindelar WF, Kinsella TJ, Tepper JE, et al. Randomized trial of intraoperative radiotherapy in carcinoma of the stomach. Am J Surg 1993; 165:178–87. 12 Calvo FA, Aristu JJ, Azinovic I, et al. Intraoperative and external radiotherapy in resected gastric cancer: updated report of a phase II trial. Int J Radiat Oncol Biol Phys 1992; 24:729–36. 13 Moertel CG, Childs DS, O’Fallon JR, Holbrook MA, Schutt AJ, Reitemeier RJ. Combined 5-fluorouracil and radiation therapy
14
15
16 17
18
19
20
21
22
23
as a surgical adjuvant for poor prognosis gastric carcinoma. J Clin Oncol 1984; 2:1249–1254. Macdonald JS, Smalley SR, Benedetti J, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med 2001; 345:725–30. Hundahl SA, Macdonald JS, Benedetti J, Fitzsimmons T. Southwest Oncology Group and the Gastric Intergroup. Surgical treatment variation in a prospective, randomized trial of chemoradiotherapy in gastric cancer: the effect of undertreatment. Ann Surg Oncol 2002; 9:278–86. van de Velde CJ, Peeters KC. The gastric cancer treatment controversy. J Clin Oncol 2003; 21:2234. Smalley SR, Gunderson L, Tepper J, et al. Gastric surgical adjuvant radiotherapy consensus report: rationale and treatment implementation. Int J Radiat Oncol Biol Phys 2002; 52:283–93. Rich TA, Shepard RC, Mosley ST 2004. Four decades of continuing innovation with fluorouracil: Current and future approaches to fluorouracil chemoraditation theraphy. J Clin Oncol 22:2214–2232. Wagner AD, Grothe W, Haerting J, et al 2006 Chemotherapy in advanced gastric cancer: A systematic review and meta-analysis based on aggregate data. J Clin Oncol 24:2903–09. Webb A, Cunningham D, Scarffe H, et al. 1997 Randomised trial comparing epirubicin, cisplatin and fluorouracil versus fluorouracil, doxorubicin and methotrexate in advanced oesophagogastric cancer. J Clin Oncol 15:261–7. Ross P, Nicolson M, Cunningham D, et al. Prospective randomised trial comparing mitomycin, cisplatin and protracted venous infusion fluorouracil (PVI-5FU) with epirubicin, cisplatin and PVI-5FU in advanced oesophagogastric cancer. J Clin Oncol 2002; 20:1996–2004. Kulke MH, Wu B, Clark JW, et al. A phase II study of doxorubicin, cisplatin and 5-fluorouracil in patients with advanced adenocarcinoma of the stomach or oesophagus. Cancer Invest 2006; 24:229–34. Cunningham, D, Allum W, Stenning S, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophgeal cancer. N Engl J Med 2006; 355:11–20.
26 Bladder cancer ROBERT HUDDART, PAULA WELLS AND ALAN HORWICH
Introduction Epidemiology Aetiology Genetics of bladder cancer Pathology Clinical presentation and investigation
614 614 615 616 616 617
Management Management of invasive bladder cancer Management of advanced disease Key references References
620 625 631 633 633
INTRODUCTION
EPIDEMIOLOGY
Bladder cancer is a disease with significant heterogeneity of outcome, ranging from tumours involving mucosa, in which metastatic disease is rare, to muscle-invasive disease with poor prognosis. Management is further complicated as bladder cancer often occurs in the elderly, who frequently suffer concurrent medical conditions (often related to smoking) and may be of poor performance status, thus limiting treatment options. Invasive bladder cancer, in particular, remains an area of controversy, with significant variability in management worldwide. As in many arenas where uncertainty exists, this controversy is fuelled by the paucity of available data from prospective randomized clinical trials. Consequently, fundamental questions regarding treatment selection for individual patients have not been answered definitively. The literature contains studies addressing the various issues in the field, but often with conflicting or inconclusive results. From these, a number of recommendations can be made, but there is still a pressing need for large randomized studies to be conducted so that clear answers to these questions can be obtained. It must be emphasized, however, that despite this, there have been significant advances in operative technique and technical radiotherapy and in the administration of chemotherapy, with attendant reduction in treatmentrelated morbidity and improvements in survival in patients suitable for radical treatment. In addition, advances in molecular biology have provided us with further insights into the mechanism of bladder cancer pathogenesis, invasion and metastasis from which new targets for therapy may be derived.
Bladder cancer is the fifth most commonly diagnosed cancer in the UK, affecting more than 10 600 individuals each year and causing around 4900 deaths in 2003. The disease is commonest in men, with an age-standardized incidence of 22/100 000, compared with 6.3/100 000 in women, giving a male:female ratio of 2.5:1.0.1 It incidence has risen during the twentieth century in developed countries, including the UK. In Britain, the age-standardized incidence rates rose throughout the 1970s and 1980s to reach a peak in the late 1980s of around 31 per 100 000 males and 9 per 100 000 females, since when they have been declining. However, time trends and regional differences in incidence rates are difficult to interpret because of changing coding and classification practices affecting the definition of invasive carcinoma of the bladder.2 Internationally, the incidence rate of bladder cancer among men varies more than ten-fold3 and has been rising in many areas of the world.4 High rates occur in Western Europe and North America, whereas the disease is less common in Eastern Europe and Asia. The disease predominates in Caucasians, and incidence rises sharply with age (about two-thirds of cases occurring from 65 years onwards). Survival rates are improving for individuals in affluent groups but most particularly for men, who, in contrast to most other common malignancies, show a consistent survival advantage over women.5 The decline in death rate for men has been significant, falling from 12.2 in 1979 to 8.7 per 100 000 in the year 2003. For women, however, the rates have changed little, from around 3.3 in the early 1970s to 3.1 in 2003, resulting in a male:female ratio of 3.7:1 in 1979 to 2.8:1 in 2003.
Aetiology 615
AETIOLOGY
Drugs
Industrial exposure
A number of drugs have also been identified as aetiological agents in bladder cancer. For instance phenacetin, which was in general use as an analgesic, has been linked to carcinoma of both the upper and lower urinary tracts.18,19 Also cytotoxic drugs used in the treatment of cancer, such as the alkylating agent cyclophosphamide, are themselves carcinogenic, as are many of their metabolites. For cyclophosphamide, the carcinogenic effect is dose dependent, with bladder cancer risk increasing with dose delivered.20–22
Environmental factors are clearly implicated in the development of bladder cancer. A number of clinical studies and mortality surveys have resulted in the identification of hazards in the dye-stuffs and rubber industries6,7 associated with a 10–50-fold increased risk of bladder cancer, primarily attributed to exposure to the aromatic amines 2-naphthylamine and benzedine.6,8,9 The risk of death from bladder cancer is associated with young age at first exposure and increasing duration of employment, whereas mortality decreases with increasing time from last exposure. This association has been supported by data from the UK, where reduced bladder cancer risk was demonstrated following the introduction of protective measures and subsequent banning of the industrial use of 2-naphthylamine and benzedine in 1950 and 1962 respectively.10,11 2-Naphthylamine is also implicated in the development of bladder cancer in the rubber, electric-cable and chemical manufacturing industries. Subsequently, studies have suggested approximately 40 potentially high-risk occupations (but strong evidence of increased risk is apparent for very few of these) and the association of several other aromatic amines (including 4-aminophenyl and 4-chloro-o-toluene) in the development of bladder carcinoma.
Radiation The risk of bladder cancer is also increased in individuals exposed to ionizing radiation during therapeutic pelvic radiotherapy for both benign (e.g. dysfunctional uterine bleeding) and malignant disorders (e.g. cervical cancer).23,24 In these studies, increased risk was experienced by women who were under 55 years when treated, with increasing dose to the bladder and increasing time since exposure. The relative risk of up to 8.7 was seen for patients treated at least 20 years earlier. A similar association has been noted post-radioactive iodine (iodine-131) treatment for thyroid cancer and in the survivors of the atomic bombs of Hiroshima and Nagasaki.
Infection/inflammation Tobacco smoking Cigarette smoking is a well-established cause of bladder cancer and its importance lies in the prevalence of the habit within the population. It has been suggested that up to 40 per cent of all male and 10 per cent of female cases may be attributable to this exposure. Overall, smokers have a threefold risk, with mortality rates paralleling smoking patterns.12,13 The causative links were established as a result of positive associations demonstrated in a metaanalysis of 43 published cohort and case-control studies,14 the best known cohort study being that of British male doctors.15 The striking consistency of the findings, the dose–response relationships and the identification of at least two known bladder carcinogens in cigarette smoke (2-naphthylamine and 4-aminobiphenyl), as well as aromatic amine-based DNA adducts,16 give confidence that there is a causative link. The risk increases with smoking intensity (packs per day), and decreases with cessation of smoking (30–60 per cent reduction in risk of cancer), although in the long term it does not decline to the level of a non-smoker.14,17 However, the aetiological associations of tobacco with bladder cancer are complex. There are only weak correlations in incidence between the sexes and very little agreement with the geographical pattern or long-term trends of lung cancer in the UK, suggesting that the associations with smoking may be confounded by other factors.
Another important group of aetiological factors is those arising secondary to chronic irritation of bladder urothelium, most particularly chronic or repeated urinary tract infections, often associated with urinary stasis. The most important infective agent in the aetiology of bladder carcinoma is Schistosoma haematobium, which is endemic in developing countries and associated with a high incidence of squamous-cell carcinoma.25 There is evidence from Egypt of the efficacy of educational and early anti-schistosomal therapy in the prevention of bladder cancer.26 Developmental abnormalities of the bladder, bladder diverticulae, chronic bladder stones or indwelling catheters may also predispose to chronic infection, inflammation and increased risk of bladder carcinoma.27
Genetic predisposition A number of studies have investigated whether a genetic predisposition to bladder cancer exists. On the whole, these studies have shown an increased relative risk for developing bladder cancer in the relatives of bladder cancer patients.28 In studies amongst Utah Mormons, there was also an increased degree of familiality in bladder cancer patients.29 A case-controlled study showed that the risk was greatest in smokers (versus non-smokers). There is also evidence that patients with deficient detoxification mechanisms, such as those with slow acetylator phenotype associated with
616 Bladder cancer
homozygosity for a mutated gene (NAT 2)30,31 or homozygous absence of a functional human glutathione transferase M1 (GSTM1) gene (which codes for a family of enzymes that detoxify a spectrum of reactive carcinogenic metabolites by catalysing their conjugation to glutathione)32 are at increased risk of bladder cancer.
Normal urothelium
9p/9q– 9p/9q– Hyperplasia
Atypia/G1 TP53
GENETICS OF BLADDER CANCER A number of different oncogenes and tumour suppressor genes have been implicated in the aetiology of bladder cancer.33 The most consistent abnormality seen in its development is homozygous deletion of chromosome 9. Loss of the entire chromosome 9 allele has been reported in up to 30–75 per cent of clinical tumours, including small, localized papillary carcinoma. In some tumours, the loss is restricted to either the short arm (especially at 9p21) or the long arm of chromosome 9 (at one of three, loci 9q13–31, 9q32–33 and 9q34). 9p21 is the site of the tumour suppressor p16INK4a (and associated genes p14 ARF and p15INK4b), which are now known to be homozygously deleted in at least 20–45 per cent of bladder tumours. A large number of other bladder genetic changes have been identified in bladder tumours by comparative genomic hybridization and other techniques. The retinoblastoma gene (Rb) is likely to be the target of the 13q deletion, which has been shown to be frequently mutated/dysregulated in bladder tumours, particularly in tumours without the 9p21 abnormality. The p53 gene (the probable target of the 17p deletion) is mutated in 10–70 per cent of bladder tumours. Both these changes are associated with increases in grade and stage and with increased risk of progression and survival.34 This risk of poor outcome is particularly marked in the presence of abnormalities of both p53 and Rb 35,36 and is seen in patients with both T1 and invasive bladder cancer. These findings suggest that p53/Rb mutations are late rather than initiating events in the genesis of bladder cancer. In addition to these, a number of oncogenes have been shown to be amplified and mutated in a number of series with varying frequencies, at least partly due to the case mix (increased frequency in advanced disease). These include the C-erb-B2 (amplified in 10–14 per cent of bladder tumours), cyclin d1 (amplified in 10–20 per cent of tumours), H-ras (mutated in 6–44 per cent of tumours) and EGFR (over-expressed in 11–48 per cent of bladder cancers).33,37 EGFR has promoted most interest, as in at least one study, over-expression was an independent prognostic factor for survival and progression.38 It is clear from this discussion that multiple genetic changes are required to generate an invasive bladder carcinoma. One possible model is illustrated in Figure 26.1. In addition to these genetic alterations, bladder cancers exhibit changes in microvasculature,39 over-expression of vascular growth factors such as vascular endothelial growth factor, changes in adhesion molecules such as E cadherin, and re-expression of telomerase.40 Cycloxysenase2 (COX2) is
9p/9q–
Atypia
Atypia/G2
TP53 (65%) Superficial papillary (Ta) G1 G2 G3 9p/9q– (34%) TP53 (3%) 10–20%
TP53
Carcinoma in-situ
T1-invading lamina propria
Muscle invasive carcinoma (T2)
3p–, 8p–, 14q–, 13q–, 9p/9q– (12%) ? 9p/9q– TP53 (32%) 9p/9q– (59%) gains: 1q, 8q, 3p, 3q, 5p, 6p, 10p losses: 2q, 8p, 11p amplifications: 1q, 3p, 6p, 8p, 8q, 10p, 11q, 12q, 13q, Xp, Xq TP53 (51%) multiple other changes
Figure 26.1 Proposed model for the multiple genetic changes required to generate invasive bladder carcinoma. (From Knowles, 1999, with permission)
also over-expressed in cancer but not in normal urothelium, and expression increases with grade and stage to invasive disease.41–44 COX2 can promote tumorigenesis by inhibiting apoptosis and increasing angiogenesis, DNA synthesis and cell proliferation. COX2 has provoked recent interest due to epidemiological evidence that it could be a potential target in bladder cancer (see ‘Superficial disease and systemic treatments’).
PATHOLOGY In Europe and the USA, more than 90 per cent of tumours are of urothelial (transitional cell) origin and approximately 5 per cent are squamous-cell carcinomas, except in areas where schistosomiasis is endemic (where they constitute up to 80 per cent of all urothelial malignancy). Primary adenocarcinoma of the bladder may arise from the urachal remnant or bladder exstrophy, and the remainder are rare tumours such as small-cell carcinoma, sarcoma, lymphoma and melanoma.
Superficial bladder tumours The term benign papillomas (including the uncommon variants the inverted and everted papillomas) should be
Clinical presentation and investigation 617
reserved for well-differentiated lesions that do not show evidence of invasion, abnormal mitotic activity or cellular atypia. The term papilloma should not be used for superficial transitional cell carcinomas (TCCs), even if they have a frond-like or papillary appearance. All superficial TCCs (Ta, T1 CIS) are superficial to the deep muscle layer and exhibit evidence of invasion, abnormal mitotic activity or cellular atypia. Important prognostic differences exist between tumours confined by the basement membrane and those invading the lamina propria (which consists of loose connective tissue with its rich vascular and lymphatic channels). Within the lamina propria lies the muscularis mucosa, which some authors suggest has further prognostic significance as invasion of this muscular layer is associated with a significantly higher incidence of tumour recurrence and progression.45 Unfortunately, the muscularis mucosa is variably present, making it an unreliable landmark for diagnosis. The prognosis of superficial bladder cancer is strongly correlated with tumour grade. Approximately 75–85 per cent of patients with primary TCC present with low-grade tumours confined to superficial mucosa (Ta). Although up to 75 per cent of these recur, the majority remain amenable to transurethral resection and the selective administration of intravesical therapy, with only 10 per cent of patients with recurrent lesions demonstrating invasion over a 10-year period.46 It seems clear from genetic analyses that these multi-focal areas are clonally related Multi-focal bladder tumours thus represent a pan-urothelial monoclonal tumour dissemination.47 A proportion of papillary tumours will eventually invade superficial connective tissue (T1 disease), characterized by decreased survival. The 5-year survival is more than 90 per cent for patients with Ta tumours (non-invasive) and 75 per cent for T1 tumours. High-grade superficial (T1) tumours are a considerably more aggressive entity, which recur in 80 per cent and progress to invasive disease in more than 50 per cent of cases.48,49
cohesion, often resulting in very few malignant cells or only denuded mucosa. The tumour cells are often large with dysplastic nuclei and prominent nucleoli.
Carcinoma in situ
Approximately 80 per cent cases of bladder cancer present with haematuria, which is usually painless and intermittent and either visible (macroscopic) or detected on routine urinalysis (microscopic). The degree of haematuria does not correlate with disease extent and therefore patients presenting in this way should be investigated as a matter of urgency. Fifteen per cent patients presenting with macroscopic haematuria and approximately 6 per cent with asymptomatic microscopic haematuria harbour bladder cancer. The presence of pain suggests an inflammatory component most commonly associated with bacterial or interstitial cystitis, but may also occur in up to 25 per cent of patients with bladder cancer. Irritative voiding symptoms, including urinary frequency, urgency and dysuria, are particularly associated with CIS and invasive bladder tumours.53 Bladder cancer may also present with bladder outlet obstruction or ipsilateral flank pain secondary to ureteral obstruction. Patients with advanced or metastatic disease often suffer constitutional symptoms, including anorexia, weight loss
Carcinoma in situ (CIS) is an important variant of superficial bladder cancer characterized by a high propensity for progression to invasive disease, more than 54 per cent of cases progressing within 5 years.50 It can be diffuse or focal, symptomatic or asymptomatic, and may or may not be associated with superficial papillary tumours. Diffuse CIS, particularly when associated with symptoms or identifiable papillary tumours, is an unfavourable prognostic finding. Riddle et al. reported a progression rate of 58 per cent in patients with diffuse CIS, whereas only 8 per cent of those with focal disease progressed.51 Furthermore, in a study of patients with CIS associated with low-grade papillary tumours, an incidence of subsequent muscle-invasive disease or metastases of 83 per cent was noted.52 Macroscopically, the lesion is red and velvety and frequently extensive, involving prostatic urethra, prostate and ureters. The urothelium is markedly abnormal, showing a marked lack of cellular
Invasive bladder carcinoma The gross appearance of invasive TCC is variable, ranging from strikingly papillary, nodular or polypoid to sessile and ulcerated lesions. Most commonly there are nests, small clusters or single cells irregularly dispersed in the lamina propria and muscularis propria, but occasionally a more diffuse pattern is present. The neoplastic cells are usually of moderate size with modest amounts of pale to slightly eosinophilic cytoplasm, with at least moderate, if not marked, cellular atypia present. Approximately 10 per cent of TCCs, particularly high-grade tumours, contain foci of glandular or squamous differentiation or even atypical spindle cells that mimic sarcomas. Other variants occur more rarely. Squamous-cell carcinomas vary in morphological appearance but are usually large and deeply invasive, even when well differentiated. Their microscopic appearance is similar to that of lesions arising elsewhere in the body; most are moderately or well differentiated and often abundantly keratinized. Adenocarcinoma is generally of poor prognosis, with tumours arising from urachal remnants having the best outlook and signet-ring carcinomas (accounting for 3–5 per cent of tumours) the worst. The mucosa is usually oedematous and ulcerated in most cases of adenocarcinoma, but some develop diffuse fibrosis and mural thickening similar to linitis plastica of the stomach. Small-cell carcinoma is microscopically similar to its lung counterpart and carries a similarly gloomy outlook.
CLINICAL PRESENTATION AND INVESTIGATION
618 Bladder cancer
and pain arising from sites of metastasis such as bone. Physical examination is often unremarkable in these patients, but a careful pelvic examination should be undertaken to determine whether a mass is palpable or to assess the fixity of tumour to adjacent organs. Urine microscopy and culture should be performed and, if negative, repeated in view of the intermittent nature of the haematuria.
Management of haematuria Urological opinion should be sought in any case of macroscopic haematuria or persistent microscopic haematuria. Investigation of the upper urinary tract is obligatory and may be achieved with intravenous urography, or ultrasound of the kidneys and bilateral retrograde ureteropyelography performed at cystoscopy for patients allergic to intravenous contrast material or with a history of renal insufficiency. Ipsilateral hydronephrosis is often associated with muscleinvasive bladder carcinoma (approximately 90 per cent of cases) and its presence should prompt rigorous investigation.54 The routine use of urinary cytology in all patients suspected of harbouring bladder carcinoma is debated. Cytology is most accurate for patients with high-grade tumours and CIS, but there remains a 20 per cent falsenegative rate, while low-grade tumours are so frequently associated with a negative result that positive cytology should raise the suspicion of a concomitant high-grade lesion. Urine cytology should be obtained from a well-hydrated patient, as cellular degeneration occurs when urine remains in the bladder for prolonged periods. Other factors that may artefactually alter urinary cytology include the presence of urinary tract infection, indwelling catheters, bladder instrumentation, radiotherapy and intra-vesical immunotherapy or chemotherapy. Saline bladder washings may be employed to improve the accuracy and sensitivity of the technique. Several tests have been developed for the diagnosis or follow-up of recurrent bladder cancer on voided urine samples. Some, including the BTA test, the BTA Stat Test, the BTA TRAK assay and the urinary nuclear matrix protein NMP22, are commercially available. The BTA test is a latex agglutination assay that detects the presence of basement-membrane antigens isolated and characterized in the urine of patients with bladder cancer.55 The BTA Stat and TRAK assays, on the other hand, are sensitive to human component factor H-related protein produced in vitro specifically by human bladder cancer cell lines.56,57 These tests were initially developed as a more sensitive alternative to urine cytology in detecting primary and recurrent superficial bladder cancer,58,59 but recent work has highlighted the risk of false-negative results and of false-positive reactions in patients with inflammatory bladder conditions or other genitourinary malignancy.60,61 Several other assays and techniques remain under development, including tests to identify nuclear matrix proteins (NMPs),62 telomerase40 and other genetic alterations.63 Most of these are limited in their detection of well-differentiated tumours and, because
of this, none has sufficient accuracy to replace standard investigations for either diagnosis or follow-up. Recently, there has been interest in the development of methods to detect proteins of the mini-chromosome maintenance (MCM) family, which are known to play a critical regulatory role in the initiation of DNA replication. It has previously been demonstrated that deregulated expression of MCM proteins is characteristic of epithelialcell carcinogenesis, resulting in the exfoliation of MCMpositive tumour cells that have been used in the screening of cervical carcinoma.64 Because MCM proteins identify both proliferating and non-proliferating cells with proliferative potential, MCM protein expression is a more sensitive marker of epithelial carcinogenesis than is expression of conventional proliferation markers. The percentage of MCM5-expressing cells in urothelial carcinoma is related to grade of TCC, averaging 78 per cent for poorly differentiated, 70 per cent for moderately differentiated and 45 per cent for well-differentiated tumours.65 Subsequent investigation has demonstrated that elevated levels of MCM5 in urine sediments are highly predictive of bladder cancer,66 raising the possibility that this technique could be used in primary staging, detection of relapse and screening for bladder cancer. Ultimately, the diagnosis of bladder carcinoma is made on cystoscopy with pathological evaluation of a resected tumour specimen. The initial examination for patients with haematuria is often by flexible cystoscopy under local anaesthetic. Abnormal findings are confirmed under general anaesthetic, with bimanual examination to confirm the presence, extent and fixation of a palpable bladder mass. The presence of induration or a palpable mass on bimanual examination following apparent complete resection of tumour implies extra-vesical extension, whereas resolution of these features implies organ-confined disease. Cystoscopic evaluation involves inspection of the entire urethra, prostate and bladder neck with retrograde pyelography when the upper urinary tract has not already been satisfactorily visualized. The number, location, size and configuration of all tumours and any associated mucosal abnormalities should be noted, and documented on a bladder map. Biopsy or resection of a tumour should be accompanied by directed biopsies of adjacent and normal-appearing bladder mucosa, and should only be performed once satisfactory evaluation of the entire urinary tract has been undertaken. General assessment of the patient should include full blood count, biochemistry and a chest X-ray and a bone scan when there is a history of bone pain.
Staging The tumour node metastasis (TNM) system was adopted by the Union Internationale Contre Cancer (UICC) in 1963. Tumours are classified by this system according to anatomical extent, using data from examination under
Clinical presentation and investigation 619
1. 2. 3. 4.
Epithelium Subepithelial connective tissue Muscle Perivesical fat
Box 26.1 Union Internationale Contre Cancer (UICC) TNM staging system for bladder carcinoma
4
3 2 1 T1
T2a
T3a
Tis Ta T4b T pT
T2b
T3b T4a
Figure 26.2 Diagnostic representation of the 2002 TMN classification for bladder carcinoma.
anaesthesia and radiological investigations. As accurate assessment of muscle invasion is not possible by clinical examination alone, a postoperative modification using the operative specimen is also required (pTNM). The latest modification, published in 2002, includes changes that reflect the prognostic implication of organ-confined, muscle-invasive disease versus perivesical extension (Fig. 26.2).67 Tumours invading the superficial (inner half) muscle wall are now designated T2a, and those of the deep (outer half) muscle layer T2b, whilst any tumour penetrating muscle into perivesical fat is classified as stage T3. Support for the redefinition came from observations by Pearce and Blandy,68,69 who noted little difference in survival between superficial and deep muscle invasion. Furthermore, the presence or absence of muscle invasion rather than its depth dictates management strategies for most centres. Another area requiring clarification is the prognostic distinction between non-invasive tumour growing into the prostate from the urethra along the prostatic ducts, and direct invasion of a bladder tumour into prostatic stroma (a true T4a tumour). Pugh first drew attention to this problem with the UICC staging system which reclassified noninvasive tumour as P4aa and disease invading into prostatic stroma as P4ab.70 Similar observations have recently been made, confirming that prostatic stromal involvement adversely affects prognosis.71 The classification of nodal status has also been modified from the 1978 system, in which it was staged according to site alone, to that of the 2002 edition, in which classification is according to number and size (Box 26.1).
Imaging Accurate staging of bladder cancer is important both to determine prognosis and for treatment decision making.
TX Primary tumour cannot be assessed. TO No evidence of primary tumour. TA Non-invasive papillary carcinoma. Tis Carcinoma-in-situ: flat tumour. T1 Invades subepithelial connective tissue. T2A Superficial (inner half) muscle invasion. T2B Deep (outer half) muscle invasion. T3 Invades perivesical fat. T3A Microscopic. T3B Macroscopic. T4A Invades prostate, uterus or vagina. T4B Invades pelvic or abdominal wall. Primary tumour (G): suffix m should be added to the appropriate T to indicate multiple tumours. Suffix IS may be added to any T to indicate the presence of associated carcinoma-in-situ. Lymph nodes NO No regional node involvement. N1 A single node, less than 2 cm in largest diameter. N2 A single node 2–5 cm or multiple nodes less than 5 cm. N3 Nodal involvement greater than 5 cm. Provisional TX: regional lymph nodes cannot be assessed.
Unfortunately, conventional imaging modalities are limited by the inability to detect microscopic disease and by inaccuracies in definition of macroscopic disease. Where an increase in sensitivity is achieved, specificity is compromised, as interpretation of abnormalities as ‘cancer’ lead to increased detection of disease but overestimate the extent and presence of disease in many instances. However, radiological investigations form a vital part of tumour staging, and an understanding of the limitations of each modality is important for the interpretation of clinical data. For many years the mainstay of pelvic staging has been computed tomography (CT), which can identify tumours that extend into the bladder lumen and perivesical tissue and the presence of bladder-wall thickening. However, these findings may reflect inflammatory processes rather than tumour infiltration, which limits the value of CT scans in defining depth of invasion and consequently stage of disease. A comprehensive review by MacVicar and Husband (1994) further addressed these issues, observing that CT was unable accurately to demonstrate deep muscle invasion due to an inability to distinguish between individual layers of the muscle wall, but was useful in defining extension of disease through the bladder wall (T3).72 The ability of CT to detect involved lymph nodes is limited by size. Nodes of greater than 1–1.5 cm in diameter are considered abnormal, with a resulting sensitivity of between 50 per cent and 85 per cent. However, nodal enlargement results from a
620 Bladder cancer
number of other causes, resulting in false positives, such that specificity rates lie between 67 per cent and 100 per cent.73–75 Magnetic resonance imaging (MRI) is generally thought to offer advantages over CT in the staging of bladder cancer, but where perivesical tumour and advanced disease invading adjacent structures exist, CT may be adequate. Computed tomography is also equivalent in the definition of involved lymph nodes. Magnetic resonance imaging is superior to CT in the delineation of organ-confined tumour, due to an increased definition of superficial from deep muscle invasion. However, its ability to distinguish between T2a and T2b tumours is still limited, with accuracy increasing for the higher tumour stages.76 The facility for multi-planar imaging is also particularly advantageous in the assessment of tumours of the bladder base. Magnetic resonance imaging is therefore recommended as the staging method of choice if facilities are available.77 Gadolinium-DTPA enhancement improves the assessment of bladder-wall invasion, but the technique is affected by late tissue changes resulting from radiotherapy.78 Dynamic MRI with contrast enhancement has resulted in further improvement in image quality in the pelvis, which may translate into more accurate preoperative staging.79 Transurethral ultrasound (TUUS) has been reported as the most sensitive method of staging bladder cancer locally, with a reported staging accuracy of 83 per cent.80 Koraitim et al.81 reported a 100 per cent correlation between preoperative TUUS and pathological staging from cystectomy specimens in non-invasive (Ta and T1) tumours, but the correlation fell with increasing depth of invasion, to 96 per cent for T2a and b and 70 per cent for T3 disease. A major disadvantage of this technique is the need for general or spinal anaesthesia at the time of cystoscopy, limiting its use to specialist centres.
Screening The fact that patients with bladder carcinoma usually suffer haematuria at some point in their disease raises the possibility of testing for its presence as a screening test. Haematuria is, however, frequently intermittent, and repeated testing is required to confirm its presence. Screening for haematuria can be achieved by microscopic analysis or using a chemical reagent strip that detects the presence of haemoglobin. Two screening studies in the general population using home testing have reported similar findings.82,83 In these, 15–20 per cent of the screened population had haematuria, and of those who completed the evaluation, 6–8 per cent had urothelial cancers, with 1.2–1.3 per cent arising in the bladder. It must be noted that neither study involved a randomized control population, and therefore results must be interpreted with caution. However, using age-matched and gender-matched unscreened controls derived from the tumour registry,82 it was demonstrated that both populations had approximately the same
proportions of superficial and high-grade carcinomas, but that the number of muscle-invasive tumours was much higher in the unscreened population. The time to death from bladder cancer was notably increased in the screened population, but this may reflect lead-time bias. Thus population screening may increase the probability of diagnosing disease at an earlier stage but it remains unclear as to whether this affects survival. The positive predictive value (proportion of patients who test positively and are found to have disease) would be improved by limiting screening to those at high risk. For bladder cancer, this may mean restricting participation to people with occupational exposure to known bladder carcinogens and to men aged 50 and over with smoking histories. Results of prospective randomized studies to determine the value of screening in this disease are awaited.
MANAGEMENT Pathological examination of a cystectomy specimen provides the most reliable means of determining stage and patient prognosis for recurrence-free and overall survival. However, the clinical behaviour of bladder cancer is not adequately predicted by stage and grade alone, and the ability precisely to define the true biological potential of a tumour would facilitate better treatment selection, which may translate into improved survival figures. Our understanding of tumour biology has evolved rapidly over the past decade, prompted by advances in molecular biology, immunobiology and cytogenetics. From this, novel tumour markers have been identified and are being evaluated as potential prognostic indicators that may be applied in a clinical setting. The main division in management is between tumours that invade muscle, where patients are at substantial risk of local and distant recurrence, and those that are restricted to the superficial lamina propria with a good prognosis. We will therefore consider management in three groups: superficial low-grade disease, superficial high-grade disease (including CIS), and muscle-invasive carcinoma.
Superficial low-grade tumours SURGERY
The vast majority of superficial bladder cancers are amenable to transurethral resection (TUR). Full clinical staging is undertaken prior to resection, including biopsies of apparently normal mucosa, particularly when urinary cytology is positive in the absence of an obvious bladder tumour and a normal upper tract. Finally, tumour resection should be deep enough to obtain muscle as part of the evaluation of the T category. Recurrent low-grade papillary tumours can also be treated by thermo-coagulation using a neodymium-Yag (Nd-Yag) laser. This usually treats 3–5 mm, and although
Management 621
Table 26.1 Prognostic factors for superficial bladder cancer EORTC trials
MRC trial
Time to first recurrence (univariate) Number of tumoursa Gradeb Prior recurrenceb Time from diagnosisb Site of tumourb a b
Recurrence rate/year (multivariate)
Time to invasion (multivariate)
Recurrence rate at 2 years (univariate)
Recurrence at 3 months Prior recurrence Number of tumours at entry Grade
Recurrence at 3 months Grade Prior recurrence Site of tumour Size of tumour Sex
Result of 3-month cystoscopyb Number of tumoursb Gradeb Maximum sizeb Site of tumour
Included in multivariate analysis if site is excluded. Positive on multivariate analysis.
Table 26.2 Risk of recurrence of patients with TaT1 bladder cancer according to prognostic groupings of Parmar et al., 198989
Group1 Group 2 Group 3
Diagnosis
3-month cystoscopy
Solitary Solitary Multiple Multiple
No recurrence Recurrence No recurrence Recurrence
transmural coagulation of the bladder wall has occasionally been observed with higher-energy systems, bladder integrity is usually preserved and perforation with extravasation is uncommon.84 Thus laser is considered a safe and effective treatment for papillary low-grade tumours. Where highgrade disease or tumour invasion is suggested by a more sessile appearance, electrocautery resection is preferable, as an adequate specimen containing detrusor muscle for histological examination can be more readily assured. More recently, attempts to have been made to improve the efficiency of local resection by using fluorescence endoscopy. Fluorescence endoscopy is based on using a fluorochrome, such as 5-amino laevulinic acid (ALA) or its ester derivative, which is then visualized by light of appropriate wavelength (a blue light for ALA). Tumours emit light of a different wavelength (red for ALA). A number of studies have demonstrated enhanced sensitivity for tumour detection, with sensitivities of 70–75 per cent for standard resection compared to more than 95 per cent for fluorescence endoscopy,85,86 with detection of CIS being of particular value. In a randomized trial of 191 patients, Filbeck et al. showed fluorescence endoscopy improved 4-year recurrence-free survival from 60.7 per cent to 85 per cent,87 although the long-term impact on progression is unclear.88 PREDICTING RECURRENCE IN PATIENTS WITH SUPERFICIAL TUMOURS
The initial management of all patients with superficial bladder cancer involves complete resection of the tumour, with subsequent treatment being dependent on the predicted
Proportion of patients (%)
2-year recurrence free rate (%)
60 30–35
74 44
5–10
21
risks of recurrence and progression. Several studies have investigated factors that may predict recurrence and define patients requiring further treatment. Two of the largest series are those of the British Medical Research Council (MRC)89 and the European Organisation for Research and Treatment for Cancer (EORTC),90 based on patients entered into their randomized trials of intra-vesical chemotherapy (see below). Although generalizations from these data must be interpreted with care, as only patients fulfilling the entry criteria for the studies are included, several important prognostic factors have been identified, as shown in Tables 26.1 and 26.2. Both studies highlighted the importance of grade and tumour number at presentation, but also of the 3-month cystoscopy as a predictor of subsequent recurrence. Indeed, in the MRC analysis, when results from the 3-month cystoscopy were combined with the number of tumours at presentation, no other factor added significantly to the prediction of relapse. Not surprisingly, disease recurrence at cystoscopy continues to be highly predictive of future recurrence, with the probability of developing recurrence decreasing with each negative cystoscopy, reaching 8 per cent at 5 years and 0 per cent at 10 years.91 In these studies, T stage (Ta versus T1) was not a strong predictor for recurrence, but it is likely, as shown by Kurth et al.90 that T stage does predict risk of progression. Over recent years, a number of new biological factors have been identified as possible predictors of either recurrence or progression (Table 26.3). However, it is uncertain, with the possible exception of EGFR, whether they add predictive strength to the currently identified factors. Therefore further evaluation is required, including their use in large patient
622 Bladder cancer
Table 26.3 Biological markers of recurrence and progression Prognostic factor
Association with bladder cancer
Limitations
References
ABO blood antigen
Loss of expression with bladder cancer
20% of normal cells do not express antigen
292
M344 antigen
Decreased expression associated with increased grade
Adds little to grade
293
Rarely positive in G3 tumours Epidermal growth factor receptor
Over-expression linked to increased cancer-specific death rate, reduced recurrence-free survival, increased recurrence and progression
Over-expression associated with higher grade, stage and ploidy Not conclusive predictive marker of superficial disease
295 296 294
Acidic fibroblast growth factor (FGF)
Expression correlates with tumour stage
Lacks specificity and is elevated in other disease processes, e.g. benign prostatic hypertrophy
297
Transforming growth factor-beta
Highest levels found in low-grade and low-stage tumours
Not independent prognostic indicator of survival
298, 299
Decreased expression associated with increased probability of progression and reduced survival Rb and p53
Excellent outcome35,36
populations and inclusion into multivariate models, before any will be accepted into routine clinical practice. INTRA-VESICAL THERAPY: CHEMOTHERAPY
Intra-vesical therapy can be given in the prophylactic, or adjuvant, setting, where it is intended to prevent recurrence after endoscopic resection of all visible tumours, or as definitive therapy where it is designed to treat unresectable papillary tumours, or CIS. The advantages of the intra-vesical route are that high concentrations of agent are in contact with tumour-bearing mucosa or bladder epithelium at risk with little or no systemic toxicity. Disadvantages include the local side effects in the bladder due to high local drug concentrations and the need for transurethral manipulation. A variety of drugs of similar efficacy and toxicity have been used for the prophylactic treatment of superficial bladder cancer and are detailed in Table 26.4. In Europe, availability and practical considerations determine the choice of treatment from adriamycin, epirubicin and mitomycin-C. A number of randomized trials, including those of the EORTC and MRC, have addressed the optimal use of intravesical chemotherapy in the prophylactic treatment of stages Ta and T1 bladder cancer.46,92–94 Most demonstrated that adjuvant treatment following TUR resulted in decreased recurrence rate or prolonged disease-free interval but had insufficient statistical power to detect differences in time to progression and survival. A meta-analysis performed by the EORTC GU Group and the MRC Working Party on Superficial Bladder Cancer 95 has confirmed the favourable impact on disease-free interval in these patients. However, no long-term benefit could be demonstrated in terms of
300
Not independent predictor of recurrence
300
increasing time to progression to invasive disease, duration of survival or progression-free survival. The benefit of early versus more delayed chemotherapy as well as short (sometimes single-instillation) chemotherapy versus long-term adjuvant therapy has also been studied. Three trials investigating a single instillation of mitomycin-C (MMC) (40 mg in 40 mL saline) or epirubicin (80 mg in 40 mL saline) given within 24 hours of TUR 96,97 demonstrated a 40–50 per cent reduction of risk of recurrence compared with TUR alone. This treatment has minimal toxicity and, as even the best prognostic groups have a 30 per cent risk of recurrence (see Table 26.2), these results suggest that a single instillation of mitomycin-C (or epirubicin) may be advantageous at diagnosis. Trials comparing single and more prolonged treatments suggest that additional benefit may be gained by repeated instillation of intra-vesical chemotherapy, although the majority of the benefit is gained from an early postoperative treatment. Such treatment is therefore advised for patients at high risk of recurrence at presentation or first cystoscopy (MRC groups 2/3; Table 26.5), patients with multiple tumours or frequently relapsing patients. A number of different regimes have been investigated, varying from the MRC regime of five courses of treatment administered at 3-monthly intervals to the EORTC regime of weekly instillation for 4 weeks followed by monthly treatment for 11 months. The optimal duration of treatment has also been addressed by the EORTC. A 6-month course is usually sufficient, but 12 months provides better results for patients in whom early intra-vesical instillation is not possible.98 However, despite this, there is little clear evidence that one regime is superior to another, and a common pragmatic
Management 623
Table 26.4 Intravesical chemotherapy Agent
Mechanism of action
Absorption
Toxicity
Response rate
Thiotepa
Alkylating agent Interferes with protein synthesis
Molecular weight: 189 Da Systemic absorption: high
Systemic: myelosuppression, rare and transient301 Local: chemical cystitis 12–69% Dependent on dose and instillation schedule302
Definitive treatment: 38%46 Adjuvant treatment: increased control rate of 17% over controls50 Other studies: no statistical difference92
Doxorubicin
Anthracycline antibiotic Binds to DNA
Molecular weight: 580 Da Systemic absorption: low
Systemic: very rare Local: chemical cystitis 20–30%303,304
Definitive treatment:46 papillary tumours: 28–56% CIS: CR 34% of patients, median time to treatment failure 5 months105 Adjuvant treatment: improved disease-free survival when given with MMC, but no difference between short-term and long-term schedules93
Mitomycin C Unknown mechanism (MMC) of action but produces a cytotoxic alkylating agent
Molecular weight: 329 Da Systemic absorption: low
Systemic: rare Local: chemical cystitis 6–41%302
Definitive treatment: papillary tumours:43%46 CIS: CR 58% of patients Adjuvant treatment: reduced recurrence rates versus control50,305 Short-term adjuvant therapy (20 instillations of 20 mg MMC in 20 weeks) is as effective as maintenance therapy306
Epirubicin
Less toxicity than doxorubicin
Systemic: rare Local: chemical cystitis 14% for 8 weekly instillations307 Chemical cystitis 14% for single instillation96
Definitive treatment: EORTC Study 30869: 56% CR in marker lesions (117) Adjuvant treatment: EORTC Study 30863: single instillation of epirubicin significantly reduces recurrence96
Anthracycline antibiotic Binds to DNA
CIS, carcinoma-in-situ; CR, complete response; MMC, mitomycin-C; Da, Daltons
Table 26.5 Current guidelines for the treatment of patients with superficial bladder cancer Ta/T1 Ta/T1
Grades 1–2 Grades 1–2
Tis T1
Grade 3
Single; treatment of mitomycin-C post-TURBT Single or multiple recurrences: resect all tumours possible Intravesical chemotherapy for: (i) first relapse 6/12 from original disease, (ii) multiple relapses, (iii) frequent relapses e.g. single treatment mitomycin-C for solitary recurrence 6/12 from original disease A course of mitomycin C (40 mg/40 mL) weekly 6 or BCG weekly 6 Intravesical BCG (consider maintenance) or cystectomy Intravesical BCG or cystectomy
TURBT; BCG, bacilli Calmette–Guérin.
regime is to treat with mitomycin-C (40 mg in 40 mL saline) weekly for 6 weeks (side effects allowing). One potential problem with mitomycin-C is variable drug delivery. Au et al.99 addressed this by using pharmacokinetic manipulations; essentially decreasing urine volume
(by pre-treatment dehydration and reducing administration volume) and urine alkalinization (by oral sodium bicarbonate) to stabilize the drug. Compared to standard treatment, patients receiving optimized treatment had a longer time to recurrence (median 29.1 months versus 11.8 months)
624 Bladder cancer
and recurrence-free fraction (41 per cent versus 24.6 per cent). An alternative technique, electromotive drug delivery, is externally to apply a potential difference across the bladder wall by a pulsed electric current. This has been reported to improve drug delivery such that peak plasma concentrations were 5.5 times higher compared to passive diffusion and increased time to recurrence in one small randomized trial.100 The use of bacille Camette–Guérin (BCG) in low-grade superficial disease is controversial. A meta-analysis of 585 patients has shown that BCG reduces the risk of recurrence versus TURBT alone,101,102 and a further meta-analysis showed that BCG was more effective (and more toxic) than mitomycin-C in intermediate-risk and high-risk Ta and T1 tumours (defined according to Kurth et al., high tumour grade 3 cm or previous recurrence) but not in lower-risk disease.103 This analysis found no reduction in progression for BCG compared to mitomycin-C, although a subsequent meta-analysis suggested that there is an odds ratio of 0.67 (95% confidence interval (CI) 0.47–0.94) for progression, but only if maintenance BCG is used.104 As the majority of this patient group have a low risk for progression and recurrence, it would seem reasonable to use intra-vesical chemotherapy as first-line treatment and reserve BCG for patients with poor-risk disease and a high risk of progression (including high grade and CIS), or for chemotherapy failures. SUPERFICIAL HIGH-GRADE TUMOURS (CIS AND T1)
High-grade T1 tumours are considerably more aggressive than their lower-grade counterparts, with over one-third of patients progressing to invasive disease within a few years. The outcome is similar to that of CIS, which in 90 per cent of cases is found in association with visible bladder tumours. The lesion may be focal but is usually diffuse, with only 10 per cent occurring as isolated pathological lesions. Even then, 20–34 per cent have concurrent micro-invasive carcinoma at cystectomy50 and are at risk of developing muscleinvasive disease in 42–83 per cent of cases. The CIS may not be visible endoscopically, and even if lesions are seen, they are often too extensive to resect and have ill-defined margins. This, in association with the frequency of concomitant invasive carcinoma and high risk of progression to invasive disease, has resulted in a historical preference for cystectomy. However, the results of immediate cystectomy are not superior to those of cystectomy performed after failure of intra-vesical treatment,105 which has led to more widespread adoption of an organ-preserving approach. Carcinoma-in-situ is the optimal target for intra-vesical therapy as close contact between agent and the tumour can be achieved. Bacille Calmette–Guérin is an active intra-vesical immunotherapy agent for bladder cancer, although its precise mechanism of action is not well understood. However, it is known that direct contact between tumour cells and BCG is necessary and that T lymphocytes are required for BCGmediated anti-tumour activity.106 It is likely that BCG activates dendritic cells (DCs) in a non-specific manner and that
potent anti-tumour Th1 responses are generated. A significant reduction in tumour recurrence and progression is noted in most prospective controlled studies comparing BCG with TUR alone.102 Nine trials have compared BCG and intra-vesical chemotherapy, of which four (3/6 v MMC and 1/3 v adriamycin/epirubicin) have demonstrated the superiority of BCG. Individually, these trials are too small to demonstrate a benefit for BCG over intra-vesical chemotherapy in reducing progression rates,107 but a metaanalysis of these studies shows that, compared to intra-vesical treatment, BCG improves complete response rate (68 per cent versus 51.5 per cent) and long-term control (46.7 per cent versus 26.2 per cent, odds ratio 0.41, p 0.0001). Patients on BCG with maintenance had an OR of 0.57 (p 0.04) for progression.108 Adjuvant therapy with BCG is frequently associated with significant toxicity, which often results in early cessation of treatment. In an attempt to reduce toxicity and improve efficacy compliance, several studies of sequential chemotherapy and BCG versus BCG alone109 or mitomycin-C alone110,111 have been performed, which demonstrated comparable efficacy and superiority in terms of toxicity for the alternating regimen.112 However, a more recent study has compared BCG alone with BCG plus electromotive mitomycin, with both arms receiving maintenance treatment. Patients receiving combination treatment had improvements in most parameters of efficacy, including recurrence rates, progression rates, overall and disease-specific survival.113 This may be an important development but will require repetition before being adopted into routine practice. Primary CIS (no previous TCC), secondary CIS (occurring after previous TCC) and concurrent CIS (associated with superficial TCC) respond similarly to BCG treatment, and today BCG is recommended as first-line therapy for this disease. The prescription of BCG is dependent on the strain used, with different doses of each required for optimum effect. No consensus exists on the optimum schedule and strain to use. Common practice in the UK would be to use 50 mg of the Tice strain or 120 mg of the Connaught strain, given weekly for 6 weeks. The need for subsequent maintenance has been controversial. A Southwest Oncology Group (SWOG) trial of ‘booster’ treatment with repeated 3-week courses at 3 months, 6 months and 6 monthly for 3 years initially showed superior results to a single 6-week course.114 This approach has been supported by the greater effect on progression in the metaanalyses discussed above104,108 and should probably now be regarded as the best way to utilize BCG. It has to be accepted that this is at the cost of significantly increased toxicity. The alternative is to repeat the 6-week course at time of treatment failure, as tested by Catalona et al.115 Bacille Calmette–Guérin has a profound inflammatory reaction on bladder mucosa and results in more pronounced local toxicity than chemotherapy. Serious systemic symptoms are uncommon but do occur, and as many as 25 per cent of patients have an influenza-like syndrome lasting between 12 and 24 hours after installation. Those patients suspected of contracting systemic tuberculosis are usually successfully
Management of invasive bladder cancer 625
treated with anti-tuberculous therapy. However, following intra-vesical treatment, a number of deaths have been reported related to bladder trauma, either secondary to traumatic instillation or when the instillation has occurred immediately after TURBT. Some investigators have advised the use of prophylactic isoniazid prior to BCG treatment. The interim results of an EORTC study (Protocol 30911) of this approach have shown no benefit in terms of local or systemic side effects116 and the treatment was associated with transient liver dysfunction, so prophylactic therapy with isoniazid is not recommended.117 The use of BCG is contraindicated in immunocompromised patients such as those with HIV infection, patients undergoing immunosuppressive therapy, patients with coexistent malignancy such as leukaemia or Hodgkin’s disease, and pregnant or lactating women.117 Intra-vesical BCG produces high and durable response rates in CIS, but 30 per cent of patients fail to respond to firstline treatment, while a further 30 per cent achieving a complete response with initial therapy relapse within 5 years. Overall, only 31 per cent of patients treated with BCG for CIS remain tumour free at 10 years.118 A considerable proportion of the failure is related to extra-vesical progression of disease, either to the lower ureters or prostatic urethra and duct. In the literature, reports of prostatic urethral involvement after BCG range from 1.5 per cent to 6.3 per cent, which is lower than that for chemotherapy (33–37 per cent).119 The treatment of refractory CIS that incompletely responds or recurs after BCG is debatable. Generally, immediate cystectomy is considered the treatment of choice, especially when CIS is associated with high-grade lamina propria invasive disease (T1G3), a scenario with a more substantial risk for progression after conservative measures. However, many patients with relatively asymptomatic CIS may prefer to retain normal bladder function even after relapse following BCG; the fact that progression usually occurs after 2 years means that in many cases a trial of second-line therapy is reasonable.120 Reports suggest that a proportion of such patients respond to a second course of BCG,121 especially if maintenance has not been used.115 The addition of interferon to BCG, techniques to improve intra-vesical drug delivery such as electromotive mitomycin-C and intra-vesical hyperthermia/chemotherapy, new intra-vesical agents such as gemcitabine, and photodynamic therapy are all being tested in this setting.122,123 A recent study of intra-vesical gemcitabine showed a high but mainly transient response rate.124 RADIOTHERAPY AND SYSTEMIC TREATMENTS FOR SUPERFICIAL DISEASE
Radiotherapy is usually considered ineffective in superficial disease because although it is effective in eradicating the primary, it has little impact on preventing recurrence. However, there has been some experience with T1G3 disease. Rodel et al. treated 74 patients with T1G3 tumours with radiotherapy concomitant chemotherapy (median dose
54 Gy, 45–60 Gy).125 A complete remission at restaging was achieved in 83.7 per cent, 47 per cent remained disease free, with 77 per cent of surviving patients retaining their bladder at 5 years. Its role in this context has been tested in a randomized MRC trial in which radiotherapy was compared to TUR alone (solitary disease) or BCG (multiple tumours). Preliminary analysis showed no advantage for radiotherapy (hazard ratio 1.16 (in favour of control), 95% CI 0.7–1.92, p 0.552). Further analysis of these data is awaited. It would suggest that radiotherapy should not be used routinely for T1G3 disease but may be an alternative for those who have failed other conservative treatments and are unable to undergo cystectomy. As intra-vesical treatment is an attractive alternative for superficial disease, little exploration of systemic therapies in the prevention of recurrence has been undertaken. Two trials of oral systemic chemotherapy, methotrexate127 and UFT,128 have shown reduced recurrence rates but have not gained acceptance due to concerns regarding safety and systemic toxicity. Epidemiological evidence suggesting increased bladder cancer risk in populations with low vitamin A levels and laboratory data demonstrating growth inhibition of retinoids129 led to the inception of two small randomized trials of retinoids as secondary prevention of bladder cancer, both of which showed an advantage in terms of recurrence rates. A further small study of high-dose vitamins, including vitamin A, had similar results, with a 40 per cent reduction in recurrence when used in addition to BCG.130 As highlighted above, COX2 over-expression has been associated with bladder cancer, with increased expression with stage and grade. Two case-control studies have shown a reduction in the risk of developing bladder cancer for patients on non-steroidal anti-inflammatory drugs (NSAIDs, odds ratio 0.81) and a reduction in tumour recurrence (cumulative tumour recurrence at 1 year of 30 per cent if taking COX2 inhibitor or NSAID versus 61.5 per cent if not).131,132 These data suggest that COX2 inhibition may have a role in preventing recurrences. It has also been suggested that COX2 inhibition may enhance the efficacy of BCG.133 The use of COX2 inhibition is due to be tested in a UK randomized trial to start in 2007 (BOXIT trial). Finally, the development of inhibitors of angiogenesis and invasion inhibitors such as matrix metalloproteinase inhibitors may be a fertile area for future research. The current guidelines for the treatment of patients with superficial bladder cancer are outlined in Table 26.5.
MANAGEMENT OF INVASIVE BLADDER CANCER Prognostic factors Stage and grade remain the most important prognostic factors for invasive disease, with the presence of extra-vesical extension101 and lymph-node positivity134,135 having the
626 Bladder cancer
SURGICAL TREATMENT
Radical cystectomy remains the treatment associated with highest local cure, with pelvic recurrence rates of less than 10 per cent in node-negative tumours and 10–20 per cent in patients with resected pelvic nodal metastases.148–150 A review comparing outcomes of surgery in National Cancer Institute (NCI) cancer centres (n 2977) with those in other American hospitals (n 2566) included patients over 65 years of age on the Medicare database treated for cancer by cystectomy in cancer centres between 1994 and 1999.151,152 As illustrated in Figure 26.3, there was no significant difference between the two hospital groups, with 5-year survival of approximately 38 per cent. In the past, cystectomy was associated with significant morbidity and high mortality, but improvements in operative technique have seen the perioperative complication rate fall from
Cystectomy Observed Survival (Proportion)
greatest impact on survival. The extent of lymph-node involvement is also of prognostic significance, with survival decreasing with increasing number of positive nodes.135 A number of potential biological prognostic markers have also been investigated in invasive bladder cancer, including the T138 antigen and the proliferative markers Ki-67 and proliferating cell nuclear antigen (PCNA). However, although expression may correspond with tumour aggressiveness, their role in routine clinical practice awaits prospective evaluation in larger groups of patients.136–139 Useful prognostic information may also be gained by measuring changes in the expression of molecules thought to be associated with invasion and metastasis. These include cellular adhesion molecules such as cadherins140 and integrins,141 angiogenic stimulators such as the FGFs142 and epidermal growth factor (EGF),143 and angiogenic inhibitors such as thrombospondin-1 (TSP-1) and angiostatin.144 For instance, reduced expression of E-cadherin and loss of α-2-integrin expression have been associated with increasing stage, and preliminary data suggest a relationship between E-cadherin expression and prognosis,145 although further studies are required to confirm these findings. As both tumour growth and invasion are dependent on the process of new vessel formation (angiogenesis), the ability to quantitate angiogenesis within a tumour may also provide important prognostic information. To this end, ‘micro-vessel density’ (MVD) has been determined histochemically using antibodies to vascular endothelial cells, such as antibodies to Factor VIII and CD34. Furthermore, a positive correlation has been demonstrated between MVD, lymph-node metastasis 39 and tumour progression.146 Finally, reduced expression of the angiogenic inhibitor TSP has been significantly correlated with increased tumour recurrence and decreased survival.147 However, although these markers have provided insights into the biological behaviour of tumours, an inconsistent relationship exists between them and patient prognosis. Further studies are therefore required to clarify their role in the clinical setting.
1.00 0.75 0.50 0.25 0.00 0
1
2
3
4
5
Years
Figure 26.3 Overall survival after cystectomy for bladder cancer in US centres. (From Birkmeyer et al., 2005, with permission).
approximately 35 per cent prior to 1970 to less than 10 per cent reported currently, with a corresponding fall in operative mortality from nearly 20 per cent to 3 per cent.151,152 The operation involves en-bloc removal of bladder, prostate and seminal vesicles in the male, or bladder, urethra, uterus, cervix, fallopian tubes, ovaries and anterior vaginal wall in the female with surrounding perivesical fat, pelvic visceral peritoneum and lymph nodes. Following cystectomy, the urine is diverted into an incontinent stoma, a continent urinary reservoir or an orthotopic bladder substitute, which allows the patient to void urethrally. The standard against which other techniques are compared is the ileal conduit, which involves formation of a ureteroileal anastomosis and is associated with a lower incidence of long-term metabolic disturbance and renal deterioration than its predecessor, the ureterosigmoidostomy. However, it is associated with significant physical and psychological morbidity and is therefore being superseded by continent diversions or orthotopic bladder substitutes. The continent urinary diversion is an intra-abdominal urinary reservoir, which is catheterizable or has an outlet controlled by the anal sphincter. The reservoir is usually fashioned from stomach, ileum or part of the large bowel, with formation of some form of mechanism to prevent reflux of urine to the kidneys. Probably the best studied ileal continent diversion is the Koch pouch,153 which uses intussuscepted ileum and has a continent catheterizable stoma. More recently, this technique been adapted for use as an orthotopic bladder substitute.154 Here, a bladder is constructed using loops of ileum and re-anastomosed onto the urethra. The drawback of this approach is the longer operating time, with high complication and re-operation rates and a high incidence of urinary incontinence. This, in combination with the age and performance status of the average bladder cancer patient, means that this procedure is only suitable in a minority of cases. It has been suggested that orthotopic bladder substitutes have a higher risk of chronic renal damage than an ileal conduit.155 Other more recent technical developments include the development of a nerve-sparing technique156 in the male,
Management of invasive bladder cancer 627
which has resulted in the preservation of sexual potency,157 and the urethra-sparing procedure in the female.158 By using a selective approach to urethrectomy, continence should be possible with an orthotopic bladder replacement providing meticulous follow-up is employed to ensure the early detection of recurrence. However, urethral preservation should be avoided in patients with multiple papillary tumours, tumours involving the bladder neck or posterior urethra159 and in the presence of CIS. Bladder-preserving surgery results in the retention of physiological bladder function, continence, potency and the ability to sample regional pelvic lymph nodes. Partial cystectomy alone is only suitable for a minority of patients with invasive bladder cancer. Suitable patients have a solitary muscle-invasive primary that is amenable to complete excision and biopsy, proven absence of cellular atypia, or CIS, in the remaining bladder. High recurrence rates (38–78 per cent) have been reported in unselected cases,160 but 5-year survival figures comparable to those of contemporary series of radical cystectomy can be achieved in carefully selected patients.160 Bladder preservation rates of 53–77 per cent with an overall survival of approximately 80 per cent have been reported following radical TUR alone.161–163 However, these were small studies in highly selected patients. Kondas and Szentogyorgyi164 have suggested, following analysis of their series of 761 patients, that the following criteria be applied in the selection of patients for curative TUR. The primary could be of any grade but should be solitary and organ confined, with a maximum diameter of 2–3 cm at the base and located at a fixed portion of the bladder. The completeness of resection is clearly an important factor in achieving control of the disease, which may be assessed by marginal and random biopsies following resection. Pathological confirmation of tumour clearance is also provided by negative urine cytology at 10 days and absence of disease at a secondary resection performed several weeks after the initial TUR.165–167 Although by following these guidelines for bladder conservation high long-term control rates can be achieved, it is probably best to reserve such approaches for patients who are not fit enough for or are unable to have local radical treatment. Local control of radical TUR may be improved by the use of chemotherapy168,169 and/or radiotherapy in combination with radical TUR and is discussed below. RADIOTHERAPY IN THE MANAGEMENT OF MUSCLE-INVASIVE BLADDER CANCER
Radiotherapy has been used in the successful treatment of bladder cancer, with reported overall 5-year survival figures ranging from 20 per cent to 40 per cent.170–176 However, there is still resistance to bladder preservation using radiotherapy in some centres, related to the lack of well-executed trials comparing radical radiotherapy to cystectomy and apparent lack of survival benefit for surgery in the salvage setting.177–179 Comparisons between institutional studies and non-randomized data have consistently shown a higher
local control rate for cystectomy and have led to fears of disease progression prior to salvage surgery, with a possible compromise in survival. However, when the effects of selection bias (e.g. fitter patients being selected for surgery), stage migration due to clinical versus pathological staging and differences in prognostic factors between patients selected for radiotherapy or surgery are taken into consideration, there is little evidence from the current data that overall survival is compromised.180 If relapse does occur after radiotherapy, the patient will require surgery in addition to radiotherapy, and this post-radiation cystectomy is a more difficult operation and associated with greater morbidity with little possibility of reconstructive surgery. All patients undergoing attempted bladder-preserving treatment should be counselled accordingly before commencing treatment. Despite these problems, the increasing interest in selective bladder preservation is likely to result in the more widespread use of radiotherapy as the primary treatment modality internationally.
Case selection for radiotherapy In deciding to attempt bladder preservation using radiotherapy, careful case selection is required as there are several situations in which primary cystectomy may be preferred. This is clearly the case for patients who have had previous pelvic radiotherapy. Another important group is that of patients whose disease has resulted in irretrievable loss of bladder function, when there is little benefit from bladder preservation. Radiotherapy does not improve incontinence or the capacity of a bladder damaged by previous interventions, and will exacerbate symptoms arising from a severely irritable bladder. Patients with active inflammatory bowel disease, extensive prior pelvic surgery or chronic pelvic infections are at high risk of severe bowel complications following radiotherapy. In these cases, alternative treatment should be carefully considered. Surgery is preferred when tumours arise in bladder diverticulae, where there is increased risk of local failure as a result of both potential difficulties in patient set-up and inhomogeneity of dose distribution, but also due to the problem of adequate cystoscopic follow-up of the region. Finally, patients whose pattern of disease suggests a low possibility of control with radiotherapy alone should be considered for alternative therapies. This includes those who have undergone multiple resections for recurrent superficial tumours or multiple courses of chemotherapy or BCG, patients with diffuse malignant involvement, multiple tumours, large tumours with extra-vesical masses 5 cm and tumours of squamous or adenocarcinoma histology. There are conflicting views as to the influence of coexistent CIS to local tumour control following radiotherapy.172,181–183 Recent reports of the use of intra-vesical BCG for persistent or recurrent CIS post-radiotherapy suggest that high response rates can be achieved,184 but long-term follow-up is required to assess the true potential of this treatment.
628 Bladder cancer
RADIOTHERAPY TECHNIQUE
PROGNOSTIC FACTORS FOR LOCAL CONTROL AFTER RADIOTHERAPY TREATMENT
A favourable response to radiotherapy may be expected in small-volume T2 (rather than T3/4 patients), in the absence of ureteral obstruction,193–195 in papillary rather than sessile tumours,194,195 with normal haemoglobin levels above 13 g196–199 and following a good response to the first 40 Gy
% probability of survival
T1 T2 T3 T4
80 60 40 20 0 0
1 2 3 4 Time since RMH presentation (years)
5
Figure 26.4 Survival according to T stage for patients with bladder carcinoma treated at the Royal Marsden Hospital between 1990 and 1999.
100 % probability of survival
Current radiotherapy technique utilizes CT-scan-associated planning to visualize the bladder. The target volume comprises the empty bladder (to minimize the irradiated volume) and any extra-vesical disease with a 1.5–2 cm margin (to allow for microscopic disease and organ movement). In most instances, an anterior and two lateral treatment fields encompass this. The pelvis is not routinely included in these treatments as this substantially adds to treatment toxicity without any clear evidence for improved pelvic control. The toxicity of treatment is cumulative and related to the dose and volume of normal tissue irradiated. The predominant acute symptoms are radiation cystitis associated with urinary frequency, urgency and dysuria, proctitis and lethargy. In the long term, bladder function may deteriorate as a result of organ shrinkage related to fibrosis. Superficial telangiectasia in the bladder may give rise to haematuria, and late bowel damage may result in bleeding, which may occasionally be profound and require operative intervention. Impotence may also occur, although the precise incidence is not well documented. There is a clear dose–effect relationship for radiotherapy, as demonstrated by the down-staging of the operative specimen after preoperative radiotherapy185–187 and improved local control and patient survival related to the total dose delivered to the primary.188,189 On this basis, doses in the region of 64 Gy in 2-Gy fractions are used at the Royal Marsden. Elsewhere in the UK, shorter fractionation schemes, such as 55 Gy in 20 fractions, are used, which, although not tested against standard fractionation schedules, are thought to be of equivalent efficacy. The results of radical radiotherapy at the Royal Marsden Hospital are shown in Figures 26.4 and 26.5 in terms of survival according to stage and lymph-node status. A prospective randomized trial has failed to show any benefit from accelerated fractionation.190 Significant proportions of patients are too frail or old to tolerate a standard course of radiotherapy and in such patients local control can be achieved by using weekly 6-Gy fractions to 30–36 Gy.191 This regime appears well tolerated, with acceptable levels of late toxicity. This sort of schema can also be useful for the control of local symptoms in patients with metastatic disease. A similar regime of 21 Gy in three fractions in1 week has been tested in this setting and 35Gy in ten fractions in a multi-centre MRC trial of 500 patients. The results demonstrated no difference in efficacy, toxicity or survival between the two arms.192
100
Node ve Node ve
80 60 40 20 0 0
1 2 3 4 Time since RMH presentation (years)
5
Figure 26.5 Survival according to nodal status for patients with bladder carcinoma treated at the Royal Marsden Hospital between 1990 and 1999.
of radical radiotherapy.69,195 De-bulking of the tumour prior to radiotherapy also improves local control, both when the radiotherapy is delivered as the sole modality181 and when it is administered concurrently with cisplatin.200 However, the excellent local control rates may be partly attributed to case selection (i.e. tumours are small enough to be de-bulked) rather than to the surgery itself. IMPROVING THE RESULTS OF RADICAL RADIOTHERAPY
A number of strategies are being explored in an attempt to improve the efficacy of radiotherapy. Following the observation that treatment breaks are associated with impaired outcome and that bladder cancer has a short potential doubling time, accelerated fractionation (AF) has been explored. However, a recently completed randomized trial of 60.8 Gy in 32 fractions over 4 weeks treating twice daily versus a standard treatment of 64 Gy in 32 daily fractions over 6.5 weeks has failed to show benefit from AF.190 The trial was limited in power, as only 229 patients were randomized. The overall survival probability at 5 years was 37 per cent for AF versus 40 per cent for conventional fractionation
Management of invasive bladder cancer 629
(CF). Late radiation toxicity equivalent to RTOG grade 2 or more had occurred in 44 per cent of AF patients versus 38 per cent after CF. Thus it was considered that CF should remain the standard of care. Radiation resistance may be related to tumour hypoxia. There are some data that suggest that reversal of this using hyperbaric oxygen can improve local control,198 but the difficulties of implementing the technique in clinical practice have prevented its routine use. More recently, carbogen (95 per cent O2, 5 per cent CO2) in combination with nicotinamide has been used to improve tumour oxygenation, with promising results in pilot studies.201 Hoskin et al.202 have evaluated 105 patients treated with radiotherapy to 50–55 Gy over 4 weeks, with either carbogen alone or carbogen plus nicotinamide (ARCON). In both treatments at 3 years, local tumour control and survival were 53 per cent and 43 per cent respectively, but approximately 25 per cent of patients had urinary frequency at least twice per hour. This technique is currently being tested in a phase III randomized trial. Increased radiation dose could improve local control, but the challenge is to maintain acceptable local morbidity. One way to achieve this is by using interstitial radiotherapy. This has been explored by several groups, most frequently in combination with TUR and low-dose external-beam radiotherapy (usually up to 30 Gy).203–207 In most of these series, impressive local control rates in excess of 80 per cent are reported, with acceptable toxicity levels, although the impact of case selection on these results is difficult to assess. The need for the surgical placement of the implant has restricted its use to a few specialist centres. The dose of external-beam radiotherapy is limited by the tolerance of normal tissues, and in the bladder retrospective data suggest that radiation tolerance is a function of the volume irradiated; the tolerance of part of the bladder greater than that of the whole bladder. The interstitial radiotherapy data have shown that satisfactory local control can be achieved by treatment focused on the tumour (compared to treating the whole bladder). Reducing the volume of bladder irradiated could reduce toxicity, allowing the overall radiation dose to be increased, which should improve control rates. The potential problems of this approach arise from the fact that the bladder is not a fixed organ, so its location and size may vary. Studies comparing the CT-planned target volume before and during radiotherapy demonstrate that a shift of target volume may occur.208 In one study, movement of the bladder wall was observed in 60 per cent patients, and although movement tended to be towards the midline and hence into the target volume, a reduction in the margin around the tumour was seen in 33 per cent of patients.209 Thus adequate margins around the tumour are required to ensure that geographical misses do not occur. Despite these problems, a small pilot study at the Royal Marsden Hospital with a modest decrease in the whole-bladder dose has shown significant reduction in toxicity and may form the basis of more extensive prospective evaluation of this concept.
Combined modality treatment SELECTIVE BLADDER PRESERVATION
An approach to seeking the advantages of bladder preservation with the efficacy of cystectomy has been to select for preservation using the criterion of initial response, an important indication of prognosis. When survival data from 125 patients treated with neoadjuvant chemotherapy and cystectomy were analysed, 91 per cent of down-staged patients survived, as opposed to only 37 per cent of those with persisting muscle-invasive disease.210 This approach was developed by Shipley,211 whose protocol required complete TUR followed by neoadjuvant and concomitant platinum chemoradiation to a limited dose. At this time, cystoscopic reassessment is undertaken and patients in remission are selected for complete radiotherapy and organ conservation, whereas those with persisting tumour have cystectomy. A substantial number of patients are able to retain their bladders and have good urinary function.212 The Radiotherapy and Oncology Group (RTOG) has explored this further in a sequence of studies.213 For example, in RTOG 97-06,214 52 patients entered a selective conservation protocol in which the initial phase of therapy was aggressive TURBT and accelerated concomitant boost radiotherapy sensitized by cisplatin. The complete response rate was 74 per cent and these patients had chemo-radiotherapy consolidation. All patients were intended to have adjuvant combination chemotherapy. At 36 months, the loco-regional failure rate was 27 per cent and the overall survival was 61 per cent. Forty-eight per cent of patients survived with intact bladder. The success of this approach suggests that it merits formal comparison with the policy of immediate cystectomy.180 PREOPERATIVE AND ADJUVANT RADIOTHERAPY
The rationale for the use of preoperative radiotherapy is to prevent the intra-operative seeding of tumour cells, to sterilize microscopic tumour deposits in the perivesical tissues and theoretically to eradicate tumour that has not been rendered hypoxic secondary to surgically mediated changes in the surrounding vasculature. A number of retrospective reviews and non-randomized studies have suggested a benefit for preoperative radiotherapy over cystectomy alone.189,215,216 However, the few randomized trials conducted showed no survival advantage of preoperative radiotherapy, although they suffered from poor accrual and had insufficient power to detect even a large treatment effect.217–219 The SWOG trial did not demonstrate a benefit for preoperative irradiation and cystectomy. Despite being the largest study to date, fewer than 150 patients were randomised, so the power of the study was limited and the confidence intervals wide.220,221 Several other factors may have contributed to this apparent lack of effect: the study used a short course of radiotherapy (20 Gy in five fractions over 1 week), and a large proportion of the patients had T2 tumours, which may not benefit from preoperative radiotherapy. Finally, the known difference in relapse between deeply muscle invasive (T2b) and
630 Bladder cancer
extra-vesical (T3b) tumours was concealed by considering these patients as one group. It is probably best to conclude that the case for preoperative radiotherapy is not proven but that it may be an area for continued research. Limited examination of postoperative treatment has also been undertaken, but is associated with a high risk of bowel toxicity.222
Neoadjuvant chemotherapy Neoadjuvant chemotherapy aims to treat micrometastatic disease present at the time of initial diagnosis. Trials are summarized in Table 26.6, and although many failed to show benefit, the largest three have suggested a survival advantage. The largest trial of neoadjuvant combination chemotherapy has been the EORTC/MRC study of three cycles of cisplatin, methotrexate and vinblastine (CMV) followed by either radical radiotherapy or cystectomy; 976 patients were randomized.223 Cystectomy was planned in 484 patients, radio-therapy in 414 and combined treatment in 77. An updated analysis224 with median follow-up of 7 years has shown that neoadjuvant chemotherapy was associated with improvement in metastasis-free survival (54 per cent versus
45 per cent, p 0.002) and in overall survival (56.5 per cent versus 50.0 per cent, p 0.03). The Nordic Collaborative Group reported an overall analysis of their two neoadjuvant trials225 based on 620 patients treated between 1985 and 1997. Platinum was combined with adriamycin in the first and with methotrexate in the second trial. In the first trial, preoperative radiotherapy (20 Gy in five fractions) was used in both arms. The combined study results showed a hazard ratio of 0.80 (95% CI 0.64–0.99) for overall survival in favour of neoadjuvant treatment. The phase III Intergroup Trial in the USA evaluated three cycles of neoadjuvant M-VAC chemotherapy.226 This trial, coordinated by SWOG with participation of the ECOG and CALGB, enrolled 307 eligible patients, randomizing 154 to cystectomy alone and 153 to cystectomy after neoadjuvant M-VAC. Thirty-eight per cent of patients on the chemotherapy arm had no residual disease in the cystectomy specimen (15 per cent if no chemotherapy was given). The median survival for patients receiving chemotherapy was longer, at 77 months compared with 46 months, which was statistically significant on unstratified two-sided log rank analysis (p 0.05) and of borderline significance after stratifying for age and tumour stage (p 0.06). This equated to a
Table 26.6 Results of trials of neoadjuvant chemotherapy versus observation Trial organization or country
Number of patients
Primary tumour management
United Kingdom308
376
Spain309
Chemotherapy survival benefit
Treatment arms
Survival rate (%)
RT, cystectomy or both
Methotrexate Observation
39 37
No
122
Cystectomy
Cisplatin Observation
41 41
No
Australia310
225
RT
Cisplatin Observation
39 40
No
Nordic I311, 312
325
RT and cystectomy
Cisplatin/doxorubicin Observation
59 51
Trend T3/T4 disease
Italy (Genoa)313
184
Cystectomy
Cisplatin/5-FU/RT Observation
NA
No
Italy (GISTV)314
171
Cystectomy
MVEC Observation
85 74
No
Italy (Guone)315
230
Cystectomy
MVAC Observation
NA
No
MRC/EORTC316
976
RT or Cystectomy
CMV Observation
56 50
Yes
RTOG317
123
Cisplatin/RT
MCV Observation
48 49
No
Nordic II318
317
Cystectomy
Cisplatin/methotrexate Observation
53 46
No
SWOG227
307
Cystectomy
MVAC Observation
57 43
Yes (trend)
RT, radiotherapy; GISTV, Gruppo Italiano per lo Studio dei Tumori de la Vesicula; MRC, Medical Research Council; EORTC, European Organisation for Research and Treatment for Cancer; RTOG, Radiotherapy and Oncology Group; SWOG, Southwest Oncology Group; 5-FU, 5-fluorouracil; MVEC, methotrexate, vinblastine, epirubicin, cisplatin; MVAC, methotrexate, vinblastine, doxorubicin, cisplatin; CMV, cisplatin, methotrexate, vinblastine.
Management of advanced disease 631
disease-specific hazard ratio of 1.66 (95% CI 1.22–2.45, p 0.002).7 The international data on neoadjuvant chemotherapy were included in an individual patient data meta-analysis.7 This confirmed that multi-agent chemotherapy reduced the risk of death by about 13 per cent (hazard ratio 0.87, 95% CI 0.78–0.97, p 0.016), equating to a 5 per cent absolute improvement in survival at 5 years. An update of the metaanalysis based on 3005 patients in trials continued to show a 5 per cent absolute improvement in survival at 5 years and for platinum-based combination chemotherapy there was a significant disease-free survival benefit of 9 per cent.227
Concurrent chemotherapy and radiotherapy In an attempt to improve the therapeutic ratio of radiotherapy, concurrent chemotherapy has been investigated as a radiosensitizer. The most frequently studied drug in this context is cisplatin, as, in addition to its cytotoxicity, it has been shown to enhance the radiation effect in vivo, in vitro and under hypoxic conditions. A large number of phase II studies have been performed, the majority utilizing cisplatin alone or in combination with 5-fluorouracil (5-FU), which have demonstrated high complete responses, with bladder preservation in more than 60 per cent of patients.200,228–234 The National Cancer Institute of Canada has conducted a prospective trial examining the effect of the addition of concurrent cisplatin to preoperative or definitive radiotherapy on tumour control and overall survival.228 Ninety-nine patients had either radiotherapy or preoperative radiotherapy and cystectomy and were randomized to receive or not receive cisplatin 100 mg/m2 every 2 weeks during radiotherapy. Despite the lack of power of this small study to detect any but large benefits, it demonstrated improved recurrence-free survival for the chemotherapy group and a non-significant trend to improved overall survival. The absolute difference in pelvic relapse-free survival was 20 per cent (p 0.038). Work is needed to address the issue of the optimum dose schedule for cisplatin, e.g. is daily administration better? Cisplatin is not necessarily the ideal agent in this group of patients, who are often elderly and have renal impairment. There have been studies using drugs that are less dependent on renal function, such as 5-FU alone235 or in combination,236 which have also shown promise in the phase II setting, and this approach is now in phase III trial. A study of weekly gemcitabine with a radiotherapy schedule of 52 Gy in 20 fractions found doselimiting toxicity at a dose of 150 mg/m2.237 ADJUVANT CHEMOTHERAPY
Advocates of adjuvant chemotherapy indicate the benefits of immediate definitive therapy of the known localized disease and, in the setting of cystectomy, the ability to stage accurately and thus judge the indication for chemotherapy more precisely. Unfortunately, trials generally have been underpowered 238 and some have used
questionable techniques of analysis,135 so the value of adjuvant treatment remains uncertain. A recent meta-analysis included only 491 patients from six trials,227 thus the finding that adjuvant chemotherapy led to an absolute reduction of risk of death of 9 per cent at 3 years (11 per cent for trials using cisplatin-based combination chemotherapy) should not be regarded as reliable. A current EORTC trial is addressing this question in patients found to have pT3/pT4 or pN disease.
MANAGEMENT OF ADVANCED DISEASE Despite the recognized chemo-sensitivity and radiosensitivity of TCC of the bladder, more than 5000 deaths are attributable to the disease in the UK each year. Approximately 50 per cent of patients with muscle invasion ultimately succumb to metastatic disease. The commonest sites of metastasis include the regional lymph nodes, bone, lung, skin and liver and, less frequently, brain and meninges and the organs within the peritoneal cavity. The distribution of metastasis is important when considering treatment, as the site(s) of involvement correlate with prognosis, with improved survival in those with disease confined to the lymph nodes or skin, and with a substantially worse prognosis with liver and bone metastases.239 Consideration of these factors and other prognostic factors, especially performance status,240 is important when comparing the results of clinical trials in this setting.
Chemotherapy Systemic chemotherapy is the main active treatment for patients with metastatic disease. Single-agent chemotherapy yields objective responses of the order of 15–20 per cent, with complete responses of 5–10 per cent 241–251 but with limited response duration of only 4–6 months.244,247,252,253 The most active agents in these historical studies included cisplatin, doxorubicin, mitomycin-C and methotrexate and then the vinca alkaloids and 5-FU. The development of two-drug, three-drug and fourdrug combinations increased response rates and survival.241,243,244,247,248,253 Until recently, the best seemed to be the MVAC and CMV regimes, incorporating methotrexate, vinblastine and cisplatin (with or without doxorubicin),254–256 described in Table 26.7. For both these regimes, high response rates, including a significant complete response rate, have been reported. For instance, in the initial report on the MVAC regimen there was an overall response rate of 72 per cent, with 36 per cent complete response,255,256 although subsequent reports, including multi-centre studies, have shown that in an average patient population response rates of 40–50 per cent with a 10 per cent complete response rate were more common.252,257 Toxicities include significant rates of mucositis (40 per cent), renal toxicity (31 per cent) and neutropenic sepsis (20 per cent), with a toxic death rate of 4 per cent.
632 Bladder cancer
Table 26.7 Summary of the chemotherapy regimens used in the treatment of invasive bladder cancer CMV
Cisplatin 100 mg/m2 day 2 Methotrexate 40 mg/m2 days 1 and 8 Vinblastine 5 mg/m2 days 1 and 8 Cycle repeated every 21 days
MVAC
Methotrexate 30 mg/m2 days 2, 15 and 22 Vinblastine 3 mg/m2 days 2, 15 and 22 Doxorubicin 30 mg/m2 day 2 Cisplatin 70 mg/m2 day 2 Cycle repeated every 28 days
GC
Gemcitabine 1000 mg/m2 days 1, 8 and 15 Cisplatin 70 mg/m2 day 2 Cycle repeated every 28 days
The median survival of patients with metastatic uroepithelial cancer not treated with chemotherapy is in the order 3–6 months,258,259 which increases to around 9–12 months with treatment.258–260 Although trials of multi-agent chemotherapy versus best supportive care are not available, studies of more intensive versus less intensive chemotherapy, including MVAC versus cisplatin,252 MVAC versus CAP (cyclophosphamide adriamycin cisplatinum) and see Table 26.7261 and CMV versus MV,240 have shown better response rates, progression-free survival and, in some cases, overall survival and symptomatic control for the more intensive regimen. The reports of a significant complete response rate raised the possibility of an increase (up to 20 per cent) in long-term survival, but more recent analysis suggests survival at 5 years is unusual except in groups receiving chemotherapy for T4 or node-positive disease (who often receive additional radiotherapy).262,263 In an attempt to improve these results, several trials have addressed the impact of moderate increases in dosing, using colony-stimulating factors.264–269 In most cases, this led to an increased intensity of the chemotherapy delivered and reduced haematological and mucosal toxicity, but it is not clear whether there are improved response rates.264,267–269 In a moderate-sized EORTC trial, an accelerated MVAC schedule administered on a 2-week cycle supported by granulocyte colony-stimulating factor (GCSF) was compared to standard MVAC; 263 patients were randomized. The accelerated schedule was associated with more complete responses (21 per cent versus 12 per cent) and after a median of 7 years there was also benefit in long-term survival (24.6 per cent versus 13.2 per cent, borderline statistical significance).270 The need for therapeutic improvements has led to the examination of several new agents. Of particular interest have been the taxanes paclitaxel (Taxol) and docetaxel (Taxotere) and the nucleoside analogue gemcitabine. Early studies of paclitaxel in chemotherapy-naive patients reported response rates of 42 per cent, with complete responses in 27 per cent of patients.246 However, although follow-up studies have confirmed activity, response rates were significantly lower, particularly in previously treated patients.271,272 Docetaxel has also shown activity in previously untreated
patients with renal insufficiency273 and as second-line chemotherapy in patients who had failed post-cisplatin.274 Gemcitabine, an analogue of cytosine arabinoside, first demonstrated activity in bladder cancer in early-phase trials in Italy275 in previously treated patients, and follow-up studies have demonstrated single-agent response rates of approximately 30 per cent.276,277 In view of the activity of paclitaxel and gemcitabine, it is not surprising that a number of phase II studies of these agents in combination with cisplatinum/carboplatin and other active drugs have been performed.278–280 It appears that weekly paclitaxel with gemcitabine causes lung toxicity.281 A trial of the gemcitabine/cisplatin doublet (GC) versus MVAC282 led to a change in practice and has recently been updated.283 Four hundred and five patients with locally advanced or metastatic TCC were randomized, and overall survival was similar in both arms (median 14 months for GC and 15 months for MVAC, 5-year survival 13.0 per cent and 15.3 per cent respectively). There was less toxicity with GC, which has become the treatment of choice in many centres. A trial from the Hellenic Oncology Group of the docetaxel/cisplatin doublet showed it was inferior to MVAC.284 The Spanish Genitourinary Group found that the triplet combination of cisplatin/gemcitabin/paclitaxel gave a complete response of 26 per cent and an overall response rate of 76 per cent,285 and an EORTC trial of this combination versus GC has completed accrual. A preliminary report suggests equivalent activity.286 Other agents that have shown moderate activity have included ifosfamide (with response rates of 20 per cent in previously treated patients)251 and gallium nitrate,287 although both of these drugs have considerable treatment-related toxicity, making them poor candidates for use in bladder cancer. Many patients are either too frail or have inadequate renal function to tolerate cisplatin-based treatment. Several groups have explored the use of the cisplatin analogue carboplatin in view of its more favourable toxicity profile and less renal dependence. Most studies have confirmed activity of carboplatin, but in the few comparative studies there is a tendency for lower efficacy.288–290 However, when cisplatin usage is contraindicated, its use is a reasonable option, possibly in combination with drugs such as methotrexate or gemcitabine.
Radiotherapy and surgery in the palliation of advanced disease In patients with symptomatic advanced local disease, surgery and radiotherapy may be needed when conservative measures fail. When patients with urinary fistulae/leakage cannot be managed by simple conservative measures such as catheterization, supravesical bladder diversions (and occasionally cystectomy) can be an effective manoeuvre, particularly in patients of good performance status and reasonable life expectancy, to relieve the problems of bladder irritability and perineal excoriation caused by constant urinary leakage.291
References 633
A further distressing symptom of bladder cancer is haematuria, which may be profound and associated with clot retention. Palliative radiotherapy may be very effective in controlling the bleeding using hypofractionated regimes, although the optimal treatment protocol has not been definitively ascertained. Radiotherapy may also provide benefit for those patients with obstructive nephropathy, urinary incontinence and bone or brain metastases. In all cases the expected benefit of any intervention in the palliative setting must be carefully weighed against its morbidity. Although treatment with surgery and multi-agent chemotherapy may result in excellent palliation for some patients, it must be remembered that this may be at the expense of significant toxicity. Most are too frail, and simple conservative measures or palliative radiotherapy in the setting of multidisciplinary care may be most appropriate.
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
●
Bladder cancer is the fifth commonest cancer in the UK, causing 5400 deaths per year. The risk of developing bladder cancer is increased by exposure to environmental carcinogens (tobacco smoking, aromatic amines), drugs, radiation and factors leading to chronic irritation of the bladder. The majority of bladder cancers present with painless macroscopic haematuria and in the UK are largely transitional-cell carcinoma subtypes. Superficial (non-invasive) tumours carry a good prognosis and can usually be controlled by local resections. Recurrence of superficial disease is common following local treatment, especially if tumours are multiple or high grade. Risk of recurrence is reduced by intra-vesical chemotherapy or immunotherapy. Approximately 20 per cent of superficial disease presentations progress to invasive cancer. Progression is associated with high-grade disease. Progression rate may be reduced by BCG treatment. Invasive disease carries a worse prognosis. Radical cystectomy and radiotherapy (with salvage cystectomy for relapse or persistent disease) are alternative strategies with similar long-term survival outcomes. No clear advantages have been demonstrated for adding adjuvant radiotherapy to cystectomy. Neoadjuvant chemotherapy confers a survival benefit of about 5 per cent; further trials are needed to confirm the possible benefit of adjuvant chemotherapy. Metastatic disease carries a poor prognosis. Multi-agent platinum-based chemotherapy is useful palliative treatment and carries survival and quality-of-life benefits.
KEY REFERENCES Hall RR Clinical Management of Bladder Cancer. London: Arnold. Oxford University Press New York. 1999. International Collaboration of Trialists. Neoadjuvant cisplatin, methotrexate and vinblastine chemotherapy for muscleinvasive bladder cancer: a randomised controlled trial. Lancet 1999; 354:533–40. Lamm DL, Blumenstein BA, Crawford ED, et al. A randomized trial of intravesical doxorubicin and immunotherapy with bacille Calmette–Guerin for transitional-cell carcinoma of the bladder. N Engl J Med 1991; 325(17):1205–9. Pawinski A, Sylvester R, Kurth KH, et al. A combined analysis of European Organization for Research and Treatment of Cancer, and Medical Research Council randomized clinical trials for the prophylactic treatment of stage TaT1 bladder cancer. European Organization for Research and Treatment of Cancer, Genitourinary Tract Cancer Cooperative Group and the Medical Research Council Working Party on Superficial Bladder Cancer. J Urol 1996; 156(6):1934–40; discussion 1940–1941. Raghavan D. Progress in the chemotherapy of metastatic cancer of the urinary tract. Cancer 2003; 97(8 Suppl.):2050–5. Shipley WU, Zietman AL, Kaufman DS, Coen JJ, Sandler HM. Selective bladder preservation by trimodality therapy for patients with muscularis propia-invasive bladder cancer and who are cystectomy candidates – the Massachusetts General Hospital and Radiation Therapy Oncology Group experiences. Semin Radiat Oncol 2005; 15(1):36–41. Skinner D, Lieskovsky G. Management of invasive and high grade bladder cancer. In: Skinner DG, Lieskovsky G (eds), Diagnosis and Management of Genituourinary Cancer. Philadelphia: WB Saunders Company, 1988, 295–312.
REFERENCES 1 Office for National Statistics. Cancer Statistics registrations: Registrations of cancer diagnosed in 2001, England. National Statistics: London. 2004; Series MB1(No.32). 2 Anderson, Stephenson J. UKACR comparison of cancer registrations. Bladder Tumours Internal Report 2004; Report No.3. 3 Parkin D, Whelan S, Ferlay Jea. Cancer incidence in five continents. Lyon: International Agency for Research on Cancer Scientific publications; 1997. 4 McCredie M. Bladder and kidney cancers. Cancer Surv 1994;19–20:343–68. 5 Coleman M, Babb P, Damiecki P, et al. Cancer survival trends in England and Wales 1971–1995 deprivation and NHS Region. The Stationary Office. 1999. 6 Case RAM, Hosker ME, McDonald DB, et al. Tumours of the urinary bladder in workmen engaged in the manufacture and use of certain dyestuff intermediates in the British chemical industry. Br J Ind Med 1954;11:75–104. 7 Case RA, Hosker ME. Tumour of the urinary bladder as an occupational disease in the rubber industry in England and Wales. Br J Prev Soc Med 1954 Apr;8(2):39–50.
634 Bladder cancer
8 Decarli A, Peto J, Piolatto G, La Vecchia C. Bladder cancer mortality of workers exposed to aromatic amines: analysis of models of carcinogenesis. Br J Cancer 1985 May;51(5):707–12. 9 Rubino GF, Scansetti G, Piolatto G, Pira E. The carcinogenic effect of aromatic amines: an epidemiological study on the role of o-toluidine and 4,4-methylene bis (2-methylaniline) in inducing bladder cancer in man. Environ Res 1982 Apr;27(2):241–54. 10 Morrison AS, Ahlbom A, Verhoek WG, et al. Occupation and bladder cancer in Boston, USA, Manchester, UK, and Nagoya, Japan. J Epidemiol Community Health 1985 Dec;39(4):294–300. 11 Boyko RW, Cartwright RA, Glashan RW. Bladder cancer in dye manufacturing workers. J Occup Med 1985 Nov;27(11):799–803. 12 Hoover R, Cole P. Population trends in cigarette smoking and bladder cancer. Am J Epidemiol 1971 Nov;94(5): 409–18. 13 McLaughlin JK, Hrubec Z, Blot WJ, Fraumeni JF, Jr. Smoking and cancer mortality among U.S. veterans: a 26-year follow-up. Int J Cancer 1995 Jan 17;60(2):190–3. 14 Zeegers MP, Tan FE, Dorant E, van Den Brandt PA. The impact of characteristics of cigarette smoking on urinary tract cancer risk: a meta-analysis of epidemiologic studies. Cancer 2000 Aug 1;89(3):630–9. 15 Doll R, Peto R. Mortality in relation to smoking: 20 years’ observations on male British doctors. Br Med J 1976 Dec 25;2(6051):1525–36. 16 Bartsch H, Caporaso N, Coda M, et al. Carcinogen hemoglobin adducts, urinary mutagenicity, and metabolic phenotype in active and passive cigarette smokers. J Natl Cancer Inst 1990 Dec 5;82(23):1826–31. 17 Hartge P, Silverman D, Hoover R, et al. Changing cigarette habits and bladder cancer risk: a case-control study. J Natl Cancer Inst 1987 Jun;78(6):1119–25. 18 McCredie M, Stewart JH, Ford JM, MacLennan RA. Phenacetin-containing analgesics and cancer of the bladder or renal pelvis in women. Br J Urol 1983 Apr;55(2): 220–4. 19 Piper JM, Tonascia J, Matanoski GM. Heavy phenacetin use and bladder cancer in women aged 20 to 49 years. N Engl J Med 1985 Aug 1;313(5):292–5. 20 Travis LB, Curtis RE, Glimelius B, et al. Bladder and kidney cancer following cyclophosphamide therapy for non-Hodgkin’s lymphoma. J Natl Cancer Inst 1995 Apr 5;87(7):524–30. 21 Levine LA, Richie JP. Urological complications of cyclophosphamide. J Urol 1989;141:1063–69. 22 Pedersen-Bjergaard J. Carcinoma of the urinary bladder after treatment with Cyclophosphamide for non-Hodgkins lymphoma. The New England Journal of Medicine 1988;318:1028–88. 23 Boice JD, Jr., Engholm G, Kleinerman RA, et al. Radiation dose and second cancer risk in patients treated for cancer of the cervix. Radiat Res 1988 Oct;116(1):3–55. 24 Inskip PD, Monson RR, Wagoner JK, et al. Cancer mortality following radium treatment for uterine bleeding. Radiat Res 1990 Sep;123(3):331–44. 25 Badawi AF, Mostafa MH, Probert A, O’Connor PJ. Role of schistosomiasis in human bladder cancer: evidence of
26
27 28 29
30
31
32
33
34
35
36
37
38
39
40
41
42
association, aetiological factors, and basic mechanisms of carcinogenesis. Eur J Cancer Prev 1995 Feb;4(1):45–59. Soliman AS, Levin B, El-Badawy S, et al. Planning cancer prevention strategies based on epidemiologic characteristics: an Egyptian example. Public Health Rev 2001;29(1):1–11. Dolin PJ, Cook-Mozaffari P. Occupation and bladder cancer: a death-certificate study. Br J Cancer 1992;66(3):568–78. Kiemeney LA, Schoenberg, M. Familial transitional cell carcinoma. J Urol 1996;156(3):867–72. Goldgar D, Easton D, Cannon-Albright L, Skolnick M. Systematic population-based assessment of cancer risk in first-degree relatives of cancer probands. J Natl Cancer Inst 1994;86:1600–8. Hein DW. Acetylator genotype and arylamine-induced carcinogenesis. Biochim Biophys Acta 1988 Aug 3;948(1):37–66. Brockmoller J, Cascorbi I, Kerb R, Roots I. Combined analysis of inherited polymorphisms in arylamine N-acetyltransferase 2, glutathione S-transferases M1 and T1, microsomal epoxide hydrolase, and cytochrome P450 enzymes as modulators of bladder cancer risk. Cancer Res 1996 Sep 1;56(17):3915–25. Ketterer B. Protective role of glutathione and glutathione transferases in mutagenesis and carcinogenesis. Mutat Res 1988 Dec;202(2):343–61. Knowles MA. The genetics of transitional cell carcinoma: progress and potential clinical application. BJU Int 1999 Sep;84(4):412–27. Esrig D, Elmajian D, Groshen S, et al. Accumulation of nuclear p53 and tumor progression in bladder cancer. New England Journal of Medicine 1994;331(19):1259–64. Grossman HB, Liebert M, Antelo M, et al. p53 and RB expression predict progression in T1 bladder cancer. Clin Cancer Res 1998 Apr;4(4):829–34. Cordon-Cardo C, Zhang ZF, Dalbagni G, et al. Cooperative effects of p53 and pRB alterations in primary superficial bladder tumors. Cancer Res 1997 Apr 1;57(7):1217–21. Reznikoff CA, Belair CD, Yeager TR, et al. A molecular genetic model of human bladder cancer pathogenesis. Semin Oncol 1996 Oct;23(5):571–84. Mellon K, Wright C, Kelly P, Wilson Horne CH, Neal DE. Long-term outcome related to epidermal growth factor receptor status in bladder cancer. J Urol 1995; 153(3):919–25. Jaeger TM, Weidner N, Chew K, et al. Tumor angiogenesis correlates with lymph node metastases in invasive bladder cancer. J Urol 1995 Jul;154(1):69–71. Yoshida K, Sugino T, Tahara H, et al. Telomerase activity in bladder carcinoma and its implication for noninvasive diagnosis by detection of exfoliated cancer cells in urine. Cancer 1997 Jan 15;79(2):362–9. Mohammed SI, Knapp DW, Bostwick DG, et al. Expression of cyclooxygenase-2 (COX-2) in human invasive transitional cell carcinoma (TCC) of the urinary bladder. Cancer Res 1999 Nov 15;59(22):5647–50. Ristimaki A. Progress in the field of selective COX-2 inhibitors—William Harvey Research Conference. Drugs 2001 Nov;4(11):1252–3.
References 635
43 Shariat SF, Kim JH, Ayala GE, et al. Cyclooxygenase-2 is highly expressed in carcinoma in situ and T1 transitional cell carcinoma of the bladder. J Urol 2003 Mar;169(3):938–42. 44 Kim SI, Kwon SM, Kim YS, Hong SJ. Association of cyclooxygenase-2 expression with prognosis of stage T1 grade 3 bladder cancer. Urology 2002 Nov;60(5):816–21. 45 Hasui Y, Osada Y, Kitada S, Nishi S. Significance of invasion to the muscularis mucosae on the progression of superficial bladder cancer. Urology 1994 Jun;43(6):782–6. 46 Bouffioux C, Denis L, Oosterlinck W, et al. Adjuvant chemotherapy of recurrent superficial transitional cell carcinoma: results of a European organization for research on treatment of cancer randomized trial comparing intravesical instillation of thiotepa, doxorubicin and cisplatin. The European Organization for Research on Treatment of Cancer Genitourinary Group. J Urol 1992 Aug;148(2 Pt 1):297–301. 47 Denzinger S, Mohren K, Knuechel R, et al. Improved clonality analysis of multifocal bladder tumors by combination of histopathologic organ mapping, loss of heterozygosity, fluorescence in situ hybridization, and p53 analyses. Hum Pathol 2006 Feb;37(2):143–51. 48 Heney NM, Ahmed S, Flanagan MJ, et al. Superficial bladder cancer: progression and recurrence. J Urol 1983 Dec;130(6):1083–6. 49 Herr HW. Tumor progression and survival of patients with high grade, noninvasive papillary (TaG3) bladder tumors: 15-year outcome. J Urol 2000 Jan;163(1):60–1; discussion 1–2. 50 Lamm DL. Carcinoma in situ. Urol Clin North Am 1992 Aug;19(3):499–508. 51 Riddle PR, Chisholm GD, Trott PA, Pugh RC. Flat carcinoma in Situ of bladder. Br J Urol 1975;47(7):829–33. 52 Althausen AF, Prout GR, Jr., Daly JJ. Non-invasive papillary carcinoma of the bladder associated with carcinoma in situ. J Urol 1976 Nov;116(5):575–80. 53 Farrow GM, Utz DC, Rife CC, Greene LF. Clinical observations on sixty-nine cases of in situ carcinoma of the urinary bladder. Cancer Res 1977 Aug;37(8 Pt 2):2794–8. 54 Hatch TR, Barry JM. The value of excretory urography in staging bladder cancer. J Urol 1986 Jan;135(1):49. 55 Fradet Y, Cordon-Cardo C. Critical appraisal of tumor markers in bladder cancer. Semin Urol 1993 Aug;11(3):145–53. 56 Ellis WJ, Blumenstein BA, Ishak LM, Enfield DL. Clinical evaluation of the BTA TRAK assay and comparison to voided urine cytology and the Bard BTA test in patients with recurrent bladder tumors. The Multi Center Study Group. Urology 1997 Dec;50(6):882–7. 57 Sarosdy MF. The use of the BTA Test in the detection of persistent or recurrent transitional-cell cancer of the bladder. World J Urol 1997;15(2):103–6. 58 D’Hallewin MA, Baert L. Initial evaluation of the bladder tumor antigen test in superficial bladder cancer. J Urol 1996 Feb;155(2):475–6. 59 Sarosdy MF, deVere White RW, Soloway MS, et al. Results of a multicenter trial using the BTA test to monitor for and
60
61
62
63
64
65
66
67 68 69
70
71
72 73
74
75
76
diagnose recurrent bladder cancer. J Urol 1995 Aug;154(2 Pt 1):379–83; discussion 83–4. Johnston B, Morales A, Emerson L, Lundie M. Rapid detection of bladder cancer: a comparative study of point of care tests. J Urol 1997 Dec;158(6):2098–101. Murphy WM, Rivera-Ramirez I, Medina CA, Wright NJ, Wajsman Z. The bladder tumor antigen (BTA) test compared to voided urine cytology in the detection of bladder neoplasms. J Urol 1997 Dec;158(6):2102–6. Soloway MS, Briggman V, Carpinito GA, et al. Use of a new tumor marker, urinary NMP22, in the detection of occult or rapidly recurring transitional cell carcinoma of the urinary tract following surgical treatment. J Urol 1996 Aug;156(2 Pt 1):363–7. Mao L, Schoenberg MP, Scicchitano M, et al. Molecular detection of primary bladder cancer by microsatellite analysis. Science 1996 Feb 2;271(5249):659–62. Williams GH, Romanowski P, Morris L, et al. Improved cervical smear assessment using antibodies against proteins that regulate DNA replication. Proc Natl Acad Sci U S A 1998 Dec 8;95(25):14932–7. Stoeber K, Halsall I, Freeman A, et al. Immunoassay for urothelial cancers that detects DNA replication protein Mcm5 in urine. Lancet 1999 Oct 30;354(9189):1524–5. Stoeber K, Swinn R, Prevost AT, et al. Diagnosis of genitourinary tract cancer by detection of minichromosome maintenance 5 protein in urine sediments. J Natl Cancer Inst 2002 Jul 17;94(14):1071–9. Sobin L, Wittekind C. Staging of urinary bladder. 6th ed. New York: Wiley-Liss; 2002. Pearse HD, Reed RR, Hodges CV. Radical cystectomy for bladder cancer. J Urol 1978 Feb;119(2):216–8. Blandy JP, England HR, Evans JW, et al. T3 bladder cancer – the case for salvage cystectomy. Br J Urol 1980;52(6): 506–10. Pugh R. Bladder Cancer. 1981. (Oliver RTD, Hendry WF, Bloom HJG) eds, Principles in combination therapy. Butterworths 1981, London. Esrig D, Freeman JA, Elmajian DA, et al. Transitional cell carcinoma involving the prostate with a proposed staging classification for stromal invasion. J Urol 1996 Sep;156(3):1071–6. MacVicar D, Husband JE. Radiology in the staging of bladder cancer. Br J Hosp Med 1994 May 4–17;51(9):454–8. Koss JC, Arger PH, Coleman BG, Mulhern CB, Jr., Pollack HM, Wein AJ. CT staging of bladder carcinoma. AJR Am J Roentgenol 1981 Aug;137(2):359–62. Weinerman PM, Arger PH, Pollack HM. CT evaluation of bladder and prostate neoplasms. Urol Radiol 1982;4(2–3):105–14. Walsh JW, Amendola MA, Konerding KF, Tisnado J, Hazra TA. Computed tomographic detection of pelvic and inguinal lymph-node metastases from primary and recurrent pelvic malignant disease. Radiology 1980 Oct;137(1 Pt 1):157–66. Kim B, Semelka RC, Ascher SM, Chalpin DB, Carroll PR, Hricak H. Bladder tumor staging: comparison of contrastenhanced CT, T1- and T2-weighted MR imaging, dynamic
636 Bladder cancer
77 78 79
80
81
82
83
84
85
86
87
88
89
90
91
92
gadolinium-enhanced imaging, and late gadoliniumenhanced imaging. Radiology 1994 Oct;193(1):239–45. Husband J, Johnson R, Reznek R. A guide to practical use of MRI in oncology. Royal College of Radiologists, London. 1999. Hawnaur JM. Staging of cervical and endometrial carcinoma. Clin Radiol 1993 Jan;47(1):7–13. Barentsz JO, Jager GJ, van Vierzen PB, et al. Staging urinary bladder cancer after transurethral biopsy: value of fast dynamic contrast-enhanced MR imaging. Radiology 1996 Oct;201(1):185–93. Schuller J, Walther V, Schmiedt E, Staehler G, Bauer HW, Schilling A. Intravesical ultrasound tomography in staging bladder carcinoma. J Urol 1982 Aug;128(2):264–6. Koraitim M, Kamal B, Metwalli N, Zaky Y. Transurethral ultrasonographic assessment of bladder carcinoma: its value and limitation. J Urol 1995 Aug;154(2 Pt 1):375–8. Messing EM, Young TB, Hunt VB, et al. Hematuria home screening: repeat testing results. J Urol 1995 Jul;154(1): 57–61. Britton JP, Dowell AC, Whelan P, Harris CM. A community study of bladder cancer screening by the detection of occult urinary bleeding. J Urol 1992 Sep;148(3):788–90. Hofstetter A, Frank F, Keiditsch E, Bowering R. Endoscopic Neodymium-YAG laser application for destroying bladder tumors. Eur Urol 1981;7(5):278–82. Jichlinski P, Guillou L, Karlsen SJ, et al. Hexyl aminolevulinate fluorescence cystoscopy: new diagnostic tool for photodiagnosis of superficial bladder cancer—a multicenter study. J Urol 2003 Jul;170(1):226–9. Schmidbauer J, Witjes F, Schmeller N, Donat R, Susani M, Marberger M. Improved detection of urothelial carcinoma in situ with hexaminolevulinate fluorescence cystoscopy. J Urol 2004 Jan;171(1):135–8. Filbeck T, Pichlmeier U, Knuechel R, Wieland WF, Rossler W. Reducing the risk of superficial bladder cancer recurrence with 5-aminolevulinic acid-induced fluorescence diagnosis. Results of a 5-year study. Urologe A 2003 Oct;42(10):1366–73. Zaak D, Karl A, Knuchel R, et al. Diagnosis of urothelial carcinoma of the bladder using fluorescence endoscopy. BJU Int 2005 Aug;96(2):217–22. Parmar MK, Freedman LS, Hargreave TB, Tolley DA. Prognostic factors for recurrence and followup policies in the treatment of superficial bladder cancer: report from the British Medical Research Council Subgroup on Superficial Bladder Cancer (Urological Cancer Working Party). J Urol 1989 Aug;142(2 Pt 1):284–8. Kurth KH, Denis L, Bouffioux C, et al. Factors affecting recurrence and progression in superficial bladder tumours. Eur J Cancer 1995;31a(11):1840–6. Fitzpatrick JM, West AB, Butler MR, Lane V, O’Flynn JD. Superficial bladder tumors (stage pTa, grades 1 and 2): the importance of recurrence pattern following initial resection. J Urol 1986 May;135(5):920–2. Schulman CC, Robinson M, Denis L, et al. Prophylactic chemotherapy of superficial transitional cell bladder carcinoma: an EORTC randomized trial comparing thiotepa,
93
94 95
96
97
98
99
100
101
102
103
an epipodophyllotoxin (VM26) and TUR alone. Eur Urol 1982;8(4):207–12. Kurth K, Tunn U, Ay R, et al. Adjuvant chemotherapy for superficial transitional cell bladder carcinoma: long-term results of a European Organization for Research and Treatment of Cancer randomized trial comparing doxorubicin, ethoglucid and transurethral resection alone. J Urol 1997 Aug;158(2):378–84. Van der Meijden A. Ta, T1 Bladder cancer: What can we learn from EORTC trials. Urol Int 1997;4:15–9. Pawinski A, Sylvester R, Kurth KH, et al. A combined analysis of European Organization for Research and Treatment of Cancer, and Medical Research Council randomized clinical trials for the prophylactic treatment of stage TaT1 bladder cancer. European Organization for Research and Treatment of Cancer Genitourinary Tract Cancer Cooperative Group and the Medical Research Council Working Party on Superficial Bladder Cancer. J Urol 1996 Dec;156(6):1934–40, discussion 40–1. Oosterlinck W, Kurth KH, Schroder F, Bultinck J, Hammond B, Sylvester R. A prospective European Organization for Research and Treatment of Cancer Genitourinary Group randomized trial comparing transurethral resection followed by a single intravesical instillation of epirubicin or water in single stage Ta, T1 papillary carcinoma of the bladder. J Urol 1993 Apr;149(4):749–52. Tolley DA, Parmar MK, Grigor KM, et al. The effect of intravesical mitomycin C on recurrence of newly diagnosed superficial bladder cancer: a further report with 7 years of follow up. J Urol 1996;155(4):1233–8. Bouffioux CH, Kurth KH, Bono A, et al. Intravesical adjuvant chemotherapy for superficial transitional cell bladder carcinoma: Results of 2 European Organization for Research and Treatment of Cancer randomised trials with mitomycin C and doxorubicin comparing early versus delayed instillations and short-term versus long-term treatment. J Urol 1995;153(3):934–41. Au JL, Badalament RA, Wientjes MG, et al. Methods to improve efficacy of intravesical mitomycin C: results of a randomized phase III trial. J Natl Cancer Inst 2001 Apr 18;93(8):597–604. Di Stasi SM, Giannantoni A, Stephen RL, et al. Intravesical electromotive mitomycin C versus passive transport mitomycin C for high risk superficial bladder cancer: a prospective randomized study. J Urol 2003 Sep;170(3):777–82. Shelley MD, Wilt TJ, Barber J, Mason MD. A meta-analysis of randomised trials suggests a survival benefit for combined radiotherapy and radical cystectomy compared with radical radiotherapy for invasive bladder cancer: are these data relevant to modern practice? Clin Oncol (R Coll Radiol) 2004 May;16(3):166–71. Shelley MD, Court JB, Kynaston H, Wilt TJ, Fish RG, Mason M. Intravesical Bacillus Calmette-Guerin in Ta and T1 Bladder Cancer. Cochrane Database Syst Rev 2000(4):CD001986. Shelley MD, Wilt TJ, Court J, Coles B, Kynaston H, Mason MD. Intravesical bacillus Calmette-Guerin is superior to
References 637
104
105
106 107
108
109
110
111
112
113
114
115
116
mitomycin C in reducing tumour recurrence in high-risk superficial bladder cancer: a meta-analysis of randomized trials. BJU Int 2004 Mar;93(4):485–90. Bohle A, Bock PR. Intravesical bacille Calmette-Guerin versus mitomycin C in superficial bladder cancer: formal meta-analysis of comparative studies on tumor progression. Urology 2004 Apr;63(4):682–6; discussion 6–7. Lamm DL, Blumenstein BA, Crawford ED, et al. A randomized trial of intravesical doxorubicin and immunotherapy with bacille Calmette-Guerin for transitional-cell carcinoma of the bladder. N Engl J Med 1991 Oct 24;325(17):1205–9. Martinez-Pineiro JA, Martinez-Pineiro L. BCG update: intravesical therapy. Eur Urol 1997;31 Suppl 1:31–41. Pagano F, Bassi P, Milani C, Piazza N, Meneghini A, Garbeglio A. BCG in superficial bladder cancer: a review of phase III European trials. Eur Urol 1992;21 Suppl 2:7–11. Sylvester RJ, van der Meijden AP, Witjes JA, Kurth K. Bacillus calmette-guerin versus chemotherapy for the intravesical treatment of patients with carcinoma in situ of the bladder: a meta-analysis of the published results of randomized clinical trials. J Urol 2005 Jul;174(1):86–91; discussion -2. Kaasinen E, Wijkstrom H, Malmstrom PU, et al. Alternating mitomycin C and BCG instillations versus BCG alone in treatment of carcinoma in situ of the urinary bladder: a Nordic study. Eur Urol 2003 Jun;43(6):637–45. Rintala E, Jauhiainen K, Rajala P, Ruutu M, Kaasinen E, Alfthan O. Alternating mitomycin C and bacillus CalmetteGuerin instillation therapy for carcinoma in situ of the bladder. The Finnbladder Group. J Urol 1995 Dec;154(6):2050–3. Witjes JA, Caris CT, Mungan NA, Debruyne FM, Witjes WP. Results of a randomized phase III trial of sequential intravesical therapy with mitomycin C and bacillus Calmette-Guerin versus mitomycin C alone in patients with superficial bladder cancer. J Urol 1998 Nov;160(5):1668–71; discussion 71–2. Ali-El-Dein B, Nabeeh A, Ismail EH, Ghoneim MA. Sequential bacillus Calmette-Guerin and epirubicin versus bacillus Calmette-Guerin alone for superficial bladder tumors: a randomized prospective study. J Urol 1999 Aug;162(2):339–42. Di Stasi SM, Giannantoni A, Giurioli A, et al. Sequential BCG and electromotive mitomycin versus BCG alone for high-risk superficial bladder cancer: a randomised controlled trial. Lancet Oncol 2006 Jan;7(1):43–51. Lamm D, Blumenstein B, Sarosdy Mea. Significant long term patient benefit with BCG maintenance therapy: a South–west Oncology Group study. J Urol 1997;157. Catalona WJ, Hudson MA, Gillen DP, Andriole GL, Ratliff TL. Risks and benefits of repeated courses of intravesical bacillus Calmette-Guerin therapy for superficial bladder cancer. J Urol 1987 Feb;137(2):220–4. van der Meijden AP, Brausi M, Zambon V, Kirkels W, de Balincourt C, Sylvester R. Intravesical instillation of epirubicin, bacillus Calmette-Guerin and bacillus CalmetteGuerin plus isoniazid for intermediate and high risk Ta, T1 papillary carcinoma of the bladder: a European Organization
117
118
119
120
121
122
123
124
125
126 127
128
129
130
131
132
for Research and Treatment of Cancer genito-urinary group randomized phase III trial. J Urol 2001 Aug;166(2):476–81. Lamm DL, van der Meijden PM, Morales A, et al. Incidence and treatment of complications of bacillus Calmette-Guerin intravesical therapy in superficial bladder cancer. J Urol 1992 Mar;147(3):596–600. Herr HW, Wartinger DD, Fair WR, Oettgen HF. Bacillus Calmette-Guerin therapy for superficial bladder cancer: a 10-year followup. J Urol 1992 Apr;147(4):1020–3. Schelhammer P. Intravesical BCG treatment superficial transitional cell carcinoma of the bladder and prostatic urethra BCG immunotherapy in superficial bladder cancer: (Pagano FABP) ed, Cleup, Badova; 1994. Herr HW, Badalament RA, Amato DA, Laudone VP, Fair WR, Whitmore WF, Jr. Superficial bladder cancer treated with bacillus Calmette-Guerin: a multivariate analysis of factors affecting tumor progression. J Urol 1989 Jan;141(1):22–9. Yamada Y, Hara I, Kumano M, Furukawa J, Yamanaka K, Kamidono S. Is second course intravesical Bacillus CalmetteGuerin therapy for recurrent carcinoma in situ of the bladder useful?. Hinyokika Kiyo 2005 Aug;51(8):539–43. Witjes JA, Melissen DO, Kiemeney LA. Current practice in the management of superficial bladder cancer in the Netherlands and Belgian Flanders: a survey. Eur Urol 2006 Mar;49(3):478–84. Joudi FN, O’Donnell MA. Second-line intravesical therapy versus cystectomy for bacille Calmette-Guerin (BCG) failures. Curr Opin Urol 2004 Sep;14(5):271–5. Gacci M, Bartoletti R, Cai T, et al. Intravesical gemcitabine in BCG-refractory T1G3 transitional cell carcinoma of the bladder: a pilot study. Urol Int 2006;76(2):106–11. Rodel C, Dunst J, Grabenbauer GG, et al. Radiotherapy is an effective treatment for high-risk T1-bladder cancer. Strahlenther Onko 2001 Feb;177(2):82–8; discussion 9. Harland SJ. A second look at the pT1 G3 bladder tumour. Clin Oncol (R Coll Radiol 2005 Oct;17(7):498–502. Nogueira March J, Ojea A, Figueiredo L, Jamardo D, Diez E, Perez Villanueve J. Evaluation of the efficacy of oral methotrexate in the prevention of recurrence of superficial bladder tumours. Br J Urol 1985;57(3):306–7. Kubota Y, Hosaka M, Fukushima S, Kondo I. Prophylactic oral UFT therapy for superficial bladder cancer. Cance 1993 Mar 1;71(5):1842–5. Nutting C, Huddart R. Rethinking the secondary prevention of superficial bladder cancer: is there a role for retinoids? BJU 2000;85(9):1023–6. Lamm DL, Riggs DR, Shriver JS, vanGilder PF, Rach JF, DeHaven JI. Megadose vitamins in bladder cancer: a doubleblind clinical trial. J Urol 1994 Jan;151(1):21–6. Sheehy O, Zhao S, Raymoundo A, Miller B, Aprikian A, Lelorier J. Celecoxib associated with reduced risk f superficial bladder cancer (SBC) recurrence. ASCO, ThirtyNinth Annual Meeting; 2003; May 31–June 3, 2003, Chicago, Illinois: JCO; 2003. p. 383 Abstract 1539. Castelao JE, Yuan JM, Gago-Dominguez M, Yu MC, Ross RK. Non-steroidal anti-inflammatory drugs and bladder cancer prevention. Br J Cancer 2000 Apr;82(7):1364–9.
638 Bladder cancer
133 Dovedi SJ, Kirby JA, Atkins H, Davies BR, Kelly JD. Cyclooxygenase-2 inhibition: a potential mechanism for increasing the efficacy of bacillus calmette-guerin immunotherapy for bladder cancer. J Urol 2005 Jul;174(1):332–7; discussion 7. 134 Skinner DG, Lieskovsky G. Contemporary cystectomy with pelvic node dissection compared to preoperative radiation therapy plus cystectomy in management of invasive bladder cancer. J Urol 1984;131(6):1069–72. 135 Skinner DG, Daniels JR, Russell CA, et al. The role of adjuvant chemotherapy following cystectomy for invasive bladder cancer: a prospective comparative trial. J Urol 1991 Mar;145(3):459–64; discussion 64–7. 136 Fradet Y, Tardif M, Bourget L, Robert J. Clinical cancer progression in urinary bladder tumors evaluated by multiparameter flow cytometry with monoclonal antibodies. Laval University Urology Group. Cancer Res 1990 Jan 15;50(2):432–7. 137 Okamura K, Miyake K, Koshikawa T, Asai J. Growth fractions of transitional cell carcinomas of the bladder defined by the monoclonal antibody Ki-67. J Urol 1990 Oct;144(4):875–8. 138 Bush C, Price P, Norton J, et al. Proliferation in human bladder carcinoma measured by Ki67 antibody labelling: its potential clinical importance. BrJ Cancer 1991;64(2): 357–60. 139 Cohen MB, Waldman FM, Carroll PR, Kerschmann R, Chew K, Mayall BH. Comparison of five histopathologic methods to assess cellular proliferation in transitional cell carcinoma of the urinary bladder. Hum Pathol 1993 Jul;24(7):772–8. 140 Takeichi M. Cadherin cell adhesion receptors as a morphogenetic regulator. Science 1991 Mar 22;251(5000):1451–5. 141 Edelman GM, Crossin KL. Cell adhesion molecules: implications for a molecular histology. Annu Rev Biochem 1991;60:155–90. 142 Jouanneau J, Gavrilovic J, Caruelle D, et al. Secreted or nonsecreted forms of acidic fibroblast growth factor produced by transfected epithelial cells influence cell morphology, motility, and invasive potential. Proc Natl Acad Sci U S A 1991 Apr 1;88(7):2893–7. 143 Yura Y, Hayashi O, Kelly M, Oyasu R. Identification of epidermal growth factor as a component of the rat urinary bladder tumor-enhancing urinary fractions. Cancer Res 1989 Mar 15;49(6):1548–53. 144 O’Reilly MS, Holmgren L, Chen C, Folkman J. Angiostatin induces and sustains dormancy of human primary tumors in mice. Nat Med 1996 Jun;2(6):689–92. 145 Lipponen PK, Eskelinen MJ. Reduced expression of E-cadherin is related to invasive disease and frequent recurrence in bladder cancer. J Cancer Res Clin Oncol 1995;121(5):303–8. 146 Bochner BH, Cote RJ, Weidner N, et al. Angiogenesis in bladder cancer: relationship between microvessel density and tumor prognosis. J Natl Cancer Inst 1995 Nov 1;87(21):1603–12. 147 Grossfeld GD, Ginsberg DA, Stein JP, et al. Thrombospondin1 expression in bladder cancer: association with p53
148
149
150
151
152
153
154
155
156
157
158
159
160 161 162
163
alterations, tumor angiogenesis, and tumor progression. J Natl Cancer Inst 1997 Feb 5;89(3):219–27. Roehrborn CG, Sagalowsky AI, Peters PC. Long-term patient survival after cystectomy for regional metastatic transitional cell carcinoma of the bladder. J Urol 1991 Jul;146(1):36–9. Wishnow KI, Dmochowski R. Pelvic recurrence after radical cystectomy without preoperative radiation. J Urol 1988 Jul;140(1):42–3. Montie JE, Straffon RA, Stewart BH. Radical cystectomy without radiation therapy for carcinoma of the bladder. J Urol 1984;131:477–82. Birkmeyer NJ, Goodney PP, Stukel TA, Hillner BE, Birkmeyer JD. Do cancer centers designated by the National Cancer Institute have better surgical outcomes? Cancer 2005 Feb 1;103(3):435–41. Quek ML, Stein JP, Daneshmand S, et al. A critical analysis of perioperative mortality from radical cystectomy. J Urol 2006 Mar;175(3 Pt 1):886–9; discussion 9–90. Kock NG, Hulten L, Myrvold HE. Ileoanal anastomosis with interposition of the ileal ‘Kock pouch’. Preliminary results. Dis Colon Rectum 1989 Dec;32(12):1050–4. Kock NG, Ghoneim MA, Lycke KG, Mahran MR. Replacement of the bladder by the urethral Kock pouch: functional results, urodynamics and radiological features. J Urol 1989 May;141(5):1111–6. Song C, Kang T, Hong JH, Kim CS, Ahn H. Changes in the upper urinary tract after radical cystectomy and urinary diversion: a comparison of antirefluxing and refluxing orthotopic bladder substitutes and the ileal conduit. J Urol 2006 Jan;175(1):185–9; discussion 9. Walsh PC, Mostwin JL. Radical prostatectomy and cystoprostatectomy with preservation of potency. Results using a new nerve-sparing technique. Br J Urol 1984 Dec;56(6):694–7. Schoenberg MP, Walsh PC, Breazeale DR, Marshall FF, Mostwin JL, Brendler CB. Local recurrence and survival following nerve sparing radical cystoprostatectomy for bladder cancer: 10-year followup. J Urol 1996 Feb;155(2):490–4. Coloby PJ, Kakizoe T, Tobisu K, Sakamoto M. Urethral involvement in female bladder cancer patients: mapping of 47 consecutive cysto-urethrectomy specimens. J Urol 1994 Nov;152(5 Pt 1):1438–42. Hendry WF, Gowing NF, Wallace DM. Surgical treatment of urethral tumours associated with bladder cancer. Proc R Soc Med 1974 Apr;67(4):304–7. Sweeney P, Kursh ED, Resnick MI. Partial cystectomy. Urol Clin North Am 1992 Nov;19(4):701–11. Herr HW. Transurethral resection in regionally advanced bladder cancer. Urol Clin North Am 1992 Nov;19(4):695–700. Henry K, Miller J, Mori M, Loening S, Fallon B. Comparison of transurethral resection to radical therapies for stage B bladder tumors. J Urol 1988 Nov;140(5):964–7. Solsona E, Iborra I, Ricos JV, et al. Corpus cavernosum invasion and tumor grade in the prediction of lymph node condition of penile carcinoma. Eur Urol 1992;22(2):115–8.
References 639
164 Kondas J, Szentgyorgyi E. Transurethral resection of 1250 bladder tumours. Int Urol Nephrol 1992;24(1):35–42. 165 Herr H. Conservative management of muscle-infiltrating bladder cancer: prospective experience. J Urol 1987;138:1162–3. 166 Koloszy Z. Histological ‘self control’ in transurethral resection of bladder tumours. Br J Urol 1991;67:162–4. 167 Alken P, Kohrmann KU. The essentials of transurethral resection of bladder tumors (TUR-B). Arch Ital Urol Androl 1993 Dec;65(6):629–32. 168 Martinez-Pineiro L, Gonzalez-Peramato P, Hidalgo L, et al. Primary bladder adenocarcinoma: retrospective study of 11 cases and general review. Arch Esp Urol 1991 Mar;44(2):131–8. 169 Thomas DJ, Roberts JT, Hall RR, Reading J. Radical transurethral resection and chemotherapy in the treatment of muscle-invasive bladder cancer: a long-term follow-up. BJU Int 1999 Mar;83(4):432–7. 170 Vale JA, A’Hern RP, Liu K, et al. Predicting the outcome of radical radiotherapy for invasive bladder cancer. Eur Urol 1993;24(1):48–51. 171 Pollack A, Zagars GK, Cole CJ, Dinney CP, Swanson DA, Grossman HB. The relationship of local control to distant metastasis in muscle invasive bladder cancer. J Urol 1995 Dec;154(6):2059–63; discussion 63–4. 172 Gospodarowicz MK, Hawkins NV, Rawlings GA, et al. Radical radiotherapy for muscle invasive transitional cell carcinoma of the bladder: failure analysis. J Urol 1989;142(6): 1448–53. 173 Duncan W, Quilty PM. The results of a series of 963 patients with transitional cell carcinoma of the urinary bladder primarily treated by radical megavoltage X-ray therapy. Radiother Oncol 1986;7(4):299–310. 174 Blandy JP, Jenkins BJ, Fowler CG, et al. Radical radiotherapy and salvage cystectomy for T2/3 cancer of the bladder. Prog Clin Biol Res 1988;260:447–51. 175 Davidson SE, Symonds RP, Snee MP, Upadhyay S, Habeshaw T, Robertson AG. Assessment of factors influencing the outcome of radiotherapy for bladder cancer. Br J Urol 1990 Sep;66(3):288–93. 176 Smaaland R, Akslen LA, Tonder B, Mehus A, Lote K, Albrektsen G. Radical radiation treatment of invasive and locally advanced bladder carcinoma in elderly patients. Br J Urol 1991 Jan;67(1):61–9. 177 Barlebo H, Steven K, Bea S. Preoperative irradiation (40 Gy) and cystectomy versus radiotherapy (60 Gy) followed by salvage cystectomy in the treatment of advanced bladder cancer. J Urol (suppl) 1990;143:291A. 178 Bloom HJ, Hendry WF, Wallace DM, Skeet RG. Treatment of T3 bladder cancer: controlled trial of preoperative radiotherapy and radical cystectomy versus radical radiotherapy, second report and review. Br J Urol 1982;54:136–51. 179 Sell A, Jakobsen A, Nerstrom B, Sorensen BL, Steven K, Barlebo H. Treatment of advanced bladder cancer category T2 T3 and T4a. A randomized multicenter study of preoperative irradiation and cystectomy versus radical
180 181
182
183
184
185 186 187 188
189
190
191
192
193
194
195
irradiation and early salvage cystectomy for residual tumor. DAVECA protocol 8201. Danish Vesical Cancer Group. Scand J Urol Nephrol Suppl 1991;138:193–201. Horwich A, Huddart R. Conservative treatment of bladder cancer. Clin Oncol (R Coll Radiol) 2004 May;16(3):163–5. Fung CY, Shipley WU, Young RH, et al. Prognostic factors in invasive bladder carcinoma in a prospective trial of preoperative adjuvant chemotherapy and radiotherapy. J Clin Oncol 1991 Sep;9(9):1533–42. Quilty PM, Hargreave TB, Smith G, Duncan W. Do normal mucosal biopsies predict prognosis in patients with transitional cell carcinoma of bladder treated by radical radiotherapy? Br J Urol 1987 Mar;59(3):242–7. Wolf H, Olsen PR, Hojgaard K. Urothelial dysplasia concomitant with bladder tumours: a determinant for future new occurrences in patients treated by full-course radiotherapy. Lancet 1985 May 4;1(8436):1005–8. Pisters LL, Tykochinsky G, Wajsman Z. Intravesical bacillus Calmette-Guerin or mitomycin C in the treatment of carcinoma in situ of the bladder following prior pelvic radiation therapy. J Urol 1991 Dec;146(6):1514–7. Sagerman RH. Pre-operative irradiation in carcinoma of the bladder. Am J Röntgenol 1968;102:577–80. Prout GR, Jr. The surgical management of bladder carcinoma. Urol Clin North Am 1976 Feb;3(1):149–75. Miller LS. T3 bladder cancer: the case for higher radiation dosage. Cancer 1980 Apr 15;45(7 Suppl):1875–8. Morrison R. The results of treatment of cancer of the bladder-a clinical contribution to radiobiology. Clin Radiol 1975 Jan;26(1):67–75. Parsons JT, Million RR. Planned preoperative irradiation in the management of clinical stage B2-C (T3) bladder carcinoma. Int J Radiat Oncol Biol Phys 1988 Apr;14(4):797–810. Horwich A, Dearnaley D, Huddart R, et al. A randomised trial of accelerated radiotherapy for localised invasive bladder cancer. Radiother Oncol 2005 Apr;75(1):34–43. Jose C, Price A, Norman A, et al. Hypofractionated Radiotherapy for Patients with Carcinoma of the Bladder. Clinical Oncology 1999;11:330–3. Duchesne G, Bolger J, Griffiths G, et al. A randomized trial of hypofractionated schedules of palliative radiotherapy in the management of bladder carcinoma: results of Medical Research Council Trial BA09. Int J Radiol Oncol Biol Phys 2000;47(2):379–88. Greiner R, Skaleric C, Veraguth P. The prognostic significance of ureteral obstruction in carcinoma of the bladder. Int J Radiat Oncol Biol Phys 1977;2:1095–100. van der Werf-Messing B. Preoperative irradiation followed by cystectomy to treat carcinoma of the urinary bladder category T3NX,0–4M0. Int J Radiat Oncol Biol Phys 1979 Mar;5(3):395–401. Shipley WU, Rose MA, Perrone TL, Mannix CM, Heney NM, Prout GRJ. Full-dose irradiation for patients with invasive bladder carcinoma: clinical and histological factors prognostic of improved survival. J Urol 1985;134(4): 679–83.
640 Bladder cancer
196 Fossa SD, Aass N, Ous S, Waehre H, Ilner K, Hannisdal E. Survival after curative treatment of muscle-invasive bladder cancer. Acta Oncol 1996;35 Suppl 8:59–65. 197 Quilty PM, Duncan W. The influence of hemoglobin level on the regression and long term local control of transitional cell carcinoma of the bladder following photon irradiation. Int J Radiat Oncol Biol Phys 1986 Oct;12(10):1735–42. 198 Overgaard J, Horsman MR. Modification of Hypoxia-Induced Radioresistance in Tumors by the Use of Oxygen and Sensitizers. Semin Radiat Oncol 1996;6(1):10–21. 199 Cole CJ, Pollack A, Zagars GK, Dinney CP, Swanson DA, von Eschenbach AC. Local control of muscle-invasive bladder cancer: preoperative radiotherapy and cystectomy versus cystectomy alone. Int J Radiat Oncol Biol Phys 1995 May 15;32(2):331–40. 200 Sauer R, Dunst J, Altendorf-Hofmann A, Fischer H, Bornhof C, Schrott KM. Radiotherapy with and without cisplatin in bladder cancer. Int J Radiat Oncol Biol Phys 1990 Sep;19(3):687–91. 201 Hoskin PJ, Saunders MI, Dische S. Hypoxic radiosensitizers in radical radiotherapy for patients with bladder carcinoma: hyperbaric oxygen, misonidazole, and accelerated radiotherapy, carbogen, and nicotinamide. Cancer 1999 Oct 1;86(7):1322–8. 202 Hoskin PJ, Rojas AM, Phillips H, Saunders MI. Acute and late morbidity in the treatment of advanced bladder carcinoma with accelerated radiotherapy, carbogen, and nicotinamide. Cancer 2005 Jun 1;103(11):2287–97. 203 Pernot M, Hubert J, Guillemin F, et al. Combined surgery and brachytherapy in the treatment of some cancers of the bladder (partial cystectomy and interstitial iridium-192). Radiother Oncol 1996 Feb;38(2):115–20. 204 Rozan R, Albuisson E, Donnarieix D, et al. Interstitial Iridium-192 for bladder cancer (a multicentric survey: 205 patients). Int J Radiat Oncol Biol Phys 1992;24(3):469–77. 205 Moonen LMF, Horenblas S, van der Voet JCM, Nuytem MJC, Bartelink H. Bladder conservation n selected T1 G3 and muscle-invasive T2-T3a bladder carcinoma using combination therapy of surgery and iridium-192 implantation. Br J Urol 1994;74(3):322–7. 206 Wijnmaalen A, Helle PA, Koper PC, et al. Muscle invasive bladder cancer treated by transurethral resection, followed by external beam radiation and interstitial iridium-192. Int J Radiat Oncol Biol Phys 1997 Dec 1;39(5):1043–52. 207 de Neve W, Lybeert MLM, Goor C, Crommelin MA, Ribot JG. T1 and T2 carcinoma of the urinary bladder: long term results with external, preoperative, or interstitial radiotherapy. Int J Radiat Oncol Biol Phys 1992;23(2):299–304. 208 Sur RK, Clinkard J, Jones WG, et al. Changes in target volume during radiotherapy treatment of invasive bladder carcinoma. Clin Oncol (R Coll Radiol) 1993;5(1):30–3. 209 Turner SL, Swindell R, Bowl N, et al. Bladder movement during radiation therapy for bladder cancer: implications for treatment planning. Int J Radiat Oncol Biol Phys 1997 Sep 1;39(2):355–60. 210 Splinter TA, Scher HI, Denis L, et al. The prognostic value of the pathological response to combination chemotherapy
211
212
213
214
215
216
217
218
219
220
221
222
223
before cystectomy in patients with invasive bladder cancer. European Organization for Research on Treatment of Cancer—Genitourinary Group. J Urol 1992 Mar;147(3):606–8. Shipley WU, Kaufman DS, Zehr E, et al. Selective bladder preservation by combined modality protocol treatment: long-term outcomes of 190 patients with invasive bladder cancer. Urology 2002;60(1):62–7. Zietman AL, Sacco D, Skowronski U, et al. Organ conservation in invasive bladder cancer by transurethral resection, chemotherapy and radiation: results of a urodynamic and quality of life study on long-term survivors. J Urol 2003 Nov;170(5):1772–6. Shipley WU, Kaufman DS, Tester WJ, Pilepich MV, Sandler HM. Overview of bladder cancer trials in the Radiation Therapy Oncology Group. Cancer 2003 Apr 15;97(8 Suppl):2115–9. Hagan MP, Winter KA, Kaufman DS, et al. RTOG 97-06: initial report of a phase I-II trial of selective bladder conservation using TURBT, twice-daily accelerated irradiation sensitized with cisplatin, and adjuvant MCV combination chemotherapy. Int J Radiat Oncol Biol Phys 2003 Nov 1;57(3):665–72. Whitmore WF, Jr., Batata M. Status of integrated irradiation and cystectomy for bladder cancer. Urol Clin North Am 1984 Nov;11(4):681–91. Cole C, Pollack A, Zagars GK, Dinney CP, Swanson DA, von Eschenbach AC. Local control of muscle-invasive bladder cancer; preoperative radiotherapy and cystectomy versus cystectomy alone. Int J Radiat Oncol Biol Phys 1994;30 Suppl 1:200. Abrahamsen JF, Fossa SD. Long-term morbidity after curative radiotherapy for carcinoma of the bladder. A retrospective study. Strahlenther Onkol 1990 Sep;166(9):580–3. Blackard CE, Byar DP. Results of a clinical trial of surgery and radiation in stages II and 3 carcinoma of the bladder. J Urol 1972 Dec;108(6):875–8. Slack NH, Bross IDJ, Prout GR. Five-year follw up results of collaborative studies of therapies for carcinoma of the bladder. J Surg Oncol 1977;9(4):393–405. Smith J, Crawford E, Blumenstein Bea. A randomized prospective trial of prospective irradiation plus radical cystectomy plus surgery alone for transitional cell carcinoma of the bladder. A Southwest Oncology Group Study. J Urol 1988;139:266A. Crawford ED, Das S, Smith JA, Jr. Preoperative radiation therapy in the treatment of bladder cancer. Urol Clin North Am 1987 Nov;14(4):781–7. Reisinger SA, Mohiuddin M, Mulholland SG. Combined preand postoperative adjuvant radiation therapy for bladder cancer—a ten year experience. Int J Radiat Oncol Biol Phys 1992;24(3):463–8. Hankey BF, Edwards BK, Ries LA, Percy CL, Shambaugh E. Problems in cancer surveillance: delineating in situ and invasive bladder cancer. J Natl Cancer Inst 1991 Mar 20;83(6):384–5.
References 641
224 Hall R, on behalf of the Int. Collaboration of Trialists of the MRC Advanced Bladder Cancer Group; MRC Clinical Trials Unit L, UK. Updated results of a randomised controlled trial of neoadjuvant cisplatin (C) methotrexate (M) and vinblastine (V) chemotherapy for muscle-invasive bladder cancer. ASCO, Thirty-Eighth Annual Meeting 2002; May 18–21, 2002, Orlando, Florida: JCO; 2002. p. 178a Abstract 710 225 Sherif A, Holmberg L, Rintala E, et al. Neoadjuvant cisplatinum based combination chemotherapy in patients with invasive bladder cancer: a combined analysis of two Nordic studies. Eur Urol 2004 Mar;45(3):297–303. 226 Grossman HB, Natale RB, Tangen CM, et al. Neoadjuvant chemotherapy plus cystectomy compared with cystectomy alone for locally advanced bladder cancer. N Engl J Med 2003 Aug 28;349(9):859–66. 227 Advanced Bladder Cancer (ABC) meta-analysis collaboration. adjuvant chemotherapy in invasive bladder cancer: A systematic review and meta-analysis of individual patient data. Eur Urol 2005 Aug;48(2):189–201 228 Coppin CM, Gospodarowicz MK, James K, et al. Improved local control of invasive bladder cancer by concurrent cisplatin and preoperative or definitive radiation. The National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 1996 Nov;14(11):2901–7. 229 Shipley WU, Kaufman DS, Heney NM, Griffin PP, Althausen AF, Prout GR, Jr. The integration of chemotherapy, radiotherapy and transurethral surgery in bladder-sparing approaches for patients with invasive tumors. Prog Clin Biol Res 1990;353:85–94. 230 Jakse G, Frommhold H, Nedden DZ. Combined radiation and chemotherapy for locally advanced transitional cell carcinoma of the urinary bladder. Cancer 1985;55: 1659–64. 231 Eapen L, Stewart D, Danjoux C, et al. Intraarterial cisplatin and concurrent radiation for locally advanced bladder cancer. J Clin Oncol 1989 Feb;7(2):230–5. 232 Rotman M, Aziz H, Porrazzo M, Choi KN, Silverstein M, Rosenthal J, et al. Treatment of advanced transitional cell carcinoma of the bladder with irradiation and concomitant 5-fluorouracil infusion. Int J Radiat Oncol Biol Phys 1990 May;18(5):1131–7. 233 Russell KJ, Boileau MA, Higano C, et al. Combined 5-fluorouracil and irradiation for transitional cell carcinoma of the urinary bladder. Int J Radiat Oncol Biol Phys 1990 Sep;19(3):693–9. 234 Housset M, Maulard C, Chretien Y, et al. Combined radiation and chemotherapy for invasive transitional-cell carcinoma of the bladder: A prospective study. J Clin Oncol 1993;11(11):2150–7. 235 Housset M, Dufour B, Maulard-Durdux C, Chretien Y, Mejean A. Concomitant fluorouracil (5-FU)-cisplatin (CDDP) and bifractionated split course radiation therapy (BSCRT) for invasive bladder cancer (Abstr 1139). Proc Am Soc Clin Oncol 1997;16:319a. 236 Hussain SA, Stocken DD, Riley P, et al. A phase I/II study of gemcitabine and fractionated cisplatin in an outpatient setting using a 21-day schedule in patients with advanced
237
238
239
240
241
242
243
244
245
246
247
248
and metastatic bladder cancer. Br J Cancer 2004 Aug 31;91(5):844–9. Sangar VK, McBain CA, Lyons J, et al. Phase I study of conformal radiotherapy with concurrent gemcitabine in locally advanced bladder cancer. Int J Radiat Oncol Biol Phys 2005 Feb 1;61(2):420–5. Lehmann J, Retz M, Wiemers C, et al. Adjuvant cisplatin plus methotrexate versus methotrexate, vinblastine, epirubicin, and cisplatin in locally advanced bladder cancer: results of a randomized, multicenter, phase III trial (AUO-AB 05/95). J Clin Oncol 2005 Aug 1;23(22):4963–74. Geller NL, Sternberg CN, Penenberg D, Scher H, Yagoda A. Prognostic factors for survival of patients with advanced urothelial tumors treated with methotrexate, vinblastine, doxorubicin, and cisplatin chemotherapy. Cancer 1991 Mar 15;67(6):1525–31. Mead G, Russell M, Clark P, et al. A randomized trial comparing methotrexate and vinblastine (MV) with cisplatin, methotrexate and vinblastine (CMV) in advanced transitional cell carcinoma: results and a report on prognostic factors in a Medical Research Council study. British Journal of Cancer 1998;78(8):1067–75. Al-Sarraf M, Frank J, Smith JA, Jr., et al. Phase II trial of cyclophosphamide, doxorubicin, and cisplatin (CAP) versus amsacrine in patients with transitional cell carcinoma of the urinary bladder: a Southwest Oncology Group study. Cancer Treat Rep 1985 Feb;69(2):189–94. Blumenreich MS, Needles B, Yagoda A, Sogani P, Grabstald H, Whitmore WF, Jr. Intravesical cisplatin for superficial bladder tumors. Cancer 1982 Sep 1;50(5):863–5. Gagliano R, Levin H, El-Bolkainy MN, et al. Adriamycin versus adriamycin plus cis-diamminedichloroplatinum (DDP) in advanced transitional cell bladder carcinoma. A Southwest Oncology Group study. Am J Clin Oncol 1983 Apr;6(2):215–8. Hillcoat BL, Raghavan D, Matthews J, et al. A randomised trial of cisplatin versus cisplatin plus methotrexate in advanced cancer of the urothelial tract. J Clin Oncol 1989;7(6):706–9. Richards B, Newling D, Fossa S, et al. Vincristine in advanced bladder cancer: a European organization for research on treatment of cancer (EORTC) phase II study. Cancer Treat Rep 1983 Jun;67(6):575–7. Roth BJ, Dreicer R, Einhorn LH, et al. Significant activity of paclitaxel in advanced transitional-cell carcinoma of the urothelium: a phase II trial of the Eastern Cooperative Oncology Group. J Clin Oncol 1994 Nov;12(11):2264–70. Soloway MS, Einstein A, Corder MP, Bonney W, Prout GR, Jr., Coombs J. A comparison of cisplatin and the combination of cisplatin and cyclophosphamide in advanced urothelial cancer. A National Bladder Cancer Collaborative Group A Study. Cancer 1983 Sep 1;52(5):767–72. Troner M, Birch R, Omura GA, Williams S. Phase III comparison of cisplatin alone versus cisplatin, doxorubicin and cyclophosphamide in the treatment of bladder (urothelial) cancer: a Southeastern Cancer Study Group trial. J Urol 1987 Apr;137(4):660–2.
642 Bladder cancer
249 Turner AG, Hendry WF, Williams GB, Bloom HJG. The treatment of advanced bladder cancer with methotrexate. Br J Urol 1977;49:673–8. 250 van Oosterom AT, Fossa SD, Mulder JH, Calciati A, de Pauw M, Sylvester R. Mitoxantrone in advanced bladder carcinoma. A phase II study of the EORTC Genito-urinary Tract Cancer Cooperative Group. Eur J Cancer Clin Oncol 1985 Sep;21(9):1013–4. 251 Witte RS, Elson P, Bono B, et al. Eastern Cooperative Oncology Group phase II trial of ifosfamide in the treatment of previously treated advanced urothelial carcinoma. J Clin Oncol 1997 Feb;15(2):589–93. 252 Loehrer PJ, Sr., Einhorn LH, Elson PJ, et al. A randomized comparison of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: a cooperative group study [published erratum appears in J Clin Oncol 1993 Feb;11(2):384]. J Clin Oncol. 1992;10(7):1066–73. 253 Khandekar JD, Elson PJ, DeWys WD, Slayton RE, Harris DT. Comparative activity and toxicity of CisDiamminedichloroplatinum (DDP) and a combination of Doxorubicin, Cyclophosphamide, and DDP in disseminated transitional cell carcinomas of the urinary tract. J Clin Oncol 1985;3(4):539–45. 254 Harker WG, Meyers FJ, Freiha FS, et al. Cisplatin, methotrexate, and vinblastine (CMV): an effective chemotherapy regimen for metastatic transitional cell carcinoma of the urinary tract. A Northern California Oncology Group Study. J Clin Oncol 1985;3(11):1463–70. 255 Sternberg CN, Yagoda A, Scher HI. Chemotherapy for advanced urothelial tract tumors: the M-VAC regimen. Prog Clin Biol Res 1988;277:45–51. 256 Sternberg CN, Yagoda A, Scher HI, et al. Methotrexate, vinblastine, doxorubicin, and cisplatin for advanced transitional cell carcinoma of the urothelium. Efficacy and patterns of response and relapse. Cancer 1989 Dec 15;64(12):2448–58. 257 Cook A, Huddart R, Jay G, Norman A, Dearnaley D, Horwich A. The utility of tumour markers in assessing the response to chemotherapy in advanced bladder cancer. BJC 2000;82(12):1952–7. 258 Raghavan D. Chemotherapy for advanced bladder cancer: ‘Midsummer Night’s Dream’ or ‘Much Ado About Nothing’? Br J Cancer 1990 Sep;62(3):337–40. 259 Loehrer P, De Mulder P. Chapter 28 Management of metastatic bladder cancer. (Raghavan DSH, Leibel SA, Scher HI, Lange PH)eds Philadelphia: Lippincott-Raven; 1997 Pg 299–305. 260 Young D, Garnick M. Chemotherapy in bladder cancer: the North American experience. London: Edward Arnold; 1988. 261 Logothetis CJ, Dexeus FH, Finn L, et al. A prospective randomized trial comparing MVAC and CISCA chemotherapy for patients with metastatic urothelial tumors. J Clin Oncol 1990;(6):1050–5. 262 Fossa SD, Sternberg C, Scher HI, et al. Survival of patients with advanced urothelial cancer treated with cisplatinbased chemotherapy. Br J Cancer 1996 Nov;74(10):1655–9.
263 Saxman SB, Propert KJ, Einhorn LH, et al. Long-term followup of a phase III intergroup study of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: a cooperative group study. J Clin Oncol 1997 Jul;15(7):2564–9. 264 Gabrilove JL, Jakubowski A, Scher H, et al. Effect of granulocyte colony-stimulating factor on neutropenia and associated morbidity due to chemotherapy for transitional cell carcinoma of the urothelium. N Engl J Med 1988;318:1414–22. 265 Loehrer P, Sr. Eelson, P, Dreicer R, et al. Escalated dosages of methotrexate, vinblastine, doxorubicin, and cisplatin plus recombinant human granulocyte colony-stimulating factor in advanced urothelial carcinoma: An Eastern Cooperative Oncology Group Trial. Journal of Clinical Oncology 1994;12(3):483–8. 266 Logothetis C, Finn L, Amato R, Hassan E, Sella A. Escalated MVAC + RHGM-CSF in metastatic transitional cell carcinoma. Proc Am Soc Clin Oncol 1992;11:202. 267 Moore MJ, Iscoe N, Tannock IF. A phase II study of methotrexate, vinblastine, doxorubicin and cisplatin plus recombinant human granulocyte-macrophage colony stimulating factors in patients with advanced transitional cell carcinoma. J Urol 1993 Oct;150(4):1131–4. 268 Seidman AD, Scher HI, Gabrilove JL, et al. Doseintensification of MVAC with recombinant granulocyte colony-stimulating factor as initial therapy in advanced urothelial cancer. J Clin Oncol 1993 Mar;11(3):408–14. 269 Sternberg C, de Mulder P, van Oosterom A, Fossa S, Giannarelli D, Soedirman J. Escalated M-VAC chemotherapy and recombinant human granulocyte-macrophage colony stimulating factor (rhGM-CSF) in patients with advanced urothelial tract tumors. Annals of Oncology 1993;4(5):403–7. 270 Sternberg CN, de Mulder P, Schornagel JH, et al. Seven year update of an EORTC phase III trial of high-dose intensity MVAC chemotherapy and G-CSF versus classic M-VAC in advanced urothelial tract tumours. Eur J Cancer 2006 Jan;42(1):50–4. 271 Dreicer R, Gustin DM, See WA, Williams RD. Paclitaxel in advanced urothelial carcinoma: its role in patients with renal insufficiency and as salvage therapy. J Urol 1996 Nov;156(5):1606–8. 272 Papamichael D, Gallagher CJ, Oliver RT, Johnson PW, Waxman J. Phase II study of paclitaxel in pretreated patients with locally advanced/metastatic cancer of the bladder and ureter. Br J Cancer 1997;75(4):606–7. 273 Dimopoulos MA, Deliveliotis C, Moulopoulos LA, et al. Treatment of patients with metastatic urothelial carcinoma and impaired renal function with single-agent docetaxel. Urology 1998 Jul;52(1):56–60. 274 McCaffrey JA, Hilton S, Mazumdar M, et al. Phase II trial of docetaxel in patients with advanced or metastatic transitional-cell carcinoma. J Clin Oncol 1997 May;15(5): 1853–7. 275 Pollera CF, Ceribelli A, Crecco M, Calabresi F. Weekly gemcitabine in advanced bladder cancer: a preliminary
References 643
276
277
278
279
280
281
282
283
284
285
286
287
288
report from a phase I study. Ann Oncol 1994 Feb;5(2):182–4. Moore MJ, Tannock IF, Ernst DS, Huan S, Murray N. Gemcitabine: a promising new agent in the treatment of advanced urothelial cancer. J Clin Oncol 1997 Dec;15(12):3441–5. Stadler WM, Kuzel T, Raghavan D, Levine E, Vogelzang NJ, Dorr F. A phase II study of gemcitabine in the treatment of patients with advanced transitional cell carcinoma. Proc Am Soc Clin Oncol 1995;14:241. Redman B, Smith D, Flaherty L, Du W, Hussain M. Phase II trial of paclitaxel and carboplatin in the treatment of advanced urothelial carcinoma. J Clin Oncol 1998;16(5):1844–8. Zielinski CC, Schnack B, Grbovic M, et al. Paclitaxel and carboplatin in patients with metastatic urothelial cancer: results of a phase II trial. Br J Cancer 1998;78(3):370–4. Vaughn D, Malkowicz S, Zoltick B, et al. Paclitaxel plus carboplatin in advanced carcinoma of the urothelium: an active and tolerable outpatient regimen. Journal of Clinical Oncology 1998;16(1):255–60. Li J, Juliar B, Yiannoutsos C, et al. Weekly paclitaxel and gemcitabine in advanced transitional-cell carcinoma of the urothelium: a phase II Hoosier Oncology Group study. J Clin Oncol 2005 Feb 20;23(6):1185–91. von der Maase H, Hansen SW, Roberts JT, et al. Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin, and cisplatin in advanced or metastatic bladder cancer: results of a large, randomized, multinational, multicenter, phase III study. J Clin Oncol 2000;17(17):3068–77. von der Maase H, Sengelov L, Roberts J, et al. Long-term survival results of a randomized trial comparing gemcitabine plus cisplatin, with methotrexate, vinblastine, doxorubicin, plus cisplatin in patients with bladder cancer. J Clin Oncol 2005;23(21):4602–8. Bamias A, Aravantinos G, Deliveliotis C, et al. Docetaxel and cisplatin with granulocyte colony-stimulating factor (G-CSF) versus MVAC with G-CSF in advanced urothelial carcinoma: a multicenter, randomized, phase III study from the Hellenic Cooperative Oncology Group. J Clin Oncol 2004 Jan 15;22(2):220–8. Bellmunt J, Guillem V, Paz-Ares L, et al. Phase I-II study of paclitaxel, cisplatin, and gemcitabine in advanced transitionalcell carcinoma of the urothelium. Spanish Oncology Genitourinary Group. J Clin Oncol 2000;18(18):3247–55. Bellmunt J, von der Maase H, Mead GM, et al. Randomized phase III study comparing paclitaxel/cisplatin/gemcitabine (PCG) and gemcitabine/cisplatin (GC) in patients with locally advanced (LA) or metastatic (M) urothelial cancer without prior systemic therapy; EORTC30987/Intergroup Study. J Clin Oncol 2007 ASCO Annual Meeting Proceedings Part I. 25 18 June 20 Supple. Seligman PA, Crawford ED. Treatment of advanced transitional cell carcinoma of the bladder with continuous-infusion gallium nitrate. J Natl Cancer Inst 1991 Nov 6;83(21):1582–4. Bellmunt J, Ribas A, Eres N, et al. Carboplatin-based versus cisplatin-based chemotherapy in the treatment of surgically
289
290
291
292
293
294
295
296
297
298
299
300
incurable advanced bladder carcinoma. Cancer 1997;10(10):1966–72. Hainsworth JD, Meluch AA, Litchy S, et al. Paclitaxel, carboplatin, and gemcitabine in the treatment of patients with advanced transitional cell carcinoma of the urothelium. Cancer 2005 Jun 1;103(11):2298–303. Bamias A, Moulopoulos LA, Koutras A, et al. The combination of gemcitabine and carboplatin as first-line treatment in patients with advanced urothelial carcinoma. A Phase II study of the Hellenic Cooperative Oncology Group. Cancer 2006 Jan 15;106(2):297–303. Montie JE, Whitmore WF, Jr., Grabstald HM, Yagoda A. Unresectable carcinoma of the bladder. Cancer 1983 Jun 15;51(12):2351–5. Sato N, Sumiya H, Isaka S, Shimazaki J, Matsuzaki O. [Prognostic factors for progression of superficial bladder cancer]. Nippon Hinyokika Gakkai Zasshi 1992 Aug;83(8):1263–9. Cordon-Cardo C, Wartinger DD, Melamed MR, Fair W, Fradet Y. Immunopathologic analysis of human urinary bladder cancer. Characterization of two new antigens associated with low-grade superficial bladder tumors. Am J Pathol 1992 Feb;140(2):375–85. Lipponen P, Eskelinen M. Expression of epidermal growth factor receptor in bladder cancer as related to established prognostic factors, oncoprotein (c-erbB-2, p53) expression and long-term prognosis. Br J Cancer 1994 Jun;69(6):1120–5. Zelefsky MJ, Hollister T, Raben A, Matthews S, Wallner KE. Five-year biochemical outcome and toxicity with transperineal CT-planned permanent I-125 prostate implantation for patients with localized prostate cancer. Int J Radiat Oncol Biol Phys 2000 Jul 15;47(5):1261–6. Liukkonen T, Rajala P, Raitanen M, Rintala E, Kaasinen E, Lipponen P. Prognostic value of MIB-1 score, p53, EGFr, mitotic index and papillary status in primary superficial (Stage pTa/T1) bladder cancer: a prospective comparative study. The Finnbladder Group. Eur Urol1999 Nov;36(5):393–400. Chopin DK, Caruelle JP, Colombel M, et al. Increased immunodetection of acidic fibroblast growth factor in bladder cancer, detectable in urine. J Urol 1993 Oct;150(4):1126–30. Izadifar V, de Boer WI, Muscatelli-Groux B, Maille P, van der Kwast TH, Chopin DK. Expression of transforming growth factor beta1 and its receptors in normal human urothelium and human transitional cell carcinomas. Hum Pathol 1999 Apr;30(4):372–7. Eder IE, Stenzl A, Hobisch A, Cronauer MV, Bartsch G, Klocker H. Expression of transforming growth factors beta1, beta 2 and beta 3 in human bladder carcinomas. Br J Cancer 1997;75(12):1753–60. Tokunaga H, Lee DH, Kim IY, Wheeler TM, Lerner SP. Decreased expression of transforming growth factor beta receptor type I is associated with poor prognosis in bladder transitional cell carcinoma patients. Clin Cancer Res 1999 Sep;5(9):2520–5.
644 Bladder cancer
301 Soloway MS, Ford KS. Thiotepa-induced myelosuppression: review of 670 bladder instillations. J Urol 1983 Nov;130(5):889–91. 302 Thrasher JB, Crawford ED. Complications of intravesical chemotherapy. Urol Clin North Am 1992 Aug;19(3):529–39. 303 Crawford ED, McKenzie D, Mansson W, et al. Adverse reactions to the intravesical administration of doxorubicin hydrochloride: report of 6 cases. J Urol 1986 Sep;136(3):668–9. 304 Akaza H, Isaka S, Koiso K, et al. Comparative analysis of short-term and long-term prophylactic intravesical chemotherapy of superficial bladder cancer. Prospective, randomized, controlled studies of the Japanese Urological Cancer Research Group. Cancer Chemother Pharmacol 1987;20 Suppl:S91-6. 305 Huland H, Otto U, Droese M, Kloppel G. Long-term mitomycin C instillation after transurethral resection of superficial bladder carcinoma: influence on recurrence, progression and survival. J Urol 1984 Jul;132(1):27–9. 306 Huland H, Kloppel G, Feddersen I, et al. Comparison of different schedules of cytostatic intravesical instillations in patients with superficial bladder carcinoma: final evaluation of a prospective multicenter study with 419 patients. J Urol 1990 Jul;144(1):68–71; discussion -2. 307 Burk K, Kurth KH, Newling D. Epirubicin in treatment and recurrence prophylaxis of patients with superficial bladder cancer. Prog Clin Biol Res 1989;303:423–34. 308 Shearer RJ, Chilvers CF, Bloom HJ, Bliss JM, Horwich A, Babiker A. Adjuvant chemotherapy in T3 carcinoma of the bladder. A prospective trial: Preliminary report. Br J Urol 1988;62(6):558–64. 309 Martinez-Pineiro JA, Jimenez Leon J, Gonzalez Martin M, et al. Neo-adjuvant cisplatinum in locally advanced urothelial bladder cancer: a prospective randomized study of the group CUETO. Prog Clin Biol Res 1990;353:95–103.
310 Wallace DM, Raghavan D, Kelly KA, et al. Neo-adjuvant (pre-emptive) cisplatin therapy in invasive transitional cell carcinoma of the bladder. Br J Urol 1991;67(6):608–15. 311 Rintala E, Hannisdahl E, Fossa SD, Hellsten S, Sander S. Neoadjuvant chemotherapy in bladder cancer: a randomized study. Nordic Cystectomy Trial I. Scand J Urol Nephrol 1993;27(3):355–62. 312 Malmstrom P, Rintala E, Wahlqvist R, Hellstrom P, Hellsten S, Hannisdal E. Five-year follow-up of a prospective trial of radical cystectomy and neoadjuvant chemotherapy: Nordic Cystectomy Trial I. The Nordic Cooperative Bladder Cancer Study Group. J Urol 1996;155(6):1903–6. 313 Curotto A, Canobbio C, Orsatti Mea. Chemoradiation in locally advanced bladder cancer: neoadjuvant treatment. Oncology: New Trends in Diagnosis and Treatment of Bladder Cancer 1994;3:433–7. 314 Cortesi E. Neoadjuvant treatment of locally advanced bladder cancer: a randomized prospective clinical trial. Proc Am Soc Clin Oncol 1995;14:A623. 315 Sternberg CN. Neoadjuvant and adjuvant chemotherapy in locally advanced bladder cancer. Semin Oncol 1996 Oct;23(5):621–32. 316 Hall RR. Neoadjuvant treatment for invasive bladder cancer: is it worthwhile? Acta Urol Belg 1996 May; 64(2):39–42. 317 Shipley W, Winter K, Kaufman D, et al. Phase III trial of neoadjuvant chemotherapy in patients with invasive bladder cancer treated with selective bladder preservation by combined radiation therapy and chemotherapy: initial results of radiation therapy Oncology Group 89–03. Journal of Clinical Oncology 1998;16(11):3576–83. 318 Sherif A, Rintala E, Mestad O, et al. Neoadjuvant cisplatinmethotrexate chemotherapy for invasive bladder cancer Nordic cystectomy trial 2. Scand J Urol Nephrol 2002;36(6):419–25.
27 Prostate cancer MALCOLM MASON
Introduction Epidemiology and aetiology Prevention and screening Pathogenesis Pathology and staging Clinical features Diagnosis and staging
645 645 646 647 647 649 650
INTRODUCTION Prostate cancer is a uniquely challenging disease. At the heart of the challenge is the disparity between the mortality rate (around 17.9 deaths per 100 000 men per annum in the UK, and 15.8 per 100 000 men per annum in the USA) and the incidence of the disease (around 52.2 cases per 100 000 in the UK, and 124.8 per 100 000 per annum in the USA).1 This disparity is not because of successful treatment. Indeed, the incidence figures underestimate the true ‘biological’ prevalence of the disease; older post-mortem studies indicate that up to 80 per cent of all men harbour foci of prostate cancer by the age of 80, although more recent studies suggest a much lower incidence of around 25 per cent overall.2,3 Rather, it is because, in the overwhelming majority of men in the population, prostate cancer is an indolent, even harmless disease. Our challenges are, first, correctly to identify those men whose cancer threatens their health and, second, to identify the best treatment for such men at an early stage. W. Whitmore’s dictum, written in 19904 and paraphrased here, still encapsulates the challenge that this chapter attempts to address: Is treatment possible for those in whom it is necessary? Is treatment necessary for those in whom it is possible?
EPIDEMIOLOGY AND AETIOLOGY Diet is of overwhelming importance. There is a striking geographical variation in the incidence of prostate cancer,
Principles of management Management of organ-confined disease Management of locally advanced disease Management of recurrent disease after radical local therapy Management of metastatic disease Concluding remarks References
651 651 656 657 657 658 659
with high levels in Western countries and low levels in Asia and the Far East.1 As with breast cancer, the changing incidence in migrants from low-incidence to high-incidence regions suggests a dietary component.5 While particular attention has been paid to dietary fat levels, particularly of saturated fat, there are other components. Around 10 000 years ago, the diet of humans changed as agriculture developed and as the dog was domesticated.6 The latter enabled humans to hunt larger animals for meat and, at this time, they also began to cook their meat and to share their food with dogs. Strikingly, as commented by Dr Don Coffey from the Johns Hopkins Hospital in Baltimore, in the entire animal kingdom, humans and dogs are the only two species that develop prostate cancer, and this suggests that our departure from our evolutionary heritage (in dietary terms) may be responsible for the disease in both animals. Taking this a step further, a link between cooked meat, saturated fat and proliferative inflammatory atrophy in the prostate (and the breast!) has been suggested,7 due perhaps to the generation of carcinogenic chemicals as a result of the cooking process and to the saturated fat content, both of which are postulated to set up a focal inflammatory process. Prostate cancer is generally a disease of old age. Its agespecific incidence curve shows a steady rise, with most cases being diagnosed in men in their sixties and seventies (Fig. 27.1). Nonetheless, this has shifted somewhat in recent years, as screening based on prostate-specific antigen (PSA) has led to the identification of men in their fifties and even their forties with prostate cancer,8 almost certainly a manifestation of lead-time bias.
646 Prostate cancer
1,000 Male cases Male rates
6 000
750
4 000
500
2 000
250
0
Rate per 100,000 males
Number of cases
8 000
85+
80–84
75–79
70–74
65–69
60–64
55–59
50–54
45–49
40–44
35–39
30–34
25–29
20–24
15–19
5–9
10–14
0–4
0
Age at diagnosis Figure 27.1 Numbers of new cases and age-specific incidence rates, male prostate cancer, UK 2002
A genetic component to the aetiology is suggested by the increased risk of the disease faced by first-degree relatives, especially when multiple first-degree relatives are affected at a young age (60 years). Although several candidate gene regions have been proposed, no definitive susceptibility gene has been described whose role can be validated in different populations.9 It seems more likely that multiple, common genetic variations (such as single nucleotide polymorphisms) combine to confer an increased risk.9 Interestingly, candidates include two genes that might be considered to be related to inflammation.7,10,11 On the other hand, there appears to be an increased risk in families carrying mutations of the BRCA-1 and BRCA-2 genes,10,12,13 which in turn might implicate a wider role for variations in ATM gene polymorphisms in the wider population.9,14 Genetic factors may be one of several that contribute to an increased risk of the disease in some races, most notably among black Americans.15,16 Furthermore, black Americans are more likely to present with high-grade disease.17 Vitamin D has also been implicated in the aetiology of prostate cancer, with the suggestion that men with low levels are at increased risk of developing the disease.18 Whereas this might at first sight appear to be strange, it is known that vitamin D plays an important role in cell signalling, and has influence on the growth of prostate cancer cells in laboratory models.18,19 Suggestions that cadmium exposure is a risk factor for prostate cancer have not been borne out by recent studies.20
PREVENTION AND SCREENING There is no topic in medicine as controversial as prostate cancer screening. It is recommended in the USA, beginning in men over 50 years old, and is common practice in some other countries, whereas in others it is not yet a formal public health programme. To the layman, it seems obvious that early detection must lead to better outcomes, and the path to early detection of prostate cancer is through screening.
This is reinforced by public health messages about screening for hypertension, or cervical cancer, or breast cancer, or diabetes. How could it be different for prostate cancer? The fall in prostate cancer mortality seen in the USA in the late 1990s has been ascribed to screening, but it is simply impossible that a reduction in mortality would appear so early, as Prostate-Specific Antigen (PSA) tests only became available in the late 1980s and screening only began in earnest the early 1990s.17 Data from the Tyrol province of Austria, where a national screening programme was introduced, have also been cited as evidence for the efficacy of screening, as this province had lower prostate cancer mortality rates than the rest of Austria, where screening did not take place, but the timescale argues against this (as the Tyrol investigators themselves recognize).21 Comparisons between the prostate cancer mortality rates in Connecticut (where screening does not occur) and in Seattle (where it does) have also, so far, failed to show evidence for the benefits of screening.22 There is a need, therefore, to demonstrate the benefits of screening in a formal study before it is introduced as a national policy in countries such as the UK. There are three trials worldwide that will provide the evidence base for screening. The European Randomised Trial of Screening for Prostate Cancer (ERSPC) is a multinational study that has been recruiting since 1993.23 In the USA, the PLCO (Prostate, Lung, Colon, and Ovary) programme has randomized about 80 000 men into the prostate cancer arm of the programme.24 Together, the ERSPC and the PLCO studies hope to recruit around 260 000 men, an enormous undertaking.23 Finally, in the UK, the Prostate testing for cancer and Treatment (ProtecT) study is shortly to launch a companion study (the CAP study), which will make a comparison of prostate cancer mortality having randomized primary-care providers to participate or not to participate in the study.25 The ProtecT study is discussed in more detail below. Can prostate cancer be prevented? The importance of diet in its aetiology suggests that, at least at one level, it can, by returning to the diet for which we have evolved.6 Other
Pathology and staging 647
components of the Asian/Far Eastern diet might also help to prevent prostate cancer, for example soy isoflavinoids.26 Intriguingly, a study has also suggested that selenium supplementation of the diet might also prevent prostate cancer,27 as might other components including lycopenes (from cooked or processed tomatoes) and green tea.16,28,29 A randomized trial of the 5-alpha reductase inhibitor finasteride as a possible chemo-preventative agent (acting by preventing the conversion of testosterone to the active dihydrotestosterone) did, indeed, show a reduction in the incidence of prostate cancer in the treated group. However, it appeared to show an increased proportion of high-risk disease in those who did develop prostate cancer, a finding that continues to excite much debate.30 Patients who remained prostate cancer free at the end of the study (7 years) were biopsied. Prostate cancer was detected in a staggering 24 per cent of the men in the control arm of the study, which must represent the best demonstration yet of our ability to overdiagnose prostate cancer using current methods. If this does not eloquently warn of the dangers of unnecessarily diagnosing and treating men for innocent prostate cancers, nothing will!
PATHOGENESIS Prostatic intra-epithelial neoplasia (PIN) is a histologically recognized condition, so named because it was presumed to be the precursor of prostate cancer, in the same way that cervical intraepithelial neoplasia (CIN) is of cervical cancer. It was graded according to the degree of cellular atypia, but is generally referred to simply as low-grade PIN or high-grade PIN.31,32 However, its relationship to invasive prostate cancer is much less clear than is the case with CIN and cervical cancer. What can be said with some confidence is that any relationship that there is, is likely to be predominantly in high-grade and not in low-grade PIN.33 On the other hand, it is equally clear that not all men with high-grade PIN will go on to develop invasive prostate cancer,33–35 and it may be that it is a marker of common risk factors rather than a true biological precursor per se. A second condition, termed atypical small acinar proliferation, is also recognized, and is said to be more likely to progress to frank carcinoma than is high-grade PIN,33,34 although whether some of these are, in fact, low-grade cancers from the outset is conjectural. The other condition, already referred to, which could be a precursor to prostate cancer is proliferative inflammatory atrophy.36 This will be more difficult to see in needle-biopsy specimens (as opposed to whole-organ mounts from radical prostatectomy specimens). In fact, contrary to expectations, cell turnover in regions of inflammatory atrophy is extremely high, explaining why such an association would make sense. Moreover, such lesions express high levels of glutathione-S-transferase, an enzyme involved in the detoxification of polycyclic aromatic hydrocarbons, consistent with the relationship to inflammation.7,36
Prostate cancer is androgen dependent, that is to say, it does not develop in the absence of testosterone (for example it is exceedingly rare in eunuchs), and growth control, mediated by the androgen receptor, is key to the pathogenesis of the disease. At the same time, other growth factors are of importance, including epidermal growth factor (EGF)37,38 and insulin-like growth factor-1 (IGF-1),39,41 and circulating levels of these can in turn be influenced by diet.28 In understanding the development of prostate cancer, it is a mistake to consider the prostatic epithelium in isolation; the prostate is an example par excellence of an organ in which the behaviour of the epithelium is governed by the stroma, and vice versa.42–45 Prostatic stromal fibroblasts and tumour cells secrete a number of growth factors that impact on the development and progression of prostate cancer, including fibroblast growth factors (FGF),46 vascular-endothelial growth factor (VEGF),47 transforming growth factor Beta(TGF-β)48 and hepatocyte growth factor (HGF).49 It has long been thought that prostate cancers, like other types of cancer, originate in stem cells. Stem cells are rare components of the prostatic epithelium, are situated in the basal layer and are thought to give rise to luminal epithelial cells, luminal secretory cells, and neuro-endocrine cells.50 The phenotypic markers that describe these cell lineages are currently being described. A prostate cancer stem cell has also recently been described, and an understanding of its biology may yield new insights and new targets for cancer therapy.51
PATHOLOGY AND STAGING Nearly all prostate cancers are adenocarcinomas. The Gleason grading system is now almost universally employed for prostatic adenocarcinomas. In its original form, it took as its starting premise that, histologically, there was a major pattern and a minor pattern in prostate cancers. Gleason ascribed a score of 1–5 to each pattern, and the Gleason sum score simply adds them together52 (Fig. 27.2). For example, a Gleason sum score of 3 3 6 implies a score of 3/5 for the major pattern and 3/5 for the minor pattern, making a sum score of 6. In the years following its first
Figure 27.2 cancer
The Gleason histologic grading system for prostate
648 Prostate cancer
Table 27.1 Distribution of Gleason grades in the Cardiff Early Prostate Cancer clinic, 1997–2006 (Courtesy of Professor HG Kynaston). Frequency 3 6 8 41 558 377 84 67 5 1149 80 1229
Percent .2 .5 .7 3.3 45.4 30.7 6.8 5.5 .4 93.5 6.5 100.0
600 500
Frequency
Gleason Sum 2 3 4 5 6 7 8 9 10 Total No details Total
400 300 200 100 0 2
description, there was a relatively even spread of Gleason scores among diagnosed prostate cancers, but there was also wide inter-observer variation in ascribing a score to an individual tumour. In recent years, this variability has lessened considerably, together with an interesting decline in the number of cases that are ascribed a Gleason sum score of less than 6. Indeed, in the Cardiff practice, Gleason 3 3 6 is the most commonly found sum score (Table 27.1). The importance of this is that care is needed in comparing outcomes in a contemporary series of patients with those in a historical series. The Gleason sum scores may be similar, but the patients may be very different. The most striking indication of this stage shift is seen in the re-analysis of the Connecticut series of patients managed by ‘watchful waiting’ – the contemporary histological assessment (which was blinded with respect to the original assessment) shows that Gleason scores 2–5, which were well represented in the original analysis, have now become uncommon, with the majority of cases now being graded as Gleason 3 3 6, even though the analysis has been done on the same series of patients.53 The World Health Organisation has also described a grading system, which is less often used now than it was formerly (when the Gleason system was less established). Tumours are graded I, II and III, (corresponding roughly to Gleason scores 2–4, 5–7 and 8–10 respectively), equivalent to well-differentiated, moderately differentiated and poorly differentiated tumours. Immunohistochemical stains used to identify prostate cancers include PSA, cytokeratins and, more recently, alphamethylacyl-CoA-racemase (AMACR),54 which appears to be a particularly promising marker for distinguishing prostate cancer from benign prostatic disease. Rare tumours of the prostate include small-cell carcinomas, which behave in a similar fashion to small-cell carcinomas at other sites, and which show features of neuro-endocrine differentiation.55 Others, such as sarcomas and lymphomas, are managed according to the principles
3
4
5
6
7
8
9
10
adopted for those tumours at other anatomical sites.56 Transitional-cell carcinomas (TCCs) can arise from the prostatic urethra, and are more likely to behave like muscle-invasive bladder cancers.57,58 Transitional-cell carcinoma can also involve the prostate due to local spread from a bladder tumour,58 and, more rarely, the prostate can be involved in local spread from other pelvic tumours, such as from the rectum or sigmoid colon. Adenocarcinomas typically spread locally, involving the peri-prostatic tissues or the seminal vesicles, and to regional lymph nodes in the pelvis. The distribution of lymph-node metastases has been mapped out using nanoparticleenhanced magnetic resonance imaging (MRI) after pathological confirmation, and shows quite extensive variation; this is an important consideration in therapeutic nodal irradiation59 (Fig. 27.3). Subsequent spread to para-aortic and, on occasions, to supra-diaphragmatic lymph nodes can occur. One of the most striking features of prostate cancer, however, is its peculiar propensity to bone metastases, which are typically osteoblastic, in contrast to the lytic metastases seen, for example, in multiple myeloma or (less strikingly) in breast cancer. This clinical feature must be a reflection of an important biological feature of prostate cancers, and it is clear that there are important interactions between bone cells – especially osteoblasts and, to a lesser extent, osteoclasts60,61 – and prostate cancer cells, with the latter secreting a number of factors, including bone morphogenetic proteins62 and osteoprotegerin.63 Bone metastases are seen particularly in the axial skeleton, in the pelvis and in the proximal femora, and the anatomical distribution was ascribed to the rich venous drainage from the prostate into Batson’s vertebral plexus. This is likely to be only part of the story, however, given the biological aspects of ‘soil and seed’, and bone metastases can also occur elsewhere, notably in the base of skull. It is, however, almost forgotten that previous generations of pathologists, who carried out meticulous post-mortem examinations, recognized that despite the clinical dominance of bone metastases, pathologically,
Clinical features 649
Figure 27.3 Incidence of lymph node metastases by site based on pathological staging in patients with locally advanced disease (See Plate Section.)
visceral metastases do, in fact, occur more widely than is supposed today.64 The TNM staging system is the one most widely in use in clinical practice in the UK65 (Box 27.1), although in the USA, the parallel American Joint Committee on Cancer (AJCC) staging system is also widely used.66 Despite the most recent 2002 TNM staging system, some publications continue to use the previous (1997) system; the important message is to be clear about which system is being used, particularly with regard to staging the primary tumour.
CLINICAL FEATURES Symptoms and signs Early, organ-confined prostate cancer is entirely asymptomatic. There is something uncomfortable in today’s public health ‘message’, at least in the way in which it is interpreted by the public, which essentially asserts that lower urinary tract symptoms (LUTS) should lead to prompt referral and investigation so that an underlying prostate cancer can be picked up as early as possible. This is inherent rubbish, as (a) most LUTS are not due to prostate cancer, and one might just as well say that LUTS should lead to urgent assessment for possible hypertension, (b) if LUTS are due to prostate
cancer, it is likely to be locally advanced (at least), and probably beyond the reach of radical therapy, and (c) the slow tempo of most prostate cancers would suggest that efficient assessment does not exactly require a ‘same-day’ service (although patient anxiety might very legitimately do so) – and the case for screening in prostate cancer has not yet been made. The best reason for prompt assessment of LUTS is that the underlying condition, most commonly benign prostatic hypertrophy (BPH), can be effectively treated and symptoms improved. The digital rectal examination (DRE) is an important part of the assessment of a patient with suspected prostate cancer, but by itself is often inaccurate.67 The TNM staging, most often used in the UK, depends on the DRE assessment of clinical stage, although this is often supplemented with information from other investigations. A nodule may be palpable, and extension of the disease beyond the prostate gland or into the seminal vesicles may also be noted. Locally advanced disease can give rise to LUTS or, more rarely, to other pelvic symptoms such as rectal pain, bleeding or constipation due to rectal involvement, and deep perineal discomfort due to peri-prostatic soft-tissue involvement. The last-mentioned seems to be a feature of ‘PSAnegative’ prostate cancers, i.e. those that produce lesser amounts of PSA, such that serum PSA levels are disproportionately low for the stage of disease.68 Haemospermia is more common as a symptom of locally advanced disease than is recognized, but most cases of haemospermia have causes other than prostate cancer.69 Lymph-node metastases may give rise to lymphoedema in the lower limbs and genitalia and, occasionally, peripheral lymphadenopathy may be palpable in the groins or in the neck. Bone metastases may be asymptomatic in the early stages, but ultimately give rise to bone pain or to other complications – typically pathological fractures and spinalcord compression. The latter is a potentially disastrous complication, the possibility of which the clinician should be perpetually alert to.70 Hypercalcaemia is relatively less common in prostate cancer than in breast cancer, reflecting a lesser degree of bone resorption (although increased bone resorption due to increased osteoclast activity most definitely does occur in prostate cancer). When hypercalcaemia does occur, it may be a reflection of a paraneoplastic syndrome with ectopic parathyroid hormone production.71
Prostate-specific antigen Prostate-specific antigen has evolved from humble beginnings72 to becoming, arguably, the most successful tumour marker in the entire field of oncology. It is produced by more than 95 per cent of all prostate cancers, mirrors disease activity extraordinarily well, and provides a sensitive guide to disease recurrence many years before it becomes clinically apparent. For patients who have had complete ablation of the prostate (i.e. after radical prostatectomy), the PSA will fall to undetectable limits,73
650 Prostate cancer
Box 27.1 The TNM staging of prostate cancer (65) T – Primary tumour TX T0 T1 T1a T1b T1c T2 T2a T2b T2c T3 T3a T3b T4 N – Regional lymph nodes NX N0 N1 M – Distant metastases MX M0 M1
Cannot evaluate the primary tumour No evidence of tumour Clinically inapparent tumour not palpable or visible by imaging Tumour incidental histological finding in 5% or less of tissue resected Tumour incidental histological finding in more than 5% of tissue resected Tumour identified by needle biopsy (e.g. because of elevated prostate-specific antigen) Tumour confined within the prostate Tumour in half or less of one lobe Tumour in more than half of one lobe, but not both Tumour in both lobes Tumour extends through the prostatic capsule Extracapsular extension (unilateral or bilateral) Tumour invades seminal vesicle(s) Tumour fixed or invades adjacent structures other than seminal vesicles Regional lymph nodes cannot be assessed No regional lymph-node metastases Regional lymph-node metastases
M1a M1b M1c
Distant metastasis cannot be assessed No distant metastasis Distant metastasis Non-regional lymph node(s) Bone(s) Other site(s)
although ultra-sensitive assays enable the detection of disease recurrence with PSA levels as low as 0.1 ng/L. Prostatespecific antigen kinetics can also give a useful window on disease behaviour; for example the use of PSA doubling time73,74 (often expressed logarithmically as the time taken for the PSA to double) gives a good indication of tumour proliferation rate. This can aid clinical decision-making.75,76 For example, suppose that a man undergoes a radical prostatectomy at the age of 65, and 7 years later his PSA begins to rise to a level of 0.2 ng/mL with no clinical evidence of disease. If his PSA doubling time was 10 years, it is extremely unlikely that he will suffer clinical consequences of his disease during his lifetime, and it may be safe to monitor him with no active treatment, unless the PSA doubling time were to accelerate. Where PSA has been less successful and more controversial is in the diagnosis of prostate cancer. This is principally due to its lack of specificity, in that most men with a raised PSA have BPH rather than prostate cancer. Furthermore, some men whose PSA is in the ‘normal’ range have prostate cancer on biopsy (maybe not surprisingly, given the incredibly high biological prevalence of the disease). In trying to overcome this, studies have asked whether any other features of PSA might improve its sensitivity/specificity profile. Two facets in particular have been postulated: the free:total ratio, or absolute levels of ‘free’ PSA77 – because
a higher risk of prostate cancer is associated with lower levels of ‘free’ as opposed to ‘complexed’ PSA78,79 – and the PSA doubling time (see above). A third profile – the PSA density, which relates the PSA level to the prostatic volume – is still advocated by some but is less promising if the total PSA is more than 4 ng/mL.80,81
Other tumour markers An older tumour marker – prostatic acid phosphatase – is occasionally helpful, for example in immunochemistry. Two other potential tissue markers, used in research, are prostate membrane-specific antigen (PSMA)82 and prostate stem-cell antigen (PSCA).83 At the moment they are not useful in routine clinical practice. More recently, the PC3a test, which directly measures genes identified in prostate cancer cells, from a urine specimen, is being investigated to determine whether its use in combination with PSA might improve diagnostic sensitivity and specificity.84
DIAGNOSIS AND STAGING The diagnosis of prostate cancer should usually be confirmed histologically, with one possible exception, to be
Management of organ-confined disease 651
discussed later. A histological diagnosis is usually made following a transrectal ultrasound (TRUS)-guided biopsy.85,86 Prostate cancer may be suspected on a TRUS on the basis of a hypo-echogenic area, particularly if it is in the peripheral zone, the most usual site for prostate cancers. If an abnormal region is seen, targeted biopsies can be performed. If not, a series of biopsies is taken (usually four to five from each side, although this varies from one centre to another), and even if targeted biopsies are taken, additional samples will also be taken from elsewhere in the gland. Computed tomography (CT) is not a good modality for diagnosing or staging prostate cancer (in regard to the assessment of the prostate gland itself, that is), being able to detect only gross abnormalities of the gland with local spread. Magnetic resonance imaging gives better information, prostate cancers being visible typically as areas of low signal on T2-weighted sequences. Extension of disease beyond the capsule or into the seminal vesicles can also be seen on MRI. As an investigation it has yet to establish itself routinely in the assessment of the prostate, although as more data are available on the sensitivity and specificity of the latest generation of scanners, it is likely to become more popular.87,88 Either CT or MRI scanning may be used to detect lymph-node spread. Their contribution is limited to a description of the size and shape of lymph nodes; as yet, they can say little about nodal architecture, but it is probable that the combination of MRI with microscopic superparamagnetic nanoparticles, which are taken up in lymph nodes, might prove more successful.89 Technetium-99 bone scanning is a sensitive means of detecting bone metastases, the likelihood of metastases being greater the higher the PSA. Conversely, patients with a PSA below 20 ng/mL have a rate of positive bone scans of less than 1 per cent,90 and many centres therefore omit bone scanning in patients with a PSA in this range. Indeed, the relatively low rate of detection of metastatic disease in general in today’s patient population argues that a more riskbased approach to staging investigations might be justified.91 Is a histological diagnosis always required? It is a good oncological principle that it is, wherever possible. However, in the case of a patient with radiologically demonstrable sclerotic bone metastases, a clinically malignant prostate gland on rectal examination, and a PSA of over 100 ng/mL, the diagnosis of metastatic prostate cancer is almost incontrovertible. Even in the context of a clinical trial, such patients are often treated on the basis of a clinical diagnosis alone,92 and can be spared the indignity and discomfort of a TRUS-guided biopsy.
PRINCIPLES OF MANAGEMENT The major distinction – in terms of management philosophy – is between organ-confined disease and anything more advanced than this. The management of more advanced disease is discussed below, but some preliminary remarks are useful before considering organ-confined disease.
The incredibly high prevalence of prostate cancer has already been stressed. One consequence of this is that the majority of men diagnosed with early (organ-confined) disease could actually have insignificant disease. On the other hand, by definition, those men who die of prostate cancer have significant disease, and a number of these men will have originally been diagnosed with (apparently) organ-confined disease. The problem is that there is, currently, no reliable way of distinguishing between the two – the ‘tiger’, which has the potential to metastasize and cause death, and the ‘pussycat’, which, if left alone, will cause no clinical problems during a patient’s lifetime. Taking this argument further, any study looking at the efficacy of treatment in early prostate cancer carries the risk that the patient population will be dominated by men with indolent disease. This, in turn, could dilute the true efficacy of treatment in men with significant disease and, in randomized trials, might even obscure important differences between different forms of treatment, unless such trials are powered carefully. The assessment of men with organ-confined disease tends to rely on a combination of clinical parameters – Tstage (including those parameters discussed above [diagnosis]), PSA level and Gleason score. It is certainly true that the higher the PSA, the higher the Gleason score, and the larger the tumour bulk, the more likely the patient is to have significant disease (a tiger). This is often a useful basis for clinical decision-making, but it is too often forgotten that this form of assessment is a guide to the probability of disease being significant, and not a measure of whether the disease is actually significant (see Fig. 27.4 for an illustration of this in regard to conservative management).
MANAGEMENT OF ORGAN-CONFINED DISEASE Can we tame the tiger without drowning the pussycat? There are three broad strategies for the management of localized (T1–2) prostate cancer: active monitoring (formerly called ‘watchful waiting’, and sometimes called ‘active surveillance’), radiotherapy (by external beam or brachytherapy) and surgery (by open or laparoscopic radical prostatectomy). For the patient diagnosed today, often solely on the basis of a raised PSA in an otherwise asymptomatic man, with a relatively small, clinically organ-confined tumour, there is no evidence that any one of these approaches is superior in terms of prolonging life. This statement needs to be made clearly and uncategorically. Selecting the ‘right’ treatment for an individual patient is a complex process, which has been plagued by a number of problems. ●
There is a tendency for specialists to recommend their own modality of treatment, which can be confusing for
652 Prostate cancer
Survival Non–Prostate Cancer Mortality Prostate Cancer Mortality
Age at Diagnosis, y 60–64 65–69 Gleason Score 2–4
55–59
70–74
100
0
80
20
60
40
40
60
20
80
0
100 Gleason Score 5
100
0
80
20
60
40
40
60
20
80 100
0 100
0
80
20
60
40
40
60
20
80
Deceased (%)
Alive (%)
Gleason Score 6
100
0 Gleason Score 7 100
0
80
20
60
40
40
60
20
80 100
0 Gleason Score 8–10 100
0
80
20
60
40
40
60
20
80
0
0
5
10
15
20
0
5
10 15 20 0 5 10 Years Following Diagnosis
15
20
0
5
10
15
100 20
Figure 27.4 Competing risk analysis, showing overall survival according to age and Gleason score, in men with localised prostate cancer managed by watchful waiting.100
●
patients when different specialists assert that their own treatment is ‘best’, and relatively few recommend active monitoring.93,95 Patients are subjected to an increasing amount of ‘hype’ around newer treatment modalities. A few years ago this happened with brachytherapy; today it is happening with cryotherapy and high-intensity focused ultrasound. It is not difficult for patients to find information on these modalities on the Internet.96,97
●
Unrandomized case series tend to show excellent results, but whether this is due to excellent treatment or to excellent case selection is unclear. However, to patients, it seems self-evident that it is due to excellent treatment, especially if this view is reinforced by the specialist.
In the UK, the ProtecT study is currently randomizing men diagnosed with organ-confined prostate cancer following an invitation to attend for PSA testing between
Management of organ-confined disease 653
active monitoring, radical prostatecomy and radical radiotherapy.25 This study will recruit from 233 000 men, with the expectation that this will yield around 2000 men with early prostate cancer, who will be randomized. This trial will be the first to inform decision-making for men diagnosed on the basis of a raised PSA. Unrandomized comparisons are dangerous. For example, the extremely large Surveillance, Epidemiology and End Results (SEER) programme published valuable data on outcomes after the treatment of prostate cancer by watchful waiting, prostatectomy or radiotherapy.98 These data cannot be used to compare modalities, nor were they intended to. However, two particularly important observations are worth stressing: first, the relatively poor outcomes for men with high-grade disease, and second, that the presentation of unrandomized results by treatment received over-estimates the benefits of that treatment, in comparison to the results by treatment intent.
Active monitoring and ‘watchful waiting’ In the late 1980s and 1990s, outcomes from patients managed conservatively began to appear in the literature.99 It was clear that, depending on how patients were selected, but especially if they had low-stage, low-grade disease, a substantial proportion would remain alive and well despite not having been treated initially by surgery or radiotherapy.100 Furthermore, for many (although not all) patients, such a treatment approach was acceptable. For some patients, the terminology itself (‘watchful waiting’ was coined in the 1980s) was unacceptable (‘you watch, while I wait to die’). There are two different scenarios that determine the philosophy, and hence the policy (the schedule of visits and the appropriate action to take). The first is a patient with lower grade disease who is elderly and/or has significant co-morbidity. In this patient, the aim of management is to avoid treatment insofar as this is possible, and it is likely that any treatment given will be palliative, or at least will be given in response to symptoms. This is the philosophy of ‘watchful waiting’. The second scenario is a younger patient with no symptoms who has had a low-stage (e.g. T1), lower grade prostate cancer diagnosed. In this patient, active monitoring implies that treatment will be given in response to evidence of disease progression, which might consist of a rising PSA or a higher Gleason score on a subsequent re-biopsy (a procedure that would never be done in the first scenario).101,102 Furthermore, such treatment might very well be radical in intent, assuming that the patient is still suitable for such treatment. There is no universal guide to the best follow-up schedule, or even to the appropriate ‘trigger’ for re-investigation and/or treatment.
Radical prostatectomy Radical prostatectomy involves the removal of the prostate, seminal vesicles and pelvic lymph nodes. The
bladder neck is then re-anastomosed to the urethra, and a catheter is placed in situ for around 10 days.103 In experienced hands, blood loss is usually below the threshold for transfusion, and the length of hospital stay can be as short as 5 days. Postoperative mortality is exceedingly low. There is undoubtedly a ‘learning curve’, the best results depending on experience, but also on the numbers of patients treated by the centre.104,105 The main complications following radical prostatectomy are incontinence and erectile failure. Following a classical radical prostatectomy, the incidence of erectile failure is between 90 per cent and 100 per cent,106 but in recent years it has been possible to perform a nerve-sparing procedure on appropriately selected patients, which is said to preserve potency in up to 80 per cent of patients.107–109 Significant incontinence rates are usually of the order of around 1–2 per cent, but more minor degrees of leakage might occur in around 9 per cent of men,109 particularly in the first year after surgery. Most centres perform radical prostatectomy as an open procedure, with an abdominal incision. However, more recently there have been two variations, which could become more popular in the future. The first is the radical perineal prostatectomy, which has been performed more extensively in continental Europe than in the UK. Its advocates claim that it has advantages in terms of pain, time to recovery, blood loss and ability to perform nervesparing surgery.110 The second variation is the laparoscopic radical prostatectomy, which may offer significant advantages in terms of blood loss,111 requires a smaller incision and is said to allow improved visualization of the neurovascular bundles and operative field. Laparoscopic surgery is technically challenging, and its major drawback is that if there are complications, an open procedure is required as an emergency. This happens to very few patients, but such patients may be somewhat worse off than they would have been with an initial open procedure, and their cancer surgery may have been sub-optimal. Cancer control rates are said to be similar with all three methods of radical prostatectomy, although this has not been tested in a formal randomized trial and relatively small numbers of patients have been treated with these variations in comparison to the numbers treated by the open operation.110,111 The results of the first randomized trial comparing any form of radical therapy to conservative management in early-stage prostate cancer have now been reported. The Scandinavian Prostate Cancer Study Group performed a randomized trial comparing radical prostatectomy to watchful waiting; 695 men were randomized between 1989 and 1999. At a median follow-up of 8.2 years, 83 men in the prostatectomy group have died, compared to 106 men in the watchful-waiting group.112 This is the first level 1 evidence that radical treatment might improve survival in men with prostate cancer. However, there are important differences between the patients enrolled into this study and contemporary patients. Patients were diagnosed somewhat later in the course of their disease than contemporary patients, with
654 Prostate cancer
the majority having T2 tumours, only around 5 per cent being diagnosed by PSA screening, and 17–20 per cent having a PSA level of more than 20 ng/mL at diagnosis. Furthermore, watchful waiting, rather than active monitoring, was performed. It has been estimated that even if reproducible in a contemporary series of patients, the true benefits of surgery in terms of survival might not become apparent for as long as 17 years, based on the time taken for the survival curves to separate in this study, plus the 12 years’ lead time achieved with PSA screening compared with symptomatic presentation.112 Is there a survival benefit for much earlier surgery (i.e. at the time of diagnosis of a PSA-detected cancer) compared to delayed surgery (based on active monitoring)? We still do not know.
Radiotherapy Radiotherapy has been used to treat primary prostate cancer since the pioneering work of Bagshaw, beginning in the 1960s.114 Most initial work, summarizing retrospective studies from this period, used external-beam radiotherapy (ERBT). More recently, prostate brachytherapy has become popular, with the potential to deliver very high doses of radiation to the prostate with an almost ideal dose distribution. EXTERNAL-BEAM RADIOTHERAPY
Patients treated with EBRT should be treated with conformal therapy, since there is level 1 evidence that this approach reduces treatment complications and side effects, and also allows for dose escalation, which cannot be performed safely with conventional open collimators.115 In the USA there has been a steady trend towards ever-increasing doses of EBRT over the last 15 years or so. This has culminated in several randomized studies, four of which have reported116–119 – one of these achieved dose escalation using protons118 – while others are yet to complete or report. These studies indicate lower risks of failure with higher radiation doses, although this is in terms of biochemical failure, as the data are not yet mature enough to comment on overall survival. However, these conclusions are also supported by unrandomized data. In a large multi-institutional review of 4839 patients, higher radiation doses were an independent factor predictive of a lower risk of biochemical failure.120 In the USA, total EBRT doses of 78 Gy or even 80 Gy are commonly given in 1.8-Gy daily fractions. Doses in the UK have remained somewhat lower than this, but there is another important difference in philosophy: in the UK, most patients receiving EBRT will have their treatment preceded by a short (3 months or so) course of androgendeprivation therapy. This policy has principally arisen as a result of the studies in locally advanced disease,121–123 but this philosophy does involve extrapolation of data from those patients to patients with somewhat earlier disease. It is likely that the use of neoadjuvant/adjuvant hormone
therapy plus a lower dose of radiotherapy (e.g. 74 Gy) may well be equivalent to a higher dose (e.g. 80 Gy) without hormone therapy. The usefulness of hormone therapy in patients with earlier stage disease is currently being tested by the European Organisation for Research and Treatment of Cancer (EORTC) 22991 study, in which patients with T1 or T2 carcinoma of the prostate are randomized to radiotherapy plus no androgen-deprivation therapy or to radiotherapy plus 6 months’ treatment with androgen deprivation. A previous EORTC study, comparing 6 months’ and 3 years’ duration of hormone therapy in patients with T1–2 prostate cancer, has completed accrual and is awaiting further maturity before the results are released. A possible approach to adjuvant/neoadjuvant hormone therapy in the future might involve taking a much more ‘risk-based’ approach, in which patients with a higher risk of disease progression (higher T stage, higher Gleason scores, higher PSAs) might be offered radiotherapy plus androgen-deprivation therapy, but patients who are in a low-risk category (e.g. lower Gleason scores and stage T1 disease with a low PSA) might be offered radiotherapy alone. There is another aspect to the radiotherapy dose/ fractionation question, which has arisen empirically, but which may be finding a scientific berth. In some parts of the world (notably in the UK and Canada), some clinicians treat prostate cancer using a high dose per fraction (e.g. a dose of 55 Gy in 20 fractions). A recent randomized trial comparing 52.5 Gy in 20 fractions with 66 Gy in 33 fractions showed no significant difference, but a trend in favour of the 66-Gy arm.124 However, there is a radiobiological argument that states that a higher dose per fraction might be advantageous for the treatment of prostate cancer,125 as it might have the characteristics associated with a ‘late-reacting’ tissue and a low alpha:beta ratio. Currently ongoing trials are evaluating the prospects for getting the best of both worlds: escalating the total radiation dose, but doing so with larger (2 Gy) doses per fraction, escalating the total dose beyond 55 Gy and towards a target of around 70 Gy.126 External-beam radiotherapy for prostate cancer should be CT planned, with the patient supine and a full bladder. There is no uniform agreement as to the optimum field size or target definition. In the ongoing ProtecT Study, the target volumes are defined according to risk status – based on the algorithm derived from data held by the Radiation Therapy Oncology Group (RTOG) in the USA127 – as follows. ●
●
Group L (low risk of seminal vesicle involvement): clinical stages T1b/c or T2a with (PSA [(Gleason score 6) 10]) 15. Group M (moderate or high risk of seminal vesicle involvement): clinical stages T1b/c or T2a with (PSA [(Gleason score 6) 10]) 15. The policy for treatment margins is given in Table 27.2.
Management of organ-confined disease 655
Table 27.2 Definitions of CTV and PTV as used in the UK ProtecT study Phase I
Needle, delivering seeds into prostate
Phase II Catheter in urethra
Group L CTV PTV Group M CTV PTV
Prostate & base SV* (CTV1) No margin beyond organ CTV1 10 mm
Prostate only (CTV2) No margin beyond organ CTV2 5 mm
Prostate & SV (CTV1) No margin beyond organ
Prostate only (CTV2) No margin beyond organ
CTV1 10 mm
CTV2 5 mm
CTV, clinical target volume; PTV, planning target volume; SV, seminal vesicle.
In recent years it has become clear that the prostate gland is capable of a substantial degree of movement, which could affect the accuracy of treatment delivery. In the future, more sophisticated forms of treatment verification with image-guided radiotherapy (IGRT) will refine the techniques for treating this disease still further.128 It has been suggested that there is an increased risk of rectal cancer following prostate radiotherapy,129 but if this occurs at all, the risk is less than 1 per cent, and it is by no means clear that the perceived increase in risk is real.130 Following high-dose conformal radiotherapy, the incidence of significant/severe (RTOG Grade III/IV) late bowel toxicity is of the order of 1–2 per cent.116–118 Erectile function can be preserved in around 60 per cent of patients who are potent before treatment,131 but some suggest that this figure declines with time.132 Incontinence is rare unless the patient has already had other surgical procedures such as a TURP.131,133 BRACHYTHERAPY
Prostate brachytherapy offers the Utopian ideal of a very high radiation dose to the prostate with a rapid fall-off in dose beyond the prostate (due to the inverse square law). The technique involves several steps to ensure a uniform dose distribution.134 Patients have a TRUS scan performed under general anaesthetic in the lithotomy position as a planning study. As a second procedure, also under general anaesthetic, permanent radioactive seeds are implanted into the prostate, usually made of iodine-125 or palladium-103. Their distribution is guided by meticulous planning based on the planning TRUS scan and information from prior CT scanning. A grid system in a perineal mount is used to guide the insertion of radioactive seeds into the prostate (Fig. 27.5). Seeds are usually placed in tandem and are linked to prevent seed migration. Doses can be manipulated in order, relatively, to spare the urethra, which may go some way towards reducing urinary complications. The treatment involves a brief stay in hospital, which is much
Ultrasound probe in rectum for needle guidance
Template to aid accurate placement of the needles delivering the seeds
Figure 27.5 Brachytherapy technique (courtesy of The Prostate Cancer Centre, Stirling Road, Guildford, Surrey)
shorter than for either EBRT or radical prostatectomy. This treatment is usually very well tolerated, requiring minor analgesics, and the reported outcomes are excellent.135–137 The main side effects are urinary, with dysuria and frequency, sometimes even amounting to acute retention due to urethral inflammation and oedema. This comes on in a few weeks as the acute reaction is maximal around 6–8 weeks later, and generally subsides thereafter, although it can sometimes persist to a degree for several months. Brachytherapy is reported to produce potency rates that are as good as, or better than, those with EBRT,138 and has a very low incidence of rectal complications. However, it has not been compared in a randomized trial to either surgery or EBRT and, in common with all other modalities of treatment for prostate cancer, it is important to separate the effects of the treatment from the effects of patient selection. Nonetheless, brachytherapy must be regarded as an important treatment modality in its own right, and on an equal footing to EBRT and surgery. A variation of permanent, low-dose-rate brachytherapy as discussed above is the use of high-dose-rate brachytherapy, which involves a similar planning technique, but in place of permanent seeds, temporary tubes are inserted into the prostate, which are then afterloaded in the same way as for gynaecological and other cancers. This can be performed either as monotherapy139,140 or in combination with EBRT and/or androgen-deprivation therapy.141–143 Indeed, it is on the basis of the results of this form of therapy that the concept of the low alpha:beta ratio (and hence the benefits of a large dose per fraction, as discussed above) for prostate cancer was proposed.142
Novel forms of therapy Several alternative forms of local treatment to the prostate gland have emerged in recent years. Cryotherapy involves the insertion of a thermal probe transperineally into the
656 Prostate cancer
prostate, in a manner analogous to the brachytherapy technique discussed above, and the creation of a region of thermal injury due to the formation of an ice ball around the probe. This is an effective way of killing both normal and tumour cells, but although it is sometimes promoted as an alternative treatment for primary prostate cancer, the data are really too sparse for it to be recommended at present, outside of a formal study setting.144 However, it may well have a role to play in the management of locally recurrent disease following radiotherapy.145 In its early years, cryotherapy was associated with substantial complications, the most severe of which was the development of a rectourethral fistula.146 This is now less common with better control of the ice probe and better selection of patients, although complication rates are inevitably higher when it is used as salvage therapy compared to primary therapy. High-intensity focused ultrasound (HIFU) is another non-invasive technique, which uses an ultrasound probe in the rectum, which can be focused to a specific region of the prostate and generate a thermal injury. The extent of injury induced by the probe can be visualized under ultrasound control and the procedure can be computer controlled. Typically, the procedure, done under a general anaesthetic, involves the creation of a ‘mosaic’ of thermal burns that cover the entire prostate, or the region required. There may be more potential for control of the ultrasound distribution, and even for sparing the neurovascular bundle. As with cryotherapy, more information is needed before this can be recommended outside of a study setting.147,148 The same applies to photodynamic therapy (PDT), for which there has been a resurgence of interest with the development of better light sources and better photo-sensitizers.149
MANAGEMENT OF LOCALLY ADVANCED DISEASE External-beam radiotherapy is the mainstay of management for locally advanced (T3 to T4) prostate cancer. The study led by the EORTC demonstrated that for patients with locally advanced disease who were being managed by EBRT, outcomes were significantly better if they were treated with adjuvant hormone therapy (for 3 years, starting on the last day of radiotherapy).121 These improve-ments were seen not only in terms of disease-free survival, but also in overall survival. Not surprisingly, this led to the standard use of some form of androgen-deprivation therapy (most usually with luteinizing hormone releasing hormone (LHRH) agonists such as Goserelin) in all such patients being treated with EBRT. The story might end there, except for a number of outstanding questions. What is the optimum scheduling of androgendeprivation therapy? What is its optimum duration? What is the optimum dose of radiotherapy? (In the study reported by Bolla,121 the dose was between 65 Gy and 69 Gy.) Additionally, a few patients in the study reported by Bolla had organconfined disease (T2) but with other unfavourable factors, such as a high histological grade.
Two studies performed by the RTOG have also been important. The RTOG 86-10 study randomized men with bulky T2 cancers to radiotherapy alone or to radiotherapy with androgen-deprivation for 2 months before and 2 months after the start of radiotherapy. Outcomes were better for men with low-grade or intermediate-grade (Gleason 2–6) disease, with a significant improvement in cause-specific survival and freedom from progression.150 Other studies have confirmed the benefits in terms of a reduction in the risk of biochemical progression, but not as yet in overall survival.151,152 A second study performed by the RTOG, 85-31, randomized men with locally advanced, and some with loco-regionally advanced, disease to radiotherapy alone or to radiotherapy plus concomitant and long-term hormone therapy. In the most recent analysis of this study, an overall survival benefit has been seen in the adjuvant therapy arm of the study. As in previous analyses, the effects were greatest for men with high Gleason scores8–10 disease, but, in addition, a benefit is now emerging in patients with Gleason score 7.122 It is perhaps understandable that a policy of androgen deprivation in all patients undergoing EBRT is often seen as the standard approach, although, strictly speaking, the evidence comes from patients such as these, with high-risk localized or locally advanced disease. In the context of locally advanced disease, radiotherapy is generally given to the prostate and the seminal vesicles. Whether or not the pelvic lymph nodes should be irradiated has been a matter of debate. An older RTOG randomized trial failed to show a significant benefit for whole pelvic radiotherapy, and undoubtedly such an approach increases toxicity compared with treatment confined to the prostate alone.153 The same was true of irradiation of the para-aortic nodes.154 However, a more recent study, the RTOG 94-13 study, has suggested that, at least in the context of patients being treated with neoadjuvant and concomitant hormone therapy, whole pelvic radiotherapy may offer improvements in progression-free survival.155 These results are the first analysis of this study, and more mature data are needed, but they are important in order to underline the message that treating regional lymph nodes in patients with certain categories of locally advanced (or high-risk localized) prostate cancer has not been absolutely ruled out. In the future, developments in radiotherapy treatment such as intensity-modulated radiotherapy (IMRT) will allow treatment of regional lymph nodes to a far higher dose than previously, and this may yet completely change the management of this disease by radiotherapy. Another approach to radiotherapy dose escalation in patients with locally advanced disease is combined EBRT plus high-dose-rate brachytherapy, as discussed above. Trials are also underway to ascertain whether the addition of systemic therapies such as docetaxel may benefit such patients with high-risk, locally advanced disease, but for the present such therapy should only be administered within the context of a clinical trial. One such trial is the Medical Research Council (MRC) STAMPEDE study, discussed later. In view of the benefits of adjuvant bisphosphonates in diseases
Management of metastatic disease 657
such as breast cancer and myeloma, it was reasonable to hope that such a benefit might also be achievable in prostate cancer. However, an MRC randomized trial of adjuvant clodronate in patients with localized or locally advanced disease has failed to show an improvement in terms of time to development of symptomatic bone metastases or survival.156 With regard to the EORTC study demonstrating the benefits of androgen-deprivation therapy in patients treated with radiotherapy, a moment of reflection is justified. The natural assumption is that combined hormone therapy and radiotherapy is the proven standard of care and must be regarded as the minimum acceptable treatment. Is it? There are no data that prove that the survival benefit in that study depended on the use of both hormones and radiotherapy. Do we know that it was not a result of hormone therapy alone? This was the basis of the National Cancer Institute of Canada/MRC randomized trial, which has now closed to recruitment and which compared long-term androgendeprivation therapy alone versus the same plus radiotherapy. The first results are expected in a few years’ time.157
MANAGEMENT OF RECURRENT DISEASE AFTER RADICAL LOCAL THERAPY Most patients who relapse after radical local therapy (radical prostatectomy or radiotherapy) will do so with a rising PSA as their first, and usually only, indication of disease. For patients relapsing after radiotherapy, the next line of treatment is usually with androgen-deprivation therapy. Before embarking on this, it is worth making some assessment of the rate of PSA climb (e.g. using the PSA doubling time). Occasionally, patients may relapse with a very slow PSA doubling time, and in such patients it is important to weigh up any benefits from early hormone therapy against the adverse effects, particularly against the background of the PSA doubling time, the patient’s age and general fitness as well as his own personal wishes.74 It is possible that novel therapy such as HIFU or cryotherapy might have a role in the management of patients relapsing with local disease following previous radiotherapy, and more data are needed on this (see above). For patients relapsing after radical prostatectomy, the situation is a little different. Following a radical prostatectomy, the expectation is that the PSA will fall to undetectable levels. The definition of what constitutes ‘undetectable’ obviously depends on the sensitivity of the PSA assay, but in most modern centres, ultra-sensitive assays are performed with a lower limit of detection of 0.1 ng/mL. This extremely sensitive assay allows for the identification of patients relapsing biochemically after radical prostatectomy at a very early stage. Prior to the use of such sensitive assays, the recommendations might have been somewhat different. A recent EORTC study randomized patients treated by radical prostatectomy and with histological evidence of disease at the resection margins to immediate postoperative radiotherapy to the prostate bed or to no immediate treatment until clinical (as opposed to
biochemical) evidence of relapse.158 This study indicated the benefit for patients treated with radiotherapy in terms of biochemical progression-free survival, but as yet no benefit in terms of overall survival. How does this translate into a treatment guide for patients with very early biochemical relapse after radical prostatectomy, or for patients today with margin-positive disease, who are at higher risk of relapse but whose relapse is likely to be picked up at a very early stage using ultra-sensitive PSA assays? Furthermore, if EBRT can be used in patients at risk of relapse, or on relapse following radical prostatectomy, should it be given alone or should it be given in combination with some form of androgen-deprivation therapy? For how long? Unsurprisingly, given that none of these questions can be answered on the basis of evidence, the result is a range of opinions, assertions and management policies.159 These questions will shortly be addressed in the international MRC-led study (RADICALS), which will randomize patients to immediate radiotherapy or radiotherapy delayed until early evidence of biochemical failure, and will also randomize patients between no additional androgen-deprivation therapy and different durations of androgen-deprivation therapy. It will, however, be many years before this study reports.
MANAGEMENT OF METASTATIC DISEASE It was through the dramatic effects of orchidectomy, and hormone therapy with oestrogens in patients with metastatic prostate cancer, that the exquisite sensitivity to androgen deprivation was first demonstrated in the 1940s.160 Patients presenting with widespread bone pain could experience dramatic symptomatic and objective improvement within a relatively short space of time. The overall response rate to first-line hormone therapy (be it by orchidectomy, LHRH analogues or, in the older case series, with stilboestrol therapy) is of the order of 85 per cent. Subsequent studies by the Veterans Administration in the USA confirmed these high response rates, but the use of stilbestrol in relatively high doses was associated with an unacceptably high rate of deaths from cardiovascular disease.161 The mainstay of androgen-deprivation therapy remained surgical orchidectomy until the introduction in the 1980s of LHRH agonists,162 which are now the most commonly used form of androgen-deprivation therapy. The LHRH agonists are associated with an initial tumour ‘flare’ in response to a transient rise in serum testosterone,163 which can be prevented using short-term oral anti-androgens (for, say, 3 weeks). Untreated, the consequences for patients with metastatic disease could be severe – incipient renal obstruction becomes renal failure, and incipient spinal cord compression becomes paraparesis. Responses to androgen deprivation last for a median time of only 12–18 months,164 following which the disease enters a hormone refractory phase. From a practical point of view, before labelling a patient as hormone refractory, it may be worthwhile attempting further therapy with second-line hormones
658 Prostate cancer
such as oral anti-androgens, in addition to continuing previous first-line therapy, and a proportion of patients will respond.165 The second-line response is likely to be shorter than the first-line response, but it is occasionally possible to elicit a third-line hormone response, most notably seen as a withdrawal response, if oral anti-androgen is subsequently withdrawn.166 Drugs such as cyproterone acetate and flutamide can result in hepatotoxicity, and should be used with care.167 More recently, there has been a resurgence of interest in stilboestrol, in much lower doses (1–2 mg daily) than was the case in the old Veterans’ Administration studies.168 All forms of first-line androgen-deprivation therapy are assumed to be equivalent in treating advanced prostate cancer, with the exception of bicalutamide, which appears to be slightly less effective than LHRH and is not licensed in that setting as monotherapy.169 For the remainder, the original trials comparing them to orchidectomy are now almost forgotten, but a small warning is in order! Different forms of androgen-deprivation therapy may not be quite as equivalent as is assumed.170,171 There are two other principal variations on the theme of androgen deprivation.171 The first is maximal (combined) androgen blockade (MAB), in which testicular androgen production is abrogated by LHRH agonists or by orchidectomy and adrenal androgen action is blocked by an oral antiandrogen. This appeared to be a significant advance for men with limited-volume metastatic disease,172 but the Prostate Cancer Trialists’ Collaborative Group (PCTCG) metaanalysis of all randomized trials comparing MAB with monotherapy suggests a mere 1.8 per cent trend in improved survival overall, which was non-significant.173 Maximal androgen blockade is little used in the UK, but is much more widely used in the USA. Why? Partly, this may be because the meta-analysis did suggest a significant (p 0.005) 2.9 per cent absolute difference in overall survival when the oral antiandrogen was flutamide or nilutamide. Partly, it is because the PCTCG study spawned a veritable industry in MAB meta-analyses – for a summary, see the literature review for the UK National Institute for Clinical Excellence174 – some of which reported more benefit than the PCTCG study. Not all of these were of the same quality. The other variation on the theme of androgen deprivation is intermittent hormone therapy. This has two theoretical advantages. First, there are undoubted quality-of-life gains to be had from a ‘holiday’ from androgen deprivation, which has substantial side effects, especially in younger men and when longer durations are used. Second, there is, at least in theory, the possibility that intermittent therapy might prolong the time window of hormone sensitivity, by allowing the re-growth of hormone-sensitive cells intermittently.175 Intermittent therapy is feasible, and so far appears to be efficacious enough to justify the randomized trials that are still ongoing, comparing it to continuous therapy.176 Until these trials report, it should be considered experimental, and should only be used in a study setting. When the disease becomes hormone refractory, other options need to be sought. Chemotherapy became a topic of
interest after Tannock showed, in a randomized trial, that the use of mitoxantrone plus prednisolone was superior in relieving pain compared to prednisolone alone.177 Singleagent docetaxel is now a standard regime in hormone-refractory disease, following the publication of two pivotal randomized trials,178,179 which additionally demonstrated, for the first time, a survival benefit for any treatment in hormone-refractory disease, albeit a modest one of a few months. The international TAX 327 trial compared docetaxel with mitoxantrone and prednisone,179 while the Southwest Oncology Group study compared docetaxel plus estramustine with mitoxantrone plus prednisone.178 A typical regime for single-agent docetaxel would be 75 mg/m2 every 3 weeks. Bisphosphonates have already been mentioned in the context of locally advanced disease. In metastatic disease affecting bone, there is every reason to expect that these agents should be beneficial, since there is a very clear element of bone resorption (osteoclastic) in prostate cancer patients, despite the overall picture of osteoblastic (sclerotic) metastatic disease.180 Despite this hope, an MRC randomized trial of clodronate versus placebo in patients with hormone-sensitive bone metastatic prostate cancer showed only a non-significant trend to an improved time to first symptomatic metastases.92 A placebo-controlled randomized trial of zoledronic acid has, however, shown a significant delay in time to a skeletal-related event (radiotherapy for bone pain, pathological fracture, spinal cord compression, or hypercalcaemia) in patients with hormone-refractory disease.181 Interestingly, a significant component of this benefit is the reduction in risk of pathological fractures. In patients with non-metastatic disease, the use of androgen deprivation is associated with a reduction in bone mineral density, particularly with orchidectomy or LHRH agonists in contrast to bicalutamide,182 and this can be reversed, or prevented, by the concomitant use of bisphosphonates.183 All men on long-term androgendeprivation therapy should be assessed for loss of bone mineral density, and should be treated with prophylactic bisphosphonates if they are found to be at risk. The new era of biological therapies is dawning in the world of prostate cancer treatment, albeit more slowly than in some other cancers.165 Despite some disappointments, for example with gefitinib,184 there are many other options, which include COX-2 inhibitors,185 angiogenesis inhibitors,186 endothelin receptor antagonists,187 bcl-2 antisense therapy,188 immunotherapy with dendritic cells189 and others that are variously in development or early-phase clinical trials.190 There are likely to be continuing developments in conventional chemotherapeutic agents, such as the epothilones, which, like docetaxel, target the microtubule and which appear to be promising.191
CONCLUDING REMARKS Is treatment possible for those in whom it is necessary? Is treatment necessary for those in whom it is possible? Sixteen
References 659
years on from the time when Whitmore posed these questions, we must admit that we still do not know the answers. However, this is not due to inactivity, but rather, on the one hand, to the extremely long natural history of prostate cancer, and on the other hand, to therapeutic resistance in patients with hormone-refractory disease. Major trials, currently in progress, should help to answer the key questions related to screening and the treatment of early disease. In advanced disease, new therapeutic options are appearing at a dizzying rate and these should bear fruit. In addition, there is a question of timing. Is it better to give systemic therapy earlier in the natural history of the disease rather than waiting until patients have hormone-refractory disease? A number of trials are addressing this issue, among them the MRC-led STAMPEDE study. In this study, patients with locally advanced or metastatic disease starting first-line androgen deprivation therapy are randomized to no additional treatment or to additional docetaxel, zoledronic acid, celecoxib, docetaxel plus zoledronic acid, or docetaxel plus celecoxib. This trial, and other trials, has just begun. Let us be hopeful that, 16 years from now, we will have some fundamental answers.
KEY LEARNING POINTS ●
●
●
●
●
Active monitoring is a valid option for men with organ-confined prostate cancer, particularly for those in a low-risk group. At its heart is a philosophy of selective, deferred treatment for men whose cancer shows subsequent evidence of progression. Both radical surgery and radical radiotherapy (by external beam or brachytherapy) result in 5-year progression-free survival rates of around 80% overall, and around 90% in low-risk men with organ-confined prostate cancer. There is no evidence that one modality is superior to another in efficacy, nor that immediate treatment is superior to deferred treatment as practiced by contemporary active monitoring, in terms of overall survival rates. There is strong evidence for improved progression-free survival (but as yet, not for overall survival) with high-dose external beam radiotherapy in organ confined disease, and for improved survival with combined radiotherapy plus hormone therapy for locally advanced disease. Radiotherapy will give biochemical control for some men relapsing after radical prostatectomy, but its optimum schedule, combination with hormone therapy, and impact on overall survival remain to be determined. Hormone therapy with some form of androgen deprivation remains the mainstay of treatment
●
for advanced disease, and ongoing trials will determine the benefits of additional systemic therapy at the time of first-line hormone therapy. The side effects of hormone therapy, particularly osteoporosis, need to be considered and prevented where possible. There remains an urgent need for better treatment for men with hormone-refractory disease, though docetaxel has been established as a standard of care for men with bone metastases who are fit enough for chemotherapy.
REFERENCES 1 Cancer Research UK. UK Prostate Cancer Statistics. http://info.cancerresearchuk.org/cancerstats/types/prostate/. 2 Franks LM. Latent carcinoma of the prostate. J Pathol Bacteriol 1954; 68:603–16. 3 Breslow N, Chan CW, Dhom G, et al. Latent carcinoma of prostate at autopsy in seven areas. The International Agency for Research on Cancer, Lyons, France. Int J Cancer 1977; 20:680–8. 4 Whitmore WF, Jr. Natural history of low-stage prostatic cancer and the impact of early detection. Urol Clin North Am 1990; 17:689–97. 5 Nguyen EV. Cancer in Asian American males: epidemiology, causes, prevention, and early detection. Asian Am Pac Isl J Health 2003; 10:86–99. 6 Coffey DS. Similarities of prostate and breast cancer: Evolution, diet, and estrogens. Urology 2001; 57:31–8. 7 Nelson WG, De Marzo AM, DeWeese TL, Isaacs WB. The role of inflammation in the pathogenesis of prostate cancer. J Urol 2004; 172:S6–11; discussion S11–2. 8 Lowe FC, Gilbert SM, Kahane H. Evidence of increased prostate cancer detection in men aged 50 to 59: a review of 324,684 biopsies performed between 1995 and 2002. Urology 2003; 62:1045–9. 9 Edwards SM, Eeles RA. Unravelling the genetics of prostate cancer. Am J Med Genet C Semin Med Genet 2004; 129:65–73. 10 Simard J, Dumont M, Labuda D, et al. Prostate cancer susceptibility genes: lessons learned and challenges posed. Endocr Relat Cancer 2003; 10:225–59. 11 Casey G, Neville PJ, Plummer SJ, et al. RNASEL Arg462Gln variant is implicated in up to 13% of prostate cancer cases. Nat Genet 2002; 32:581–3. 12 van Asperen CJ, Brohet RM, Meijers-Heijboer EJ, et al. Cancer risks in BRCA2 families: estimates for sites other than breast and ovary. J Med Genet 2005; 42:711–9. 13 Liede A, Karlan BY, Narod SA. Cancer risks for male carriers of germline mutations in BRCA1 or BRCA2: a review of the literature. J Clin Oncol 2004; 22:735–42. 14 Angele S, Falconer A, Edwards SM, et al. ATM polymorphisms as risk factors for prostate cancer development. Br J Cancer 2004; 91:783–7.
660 Prostate cancer
15 Hsieh K, and Albertsen PC. Populations at high risk for prostate cancer. Urol Clin North Am 2003; 30:669–76. 16 Crawford ED. Epidemiology of prostate cancer. Urology 2003; 62:3–12. 17 Thompson IM, Ankerst DP, Chi C, et al. Assessing prostate cancer risk: results from the Prostate Cancer Prevention Trial. J Natl Cancer Inst 2006; 98:529–34. 18 Polek TC, Weigel NL. Vitamin D and prostate cancer. J Androl 2002; 23:9–17. 19 Krishnan AV, Peehl DM, Feldman D. The role of vitamin D in prostate cancer. Recent Results Cancer Res 2003; 164:205–21. 20 Verougstraete V, Lison D, Hotz P. Cadmium, lung and prostate cancer: a systematic review of recent epidemiological data. J Toxicol Environ Health B Crit Rev 2003; 6:227–55. 21 Horninger W, Berger A, Pelzer A, et al. Screening for prostate cancer: updated experience from the Tyrol study. Can J Urol 2005; 12 Suppl 1:7–13; discussion 92–3. 22 Lu-Yao G, Albertsen PC, Stanford JL, Stukel TA, WalkerCorkery ES, Barry MJ, et al. Natural experiment examining impact of aggressive screening and treatment on prostate cancer mortality in two fixed cohorts from Seattle area and Connecticut. Bmj 2002; 325:740. 23 Schroder FH, Denis LJ, Roobol M, et al. The story of the European Randomized Study of Screening for Prostate Cancer. BJU Int 2003; 92 Suppl 2:1–13. 24 Andriole GL, Levin DL, Crawford ED, et al. Prostate Cancer Screening in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial: findings from the initial screening round of a randomized trial. J Natl Cancer Inst 2005; 97:433–8. 25 Donovan J, Hamdy F, Neal D, et al. Prostate Testing for Cancer and Treatment (ProtecT) feasibility study. Health Technol Assess 7:2003; 1–88. 26 Ravindranath MH, Muthugounder S, Presser N, Viswanathan S. Anticancer therapeutic potential of soy isoflavone, genistein. Adv Exp Med Biol 2004; 546:121–65. 27 Clark LC, Dalkin B, Krongrad A, et al. Decreased incidence of prostate cancer with selenium supplementation: results of a double-blind cancer prevention trial. Br J Urol 1998; 81:730–4. 28 Wolk A. Diet, lifestyle and risk of prostate cancer. Acta Oncol 2005; 44:277–81. 29 Neill MG, Fleshner NE. An update on chemoprevention strategies in prostate cancer for 2006. Curr Opin Urol 2006; 16:132–7. 30 Thompson IM, Goodman PJ, Tangen CM, et al. The influence of finasteride on the development of prostate cancer. N Engl J Med 2003; 349:215–24. 31 Colanzi P, Santinelli A, Mazzucchelli R, Pomante, R, Montironi, R. Prostatic intraepithelial neoplasia and prostate cancer: analytical evaluation. Adv Clin Path 1998; 2:271–284. 32 Bostwick DG, Qian, J. High-grade prostatic intraepithelial neoplasia. Mod Pathol 2004; 17:360–79. 33 Schlesinger C, Bostwick DG, Iczkowski KA. High-grade prostatic intraepithelial neoplasia and atypical small acinar
34
35
36
37
38
39
40
41
42
43
44
45 46
47
proliferation: predictive value for cancer in current practice. Am J Surg Pathol 2005; 29:1201–7. Girasole CR, Cookson MS, Putzi MJ, et al. Significance of atypical and suspicious small acinar proliferations, and high grade prostatic intraepithelial neoplasia on prostate biopsy: implications for cancer detection and biopsy strategy. J Urol 2006; 175:929–33. Gokden N, Roehl KA, Catalona WJ, Humphrey PA. High-grade prostatic intraepithelial neoplasia in needle biopsy as risk factor for detection of adenocarcinoma: current level of risk in screening population. Urology 2005; 65:538–42. De Marzo AM, Marchi VL, Epstein JI, Nelson WG. Proliferative inflammatory atrophy of the prostate: implications for prostatic carcinogenesis. Am J Pathol 1999; 155:1985–92. El Sheikh SS, Domin J, Abel P, Stamp G, Lalani el N. Phosphorylation of both EGFR and ErbB2 is a reliable predictor of prostate cancer cell proliferation in response to EGF. Neoplasia 2004; 6:846–53. Gregory CW, Whang YE, McCall W, et al. Heregulin-induced activation of HER2 and HER3 increases androgen receptor transactivation and CWR-R1 human recurrent prostate cancer cell growth. Clin Cancer Res 2005; 11:1704–12. Wang Y, Sun Y. Insulin-like growth factor receptor-1 as an anti-cancer target: blocking transformation and inducing apoptosis. Curr Cancer Drug Targets 2002; 2:191–207. Shi R, Berkel HJ, Yu H. Insulin-like growth factor-I and prostate cancer: a meta-analysis. Br J Cancer 2001; 85:991–6. Oliver SE, Barrass B, Gunnell DJ, et al. Serum insulin-like growth factor-I is positively associated with serum prostatespecific antigen in middle-aged men without evidence of prostate cancer. Cancer Epidemiol Biomarkers Prev 2004; 13:163–5. Sung SY, Chung LW. Prostate tumor-stroma interaction: molecular mechanisms and opportunities for therapeutic targeting. Differentiation 2002; 70:506–21. Hall JA, Maitland NJ, Stower M, Lang SH. Primary prostate stromal cells modulate the morphology and migration of primary prostate epithelial cells in type 1 collagen gels. Cancer Res 2002; 62:58–62. Dawson LA, Maitland NJ, Turner AJ, Usmani BA. Stromal-epithelial interactions influence prostate cancer cell invasion by altering the balance of metallopeptidase expression. Br J Cancer 2004; 90:1577–82. Condon MS. The role of the stromal microenvironment in prostate cancer. Semin Cancer Biol 2005; 15:132–7. Cronauer MV, Schulz WA, Seifert HH, Ackermann R, Burchardt M. Fibroblast growth factors and their receptors in urological cancers: basic research and clinical implications. Eur Urol 2003; 43:309–19. Pallares J, Rojo F, Iriarte J, Morote J, Armadans LI, de Torres I, et al. Study of microvessel density and the expression of the angiogenic factors VEGF, bFGF and the receptors Flt-1 and FLK-1 in benign, premalignant and malignant prostate tissues. Histol Histopathol 2006; 21:857–65.
References 661
48 Kaminska B, Wesolowska A, Danilkiewicz, M. TGF beta signalling and its role in tumour pathogenesis. Acta Biochim Pol 2005; 52:329–37. 49 Hurle RA, Davies G, Parr C, Mason MD, Jenkins SA, Kynaston HG, Jiang WG, et al. Hepatocyte growth factor/scatter factor and prostate cancer: a review. Histol Histopathol 2005; 20:1339–49. 50 Collins AT, Habib FK, Maitland NJ, Neal DE. Identification and isolation of human prostate epithelial stem cells based on alpha(2)beta(1)-integrin expression. J Cell Sci 2001; 114:3865–72. 51 Collins AT, Berry PA, Hyde C, Stower M.J, Maitland NJ. Prospective identification of tumorigenic prostate cancer stem cells. Cancer Res 2005; 65:10946–51. 52 Gleason DF. The Veteran’s Administration Cooperative Urologic Research Group: histological grading and clinical staging of prostatic carcinoma. in Urologic Pathology: The Prostate 1977; (ed. Tannenbaum, M.) 171–198 (Lea and Febinger, Philadelphia,. 53 Albertsen PC, Hanley JA, Barrows GH, Penson DF, Kowalczyk PD, Sanders MM, Fine J, et al. Prostate cancer and the Will Rogers phenomenon. J Natl Cancer Inst 2005; 97:1248–53. 54 Jiang Z, Wu CL, Woda BA, et al. Alpha-methylacyl-CoA racemase: a multi-institutional study of a new prostate cancer marker. Histopathology 2004; 45:218–25. 55 Rubenstein JH, Katin MJ, Mangano MM, Dauphin J, Salenius SA, Dosoretz DE, Blitzer PH, et al. Small cell anaplastic carcinoma of the prostate: seven new cases, review of the literature, and discussion of a therapeutic strategy. Am J Clin Oncol 1997; 20:376–80. 56 Shabaik A. Nonepithelial tumors and tumor-like lesions of the prostate gland. Crit Rev Clin Lab Sci 2003; 40:429–72. 57 Cheville JC, Dundore PA, Bostwick DG, Lieber MM, Batts KP, Sebo TJ, Farrow GM, et al. Transitional cell carcinoma of the prostate: clinicopathologic study of 50 cases. Cancer 1998; 82:703–7. 58 Njinou Ngninkeu B, Lorge F, Moulin P, Jamart J, Van Cangh PJ. Transitional cell carcinoma involving the prostate: a clinicopathological retrospective study of 76 cases. J Urol 2003; 169, 149–52. 59 Shih HA, Harisinghani M, Zietman AL, Wolfgang JA, Saksena M, Weissleder R, et al. Mapping of nodal disease in locally advanced prostate cancer: rethinking the clinical target volume for pelvic nodal irradiation based on vascular rather than bony anatomy. Int J Radiat Oncol Biol Phys 2005; 63:1262–9. 60 Blaszczyk N, Masri BA, Mawji NR, et al. Osteoblast-derived factors induce androgen-independent proliferation and expression of prostate-specific antigen in human prostate cancer cells. Clin Cancer Res 2004; 10:1860–9. 61 Loberg RD, Logothetis CJ, Keller ET, Pienta KJ. Pathogenesis and treatment of prostate cancer bone metastases: targeting the lethal phenotype. J Clin Oncol 2005; 23:8232–41. 62 Feeley BT, Gamradt SC, Hsu WK, et al. Influence of BMPs on the formation of osteoblastic lesions in metastatic prostate cancer. J Bone Miner Res 2005; 20:2189–99.
63 Corey E, Brown LG, Kiefer JA, et al. Osteoprotegerin in prostate cancer bone metastasis. Cancer Res 2005; 65:1710–8. 64 Bubendorf L, Schopfer A, Wagner U, et al. Metastatic patterns of prostate cancer: an autopsy study of 1,589 patients. Hum Pathol 2000; 31:578–83. 65 International Union Against Cancer. TNM Classification of Malignant Tumours, (Wiley-Liss, New York, 2002). 66 American Joint Committee on Cancer. AJCC Cancer Staging Manual, (Springer, New York, 2002). 67 Mistry K, Cable G. Meta-analysis of prostate-specific antigen and digital rectal examination as screening tests for prostate carcinoma. J Am Board Fam Pract 2003; 16:95–101. 68 Birtle AJ, Freeman A, Masters JR, Payne HA, Harland SJ. Clinical features of patients who present with metastatic prostate carcinoma and serum prostate-specific antigen (PSA) levels 10 ng/mL: the “PSA negative” patients. Cancer 2003; 98:2362–7. 69 Han M, Brannigan RE, Antenor JA, Roehl KA, Catalona WJ. Association of hemospermia with prostate cancer. J Urol 2004; 172:2189–92. 70 Chen TC. Prostate cancer and spinal cord compression. Oncology 2001; (Williston Park) 15:841–55; discussion 855,859–61. 71 Patel S, Rosenthal JT. Hypercalcemia in carcinoma of prostate. Its cure by orchiectomy. Urology 1985; 25:627–9. 72 Ablin RJ, Pfeiffer L, Gonder MJ, Soanes WA. Precipitating antibody in the sera of patients treated cryosurgically for carcinoma of the prostate. Exp Med Surg 1969; 27:406–10. 73 Small EJ, Roach M, 3rd. Prostate-specific antigen in prostate cancer: a case study in the development of a tumor marker to monitor recurrence and assess response. Semin Oncol 2002; 29:264–73. 74 Klotz LH. PSA recurrence: definitions, PSA kinetics, and identifying patients at risk. Can J Urol 2006; 13 Suppl 2:43–7. 75 Carroll P. Rising PSA after a radical treatment. Eur Urol 2001; 40 Suppl 2:9–16. 76 Eastham JA. Prostate-specific antigen doubling time as a prognostic marker in prostate cancer. Nat Clin Pract Urol 2005; 2:482–91. 77 Saraiya M, Kottiri BJ, Leadbetter S, Blackman D, Thompson T, McKenna MT, Stallings FL, et al. Total and percent free prostate-specific antigen levels among U.S. men, 2001–2002. Cancer Epidemiol Biomarkers Prev 2005; 14:2178–82. 78 Froehner M, Hakenberg OW, Koch R, Schmidt U, Meye A, Wirth MP, et al. Comparison of the clinical value of complexed PSA and total PSA in the discrimination between benign prostatic hyperplasia and prostate cancer. Urol Int 2006; 76:27–30. 79 Jung K, Lein M, Butz H, Stephan C, Loening SA, Keller T, et al. New insights into the diagnostic accuracy of complexed and total prostate specific antigen using discordance analysis characteristics. J Urol 2006; 175:1275–80. 80 Jones TD, Koch MO, Bunde PJ, Cheng L. Is prostate-specific antigen (PSA) density better than the preoperative PSA level in predicting early biochemical recurrence of prostate cancer after radical prostatectomy? BJU Int 2006; 97:480–4.
662 Prostate cancer
81 Stephan C, Stroebel G, Heinau M, et al. The ratio of prostatespecific antigen (PSA) to prostate volume (PSA density) as a parameter to improve the detection of prostate carcinoma in PSA values in the range of 4 ng/mL. Cancer 2005; 104:993–1003. 82 Ghosh A, Heston WD. Tumor target prostate specific membrane antigen (PSMA) and its regulation in prostate cancer. J Cell Biochem 2004; 91:528–39. 83 Jalkut MW, Reiter RE. Role of prostate stem cell antigen in prostate cancer research. Curr Opin Urol 2002; 12:401–6. 84 Haese A, Van Poppel H, Marberger M, et al. The value of the PCA3 assay in guiding decision which men with a negative prostate biopsy need immediate repeat biopsy: Preliminary European data. 2007 European Urology Supplements, 6: 48–48. 85 Spirnak JP, Resnick MI. Transrectal ultrasonography. Urology 1984; 23:461–7. 86 Trabulsi EJ, Merriam WG, Gomella LG. New imaging techniques in prostate cancer. Curr Urol Rep 2006; 7:175–80. 87 Yu KK, Hricak H. Imaging prostate cancer. Radiol Clin North Am 2000; 38:59–85, viii. 88 Jackson AS, Parker CC, Norman AR, et al. Tumour staging using magnetic resonance imaging in clinically localised prostate cancer: relationship to biochemical outcome after neo-adjuvant androgen deprivation and radical radiotherapy. Clin Oncol (R Coll Radiol) 2005; 17:167–71. 89 Harisinghani MG, Barentsz J, Hahn PF, et al. Noninvasive detection of clinically occult lymph-node metastases in prostate cancer. N Engl J Med 2003; 348:2491–9. 90 Chybowski FM, Keller JJ, Bergstralh EJ, Oesterling J.E. Predicting radionuclide bone scan findings in patients with newly diagnosed, untreated prostate cancer: prostate specific antigen is superior to all other clinical parameters. J Urol 1991; 145:313–8. 91 Albertsen PC, Hanley JA, Harlan LC, et al. The positive yield of imaging studies in the evaluation of men with newly diagnosed prostate cancer: a population based analysis. J Urol 2000; 163:1138–43. 92 Dearnaley DP, Sydes MR, Mason MD, et al. A double-blind, placebo-controlled, randomized trial of oral sodium clodronate for metastatic prostate cancer (MRC PR05 Trial). J Natl Cancer Inst 2003; 95:1300–11. 93 Fowler FJ Jr, McNaughton Collins M, Albertsen PC, Zietman A, Elliott DB, Barry MJ, et al. Comparison of recommendations by urologists and radiation oncologists for treatment of clinically localized prostate cancer. Jama 2000; 283:3217–22. 94 Hanna CL, Mason MD, Donovan JL, Barber JP. Clinical oncologists favour radical radiotherapy for localized prostate cancer: a questionnaire survey. BJU Int 2002; 90:558–60. 95 Donovan JL, Frankel SJ, Faulkner A, Selley S, Gillatt D, Hamdy FC, et al. Dilemmas in treating early prostate cancer: the evidence and a questionnaire survey of consultant urologists in the United Kingdom. Bmj 1999; 318:299–300. 96 http://www.internationalhifu.com/. 97 http://www.galilmedical.com/Seednet/main.htm. 98 Lu-Yao GL, Yao SL. Population-based study of long-term survival in patients with clinically localised prostate cancer [see comments]. Lancet 1997; 349:906–10.
99 Chodak GW, Thisted RA, Gerber GS, et al. Results of conservative management of clinically localized prostate cancer. N Engl J Med 1994; 330:242–8. 100 Albertsen PC, Hanley JA, Fine, J. 20-year outcomes following conservative management of clinically localized prostate cancer. Jama 2005; 293:2095–101. 101 Klotz L. Active surveillance for prostate cancer: for whom? J Clin Oncol 2005; 23:8165–9. 102 Hardie C, Parker C, Norman A, Eeles R, Horwich A, Huddart R, Dearnaley D, et al. Early outcomes of active surveillance for localized prostate cancer. BJU Int 2005; 95:956–60. 103 Walsh PC. Radical prostatectomy: a procedure in evolution. Semin Oncol 1994; 21:662–71. 104 Konety BR, Allareddy V, Modak S, Smith B. Mortality after major surgery for urologic cancers in specialized urology hospitals: are they any better? J Clin Oncol 2006; 24:2006–12. 105 Joudi FN, Konety BR. The impact of provider volume on outcomes from urological cancer therapy. J Urol 2005; 174:432–8. 106 Borchers H, Brehmer B, Kirschner-Hermanns R, Reineke T, Tietze L, Jakse G, et al. Erectile function after non-nervesparing radical prostatectomy: fact or fiction? Urol Int 2006; 76, 213–6. 107 McCullough AR. Prevention and management of erectile dysfunction following radical prostatectomy. Urol Clin North Am 2001; 28:613–27. 108 Miyao N, Adachi H, Sato Y, et al. Recovery of sexual function after nerve-sparing radical prostatectomy or cystectomy. Int J Urol 2001; 8:158–64. 109 Stanford JL, Feng Z, Hamilton AS, et al. Urinary and sexual function after radical prostatectomy for clinically localized prostate cancer: the Prostate Cancer Outcomes Study. Jama 2000; 283:354–60. 110 Janoff DM, Parra RO. Contemporary appraisal of radical perineal prostatectomy. J Urol 2005; 173:1863–70. 111 Tooher R, Swindle P, Woo H, Miller J, Maddern G. Laparoscopic radical prostatectomy for localized prostate cancer: a systematic review of comparative studies. J Urol 2006; 175:2011–7. 112 Bill-Axelson A, Holmberg L, Ruutu M, et al. Radical prostatectomy versus watchful waiting in early prostate cancer. N Engl J Med 2005; 352:1977–84. 113 Draisma G, Boer R, Otto SJ, van der Cruijsen IW, Damhuis RA, Schroder FH, de Koning HJ, et al. Lead times and overdetection due to prostate-specific antigen screening: estimates from the European Randomized Study of Screening for Prostate Cancer. J Natl Cancer Inst 2003; 95:868–78. 114 Bagshaw MA, Cox RS, Hancock SL. Control of prostate cancer with radiotherapy: long-term results. J Urol 1994; 152:1781–5. 115 Dearnaley DP, Khoo VS, Norman AR, et al. Comparison of radiation side-effects of conformal and conventional radiotherapy in prostate cancer: a randomised trial. Lancet 1999; 353:267–72.
References 663
116 Pollack A, Zagars GK, Smith LG, et al. Preliminary results of a randomized radiotherapy dose-escalation study comparing 70 Gy with 78 Gy for prostate cancer. J Clin Oncol 2000; 18:3904–11. 117 Dearnaley DP, Sydes MR, Graham JD, et al. Escalated-dose versus standard-dose conformal radiotherapy in prostate cancer: first results from the MRC RT01 randomised controlled trial. Lancet Oncology 2007; 8:475–487. 118 Zietman AL, DeSilvio ML, Slater JD, Rossi CJ, Jr, Miller DW, Adams JA, Shipley WU, et al. Comparison of conventionaldose vs high-dose conformal radiation therapy in clinically localized adenocarcinoma of the prostate: a randomized controlled trial. Jama 2005; 294:1233–9. 119 Peeters STH, Heemsbergen WD, Koper PCM, et al. Doseresponse in radiotherapy for localized prostate cancer: Results of the Dutch multicenter randomized phase III trial comparing 68 Gy of radiotherapy with 78 Gy. Journal of Clinical Oncology 2006; 24:1990–1996. 120 Kuban DA, Thames HD, Levy LB, et al. Long-term multiinstitutional analysis of stage T1-T2 prostate cancer treated with radiotherapy in the PSA era. Int J Radiat Oncol Biol Phys 2003; 57:915–28. 121 Bolla M, Collette L, Blank L, et al. Long-term results with immediate androgen suppression and external irradiation in patients with locally advanced prostate cancer (an EORTC study): a phase III randomised trial. Lancet 2002; 360:103–6. 122 Pilepich MV, Winter K, Lawton CA, et al. Androgen suppression adjuvant to definitive radiotherapy in prostate carcinoma – long-term results of phase III RTOG 85–31. Int J Radiat Oncol Biol Phys 2005; 61:1285–90. 123 Hanks GE, Pajak TF, Porter A, et al. Phase III trial of longterm adjuvant androgen deprivation after neoadjuvant hormonal cytoreduction and radiotherapy in locally advanced carcinoma of the prostate: the Radiation Therapy Oncology Group Protocol 92-02. J Clin Oncol 2003; 21:3972–8. 124 Lukka H, Hayter C, Julian JA, et al. Randomized trial comparing two fractionation schedules for patients with localized prostate cancer. J Clin Oncol 2005; 23:6132–8. 125 Fowler JF. The radiobiology of prostate cancer including new aspects of fractionated radiotherapy. Acta Oncol 2005; 44:265–76. 126 Kupelian PA, Thakkar VV, Khuntia D, Reddy CA, Klein EA, Mahadevan A, et al. Hypofractionated intensity-modulated radiotherapy (70 gy at 2.5 Gy per fraction) for localized prostate cancer: long-term outcomes. Int J Radiat Oncol Biol Phys 2005; 63:1463–8. 127 Roach M, 3rd, Marquez C, Yuo HS, et al. Predicting the risk of lymph node involvement using the pre-treatment prostate specific antigen and Gleason score in men with clinically localized prostate cancer. Int J Radiat Oncol Biol Phys 1994; 28, 33–7. 128 Wu Q, Liang J, Yan D. Application of dose compensation in image-guided radiotherapy of prostate cancer. Phys Med Biol 2006; 51:1405–19. 129 Baxter NN, Tepper JE, Durham SB, Rothenberger DA, Virnig BA. Increased risk of rectal cancer after prostate radiation: a
130
131
132
133
134
135
136
137
138
139
140
141
142
population-based study. Gastroenterology 2005; 128:819–24. Kendal WS, Eapen L, Macrae R, Malone S, Nicholas G. Prostatic irradiation is not associated with any measurable increase in the risk of subsequent rectal cancer. Int J Radiat Oncol Biol Phys 2006. Mantz CA, Song P, Farhangi E, et al. Potency probability following conformal megavoltage radiotherapy using conventional doses for localized prostate cancer. Int J Radiat Oncol Biol Phys 1997; 37:551–7. Litwin MS, Flanders SC, Pasta DJ, Stoddard ML, Lubeck DP, Henning JM, et al. Sexual function and bother after radical prostatectomy or radiation for prostate cancer: multivariate quality-of-life analysis from CaPSURE. Cancer of the Prostate Strategic Urologic Research Endeavor. Urology 1999; 54:503–8. Lee WR, Schultheiss TE, Hanlon AL, Hanks GE. Urinary incontinence following external-beam radiotherapy for clinically localized prostate cancer. Urology 1996; 48:95–9. Blasko JC, Mate T, Sylvester JE, Grimm PD, Cavanagh W. Brachytherapy for carcinoma of the prostate: techniques, patient selection, and clinical outcomes. Semin Radiat Oncol 2002; 12:81–94. Herstein A, Wallner K, Merrick G, I-125 versus Pd-103 for low-risk prostate cancer: long-term morbidity outcomes from a prospective randomized multicenter controlled trial. Cancer J 2005; 11:385–9. Stone NN, Stock RG, Unger P. Intermediate term biochemical-free progression and local control following 125iodine brachytherapy for prostate cancer. J Urol 2005; 173:803–7. Sylvester JE, Blasko JC, Grimm PD, Meier R, Malmgren JA. Ten-year biochemical relapse-free survival after external beam radiation and brachytherapy for localized prostate cancer: the Seattle experience. Int J Radiat Oncol Biol Phys 2003; 57:944–52. Robinson JW, Moritz S, Fung T. Meta-analysis of rates of erectile function after treatment of localized prostate carcinoma. Int J Radiat Oncol Biol Phys 2002; 54:1063–8. Grills IS, Martinez AA, Hollander M, Huang R, Goldman K, Chen PY, Gustafson GS, et al. High dose rate brachytherapy as prostate cancer monotherapy reduces toxicity compared to low dose rate palladium seeds. J Urol 2004; 171:1098–104. Sylvester JE. High-dose-rate versus low-dose-rate monotherapy in the treatment of localized prostate cancer. The case for high-dose-rate monotherapy: An up and coming treatment option for low-risk prostate cancer. Brachytherapy 2006; 5:1–4; discussion 7. Morton GC. The emerging role of high-dose-rate brachytherapy for prostate cancer. Clin Oncol (R Coll Radiol) 2005; 17:219–27. Wang JZ, Li XA, Yu CX, DiBiase SJ. The low alpha/beta ratio for prostate cancer: what does the clinical outcome of HDR brachytherapy tell us? Int J Radiat Oncol Biol Phys 2003; 57:1101–8.
664 Prostate cancer
143 Deger S, Boehmer D, Roigas J, et al. High dose rate (HDR) brachytherapy with conformal radiation therapy for localized prostate cancer. Eur Urol 2005; 47:441–8. 144 Cresswell J, Asterling S, Chaudhary M, Sheikh N, Greene D. Third-generation cryotherapy for prostate cancer in the UK: a prospective study of the early outcomes in primary and recurrent disease. BJU Int 2006; 97:969–74. 145 Touma NJ, Izawa JI, Chin JL. Current status of local salvage therapies following radiation failure for prostate cancer. J Urol 2005; 173:373–9. 146 Shinohara K. Prostate cancer: cryotherapy. Urol Clin North Am 2003; 30:725–36, viii. 147 Uchida T, Ohkusa H, Yamashita H, Shoji S, Nagata Y, Hyodo T, Satoh T, et al. Five years experience of transrectal highintensity focused ultrasound using the Sonablate device in the treatment of localized prostate cancer. Int J Urol 2006; 13:228–33. 148 Pickles T, Goldenberg L, Steinhoff G. Technology review: high-intensity focused ultrasound for prostate cancer. Can J Urol 2005; 12:2593–7. 149 Moore CM, Hoh IM, Bown SG, Emberton M. Does photodynamic therapy have the necessary attributes to become a future treatment for organ-confined prostate cancer? BJU Int 2005; 96:754–8. 150 Pilepich MV, Winter K, John MJ, et al. Phase III radiation therapy oncology group (RTOG) trial 86–10 of androgen deprivation adjuvant to definitive radiotherapy in locally advanced carcinoma of the prostate. Int J Radiat Oncol Biol Phys 2001; 50:1243–52. 151 Granfors T, Modig H, Damber JE, Tomic R. Combined orchiectomy and external radiotherapy versus radiotherapy alone for nonmetastatic prostate cancer with or without pelvic lymph node involvement: a prospective randomized study. J Urol 1998; 159:2030–4. 152 Fellows GJ, Clark PB, Beynon LL, et al. Treatment of advanced localised prostatic cancer by orchiectomy, radiotherapy, or combined treatment. A Medical Research Council Study. Urological Cancer Working Party – Subgroup on Prostatic Cancer. Br J Urol 1992; 70:304–9. 153 Asbell SO, Krall JM, Pilepich MV, et al. Elective pelvic irradiation in stage A2, B carcinoma of the prostate: analysis of RTOG 77-06. Int J Radiat Oncol Biol Phys 1988; 15:1307–16. 154 Pilepich MV, Krall JM, Johnson RJ, Sause WT, Perez CA, Zinninger M, Martz K, et al. Extended field (periaortic) irradiation in carcinoma of the prostate – analysis of RTOG 75-06. Int J Radiat Oncol Biol Phys 1986; 12:345–51. 155 Roach M, 3rd, DeSilvio M, Lawton C, et al. Phase III trial comparing whole-pelvic versus prostate-only radiotherapy and neoadjuvant versus adjuvant combined androgen suppression: Radiation Therapy Oncology Group 9413. J Clin Oncol 2003; 21:1904–11. 156 Mason MD, Sydes MR, Glaholm J, et al. Oral sodium clodronate for nonmetastatic prostate cancer – results of a randomized double-blind placebo-controlled trial: Medical Research Council PR04 (ISRCTN61384873). J Natl Cancer Inst, 2007; 99:765–776.
157 Mason M, Warde P, Sydes M, et al. Defining the need for local therapy in locally advanced prostate cancer: an appraisal of the MRC PR07 study. Clin Oncol (R Coll Radiol) 2005; 17:217–8. 158 Bolla M, van Poppel H, Collette L, et al. Postoperative radiotherapy after radical prostatectomy: a randomised controlled trial (EORTC trial 22911). Lancet 2005; 366:572–8. 159 Morris SL, Parker C, Huddart R, Horwich A, Dearnaley D. Current opinion on adjuvant and salvage treatment after radical prostatectomy. Clin Oncol (R Coll Radiol) 2004; 16:277–82. 160 Huggins C, Hodges CV. Studies on prostatic cancer. I. The effect of castration, of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. 1941. J Urol 2002; 167:948–51; discussion 952. 161 Byar DP. Proceedings: The Veterans Administration Cooperative Urological Research Group’s studies of cancer of the prostate. Cancer 1973; 32:1126–30. 162 Ahmed SR, Brooman PJ, Shalet SM, Howell A, Blacklock NJ, Rickards D, et al. Treatment of advanced prostatic cancer with LHRH analogue ICI 118630: clinical response and hormonal mechanisms. Lancet 1983; 2:415–9. 163 Huben RP. Hormone therapy of prostatic bone metastases. Adv Exp Med Biol 1992; 324:305–16. 164 Vogelzang NJ, Chodak GW, Soloway MS, et al. Goserelin versus orchiectomy in the treatment of advanced prostate cancer: final results of a randomized trial. Zoladex Prostate Study Group. Urology 1995; 46:220–6. 165 Sternberg CN. What’s new in the treatment of advanced prostate cancer? Eur J Cancer 2003; 39:136–46. 166 Caldiroli M, Cova V, Lovisolo JA, Reali L, Bono AV. Antiandrogen withdrawal in the treatment of hormonerelapsed prostate cancer: single institutional experience. Eur Urol 2001; 39 Suppl 2:6–10. 167 Thole Z, Manso G, Salgueiro E, Revuelta P, Hidalgo A. Hepatotoxicity induced by antiandrogens: a review of the literature. Urol Int 2004; 73:289–95. 168 Scherr DS, Pitts WR, Jr. The nonsteroidal effects of diethylstilbestrol: the rationale for androgen deprivation therapy without estrogen deprivation in the treatment of prostate cancer. J Urol 2003; 170:1703–8. 169 Tyrrell CJ, Kaisary AV, Iversen P, et al. A randomised comparison of ‘Casodex’ (bicalutamide) 150 mg monotherapy versus castration in the treatment of metastatic and locally advanced prostate cancer. Eur Urol 1998; 33:447–56. 170 Seidenfeld J, Samson DJ, Hasselblad V, Aronson N, Albertsen PC, Bennett CL, Wilt TJ, et al. Single-therapy androgen suppression in men with advanced prostate cancer: a systematic review and meta-analysis. Ann Intern Med 2000; 132:566–77. 171 Carroll PR, Kantoff PW, Balk SP, et al. Overview consensus statement. Newer approaches to androgen deprivation therapy in prostate cancer. Urology 2002; 60, 1–6. 172 Crawford ED, Eisenberger MA, McLeod DG, et al. A controlled trial of leuprolide with and without flutamide in prostatic carcinoma. N Engl J Med 1989; 321:419–24.
References 665
173 Prostate Cancer Trialist Collaborative Group. Maximum androgen blockade in advanced prostate cancer: an overview of the randomised trials. Lancet 2000; 355:1491–1498. 174 National Institute of Clinical Excellence. Improving Outcomes in Urological Cancers. The Research Evidence. (2002). 175 Bruchovsky N, Klotz LH, Sadar M, et al. Intermittent androgen suppression for prostate cancer: Canadian Prospective Trial and related observations. Mol Urol 2000; 4:191–9;discussion 201. 176 Hurtado-Coll A, Goldenberg SL, Gleave ME, Klotz L. Intermittent androgen suppression in prostate cancer: the Canadian experience. Urology 2002; 60:52–6; discussion 56. 177 Tannock IF, Osoba D, Stockler MR, et al. Chemotherapy with mitoxantrone plus prednisone or prednisone alone for symptomatic hormone-resistant prostate cancer: a Canadian randomized trial with palliative end points. J Clin Oncol 1996; 14:1756–64. 178 Petrylak DP, Tangen CM, Hussain MH, et al. Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N Engl J Med 2004; 351:1513–20. 179 Tannock IF, de Wit R, Berry WR, et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004; 351:1502–12. 180 Clarke NW, McClure J, George NJ. Morphometric evidence for bone resorption and replacement in prostate cancer. Br J Urol 1991; 68:74–80. 181 Saad F, Gleason DM, Murray R, et al. Long-term efficacy of zoledronic acid for the prevention of skeletal complications in patients with metastatic hormone-refractory prostate cancer. J Natl Cancer Inst 2004; 96:879–82. 182 Smith MR, Goode M, Zietman AL, McGovern FJ, Lee H, Finkelstein JS, et al. Bicalutamide monotherapy versus leuprolide monotherapy for prostate cancer: effects on bone mineral density and body composition. J Clin Oncol 2004; 22:2546–53.
183 Smith MR, Eastham J, Gleason DM, Shasha D, Tchekmedyian S, Zinner N, et al. Randomized controlled trial of zoledronic acid to prevent bone loss in men receiving androgen deprivation therapy for nonmetastatic prostate cancer. J Urol 2003; 169:2008–12. 184 Canil CM, Moore MJ, Winquist E, et al. Randomized phase II study of two doses of gefitinib in hormone-refractory prostate cancer: a trial of the National Cancer Institute of Canada-Clinical Trials Group. J Clin Oncol 2005; 23:455–60. 185 Pruthi RS, Derksen JE, Moore D, Carson CC, Grigson G, Watkins C, Wallen E, et al. Phase II trial of celecoxib in prostate-specific antigen recurrent prostate cancer after definitive radiation therapy or radical prostatectomy. Clin Cancer Res 2006; 12:2172–7. 186 Longoria RL, Cox MC, Figg WD. Antiangiogenesis: a possible treatment option for prostate cancer? Clin Genitourin Cancer 2005; 4:197–202. 187 Nelson JB. Endothelin inhibition: novel therapy for prostate cancer. J Urol 2003; 170:S65–7; discussion S67–8. 188 Tolcher AW, Chi K, Kuhn J, et al. A phase II, pharmacokinetic, and biological correlative study of oblimersen sodium and docetaxel in patients with hormone-refractory prostate cancer. Clin Cancer Res 2005; 11:3854–61. 189 Beinart G, Rini BI, Weinberg V, Small EJ. Antigen-presenting cells 8015 (Provenge) in patients with androgen-dependent, biochemically relapsed prostate cancer. Clin Prostate Cancer 2005; 4:55–60. 190 Kantoff P. Recent progress in management of advanced prostate cancer. Oncology (Williston Park) 2005; 19:631–6. 191 Galsky MD, Small EJ, Oh WK, et al. Multi-institutional randomized phase II trial of the epothilone B analog ixabepilone (BMS-247550) with or without estramustine phosphate in patients with progressive castrate metastatic prostate cancer. J Clin Oncol 2005; 23:1439–46 .
28 Colorectal cancer RICHARD H.J. BEGENT, ASTRID MAYER, CHRISTOPHER H. COLLIS AND RUSSELL N. MOULE
Introduction Incidence and cause Diagnosis Pathological features Molecular pathogenesis of colorectal carcinoma Pathology of invasive colorectal cancers
666 666 667 668 670 671
INTRODUCTION Although colorectal cancer remains one of the major killers in developed countries, there are significant advances in risk prediction, screening, surgery, radiotherapy and systemic treatment. This is reflected in improving survival rates, but there is much more to do to overcome the massive mortality still caused by colorectal cancer.
INCIDENCE AND CAUSE In the UK, 34 889 people were diagnosed with colorectal cancer in 2002, with a male-to-female ratio of 1.16:1. In the
Treatment Follow-up after initial treatment Future developments National Health Service guidelines Conclusions References
same year, 16 148 people died of the disease.1 Lifetime incidence is 1 in 18 for men and 1 in 20 for women and is related to age, with nearly 85 per cent of cases being diagnosed in people aged 60 years or over. Survival has doubled over the last 30 years because of advances in early diagnosis and treatment (Fig. 28.1).
Genetic damage causes colorectal cancer Colorectal carcinoma appears to result from the accumulation over time of separate items of genetic damage in a colonic epithelial cell, explaining the increasing incidence with age of sporadic cases and the rarity before the age of
60 Female Male 50
% survival
40
30
20
10
0
1971–75
1976–80
1981–85
1986–90
672 685 686 687 688 688
1991–95 1996–2000 2001–05
Figure 28.1 Age standardised 5 year survival of patients diagnosed with colon carcinoma in England and Wales.
Diagnosis 667
are physical obesity and reduced physical exercise.10 Cigarette smoking is associated with colorectal adenoma formation.11 Regular aspirin consumption has been shown to confer a reduced cancer risk, although the exact dosage and duration are as yet unknown and must be balanced against the possible side effects of stroke and gastrointestinal haemorrhage.12 Obesity has been associated with an increased risk of colon cancer; the risk increasing by an estimated 15 per cent in overweight and 33 per cent in obese people. The conclusions of the Working Group on Diet and Cancer of the Committee on Medical Aspects of Food and Nutrition Policy (COMA) still hold true: ‘There is moderately consistent evidence that diets with less red and processed meat and more vegetables and fibre are associated with reduced risk of colorectal cancer’.13
40 years. Figure 28.2 shows the relationship of incidence to age in England and Wales.2 The inheritance of specific gene defects accounts for 5–15 per cent of cases in recognized family cancer syndromes and results in the development of a carcinoma at an earlier age, usually before 50 years. The molecular genetics by which colonic mucosa progresses through the normal–adenoma–carcinoma sequence were described by Fearon and Vogelstein.3
Diet The association between a high-fibre diet and low incidence of colorectal carcinoma has long been described. Possible explanations include a high-fibre diet leading to increased faecal bulk, with possible dilution of any potential carcinogen and re-absorption and hydroxylation of bile acids.4 Some animal studies support this hypothesis, although it is doubtful whether results based on animals being fed a single fibre component can be extrapolated to humans, given the range and complexity of our diet. Furthermore, the definition of a ‘high-fibre diet’ is not strict in the literature and comprises a heterogenous group of vegetable and plant polymers. However, meta-analysis of more than 60 human dietary studies did support the overall high-fibre hypothesis.5 Interest has also focused on diets rich in vegetables, fruit and plants. A prospective study of 764 343 subjects showed a reduced risk of colorectal cancer in 25 per cent of men and 38 per cent of women with the highest level of consumption of vegetables.6 Garlic appears to have the strongest protective effect against distal colonic cancers in two large cohort series.7,8 Potential anti-carcinogens in such diets include folate, vitamins A, C and E, carotenoids, plant sterols and selenium. A diet rich in red meat7 and high in saturated fat9 is associated with increased risk of colorectal cancer as
DIAGNOSIS Presentation by change in bowel habit, rectal bleeding, abdominal pain, weight loss or symptoms of anaemia should lead to rectal examination and endoscopy. It is important that the whole colon is examined if a carcinoma is located in the lower bowel, because of the possibility of a synchronous tumour elsewhere in the colon. In the elderly or frail, the use of air-contrast computed tomography (CT) colonography is a sensitive and specific alternative,14 and a case can also be made for this technique as an alternative in all adults. Computed tomography or magnetic resonance imaging (MRI) of the abdomen and pelvis and CT of the thorax provide the most accurate practical staging investigation. Serum carcinoembryonic antigen (CEA) should be measured as a prognostic indicator and as a basis for future monitoring.
4,000
400
2,000
200 1,000
0
Age at diagnosis
85+
80–84
75–79
70–74
65–69
60–64
55–59
50–54
45–49
40–44
35–39
30–24
25–29
20–24
15–19
10–14
5–9
0 0–4
Number of cases
3,000
Rate per 100,000 population
600 Male cases Female cases Male rates Female rates
Figure 28.2 Number of new cases and agespecific incidence rates, by sex, bowel cancer, UK 2002. From Cancer Research UK Cancer Statistics (http://info.cancerresearchuk.org/cancerstats/)
668 Colorectal cancer
PATHOLOGICAL FEATURES Polyps A polyp is a protrusion of a circumscribed lesion into a hollow viscus. Colorectal polyps may be classified according to their origin: epithelial (adenomatous and hyperplastic), hamartomatous (such as Peutz–Jeghers and juvenile polyps), inflammatory, lymphoid and mesenchymal (such as lipomas). The commonest polyps reported by pathologists are adenomatous and hyperplastic lesions. Normal large-bowel epithelium has a turnover of approximately 6 days. The proliferative compartment is restricted to the lower third of the crypt. As cells divide and migrate from this zone they differentiate and lose the ability to divide. In adenomatous lesions, the proliferative compartment enlarges so that the entire crypt and surface may be involved, which, combined with increased longevity of the cells/reduced cell death, results in crypt extension and branching, producing a tubular architecture. In those lesions where there is mesenchymal proliferation as well, a villous architecture results in a tubulovillous adenoma. Adenomatous polyps are dysplastic by definition. Whilst many polyps are stalked, some are sessile lesions. They are classified according to their architecture and degree of dysplasia. Tubular adenomas are composed of 75 per cent tubular glands, villous lesions 50 per cent villous architecture and tubulovillous adenomas 25–50 per cent villous pattern.15 The degree of dysplasia is classified as mild, moderate or severe and is based on the cytological features of the lining epithelial cells together with assessment of the neoplastic architecture. Thus severely dysplastic adenomas show loss of polarity, nuclear enlargement, prominent nucleoli, numerous mitoses and an often complex branching and cribriform pattern of the glands.16 Adenomas are benign neoplasms with a malignant potential. The risk of invasive malignancy is related to the size, architectural subtype, degree of dysplasia and number of adenomatous lesions in any one patient, although not all adenomas will develop into cancer and the exact risk and time course are not known.17 Thus those adenomas that are less than 1 cm in maximal dimension have less than 1 per cent risk of invasive malignancy, compared to 10 per cent of adenomas between 1 cm and 2 cm in diameter and 50 per cent of adenomas larger than 2 cm.18 Increasing malignant potential is noted for those lesions with a villous architecture and severe dysplasia – the latter is associated with a greater incidence of aneuploidy.19 Any polyp removed endoscopically should always be submitted for histological examination. Those polyps classified as adenomatous should be categorized by the pathologist in terms of architecture and dysplasia. Completeness of excision cannot always be evaluated due to poor orientation and fragmentation of the sample. The reporting pathologist should exclude co-existent invasive malignancy. Features favouring invasive transformation include a desmoplastic stroma surrounding
infiltrating neoplastic glands with a destructive growth pattern including invasion through the muscularis mucosae. It is widely accepted that typical invasive colorectal carcinomas develop through the normal-adenoma–carcinoma sequence – a morphological corollary of the multistep theory of neoplasia. Evidence supporting this sequence includes the following. ●
●
●
●
Subjects with familial adenomatous polyposis (FAP) syndrome develop adenomas at an earlier age than is the case with carcinoma. Adenomas are six times more common in colorectal cancer resections than in control subjects without carcinoma (matched for age, sex and segment).20 Adenomas are found in 75 per cent of resection specimens for synchronous cancers21 and patients with previous adenomas have a greater risk of metachronous cancer development.22 Residual adenomatous change may be found adjacent to invasive carcinoma. This is unusual when examining large tumours histologically, presumably reflecting destruction of the ‘parent’ adenoma by the infiltrating tumour edge.
Hyperplastic polyps are not neoplastic and result from a defect in epithelial cell maturation, which produces their characteristic histological appearance. These are distinct from serrated adenomas (mixed hyperplastic adenomatous polyps), which combine the low-power architectural appearance of hyperplastic polyp with the cytological atypia of an adenoma. Longacre and Fenoglio-Preiser23 described malignant transformation in 10 per cent of serrated adenomas in a series of 110 cases. However, it must be stressed that these were described as ‘intramucosal carcinomas’, which is a term not recognized in the UK and is equivalent to severe dysplasia in our nomenclature. Consequently the risk of invasive neoplastic transformation is still unknown.
Associated conditions ULCERATIVE COLITIS
It is well recognized that patients with early-onset, longstanding and extensive ulcerative colitis (UC) have an increased risk of colorectal carcinoma.24 Tumours arising in this setting are multifocal, flat with diffusely infiltrating edge and poorly differentiated microscopically. Although these tumours typically arise from a pre-invasive dysplastic stage, more than 95 per cent of dysplasias occur in endoscopically normal areas or foci with ill-defined velvety nodular mucosal thickening and are quite distinct from the adenomatous polyps described above in the non-colitic population.25 Microscopically these mucosal areas may be completely flat or have a mixed villiform and tubular architecture and are lined by variably dysplastic epithelium. Making a confident diagnosis of dysplasia can be difficult for the pathologist, as atypical areas often arise in bowel
Pathological features 669
involved in active inflammation. In this situation regenerative epithelial atypia may mimic dysplasia and the pathologist may be wise to issue a guarded report with advice to re-biopsy following treatment. A report of unequivocal high-grade dysplasia should prompt consideration of that patient for colectomy, so close liaison between pathologist and clinician is advisable. Less than 5 per cent of patients with ulcerative colitis develop polypoid dysplasia, known as ‘dysplasia-associated lesion or mass’ (DALM). These are endoscopically complex villiform mucosal lesions and are distinct from the usual adenomatous polyps. Histologically they are characterized by a complex arborizing villiform and tubular structure lined by variably dysplastic epithelium. The identification of DALM carries an ominous prognosis – Blackstone et al.26 described co-existent carcinoma in 67 per cent of patients with DALM – and therefore patients should be considered for urgent colectomy. HEREDITARY NON-POLYPOSIS COLORECTAL CANCER
Hereditary non-polyposis colorectal cancer (HNPCC) is an autosomal dominant hereditary disorder associated with cancers that may occur in multiple sites but that share the feature of DNA mismatch repair deficiency genes hMSH2,
hMLH1, hPMS1 and hPMS2 on chromosomes 2p and 3p. Subjects are at an increased risk of developing colorectal cancer. The average age of diagnosis is 45 years and the overall lifetime risk is 80 per cent in gene carriers.27 Although the disorder is not associated with the almost carpeting of adenomas seen in FAP, it is still thought that the tumours arise through the adenoma–carcinoma sequence, with the caveat that the adenomas appear more ‘aggressive’, with a higher incidence of large size, villous architecture and severe dysplasia.28 In addition, the resultant invasive carcinoma tends to be right-sided, with mucinous differentiation and tumour-infiltrating lymphocytes on microscopic examination, which, in a young patient, should prompt the reporting histopathologist to suggest a diagnosis of HNPCC if this has not already been considered clinically.29 Hereditary non-polyposis colorectal cancer is also associated with endometrial, ovarian, gastric, pancreatic and renal malignancies. The Muir–Torre syndrome is a particular variant in which colorectal cancer is associated with sebaceous tumours and keratoacanthomas.30 The value of several of the newer proposed diagnostic criteria in identifying subjects with mutations in HNPCC-associated mismatch repair genes has been evaluated by Syngal et al.31 (Table 28.1). Of 70 families, 28 fulfilled the Amsterdam Criteria, 39 the Modified Amsterdam Criteria, 34 the Amsterdam II
Table 28.1 Clinical criteria for hereditary non-polyposis colorectal cancer Name
Criteria
Amsterdama
Three relatives with CRC, one of whom is a first-degree relative of the other two; CRC involving at least two generations; one or more CRC cases diagnosed before the age of 50
Modified Amsterdamb
(1) Very small families, which cannot be further expanded, can be considered as HNPCC even if only two CRCs in first-degree relatives; CRC must involve at least two generations, and one or more CRC cases must be diagnosed under age 55 or (2) In families with two first-degree relatives affected by CRC, the presence of a third relative with an unusual early-onset neoplasm or endometrial cancer is sufficient
Amsterdam II
Three relatives with an HNPCC-associated tumour (CRC, endometrial, small bowel, ureter or renal pelvis), one of whom is a first-degree relative of the other two; involving at least two generations; one or more cases diagnosed before the age of 50
Bethesdab
(1) Subjects with cancer in families that fulfil Amsterdam criteria (2) Subjects with two HNPCC-related cancers, including synchronous and metachronous CRCs or associated extra-colonic cancers (3) Subjects with CRC and a first-degree relative with CRC and/or HNPCC-related extra-colonic cancer and/or colorectal adenoma; one of the cancers diagnosed at age 45 years and the adenoma diagnosed at age 40 years (4) Subjects with CRC or endometrial cancer diagnosed at age 45 years (5) Subjects with right-sided CRC with an undifferentiated pattern (solid/cribriform) on histopathology diagnosed at age 45 years (6) Subjects with signet-ring cell-type CRC diagnosed at age 45 years (7) Subjects with adenomas diagnosed at age 40 years
Modified from Syngal et al.31 a All criteria must be met. b Meeting all features listed under any of the numbered criteria is sufficient. HNPCC, hereditary non-polyposis colorectal cancer; CRC, colorectal cancer.
670 Colorectal cancer
Criteria, and 56 fulfilled at least one of the seven Bethesda Guidelines for the identification of HNPCC patients. The sensitivity and specificity of the Amsterdam Criteria were 61 per cent and 67 per cent. The sensitivities of the Modified Amsterdam and Amsterdam II Criteria were 72 and 78 per cent, respectively. Overall, the most sensitive criteria for identifying families with pathogenic mutations were the Bethesda Criteria, with a sensitivity of 94 per cent, but a specificity of only 25 per cent. Use of only the first three criteria of the Bethesda Guidelines was associated with a sensitivity of 94 per cent and a specificity of 49 per cent. FAMILIAL ADENOMATOUS POLYPOSIS SYNDROME
This is an autosomal dominant inherited disorder characterized by mutations in the APC gene located on chromosome 5.32 The function of the APC protein is unknown, although it does interact with the cell adhesion molecule catenin.30 APC gene mutation is an early event in the development of colorectal cancer. In FAP, the colorectum is literally carpeted with hundreds and thousands of adenomatous polyps. These typically occur at a mean age of 16 years, and unless prophylactic colectomy supervenes, 90 per cent of affected subjects will have one or more invasive cancers by 45 years of age.33 Familial adenomatous polyposis has three variants. In Gardner’s syndrome, colorectal adenomatous polyps are associated with osteomas and desmoid tumours. Turcot’s syndrome is characterized by colonic polyposis associated with central nervous system tumours, which include medulloblastomas and ependymomas. Interestingly, one-third of these subjects have mutations in DNA mismatch repair genes and develop glioblastoma multiforme.34 Attenuated FAP is a variant of FAP characterized by far fewer adenomas, which are typically right-sided.35 The risk of colonic cancer is high but not absolute, as compared to FAP. In addition, colorectal cancer commonly occurs in families not otherwise affected by polyposis syndromes. There is a two-fold to four-fold increased risk for a family member if a first-degree relative has colorectal cancer,33,36 which is further increased if more than one first-degree relative is affected and they are less than 50 years of age. As yet, a specific genetic mutation(s) is not known. Pilot studies in colorectal cancer patients under 30 years old have shown that 41 per cent have MMR gene mutations.37 The risk for people with these mutations of developing colorectal cancer by the age of 70 is approximately 91 per cent for men and 69 per cent for women.38 A number of polymorphisms with a moderate effect on risk have been described, including variants in the APC, HRAS1-VNTR and MTHFR genes.39
MOLECULAR PATHOGENESIS OF COLORECTAL CARCINOMA The histological progression of normal large-bowel mucosa through adenoma to carcinoma is a morphological corollary
of a multistep sequence of genetic mutations. The earliest mutation is that of the tumour suppressor gene APC, located on the long arm of chromosome 5. Wild-type APC encodes a cytoplasmic protein that is found in highest density in the upper portion of crypt epithelium. It interacts with beta-catenin and is thought to be important in the transmission of contact-inhibition stimuli into cells.40 APC is a tumour suppressor gene and therefore acts in a recessive manner. Familial adenomatous polyposis subjects inherit a germline mutation of one APC allele. The mutant APC gene protein product may form a heterodimer with the APC gene protein product from the unaffected allele, which prevents normal function. This mechanism may also occur in sporadic colorectal cancer cases following a somatic mutation of one allele. In addition, mutation of one APC gene allele may be followed by deletion of the other normal allele. This may be a consequence of genomic instability with flawed segregation of chromosomal material during mitosis. This loss of heterozygosity (LOH) results in loss of normal tumour suppressor function with clonal expansion and proliferation. Loss of heterozygosity is an important mechanism in the abolition of a normal tumour suppressor gene product function. APC mutations have been found in adenomas less than 1 cm in size, and the rate of mutation is similar in adenomas and carcinomas, further endorsing the role of APC as the initial step in tumorigenesis.41 This is then followed by mutation of the K-ras proto-oncogene. The wild-type gene encodes a protein with GDP/GTP-binding domains and is important in intracellular signal transduction.42 As proto-oncogenes act in a dominant manner, mutation in only one allele is required to inhibit normal function (conversion to oncogene). Mutant K-ras protein permits stabilization of the active gene product, with unregulated cell growth and therefore clonal expansion of the adenoma and an overall increase in size. Subclones then develop within this enlarging adenoma that contain further mutations, including deleted in colorectal cancer gene (DCC) and P53 genes. The DCC tumour suppressor gene is located on chromosome 18 and encodes a transmembrane protein with structural homologies to the cell adhesion molecule NCAM.43 Mutation in this gene confers a metastatic capacity to the tumour cells.43,44 Wild-type P53 is a tumour suppressor gene and is the most widespread mutation found in human cancers.45 It is located on the short arm of chromosome 17 and is important in the control of the cell cycle, programmed cell death and DNA repair and synthesis.46 Thus mutation in P53 allows unregulated cell growth. Genetic studies indicate that P53 loss is a late event, occurring after mutation in DCC and corresponding histologically to the development of invasive carcinoma in an adenomatous polyp.47 OMIM48 lists 288 genetic variations (mutations, deletions, altered copy number or chromosomal aberrations) in colorectal cancer and the NCBI Single Nucleotide Polymorphism Database lists 6772 single nucleotide polymorphisms (SNPs) related to colon cancer.49 Activating mutations in the oncogene KRAS and mutation or loss of
Pathology of invasive colorectal cancers 671
the tumour suppressor genes APC, SMAD4 and TP53 are reported in colorectal cancers, and APC mutations appear to be the earliest step in colorectal carcinogenesis, with both copies of the APC gene inactivated in up to 80 per cent of sporadic cases.41 Groups of related mutations (haplotypes) also appear to be important in many biological states and present a rich area for research in colorectal cancer.50
commoner gastric signet-cell carcinoma or lobular carcinoma of the breast in women should be excluded before rendering a diagnosis.
Gene expression microarrays, epigenetics and proteomics
STAGING
Although the technology and methods of analysis are not fully developed, it is becoming clear that patterns of gene expression have the potential to discriminate between normal tissue, adenomas and carcinomas and to help in the prediction of tumour behaviour in terms of metastaic potential and response to therapy.51,52 Similarly, epigenetics has a role in carcinogenesis and, alongside genetic changes and proteomics, provides an analysis of the cancer characteristics that includes functionally important post-translational modification. Rapid progress is being made in these fields and it is very likely that epigenetic and proteomic signatures of colorectal cancer will be elucidated (for review see references 53 and 54).
PATHOLOGY OF INVASIVE COLORECTAL CANCERS Macroscopic appearance The gross morphology of colorectal cancers may be polypoid, fungating, ulcerative or diffusely infiltrating. Approximately 60 per cent of invasive carcinomas arise in the sigmoid colon and rectum, although an increase of rightsided colonic cancers from 18.7 per cent to 27.5 per cent and a decrease in left-sided tumours from 72.1 per cent to 62.5 per cent have been described in recent years.55
Microscopic appearance Microscopically, the tumours may be classified into well, moderate or poorly differentiated adenocarcinomas. Scattered neuroendocrine and Paneth cells may be evident, although typically they are few in number. Rare variants include the poor-prognosis mucinous carcinoma characterized by strips and sheets of neoplastic glandular epithelium bathed in lakes of mucin (the latter more than 50 per cent of the tumour). Undifferentiated carcinoma is characterized by sheets of undifferentiated malignant cells with vesicular nuclei and prominent nucleoli typically located in the caecum. Despite its ‘aggressive’ microscopic appearance, the tumour has a good prognosis.56 Small-cell carcinoma is an aggressive tumour histologically indistinguishable from its pulmonary counterpart and with a similar biological profile.57 Some 50 cases have been reported in the literature of primary colorectal signet-cell carcinoma.58 The
Prognostic factors
The mucosal origin allows tumour to infiltrate cirumferentially and longitudinally along the bowel. Typically, however, macroscopic assessment correlates well with microscopic spread such that a 2 cm distal margin of excision is adequate, as the risk of unrecognized tumour growth beyond this is negligible. Indeed, Royal College of Pathology recommendations state that distal resection margins greater than 2 cm from the colorectal primary do not need to be routinely sampled by the pathologist.59 The importance of complete circumferential excision for rectal tumours must be stressed and represents total mesorectal excision by the surgeon. The circumferential margin is the area below the peritoneal reflection anteriorly that continues on to the posterior wall, where it extends superiorly into the triangular ‘bare area’ that runs up to the sigmoid mesocolon. In pathological terms, the circumferential margin is said to be involved if tumour is less than 1 mm from it, be it through direct continuity of spread from the main lesion, tumour emboli in lymphatics or lymph nodes or by tumour deposits discontinuous from the main growth.59 Positive circumferential margin involvement is the main determinant of local recurrence in rectal carcinoma.60 The 5-year survival rate of rectal carcinoma varies between 40 per cent and 70 per cent and is due mainly to local recurrence of between 5 per cent and 40 per cent in potentially curative procedures.61 The high spatial resolution MRI of the rectum makes it an accurate method of staging rectal cancer. The technique requires attention to detail so that correct planes and scan parameters are used to obtain the best images. A detailed understanding of the pathological features of these tumours is required for image interpretation so that prognostic information beyond the basic T and N staging of the tumour can be obtained. The use of standardized criteria for reporting is reproducible in the multicentre setting and preoperative multidisciplinary discussion of the MRI features increases the number of operations performed with tumourfree resection margins.62 Colorectal cancer may spread by one of several routes, including locally with infiltration of contiguous tumour into adjacent organs, transcoelomically, via lymphatics and veins. The vast majority of the colon (excepting part of the caecum) and the anterior wall of the rectum to a variable length are covered by serosa. Ulceration of this either macroscopically or microscopically by tumour constitutes a breach of the peritoneum and a route for the dissemination of malignant cells into the abdominal cavity. Shepherd et al. showed that those colonic tumours with microscopic serosal surface ulceration by tumour resulted in peritoneal
672 Colorectal cancer
recurrence in 50 per cent of cases.63 Tumour spread via lymphatics to regional nodes forms an integral component of the various staging systems. Although typically metastases involve nodes in a sequential fashion corresponding to the anatomical sequence of drainage, in advanced carcinoma cases blockage ‘upstream’ by tumour of a node may result in retrograde lymph-node spread. The number of lymph nodes involved may affect the prognosis. The 5-year survival for Dukes C with one lymph node positive is 63.6 per cent, but falls to 2.1 per cent when more than ten nodes are involved, and further endorses the care needed in the initial sampling of the specimen by the pathologist.64 Although interest has recently focused on the use of ancillary techniques to detect nodal micro-metastases, O’Brien et al. showed that routine analysis of haematoxylin and eosin sections of lymph nodes was equally as sensitive in picking up tumour involvement as the use of CEA and EMA immunohistochemical techniques.65 Moreover, those studies that have gone on to re-stage patients after immunohistochemical examination for micro-metastases have failed to show any difference in 5-year survival.66 Furthermore, tumour may embolize via the portal system to the liver and, in low rectal carcinomas, via inferior haemorrhoidal veins direct to the lungs. Prognosis is adversely affected when thickwalled calibre extra-mural veins are involved, although this does not appear to be an independent prognostic variable.67 Dukes staging is the most important prognostic indicator for subsequent patient management. This eponymous system, described in 1928, initally referred to rectal carcinomas.68 It is now applied to all colorectal cancers. Dukes A tumours are those invasive carcinomas that infiltrate into the submucosa or muscularis propria but do not breach the bowel wall and are without lymph-node metastases (with a 93 per cent 5-year survival); Dukes B refers to those tumours that extend into mesorectal or pericolic fat in continuity without lymph-node involvement (associated 65 per cent 5-year survival); and Dukes C are those tumours with regional lymph-node involvement (23 per cent 5-year survival).69 Subsequent to this, the Dukes classification has been modified and it is now accepted for all colorectal carcinomas. Dukes C may be further subdivided into C1 (apical nodenegative) and C2 (apical node-positive) tumours with concomitant 5-year survival of 40.9 per cent and 13.6 per cent respectively.64 Turnbull et al. added the Dukes D category, representing the presence or absence of distant metastases.70 The American Joint Commission on Cancer (AJCC) has developed a staging system based on similar parameters and assessed by the extent of local invasion (T1, submucosa; T2, muscularis propria; T3, beyond muscularis propria; and T4, tumour cells breaching the peritoneal surface or invading adjacent organs), lymph-node status (N0, no lymph nodes involved; N1, fewer than three lymph nodes involved; and N2, more than three lymph nodes positive), and presence or absence of distant metastases (M0/M1).71 The AJCC72 reported a multidisciplinary consensus conference using published literature to develop an arbitrary classification system of prognostic markers. They concluded
that several T categories should be subdivided: pTis into intraepithelial carcinoma (pTie) and intramucosal carcinoma (pTim); pT1 into pT1a and pT1b, corresponding to the absence or presence of blood or lymphatic vessel invasion, respectively; and pT4 into pT4a and pT4b according to the absence or presence of tumour involving the surface of the specimen, respectively. OTHER PROGNOSTIC FACTORS
The AJCC Working Party72 also recommended that TNM groups be stratified based on the presence or absence of elevated serum levels of CEA (5 ng/mL) on preoperative examination. In addition, the working party concluded that carcinoma of the appendix should be excluded from the colorectal carcinoma staging system because of fundamental differences in natural history. Although many molecular and oncogenic markers showed promise to supplement or modify the current staging systems eventually, to the authors’ knowledge none has yet been evaluated sufficiently to recommend its inclusion in the TNM system. The pathologist plays a major role in the staging of colorectal cancers. It is only with meticulous examination of the gross specimen that all the prognostic parameters just described can be assessed. The most advanced molecular techniques employed on microscopic sections will not make up for poor sampling of the initial resection specimen. Staging systems are shown in Boxes 28.1 and 28.2 and relationship to serum CEA in Table 28.2.
TREATMENT Screening SIGMOIDOSCOPY
A case-control study73 showed a 59 per cent reduction in mortality from bowel cancer in people screened by sigmoidoscopy, with the reduction only applying to tumours arising in the screened part of the bowel. The potential benefit comes largely from identifying and removing pre-malignant adenomas, although carcinomas may also be diagnosed. Evidence that sigmoidoscopy rather than colonoscopy is appropriate for initial screening comes from a study74 in which the standardized incidence ratio of colonic cancer was 3.6 if a single villous, tubulovillous or large (1 cm) adenoma was found in the rectosigmoid by rigid sigmoidoscopy. If the adenomas were multiple, the standardized incidence ratio was 6.6. This shows that the presence of a large, tubulovillous or villous adenoma in the rectosigmoid appears to be predictive of carcinoma at remote proximal sites in the colon and identifies a group needing follow-up by colonoscopy. A pilot for a randomized British trial is reexamining this issue by performing a single flexible sigmoidoscopy between 55 and 64 years of age. Preliminary data
Treatment 673
Box 28.1 Staging systems for colorectal cancer
Box 28.2 Dukes staging system correlated with TNM
Primary tumour (T) TX Primary tumour cannot be assessed T0 No evidence of primary tumour T1s Carcinoma-in-situ T1 Tumour invades submucosa T2 Tumour invades muscularis propria T3 Tumour invades through the muscularis propria into the subserosa, or into non-peritonealized pericolic or peri-rectal tissues T4 Tumour perforates the visceral peritoneum or directly invades other organs or structures*
Dukes A
*Direct invasion of other organs or structures includes invasion of other segments of colorectum by way of serosa (e.g. invasion of the sigmoid colon by a carcinoma of the caecum) Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph-node metastasis N1 Metastasis in 1–3 peri-colic or peri-rectal nodes N2 Metastasis in 4 or more peri-colic or peri-rectal lymph nodes N3 Metastasis in any lymph node along the course of a named vascular trunk
Dukes B Dukes C Dukes D
Stage I Stage I Stage II Stage II Stage III Stage III Stage IV
Modified Astler–Coller (MAC) system correlated with TNM (AJCC stage) MAC A MAC B1 MAC B2 MAC B3 MAC C1 MAC C2 MAC C3
T1N0M0 T2N0M0 T3N0M0, T4N0M0 T4N0M0 T2N1M0, T2N2M0, T2N3M0 T3N1M0, T3N2M0, T3N3M0 T4N1M0, T4N2M0, T4N3M0
Stage I Stage I Stage II Stage II Stage III Stage III Stage III
Table 28.2 Incidence of positive tumour markers at different stages of disease Dukes stage A B C Distant metastases
Distant metastasis (M) MX Presence of distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis
T1N0M0 T2N0M0 T3N0M0 T4N0M0 T(any)N1M0, T(any)N2M0 T(any)N3M0 T(any)N(any)M1
5-year survival (%)
Raised CEA (%)
Raised CA19-9 (%)
80 50 30 5
4 26 44 65
11 31 26 49
CEA, carcinoembryonic antigen; CA19-9, carbohydrate antigen 19–9.
Stage grouping Dukes Stage 0 Stage I Stage Stage III Stage IV
Tis T1 T2 T3 T4 Any T Any T Any T
N0 N0 N0 N0 N0 N1 N2, N3 Any N
M0 M0 M0 M0 M0 M0 M0 M1
A B
proportion than in unscreened patients, and this suggests that the prospects of cure are higher. However, the generally accepted British (but not the North American) view is that it is still necessary to conduct a randomized clinical trial to investigate the effect of flexible sigmoidoscopy on survival before population screening is considered.75
C D
Note: Dukes B is a composite of better (T3, N0, M0) and worse (T4, N0, M0) prognostic groups, as is Dukes C (Any T, N1, M0 and Any T, N2, N3, M0).
on 23 000 individuals screened indicate good compliance rates with the expected incidence of adenomas (9–10 per cent) and referral for colonoscopy. The incidence of cancers was higher than expected, at 7 per thousand. Fiftyfive per cent of the cancers were Dukes stage A, a higher
FAECAL OCCULT BLOOD
Three large randomized trials76–78 show a survival advantage of up to 20 per cent for screening by faecal occult blood test every 2 years. The impact of a national screening programme in the UK would be to prevent about 1200 deaths a year with screening of people aged 50–69 years and an expected compliance rate of 60 per cent.79 About 2 per cent of such individuals test positive; 10 per cent of these will be found to have a carcinoma, 30 per cent an adenoma larger than 1 cm, 10 per cent a small adenoma, and 50 per cent no abnormality. Those testing positive need either a colonoscopy or double contrast barium enema and a flexible
674 Colorectal cancer
sigmoidoscopy. Colonoscopy in expert hands is probably the most sensitive, although it has a higher mortality (about 0.01 per cent) than the other investigations. Much of this risk is associated with polypectomy, which may in any event be required after a barium enema. The feasibility of a national screening programme is being investigated in a study of 100 000 people aged 50–69 years in the UK. SCREENING IN PEOPLE WITH HNPCCC AND OTHER RISK FACTORS
Colonoscopic screening every 5 years with 3-yearly examinations when a polyp is found reduces the risk of carcinoma depending on the level of risk, patients with HNPCC meeting the Amsterdam Criteria achieving a reduction of greater than 50 per cent.80
Surgery TREATMENT OF POLYPS
All polyps, with the exception of those that are sessile or very large, should be removed with a diathermy snare at the time of colonoscopy. Attempts to remove large sessile polyps may cause heavy bleeding or perforation. They are often malignant and should be treated by open surgery.81 Very small sessile polyps may be biopsied and destroyed with the ‘hot biopsy’ forceps. After colonoscopic removal of polyps, every effort should be made to retrieve them, as management thereafter will depend on histological examination. If a polyp is benign, no further surgical procedure is indicated. The patient should be followed up by colonoscopic screening, as the risk of further polyps may be as high as 40 per cent,82 although carcinoma after complete clearance of polyps is rare. The principles for the safe management of patients with malignant colonic polyps have been well established.83–87 Endoscopic polypectomy must be complete and on histological examination the excision margin or stalk must be free from invasion, the tumour moderately or well differentiated and with no evidence of malignant lymphatic or venous invasion. If the above criteria cannot be met, formal surgical resection is indicated,87 especially if the surgical risk is low. A better than 95 per cent 5-year survival rate has been achieved for malignant polyps removed by colonoscopic snaring, with no operative mortality and a cumulative recurrence rate of 11.3 per cent after 5 years.85 After surgical excision, the corresponding recurrence rate was 8.9 per cent, with a 5-year survival rate of 86 per cent. In the case of FAP, surgical resection is mandatory. Total colectomy and ileorectal anastomosis avoids the need for permanent ileostomy, but the lifetime risk of cancer in the rectal stump is at least 10 per cent and restorative proctocolectomy may be preferred. The choice of operation may be influenced by molecular genetic tests.27
OPERATIONS FOR CANCER
The purpose of surgical treatment is to remove the primary tumour together with all lymph nodes following the course of the named arterial blood supply and any removable adjacent organs that have been directly invaded. For tumours involving the caecum and ascending colon, the ileocolic, right colic and right branch of the middle colic arteries are divided at their origins. The mesocolon is dissected from the right ureter, duodenum, pancreas and gonadal vessels and the entire right colon is then removed together with the greater omentum. An anastomosis is then made between the terminal ileum and mid-transverse colon. Similar principles govern the surgery for transverse and descending colon tumours, the anastomosis being made between ascending and descending colon or transverse and sigmoid colon respectively. In the case of the sigmoid colon, the inferior mesenteric artery is ligated at its origin with aorta and the anastomosis made, after mobilization of the splenic flexure, between the descending colon and the rectum. The principles applied to curative surgery for rectal cancer are governed by the blood supply to the rectum. This is not only from the inferior mesenteric artery, but also from the middle rectal, branches of the internal iliac artery laterally and the pudendal vessels from below. There is a need to preserve anal sphincter function whenever possible and to avoid damage to the parasympathetic pelvic nerves, which would result in impotence and bladder dysfunction. Tumours lying above the levator ani muscle and anal sphincter are resected by anterior restorative resection through the abdomen. The operation involves ligation of the inferior mesenteric artery and removal of the rectum and the meso-rectum, but without formal node dissection on the lateral pelvic wall. Local recurrence can be minimized by total mesorectal excision, although faecal continence may be impaired if the anastomosis is very low.88 Construction of a colonic reservoir immediately proximal to the anastomosis may improve the functional results,89 although there may be some difficulty in spontaneous defaecation after this modification. The rectum is divided at least 1 cm90 below the tumour, and anastomosis made between the sigmoid colon and the rectum or anal canal. Most surgeons now use a circular stapler to achieve a secure low colo–rectal or colo–anal anastomosis, although excellent results may still be achieved by hand-sewn anastomosis.91 The place of protective colostomy or ileostomy after low anastomosis remains contentious. There appears to be little evidence that either the incidence of anastomotic leakage or the mortality following such a leakage is improved, but it is likely that pelvic-floor contamination may impair later continence. Laparoscopic-assisted surgery for colorectal cancer has been widely adopted, but only recently have data from largescale randomized trials supporting its use been published. In the Medical Research Council (MRC) CLASSICC Trial,92 apart from patients undergoing laparoscopic anterior resection for rectal cancer, rates of positive resection margins
Treatment 675
were similar between treatment groups. Patients with converted treatment had raised complication rates. It was concluded that laparoscopic-assisted surgery for cancer of the colon is as effective as open surgery in the short term and is likely to produce similar long-term outcomes. However, impaired short-term outcomes after laparoscopic-assisted anterior resection for cancer of the rectum do not yet justify its routine use for this purpose. Leung et al.93 investigated patients with rectosigmoid carcinoma and found that laparoscopic resection did not jeopardize survival (5-year survivals for the laparoscopic and open resection groups were 76.1 per cent [SE 3.7] and 72.9 per cent [SE 4.0] respectively). The probabilities of being disease free at 5 years were 75.3 per cent [SE 3.7] and 78.3 per cent [SE 3.7], respectively. Patient selection, surgical training and a need for further information about long-term outcomes are outstanding issues. LOW RECTAL CARCINOMAS
Rectal cancers adjacent to or invading the levator ani and anal sphincter muscles can only be cured by removal of these structures. The operation of abdomino-perineal resection is identical to anterior restorative resection as far as the abdominal part of the operation is concerned. Synchronously, the anal canal together with the contents of the ischio-rectal fossa and the entire levator ani muscle are removed by a perineal approach and the perineal wound closed with suction drainage. A permanent colostomy is constructed, usually in the left iliac fossa, but preoperative fitting is essential so that the precise position of the stoma can be tailored to the needs of the patient. LOCALLY EXTENSIVE TUMOUR
About 10 per cent of large-bowel tumours, particularly those within the pelvis, will be adherent to adjacent organs due to either inflammatory fibrosis or malignancy. It has been shown that resection of adjacent organs may be achieved with reasonable safety in the absence of metastatic disease, and the survival results can compare with those for tumours confined to the rectum or colon.94,95 Complete pelvic exenteration has a high morbidity and mortality, but careful selection may achieve worthwhile survival times.96 Preoperative radiotherapy may allow the removal of advanced tumours with reduced morbidity97 and with a lower incidence of local recurrence.98
the best palliation and allows radiotherapy and/or chemotherapy to be used with the best chance of benefit. In the presence of distant metastases, if the tumour is not removable or can only be removed by mutilating surgery, it is nearly always best to close the abdomen and rely on palliative non-surgical measures. Expandable metallic stents provide a new modality of palliation for patients with noncurable metastatic colorectal adenocarcinoma distal to the splenic flexure.99,100 Endoscopic laser ablation101 may also relieve obstruction.
OPERATIVE MORTALITY AND MORBIDITY
Operative mortality for non-emergency resections of potentially curable cancers lies between 1 per cent and 10 per cent. Complications include those common to all surgery, but septic problems predominate. In particular, there is a wide variation in the rate of anastomotic leakage amongst surgeons, of between 1 per cent and 10 per cent in most centres, according to whether is it measured radiologically or clinically. Temporary and, rarely, permanent impotence in male patients may occur if there is damage to the pelvic parasympathetic nerves, and patients of both sexes may suffer bladder dysfunction for the same reason.
OBSTRUCTING AND PERFORATING CANCERS
Emergency operations for both obstructed and perforated carcinomas have a significantly higher peri-operative mortality and worse long-term survival.102 The management of obstructing or perforated carcinoma of the left colon remains a controversial matter, with some preferring a staged resection and others a primary resection and anastomosis. The St Mary’s Hospital Large Bowel Cancer Project103 included 713 patients with malignant large-bowel obstruction. Immediate anastomosis in the obstructed left colon had a high clinical leak rate, but the mortality of primary resection was no greater than the cumulative mortality of the staged procedure. Primary resection has the advantage of a much reduced hospital stay in a group of patients with poor long-term survival prospects and is now generally preferred.104–106 Better results are achieved by more experienced surgeons whichever policy is adopted.107–109
LOCAL RECURRENCE INCOMPLETELY RESECTABLE TUMOUR
The management of incurable or irremovable large-bowel cancers is often difficult. In general, incurable disease should be treated by palliative resection whenever possible, even if, in the case of rectal tumours, this requires a permanent colostomy. Reduction of tumour bulk and removal of tumour likely to cause obstruction or fungation provides
Although the place of routine follow-up after surgery for colorectal cancer has been questioned,110 there is evidence that an aggressive approach and attempted surgical resection of any local recurrence can be worthwhile.111,112 In particular, a rise in CEA level above the baseline level established after a primary resection113,114 seems to be a sensitive indicator of recurrence.
676 Colorectal cancer
Local or anastomotic recurrence after operation for rectal cancer often presents with local symptoms such as recurrent bleeding, tenesmus, pain, incontinence and eventual obstruction. Many patients do not have disseminated cancer, and CEA levels are often initially normal. If the initial procedure was an anterior restorative resection, these patients can sometimes be treated by abdomino-perineal resection,115 but often further radical surgery is impossible.116 Small recurrences may be palliated with the NdYAG laser, but poor results have been reported for large tumours, especially those associated with obstruction or involving the sphincters.117
individuals after resection or ablative treatment, for instance in breast cancer,121,122 cancers of unknown primary site123 and neuroendocrine tumours.124 LUNG METASTASES
Low rectal carcinomas may metastasize through the systemic circulation directly to the lung, but the more common picture is of lung and liver metastases together with primary tumours at any site in the colon or rectum. Longterm survival is reported after resection when the number of metastases is small, even if there are also resectable liver metastases.125
HEPATIC METASTASES
It is generally agreed that untreated multiple hepatic metastases lead to an inevitably fatal prognosis, with an average survival of about 6 months from the time of diagnosis. Liver resection is the only established curative option for liver metastases, with one-third of resected patients alive at 5 years. In those developing recurrence in the liver following resection, further liver surgery may be curative, with similar 5-year survival rates of about 30 per cent. Until recently, surgery was feasible in only 15–25 per cent of patients with colorectal liver metastases. New strategies, such as downstaging chemotherapy, portal vein embolization and two-stage hepatectomy, may increase the resectability rate by 15 per cent. Earlier detection of liver metastases would increase resectability, although good follow-up trials are lacking. Once suspected, colorectal liver metastases are staged by CT, MRI and 18F-fluoro-2deoxyglucose positron emission tomography (FDG-PET). The last-mentioned probably has the greatest overall accuracy for hepatic and extra-hepatic metastases,118 but CT and MRI provide essential anatomical information for planning of surgery. Mortality following liver resection is less than 5 per cent in major centres, with a morbidity rate of 20–50 per cent. Prognostic scoring systems can be used to predict the likely cure rate. Pulmonary metastases occur in 10–25 per cent of patients with resected colorectal cancer, but are limited to the lung in only 2 per cent of cases. Surgery provides long-term survival in 20–40 per cent of these selected cases, and repeat lung resection has shown similar rates. For patients with unresectable disease, chemotherapy and ablation techniques have been demonstrated to prolong survival (for review see references 119 and 120). Radiofrequency ablation is proving effective in the management of small numbers of unresectable liver metastases that are not amenable to surgery. LIVER METASTASES FROM OTHER PRIMARY SITES
When metastases from sites other than the bowel are confined to the liver, there may also be a case for liver resection, radiofrequency ablation or laser therapy. There are no randomized trials showing benefit over systemic treatment, but there are numerous reports of prolonged survival in
Radiotherapy The potential clinical indications for radiotherapy and chemotherapy may be listed as follows. ●
● ●
●
To enhance the outcome of operable disease in terms of survival and local control. To facilitate potentially difficult surgery. To treat extensive inoperable disease: (i) possibly to render it operable, (ii) to obtain long-term local control or possibly eradication, (iii) purely for control of symptoms. To conserve anal function: (i) as adjunct to surgery, (ii) as exclusive treatment.
It may achieve this by the following mechanisms. ●
●
●
●
Reduction of tumour mass (to facilitate surgery or palliate symptoms). Local elimination of small-volume disease to reduce local recurrence. Local elimination of small-volume disease, a source of potential metastasis. Elimination of micro-metastases.
The commonly considered applications are summarized in Table 28.3. CARCINOMA OF THE RECTUM
Radiotherapy in combination with surgery Local recurrence in the pelvis and metastasis may be a major problem after anterior or abdominoperineal resection. Reviewing their results for the likelihood of recurrence, Hager et al.126 found that patients with only mucosal involvement had a local recurrence rate of 8 per cent, with a 90 per cent 5-year survival. Patients with muscular invasion had a recurrence rate of 15 per cent with a 58 per cent 5-year survival. A high-risk group where there was incomplete resection, had a recurrence rate of 24 per cent, with
Treatment 677
Table 28.3 Applications of radiotherapy Location in rectum Low-risk T1, T2, operable Low-risk T1, T2, operable Tethered T2, T3, operable Tethered T2, T3, operable Inoperable/fixed Macroscopic or microscopic residual tumour Recurrence Node positive
High rectal Low rectal High rectal Low rectal Any
Preoperative RT, short course
Preoperative RT, long course or chemo/RT
Postoperative RT
?
RT
Chemotherapy alone or chemo/RT
?
?
Surgery ?
RT, radiotherapy; chemo, chemotherapy.
61 per cent surviving at 4 years. Other studies show similar overall results. Approximately a third of patients have Dukes A (TM1 N0 and T2N0) tumours, for which the local recurrence rate is less than 5 per cent. For non-adherent Dukes B (again about a third of cases) the recurrence rate is approximately 10 per cent, and for adherent tumours or node-positive tumours (B3 and C), approximately the remaining third of cases, recurrence rates are reported of 20–70 per cent. The desired effect of adjuvant treatment is to reduce local recurrence and increase survival. The former can be achieved but the latter has proved more elusive. Some argue that the very low recurrence rates achieved with excellent surgery mean that adjuvant radiotherapy is not justified, but, as described below, there are many studies to indicate that, given adequate radiotherapy, local recurrence rates are reduced, and one large study reported a significant improvement in survival. Preoperative radiotherapy The Swedish Rectal Cancer Group127 showed benefits in terms of local control and survival. A short fractionation schedule was used: 1168 patients were randomized to receive either preoperative radiotherapy, 25 Gy in five fractions, followed by surgery (within 7 days) or surgery alone. At 5 years the local recurrence rate was reduced from 27 per cent in the surgery-only group to 11 per cent, and the 5-year survival was increased from 48 per cent to 58 per cent. The Swedish group also reviewed 780 cases registered for the study but not entered into it. Of these, only 67 had radiotherapy and the 5-year survival was 48 per cent overall, i.e. identical to the no-treatment group arm of the trial,47 confirming the truly representative nature of the control group. The CR07, phase III randomized trial from the MRC closed in August 2005 and final analysis is awaited. The endpoints included assessment of local recurrence-free survival, overall survival, time to appearance of distant metastases, disease-free survival and treatment-related morbidity. This study recruited 1350 patients with
histologically proven adenocarcinoma of the rectum over 8 years. Patients were randomized to either preoperative radiotherapy (25 Gy in five fractions, followed by surgery within 7 days of completion of radiotherapy), or postoperative chemo-radiation (4–12 weeks after surgery if the circumferential resection margin (CRM) is positive, 45 Gy in 25 fractions, with concurrent chemotherapy of either weekly bolus 5-fluorouracil (5-FU) and folinic acid, 5-day bolus schedule, or prolonged venous infusion). The median follow-up was 3 years, with 4.7 per cent (preoperative) and 11.1 per cent (postoperative) local recurrence rates as the primary endpoints, disease-free survival rates of 79.5 per cent and 74.9 per cent, respectively, and overall survival of 80.8 and 78.7 per cent in favour of preoperative radiotherapy.128 Also reported from the CR07 trial at ASCO in 2006 was that the plane of surgical dissection impacted on local recurrence, which was further reduced in the presence of preoperative radiotherapy. The CRM was involved in 11 per cent of surgical resection specimens and for these patients the 3-year local recurrence, disease-free survival and overall survival rates were 18 per cent, 50 per cent and 57 per cent, respectively. For the 1093 patients with CRMnegative, the respective rates were 7 per cent, 81 per cent and 84 per cent. Local recurrence was lowest when resection occurred in the mesorectal plane. In the preoperative radiotherapy group, the 3-year local recurrence rate was 1 per cent compared to 6 per cent (hazard ratio 4.47), with disease-free survival being 87 per cent compared to 80 per cent for the postoperative group (hazard ratio 1.53). Similary, both local recurrence and disease-free survival were improved in the other planes of surgical dissection (muscularis propria, intra-mesorectal) in the preoperative group.129 The Swedish Rectal Cancer Trial on preoperative radiotherapy has recently published data regarding side effects following a 14-year follow-up period. In the early postoperative period, those who had received radiotherapy had
678 Colorectal cancer
an increased risk of being admitted to hospital compared to the surgery-alone group. An increased risk was also noted of, especially, bowel obstruction (after approximately 8 years), nausea and non-specific abdominal pain,130 again in the irradiated group. However, the volumes that were irradiated in the Swedish trial initially were larger (upper limit above L4) than those used today (secondary to improved conformal techniques). Folkesson et al., in 2005,131 reported that of 1168 patients randomized to the Swedish Rectal Cancer Trial from 1987 to 1990, those who received preoperative radiotherapy of 25 Gy in five fractions followed by surgery had an overall survival of 38 per cent, compared to 30 per cent (p 0.008) in the surgery-alone group. Local recurrence was significantly reduced in the radiotherapy arm – 9 per cent compared to 26 per cent (p 0.001) with surgery alone. Also the cancer-specific survival was significant, at 72 per cent compared to 26 per cent in the surgery-only group. Recently, a consensus was published that aimed to establish the risk and benefits of preoperative radiotherapy followed by total mesorectal excision, using the Delphi technique, in the UK, Netherlands and Sweden. Major and minor complication criteria, including, for example, anastomotic leak, haemorrhage, fistula and sepsis, were suggested by colorectal surgeons, which would lead them to reconsider pre-operative radiotherapy.132 The aim of these criteria is to allow a risk–benefit assessment of the application of preoperative radiotherapy, and therefore to evaluate treatment. A meta-analysis involving 22 randomized trials and 8507 patients reported that the yearly risk of local recurrence was 46 per cent, and 37 per cent lower in those who had preoperative (BED 30 Gy) or postoperative radiotherapy respectively, compared to surgery alone. However, there was no significant difference in the overall survival amongst those who had received radiotherapy (62 per cent died) versus surgery alone (63 per cent died).133 However, the preoperative group did have reduced cancer-related mortality of 45 per cent compared to 50 per cent. The aim is to increase sphincter preservation, especially in low-lying rectal tumours, and therefore avoid AP resection. Overall, these data suggest that preoperative radiotherapy is beneficial in terms of overall survival and local recurrence rates compared to surgery alone. Chemo-radiation, preoperative versus postoperative Sauer et al. in 2003,134 compared preoperative versus postoperative chemo-radiation. In this phase III randomized trial, 805 patients were allocated to either preoperative arm 50.4 Gy in 28 fractions with 5-day infusional 5-FU on weeks 1 and 5, followed by maintenance chemotherapy, or same regimen to postoperative group with a further 5.4-Gy boost. The acute toxicity was tolerated well in both groups, with the postoperative period having complications of 12 per cent in both instances. Following this, in 2004, Sauer et al. with an identical protocol, the overall survival data was
published at 76 per cent versus 74 per cent for the preoperative and postoperative arms respectively (p 0.80), and 5-year disease-free survival of 68 per cent versus 65 per cent. Although no survival benefit was achieved, there was again minimal difference in surgical morbidity in both groups. However, the question to be addressed is whether the reduction in local recurrence and tumour downstaging following preoperative chemo-radiation translates into a reduction in the use of abdomino-perineal resections, and therefore sphincter-preserving surgical techniques. Use of capecitabine as part of chemo-radiation Whereas 5-FU has been the mainstay of treatment in recent years, oral preparations of fluoropyrimidines are being used increasingly. The oral prodrug capecitabine is part of the management in the adjuvant and metastatic settings of colorectal cancer. However, its use with radiation is yet to be established. To date, there are no published phase III randomized controlled trials to assess response rates and survival benefits, when compared to either bolus or prolonged intravenous infusion of 5-FU. An initial phase I trial with 36 patients investigated escalating doses of capecitabine, from 350 mg/m2 to 1250 mg/m2, twice daily, including weekends, for the duration of radiotherapy (50.4 Gy in 28 fractions, 1.8 Gy per fraction, once a day). This was given for rectal carcinoma in the adjuvant, neoadjuvant and palliative settings. The dose-limiting toxicity was hand–foot syndrome at 100 mg/m2, and the maximum myelosuppression was grade 2; also diarrhoea was not dose limiting and occurred in one patient only. The recommended dose was 825 mg/m2 twice daily. Although further feasibility and survival data are required, this initial report suggests that capecitabine as part of chemo-radiation is well tolerated.135 However, the timing of administration to maximize radio-sensitizsation has yet to be addressed. Following on from this, a further preoperative phase II trial137 has been conducted in the use of concurrent capecitabine and oxaliplatin with radiation. In these 31 patients, capecitabine, at 825 mg/m2 twice a day on days 1–14 and 22–35, and oxaliplatin, with an initial dose of 50 mg/m2 on days 1, 8, 22 and 29, which was then escalated in steps of 10 mg/m2, were given concurrently with 50.4 Gy in 28 fractions of radiotherapy. The patients enrolled had T3–4 adenocarcinoma of the rectum, tumours close to the levator sling, any T stage with M1 disease requiring surgery to primary tumour, or locally recurrent disease. The dose-limiting toxicity of diarrhoea occurred at 60 mg/m2 and therefore the oxaliplatin dose was capped at this level. Surgery was measured at 6 weeks following completion of chemo-radiotherapy. Primary endpoints included downstaging (59 per cent), negative lymph nodes (68 per cent), and pathological complete response at the resection margin (19 per cent). This suggested that significant tumour reduction occurred, which facilitated sphincter-preserving surgery in 36 per cent of cases. The oxaliplatin toxicity experienced included sensory neuropathy and grade 2 leucopenia. Compliance was noted
Treatment 679
Table 28.4 Preoperative radiotherapy alone Number of cases
Dose (Gy)
51
45
48 74 35 56
40–45 (? 10) 45 46.8 3.6 36 in 12 fractions
CR (%) 8.5
4 14 4
5-year survival (%)
5-year anal function (%)
74 (2 years)
74
82 73 64 76
86 80 79 72
Reference Bozzetti et al., 1999137 (hyperfractionated) Mohiuddin et al., 1998138 Ahmad & Nagle, 1997139 Wagman et al., 1998140 Corsa, 1997141
CR, complete response.
in 89 per cent, and all patients received full-dose radiation. In general, the regime was well tolerated, and the results of this study suggest that such a combination may have a role in future patient management. However, further phase III randomized controlled trials are required to assess the survival benefits of chemo-radiation with capecitabine and/or oxaliplatin compared to prolonged venous infusions of 5-FU. Preoperative radiotherapy to allow anus preservation Preoperative radiotherapy for low and/or locally advanced (T3, T4) tumours has been reported by a number of groups to result in anal sphincter preservation in about 80 per cent of cases and a 5-year survival of about 80 per cent, as shown in Table 28.4. As discussed above, chemo-radiation may well be even more effective. COLONIC CARCINOMA
Adjuvant radiotherapy (postoperative) No benefit from adjuvant radiotherapy has been shown for early colonic carcinoma. However, several studies have been performed, and the possibility of benefit has not been convincingly refuted. In general, local recurrence is less of a problem than with rectal carcinoma. Thus local radiotherapy has less theoretical opportunity for benefit. Although tried, whole abdominal irradiation has been of minimal benefit, since the marked toxicity severely restricted the dose. However, with T3N0M0 tumours where there may be a particularly high risk, radiotherapy may have a role. One hundred and thirty-three cases of locally advanced tumours (T3 and T4) were reported from the Massachussets General Hospital treated with radiotherapy to a dose of 45Gy together with a boost of 5–10 Gy to the tumour bed (22 of these patients were also given concurrent fluorouracil). Compared with 395 historical controls, the local failure rate was reduced from 26 per cent to 18 per cent.142 Dose–response to radiotherapy The effect of dose has also been examined in a number of retrospective studies, and indeed, looking at the various studies presented here, low doses were generally ineffective compared with higher doses. In one report, the relapse rate
was only 10 per cent if the radiotherapy dose was equal to or greater than 45 Gy, but 50 per cent if the dose was less than 40 Gy.143 There seems little doubt that there is a dose–response effect.139 Timing of surgery following radiotherapy The optimal timing of surgery following radiotherapy is somewhat controversial. Where there is bulk tumour for which regression is required to facilitate surgery, it is important to leave at least 4–6 weeks for this to occur. However, if the tumour is operable, we have favoured surgery at 7–10 days post-radiation, at which time the tissues are relatively easy to dissect, and the earlier timing ensures that any residual tumour is seen and resected. Likewise, the Stockholm studies127 recommend surgery at less than a week from completion of radiotherapy. On the other hand, many surgeons prefer to operate later, once the reaction has settled and the tumour has had a chance to shrink further. On the other side of the Atlantic, the generally favoured timing is 6 weeks after radiotherapy, it being argued that this allows for better tumour regression and normal tissue recovery. However, in Europe, the time intervals have usually been about 2–3 weeks, when it is felt that surgery is easiest and, furthermore, the shorter the time between radiotherapy and surgery, the less chance there is for progression and proliferation of the clonogenic cells. There are no studies specifically addressing this problem. Localized inoperable tumours This is a rather heterogeneous group and includes patients presenting with primary advanced disease previously untreated and also those with recurrent disease having received various treatments, either surgery alone or surgery and radiotherapy, or all three modalities, and the approach will clearly depend on what treatment has already been given as well as on the disease extent and general status of the patient. Supra-radical surgery with extended dissections can be carried out to attempt to excise extensive tumour, and occasionally long-term control may be achieved. Radical radiotherapy Radiotherapy alone may be as effective as surgery in controlling disease, and in one retrospective report of patients
680 Colorectal cancer
with recurrent disease, both the time to recurrence and the overall survival were similar in both groups. Therefore radiotherapy would seem to be a reasonable alternative to surgery if major surgery is contraindicated. Preoperative radiotherapy and chemo-radiation for localized inoperable tumours Doses of 45 Gy or more have been used, allowing resection in 50–75 per cent of cases, but long-term control is only achieved in 25–35 per cent. For example, Mendenhall et al., using 33–60 Gy, reported that resection became possible in most cases (37 out of 42); half were complete and half incomplete, resulting in 5-year survivals of 14 per cent in those with incomplete resection and 29 per cent where a complete resection was achieved.144 Most would now use chemo-radiation in this situation, as described above.
(IgG1) monoclonal antibody to the epidermal growth factor receptor (EGFR) ligand-binding domain) is currently being used in a trial setting in metastatic colorectal cancer. However, its combination with radiation as a multi-modality approach in non-metastatic stage 3 or 4 squamous carcinoma of the head and neck (oropharynx, hypopharynx and larynx) is already documented.148 In this site, a loading dose of 400 mg/m2 was used the week before radiotherapy, and 250 mg/m2 as a weekly infusion during radiotherapy. The expression of EGFR in colorectal cancer is similarly amplified, and therefore extrapolation of the head and neck data may suggest an emerging role in colorectal cancer. However, this is still awaiting exploration in clinical trials. Further developments may also occur in combination with bevacizumab, although the effect on radiaton acute and late effects would need to be established.
Intra-operative radiotherapy Intra-operative radiotherapy has also been used for the treatment of localized inoperable tumours. The Mayo Clinic145 used 45–55 Gy given by external radiotherapy and an intra-operative boost of 10–20 Gy. There was a local failure rate of 17 per cent, and the overall 5-year survival rate of 25 per cent was not greatly enhanced compared with the previous studies of preoperative external-beam therapy alone. Likewise, the Massachusetts General Hospital reports using 50.4 Gy in 5.5 weeks, with an intra-operative boost of 10–20 Gy. Seventy per cent of patients went on to achieve complete surgical clearance and the early report of 3-year survival was 60 per cent.146 Both studies had worse results with recurrent disease than with primary locally advanced disease. More recent studies combined chemoradiation as primary treatment with intra-operative radiotherapy and report improved overall survival.
Radiotherapy techniques in colorectal carcinoma PRINCIPLES OF RADIOTHERAPY
Where residual disease remains apparent and preoperative radiotherapy has not been given, radiotherapy can be offered. Many of the early studies of radiotherapy were related to treatment given in this way. There are no randomized studies. Reported local recurrence rates vary from 15 per cent to 76 per cent, but the 5-year survival is more consistent at around 25 per cent.147 Perhaps the most promising approach remains with preoperative radiotherapy and chemotherapy, but randomized studies are needed to establish this.
The treatment volume and dose given are dependent on the amount of tumour present and/or the risk of tumour recurrence, and the normal-tissue tolerance. For ‘standard’ preoperative radiotherapy, it has been usual to use a uniform dose to the whole pelvis. If there is a bulky or inoperable tumour, a proportionally higher dose may be needed for the tumour itself, and this is achieved using a shrinking volume technique. Likewise, if surgery cannot be contemplated, a shrinking volume technique is again appropriate. For postoperative radiotherapy, the treatment can be tailored to the operative findings. Where excellent local clearance is achieved, a uniform field to the pelvis, to include the tumour bed and pelvic nodes at risk, is appropriate, but again, if there is an area of high risk of local recurrence, this can be boosted by the shrinking volume technique. For palliative radiotherapy, lower doses, such as 3000 cGy in ten fractions or 2000 cGy in five fractions, can be given by the simple technique of anterior and posterior fields. For the treatment of colonic carcinoma, similar principles apply, although in general it is difficult to avoid small bowel and therefore either reduced doses or a reduced margin to the tumour has to be used so as to limit the exposure of the small bowel, but small-bowel problems are likely in some 5 per cent of patients.149
Locally advanced disease: palliative radiotherapy
Regions at risk and tissue tolerance
Patients with relapsed locally advanced disease may still achieve good local response and good palliation. Even patients previously irradiated will tolerate up to 30 Gy.
The use of anterior and posterior fields for postoperative radiotherapy at the M.D. Anderson Cancer Center, with the upper margin of the fields extending up to L2, resulted in a 17.5 per cent incidence of small-bowel obstruction, compared with only 5 per cent in those who had surgery alone. When the fields were altered to an upper limit of just below L5, the incidence dropped to below 10 per cent.150 In practice, therefore, the para-aortic and high pelvic nodes are seldom included unless there are special indications.
Postoperative radiotherapy for residual macroscopic disease and/or recurrent inoperable disease
Potential advances in radiotherapy Monoclonal antibodies and other biological agents are increasingly being used. Cetuximab (immunoglobulin G1
Treatment 681
The internal iliac and pre-sacral nodes are at high risk of metastatic involvement and should be included in the radiotherapy field. External iliac nodes are not usually involved unless there has been invasion of pelvic organs such as the prostate, bladder, vagina or uterus. Recommended treatment volume To cover the regional nodes, the pelvic volume extends from the bottom of L5 to the upper side of the anal canal for high tumours, and to include the anal canal with lower tumours. The perineum need only be included if it is considered to be at special risk or after an abdomino-perineal (AP) resection. If it is included, bolus may help to make the dose more uniform. Posteriorly, the field includes the sacrum, thus encompassing the pre-sacral nodes and sacral canal. Anteriorly, the margin may be more difficult to define with a plain film only. If the anterior margin is taken to just behind the symphysis pubis, this will ensure covering the primary lesion, the adjacent prostate, base of bladder and vagina and internal iliac nodes, although to include the external iliac nodes, the field would generally need to come to the front of the symphysis pubis. The extent of the tumour and the localization of these areas at risk are more easily confirmed with CT planning. If a high-risk area requires an additional dose over and above the small-bowel tolerance of 40–50 Gy, or if palliative radiotherapy is the objective where minimal bowel toxicity is required, the reduced high-dose volume is best defined by CT planning. Inclusion of the perineum adds considerably to the acute morbidity, but there has not been a reported increase in chronic complications. After abdomino-perineal resection without radiotherapy, a perineal recurrence rate as high as 23 per cent has been reported, but following radiotherapy including the perineum, the incidence was only 2 per cent at the Mayo Clinic. It is wise, therefore, to include the perineum after AP resections, although for other low tumours, or for preoperative radiotherapy, the need is not established. Radiotherapy planning, field arrangements and doses Localization may be achieved by simulator or CT planning. The patient is best treated prone and the bladder should be full to minimize small bowel in the pelvis. A marker wire is placed over the lower anal margin to help identify the anal canal. On the lateral projection, the rectum is generally clearly outlined by the normal rectal air, although a contrast tampon in the vagina may be a useful aid. Immobilization body moulds can also help with accuracy, and small-bowel contrast can help to identify small bowel where a large amount of small bowel is potentially being irradiated. A direct posterior field and two lateral fields are commonly used, or two post-oblique fields may be used. Lateral fields allow perhaps slightly easier assessment of shielding of normal tissues. Such areas prone to a brisk reaction and delayed healing include the natal cleft and perineum, and providing these areas are not at risk, they can be easily shielded by appropriate leading of the lateral fields.
A wide range of doses has been used, as discussed above. Using fractionated radiotherapy, doses of 40 Gy in 20 fractions to 50 Gy in 30 fractions have been used, to which a boost to a reduced volume of a further 10–20 Gy may be added provided small bowel is excluded from the field. COMPLICATIONS OF RADIOTHERAPY
Perhaps the most common and serious complications of radiotherapy are to the small bowel, although particularly unpleasant and prolonged reactions can be seen in the perineum. Factors contributing to increased complications are shown in Table 28.5. Acute reactions include the small-bowel symptoms of diarrhoea and abdominal cramps, and the large-bowel effects of acute proctitis with urgency and frequency, tenesmus and occasionally a bloody or mucous discharge. Urinary frequency and dysuria may occur. Skin erythema may be particularly troublesome in the skinfolds of the natal cleft and perineum. These acute effects occur 2–3 weeks into a course of fractionated radiotherapy and generally resolve within a few weeks of stopping the treatment. Most serious are the delayed reactions, which may persist carrying on from the acute reactions or may develop after a latent period of some 6–18 months (Box 28.3). In Table 28.5 Factors increasing normal-tissue damage from radiation Radiotherapy related Total dose Large fraction size Short overall treatment time Low radiation energy Large treatment volume Prior pelvic irradiation Concomitant radiotherapy
Non-radiotherapy related Pelvic inflammatory disease Hypertension Diabetes mellitus Obesity Concomitant chemotherapy Immunocompromise (primary or secondary) Genetic disorders, e.g. Li Fraumeni, ataxia telangiectasia
Box 28.3 Late radiation complications Chronic diarrhoea Proctitis Rectal blood loss Rectal pain Small-bowel obstruction Small-bowel perforation Perineal and scrotal tenderness Delayed perineal healing Bladder atrophy and bleeding Sacral necrosis Infertility
682 Colorectal cancer
one retrospective study, ileus occurred in only 8 per cent of cases without radiotherapy, but in 23 per cent of those receiving radiotherapy, and surgery was required in only 4 per cent of those without radiotherapy versus 21 per cent in those receiving radiotherapy.151 Infertility is inevitable if the gonads are included in the radiation field and may occur even if close to the edge of the field. To minimize the risk of reactions, the factors listed in Table 28.5 should be taken into account. Total dose is the ultimate criterion, but for maximum therapeutic benefit and minimum toxicity, fractions should probably be 200 cGy or less. A linear accelerator with 8–15 MV electrons is optimal, and the treatment volume should be carefully assessed, being planned by computer to avoid hot spots and to limit the amount of bowel in the field. The other nonradiotherapy-related factors and medical conditions are difficult to quantify, but some allowance may need to be made. Where radiation tissue damage has developed, a conservative approach is often successful. For radiation proctitis with troublesome intermittent bleeding, which may demand recurrent transfusions, local instillation of formalin has been reported to be effective (and simple) in controlling the bleeding.152 If surgery has to be embarked upon, fresh tissues are needed to assist healing, such as unirradiated omentum or full-thickness pedicle skin flaps. ●
Chemotherapy Chemo-radiotherapy in localized rectal carcinoma is addressed above. The same principles as for colon cancer apply to patients with rectal cancer who have not undergone chemo-radiation. In patients who have received chemo-radiation, the role of adjuvant chemotherapy remains controversial. COLONIC CARCINOMA
Adjuvant chemotherapy for colon carcinoma Adjuvant chemotherapy in colon cancer is given to eliminate micro-metastatic residual disease to reduce the risk of recurrence and improve survival. Until recently, 5-FU formed the mainstay of adjuvant treatment. More recently, capecitabine, oxaliplatin and irinotecan have been studied for adjuvant treatment of colon cancer. ●
Dukes stage C. The first large study reporting a significant benefit of adjuvant chemotherapy with 5-FU in patients with Dukes stage C colon carcinoma was reported in 1990 by Moertel et al.153 At a median follow-up of 3 years, adjuvant chemotherapy with 5-FU and levamisole was associated with a 33 per cent reduction in the risk of death and a 41 per cent reduction in the risk of relapse. Levamisole has since been abandoned due to its lack of efficacy and has widely been replaced by folinic acid (calcium folinate) (FA).154 The 5 days every 4 weeks schedule used initially has since been shown to be more toxic than other 5-FU/FA-based schedules. Equal
efficacy and less toxicity of weekly 5-FU/FA compared to a daily 5 5-FU/FA every 4 weeks schedule was confirmed in the large QUASAR Colorectal Cancer Group Study,155 although comparison was non-randomized. Similarly, less toxicity and equal efficacy were found with an infusional regimen of 5-FU and FA given over 2 days every 2 weeks.156 Capecitabine, a 5-FU prodrug given orally, is convenient, and comparable efficacy to that of 5-FU/FA has recently been shown.157 The addition of oxaliplatin to 5-FU/FA for adjuvant treatment was investigated in the MOSAIC trial. In this trial the combination was reported significantly to improve disease-free survival in patients with Dukes stage C cancer.158,159 Peripheral sensory neuropathy is a common side effect of oxaliplatin. Although reversible in most patients, 1 per cent of patients experienced persistent grade 3 peripheral sensory neuropathy 1 year after completion of treatment. Improvement of 3-year disease-free survival with 5-FU/FA and oxaliplatin was confirmed in a second trial using a weekly bolus regimen.160 In contrast to oxaliplatin, the addition of irinotecan to 5-FU failed to reveal a significant improvement in disease-free survival.161,162 This has led to the recommendation to offer oxaliplatin to high-risk Dukes C cancer patients, in particular those with T4N2 disease. Dukes stage B. Dukes stage B has been included in several randomized trials of adjuvant chemotherapy, along with stage C. However, proving unequivocal benefit has been difficult. The reason behind this is that the rate of recurrence in Dukes B is about half that in Dukes C, so many more patients are required to demonstrate a benefit. The only study so far showing an unequivocal benefit for adjuvant treatment for Dukes stage B cancer is the QUASAR trial.
With a median follow-up of 4.2 years, the relative risks of death and recurrence in the chemotherapy group were 0.88 and 0.82 respectively.163 The benefit of adjuvant treatment in Dukes stage B cancer is supported by two meta-analyses. The NSABP investigators164 analysed four trials, including 1565 patients with Dukes B tumours. They compared the patient groups receiving the best regimen (usually 5-FU and FA) with the patients with the worst outcome (either no treatment or other chemotherapy regimens). The cumulative odds of death in the Dukes B group were 0.7, which was statistically significant. The IMPACT B2 investigators pooled data from five trials with 1016 patients.165 At a median follow-up of 5.75 years, they found hazards ratios for event-free survival and overall survival of 0.83 and 0.86, respectively, but this benefit did not reach statistical significance. In addition, results from the MOSAIC trial showed a small and statistically non-significant improvement in disease-free survival in Dukes stage B disease. Although adjuvant treatment for all Dukes stage B cancer still remains controversial, there is increasing evidence of benefit for this disease stage. Particularly patients with adverse prognostic features, including bowel obstruction and perforation,
Treatment 683
(a)
Figure 28.3 Response of recurrent rectal carcinoma to cetuxumab and irinotecan after becoming refractory to irinotecan, fluorouracil and folinic acid. The arrow shows a lung metastasis which was no longer detectable after treatment.
(b)
should be considered, since this group of patients has a worse prognosis than some Dukes stage C patients. New therapies currently being investigated include the anti-EGFR antibody cetuximab and the anti-vascular endothelial growth factor (VEGF) antibody bevacuzimab, which are discussed under metastatic tumour therapy below. METASTATIC AND LOCALLY RECURRENT TUMOUR
Choice of treatment: chemotherapy, radiotherapy and palliative treatment Chemotherapy forms the mainstay of treatment for inoperable metastatic colonic and rectal cancer. For decades 5-FU had been the only drug with proven efficacy. A Cochrane collaborative meta-analysis of a subset of trials that provided individual patient data compared palliative treatment to chemotherapy with 5-FU. Chemotherapy with 5FU resulted in a median survival of 11.7 months, compared to 8 months in the control group, and improved survival at 6 and 12 months by 16 per cent.166 The systemic treatment options for patients with metastatic colon and rectal cancer have changed considerably following the introduction of new drugs, including the cytotoxics oxaliplatin, irinotecan and capecitabine, the anti-EGFR antibody cetuximab (Fig. 28.3) and the anti-VEGF antibody bevazicumab. This has resulted in improvements of median survival, with up to 20 months being reported by some investigators.167 The role of radiotherapy is limited in metastatic disease, but patients with relapsed locally advanced disease may still achieve a good local response and good palliation. Chemotherapy regimens Chemotherapy regimens are summarized in Box 28.4. ●
Fluoropyrimidines. 5-Fluorouracil inhibits the enzyme thymidilate synthetase (TS), which leads to depletion of deoxythymidine triphosphate, thus interfering with DNA synthesis and repair. An enhancement of this effect has been seen when 5-FU is used with FA. FA prolongs the inhibition of TS activity and hence DNA synthesis produced by 5-FU through stabilization of the ternary complex of 5,10-methylene tetrahydrofolic acid with
●
●
the 5-FU metabolite FdUMP and TS. Response rates with 5-FU alone are low (10–15 per cent). The effect is dose related, and infusion over a few days increases response rates. Biomodulation with FA has been reported to increase response rates to 20–40 per cent and improve symptom control in up to 75 per cent of cases. Metaanalysis has confirmed a two-fold increase in tumour response of 21 per cent compared to 11 per cent, and a small but statistically significant overall survival benefit of 11.7 months compared to 10.5 months.168 The side effects of 5-FU and FA include mucositis; this manifests principally as diarrhoea, which may be severe and even life threatening. It should be treated promptly and dehydration avoided. Modest dose reduction will normally prevent recurrence. Myelosuppression is uncommon. Most patients become fatigued during chemotherapy lasting over 3–6 months. Infusion of 5-FU is associated with a dermatitis involving the hands and feet (palmar–plantar syndrome), which may be alleviated by oral pyridoxine administration. The potential cardiac complications of 5-FU indicate the need for caution when using it in patients with pre-existing heart disease. Raltitrexed. Raltitrexed is a potent TS inhibitor, which, due to its pharmacokinetics, can be given at 3-weekly intervals. One trial was terminated early due to an unexpectedly high treatment-related death rate.169 Because of this, there are reservations about the use of this drug, as there are reasonable alternatives in the form of 5-FU and FA and capecitabine. Oral fluoropyrimidines. 5-Fluorouracil is absorbed erratically (0–80 per cent) and is not generally considered useful by this route. Capecitabine and UFT (uracil and Tegafur) have been developed to give sustained and predictable serum levels after oral administration. Capecitabine is an orally bioavailable prodrug of 5-FU. Thymidine phosphorylase (TP) is required for activation, and this enzyme is overexpressed in some colorectal carcinomas, particularly in response to tumour hypoxia. Treatment with capecitabine in metastatic colon and rectal cancer has been shown to be equally effective compared with 5-FU/FA171. Myelotoxicity was found to occur comparably with less intensity than with 5-FU/FA,
684 Colorectal cancer
Box 28.4 Chemotherapy regimens Modified DeGramont FA 350 mg (flat dose) i.v. in 250 mL 0.9% sodium chloride over 2 hours 5-FU 400 mg/m2 i.v. bolus. 5-FU 2400 mg/m2 i.v. via infusion pump over 46 hours Repeat on day 15 Weekly 5-FU/FA FA 30 mg i.v. via side-arm of fast-running 250 mL 0.9% sodium chloride over 5 minutes 5-FU 425 mg i.v. bolus Irinotecan single agent Irinotecan 350 mg/m2 i.v. in 250 mL 0.9% sodium chloride over 30–90 minutes In the event of acute cholinergic symptoms after administration of irinotecan, atropine 0.25 mg is given subcutaneously. The relevant symptoms are early diarrhoea with abdominal cramps, sweating, miosis, salivation or lacrimation Repeat on day 21 FOLFIRI Irinotecan 180 mg/m2 i.v. in 250 mL 0.9% sodium chloride over 30–60 minutes In the event of acute cholinergic symptoms after administration of irinotecan, atropine 0.25 mg is given subcutaneously. The relevant symptoms are early diarrhoea with abdominal cramps, sweating, miosis, salivation or lacrimation FA 350 mg (flat dose) i.v. in 250 mL 0.9% sodium chloride over 2 hours 5-FU 400 mg/m2 i.v. bolus 5-FU 2400 mg/m2 i.v. via infusion pump over 46 hours Repeat on day 15
but diarrhoea and palmar–plantar syndrome were experienced more frequently. On the basis of these data, capecitabine is increasingly replacing 5-FU because of greater patient convenience and reduced hospital costs. Irinotecan is a topoisomerase-I inhibitor. It blocks DNA replication mediated by the enzyme and induces singlestrand DNA breaks, which inhibit replication. Treatment with this topoisomerase-I inhibitor produced response rates of 11–23 per cent in phase II studies of patients with colorectal cancer whose tumours were resistant to 5-FU. A further 40 per cent of patients showed stabilization of tumour for a median of 5 months.171 The most common adverse effects include diarrhoea, nausea and vomiting, myelosuppression, asthenia, alopecia and cholinergic syndrome, the last-mentioned occurring at the time of administration. Two randomized trials of irinotecan have shown survival benefit, with the 1-year survival increased by 13 per cent and
CAPIRI Irinotecan 180 mg/m2 i.v. in 250 mL 0.9% sodium chloride over 30–60 minutes In the event of acute cholinergic symptoms after administration of irinotecan, atropine 0.25 mg is given subcutaneously. The relevant symptoms are early diarrhoea with abdominal cramps, sweating, miosis, salivation or lacrimation Capecitabine 800 mg/m2 b.d. daily day 1–9 Repeat on day 15 Consider increasing capecitabine to 1000 mg/m2 after 3 cycles if no toxicity 2 FOLFOX Oxaliplatin 85 mg/m2 i.v. in 250 mL 5% dextrose over 2 hours FA 350 mg (flat dose) i.v. in 250 mL 0.9% sodium chloride over 2 hours 5-FU 400 mg/m2 i.v. bolus 5-FU 2400 mg/m2 i.v. via infusion pump over 46 hours Repeat on day 15 CAPOX Oxaliplatin 85 mg/m2 i.v. in 500 mL dextrose 5% over 2 hours Capecitabine 800 mg/m2 b.d. daily day 1–9 Repeat on day 15 Consider increasing capecitabine to 1000 mg/m2 after 3 cycles if no toxicity 2 Capecitabine Capecitabine 1250 mg/m2 b.d. p.o. for 14 days Repeat on day 21 i.v., intravenous; b.d., twice daily; p.o., by mouth; FA, folinic acid; 5-FU, 5-fluorouracil.
22.4 per cent compared with palliative therapy172 or 5-FU infusion.173 Patients whose tumours progressed within 6 months of 5-FU showed a significant, though small, survival advantage following treatment with irinotecan. Combination of irinotecan with 5-FU and FA shows synergy in pre-clinical studies, and clinical combination regimens have been developed. Randomized comparison of the combination with 5-FU and FA showed significantly higher response rates for the combination in two trials,174,175 and a survival improvement of 17.4 versus 14.1 months has been reported. In previously untreated patients, the weekly regimen reported by Saltz has produced severe toxicity and is probably best avoided. Two-weekly combination with infusional 5-FU/FA (FOLFIRI) and a 3-weekly single administration schedule are currently frequently used. In patients with disease refractory to irinotecan, combination with the antiEGFR antibody cetuximab has been shown to overcome resistance to irinotecan, with a reponse rate of 22.9 per cent.177
Follow-up after initial treatment 685
Oxaliplatin belongs to a group of platinum compounds synthesized to include the 1,2-diaminocyclohexane carrier ligand. Although they bind to DNA in a way broadly similar to cisplatin, they are effective in cisplatin-resistant cell lines, possibly through greater resistance to DNA repair. Single-agent response rates of 10–20 per cent were reported. In phase I trials, the reversible peripheral neuropathy was dose limiting, but the ototoxicity and renal toxicity seen with cisplatin were largely absent. Synergy with TS inhibitors and with topoisomerase-I poisons was reported in pre-clinical studies, and much of the subsequent clinical work has been with oxaliplatin in combination with 5-FU and FA. This combination has produced response rates of 34–67 per cent in previously untreated patients and 21–58 per cent in patients who had relapsed after 5-FU/FA. In randomized controlled trials, oxaliplatin with 5-FU/FA has demonstrated significantly higher response rates, of 51 per cent versus 22 per cent, and disease-free survival of 9 versus 6.2 months compared to 5FU/FA alone. A small number of complete responses have been reported (for review see reference 177). There is evidence that oxaliplatin given for metastases confined to the liver may increase survival when combined with subsequent resection of liver metastases.178 Choice of first-line chemotherapy With both oxaliplatin and irinotecan showing efficacy in colon cancer, the question as to which of these agents should be used first has been addressed. In a phase III trial, Tournigand et al.179 compared oxaliplatin and 5-FU followed by irinotecan and 5-FU with the reverse sequence and found no significant difference in median survival. A larger trial in the USA showed modest superiority for oxaliplatin in terms of response rate and time to progression; however, the availablility of second-line chemotherapy was imbalanced.180 It appears, therefore, that the choice of drug for first-line treatment remains a question of patient preference after discussion of the different toxicity profiles. Prediction of response to chemotherapy The effect of drugs acting on TS, particularly 5-FU, may be expected to depend on the tumour concentration of the enzyme. High levels of TS have been found to predict poor response to treatment with 5-FU,181,182 and a number of studies have shown that high levels of TS predict for worse survival in Dukes stage B and C patients.181,183 A subset of patients has been described with low TS expression who do not respond to 5-FU and for whom a different mechanism of resistance appears likely. TS gene polymorphisms have the potential to be useful for the prediction of outcome and toxicity, since this may affect transcription and translation of the TS gene.184 Dihydropyrimidine dehydrogenase (DPD) catalyses the ratelimiting step in the catabolism of 5-FU, and tumour levels of DPD have been reported to determine response,185 whilst patients with DPD deficiency experience profound systemic
toxicity. Genetic instability has been recognized as a central element in the development of malignant lesions. Microsatellite instability (MSI) is commonly found in colonic cancers. Microsatellites are sections of DNA in which a short sequence of DNA is repeated multiple times. Microsatellite instability is a situation in which a microsatellite has gained or lost repeat units and has undergone a change in length, which results in frameshift mutations or base-pair mutations, or both. It is typically associated with DNA mismatch repair defects, in particular with the mismatch repair genes MLH1, MSH2 and MSH6. Studies exploring the correlation with response to chemotherapy are currently controversial.186,187 The development of high-throughput techniques such as DNA microarray profiling, proteomic profiling, comparative genomic hybridization analysis and metabolomics has significantly enhanced the potential to predict response to treatment. A study in cell lines identified a panel of genes with a higher likelihood to predict response to 5-FU than classical markers of 5-FU response including TS.188
FOLLOW-UP AFTER INITIAL TREATMENT After curative treatment of colorectal cancer, patients are at risk of developing metastatic disease, but are also at a higher risk of developing a second primary colorectal cancer than the general population. The aim of follow-up is therefore the detection of second primary tumours in the colon and of resectable recurrent disease, as the 5-year survival of patients with resectable liver or lung metastases is 30–40 per cent. Colonoscopy should be performed every 3–5 years after surgery. In patients with rising CEA and non-detectable disease on CT, FDG-PET scanning can be considered. The latter appears particularly effective in locating recurrences that may be missed by CT or MRI.189,190 Immunoscintigraphy can be helpful in locating metastases when serum CEA levels rise.191,192 If there was no effective treatment for metastases or local recurrence, it could be argued that minimal followup is all that is necessary and that if recurrence occurs, the symptoms should be treated palliatively. It was estimated in the 1980s193 that as many as 20 per cent of patients relapsing after apparently curative resection can be cured by resection of hepatic or pulmonary metastases. Advances in imaging, surgical technique and supportive measures have probably improved on this. Results with liver metastases are best when there is a solitary liver deposit, but multiple deposits can be resected, with long-term survival.194,195 The survival benefits of chemotherapy for metastatic disease are greatest when deposits are detected and treated at an early stage. There is therefore a strong case for early detection of either resectable or unresectable recurrence. Intensive follow-up can usually detect recurrence before it produces symptoms. Monthly serum CEA measurements predict clinically evident recurrence by an average of 6
686 Colorectal cancer
months in about 60 per cent of patients and trigger other investigations or exploratory laparotomy.196 Routine chest radiographs, liver ultrasound and abdominal CT and colonoscopy are also used for early detection. A regimen of bi-annual spiral CT with intravenous contrast for 2 years is recommended. Immunoscintigraphy can be helpful in locating metastases when serum CEA levels rise.191,192 18 F-fluorodeoxyglucose-PET appears particularly effective in locating recurrences that may be missed by CT or MRI.189,190
Radioimmunotherapy Radioimmunotherapy with an anti-CEA antibody has shown evidence of response.201 In-vivo experiments showed significantly better results when radiolabeled antibody was combined with the anti-vascular drug combretastatin, as antibodies preferentially localize in the viable tumour rim and combretastatin causes central necrosis.202 A Phase I study exploring this combination is currently ongoing. Gene therapy
FUTURE DEVELOPMENTS There has been an improvement in the outlook for people with colorectal cancer, and a number of new approaches are in clinical trials, which may have further impact.
Experimental approaches
Suicide gene therapy involves an enzyme such as CPG2 placed under control of the CEA promoter and expressed in CEA-positive tumour cells. A prodrug, such as that described above for ADEPT, is then activated at the tumour site after intravenous administration. This approach requires successful vector development for delivery of the enzyme gene to tumours in humans.203 Promising results for new prodrugs for this concept have recently been described.204 Vaccines
FUTURE PROSPECTS
As post-genomic knowledge of the molecular basis of cancer develops, there is scope for novel therapies directed against many cell-surface and intracellular targets, with a reasonable prospect of useful progress in the coming decade. Several new approaches are already in development. Antibody-directed enzyme prodrug therapy Antibody-directed enzyme prodrug therapy (ADEPT) uses the principle of pre-targeting, but adds augmentation of the therapeutic effect by targeting an enzyme, each molecule of which has the ability to activate many molecules of cytotoxic drug in the cancer. An antibody directed against a tumour-associated antigen is linked to an enzyme and given intravenously, resulting in selective accumulation of enzyme in tumour. When the discrimination between tumour and normal-tissue enzyme levels is sufficient, a prodrug is given intravenously and is converted to an active cytotoxic drug by enzyme within the tumour. This gives higher tumour-to-normal-tissue ratios at the time when therapy is given than can be achieved with direct tumour targeting. High levels of a cytotoxic drug are potentially generated in the tumour through the capacity of each enzyme molecule to convert many molecules of prodrug into drug. Several enzyme and prodrug systems have been investigated. An early-phase clinical trial of ADEPT with antibody to CEA conjugated to carboxypeptidase G2 (CPG2) and a benzoic acid mustard–glutamate prodrug showed evidence of response in colorectal cancer.197 A galactosylated antibody directed against the active site of CPG2 was required to increase clearance of the antibody–enzyme. To address slow clearance of the antibody–enzyme constructs, a genetically engineered antibody–enzyme fusion protein was designed. Antibody-directed enzyme prodrug therapy with the fusion protein of an anti-CEA scFv antibody and CPG2 are currently being tested in a Phase 1 trial.192,198–200
Tumor-associated antigens have been targeted by therapeutic vaccination in colorectal carcinoma patients. The results demonstrate that tumor antigen-specific immune responses are reproducibly induced and clinical benefit has been described in some studies.205
Potential advances in radiotherapy Monoclonal antibodies and other biological agents are increasingly being used. Cetuximab (IgG1 monoclonal antibody to the EGFR ligand-binding domain) is currently in use in a trial setting for metastatic colorectal cancer. However, its combination with radiation, as a multi-modality approach in non-metastatic stage 3 or 4 squamous carcinoma of the head and neck (oropharynx, hypopharynx and larynx), is already documented.148 In this site, a loading dose of 400 mg/m2 was used the week before radiotherapy, and 250 mg/m2 as a weekly infusion during radiotherapy. The expression of EGFR in colorectal cancer is similarly amplified, and therefore extrapolation of the head and neck data may suggest an emerging role in colorectal cancer. However, this is still awaiting exploration in clinical trials. Also, in combination with bevacizumab, further developments may occur, although the effect on radiation acute and late effects would need to be established. RADIOTHERAPY TECHNIQUES
Hyperfractionated and accelerated radiotherapy have been used in a variety of other sites, including the lung and head and neck. The theory underlying this approach is that by keeping the overall treatment time as short as possible, local control is not compromised. Coucke et al.206 reported on a phase II trial in locally advanced resectable stage 2 and 3 rectal cancer. Patients received 41.6 Gy in 2.5 weeks at 1.6 Gy per fraction with a 6-hour inter-fractionation interval, and two fractions per day. Of the 279 patients entered, 250 were
National health service guidelines 687
assessed, and results showed a 91.7 per cent 5-year actuarial local control, and overall survival of 59.6 per cent, with 38 per cent being down-staged. However, the trial mainly involved early-stage tumours, which were distally located. The RTOG-0012 phase II trial in 2006207 reported on patients randomized to either 5-FU as prolonged venous infusion (225 mg/m2 per day) and hyperfractionated radiotherapy (55.2–60 Gy, 1.2 Gy per fraction, twice daily) or 5-FU and irinotecan (225 mg/m2 Monday to Friday and 50 mg/m2 once a week for 4 weeks) and conventional fractionation 50.4–54 Gy in 28–30 fractions. The pathological complete response was similar, at 28 per cent, and 78 per cent of patients were down-staged in both arms as well. Toxicity, both non-haematological and haematological, was worse in the irinotecan arm. Although it is difficult to draw conclusions from this preoperative study due to differences in chemotherapy schedules used in the UK, it does add further weight to down-staging and the subsequent possibility of sphincter preservation with combination chemotherapy regimes in the neoadjuvant setting. This may, however, be at the expense of acute and long-term morbidity. Radiotherapy techniques are constantly evolving to improve tumour delineation, allowing dose escalation and reducing the dose to surrounding tissue and organs at risk. Positron emission tomography has an established role in lung cancer patients being considered for radical treatment. In a small series of 11 patients, with specially developed software for FDG-PET algorithms, the possibility of tumourvolume definition was assessed in patients with rectal cancer undergoing pre-operative radiotherapy. Ciernik et al.208 reported that the use of PET in planning provided improved accuracy in planning target volume (PTV) definition. This is a small series, but it highlights the fact that improved tumour delineation may provide more effective planning, with applications, for example, to IMRT to improve ultimate survival and local control. Similarly, IMRT may allow a reduction in dose to organs at risk. In 2004, Nuyttens et al.209 reported on 12 patients with rectal cancer who received small-bowel contrast both for planning and once a week during treatment. They reported that the total small bowel irradiated was reduced in IMRT when compared to conventional three-field plans of the same patients. Again conclusions are difficult to draw, but resource implications would be significant, and therefore cost effectiveness becomes an increasingly important issue. INTRA-OPERATIVE BRACHYTHERAPY BOOST
In 2003, the Mayo Clinic published their 10-year results for 35 patients treated with endocavitary radiotherapy to a total of 20–155 Gy using 50-Kv photons. A fractionated schedule was used that depended on the intention of treatment. For those treated curatively, 5-year and 10-year survival was 65 per cent and 42 per cent, respectively, with local control at 10 years of 76 per cent. Of those treated radically, 80 per cent had toxicity symptoms after 90 days.210
Other groups have reported feasibility studies of preoperative brachytherapy using 26 Gy in four fractions, followed by surgery and subsequently 45 Gy in 25 fractions in combination with 5-FU in T2 to early T4 disease, with postoperative adjuvant indications of positive nodal disease. Sixty-eight per cent of these patients had clinical responses, with 32 per cent positive for residual disease.211 The treatment was well tolerated, and may in future provide a further mechanism, if resources are available, of sphincter-preserving surgery. With early T1 tumours, endocavitary radiotherapy followed by a brachytherapy boost may provide sphincter preservation in 80 per cent of cases.212 Alternatively, brachytherapy may be helpful in those with locally recurrent disease involving the lateral pelvic side walls, or sacrum. However, these reports are of small series, and brachytherapy in this setting is not routinely used in the UK.
NATIONAL HEALTH SERVICE GUIDELINES There are key recommendations for colorectal cancer services.213 ●
●
●
●
●
Patient focus. Patients should be given full verbal and written information about their condition and about any treatment that may be offered. They should have continuing access to a member of the core team who can offer guidance and support. Multidisciplinary teams. Management of cancer care by multidisciplinary teams, which work to agreed protocols, is likely to facilitate the implementation of the Clinical Outcomes Group (COG) guidelines and improve the quality and coordination of care. These teams should include clinicians with up to-date knowledge of diagnosis and treatment of colorectal cancer, and specialized nursing staff who can support and advice patients. Endoscopy facilities. Adequate endoscopy facilities should be provided to help ensure accurate and timely diagnosis. The quality of endoscopy facilities – particularly colonoscopy completion and complication rates – should be monitored and staff given additional training when necessary to improve standards. Surgery for rectal cancer. Surgery for rectal cancer should be concentrated in the hands of surgeons who can demonstrate good results, particularly in terms of low recurrence rates. Surgeons should monitor their response by working closely with histopathologists. Improved pathology reporting. Pathology reporting should be sufficiently detailed to give comprehensive feedback on the adequacy of surgery, particularly for rectal cancer. Reports on surgical specimens should include data related to the size, type, grade and Dukes stage of tumour, and the involvement of lymph nodes and surgical margins. This information is important to guide treatment decisions, for routine collection of data on case-mix by cancer registries, and for monitoring long-term outcomes.
688 Colorectal cancer
●
Adjuvant therapies. Preoperative radiotherapy should be available for patients with rectal cancer. Adjuvant therapy can improve survival in some groups of patients and should be more widely available. Large-scale, nationally or internationally co-ordinated randomized trials should be supported in order to determine the best management of patients with colorectal cancer.
CONCLUSIONS Early diagnosis through case finding and screening of highrisk groups can make it possible to treat colorectal cancer at a stage at which prognosis is relatively good. Adjuvant chemotherapy and radiotherapy play a major role in improving survival. Particular attention should be paid to measuring quality of life in clinical trials and in routine practice. Teams including the relevant specialists are likely to be most effective in implementing these measures in an efficient and cost-effective way. Improved understanding of cancer biology and of therapeutics promises continuing progress in diagnosis and management in the coming decade.
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
Early diagnosis through case finding and screening of high-risk groups can make it possible to treat at a stage when prognosis is relatively good. Optimal treatment is best determined in an efficient and cost-effective way by multidisciplinary teams including the relevant specialists. Evidence has been provided that preoperative chemo-radiotherapy for rectal cancer is preferable to postoperative treatment. Adjuvant chemotherapy has been shown to be beneficial in localized disease. Due to the increasing choice of drugs, the treatment of individual patients should be tailored according to the relevant risk factors. Careful follow-up after initial surgery is justified by the potential for cure of some patients with metastases by surgical resection. Systemic treatment of metastatic disease has been shown to prolong survival. The choice of drug for first-line treatment remains a question of patient preference after discussion of the different toxicity profiles. Particular attention should be paid to measuring quality of life in clinical trials and in routine practice. Improved understanding of cancer biology and of therapeutics promises continuing progress in diagnosis and management in the coming decade.
REFERENCES 1 http://info.cancerresearchuk.org/cancerstats/ 2 OPCS. 1988 Cancer Statistics: Registrations, England and Wales. London: Government Statistical Service Series MB1 No 21, 1994. 3 Fearon ER, Vogelstein B. A genetic model of carcinogenesis. Cell 1990; 61:759. 4 Jacobs LR. Fiber and colon cancer. Gastroenterol Clin North Am 1988; 17:747–60. 5 Trock B, Lanza E, Greenwald P. Dietary fibre, vegetables and colon cancer: Critical review and meta-analysis of the epidemiologic evidence. J Natl Cancer Inst 1990; 82:650. 6 Thun MJ, Calle EE, Namboodiri MM, et al. Risk factors for fatal colon cancer in a large prospective study. J Natl Cancer Inst 1992; 84:1491–500. 7 Giovannucci E, Rimm EB, Stampfer MJ, et al. Intake of fat, meat and fiber in relation to risk of colon cancer in men. Cancer Res 1994; 54:2390–7. 8 Steinmetz KA, Potter JD. Vegetables, fruit and cancer: II. Mechanisms (Review). Cancer Causes Control 1991; 2:427–42. 9 Willett WC, Stampfer MJ, Colditz GA, et al. Relation of meat, fat and fiber intake to the risk of colon cancer in a prospective study amongst women. N Engl J Med 1990; 323:1664–72. 10 Gerhardsson M, Floderus B, Norell SE. Physical activity and colon cancer risk. Int J Epidemiol 1989; 18(3):728–9. 11 Peipins LA, Sandler RS. Epidemiology of colorectal adenomas. Epidemiol Rev 1994; 16(2):273–97. 12 Greenberg ER, Baron JA. Prospects for preventing colorectal cancer death. J Natl Cancer Inst 1993; 85:1182–4. 13 COMA. Nutritional Aspects of the Development of Cancer. London: Department of Health, 1998. 14 Halligan S, Altman DG, Taylor SA, Mallett S, Deeks JJ, Bartram CI, Atkin W. CT colonography in the detection of colorectal polyps and cancer: systematic review, meta-analysis, and proposed minimum data set for study level reporting. Radiology 2005; 237(3):893–904. 15 Crawford JM. The gastrointestinal tract. In: Cotran RS, Kumar V, Robbins S (eds) Robbins Pathologic Basis of Disease, 5th edn. Philadelphia: W.B. Saunders, 1994, 811–13. 16 Konishi F, Morson BC. Pathology of colorectal adenomas. J Clin Pathol 1982; 35:830–41. 17 Morson BC, Bussey HJR. Magnitude of risk for cancer patients and colorectal adenomas. Br J Surg 1985; 72(suppl):523–5. 18 Muto T, Bussey HJR, Morson BC. The evaluation of cancer of the colon and rectum. Cancer 1975; 36:2251–70. 19 Goh HS, Jass JR. DNA content and the adenoma-carcinoma sequence in the colorectum. J Clin Pathol 1986; 39:387–92. 20 Eide TJ. Prevalence and morphological features of adenomas of the large intestine with and without colorectal carcinoma. Histopathology 1986; 10:111–18. 21 Heald RJ, Lockhart-Mummery HE. The lesion of the second cancer of the large bowel. Br J Surg 1972; 59:16–19. 22 Bussey HJR, Wallace MH, Morson BC. Metachronous carcinoma of the large intestine and intestinal polyps. Proc R Soc Med 1967; 60:208–10.
References 689
23 Longacre TA, Fenoglio-Preiser CM. Mixed hyperplastic adenomatous polyps/serrated adenomas. Am J Surg Pathol 1990; 14:524–37. 24 Lennard-Jones JE, Morson BC, Ritchie JK, Williams CB. Cancer surveillance in ulcerative colitis. Experience over 15 years. Lancet 1983; 2:149–52. 25 Dhir V, Gopinath N. Endoscopic appearance of dysplasia and cancer in JBD Tytgat GNJ. Eur J Cancer 1995; 31:1174–7. 26 Blackstone M, Riddell R, Gerald Rogers BH, et al. Dysplasiaassociated lesion or mass (DALM) detected by colonoscopy in long-standing ulcerative colitis: an indication for colectomy. Gastroenterology 1981; 80:366–74. 27 Vasen HF, van der Luijt RB, Slors JF, et al. Molecular genetic tests as a guide to surgical management of familial adenomatous polyposis. Lancet 1996; 348(9025):433–5. 28 Jass JR, Smyrk TC, Stewart SM, et al. Pathology of hereditary non polyposis colorectal carcinoma. Anticancer Res 1994; 14:1631–4. 29 Jass JR. Diagnosis of hereditary non polyposis colorectal carcinoma. Histopathology 1998; 32:491–7. 30 Burt RW. Familial risk and colorectal cancer. Gastroenterol Clin North Am 1996; 25:793–803. 31 Syngal S, Fox EA, Eng C, Kolodner RD, Garber JE. Clinical criteria for HNPCC (Amsterdam II, Sensitivity and specificity of clinical criteria for hereditary non-polyposis colorectal cancer associated mutations in MSH2 and MLH1Sapna). J Med Genet 2000; 37:641–5. 32 Hamilton SR. Colon cancer testing and screening. Arch Pathol Lab Med 1999; 123(11):1027–9. 33 Burt RW, Di Sario JA, Cannon-Albright L. Genetics of colon cancer: Impact of inheritance on colon cancer risk. Annu Rev Med 1995; 46:371–9. 34 Hamilton SR, Liu B, Parsons RE, et al. The molecular basis of “Turcot syndrome”. N Engl J Med 1995; 332:839–47. 35 Spirio L, Olschwang S, Groden J, et al. Alleles of the APC gene: An attenuated form of familial polyposis. Cell 1993; 75:951–7. 36 Fuchs CS, Giovannucci E, Colditz GA, et al. A prospective study of family history and the risk of colorectal cancer. N Engl J Med 1994; 331:1669–94. 37 Dunlop MG, Farrington SM, Carothers AD, et al. Cancer risk associated with germline DNA mismatch repair gene mutations. Hum Mol Genet 1997; 6(1):105–10. 38 Houlston RS, Tomlinson IP. Polymorphisms and colorectal tumor risk. Gastroenterology 2001; 121(2):282–301. 39 Fearnhead NS, Wilding JL, Bodmer WF. Genetics of colorectal cancer: hereditary aspects and overview of colorectal tumorigenesis. Br Med Bull 2002; 64:27–43. 40 Rubinfeld B, Souza B, Albert I, et al. Association of the APC gene product with β catenin. Science 1993; 262:1731–4. 41 Powell SM, Zilz N, Beazer-Barclay Y, et al. APC mutations occur early during colorectal tumourigenesis. Nature 1992; 359:235–7. 42 Barbacid M. Ras genes. Annu Rev Biochem 1987; 56:779–827. 43 Fearon ER, Cho KR, Nigro JM. Identification of a chromosome 18q gene that is altered in colorectal cancers. Science 1990; 247:49–58.
44 Vogelstein B, Fearon ER, Hamilton SR, et al. Genetic alterations during colorectal tumour development. N Engl J Med 1988; 319:525–32. 45 Greenblatt MS, Bennett WP, Hollstein M, et al. Mutations in the P53 tumour suppressor gene: clues to cancer etiology and molecular pathogenesis. Cancer Res 1994; 54:4855–78. 46 Lane DP. Cancer, P53, guardian of the genome. Nature 1992; 358:15–16. 47 Boland SR, Sato J, Appelman HD, et al. Microallelotyping defines the sequence and tempo of allelic losses at tumour suppressor gene loci during colorectal cancer progression. Nat Med 1995; 1:902–9. 48 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?dbSnp 49 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?dbOMIM 50 Couzin J. Genomics. The HapMap gold rush: researchers mine a rich deposit. Science 2006; 312(5777):1131. 51 Chung CH, Bernard PS, Perou CM. Molecular portraits and the family tree of cancer. Nat Genet 2002; 32:533–40. 52 Ramaswamy S, Ross KN, Lander ES, Golub TR. A molecular signature of metastasis in primary solid tumors. Nat Genet 2002; 33:49–54. 53 Feinberg AP, Ohlsson R, Henikoff S. The epigenetic progenitor origin of human cancer. Nat Rev Genet 2006; 7(1):21–33. 54 Friedman D, Hill S, Keller S, Merchant NP, Levy SE, Coffey R, Caprioli RM. Proteome analysis of human colon cancer by two-dimensional difference gel electrophoresis and mass spectrometry. Proteomics 2004; 4:793–811. 55 Slater GI, Haber RH, Aufses AH. Changing distribution of carcinoma of the colon and rectum. Surg Gynaecol Obstet 1984; 158:716–18. 56 Gibbs NM. Undifferentiated carcinoma of the large intestine. Histopathology 1977; 1:77–84. 57 Mills SE, Allen MS Jr, Cohen AR. Small cell undifferentiated carcinoma of the colon. Am J Sur Pathol 1983; 7:643–51. 58 Sasaki O, Atkin WS, Jass JR. Mucinous carcinoma of the rectum. Histopathology 1987; 11:259–72. 59 Quirke P, Williams GT (Co-ordinators). Standards and Minimum Datasets for Reporting Common Cancers. Minimum Dataset for Colorectal Cancer Histopathology Reports. London: Royal College of Pathologists, 1998. 60 Adam IJ, Mohamdee MO, Martin IG, et al. Role of circumferential margin involvement in the local recurrence of rectal carcinoma. Lancet 1994; 344:707–10. 61 Quirke P. Assessing the quality of rectal surgery. Bulletin Royal College of Pathologists 1998; 104: Abstract VII. 62 Brown G. Thin section MRI in multidisciplinary pre-operative decision making for patients with rectal cancer. Br J Radiol 2005; 78 Spec No 2:S117–27. 63 Shepherd NA, Baxter K, Love S. The prognostic importance of peritoneal involvement in colonic carcinoma: A prospective evaluation. Gastroenterology 1997; 112: 1096–1102. 64 Dukes CE, Bussey HJR. The spread of rectal cancer and its effect on prognosis. Br J Cancer 1958; 12:309–20. 65 O’Brien MJ, Zamcheck N, Burke B, et al. Immunocytochemical localisation of carcinoembryonic antigen in benign and
690 Colorectal cancer
66
67
68 69 70
71 72
73
74
75
76
77
78
79 80
81
82
83 84
malignant colorectal tissues. Am J Clin Pathol 1981; 75:283–90. Cutait, R, Alvee VAF, Lopes LC, et al. Restaging of colorectal carcinoma based on the identification of lymph node micrometastases through immunoperoxidase staining of CEA and cytokeratins. Dis Colon Rectum 1991; 34:917–20. Jass JR, Atkin WS, Cuzick J, et al. The grading of rectal carcinoma. Historical perspectives and a multi variate analysis of 447 cases. Histopathology 1986; 10:437–59. Dukes CE. The classification of cancer of the rectum. J Pathol Bacteriol 1932; 35:373. Dukes CE. Cancer of the rectum: an analysis of 1000 cases. J Pathol Bacteriol 1940; 50:527–39. Turnbull RB, Kyle K, Watson FR, et al. Cancer of the rectum. The influence of the no touch isolation technique on survival rates. Ann Surg 1967; 166:420–7. Beahrs ON, Henson DE, Hutter RVP, et al. Manual for staging for cancer (4th ed). Philadelphia: JB Lippincott, 69-73. Compton C, Fenoglio-Preiser CM, Pettigrew N, Fielding LP. AJCC, American Joint Committee on Cancer Prognostic Factors Consensus Conference: Colorectal Working Group. Cancer 2000; 88:1739–57. Selby JV, Friedman GD, Queensbury CP, et al. A case control study of screening sigmoidoscopy and mortality from colorectal cancer. N Engl J Med 1992; 326:653–5. Aitken WS, Morson BC, Cuzick J. Long term risk of colorectal cancer after excision of rectosigmoid adenomas. N Engl J Med 1992; 326:658–62. Aitken WS, Hart A, McIntyre P, et al. Uptake, yield of neoplasia and adverse effects of flexible sigmoidoscopy screening. Gut 1998; 42:560–5. Hardcastle J, Chamberlain J, Robinson M, Moss S, Amar S, Balfour T. Randomised controlled trial of faecal-occult-blood screening for colorectal cancer. Lancet 1996; 348:1472–7. Kronborg O, Fenger C, Olsen J, Jorgensen O, Sondergaard O. Randomised study of screening for colorectal cancer with faecal-occult-blood test. Lancet 1996; 348:1467–71. Mandel J, Church T, Ederer F, Bond J. Colorectal cancer mortality: effectiveness of biennial screening for fecal occult blood. N Engl J Med 1999; 91:434–7. Aitken W. Implementing screening for colorectal cancer. Br Med J 1999; 319:1212–13. Dove-Edwin I, Sasieni P, Adams J, Thomas HJW. Prevention of colorectal cancer by colonoscopic surveillance in individuals with a family history of colorectal cancer: 16 year, prospective, follow-up study. BMJ 2005; 331:1047–9. Russell JB, Chu DZ, Russell MP, et al. When is polypectomy sufficient treatment for colorectal cancer in a polyp? Am J Surg 1990; 160:665–8. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. The National Polyp Study Workgroup. N Engl J Med 1993; 329(27):1977–81. Morson BC. The polyp story. Postgrad Med J 1984; 60: 820–4. Nivatvongs S, Rojanasakul A, Reiman HM, et al. The risk of lymph node metastasis in colorectal polyps with invasive adenocarcinoma. Dis Colon Rectum 1991; 34:323–8.
85 Chantereau MJ, Faivre J, Boutron MC, et al. Epidemiology, management, and prognosis of malignant large bowel polyps within a defined population. Gut 1992; 33:259–63. 86 Nozaki R, Takagi K, Takano M, Miyata M. Clinical investigation of colorectal cancer detected by follow-up colonoscopy after endoscopic polypectomy. Dis Colon Rectum 1997; 40(10 Suppl):S16–22. 87 Netzer P, Forster C, Biral R, et al. Risk factor assessment of endoscopically removed malignant colorectal polyps. Gut 1998; 43(5):669–74. 88 Karanjia ND, Schache DJ, Heald RJ. Function of the distal rectum after low anterior resection for carcinoma. Br J Surg 1992; 79:114–6. 89 Dehni N, Tiret E, Singland JD, Cunningham C, Schlegel RD, Guiguet M, Parc R. Long-term functional outcome after low anterior resection: comparison of low colorectal anastomosis and colonic J-pouch-anal anastomosis. Dis Colon Rectum 1998; 41(7):817–22; discussion 822–3. 90 Williams NS, Dixon MF, Johnston D. Reappraisal of the 5 centimetre rule of distal excision for carcinoma of the rectum: A study of distal intramural spread and of patients survival. Br J Surg 1983); 70:150–4. 91 Matheson NA, McIntosh CA, Krukowski Z. Continuing experience with single layer appositional anastomosis in the large bowel. Br J Surg 1985; 72:S104–6. 92 Guillou PJ, Quirke P, Thorpe H, et al. Short-term endpoints of conventional versus laparoscopic-assisted surgery. Lancet 2005; 365(9472):1718–26. 93 Leung KL, Kwok SP, Lam SC, et al. Laparoscopic resection of rectosigmoid carcinoma: prospective randomised trial. Lancet 2004; 363(9416):1187–92. 94 Izbicki JR, Hosch SB, Knoefel WT, Passlick B, Bloechle C, Broelsch CE. Extended resections are beneficial for patients with locally advanced colorectal cancer. Dis Colon Rectum 1995; 38(12):1251–6. 95 Poeze M, Houbiers JG, van de Velde CJ, Wobbes T, von Meyenfeldt MF. Radical resection of locally advanced colorectal cancer. Br J Surg 1995; 82(10):1386–90. 96 Shirouzu K, Isomoto H, Kakegawa T. Total pelvic exenteration for locally advanced colorectal carcinoma. Br J Surg 1996; 83(1):32–5. 97 Saito N, Sarashina H, Nunomura M, Koda K, Takiguchi N, Nakajima N. Clinical evaluation of nerve-sparing surgery combined with preoperative radiotherapy in advanced rectal cancer patients. Am J Surg 1998; 175(4):277–82. 98 Marsh PJ, James RD, Schofield PF. Adjuvant preoperative radiotherapy for locally advanced rectal carcinoma. Results of a prospective, randomized trial. Dis Colon Rectum 1994; 37(12):1205–14. 99 Ptok H, Meyer F, Marusch F, Steinert R, Gastinger I, Lippert H, Meyer L. Palliative stent implantation in the treatment of malignant colorectal obstruction. Surg Endosc 2006; 20(6):909–14. 100 Carne PW, Frye JN, Robertson GM, Frizelle FA. Stents or open operation for palliation of colorectal cancer: a retrospective, cohort study of perioperative outcome and long-term survival. Dis Colon Rectum 2004; 47(9):1455–61.
References 691
101 Daneker GWJ, Carlson GW, Hohn DC, Lynch P, Roubein L, Levin B. Endoscopic laser recanalization is effective for prevention and treatment of obstruction in sigmoid and rectal cancer. Arch Surg 1991; 126:1348–52. 102 Garcia Valdecasas JC, Llovera JM, deLacy AM, et al. Obstructing colorectal carcinomas. Prospective study. Dis Colon Rectum 1991; 34:759–62. 103 Phillips RKS, Hittinger R, Fry JS, et al. Malignant large bowel obstruction. Br J Surg 1985; 72:296–302. 104 Anonymous. Single-stage treatment for malignant left-sided colonic obstruction: a prospective randomized clinical trial comparing subtotal colectomy with segmental resection following intraoperative irrigation. The SCOTIA Study Group. Subtotal Colectomy versus On-table Irrigation and Anastomosis. Br J Surg 1995; 82(12):1622–7. 105 Lopez-Kostner F, Hool GR, Lavery IC. Management and causes of acute large-bowel obstruction. [Review]. Surg Clin North Am 1997; 77(6):1265–90. 106 Poon RT, Law WL, Chu KW, Wong J. Emergency resection and primary anastomosis for left-sided obstructing colorectal carcinoma in the elderly. Br J Surg 1998; 85(11):1539–42. 107 Anderson JH, Hole D, McArdle CS. Elective versus emergency surgery for patients with colorectal cancer. Br J Surg 1992; 79:706–9. 108 Sjodahl R, Franzen T, Nystrom PO. Primary versus staged resection for acute obstructing colorectal carcinoma. Br J Surg 1992; 79:685–8. 109 Deans GT, Krukowski ZH, Irwin ST. Malignant obstruction of the left colon. [Review]. Br J Surg 1994; 81(9):1270–6. 110 Bohm B, Schwenk W, Hucke HP, et al. Does methodic longterm follow-up affect survival after curative resection of colorectal carcinoma? Dis Colon Rectum 1993; 36:280–6. 111 Abulafi AM, Williams NS. Local recurrence of colorectal cancer: the problem, mechanisms, management and adjuvant therapy. [Review]. Br J Surg 1994; 81:7–19. 112 Pietra N, Sarli L, Costi R, Ouchemi C, Grattarola M, Peracchia A. Role of follow-up in management of local recurrences of colorectal cancer: a prospective, randomized study. Dis Colon Rectum 1998; 41(9):1127–133. 113 Chu DZ, Erickson CA, Russell MP, et al. Prognostic significance of carcinoembryonic antigen in colorectal carcinoma. Serum levels before and after resection and before recurrence. Arch Surg 1991; 126:314–6. 114 Rocklin MS, Senagore AJ, Talbott TM. Role of carcinoembryonic antigen and liver function tests in the detection of recurrent colorectal carcinoma. Dis Colon Rectum 1991; 34:794–7. 115 Rodriguez Bigas MA, Stulc JP, Davidson B, et al. Prognostic significance of anastomotic recurrence from colorectal adenocarcinoma. Dis Colon Rectum 1992; 35:838–42. 116 Killingback MJ. Indications for local excision of rectal cancer. Br J Surg 1985; 72:S54–6. 117 Bright N, Hale P, Mason R. Poor palliation of colorectal malignancy with the neodymium yttrium-aluminium-garnet laser. Br J Surg 1992; 79:308–9. 118 Wiering B, Krabbe PF, Jager GJ, Oyen WJ, Ruers TJ. The impact of fluor-18-deoxyglucose-positron emission
119 120
121
122
123
124
125
126 127
128
129
130
131
132
tomography in the management of colorectal liver metastases. Cancer 2005; 104(12):2658–70. Fusai G, Davidson BR. Management of colorectal liver metastases. Colorectal Dis 2003; 5(1):2–23. Simmonds PC, Primrose JN, Colquitt JL, Garden OJ, Poston GJ, Rees M. Surgical resection of hepatic metastases from colorectal cancer: a systematic review of published studies. Br J Cancer 2006; 94(7):982–99. Mack MG, Straub R, Eichler K, Sollner O, Lehnert T, Vogl TJ. Breast cancer metastases in liver: laser-induced interstitial thermotherapy – local tumor control rate and survival data. Radiology 2004; 233(2):400–9. Diaz R, Santaballa A, Munarriz B, Calderero V. Hepatic resection in breast cancer metastases: should it be considered standard treatment? Breast 2004; 3(3):254–8. Hawksworth J, Geisinger K, Zagoria R, Kavanagh P, Howerton R, Levine EA, Shen P. Surgical and ablative treatment for metastatic adenocarcinoma to the liver from unknown primary tumor. Am Surg 2004; 70(6):512–7. Weitz J, Blumgart LH, Fong Y, Jarnagin WR, D’Angelica M, Harrison LE, DeMatteo RP. Partial hepatectomy for metastases from noncolorectal, nonneuroendocrine carcinoma. Ann Surg 2005; 241(2):269–76. Shah SA, Haddad R, Al-Sukhni W, et al. Surgical resection of hepatic and pulmonary metastases from colorectal carcinoma. J Am Coll Surg 2006; 202(3):468–75. Hager, TH, Gall FP, Hermanek P, Local excision of cancer of the rectum. Dis Colon Rectum 1983; 26:149. Stockholm Rectal Cancer Study Group. Preoperative shortterm radiation therapy in operable rectal cancer: A randomized trial. Cancer 1990; 66:49–55. Sebag-Montefiore D, Steele R, Quirke P, et al. for the NCRI colorectal cancer study group and CR0. Routine short course pre-op radiotherapy or selective post-op chemoradiotherapy for resectable rectal cancer? Preliminary results of the MRC CR07 randomised trial. J Clin Oncol 2006 ASCO Annual Meeting Proceedings Part I. Vol 24, No. 18S (June 20 Supplement), 2006:3511. Quirke P, Sebag-Montefiore D, Steele R, et al., NCRI colorectal cancer study group and CR07. Local recurrence after rectal cancer resection is strongly related to the plane of surgical dissection and is further reduced by preoperative short course radiotherapy. Preliminary results of the Medical Research Council (MRC) CR07 trial. J Clin Oncol 2006 ASCO Annual Meeting Proceedings Part I. Vol 24, No. 18S (June 20 Supplement), 2006:3512. Birgisson H, Påhlman L, Gunnarsson U, Glimelius B. Adverse effects of preoperative radiation therapy for rectal cancer: long-term follow-up of the Swedish Rectal Cancer Trial. J Clin Oncol 2005; 23:8697–705. Folkesson J, Birgisson H, Påhlman L, Cedermark B, Glimelius B, Gunnarsson U. The Swedish Rectal Cancer Trial: long lasting benefits from radiotherapy on survival and local recurrence rate. J Clin Oncol 2005; 23:5644–50. Bakx R, Emous M, Legemate DA, et al. Categorization of major and minor complications in the treatment of patients with resectable rectal cancer using short-term
692 Colorectal cancer
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
pre-operative radiotherapy and total mesorectal excision: a Delphi round. Colorectal Dis 2006; 8(4):302–8. Colorectal Cancer Collaborative Group. Adjuvant radiotherapy for rectal cancer: a systematic overview of 8,507 patients from 22 randomised trials. Lancet 2001; 358:1291–4. Sauer R, Fietkau R, Wittekind C, et al; German Rectal Cancer Group. Adjuvant vs. neoadjuvant radiochemotherapy for locally advanced rectal cancer: the German trial CAO/ARO/AIO-94. Colorectal Dis 2003; 5:406–15. Dunst J, Reese T, Sutter T, Zuhlke H, Hinke A, KollingSchlebusch K, Frings S. Phase I trial evaluating the concurrent combination of radiotherapy and capecitabine in rectal cancer. J Clin Oncol 2002; 20:3983–91. Rödel C, Grabenbauer GG, Papadopoulos T, et al. Phase I/II trial of capecitabine, oxaliplatin, and radiation for rectal cancer. J Clin Oncol 2003; 15; 21(16):3098–104. Bozzetti F, Baratti D, Andreola S, et al. Preoperative radiation therapy for patients with T2-T3 carcinoma of the middle-to-lower rectum. Cancer 1999; 86(3):398–404. Mohiuddin M, Regine WF, Marks GJ, Marks JW. High-dose preoperative radiation and the challenge of sphincterpreservation surgery for cancer of the distal 2 cm of the rectum. Int J Radiat Oncol Biol Phys 1998; 40(3):569–74. Ahmad NR, Nagle D. Long-term results of preoperative radiation therapy alone for stage T3 and T4 rectal cancer. Br J Surg 1997; 84(10):1445–8. Wagman R, Minsky BD, Cohen AM, Guillem JG, Paty PP. Sphincter preservation in rectal cancer with preoperative radiation therapy and coloanal anastomosis: long term follow-up. Int J Radiat Oncol Biol Phys 1998; 42(1):51–7. Rosenthal SA, Yeung RS, Weese JL, Eisenberg BL, Hoffman JP, Coia LR, Hanks GE. Conservative management of extensive low-lying rectal carcinomas with transanal local excision and combined preoperative and postoperative radiation therapy. A report of a phase I-II trial. Cancer 1992; 69(2):335–41. Willet CG, Tepper JE, Donnely S, et al. Patterns of failure following local excision and local excision and postoperative radiation therapy for invasive rectal adenocarcinoma. J Clin Oncol 1989; 7:1003–8. Brizel HE, Tepperman BS. Postoperative adjuvant irradiation for adenocarcinoma of the rectum and sigmoid. Am J Clin Oncol 1984; 28:3. Mendenhall WM, Bland KI, Copeland EM, Summers GE, Pfaff WW, Souba WW, Million RR. Does preoperative radiation therapy enhance the probability of local control and survival in high risk distal rectal cancer? Ann Surg 1992; 215:696–705. Gunderson LK, Martin JK, Beart RW, et al. Intraoperative and external beam irradiation for locally advanced colorectal cancer. Ann Surg 1988; 207:52. Tepper JE, Cohen AM, Wood WC, et al. Intraoperative electron beam radiotherapy in the treatment of unresectable rectal cancer. Arch Surg 1986; 121:421 Whiting JF, Howes A, Osteen RT. Preoperative irradiation for unresectable carcinoma of the rectum. Surg Gynaecol Obstet 1993; 176:203–7.
148 Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 2006; 354:567–78. 149 Duttenhaver JR, Hoskins RB, Gunderson LK, et al. Adjuvant postoperative radiation therapy in the management of of adenocarcinoma of the colon. Cancer 1986; 57:955. 150 Withers HR, Cuasay L, Mason KA, et al. Elective radiation therapy in the curative treatment of cancer of the rectum and rectosigmoid colon. In: Stroehlein JR, Romsdahl MM (eds) Gastrointestinal Cancer. New York: Raven Press, 1981, 351. 151 Els M, Gross T, Ackermann C, Tondelli P. Incidence of ileus after rectal resection for rectal carcinoma, with and without adjuncitve radiation therapy. Schweiz Med Wochenschr 1992; 122:745–7. 152 Seow Choen F, Goh HS, Eu KW, Ho YH, Tay SK. A simple and effective treatment for hemorrhagic radiation proctitis using formalin. Dis Colon Rectum 1993; 36:135–8. 153 Moertel CG, Fleming TR, Macdonald JS, et al. Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. N Engl J Med 1990; 322(6):352–8. 154 Wolmark N, Rockette H, Mamounas E, et al. Clinical trial to assess the relative efficacy of fluorouracil and leucovorin, fluorouracil and levamisole, and fluorouracil, leucovorin, and levamisole in patients with Dukes’ B and C carcinoma of the colon: results from National Surgical Adjuvant Breast and Bowel Project C-04. J Clin Oncol 1999; 17(11):3553–9. 155 Kerr DJ, Gray R, McConkey C, Barnwell J. Adjuvant chemotherapy with 5-fluorouracil, L-folinic acid and levamisole for patients with colorectal cancer: nonrandomised comparison of weekly versus four-weekly schedules — less pain, same gain. QUASAR Colorectal Cancer Study Group. Ann Oncol 2000; 11(8):947–55. 156 Andre T, Colin P, Louvet C, Gamelin E, Bouche O, Achille E, et al. Semimonthly versus monthly regimen of fluorouracil and leucovorin administered for 24 or 36 weeks as adjuvant therapy in stage II and III colon cancer: results of a randomized trial. J Clin Oncol 2003; 21(15):2896–903. 157 Twelves C, Wong A, Nowacki MP, et al. Capecitabine as adjuvant treatment for stage III colon cancer. N Engl J Med 2005; 352(26):2696–704. 158 Andre T, Boni C, Mounedji-Boudiaf L, et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N Engl J Med 2004; 350(23):2343–51. 159 De Gramont A, Boni C, Navarro M, et al. Oxaliplatin/5FU/LV in the adjuvant treatment of stage II and stage III colon cancer: efficacy results with a median follow-up of 4 years. J Clin Oncol 2005; 23(16S):246. 160 Wolmark N, Wieand HS, Kuebler JP, Colangelo L, Smith RE. A phase III trial comparing FULV to FULV oxaliplatin in stage II or III carcinoma of the colon: results of NSABP Protocol C-07. J Clin Oncol 2005; 23(16S):246. 161 Van Cutsem E, Labianca R, Hossfeld D, et al. PETACC 3 Randomized phase III trial comparing infused irinotecan / 5fluorouracil (5-FU)/folinic acid (IF) versus 5-FU/FA (F) in stage III colon cancer patients (pts). (PETACC 3). J Clin Oncol 2005; 23(16S):3.
References 693
162 Ychou M, Raoul J, Douillard J, et al. for the GI Group of the FNCLCC and the FFCD A phase III randomized trial of LV5FU2CPT-11 vs. LV5FU2 alone in adjuvant high risk colon cancer (FNCLCC Accord02/FFCD9802). J Clin Oncol 2005; 23(16S):246. 163 Gray RG, Barnwell J, Hills R, McConkey C, Williams N, Kerr D on behalf of QUASAR Collaborative Group. QUASAR: a randomised study of adjuvant chemotherapy (CT) vs observation including 3238 colorectal cancer patients. J Clin Oncol 2004; 22(14S):245. 164 Mamounas E, Wieand S, Wolmark N, et al. Comparative efficacy of adjuvant chemotherapy in patients with Dukes’ B versus Dukes’ C colon cancer: results from four National Surgical Adjuvant Breast and Bowel Project adjuvant studies (C-01, C-02, C-03, and C-04). J Clin Oncol 1999; 17(5):1349–55. 165 Impact B2 investigators. International multicentre pooled analysis of B2 colon cancer trials. Efficacy of adjuvant fluorouracil and folinic acid in B2 colon cancer. J Clin Oncol 1999; 17:1356–63. 166 Cochrane review. Colorectal meta-analysis collaboration: Palliative chemotherapy for advanced or metastatic colorectal cancer (Cochrane Review). The Cochrane Library, Issue 2. Oxford: Update Software, 2000. 167 Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004; 350(23):2335–42. 168 Thirion P, Michiels S, Pignon JP, et al. Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: an updated meta-analysis. J Clin Oncol 2004; 22(18):3766–75. 169 Maughan TS, James RD, Kerr D, et al. Preliminary results of a multicentre randomised trial comparing 3 chemotherapy regimens (de Gramont, Lokich and Raltitrexed) in metastatic colorectal cancer. J Clin Oncol 1999; 18:262. 170 Van Cutsem E, Twelves C, Cassidy J, et al. Oral capecitabine compared with intravenous fluorouracil plus leucovorin in patients with metastatic colorectal cancer: results of a large phase III study. J Clin Oncol 2001; 19(21):4097–106. 171 Von Hoff DD, Rothenberg ML, Pitot HC. Irinotecan therapy for patients with previously treated metastatic colorectal cancer. Overall results of FDA approved pivotal US clinical trials. J Clin Oncol 1997; 16:803. 172 Cunningham G, Pyrhonen S, James R, et al. Randomised trial of irinotecan versus supportive care alone after fluorouracil failure for patients with metastatic colorectal cancer. Lancet 1998; 352:1407–12. 173 Rougier P, Van Cutsem E, Bajetta E, et al. Randomised trial of irinotecan versus fluorouracil by continuous infusion after fluorouracil failure in patients with metastatic colorectal cancer. Lancet 1998; 352(9138):1407–12. 174 Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. Irinotecan Study Group. N Engl J Med 2000; 343(13):905–14. 175 Douillard JY, Cunningham D, Roth AD, et al. Irinotecan combined with fluorouracil combined with fluorouracil alone
176
177
178
179
180
181
182
183
184
185
186
187
188
as first line treatment for metastatic colorectal cancer: a multicentre randomised trial. Lancet 2000; 355:1041–7. Cunningham D, Humblet Y, Siena S, et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecanrefractory metastatic colorectal cancer. N Engl J Med 2004; 351(4):337–45. Raymond E, Chaney SG, Taamma A, Cvitkovic E. Oxaliplatin: a review of preclinical and clinical studies. Ann Oncol 1998; 9(10):1053–71. Giacchetti S, Itzhaki M, Gruia G, et al. Long-term survival of patients with unresectable colorectal cancer liver metastases following infusional chemotherapy with 5fluorouracil, leucovorin, oxaliplatin and surgery. Ann Oncol 1999; 10(6):663–9. Tournigand C, Andre T, Achille E, et al. FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomized GERCOR study. J Clin Oncol 2004; 22(2):229–37. Goldberg RM, Sargent DJ, Morton RF, et al. A randomized controlled trial of fluorouracil plus leucovorin, irinotecan, and oxaliplatin combinations in patients with previously untreated metastatic colorectal cancer. J Clin Oncol 2004; 22(1):23–30. Johnston PG, Fisher ER, Rockette HE, et al. The role of thymidylate synthase expression in prognosis and outcome of adjuvant chemotherapy in patients with rectal cancer. J Clin Oncol 1994; 12(12):2640–7. Lenz HJ, Danenberg KD, Leichman CG, et al. p53 and thymidylate synthase expression in untreated stage II colon cancer: associations with recurrence, survival, and site. Clin Cancer Res 1998; 4(5):1227–34. Edler D, Hallstrom M, Johnston PG, Magnusson I, Ragnhammar P, Blomgren H. Thymidylate synthase expression: an independent prognostic factor for local recurrence, distant metastasis, disease-free and overall survival in rectal cancer. Clin Cancer Res 2000; 6(4):1378–84. Kawakami K, Omura K, Kanehira E, Watanabe Y. Polymorphic tandem repeats in the thymidylate synthase gene is associated with its protein expression in human gastrointestinal cancers. Anticancer Res 1999; 19(4B):3249–52. Salonga D, Danenberg KD, Johnson M, et al. Colorectal tumors responding to 5-fluorouracil have low gene expression levels of dihydropyrimidine dehydrogenase, thymidylate synthase, and thymidine phosphorylase. Clin Cancer Res 2000; 6(4):1322–27. Liang JT, Huang KC, Lai HS, et al. High-frequency microsatellite instability predicts better chemosensitivity to high-dose 5-fluorouracil plus leucovorin chemotherapy for stage IV sporadic colorectal cancer after palliative bowel resection. Int J Cancer 2002; 101(6):519–25. Ribic CM, Sargent DJ, Moore MJ, et al. Tumor microsatelliteinstability status as a predictor of benefit from fluorouracilbased adjuvant chemotherapy for colon cancer. N Engl J Med 2003; 349(3):247–57. Mariadason JM, Arango D, Shi Q, et al. Gene expression profiling-based prediction of response of colon carcinoma
694 Colorectal cancer
189
190
191
192
193
194
195
196
197
198
199
200
201
cells to 5-fluorouracil and camptothecin. Cancer Res 2003; 63(24):8791–812. Ogunbiyi OA, Flanagan FL, Dehdashti F, et al. Detection of recurrent and metastatic colorectal cancer: comparison of positron emission tomography and computed tomography. Ann Surg Oncol 1997; 4(8):613–20. Imdahl A, Reinhardt MJ, Nitzsche EU, et al. Impact of 18FFDG-positron emission tomography for decision making in colorectal cancer recurrences. Langenbecks Arch Surg 2000; 385:129–34. Begent RH, Keep PA, Searle F, et al. Radioimmunolocalization and selection for surgery in recurrent colorectal cancer. Br J Surg 1986; 73(1):64–7. Begent RH, Verhaar MJ, Chester KA, et al. Clinical evidence of efficient tumor targeting based on single-chain Fv antibody selected from a combinatorial library. Nat Med 1996; 2(9):979–84. August DA, Ottow RT, Sugarbaker A. Clinical perspective of human colorectal cancer metastasis. Cancer Metastasis Rev 1984; 3:303–24. Taylor I. Liver metastases from colorectal cancer: lessons from past and present clinical studies. Br J Surg 1996; 83(4):456–60. Millikan KW, Staren ED, Doolas A. Invasive therapy of metastatic colorectal cancer to the liver. Surg Clin North Am 1997; 77(1):27–48. Begent RHJ, Rustin GJS. Tumour markers: from carcinoembryonic antigen to products of hybridoma technology. Cancer Surv 1989; 8:108–21. Napier MP, Sharma SK, Springer CJ, et al. Antibody-directed enzyme prodrug therapy: efficacy and mechanism of action in colorectal carcinoma. Clin Cancer Res 2000; 6(3):765–72. Chester KA, Begent RH, Robson L, et al. Phage libraries for generation of clinically useful antibodies. Lancet 1994; 343(8895):455–6. Mayer A, Tsiompanou E, O’Malley D, et al. Radioimmunoguided surgery in colorectal cancer using a genetically engineered anti-CEA single-chain Fv antibody. Clin Cancer Res 2000; 6(5):1711–9. Sharma SK, Pedley RB, Bhatia J, et al. Sustained tumor regression of human colorectal cancer xenografts using a multifunctional mannosylated fusion protein in antibodydirected enzyme prodrug therapy. Clin Cancer Res 2005; 11(2 Pt 1):814–25. Lane DM, Eagle KF, Green AJ, Keep PA, Begent RHJ. Radioimmunotherapy of metastatic colorectal tumours with 131-Iodine antibody to CEA: Phase I/II study with
202
203 204
205
206
207
208
209
210
211
212
213
comparative bio-distribution of intact and F(ab’)2 antibodies. Br J Cancer 1994; 70:521–5. Pedley RB, Hill SA, Boxer GM, et al. Eradication of colorectal xenografts by combined radioimmunotherapy and combretastatin a-4 3-O-phosphate. Cancer Res 2001; 61(12):4716–4722. Springer CJ, Niculescu-Duvaz I. Prodrug-activating systems in suicide gene therapy. J Clin Invest 2000; 105(9):1161–7. Davies LC, Friedlos F, Hedley D, et al. Novel fluorinated prodrugs for activation by carboxypeptidase G2 showing good in vivo antitumor activity in gene-directed enzyme prodrug therapy. J Med Chem 2005; 48(16):5321–8. Mosolits S, Nilsson B, Mellstedt H. Towards therapeutic vaccines for colorectal carcinoma: a review of clinical trials. Expert Rev Vaccines 2005; 4(3):329–50. Coucke PA, Notter M, Stamm B, et al. On Behalf Of All Surgeons From Public Hospitals And Private Clinics. Preoperative hyperfractionated accelerated radiotherapy (HART) in locally advanced rectal cancer (LARC) immediately followed by surgery. A prospective phase II trial. Radiother Oncol 2006; (awaiting online print). Mohiuddin M, Winter K, Mitchell E, et al. Radiation Therapy Oncology Group Trial 0012. Randomized phase II study of neoadjuvant combined-modality chemoradiation for distal rectal cancer: Radiation Therapy Oncology Group Trial 0012. J Clin Oncol 2006; 24:650–5. Ciernik IF, Huser M, Burger C, Davis JB, Szekely G. Automated functional image-guided radiation treatment planning for rectal cancer. Int J Radiat Oncol Biol Phys 2005; 62:893–900. Nuyttens JJ, Robertson JM, Yan D, Martinez A. The influence of small bowel motion on both a conventional three-field and intensity modulated radiation therapy (IMRT) for rectal cancer. Cancer Radiother 2004; 8(5):297–304. Lavertu S, Schild SE, Gunderson LL, Haddock MG, Martenson JA. Endocavitary irradiation for rectal carcinoma: 10 year results. Am J Clin Oncol 2003; 26(5):508–12. Jakobsen A, Mortensen JP, Bisgaard C, Lindebjerg J, Hansen JW, Rafaelsen SR. Preoperative chemoradiation of locally advanced T3 rectal cancer combined with an endorectal boost. Int J Radiat Oncol Biol Phys 2006; 64:461–5. Coatmeur O, Truc G, Barillot I, Horiot JC, Maingon P. Treatment of T1-T2 rectal tumors by contact therapy and interstitial brachytherapy. Radiother Oncol 2004; 70:177–82. Cancer Guidance Subgroup of the Clinical Outcomes Group 1997
29 Anus BERNARD J. CUMMINGS
Anatomy Pathology Epidemiology and risk factors Presentation
695 695 696 696
ANATOMY The ‘anus’ is a general term for the outlet of the bowel and includes both the anal canal and peri-anal skin. The anal canal is defined in the major international cancer staging systems as that part of the intestine that extends from the rectum to the junction with the hair-bearing skin of the peri-anal region.1,2 The canal is 3–4 cm long, the superior limit being the palpable upper border of the anal sphincters and puborectalis muscles of the anorectal ring, and the distal limit, or anal verge, the level at which the walls of the canal come into contact in their normal resting state.3 The peri-anal area is the skin within a 5 cm radius of the anal verge. The term anal margin is used, often ambiguously, as a synonym for the distal canal and anal verge, the peri-anal skin immediately adjacent to the distal limit of the anal canal, or the peri-anal area as a whole. The peri-anal skin is similar histologically to hair-bearing skin elsewhere. At the anal verge, the skin blends with the modified squamous epithelium of the distal canal, which lacks hair and cutaneous glands.3 This modified squamous epithelium merges just below the pectinate or dentate line, the level of the anal valves, with a transitional-type epithelium that includes features of rectal, urothelial and squamous epithelia. The transitional zone extends proximally for about 2 cm, where it blends irregularly with the glandular mucosa of the rectum. There are three major lymphatic pathways from the anal tissues. These pathways are not exclusive, and there are numerous lymphatic connections between the various levels of the canal and the peri-anal skin close to the anal verge. Lymphatics from the uppermost part of the canal drain predominantly to the peri-rectal and superior haemorrhoidal
Diagnostic work-up and staging Management Key references References
696 698 702 703
nodes of the inferior mesenteric system. Those from the area around and above the dentate line flow to the internal pudendal, hypogastric and obturator nodes of the internal iliac system. Lymphatics from the distal canal, anal verge and peri-anal skin drain to the superficial inguinal nodes, and occasionally to the femoral nodes, of the external iliac system.
PATHOLOGY Malignant tumours of the anal region arise in the canal about four times as frequently as in the peri-anal skin. When the origin is in doubt, it is usual to classify the tumour as an anal-canal cancer. The current World Health Organization (WHO) histological classification of anal tumours describes intra-epithelial and invasive neoplasms.4 The term anal intra-epithelial neoplasia (AIN) is applied to pre-cancerous changes in the epithelium of the anal canal and peri-anal skin. The pathology and natural history of AIN have been reviewed elsewhere.5 Under the current WHO classification, the term squamous cell carcinoma is applied to all of the various subtypes, basaloid or cloacogenic, and large-cell keratinizing and non-keratinizing, previously used. Clinicians have for some time grouped the various subtypes of anal squamouscell cancer as ‘epidermoid’ cancers because of their similar natural history. About 85–90 per cent of primary anal-canal cancers are squamous cell type. The remaining 10–15 per cent are predominantly adenocarcinomas, most of which arise from anal glands. Adenocarcinomas from rectal-type mucosa in the proximal canal are usually classified as primary rectal cancers. The WHO system also lists rare variants,
696 Anus
including squamous cell carcinoma with mucinous microcysts, small-cell and undifferentiated cancers. Primary cancers of the peri-anal skin are similar to cancers of the skin arising in other sites. Most are squamouscell cancers, with occasional basal-cell cancers and skin adnexal adenocarcinomas.
EPIDEMIOLOGY AND RISK FACTORS Cancers of the anal region are about one-tenth as common as cancers of the rectum, although there is considerable geographic variation. The incidence has been rising in many countries over the past 30 years.6 Although the annual incidence according to most cancer registries is just below 1 per 100 000, the annual age-adjusted rates in the USA Surveillance, Epidemiology and End Results (SEER) Registry for 1994–2000 had risen to 1.59 per 100 000 for males and 1.84 per 100 000 for females. The long-time female predominance declined between 1973–1979 and 1994–2000. The incidence of AIN almost doubled in both men and women in that interval.7 Epidemiological studies have identified a number of factors associated with squamous cell anal cancer, although no unequivocal aetiologic pathway. Benign conditions such as fistulae, fissures and haemorrhoids do not predispose to anal cancer, nor do inflammatory bowel diseases such as Crohn’s disease and ulcerative colitis.8 Sexually transmitted factors are thought to be a significant cause of cancers of the anogenital area. The risk of anal cancer is increased in men, and to a lesser extent in women, who give a history of anoreceptive intercourse.9,10 Other significant factors include multiple sexual partners, several sexually transmitted viral and bacterial infections,9 and a history of cancer or intra-epithelial neoplasia of the vulva, vagina or cervix.11 Compromised cell-mediated immunity in patients infected with human immunodeficiency virus (HIV) is associated with an increased risk of anal cancer, although anal cancer is not at this time an AIDS-defining neoplasm. In the USA AIDS-Cancer Registry Study, the relative risk for anal cancer and AIN was 37.9 in HIV-positive men and 6.8 in HIV-positive women, compared with the general population.12 Risk factors not associated directly with a potentially sexually transmitted agent include cigarette smoking10 and iatrogenic immunosuppression following organ transplant.13 The sexually transmissible agent implicated as the most likely basis for these various epidemiological observations is human papillomavirus (HPV), although it is at present thought that HPV infection is not of itself sufficient to cause malignant conversion.14 Seventy per cent or more of analcanal and peri-anal cancers are HPV positive, with higher frequencies in anal-canal than in peri-anal cancers, in women and in patients who are HIV positive. Type 16 and, to a lesser extent, type 18 are the most commonly found genotypes, and the viral early genes E6 and E7 appear to play an important role in the initiation and maintenance of the malignant phenotype in HPV-positive cancers.14 There
is some geographical variation in the prevalence of the HPV types identified in anal-cancer tissue.15 The presence of HPV genetic material in anal-cancer cells does not appear to influence survival.16 The progression rate of AIN to invasive cancer is slow. It has been estimated at no more than 1 per cent per year in HIV-infected patients, a rate about 3 logs higher than that in the general population.17 Key events in the pathogenesis of anal squamous cell cancer include, but are not restricted to, integration of HPV DNA into the cellular genome, E6mediated functional inactivation of p53 protein, and loss of heterozygosity on chromosome 11q.18
PRESENTATION Most symptoms are non-specific. Bleeding, discharge and anal discomfort are reported by about half the patients with cancers of the canal, and about a quarter are aware of a mass. A palpable mass, discomfort and discharge are the most common presenting features of peri-anal tumours. A few asymptomatic primary cancers are found during physical examinations for other conditions, or during investigation of an enlarged inguinal node. Unsuspected superficial cancer or high-grade intra-epithelial neoplasia is sometimes found on histological examination of haemorrhoidectomy specimens or peri-anal condylomata. Gross faecal incontinence due to sphincter destruction or vaginal fistula formation occurs in fewer than 5 per cent, even in neglected cancers, which may reach considerable size. Symptomatic extra-pelvic metastases are a very rare first presenting indication of anal cancer.
DIAGNOSTIC WORK-UP AND STAGING Anal-canal cancers The features of greatest prognostic significance for survival are, in ascending order of seriousness, the size of the primary cancer, spread to regional lymph nodes, and extrapelvic metastases. The probability of retaining anal function depends on sphincter competence at presentation and the size of the primary cancer. Inguinal lymph-node metastases are clinically detectable in up to about 15 per cent of patients at presentation. In series managed by radical surgery, pelvic-node metastases were found in about 30 per cent, with about equal risk of involvement in the internal iliac and peri-rectal–superior haemorrhoidal node pathways.19,20 Extra-pelvic metastases are uncommon at presentation, being found in no more than 5 per cent, usually in the liver or lungs and occasionally in bone. The current UICC staging system is shown in Box 29.1. The primary tumour should be biopsied, if necessary under general anaesthesia. Since non-specific or reactive enlargement of the inguinal nodes is common, metastasis should be confirmed by fine-needle aspiration or excision
Diagnostic work-up and staging 697
Box 29.1 Anal canal TNM classification T – Primary tumour TX Primary tumour cannot be assessed T0 No evidence of primary tumour Tis Carcinoma-in-situ T1 Tumour 2 cm or less in greatest dimension T2 Tumour more than 2 cm but not more than 5 cm in greatest dimension T3 Tumour more than 5 cm in greatest dimension T4 Tumour of any size invades adjacent organ(s), e.g. vagina, urethra, bladder (involvement of sphincter muscle(s) alone is not classified as T4) N – Regional lymph nodes NX Regional lymph nodes cannot be assessed N0 No regional lymph-node metastasis N1 Metastasis in peri-rectal lymph node(s) N2 Metastasis in unilateral internal iliac and/or inguinal lymph node(s) N3 Metastases in peri-rectal and inguinal lymph nodes and/or bilateral internal iliac and/or inguinal lymph nodes M – Distant metastasis MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage grouping Stage 0 Stage I Stage II Stage IIIA
Stage IIIB Stage IV
Tis T1 T2 T3 T1 T2 T3 T4 T4 Any T Any T
N0 N0 N0 N0 N1 N1 N1 N0 N1 N2,N3 Any N
present none provides information to assess prognosis or guide the selection of treatment more effectively than the simple features of tumour size and the presence of lymphnode or other metastases outlined above.21,22
Peri-anal cancers As with cancers of the canal, the strongest prognostic factors for survival are the size of the primary tumour and the presence of inguinal-node or more distant metastases. Sphincter competence at presentation and the size of the primary cancer affect the likelihood of preservation of anal function. Inguinal-node metastases are found in no more than 5–10 per cent, and should be confirmed histologically. Computed tomography or MRI examination of the abdomen and pelvis is not indicated unless inguinal-node metastases are present or the cancer involves the anal canal. A chest X-ray is usually performed, but rarely discloses metastases. The same panel of blood tests as for anal-canal cancer should be obtained. The staging system for cancers of the peri-anal skin is the same as that for skin cancers elsewhere (Box 29.2).
Box 29.2 Peri-anal skin TNM classification
M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1
biopsy. Abdominal and pelvic computed tomography (CT) scans will disclose liver and large nodal metastases. Magnetic resonance imaging (MRI) and trans-anal ultrasound provide good detail of the depth of penetration of the primary cancer. Half of all node metastases are less than 0.5 cm in size, but small nodal metastases are not currently identified reliably by any available imaging modality. A standard chest X-ray is adequate to screen for pulmonary metastases. Localized skeletal symptoms should also be evaluated radiologically. Full blood count, renal and liver function tests and, if risk factors are present, HIV antibody tests should be performed. A variety of other tumour marker, cellular, molecular and histopathological features have been studied, but at
T – Primary tumour TX Primary tumour cannot be assessed T0 No evidence of primary tumour Tis Carcinoma-in-situ T1 Tumour 2 cm or less in greatest dimension T2 Tumour more than 2 cm but not more than 5 cm in greatest dimension T3 Tumour more than 5 cm in greatest dimension T4 Tumour invades deep extradermal structures, i.e. cartilage, skeletal muscle, or bone N – Regional lymph nodes (ipsilateral inguinal nodes) NX Regional lymph nodes cannot be assessed N0 No regional lymph-node metastasis N1 Regional lymph-node metastasis M – Distant metastasis MX Distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis Stage grouping Stage 0 Stage I Stage II Stage III Stage IV
Tis T1 T2 T3 T4 Any T Any T
N0 N0 N0 N0 N0 N1 Any N
M0 M0 M0 M0 M0 M0 M1
698 Anus
MANAGEMENT Anal-canal cancers Combined-modality radiation and chemotherapy, with radical surgery reserved for the management of recurrent cancer, is firmly established as the initial treatment of choice for squamous cell cancers of the anal canal. Except where stated, the following comments related to the management of squamous cell cancers. Prior to the adoption of combined-modality treatment, most patients were treated by radical surgery such as abdominoperineal resection, although some centres favoured radiation therapy. Radical resection has not been compared with radiation alone or with combined-modality treatment in formal randomized trials, but reviews of published results indicate that cure rates are comparable.23,24 Radiation-based treatment also permits the preservation of anorectal function in the majority of patients.24 Following the initial description in 1974 by Nigro, Vaitkevicius and Considine of the ability of concurrent radiation, 5-fluorouracil (5-FU) and mitomycin-C (MMC) to produce durable complete regression of anal cancers,25 a series of similar favourable reports led to the performance of three major randomized trials. These trials established that delivering 5-FU and MMC concurrently with radiation gave outcomes superior to those achieved by the same schedule of radiation alone or radiation and 5-FU.26–28 In the trial undertaken by the UK Co-ordinating Committee for Cancer Research (UKCCCR), 577 patients with all stages of epidermoid cancer of the anal canal or anal margin were randomly assigned to receive radiation alone (45 Gy in 20–25 fractions in 4–5 weeks) or radiation plus 5-FU (1000 mg/m2 per 24 hours for 96 hours or 750 mg/m2 per 24 hours for 120 hours) by continuous intravenous infusion in the first and final weeks of radiation treatment, and MMC (12 mg/m2) by bolus injection on day 1 of the first course of 5-FU only.27 Six weeks after the initial phase of treatment, patients received additional radiation without chemotherapy (15 Gy in six fractions by external-beam therapy or 25 Gy over 2–3 days by 192Ir implant). For those patients whose tumour showed less
than a partial response at the 6 weeks evaluation point (about 10 per cent of those treated by either regimen), surgery was performed rather than additional radiation. Local failure was diagnosed by 3 years in 39 per cent of the patients treated by radiation and chemotherapy, compared to 61 per cent of those who received radiation only. The definition of local failure was broader in this trial than in most: in addition to the presence of residual or recurrent cancer in the primary site or regional nodes, treatmentrelated morbidity requiring surgery, and inability to close a pre-treatment colostomy were also considered local failure. Surgery with colostomy was required for late treatment-related toxicity for ten patients in each study arm. Six patients in the combined-modality group and two in the radiation-alone group died of treatment-related morbidity. Primary tumour control, colostomy-free survival and cause-specific survival rates were all significantly improved by combined-modality treatment. The overall survival rate was also improved, although not to conventional levels of statistical significance (Table 29.1). The European Organisation for Research and Treatment of Cancer (EORTC) performed a similar study on 103 patients with advanced anal cancers.28 Pelvic radiation treatment of 45 Gy in 25 fractions over 5 weeks was combined in half the patients with 5-FU (750 mg/m2 per 24 hours for 120 hours by continuous intravenous infusion) in both the first and fifth weeks of radiation. In the first week only, a bolus injection of MMC (15 mg/m2) was given on the first day of the 5-FU infusion. Six weeks later, additional radiation was delivered by external-beam or interstitial techniques (15 Gy if there had been complete clinical response to the initial course of treatment, 20 Gy if response had been partial). Acute and late toxicity rates were similar in the treatment groups. The local control and colostomy-free survival rates were significantly better following combined-modality treatment, but, as in the UK trial, the improvement in overall survival rates was not statistically significant (see Table 29.1). In North America, the Radiation Therapy Oncology Group (RTOG) and the Eastern Cooperative Oncology Group (ECOG) combined to determine whether MMC could be omitted from the treatment regimen.26,29 In this
Table 29.1 Three-year results (percentages) of randomized trials of radiation alone (RT) versus radiation, 5-fluorouracil and mitomycin-C (RTCT) UKCCCR (n ⴝ 557)
Local control Cause-specific survival Overall survival
EORTC (n ⴝ 103)
RT
RTCT
p
RT
RTCT
p
39 61 58
61 72 65
0.0001 0.02 0.25
55 NA 65
65 NA 70
0.02 NA 0.17
UKCCCR, UK Co-ordinating Committee for Cancer Research; EORTC, European Organisation for Research and Treatment of Cancer.
Management 699
trial, 291 patients with cancers of the anal canal of any T and N category, without evidence of extra-pelvic metastases, were treated with external-beam pelvic radiation (45–50.4 Gy in 25–28 fractions over 5 weeks) plus two courses of 5-FU (1000 mg/m2 per 24 hours by continuous intravenous infusion for 96 hours), with or without MMC (10 mg/m2 by bolus injection on the first day of each 5-FU infusion), in the first and fifth weeks of radiation treatment. Biopsy of the primary tumour site 6 weeks after radiation was a prescribed part of the protocol. Biopsies were positive in 15 per cent of those who received 5-FU only with radiation and in 8 per cent of those treated with both 5-FU and MMC (not significant). Patients with residual cancer in whom it was thought preservation of anal function was still possible received additional radiation (9 Gy in five fractions in 1 week) concomitantly with a 96-hour infusion of 5-FU (1000 mg/m2 per 24 hours) and a short infusion of cisplatin (100 mg/m2 over 4–6 hours) on the second day of the infusion of 5-FU. Acute haematologic toxicity was more common after MMC, but otherwise the rates of acute and late toxicity were comparable in each group. The rates of colostomy-free and disease-free survival were significantly better in those treated with radiation, 5-FU and MMC, but overall survival rates at 5 years were identical (Table 29.2). The success of these combinations of radiation, 5-FU and MMC allowed substantial numbers of patients to retain anorectal function. In general, these and other studies did not include detailed prospective evaluation of the quality of anorectal function. Function is impaired in some patients.30–32 However, the general finding has been of ‘adequate’ function, and few patients have required surgery for incontinence. Low-grade symptomatic morbidity affecting the peri-anal skin, anorectum and other pelvic organs is usually managed medically, although with varying levels of success.30 The effectiveness of surgical salvage of residual cancer varies considerably,33,34 although overall, pelvic failure rates following combined-modality treatment and salvage therapy are lower than those previously recorded with radical surgery alone. There is no evidence that efforts to identify radiation–chemotherapy-resistant cancers by elective Table 29.2 Five-year results (percentages) of randomized trial of radiation and 5-fluorouracil (5FU) with/without mitomycin-C (MMC) RTOG/ECOG (n ⴝ 291)
Loco-regional control Disease-free survival Overall survival
RT5FU
RT5FUMMC
p
64 50 65
83 67 67
0.001 0.006 0.70
RTOG, Radiation Therapy Oncology Group; ECOG, Eastern Cooperative Oncology Group.
biopsy or by evaluation of the extent of clinical response at about 6 weeks leads to a more successful outcome than delaying salvage surgery until cancer re-growth is clinically apparent. Efforts have been made to improve local control rates by intensifying treatment, particularly radiation therapy. Radiation schedules can be intensified by increasing total dose and/or shortening the overall time in which treatment is delivered. Neither the randomized trials described earlier nor the numerous non-randomized studies have established the optimum schedules. When combined with 5-FU and MMC, radiation doses of as low as 30 Gy in 15 fractions in 3 weeks are capable of eradicating up to about 90 per cent of anal cancers 3 cm or less in size.24,35 Higher doses, from 45 Gy in 25 fractions in 5 weeks to 54 Gy in 30 fractions in 6 weeks, often supplemented by further radiation after an interval of 6–8 weeks to a total of 60–65 Gy, have controlled from 65 per cent to 75 per cent of primary tumours larger than 4 cm.24 Current studies of intensified treatment generally exclude small cancers (typically T1 category, up to 2 cm), but it is likely that the breakpoint by tumour size could be set higher and that some patients are over-treated. Interruptions in the radiation schedules, either elective or clinically necessary, are common in the treatment of anal cancer. The shorter interruptions are related principally to the discomfort associated with acute dermatitis and anoproctitis. Longer intervals have been introduced where decisions regarding further treatment are predicated on the extent of clinical or histopathological response to the initial phase of combined radiation and chemotherapy. Interruptions run counter to radiobiological theory, in that extended overall time of treatment may increase the risk of cancer re-growth. Anal cancers appear to have a short potential doubling time of about 4 days (range 1–30 days, n 26),36 an index that would argue against prolonged treatment times. Some investigators who have attempted to increase total radiation dose and shorten overall treatment time have encountered poor tolerance. For example, half the patients (9 of 18) in an RTOG phase II trial of 59.4 Gy in 33 fractions over 6.5 weeks required a break in treatment of 2 weeks or more.37 When overall treatment time has been considered in multivariate analyses of risk factors for failure, most have found that shorter treatment times are to be preferred.38 However, it is difficult to separate considerations of overall treatment time from those of total radiation dose and other components of the treatment schedule; the non-randomized nature of the studies means also that other unrecognized factors may have affected outcome. Some current trials seek to shorten overall treatment time by allowing interruptions only for those patients who experience severe toxicity, or by incorporating briefer intervals than the 6–8 weeks before the second phase of treatment used in earlier studies. The rationale for increasing the total radiation dose is based on analysis of non-randomized studies, which suggest
700 Anus
a dose–control relationship,39,40 and on subset analyses of the randomized trials, which sometimes show relatively low control rates even for smaller tumours with the doses used. In the UKCCCR trial, there was a 3-year local failure rate of 25 per cent in the T1T2N0 cancers treated with chemotherapy and radiation doses of up to 60 Gy (external beam) or 70 Gy (external beam plus interstitial radiation) over 12 weeks.41 Some groups are now using schedules that call for up to 60 Gy in 7 weeks without interruption, particularly for patients with larger tumours.37,42 However, a considerable number of patients treated with those protocols need unscheduled breaks in treatment. Most radiation protocols use techniques that encompass the whole lower pelvis in order to include the inguinal and posterior pelvic lymph nodes and the primary cancer in a single radiation volume. A more limited posterior pelvic radiation volume is also effective, and can be combined with separate radiation fields to the inguinal nodes if desired. There is some evidence that both acute and late morbidity are influenced by technique as well as by radiation dose-time–fractionation parameters, and the ideal technique has not yet been established.24 Recently developed techniques such as multiple coplanar oblique fields43 and intensity-modulated radiation therapy (IMRT)44 have been shown to reduce the extent and severity of peri-anal skin reactions and allow uninterrupted treatment to at least 54 Gy in 6 weeks, with concurrent chemotherapy. The dose intensity of chemotherapy may also affect outcome.45 The three randomized trials described above used short continuous infusions of 5-FU. These 4-day or 5-day infusions of 1000 mg/m2 per day generally deliver more 5FU in the same overall time period (and with much higher short-term serum and tissue levels) than infusions continued throughout the several weeks of radiation. Extended infusions, typically of about 225 mg/m2 per day for 5 weeks, are favoured by those who argue that it is desirable to seek a potentially synergistic radio-sensitizing interaction with 5-FU for every radiation fractional treatment.39 Most protocols continue to employ short-term infusions of 5-FU, although formal trials of protracted continuous infusions are in progress. The cytotoxic drugs that have received the greatest attention are 5-FU and MMC, but other drugs are being studied. Bleomycin was given concurrently with radiation in Sweden, but the results of non-randomized studies did not suggest benefit.46 There is currently widespread interest in the combination of radiation, 5-FU and cisplatin, based on its effectiveness against squamous cell cancers in other sites and on the potential of cisplatin as a radiation sensitizer. Non-randomized studies in anal cancer have described levels of local tumour control comparable to those produced by 5-FU, MMC and radiation.47,48 Several current randomized trials incorporate cisplatin. The UKCCCR Anal Cancer Trial II includes a double randomization, the first of which compares 5-FU plus MMC with 5-FU plus cisplatin concurrently with 50.4 Gy over 5.5 weeks without interruption. The second randomization
compares two courses of adjuvant 5-FU and cisplatin with no adjuvant therapy. The North American RTOG 9811 trial randomizes patients to radiation (59 Gy over 6.5 weeks, uninterrupted) with concurrent 5-FU/MMC or 5-FU/Cisplatin. Those in the latter arm also receive two courses of neoadjuvant 5-FU and cisplatin. The EORTC 22011 trial compares single-dose MMC and weekly cisplatin given concurrently with radiation (59.4 Gy, split course over 8.5 weeks) to MMC plus prolonged continuous infusion of 5-FU given concurrently with radiation. The Federation Francaise de Cancerologie Digestive (FFCD) 9804 trial includes only 5-FU and cisplatin and evaluates the merits of neoadjuvant chemotherapy and of different doses of radiation with concurrent chemotherapy.
LYMPH-NODE METASTASES
Metastases in regional lymph nodes can be eradicated by the same chemotherapy and radiation schedules used to treat the primary anal cancer. When patients are first diagnosed with anal-canal cancers, pelvic-node metastases are present in about 30 per cent and inguinal-node metastases in about 15 per cent.19,20 The probability of control of node metastases is similar to that of the primary cancer, although some consider it preferable locally to excise inguinal-node metastases prior to radiation and chemotherapy.49,50 Radical dissection of the inguino-femoral nodes either prior to or following high-dose radiation treatment should be avoided if possible because of the risk of poor healing and persistent leg oedema. Although relapse of treated node masses is uncommon, with local nodal control rates of 80 per cent or better, patients with nodal metastases are at increased risk of extra-pelvic metastases, and 5-year survival rates are usually 10–20 per cent less than in patients who do not have demonstrable inguinal-node or pelvic-node metastases. Late failure in inguinal nodes that were not enlarged at the time of first presentation has been reported in 15–25 per cent of patients whose primary cancer was treated surgically.19,51 Elective irradiation of clinically normal inguino-femoral nodes reduces the risk of late recurrence in these nodes to less than 5 per cent.24 Occasional authors prefer not to treat clinically negative inguinal nodes electively, but to reserve treatment for node metastases diagnosed later.52 The role of inguinal sentinel node biopsy is still being evaluated.53
EXTRA-PELVIC METASTASES
Extra-pelvic metastases are found in from 10 per cent to 20 per cent of patients. In the UKCCCR trial, metastases were the only recognized site of cancer in about one-quarter of those dying of anal cancer.27 The median survival time after the diagnosis of metastases is from about 8 to 12 months.7,54,55 Cisplatin and 5-FU produce responses more often than other drugs, although these responses are usually partial and short lived.56 Combinations of radiation, 5-FU and MMC or cisplatin appear relatively ineffective when used
Management 701
to treat metastases, perhaps because the doses of radiation that can be tolerated by organs such as the liver or lung are less than those directed to the pelvis. The role of adjuvant chemotherapy has not been determined, although trials of platinum-based induction and post-radiation chemotherapy are being undertaken, both to intensify treatment to the pelvic cancer and to evaluate the merits of systemic adjuvant therapy.57,58 RADIATION THERAPY ALONE
The preference for combined-modality treatment has meant that it has largely supplanted the use of radiation therapy alone. Combined-modality protocols have been shown to produce higher control rates than radiation alone for all sizes of anal-canal tumours treated by the relatively protracted radiation schedules tested. However, uninterrupted courses of radical radiation (typically equivalent to 50 Gy in 4 weeks or 60–65 Gy in 6 weeks) control 80 per cent or more cancers up to about 4 cm in size and about 50 per cent of larger tumours.24 Radiation alone should be considered when chemotherapy is contraindicated, since survival rates are comparable to those of surgery, and anorectal function can be preserved. The most common contraindications to standard doses of chemotherapy are age and co-morbid conditions, or concern about treatment-related toxicities. Some authors use modified doses59 rather than withhold chemotherapy altogether.60 SURGERY
The main role of surgery now is as salvage following unsuccessful radiation with/without chemotherapy, and as treatment for radiation-induced toxicity not amenable to conservative measures. Surgery is indicated for patients who cannot receive pelvic radiation, most commonly because of previous radiation for other malignancy. Those few patients who present with incontinence due to destruction of the anal sphincters or to vaginal fistula are probably best managed surgically. Preoperative or postoperative adjuvant radiation with/without chemotherapy should be considered to reduce the risk of pelvic recurrence. In series managed exclusively by surgery, 5-year survival rates ranged from about 50 per cent to 70 per cent, with loco-regional recurrence rates of 25–35 per cent.20,54,61,62 Most patients required excision of the anorectum and permanent colostomy. Local excision was generally considered only for superficial squamous cell cancers up to about 2–3 cm in size, since tumours of this size that have not penetrated the sphincter muscles are associated with a less than 10 per cent risk of nodal metastases,20 and adequate resection margins can be established without sacrificing the competence of the sphincters. From 25 per cent to 50 per cent of those who suffer failure of chemo-radiation are deemed unsuitable for attempted surgical salvage, usually because of the local extent of cancer. When salvage surgery is performed, the strongest negative predictor of survival is a
positive resection margin.34 Mesorectal nodal status in the salvage surgery specimen was not a useful predictor of further local recurrence or death from anal cancer.33 NON-SQUAMOUS CELL CANCERS
Adenocarcinomas of the anal canal are usually managed surgically. The choice between abdomino-anal excision and colostomy or local excision is based on the size of the cancer and general principles of oncologic surgery. Most series are small, but 5-year survival rates of less than 50 per cent, with local recurrence rates of about 25 per cent, are typical.62–64 Because anal adenocarcinomas are encountered infrequently, they are often managed by the protocols in place for adenocarcinomas of the rectum, although the roles of adjuvant radiation and chemotherapy for anal cancers are uncertain. Some centres have treated anal adenocarcinomas up to about 4–5 cm in size with radiation and a 5-FU-based regimen, using protocols similar to those for squamous cancer, and have preserved anorectal function in some.65 Small-cell and undifferentiated cancers metastasize early. No successful approach to treatment has been devised. Systemic chemotherapy similar to that used for small-cell cancers that arise elsewhere is sometimes given, but is of unproven value. The primary cancer may be managed by radiation-based protocols or by surgery.
Peri-anal cancer Anal intra-epithelial neoplasia is frequently multi-focal and affects both the peri-anal skin and anal canal, but is often first diagnosed in the peri-anal skin. A variety of treatments are used.5 Local excision, sometimes coupled with skin graft, is the preferred treatment of all histological subtypes of invasive peri-anal cancer when an adequate margin can be achieved without compromising anal function. The high local recurrence rates reported by some authors66 following resection of squamous cell cancers are probably due to failure to achieve clear resection margins and/or to multi-centric intra-epithelial neoplasia. Further local excision of recurrence is often sufficient to effect local control.66 Radiation therapy is indicated when surgery would compromise anorectal function. Whereas radiation therapy alone is effective for both limited and extensive squamous cell cancers,67,68 some centres prefer protocols of combined radiation and chemotherapy identical to those used for anal-canal cancer.69,70 The UK randomized trial included patients with cancers of the canal (about 75 per cent of cases) or anal margin (the latter when local excision was not possible). Results by site of origin were not reported, but local control and cause-specific survival rates in that trial favoured combined-modality treatment.27 Inguinal-node metastases are uncommon, except from large squamous cell or poorly differentiated cancers, and
702 Anus
may be managed by the same principles as node metastases from anal-canal cancer. Elective treatment of the groin nodes is favoured by some when the primary tumour is larger than about 4 cm or poorly differentiated. Death from peri-anal cancer is uncommon, and is generally associated with large and deeply infiltrating primary cancers or with distant metastases, the latter often preceded by involvement of inguinal nodes. The principles of management for adenocarcinomas and basal-cell cancers of the peri-anal skin are similar to those for squamous cell cancers. Anorectal function is preserved where possible by local excision or radiation therapy.
KEY LEARNING POINTS ●
●
Patients with HIV infection Squamous cell cancers of the anal canal or peri-anal skin are more common in patients with HIV infection, both prior to and after the diagnosis of AIDS.71 The treatment policies adopted for these patients are influenced not only by the presence and severity of any other illness affecting the patient, but also by the unusual sensitivity to radiation, and, to a lesser extent, to chemotherapy, demonstrated by some patients with AIDS. The risk of increased toxicity from radiation and/or chemotherapy and of a lower probability of local control of the anal cancer may be greater in patients with a lymphocyte CD4 count less than 200/μL at the time of starting treatment, or with AIDS.72,73 Management should be individualized, with judicious selection of observation, local excision, and radiation therapy alone or with chemotherapy, whenever possible reserving radical surgery for persistent local cancer causing symptoms or affecting continence. Overall local tumour control rates are variously reported as similar to or a little less than those for the general population with anal cancers of similar size. The introduction of highly active anti-retroviral therapy (HAART) has led to improved immune function and prolonged survival in HIV-infected patients. However, most studies have not shown an associated decrease in the prevalence of anogenital HPV infection or high-grade AIN, although in one study there was some reduction in low-grade AIN.74 Also, although the incidence of some AIDS-related malignancies has fallen since the introduction of HAART, that of anal cancer has not.75 It is likely that the incidence of HIV-positive patients with anal cancer will increase, as these patients survive longer when treated with HAART.
KEY REFERENCES Bartelink H, Roelofsen F, Eschwege F, et al. Concomitant radiotherapy and chemotherapy is superior to radiotherapy alone in the treatment of locally advanced anal cancer: results of a phase III randomized trial of the European Organization for Research and Treatment of Cancer Radiotherapy and Gastrointestinal Cooperative Groups. J Clin Oncol 1997; 15:2040–9.
●
●
Consider cancers of the anal canal and peri-anal skin separately. Anal-canal cancers Most are squamous cell. All variants of squamous cell cancer are managed similarly. Concurrent radiation, 5-fluorouracil and mitomycin-C is the standard treatment. Radical surgery is reserved for residual cancer, severe treatment-related toxicity, incontinence at presentation. Five-year survival rates are about 65 per cent. Local cancer control, with preservation of anal function, occurs in about 75 per cent of patients following radiation and chemotherapy. Current trials are investigating intensified radiation: higher doses, shorter overall time; cisplatin in place of mitomycin-C; induction and adjuvant chemotherapy. Peri-anal cancers Most are squamous cell. Local excision is the standard treatment if anorectal function can be preserved. Radiation alone, or concurrent radiation, 5fluorouracil and mitomycin-C is preferred for more extensive cancers. Radical surgery is reserved for residual cancer, severe treatment-related toxicity, incontinence at presentation. Five-year survival rates are about 80 per cent. Local cancer control, with preservation of anal function, occurs in about 85 per cent of patients. Whenever possible, cancer treatment should cure with as little interference with normal quality of life as possible. Most patients who present with anal cancer now have good prospects for both cure and retention of anorectal function. The preferred management is based on combinations of radiation and chemotherapy, with radical surgery reserved for residual cancer.
Cummings BJ, Brierley JD. Anal canal. In: Perez CA, Brady LW, Halperin EC, Schmidt-Ullrich RK (eds), Principles and Practice of Radiation Oncology, 4th edn. Philadelphia: Lippincott, Williams and Wilkins, 2003, 1630–48. Flam M, John M, Pajak TF, et al. Role of mitomycin in combination with fluorouracil and radiotherapy, and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: results of a phase III randomized intergroup study. J Clin Oncol 1996; 14:2527–39.
References 703
UKCCCR Working Party. Epidermoid anal cancer: results from the UKCCCR randomised trial of radiotherapy alone versus radiotherapy, 5-fluorouracil, and mitomycin. UKCCCR Anal Cancer Trial Working Party. UK Co-ordinating Committee on Cancer Research. Lancet 1996; 348:1049–54.
REFERENCES 1 Sobin LH, Wittekind C (eds). TNM Classification of Malignant Tumours, 6th edn. New York: Wiley-Liss, 2002. 2 Greene FL, Page DL, Fleming D, et al. (eds). AJCC Cancer Staging Manual, 6th edn. Philadelphia: Lippincott Raven, 2002. 3 Fenger C. Histology of the anal canal. AJSP 1988; 12:41–55. 4 Fenger C, Frisch M, Marti MC, Parc R. Tumours of the anal canal. In: Hamilton SR, Aaltonen LA (eds), Pathology and Genetics of Tumors of the Digestive System. Lyon: IARC Press, 2000, 145–55. 5 Abbasakoor F, Boulos PB. Anal intraepithelial neoplasia. Br J Surg 2005; 93:277–90. 6 Melbye M, Rabkin C, Frisch M, Biggar RJ. Changing patterns of anal cancer incidence in the United States, 1940–1989. Am J Epidemiol 1994; 139:772–80. 7 Johnson LG, Madeleine MM, Newcomer, LM, Schwartz SM, Daling JR. Anal cancer incidence and survival: the Surveillance, Epidemiology, and End Results experience, 1973–2000. Cancer 2004; 101:281–8. 8 Frisch M, Olsen, JH, Bautz A, Melbye M. Benign anal lesions and the risk of anal cancer. N Engl J Med 1994; 331:300–2. 9 Frisch M, Glimelius B, van den Brule AJ, et al. Sexually transmitted infection as a cause of anal cancer. N Engl J Med 1997; 337:1350–8. 10 Daling JR, Madeleine MM, Johnson LG, et al. Human papillomavirus, smoking and sexual practices in the etiology of anal cancer. Cancer 2004; 101:270–80. 11 Frisch M, Olsen JH, Melbye M. Malignancies that occur before and after anal cancer: clues to their etiology. Am J Epidemiol 1994; 140:12–19. 12 Frisch M, Biggar RJ, Engels EA, Goedert JJ. Association of cancer with AIDS-related immunosuppression in adults. JAMA 2001; 285:1736–45. 13 Penn I. Cancer is a complication of severe immunosuppression. Surg Gynecol Obstet 1986; 162:603–10. 14 zur Hausen H. Papillomaviruses in human cancers. Proc Assoc Am Phys 1999; 111:581–7. 15 Scholefield JH, Palmer JG, Shepherd NA, Love S, Miller KJ, Northover JM. Clinical and pathological correlates of HPV type 16 DNA in anal cancer. Int J Colorectal Dis 1990; 5:219–22. 16 Williams GR, Lu QL, Love SB, Talbot IC, Northover JM. Properties of HPV-positive and HPV-negative anal carcinomas. J Pathol 1996; 180:378–82. 17 Klencke BJ, Palefsky JM. Anal cancer: an HIV-associated cancer. Hematol/Oncol Clin North Am 2003; 17:859–72. 18 Gervaz P, Hahnloser D, Wolff BG, et al. Molecular biology of squamous cell carcinoma of the anus: a comparison of HIVpositive and HIV-negative patients. J Gastrointest Surg 2004; 8:1024–31.
19 Golden GT, Horsley JS III. Surgical management of epidermoid carcinoma of the anus. Am J Surg 1976; 131:275–80. 20 Boman BM, Moertel CG, O’Connell MJ, et al. Carcinoma of the anal canal. A clinical and pathologic study of 188 cases. Cancer 1984; 54:114–25. 21 Cummings BJ. Anal cancer. In: Gospodarowicz MK, Henson DE, Hutter RV, O’Sullivan B, Sobin LH, Wittekind C (eds), Prognostic Factors in Cancer, 2nd edn. New York: Wiley-Liss, 2001, 281–96. 22 Fenger C. Prognostic factors in anal carcinoma. Pathology 2002; 34:573–8. 23 Myerson RJ, Karnell LH, Menck HR. The National Cancer Data Base report on carcinoma of the anus. Cancer 1997; 80:805–15. 24 Cummings BJ, Brierley JD. Anal canal. In: Perez CA, Brady LW, Halperin EC, Schmidt-Ullrich RK (eds), Principles and Practice of Radiation Oncology, 4th edn. Philadelphia: Lippincott, Williams and Wilkins, 2003, 1630–48. 25 Nigro ND, Vaitkevicius VK, Considine B Jr. Combined therapy for cancer of the anal canal: a preliminary report. Dis Colon Rectum 1974; 15:354–6. 26 Flam M, John M, Pajak TF, et al. Role of mitomycin in combination with fluorouracil and radiotherapy, and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: results of a phase III randomized intergroup study. J Clin Oncol 1996; 14:2527–39. 26 UKCCCR Working Party. Epidermoid anal cancer: results from the UKCCCR randomised trial of radiotherapy alone versus radiotherapy, 5-fluorouracil, and mitomycin. UKCCCR Anal Cancer Trial Working Party. UK Co-ordinating Committee on Cancer Research. Lancet 1996; 348:1049–54. 27 Bartelink H, Roelofsen F, Eschwege F, et al. Concomitant radiotherapy and chemotherapy is superior to radiotherapy alone in the treatment of locally advanced anal cancer: results of a phase III randomized trial of the European Organization for Research and Treatment of Cancer Radiotherapy and Gastrointestinal Cooperative Groups. J Clin Oncol 1997; 15:2040–9. 28 Flam M, John M, Pajak TF, et al. Role of mitomycin in combination with fluorouracil and radiotherapy, and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: results of a phase III randomized intergroup study (author update). Classic Papers and Current Comments. Highlights of Clinical Gastrointestinal Research. Philadelphia: Lippincott Williams and Wilkins, 1999; 3:539–52. 29 Cummings BJ. Preservation of structure and function in epidermoid cancer of the anal canal. In: Rosenthal CJ, Rotman M (eds), Infusion Chemotherapy–Radiation Therapy Interactions: its Biology and Significance for Organ Salvage and Prevention of Second Primary Neoplasms. Amsterdam: Elsevier Science Publishing Co., 1998, 167–78. 30 Vordermark D, Sailer M, Flentje M, Thiede A, Kolbl O. Curative-intent radiation therapy in anal carcinoma: quality of life and sphincter function. Radiother Oncol 1999; 52:239–43.
704 Anus
31 Jephcott CR, Paltiel C, Hay J. Quality of life after non-surgical treatment of anal carcinoma: a case control study in long-term survivors. Clin Oncol 2004; 16:530–5. 32 Hill J, Meadows H, Haboubi N, Talbot IC, Northover JM. Pathologic staging of epidermoid anal carcinoma for the new era. Colorectal Dis 2003; 5:206–13. 33 Renehan AG, Saunders MP, Schofield PF, O’Dwyer ST. Patterns of local disease-failure and outcome after salvage surgery in patients with anal cancer. Br J Surg 2005; 92:605–14. 34 Nigro ND. An evaluation of combined therapy for squamous cell cancer of the anal canal. Dis Colon Rectum 1984; 27:763–6. 35 Wong CS, Tsang RW, Cummings BJ, et al. Proliferation parameters in epidermoid carcinomas of the anal canal. Radiother Oncol 2000; 56:349–53. 36 John M, Pajak T, Flam M, et al. Dose escalation in chemoradiation for anal cancer: preliminary results of RTOG 92–08. Cancer J Sci Am 1996; 2:205–11. 37 Graf R, Wust P, Hildebrandt B, et al. Impact of overall treatment time on local control of anal cancer treated with radiochemotherapy. Oncology 2003; 65:14–22. 38 Rich TA, Ajani JA, Morrison WH, Ota D, Levin B. Chemoradiation therapy for anal cancer: radiation plus continuous infusion of 5-fluorouracil with or without cisplatin. Radiother Oncol 1993; 27:209–15. 39 Constantinou EC, Daly W, Fung CY, Willett CG, Kaufman DS, Delaney TF. Time–dose considerations in the treatment of anal cancer. Int J Radiat Oncol Biol Phys 1997; 39:651–7. 40 Northover J, Meadows H, Ryan C, Gray R. Combined radiotherapy and chemotherapy for rectal cancer (letter). Lancet 1997; 349:205–6. 41 Martenson JA, Lipsitz SR, Wagner H Jr, et al. Initial results of a phase II trial of high dose radiation therapy, 5-fluorouracil, and cisplatin for patients with anal cancer (E4292): an Eastern Cooperative Oncology Group. Int J Radiat Oncol Biol Phys 1996; 35:745–9. 42 Vuong T, Devic S, Belliveau P, Muanza T, Hegyi G. Contribution of conformal therapy in the treatment of anal canal carcinoma with combined chemotherapy and radiotherapy: results of a phase II study. Int J Radiat Oncol Biol Phys 2003; 56:823–31. 43 Milano M, Jani AB, Farrey KJ, Rash C, Heimann R, Chmura SJ. Intensity-modulated radiation therapy (IMRT) in the treatment of anal cancer: toxicity and clinical outcome. Int J Radiat Oncol Biol Phys 2005; 63:354–61. 44 Ceresoli GL, Ferreri AJ, Cordio S, Villa E. Role of dose intensity in conservative treatment of anal canal carcinoma. Report of 35 cases. Oncology 1998; 55:525–32. 45 Glimelius B, Pahlman L. Radiation therapy of anal epidermoid carcinoma. Int J Radiat Oncol Biol Phys 1987; 13:305–12. 46 Gerard JP, Ayzac L, Hun D, et al. Treatment of anal canal carcinoma with high dose radiation therapy and concomitant fluorouracil–cisplatinum. Long-term results in 95 patients. Radiother Oncol 1998; 46:249–56. 47 Hung A, Crane C, Declos M, et al. Cisplatin-based combined modality therapy for anal carcinoma: a wider therapeutic index. Cancer 2003; 97:1195–202.
48 Papillon J. Rectal and Anal Cancers: Conservative Treatment by Irradiation. An Alternative to Radical Surgery. Berlin: SpringerVerlag, 1982. 49 Cummings BJ, Keane TJ, O’Sullivan B, Wong CS, Catton CN. Epidermoid anal cancer: treatment by radiation alone or by radiation and 5-fluorouracil with and without mitomycin C. Int J Radiat Oncol Biol Phys1991; 21:1115–25. 50 Stearns MW Jr, Urmacher C, Sternberg SS, Woodruff J, Attiyeh F. Cancer of the anal canal. Curr Probl Cancer 1980; 4:1–44. 51 Gerard JP, Chapet O, Samiei F, et al. Management of inguinal lymph node metastases in patients with carcinoma of the anal canal. Experience in a series of 270 patients treated in Lyon and review of the literature. Cancer 2001; 92:77–84. 52 Damin DC, Rosito MA, Schwartsmann G. Sentinel lymph node in carcinoma of the anal canal: a review. Eur J Surg Oncol 2006; 32:247–52. 53 Greenall MJ, Quan SH, Decosse JJ. Epidermoid cancer of the anus. Br J Surg 1985; 72:S97–103. 54 Tanum G. Treatment of relapsing anal carcinoma. Acta Oncol 1993: 32:33–5. 55 Flam MS. Chemotherapy of persistent, recurrent or metastatic cancer. In: Cohen AM, Winawer SJ, Friedman MA, Gunderson LL (eds), Cancer of the Colon, Rectum and Anus. New York: McGraw Hill Inc., 1995, 1051–60. 56 Meropol NJ, Niedzwiecki D, Shank B. Combined modality therapy of poor risk anal canal carcinoma: a phase II study of the Cancer and Leukemia Group B (CALGB). Proceedings of the Annual Meeting of the American Society of Clinical Oncology 1999; 18:A909. 57 Peiffert D, Giovannini M, Ducreux M, et al. High dose radiation therapy and neoadjuvant plus concomitant chemotherapy with 5-fluorouracil and cisplatinum in patients with locally advanced squamous cell anal canal cancer: final results of a Phase II study. Ann Oncol 2001; 12:397–404. 58 Charnley N, Choudhury A, Chesser P, Cooper RA, SebagMontefiore D. Effective treatment of anal cancer in the elderly with low-dose chemoradiotherapy. Br J Cancer 2005; 92:1221–5. 59 Chauveinc L, Buthaud X, Falcou MC, et al. Anal canal cancer treatment: practical limitations of routine prescription of concurrent chemotherapy and radiotherapy. Br J Cancer 2003; 89:2057–61. 60 Pintor MP, Northover JM, Nicholls RJ. Squamous cell carcinoma of the anus at one hospital from 1948 to 1984. Br J Surg 1989; 76:806–10. 61 Klas JV, Rothenberger DA, Wong WD, Madoff RD. Malignant tumors of the anal canal. The spectrum of disease, treatment and outcomes. Cancer 1999; 85:1686–93. 62 Abel ME, Chiu YSY, Russell TR, Volpe PA. Adenocarcinoma of the anal glands. Results of a survey. Dis Colon Rectum 1993; 36:383–7. 63 Tarazi R, Nelson RL. Anal adenocarcinoma: a comprehensive review. Semin Surg Oncol 1994; 10:235–40. 64 Belkacemi Y, Berger C, Poortmans P, et al. Management of anal canal adenocarcinoma: a large retrospective study from the Rare Cancer Network. Int J Radiat Oncol Biol Phys 2003; 56:1274–83.
References 705
65 Greenall MJ, Quan SH, Stearns MW, Urmacher C, DeCosse JJ. Epidermoid cancer of the anal margin. Pathologic features, treatment, and clinical results. Am J Surg 1985; 149:95–101. 66 Papillon J, Chassard JL. Respective roles of radiotherapy and surgery in the management of epidermoid carcinoma of the anal margin. Series of 57 patients. Dis Colon Rectum 1992; 35:422–9. 67 Peiffert D, Bey P, Pernot M, et al. Conservative treatment by irradiation of epidermoid carcinomas of the anal margin. Int J Radiat Oncol Biol Phys 1997; 39:57–66. 68 Cummings BJ, Keane TJ, Hawkins NV, O’Sullivan B. Treatment of perianal carcinoma by radiation plus chemotherapy (abstract). Proceedings of the American Society for Therapeutic Radiology and Oncology. Int J Radiat Oncol Biol Phys 1986; 12(Suppl. 1):170.
69 Melbye M, Cote TR, Kessler L, Gail M, Biggar RJ. High incidence of anal cancer among AIDS patients. The AIDS/Cancer Working Group. Lancet 1994; 343:636–9. 70 Hoffman R, Welton ML, Klencke B, Weinberg V, Krieg R. The significance of pretreatment CD4 count on the outcome and treatment tolerance of HIV-positive patients with anal cancer. Int J Radiat Oncol Biol Phys 1999; 44:127–31. 71 Place RJ, Gregorcyk SG, Huber PJ, Simmang CL. Outcome analysis of HIV-positive patients with anal squamous cell carcinoma. Dis Colon Rectum 2001; 44:505–12. 72 Heard I, Palefsky JM, Kazatchkine MD. The impact of HIV antiviral therapy on human papillomavirus (HPV) infections and HPV-related diseases. Antivir Ther 2004; 9:13–22. 73 Bower M, Powles T, Newsom-Davis T, et al. HIV-associated anal cancer. J Acquir Immune Defic Syndr 2004; 37:1563–5.
30 Germ-cell cancer of the testis and related neoplasms GRAHAM M. MEAD
Introduction Conclusions
706 725
INTRODUCTION Germ-cell cancers are rare malignancies of young adult life which occur in males in 97–98 per cent of cases. Histologically identical neoplasms may arise in multiple sites. Whilst the most common primary site is the testis, other recognized primary sites include the retroperitoneum, mediastinum, pineal/suprasellar area, the ovary and (in infants) the sacro-coccygeal region. This chapter focuses predominantly on testicular germ-cell cancers, but also includes sections on primary mediastinal and retroperitoneal disease. Over 95 per cent of patients presenting with a germ-cell tumour can expect to be cured.
Incidence This is the most common cancer of young men aged less than 35 years of age, but still comprises of only 1 per cent of male cancers. Approximately 1:400–450 males in the UK will develop a testicular germ-cell cancer in their lifetime, comprising, in the year 2000, approximately 2000 cases (7 cases per 100 000 population per year). Germ-cell cancers of the testis are divided histologically into testicular seminoma and teratoma; these two subtypes occur with approximately equal incidence. The median age of onset of teratoma is 25–30 years, and of seminoma approximately a decade later. Germ-cell tumours of the testis are rare before the age of 15 and comparatively rare after the age of 60, when large-cell nonHodgkin’s lymphoma becomes the commonest type of testicular cancer. Evidence from a variety of Western nations has shown a markedly increasing incidence of both types of germ-cell cancer in the last 40–50 years, with many
References
725
countries reporting a two-fold to three-fold increase in rates during this period.1* Simultaneously there is strong evidence to suggest that sperm counts are falling, and it seems likely that the same aetiological events underlie both abnormalities.
Aetiological factors (Table 30.1) Germ-cell tumours of the testis occur more commonly in white populations. The highest recorded incidence is seen in Nordic countries, in particular Denmark and Norway; however, paradoxically, the incidence in Finland is low. The main predisposing factor in the development of germ-cell cancer is testicular maldescent.2* If either testis fails to descend normally, the incidence of testicular cancer rises approximately five-fold (both testes are at increased risk), and if both are maldescended, the incidence rises tenfold. Patients undergoing orchidectomy for a germ-cell cancer have a markedly increased risk of developing contralateral disease – estimated at 4–5 per cent – and those with a family history of testicular germ-cell cancer (particularly siblings) have an increased risk of developing a tumour themselves of approximately five-fold to ten-fold.3* Additional risk factors are Down’s syndrome, the presence of testicular atrophy or dysgenesis, and a previous history of mumps orchitis. Other possible risk factors include higher social class, high maternal and low birth weight, and infection with human immunodeficiency virus (HIV). The incidence of testicular maldescent is rising in the UK. Whereas this abnormality is present in only 10 per cent of patients presenting with testicular cancer, it is thought that common prenatal influences underlie the rising incidence of both abnormalities. Interestingly, the incidence
Introduction 707
of testicular cancer among Danish males born during the period of invasion in World War II was markedly lower, suggesting a possible nutritional influence on the mother and developing male fetus.
Pathogenesis Linkage studies in patients with a family history of testis cancer have identified an abnormal gene on the X chromosome (denoted Xq27 or TGCT1), hypothesized to be one of the genetic causes of this disease in perhaps 10–20 per cent of cases.4* Histologically, carcinoma-in-situ (CIS)5 – the presence of atypical intratubular germ cells (Fig. 30.1) – is the uniform histological precursor of testicular germ-cell tumours (except for spermatocytic seminoma, a non-germ-cell cancer). Carcinoma-in-situ is present adjacent to recognizable cancer in almost all resected testes containing malignant germ-cell cancer. In addition, there is an increased incidence of CIS (of approximately 5 per cent) in the contralateral Table 30.1 Risk factors for testicular cancer Factor Unilateral testicular maldescent Bilateral testicular maldescent Family history: sib father Contralateral testicular cancer Mumps orchitis/testicular atrophy Childhood inguinal hernia or hydrocoele Down’s syndrome
Approximate increase in risk
5 10 10 2–3 10 unknown unknown unknown
Lifetime risk of testicular cancer in UK 1 in 450.
testis of men with testicular germ-cell tumours, particularly if the testis is atrophic. Carcinoma-in-situ generally occurs as a field change throughout the testis and can be detected by a biopsy of a single site. A longitudinal study on patients with this histological abnormality suggested an increasing incidence of invasive cancer with time, rising to 50 per cent by 5 years.6* Testicular cancer is extremely rare in patients in whom a previous biopsy has demonstrated the absence of in-situ cancer. Carcinoma-in-situ has never been shown to resolve spontaneously, although the histological abnormalities will regress, probably temporarily, with chemotherapy.7* and permanently with irradiation.8* It is currently hypothesized that CIS develops in early (probably prenatal) life and that promotional events, probably associated with puberty, cause later development of testicular cancer.5 TUMOUR CYTOGENETICS
Germ-cell cancers arising at any site are characterized in a high proportion of cases (around 80 per cent) by the presence of an isochromosome of the short arm of chromosome 12 – denoted i (12p). The precise role of this chromosomal abnormality in the pathogenesis of germcell tumours is at present unknown.
Pathology Germ-cell tumours of the testis have diverse and complex histopathological features, and no entirely satisfactory classification exists that adequately describes the potential admixtures of malignant and benign elements that may be present (and this heterogeneity is probably best described individually). Two histological classifications are in use at present, which are often used interchangeably: the British Testicular Tumour Panel and the World Health Organization (WHO) classifications (used in the USA) (Table 30.2).
Table 30.2 Pathology of male germ-cell cancer UK Seminoma Spermatocytic seminoma Malignant teratoma:
Figure 30.1 H&E section of testicular biopsy. Areas of carcinoma-in-situ are seen (ballooned germ cell, arrowed).
differentiated (TD) undifferentiated (MTU) intermediate (MTI) trophoblastic (MTT) yolk-sac tumour (YST) Combined tumour: seminoma and teratoma
USA Seminoma Spermatocytic seminoma Non-seminomatous germ-cell cancer: teratoma (mature, immature) embryonal carcinoma teratocarcinoma choriocarcinoma yolk-sac tumour Combined tumour: seminoma and teratoma
708 Germ-cell cancer of the testis and related neoplasms
Both classifications describe four main malignant cell types that can be recognized cytologically: seminoma, malignant teratoma undifferentiated (MTU, also known as embryonal carcinoma), malignant teratoma trophoblastic (MTT, choriocarcinoma), and yolk-sac tumour (YST). In addition, histologically ‘benign’ elements – teratoma differentiated (TD, mature teratoma) – can commonly be recognized. Testicular germ-cell cancers can potentially contain any of these elements in any proportion. Although metastatic disease normally reflects the tissue types found in the primary, it is not rare for histologies to differ between the two sites. Germ-cell cancers are commonly divided, for clinical purposes, into seminoma and non-seminoma germ-cell tumours. In a proportion of cases (at least 15 per cent), elements of both tumour types are present in the primary. For practical purposes, these patients are managed as non-seminoma. Available pathological and clinical data suggest that occasional patients with seminoma ‘transform’ histologically into teratoma – particularly of yolksac type.
Seminoma Seminoma of the testis has a characteristic uniform, solid, greyish appearance (Fig. 30.2). Histologically, the appearances are quite characteristic, and immuno-stains are not usually required. However, stains for placental alkaline phosphatase (PLAP, staining seminoma) and cytokeratin and alpha-fetoprotein (AFP, staining teratoma) will help distinguish difficult cases from teratoma. Anaplastic seminoma is a histologically less well-differentiated tumour; however, there are no clinical management implications in the recognition of such cases. Spermatocytic seminoma, a non-germ-cell tumour, has a different and also characteristic appearance in which CIS is absent. This is usually a tumour of elderly males, which does not metastasize from the primary site.
Figure 30.2 Orchidectomy specimen. Typical cross-section of seminoma of the testis.
Non-seminomatous tumours Malignant teratomas are commonly mixed tumours with different proportions of each malignant element (Fig. 30.3). In a proportion of cases with metastatic disease, spontaneous regression of the primary occurs, usually leaving a recognizable scar, although in-situ cancer may persist. Similar appearances are often seen if orchidectomy is performed after chemotherapy for metastatic disease. MALIGNANT TERATOMA UNDIFFERENTIATED
Malignant teratoma undifferentiated is known as embryonal carcinoma in the USA. This relatively featureless tumour is uncommonly found in pure form and has a higher propensity for vascular invasion and metastatic spread than the other tumour types. TERATOMA DIFFERENTIATED
Teratoma differentiated – known as mature teratoma in the WHO classification – is commonly a cystic neoplasm containing mesodermal, endodermal and ectodermal elements, which are histologically benign. Varying degrees of cellular atypia are usually present (immature teratoma). This tumour may be present in the testis in pure form in 2–3 per cent of cases,9 but more commonly, elements of TD are interspersed among malignant elements. The component tissues of TD are clearly unstable and secondary (non-germ-cell) cancers are well described as arising within the primary tumour or metastatic deposits of TD.10 Teratoma differentiated commonly metastasizes despite its benign appearance, most often to the retroperitoneum, and should be regarded as malignant neoplasm.9 MALIGNANT TERATOMA INTERMEDIATE
Malignant teratoma intermediate (MTI) is known as teratocarcinoma in the WHO classification and comprises a mixture of TD and MTU in varying degree.
Figure 30.3 Orchidectomy specimen. Typical teratoma of the testis. Complex cystic/solid tumour with areas of cartilage formation.
Staging 709
MALIGNANT TERATOMA TROPHOPLASTIC
In the UK classification of testicular teratoma, the presence of recognizable trophoblastic elements (cytotrophoblast and syncytiotrophoblast) is diagnostic of MTT. In the WHO classification, the equivalent diagnosis is choriocarcinoma, although this implies the presence of a pure trophoblastic tumour. Trophoblastic elements are responsible for human chorionic gonadotrophin (hCG) production (an essential marker of teratoma); they commonly spread widely via vascular channels.11
YOLK-SAC TUMOUR
Yolk-sac elements may be present in any teratoma. In addition, pure YSTs are recognized, particularly in the testis of infants and as a primary tumour of the anterior mediastinum in adults. Yolk-sac tumour is responsible for the production of AFP. These tumours tend to metastasize less commonly and less widely than some of the other malignant components.
Pathological staging The pathology report should include an assessment of gross tumour size. It is generally recommended that histological assessment should include examination of one block per cubic centimetre of resected cancer. The report should include details of the different histological types present, together with an assessment of their
proportion. Tumour extension through the tunica albuginea with involvement of the tunica vaginalis or involvement of the rete testis should be reported, as should spermatic cord or scrotal invasion. An assessment should be made as to whether vascular/lymphatic invasion is present, and a TNM stage should be allocated (Box 30.1).
Staging Once a diagnosis of testicular cancer has been made, the patient should be staged. This should include the following. 1. A thorough clinical examination including a careful search for gynaecomastia (first apparent as a preareolar breast ‘bud’) and for evidence of cervical, abdominal and inguinal masses. The remaining testis should also be examined and its size assessed. 2. Measurement of the serum markers AFP/hCG/lactate dehydrogenase (LDH). If these are found to be elevated, measurement should be repeated weekly to determine if they are rising or falling. 3. A chest X-ray and computed tomography (CT) scan of the chest, abdomen and pelvis. Brain CT or magnetic resonance imaging (MRI) should also be performed in high-risk patients. Positron emission tomography (PET) scanning has, as yet, no definite role in initial staging, but may be important after chemotherapy (see below). Individual components of the staging examinations are now described in more detail.
Box 30.1 TNM classification of primary tumour pT – Primary tumour pTX Primary tumour cannot be assessed (if no radical orchidectomy has been performed, TX is used) pT0 No evidence of primary tumour (e.g. histological scar in testis) pTis Intra-tubular germ-cell neoplasia (carcinoma-insitu) pT1 Tumour limited to testis and epididymis without vascular/lymphatic invasion; tumour may invade tunica albuginea but not tunica vaginalis pT2 Tumour limited to testis and epididymis with vascular/lymphatic invasion, or tumour extending through tunica albuginea with involvement of tunica vaginalis pT3 Tumour invades spermatic cord with or without vascular/lymphatic invasion pT4 Tumour invades scrotum with or without vascular/lymphatic invasion
SERUM MARKERS
Tumour marker (AFP, hCG or LDH) elevation is common in patients with germ-cell cancers (Table 30.3) and is of vital importance in their practical management. These markers have a number of uses. Diagnostically, elevation of both AFP and hCG in a male is pathognomonic of a diagnosis of malignant teratoma. The height of each tumour marker relates to prognosis in teratoma and helps to determine therapy. For example, very high and rising levels of any of the markers are adverse features in malignant teratoma that may indicate the need for a more intensive treatment approach. Markers are also used during therapy, to confirm response, and on follow-up should remain normal as useful (although not definitive) evidence of continued remission. Tumour markers (at least AFP and hCG levels) should initially be measured before orchidectomy. The presence of an elevated AFP level indicates that elements of malignant teratoma are present in the tumour. Post-orchidectomy marker levels (if elevated pre-orchidectomy) should be measured at least weekly and should, of course, return rapidly to normal in the absence of metastatic disease.
710 Germ-cell cancer of the testis and related neoplasms
Table 30.3 Serum markers (alpha-fetoprotein [AFP] and human chorionic gonadotrophin [hCG]) in groups of 50 patients (Southampton data) AFP % elevated
Range (ku/L)
hCG % elevated
Range (IU/L)
Seminoma Pre-orchidectomy All patients Stage I Pre-chemotherapy Metastatic disease
– –
– –
36 31
4–77 4–77
–
–
53
4–1647
Teratoma Pre-orchidectomy All patients Stage I Pre-chemotherapy Metastatic disease
68 63
11–33 000 11–32 000
70 67
4–57 000 4–61 800
70
12–218 000
72
4–777 000
Marker levels should always be obtained immediately before and after chemotherapy as they provide an essential guide to response. Alpha-fetoprotein Alpha-fetoprotein is an oncofoetal antigen produced in the fetal liver, yolk sac and gastrointestinal tract. Elevated levels of AFP occur in malignant teratoma (produced by yolksac elements), in patients with malignant hepatoma and occasionally other gastrointestinal malignancies, and in patients with liver-cell damage and accompanying proliferation (e.g. cirrhosis with regeneration). Minor elevations of AFP can also occur after chemotherapy with the POMB ACE chemotherapy regimen, perhaps related to the use of methotrexate. Alpha-fetoprotein is never elevated in patients with pure seminoma. It has a half-life of approximately 5–6 days. Elevation of AFP is found in approximately 70 per cent of teratoma patients pre-orchidectomy and a comparable number of cases pre-chemotherapy for metastatic disease. Beta subunit of human chorionic gonadotrophin Human chorionic gonadotrophin is a glycopeptide secreted by the placenta in pregnancy. It is comprised of an alpha and a beta subunit, the former common to luteinizing hormone (LH), follicle-stimulating hormone (FSH) and thyroidstimulating hormone (TSH), and the latter specific to hCG only. Elevation of hCG occurs in seminoma, usually at low levels (elevated in around 35 per cent of patients preorchidectomy), and malignant teratoma (elevated in approximately 70 per cent of patients pre-orchidectomy). It also occurs in a variety of other cancers, including lung cancer, stomach and bladder cancer. An elevated hCG is therefore not specific to germ-cell cancer. It has also been described in hypogonadal patients with high FSH or LH
levels, presumably reflecting cross-reactivity in the immunoassay. Elevation of hCG has also, rarely, been described in marijuana users. Human chorionic gonadotrophin is produced by the trophoblastic tumour elements; its half-life is 24–36 hours. Placental alkaline phosphatase Placental alkaline phosphatase is an iso-enzyme of alkaline phosphatase that is normally expressed by placental syncytiotrophoblasts. It has predominantly been used as a marker of seminoma. However, it has not found widespread use, particularly as its specificity is marred by an increase in levels in patients who are cigarette smokers. Lactate dehydrogenase Lactate dehydrogenase (or its iso-enzyme hydroxybutyrate dehydrogenase, HBD) is commonly elevated in both seminoma and teratoma. This glycolytic enzyme is present in all living cells, and the precise aetiology of its elevation in germcell tumours is unknown. The main use of LDH is in relationship to the assessment of prognosis, particularly in malignant teratoma, where increasing levels in patients with metastatic disease are related to a worsening outlook. It may be a useful marker in the follow-up of patients following chemotherapy for metastatic seminoma. Minor elevations of LDH are, however, rather non-specific, and sensitivity and specificity tend to be low. Lactate dehydrogenase is not a useful marker during or shortly after chemotherapy, as levels are increased at the time of bonemarrow recovery. Its assays vary widely and are not standardized. It is conventional to quote the degree of elevation of LDH as a ratio when compared with the upper limit of normal of each laboratory (e.g. level 900 IU, normal range 200–450 L, LDH ratio 2).
Staging 711
RADIOLOGY
Chest X-ray All patients with testicular cancer should have a chest X-ray. Peripheral lung metastases may be evidence of metastatic teratoma (Fig. 30.4) and a mediastinal mass evidence of seminoma or teratoma. Lung metastases occur much less commonly in seminoma than in teratoma, and this diagnosis should be reviewed if such lesions are present. An association has been described between malignant germ-cell tumours and sarcoidosis. The presence of isolated and otherwise unexplained mediastinal or hilar nodal enlargement should raise suspicion of the possibility of this lesion, and consideration should be given to a biopsy being performed if other confirmatory evidence of metastatic germ-cell cancer is unavailable (e.g. rising serum markers).
Figure 30.4 Widespread lung metastases secondary to malignant teratoma of the testis.
Computed tomography scanning Computed tomography scanning remains the imaging method of choice in all patients with a germ-cell tumour. Abdominal lymphatic spread in both tumour types occurs to characteristic sites (the so-called ‘landing zones’); these are the para-aortic area for left-sided primary tumours and the inter-aorto-caval and para-caval area for right-sided tumours (Fig. 30.5). Computed tomography scanning is capable only of detecting nodal enlargement, and the interpretation of scans relies on assessment of nodal size and number. Using 10 mm as the upper limit of normal nodal size in the retroperitoneum results in an acceptable level of sensitivity and specificity of detection of metastatic disease. Nodal enlargement in the ‘landing zones’ can be interpreted more definitely than that occurring elsewhere. Equivocal or atypical retroperitoneal nodal enlargement may occur in patients with malignant teratoma and should not be over-interpreted. If supporting evidence of malignant spread is present (e.g. rising tumour markers), the patient should proceed with treatment. However, if doubt exists and tumour markers are normal or falling, it is reasonable to delay therapy and repeat the abdominal scan after 4–6 weeks with little adverse effect on the patient as treatment of these diseases is so effective. Abdominal adenopathy may be massive and may then be associated with compression of one or both ureters or the inferior vena cava (characteristically occurring with right-sided tumours and sometimes associated with thrombosis and perhaps embolism). Contiguous spread may occur into the epidural space with spinal-cord compression. Rarely, patients may develop high gastrointestinal obstruction or pancreatitis because of the size of tumour deposits. Computed tomography scanning is also most effective at detecting lymphatic spread beyond the abdomen – characteristic initially to the retrocrural than posterior mediastinal nodes. This type of lymphatic spread is most characteristic of seminoma, which tends to metastasize in a predictable and contiguous fashion. Patients with retroperitoneal nodal spread may also develop retrograde
(a)
(b) Figure 30.5 Abdominal CT scans showing typical sites of spread of (a) left-sided and (b) right-sided metastatic disease.
spread to the contralateral ‘landing zone’, iliac or even inguinal regions, which should be routinely scanned in all patients at presentation. Whereas seminoma characteristically spreads by lymphatic routes, testicular teratoma commonly spreads by both lymphatic and vascular channels.
712 Germ-cell cancer of the testis and related neoplasms
Vascular spread occurs most commonly to the lungs, but occasionally also to the liver, brain, gastrointestinal tract, adrenals, kidneys and bones. It is not unusual in patients with malignant teratoma for spread to occur exclusively to the lungs in the absence of retroperitoneal adenopathy. However, small, single lung nodules should not be overinterpreted, and repeat scanning may be necessary in the absence of diagnostic marker elevation to determine whether these are benign or malignant in nature. A CT scan or MRI of the brain should be obtained in all patients with a high hCG level (e.g. 10 000 IU/L) and in those with florid lung metastases. Brain metastases are rare at presentation, but their early detection in high-risk cases is important. Metastatic masses from both teratoma and seminoma may show low attenuation areas secondary to necrosis or the presence of teratoma differentiated. Calcification may also be seen within either tumour type. Magnetic resonance imaging A potential advantage of this technique is the absence of irradiation in this young patient population. However, slow scan times, lack of availability, lack of oral contrast agents and claustrophobia limit the use of this technique, which has not been well validated in comparison with CT scanning. Magnetic resonance imaging scanning is, however, clearly the diagnostic imaging investigation of choice in patients with brain metastases or spinal-cord compression. However, it is not as effective as CT at examining the lungs, although there may be advantages in some cases with regard to the examination of mediastinal structures. No effective oral contrast medium is available for MRI. This considerably reduces its usefulness for examining the retroperitoneum – the main sites of spread of germ-cell cancer are closely related to the gastrointestinal tract, and oral contrast is used with CT scanning to aid differentiation from malignant masses. Positron emission tomography scanning Positron emission tomography scanning relies on the detection of enhanced uptake of radiolabelled fluorodeoxyglucose (FDG) by malignant tissue. A theoretical advantage of this technique when compared with CT or MRI is the possibility of detecting malignant infiltration of normal-size lymph nodes. Resolution is limited, however, as lesions less than 5 mm cannot be visualized. Positron emission tomography scanners are not at present widely available and scans are expensive. However, it is clear that PET scans can be highly effective and superior to CT in detecting metastatic disease, although such scans are not routinely necessary or indicated. Data from patients with stage I disease have shown that PET scanning can detect a proportion of patients destined to relapse as a result of low-volume disease ahead of similar detection by CT. Similarly, metastatic sites can occasionally be detected in patients with elevated markers but normal CT scans.
Positron emission tomography has also been evaluated post-chemotherapy in an attempt to judge whether residual masses contain viable cancer or only teratoma differentiated/fibrosis. Results have been variable in this setting, as necrotic tissue can occasionally be imaged by PET, as, virtually always, is cancer. Teratoma differentiated is not, however, associated with increased FDG uptake. Early data suggest an important role for PET in the evaluation of residual masses following chemotherapy in patients with bulky seminoma12 (see below). Positive FDG uptake appears to be strongly associated with residual active seminoma.
Clinical presentation The great majority of patients with a testicular germ-cell tumour present as a result of symptoms from their primary. Most commonly, patients notice a lump or change of texture or weight of the testis, or generalized swelling, which may be intermittently or progressively painful. Occasionally, patients present with acute onset of local pain, swelling and infective symptoms, with failure of the testis to return to normal after antibiotics, or with a hydrocoele obscuring a tumour mass. An unusual though characteristic presentation is with a shrinking, often painless, testis. Finally, the presence of a testicular cancer may first be manifest following trauma with associated swelling, which may have failed to settle satisfactorily. The duration of testicular symptoms varies from days to many years (the latter usually in patients with teratoma differentiated or seminoma). In approximately 10 per cent of patients, there will be a previous history of testicular maldescent (of either the affected or the non-affected testis) and occasionally a history of preceding childhood inguinal hernia repair or hydrocoele or a positive family history. Patients with an elevated hCG commonly notice nipple tenderness, sometimes together with gynaecomastia. A number of patients with germ-cell tumours present with symptoms from extra-gonadal disease – the testicular primary may not have been noticed, be clinically unapparent (and subsequently revealed by testicular ultrasound) or be absent (extra-gonadal primary, Fig. 30.6). The most common symptoms in this situation are lumbar backache secondary to metastatic para-aortic lymphadenopathy. Where this disease is bulky, obstruction of the kidney(s), bowel or inferior vena cava may occur (with possible pulmonary emboli), with appropriate symptoms and signs. Rarely, patients present with an inguinal lymph-node mass – usually a result of previous inguinal/scrotal surgery with resulting anomalous lymphatic drainage – or with a cervical nodal mass. Patients with metastatic teratoma may present with lung metastases, manifest by chest pain, dyspnoea or haemoptysis. Metastatic disease in the lung can be rapidly progressive, particularly in the choriocarcinoma syndrome,11 in which widespread vascular dissemination is common, and the outlook sometimes poor.
Management of malignant teratoma 713
Figure 30.6 Primary retroperitoneal choriocarcinoma. The patient presented with the abdominal mass shown and an human chorionic gonadotrophin (hCG) of 750 000 IU/L. Testicular ultrasound was normal. Chemotherapy resulted in cure.
ASSESSMENT AND MANAGEMENT OF THE PRIMARY
Once the diagnostic suspicion of germ-cell tumour has been raised, all patients should have bilateral testicular ultrasound to assess testicular size and intra-testicular abnormalities (e.g. calcification) and to look for a neoplasm within the body of the testis. The tumour markers AFP, hCG and LDH should be measured. Where a suspicious lesion is present in the testis, testicular exploration should be performed by an inguinal incision, and orchidectomy should be performed if a tumour is found. Rarely, scrotal exploration will be carried out in error as cancer has not been considered. In these cases, the wound should be closed and inguinal orchidectomy performed. More radical surgery is unnecessary. Occasionally, patients presenting with metastatic disease are diagnosed as a result of biopsy of one of these sites or are found to have diagnostic elevation of AFP or hCG. In such cases, particularly if performance status is poor or metastatic disease widespread, orchidectomy should be delayed and chemotherapy should be initiated at once. It is recommended, however, that orchidectomy is performed once treatment is completed.13
Management of malignant teratoma (non-seminomatous germ-cell tumour) Forty-five to 50 per cent of patients with germ-cell cancer of the testis have either pure malignant teratoma or (in approximately 15 per cent of cases) a combined malignant teratoma and seminoma. These latter cases are managed exclusively as teratoma. STAGE I TERATOMA
Approximately 60 per cent of patients with testicular teratoma will have stage I disease at first presentation, i.e.
normal physical examination and CT scans of chest, abdomen and pelvis; serum markers if initially elevated should fall to within the normal range. There is currently no proven role for PET scanning in these patients. Although this technique will aid early detection of metastatic disease in some cases, this is not likely to affect the already excellent outcome for these patients. Management approaches for these patients vary and may include surveillance, adjuvant chemotherapy or retroperitoneal lymph-node dissection (RPLND), although the last-mentioned approach is not used in the UK. There is no role for irradiation. Treatment is guided by the known risk factors for metastatic disease, and patients are generally divided into high-risk and low-risk groups, determined by the pathological features in the resected testis. The major factor predicting relapse is vascular (i.e. venous or lymphatic) invasion by cancer. This feature is present in approximately 25 per cent of cases and, when present, predicts a risk of subsequent relapse of approximately 50 per cent. Surveillance Following orchidectomy for stage I teratoma, the overall risk of relapse is 25–30 per cent.14 Relapses usually occur within a few months and are rare beyond 18–24 months. No standard surveillance protocol has been agreed; however, most oncologists recommend tumour marker measurements (AFP and hCG) and a chest X-ray monthly for 1 year, then 2-monthly for 1 year. In a Medical Research Council (MRC) trial (TE-08), the necessary frequency of CT scanning was evaluated, comparing a total of three whole-body scans (0, 3 and 12 months after orchidectomy) and six (0, 3, 6, 9, 12 and 24 months). No difference in outcome was found, and this study also confirmed the results of an earlier series suggesting that chest CT scanning post-staging could be replaced by a chest X-ray without detriment to the patient.15** This CT surveillance programme – initial whole-body staging CT scan followed by abdominal scanning only at 3 months and 1 year – can now be regarded as the standard approach. If patients without evidence of vascular invasion are selected for surveillance, the risk of relapse is only 15 per cent, versus 40–50 per cent for those with this feature. Follow-up, however, should be the same. The overall cure rate should approach 100 per cent, as salvage chemotherapy is highly effective (see ‘Treatment of metastatic teratoma’ below). Adjuvant chemotherapy Adjuvant chemotherapy has been tested in patients with stage I teratoma with a high risk of subsequent spread, identified by the presence of vascular invasion. Two cycles of modified bleomycin, etoposide and cisplatin (BEP, see ‘Treatment of metastatic teratoma’) chemotherapy, given with a total dose of etoposide of 360 mg/m2 in each course (Box 30.2), have been shown to reduce the relapse rate from 40 per cent to 1 per cent.16* This treatment approach has
714 Germ-cell cancer of the testis and related neoplasms
been widely adopted in the UK, but should be regarded as a matter of choice for patients, as cure rates should be exceptionally high whether or not this approach is taken. Those rare patients relapsing after adjuvant chemotherapy are reported as having a poor prognosis, but there are inadequate data in this area at present.
Box 30.2 Bleomycin, etoposide and cisplatin (BEP) for teratoma Adjuvant BEP16* Two courses Bleomycin 30 000 units i.v./i.m.weekly 6 Etoposide 120 mg/m2 i.v. daily 3 every 3 weeks Cisplatin 50 mg/m2 i.v. daily 2 every 3 weeks Good prognosis teratoma23** Three courses Bleomycin 30 000 units i.v./i.m. weekly 9 Etoposide 165 mg/m2 i.v. daily 3 every 3 weeks Cisplatin 50 mg/m2 i.v. daily 2 every 3 weeks Intermediate-prognosis or poor-prognosis metastatic teratoma24* Four courses Bleomycin 30 000 units i.v./i.m. weekly 12 Etoposide 100 mg/m2 i.v. daily 5 every 3 weeks Cisplatin 20 mg/m2 i.v. daily 5 every 3 weeks i.v., intravenous; i.m., intramuscular.
Retroperitoneal lymph-node dissection In the USA and many European countries, RPLND remains a routine treatment approach to stage I teratoma.17,18 The advent of more limited and therefore nerve-sparing surgery has dramatically reduced the previously common complication of retrograde ejaculation. Approximately 70–80 per cent of cases will have negative histology (pathological stage I); however, despite this finding, 10 per cent of unselected patients will relapse, almost always outside the abdomen. Patients in whom nodal disease is found and completely resected (stage II) have been reported in early series as having a 50 per cent relapse rate if no further treatment is given. However, if patients with raised tumour markers prior to surgery (stage IM) are excluded from surgery18, results can be excellent in experienced hands, with relapse rates of less than 5 per cent recorded for pathological stage I, and 10 per cent for low-volume pathological stage II. Adjuvant chemotherapy (e.g. two cycles of BEP) is recommended if malignant nodes larger than 2 cm are found; for remaining cases, chemotherapy given at the time of relapse may well suffice, and should result in virtually universal cure.
MANAGEMENT OF METASTATIC TERATOMA
In the USA and many European countries, RPLND remains the treatment of choice for early stage II patients,18,19 although increasingly patients with elevated serum markers, vascular invasion in the primary or tumour masses greater than 3 cm in diameter are treated with primary chemotherapy. In the UK, combination chemotherapy is regarded as the treatment of choice, and patients who have residual masses at the completion of treatment have these surgically excised at that time. Staging and prognostic factor classifications: teratoma Metastatic testicular teratoma is a heterogeneous disease in which spread frequently occurs by lymphatic and/or vascular routes. Unlike other malignancies, disease bulk may have no relevance to prognosis, as many of the more bulky metastatic masses largely comprise teratoma differentiated. In addition, these neoplasms are characterized by elevation of serum markers (AFP, hCG and LDH) – it has been know for many years that the degree of elevation of these markers relates to prognosis. A wide variety of different prognostic factor classifications were in use in the 1980s and 1990s; these bore little relationship to each other, and it proved increasingly difficult to translate clinical trial data between groups because of their use. In 1997, this situation was finally resolved when the International Germ Cell Consensus Classification (IGCCC) – a prognostic factor-based staging system – was introduced and widely adopted.20*** This classification resulted from worldwide collaboration. Retrospective clinical data were collected on more than 5000 patients with metastatic testicular teratoma or primary mediastinal or retroperitoneal teratoma who had been treated with cisplatin-containing chemotherapy. The median follow-up time was 5 years. Extensive data were analysed with regard to tumour markers, site of primary and sites and extent of metastatic disease. It rapidly became apparent that the main determinant of prognosis was the degree of elevation of all three tumour markers. Two additional adverse features were the presence of a mediastinal primary and spread to visceral sites (e.g. liver, bone and brain). The IGCCC collaboration resulted in the staging classification shown in Box 30.3. Clinical trials initiated since the introduction of this classification have stratified patients using the criteria shown. However, studies completed prior to the introduction of this classification have used a wide variety of staging classifications, which divided the metastatic testicular teratoma population into two or three groups. The older Royal Marsden classification continues to be helpful as a description of the extent of abdominal disease (stage IIA mass, 2 cm, IIB 2–5 cm, IIC 5 cm), as larger tumour masses pre-chemotherapy more often require resection once chemotherapy is completed.
Management of malignant teratoma 715
Box 30.3 Prognosis factors in metastatic teratoma, International Germ Cell Consensus Classification Good prognosis 56% of cases 5-year progression-free survival 89%; survival 92% Testis/retroperitoneal primary and No non-pulmonary visceral metastases and Good markers – all of: AFP 1000 KU/L and hCG 5000 IU/L and LDH 1.5 upper limit of normal Intermediate prognosis 28% of cases 5-year progression-free survival 75%; survival 80% Testis/retroperitoneal primary and No non-pulmonary visceral metastases and Intermediate markers – any of: AFP 1000 and 10 000 KU/L or hCG 5000 IU/L and 50 000 IU/L or LDH 1.5 N and 10 N upper limit of normal Poor prognosis 16% of cases 5-year progression-free survival 41%; survival 48% Mediastinal primary or Non-pulmonary visceral metastases or Poor markers – any of: AFP 10 000 KU/L or hCG 50 000 IU/L (10 000 IU/L) or LDH 10 upper limit of normal AFP, alpha-fetoprotein; hCG, human chorionic gonadotrophin; LDH, lactate dehydrogenase.
TREATMENT OF METASTATIC TERATOMA: BLEOMYCIN, ETOPOSIDE AND CISPLATIN AS STANDARD THERAPY
Metastatic testicular teratoma is a highly chemo-sensitive disease. The introduction of cisplatin, vinblastine and bleomycin (PVB) by the Indiana group in the late 1970s ranks as one of the single most important advances in oncology. Thirty years ago, 5–10 per cent of patients with metastatic teratoma were cured, whereas at present 85–90 per cent of such patients should be cured. This neoplasm is rare and its treatment complex. There is clear evidence that referral to specialist treatment centres, particularly for patients with intermediate-prognosis and poor-prognosis
disease, improves survival.21** Chemotherapy protocols should be strictly adhered to and surgery considered in all cases post-chemotherapy where indicated. Patients with non-threatening metastatic disease may wish to consider banking sperm prior to chemotherapy, although sperm counts frequently recover despite such treatment. The standard therapy for patients with all stages of metastatic testicular teratoma is BEP chemotherapy, which is given at 21-day intervals at a variety of doses and in a variety of schedules (see Box 30.2), including the following. ●
● ●
Adjuvant therapy for high-risk stage I disease. In these cases, BEP is given for two courses with reduced-dose etoposide.16 Good-risk teratoma: BEP is given for three courses.22**,23** Intermediate-prognosis or poor-prognosis disease: BEP is given for four courses.24**
Inpatient admission is required for BEP chemotherapy, although the possibility of delivering this treatment on an outpatient basis is being explored. The treatment can almost always be delivered at full dose, on time (every 3 weeks). Certainly full-dose therapy should be given on day 1 if the white cell count is 2.0 109/L with a platelet count 120 109/L. Cisplatin, given as part of BEP, is generally administered in divided dose, and is equally effective given twice daily (50 mg/m2) or five times per day (20 mg/m2).23** Intravenous bleomycin can be given virtually regardless of blood counts, but should be used with considerable care in older patients (40 years) or if the serum creatinine is elevated or the creatinine clearance reduced below 60 mL/min (see below). There is no evidence that the use of additional growth factors (e.g. filgrastim) is necessary or affects treatment results.25** Good-prognosis disease Cure rates for good-prognosis disease (see Box 30.3) should exceed 95 per cent in the group defined by IGCCC criteria. Commencing from early studies in which four courses of BEP were regarded as standard, a number of important studies have attempted to reduce treatment toxicity while maintaining efficacy in these patients. These approaches have included the following. ●
Omission of bleomycin. Bleomycin is a well-tolerated and long-established treatment for testicular teratoma. Although this drug is relatively non-myelosuppressive, a feared, though infrequent, complication of its use is pulmonary toxicity (see section Chemotherapy-related toxicity). Because of this complication, attempts have been made to omit bleomycin from BEP chemotherapy. Indeed, a standard treatment approach advocated by the Memorial Sloan Kettering26 is the use of EP therapy (cisplatin 20 mg/m2 per day 5 together with etoposide 100 mg/m2 intravenously daily 5 on four occasions). Bleomycin toxicity occurs with increasing dose, with a
716 Germ-cell cancer of the testis and related neoplasms
recommended maximum in this group of 270 000 IU, but also occurs when renal clearance is impaired (bleomycin should be given with caution in any patient with an elevated creatinine or creatinine clearance 60 mL/min) or in older (40 years) patients, for whom lower doses may be considered (see below). Four large trials have examined the role of bleomycin in BEP (three or four cycles) in metastatic teratoma.27**,28**,29**,30** In summary, these studies have shown that whereas bleomycin clearly increases myelotoxicity, and causes occasional mortality from bleomycin lung, overall failure-free survival is improved when this drug is used. One of these studies30** compared BEP 3 against EP 4. Although survival in this study was equivalent, more adverse cancer events occurred in the group of patients receiving EP, and BEP was recommended as standard therapy. It is now generally recommended that if three cycles of BEP are to be used, bleomycin should be given weekly to a total dose of 270 000 units, a cumulative dose that should rarely be complicated by bleomycin lung. ●
Reduction in number of treatment courses from four to three. The South Eastern Cancer Study Group (SECSG),31** in a comparatively small study (184 patients), randomized between BEP 4 and BEP 3 courses. Three courses of treatment proved as effective as four, with reduced toxicity. Concerns were raised, however, about the statistical power with such relatively small numbers, and therefore this trial was repeated by the MRC/EORTC (European Organisation for Research into Treatment of Cancer) group. The results of this study supported the SECSG result, with 812 patients randomized between BEP 3 courses and BEP 4 courses. Failure-free and overall survivals were respectively 90.4 per cent and 89.4 per cent and 97.0 per cent and 97.1 per cent.23** This study also examined whether cisplatin and etoposide should be administered over a 3-day versus 5-day period, and again found equivalence with regard to cure rates. Toxicity did vary, however, with increased gastrointestinal toxicity and tinnitus in patients treated over 3 rather than 5 days, particularly if treated for four courses.32**
●
Substitution of carboplatin for cisplatin. Carboplatin is a much better tolerated drug than cisplatin and can be given on an outpatient basis. Preliminary phase II data suggested equivalence when carboplatin was substituted for cisplatin in BEP, however two randomised trials33,34** clearly demonstrated that carboplatin is inferior to cisplatin at the doses used in these studies. There is at present no role for carboplatin in the initial therapy of metastatic teratoma.
Good-prognosis disease: summary Three cycles of BEP chemotherapy with bleomycin given to a total dose of 270 000 units have clearly been shown to be standard therapy for these patients.23** In occasional
patients, when there are concerns about bleomycin dose, in particular older patients, it remains reasonable to treat with four cycles of EP (which, in the author’s nonevidence-based practice, is supplemented with a single dose of bleomycin 30 000 units at 3-weekly intervals). Occasional patients intolerant of intravenous infusions because of severe learning difficulties can still be treated by substituting carboplatin for cisplatin, although it has to be accepted that cure rates will be lower. INTERMEDIATE-PROGNOSIS AND POOR-PROGNOSIS TERATOMA
Approximately 30 per cent of patients with teratoma fall into the intermediate-prognosis and poor-prognosis group. These patients often have rapidly progressive disease and treatment should proceed with relative urgency. Disease in the chest can be bulky and associated with intrapulmonary haemorrhage. Occasional patients have ensuing rapidly progressive dyspnoea and require urgent therapy, with careful attention to the avoidance of excess hydration. Abdominal disease can be associated with inferior vena cava obstruction (particularly with right-sided tumours) or with ureteric obstruction that may require stenting. Clinicians should be aware of the possibility of central nervous system (CNS) disease in these patients. Standard chemotherapy remains BEP (etoposide dose 500 mg/m2 per course) given for four courses with a total bleomycin dose (30 000 units weekly) of 360 000 units. There is no evidence that filgrastim or other growth factors add to the efficacy of this regime.25** A number of phase III trials have explored dose escalation of cisplatin, increased dose intensity and addition/substitution of other drugs (particularly ifosfamide), but have failed to improve on the results achieved with BEP – whilst at the same time increasing toxicity.24**,35** A number of more intensive treatment approaches to these patients have been explored. In a current MRC/EORTC study in patients with intermediate-prognosis disease, four cycles of BEP are being compared with four cycles of T-BEP (paclitaxel BEP, supported by granulocyte colony-stimulating factor, G-CSF). POMB ACE chemotherapy (cisplatin, vincristine, methotrexate and bleomycin alternating with dactinomycin, cyclophosphamide and etoposide) has been extensively evaluated at Charing Cross Hospital, London, with excellent results,36* but has not been tested in a phase III trial. CBOP-BEP, a regimen employing initial intensive induction with both carboplatin and cisplatin with vincristine and infusional bleomycin, followed by BEP, has also given promising results;37* again, only in phase II evaluation, although a phase III comparison with BEP has now started in the UK in patients with poor-prognosis disease. Further dose escalation has been explored. The Memorial Sloan Kettering Hospital in New York has pioneered the use of high-dose chemotherapy with stem-cell support as part of the initial treatment approach. A recently
Management of malignant teratoma 717
completed inter-group phase III study in poor-prognosis patients has compared four cycles of standard BEP with two cycles of BEP followed by two cycles of high-dose carboplatin, etoposide and cyclophosphamide with peripheral stem-cell support. A German group has taken a different approach to this patient group by administering an initial cycle of VIP (etoposide, ifosfamide and cisplatin). Three more cycles of chemotherapy with VIP were then given with stem-cell and growth-factor support, enabling considerable dose escalation to take place.38* In an ongoing phase III EORTC trial, four cycles of standard BEP chemotherapy are being compared with this regimen in patients with poor-prognosis disease. Management of CNS disease Malignant teratoma can involve the CNS at presentation or at relapse, then either in isolation (sanctuary site relapse) or in association with progressive systemic disease.39*,40* Metastases may be single or multiple. Spinalcord compression may also occur, usually secondary to contiguous spread of tumour from an adjacent vertebral body or para-aortic mass.41* Meningeal disease occurs but is rare. Central nervous system disease is most commonly seen in patients with a high presenting hCG (e.g. 10 000 IU/L) or with multiple (20) lung metastases, and a routine CT or MRI brain scan is indicated at initial evaluation in such cases. The management of CNS disease at presentation is controversial.40* It has been proposed that if a single metastasis is present and potentially resectable, and the systemic disease is clinically non-threatening, craniotomy and resection should be considered as an initial measure. In practice, such patients are extremely rare, and for the majority BEP chemotherapy should be commenced urgently. It is usually recommended that this treatment is supplemented by whole-brain irradiation,39* but there is little evidence to support the use of this additional treatment. The cure rates for patients presenting with CNS disease approach 50–60 per cent.20*** Similar results should be achieved in patients with sanctuary site relapse after chemotherapy, in which there is no systemic disease.39*,40* These cases commonly occur at a single site, which ideally should be resected then irradiated. Patients with end-stage disease involving CNS and peripheral sites should usually be treated palliatively.
not fall at half-life rates. The tumour markers fall into the normal range in the great majority of patients. However, those patients with very high initial serum markers, particularly those with choriocarcinoma and a very high hCG, commonly develop a plateau in the range of 10–50 IU/L. Whereas rising tumour markers clearly represent relapse of cancer, this is not necessarily the case with patients who have achieved a plateau. These patients can be observed – or, where necessary, operated on to resect residual masses – after which the tumour markers fall permanently to normal in around 50 per cent of cases, often despite the absence of histological evidence of cancer in resected masses. In the remaining cases, rapid relapse generally occurs, requiring combination chemotherapy. Radiologically, tumour masses usually respond promptly to chemotherapy, although, paradoxically, those of choriocarcinoma are often slow to respond. This may be a particular problem where gross pulmonary disease is present. It is important to be aware that growth of metastatic tumour (most commonly in the abdomen) may occur whilst tumour markers are falling. This is known as the ‘growing teratoma syndrome’ and is commonly due to expansion of masses of differentiated teratoma (Fig. 30.7).42 This should not be regarded as failure of chemotherapy, and surgery is indicated. Timing this relative to chemotherapy administration must be subject to the site and extent of tumour masses, the level of tumour markers and the response of disease at other sites. POST-CHEMOTHERAPY SURGERY
Once chemotherapy is completed, complete radiological and marker re-evaluation is indicated. Residual masses are common, particularly in patients with initially bulky disease and where elements of teratoma differentiated were present in the primary tumour (and therefore possibly in a proportion of metastases). It is not possible absolutely to
MONITORING CHEMOTHERAPY RESPONSE
Patients receiving chemotherapy for metastatic teratoma should be monitored by sequential tumour marker measurements and simple radiological examinations (e.g. chest X-ray). A common phenomenon following the first chemotherapy administration is for the tumour markers to rise for 7–10 days (tumour ‘flare’), presumably as a result of tumour-cell necrosis. Tumour markers should then fall consistently, although, not uncommonly, marker levels do
Figure 30.7 Growing teratoma syndrome. The retroperitoneal mass shown in these X-rays continue to increase in size despite return of the serum markers to normal. The patient underwent thoracolaporatomy with removal of multiple lung metastases and the retroperitoneal mass illustrated. Pathologically, all the resected tissue comprised teratoma differentiated.
718 Germ-cell cancer of the testis and related neoplasms
predict the composition of residual masses; these may contain necrotic tissue, teratoma differentiated or cancer (although the last-mentioned is uncommon in the absence of rising tumour markers). Many masses comprise a mixture of these histological elements. Except in special circumstances (marker plateau), surgery is not recommended in this last group. In all other cases, it is recommended that all abnormal masses visualized on abdominal CT scan be excised, certainly if these are more than 1.0 cm in diameter. However, clinical sense dictates that huge tumour masses at presentation that are resolving satisfactorily (e.g. 20 cm retroperitoneal mass reducing to 2 cm) may be observed for a short while to see if they regress further spontaneously. In the UK, template RLND is used where possible, with, in addition, the removal of all visibly abnormal disease sites. When bulky disease is present, this surgery can be complex, and occasionally patients will require resection of a kidney, the inferior vena cava or a length of aorta in order to achieve complete removal of tumour masses. Residual lung masses seen on chest X-ray should also be resected if this is technically feasible, and combined thorocolaparotomy is a possibility if both abdominal and lung masses require resection. When this combined approach is not reasonable, laparotomy is generally performed first, with observation of the pulmonary masses during the time of recovery. If the primary tumour is still present, orchidectomy is recommended post-chemotherapy.13* Careful histological evaluation of resected tissue by an experienced histopathologist is essential. Completely resected necrotic tissue or teratoma differentiated (which may appear quite dysplastic) requires no further therapy (Fig. 30.8). Patients in whom cancer is resected should be considered for further chemotherapy – generally with two more courses of BEP. This is, however, an individualized patient decision.43* POST-TREATMENT FOLLOW-UP: TERATOMA
Following chemotherapy and surgery (where necessary), a baseline CT scan of all previously abnormal areas should
Figure 30.8 Mass resected at post-chemotherapy laparotomy. Pathologically, all the resected tissue comprised teratoma differentiated.
be obtained. Further short-interval CT is not necessary if this study is normal. Patients should initially be monitored monthly with marker estimations. Chest X-rays are only worthwhile if metastatic disease is still visible in the chest after chemotherapy. Relapse from complete remission is relatively unusual (10–15 per cent of cases), occurring most commonly in patients with initial poor-prognosis disease or with residual unresectable disease in the chest or abdomen. In Southampton, the relapse rate from complete remission is 12 per cent. Half these relapses occur within 2 years of completing chemotherapy and half later. Relapses at up to 20–25 years post-chemotherapy have been recorded, particularly in patients with initial bulky disease, unresectable masses or resected teratoma differentiated. After an initial close surveillance policy, post-chemotherapy follow-up can be rapidly reduced to 6 monthly then yearly. Currently patients treated with chemotherapy alone who achieve complete remission should probably be followed for at least 5 years. Those in whom residual masses remain, or in whom teratoma differentiated has been resected, should be followed for life. There may be a role for follow-up CT scans in these latter patients, perhaps at 5-year intervals.
Management of malignant seminoma Pure seminoma comprises approximately 50 per cent of testicular cancers; mixed tumours (seminoma and teratoma) should be managed as malignant teratoma. STAGE I SEMINOMA
Approximately 80 per cent of patients with seminoma have stage I disease at presentation. This patient group therefore represents the most common management problem in this tumour type. Patients with spermatocytic seminoma (non-germ-cell tumour) have a very low risk of spread, and no further staging or therapy is indicated. It is known from surveillance studies in stage I seminoma that occult metastatic disease is present in 15–20 per cent of patients who, in the absence of therapy, develop relapse. Retrospective analyses of pooled series of patients managed by surveillance suggest that the risk factors for relapse are increasing size of the primary tumour (with those 4 cm in diameter at most increased risk) together with tumour involvement of the rete testis.44** Patients with both of these features in this retrospective study had a 31.5 per cent risk of relapse, a patient with either feature around a 15 per cent chance of relapse, and patients with neither feature a 12 per cent risk of relapse. A variety of management approaches have been used for stage I patients; cure rates for stage I disease should approach 100 per cent.
Management of malignant seminoma 719
RADIOTHERAPY
Adjuvant radiotherapy remains a valid, although decreasingly used, management approach, which has been much modified by recent trial experience. Historically, stage I seminoma was managed by treatment with dog-leg radiotherapy to a dose of 30 Gy. Seminoma is an exquisitely radiosensitive tumour, and reduced radiation doses (and fields) were obviously considered desirable to minimize early and late (second malignancy, infertility from contralateral testicular irradiation) treatment effects. A pivotal study conducted by the MRC45** randomized 478 patients (T1–T3, no previous ipsilateral inguinal surgery) to either dog-leg or para-aortic radiotherapy to a dose of 30 Gy. Three-year relapse-free survival (96.6 per cent versus 96 per cent) and survival were equivalent, with only one cancer-related death. Gastrointestinal toxicity and sperm-count toxicity were less in the para-aortic arm, which this trial group therefore recommended. A number of relapses in the para-aortic arm occurred in the ipsilateral pelvic nodes, prompting concerns about appropriate CT follow-up in this group. (CT scan is not routinely recommended after dog-leg irradiation for stage I disease.) Appropriate CT follow-up schedules remain a matter of individual clinical and patient choice when this radiotherapy technique is used. In a second study, the MRC46** led a trial comparing two different radiation doses (30 Gy versus 20 Gy) in patients treated with para-aortic radiotherapy (or dog-leg where previous ipsilateral inguinal surgery had been performed). Six hundred and twenty-five patients were randomized and 5-year relapse-free survival (97 per cent versus 96.4 per cent) and survival rates were equivalent. Again, only one death resulted from seminoma. This treatment has therefore become the standard radiotherapy adjuvant treatment approach to this cancer. ADJUVANT CHEMOTHERAPY
Seminoma is an exquisitely chemo-sensitive disease. A number of phase II trials have evaluated carboplatin given as one or two intravenous adjuvant treatment cycles at varying dose. In another pivotal study, the MRC47** compared a single intravenous injection of carboplatin (AUC-7) with para-aortic or dog-leg irradiation, as appropriate, at a dose of 20 Gy or 30 Gy. One thousand four hundred and seventyseven patients were randomized, 580 of whom were treated with carboplatin. Results were again equivalent (3-year relapse-free survival 94.8 per cent versus 95.9 per cent), and the toxicity of carboplatin was less than that of radiotherapy. Remarkably, contralateral germ-cell cancer occurred less commonly in the group receiving chemotherapy than in those receiving para-aortic irradiation (1.9 per cent versus 0.54 per cent). However, this was considered a preliminary observation requiring further follow-up. Relapse patterns differed between patients treated with radiotherapy and chemotherapy, with most chemotherapy
patients relapsing in the para-aortic nodes, justifying at least one routine follow-up CT scan of the abdomen and pelvis 1 year after treatment, whereas patients receiving paraaortic radiotherapy relapsed most commonly in the pelvis or supra-diaphragmatic nodes. Again, only one patient in the trial died from germ-cell cancer. SURVEILLANCE
A final management approach to stage I seminoma is surveillance. On superficial examination, this is the obvious approach, since treatment is reserved for those 15–20 per cent of patients harbouring metastatic disease. However, there are disadvantages to this approach. Seminoma is less commonly associated with marker elevation than teratoma, thereby placing an increased emphasis on radiological (CT scan) follow-up. Currently there are no standard CT follow-up schedules, but many groups continue to perform these scans for more than 5 years, as late relapse has been recorded beyond this time. In addition, relapse can occur outside standard irradiation fields and then involve treatment with chemotherapy. Surveillance is used in some centres, however, many of which select patients for this management approach on the basis of the prognostic criteria for relapse described above. Cure rates approaching 100 per cent should be achieved. Patients under surveillance should be seen at least at 3-monthly intervals for the first year. Follow-up CT scanning can be confined to the abdomen, with accompanying chest X-rays and serum markers, but no definitive schedule has yet been described. SUMMARY OF MANAGEMENT OPTIONS FOR STAGE I SEMINOMA
It cannot be over-emphasized that 80–85 per cent of patients with stage I seminoma do not harbour cancer and therefore require no treatment. Thus any adjuvant approach should be maximally effective for patients with sub-clinical cancer but should cause minimal harm to the remaining patients in this young population. On these criteria, surveillance or carboplatin seems the rational management approach. Individual patients and their clinicians need to decide which approach is most suitable, and this may vary by risk of relapse, patient’s temperament, geography and resources. What is absolutely clear is that all patients should be followed carefully for 5 years, and probably longer for patients undergoing surveillance. STAGES IIA AND IIB SEMINOMA
In perhaps 10 per cent of patients with seminoma, evidence of low-volume para-aortic adenopathy ( 5 cm) will be found at presentation. A majority of these patients will have normal serum markers. Two management options are available.48* Dog-leg radiotherapy is probably the management of choice for
720 Germ-cell cancer of the testis and related neoplasms
patients with stage IIA disease. Relapse-free survival of 95 per cent should be achievable, and virtually all patients should be cured, as salvage chemotherapy is so effective. The management of stage IIB disease (2–5 cm para-aortic nodes) will vary by bulk and site, and either dog-leg radiotherapy or combination chemotherapy (BEP 3) is a reasonable treatment option. Either should result in a relapse-free survival rate of approximately 90 per cent, with overall survival approaching 100 per cent. Again, individual clinicians and their patients need to decide which treatment option is appropriate for their given circumstances. Patients with IIA and IIB seminoma require posttreatment CT scanning. Residual masses are unusual, but are almost always sterile and can be watched. There is no indication for prophylactic treatment of the mediastinum with irradiation, a common practice in the past that was of doubtful benefit and resulted in cardiac toxicity. Relapse, when it occurs following radiotherapy, is almost always sited in the mediastinum or left cervical region. However, relapse can occur to other sites (lung, pleura or bone) but is relatively rare.
PROGNOSTIC FACTORS: METASTATIC SEMINOMA
Patients with metastatic seminoma treated with platinumcontaining chemotherapy were included in the IGCCC study,20*** although because these cases were comparatively rare, only 660 were available for analysis (Box 30.4).
Box 30.4 Prognosis factors in metastatic seminoma, International Germ Cell Consensus Classification Good prognosis 90% of patients 5-year progression-free survival 82%; survival 86% Any primary site and No non-pulmonary visceral metastases and Normal AFP, any hCG and LDH Poor prognosis 10% of patients 5-year progressive-free survival 67%; survival 72% Seminoma Any primary site and Non-pulmonary visceral metastases and normal AFP, any hCG any LDH AFP, alpha-fetoprotein, hCG, human chorionic gonadotrophin; LDH, lactate dehydrogenase.
In marked contra-distinction to malignant teratoma, seminoma arising in extra-gonadal sites (particularly the mediastinum) has no adverse prognostic implications and an identical (and excellent) prognosis, comparable to other types of metastatic testicular seminoma. The main adverse features identified in this study were the presence of nonpulmonary visceral metastases. Patients were divided into a good-prognosis or intermediate-prognosis group using this factor alone. No poor-prognosis group could be identified in this study. SEMINOMA STAGE IIC, D, RELAPSE POST-RADIOTHERAPY AND ALL OTHER METASTATIC PATIENTS
It is generally accepted that all the above patients with seminoma should be treated with combination chemotherapy. The great majority of patients with metastatic seminoma present with nodal disease in the retroperitoneum and/or posterior mediastinum and/or left cervical area. Occasional patients, however, have disease spread to the lungs, pleura and bones, although these latter sites are most commonly seen at relapse after chemotherapy. Metastatic seminoma is exquisitely sensitive to treatment with cisplatin and carboplatin used as single agents. Two trials have addressed whether single-agent carboplatin, which is extremely well tolerated, is sufficient therapy or whether combination chemotherapy is advantageous. The data from these trials have now been combined and published.49*** The MRC randomized 130 patients with advanced seminoma between either single-agent carboplatin (400 mg/m2 intravenously) or combination chemotherapy with etoposide and cisplatin.50** This trial was abandoned prematurely because of inferior treatment results seen in patients with metastatic teratoma who received carboplatin combination chemotherapy. During a similar period, a German national study randomized patients between single-agent carboplatin (400 mg/m2 intravenously) and combination chemotherapy with cisplatin, etoposide and ifosfamide. The combined study data49*** clearly indicated that both progression-free survival (72 per cent versus 92 per cent) and survival (89 per cent versus 94 per cent) were inferior in the patients treated with carboplatin. Therefore the treatment of choice for patients with metastatic seminoma should probably be three cycles of BEP chemotherapy (as described above for teratoma) or, in older patients where there are concerns about bleomycin toxicity, four cycles of cisplatin, etoposide bleomycin (as also described above). Although single-agent carboplatin is less effective, it may have a role in special circumstances, e.g. in patients with severe renal failure or profound learning difficulties, for whom cisplatin may be impracticable. Virtually all patients with metastatic seminoma will respond rapidly and satisfactorily to chemotherapy, although a substantial proportion of patients (again, particularly those with bulky disease) will have residual masses on completion of treatment – most commonly in the
Management of relapsed germ-cell cancer 721
retroperitoneum. Attempts have been made to resect such disease; however, the desmoplastic response evoked by metastatic seminoma, and the close relationship of the disease to vital retroperitoneal structures, has rendered such surgery fraught with difficulties. In fact, approximately 90 per cent of resected masses have proved sterile histologically – this finding being confirmed by the very low relapse rate when such masses are observed rather than resected. Recent reports suggest that PET scanning may be able to differentiate between residual active seminoma and necrosis.51* If this experience is confirmed, it may modify the current treatment approach to these patients, perhaps reserving observation for patients with a negative PET scan and subjecting patients with a positive scan (and probably biopsy) to either retroperitoneal surgery or irradiation, as appropriate. The alternative management approach is close observation, reserving therapy for increase in residual mass size, in which event a biopsy should be performed to exclude the presence of metastatic teratoma. Certainly, adjuvant irradiation used on a routine basis has been shown in a large retrospective study to be of no benefit. Overall, approximately 90 per cent of patients with bulky metastatic seminoma should remain failure free following treatment with intravenous chemotherapy as described above. Patients requiring salvage chemotherapy should be treated as being comparable to patients with teratoma – anecdotal evidence suggests that cure rates are higher. FOLLOW-UP OF SEMINOMA POST-TREATMENT
Patients with stage I seminoma have very low relapse rates, and follow-up intervals of 3 months for the first year and then 4 months for the second year should suffice, with less frequent follow-up thereafter. Computed tomography scanning is not required if adjuvant dog-leg irradiation is used. Pelvic CT scanning is optional in patients treated with para-aortic irradiation, particularly to detect occult pelvic relapse. There are no current follow-up scan recommendations for patients treated with adjuvant carboplatin. In the clinical trials described in the section Stage I Seminoma46**,47**, yearly whole-body scans were performed for 3 years after treatment. However, the author currently recommends a single abdominal/pelvic CT scan at 1 year. Patients receiving radiotherapy for stages IIA and B seminoma should have a single post-treatment CT scan to ensure that disease has entirely resolved. Patients with bulky abdominal seminoma treated with chemotherapy require follow-up scans of residual abdominal masses, where present, until these have stabilized or resolved. The masses commonly calcify and remain present for years. Late relapse of seminoma is rare, and follow-up for 5 years is probably sufficient. Patients relapsing after chemotherapy or radiotherapy for seminoma should have serum AFP and hCG levels measured and preferably a biopsy to confirm the relapsing masses comprise seminoma – histological ‘progression’ to
teratoma or relapse of an occult mixed tumour can occur, often accompanied by a rising AFP level.
Management of relapsed germ-cell cancer CONVENTIONAL-DOSE SALVAGE CHEMOTHERAPY AND SURGERY
Teratoma relapse is often detected because of pathological marker elevation – or because of progression of residual masses or the appearance of new masses. In the latter cases, and if the tumour markers remain normal, biopsy or surgical removal (where technically feasible) should be considered, as these may comprise teratoma differentiated – or, rarely, a new non-germ-cell malignant neoplasm derived from residual elements of teratoma differentiated. Occasionally relapses occur in the CNS, which can act as a sanctuary site from chemotherapy; patients at risk are those with a very high presentation hCG or multiple (20) lung metastases (see section Management of CNS disease). Patients with seminoma relapse in a number of different contexts. Patients originally diagnosed with stage I disease and managed with surveillance or single-agent carboplatin most commonly relapse in the retroperitoneum and/or pelvis.47** If disease is non-bulky and isolated, it may be reasonable to consider treatment with dog-leg irradiation. Otherwise, treatment with BEP chemotherapy (Box 30.2) is indicated. Patients relapsing after prophylactic radiotherapy should always be treated with BEP, with a very high chance of cure.46**,47** Patients who have received previous combination chemotherapy for advanced stage presentation should be managed in a fashion analogous to that for patients treated with BEP chemotherapy for metastatic teratoma, with one difference: if disease relapse is confined to one site, consideration should be given to the use of postchemotherapy irradiation, as this tumour is so sensitive to this modality. This decision should be taken on an individual patient basis. Most data about prognosis at the time of relapse relate to patients with metastatic teratoma. In this population, the prognosis relates to the initial response to chemotherapy (patients with primary refractory disease have a far worse outlook than those with initial complete response), time to relapse (late relapse is more favourable), bulk of disease and height of tumour marker elevation at relapse. Patients are generally divided into those with early relapse (2 years from achieving complete remission with BEP) or late relapse (the remainder). Data from Southampton suggest that 12 per cent of patients relapse overall, half in each of these two clinical contexts. There are two possible approaches to combination salvage chemotherapy. These are the use of a conventional-dose alternative regime or, as an alternative, brief re-induction with second-line chemotherapy followed by planned high-dose chemotherapy with peripheral stem-cell support (one or two cycles). It is generally recommended that patients with
722 Germ-cell cancer of the testis and related neoplasms
seminoma are managed with second-line conventional-dose chemotherapy ( radiotherapy as described above). Patients with primary refractory teratoma (i.e. no initial complete remission and rapid relapse) are probably best managed with salvage high-dose chemotherapy.52* In a recently published randomized trial, relapsing patients (all of whom achieved initial complete response or partial response marker negative with BEP) were randomized to receive either four cycles of VeIP (vinblastine, ifosfamide and cisplatin) or three cycles of VeIP followed by one cycle of high-dose chemotherapy with carboplatin, etoposide and cyclophosphamide supported by peripheral stem cells.53* Remarkably, survival in both arms of this study was identical, at just over 50 per cent. Most treatment centres in the UK currently recommend conventional-dose second-line chemotherapy, utilizing high-dose chemotherapy as a third treatment approach. Until recently, a most widely used regimen worldwide in this context was VeIP,54* although latterly TIP (paclitaxel, ifosfamide and cisplatin) has become more popular.55*,56* These regimens are particularly myelosuppressive, but despite this should be, wherever possible, given at full doses and on time, where necessary with growth factor support. Patients responding to conventional-dose salvage chemotherapy should receive four courses of treatment, after which complete re-evaluation should take place to see if salvage surgery is possible or necessary. Where tumour markers return to normal and residual teratomatous masses remain, these should always be resected. Approximately half of patients with relapse from BEP are described as having late relapse, i.e. relapse occurring more than 2 years and described on occasion over 20 years following complete remission.57,58 These patients are increasingly identified with non-bulky disease, when relapse is often present at a single site, often associated with elevated AFP and sometimes in unusual locations. Serum markers can occasionally rise over years. Current evidence suggests that these patients are relatively chemo-refractory, and that radical surgery, wherever possible, should be the initial treatment approach, which can result in cure. Where this is impossible, initial chemotherapy (probably with TIP)56* should be given, although available data suggest that surgery will almost always be necessary to achieve cure. HIGH-DOSE CHEMOTHERAPY WITH STEM-CELL SUPPORT
A number of studies have evaluated high-dose chemotherapy as second-line or third-line treatment for metastatic germ-cell cancer. Commonly used regimes are high-dose carboplatin and etoposide59 or carboplatin, etoposide and an alkylating agent (cyclophosphamide or ifosfamide).60 In general, where two drug regimes have been used, two cycles of high-dose chemotherapy have been used. The advent of peripheral stem-cell support and the availability of growth factors have made these treatments much safer. Prognostic factors have been identified that determine
outcome in these patients, for example patients with cisplatin-refractory disease or a mediastinal primary have lower cure rates.61* A single randomized trial53*** has evaluated conventional versus high-dose chemotherapy as second-line treatment for metastatic teratoma and found no differences in survival. Despite this, a number of centres prefer to treat patients initially with high-dose chemotherapy utilizing one or two ‘tandem’ cycles of high-dose therapy as above.59 Mortality for this treatment approach should now be less than 5 per cent, and cure rates of 40–50 per cent are described when it is used as second-line therapy. However, treatment options following relapse from high-dose therapy can be limited and survival often poor.
New drugs Since at least 85 per cent of patients with metastatic germcell cancer will be cured with initial chemotherapy, and a substantial proportion of the remainder will be cured with salvage regimens, there are few opportunities to test new drugs, which are thereby disadvantaged in this heavily pretreated population. However, it is clear that paclitaxel, gemcitabine, oxaliplatin, epirubicin and possibly irinotecan are active and that published combinations such as paclitaxel and gemcitabine,62 oxaliplatin and gemcitabine,63 and cisplatin and epirubicin64 can result in longstanding remissions in a small proportion of cases. It is suggested that readers consult the original text for details of these treatments.
Chemotherapy-related toxicity The treatment of germ-cell tumours with cisplatincontaining chemotherapy is associated with a wide range of toxicities, particularly for patients receiving salvage treatment for relapsed disease. Virtually all patients suffer problems with nausea and vomiting despite standard prophylactic treatment with a 5-HT3 antagonist and dexamethasone; alopecia occurs universally. Myelosuppression is also virtually universal; this predominantly comprises neutropenia, sometimes complicated by a fever, with the requirement for intravenous antibiotics (gentamicin should be avoided in these circumstances as nephrotoxicity may result). As described previously, there is no role for routine prophylactic G-CSF administration. Perhaps the most common life-threatening toxicity in relationship to BEP is bleomycin lung.65* Bleomycin causes pneumonitis and progressive pulmonary fibrosis with an overall incidence of 3–4 per cent in patients receiving BEP chemotherapy at full dose (360 000 units), with a mortality of 1–2 per cent. Bleomycin toxicity is dose related, but can occur after quite low doses have been given. Additional risk factors are increasing age (40 years) and deteriorating renal function. Bleomycin should not be administered to
Contralateral testicular cancer 723
patients with significant elevation of serum creatinine. Bleomycin lung is difficult to anticipate – the findings of bilateral basal crepitations, bilateral basal changes on chest X-ray or persistent cough or dyspnoea may all indicate early toxicity and may be an indication for discontinuation of the drug. Cisplatin commonly causes tinnitus and high-tone hearing loss, which may become apparent during or after treatment. These symptoms can settle with time, although one or both will persist in 20 per cent of patients.66* In addition, impaired renal function, with an overall loss of 10–20 per cent, commonly occurs when BEP is given at full dose.67* Once chemotherapy has been completed, recovery generally occurs rapidly and the great majority of patients resume their previous lifestyle. However, it is at this time that symptoms relating to peripheral neuropathy (caused by cisplatin) and sometimes transient Lhermitte’s syndrome may occur. Infertility is probably universal during chemotherapy, but sperm counts generally gradually recover to their previous levels, although this may take years if sperm counts are low at the initiation of chemotherapy or if more than four courses of cisplatin are given.68* There have been no reports of an increase in congenital abnormalities in the children of survivors of testicular cancer. On long-term follow-up, there is a clear-cut increased risk of both systolic and diastolic hypertension, presumably related to kidney damage from cisplatin. In addition, an increased proportion of patients who have received BEP chemotherapy have an increased body mass index and probably an increased risk of hyperlipidaemia and coronary artery disease,69*,70*,71* although these risks have not yet been adequately quantified Occasional cases of aseptic necrosis of the hips are well described. A great majority of patients return to their life and work, and the side effects described should be set in the context of the previous appalling prognosis for this disease.
Management of the contralateral testis The lifetime risk of developing testicular cancer is 1:400 to 1:450. Patients diagnosed with a germ-cell cancer are, however, at much increased risk of contralateral disease. Rarely, patients present with synchronous bilateral tumours; much more commonly, the second tumour is delayed by 5–10 years. The underlying pathological event is the presence of CIS, recognizable only by testicular biopsy. In a large German series,72* routine testicular biopsy was performed at the time of orchidectomy for cancer, and CIS was detected in the contralateral testis in 4.9 per cent of cases, which equates approximately to the recognized risk of contralateral disease. The main risk factor for contralateral CIS is the presence of an atrophic contralateral testis – probably best defined as a testicular volume of less than 16 mL
(measured by ultrasound). The second main risk factor is decreasing age, particularly below 30 years. A history of testicular maldescent without accompanying atrophy is probably not a risk factor, as was previously thought. In an MRC study,73* a particularly high-risk group was identified: patients aged less than 30 years with a small remaining testis, as defined above, were found to have a risk of contralateral CIS of more than 30 per cent; for the remaining patients, the risk was between 2 per cent and 4 per cent. There is strong clinical evidence that CIS carries a very high risk of subsequent malignancy (50 per cent risk by 5 years)6* and treatment is generally recommended once this abnormality is found. Carcinoma-in-situ is generally associated with poor sperm production, and azoospermia is not unusual when CIS is gross. CIS can be effectively treated with local radiation. The currently recommended dose is 20 Gy in ten daily fractions.8* Lower doses have been used in an attempt to prevent hypogonadism, but have been associated with an increased risk of relapse.8* Fertile patients may wish to bank sperm prior to such treatment, which will universally result in sterility, or may prefer to attempt to father children prior to treatment. In such cases, where CIS is known to be present, a close watch on the remaining testis should be maintained, with frequent routine follow-up ultrasound and examination. Radiotherapy commonly causes delayed Leydig-cell damage, with resulting hypogonadism, in which case hormone replacement will be required.
Contralateral testicular cancer Synchronous or metachronous testicular cancers occur in the remaining testis in 2–5 per cent of cases. Most commonly, both tumours comprise seminoma. When such patients are seen, urgent sperm banking should be offered, where appropriate, and followed by clinical staging. When metastatic disease is found in patients with bilateral tumours at presentation, the preferred management option may be orchidectomy of the most severely affected testis followed by combination chemotherapy and then re-evaluation of both the metastatic disease sites and the remaining testis. If preservation of this testis is considered desirable, this may be reasonable as long as the cancer has resolved, and the testis is of reasonable size and has adequate hormonal function. Case reports have described recovery of fertility in such cases, but late relapse as a result of persistence or recurrence of CIS or cancer remains a worrying possibility. If preservation of fertility is not important, testicular irradiation to a dose of 20 Gy may be considered post-chemotherapy. This dose of irradiation would effectively sterilize CIS or residual seminoma. However, this is not an evidence-based approach. If a contralateral tumour is found in the absence of metastatic disease, partial orchidectomy may be considered reasonable in specific circumstances. A current
724 Germ-cell cancer of the testis and related neoplasms
clinical series74* suggests that these include the presence of a tumour less than 2 cm in diameter with normal preoperative LH and testosterone levels. Multiple biopsies of the tumour bed to ensure complete resection together with adjuvant local irradiation should be considered unless ongoing fertility is a requirement, in which case close observation should be pursued. For the majority of patients, bilateral orchidectomy will be necessary or requested. In these cases, hormonal replacement is essential. Hypogonadism commonly causes flushes, sweats, swinging moods and impotence, with a long-term complication of osteoporosis. Hormone replacement can be achieved by a number of means. These include intramuscular testosterone (generally given at 10–12-week intervals) and subcutaneous testosterone implants (inserted via a cannula at 4-monthly intervals). However, many patients are now satisfactorily treated with a transdermal preparation (Testogel) at a dose of 50–100 mg daily applied to the skin. Patients on hormone replacement should have follow-up LH and testosterone levels measured to ensure that replacement is adequate.
Extra-gonadal germ-cell tumours MEDIASTINAL GERM-CELL TUMOUR
Approximately 2–3 per cent of germ-cell tumours apparently arise in the mediastinum – testicular biopsies in such patients are normal (and unnecessary if testicular ultrasound is normal). Patients with mediastinal germ-cell tumours commonly have gross advanced disease at presentation (Fig. 30.9). The common presenting features include dyspnoea, chest pain (often pleuritic), superior vena caval obstruction and pericardial tamponade. Mediastinal seminoma and teratoma occur; the latter is most common. Mediastinal seminoma carries no increased risk of death when compared with comparable patients with metastatic testicular seminoma20*** and should be routinely treated with three cycles of BEP alone. There is no evidence to support the additional use of postchemotherapy surgery or irradiation.75* Mediastinal teratoma is an aggressive and relatively poor-prognosis form of germ-cell cancer.74* A variety of histologies may be seen, including pure yolk-sac tumour or choriocarcinoma. Mediastinal teratoma is particularly associated with the development of a second malignancy (e.g. sarcoma) within the tumour mass. These cases are also rarely associated with Klinefelter’s syndrome. In addition, the co-existence of mediastinal teratoma and haematological malignancy (myelodysplasia, acute myeloid leukaemia or malignant histiocytosis) is well recorded,77* both diseases arising from a common malignant clone characterized by i(12p). Haematological malignancy in these cases generally presents either simultaneously with the mediastinal tumour or within 18 months to 2 years. Prognosis is extremely poor.
Figure 30.9 Chest X-ray showing very large anterior mediastinal mass. The patient's serum alpha-fetoprotein (AFP) was grossly elevated and biopsy was considered unnecessary. Bleomycin, etoposide and cisplatin (BEP) chemotherapy was given.
As mediastinal teratoma usually presents with gross advanced disease, urgent therapy is necessary, and the diagnosis can often be based on serum marker elevations alone in a patient with a large anterior mediastinal mass (see Fig. 30.9). Mediastinal teratoma has the worst prognosis of any form of germ-cell cancer, with a cure rate of approximately 40–50 per cent in most studies.76* Bleomycin, etoposide and cisplatin chemotherapy remains the standard treatment, although some authorities believe that more intensive chemotherapy regimens may produce better results. Following four courses of chemotherapy, patients should be re-evaluated, when virtually all will be found to have residual masses in the anterior mediastinum. These should always, where technically feasible (and despite marker elevation), be resected.78* These masses may comprise necrosis, differentiated teratoma or active cancer (areas of sarcoma may also be seen). Post-chemotherapy follow-up should be close. Unfortunately, those patients who develop relapse of their disease manifest by increasing masses or elevated markers have an extremely poor prognosis, as they are particular refractory to salvage therapy. High-dose therapy with stem-cell support is largely ineffective, and experimental treatment protocols are appropriate.79*
References 725
RETROPERITONEAL PRIMARY GERM-CELL TUMOURS
A small proportion of patients with germ-cell cancer present as a result of retroperitoneal lymph-node enlargement. This is often massive and sometimes associated with obstruction of one or both ureters or the inferior vena cava. Biopsy or serum markers may reveal that the tumour is either a seminoma or malignant teratoma. Testicular ultrasound should be performed in all such cases. It is sometimes possible to determine from the presenting CT scan whether the metastatic disease has derived from the left or right testis – ultrasound will commonly show abnormalities on this side. Even when testicular ultrasound is normal, bilateral testicular biopsies have shown an incidence of CIS of approximately 40 per cent.80* It seems likely that the testis is the primary site in such cases, and orchidectomy should be performed once treatment is completed. Retroperitoneal germ-cell cancers are managed in the same way as their testicular counterparts. The prognosis relates to recognized prognostic factors, particularly the degree of tumour marker elevation. Many patients with teratoma will require post-chemotherapy surgery.
CONCLUSIONS The germ-cell tumours are a complex and fascinating group of malignancies with rapidly increasing incidence. Their modern interdisciplinary management represents a triumph of clinical science. Evidence-based management protocols are now available for stage I and good-prognosis metastatic disease, for which clinical trials have largely discontinued. Patients with intermediate-prognosis or poor-prognosis disease should, wherever possible, continue to be entered into randomized trials. Salvage chemotherapy can cure many of these patients, and the outlook continues to improve with time. The epidemiology of these diseases remains a challenge. Genetic susceptibility is increasingly defined, and screening for high-risk families may be a possibility in the future. Increasingly, patients with germ-cell cancers are managed in specialist centres. There is good evidence that this type of specialist approach improves management and outlook.
KEY LEARNING POINTS ●
Other testicular tumours SEX-CORD/STROMAL TUMOURS
Leydig-cell and Sertoli-cell tumours Patients with Leydig-cell and Sertoli-cell tumours together comprise perhaps 2–3 per cent of patients with testicular malignancy.81* These tumours can occur at any age, and at least 90 per cent of cases are benign. Leydig-cell tumours are much more common. It is not possible pathologically to differentiate benign from malignant cases; the latter is more likely in the presence of increased mitotic activity, capsular invasion and cellular pleomorphism. Ultimately, malignancy will only be confirmed if metastatic disease develops. Most cases of sex-cord stromal tumours present with testicular swelling. These tumours may occasionally be hormone secreting, with associated elevation of oestrogen levels and gynaecomastia. Management is with orchidectomy. Post-orchidectomy CT scanning should be performed. The great majority of patients are stage I and require no further therapy. No evidence-based surveillance protocol is available. However, the author follows patients with two further abdominal CT scans at approximately 6 months and 2 years, with intervening chest X-rays and physical examination. Metastatic disease rarely occurs. When this is seen, chemotherapy is ineffective and management relies on radical surgery. The prognosis is very poor in such cases, but occasional cures have been recorded.
●
●
●
Stage I seminoma can be managed with adjuvant carboplatin, surveillance or, decreasingly, adjuvant radiotherapy. Stage I teratoma may be managed with surveillance, retroperitoneal surgery, or, in high risk cases, with adjuvant chemotherapy. Standard follow up protocols are increasingly defined for these patients. Cure rates should approach 100%. Bulky metastatic seminoma is managed with combination chemotherapy using at least cisplatin and etoposide. Cure rates of at least 90% should be achieved. The standard chemotherapy for metastatic non-seminoma is BEP (bleomycin, etoposide and cisplatin). A variety of different schedules are used. For good prognosis disease cure rates should exceed 95%. There is a 2%–5% risk of developing contralateral testicular cancer. High risk patients should be screened for CIS (carcinoma in-situ) with testicular biopsy.
REFERENCES ◆1
Huyghe E, Matsuda T, Thonneau P. Increasing incidence of testicular cancer worldwide: a review. J Urol 2003; 170:5–11. 2 United Kingdom Testicular Cancer Study Group. Aetiology of testicular cancer: association with congenital abnormalities, age at puberty, infertility, and exercise. BMJ 1994; 308:1393–9.
726 Germ-cell cancer of the testis and related neoplasms
3 Forman D, Oliver RTD, Brett AR, et al. Familial testicular cancer: a report of the UK family register, estimation of risk and an HLA class I sib-pair analysis. Br J Cancer 1992; 65:255–62. ●4 Rapley EA, Crockford G, Teare D, et al. Localisation to chromosome Xq27 of a susceptibility gene to testicular germ cell tumours. Nat Gen 2000; 24:197–200. ◆5 Hoei-Hansen CE, Rajpert-De Meyts E, Daugarrd G, et al. Carcinoma in situ testis, the progenitor of testicular germ cell tumours: a clinical review. Ann Oncol 2005; 16:863–8. ●6 Von der Maase H, Rorth M, Walbom-Jorgensen S, et al. Carcinoma in situ of contralateral testis in patients with testicular germ cell cancer: study of 27 cases in 500 patients. BMJ 1986; 293:1398–401. 7 Christensen TB, Daugaard G, Geertsen PF, et al. Effect of chemotherapy on carcinoma in situ of the testis. Ann Oncol 1998; 9:657–60. ●8 Petersen PM, Giwercman A, Daugaard G, et al. Effect of graded testicular doses of radiotherapy in patients treated for carcinoma-in-situ in the testis. J Clin Oncol 2002; 20:1537–43. 9 Simmonds PD, Lee AHS, Theaker JM, et al. Primary pure teratoma of the testis. J Urol 1996; 155:939–42. 10 Motzer RJ, Amsterdam A, Prieto V, et al. Teratoma with malignant transformation: diverse malignant histologies arising in men with germ cell tumors. J Urol 1998; 159:133–8. 11 McKendrick JJ, Theaker J, Mead GM. Nonseminomatous germ cell tumour with very high serum human chorionic gonadotropin. Cancer 1991; 67:684–9. 12 De Santis M, Bokemeyer C, Becherer A, et al. Predictive impact of 2-18 fluoro-2-deoxy-D-glucose positron emission tomography for residual postchemotherapy masses in patients with bulky seminoma. J Clin Oncol 2001; 19:3740–4. 13 Geldart TR, Simmonds PD, Mead GM. Orchidectomy after chemotherapy for patients with metastatic testicular germ cell cancer. BJU International 2002; 90:451–5. ●14 Read G, Stenning S, Cullen MH, et al. Medical Research Council prospective study of surveillance for stage I testicular teratoma. J Clin Oncol 1992; 10:1762–8. 15 Rustin GJ, Mead GM, Stenning SP, et al. Randomised trial of two or five computed tomography scans in the surveillance of patients with stage 1 nonseminomatous germ cell tumors of the testis: Medical Research Council Trial TE08, ISRCTN56475197 – The National Cancer Research Institute Cancer Clinical Studies Group. J Clin Oncol 2007; 25:1310–15. ●16 Cullen MH, Stenning SP, Parkinson MC, et al. Short-course adjuvant chemotherapy in high risk stage I nonseminomatous germ cell tumors of the testis: a Medical Research Council report. J Clin Oncol 1996; 14:1106–13. 17 Donohue JP, Thornhill, JA, Foster RS, et al. Primary retroperitoneal lymph node dissection in clinical stage A non-seminomatous germ cell testis cancer. Review of the Indiana University experience 1965–1989. Br J Urol 1993; 71:326–35.
18 Stephenson AJ, Bosl GJ, Motser RJ, et al. Retroperitoneal lymph node dissection for nonseminomatous germ cell testicular cancer: impact of patient selection factors on outcome. J Clin Oncol 2005; 23:2781–8. 19 Donohue JP, Thornhill JA, Foster RS, et al. The role of retroperitoneal lymphadenectomy in clinical stage B testis cancer: The Indiana University experience (1965 to 1989). J Urol 1995; 153:85–9. ●20 International Germ Cell Cancer Collaborative Group. International Germ Cell Consensus Classification: a prognostic factor based staging system for metastatic germ cell cancers. J Clin Oncol 1997; 15:594–603. 21 Mead GM. Who should manage germ cell tumours of the testis? BJU International 1999; 84:61–7. ●22 Saxman SB, Finch D, Gonin R, et al. Long-term follow-up of a phase III study of three versus four cycles of bleomycin, etoposide and cisplatin in favorable prognosis germ cell tumors: the Indiana University experience. J Clin Oncol 1998; 16:702–6. ●23 de Wit R, Roberts JT, Wilkinson PM, et al. Equivalence of three or four cycles of bleomycin, etoposide and cisplatin chemotherapy and of a 3- or 5-day schedule in good prognosis germ cell cancer: a randomized study of the European Organization for Research and Treatment of Cancer Genitourinary Tract Cancer Cooperative Group and the Medical Research Council. J Clin Oncol 2001; 19:1629–40. ●24 Hinton S, Catalano PJ, Einhorn LH, et al. Cisplatin, etoposide and either bleomycin or ifosfamide in the treatment of disseminated germ cell tumors. Cancer 2003; 97:1869–75. ●25 Fossa SD, Kaye SB, Mead GM, et al. Filgrastim during combination chemotherapy of patients with poor prognosis metastatic germ cell malignancy. J Clin Oncol 1998; 16:716–24. 26 Xiao H, Mazumdar M, Bajorin DF, et al. Long-term follow-up of patients with good risk germ cell tumors treated with etoposide and cisplatin. J Clin Oncol 1997; 15:2553–8. 27 Levi JA, Raghaven D, Harvey V, et al. The importance of bleomycin in combination chemotherapy for good prognosis germ cell carcinoma. J Clin Oncol 1993; 11:1300–5. ●28 Loehrer PJ, Johnson D, Elson P, et al. Importance of bleomycin in favorable prognosis disseminated germ cell tumors: an Eastern Cooperative Oncology Group trial. J Clin Oncol 1995; 13:470–6. 29 de Wit R, Stoter G, Kaye SB, et al. Importance of bleomycin in combination chemotherapy for good prognosis testicular nonseminoma: a randomized study of the European Organization for Research and Treatment of Cancer Genitourinary Tract Cancer Cooperative Group. J Clin Oncol 1997; 15:1837–43. 30 Culine S, Kerbrat P, Bousy J, et al. The optimal chemotherapy regimen for good risk metastatic non-seminomatous germ cell tumors (MNSGCT) is 3 cycles of bleomycin, etoposide and cisplatin: mature results of a randomized trial. Proc ASCO 2003; Abstr 1536. 31 Einhorn LH, Williams SD, Loehrer PJ, et al. Evaluation of optimal duration of chemotherapy in favorable prognosis
References 727
32
●33
●34
●35
36
37
38
39
40
41
42 43
disseminated germ cell tumors: a Southeastern Cancer Study Group protocol. J Clin Oncol 1989; 7:387–91. Fossa SD, de Wit R, Roberts T, et al. Quality of life in good prognosis patients with metastatic germ cell cancer: a prospective study of the European Organization for Research and Treatment of Cancer Genitourinary Group/Medical Research Council Testicular Cancer Study Group (30941/TE20). J Clin Oncol 2003; 21:1107–18. Horwich A, Sleijfer DT, Fossa SD, et al. Randomized trial of bleomycin, etoposide and cisplatin compared with bleomcyin, etoposide and carboplatin in good prognosis metastatic nonseminomatous germ cell cancer: a multiinstitutional Medical Research Council/European Organization for Research and Treatment of Cancer trial. J Clin Oncol 1997; 15:1844–52. Bajorin DF, Sarosdy MF, Pfister DG, et al. Randomized trial of etoposide and cisplatin versus etoposide and carboplatin in patients with good risk germ cell tumors: a multiinstitutional study. J Clin Oncol 1993; 11:598–606. Kaye SB, Mead GM, Fossa S, et al. Intensive induction sequential chemotherapy with BOP/VIP-B compared with treatment with BEP/EP for poor prognosis metastatic nonseminomatous germ cell tumor: a randomized Medical Research Council/European Organization for Research and Treatment of Cancer study. J Clin Oncol 1998; 16:692–701. Bower M, Newlands ES, Holden L, et al. Treatment of men with metastatic non-seminomatous germ cell tumours with cyclical POMB/ACE chemotherapy. Ann Oncol 1997; 8:477–83. Christina JA, Huddart RA, Norman A, et al. Intensive induction chemotherapy with CBOP/BEP in patients with poor prognosis germ cell tumors. J Clin Oncol 2003; 21:871–7. Bokemeyer C, Schmoll HJ, Harstrick A, et al. A phase I/II study of a stepwise dose escalated regimen of cisplatin, etoposide and ifosfamide plus granulocyte macrophage colony stimulating factor (GM-CSF) in patients with advanced germ cell tumours. Eur J Cancer 1993; 29A:2225–31. Bokemeyer C, Nowak P, Haupt A, et al. Treatment of brain metastases in patients with testicular cancer. J Clin Oncol 1997; 15:1449–54. Fossa SD, Bokemeyer C, Gerl A, et al. Treatment outcome of patients with brain metastases from malignant germ cell tumors. Cancer 1998; 85:988–97. Gale J, Mead GM, Simmonds PD, et al. Management of spinal cord and cauda equina compression secondary to epidural metastatic disease in adults with malignant germ cell tumours. Clin Oncol 2002; 14:481–90. Jeffery GM, Theaker JM, Lee AHS, et al. The growing teratoma syndrome. Br J Urol 1991; 67:195–202. Fizazi K, Tjulandin S, Salvioni R, et al. Viable malignant cells after primary chemotherapy for disseminated nonseminomatous germ cell tumors: prognostic factors and role of postsurgery chemotherapy – results from an international study group. J Clin Oncol 2001; 19:2647–57.
◆44
●45
●46
●47
●48
●49
●50
◆51
52
●53
54
●55
56
57
Warde P, Specht L, Horwich A, et al. Prognostic factors for relapse in stage I seminoma managed by surveillance. A pooled analysis. J Clin Oncol 2002; 20:4448–52. Fossa SD, Horwich A, Russell JM, et al. Optimal planning target volume for stage I testicular seminoma: a Medical Research Council randomized trial. J Clin Oncol 1999; 17:1146–54. Jones WG, Fossa SD, Mead GM, et al. Randomized trial of 30 versus 20 Gy in the adjuvant treatment of stage I testicular seminoma: a report on Medical Research Council trial TE18, European Organisation for the Research and Treatment of Cancer trial 30942 (5328). J Clin Oncol 2005; 23:1200–8. Oliver RTD, Mason MD, Mead GM, et al. Radiotherapy versus single-dose carboplatin in adjuvant treatment of stage I seminoma: a randomised trial. Lancet 2005; 366:293–300. Classen J, Schmidberger H, Meisner C, et al. Radiotherapy for stages IIA/B testicular seminoma: final report of a prospective multicenter clinical trial. J Clin Oncol 2003; 21:1101–6. Bokemeyer C, Kollmannsberger C, Stenning C, et al. Metastatic seminoma treated with either single agent carboplatin or cisplatin based combination chemotherapy: a pooled analysis of two randomised trials. Br J Cancer 2004; 91:683–7. Horwich A, Oliver RTD, Wilkinson PM, et al. A Medical Research Council randomized trial of single agent carboplatin versus etoposide and cisplatin for advanced metastatic seminoma. Br J Cancer 2000; 83:1623–9. De Santis M, Bechere A, Bokemeyer C, et al. 2-18 fluorodeoxy-D-glucose positron emission tomography is a reliable predictor for viable tumor in postchemotherapy seminoma: an update of the prospective multicentric SEMPET trial. J Clin Oncol 2004; 22:1034–9. Vaena DA, Abonour R, Einhorn LH. Longterm survival after high dose salvage chemotherapy for germ cell malignancies with adverse prognostic variables. J Clin Oncol 2003; 21:4100–4. Pico JL, Rosti G, Kramar A, et al. A randomised trial of high dose chemotherapy in the salvage treatment of patients failing first line platinum chemotherapy for advanced germ cell tumours. Ann Oncol 2005; 16:1152–9. Loehrer PJ, Gonin R, Nichols CR, et al. Vinblastine plus ifosfamide plus cisplatin as initial salvage therapy in recurrent germ cell tumor. J Clin Oncol 1998; 16:2500–4. Kondagunta GV, Bacik J, Donadio A, et al. Combination of paclitaxel, ifosfamide and cisplatin is an effective second line therapy for patients with relapsed testicular germ cell tumors. J Clin Oncol 2005; 23:6549–55. Mead GM, Cullen MH, Huddart R, et al. A phase II trial of TIP (paclitaxel, ifosfamide and cisplatin) given as second line (post-BEP) salvage chemotherapy for patients with metastatic germ cell cancer: a Medical Research Council trial. Br J Cancer 2005; 93:178–84. Gerl A, Clemm C, Schmeller N, et al. Late relapse of germ cell tumors after cisplatin based chemotherapy. Ann Oncol 1997; 8:41–7.
728 Germ-cell cancer of the testis and related neoplasms
58 George DW, Foster RS, Hromas RA, et al. Update on late relapse of germ cell tumor: a clinical and molecular analysis. J Clin Oncol 2003; 21:113–22. 59 Bhatia S, Abonour R, Porcu P, et al. High dose chemotherapy as initial salvage chemotherapy in patients with relapsed testicular cancer. J Clin Oncol 2000; 18:3346–51. 60 Motzer RJ, Mazumdar M, Bosl J, et al. High dose carboplatin, etoposide and cyclophosphamide for patients with refractory germ cell tumors: treatment results and prognostic factors for survival and toxicity. J Clin Oncol 1996; 14:1098–105. ●61 Beyer J, Kramar A, Mandanas R, et al. High dose chemotherapy as salvage treatment in germ cell tumours: a multivariate analysis of prognostic variables. J Clin Oncol 1996; 14:2638–45. 62 Hinton S, Catalona P, Einhorn LH, et al. Phase II study of paclitaxel plus gemcitabine in refractory germ cell tumors (E9897): a trial of the Eastern Cooperative Oncology Group. J Clin Oncol 2002; 20:1859–63. 63 Killmannsberger C, Beyer J, Leiersch R, et al. Combination chemotherapy with gemcitabine plus oxaliplatin in patients with intensively pretreated or refractory germ cell cancer: a study of the German Testicular Cancer Study Group. J Clin Oncol 2004; 22:108–14. 64 Bedano PM, Brames MJ, Williams SW, et al. A phase II study of cisplatin plus epirubicin salvage chemotherapy in refractory germ cell tumors. Proc ASCO 2003; Abstr. 4526. 65 O’Sullivan JM, Huddart RA, Norman AR, et al. Predicting the risk of bleomycin lung toxicity in patients with germ cell tumours. Ann Oncol 2003; 14:91–6. 66 Bokemeyer C, Berger CC , Hartmann JT, et al. Analysis of risk factors for cisplatin induced ototoxicity in patients with testicular cancer. Br J Cancer 1998; 77:1355–62. 67 Fossa SD, Aass N, Winderen M, et al. Long term renal function after treatment for malignant germ cell tumours. Ann Oncol 2002; 13:322–8. 68 Huddart RA, Norman A, Moyunihan C, et al. Fertiity, gonadal and sexual function in survivors of testicular cancer. Br J Cancer 2005; 93:200–7. 69 Sagstuen H, Aass N, Fossa SD, et al. Blood pressure and body mass index in longterm survivors of testicular cancer. J Clin Oncol 2005; 23:4980–90.
70 Meinardi MT, Gietema JA, van der Graff WTA, et al. Cardiovascular morbidity in long term survivors of metastatic testicular cancer. J Clin Oncol 2000; 18:1725–32. 71 Huddard RA, Norman A, Shahidi M, et al. Cardiovascular disease as a long term complication of treatment for testicular cancer. J Clin Oncol 2003; 21:1513–23. 72 Dieckmann KP, Loy V. Prevalence of contralateral testicular intraepithelial neoplasia in patients with testicular germ cell neoplasms. J Clin Oncol 1996; 14:3126–32. 73 Harland SJ, Cook PA, Fossa SD, et al. Intratubular germ cell neoplasia of the contralateral testis in testicular cancer: defining a high risk group. J Urol 1998; 160:1353–7. 74 Heidereich A, Albers P, Weibbach L, et al. Organ preserving surgery in testicular cancer – long term results of the German Testicular Cancer Study Group. Proc ASCO 2005; Abstr. 4523. 75 Bokemeyer C, Droz JP, Horwich A, et al. Extragonadal seminoma. An international multicenter analysis of prognostic factors and long term treatment outcome. Cancer 2001; 91:1394–401. 76 Ganjoo KN, Rieger KM, Kesler KA, et al. Results of modern therapy for patients with mediastinal nonseminomatous germ cell tumors. Cancer 2000; 88:1051–6. 77 Hartmann JT, Nichols CR, Droz JP, et al. Hematologic disorders associated with primary mediastinal nonseminomatous germ cell tumors. J Natl Canc Inst 2000; 92:54–61. 78 Vuky J, Bains M, Bacik J, et al. Role of postchemotherapy adjunctive surgery in the management of patients with nonseminoma arising from the mediastinum. J Clin Oncol 2001; 19:682–8. 79 Hartmann JT, Einhorn L, Nichols CR, et al. Second line chemotherapy in patients with relapsed extragonadal nonseminomatous germ cell tumors: results of an international multicenter analysis. J Clin Oncol 2001; 19:1641–8. 80 Daugaard G, Rorth M, von der Maase H, et al. Management of extragonadal germ cell tumors and the significance of bilateral testicular biopsies. Ann Oncol 1992; 3:283–9. ◆81 Conkey DS, Howard GCW, Grigor KM, et al. Testicular sex cord stromal tumours: the Edinburgh experience 1988–2002, and a review of the literature. Clin Oncol 2005; 17:322–7.
Figure 6.8c CT-PET fusion image accurately delineates the primary tumour lying in the left Waldeyer’s fossa (arrow).
Figure 6.11c The PET-CT fusion image provides anatomical correlation and demonstrates, that the lymph nodes have no increased tracer uptake, thus excluding mediastinal metastases.
Figure 14.4 Single photon emission computerized tomographic (SPECT) scan showing both thallium uptake and Tchexamethylpropyleneamine oxime (HMPAO) images in a glioblastoma.
Figure 6.10c PET-CT fusion provided anatomical and functional correlation of the active tumour sites.
Figure 17.11 Three-dimensional plan (anterior and pair of antero-oblique fields) for phase II volume covering microscopic disease in the thyroid bed.
Figure 19.6a The apex of the heart (green) exposed to radiotherapy in a standard open tangential field.
(a)
(a)
Figure 19.6b The apex of the heart (green) protected throughout every exposure using one leaf of a multi-leaf collimator that moves into the beam in the longitudinal axis.
(b)
Figure 19.7 Poor two-dimensional dosimetry in a standard breast plan using an open tangential field: (a) transverse view; (b) sagittal view.
(b)
Figure 19.8 Improved dosimetry using IMRT avoids hotstops and ensures better coverage of chest wall: (a) transverse view; (b) sagittal view.
Figure 21.2 Four-field conformal plan for radical oesophageal radiotherapy – coaxial view.
Figure 21.3 Four-field conformal plan for radical oesophageal radiotherapy – sagittal view.
Figure 27.3 Incidence of lymph nodes by patient showing nodal groups affected.
Figure 36.5 Ultrasonography with colour Doppler showing persistent gestational trophoblastic disease following a CM within the body and wall of the uterus. A typical vesicular or ‘snow storm’ appearance of residual molar tissue can be seen within the uterus, together with a rich blood supply through the endometrium and myometrium. There is no evidence of a fetus.
Figure 40.3 PET-CT scan of patient shown in Figure 40.1, obtained after completion of three cycles of interdigitated neoadjuvant MAID (mesna, doxorubicin, ifosfamide and dacarbazine) chemotherapy and 44 Gy of radiation therapy.
(a)
Figure 40.4a-c
(b)
(c)
Photomicrographs showing (a) low, (b) intermediate, and (c) high grade soft tissue myxoid liposarcoma.
Figure 51.2 This is a beam’s-eye view (BEV) of a single treatment portal used in prostate radiotherapy. The target volume of the prostate gland (green) is located between the organs at risk, which are the bladder (blue) anteriorly and the rectum (red) posteriorly. Using conformal radiotherapy (CFRT) techniques, the treatment field (blue line) is shaped around the profile of the prostate gland in the BEV compared with the conventional treatment field (black dotted line), which is an unshaped rectangular field. It is clear that the unshaped rectangular field will irradiate substantially more of the bladder and rectum compared with the conformal field.
Father, has two alleles of each gene
Figure 51.3 A model view of a 3-field coplanar conformal radiotherapy (CFRT) plan within the abdomen. This 3D reconstruction of the patient shows the orientation of the beams and their coverage of the clinical target volume (solid blue colour) located in the middle of the picture and covered by the planning target volume (red wire colour). The surrounding normal structures have been outlined: the bowel (translucent brown colour), spinal cord (solid light blue colour), right kidney (solid yellow colour) and liver (translucent green colour). The conformal field shapes are shown as they should appear projected against the bony anatomy of the patient and are used to verify the positioning of the beams daily during patient set-up with electronic portal imaging.
Mother, has two alleles of each gene
Gametes, contain single allele
Offspring, carry two alleles
Figure 62.2
Figure 55.1 inheritance.
Principles of genetic
Arterial reconstruction for renal vessel anatomy prior to surgery in kidneys with double renal arteries bilaterally.
31 Renal cell cancer TOM R. GELDART
Introduction and epidemiology Histopathology Inherited syndromes Risk factors Clinical features Clinical evaluation and staging Localized disease
729 730 731 732 733 733 734
INTRODUCTION AND EPIDEMIOLOGY Malignant renal tumours account for about 3 per cent of all cancers worldwide. In the United Kingdom, over 6000 new cases of kidney cancer are diagnosed each year with a crude incidence rate of about 9 per 100 000, a figure which has been steadily increasing over the last three decades1 (Fig. 31.1). There has been a corresponding increase in 5-year, age-standardized and relative survival from a figure of 30 per cent in the 1970s to a current level of 50 per cent (Fig. 31.2). Much of this apparent increase in incidence
Disease relapse Metastatic disease – surgical management Metastatic disease – non-surgical management Prognostic factors Novel agents Conclusion References
735 735 736 740 741 743 743
and survival is likely to be due to the widespread use of ultrasound and cross-sectional imaging resulting in more cases of kidney cancer being detected at an earlier and often asymptomatic stage. Kidney cancer affects men more commonly than women (in a ratio of approximately 3:2) and is predominantly a disease of the elderly. Age-specific incidence rates increase consistently with age with the absolute numbers of cases demonstrating a plateau during the seventh decade (Fig. 31.3).
100
12
Males Females Persons
80 Percentage
Rate per 100 000 population
14
10 8
Men Women
60 40
6 20 4 0
2 0 1975 1978 1981 1984 1987 1990 1993 1996 1999
1971– 1976– 1981– 1986– 1991– 1996– 2000– 1975 1980 1985 1990 1995 1999 2001 Period of diagnosis
Year of diagnosis
Figure 31.1 Age-standardized (European) incidence rates kidney cancer, by sex, 1975–20011
Figure 31.2 Five-year age-standardized survival for kidney cancer by sex for patients diagnosed in England and Wales during 1971–19991
730 Renal cell cancer
80
Male cases Female cases
600
70
Male rates Number of cases
500
60
Female rates 50
400 40 300 30 200
20
100
Rate per 100 000 population
700
10
0 85
80–84
75–79
70–74
65–69
60–64
55–59
50–54
45–49
40–44
35–39
30–34
25–29
20–24
15–19
10–14
5–9
0–4
0
Age at diagnosis
Figure 31.3 Numbers of new cases and age-specific incidence rates by sex, kidney cancer, UK 20011
HISTOPATHOLOGY Eighty to ninety per cent of malignant renal tumours arise from the renal parenchymal tissue with the remainder being uro-epithelial in origin. The vast majority of uroepithelial tumours are transitional cell carcinomas arising from the renal pelvis, and treatment principles are those of uro-epithelial cancer occurring at other sites such as the bladder and ureter (see Chapter 26, Bladder cancer). Renal parenchymal tumours were originally considered to arise from ectopic adrenal rests, hence the coining of the term hypernephroma. In 1960, on the basis of electron microscopic features, Oberling et al demonstrated a proximal renal tubular origin and tumours were renamed renal cell adenocarcinoma or renal cell carcinoma (RCC).2 Subsequently, RCC has been further reclassified and subdivided into conventional (clear cell), papillary (chromophil), chromophobe and collecting duct (Bellini’s duct) renal cell carcinoma3 (Table 31.1).
Conventional (clear cell) renal cell carcinoma Conventional (clear cell) renal carcinomas account for the vast majority (75 per cent or more) of malignant renal cancers in surgical series.4 Macroscopically, tumours have a characteristic appearance with solid areas often interspersed with areas of cystic degeneration. These tumours are richly vascular with numerous vessels throughout the supporting stroma. Microscopically, tumour cells have a clear cytoplasm with a low nuclear to cytoplasmic ratio. Four grades (I–IV) are recognized on the basis of nuclear size, contour and the presence of nucleoli. Sarcomatoid change may be seen in grade IV tumours and represents a poor prognostic feature; it occurs in about 5 per cent of
Table 31.1 Subclassification of malignant renal carcinomas Tumour Type Conventional (clear cell) Papillary (chromophil) Chromophobe Collecting duct
Frequency Cell of Origin 70–80% 15% 5% 2%
Proximal tubule Proximal tubule Cortical collecting duct Medullary collecting duct
tumours. The presence of tumour necrosis is also of prognostic significance and is associated with a worse survival.
Papillary (chromophil) renal carcinoma Papillary (chromophil) carcinomas account for approximately 15 per cent of renal tumours with a male predominance of up to 8:1. Papillary architecture predominates in these tumours. Papillary tumours may be further subclassified into type I or type II. Type I (basophilic) tumours consist of small cells with scanty cytoplasm and small nuclei; type II (eosinophilic) tumours have large cells with abundant eosinophilic cytoplasm and large nuclei.5 Papillary renal cell cancers appear to have a better prognosis than renal cell cancers overall and stage for stage, with a 5-year overall survival rate of about 85 per cent compared to 50 per cent for conventional renal cell carcinoma.6,7 Papillary carcinomas appear to be more common in patients on dialysis for end stage renal disease.
Chromophobe renal carcinoma Chromophobe carcinomas account for approximately 5 per cent of renal carcinomas. In contrast to clear cell and
Inherited syndromes 731
papillary tumours which characteristically arise from the proximal tubule, chromophobe carcinomas are thought to arise from the cortical collecting duct.8 Microscopically, chromophobe tumours consist of large polygonal cells forming a compact growth pattern with pale cytoplasm and prominent cell membranes.9
Hereditary conventional renal cell carcinoma Several familial syndromes are associated with the development of conventional renal cell carcinoma, the most common being von Hippel–Lindau (VHL) disease. VON HIPPEL–LINDAU DISEASE
Collecting duct (Bellini’s duct) and medullary carcinoma Collecting duct (Bellini’s duct) carcinoma is a rare form of renal carcinoma thought to arise from the medullary collecting ducts. It accounts for less than 2 per cent of renal carcinomas. Historically, the term has been used to describe carcinomas with a wide range of appearances. Histologically, tumours may resemble papillary renal cell carcinoma exhibiting tubular papillary growth combined with a solid/ microcystic pattern.10 Medullary carcinoma of the kidney is a rare variant that appears to be found most commonly in African Americans with sickle cell disease or trait.11 Marked expression of vascular endothelial growth factor (VEGF) is seen histopathologically and chronic medullary hypoxia secondary to sickle cell disease may be a factor in the underlying pathogenesis. Collecting duct carcinomas characteristically behave in an aggressive fashion with early local invasion and/or metastatic spread and a poor prognosis.12–14*
Benign renal tumours Benign renal tumours include papillary and metanephric adenomas, renal oncocytomas, angiomyolipomas, fibromas and lipomas. Although occasionally symptomatic, these benign tumours are often detected incidentally. Surgical excision, where appropriate, represents definitive treatment.
Nephroblastoma (Wilms’ tumour) Nephroblastomas are malignant mixed embryonic tumours of the kidney and one of the most common neoplasms in young children, occurring with a peak incidence in the second year of life. Their diagnosis and treatment is discussed in Chapter 46.
INHERITED SYNDROMES A number of hereditary syndromes are associated with the development of kidney cancer which tends to present at a younger age with multifocal and/or bilateral disease. Significant progress has been made in the identification of their underlying genetic abnormalities with important implications for the treatment of both hereditary and sporadic cancers.
Von Hippel–Lindau disease is a rare, autosomal dominant inherited syndrome characterized by the development of multiple, vascular, benign and malignant tumours in the eyes, inner ears, brain, spine, pancreas, adrenal glands and kidneys.15 It is estimated to affect 1:36 000 people with an estimated penetrance of 80–90 per cent by age 65. Renal carcinomas develop in up to half of those affected and are always conventional in subtype and typically multicentric and bilateral. An inherited germline mutation in one allele of the VHL gene, a tumour suppressor gene located on chromosome 3p25, underlies the clinical syndrome. According to a two-hit model, clinical abnormalities such as renal cell carcinomas are thought to arise following loss of the remaining wild-type allele (inherited from the unaffected parent) through somatic mutation in the target organ.16–18 Of considerable importance and wider clinical relevance is demonstration of VHL gene inactivation through deletion, mutation or methylation in the majority of sporadic conventional RCCs.19 The VHL gene encodes a 213 amino acid protein (pVHL) that forms part of a larger ubiquitin ligase complex. This complex is critically involved in the degradation of hypoxia-inducible factor (HIF), a protein transcription factor. HIF induces transcription of a variety of genes that allow cells to grow and survive in hypoxic conditions. In conditions of normal oxygenation and normal pVHL function the pVHL protein complex binds and ubiquinates the alpha subunit of HIF, targeting it for destruction via the proteosome and thereby regulating its activity (Fig. 31.4). In hypoxic conditions or in the presence of VHL mutation, HIF escapes degradation by VHL and remains constitutively active leading to increased transcription of a number of genes important for normal and malignant cellular growth and proliferation (transforming growth factor alpha and beta (TGF-α/-β)), glucose uptake (glut1), angiogenesis (vascular endothelial growth factor (VEGF), platelet derived growth factor (PDGF)) and acid–base balance (carbonic anhydrase IX (CAIX)) amongst others.20 A number of small molecule tyrosine kinase inhibitors that target receptors of HIF inducible proteins such as VEGF have completed early phase studies and are showing considerable activity in patients with conventional RCC. Their use is discussed later in this chapter. TUBEROUS SCLEROSIS
Tuberous sclerosis, an inherited syndrome of epilepsy, mental retardation and adenoma sebaceum, may also be associated with the development of conventional renal cell carcinoma although this is said to occur in only 1–2 per cent
732 Renal cell cancer
pVHL
HIF
Proteasome
HIF degradation Ubiquitination
(a)
pVHL
HIF
ⴙⴙ
Transcription of hypoxiainducible genes, e.g. VEGF
(b) Figure 31.4 Von Hippel–Lindau. A: In normoxic conditions with normal VHL gene function, the VHL protein (pVHL) complex binds to the alpha subunit of hypoxia-inducible factor (HIF) leading to attachment of ubiquitin and degradation via the proteasome. B: In hypoxic conditions or with abnormal pVHL function, HIF escapes regulation leading to increased transcription of hypoxia-inducible genes
of patients. Two underlying genes, tuberous sclerosis complex one and two (TSC1 and TSC2), have been identified with TSC2 thought to act as a tumour suppressor gene.
Hereditary papillary renal carcinoma Mutation in the proto-oncogene met, located on chromosome 7q31, is associated with the rare syndrome of hereditary papillary renal carcinoma. Mutations are inherited in an autosomal dominant fashion and tumours are commonly bilateral and multicentric. The met gene encodes for a tyrosine kinase receptor whose ligand is hepatocyte growth factor.21,22 Although the exact function of the met receptor is unclear, normal organogenesis appears to be dependent on met signalling with activation of the met receptor leading to increased cellular proliferation. A second type of hereditary papillary renal carcinoma relates to mutation in the fumarate hydratase gene located on chromosome 1. This mutation appears to be associated with a more aggressive form of the disease occurring in a younger age group. Other clinical manifestations include skin leiomyomas, and uterine fibroids are seen in women.
Hereditary chromophobe renal carcinoma Patients with Birt–Hogg–Dube syndrome, a rare autosomal dominant syndrome, characteristically develop chromophobe renal cell carcinomas in association with colonic neoplasms, skin lesions and recurrent pneumothoraces.
As is seen with other inherited syndromes, renal tumours in this condition are often multiple and bilateral. The Birt–Hogg–Dube gene has been mapped to chromosome 17p11.2, a mutation that is thought to be rare in sporadic tumours.23
RISK FACTORS A number of risk factors have been implicated in the development of RCC, including cigarette smoking, obesity and hypertension. Case-control and cohort studies have consistently reported a relationship between these risk factors and RCC although many studies have lacked large sample sizes and therefore statistical power.24 A recent large metaanalysis of 19 case-control and 5 cohort studies has suggested an increased relative risk of RCC for ‘ever’ smokers as compared to lifetime never smokers of 1.38 (95 per cent CI 1.27–1.50), with a strong dose-dependent increase in risk.25 A reduction in relative risk was seen in those individuals who had quit smoking for at least 10 years. Obesity appears to be an independent risk factor for the development of RCC with increasing body mass index (BMI) conferring an increased risk of RCC, particularly amongst women (RR 2.25; 95 per cent CI 1.14–4.44; p-trend 0.009).26 The mechanism underlying this link remains unclear but may be related to increased levels of endogenous oestrogens. A further large study has examined the inter-relationship of obesity, hypertension and risk of RCC in a large cohort of men.27 Men with the highest BMI (27) had a significantly elevated RCC risk compared
Clinical evaluation and staging 733
with men with a lower BMI (RR 1.9; 95 per cent CI 1.3–2.7; p-trend 0.001). There was also a direct and independent association between higher diastolic and systolic blood pressures and a higher risk of renal cell cancer (RR 2.2; 95 per cent CI 1.1–4.5; p-trend 0.001). Additional risk factors for renal cell carcinoma include pre-existent renal failure and dialysis, presumably relating to the development of acquired renal cystic disease and occupational exposure to a variety of carcinogens including asbestos, aromatic hydrocarbons, heavy metals, petroleum products and trichloroethylene.28,29
CLINICAL FEATURES Kidney cancer remains occult for much of its clinical course. As many as 25 per cent of patients with kidney cancer are now diagnosed incidentally following diagnostic imaging performed for another indication, a phenomenon that has led to a clear size and stage migration with increasing numbers of earlier, smaller tumours now being detected.30–32 Approximately 50 per cent of patients will present with localized disease, with remaining patients presenting with locally advanced or metastatic disease in roughly equal proportions. Symptoms relating to the primary tumour tend to occur late. Macroscopic haematuria, flank pain and a palpable mass represent the classical triad of symptoms although this presentation occurs in a minority of patients and generally indicates locally advanced disease. Disease may be locally invasive into adjacent structures including ipsilateral adrenal gland, liver, pancreas, spleen and abdominal wall. Local vascular invasion is a typical feature of renal cell carcinoma and tumour may extend into the renal vein and subsequently onwards into the inferior vena cava. Vena caval extension may be longitudinally extensive and direct invasion into the caval wall is also seen. Metastatic spread may occur via lymphatic and haematogenous routes. Early lymphatic spread typically gives rise to loco-regional/retroperitoneal lymphadenopathy and is an adverse prognostic feature. Involvement of more distant nodal groups including mediastinal and peripheral nodes is not uncommonly seen. Haematogenous spread is frequently encountered with common sites of metastatic disease including lung, bone, liver and brain. Lung metastases may present as an incidental finding on a chest X-ray or computed tomography (CT) or with symptomatic breathlessness, cough and occasionally haemoptysis. In patients with involvement of the inferior vena cava, pulmonary embolic disease may occur as a presenting feature. Patients with isolated pulmonary metastases are thought to comprise a better prognostic group compared to those with spread to other visceral sites and such patients may remain asymptomatic for considerable periods of time without treatment. Bone metastases are typically lytic and commonly present with pain or pathological fracture. Metastases to the spine are a particularly difficult
Table 31.2 Survival estimates for renal carcinoma by stage grouping, 1988–2001.35 Survival 1-year 2-year 5-year 10-year
Localized
Regional
Distant
97% 95% 90% 83%
84% 74% 61% 50%
32% 18% 10% 7%
management problem with the associated risk of spinal cord compromise. Hepatic spread may cause right upper quadrant pain and associated hepatomegaly. Brain metastases may cause symptoms of raised intracranial pressure such as headache, visual blurring and nausea or alternatively may present with sudden-onset neurological deficit or fits. In contrast to other epithelial malignancies, metastatic spread to other intra-abdominal organs such as the gastrointestinal tract (stomach, small and large bowel), pancreas and contralateral adrenal gland is not uncommon. Metastatic involvement of the gastrointestinal tract may present acutely with bleeding, obstruction or intussusception. Systemic symptoms are common in patients with locally advanced or metastatic kidney cancer and include fatigue, anorexia, weight loss and fever. Symptoms may be profound and extremely debilitating. Anaemia, hypercalcaemia and an elevated lactate dehydrogenase are also seen and are of adverse prognostic significance.33 Paraneoplastic syndromes relating to hypersecretion of humoral factors such as parathormone-related peptide or erythropoietin may occur.34
CLINICAL EVALUATION AND STAGING As with most cancers, prognosis is clearly related to stage at diagnosis. Small operable tumours are associated with an excellent 5-year survival whilst patients presenting with metastatic disease have a much poorer prognosis (Table 31.2 and Figure 31.5). The International Union Against Cancer (UICC) and American Joint Committee on Cancer (AJCC) staging system is in common use world-wide36,37 (Table 31.3). A thorough clinical examination should be performed in all patients with a suspected diagnosis of kidney cancer. Signs of locally advanced disease may include the presence of a palpable abdominal mass, a varicocoele or lower limb oedema. Supraclavicular or other peripheral adenopathy may be clinically palpable, as may metastatic skin nodules. Urine analysis may demonstrate macroscopic or microscopic haematuria. A full blood count and serum biochemistry including a renal, bone and liver profile plus lactate dehydrogenase are considered routine. Ultrasound and contrast-enhanced CT of abdomen, pelvis and chest (with bone windows) are the most
734 Renal cell cancer
Table 31.3b Stage grouping: International Union Against Cancer (UICC) and American Joint Committee on Cancer (AJCC) staging for kidney cancer36,37
100 90 80
AJCC Stage Grouping
Survival (%)
70 60
Stage I Stage II Stage III
50 40 30
Localized Regional
20
Distant
10 0 1
2
3
4 5 6 7 8 Years from diagnosis
9
10
11
Figure 31.5 Survival estimates for renal carcinoma by stage grouping, 1988–2001.35 Stage IV Table 31.3a TNM staging: International Union Against Cancer (UICC) and American Joint Committee on Cancer (AJCC) staging for kidney cancer36,37
T1 N0 M0 T2 N0 M0 T1 N1 M0 T2 N1 M0 T3 N0 M0 T3 N1 M0 T3a N0 M0 T3a N1 M0 T3b N0 M0 T3b N1 M0 T3c N0 M0 T3c N1 M0 T4 N0 M0 T4 N1 M0 Any T N2 M0 Any T Any N M1
Primary Tumor (T) TX T1 T1a T1b T2 T3
T3a T3b
T3c T4
Primary tumor cannot be assessed Tumor 7 cm or less in greatest dimension, limited to the kidney Tumor 4 cm or less in greatest dimension Tumor more than 4 cm but not more than 7 cm Tumor more than 7 cm in greatest dimension, limited to the kidney Tumor extends into major veins or invades adrenal gland or perinephric tissues but not beyond Gerota’s fascia Tumor directly invades the adrenal gland or perirenal and/or renal sinus fat but not beyond Gerota’s fascia Tumor extends into renal vein or its segmental (muscle-containing) branches, or vena cava below the diaphragm Tumor extends into vena cava above diaphragm or invades the wall of the vena cava Tumor invades beyond Gerota’s fascia
Regional Nodal (N)* NX N0 N1 N2
Regional lymph nodes cannot be assessed No regional lymph node metastases Metastases in a single regional lymph node Metastases in more than one regional lymph node
Distant Metastasis (M) MX M0 M1
Distant metastasis cannot be assessed No distant metastasis Distant metastasis
*Laterality does not affect the N classification
commonly used imaging modalities and staging tools for renal tumours. Particular attention should be paid to the assessment of direct extension into adjacent organs, enlargement of retroperitoneal lymph nodes, adrenal involvement, invasion of the ipsilateral renal vein (with or without extension into the inferior vena cava), and the presence or absence of distant metastatic disease. Technetium-99m bone scanning typically underestimates or understages bony disease but may be helpful in evaluating individuals with symptomatic bone pain or elevated alkaline phosphatase.38,39 Plain films are helpful in evaluation of local bony disease. Magnetic resonance imaging (MRI) is a particularly sensitive modality for evaluating sites of bony metastatic disease in the spine and may also be used as an alternative to CT for local staging.
LOCALIZED DISEASE Surgical management Patients with early-stage renal cell carcinoma who are medically fit should undergo surgery with curative intent. Radical nephrectomy comprising complete excision of Gerota’s fascia and its contents (kidney and ipsilateral adrenal gland) represents standard surgical management although the need for ipsilateral adrenalectomy in macroscopically normal glands has been questioned.40 Regional lymphadenectomy is often performed at the time of nephrectomy and although it has been difficult to conclusively demonstrate a survival benefit for such an approach, important prognostic information may be obtained from
Metastatic disease – surgical management 735
histopathological evaluation of resected nodes since the majority of patients with local nodal involvement will ultimately relapse.41,42* A variety of surgical techniques are currently practised and include anterior transperitoneal, flank, thoracoabdominal and laparoscopic approaches with the choice of approach relating to the size and location of the tumour, presence or absence of inferior venal caval invasion and the body habitus of the patient. Image-guided renal vessel embolization may be performed electively prior to operative intervention and may help to facilitate surgery by reducing tumour vascularity and minimizing surgical blood loss. In patients with symptomatic primary tumours where surgery is considered inappropriate or where the primary is felt to be inoperable, renal embolization may be used alone as effective palliation.43* Where tumour extends into vena cava, caval removal may be required. Extensive caval infiltration particularly with supradiaphragmatic extension does not preclude surgical intervention but may require intraoperative cardiopulmonary bypass. Where technically possible, an alternative to radical nephrectomy for small (less than 4 cm) localized tumours is partial nephrectomy performed via an open or laparoscopic approach. There is an increasing trend towards the use of such nephron-sparing surgery, particularly in patients with a solitary kidney, when tumours occur in a transplanted kidney and in hereditary kidney cancer where tumours may be multiple and/or bilateral.
Radiofrequency ablation This treatment modality is gaining interest as an alternative in small (generally less than 3 cm), localized, earlystage renal cell cancer, particularly in patients with operable disease who are considered high risk due to significant comorbidity or advancing age or in patients with bilateral disease. A needle electrode is inserted into the lesion with imaging guidance following local anaesthesia. When the electrode is connected to an energy source, local ionic agitation and frictional heat leads to cell death and coagulation necrosis. The treated site is then gradually replaced by fibrosis/scar tissue. Treatment is generally carried out over a 10–60 minute period. Confirmatory, pre-ablational biopsies should be performed at the same sitting to obtain histological diagnosis which would not otherwise be available with such an approach.
may be very wide with some patients relapsing many years after initial surgery. Pulmonary metastatic disease is the most common site of spread, occurring in up to two thirds of patients.44,45
METASTATIC DISEASE – SURGICAL MANAGEMENT Cytoreductive nephrectomy For individuals presenting with metastatic disease, surgical removal of the primary tumour (cytoreductive nephrectomy) has historically been performed in patients with symptomatic primary tumours or in patients with operable primary tumours and isolated metastatic disease amenable to resection. Support for such an approach included the observation that primary tumours respond relatively poorly to immunotherapy and that very occasionally metastatic disease could spontaneously regress following surgical removal of the primary.46* Two randomized controlled trials have now demonstrated an improvement in survival following cytoreductive nephrectomy in selected patients presenting with metastatic disease prior to a planned course of interferon alpha.47,48** In both trials, patients were highly selected and of good performance status (ECOG 0–1). Combined analysis of the two studies demonstrated a weighted mean difference in median survival to be 4.8 months (12.6 vs. 7.8 months) in favour of the surgical arm. Both arms demonstrated a similarly low rate of partial or complete remission to systemic therapy (19/331 patients; 6 per cent) with no significant difference between arms.49 Despite the demonstration of these modest improvements in median survival, such potential gains need to be carefully considered against the potential morbidity, mortality and recovery time from surgery. These randomized trials do not support a blanket policy of cytoreductive nephrectomy in all patients presenting with metastatic RCC. Indeed, many patients presenting with metastatic disease are unfit and of poor performance status (PS). It is informative to note that even within the restrictive entry criteria of performance status 0–1 in the randomized trial, subgroup analysis suggested performance status to have significantly greater impact upon survival than therapy: median survival for patients of PS 0 vs. PS 1, receiving interferon plus nephrectomy was 17.4 vs. 6.9 months, for interferon alone it was 11.7 vs. 4.8 months.47
DISEASE RELAPSE Approximately 25 per cent of patients will relapse following radical nephrectomy performed with curative intent. Despite these figures, standard postoperative management is observation alone with no benefit seen in a number of trials of adjuvant cytokine therapy. The median time to relapse is between one and two years although the range
Resection of loco-regionally recurrent or metastatic disease Patients with kidney cancer may present or relapse with macroscopically isolated, resectable loco-regional or metastatic disease. Given the occasionally indolent nature
736 Renal cell cancer
of renal cell cancer and the limited efficacy of systemic treatment, metastectomy is occasionally performed, particularly in patients who have experienced long treatmentfree intervals. Case series have examined the role of surgical excision of resectable recurrent renal fossa disease following previous nephrectomy. Such studies suggest that such an approach may be associated with long-term local control and prolonged progression-free survival although randomized evidence is lacking.50–52* Resection of pulmonary metastatic disease is also widely practised and the subject of numerous case series.53–56* Randomized evidence is again lacking but such series report 5-year survival figures in the region of 40 per cent; ultimately, however, virtually all patients will relapse. Isolated or threatening bone disease is a particular management problem. Good, early local disease control is important and where appropriate and technically feasible, aggressive surgical intervention is warranted; conservative treatment in this setting is almost always complicated by further local progression and particularly difficult ongoing management.57* Spinal metastatic disease may prove particularly problematic, with the associated risk of spinal cord compression. Randomized evidence suggests that early surgical intervention followed by radiotherapy is superior to radiotherapy alone in malignant cord compression, and in a relatively radioresistant tumour such as renal cell cancer, this may be particularly true.58** Where disease appears to be isolated to a single, symptomatic vertebral body then aggressive local surgery may decrease the chance of subsequent local progression and the possibility of cord compression. Given the exceptional vascularity of metastatic kidney cancer, pre-surgical tumour embolization is commonly utilized to minimize surgical blood loss. Surgical intervention has been practised for single or multiple metastatic lesions and may be performed on more than one occasion. Patients with clinically evident synchronous metastatic disease appear to experience decreased survival when compared with patients who develop metastasis after the primary tumour is removed.59*
METASTATIC DISEASE – NON-SURGICAL MANAGEMENT The use of non-surgical approaches for the treatment of kidney cancer is currently restricted to patients with locally advanced or metastatic disease. There is currently insufficient evidence to support the use of adjuvant therapy outside the confines of a clinical trial. Metastatic renal cell carcinoma may behave in a biologically variable fashion with a large variation in observed survival although overall, metastatic disease is associated with poor survival and limited response to systemic therapy. As such, it remains a priority to identify more effective therapy through involvement in clinical trials.
Radiotherapy Although renal cell carcinomas are thought to be relatively radioresistant, radiotherapy can still play an important part in the management of local treatment for symptomatic (predominantly bony) metastatic disease. Case series suggest that the vast majority of patients will experience symptomatic benefit from treatment although duration of response may be short lived. Higher biologically effective doses of radiotherapy may be associated with improved palliation.60–62* Post-surgical radiotherapy is commonly used to decrease local recurrence/progression rates following orthopaedic intervention in the spine or long bones.
Chemotherapy Historically, the use of cytotoxic chemotherapy in advanced RCC has proven to be disappointing with low response rates and little impact on survival. In an early review by Yagoda et al, an analysis of trials of 72 cytotoxic agents in over 3500 patients suggested a response rate of only 5.6 per cent.63* More recent reviews have revealed similarly disappointing results.64,65 The use of more novel chemotherapy agents such as gemcitabine has been associated with some enthusiasm secondary to small early phase studies suggesting limited activity in selected patients.66,67 In general, however, RCC is considered to be an inherently chemorefractory tumour and chemotherapy is not in routine use as a single modality of treatment.*
Immunotherapy The variable natural history of advanced kidney cancer and the observation that some patients with metastatic disease occasionally undergo spontaneous remission lends support to the hypothesis that the host immunity may play an important role in the regulation of tumour progression and also provides some support for the therapeutic rationale in using immunotherapy approaches. Following the disappointing results seen with chemotherapy, immunotherapy has become the mainstay of treatment for advanced kidney cancer. The main immunotherapeutic approach has involved the administration of recombinant cytokines including interferon alpha (IFN-α) and interleukin-2 (IL-2) in a variety of doses and schedules. The use of these agents has become the standard of care for advanced RCC*** (Table 31.4). A variety of other immunotherapeutic approaches have also been undertaken in the treatment of renal cell carcinoma including adoptive immunotherapy, vaccination and allogeneic stem cell transplantation. INTERFERON ALPHA
The three major interferons (IFN-α, -β and -γ) have a number of immunoregulatory effects including stimulatory
Metastatic disease – non-surgical management 737
Table 31.4 Selected trials of immunotherapy in renal carcinoma Trial
Size
RR
MS (Months)
Comment
Interferon alpha IFN-α vs. MPA68,69 IFN-α vs. IFN-α CRA70
n 335 n 320
14% vs. 2% NR
8.5 vs. 6 17.3 vs. 13.2
Survival benefit for IFN-α therapy Survival benefit following addition of CRA but increased toxicity
Interleukin-2 HD IL-2 single arm71,72
n 255
15%
16.3
HD IL-2 vs. LD IL-2 IV vs. LD IL-2 SC73
n 400
21% vs. 13% vs. 10%
NR NS
Prolonged survival in complete responders (n 17). High toxicity, 4% treatmentrelated mortality LD IL-2 associated with decreased toxicity but inferior response duration and survival for complete responders
Combination cytokine therapy CVI IL-2 vs. IFN-α vs. CVI IL-2 IFN-α74
n 425
6.5% vs. 7.5% vs. 18.6%
12 vs. 13 vs. 17 (NS)
n 492
4.1% vs. 4.4% vs. 10.9% vs. 2.5%
15.3 vs. 15.2 vs. 16.8 vs.14.9 (NS)
n 160
16.5 vs. 2.5%
17.9 vs. 9.5
n 78
38% vs. 0%
24 vs. 14
n 341
31% vs. 26% vs. 20% (NS)
25 vs. 27 vs. 16
CVI IL-2 vs. IFN-α vs. CVI IL-2 IFN-α vs. MPA75 Biochemotherapy IFN-α VBL vs. VBL76 IL-2 IFN-α 5-FU vs. tamoxifen77 IL-2 IFN-α 5-FU vs. IL-2 IFN-α 5-FU 13-CRA vs. IFN-α VBL78
Improved response rate for combination but no survival difference Intermediate prognosis group only. Increased toxicity with no survival benefit from cytokine treatment Survival benefit for IFN-α therapy. Contribution of VBL uncertain High response rate seen with combination regimen Inferior survival in IFN-α/VBL arm. 13-CRA not of benefit
Key: RR, response rate; MS, median survival; IFN-α, interferon alpha; MPA, medroxyprogesterone acetate; 13-CRA, 13-cis-retinoic acid; NR, not reported; HD IL-2, high-dose interleukin-2; LD IL-2, low-dose interleukin-2; IV, intravenous; SC, subcutaneous; NS, not significant; CVI, continuous venous infusion; VBL, vinblastine; 5-FU, 5-fluorouracil.
effects on macrophages, NK cells, T-lymphocytes and neutrophils and facilitation of antigen presentation through up-regulation of class I and class II major histocompatibility complex expression. Interferons may also inhibit angiogenesis, induce cellular differentiation and inhibit the proliferation of tumour cells. Interferon alpha (IFN-α) appears to have the greatest activity against renal cell carcinoma and as such has been most intensively studied. Several randomized studies have addressed the efficacy of IFN-α in the treatment of kidney cancer with or without the use of additional nonimmunotherapeutic agents.68,76,79,80** A key randomized phase III study from the Medical Research Council evaluated the use of IFN-α at a dose of 10 million units subcutaneously, three times weekly against medroxyprogesterone acetate (MPA) 300 mg orally once daily for 12 weeks in 335 patients with metastatic RCC (68, updated in ref. 69). The primary endpoint was survival and IFN-α was associated
with a 28 per cent reduction in the risk of death (HR 0.72 (95 per cent CI 0.55–0.94), p 0.017). Both median survival (8.5 vs. 6 months) and 1-year survival (43 per cent vs. 31 per cent) were improved in the IFN-α arm. The use of IFN-α was associated with an overall response rate of 14 per cent (CR 2 per cent) at 3 months; 42 per cent of patients had stable disease. A recent Cochrane Collaboration review of immunotherapy for advanced RCC has been undertaken.81*** Analysis of the randomized trials of IFN-α against control regimens which currently would be considered largely ineffective (vinblastine, MPA), suggests a median survival advantage of 2.6 months with a reduction in the risk of death for at least the first two years (pooled HR 0.79, 95 per cent CI 0.68–0.91) compared to control therapy. Despite the rather modest effects of IFN-α seen in the above trials, it remains a standard treatment option for advanced renal cell cancer. However, IFN-α is associated with significant, dose-related
738 Renal cell cancer
Approval for the use of IL-2 as treatment for advanced renal cell cancer in the US was granted following the publication of data collated from a series of seven phase II trials from the National Cancer Institute (NCI).71 Two hundred and fifty-five highly selected patients were treated with high-dose IL-2 (600 000 or 720 000 IU/kg intravenously every 8 hours over 5 days) followed by a second identical cycle of treatment after 5–9 days of rest. Courses were repeated every 6–12 weeks in stable or responding patients. Initial results were updated in a subsequent publication.72* Objective responses were seen in 37 of 255 patients (15 per cent) with 17 complete responses (7 per cent) and 20 partial responses (8 per cent). Median survival time for all 255 patients was 16.3 months (Fig. 31.6a). For patients entering a partial response, the median duration of response was 19.0 months. The median duration of response for complete responders was not reached, but was at least 80 months (range, 7 to at least 131 months). Despite the highly selected cohort of patients, treatment was associated with severe acute toxicity. Severe vascular leak syndrome predominated with patients at risk of developing peripheral and pulmonary oedema, hypotension, oliguria and
Probability of survival
1.0
0.6 0.4 0.2
0
10 20 30 40 50 60 70 80 90 100 110 120 130 140 Months
(a) Figure 31.6a Overall survival for all patients treated with highdose IL-2 therapy71,72
1.0 0.8 0.6 0.4 0.2 0.0
INTERLEUKIN-2
Interleukin-2 (IL-2) is a critical cytokine in the activation of cellular and humoral immune responses. IL-2 induces proliferation of antigen primed CD4 and CD8 positive T cells and enhances the cytotoxic activity of NK cells.
0.8
0.0
Probability of continuing response
toxicity which commonly includes anorexia, mild nausea, influenza-like symptoms and lethargy. Other toxicity may include reversible myelosuppression, depression (including suicidal behaviour), hypersensitivity reactions, thyroid abnormalities, alopecia, hyperglycaemia (or worsening of diabetic control), hypertriglyceridaemia and skin rash. Nephrotoxicity and hepatotoxicity have also been reported. Side effects may be severe and dose limiting although the severity of these symptoms commonly diminishes over time. Evening administration of IFN-α along with co-administration of paracetamol is routinely recommended and may help to ameliorate toxicity. The concurrent use of prednisolone may be dramatic in helping to diminish toxicity although theoretically corticosteroids might also diminish the immunomodulatory effects of IFN-α therapy. Despite such manoeuvres, a significant minority of patients go on to require dose reductions or even cessation of treatment as a result of excessive toxicity. The benefit of interferon appears to be greatest in patients with good performance status and good prognosis disease.82** (see Prognostic factors, below). The low response rate of interferon and its significant toxicity profile mean that treatment should be reserved for fitter patients of good performance status (ECOG 2). Response to IFN-α is not rapid and may take several months. However, the majority of patients who are going to respond are likely to have done so within three months of treatment.68** Patients with conventional (clear cell) histology appear more likely to respond than patients with other histological subtypes. The optimal duration of IFN-α therapy in responding patients is unknown. Treatment may be continued indefinitely for those patients who are deriving clinical benefit but, in practice, chronic IFN-α related toxicity will often result in discontinuation of treatment. In the small proportion of patients who achieve a good response and have a prolonged treatment-free interval following cessation of therapy, rechallenge with IFN-α may prove beneficial. Attempts have been made to improve response rates and median survival following IFN-α therapy through the addition of agents such as 13-cis-retinoic acid (13-CRA), which is postulated to increase the antiproliferative effects of IFN-α. A phase III study from Motzer et al in 284 patients failed to show any significant improvement in response rate or overall survival although progression-free survival was prolonged in the combination arm (33 vs. 22 months, p 0.03).83 A subsequent study of similar design in 320 patients demonstrated an improvement in overall survival (17.3 vs. 13.2 months, p 0.048) which was marginally statistically significant albeit at the cost of increased treatment-related toxicity.70
0
10
20
30
40
50
60
70
80
90 100 110 120 130
Duration of response (months)
(b) Figure 31.6b Duration of response for patients entering complete response to high-dose IL-2 therapy71,72
Metastatic disease – non-surgical management 739
multi-organ failure. The majority of patients required intravenous resuscitation with fluids and inotropic/ vasopressor support and there was a high level (4 per cent) of treatment-related mortality. For many clinicians, the severe toxicity, potential morbidity and mortality of high-dose IL-2 is seen as unacceptable. High-dose IL-2 treatment is limited to highly selected patients treated in a limited number of centres where persisting enthusiasm for its use relates mainly to the potentially prolonged remission seen in a small minority of patients who enter CR71* (Fig. 31.6b). A variety of modified IL-2 schedules and dose levels have been evaluated with the intent of attempting to decrease treatment-related toxicity whilst maintaining efficacy. Further research from the NCI evaluated the use of standard high-dose (HD) IL2 against low-dose (LD) (10 per cent dose) IL-2 treatment given intravenously (IV) or subcutaneously (SC) (73,84). Four hundred selected, good performance status patients were evaluated. The overall response rate to HD IV IL-2 was significantly higher (21 per cent; 7 per cent CR) than LD IV IL-2 (13 per cent) or SC IL-2 (10 per cent). After a median follow-up of 7.4 years, there was no overall survival difference seen between the three arms. As expected, toxicity was significantly less severe in the LD IV and SC IL-2 arm and, in contrast to earlier studies, there were no IL-2 related deaths in any arm. However, reduction in toxicity occurred at the cost of inferior response duration and survival in the small numbers of completely responding patients in the LD IV therapy compared with HD IV therapy (p .04). These results led the authors to conclude that when IL-2 therapy is being considered, HD IL-2 therapy should remain the treatment of choice where treating physicians are able to pursue such a regimen bearing in mind the risk of irreversible toxicity or death that appears to be less than 1 per cent. COMBINATION CYTOKINE THERAPY
The French Groupe Francais d’Immunothérapie have conducted several large-scale randomized studies comparing the role of IFN-α, IL-2 or combinations of both cytokines. In an early study, 425 patients with metastatic RCC and good performance status were randomized to receive one of three arms: single agent IL-2 was given as a 5-day continuous venous infusion at a dose of 18 million units/m2 per day; IFN-α as a single agent was given at a dose of 18 million units subcutaneously three times a week. In the combination arm, IFN-α was given at a lower dose of 6 million units, subcutaneously, three times a week during each IL-2 infusion.74 Response rates between single agent IL-2 and IFN-α were not significantly different (6.5 per cent vs. 7.5 per cent). A higher response rate (18.6 per cent, p 0.01) and progression-free survival (20 per cent at one year vs. 15 per cent for IL-2 and 12 per cent for IFN-α, p 0.01) was seen for the combination arm but this did not translate into a difference in overall survival between the three arms. Toxicity was dramatically higher in the IL-2
containing arms. A poor prognostic subgroup of patients with more than one metastatic site, liver involvement and a short (1 year) interval from nephrectomy to metastatic disease were identified in this study. Such patients were seen to have a poor survival (median 6 months) and virtually no chance of benefiting from cytokine therapy. A further publication analysed the benefit of crossover therapy in the above patient group following failure of interferon or IL-2 treatment. One hundred and thirteen patients went on to receive second-line crossover therapy. Only four patients subsequently responded to treatment (three with IL-2, one with IFN) suggesting that second-line crossover cytokine therapy is unlikely to be of benefit.74a* A further large randomized study has evaluated the role of cytokine treatment in patients of intermediate prognosis, excluding those poor prognosis patients identified by the previous study and patients considered to be of good prognosis with only a single site of metastatic disease.75 At the time of writing this study has not yet been published as a full manuscript. The last update of 492 patients with good performance status but intermediate prognosis disease were randomized to 18 million units (MU) of IL-2 subcutaneously (SC) for 5 days a week for two cycles of 4 weeks, or 9 MU IFN-α SC three times per week, or a combination of both cytokines using the same IL-2 schedule with reduced-dose 6 MU IFN-α SC three times per week or medroxyprogesterone acetate 200 mg daily. Response rates were low (less than 10 per cent) in all four arms at three and six months and median survival (OS 15 months; NS) did not differ between arms. As expected, toxicity was considerably higher and quality of life poorer in patients receiving cytokines, particularly if they were IL-2 containing. These trials fail to provide good evidence to support the use of combination cytokine therapy in the patient groups studied and question the utility of any cytokine therapy in patients judged to be of intermediate and poor prognosis.75 Patient selection is clearly of considerable importance when considering the use of cytokine therapy and where cytokines are utilized outside the confines of a clinical trial, treatment should be with a single agent.74 BIOCHEMOTHERAPY
A number of trials have evaluated the combination of chemotherapy with cytokine therapy. The rationale for such an approach has been the lack of cross-resistance and overlapping toxicity between the two modalities of treatment and the proposal that chemotherapy may lead to an increase in tumour antigens available for recognition by the host immune system. Critics of such an approach cite the inherently chemo-refractory nature of kidney cancer and that the administration of chemotherapy is likely to be immunosuppressive, down-regulating any potential immune response that cytokine therapy might augment. Pyrhonen et al evaluated the role of vinblastine in combination with IFN-α against vinblastine alone in a phase III trial in 160 patients.76 Both response rate (16.5 per cent vs.
740 Renal cell cancer
2.5 per cent) and median survival (68 vs. 38 weeks, p 0.0049) were improved in the combination arm. However critics have argued that the benefit seen was solely due to IFN-α rather than combination treatment and a further phase III study comparing IFN-α with or without vinblastine found no statistically significant improvement in overall survival although the combination was associated with a modest improvement in response rates (24 per cent vs. 11 per cent).85 The highest published response rates to biochemotherapy have been obtained using combination IL-2, IFN-α and bolus 5-FU in a regimen initially described by Atzpodien et al. In a randomized study of this combination against tamoxifen in 78 patients, the combination arm led to a response rate of 39 per cent compared to 0 per cent for tamoxifen.77** Overall survival was significantly improved in the combination arm (24 months vs. 14 months, p 0.04). A subsequent publication formally compared the combination regimen to the same treatment with addition of 13-cis-retinoic acid against a control arm of IFN-α plus vinblastine.78 Three hundred and forty-one selected good and intermediate prognosis patients were randomized within the study. Response rates were not significantly different at 31 per cent, 26 per cent and 20 per cent respectively. The interleukin containing arms were associated with an improvement in overall survival (25 months vs. 27 months vs. 16 months; p 0.025). The addition of 13cis-retinoic acid did not improve efficacy but did increase toxicity. Although patients were stratified at study entry, fewer patients in the IFN-α/vinblastine arm had undergone nephrectomy and there was a higher incidence of patients with poor prognostic metastatic sites (bone and CNS) in this arm. Despite the above trials, immunochemotherapy is not in widespread routine clinical use. A number of other centres have attempted to replicate the results published by Atzpodien’s group, however response rates and survival have varied widely and in general have been inferior.86–90 Differences in patient selection and the dosing/scheduling of biochemotherapy may explain some of these findings. A large MRC/EORTC randomised trial (RE04) comparing Atzpodien’s regimen against single agent IFN-α in advanced renal cell cancer is currently nearing completion with the aim of recruiting 1100 patients.
379 patients were assessable for the intention-to-treat analysis. Patients were excluded for a variety of reasons including incorrect histological diagnosis, clinical stage or an inability to prepare tumour vaccine. Five-year progression-free survival rates were 77.4 per cent in the vaccine group and 67.8 per cent in the control group. Overall survival was not reported. A similar approach using an autologous vaccine derived from heat shock protein plus tumour derived peptides (Oncophage, HSPCC-96; Antigenics) has also been undertaken and a large adjuvant clinical trial is underway. For the present time such an approach remains experimental.
VACCINATION
Renal cell carcinoma can behave in a biologically diverse fashion with survival for patients with metastatic disease ranging from a few weeks to many years or even decades. Given the heterogeneity of this disease, the relatively low response rate to cytokine therapy and the potential toxicities of treatment, a number of attempts have been made to identify prognostic factors that might help to predict response to treatment, stratify patients for clinical trials and help guide estimation of overall survival. The Memorial Sloan Kettering group identified five separate variables which appeared to act as risk factors for short survival in a heterogeneous group of 670 patients with advanced RCC treated in a variety of prospective
Adjuvant therapy is not in routine use following resection of renal cell carcinoma performed with curative intent. However, data from a German group have suggested that individualized adjuvant treatment with an autologous tumour cell vaccine may be beneficial. Prior to surgery, patients with pT2–3b pN0–3 M0 renal cell carcinoma were randomized to receive autologous cell vaccine (six intradermal injections at 4-week intervals) or no adjuvant treatment (control group). The primary endpoint of the trial was to reduce the risk of tumour progression, defined as progression or death. Of 558 patients initially identified,
TRANSPLANTATION
Interesting data concerning the use of non-myeloablative allogeneic peripheral blood stem cell transplantation were published in 2000.91* This case series of 19 patients evaluated the role of a graft-versus-tumour effect in patients with cytokine refractory advanced RCC treated with a highly immunosuppressive conditioning regimen of cyclophosphamide and fludarabine followed by infusion of stem cells from an HLA-matched sibling. Post-transplant immunosuppression with cyclosporin was then withdrawn early in patients with disease progression to induce a graftversus-tumour effect, with unresponding patients receiving up to three further infusions of donor lymphocytes. Ten patients responded to treatment with 3 patients obtaining a complete response. Typically responses occurred late (at a median of 4 months following initial chemotherapy) and following withdrawal of ciclosporin. Nine of the 10 responding patients developed evidence of graft-versus-host disease, the only factor predictive of a response to treatment. Although this study provides a further insight into the potential efficacy of immune manipulation in the treatment of RCC, such an approach is not applicable to the vast majority of patients presenting with metastatic RCC. Treatment was associated with substantial toxicity and in this small series two out of 19 treated patients died of transplant-related causes.
PROGNOSTIC FACTORS
Novel agents 741
clinical trials.92 These included low performance status, elevated lactate dehydrogenase, low serum haemoglobin, elevated corrected serum calcium and absence of prior nephrectomy. Using these variables, patients were classified into one of three risk groups. Patients classified as favourable risk (zero risk factors) had a median survival of 20 months in comparison to an intermediate risk group (one or two risk factors) with a survival of 10 months and a poor risk group (three or more risk factors) whose median survival was particularly poor at 4 months. Following the acceptance of IFN-α as a standard treatment regimen and randomized evidence supporting the role of cytoreductive nephrectomy prior to cytokine therapy in appropriate patients with metastatic disease, a further analysis was undertaken in 463 patients treated with IFN-α in a variety of prospective randomized trials.33 In this analysis, interval from diagnosis to introduction of IFN-α therapy replaced absence of prior nephrectomy as one of the five risk factors used to create a risk model. Median survival for patients in the favourable risk group was 30 months compared to 14 months for the intermediate risk group and 5 months for poor risk patients (Fig. 31.7). The French Groupe Francais d’Immunothérapie have conducted a similar study in 782 patients treated with a variety of cytokine regimens.93 Amongst a variety of poor prognostic factors identified, four independent factors were predictive of rapid unfavourable progression under cytokine treatment including presence of hepatic metastases, short interval (1 year) from diagnosis of primary tumour to metastatic disease, more than one metastatic site and elevated neutrophil count. Patients with three or more of these factors had a 80 per cent chance of rapid progression despite cytokine treatment and as such, patients in this subgroup should be considered poor candidates for such treatment. An international kidney cancer working group (IKCWG) has been set up to establish a comprehensive database of
1.0 0 risk factor (80 patients, 21 alive) 1 or 2 risk factors (269 patients, 36 alive) 3, 4, or 5 risk factors (88 patients, 0 alive)
0.9 Proportion of surviving
0.8 0.7 0.6 0.5 0.4
potential prognostic factors and derive a single validated model to describe outcome. The use of prognostic groupings is encouraged and may be particularly helpful in influencing treatment decisions and advising patients about prognosis.
NOVEL AGENTS An understanding of the molecular pathways underlying the development and progression of renal cell carcinoma has led to the identification of a number of key therapeutic targets (Table 31.5). As discussed earlier, defective VHL gene function is apparent in the majority of sporadic conventional renal cell carcinomas and the targeting of VHL regulated genes such as VEGF through the use of monoclonal antibodies or small molecule tyrosine kinase inhibitors has represented the main thrust of development to date. Early clinical trial data are very encouraging and numerous phase II and III clinical trials are currently underway or on the verge of reporting. Considerable optimism has surrounded the development of new agents that are beginning to demonstrate substantial activity in cytokine refractory disease. The number of active systemic treatments available for the treatment of advanced kidney cancer is likely to increase dramatically in the near future, and in the long term targeted agents may well replace cytokine therapy as the first-line treatment of choice.
Bevacizumab Bevacizumab, a humanized anti-VEGF monoclonal antibody, was one of the first novel agents to demonstrate activity in conventional RCC. A randomized, double-blind phase II trial from the NCI evaluated the efficacy of two doses of bevacizumab against placebo in patients with progressive metastatic conventional RCC, the majority of whom had received prior IL-2 therapy.94 The trial was stopped early after an interim analysis of data from 116 patients revealed a statistically significant progression-free survival advantage of 2.3 months (4.8 vs. 2.5 months, p 0.001) for the highdose (10 mg/kg) arm compared to placebo. The response rate (all partial responses) to high-dose bevacizumab was 10 per cent. No difference in overall survival was seen although crossover from placebo to bevacizumab was allowed following disease progression. Toxicity was minimal with hypertension and asymptomatic proteinuria predominating.
0.3 0.2
Bevacizumab and erlotinib
0.1 0.0 0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16
Years from start of interferon alpha therapy
Figure 31.7 Overall survival for good (0 risk factors), intermediate (1–2 risk factors) or poor (3 or more risk factors) prognosis disease by Motzer criteria33
A further single arm phase II trial provides some additional support for the efficacy of VEGF blockade by bevacizumab.95 Sixty-three patients with progressive metastatic conventional RCC were treated with bevacizumab in combination with erlotinib, an oral epidermal growth factor (EGFR) receptor tyrosine kinase inhibitor. Previous
742 Renal cell cancer
Table 31.5 Selected novel agents in the treatment of renal cell carcinoma Agent
Description
Target
Route
Current Status
Bevacizumab Sorafenib
MAb SM TKI
Intravenous, fortnightly Oral, daily
Phase III Phase III
Sunitinib Temsirolimus Anti-caIX/g250 ABX-EGF Erlotinib/gefitinib
SM TKI SM MAb MAb SM TKIs
VEGF VEGFR, PDGFR Raf Kinase VEGFR, PDGFR Flt 3, KIT mTOR caIX EGFR EGFR
Oral, daily for 4/52, q 6/52 Intravenous, weekly Intravenous Intravenous, weekly Oral
Phase III Phase III Phase III Phase II Phase II
Mab, monoclonal antibody; VEGF, vascular endothelial growth factor; SM, small molecule; TKI, tyrosine kinase inhibitor; PDGFR, platelet derived growth factor receptor; mTOR, mammalian target of rapamycin; caIX, carbonic anhydrase IX; EGFR, epidermal growth factor receptor.
clinical experience of EGFR blockade through the use of small molecules (erlotinib/gefitinib) or by monoclonal antibody (cetuximab/ABX-EGF) had been disappointing with little clinical activity seen.96–99* The combination of bevacizumab and erlotinib however was associated with a 25 per cent response rate suggesting that VEGF blockade may have some role in overcoming resistance to EGF inhibition. Further bevacizumab treatment combination studies are underway and a large, randomized phase III trial of combination IFN-α plus bevacizumab versus IFN-α alone has recently been completed.
Temsirolimus
Anti-g250/caIX antibody
Multi-targeted tyrosine kinase inhibitors
Carbonic anhydrase IX (caIX; g250) is an HIF-inducible cell surface antigen absent on most normal tissues but over-expressed on the majority of RCCs and levels of expression are thought to relate to prognosis. Early phase data using naked antibody, radiolabelled antibody or antibody in combination with cytokine therapy have been published.100–103 Overall, response rates were low although a number of patients did achieve disease stabilization and treatment was well tolerated. A large phase III trial in the adjuvant setting is underway.
Much of the current enthusiasm for novel agents in RCC relates to the use of oral, multi-targeted, small molecule tyrosine kinase inhibitors. Such drugs are thought to inhibit a number of HIF-inducible, VHL-regulated receptor tyrosine kinases such as VEGF and PDGF and also Raf kinase, FLT3 and cKIT mediated signalling. Several agents are in clinical development with sorafenib (Bay 43-9006; Bayer) and sunitinib (SU11248; Pfizer) currently in phase III evaluation. Phase II data for both sorafenib and sunitinib have suggested an impressive 70–80 per cent clinical benefit rate for patients with progressive cytokine refractory RCC with up to 40 per cent of patients achieving a partial response to treatment.106–108* Such patients would previously have been treated with supportive care, palliative radiotherapy or in clinical trials of experimental agents. Preliminary phase III data evaluating sorafenib in cytokine refractory RCC have also been presented in abstract form.109 Nine hundred and five patients with cytokine refractory advanced RCC were randomized to either placebo or sorafenib therapy. Data were available for 769 patients. Compared to phase II data, the response rate according to strict RECIST (Response Evaluation Criteria in Solid Tumors) criteria110 was disappointingly low (2 per cent). However, the vast majority of patients achieved significant and rapid improvements in symptoms relating to minor reductions in tumour volume (stable disease rate 78 per cent,
Thalidomide/lenalidomide Thalidomide was developed in the 1950s as a sedative and anti-emetic. Its use was withdrawn following observation of its teratogenic properties in the embryos of women who were exposed to the drug in early pregnancy. In the mid 1990s, it was postulated that the teratogenic effects of thalidomide were the result of an anti-angiogenic effect and as such its efficacy has been studied in a variety of malignancies. A variety of phase II studies in RCC have evaluated doses ranging from 100 mg to 1200 mg (reviewed in ref 104). Response rates are low (⬃10 per cent or less) although some patients may achieve disease stabilization. A novel thalidomide analogue, lenalidomide, is currently under investigation.
Temsirolimus (CCI-779; Wyeth) is a novel inhibitor of the mammalian target rapamycin (mTOR) kinase, interfering with cell cycle control and leading to cell cycle arrest. Nonrandomized phase II data have suggested some activity in cytokine refractory RCC, particularly in patients with poor prognostic features.105* A large phase III international trial comparing temsirolimus to interferon or a combination of both treatments has recently completed accrual and results are awaited.
References 743
predominantly comprised of minor responses). Progressionfree survival was significantly extended from a median of 3 months for placebo to 6 months for sorafenib. Overall survival data are pending. Common toxicities for these drugs include fatigue, gastrointestinal upset and skin rash. Randomized studies comparing them with interferon as first-line treatment for metastatic disease are underway.
CONCLUSION Kidney cancer currently accounts for 3 per cent of the total global cancer burden although its incidence is rising. Kidney cancer remains notoriously difficult to detect early and many patients present with advanced disease or will relapse following nephrectomy. Management of advanced disease remains multidisciplinary although the efficacy of conventional treatment options remains limited. Cytotoxic chemotherapy is largely ineffective and renal cancers are relatively radioresistant tumours. Response rates to conventional immunotherapy with IFN-α or IL-2 are low and treatment may be associated with significant toxicity. Considerable optimism however surrounds the development of the novel targeted agents, particularly the oral tyrosine kinase inhibitors which are demonstrating promising efficacy in patients with cytokine refractory disease. Their use has underlined the importance of understanding molecular and genetic abnormalities of RCC such as VHL gene inactivation that underpin the development and progression of the various subtypes of renal carcinoma. There has been a dramatic increase in global clinical trial activity and it now seems likely that our increased knowledge of the tumour biology will lead to substantial advances in systemic treatment of renal cell cancer over the coming years.
KEY LEARNING POINTS ●
●
●
●
Malignant kidney tumours account for 3 per cent of all cancers; detection of asymptomatic, earlystage disease through the increasing use of crosssectional imaging has led to a significant rise in incidence over the last three decades. Nephrectomy is curative for many patients with early-stage kidney cancer and has been shown to improve survival in selected patients with metastatic disease. Immunotherapy with interferon alpha or interleukin-2 remains the mainstay of treatment for advanced disease in selected patients. An understanding of the genetic basis of hereditary and sporadic conventional renal cell carcinoma has contributed significantly to the development of effective, novel therapies such as the oral multi-targeted tyrosine kinase inhibitors.
REFERENCES 1 http://info.cancerresearchuk.org/cancerstats/ Oberling C, Riviere M, Haguenau F. Ultrastructure of the clear cells in renal carcinomas and its importance for the demonstration of their renal origin. Nature 1960; 186:402–3. 3 Storkel S, Eble JN, Adlakha K, et al. Classification of renal cell carcinoma: Workgroup No. 1. Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer 1997; 80(5):987–9. 4 Zambrano NR, Lubensky IA, Merino MJ, Linehan WM, Walther MM. Histopathology and molecular genetics of renal tumors toward unification of a classification system. J Urol 1999; 162(4):1246–58. 5 Delahunt B, Eble JN. Papillary renal cell carcinoma: a clinicopathologic and immunohistochemical study of 105 tumors. Mod Pathol 1997; 10(6):537–44. 6 Guinan PD, Vogelzang NJ, Fremgen AM, Chmiel JS, Sylvester JL, Sener SF, Imperato JP. Renal cell carcinoma: tumor size, stage and survival. Members of the Cancer Incidence and End Results Committee. J Urol 1995; 153 (3 Pt 2):901–3. 7 Mancilla-Jimenez R, Stanley RJ, Blath RA. Papillary renal cell carcinoma: a clinical, radiologic, and pathologic study of 34 cases. Cancer 1976; 38(6):2469–80. 8 Storkel S, Steart PV, Drenckhahn D, Thoenes W. The human chromophobe cell renal carcinoma: its probable relation to intercalated cells of the collecting duct. Virchows Arch B Cell Pathol Incl Mol Pathol 1989; 56(4):237–45. 9 Crotty TB, Farrow GM, Lieber MM. Chromophobe cell renal carcinoma: clinicopathological features of 50 cases. J Urol 1995; 154(3):964–7. 10 Kuroda N, Toi M, Hiroi M, Enzan H. Review of collecting duct carcinoma with focus on clinical and pathobiological aspects. Histol Histopathol 2002; 17(4):1329–34. 11 Swartz MA, Karth J, Schneider DT, Rodriguez R, Beckwith JB, Perlman EJ. Renal medullary carcinoma: clinical, pathologic, immunohistochemical, and genetic analysis with pathogenetic implications. Urology 2002; 60(6):1083–9. 12 Fleming S, Lewi HJ. Collecting duct carcinoma of the kidney. Histopathology 1986; 10(11):1131–41. 13 Kennedy SM, Merino MJ, Linehan WM, Roberts JR, Robertson CN, Neumann RD. Collecting duct carcinoma of the kidney. Hum Pathol 1990; 21(4):449–56. 14 Chao D, Zisman A, Pantuck AJ, et al. Collecting duct renal cell carcinoma: clinical study of a rare tumor. J Urol 2002; 167(1):71–4. 15 Joerger M, Koeberle D, Neumann HP, Gillessen S. Von Hippel–Lindau disease – a rare disease important to recognize. Onkologie 2005; 28(3):159–63. 16 Knudson AG, Jr. Genetics of human cancer. Annu Rev Genet 1986; 20:231–51. ●17 Latif F, Tory K, Gnarra J, et al. Identification of the von Hippel–Lindau disease tumor suppressor gene. Science 1993; 260(5112):1317–20. ●2
744 Renal cell cancer
18 Crossey PA, Foster K, Richards FM, et al. Molecular genetic investigations of the mechanism of tumorigenesis in von Hippel–Lindau disease: analysis of allele loss in VHL tumours. Hum Genet 1994; 93(1):53–8. 19 Gnarra JR, Tory K, Weng Y, et al. Mutations of the VHL tumour suppressor gene in renal carcinoma. Nat Genet 1994; 7(1):85–90. ◆20 Rini BI, Small EJ. Biology and clinical development of vascular endothelial growth factor-targeted therapy in renal cell carcinoma. J Clin Oncol 2005; 23(5): 1028–43. 21 Zbar B, Tory K, Merino M, et al. Hereditary papillary renal cell carcinoma. J Urol 1994; 151(3):561–6. 22 Schmidt L, Duh FM, Chen F, et al. Germline and somatic mutations in the tyrosine kinase domain of the MET protooncogene in papillary renal carcinomas. Nat Genet 1997; 16(1):68–73. 23 Nickerson ML, Warren MB, Toro JR, et al. Mutations in a novel gene lead to kidney tumors, lung wall defects, and benign tumors of the hair follicle in patients with the Birt–Hogg–Dube syndrome. Cancer Cell 2002; 2(2): 157–64. 24 Moore LE, Wilson RT, Campleman SL. Lifestyle factors, exposures, genetic susceptibility, and renal cell cancer risk: a review. Cancer Invest 2005; 23(3):240–55. 25 Hunt JD, van der Hel OL, McMillan GP, Boffetta P, Brennan P. Renal cell carcinoma in relation to cigarette smoking: meta-analysis of 24 studies. Int J Cancer 2005; 114(1):101–8. 26 Pischon T, Lahmann PH, Boeing H, et al. Body size and risk of renal cell carcinoma in the European Prospective Investigation into Cancer and Nutrition (EPIC). Int J Cancer 2005; 118(3):728–738. 27 Chow WH, Gridley G, Fraumeni JF, Jr., Jarvholm B. Obesity, hypertension, and the risk of kidney cancer in men. N Engl J Med 2000; 343(18):1305–11. 28 Matson MA, Cohen EP. Acquired cystic kidney disease: occurrence, prevalence, and renal cancers. Medicine (Baltimore) 1990; 69(4):217–26. 29 Mandel JS, McLaughlin JK, Schlehofer B, et al. International renal-cell cancer study. IV. Occupation. Int J Cancer 1995; 61(5):601–5. 30 Thompson IM, Peek M. Improvement in survival of patients with renal cell carcinoma – the role of the serendipitously detected tumor. J Urol 1988; 140(3):487–90. 31 Luciani LG, Cestari R, Tallarigo C. Incidental renal cell carcinoma – age and stage characterization and clinical implications: study of 1092 patients (1982–1997). Urology 2000; 56(1):58–62. 32 Tsui KH, Shvarts O, Smith RB, Figlin RA, deKernion JB, Belldegrun A. Prognostic indicators for renal cell carcinoma: a multivariate analysis of 643 patients using the revised 1997 TNM staging criteria. J Urol 2000; 163(4):1090–5; quiz 1295. 33 Motzer RJ, Bacik J, Murphy BA, Russo P, Mazumdar M. Interferon-alfa as a comparative treatment for clinical trials of new therapies against advanced renal cell carcinoma. J Clin Oncol 2002; 20(1):289–96.
34 Kim HL, Belldegrun AS, Freitas DG, Bui MH, Han KR, Dorey FJ, Figlin RA. Paraneoplastic signs and symptoms of renal cell carcinoma: implications for prognosis. J Urol 2003; 170(5):1742–6. 35 Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov) SEER*Stat Database: April 2005. 36 International Union against Cancer TNM Classification of Malignant Tumours, 6th edn. Hoboken, New Jersey: John Wiley & Sons, 2002. 37 American Joint Committee on Cancer Cancer Staging Manual, 6th edn. New York, NY: Springer, 2002. 38 Koga S, Tsuda S, Nishikido M, Ogawa Y, Hayashi K, Hayashi T, Kanetake H. The diagnostic value of bone scan in patients with renal cell carcinoma. J Urol 2001; 166(6):2126–8. 39 Staudenherz A, Steiner B, Puig S, Kainberger F, Leitha T. Is there a diagnostic role for bone scanning of patients with a high pretest probability for metastatic renal cell carcinoma? Cancer 1999; 85(1):153–5. 40 Wunderlich H, Schlichter A, Reichelt O, Zermann DH, Janitzky V, Kosmehl H, Schubert J. Real indications for adrenalectomy in renal cell carcinoma. Eur Urol 1999; 35(4):272–6. 41 Schafhauser W, Ebert A, Brod J, Petsch S, Schrott KM. Lymph node involvement in renal cell carcinoma and survival chance by systematic lymphadenectomy. Anticancer Res 1999; 19(2C):1573–8. 42 Phillips E, Messing EM. Role of lymphadenectomy in the treatment of renal cell carcinoma. Urology 1993; 41(1):9–15. 43 Munro NP, Woodhams S, Nawrocki JD, Fletcher MS, Thomas PJ. The role of transarterial embolization in the treatment of renal cell carcinoma. BJU Int 2003; 92(3):240–4. 44 Rabinovitch RA, Zelefsky MJ, Gaynor JJ, Fuks Z. Patterns of failure following surgical resection of renal cell carcinoma: implications for adjuvant local and systemic therapy. J Clin Oncol 1994; 12(1):206–12. 45 Stephenson AJ, Chetner MP, Rourke K, et al. Guidelines for the surveillance of localized renal cell carcinoma based on the patterns of relapse after nephrectomy. J Urol 2004; 172(1):58–62. 46 Oliver RT, Nethersell AB, Bottomley JM. Unexplained spontaneous regression and alpha-interferon as treatment for metastatic renal carcinoma. Br J Urol 1989; 63(2):128–31. ●47 Flanigan RC, Salmon SE, Blumenstein BA, et al. Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N Engl J Med 2001; 345(23):1655–9. ●48 Mickisch GH, Garin A, van Poppel H, de Prijck L, Sylvester R. Radical nephrectomy plus interferon-alfa-based immunotherapy compared with interferon alfa alone in metastatic renal-cell carcinoma: a randomised trial. Lancet 2001; 358(9286):966–70. 49 Flanigan RC, Mickisch G, Sylvester R, Tangen C, Van Poppel H, Crawford ED. Cytoreductive nephrectomy in patients with metastatic renal cancer: a combined analysis. J Urol 2004; 171(3):1071–6. 50 Sandhu SS, Symes A, A’Hern R, Sohaib SA, Eisen T, Gore M, Christmas TJ. Surgical excision of isolated renal-bed
References 745
51
52
53
54
55
56
57
58
59
60
61
62
63
64 65 66
67
recurrence after radical nephrectomy for renal cell carcinoma. BJU Int 2005; 95(4):522–5. Schrodter S, Hakenberg OW, Manseck A, Leike S, Wirth MP. Outcome of surgical treatment of isolated local recurrence after radical nephrectomy for renal cell carcinoma. J Urol 2002; 167(4):1630–3. Master VA, Gottschalk AR, Kane C, Carroll PR. Management of isolated renal fossa recurrence following radical nephrectomy. J Urol 2005; 174(2):473–7; discussion 477. Piltz S, Meimarakis G, Wichmann MW, Hatz R, Schildberg FW, Fuerst H. Long-term results after pulmonary resection of renal cell carcinoma metastases. Ann Thorac Surg 2002; 73(4):1082–7. Fourquier P, Regnard JF, Rea S, Levi JF, Levasseur P. Lung metastases of renal cell carcinoma: results of surgical resection. Eur J Cardiothorac Surg 1997; 11(1):17–21. Cerfolio RJ, Allen MS, Deschamps C, Daly RC, Wallrichs SL, Trastek VF, Pairolero PC. Pulmonary resection of metastatic renal cell carcinoma. Ann Thorac Surg 1994; 57(2):339–44. Friedel G, Hurtgen M, Penzenstadler M, Kyriss T, Toomes H. Resection of pulmonary metastases from renal cell carcinoma. Anticancer Res 1999; 19(2C):1593–6. Fuchs B, Trousdale RT, Rock MG. Solitary bony metastasis from renal cell carcinoma: significance of surgical treatment. Clin Orthop Relat Res 2005 (431):187–92. Patchell RA, Tibbs PA, Regine WF, et al. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomised trial. Lancet 2005; 366(9486):643–8. O’Dea M J, Zincke H, Utz DC, Bernatz PE. The treatment of renal cell carcinoma with solitary metastasis. J Urol 1978; 120(5):540–2. Wilson D, Hiller L, Gray L, Grainger M, Stirling A, James N. The effect of biological effective dose on time to symptom progression in metastatic renal cell carcinoma. Clin Oncol (R Coll Radiol) 2003; 15(7):400–7. DiBiase SJ, Valicenti RK, Schultz D, Xie Y, Gomella LG, Corn BW. Palliative irradiation for focally symptomatic metastatic renal cell carcinoma: support for dose escalation based on a biological model. J Urol 1997; 158(3 Pt 1):746–9. Onufrey V, Mohiuddin M. Radiation therapy in the treatment of metastatic renal cell carcinoma. Int J Radiat Oncol Biol Phys 1985; 11(11):2007–9. Yagoda A, Petrylak D, Thompson S. Cytotoxic chemotherapy for advanced renal cell carcinoma. Urol Clin North Am 1993; 20(2):303–21. Amato RJ. Chemotherapy for renal cell carcinoma. Semin Oncol 2000; 27(2):177–86. Motzer RJ, Russo P. Systemic therapy for renal cell carcinoma. J Urol 2000; 163(2):408–17. Rini BI, Vogelzang NJ, Dumas MC, Wade JL, 3rd, Taber DA, Stadler WM. Phase II trial of weekly intravenous gemcitabine with continuous infusion fluorouracil in patients with metastatic renal cell cancer. J Clin Oncol 2000; 18(12):2419–26. Nanus DM, Garino A, Milowsky MI, Larkin M, Dutcher JP. Active chemotherapy for sarcomatoid and rapidly progressing renal cell carcinoma. Cancer 2004; 101(7):1545–51.
●68
Interferon-alpha and survival in metastatic renal carcinoma: early results of a randomised controlled trial. Medical Research Council Renal Cancer Collaborators. Lancet 1999; 353(9146):14–7. 69 Hancock B, Griffiths G, Ritchie A, et al. Updated results of the MRC randomised controlled trial of alpha interferon vs. MPA in patients with metastatic renal carcinoma. Proc Am Soc Clin Onc 2000; 1336. 70 Aass N, De Mulder PH, Mickisch GH, et al. Randomized phase II/III trial of interferon Alfa-2a with and without 13-cis-retinoic acid in patients with progressive metastatic renal cell carcinoma: the European Organisation for Research and Treatment of Cancer Genito-Urinary Tract Cancer Group (EORTC 30951). J Clin Oncol 2005; 23(18):4172–8. 71 Fyfe G, Fisher RI, Rosenberg SA, Sznol M, Parkinson DR, Louie AC. Results of treatment of 255 patients with metastatic renal cell carcinoma who received high-dose recombinant interleukin-2 therapy. J Clin Oncol 1995; 13(3):688–96. 72 Fisher RI, Rosenberg SA, Fyfe G. Long-term survival update for high-dose recombinant interleukin-2 in patients with renal cell carcinoma. Cancer J Sci Am 2000; 6 Suppl 1:S55–7. 73 Yang JC, Sherry RM, Steinberg SM, et al. Randomized study of high-dose and low-dose interleukin-2 in patients with metastatic renal cancer. J Clin Oncol 2003; 21(16):3127–32. ●74 Negrier S, Escudier B, Lasset C, et al. Recombinant human interleukin-2, recombinant human interferon alfa-2a, or both in metastatic renal-cell carcinoma. Groupe Francais d’Immunotherapie. N Engl J Med 1998; 338(18):1272–8. 74a Escudier B, Chevreau C, Lasset C, et al. Cytokines in metastatic renal cell carcinoma: is it useful to switch to interleukin-2 or interferon after failure of a first treatment? Groupe Francais d’Immunotherape. J Clin Oncol 1999; 17(7):2039–43. 75 Negrier S, Perol D, Ravaud A, et al. Do cytokines improve survival in patients with metastatic renal cell carcinoma (MRCC) of intermediate prognosis? Results of the prospective randomized PERCY Quattro trial. Proc Am Soc Clin Onc 2005; #4511. ●76 Pyrhonen S, Salminen E, Ruutu M, et al. Prospective randomized trial of interferon alfa-2a plus vinblastine versus vinblastine alone in patients with advanced renal cell cancer. J Clin Oncol 1999; 17(9):2859–67. ●77 Atzpodien J, Kirchner H, Illiger HJ, et al. IL-2 in combination with IFN- alpha and 5-FU versus tamoxifen in metastatic renal cell carcinoma: long-term results of a controlled randomized clinical trial. Br J Cancer 2001; 85(8):1130–6. 78 Atzpodien J, Kirchner H, Jonas U, et al. Interleukin2- and interferon alfa-2a-based immunochemotherapy in advanced renal cell carcinoma: a Prospectively Randomized Trial of the German Cooperative Renal Carcinoma Chemoimmunotherapy Group (DGCIN). J Clin Oncol 2004; 22(7):1188–94.
746 Renal cell cancer
79 Steineck G, Strander H, Carbin BE, et al. Recombinant leukocyte interferon alpha-2a and medroxyprogesterone in advanced renal cell carcinoma. A randomized trial. Acta Oncol 1990; 29(2):155–62. 80 Kriegmair M, Oberneder R, Hofstetter A. Interferon alfa and vinblastine versus medroxyprogesterone acetate in the treatment of metastatic renal cell carcinoma. Urology 1995; 45(5):758–62. ◆81 Coppin C, Porzsolt F, Awa A, Kumpf J, Coldman A, Wilt T. Immunotherapy for advanced renal cell cancer. Cochrane Database Syst Rev 2005; (1):CD001425. 82 Motzer RJ, Mazumdar M, Bacik J, Russo P, Berg WJ, Metz EM. Effect of cytokine therapy on survival for patients with advanced renal cell carcinoma. J Clin Oncol 2000; 18(9):1928–35. 83 Motzer RJ, Murphy BA, Bacik J, et al. Phase III trial of interferon alfa-2a with or without 13-cis-retinoic acid for patients with advanced renal cell carcinoma. J Clin Oncol 2000; 18(16):2972–80. ●84 Yang JC, Topalian SL, Parkinson D, et al. Randomized comparison of high-dose and low-dose intravenous interleukin-2 for the therapy of metastatic renal cell carcinoma: an interim report. J Clin Oncol 1994; 12(8):1572–6. 85 Fossa SD, Martinelli G, Otto U, et al. Recombinant interferon alfa-2a with or without vinblastine in metastatic renal cell carcinoma: results of a European multi-center phase III study. Ann Oncol 1992; 3(4):301–5. 86 Ravaud A, Audhuy B, Gomez F, et al. Subcutaneous interleukin-2, interferon alfa-2a, and continuous infusion of fluorouracil in metastatic renal cell carcinoma: a multicenter phase II trial. Groupe Francais d’Immunotherapie. J Clin Oncol 1998; 16(8):2728–32. 87 Negrier S, Caty A, Lesimple T, et al. Treatment of patients with metastatic renal carcinoma with a combination of subcutaneous interleukin-2 and interferon alfa with or without fluorouracil. Groupe Francais d’Immunotherapie, Federation Nationale des Centres de Lutte Contre le Cancer. J Clin Oncol 2000; 18(24):4009–15. 88 Ellerhorst JA, Sella A, Amato RJ, et al. Phase II trial of 5-fluorouracil, interferon-alpha and continuous infusion interleukin-2 for patients with metastatic renal cell carcinoma. Cancer 1997; 80(11):2128–32. 89 Tourani JM, Pfister C, Berdah JF, et al. Outpatient treatment with subcutaneous interleukin-2 and interferon alfa administration in combination with fluorouracil in patients with metastatic renal cell carcinoma: results of a sequential nonrandomized phase II study. Subcutaneous Administration Propeukin Program Cooperative Group. J Clin Oncol 1998; 16(7):2505–13. 90 Elias L, Lew D, Figlin RA, et al. Infusional interleukin-2 and 5-fluorouracil with subcutaneous interferon-alpha for the treatment of patients with advanced renal cell carcinoma: a South West Oncology Group Phase II study. Cancer 2000; 89(3):597–603. ●91 Childs R, Chernoff A, Contentin N, et al. Regression of metastatic renal-cell carcinoma after nonmyeloablative
92
93
●94
95
96
97
98
99
100
101
102
103
104
allogeneic peripheral-blood stem-cell transplantation. N Engl J Med 2000; 343(11):750–8. Motzer RJ, Mazumdar M, Bacik J, Berg W, Amsterdam A, Ferrara J. Survival and prognostic stratification of 670 patients with advanced renal cell carcinoma. J Clin Oncol 1999; 17(8):2530–40. Negrier S, Escudier B, Gomez F, et al. Prognostic factors of survival and rapid progression in 782 patients with metastatic renal carcinomas treated by cytokines: a report from the Groupe Francais d’Immunotherapie. Ann Oncol 2002; 13(9):1460–8. Yang JC, Haworth L, Sherry RM, et al. A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody, for metastatic renal cancer. N Engl J Med 2003; 349(5):427–34. Hainsworth JD, Sosman JA, Spigel DR, Edwards DL, Baughman C, Greco A. Treatment of metastatic renal cell carcinoma with a combination of bevacizumab and erlotinib. J Clin Oncol 2005; 23(31):7889–96. Motzer RJ, Amato R, Todd M, et al. Phase II trial of antiepidermal growth factor receptor antibody C225 in patients with advanced renal cell carcinoma. Invest New Drugs 2003; 21(1):99–101. Rowinsky EK, Schwartz GH, Gollob JA, et al. Safety, pharmacokinetics, and activity of ABX-EGF, a fully human anti-epidermal growth factor receptor monoclonal antibody in patients with metastatic renal cell cancer. J Clin Oncol 2004; 22(15):3003–15. Drucker B, Bacik J, Ginsberg M, Marion S, Russo P, Mazumdar M, Motzer R. Phase II trial of ZD1839 (IRESSA) in patients with advanced renal cell carcinoma. Invest New Drugs 2003; 21(3):341–5. Beeram M, Rowinsky E, Weiss G, et al. Durable disease stabilization and antitumor activity with OSI-774 in renal cell carcinoma: A phase II, pharmacokinetic (PK) and biological correlative study with FDG-PET imaging. Proc Am Soc Clin Oncol 2004; 22 (14S Suppl):3050. Varga Z, de Mulder P, Kruit W, et al. A prospective openlabel single-arm phase II study of chimeric monoclonal antibody cG250 in advanced renal cell carcinoma patients. Folia Biol (Praha) 2003; 49(2):74–7. Bleumer I, Knuth A, Oosterwijk E, et al. A phase II trial of chimeric monoclonal antibody G250 for advanced renal cell carcinoma patients. Br J Cancer 2004;90(5):985–90. Bleumer I, Oosterwijk E, Oosterwijk-Wakka JC, et al. A clinical trial with chimeric monoclonal antibody WX-G250 and low dose interleukin-2 pulsing scheme for advanced renal cell carcinoma. J Urol 2006; 175(1):57–62. Brouwers AH, Mulders PF, de Mulder PH, et al. Lack of efficacy of two consecutive treatments of radioimmunotherapy with 131I-cG250 in patients with metastasized clear cell renal cell carcinoma. J Clin Oncol 2005;23(27):6540–8. Kumar S, Witzig TE, Rajkumar SV. Thalidomid: current role in the treatment of non-plasma cell malignancies. J Clin Oncol 2004; 22(12):2477–88.
References 747
105 Atkins MB, Hidalgo M, Stadler WM, et al. Randomized phase II study of multiple dose levels of CCI-779, a novel mammalian target of rapamycin kinase inhibitor, in patients with advanced refractory renal cell carcinoma. J Clin Oncol 2004; 22(5):909–18. 106 Ratain M, Eisen T, Stadler W, et al. Final findings from a Phase II, placebo-controlled, randomized discontinuation trial (RDT) of sorafenib (BAY 43-9006) in patients with advanced renal cell carcinoma (RCC). Proc Am Soc Clin Oncol 2005; 4544. 107 Faivre S, Delbaldo C, Vera K, et al. Safety, pharmacokinetic, and antitumor activity of SU11248, a novel oral multitarget tyrosine kinase inhibitor, in patients with cancer. J Clin Oncol 2006; 24(1):25–35. ●108 Motzer RJ, Michaelson MD, Redman BG, et al. Activity of SU11248, a multitargeted inhibitor of vascular endothelial
growth factor receptor and platelet-derived growth factor receptor, in patients with metastatic renal cell carcinoma. J Clin Oncol 2006; 24(1):16–24. 109 Escudier B, Szczylik C, Eisen T, Stadler W, Schwartz B, Shan M, Bukowski R. Randomized phase III trial of the Raf kinase and VEGFR inhibitor sorafenib (BAY 43-9006) in patients with advanced renal cell carcinoma (RCC). Proc Am Soc Clin Oncol 2005; (4510). 110 Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92(3):205–16.
32 Ovary and fallopian tubes DANIELA D. ROSA, ANDREW R. CLAMP AND GORDON C. JAYSON
Epithelial tumours of the ovary Low malignant potential tumours of the ovary
748 756
EPITHELIAL TUMOURS OF THE OVARY Introduction The incidence of ovarian cancer varies across the world. It is highest in the USA, Europe and Israel, and lowest in Japan and in developing countries. Ovarian cancer is the fifth most common malignancy in North American women, with more than 23 000 new cases diagnosed and 14 000 deaths annually. Ovarian adenocarcinoma is a disease of peri- and post-menopausal women with peak incidence at age 60. Less than 1 per cent of epithelial ovarian cancers occur before the age of 20 and two thirds of these malignancies are germ cell tumours. Hereditary ovarian cancers occur about 10 years earlier than sporadic cancers. Although new chemotherapeutic agents have significantly improved the median survival rate, the overall mortality of ovarian cancer has remained unchanged. In the West, more women die from ovarian cancer than from all other gynaecological malignancies combined, largely due to the fact that it is often not diagnosed until late stage when symptoms develop due to invasion of other pelvic organs or from metastases.
Aetiology and risk factors The most significant risk factors for ovarian cancer are age and a family history of the disease.1 There is an increase of 12-fold in the risk of malignancy from ages 20–29 to 60–69. The odds ratio for ovarian cancer in relatives of ovarian cancer cases compared with controls is 3.6 for
Non-epithelial tumours of the ovary Cancer of the fallopian tube References
756 758 759
first-degree relatives and 2.9 for second-degree relatives. Women who carry deleterious mutations in BRCA1 or BRCA2 genes have a lifetime risk of developing ovarian cancer ranging from 7 per cent to 60 per cent.2 Late menopause may be associated with a slightly higher risk of ovarian cancer. Nulliparity is consistently associated with an increase in risk as is infertility. The increased risk of development of borderline and invasive epithelial ovarian tumours in women who received prolonged treatments with fertility drugs is controversial.3 Oral contraceptive use, pregnancy and lactation are associated with a reduced risk. These observations suggest that repeated stimulation of the ovarian surface epithelium, which occurs in the nulliparous state as a result of uninterrupted ovulation, may predispose the epithelium to malignant transformation. Evidence is growing that the use of menopausal hormone replacement therapy is associated with an increased risk of ovarian cancer, mainly in long-time users and users of sequential estrogen-progesterone schedules.4 The risk of developing ovarian cancer increases with cumulative oral oestrogen intake but not with duration of therapy, indicating that the risk may be diminished by minimizing the daily dose of oestrogen.5 Tobacco smoke, radiation exposure, use of psychotropic medication, the mumps virus, high-level physical activity, and dietary factors such as lactose or galactose and caffeine consumption may be associated with increased ovarian cancer risk. The use of talc in genital hygiene and risk of ovarian cancer has been controversial.6,7 There appears to be a relationship between increasing body mass index and ovarian cancer risk. Prior histories of pelvic inflammatory disease, polycystic ovary
Epithelial tumours of the ovary 749
syndrome and endometriosis have also been associated with an increased risk for ovarian cancer.8 However, most of these observations are based on retrospective studies and further research is needed before conclusions and recommendations can be made. GENETIC FACTORS
Only about 5–10 per cent of ovarian cancers are familial. The average lifetime risk for women is about 1 in 70. The lifetime risk for ovarian cancer for women with a family history is estimated at 9.4 per cent.9 There are three distinct hereditary patterns described: ovarian cancer alone, ovarian and breast cancers, or ovarian and colon cancers. A strong family history of breast and/or ovarian cancer may be related to an inherited mutation in one of two genes, BRCA1 and BRCA2, whose products are involved in DNA repair and transcriptional regulation in response to DNA damage.10 Women with a germ-line mutation in BRCA1 have a lifetime risk of ovarian cancer that ranges from 16 to 44 per cent and a lifetime risk of breast cancer of 56–87 per cent.11 Certain ethnic groups such as Ashkenazi Jews have an increased probability of carrying mutations in BRCA1 or BRCA2. The three most common mutations have a prevalence of 2.5 per cent in the healthy Ashkenazi population but are found in 26–41 per cent of Ashkenazi women with ovarian cancer.12 In women with BRCA1 mutations, ovarian cancer may develop at an earlier age than in those with the sporadic form of the disease. Women with mutations in BRCA2 have a 10 per cent lifetime risk of developing ovarian cancer.13 Frequently familial ovarian cancer is a more indolent disease than sporadic ovarian cancer.14 The other spectrum of inherited disease associated with ovarian cancer is the Lynch syndrome, which is caused by mutations in DNA repair genes hMLH1, hMSH2 and hMSH6. Mutations of these genes could account for 1–2 per cent of ovarian cancers. Women with Lynch syndrome have a 40–60 per cent lifetime risk of developing colon cancer, a 40–60 per cent lifetime risk of endometrial cancer and a 12 per cent lifetime risk of ovarian cancer.15 Cases of ovarian cancer occur at a markedly earlier age and are more likely to be well or moderately differentiated and be diagnosed at an early stage. MANAGEMENT OF WOMEN WITH A FAMILY HISTORY
Women with mutations in BRCA1 or BRCA2 who have completed childbearing may reduce the risk of developing ovarian cancer by undergoing bilateral salpingo-oophorectomy. However, the peritoneum appears to remain at risk, with the subsequent development of primary peritoneal carcinoma in about 1.8 per cent of woman who have undergone risk-reducing oophorectomy.16 Bilateral oophorectomy is also associated with reductions in breast cancer risk of 56 per cent for BRCA1 carriers and 46 per cent for BRCA2 carriers.
The benefit of using oral contraceptives to reduce the risk of ovarian cancer in mutation carriers is uncertain.17,18 MOLECULAR BIOLOGY
Multiple defects in pathways controlling the hallmarks of cancer19 have been implicated in the molecular pathogenesis of epithelial ovarian cancer. These include defects in cell cycle and apoptosis control, up-regulation of growth factor receptors such as EGF and HER-2 and creation of a proangiogenic environment. However, progress in identifying the key initiating defects in ovarian carcinogenesis has been hampered by the lack of a clear stepwise model of ovarian cancer progression. One major reason behind this is the difficulty in documenting early precursor lesions although it is likely that most ovarian cancers arise from the ovarian surface epithelium (OSE).20 It is becoming clear that the histological subtypes of ovarian cancer may arise via different molecular pathways. This concept is supported both by the identification of specific molecular changes preferentially in one diagnostic grouping, e.g. loss of PTEN expression in endometrioid carcinomas,21 and also by more recent work identifying distinct gene expression profiles for each subtype.22 p53/Rb mutations lead to the development of serous carcinoma whereas Ras/PTEN mutations were associated with endometrioid carcinomas in two transgenic mouse models.21,23 These two models, as well as contributing significantly to our understanding of ovarian cancer pathogenesis, may also play important roles in the pre-clinical development of novel targeted treatment strategies. The functions of the key regulators of DNA repair that are mutated in hereditary ovarian cancer are also altered in a high proportion of sporadic cases. For instance BRCA-1 and hMLH1 activities are lost in 70 per cent and 10 per cent of cases by promoter hypermethylation. This increased understanding of the molecular biology of ovarian cancer has provided us with a host of potential targets for novel therapies as well as murine models in which to test these.23,24 While early clinical trials of EGFR and HER-2 targeted therapy have shown modest levels of anti-cancer activity,25,26 much research needs to be done exploring combinations of such targeted therapies with conventional cytotoxics.
Natural history Ovarian cancer can spread via local shedding into the peritoneal cavity and via local invasion of bowel and bladder. It commonly involves the pelvic and para-aortic lymph nodes. Tumour cells may also block diaphragmatic lymphatics, resulting in impairment of lymphatic drainage of the peritoneum. Transdiaphragmatic spread to the pleura is common. The overall 5-year survival rate for ovarian cancer is less than 50 per cent.27
750 Ovary and fallopian tubes
Pathology More than 90 per cent of ovarian tumours arise from the OSE. A similar cell type lines the visceral and parietal peritoneum. Several different histological types of ovarianneoplasia are recognized (Table 32.1). These are not only histologically distinct but may exhibit different clinical behaviour, tumorigenesis and probably gene expression pattern.28
Staging system Definitive staging of ovarian cancer requires laparotomy. It is essential to examine and biopsy or to obtain cytological brushings of the diaphragm, both paracolic gutters, the pelvic peritoneum, para-aortic and pelvic nodes, and infracolic omentum, and to obtain peritoneal washings. In this staging operation, about one third of patients thought to have either stage I or II tumours will be found to have advanced-stage disease.31 Systematic pelvic and para-aortic lymphadenectomy has no impact in overall survival in women with optimally debulked advanced ovarian carcinoma.32 The Federation Internationale de Gynecologie et d’Obstetrique (FIGO) and the American Joint Committee on Cancer (AJCC) have designated staging as in Table 32.2.33
Diagnosis Ninety-five per cent of women with ovarian cancer experience some symptoms prior to diagnosis. Symptoms initially tend to be vague but may include abdominal distension or pain, alterations in bowel or bladder habits and gynaecological complaints such as pain on intercourse. A diagnosis of ovarian cancer should be considered in any pre-menopausal woman with an unexplained enlargement of the ovary or any post-menopausal woman with a palpable ovary. Ascites, pleural effusions and an umbilical mass referred to as a Sister Mary Joseph’s nodule may also be found on physical examination.34 Paraneoplastic syndromes may also be present, including hypercalcaemia, in the case of ovarian cancer with clear cell histology and subacute cerebellar degeneration associated with anti-Purkinje cell antibodies. The sudden appearance of seborrhoeic keratoses (Leser–Trélat sign) rarely has been reported to herald the development of ovarian cancer. Other paraneoplastic diseases that can be observed in ovarian cancer patients are migratory superficial thrombophlebitis (Trousseau’s syndrome), palmar fasciitis, dermatomyositis and polyarthritis. TUMOUR MARKERS
CA 125 is a membrane glycoprotein that is released into the plasma of approximately 75 per cent of patients with ovarian adenocarcinoma. CA 125 levels alone are neither sufficiently sensitive nor specific enough to be diagnostic
Table 32.1 Histological classification of ovarian neoplasms29,30 Epithelial tumours Serous cystomas Serous benign cystadenomas Serous cystadenomas with low malignant potential (borderline tumours)* Serous cystadenocarcinomas Mucinous cystomas Mucinous benign cystadenomas Mucinous cystadenomas with low malignant potential (borderline tumours)* Mucinous cystadenocarcinomas Endometrioid tumours (similar to adenocarcinomas in the endometrium) Endometrioid benign cysts Endometrioid tumours with low malignant potential (borderline tumours)* Endometrioid adenocarcinomas Clear cell (mesonephroid) tumours Benign clear cell tumours Clear cell tumours with low malignant potential (borderline tumours)* Clear cell cystadenocarcinomas Transitional cell carcinomas Mixed epithelial tumour (two components each making up at least 10 per cent of the tumour) Small cell carcinoma Germ cell tumours Dysgerminoma Other germ cell tumours Endodermal sinus tumour (yolk sac tumour) Embryonal carcinoma Polyembryoma Choriocarcinoma Teratoma Immature Mature solid Mature cystic Monodermal and highly specialized (struma ovarii, carcinoid, malignant neuroectodermal and ependymoma) Mixed forms Sex-cord stromal tumours Granulosa cell tumour (oestrogen-producing) Thecoma Sertoli–Leydig tumour Sertoli cell tumour Leydig cell tumour Arrhenoblastoma (virilizing) Fibroma Fibrosarcoma Sclerosing stromal tumour Metastasis to the ovaries *Histologically defined as possessing proliferating activity of the epithelial cells and nuclear abnormalities but with no infiltrative destructive growth.
Epithelial tumours of the ovary 751
Table 32.2
Ovarian cancer staging according to FIGO and AJCC33
Stage Description I IA IB IC** II
Tumour involving one or both ovaries with pelvic extension and/or implants IIA Extension and/or implants on the uterus and/or fallopian tubes. No malignant cells in ascites or peritoneal washings IIB Extension to and/or implants on other pelvic tissues. No malignant cells in ascites or peritoneal washings IIC** Pelvic extension and/or implants (stage IIA or stage IIB) with malignant cells in ascites or peritoneal washings
III
IIIA IIIB IIIC IV
Ovarian cancer is limited to the ovaries Tumour limited to one ovary; capsule intact, no tumour on ovarian surface. No malignant cells in ascites or peritoneal washings* Tumour limited to both ovaries; capsules intact, no tumour on ovarian surface. No malignant cells in ascites or peritoneal washings* Tumour limited to one or both ovaries with any of the following: capsule ruptured, tumour on ovarian surface, malignant cells in ascites or peritoneal washings*
Tumour involving one or both ovaries with microscopically confirmed peritoneal implants outside the pelvis. Superficial liver metastasis equals stage III. Tumour is limited to the true pelvis but with histologically verified malignant extension to small bowel or omentum Microscopic peritoneal metastasis beyond pelvis (no macroscopic tumour) Macroscopic peritoneal metastasis beyond pelvis 2 cm in greatest dimension Peritoneal metastasis beyond pelvis 2 cm in greatest dimension and/or regional lymph node metastasis Tumour involving one or both ovaries with distant metastasis. If pleural effusion is present, positive cytological test results must exist to designate a case to stage IV. Parenchymal liver metastasis equals stage IV
*The presence of ascites does not affect staging unless malignant cells are present. **Different criteria for allotting cases to stage IC and stage IIC have an impact on diagnosis. To assess this impact, it is important to know if rupture of the capsule was (1) spontaneous or (2) caused by the surgeon, and if the source of malignant cells detected was (1) peritoneal washings or (2) ascites.
of ovarian cancer35 as CA 125 may be elevated in other conditions including pregnancy, endometriosis, adenomyosis, uterine fibroids, pelvic inflammatory disease, menstruation and benign cysts, as well as in pancreatic, breast, lung, gastric and colon cancers. In the prediction of ovarian cancer, serum levels of CA 125 above 35 U/mL have a specificity of 82.0 per cent, a sensitivity of 78.3 per cent and an accuracy of 80.4 per cent; serum levels above 65 U/mL have a specificity of 92.5 per cent, a sensitivity of 71.7 per cent and an accuracy of 83.3 per cent. A serum CA 125 value greater than 65 U/mL can accurately predict ovarian cancer in 75 per cent of cases and, in general, can differentiate 83 per cent of benign from malignant ovarian tumours.36 A number of other candidate tumour markers for ovarian cancer have been identified, including CA 15.3, CA 72.4, CA 19.9 and soluble epidermal growth factor. However, there is little evidence to suggest that their use in a multi-modality screening panel is superior to CA 125 alone in differentiating benign from malignant ovarian tumours. At present, we have sufficient evidence to recommend CA 125 levels only as a component of a multi-modality approach to the diagnosis of ovarian cancer.37 IMAGING
Sonographic morphological indexing, based on tumour volume and wall structure, is a relatively accurate and costeffective method to predict risk of malignancy in an ovarian
tumour.38 A ten-point scale based on tumour volume and morphological complexity can designate a risk of malignancy of 0.3 per cent to tumours with an index less than 5 and of 84 per cent to tumours with an index of 8 or greater. A serum CA 125 assay may assist in eliminating false-positive results. The diagnostic accuracy of pelvic ultrasonography (US) when combined with a serum CA 125 value of 65 U/mL is 94 per cent (compared to 83 per cent when CA 125 was used alone).36 Although transvaginal ultrasonography (TVS) can detect smaller lesions and internal features of masses, it is not clear whether it should always be performed in preference to transabdominal US. The more limited field of view and scanning windows used in TVS may result in failure to identify abnormalities lying higher in the pelvis, particularly in patients with enlarged myomatous uteri.39 Doppler flow studies can be used to calculate the pulsatility (PI) and resistivity (RI) indices as markers of tumour vascularity. These indices differ between benign and malignant ovarian tumours although the values obtained may overlap, making it difficult to identify ovarian malignancies clearly on the basis of pulsed Doppler findings alone.40 Contrast-enhanced power Doppler may allow more accurate quantitation of vascularity41 but is time-consuming and requires sophisticated ultrasound equipment and sonographer skill. Computed tomography (CT) can help assess the extent of disease in patients before and after primary cytoreductive surgery. Magnetic resonance imaging (MRI) is better
752 Ovary and fallopian tubes
reserved for problem solving when ultrasonographic findings are non-diagnostic or equivocal because, although it is more accurate for diagnosis, it is also more expensive.42 PROTEOMICS
Proteomic profiling using mass spectroscopy has been proposed as a method to detect ovarian cancer using blood samples.43 The initial results of a proteomic pattern that could discriminate completely patients with ovarian cancers from normal controls or from patients with benign gynaecological disease44 have proven difficult to reproduce, and further work is required before proteomic profiling can provide reproducible data to identify ovarian cancer patients specifically and sensitively.45 MULTI-MODALITY APPROACH TO THE DIAGNOSIS OF OVARIAN CANCER
Determination of the degree of suspicion for malignancy in an adnexal mass is of critical importance, particularly as the incidence of benign masses is far higher than that of malignant masses. The algorithm for the investigation of an adnexal mass should be constructed in a way that allows follow-up without surgery in women with minimal risk for ovarian neoplasia and identifies women at intermediate or high risk so that appropriate surgery can be performed. In addition to a complete history and physical examination, US imaging is almost always indicated when the patient’s only finding is a suspected ovarian mass.46 Combining the clinical impression with the age of the patient and the findings of US imaging improves both the positive and negative predictive values of diagnosing an ovarian malignancy. Serum CA 125 is often included in the preoperative evaluation of patients with an adnexal mass.47 Serum tumour markers for malignant germ cell tumours should always be included in the preoperative evaluation of the young patient with a solid adnexal mass.46 Exploratory laparotomy is necessary in all cases of suspected ovarian cancer to help confirm the diagnosis, determine the extent of the disease by staging, and resect the tumour. It is well known that expedited referral of patients with suspect masses to a gynaecological oncologist for definitive staging with laparotomy correlates with better survival rates.48
to ovarian cancer, there is no conclusive evidence for routine screening.49 An initial screening abdominal or vaginal US in asymptomatic women can be abnormal in 4–20 per cent of cases, depending on the presence or absence of risk factors for ovarian cancer. A diagnosis of ovarian neoplasm is usually made in only 0.1–0.4 per cent of screened women.50 The sensitivity of CA 125 levels of greater than or equal to 35 U/mL for the detection of ovarian cancer ranged from 20 to 58 per cent in cases occurring within the first 3 years of follow-up; the specificity is around 95–99.9 per cent.51 Compared to the women diagnosed with ovarian cancer after the screening period, those diagnosed during CA 125 screening are more likely to have early-stage disease, but with no differences in survival.52 A risk calculation based on progressively rising serum CA 125 levels showed an increase in the sensitivity of ovarian cancer detection from 62 per cent to 86 per cent and deserves further study.53 The UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) aims to establish the impact of screening on ovarian cancer mortality through the recruitment of 200 000 postmenopausal women aged between 50 and 74 years.
Treatment If ovarian cancer is suspected on the basis of the physical examination and the results of imaging and laboratory studies, an exploratory laparotomy is usually performed for histological confirmation, staging and tumour debulking. Surgical staging provides important information that can guide postoperative decision making, especially for early-stage disease. The standard surgical cytoreduction involves a vertical midline incision to permit adequate exposure of the upper abdomen and pelvis. A total abdominal hysterectomy and bilateral salpingo-oophorectomy are typically performed, along with careful examination of all peritoneal surfaces, and omentectomy.34 Full staging surgery has been described under Staging system, above. STAGES I AND II OVARIAN CANCER
Cytoreductive surgery In selected patients who desire childbearing and have stage IA and grade 1 tumours, unilateral salpingo-oophorectomy may be associated with a low risk of recurrence.54 Adjuvant chemotherapy
Screening Despite considerable efforts directed at early detection, no cost-effective screening tests have been developed. Suggested screening tests for ovarian cancer include bimanual pelvic examination, TVS and CA 125 antigen. The Pap smear and cytological examination of peritoneal lavage obtained by culdocentesis have low sensitivities and are not recommended. Similarly, for women with no known genetic predisposition
The majority of patients with epithelial ovarian cancer will require postoperative adjuvant chemotherapy in an attempt to eradicate residual disease. Two large trials55,56 randomized patients with earlystage epithelial ovarian carcinoma to adjuvant platinumbased chemotherapy or observation (Table 32.3). When analysed together, these studies demonstrate significant improvements in recurrence-free (HR 0.64; 95 per cent CI, 0.50–0.82; p .001) and overall survival (HR 0.67;
Epithelial tumours of the ovary 753
Table 32.3 Summary of ACTION56 and ICON155 trials
Inclusion criteria Chemotherapy
Surgical staging Recurrence-free survival in the adjuvant chemotherapy arm Overall survival in the adjuvant chemotherapy arm
EORTC-ACTION trial
ICON1
Stage IA and IB with grade II or III; all grades stages IC–IIA; all clear cell carcinomas. 4 cycles of carboplatin or cisplatin-based chemotherapy Inadequate staging was not an exclusion criteria Hazard ratio: 0.63 (95% CI 0.44–0.92)
All stages were potentially eligible (most patients were stage I or stage II) 6 cycles of single-agent carboplatin or cisplatin or platinum-based chemotherapy (usually cyclophosphamide, doxorubicin and cisplatin) Did not monitor whether adequate surgical staging was performed Hazard ratio: 0.65 (95% CI 0.46–0.91)
Hazard ratio: 0.69 (95% CI 0.44–1.08)
Hazard ratio: 0.66 (95% CI 0.45–0.97)
95 per cent CI, 0.50–0.90; p .008) in the chemotherapy group.57 However, one trial showed that when full staging surgery had been performed the benefit of adjuvant platinum chemotherapy was lost.56 No patient subgroup particularly benefited from adjuvant chemotherapy.56 A meta-analysis of all randomized trials comparing adjuvant chemotherapy with observation in early-stage ovarian cancer showed a hazard ratio of 0.72 (95 per cent CI 0.55– 0.94, p .017) for overall survival in favour of chemotherapy with no evidence of heterogeneity between the trials.58 In conclusion, except for the most favourable subset (patients with stage IA grade 1 disease, who have a 5-year survival rate of 90–95 per cent after surgery alone) the evidence, based on double-blind, randomized controlled trials with total mortality endpoints, supports treatment with adjuvant platinum-based chemotherapy for early-stage ovarian cancer.57 STAGES III AND IV OVARIAN CANCER
The outcome of patients with stage IV disease is less favourable than that of patients with stage III disease. Usually, patients with both stages are treated with the same chemotherapeutic strategies; however, the role of surgery for patients with stage IV disease is still unclear. Primary cytoreductive surgery Considerable evidence indicates that the volume of disease left at the completion of the primary cytoreductive surgery is related to patient survival.59 Patients with optimal cytoreduction ( 1–2 cm of residual disease) have a median survival of 39 months compared with a survival of only 17 months for patients with suboptimal residual disease. As primary cytoreduction does not correct the tumoral biology, patients with large-volume disease who achieve smallvolume disease by surgical debulking have poorer outcomes than similar patients who present with smallvolume disease at diagnosis.60
Interval cytoreductive surgery Interval surgical cytoreduction after two to six cycles of chemotherapy in patients with residual lesions measuring more than 1 cm in diameter after primary surgery is a potential strategy for reducing tumour burden.61 Chemotherapy can reduce residual disease to 1 cm or less in 50–90 per cent of these patients. However the benefit of this approach is unclear as two phase III studies have shown divergent results.61,62 Interval debulking surgery is therefore not considered part of current best practice in patients who have undergone adequate primary surgery. Neoadjuvant approach Cytoreductive surgery achieves optimal cytoreduction in less than half the patients with advanced ovarian cancer. Neoadjuvant chemotherapy with delayed surgery has been proposed as an alternative approach in bulky ovarian cancer with the aim of reducing surgical morbidity and improving quality of life.63 Several retrospective and non-randomized prospective studies of neoadjuvant chemo-therapy have been reported. There is no conclusive evidence as yet for a survival advantage of this approach when compared with conventional treatment and its role therefore remains limited. The EORTC 55971 trial, a randomized phase III study comparing initial debulking surgery with neoadjuvant chemotherapy in patients with stage IIIC or IV epithelial ovarian cancer, is currently addressing this issue.64 Neoadjuvant chemotherapy may however be considered for patients where initial optimum cytoreductive surgery is not possible, for instance in the presence of disease around nerves and vessels or at the porta hepatis. Second-look laparotomy Second-look laparotomies were routinely performed after completion of chemotherapy, for stage III patients who had a CT scan not suggestive of residual active disease, who were clinically without evidence of disease and whose CA 125 was normal. Most studies failed to show any survival advantage in patients who received a second-look operation and
754 Ovary and fallopian tubes
therefore such an approach should only be performed within a clinical research study.65 Intraperitoneal treatments The use of intraperitoneal radioactive phosphate after negative second-look surgery does not appear to increase overall or disease-free survival rates.66 However, intraperitoneal chemotherapy has been shown to improve these parameters in selected patients. The pharmacological principle is to deliver more drug direct to the tumour by repeated ‘baths’ that strip away cancer cells layer by layer with each cycle of therapy.67 Early reports suggested a role for intraperitoneal chemotherapy by demonstrating surgically defined complete response rates and prolonged survival in approximately 25–35 per cent of patients with persistent small-volume residual disease after a variety of intraperitoneal regimens.68,69 Outcome was particularly favourable in patients defined as having platinum-sensitive disease, a feature also indicative of a greater overall responsiveness to intravenous chemotherapy. Until 2002, three phase III trials had documented superior overall survival or progression-free survival associated with the administration of cisplatin by the intraperitoneal route in patients with stage III, optimally resected disease.70–72 Intraperitoneal therapy, however, had not been routinely adopted, in part because of issues relating to greater toxicity and inconvenience. Recently, a randomized trial73 showed a statistically significant prolongation of median progression-free survival and overall survival in the intraperitoneal group (benefits of 5.5 and 15.9 months, respectively) after a median follow-up of 50 months. Only 42 per cent of patients completed the planned intraperitoneal therapy because of toxic effects, often catheter related (infection, blockage or leak). Despite this, intraperitoneal chemotherapy was established as an important choice in the first-line treatment of patients with optimally debulked stage III disease. A meta-analysis of 8 randomized studies evaluating 1819 patients receiving primary treatment for ovarian cancer74 showed a hazard ratio for overall survival of 0.79 (0.70–0.90) in favour of intraperitoneal chemotherapy. It should be stressed, however, that the potential for catheter-related complications and toxicity needs to be considered when deciding on the most appropriate treatment for each individual woman. Chemotherapy regimens The platinum analogues carboplatin and cisplatin are the most active agents in the treatment of ovarian cancer. In initial trials, a relationship between survival and dose-intensity of cisplatin was noted. However, escalation of cisplatin dose to 100 mg/m2 every 3 weeks was not associated with a survival advantage over 50 mg/m2 every 3 weeks and subsequently 75 mg/m2 every 3 weeks was adopted as the standard.75,76 For carboplatin, similar results were found77 with an improved outcome shown for a target area under the curve (AUC) of 5 and 6. It has also been well established that cisplatin and carboplatin yield equivalent results,78
whether administered as single agents or in combination regimens.79 The approach of administering lower dose intensity combination chemotherapy over a longer time has also been tested in the randomized setting.80 The chemotherapy was more effective when given at the higher dose intensity. Therefore, for first-line therapy it is advised to maintain standard dose intensities. The current gold standard for postoperative treatment of patients with advanced ovarian cancer is the intravenous administration of taxane- and platinum-based chemotherapy.79 Four studies have assessed the impact on progression-free survival and overall survival when paclitaxel was added to either cisplatin or carboplatin.81–84 Two showed that combination chemotherapy containing paclitaxel prolonged both progression-free and overall survival in patients with advanced disease as compared with regimens that did not contain taxanes,81 resulting in a 30 per cent reduction in the risk of death. The other two studies, however, showed no benefit with the addition of paclitaxel.83,84 More recently, the combination of paclitaxel and carboplatin was shown to be as effective as paclitaxel and cisplatin as first-line therapy, but with less emesis, leucopenia and nephropathy.79 Docetaxel may also be substituted for paclitaxel, with similar survival outcomes and less neurotoxicity, but at the expense of greater neutropenia.85 About half of the patients with advanced disease who receive chemotherapy with paclitaxel and carboplatin obtain a complete clinical remission.79 However, despite this, relapse still occurs in more than 50 per cent of these and only 10–30 per cent of such patients have long-term survival. The majority of responding patients relapse within 14–28 months from the initial platinum-based chemotherapy. A number of consolidation and/or maintenance strategies have been studied with little success. These include lengthening the number of cycles of the induction platinum-based chemotherapy, continuing paclitaxel after a complete clinical response to induction chemotherapy has been achieved, integrating other active drugs, consolidation with high-dose chemotherapy and consolidation with intraperitoneal chemotherapy following both negative or positive reassessment. Maintenance paclitaxel given for 12 cycles, when compared to only 3 cycles, resulted in significantly prolonged disease-free survival (p 0.0023), despite a higher toxicity.86 Differences in overall survival could not be found, as the study was prematurely discontinued. Therefore, a maintenance approach has not been widely adopted. RECURRENT OVARIAN CANCER
Since recurrent disease is generally not curable, palliation of symptoms and prevention of complications are the goals of management. A common sign of relapse is a rise in the serum CA 125 level in the absence of both symptoms and abnormalities on physical examination or CT scanning.87 Frequently, the development of disease-related symptoms
Epithelial tumours of the ovary 755
is delayed (by a median of about 2 months) behind a doubling of CA 125 levels, and currently it is not clear whether commencing treatment solely on the basis of rising CA 125 levels rather than waiting for symptoms impacts on longterm outcome. This is currently being addressed in the Medical Research Council OVO-5 study.88 The choice of second-line cytotoxic agents generally depends on the treatment-free interval. Patients who have a relapse more than 5–6 months after the completion of first-line therapy are considered to have potentially platinum-sensitive disease, with response rates of 30 per cent or greater to platinum-based regimens.89 Patients with a short remission, lasting less than 6 months after first-line chemotherapy, usually (but not always) have platinumresistant disease. Platinum-sensitive disease, minimal symptoms and a small tumour burden Retreatment with platinum should be considered. The ICON4 and GEICO studies89 showed improved overall survival and progression-free survival with the addition of paclitaxel to platinum therapy. Single-agent chemotherapy with carboplatin however is still a reasonable treatment option that is often well tolerated, without appreciable alopecia. Other platinum agents, such as oxaliplatin, have activity and could be considered. In patients with severe symptoms and rapidly progressive disease, the combination of platinum and taxanes is probably the most appropriate treatment option.90,91 Relapse more than 6–12 months after first remission Tumour debulking (secondary cytoreduction) is sometimes considered before chemotherapy if the tumour is technically resectable, although this approach has never been tested in a randomized trial.92 Platinum-resistant disease When disease recurs within 6 months of previous platinumbased treatment or progresses on treatment, there is no standard second-line therapy although several other drugs have shown low levels of activity as single agents in this population. Paclitaxel should be considered if not previously given.93 Because of the chemo-resistant nature of the tumours, patients with platinum-refractory ovarian cancer who have not responded to paclitaxel may be candidates for phase I and II clinical trials evaluating novel therapies. Other agents that can be considered for patients with platinum-resistant disease include liposomal doxorubicin, topotecan, gemcitabine, oral etoposide, fluorouracil and leucovorin, ifosfamide, capecitabine and vinorelbine with response rates in the 10–20 per cent range. As no drug has been shown to be more effective than any other in platinum-resistant disease, the regimen chosen should take into account any co-morbidities and potential adverse impact on the patient’s quality of life. Several studies have investigated the use of combination chemotherapy in patients with platinum-resistant disease94,95
with response rates in the range of 40 per cent. However, such response rates have not yet been translated into improved overall survival. Additional clinical trials are required to identify more effective options for patients with relapsed disease. About 20 per cent of patients will respond to tamoxifen (20 mg twice daily), mainly when there are detectable levels of cytoplasmic oestrogen receptor on their tumours,96 and this treatment option can be considered in patients who are unlikely to tolerate well further chemotherapy.
Follow-up The ideal follow-up protocol for asymptomatic women who have completed primary debulking surgery and chemotherapy is unclear. However, follow-up should include routine complete history, physical examination including rectovaginal pelvic examination and measurement of serum CA 125 levels.49 The serum CA 125 level is valuable in patients who have elevated levels at the time of diagnosis.97 After complete surgical removal of all ovarian cancer, the half-life of CA 125 is about 6 days and several studies have demonstrated that those patients with a CA 125 half-life of 20 days have a worse prognosis.98 While an elevated or rising CA 125 level indicates a high probability of the presence of epithelial ovarian cancer, a negative CA 125 level cannot be used to exclude the presence of residual disease or relapse. The sensitivity and specificity of CA 125 in detection of disease recurrence is 62 per cent and 93 per cent, respectively.99 The usefulness of the area under the CA 125 concentration-time curve (AUC) as a new kinetic parameter for predicting overall survival in patients with ovarian cancer has also been evaluated.100 There was an association of higher CA 125 AUC with shorter survival and worse tumoral response, and the accuracy in predicting complete response to chemotherapy was 82 per cent. Further studies are required however before these results can be extrapolated into standard clinical practice. A combination of CA 125 and general physical and pelvic examination may detect progression of disease in 90 per cent of patients. Routine radiological investigations do not improve the detection of recurrence and their use should be individualized.49 Although optimal intervals for monitoring have not been determined, current practice is to follow the patient every 3–4 months. After 2 years, less frequent follow-up intervals can be considered49 with visits every 6 months until the fifth year, after which annual visits are recommended.101
Prognostic factors and survival The most important favourable prognostic factors at presentation in ovarian cancer patients include earlier disease stage, smaller disease volume prior to any surgical debulking, smaller residual tumour following primary cytoreductive surgery, good performance status, absence of ascites, a
756 Ovary and fallopian tubes
younger age, histological subtype other than mucinous/clear cell carcinoma and a well-differentiated tumour.102 Grade is the most important prognostic factor for patients with stage I ovarian cancer, followed by dense adherence and large-volume ascites.103 Patients with clear cell histology appear to have a worse prognosis whereas tumours with a significant component of transitional cell carcinoma appear to have a better prognosis. CA 125 levels also seem to be an independent prognostic factor for early-stage ovarian cancer, although the importance of this is weaker than tumoral grade and adequate surgical staging.104 The yearly mortality in ovarian cancer is approximately 65 per cent of the incidence rate, mainly because the disease is frequently diagnosed in an advanced stage. The 5-year survival rate is 76–89 per cent for stage I, 42–59 per cent for stage II, and 10–20 per cent for stage III disease. The 5-year survival rate for suboptimally debulked stage III and stage IV patients treated with platinum-based chemotherapy is 10 per cent102 although the median survival in patients with stage III disease that has been optimally debulked is 52–63 months.71
LOW MALIGNANT POTENTIAL TUMOURS OF THE OVARY Introduction Tumours of low malignant potential (borderline tumours) are not benign because they show cytological and architectural atypia. They are not malignant because there is no stromal invasion. They account for 15 per cent of all epithelial ovarian tumours and often develop at an earlier age than invasive cancer. About 75 per cent of these tumours are stage I at the time of diagnosis. Their prognosis and treatment are markedly different from those of carcinomas. The 20-year survival rate is around 90 per cent for all cases but disease-related mortality depends on several factors. According to stage, mortality ranges from 0.7 per cent for stage I to 26.8 per cent for stage III disease. Serous histology and younger age are associated with a better prognosis.105
demonstrated to be of benefit, although the contralateral ovary should be carefully examined. In general, total abdominal hysterectomy and bilateral salpingo-oophorectomy is appropriate therapy in patients who have completed their families while unilateral salpingo-oophorectomy or a partial oophorectomy can be employed to preserve fertility. Once these patients have completed their families, consideration should be given to removal of the remaining ovarian tissue due to the documented risk of disease recurrence.107 No additional treatment is indicated. Patients with localized intraperitoneal disease and negative lymph nodes have around a 5 per cent chance of developing recurrence, whereas those with localized intraperitoneal disease and positive lymph nodes have recurrence rates of 50 per cent.108 Patients with newly diagnosed stage III disease should undergo aggressive tumour debulking surgery as observational studies have indicated that when there is no gross residual tumour the survival rate can be as high as 100 per cent. Patients with bulk residual disease after surgery have a better prognosis than those with stage III epithelial ovarian cancer with a reported 7-year survival in excess of 60 per cent.109 There is no randomized evidence to support the use of postoperative chemotherapy110 and/or radiation therapy and as disease relapse may occur many years after initial diagnosis the most appropriate management approach is currently careful observation with consideration for secondary cytoreductive surgery in the face of tumour progression. The role of chemotherapy is not well defined but may be beneficial in patients whose disease has recurred rapidly or in whom further surgery is not possible.
NON-EPITHELIAL TUMOURS OF THE OVARY Germ cell tumours Germ cell tumours of the ovary are aggressive tumours, occurring more frequently in young women or adolescent girls. They are frequently unilateral, and are generally curable if treated early.111
Pathology and staging Pathology Refer to Staging system. If peritoneal implants are present, it is important to make the distinction between invasive and non-invasive implants as the presence of invasion adversely affects prognosis.106 The transformation rate from borderline tumours to cancer is about 1–2 per cent. Staging follows that used for carcinomas (see Diagnosis).
Treatment For stages I and II, overall survival after resection is close to 100 per cent. If the diagnosis is suspected or confirmed at surgery by frozen section, full staging surgery has not been
Refer to Staging system.
Staging Surgical staging of germ cell tumours of the ovary follows the same principles as for epithelial ovarian cancer (see Table 32.2). While unilateral oophorectomy is appropriate fertility-sparing surgery in a young woman, the contralateral ovary should always be examined and biopsied if suspicious. As in epithelial ovarian cancer, ascitic fluid or peritoneal washings should be examined cytologically.
Non-epithelial tumours of the ovary 757
As well as being of potential diagnostic importance, an assessment of the pre-surgical serum levels of alpha fetoprotein (AFP) and human chorionic gonadotropin (hCG) is useful as the persistent elevation of these markers after surgery indicates unresected tumour.
Treatment All patients except those with stage I, grade I immature teratoma and stage IA dysgerminoma require postoperative chemotherapy.112 DYSGERMINOMAS (SEE ALSO CHAPTER 30)
Unilateral salpingo-oophorectomy conserving the uterus and opposite ovary is accepted treatment of patients with stage I disease and may be considered for those with stages II–IV disease who wish to preserve fertility or a pregnancy. Postoperative radiological staging investigations should be performed before treatment decisions are made for patients who have not had careful surgical and pathological examination of pelvic and para-aortic lymph nodes during surgery. In stages II–IV disease, total abdominal hysterectomy and bilateral salpingo-oophorectomy are usually performed.113 Patients with stage IA tumours may be observed carefully after surgery without adjuvant treatment. Although 15–25 per cent will develop recurrent disease, this can be treated successfully at the time of recurrence with a high likelihood of cure. Incompletely staged patients or stage IB/IC should receive adjuvant treatment. Options include radiation therapy (which carries a risk of ovarian failure) or chemotherapy. Although experience with adjuvant chemotherapy is limited, its use can be justified by extrapolation from cases with more advanced disease.113 For patients with stages II–IV disease, three to four courses of a cisplatin-containing combination such as BEP (bleomycin, etoposide, cisplatin) are recommended and most patients are cured with this approach.114 If patients managed conservatively experience disease recurrence, cisplatin-based chemotherapy is effective.115 If disease relapses after first-line chemotherapy, options include abdominal radiotherapy, alternative chemotherapy regimens or consideration for high-dose therapy. The management of germ cell tumours is more fully covered in Chapter 30. OTHER GERM CELL TUMOURS
Surgical principles are similar to those described for dysgerminomas. For stage III patients with extensive intraabdominal disease whose clinical condition precludes debulking surgery, chemotherapy can be considered prior to surgery.112 Stage I tumours other than low-grade (grade I) immature teratoma are usually managed with adjuvant platinum-based chemotherapy due to a high probability of
disease relapse after surgery alone, although it is possible that an initial surveillance strategy with the use of chemotherapy at disease recurrence may be an acceptable alternative. Patients with stage II–IV disease should receive chemotherapy with three or four cycles of BEP.112 In the rare patient who does not respond to BEP, a durable remission may still be obtained with VAC (vincristine, dactinomycin, cyclophosphamide) as salvage therapy.115 All patients who do not respond to standard therapy are candidates for clinical trials. Second-look laparotomy after adjuvant chemotherapy seems to be of no benefit in patients with initially completely resected tumours. However, it may play a role in the minority of patients whose primary tumours were initially incompletely resected or contained teratomatous elements,116 since reversion to germ cell tumour has been described. In cases of recurrent tumours, treatment with chemotherapy is indicated, the type of which is determined by previous treatment.117 While cisplatin-based combination chemotherapy is effective in this setting, radiation therapy is not. Patients who do not respond to a cisplatinbased combination may still attain a durable remission with VAC or ifosfamide/cisplatin as salvage therapy.115 Ifosfamide combinations or high-dose chemotherapy and autologous marrow rescue are being investigated as potential new treatments for recurrent disease. Secondary cytoreductive surgery may have some benefit for those patients with immature teratoma. After maximal effort for surgical cytoreduction, chemotherapy should be considered. PROGNOSIS
For dysgerminoma confined to the ovary, less than 10 cm in size, with an intact capsule, unattached to other organs and without ascites, the 10-year survival following conservative surgery is about 90 per cent.113 With multi-modality therapy, overall survival for all stages of dysgerminoma is about 85 per cent at 5 years. Mature teratoma has good long-term survival; survival for immature teratoma treated by resection alone is closely related to the grade of the tumour with recurrence rates reported in historical series of 18 per cent, 37 per cent and 70 per cent for grade 1, 2 and 3 disease respectively.118 Other important prognostic factors include disease stage and extent of tumour at diagnosis. With current chemotherapy regimens, up to 97 per cent of patients are in disease remission at 10–54 months. Endodermal sinus tumours of the ovary are more aggressive. About 50 per cent of patients die within a year of diagnosis and only about 30 per cent of patients with stage I disease are alive at 2 years without chemotherapy.119 For malignant mixed germ cell tumours, size and histology are the main prognostic factors. When more than one third of the tumour is composed of endodermal sinus elements, choriocarcinoma or grade 3 immature teratoma, the prognosis is poor. When the tumour is less than 10 cm
758 Ovary and fallopian tubes
in diameter, the prognosis is good regardless of the composition of the tumour.120
Sex cord stromal tumours Sex cord stromal tumours of the ovary account for 5–10 per cent of all ovarian tumours and their clinical presentation is often characterized by their ability to secrete active sex hormones.121 While granulosa cell tumours (GCT) account for up to 70 per cent of these tumours, several other well-defined histological subtypes are recognized (see Table 32.1). Oestrogen secretion is noted in at least 70 per cent of GCTs and as the median age of diagnosis is in the early 50s, patients often present with post-menopausal bleeding or irregular menses. In such patients it is important to exclude a synchronous endometrial adenocarcinoma caused by oestrogenic hyperstimulation. Precocious puberty may be the presenting complaint in the case of juvenile GCTs. Patients with Sertoli and Leydig cell tumours often show features of virilization. These tumours may also present only when they have reached a large size with symptoms related to tumour bulk such as abdominal pain and distension. Due to their vascular nature, emergency presentation with an acute abdomen due to intraperitoneal bleeding is also not uncommon. Sex cord stromal tumours are notably associated with two hereditary malignant predisposition syndromes, Peutz–Jeghers syndrome122 and Gorlin syndrome.123 Of note, solitary ovarian fibromas may present due to the development of ascites and pleural effusion, a triad of signs known as Meig syndrome. Most sex cord stromal tumours behave in a benign fashion and present with early-stage disease. Surgical resection of the primary tumour is generally associated with high long-term survival rates and due to the rarity of most subtypes apart from GCTs, there is little published evidence to guide the management of cases with advanced or recurrent disease. MANAGEMENT OF GRANULOSA CELL TUMOURS
The majority of patients with GCTs present with stage I disease and these cases are associated with good long-term survival. It is reasonable to offer unilateral salpingooophorectomy to those patients who wish to preserve their fertility. The most important prognostic factor for GCTs is the stage of disease at clinical presentation, with around 90 per cent of stage I patients alive at 10 years compared to 17–33 per cent of patients with stage III/IV disease.124 However, up to 40 per cent of patients with stage I disease have been reported to experience disease relapse at a median of 76 months after presentation, and relapses more than 20 years after initial diagnosis have been reported. The use of adjuvant radiotherapy125 or chemotherapy in patients with advanced-stage disease or other poor prognostic features has been proposed by some experts on the
basis of anti-tumour activity in recurrent disease but no randomized studies have been conducted and not all retrospective case series support this approach. Because of the indolent nature of this condition, longterm follow-up is required. Disease relapse often occurs intra-abdominally but distant metastases in lung, liver and bone have been reported.124 In patients with recurrent disease, surgical debulking should be considered. In the presence of inoperable disease, platinum-based chemotherapy should be given. The BEP regimen shows a response rate of 64 per cent, with several responses of more than 2 years’ duration.126 Single-agent platinum may be appropriate for selected cases. There is also some evidence, albeit conflicting, to support the use of hormonal manipulation strategies in recurrent GCTs.127
CANCER OF THE FALLOPIAN TUBE Introduction Primary fallopian tube carcinoma represents about 1% of all gynaecological malignancies. It is a treatable disease but cure can be only achieved in patients with early tumours. Principles of management follow those of epithelial ovarian cancer. The presence of a second primary tumour may be found in about 20% of patients.128 It seems that fallopian tube carcinoma occurs at a higher frequency in carriers of BRCA mutations.129
Diagnosis and staging The disease is rarely diagnosed preoperatively and is often mistaken for benign pelvic disease or ovarian cancer. Often it presents at an early stage, probably due to abdominal pain resulting from tubal dilatation.130 However, it seems to have a worse prognosis than epithelial ovarian cancer despite the stage of presentation. Abnormal vaginal bleeding, abdominal pain and vaginal discharge are the most common presenting symptoms at the time of diagnosis. TVS may show normal ovaries in association with a discrete solid adnexal mass.131 Serum CA 125 levels may be elevated. The staging system is based on the ovarian carcinoma classification.132
Treatment The surgical management principles follow those adopted for ovarian carcinoma. Early lymphatic spread is recognized as a frequent cause of treatment failure and radical lymphadenectomy may prolong survival.133 Patients with apparently early-stage low-risk fallopian tube carcinoma not submitted to complete surgical staging, as well
References 759
as those with early-stage high-risk disease, should receive adjuvant single-agent carboplatin. Patients with advanced disease should undergo carboplatin-based chemotherapy with or without paclitaxel.134 Platinum-based chemotherapy has been shown to produce response rates of 87 per cent in patients with advanced-stage fallopian tube cancer. Second-line treatment for persistent or recurrent disease should be mainly based on the platinum-free interval, whereas secondary cytoreduction should be considered only for highly selected patients with localized, late relapse.134
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
●
●
Overall mortality of ovarian cancer is high although median survival has improved. No cost-effective screening tests have been developed and randomized trials are ongoing. Multiple defects in cell cycle and apoptosis control, DNA repair mechanisms, up-regulation of growth factor receptors and creation of a proangiogenic environment have been implicated in the molecular pathogenesis of epithelial ovarian cancer. Cytoreductive surgery is a key initial component in the management of patients with ovarian cancer and optimal surgical debulking is an important prognostic factor. Except for patients with stage IA grade 1 ovarian cancer, randomized controlled trial evidence supports treatment with adjuvant platinum-based chemotherapy for early-stage disease (stage I/II). Adjuvant intraperitoneal chemotherapy can increase overall survival of patients with optimally debulked ovarian cancer, but is associated with a high rate of catheter-related complications and toxicities. Platinum–taxane combinations or single-agent platinum regimens should be considered as standard first-line chemotherapy options for advanced epithelial carcinoma (stage III/IV). The choice of second-line chemotherapy for recurrent ovarian cancer should be based on the duration of response to first-line chemotherapy, and the patient’s symptoms and co-morbidities. Ovarian borderline tumours commonly present with early-stage disease and are curable by surgical resection. Germ cell tumours should be considered in young women who present with an ovarian mass and are often cured using a multi-modality approach. The principles of management of fallopian tube and primary peritoneal carcinoma follow those of epithelial ovarian cancer.
Prognosis Because of the low incidence, information regarding factors influencing clinical outcome is limited. Surgical stage I disease and the absence of residual tumour after operation seem to be the main factors associated with improved survival.128 Presence of extra-tubal disease at initial surgery is a poor prognostic factor. About 55 per cent of patients present with stage I disease and survival is around 60 per cent at 5 years. Survival at 5 years is 27 per cent for stage II, 14 per cent for stage III and 0 per cent for stage IV disease.135 Within stage I cases, the presence or absence of invasion of the tubal wall, the depth of invasion when present, and the location of the tumour within the tube (fimbrial or non-fimbrial) appear to have prognostic importance.136 Usually no patient with residual disease is alive at 5 years, stressing the need for radical surgery with maximum debulking at diagnosis.
REFERENCES 1 Holschneider CH, Berek JS. Ovarian cancer: epidemiology, biology, and prognostic factors. Semin Surg Oncol 2000; 19(1):3–10. 2 Struewing JP, Hartge P, Wacholder S, et al. The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N Engl J Med 1997; 336(20):1401–8. 3 Whittemore AS, Harris R, Itnyre J. Characteristics relating to ovarian cancer risk: collaborative analysis of 12 US casecontrol studies. II. Invasive epithelial ovarian cancers in white women. Collaborative Ovarian Cancer Group. Am J Epidemiol 1992; 136(10):1184–203. 4 Riman T, Dickman PW, Nilsson S, et al. Hormone replacement therapy and the risk of invasive epithelial ovarian cancer in Swedish women. J Natl Cancer Inst 2002; 94:497–504. 5 Glud E, Kjaer SK, Thomsen BL, et al. Hormone therapy and the impact of estrogen intake on the risk of ovarian cancer. Arch Intern Med 2004; 164(20):2253–9. 6 Cramer DW, Liberman RF, Titus-Ernstoff L, et al. Genital talc exposure and risk of ovarian cancer. Int J Cancer 1999; 81:351–6. 7 Gertig DM, Hunter DJ, Cramer DW, et al. Prospective study of talc use and ovarian cancer. J Natl Cancer Inst 2000; 92(3):249–52. 8 Brinton LA, Gridley G, Persson I, et al. Cancer risk after a hospital discharge diagnosis of endometriosis. Am J Obstet Gynecol 1997; 176:572–9. 9 Hartge P, Whittemore AS, Itnyre J. Rates and risks of ovarian cancer in subgroups of white women in the United States. Obstet Gynecol 1994; 84:760–4. 10 King MC, Marks JH, Mandell JB. Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science 2003; 302:643–6. 11 Ford D, Easton DF, Bishop DT, Narod SA, Goldgar DE. Risks of cancer in BRCA1-mutation carriers. Lancet 2005; 343:692–5.
760 Ovary and fallopian tubes
12 Muto MG, Cramer DW, Tangir J, Berkowitz R, Mok S. Frequency of the BRCA1 185delAG mutation among Jewish women with ovarian cancer and matched population controls. Cancer Res 1996; 56:1250–2. 13 Tonin P, Weber B, Offit K, et al. Frequency of recurrent BRCA1 and BRCA2 mutations in Ashkenazi Jewish breast cancer families. Nat Med 1996; 2:1179–83. 14 Ben David Y, Chetrit A, Hirsh-Yechezkel G, et al. Effect of BRCA mutations on the length of survival in epithelial ovarian tumors. J Clin Oncol 2002; 20:463–6. 15 Lu KH, Dinh M, Kohlmann W, et al. Gynecologic cancer as a “sentinel cancer” for women with hereditary nonpolyposis colorectal cancer syndrome. Obstet Gynecol 2005; 105:569–74. 16 Piver MS, Jishi MF, Tsukada Y, Nava G. Primary peritoneal carcinoma after prophylactic oophorectomy in women with a family history of ovarian cancer. A report of the Gilda Radner Familial Ovarian Cancer Registry. Cancer 1993; 71(9):2751–5. 17 Modan B, Hartge P, Hirsh-Yechezkel G, et al. Parity, oral contraceptives, and the risk of ovarian cancer among carriers and noncarriers of a BRCA1 or BRCA2 mutation. N Engl J Med 2001; 345(4):235–40. 18 Narod SA, Risch H, Moslehi R, et al. Oral contraceptives and the risk of hereditary ovarian cancer. Hereditary Ovarian Cancer Clinical Study Group. N Engl J Med 1998; 339(7):424–8. 19 Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100(1):57–70. 20 Feeley KM, Wells M. Precursor lesions of ovarian epithelial malignancy. Histopathology 2001; 38(2):87–95. 21 Obata K, Morland SJ, Watson RH, et al. Frequent PTEN/MMAC mutations in endometrioid but not serous or mucinous epithelial ovarian tumors. Cancer Res 1998; 58:2095–7. 22 Schaner ME, Ross DT, Ciaravino G, et al. Gene expression patterns in ovarian carcinomas. Mol Biol Cell 2003; 14(11):4376–86. 23 Flesken-Nikitin A, Choi KC, Eng JP, Shmidt EN, Nikitin AY. Induction of carcinogenesis by concurrent inactivation of p53 and Rb1 in the mouse ovarian surface epithelium. Cancer Res 2003; 63(13):3459–63. 24 Dinulescu DM, Ince TA, Quade BJ, Shafer SA, Crowley D, Jacks T. Role of K-ras and Pten in the development of mouse models of endometriosis and endometrioid ovarian cancer. Nat Med 2005; 11(1):63–70. 25 Schilder RJ, Sill MW, Chen X, et al. Phase II study of gefitinib in patients with relapsed or persistent ovarian or primary peritoneal carcinoma and evaluation of epidermal growth factor receptor mutations and immunohistochemical expression: a Gynecologic Oncology Group Study. Clin Cancer Res 2005; 11(15):5539–48. 26 Gordon AN, Finkler N, Edwards RP, et al. Efficacy and safety of erlotinib HCl, an epidermal growth factor receptor (HER1/EGFR) tyrosine kinase inhibitor, in patients with advanced ovarian carcinoma: results from a phase II multicenter study. Int J Gynecol Cancer 2005; 15(5):785–92. 27 Ries LA, Kosary CL, Hankey BF, et al. SEER Cancer Statistics Review 1973–1995 Bethesda, Md: National Cancer Institute, 1998.
28 Hough CD, Cho KR, Zonderman AB, Schwartz DR, Morin PJ. Coordinately up-regulated genes in ovarian cancer. Cancer Res 2001; 61(10):3869–76. 29 American Cancer Society. Cancer Facts and Figures 2005. Atlanta: American Cancer Society, 2005. 30 Serov SF, Scully RE, Robin IH. International Histologic Classification of Tumours: Histological Typing of Ovarian Tumours. Geneva: World Health Organization, 1973. 31 Helewa ME, Krepart GV, Lotocki R. Staging laparotomy in early epithelial ovarian carcinoma. Am J Obstet Gynecol 1986; 154(2):282–6. 32 Panici PB, Maggioni A, Hacker N, et al. Systematic aortic and pelvic lymphadenectomy versus resection of bulky nodes only in optimally debulked advanced ovarian cancer: a randomized clinical trial. J Natl Cancer Inst 2005; 97(8):560–6. 33 Shepherd JH. Revised FIGO staging for gynaecological cancer. Br J Obstet Gynaecol 1989; 96:889–92. 34 Cannistra SA. Cancer of the ovary. N Engl J Med 2004; 351(24):2519–29. 35 Bast RC, Jr., Knapp RC. Use of the CA 125 antigen in diagnosis and monitoring of ovarian carcinoma. Eur J Obstet Gynecol Reprod Biol 1985; 19(6):354–6. 36 Maggino T, Gadducci A, D’Addario V, et al. Prospective multicenter study on CA 125 in postmenopausal pelvic masses. Gynecol Oncol 1994; 54(2):117–23. 37 Jacobs IJ, Skates SJ, MacDonald N, et al. Screening for ovarian cancer: a pilot randomised controlled trial. Lancet 1999; 353(9160):1207–10. 38 Ueland FR, DePriest PD, Pavlik EJ, Kryscio RJ, van Nagel JR, Jr. Preoperative differentiation of malignant from benign ovarian tumors: the efficacy of morphology indexing and Doppler flow sonography. Gynecol Oncol 2003; 91(1):46–50. 39 DiSantis DJ, Scatarige JC, Kemp G, Given FT, Hsiu JG, Cramer MS. A prospective evaluation of transvaginal sonography for detection of ovarian disease. Am J Roentgenol 1993; 161(1):91–4. 40 Tekay A, Jouppila P. Validity of pulsatility and resistance indices in classification of adnexal tumors with transvaginal color Doppler ultrasound. Ultrasound Obstet Gynecol 1992; 2(5):338–44. 41 Marret H, Sauget S, Giraudeau B, et al. Contrast-enhanced sonography helps in discrimination of benign from malignant adnexal masses. J Ultrasound Med 2004; 23(12):1629–39. 42 Jeong YY, Outwater EK, Kang HK. Imaging evaluation of ovarian masses. Radiographics 2000; 20(5):1445–70. 43 van Nagell JR, DePriest PD. Management of adnexal masses in postmenopausal women. Am J Obstet Gynecol 2005; 193(1):30–5. 44 Petricoin EF, Ardekani AM, Hitt BA, et al. Use of proteomic patterns in serum to identify ovarian cancer. Lancet 2002; 359(9306):572–7. 45 Ransohoff DF. Lessons from controversy: ovarian cancer screening and serum proteomics. J Natl Cancer Inst 2005; 97(4):315–9. 46 Curtin JP. Management of the adnexal mass. Gynecol Oncol 1994; 55(3 Pt 2):S42–6. 47 Finkler NJ, Benacerraf B, Lavin PT, Wojciechowski C, Knapp RC. Comparison of serum CA 125, clinical impression, and
References 761
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
ultrasound in the preoperative evaluation of ovarian masses. Obstet Gynecol 1988; 72(4):659–64. Mayer AR, Chambers SK, Graves E, et al. Ovarian cancer staging: does it require a gynecologic oncologist? Gynecol Oncol 1992; 47(2):223–7. NIH consensus conference. Ovarian cancer. Screening, treatment, and follow-up. NIH Consensus Development Panel on Ovarian Cancer. JAMA 1995; 273(6):491–7. Campbell S, Bhan V, Royston P, Whitehead MI, Collins WP. Transabdominal ultrasound screening for early ovarian cancer. BMJ 1989; 299(6712):1363–7. Zurawski VR, Jr., Orjaseter H, Andersen A, Jellum E. Elevated serum CA 125 levels prior to diagnosis of ovarian neoplasia: relevance for early detection of ovarian cancer. Int J Cancer 1988; 42(5):677–80. Einhorn N, Bast R, Knapp R, Nilsson B, Zurawski V, Jr., Sjovall K. Long-term follow-up of the Stockholm screening study on ovarian cancer. Gynecol Oncol 2000; 79(3):466–70. Skates SJ, Menon U, MacDonald N, et al. Calculation of the risk of ovarian cancer from serial CA-125 values for preclinical detection in postmenopausal women. J Clin Oncol 2003; 21:(10 Suppl):206–10. Zanetta G, Chiari S, Rota S, et al. Conservative surgery for stage I ovarian carcinoma in women of childbearing age. Br J Obstet Gynaecol 1997; 104:1030–5. International Collaborative Ovarian Neoplasm Trial 1: a randomized trial of adjuvant chemotherapy in women with early-stage ovarian cancer. J Natl Cancer Inst 2003; 95:125–32. Trimbos JB, Vergote I, Bolis G, et al. Impact of adjuvant chemotherapy and surgical staging in early-stage ovarian carcinoma: European Organisation for Research and Treatment of Cancer-Adjuvant ChemoTherapy In Ovarian Neoplasm trial. J Natl Cancer Inst 2003; 95:113–25. Trimbos JB, Parmar M, Vergote I, et al. International Collaborative Ovarian Neoplasm trial 1 and Adjuvant Chemotherapy In Ovarian Neoplasm trial: two parallel randomized phase III trials of adjuvant chemotherapy in patients with early-stage ovarian carcinoma. J Natl Cancer Inst 2005; 95:105–12. Winter-Roach B, Hooper L, Kitchener H. Systematic review of adjuvant therapy for early stage (epithelial) ovarian cancer. Int J Gynecol Cancer 2003; 13(4):395–404. Bristow RE, Tomacruz RS, Armstrong DK, Trimble EL, Montz FJ. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. J Clin Oncol 2002; 20(5):1248–59. Hoskins WJ, Bundy BN, Thigpen JT, Omura GA. The influence of cytoreductive surgery on recurrence-free interval and survival in small-volume stage III epithelial ovarian cancer: a Gynecologic Oncology Group study. Gynecol Oncol 1992; 47(2):159–66. Rose PG, Nerenstone S, Brady MF, et al. Secondary surgical cytoreduction for advanced ovarian carcinoma. N Engl J Med 2004; 315:2489–97. van der Burg ME, van Lent M, Buyse M, et al. The effect of debulking surgery after induction chemotherapy on the prognosis in advanced epithelial ovarian cancer. Gynecological Cancer Cooperative Group of the European
63
64
65
66
67
68
69
70
71
72
73
74
75
76
Organization for Research and Treatment of Cancer. N Engl J Med 1995; 332:629–34. Chan YM, Ng TY, Ngan HY, Wong LC. Quality of life in women treated with neoadjuvant chemotherapy for advanced ovarian cancer: a prospective longitudinal study. Gynecol Oncol 2003; 88:9–16. Vergote I, De Wever I, Tjalma W, Van Gramberen M, Decloedt J, van Dam P. Interval debulking surgery: an alternative for primary surgical debulking? Semin Surg Oncol 2000; 19(1):49–53. Greer BE, Bundy BN, Ozols RF, et al. Implications of secondlook laparotomy in the context of optimally resected stage III ovarian cancer: a non-randomized comparison using an explanatory analysis: a Gynecologic Oncology Group study. Gynecol Oncol 2005; 99(1):71–9. Vergote IB, Winderen M, De Vos LN, et al. Intraperitoneal radioactive phosphorus therapy in ovarian carcinoma. Analysis of 313 patients treated primarily or at second-look laparotomy. Cancer 1993; 71:2250–60. Alberts DS, Markman M, Armstrong D, et al. Intraperitoneal therapy for stage III ovarian cancer: a therapy whose time has come! J Clin Oncol 2002; 20:3944–6. Markman M, Hakes T, Reichman B, et al. Intraperitoneal cisplatin and cytarabine in the treatment of refractory or recurrent ovarian carcinoma. J Clin Oncol 1991; 9:204–10. Howell SB, Kirmani S, McClay EF, et al. Intraperitoneal cisplatin-based chemotherapy for ovarian carcinoma. Semin Oncol 1991; 18:(1 Suppl 3):5–10. Alberts DS, Liu PY, Hannigan EV, et al. Intraperitoneal cisplatin plus intravenous cyclophosphamide versus intavenous cisplatin plus intravenous cyclophosphamide for stage III ovarian cancer. N Engl J Med 1996; 335:1950–5. Markman M, Bundy BN, Alberts DS, et al. Phase III trial of standard-dose intravenous cisplatin plus paclitaxel versus moderately high-dose carboplatin followed by intravenous paclitaxel and intraperitoneal cisplatin in small-volume stage III ovarian carcinoma: an intergroup study of the Gynecologic Oncology Group, Southwestern Oncology Group, and Eastern Cooperative Oncology Group. J Clin Oncol 2001; 19(4):1001–7. Armstrong DK, Bundy BN, Baergen R, et al. Randomized phase III study of intravenous (IV) paclitaxel and cisplatin versus IV paclitaxel, intraperitoneal (IP) cisplatin and IP paclitaxel in optimal stage III epithelial ovarian cancer (OC): A Gynecologic Oncology Group trial (GOG 172). Proc Am Soc Clin Oncol 2002; 21:201a (abstr 803). Armstrong DK, Bundy B, Wenzel L, et al. Intraperitoneal cisplatin and paclitaxel in ovarian cancer. N Engl J Med 2006; 354(1):34–43. Jaaback K, Johnson N. Intraperitoneal chemotherapy for the initial management of primary epithelial ovarian cancer. Cochrane Database Syst Rev 2006; (1):CD005340. McGuire WP, Hoskins WJ, Brady MF, et al. Assessment of dose-intensive therapy in suboptimally debulked ovarian cancer: a Gynecologic Oncology Group study. J Clin Oncol 1995; 13:1589–99. Kaye SB, Paul J, Cassidy J, et al. Mature results of a randomized trial of two doses of cisplatin for the treatment of ovarian cancer. Scottish Gynecology Cancer Trials Group. J Clin Oncol 1996; 14:2113–9.
762 Ovary and fallopian tubes
77 Gore M, Mainwaring P, A’Hern R, et al. Randomized trial of dose-intensity with single-agent carboplatin in patients with epithelial ovarian cancer. London Gynaecological Oncology Group. J Clin Oncol 1998; 16:2426–34. 78 Calvert AH, Horwich A, Newlands ES, et al. Carboplatin or cisplatin? Lancet 1988; 2:577–8. 79 Ozols RF, Bundy BN, Greer BE, et al. Phase III trial of carboplatin and paclitaxel compared with cisplatin and paclitaxel in patients with optimally resected stage III ovarian cancer: a Gynecologic Oncology Group study. J Clin Oncol 2003; 21:3194–200. 80 Murphy D, Crowther D, Renninson J, et al. A randomised dose intensity study in ovarian carcinoma comparing chemotherapy given at four week intervals for six cycles with half dose chemotherapy given for twelve cycles. Ann Oncol 1993; 4:377–83. 81 Piccart MJ, Bertelsen K, James K, et al. Randomized intergroup trial of cisplatin-paclitaxel versus cisplatincyclophosphamide in women with advanced epithelial ovarian cancer: three-year results. J Natl Cancer Inst 2000; 92(9):699–708. 82 McGuire WP, Hoskins WJ, Brady MF, et al. Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 1996; 334(1):1–6. 83 Muggia FM, Braly PS, Brady MF, et al. Phase III randomized study of cisplatin versus paclitaxel versus cisplatin and paclitaxel in patients with suboptimal stage III or IV ovarian cancer: a gynecologic oncology group study. J Clin Oncol 2000; 18(1):106–15. 84 The International Collaborative Ovarian Neoplasm Group. Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide, doxorubicin, and cisplatin in women with ovarian cancer: the ICON3 randomised trial. Lancet 2002; 360:505–15. 85 Vasey PA, Jayson GC, Gordon A, et al. Phase III randomized trial of docetaxel-carboplatin versus paclitaxel-carboplatin as first-line chemotherapy for ovarian carcinoma. J Natl Cancer Inst 2004; 96:1682–91. 86 Markman M, Liu PY, Wilczynski S, et al. Phase III randomized trial of 12 versus 3 months of maintenance paclitaxel in patients with advanced ovarian cancer after complete response to platinum and paclitaxel-based chemotherapy: a Southwest Oncology Group and Gynecologic Oncology Group trial. J Clin Oncol 2003; 21(13):2460–5. 87 Niloff JM, Knapp RC, Lavin PT, et al. The CA 125 assay as a predictor of clinical recurrence in epithelial ovarian cancer. Am J Obstet Gynecol 1986; 155(1):56–60. 88 Rustin GJ, Nelstrop AE, Tuxen MK, Lambert HE. Defining progression of ovarian carcinoma during follow-up according to CA 125: a North Thames Ovary Group Study. Ann Oncol 1996; 7(4):361–4. 89 Markman M, Rothman R, Hakes T, et al. Second-line platinum therapy in patients with ovarian cancer previously treated with cisplatin. J Clin Oncol 1991; 9(3):389–93. 90 Parmar MK, Ledermann JA, Colombo N, et al. Paclitaxel plus platinum-based chemotherapy versus conventional
91
92
93
94
95
96
97
98
99
100
101 102
103
104
105
platinum-based chemotherapy in women with relapsed ovarian cancer: the ICON4/AGO-OVAR-2.2 trial. Lancet 2003; 361(9375):2099–106. Pfisterer J, Plante M, Vergote I, et al. Gemcitabine/carboplatin (GC) vs. carboplatin (C) in platinum sensitive recurrent ovarian cancer (OVCA): results of a Gynecologic Cancer Intergroup randomized phase III trial of the AGO OVAR, the NCIC CTG and the EORTC GCG. Prog Proc Am Soc Clin Oncol 2004; 23:449. Segna RA, Dottino PR, Mandeli JP, Konsker K, Cohen CJ. Secondary cytoreduction for ovarian cancer following cisplatin therapy. J Clin Oncol 1993; 11(3):434–9. Trimble EL, Adams JD, Vena D, et al. Paclitaxel for platinumrefractory ovarian cancer: results from the first 1,000 patients registered to National Cancer Institute Treatment Referral Center 9103. J Clin Oncol 1993; 11(12):2405–10. Gonzalez-Martin AJ, Calvo E, Bover I, et al. Randomized phase II trial of carboplatin versus paclitaxel and carboplatin in platinum-sensitive recurrent advanced ovarian carcinoma: a GEICO (Grupo Espanol de Investigacion en Cancer de Ovario) study. Ann Oncol 2005; 16(5):749–55. Meyer T, Nelstrop AE, Mahmoudi M, Rustin GJ. Weekly cisplatin and oral etoposide as treatment for relapsed epithelial ovarian cancer. Ann Oncol 2001; 12(12):1705–9. Hatch KD, Beecham JB, Blessing JA, Creasman WT. Responsiveness of patients with advanced ovarian carcinoma to tamoxifen. A Gynecologic Oncology Group study of secondline therapy in 105 patients. Cancer 1991; 68(2):269–71. Hogberg T, Kagedal B. Long-term follow-up of ovarian cancer with monthly determinations of serum CA 125. Gynecol Oncol 1992; 46(2):191–8. Rustin GJ, Bast RC, Jr., Kelloff GJ, et al. Use of CA-125 in clinical trial evaluation of new therapeutic drugs for ovarian cancer. Clin Cancer Res 2004; 10(11):3919–26. Benedetti PP, Scambia G, Baiocchi G, Iacobelli S, Mancuso S. Predictive value of multiple tumor marker assays in secondlook procedures for ovarian cancer. Gynecol Oncol 1989; 35(3):286–9. Mano A, Falcao A, Godinho I, et al. CA-125 AUC as a new prognostic factor for patients with ovarian cancer. Gynecol Oncol 2005; 97(2):529–34. Vaidya AP, Curtin JP. The follow-up of ovarian cancer. Semin Oncol 2003; 30(3):401–12. Omura GA, Brady MF, Homesley HD, et al. Long-term followup and prognostic factor analysis in advanced ovarian carcinoma: the Gynecologic Oncology Group experience. J Clin Oncol 1991; 9(7):1138–50. Dembo AJ, Davy M, Stenwig AE, Berle EJ, Bush RS, Kjorstad K. Prognostic factors in patients with stage I epithelial ovarian cancer. Obstet Gynecol 1990; 75(2):263–73. Paramasivam S, Tripcony L, Crandon A, et al. Prognostic importance of preoperative CA-125 in International Federation of Gynecology and Obstetrics stage I epithelial ovarian cancer: an Australian multicenter study. J Clin Oncol 2005; 23(25):5938–42. Leake JF, Currie JL, Rosenshein NB, Woodruff JD. Long-term follow-up of serous ovarian tumors of low malignant potential. Gynecol Oncol 1992; 47(2):150–8.
References 763
106 Seidman JD, Kurman RJ. Ovarian serous borderline tumors: a critical review of the literature with emphasis on prognostic indicators. Hum Pathol 2000; 31(5):539–57. 107 Snider DD, Stuart GC, Nation JG, Robertson DI. Evaluation of surgical staging in stage I low malignant potential ovarian tumors. Gynecol Oncol 1991; 40(2):129–32. 108 Trope C, Kaern J, Vergote IB, Kristensen G, Abeler V. Are borderline tumors of the ovary overtreated both surgically and systemically? A review of four prospective randomized trials including 253 patients with borderline tumors. Gynecol Oncol 1993; 51(2):236–43. 109 Bostwick DG, Tazelaar HD, Ballon SC, Hendrickson MR, Kempson RL. Ovarian epithelial tumors of borderline malignancy. A clinical and pathologic study of 109 cases. Cancer 1986; 58(9):2052–65. 110 Fort MG, Pierce VK, Saigo PE, Hoskins WJ, Lewis JL, Jr. Evidence for the efficacy of adjuvant therapy in epithelial ovarian tumors of low malignant potential. Gynecol Oncol 1989; 32(3):269–72. 111 Segelov E, Campbell J, Ng M, et al. Cisplatin-based chemotherapy for ovarian germ cell malignancies: the Australian experience. J Clin Oncol 1994; 12:378–84. 112 Gershenson DM, Morris M, Cangir A, et al. Treatment of malignant germ cell tumors of the ovary with bleomycin, etoposide, and cisplatin. J Clin Oncol 1990; 8(4):715–20. 113 Thomas GM, Dembo AJ, Hacker NF, et al. Current therapy for dysgerminoma of the ovary. Obstet Gynecol 1987; 70:268–75. 114 Williams S, Blessing JA, Liao SY. Adjuvant therapy of ovarian germ cell tumors with cisplatin, etoposide, and bleomycin: a trial of the Gynecologic Oncology Group. J Clin Oncol 1994; 12:701–6. 115 Williams SD, Blessing JA, Moore DH, Homesley HD, Adcock L. Cisplatin, vinblastine, and bleomycin in advanced and recurrent ovarian germ-cell tumors. A trial of the Gynecologic Oncology Group. Ann Intern Med 1989; 111(1):22–7. 116 Gershenson DM. The obsolescence of second-look laparotomy in the management of malignant ovarian germ cell tumors. Gynecol Oncol 1994; 52(3):283–5. 117 Williams SD, Blessing JA, Hatch KD, Homesley HD. Chemotherapy of advanced dysgerminoma: trials of the Gynecologic Oncology Group. J Clin Oncol 1991; 9(11):1950–5. 118 Norris HJ, Zirkin HJ, Benson WL. Immature (malignant) teratoma of the ovary: a clinical and pathologic study of 58 cases. Cancer 1976; 37:2359–72. 119 Gallion H, van Nagell JR Jr, Powell DF. Therapy of endodermal sinus tumor of the ovary. Am J Obstet Gynecol 1979; 135:447–51. 120 Kurman RJ, Norris HJ. Malignant germ cell tumors of the ovary. Hum Pathol 1977; 8:551–64.
121 Outwater EK, Wagner BJ, Mannion C, McLarney JK, Kim B. Sex cord-stromal and steroid cell tumors of the ovary. Radiographics 1998; 18(6):1523–46. 122 Giardiello FM, Welsh SB, Hamilton SR, et al. Increased risk of cancer in the Peutz-Jeghers syndrome. N Engl J Med 1987; 316(24):1511–4. 123 Gorlin RJ. Nevoid basal-cell carcinoma syndrome. Medicine (Baltimore) 1987; 66(2):98–113. 124 Schumer ST, Cannistra SA. Granulosa cell tumor of the ovary. J Clin Oncol 2003; 21:1180–9. 125 Kietlinska Z, Pietrzak K, Drabik M. The management of granulosa-cell tumors of the ovary based on long-term follow up. Eur J Gynaecol Oncol 1993; 14 Suppl:118–27. 126 Homesley HD, Bundy BN, Hurteau JA, Roth LM. Bleomycin, etoposide, and cisplatin combination therapy of ovarian granulosa cell tumors and other stromal malignancies: A Gynecologic Oncology Group study. Gynecol Oncol 1999; 72(2):131–7. 127 Hardy RD, Bell JG, Nicely CJ, Reid GC. Hormonal treatment of a recurrent granulosa cell tumor of the ovary: case report and review of the literature. Gynecol Oncol 2005; 96(3):865–9. 128 Obermair A, Taylor KH, Janda M, Nicklin JL, Crandon AJ, Perrin L. Primary fallopian tube carcinoma: the Queensland experience. Int J Gynecol Cancer 2001; 11(1):69–72. 129 Brose MS, Rebbeck TR, Calzone KA, Stopfer JE, Nathanson KL, Weber BL. Cancer risk estimates for BRCA1 mutation carriers identified in a risk evaluation program. J Natl Cancer Inst 2002; 94(18):1365–72. 130 Schneider C, Wight E, Perucchini D, Haller U, Fink D. Primary carcinoma of the fallopian tube. A report of 19 cases with literature review. Eur J Gynaecol Oncol 2000; 21(6):578–82. 131 Patlas M, Rosen B, Chapman W, Wilson SR. Sonographic diagnosis of primary malignant tumors of the fallopian tube. Ultrasound Q 2004; 20(2):59–64. 132 Creasman WT. Revision in classification by International Federation of Gynecology and Obstetrics. Am J Obstet Gynecol 1992; 167(3):857–8. 133 Klein M, Rosen AC, Lahousen M, Graf AH, Rainer A. Lymphadenectomy in primary carcinoma of the Fallopian tube. Cancer Lett 1999; 147(1–2):6366. 134 Gadducci A. Current management of fallopian tube carcinoma. Curr Opin Obstet Gynecol 2002; 14(1):27–32. 135 Rosen AC, Klein M, Hafner E, Lahousen M, Graf AH, Reiner A. Management and prognosis of primary fallopian tube carcinoma. Austrian Cooperative Study Group for Fallopian Tube Carcinoma. Gynecol Obstet Invest 1999; 47(1):45–51. 136 Alvarado-Cabrero I, Young RH, Vamvakas EC, Scully RE. Carcinoma of the fallopian tube: a clinicopathological study of 105 cases with observations on staging and prognostic factors. Gynecol Oncol 1999; 72(3):367–79.
33 Uterus BLEDDYN JONES
Epidemiology Aetiology Pathology Symptoms Investigation Staging Treatment
764 764 765 766 766 766 766
EPIDEMIOLOGY Cancer of the uterus occurs most frequently in the sixth and seventh decade, with an approximate median age at presentation of 60. It is, with rare exceptions, a post-menopausal condition. Uterine cancers account for approximately 3800 deaths per year in England and Wales and are the fifth commonest cause of cancer deaths in women1*** and the fourth in the USA. The incidence figures are relatively stable over the last three decades despite the increase in age distribution of the population, the increasing use of oestrogen-only hormone replacement (HRT) and partial oestrogen agonists for breast cancer, although these statistics might be influenced by the protective effect of combined oral contraceptives and increasing practice of prophylactic hysterectomies.2*** The mortality rates are only marginally greater in women from lower socio-economic groups. There are some geographical variations with earlier age presentations in those of Japanese descent3* and possible enhanced rates of mixed mesodermal sarcomas in patients from the Indian subcontinent. Uterine cancers are likely to become the commonest curable invasive cancer in women, due to the impact of screening on the incidence of invasive cervix cancer.
AETIOLOGY The precise aetiology is unknown. There is little evidence of an inherited predisposition in most patients.4*
Follow-up Re-treatment Future prospects Sarcomas of the uterus Key references References
772 772 772 772 773 773
Environmental factors are therefore implicated, particularly the interaction between the promoter effect of oestrogens – allowing enhanced proliferation – and environmental lipidsoluble chemical carcinogens ingested with fat-rich foods. There are several predisposing factors, some of which are also recognized in the development of carcinoma of the breast, such as nulliparity: these include age, obesity, diabetes mellitus and hypertension. The synthesis of oestrogen in aromatase-containing adipose cells is an obvious source of high body oestrogen levels that would favour uterine carcinogenesis. Hormonal over-stimulation of the endometrium by oestrogens, either endogenous or exogenous, for example by the use of oestrogen-only HRT, predisposes to the development of endometrial hyperplasia. Prolonged use of tamoxifen is associated with a sevenfold increase in the development of endometrial carcinoma and also of bowel cancer5* in those patients who have previously had a carcinoma of the breast. However, other studies such as the national Surgical Adjuvant Breast Cancer Prevention Trial found a 29 per cent increase in endometrial cancer but this was not statistically significant.6** Other confounding variables (genetic, environmental carcinogens and similar risk factors such as nulliparity) could partly explain this effect although there is ample evidence that endometrial hypertrophy does develop after the long-term use of tamoxifen. Mutations of the genetic code can be produced by tamoxifen–DNA adducts.7 One paradox is that in recurrent endometrial cancer, tamoxifen can cause sustained tumour regression after failure of high-dose progestagen therapy, although these patients are usually tamoxifen naive.
Pathology 765
Polycystic ovarian disease, associated with high oestradiol levels, can predispose younger women to develop endometrial cancer.8*
PATHOLOGY As well as in non-malignant adenomas and fibroadenomas, the pre-malignant condition of endometrial hyperplasia needs careful assessment to exclude the presence of atypical cells or a frank carcinoma. There is some controversy as to the implication of the term atypical hyperplasia with respect to the actual incidence of cancer.9 Presentations within polyps usually confer a good prognosis, although careful assessment is necessary for invasion of the mucosa. Malignant tumours Adenocarcinoma of endometrioid type (non-mucus-secreting) is typical and there are numerous variants, e.g. clear cell cancer. The papillary serous variant frequently leads to rapid peritoneal spread and also appears to be the most radioresistant subtype.10 Squamous cell carcinoma can occur and more rarely small cell cancer. The histopathologist carefully determines: 1. Tumour grade,1–3 the most important factor.11* The higher-grade categories have an increased risk of local (vault) and pelvic nodal recurrence. 2. Extent of myometrial invasion: tumours invading the outer half are more likely to spread to pelvic nodes. 3. Vascular/lymphatic invasion. 4. Atypical histological types such as clear cell and the serous papillary variant confer a relatively poor prognosis. The latter frequently recur intra-abdominally and the natural history and mode of spread resembles that of ovarian cancer. The position of the tumour, whether in the upper or lower segment, may also influence outcome. The incidence of lymphatic spread is greater in the case of tumours in the lower segment of the uterus, which generally spread via the same lymphatic pathways as carcinoma of the cervix; in rare cases spread to the inguinal lymph nodes occurs (possibly via the broad ligament). Sometimes there is diagnostic uncertainty if tumours of similar histology occur within the ovary and endometrium due to either synchronous primaries or metastatic cancer. Where there is such doubt, clinicians tend to offer treatment based on the assumption that the primary tumour is ovarian, although the possibility of an endometrial primary should never be dismissed, particularly if the treatment response is not typical of ovarian cancer. The other important differential diagnosis is that of endocervical adenocarcinoma, which is normally mucus secreting, unlike endometrial carcinoma. Cellular hormone receptor status can sometimes help to differentiate the origin.
The role of peritoneal cytology as a strong independent prognostic factor remains to be determined in the case of stage I cancers but appears predictive for extra-uterine cancers.12 Some authorities classify the uterine cancers into the ‘usual type’ that are endocrine sensitive, while the other cancers are regarded as ‘special variant’ types and are oestrogen independent and poorly differentiated, arising in nonhyperplastic endometrium.
Genetic changes Aneuploidy confers a poor prognosis.13* Special variant tumours contain a far greater incidence of genetic changes as assessed by loss of chromosomal heterozygosity.14 P53 mutations are the commonest reported genetic change, occurring in 20 per cent, but are probably not an early event in cancer evolution since they are mainly found in more advanced tumours with a poor prognosis.15* Similarly, overexpression of HER-2/neu oncogene can be found in 10–15 per cent of advanced stages, whereas K-ras mutations are reported in 15 per cent of stage I–II cancers.16 Lack of PTEN gene expression is correlated with cancer progression. Multiple gene expression profiles are producing characteristic patterns of perturbations17 that could improve allocation of the precise histopathological subtype, for example if the specimen has excessive crush artefact. Such techniques may help with defining the prognosis and be useful in treatment selection, although sampling error from biopsies will remain a problem in that more biologically aggressive areas may not have been included in the assessment. One genetic condition of current practical importance is the Lynch syndrome/hereditary non-polyposis colorectal cancer syndrome (HNPCC). Recent studies have identified a 50 per cent lifetime risk of endometrial cancer; this exceeds that of colonic cancer which was responsible for the identification of the syndrome.18 This finding has important implications for the follow-up of female patients. There is also evidence of DNA mismatch repair deficiency in endometrial cancers. Repair deficiency can potentially be exploited by novel repair-inhibitor drugs: therapeutic approaches are being tested in cellular and animal models.19
Serum markers CA 125 is sometimes elevated in recurrent disease or where there is a large tumour burden that causes reactive inflammatory changes in the peritoneal surfaces. This marker is very infrequently raised at presentation and cannot be regarded as tumour specific.
Tumour receptor status There is evidence that higher nuclear progesterone receptor levels are associated with improved survival. In contrast
766 Uterus
there is conflicting evidence as to the relevance of oestrogen receptors.20*
Routes of spread The tumour may extend to fill the endometrial cavity and prolapse beyond the os cervix as a pseudo-polyp. The depth of myometrial invasion correlates with an increase in pelvic lymph node spread. Approximately 9 per cent of patients in the stage I category have pelvic nodal involvement. Direct extension to the stroma of the cervix also results in increased pelvic lymphatic involvement. Very occasionally, submucosal vaginal extension may occur, resulting in an isolated metastatic deposit in the lower third of the vagina, usually in the suburethral region, and which may cause urethral obstruction. It is consequently important to investigate the uterus thoroughly in patients that present with isolated vaginal tumours. Para-aortic nodal spread occurs in only 6 per cent of stage I tumours but in 18 per cent for the non-endometrioid adenocarcinoma variants, and the incidence also increases with stage.21* Isolated relapse in the para-aortic region does infrequently occur after successful pelvic surgery and radiotherapy. Peritoneal spread may follow extension through the serosal surface of the uterus, as may direct infiltration of the surrounding pelvic tissues. Blood stream spread is unusual but patients occasionally develop lung, liver and bone metastases. Rarely, a patient may develop a nodal metastasis in the inguinal region.
SYMPTOMS The characteristic symptoms of post-menopausal bleeding and vaginal discharge should be investigated thoroughly in all women. Other symptoms due to a pelvic mass lesion (see Chapter 32) may occur in higher stage tumours.
aperture and claustrophobic or psychotic patients may fare badly or refuse to proceed. Pelvic CT imaging is not as helpful as the above techniques in uterine tumour imaging, but abdominal CT imaging is more sensitive in the assessment of retroperitoneal lymphadenopathy. Difficulties can occur when there are multiple fibroids which can obscure a cancer or sarcoma situated in the fundal or cornual region at hysteroscopy and also cause confusion in scan interpretation.
STAGING The current Federation Internationale de Gynecologie et d’Obstetrique (FIGO) staging system differs in several ways from that of the uterine cervix. A simplified version is given in Table 33.1.
TREATMENT The treatment plan should be decided by a multidisciplinary team which takes full account of diagnosis, stage, grade and medical history and present health status of each patient. The patient should be informed of the risks and benefits associated with each method of treatment.
Surgery A total abdominal hysterectomy and bilateral salpingooophorectomy (TAH and BSO) is the traditional cornerstone of management in 90 per cent of patients, with average survival rates that exceed 70 per cent. Subtotal hysterectomy is sometimes performed, but this is not advisable as the uterine incision may contain cancer cells and subsequent radiotherapy may be severely compromised by bowel adhesions. These operations are normally done via a
INVESTIGATION Clinical assessment will include abdominal and pelvic examination. A direct biopsy can be taken in the clinic, e.g. by Pipelle or other commercial aspirators, when the endometrial thickness is greater than 4 mm.22* Should there be diagnostic difficulties an examination under anaesthetic (EUA), hysteroscopy (and further biopsy) and curettage can be performed. Several imaging methods have also improved the diagnostic and staging accuracy of uterine tumours. Transvaginal ultrasound can be used to virtually exclude the presence of endometrial cancer and to assess the extent of myometrial invasion (70–80 per cent accuracy). Greater accuracy (95–97 per cent) in staging, including the assessment of lymphadenopathy, may be obtained with magnetic resonance imaging (MRI) although very obese patients may not be accommodated within the scanner
Table 33.1 FIGO staging system Stage IA IB IC IIA IIB IIIA IIIB IV
Anatomical description of involvement Limited to endometrium Invasion of inner half of myometrium Invasion of outer half of myometrium Endocervical glandular extension Cervical stromal invasion Invasion of serosa and/or adnexa and/or positive peritoneal cytology Pelvic or para-aortic lymph nodes Bladder/bowel/inguinal nodes/abdominal or distant metastases
Treatment 767
Pfannenstiel incision and there is virtually no risk of scar recurrence. In some instances, endometrial cancer or a sarcoma is found incidentally by the histopathologist on microscopical examination of a hysterectomy specimen after elective surgery for a presumed benign condition; in other rare cases a vaginal hysterectomy will have been performed for prolapse. In such cases, it is considered wise to proceed with a bilateral salpingo-oophorectomy if this has not already been performed. Alternatively, where the latter is not feasible, the management will then follow that of a standard TAH and BSO procedure. There is increasing interest in laparoscopic procedures as a less morbid alternative to laparotomy. With experienced operators there appears to be no detriment in survival in one small randomized study.23** Some authorities continue to recommend vaginal hysterectomy as a less invasive and safer procedure than laparotomy in elderly patients.24 The role of additional pelvic lymphadenectomy remains unproven.25* In the UK, the Medical Research Council (MRC) ASTEC (A Study in the Treatment of Endometrial Cancer) trial has been designed to determine whether this additional surgical procedure is of benefit. So far the preliminary indications are that extended piecemeal nodal dissections provide no advantage and could be deleterious in terms of morbidity and survival.
Radiotherapy Radiotherapy may be used in several circumstances.26,27 These are essentially postoperative (adjuvant) following hysterectomy or primary radical or palliative treatment. The rationale of the former is to prevent local recurrence at the: 1. vaginal vault 2. pelvic lymph nodes. VAULT BRACHYTHERAPY ALONE
This approach will only prevent recurrence in the vaginal vault. This policy continues to be advocated by some specialists for surgical stages IB G3 and IC,28*,29* for example by three 7 Gy fractions or 30 Gy in five fractions delivered at 0.5 cm from a vaginal applicator surface. The potential advantages are that patients who have a low risk of nodal metastases can be treated without external beam radiotherapy and therefore avoid the potential side effects. The potential disadvantages include the possible poor intracavitary dose distribution at the vault where there may be bilateral recesses, for which the use of two ovoid shaped applicators may be preferable to the use of a conventional single vaginal cylinder technique. Another potential problem is that a later vault or isolated pelvic nodal relapse is difficult to treat, since the previously treated brachytherapy volume requires to be shielded. For these reasons, the present author prefers not to use brachytherapy as the sole therapy in this situation.
EXTERNAL BEAM AND BRACHYTHERAPY
Although the indications remain controversial, external radiotherapy is usually given to stage IC cancers and all those in the grade 3 category. Most radiotherapists will advise the same for all stage II cancers although there again remains some controversy in that local brachytherapy may suffice for the IIA well differentiated category, particularly if pelvic node sampling is negative.30,31*** A four-field treatment plan is usually used to deliver 40–45 Gy in 20–25 fractions over 4–5 weeks on a linear accelerator of at least 6 MV energy. The field dimensions are similar to those for carcinoma of the cervix, but the lateral field dimensions do not need to extend so far anteriorly since the uterine fundus is absent. Usually, the antero-posterior (AP) fields extend from the mid or lower border of the L5 vertebra to the lower aspect of the obturator foramen and laterally to 1 cm beyond the pelvic sidewall. The dimensions are normally 14 cm2 or slightly larger for the AP fields. The lateral posterior field limit can be marked by a line positioned 2.5 cm anterior to the sacral hollow or passing through the S2/3 junction where this is visible. The lateral anterior field limit normally passes through the mid symphysis pubis and should always pass just anterior to the acetabulum and provide a reasonable margin for lymph node cover in front of the sacrum/L5 vertebra. The field width is usually 8–10 cm; field sizes above this limit should be avoided because of acute toxicity. Some specialists use radio-opaque vaginal and rectal markers, but these devices are not essential as there is little risk of geographic miss in this situation, unlike in cervix cancer or for intact uterine cancers with prolapse. Shielding of the following regions is advisable: ●
●
corners of anterior and posterior fields – this will reduce the integral dose to the intestine and hip joints posterior aspect of sacrum in lateral fields – this will potentially reduce the risk of radiation-induced sacral insufficiency fractures.
These are essentially similar to cervix cancer treatments (see Chapter 34). Treatment difficulties can occur in very obese patients: additional applied dose, skin fold reactions and problems with reproducibility of daily set-up can be sufficient to restrict the prescribed dose to 20–40 Gy in some patients. After hysterectomy, there is a greater volume of small intestine within the pelvis and there may be postoperative adhesions. Other previous pelvic surgical procedures such as appendicectomy may also lead to adhesions that influence pelvic radiotolerance. The bladder position is more posterior following hysterectomy and there may be incomplete emptying. These aspects, together with the associated medical conditions referred to, may all contribute to a reduction in pelvic radiation tolerance such that doses higher than 40–45 Gy in 20–25 fractions should not be exceeded unless there are good reasons to do so. If there is definite inoperable malignant pelvic lymphadenopathy,
768 Uterus
an external beam dose of 50.4 Gy in 28 fractions can be justified. INTRA-CAVITARY THERAPY FOLLOWING EXTERNAL BEAM
A relatively small additional component of vault brachytherapy is usually given to cover the upper few centimetres of the vagina (2–3 cm active length to 6 Gy in one fraction at 0.5 cm from applicator surface or 4 cm active length to 8 Gy in two fractions at 0.5 cm). A distance of 0.5 cm is used because this represents the maximum dose to the adjacent urethral or rectal serosal surface. These are safe doses in that the present author has not experienced any grade 3 complications after treating over 500 patients. The brachytherapy volume is sufficiently small and the overall biological effective dose (BED) is below standard rectal tolerances of around 110 Gy3. When brachytherapy is used as a boost in this way, the effect of dose inhomogeneity in the vaginal vault is not likely to be as important as when brachytherapy only is used. If low/medium dose rate brachytherapy is to be used then a further 10–15 Gy using low/medium dose rate caesium can be given providing the bladder or rectal dose estimates are not greater than 1 Gy per hour. In these patients, the brachytherapy is intended to deliver a relatively small additional dose to the vault tissues that may potentially contain hypoxic scar tissue embedded with malignant cells following surgery. The technique most commonly used is that of a vaginal cylinder with radioactive loading of the upper 2–4 cm only. The total BED used in the UK in 1996 ranged between 90 and 124 Gy3; doses in the lower end of this range appear to give satisfactory local control and very low toxicity. EXTERNAL BEAM THERAPY ALONE FOLLOWING HYSTERECTOMY
Some authors prefer to avoid brachytherapy and report good results32* following external beam only, although relatively high doses such as 50.4 Gy in 28 fractions are used. This may be satisfactory in fit patients, but there remains some concern as to the late side effects of such doses to a large volume of pelvic tissues, particularly the small intestine, in patient subgroups where pelvic tolerance may be compromised. The alternative policy is frequent surveillance by means of pelvic examinations and biopsy where recurrent tumour is suspected. Prompt treatment, by external beam plus vaginal intra-cavitary brachytherapy, of recurrent pelvic carcinoma can be curative in over 50 per cent of patients.33* A higher prescribed dose of radiotherapy is usually necessary: 45–50.4 Gy in 25–28 fractions plus two intra-cavitary high dose rate insertions of 5–6 Gy at 0.5 cm or equivalent continuous radiation (e.g. 15–20 Gy at 1 Gy/h). The intra-cavitary treatment length may be longer in these patients, depending on the extent of vaginal involvement. If several intra-cavitary fractions are used, a shrinking-volume brachytherapy technique can be used, so that the whole vagina can be treated initially, with a smaller active treatment length for the final 1–2
fractions. Such clinical management can only be confidently followed if adequate staff resources and reasonably rapid access to radiotherapy facilities are available. Patients should be fully informed of the rationale for postoperative treatment and may have strong personal feelings as to whether they wish to proceed with adjuvant treatments. A full discussion, including risk of recurrence assessment and explanation of the potential side effects of radiotherapy, is essential in this context. The reported outcomes vary, but there is a general consensus that overall pelvic recurrence rates are reduced from the 30–40 per cent range to around 1–5 per cent, although there is no proven advantage in terms of survival.34*,35* The Dutch randomized trial reported by Creutzberg et al.36** rather surprisingly used only external beam radiotherapy to a total dose of 46 Gy in 23 fractions, with simple twofield arrangements in the initial years of the study. This postoperative pelvic radiotherapy did reduce the locoregional recurrence rate, but the lack of survival benefit coupled with significant serious radiation toxicity – 3 per cent grade 3–4 complications37** – resulted in a recommendation that postoperative radiotherapy may no longer be indicated for stage I cancers apart from those with high-risk features (i.e. the stage IC and grade 3 categories). Further publications by the same group provide broadly similar conclusions.38**,39** These conclusions are valid only for the radiotherapy technique and schedules used within the Dutch trial and should not be extrapolated to other treatment techniques such as combinations of brachytherapy and external radiotherapy to different doses. However, these findings and the suggested change in management undoubtedly influenced many surgeons who were unaware of the subtleties of radiation oncology and who used extrapolated logic to imply that these results should apply to all radiotherapy regimens. For example, the relatively good results obtained in the unirradiated group reflect the high efficiency of the early detection and excellent salvage radiotherapy given to patients that had recurrences in the vaginal vault. Also, it remains to be seen whether a smaller dose per fraction (e.g. 1.8 Gy) using only four field treatment plans to a slightly lower total dose of 45 Gy (a much lower biological effective dose to the wider pelvis) coupled with a low dose of vault brachytherapy to a total of just above 50 Gy may significantly improve the therapeutic ratio relative to the Dutch trials. The results of further randomized trials (such as the MRC ASTEC trial where the 1.8 Gy schedule was recommended) are awaited with interest, particularly in the trial arm where no lymphadenectomy has been performed; the wisdom of combining radical lymphadenectomy with postoperative radiotherapy remains doubtful because of the obvious reduction in pelvic radiotolerance that follows extended surgery. There is also experience of whole abdominal radiotherapy in an adjuvant setting: prolonged survival in stage III and IV cancers is described only after complete resection of gross cancer.40* This cannot be regarded as routine treatment without further research.
Treatment 769
the uterine canal, which represents the serosal surface, and this dose cannot then be exceeded in the surrounding normal tissues such as sigmoid colon or small bowel. The intra-cavitary techniques are variable and include:
RADICAL RADIOTHERAPY
Although relatively infrequently performed, this can be effective in those patients deemed inappropriate for medical or anatomical reasons for pelvic surgery. Radical radiotherapy is probably underused and carries less immediate mortality risk than surgery. With the increasing age structure of the population and also the increased prevalence of obesity with time, this approach is likely to increase in the future. Patient choice may also have a bearing on this decision, although very few patients are actually offered radiotherapy as the primary management, although some may be offered radiotherapy after they have refused a surgical procedure. Treatment can be given:
1. A single line source.41* This provides a high dose largely confined to uterus, and the length of treatment below the cervix can be adjusted according to the tumour extent. If there is cervical involvement, the active length is extended to include the upper 2 cm of vagina, and it may be necessary to include a longer portion of the vagina, or the whole vagina, according to the degree of spread. Following 45 Gy in 25 fractions external beam, a dose of 6 Gy 2 can be given at point A providing that the rectal and bladder predicted doses are less than 80 per cent of the prescribed dose. A higher dose of 7–8 Gy per fraction may be prescribed at the distance of the uterine serosal surface from the uterine canal, if this is known with reasonable accuracy. The option of variable source positions or dwell times allows for some degree of isodose curve variation, which can be useful if the tumour position is known with precision. For example, if the tumour is confined to the fundal region, then increased radiation loading at this level can be used (see Fig. 33.1); for those in mid uterus, then the corresponding portion of the intra-uterine catheter can be loaded usually to 25–50 per cent higher activity.
1. by intra-cavitary treatment alone; or 2. by combination of external beam radiotherapy and intra-cavitary brachytherapy. The 5-year pelvic control rates are good (70–80 per cent at 5 years) for small tumours, but are less satisfactory (50 per cent) in the higher stages as is the case with surgery. The tumour stage in this category can only be assessed by clinical and imaging techniques. Recent advances in pelvic imaging have allowed the oncologist to select a treatment dose that is appropriate to the dimensions of the uterus. For example, if there is a relatively small superficial cancer and the uterine wall thickness is only 1.5 cm, then the brachytherapy dose can be prescribed at this distance from
(a)
(b)
Uterine Uterine 800
800
600 400
600
400 Flange Flange
200
200 Vaginal
Vaginal
Figure 33.1 Schematic diagram of source position (indicated by the black circles) and isodose distributions for a single intra-uterine catheter. In (a) only the intra-uterine portion of the catheter is loaded and each dwell position has an equal dwell time: this produces the cigar-shaped isodose distribution. In (b) the vagina is also included but the intra-vaginal dwell times are 50 per cent of those in the intrauterine portion. Here, the isodose width is sufficient to cover the uterine fundus but narrower in order to cover a minimal amount of the paravaginal tissues. The dwell times at the fundal region may require to be increased relative to those at the centre of the catheter if there is a large fundal tumour.
770 Uterus
2. A triple arrangement (uterine tube and ovoids) as in cervix cancer: the classical pear shaped isodose distribution seems inappropriate apart from in cases where there is direct tumour extension to the cervix. 3. A double line source in uterus, one line per cornua. 4. The Heymans capsule, or equivalent technique, where multiple small source carriers or catheters are used to fill the remaining uterine cavity.42*,43* This has been extended in the modern era at Leeds and Vienna, although whether such sophisticated brachytherapy will make an impact on survival and tissue side effects remains to be shown. The technique cannot ensure that the deepest infiltrating part of the cancer is exposed to the highest brachytherapy dose. 5. The umbrella type applicator, where several catheters splay outwards towards the uterine fundus. This technique requires wide dilatation of the endocervical and uterine canal.44 The results of several reports of radical radiotherapy used alone are summarized in Table 33.2. RADIATION SIDE EFFECTS
Equivalent doses to 75 Gy radium at point A appear to result in a 17 per cent severe complication rate.51* Techniques that attempt to give a high biological dose of external beam and brachytherapy will also inevitably result in a high incidence of serious morbidity.56* High dose rate fraction sizes above 8 Gy are probably best avoided. Some series have used up to 10 Gy boost after fractionated external radiotherapy, with reports of under 5 per cent severe complications at 5 years.52* Modest brachytherapy fractionation and
care in choice of prescription point may produce a further improvement in therapeutic ratio. BED values of less than 110 Gy3 appear safe in the context of cervix cancer57* and BED doses within this range seem to be safe in intact uterine cancers. More recent use of lower prescribed doses and in some instances brachytherapy dose prescription at the serosal surface for early-stage cancers has produced very low morbidity and adequate tumour control.41* EXTERNAL BEAM ONLY
Where intra-cavitary techniques are not feasible after a standard dose of 45 Gy, it is reasonable to continue the prescription to 50.4 Gy (or higher using smaller CT planned fields). Following this, continuing tumour regression may permit successful brachytherapy, although to a lower dose than originally intended in order to respect pelvic tolerance. For extra-uterine extension, the use of conformal and intensity modulated X-ray therapy or proton therapy aided by MRI and CT treatment planning could, in principle, deliver high local doses providing the irradiated volume of normal tissue is sufficiently small. RECURRENT CANCERS
Pelvic relapses after central brachytherapy can theoretically be treated by intensity modulated proton beams without delivery of further central dose: intensity modulated radiotherapy (IMRT) using X-rays would inevitably result in further central pelvic dose. In some patients, the para-aortic lymph nodes are the first site of relapse and may be treated by radiotherapy with either radical or palliative intent; in the
Table 33.2 Inoperable uterine cancers treated only by radical radiotherapy Five-year survival (%) (number of patients)
Grade 3 or severe toxicity (%)
Brachytherapy technique
Patanaphan et al45* Wang et al46* Fishman et al47* Varia et al48* Kupelian et al49* Rouanet et al50* Taghian et al51* Kucera et al52* Grigsby et al53* Nguyen et al54* Churn & Jones41*
46 (54) 46 (41) 80–100 (54) 26–57 (73) 49–88 (152) 77 (250) 52 (104) 53 (280) 77 (69) 76 at 8 years (27) Stage1 2 64 Stage 3 4 33 (37)
N/A N/A N/A N/A 3 N/A 17 5 16 11 3
Niazi et al55*
Disease-free survival 78 per cent at 15 years (38)
Mainly intracavitary LDR Intracavitary radium Intracavitary external beam Intracavitary external beam Intracavitary external beam Intracavitary external beam LDR Single line source, HDR Intracavitary external beam HDR external beam Single line source; LDR or HDR external beam HDR with or without external beam
Author(s)
8
Ranges indicate spread of results with variations in tumour stage or grade; HDR and LDR, high and low dose rate brachytherapy respectively; NA, not applicable. The references for the above are not discussed further in the text.
Treatment 771
latter case neoadjuvant chemotherapy (e.g. using single agent carboplatin) may be considered. Particle beam radiotherapy should be less toxic in this situation. Younger and fitter patients can tolerate combined cytotoxic chemotherapy: ovarian cancer schedules are probably the best in this context, e.g. carboplatin and paclitaxel,58* but dose reduction may be necessary because of previous pelvic irradiation. PALLIATIVE TREATMENT
Analgesics, steroids and other supportive medications and care will be necessary in some patients. Palliative radiotherapy can be applied to painful bone metastases and can sometimes – at relatively low dosage – reduce recurrent pelvic bleeding. Small external beam fields, e.g. 10 10 cm fields or less, with or without brachytherapy cylinders, can be used for vaginal bleeding. In some instances brachytherapy alone may be effective.
Hormones The normal endometrium proliferates in response to oestrogen whereas increasing concentrations of progestagens secreted by the corpus luteum of pregnancy allow atrophy and metaplasia of the endometrium to form the decidua. Similar effects in endometrial cancer follow the use of supra-physiological doses of progestagens (e.g. Megace 160 mg daily, medroxyprogesterone acetate 200 bid or tds), with induction of apoptosis that requires intact apoptotic pathways. There is evidence of disordered apoptosis due to deficient expression of Bcl-2 and BAX proteins in nonresponding endometrial hyperplasias after progestagen exposure.59 Such treatment can result in partial and complete remission of recurrent cancers in patients with well and moderately well differentiated endometrial cancers.60 Parenteral depot preparations are sometimes useful in relief of symptoms in patients who may not be compliant or suitable for conventional oral treatment because of psychotic conditions or swallowing disorders. The potent mineralocorticoid effect can cause sodium and water retention and may precipitate hypertension, heart failure and cerebrovascular accidents. There is no evidence that adjuvant cytotoxic chemotherapy or high-dose progestagens improve survival when used with primary therapy,61***,62** although less toxic adjuvant therapies such as aromatase inhibitors have not been tested on a large scale. Studies of adjuvant progestagens following hysterectomy have in some instances shown increasing death rates in those treated versus control patients. Careful cardiovascular follow-up is essential in such patients. Obesity may be exacerbated due to appetite stimulation. Tamoxifen may sometimes cause remission and aromatase inhibitors such as anastrozole should, theoretically, also be effective. There is evidence of aromatase expression in stromal cells within endometrioid type cancers.63 Newer non-steroidal pure
anti-oestrogen compounds64 and other derivatives of tamoxifen65 are under development and may have a future preventative and therapeutic role. Another approach has been the use of gonadotrophin-releasing hormones in recurrent cancers: the response rate of 28 per cent does not depend on tumour grade, but is paradoxically highest in previously irradiated sites.66* Some authorities advocate the use of adjuvant progestagens where peritoneal cytology is positive, but there are no randomized trials that address this issue. There is also some interesting work in young patients with endometrial cancer who wish to preserve fertility. Surprisingly good results (50 per cent full term deliveries) can be achieved after medroxyprogesterone acetate and regular curettage, although tumours tend to recur but without evidence of wider spread.67* It should be ensured that endometrial cancer patients do not take full or partial agonist oestrogenic drugs while they have a risk of there being any recurrent cancer. The question of hormone replacement therapy (HRT) after successful therapy can be addressed at a suitable interval of time: there is no agreed interval at the present time. For example, after 2–3 years, if the prognosis is good and with no indications of recurrence, the patient can take an informed risk that HRT will not change her actual risk of recurrence, although it might accelerate a recurrence should she continue to harbour responsive endometrial cancer cells.
Cytotoxic chemotherapy and biological agents Younger and fitter patients with advanced tumour on presentation may benefit from initial chemotherapy and, if disease is confined to the pelvis and para-aortic nodes, this may be followed by radiotherapy. There are reports of possible improvements in survival when high-risk patients with papillary serous carcinoma of the uterus are given postoperative cisplatin/carboplatin plus cyclophosphamide.68 Palliation of metastatic disease by chemotherapy is frequently disappointing.69 Doxorubicin is reported to be the best single agent (20–40 per cent response rates) and although response rates of up to 30–50 per cent have been reported following combination chemotherapy using adriamycin (up to 60 mg/m2) and cis-platinum 50 mg/m2 given every 3 weeks, remissions are short in duration. For older patients, the use of carboplatin (IV 350 mg/m2 or AUC 5–6) should be considered, as this agent is fairly well tolerated and responses have been reported in stage IV patients.70 There are also reports of overall responses around 30 per cent using paclitaxel or docetaxel.71* There are isolated reports that addition of trastuzumab can also provide useful remissions in tumours that over express HER-2/neu.72
Very young patients Although rare, the condition can arise in premenopausal patients, where there are associations with polycystic ovaries,
772 Uterus
unopposed oestrogen oral contraceptives, the use of oestrogens in gonadal dysgenesis and inherited tumour suppressor gene mutations. The cancers are usually well differentiated with minimal invasion and in some instances progestagens may be used to induce regression if fertility is to be preserved.
FOLLOW-UP After standard adjuvant treatment, follow-up should be 3–6 monthly for 2 years then annually. For ‘surveillance’ patients, 2–3 monthly for 2 years followed by longer intervals is suggested. Vault smears can be used in surveillance follow-up, but can cause significant and disruptive diagnostic uncertainty after adjuvant radiotherapy. Where careful vaginal examination reveals an apparent recurrence, a formal incisional biopsy should be taken. Patients who have recurrent tumours should always see a clinical oncologist with an interest in this condition. CA 125 antigen levels are unhelpful apart from in some papillary serous variants.
RE-TREATMENT This can sometimes be attempted, particularly if previous treatment has not been given to a full radical dose or to a limited volume. For example, if vaginal brachytherapy had previously been used to give vault irradiation only, a recurrence in the lower third of the vagina may be treated by further brachytherapy (or a combination of external beam and brachytherapy) with activity restricted to the tumour site and appropriate margin. If the patient subsequently has tumour recurrence at the pelvic sidewall or in the lower half of the vagina, treatment may be attempted but is often of palliative intent. It may then be necessary to shield the central pelvic structures adjacent to the previously treated region. Recurrences in the lower vagina can be treated by further brachytherapy, although it may be possible to give some component of dose by external beam in addition (e.g. by use of a half beam blocked field arrangement, treating below the previously irradiated region). Alternatively, interstitial brachytherapy combined with external beam radiotherapy can be used to treat small volume vault recurrences,73* although good results have also been achieved (over 50 per cent 5-year survival rates) by external beam and brachytherapy using vaginal cylinders.74* Limited doses of further external radiation may be given to palliate bleeding and pain, although the patient must understand that there is a relatively high risk of serious normal tissue damage or reactions.
FUTURE PROSPECTS Greater use of modern imaging with conformal radiotherapy techniques including multiple coplanar beams and
IMRT may allow better normal tissue sparing, e.g. of small bowel after hysterectomy. These techniques have their own limitations and can lead to unnecessary low-dose irradiation of additional tissues especially when target volumes are large.75 CT/MRI compatible brachytherapy applicators can also be used to improve the specificity of treatment volumes but are time-consuming. Gene-specific therapies may allow effective tumour control with lower doses of radiation as well as better control of metastatic spread but so far attempts at using growth factor inhibitors have been disappointing in gynaecological cancers.76 In terms of radiotherapy, the advent of more proton and ion beam facilities offers the prospect for improved radiotherapy, particularly in inoperable patients: preliminary results with cervix cancer in Japan might be usefully extended to endometrial cancers and sarcomas.77*
SARCOMAS OF THE UTERUS These rare tumours are managed by similar general principles to those already discussed for epithelial malignant tumours of the uterus.78 There are several distinct pathological entities that mainly comprise the carcinosarcomas or mixed mesodermal sarcomas, the leiomyosarcomas and the stromal carcinomas. Grading is as for sarcomas at other sites. Myxoid leiomyosarcomas appear to be of low grade but can behave in a malignant fashion. Positive peritoneal cytology can be prognostically important as peritoneal spread may occur with greater ease. Patterns of spread may differ, but adenocarcinoma cells within a mixed carcinosarcoma are associated with a greater tendency to spread via lymphatics rather than via the blood stream. There is no proven role for adjuvant cytotoxic chemotherapy, but chemotherapy may provide palliation of symptoms caused by systemic spread. Adriamycin at a dose of up to 60 mg/m2 is the best single agent and there is insufficient evidence to support the use of combination chemotherapy. The uterine sarcomas should not be regarded as highly radioresistant, for although leiomyosarcomas are correctly included in this category, the other types may shrink in a slow but protracted manner after radical radiotherapy. Durable control of symptoms can occasionally be achieved after either conventionally fractionated or relatively low-dose hypofractionated palliative doses have been given. In general, the local recurrence rates after surgery (approximately 50–80 per cent) are significantly greater than for carcinomas, even in the case of relatively welldifferentiated sarcomas. Postoperative radiotherapy is usually given to prevent recurrence. Many retrospective series show a 15–30 per cent reduction in pelvic recurrence, but there is no definite evidence of enhanced survival. Care must be taken that there is no previous history of radiation exposure, which occurs in approximately 10 per cent of patients. Radical radiotherapy, rarely given in the case of inoperable or medically unfit patients, is reported to result in a 14 per cent
References 773
5-year survival for all stages but 44 per cent in the stage 1 category, although the numbers of patients treated are small. A particularly difficult clinical problem is that of radiation-induced sarcomas following radical radiotherapy for a cancer of the cervix or other pelvic malignancy many years previously: further surgery may be difficult and other re-treatment options (see above) or palliative approaches may be considered as the prognosis is usually poor.79 It is important to remember that low-grade endometrial stromal sarcomas may regress with high-dose progestagen therapy, as may the adenocarcinoma component of a carcinosarcoma. There is also limited clinical experience using aromatase inhibitors.80
KEY LEARNING POINTS ●
●
●
●
●
●
Because of age and co-morbidities, careful selection of patients is required for primary surgery or radical radiotherapy. After primary surgery, the indications for postoperative radiotherapy must be carefully considered. Recurrence at the vaginal vault following surgery alone is curable after radiotherapy in approximately 50 per cent of patients. Radical radiotherapy is an alternative to surgery and if carefully given has a low morbidity rate. Research for a safe form of adjuvant hormone therapy is required. Careful use of progestagens, or other alternatives including tamoxifen, can provide valuable remissions.
KEY REFERENCES Thomas R, Blake P. Endometrial carcinoma: Adjuvant locoregional therapy. Clin Oncol 1996; 8:140–5. Aalders J, Abeler V, Kolstad P et al. Postoperative external irradiation and prognostic parameters in stage I endometrial carcinoma; clinical and histopathological study of 540 patients. Obstet Gynaecol 1980; 56:419–27. Creutzberg CI, Van Putten W, Koper PC, Lybert ML et al. Surgery and post operative radiotherapy for patients with stage 1 endometrial carcinoma: multicentre randomised trial: PORTEC Study Group. Lancet 2000; 355:1404–11. Wylie J, Irwin C, Putulie M, Levin W et al. Results of radical radiotherapy for recurrent endometrial cancer. Gynaecol Oncol 2000; 77:66–72.
REFERENCES 1 Cancer Survival Trends, 1971–1995. London: The Stationery Office, 1999.
2 Sherman ME, Carreon JD, Lacey JV, Jr, Devesa SS. Impact of hysterectomy on endometrial carcinoma rates in the United States. J Natl Cancer Inst 2005; 97:1700–2. 3 Yamagami W, Susumu N, Banno K, et al. Clinicopathologic manifestations of early-onset endometrial cancer in Japanese women with a familial predisposition to cancer. J Obstet Gynaecol Res 2005; 31:444–51. 4 Olson JE, Sellers TA, Anderson KE, Folsom AR. Does a family history of cancer increase the risk for postmenopausal endometrial carcinoma? A prospective cohort study and a nested case-control family study of older women. Cancer 1999; 85:2444–9. 5 Newcomb PA, Solomon C, White E. Tamoxifen and risk of large bowel cancer in women with breast cancer. Breast Cancer Res Treat 1999; 53:271–7. 6 Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for the prevention of breast cancer: current status of the National Surgical Adjuvant Breast and Bowel Project P-1 study. J Natl Cancer Inst 2005; 97:1652–62. 7 Terashima I, Suzuki N, Shibutani S. Mutagenic potential of alpha-(N2-deoxyguanosinyl)tamoxifen lesions, the major DNA adducts detected in endometrial tissues of patients treated with tamoxifen. Cancer Res 1999; 59:2091–5. 8 Schmeler KM, Soliman PT, Sun CC, Slomovitz BM, Gershenson DM, Lu KH. Endometrial cancer in young, normal-weight women. Gynecol Oncol 2005; 99:388–92. 9 Shutter J, Wright TC, Jr. Prevalence of underlying adenocarcinoma in women with atypical endometrial hyperplasia. Int J Gynecol Pathol 2005; 24:313–8. 10 Martin JD, Gilks B, Lim P. Papillary serous carcinoma – a less radio-sensitive subtype of endometrial cancer. Gynecol Oncol 2005; 98:299–303. 11 Mundt AJ, Connell PP. Do conventional pathologic features lose their prognostic significance following postoperative radiation therapy in pathologic stage I–II endometrial adenocarcinoma? Int J Cancer 2000; 90:224–30. 12 Kadar N, Homesley HD, Malfetano JH. Positive peritoneal cytology is an adverse factor in endometrial carcinoma only if there is other evidence of extrauterine disease. Gynecol Oncol 1992; 46:145–9. 13 Larson DM, Berg R, Shaw G, Krawisz BR. Prognostic significance of DNA ploidy in endometrial cancer. Gynecol Oncol 1999; 74:356–60. 14 Tritz D, Pieretti M, Turner S, Powell D. Loss of heterozygosity in usual and special variant carcinomas of the endometrium. Hum Pathol 1997; 28:607–12. 15 Kihana T, Hamada K, Inoue Y, et al. Mutation and allelic loss of the p53 gene in endometrial carcinoma. Incidence and outcome in 92 surgical patients. Cancer 1995; 76:72–8. 16 Semczuk A, Berbec H, Kostuch M, Kotarski J, Wojcierowski J. Detection of K-ras mutations in cancerous lesions of human endometrium. Eur J Gynaecol Oncol 1997; 18:80–3. 17 Zorn KK, Bonome T, Gangi L, et al. Gene expression profiles of serous, endometrioid, and clear cell subtypes of ovarian and endometrial cancer. Clin Cancer Res 2005; 11:6422–30. 18 Lu HK, Broaddus RR. Gynecologic cancers in Lynch syndrome/HNPCC. Fam Cancer 2005; 4:249–54.
774 Uterus
19 McKeown SR, Jones B. DNA repair: therapeutic implications. Br J Radiol 2006; 79:91–3. 20 Creasman WT. Prognostic significance of hormone receptors in endometrial cancer. Cancer 1993; 71:1467–70. 21 Hicks ML, Piver MS, Puretz JL, Hempling RE, Baker TR, McAuley M, Walsh DL. Survival in patients with paraaortic lymph node metastases from endometrial adenocarcinoma clinically limited to the uterus. Int J Radiat Oncol Biol Phys 1993; 26:607–11. 22 Elsandabesee D, Greenwood P. The performance of Pipelle endometrial sampling in a dedicated postmenopausal bleeding clinic. J Obstet Gynaecol 2005; 25:32–4. 23 Tozzi R, Malur S, Koehler C, Schneider A. Laparoscopy versus laparotomy in endometrial cancer: first analysis of survival of a randomized prospective study. J Minim Invasive Gynecol 2005; 12:130–6. 24 Susini T, Massi G, Amunni G, Carriero C, Marchionni M, Taddei G, Scarselli G. Vaginal hysterectomy and abdominal hysterectomy for treatment of endometrial cancer in the elderly. Gynecol Oncol 2005; 96:362–7. 25 Kilgore LC, Partridge EE, Alvarez RD, Austin JM, Shingleton HM, Noojin F, 3rd, Conner W. Adenocarcinoma of the endometrium: survival comparisons of patients with and without pelvic node sampling. Gynecol Oncol 1995; 56:29–33. ◆26 Thomas R, Blake P. Endometrial carcinoma: adjuvant locoregional therapy. Clin Oncol (R Coll Radiol) 1996; 8:140–5. 27 Jones B, Pryce PL, Blake PR, Dale RG. High dose rate brachytherapy practice for the treatment of gynaecological cancers in the UK. Br J Radiol 1999; 72:371–7. 28 Chadha M, Nanavati PJ, Liu P, Fanning J, Jacobs A. Patterns of failure in endometrial carcinoma stage IB grade 3 and IC patients treated with postoperative vaginal vault brachytherapy. Gynecol Oncol 1999; 75:103–7. 29 Solhjem MC, Petersen IA, Haddock MG. Vaginal brachytherapy alone is sufficient adjuvant treatment of surgical stage I endometrial cancer. Int J Radiat Oncol Biol Phys 2005; 62:1379–84. ◆30 Menczer, J. Management of endometrial carcinoma with cervical involvement. An unsettled issue. Eur J Gynaecol Oncol 2005; 26:245–55. 31 Rittenberg PV, Lotocki RJ, Heywood MS, Krepart GV. Stage II endometrial carcinoma: limiting post-operative radiotherapy to the vaginal vault in node-negative tumors. Gynecol Oncol 2005; 98:434–8. 32 Weiss MF, Connell PP, Waggoner S, Rotmensch J, Mundt AJ. External pelvic radiation therapy in stage IC endometrial carcinoma. Obstet Gynecol 1999; 93:599–602. 33 Lin LL, Grigsby PW, Powell MA, Mutch DG. Definitive radiotherapy in the management of isolated vaginal recurrences of endometrial cancer. Int J Radiat Oncol Biol Phys 2005; 63:500–4. 34 Lybeert ML, van Putten WL, Brolmann HA, Coebergh JW. Postoperative radiotherapy for endometrial carcinoma. Stage I. Wide variation in referral patterns but no effect on
●35
●36
37
38
39
40
41
42
43
44
45
46
47
long-term survival in a retrospective study in the southeast Netherlands. Eur J Cancer 1998; 34:586–90. Aalders J, Abeler V, Kolstad P, Onsrud M. Postoperative external irradiation and prognostic parameters in stage I endometrial carcinoma: clinical and histopathologic study of 540 patients. Obstet Gynecol 1980; 56:419–27. Creutzberg CL, van Putten WL, Koper PC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000; 355:1404–11. Creutzberg CL, van Putten WL, Koper PC, et al. The morbidity of treatment for patients with Stage I endometrial cancer: results from a randomized trial. Int J Radiat Oncol Biol Phys 2001; 51:1246–55. Creutzberg CL, van Putten WL, Warlam-Rodenhuis CC, et al. Outcome of high-risk stage IC, grade 3, compared with stage I endometrial carcinoma patients: the Postoperative Radiation Therapy in Endometrial Carcinoma Trial. J Clin Oncol 2004; 22:1234–41. Scholten AN, van Putten WL, Beerman H, et al. Postoperative radiotherapy for Stage 1 endometrial carcinoma: long-term outcome of the randomized PORTEC trial with central pathology review. Int J Radiat Oncol Biol Phys 2005; 63:834–8. Sutton G, Axelrod JH, Bundy BN, et al. Whole abdominal radiotherapy in the adjuvant treatment of patients with stage III and IV endometrial cancer: a gynecologic oncology group study. Gynecol Oncol 2005; 97:755–63. Churn M, Jones B. Primary radiotherapy for carcinoma of the endometrium using external beam radiotherapy and single line source brachytherapy. Clin Oncol (R Coll Radiol) 1999; 11:255–62. Bond MG, Workman G, Martland J, et al. Dosimetric considerations in the treatment of inoperable endometrial carcinoma by a high dose rate afterloading packing technique. Clin Oncol (R Coll Radiol) 1997; 9:41–7. Weitmann HD, Potter R, Waldhausl C, Nechvile E, Kirisits C, Knocke TH. Pilot study in the treatment of endometrial carcinoma with 3D image-based high-dose-rate brachytherapy using modified Heyman packing: clinical experience and dose-volume histogram analysis. Int J Radiat Oncol Biol Phys 2005; 62:468–78. Bauer M, Schulz-Wendtland R. Technical note: a new afterloading applicator for primary brachytherapy of endometrial cancer. Br J Radiol 1993; 66:256–9. Patanaphan V, Salazar OM, Chougule P. What can be expected when radiation therapy becomes the only curative alternative for endometrial cancer? Cancer 1985, 55:1462–7. Wang ML, Hussey DH, Vigliotti AP, Benda J, Wen BC, Doornbos JF, Anderson B. Inoperable adenocarcinoma of endometrium: radiation therapy. Radiology 1987; 165:561–5. Fishman DA, Roberts KB, Chambers JT, Kohorn EI, Schwartz PE, Chambers S. K. Radiation therapy as exclusive treatment
References 775
48
49
50
51
52
53
54
55
56
57
●58
59
◆60
61
for medically inoperable patients with stage I and II endometrioid carcinoma with endometrium. Gynecol Oncol 1996; 61:189–96. Varia M, Rosenman J, Halle J, Walton L, Currie J, Fowler W. Primary radiation therapy for medically inoperable patients with endometrial carcinoma – stages I–II. Int J Radiat Oncol Biol Phys 1987; 13:11–15. Kupelian PA, Eifel PJ, Tornos C, Burke TW, Delclos L, Oswald MJ. Treatment of endometrial carcinoma with radiation therapy alone. Int J Radiat Oncol Biol Phys 1993; 27:817–24. Rouanet P, Dubois JB, Gely S, Pourquier H. Exclusive radiation therapy in endometrial carcinoma. Int J Radiat Oncol Biol Phys 1993; 26:223–8. Taghian A, Pernot M, Hoffstetter S, Luporsi E, Bey P. Radiation therapy alone for medically inoperable patients with adenocarcinoma of the endometrium. Int J Radiat Oncol Biol Phys 1988; 15:1135–40. Kucera H, Knocke TH, Kucera E, Potter R. Treatment of endometrial carcinoma with high-dose-rate brachytherapy alone in medically inoperable stage I patients. Acta Obstet Gynecol Scand 1998; 77:1008–12. Grigsby PW, Kuske RR, Perez CA, Walz BJ, Camel MH, Kao MS, Galakatos A. Medically inoperable stage I adenocarcinoma of the endometrium treated with radiotherapy alone. Int J Radiat Oncol Biol Phys 1987; 13:483–8. Nguyen C, Souhami L, Roman TN, Clark BG. High-dose-rate brachytherapy as the primary treatment of medically inoperable stage I–II endometrial carcinoma. Gynecol Oncol 1995; 59:370–5. Niazi TM, Souhami L, Portelance L, Bahoric B, Gilbert L, Stanimir G. Long-term results of high-dose-rate brachytherapy in the primary treatment of medically inoperable stage I–II endometrial carcinoma. Int J Radiat Oncol Biol Phys 2005; 63:1108–13. Jereczek-Fossa B, Jassem J, Nowak R, Badzio A. Late complications after postoperative radiotherapy in endometrial cancer: analysis of 317 consecutive cases with application of linear-quadratic model. Int J Radiat Oncol Biol Phys 1998; 41:329–38. Leborgne F, Fowler JF, Leborgne JH, Zubizarreta E, Chappell R. Biologically effective doses in medium dose rate brachytherapy of cancer of the cervix. Radiat Oncol Investig 1997; 5:289–99. Akram T, Maseelall P, Fanning J. Carboplatin and paclitaxel for the treatment of advanced or recurrent endometrial cancer. Am J Obstet Gynecol 2005; 192:1365–7. Vereide AB, Kaino T, Sager G, Orbo A. Bcl-2, BAX, and apoptosis in endometrial hyperplasia after high dose gestagen therapy: a comparison of responses in patients treated with intrauterine levonorgestrel and systemic medroxyprogesterone. Gynecol Oncol 2005; 97:740–50. Markman M. Hormonal therapy of endometrial cancer. Eur J Cancer 2005; 41:673–5. Martin-Hirsch PL, Lilford RJ, Jarvis GJ. Adjuvant progestagen therapy for the treatment of endometrial
62
63
64
65
66
67
68
69 70
71
72
73
74
75
cancer: review and meta-analyses of published randomised controlled trials. Eur J Obstet Gynecol Reprod Biol 1996; 65:201–7. Groups, C.-N.-U. E. C. S. Adjuvant medroxyprogesterone acetate in high-risk endometrial cancer. Int J Gynecol Cancer 1998; 8:387–91. Segawa T, Shozu M, Murakami K, et al. Aromatase expression in stromal cells of endometrioid endometrial cancer correlates with poor survival. Clin Cancer Res 2005; 11:2188–94. Labrie F, Labrie C, Belanger A, et al. EM-652 (SCH 57068), a third generation SERM acting as pure antiestrogen in the mammary gland and endometrium. J Steroid Biochem Mol Biol 1999; 69:51–84. O’Regan RM, Cisneros A, England GM, et al. C. Effects of the antiestrogens tamoxifen, toremifene, and ICI 182,780 on endometrial cancer growth. J Natl Cancer Inst 1998; 90:1552–8. Jeyarajah AR, Gallagher CJ, Blake PR, Oram DH, Dowsett M, Fisher C, Oliver RT. Long-term follow-up of gonadotrophinreleasing hormone analog treatment for recurrent endometrial cancer. Gynecol Oncol 1996; 63:47–52. Niwa K, Tagami K, Lian Z, Onogi K, Mori H, Tamaya T. Outcome of fertility-preserving treatment in young women with endometrial carcinomas. Br J Obst Gynecol 2005; 112:317–20. Bancher-Todesca D, Neunteufel W, Williams KE, Prainsack D, Breitenecker G, Friedlander ML, Hacker NF. Influence of postoperative treatment on survival in patients with uterine papillary serous carcinoma. Gynecol Oncol 1998; 71:344–7. Muss HB. Chemotherapy of metastatic endometrial cancer. Semin Oncol 1994; 21:107–13. Cook AM, Lodge N, Blake P. Stage IV endometrial carcinoma: a 10 year review of patients. Br J Radiol 1999; 72:485–8. Katsumata N, Noda K, Nozawa S, et al. Phase II trial of docetaxel in advanced or metastatic endometrial cancer: a Japanese cooperative study. Br J Cancer 2005; 93:999–1004. Jewell E, Secord AA, Brotherton T, Berchuck A. Use of trastuzumab in the treatment of metastatic endometrial cancer. Int J Gynecol Cancer 2006; 16:1370–3. Charra C, Roy P, Coquard R, Romestaing P, Ardiet JM, Gerard JP. Outcome of treatment of upper third vaginal recurrences of cervical and endometrial carcinomas with interstitial brachytherapy. Int J Radiat Oncol Biol Phys 1998; 40:421–6. Wylie J, Irwin C, Pintilie M, Levin W, Manchul L, Milosevic M, Fyles A. Results of radical radiotherapy for recurrent endometrial cancer. Gynecol Oncol 2000; 77:66–72. Ahamad A, D’Souza W, Salehpour M, Iyer R, Tucker SL, Jhingran A, Eifel PJ. Intensity-modulated radiation therapy after hysterectomy: comparison with conventional treatment and sensitivity of the normal-tissue-sparing effect to margin size. Int J Radiat Oncol Biol Phys 2005; 62:1117–24.
776 Uterus
◆76
Vaidya AP, Parnes AD, Seiden MV. Rationale and clinical experience with epidermal growth factor receptor inhibitors in gynecologic malignancies. Curr Treat Options Oncol 2005; 6:103–14. 77 Kagei K, Tokuuye K, Okumura T, Ohara K, Shioyama Y, Sugahara S, Akine Y. Long-term results of proton beam therapy for carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 2003; 55:1265–71. 78 Acharya S, Hensley ML, Montag AC, Fleming GF. Rare uterine cancers. Lancet Oncol 2005; 6:961–71.
79 Hagiwara T, Mori T, Kaku T. Development of endometrial cancer following radiation therapy for cervical carcinoma. Eur J Gynaecol Oncol 2005; 26:191–5. 80 Leunen M, Breugelmans M, De Sutter P, Bourgain C, Amy JJ. Low-grade endometrial stromal sarcoma treated with the aromatase inhibitor letrozole. Gynecol Oncol 2004; 95:769–71.
34 Cervix PETER BLAKE
Incidence and causes Pathology Clinical features of invasive carcinoma of the cervix Investigations Staging and prognostic factors
777 779 780 781 781
INCIDENCE AND CAUSES Incidence INVASIVE CARCINOMA OF THE CERVIX
World-wide, cervical cancer is the second most common female malignancy after breast cancer and far exceeds cancer of either the ovary or endometrium in incidence. However, the incidence of cervical malignancy varies widely from one country to another and between cultures and social classes within the same country. In many developing countries it is the most common cancer in females whilst in UK women it is the eleventh commonest with an incidence of only 0.9 per cent.1 In the west of Scotland, there was shown a clear association with socio-economic status with a threefold increased incidence in women from low socio-economic groups compared to more affluent women.2 In the USA, the incidence in black women is approximately twice that in white women. In the UK, gynaecological malignancies account for approximately 17 000 cases per annum of cancer in females, in contrast to breast cancer with an approximate incidence of 40 000 cases per annum. Invasive cervical cancer in the UK comprises 3000 of these – approximately half the incidence of either ovarian or endometrial cancer – vulval and vaginal cancers making up the remainder.1 PRE-INVASIVE DISEASE OF THE CERVIX
Cervical intra-epithelial neoplasia (CIN) precedes virtually every case of invasive squamous carcinoma of the cervix.
Treatment Complications of treatment Potential developments Conclusions References
784 790 792 793 794
Pre-invasive disease is commonly asymptomatic but may be detected by cytological examination of exfoliated cells. These are taken from the transformation zone of the cervix with a spatula and spread on a glass slide, or with a brush which is then shaken in a transport medium. These cells are studied microscopically using the Papanicolaou stain and an assessment made of the size, shape and mitotic activity of the nuclei and the nuclear–cytoplasmic ratio. On the basis of this assessment the cells will be reported as being normal, inflammatory or showing mild atypia, as showing mild, moderate or severe dyskaryosis or as being characteristic of invasive disease. Those patients with moderate or severe changes require biopsy of the cervix for histological assessment.3 Abnormal glandular cells are much more difficult to sample and detect on exfoliative cytology due to their position higher in the endocervical canal and their situation deep within the glandular crypts of the endocervix. Criteria for the diagnosis of glandular intra-epithelial neoplasia (CGIN) are more complex than for the squamous counterpart because of the lack of a clearly identifiable basement membrane around endocervical glands. The ease and reliability of exfoliative cytology has resulted in the establishment of many screening programmes. In some areas well organized programmes with high compliance rates have resulted in a large proportion of cases of cervical neoplasia being found in the pre-invasive stage or in the early, curable stages of invasive disease. This has reduced both the incidence of invasive cervical cancer and the mortality from the disease. In the UK, a decrease in mortality has been seen since improved coverage of the population was achieved in the early 1990s.1 Prior to that
778 Cervix
the poor compliance rate and a less well organized service had produced little overall effect.4 If screening is to have a major impact on disease incidence an effective patient call and re-call system has to exist, in addition to a high quality cytological service and adequate treatment facilities. Such a system can be difficult to operate in underdeveloped areas, where there are communication problems, and in developed areas where there is a large, shifting population of women at risk, as in metropolitan areas. The peak incidence of CIN occurs between the ages of 25 and 40, whilst the peak incidence of invasive carcinoma occurs in a group of patients approximately 10 years older.
It appears that normal women may have episodes of infection with HPV which they eradicate without neoplasia developing. It is those women who fail to eradicate HPV infection, or who are repeatedly infected, who are at risk of developing cervical neoplasia.9 HPV produces two proteins, E6 and E7, which bind to the protein p53, a product of the p53 tumour suppressor gene, and pRb, a product of the retinoblastoma gene, which is also involved in cell regulation.10 The normal function of these genes is to prevent cells with abnormal DNA from replicating, and DNA damage may be caused by several factors, most notably age and smoking. It is likely that carcinogenesis in cervical squamous cells is a multi-factorial process with HPV clearly involved in this process.
Causes of cervical neoplasia
Site of action of viruses
There are several aetiological factors associated with carcinoma of the cervix, including sexual behaviour, parity, genital wart disease and smoking. The most marked of these is the sexual behaviour of both the woman and her partner and exposure to the human papilloma virus (HPV). SEXUAL BEHAVIOUR
For many years an aetiological link between sexual intercourse and the development of cervical neoplasia was recognized.5 There is a higher incidence of both CIN and of invasive disease in girls who commence intercourse in their teens rather than at a later age. This suggests that the adolescent cervix is more vulnerable to potential oncogenic agents. In addition, the number of sexual partners appears to be important, with some studies having found a history of six or more sexual partners to be a significant risk factor. There is also an association between cervical neoplasia and the male partner having multiple other sexual partners.6 This gave rise to the concept of the ‘high risk’ male whose female partners are at a higher risk than average of developing cervical neoplasia due to the male being infected with a transmissible agent.7 A lower incidence of cervical neoplasia has been found in women using barrier methods of contraception than in those using the oral contraceptive pill. It is known that seminal fluid contains potent local immunosuppressants, against which the cervix would be protected by barrier methods. These observations support the theory that there is at least one transmissible agent involved in cervical neoplasia.
The great majority of neoplasias of the cervix arise at the squamo-columnar junction, an area known as the ‘transformation zone’, where columnar epithelium undergoes the metaplastic process of becoming squamous epithelium (Fig. 34.1). This transformation zone is larger in puberty, pregnancy and when taking the oral contraceptive pill. This area of the cervix is seen as the ‘target’ on which a transmissible agent has its effect and explains why sexual intercourse in puberty, multiparity and the use of the oral contraceptive pill are associated with cervical neoplasia. SMOKING
The products of smoking are concentrated in cervical mucus and produce damage to the DNA of cervical squamous cells. In addition, smoking decreases the population of Langerhans cells responsible for cell-mediated immunity in the cervix. Smoking is implicated in the development
Columnar (glandular) epithelium Endocervix
Transformation zone
VIRUSES
There is a clear link between certain types of papilloma virus and cervical neoplasia. Over 100 types of HPV have now been isolated and at least 20 of these infect the lower female genital tract. Several of these are associated with neoplasia but type 16 is particularly associated with squamous CIN and type 18 with CGIN.8
Ectocervix
Squamous epithelium
Figure 34.1 Diagram of the cervix, showing the ‘transformation ‘zone’, where the columnar epithelium of the endocervix meets the squamous epithelium of the ectocerix.
Pathology 779
of cervical neoplasia, both by direct carcinogenesis and by rendering the cervix more vulnerable to infection by a transmissible agent.11 IMMUNOSUPPRESSION
Systemic immunosuppression of patients who have undergone renal transplant is recognized as being a risk factor for cervical cancer. Cervical neoplasia is also more common in women with HIV, and the onset of invasive disease is an ‘AIDS defining illness’.12
PATHOLOGY Anatomy ANATOMY OF THE CERVIX
The female reproductive organs are represented in Figures 34.2 and 34.3. The uterus comprises the corpus and the cervix, which is the lower one third and which enters the vagina. The cervix is supported by the cardinal ligaments laterally, the uterovesical ligaments anteriorly and the uterosacral ligaments posteriorly (Fig. 34.4). The endocervical canal connects the vagina with the uterine cavity and is lined by columnar epithelium. The part of the cervix protruding into the vagina, the exocervix, is covered by stratified squamous epithelium. The cervix and body of the uterus are
small in childhood, enlarge during puberty and the reproductive years and then atrophy after the menopause. In addition to changing in size during a woman’s lifetime, the transformation zone also changes in position, being usually just inside the external cervical os in young women, but rising up the endocervical canal after the menopause. LYMPHATIC DRAINAGE
There is a well defined pattern for the lymphatic drainage of the cervix with direct drainage to the internal, external and common iliac nodes. Other node groups which may be involved by direct lymphatic spread are the parametrial, obturator and presacral nodes. Spread to the para-aortic nodes is uncommon without pelvic node involvement and the supraclavicular nodes may be involved subsequent to para-aortic node disease (Fig. 34.5). Pubic symphysis
Pubovesical ligament Obturator foramen
Bladder
Vesico-uterine ligament Cardinal ligament
Cervix
Ischium Rectum
Uterosacral ligament
Sciatic foramen
Uterine cavity
Sacrum
Internal os Cervix External os
Vagina
Ureter Cardinal ligament Endocervical canal Lateral fornix
Figure 34.4 the cervix.
The ligaments supporting the uterus at the level of
8
Figure 34.2 Diagram of the cervix, showing the relationship to the ureters and the cardinal ligaments (parametria).
7 6 5 3 2 1
Uterus
Bladder Cervix Anterior fornix
4
Pouch of Douglas Posterior fornix Rectum Vagina
Urethra
Figure 34.3 Lateral view of the pelvis, showing the relationship of the cervix to the bladder, rectum and pouch of Douglas.
Figure 34.5 The lymph node drainage of the cervix: (1) paracervical, (2) parametrial, (3) internal iliac, (4) obturator, (5) external iliac, (6) pre-sacral, (7) common iliac and (8) para-aortic nodes.
780 Cervix
Spread of disease Cervical carcinoma spreads predominantly by direct invasion and lymphatic permeation as previously described. Direct spread is superiorly into the body of the uterus and inferiorly into the vaginal mucosa. Laterally, the parametrial tissues, ligaments of the uterus and pelvic sidewall may be involved. Rarely, the bladder anteriorly or rectum posteriorly can be invaded by advanced disease. Spread is usually continuous, but seedlings from cervical cancer can occasionally be noted in the lower vagina. Blood-borne spread is unusual but, when it does occur, is to the lungs, bone and liver.
Pathology of pre-invasive disease Changes in the metaplastic process at the transformation zone may lead to dysplasia, which is known as cervical intra-epithelial neoplasia (CIN). The terms CIN1, CIN2 and CIN3 are histological terms used to describe increasing degrees of dysplasia from mild to moderate to severe. Cervical epithelium showing severe dysplasia (CIN3) was previously known as carcinoma in situ. It is known that a proportion of cases of CIN3 can develop into invasive carcinoma, whilst the great majority of CIN1 and most of CIN2 may revert to normal epithelium if left untreated. Most cases of CIN3 and many of CIN2 are treated. It is thought that, in most women, the transformation of benign epithelium to invasive cancer takes 10–15 years.
Micro-invasive carcinoma of the cervix Once the basement membrane beneath the epithelium is breached by the neoplastic process the disease is referred to as being micro-invasive. In the Federation Internationale de Gynecologie et d’Obstetrique (FIGO) classification13 this is stage Ia, which is subdivided into stage Ia1 and Ia2 (see Table 34.1) according to whether the depth of invasion is up to 3 mm or 5 mm respectively. When disease is more than 5 mm deep or wider than 7 mm, it falls within the category of FIGO stage Ib invasive carcinoma of the cervix.
Invasive carcinoma of the cervix Between 85 and 95 per cent of cervical carcinomas are squamous, the remainder being predominantly adenocarcinoma, adenosquamous carcinoma or, very rarely, small cell neuroendocrine tumours, sarcoma, lymphoma or melanoma. SQUAMOUS CARCINOMA
Most squamous carcinomas involve the exocervix and are visible on a speculum examination. Some develop within the endocervical canal and remain occult until reaching
quite a large size (a barrel carcinoma). Visible tumours may be either exophytic or ulcerating with underlying infiltration of surrounding structures. Histologically, tumours are graded as well differentiated (grade 1), moderately differentiated (grade 2) and poorly differentiated (grade 3). Alternatively, squamous tumours may be described as large cell keratinizing, large cell nonkeratinizing, and small cell non-keratinizing carcinomas, which equate approximately to grades 1, 2 and 3 respectively. Occasionally, well differentiated squamous carcinomas having the appearance of condylomata acuminata are seen; these are called verrucous carcinomas. ADENOCARCINOMA
Adenocarcinomas arise from the glandular epithelium lining the endocervical canal and the endocervical glands. Adenocarcinoma has often been thought to carry a worse prognosis than squamous carcinoma, but this is probably due to the origin, deep in the endocervical canal, leading to a later presentation of disease on the exocervix and an increased bulk of tumour, stage for stage. As the screening campaigns have reduced the incidence of squamous neoplasia, the incidence of adenocarcinoma has risen. This seems not solely to be due to the selective reduction in squamous neoplasia but also to a genuine increase in the incidence of glandular neoplasia, possibly associated with HPV type 18. ADENOSQUAMOUS CARCINOMA
This histological type contains both malignant squamous and adenocarcinomatous elements.
CLINICAL FEATURES OF INVASIVE CARCINOMA OF THE CERVIX Symptoms Whilst an increasing number of invasive cervical carcinomas are detected in the pre-symptomatic stages by cervical screening, the majority of patients still present with symptoms. The most common symptom of invasive cervical cancer is bleeding which may be post-coital, intermenstrual or post-menopausal. Vaginal discharge is the second commonest symptom. These symptoms should always be investigated by clinical examination, inspection of the cervix and exfoliative cytology and should lead to referral to a rapid referral unit according to the current referral criteria.14 Pain usually indicates late-stage disease.
Signs Visual inspection of the cervix may reveal it to be enlarged, have obvious growth on the surface or to be ulcerated. In
Staging and prognostic factors 781
addition, node masses may be palpable in the groins or abdomen and in the left supraclavicular fossa. Patients with advanced disease may have tumour in the vagina or vulva and may have symptoms and signs of anaemia, jaundice or uraemia.
INVESTIGATIONS Investigations for CIN SCREENING
In the UK, the recommendation in most areas in the 1970s and early 1980s was that screening should be repeated every 5 years from the age of 35–60 years. However, sexually active women younger than 35 years were also at risk and it was these younger women who were more prepared to attend for screening.15* It was recognized that screening should take place throughout a woman’s sexually active life. A 3–5-yearly smear from the age of commencement of sexual activity to the age of 70 years was advised.16 However, cost-effectiveness considerations now dictate that screening in the UK commences at 25 years and is repeated 5-yearly until 60 years.17 Cervical cytology can give rise to both falsepositive and false-negative results and confirmation of positive or unexpectedly negative cytological findings should be sought by biopsy. COLPOSCOPY
This technique of examination involves looking at the cervix with a low-power microscope and can be carried out without the need for a general anaesthetic. By staining the cervix and upper vagina with acetic acid or Lugol’s iodine, areas of abnormal epithelium can be identified and directed biopsies taken. In the case of extensive abnormalities or ones which extend up the endocervical canal and which cannot be entirely visualized, cone biopsy may be necessary.
Investigations for invasive disease CLINICAL EXAMINATION
Clinical examination of the patient should involve palpation of the abdomen to detect enlarged kidneys, liver, palpable para-aortic nodes or an enlarged bladder. The inguinal and supraclavicular areas should be examined for metastatic lymph nodes. The vulva should be inspected and then the vagina and cervix examined using a speculum. If an abnormality is seen, a smear or punch biopsy should be taken for diagnosis. The cervix should be examined bimanually (per vaginam and per rectum) to assess the size, shape and mobility of the uterus and any extension of tumour into surrounding tissues. Rectal examination gives additional
information on posterior spread along the uterosacral ligaments, lateral parametrial invasion and involvement of the pelvic sidewall. EXAMINATION UNDER ANAESTHETIC
Under anaesthetic, both abdominal and pelvic examination will be carried out in addition to cystoscopy, to rule out bladder involvement and, if there is any suspicion of posterior spread, proctoscopy and sigmoidoscopy. Biopsies should be taken from the cervix and any other abnormality in the vagina or vulva. Placing an inert radioopaque seed in the tissues adjacent to the tumour may be useful in locating the caudal extent of the tumour in the radiotherapy planning process. HAEMATOLOGICAL AND RADIOLOGICAL TESTS
Serum electrolytes, a full blood count, renal and liver function tests will give an indication of ureteric involvement, liver metastases or anaemia from tumour haemorrhage. Further examination of the renal tract can be by intravenous urography (IVU), ultrasound scan or computed tomography (CT) scan, and a chest X-ray can demonstrate metastatic disease. Nodal disease can be imaged by CT or magnetic resonance imaging (MRI). Most commonly a CT scan will be used to assess the state of the liver, renal tract and para-aortic lymph nodes. However, in the pelvis, MRI is a more effective method of imaging both primary disease and nodal involvement.
STAGING AND PROGNOSTIC FACTORS The most widely used staging system is that of FIGO. This is listed in Table 34.1. Apart from micro-invasive disease, which is defined histologically, the system depends largely on examination under anaesthetic. The results of cystoscopy, proctoscopy, chest X-ray and IVU can all be used in determining FIGO stage, but other imaging techniques do not alter tumour stage and the findings of these investigations should be incorporated into the tumour/node/metastasis (TNM) system or stated in addition to the FIGO staging. This failure of the FIGO system to take note of modern imaging findings, particularly in relation to nodal staging, is currently under discussion (early 2006) and may lead to inclusion of these findings in future FIGO staging of cervical cancer.
Stage Approximately 60 per cent of patients with carcinoma of the cervix can be cured of their disease (Table 34.2). Most recurrences occur within the first 3 years, and 5-year survival rates are a good measure of the effectiveness of therapy. More than 90 per cent of patients with small stage I
782 Cervix
Table 34.1 Staging for carcinoma of the cervix (FIGO)13 Stage O Ia1 Ia2 Ib1 Ib2 IIa IIb IIIa IIIb IVa IVb
Features Pre-invasive carcinoma (CIN) Pre-clinical carcinoma, minimal microinvasion: diagnosed by microscopy only Pre-clinical carcinoma, microinvasion 5 mm deep and 7 mm wide: diagnosed by microscopy only Carcinoma more extensive than Ia2 but confined to the uterus (including the uterine body) 4 cm diameter Carcinoma more extensive than Ia2 but confined to the uterus (including the uterine body) 4 cm diameter Carcinoma extending beyond the cervix into the upper two thirds of the vagina Carcinoma extending into the parametria but not reaching the pelvic sidewall Carcinoma involving the lower third of the vagina, but not reaching the pelvic sidewall Carcinoma extending to the pelvic sidewall or causing ureteric stenosis or obstruction Carcinoma involving the mucosa of the bladder or rectum, or extending beyond the true pelvis Distant metastasis
(a) Stage Ib (minimal disease)
(b) Stage Ib (whole cervix involved)
Table 34.2 Results of treatment of cervical carcinoma18 Stage I II III IV
Percentage of total
Five-year survival (%)
35 34 26 4
76 55 30 7
(c) Stage Ib (uterine body involved)
Figure 34.6 Stage Ib carcinoma of the cervix, showing how this one stage includes a wide range of tumour volumes from minimal disease of just over 0.5 cm3 (a) to huge tumours involving the uterine body (c). Although the former would not be stage as Ib1 and the latter Ib2, there is still a wide range of tumour volumes within each substage.
tumours, with uninvolved lymph nodes, can be cured of their disease but results remain disappointing for stage III and IV tumours, with 5-year survival rates of only approximately 30 per cent and 10 per cent respectively. Results for stage II disease are, perhaps, the most variable between treatment centres. Five-year survival rates of between 40 and 75 per cent are quoted.
for a ‘barrel’ tumour arising in the endocervical canal.20 In addition to recording FIGO stage it is also important to record tumour size or volume as measured at the time of examination under anaesthetic or, more accurately, on CT or MRI scans.
Tumour volume
Nodal status
The FIGO system divides cervical tumours into broad prognostic groups and now takes account of tumour size in early-stage disease. Stage is the single most important factor related to prognosis (Table 34.2 and Figs 34.6 and 34.7). However, tumour volume is also an important prognostic factor and a wide range of volumes can occur in any one FIGO stage.19 For example, a stage Ib tumour can be anything from 0.25 cm3, for a tumour just exceeding the criteria for stage Ia (Fig. 34.6a), to as much as 100–200 cm3
Nodal status has a profound effect on survival. Those patients having stage Ib carcinoma of the cervix with positive pelvic nodes have a 5-year survival only half that of those with negative pelvic nodes. It is very unusual for patients with involved para-aortic nodes to survive 5 years as this is commonly a marker of widespread dissemination.21 In addition to the distribution of nodes, the total number of nodes involved is also a powerful prognostic factor in surgically staged series.22*
Staging and prognostic factors 783
Stage IIa
Stage IIIa
Upper two-thirds vagina
Upper two-thirds vagina
Lower third vagina
Lower third vagina
Stage IIb (early, unilateral)
Lateral parametrium
Stage IIb (advanced, bilateral)
Lateral parametrium Medial parametrium
Medial parametrium
Stage IVa
Stage IIIb (unilateral)
Uterus
Bladder
Rectum
Figure 34.7 Stages IIa–IVa carcinoma of the cervix. Stage IIa and IIIa tumours involve the vaginal mucosa only. Stage IIb and IIIb tumours involve the parametria and pelvic side-wall, respectively. A distinction is occasionally made between medial and lateral parametrial involvement and unilateral and bilateral fixation to the pelvic side-wall. Ureteric obstruction also stages a tumour as IIIb. Stage IVA tumours involve the rectum or bladder.
Lymphovascular permeation Depth of invasion of the cervix Lymphovascular space involvement has a prognostic significance in being an indicator of the likelihood of pelvic nodal involvement.23* However, it also has a prognostic effect even in the absence of pelvic nodal disease.
Tumours invading more than 15 mm through the cervical stroma have a worse prognosis than less deeply invasive tumours.24*
784 Cervix
Histology In some series the degree of differentiation of squamous cancers has been shown to be of prognostic significance with patients with poorly differentiated tumours faring worse than those with well or moderately differentiated tumours. However, this has not been a universal finding. Adenocarcinomas and adenosquamous carcinomas have also been thought to have a worse prognosis, stage for stage, than squamous carcinomas. However, it is most probably the increased volume of tumours of these histological types that causes them to have a worse prognosis. Small cell neuro-endocrine tumours of the cervix have a particularly bad prognosis because of their propensity to lymphatic and blood-borne spread, even at an early stage. As at other sites of the body, survival from this type of tumour is rare.
HPV status It has been found that patients with squamous carcinoma of the cervix that is positive for HPV, in which the p53 tumour suppressor gene is blocked, have a better prognosis than patients who are HPV negative but who express an abnormal form of p53. This implies that a tumour arising from cells with an intrinsic abnormality removing all effective function of p53 behaves in a more malignant manner than a tumour in which normal p53 is still present but is blocked from action. It may be that such blocking of the p53 gene by HPV proteins, E6 and E7, is not complete.
Age In some historical series young age was shown to convey a survival advantage over old age whilst in other series no difference was found between different age groups.25 Furthermore, a poor prognostic effect of young age has also been seen.26 The impression is that cancer of the cervix in young women is occasionally a disease of rapid onset with a high propensity for metastasis and rapid progression.2
narrow surgical margins 4 mm with the presence of any two indicating a need for adjuvant therapy.
TREATMENT Cervical intra-epithelial neoplasia PREVENTION WITH VACCINES
Vaccines to parts of the HPV virus types 16 and 18 or its products E6 and E7 are now being developed. The intention of these is to prevent infection of the female genital tract with HPV and therefore prevent the formation of CIN. Early results using artificial virus-like particles are very encouraging and these will enter clinical use in the latter half of the 2000–10 decade. At the moment the existing vaccines are thought only to be effective if given to the girl before exposure to HPV, raising issues of consent to vaccination in pre-pubertal girls.27 In addition the vaccines seem to be less effective in preventing boys from contracting the virus. Ultimately, the ideal vaccine would protect girls from CIN and reverse the neoplastic process in girls already infected. It would also protect boys from penile cancer and reduce the incidence of the virus in the community, thereby reducing the risk to non-immunized individuals as well. TREATMENT
Once diagnosed histologically, cervical intra-epithelial neoplasia may be removed under colposcopic control by surgical excision by scalpel, cutting laser or hot wire loop (LLETZ). Excision is favoured over destructive methods as it allows more accurate histological assessment of the entire abnormality than is possible on a biopsy alone. This particularly relates to occult foci of invasion which, if not recognized, may lead to undertreatment. Lesions that cannot be delineated on all sides, especially if they involve the endocervical canal, must be removed by cone biopsy (Fig. 34.8).
Prognostic indices and scores The Gynecologic Oncology Group has defined a system of assigning scores to a number of prognostic factors, mostly histological, to produce an overall prognostic score. This is particularly relevant for patients treated by surgery, where the score may be used to define the need for postoperative adjuvant chemoradiotherapy. The factors taken into consideration are lymphovascular space invasion, clinical tumour size and depth of invasion.24 Many oncologists use a simpler system with absolute indicators for postoperative therapy being residual disease, positive operative margins or lymph node involvement. Relative indicators are lymphovascular permeation, depth of stromal invasion and
Cervix Transformation zone
Anterior
Stitch to mark orientation
Glandular epithelium
Squamous epithelium
Left
Right Posterior
Figure 34.8 A cone biopsy, cut either by laser or scalpel, must include all the transformation zone and be marked to allow orientation of the specimen in the pathology laboratory.
Treatment 785
Stage Ia micro-invasive disease Cone biopsy, completely excising a Ia1 lesion, should be adequate treatment, providing that the depth of invasion is less than 3 mm with no lymphatic vessel involvement. If a cone biopsy cannot completely encompass a lesion, then simple hysterectomy should be undertaken with conservation of the ovaries in women not concerned about retaining fertility, or radical trachelectomy in those wishing to retain the chance of further pregnancy. If the lesion invades further than 3 mm (stage Ia2), or if there is lymphovascular invasion, then the risk of involved lymph nodes rises and radical hysterectomy and lymphadenectomy is the treatment of choice for women not concerned about fertility. Where this is an issue, a large cone biopsy may suffice or a trachelectomy, but the pelvic lymph nodes should be assessed surgically either at an open retroperitoneal procedure or via a laparoscope. For older women or for those unfit for radical surgery, radiotherapy is an option.
Invasive cervical cancer The treatment of invasive cervical cancer will depend on the stage of disease, the size of the tumour and the fitness of the patient. It can incorporate chemotherapy, radiotherapy and surgery. The treatment strategy for individual patients should be arrived at after discussion between specialists in all three disciplines.
preserve their fertility and whose tumours are small and limited to the cervix. This is radical trachelectomy, in which the cervix and paracervical tissues are removed together with a cuff of vagina. The body of the uterus is left in place and is anastomosed to the vagina. The new opening into the uterine cavity is narrowed with a circumferential stitch. This allows a pregnancy to be carried but requires delivery by caesarean section. The technique should be regarded as experimental as long term results of tumour cure and successful pregnancies are still awaited. The procedure is usually accompanied by a pelvic lymphadenectomy carried out laparoscopically.
RADIOTHERAPY
For older women, or women with more bulky tumours, radiotherapy is the treatment of choice as the results of treatment are equal to those of surgery and the treatment is better tolerated. However, there is morbidity in terms of fibrosis in the normal tissues causing some reduction in the size of the vagina and its pliability and lubrication. Women with a high risk of nodal involvement because of large volume disease, poor differentiation or lymphatic vessel involvement should also be treated by radiotherapy. The addition of weekly cisplatin 40 mg/m2 to the radiotherapy regimen has become standard following the results of several trials in the late 1990s. Cisplatin improves disease-free survival rates by 10–15 per cent in early-stage disease.29**,30**,31**,32,33 Overall treatment time should not exceed 56 days or a decrease in effectiveness of approximately 1 per cent per day will occur.34
SURGERY
Surgery is the treatment of choice for young patients with small volume stage Ib disease in whom there is a low risk of nodal metastases. Good histological differentiation and the absence of lymphatic vessel invasion, small tumour volume and normal-sized nodes on MRI are indicators that lymph node metastases are unlikely and surgery should be considered. Pelvic lymph nodes may be assessed by laparoscopy, and para-aortic nodes also biopsied if there are positive pelvic nodes.26 Surgery for invasive cervical carcinoma should include a radical hysterectomy and a pelvic lymphadenectomy (Fig. 34.9). This is often colloquially called a ‘Wertheim’s’ hysterectomy, although this is a misnomer as Wertheim’s original operation did not include a lymphadenectomy. A long vaginal cuff should be taken as a routine in this operation, but in a young woman with a squamous carcinoma it should be possible to conserve the ovaries and avoid the menopause. Advantages of surgery over radiotherapy include the avoidance of further shrinkage of the vagina after treatment, and the maintenance of pliability and lubrication of the vaginal mucosa. In addition, the small risk of late induction of a second malignancy is avoided. A less extensive radical surgical technique is being developed for women of child-bearing age who want to
EXTERNAL BEAM THERAPY
Carcinoma of the cervix is treated by a combination of both external and intracavitary radiotherapy. The volume encompassed by external radiotherapy should include the
Margins of a radical hysterectomy
Ureter
Figure 34.9 A radical hysterectomy in which the ureters are mobilized to allow wide excision of the uterus and cervix. The cardinal ligaments and a wide cuff of vagina are also removed. A pelvic lymphadenectomy would usually accompany this procedure.
786 Cervix
L4/L5 junction
100% 50% 80%
100%
90%
Lateral field no.2
Lateral field no.1
Anterior field
Field margins Bottom of obturator foramen
Figure 34.10 The anterior external-beam radiotherapy field used to treat carcinoma of the cervix. To cover the ‘first station’ lymph nodes, the field extends from the junction of the fourth and fifth lumbar vertebrae to the bottom of the obturator foramina.
Figure 34.11 The final plan, using an anterior and two wedged lateral fields, to deliver a homogeneous dose to the tumour volume to include the primary tumour and the draining lymph nodes.
be some prospect of long term disease control. There is no clear evidence that prophylactic para-aortic irradiation is of benefit.35** INTRACAVITARY BRACHYTHERAPY
loco-regional nodes with the primary tumour (Fig. 34.5). This will commonly be from the junction of the 4th and 5th lumbar vertebra to the bottom of the obturator foramina, and laterally to 2 cm outside the bony margin of the pelvis (Fig. 34.10). If it is appropriate to try to spare the posterior half of the rectum, in the absence of uterosacral ligament involvement, this volume can be encompassed by either a parallel opposed pair, a 4-field ‘box’ technique or a technique using three fields with an anterior and two wedged lateral fields. The volume may be more individually designed with shielding of the upper corners of the antero-posterior fields to protect small bowel and laterally to shield the posterior half of the sacrum. More individualized treatment can be achieved with CT conformal planning where the clinical tumour volume is the primary tumour, cervix, body of uterus and the nodal areas as defined by the major blood vessels if the nodes are not visible (which is usually the case). Ideally an 8–10 MV linear accelerator should be used, in view of the depth of the tumour volume below the surface of the lateral fields (Fig. 34.11). Those patients with disease in the vaginal mucosa below the upper third should have the field extended to cover the full length of the vagina. This will involve marking the introitus with a lead marker inserted just inside the labia at the time of simulation. If para-aortic nodes are involved then a long, ‘spadeshaped’ field may be used to cover the para-aortic nodes (usually with an upper border of T12/L1) and the pelvis. This volume is treated by a parallel opposed pair to a maximum dose of 45 Gy in 1.8–2 Gy fractions. Large volumes of small bowel will seldom tolerate doses in excess of this and treatment is, therefore, largely palliative. The borders of the para-aortic field should cover all enlarged nodes but should not encroach upon the kidneys, as outlined on an IVU, unless this cannot be avoided and there is thought to
Historically, cervical cancer was one of the first tumours to be treated by radiotherapy, when radium was inserted into the endocervical canal and upper vagina to irradiate local disease. A technique commonly used these days is the ‘Manchester’ technique involving an intra-uterine tube and two vaginal ovoids placed in the lateral vaginal fornices (Figs 34.12 and 34.13). The proportions of radio-isotope, initially radium and more latterly caesium, within the intrauterine tube and the vaginal ovoids, were calculated to give a constant dose rate to a geometrical point ‘A’ when using different lengths of intrauterine tube and different sizes of vaginal ovoids. Point ‘A’ was originally defined as being 2 cm above the lateral fornix of the vagina, identified on check films as being the top of the ‘ovoid’, and 2 cm lateral to the axis of the intrauterine sources. With modern afterloading applicators the ‘ovoids’ may be radiolucent and may not be clearly seen on the check radiographs. It has become common practice, therefore, to position point ‘A’ 2 cm above the cervical flange on the intrauterine tube at the cervical os and 2 cm lateral to the midline of the intra-uterine applicator. However, calculating the position of point ‘A’ in this way can lead to high dose rates at point ‘A’ if the cervix is long and the ovoids are high in relation to the flange. Judgement must be used in these situations in repositioning point ‘A’ more cranially whilst ensuring that the rectal dose remains within the tolerance range of two thirds of the point ‘A’ dose. Active sources The radio-isotope initially used for intracavitary brachytherapy was radium which, because of its gaseous daughter product radon, is hazardous. In most developed countries radium has been replaced by caesium which decays to solid, non-radioactive products. The hazards of handling active sources have led to the development of afterloading
Treatment 787
Intra-uterine tube Vaginal applicators Intra-uterine tube Bladder Vaginal applicators
Afterloading catheters Rectum
(a)
Afterloading catheters
(b)
Figure 34.12 (a) The arrangement of brachytherapy sources in the uterus and vagina for the treatment of cervical carcinoma. The sources may be active or, more usually these days, afterloaded into the applicators along catheters protruding from the vagina. (b) A lateral view of an intra-uterine tube and vaginal applicators for the treatment of carcinoma of the cervix.
Table 34.3 ICRU definition of intra-cavitary dose rate to prescription point36 Dose rate Low Medium High
(a)
0.4–2 Gy/h 2–12 Gy/h 0. 2 Gy/min
(b)
Figure 34.13 The classic ‘pear-shape’ isodose curves produced in the lateral plane by ‘Manchester’ tube and ovoids (a). In the antero-posterior plane (b) there is little expansion of the isodose curves at the level of the cervix.
techniques which minimize source handling and staff exposure. Afterloading brachytherapy The basis of afterloading brachytherapy is that hollow applicators are placed within the cervix and vaginal fornices. The radio-isotope is only introduced into these when the applicators are correctly positioned, check radiographs have been taken and the patient is comfortable and in a radiationprotected environment. The sources may then be inserted by remote control. In remote afterloading systems the sources are driven into the applicators either pneumatically or on the end of cables. Cable-driven systems are generally motor-driven although there are a few hand-driven systems for use in areas
where the power supply is erratic. This process can only take place in a protected room when all staff are away from the area. Although remote systems have the advantage of complete protection of staff, they do have the disadvantage of cost and frequent source change (3-monthly for 191Ir). Low dose rate afterloading Remote afterloading systems allow the dose rate of brachytherapy to be increased. Classically, the dose rate with the Manchester radium system was approximately 50 cGy/h to point ‘A’. Caesium-137 pellets can be produced which will allow a dose rate of between 150 and 200 cGy/h to point ‘A’. The intracavitary component of mixed external beam and intracavitary treatment would seldom take more than one day at these higher dose rates and greatly improves the cost-effectiveness and patient acceptability of the treatment. Many systems now use sources that deliver a higher than standard dose rate whilst remaining in the low dose rate range as defined by the International Committee on Radiological Units36 (Table 34.3). Whilst this has the advantage of reducing treatment time it does have radiobiological
788 Cervix
consequences, necessitating a small reduction in dose of 10–15 per cent.37 Low dose rate brachytherapy is on the decline due to a reduction in the manufacturing of low dose rate equipment and caesium sources. It is likely that high dose rate systems will become the usual method of brachytherapy delivery over the period 2005–15. High dose rate afterloading If the concept of increasing dose rate is taken further, then high dose rate brachytherapy, using high activity 191Ir, delivering doses at rates in excess of 1 Gy/min to point ‘A’, gives the opportunity of very short treatment times. This allows complete geometrical stability of the applicator during the treatment and the possibility of a high patient throughput. High patient throughput has obvious advantages in areas of high incidence of cervical cancer but requires that there is a full and efficient infrastructure for the treatment process. In addition to the provision of a remote afterloading HDR facility, including machine, protected treatment room, control room and dosimetry facilities, consideration must be given to patient transport, waiting and recovery areas, the availability of staff and rapid treatment planning if a high throughput is to be realized. Geometrical stability means that, if the applicator is well positioned for the treatment and clamped in place, there is no change in the position during the few minutes of treatment. This has advantages, in terms of normal tissue sparing, over a low dose rate insertion lasting several hours, during which time both the applicators and the tissues move. Equally, a bad insertion will remain bad throughout the short treatment time, whereas the more protracted low dose rate insertion may allow some of the normal tissues in the high dose regions to move and thus be spared damage. There is considerably less time for repair of radiation damage in the normal tissues in a high dose rate treatment compared to during the continuous treatment given by a low dose rate brachytherapy implant. Therefore, such treatments have to be fractionated over several days to allow repair in normal tissues between fractions. Apart from the operational consequences of this, in terms of the patient needing several treatment sessions instead of one or two, it also means that the applicators have to be positioned consistently for each treatment, in order to take advantage of the geometrical stability. In addition, radiobiological models of repair predict that a dose reduction of 35–45 per cent from the dose that would be given at standard dose rates needs to be made in order to avoid excessive late normal tissue damage. This need for a dose reduction at high dose rates, together with the need to fractionate treatment over several days or weeks, introduces another area of uncertainty into intracavitary brachytherapy, a treatment modality that has almost entirely been developed empirically by years of clinical experience rather than by radiobiological considerations. However, early clinical results show no difference between treatment at low dose rate and high dose rate in
terms of local control of disease or late normal tissue complications.38,39** This is in spite of mathematical modelling indicating a theoretical increased risk of late normal tissue damage, for the same anti-tumour effect, from fractionated high dose rate brachytherapy compared to a continuous low dose rate insertion. This lack of difference may be due to the radiobiological disadvantages of high dose rate therapy being balanced by the geometrical and dosimetric advantages of a rigid applicator system. The integration of external beam and intracavitary brachytherapy External beam therapy is used prior to intracavitary brachytherapy and the brachytherapy dose is designed so as to avoid the need for any central shielding in the external beam fields. Typically, a dose of 45–50 Gy would be given to the pelvis over 41⁄2 to 51⁄2 weeks in 1.8 Gy or 2 Gy fractions as described earlier. An intracavitary insertion would then be undertaken to give a further 25–30 Gy to point ‘A’ at standard dose rate. If higher than standard dose rates were used, then a lower dose would be delivered. Parametrial boosts may then be used to give a further 5 Gy to bulky disease within the parametria or involved nodes on the pelvic sidewalls. These boost fields may have the medial edge, abutting the intracavitary treatment volume, shaped by rectangular blocks, custom-made blocks to match the intracavitary isodose curves or by a multi-leaf collimator to achieve approximately the same matching. The dose to the rectum from the intracavitary insertion should not exceed two thirds of the dose given to point ‘A’ and care must be taken, when using rigid applicators, that the uterus is not forcibly pushed posteriorly in the pelvis against the rectum, sigmoid colon or small bowel. Intrauterine applicator tubes should either be made of flexible plastic, to conform to the curvature and anteversion of the uterus or, if made of metal, should have a forward curvature in the antero-posterior plane. Equally, an overly curved, rigid intra-uterine tube could overdose the dome of the bladder whilst sparing the bowel. A forward curvature of 20–30° is probably suitable in most cases, depending on the patient’s anatomy. The maintenance of haemoglobin levels during radiotherapy Studies in the late 1970s showed that women with a haemoglobin level below 10 g/dL had a lower cure rate of cervical cancer by radiotherapy than women with a higher haemoglobin level. Recent analysis of the chemoradiotherapy trials has indicated a similar influence of anaemia. Women with a haemoglobin level above 12 g/dL fared better than those with a lower level and this appeared to be independent of tumour stage and size. This effect would therefore appear to be due to a radiosensitizing effect of oxygen on the tumour40 and it is now recommended that the haemoglobin level is kept above 12 g/dL throughout radiotherapy.41*
Treatment 789
Chemotherapy for carcinoma of the cervix Carcinoma of the cervix is not a highly chemosensitive tumour.42 Response rates to single agents have seldom been reported in excess of 40 per cent and the most effective drug appears to be cisplatin. However, the taxanes may have a similar level of activity when combined with carboplatin 3-weekly or used alone on a weekly basis.43 Currently, other agents under trial include gemcitabine and docetaxel. These drugs may be used alone or in combination with others for recurrent disease,44 in the neoadjuvant setting prior to surgery or radiotherapy45** and during radiotherapy as concomitant therapy for advanced, bulky disease.32 The best response rate to chemotherapy is seen in primary untreated disease (⬃70 per cent). However, the response rate in metastatic disease is lower (⬃50 per cent) and in recurrent disease, within an irradiated area, is very low indeed (⬃15 per cent).46 Concomitant chemoirradiation In 1999 the National Cancer Institute issued a statement that ‘serious consideration’ should be given to the use of concomitant chemoirradiation in the radical treatment of locally advanced carcinoma of the cervix.32 This statement was based on three published papers and two abstracts reporting randomized controlled trials of therapies including chemoirradiation in at least one arm;29**,30**,31** all showed a statistically significant improvement in survival for patients in the chemoirradiation arm in comparison to the standard therapy arm. However, follow-up varied widely and toxicity data were reported in different ways. No study compared primary radiotherapy alone with the same regimen plus concomitant chemotherapy, and a wide range of drugs, doses and schedules were used in the five trials. Only three of the five trials related to radical non-surgical treatment alone and the other two were regimens delivered either pre- or post-surgery. Despite this flimsy evidence there was a major change in practice in this area and concomitant chemoirradiation was adopted as standard therapy with cisplatin given prior to several of the radiotherapy fractions. The frequency of administration has varied between once per week in the first four weeks to twice per week throughout the whole course of radiotherapy. In general the weekly dose of cisplatin is 40 mg/m2.33 Toxicity data suggest that this addition of cisplatin has an effect on the haematological toxicity of treatment and that failure to correct anaemia may reverse the benefit of chemoradiotherapy.46 Some patients have experienced worse gastrointestinal toxicity than would be expected with radiotherapy alone but, in general, this change in practice has not been accompanied by the increase in toxicity that was feared.
chemoradiotherapy, despite the evidence base for this being in the treatment of squamous cancers. Small cell neuroendocrine tumours of the cervix behave like small cell tumours at any other site. The most effective treatment is with chemotherapy (e.g. cyclophosphamide and etoposide) followed by radiotherapy, surgery or both to the sites of bulk disease. There is usually a dramatic and rapid response but the disease metastasizes and recurs equally rapidly. Therefore, the prognosis is very poor. THE INCIDENTAL FINDING OF CERVICAL CANCER
Occasionally invasive cervical cancer is found in the specimen following a simple hysterectomy. If the depth of invasion indicates a risk of lymphatic spread (stage Ia2 or more, or if there are other poor prognostic features), then postoperative pelvic chemoradiotherapy should be prescribed. The pelvic nodes should be imaged or undergo biopsy as positive nodes will require surgery or a higher dose of radiotherapy than is used adjuvantly. If the cuff of vagina is inadequate then vault brachytherapy also should be delivered. With this technique the results of treatment are not worse than radical surgery. Although the pelvis tolerates radiotherapy less well after surgery than when radiotherapy is given as primary treatment, it is usually possible to deliver 40–50 Gy over 5–6 weeks postoperatively, usually with concomitant cisplatin, without undue late morbidity. If there is residual disease at the vaginal vault or at the margins of excision, then a second phase of treatment is needed to take the total tumour dose to at least 60–65 Gy. In this circumstance it must be accepted that there is a greater likelihood of late morbidity. This may be minimized by restricting the second phase to the smallest volume of tissue necessary to encompass the tumour using a CT planned small volume of external beam therapy or intracavitary or interstitial brachytherapy, depending on the site and size of residual disease. CERVICAL CARCINOMA DURING PREGNANCY
This difficult situation arises uncommonly and treatment depends on the wishes of the parents as well as on the stage of disease. Treatment would be similar to that for nonpregnant patients in the first and second trimester of pregnancy, with treatment preceded by abortion or hysterotomy. Caesarean section should precede treatment in the third trimester when there is the chance of producing a viable child. The commonly perceived view that carcinoma of the cervix behaves more aggressively in pregnant patients than in non-pregnant patients is not proven in long term studies of age-matched patients.
Special situations HAEMORRHAGE NON-SQUAMOUS CARCINOMA OF THE CERVIX
Adenocarcinoma and mixed carcinomas of the cervix are treated similarly to squamous cancers, often with
Carcinoma of the cervix can present with severe haemorrhage which may require firm vaginal packing, bed rest and a blood transfusion. Tranexamic acid, 500–1000 mg tds,
790 Cervix
may be helpful in stopping bleeding. Urgent external beam radiotherapy may well produce haemostasis within 24–48 hours. Classically, large fractions of at least 4 Gy have been used, without any clear evidence that this is superior to standard-size fractions. In intractable cases, hysterectomy or embolization or ligation of the internal iliac arteries should be considered. RECURRENT CARCINOMA OF THE CERVIX
Recurrence following either inadequate or radical surgery can be treated by chemoradiotherapy if a dose to the tumour in excess of 63–65 Gy can be achieved. This usually involves at least a two-phase approach to treatment with external radiotherapy followed by a CT planned small volume boost, intracavitary brachytherapy or interstitial brachytherapy. Occasionally, recurrence after radical radiotherapy, if central, can be treated by either posterior, anterior or total pelvic exenteration involving diversion of the urinary and gastrointestinal tracts. Selection of patients for this procedure must include both the physical and psychological assessment of the patient’s ability to cope with the resulting stomas. Inoperable recurrent carcinoma of the cervix within a previously irradiated area has been referred to earlier as an indication for considering chemotherapy. In general these tumours do not respond objectively although there may be some subjective improvement. Therefore, with careful patient selection, chemotherapy may be used for palliation and, in a small number of patients, long term control of disease can be achieved. As cure of recurrent disease by chemotherapy is virtually unheard of, it is important to consider the toxicity of treatment in designing a palliative regimen for a patient.
COMPLICATIONS OF TREATMENT Surgery In addition to the morbidity and mortality associated with any major pelvic operation, radical hysterectomy can also cause specific problems in the pelvis. The most common of these is a degree of flaccidity of the bladder. This symptom is apparent soon after surgery and often improves with bladder drainage over a period of a few weeks. Rarely the bladder remains flaccid and the patient has to practise intermittent self-catheterization. Other, more serious problems that can arise are fistulae between the vagina, bladder, ureter or rectum. Very rarely such fistulae may also involve the small bowel. These problems are much increased in incidence if surgery is as salvage for recurrence after radical radiotherapy. Pelvic lymphoceles may collect after a lymphadenectomy and may be asymptomatic or cause problems of pain or of obstruction to the ureters and bowel. If they become infected then eradication of the infection may be difficult and the risk of lower limb oedema is increased.
Radiotherapy Treatment regimens are designed with the assumption that all patients have a similar sensitivity to radiation although we know this not to be the case as a wide range of both acute and late reactions are seen at standard doses. The prediction of individual normal tissue sensitivity is in its infancy but, when developed, should allow a degree of ‘tailoring’ of radiotherapy regimens to suit individual patients. In the meantime, only crude criteria can be used to predict longterm toxicity, with the exception of those very rare patients with conditions such as telangiectasia-ataxia or xeroderma pigmentosum, who are highly radiosensitive. Of particular importance is smoking, which can not only reduce the chance of cure but also increase the risk of severe late effects.47 Other crude criteria include identifying those who sunburn and blister without tanning and those with rheumatoid arthritis or other connective tissue disorders, who may have a defect of DNA damage repair and who may show exaggerated radiation sensitivity.48 In addition, those patients with a history of pelvic inflammatory disease or diabetic vascular changes and those with inflammatory bowel disease, such as Crohn’s disease and ulcerative colitis, may also show excessive radiation reactions in the bowel. EXTERNAL BEAM THERAPY
Acute effects At the dose levels necessary to treat carcinoma of the cervix with curative intent it is uncommon not to cause some acute radiation reaction, usually in the bowel. In most patients this results in diarrhoea and urgency, which becomes apparent in the second or third weeks of a course of treatment. With fraction sizes of 1.8–2 Gy, diarrhoea can usually be controlled by a low roughage diet and an anti-motility drug such as codeine phosphate or loperamide, without disruption to the course of treatment. Some patients find a bulking agent helpful. When diarrhoea cannot be controlled, or if there is passage of copious mucus or blood, fraction size may have to be reduced or the treatment suspended. Radiation cystitis occurs less commonly than bowel disturbance and must be distinguished from infection by culture of a specimen of urine. Treatment of any infection and the maintenance of a high urine output of an alkaline pH, which can be achieved with oral potassium citrate, may help to relieve symptoms. With modern megavoltage radiotherapy it is unusual to cause skin reactions that are sufficient to warrant a delay in treatment. However, moist desquamation of the skin in the natal cleft may develop. Topical steroids can help before the skin is actually broken but, once this happens, soothing lotions, Geliperm and antibacterial preparations may be necessary. Late effects Late radiation effects usually become apparent months or years after treatment. Most arise gradually but some may
Complications of treatment 791
occur acutely following another pelvic event, such as a ruptured bowel diverticulum or appendicitis. Severe late effects should not occur in more than 6–8 per cent of patients treated with radical radiotherapy.49* However, in up to 50 per cent of patients, bowel habit may be permanently altered towards looseness and increased frequency of evacuation. This may be due to decreased absorption of water in the large bowel, malabsorption of bile acids from the bowel contents, bacterial overgrowth or a number of other factors, and specialist investigation is required to determine the cause.50 Some patients have a permanently restricted diet and should be referred to a dietician for specialist advice. Nondietary methods include medication with codeine phosphate or loperamide to decrease bowel motility, Questran which binds to bile acids in the bowel, or antibiotics for bacterial overgrowth. Over a long period of time, malabsorption of vitamin B12 from the terminal ileum may become apparent by an increasing mean cell volume (MCV), anaemia or, very rarely, neurological signs. If these symptoms occur then serum B12 and folate levels should be estimated and any deficiency treated with parenteral B12 and oral folate. Rectal bleeding may result from either ulceration of the bowel mucosa or from telangiectasia. Recurrent cervical cancer and primary neoplasms of the bowel should be excluded by colonoscopy. Treatment of rectal bleeding may be attempted by sucralfate enemas, and tranexamic acid 500–1000 mg tds may help in intractable cases. However, in a very small number of patients the only effective solution is ablation of the abnormal vessels by argon beam or formalin instillation. Radiation fibrosis may also cause stenosis or obstruction of the intestine, most commonly of the small bowel or sigmoid colon, and may occur at more than one level. Surgery to remove the affected bowel and anastomose the remaining lengths is skilled, as the vascularity of irradiated bowel is often poor, making the healing of anastomoses difficult and risking ‘short bowel’ syndrome of malnutrition and dehydration. Late damage to the bladder may result in only a small volume of urine being tolerated. This can be due not only to fibrosis in the bladder, restricting expansion, but also to a neurological disturbance resulting in lower pressures within the bladder triggering the desire to micturate. Haematuria may be microscopic and thought to be due to infection. Treatment of severe or frequent haematuria with tranexamic acid 500–1000 mg tds may help once it is confirmed not to be due to recurrent tumour or to a new primary bladder cancer, but formalin instillation or urinary diversion may be needed. Radiotherapy is always listed as a possible cause of ureteric stenosis. Unfortunately, this sign is far more often caused by recurrent tumour on the pelvic sidewall compressing or invading the ureter. Avascular necrosis of the femoral heads is uncommon with modern high energy radiotherapy. High energy X-rays
are not absorbed to a significantly greater degree by bone than soft tissue, unlike lower energy orthovoltage X-rays. However, fine ‘pelvic insufficiency fractures’ can lead to pelvic pain and may be misdiagnosed as either bony metastatic disease or direct invasion, both of which are uncommon. Lower limb oedema is seldom due to radiotherapy alone and is often a symptom of recurrent tumour in the pelvis or of a deep vein thrombosis. However, this symptom may occasionally be seen due to a combination of radiotherapy and radical surgery, especially if the latter was associated with infection or with pelvic lymphoceles. INTRACAVITARY THERAPY
Intracavitary brachytherapy produces a dose distribution which falls off rapidly with distance from the sources. Small changes in the relationship of normal tissues to the applicators may produce very large changes in the dose received by those tissues. Doses may be so large as to cause necrosis, which may not be functionally important in some tissues, such as the cervix itself. However, it may provide a source of infection in devitalized pelvic tissues, which may produce an unpleasant discharge. Necrosis can be disastrous if it occurs in the vaginal mucosa, rectum, bladder or terminal ureter. Attention to the accurate placement of the applicators and packing, together with a restriction on the total dose delivered to the sensitive tissues, such as the rectum and bladder, should help avoid these serious problems. Guidelines on these aspects of treatment are detailed earlier in this chapter. Dose reduction factors, to take account of non-standard dose rates, must also be used.
Chemotherapy The toxic effects of cytotoxic drugs in the treatment of cancer of the cervix, mainly the platinum agents, taxanes, methotrexate and bleomycin, are no different from those caused in the treatment of other malignancies. The main toxicities are renal, haematological and neurological. RENAL TOXICITY
The platinum agents and methotrexate are excreted, unchanged or as toxic metabolites, in the urine. Adequate renal function is essential to ensure that systemic effects are not exaggerated by delayed excretion. In addition, all three drugs can themselves cause damage to the kidneys or urothelium and worsen renal function. To avoid further damage to the renal system, cisplatin and carboplatin should only be given if a measure of the glomerular filtration rate, by creatinine clearance or EDTA clearance, shows there to be near-normal function. If not, then the dose of platinum agent should be reduced and individual chemotherapy protocols have criteria for this.
792 Cervix
Table 34.4 Dose alteration of cisplatin, methotrexate, bleomycin (PMB) according to renal function (GFR) Creatinine or EDTA clearance (mL/min) 80 60–79 40–59 25–39 25
Cisplatin and methotrexate
Folinic acid
100% 100% 50% 25% None
15 mg qds 3 d 15 mg qds 4 d 15 mg qds 4 d 15 mg qds 4 d
An example for criteria for dose reduction is shown in Table 34.4 for a regimen consisting of cisplatin, methotrexate and bleomycin. Methotrexate can damage renal tubules by crystallizing in acid urine and physically disrupting them. If the urine is maintained alkaline, by the oral administration of sodium bicarbonate, methotrexate is fully soluble and no crystals are formed. MYELOSUPPRESSION
Carboplatin, the taxanes and methotrexate can cause myelosuppression. This toxic side effect of methotrexate can be avoided by the administration of folinic acid, which must be given for sufficiently time to ensure that the serum level of methotrexate has dropped to a safe level, usually 48–72 hours. If there is a risk of the excretion of methotrexate being delayed by poor renal function then serum methotrexate levels should be estimated and folinic acid continued until this has fallen into the safe range of 0.05 micromoles/L.
Whilst unopposed oestrogens may be used safely in patients who have undergone a hysterectomy, those who have not should be prescribed a preparation with both oestrogenic and progestogenic activity. Although radical radiotherapy ablates the endometrium in the great majority of patients, a few will still have withdrawal bleeds.
Sexuality Both the diagnosis and the treatment of carcinoma of the cervix can have a devastating effect on a patient’s sexuality and sexual functioning. Many patients fear that the cancer has been caused by sexual intercourse and may return if they re-commence intercourse. In addition, they or their partners may fear that the disease is transmissible and that the male partner is at risk. Most of these fears may be alleviated by discussion and explanation. However, some patients may still choose not to resume sexual activity.51 Radical hysterectomy should not alter sexual function markedly, although patients do report altered sensation. However, radiotherapy can lead to shortening and drying of the vagina with loss of lubrication and pliability. Shortening due to the formation of adhesions in the vagina can be avoided by regular douching during treatment and using a douche or dilator for a few months after treatment. However, radiation fibrosis in the walls of the vagina will still lead to some loss of length and pliability. Continued use of a vaginal dilator, with a lubricant gel, in those patients who have not resumed sexual activity, will maintain patency of the vagina, enabling resumption of sexual activity at a later date and also examination in the follow-up clinic. Dryness of the vagina may be helped by a lubricant gel and also by HRT which, in addition, can help increase libido which is often low in this group of patients.
NEUROLOGICAL TOXICITY
Cisplatin and taxane toxicities include peripheral neuropathy, tinnitus and damage to high tone hearing. These symptoms must be enquired after at every course of chemotherapy and the drug dose reduced or stopped if they occur. Unfortunately, the peripheral neuropathy and hearing changes can sometimes not be evident until several weeks after finishing chemotherapy and are often irreversible.
Hormonal changes Surgical removal of the ovaries at radical hysterectomy can usually be avoided unless there is involvement of the body of the uterus by tumour. However, should removal be necessary or should the ovaries lie within a radiation field, then an early menopause will occur in pre-menstrual women. There is no evidence that hormone replacement therapy (HRT) has an adverse effect on survival and it may be used to relieve menopausal symptoms.
POTENTIAL DEVELOPMENTS Screening and diagnosis Screening for cervical cancer is likely to be based on liquid based cytology rather than spatula smears. The detection of HPV may also be used as a first test. Most importantly, public education is improving both regarding the early symptoms of disease and the value of cervical screening, and family practitioners are referring patients for investigation more promptly.
Treatment SURGERY
Laparoscopic surgery has been developed for both pelvic and para-aortic lymphadenectomy. This allows early carcinoma of the cervix to be staged and patients with stage
Conclusions 793
Ia2 and early-stage Ib to be treated by radical trachelectomy and pelvic lymphadenectomy. Those patients who prove to have involved nodes would go on to radical chemoradiotherapy. This approach should reduce the morbidity of treatment.
drugs affecting the EGF receptor – is currently under study.
EXTERNAL BEAM RADIOTHERAPY
Cytological screening programmes and rapid referral for colposcopy should continue to reduce the incidence of invasive carcinoma of the cervix. However, overall there is an increasing incidence in pre-invasive neoplasia. The human papilloma virus is involved in the aetiology of cervical cancer, as is smoking. Anti-HPV vaccines may prevent the formation of cervical intra-epithelial neoplasia, the forerunner to invasive disease. Treatment of early disease with both radiotherapy and surgery is equally effective, but for more advanced tumours radiotherapy is preferable. Radiotherapy should consist of both an external beam phase and an intracavitary brachytherapy phase if curative doses are to be delivered without major normal tissue toxicity. Concomitant chemotherapy may benefit patients. Brachytherapy may be at high or low dose rate and, so far, little difference has been seen between the two when the HDR regimen is fractionated and the total dose reduced from that given at low dose rate. Neoadjuvant chemotherapy has still not got a proven place in the treatment of cancer of the cervix. Currently, concomitant chemoirradiation is the main area for study. Palliation of metastatic and recurrent disease may be achieved with carefully chosen chemotherapy regimens.
Technical developments include collimation devices that aim to improve the ratio of tumour volume to normal tissue volume and verification devices, to confirm that the treatment is being given to the intended volume. In the first category is the development of CT planned conformal therapy and multi-leaf collimators which remove the need for lead blocks to shape fields. In addition to making treatment set-up much easier and more reproducible, these collimators allow irregular fields to be treated. With appropriate computer software to drive the movement of the collimators it is possible to use them in IMRT to allow very complex tumour volumes to be treated with sparing of normal tissues. In the category of treatment verification devices fall ‘beams-eye view’ and ‘transmission dose’ devices. The former images each treatment field and compares the image taken with that of the confirmatory planning film. Any deviation from a pre-set norm would be signalled to the therapy radiographers to allow checking of the patient setup. Transmission dose devices similarly measure the exit dose of each field and may be used both to calculate absorbed dose in different density tissues and to confirm consistent patient set-up. Again, a significant deviation from a pre-set norm would alert radiographers to check the set-up and, if necessary because of patient weight loss or change in patient contour, re-plan the patient. Image-guided tomotherapy may allow tissue compensation to be made on a day-to-day basis.
CONCLUSIONS
KEY LEARNING POINTS ●
INTRACAVITARY THERAPY
High dose rate (HDR) brachytherapy is likely to overtake the use of low dose rate (LDR) systems. With computer controlled micro-sources, as used in HDR therapy, it is possible to create dose distributions to optimize treatment to spare sensitive tissues. Clinical experience is needed to find how far from the classical dose distributions brachytherapy can deviate in order to decrease morbidity without sacrificing effectiveness. Accurate recording of both treatment parameters and complications is essential for this process. CHEMOTHERAPY
Studies into the role of existing drugs, alone or in combination, for the palliation of advanced disease, as neoadjuvant therapy prior to radiotherapy or surgery and as concomitant chemoradiotherapy, are likely to continue. However, a new area of research – targeted therapies and
●
●
●
Predisposing factors to cervical neoplasia include early sexual intercourse, multiple partners and human papilloma viruses (particularly type 16); smoking may also be implicated. Cervical neoplasia usually starts in the ‘transformation zone’ and pre-invasive disease may be detected by exfoliative cytology of the cervix. Invasive disease is usually preceded by CIN for several years and spreads predominantly by direct invasion and lymphatic spread to pelvic nodes. The FIGO staging system is the most commonly used system, based on clinical examination, chest X-ray and imaging of the renal tract. Other important prognostic factors include nodal status and tumour bulk. Stage is the most significant prognostic factor as it probably represents tumour bulk. Pelvic node metastasis halves the prognosis of patients with early-stage disease whilst para-aortic disease is often a marker of widespread dissemination. Lymphovascular permeation increases the
794 Cervix
◆8
●
●
●
●
likelihood of nodal metastases and is a prognostic factor in itself. Excision of pre-invasive disease is preferable to ablation if occult invasive disease is not to be missed. Stage I disease may be treated by cone biopsy, total hysterectomy, radical trachelectomy with laparoscopic pelvic lymphadenectomy or radical hysterectomy and lymphadenectomy depending on the size of tumour, depth of invasion, the presence of lymphovascular permeation and the patient’s wishes to preserve fertility. Apart from a few patients with stage IIa disease with minimal involvement of the fornices of the vagina, all patients with more advanced tumours and those unfit for surgery should be treated with primary radiotherapy. Chemoradiotherapy using concomitant cisplatin and radiotherapy offers a benefit in survival over radiotherapy alone. External radiotherapy should be individually planned for each patient to take account of the size and distribution of disease and the nodal groups at risk of metastasis. Ideally it should be conformal and CT planned. Intracavitary therapy is vital if the primary tumour is to be controlled as cure without this modality is unusual, even in early-stage disease. Low, medium or high dose rate intracavitary systems may be used. Chemotherapy can play a useful role as palliation for advanced disease and, occasionally, recurrent disease. Response rates of up to 70 per cent can be seen in previously untreated disease when using cisplatin based combination therapy.
9
10 11
12
13 ❉ 14
15
16 ❉ 17
18 19
20
21
REFERENCES 22 1 CancerStats. Cervix uteri cancer – UK. London: Cancer Research UK, 2005. 2 Lammont DW, Symonds RP, Brodie MM, et al. Age, socioeconomic status and survival from cancer of the cervix in the West of Scotland 1980–87. Br J Cancer 1993; 67:351–7. 3 Anderson MC. The pathology of cervical cancer. Clin Obstet Gynaecol 1985; 12:87–119. 4 Quinn M, Babb P, Jones J, et al. Effects of screening on incidence and mortality from cancer of the cervix in England: evaluation based on routinely collected statistics. Br Med J 1999; 318:904–8. 5 Beral V. Cancer of the cervix: a sexually transmitted infection? Lancet 1974; 1:1037–42. 6 Zunzunegui MV, King MC, Coria CF, Charlet J. Male influences on cervical cancer risk. Am J Epidemiol 1986; 123:302–7. ●7 Singer A, Reid BL, Coppleson M. A hypothesis: The role of a high risk male in the aetiology of cervical carcinoma. A correlation of epidemiology and molecular biology. Am J Obstet Gynecol 1976; 126:110–15.
23
24
25
26
Munoz N, Bosch FX, de SanJose S, et al. Epidemiologic classification of human papilloma virus types associated with cervical cancer [comment]. N Engl J Med 2003; 348:518–27. Snijders PJ, Steenhergen RD, Heiderman DA, Meijer CS. HPV mediated cervical carcinogenesis: concepts and clinical implications. J Pathol 2006; 208:152–64. Tidy JA, Wrede D. Tumour suppressor genes: new pathways in gynecological cancer. Int J Gynecol Cancer 1992; 2:1–8. Barton SE, Maddox PH. Jenkins D, et al. Effect of cigarette smoking on cervical epithelial immunity: a mechanism for neoplastic change? Lancet 1988; 2:652–4. Maiman M, Fruchter RF, Serur E. Human immunodeficiency virus infection and cervical neoplasia. Gynecol Oncol 1990; 38:377–82. FIGO. In: TNM Atlas. Third edition, second revision. UICC. Heidelberg: Springer-Verlag, 1992, 196. Department of Health. Cancer waiting targets: A guide (version 4). London: Department of Health, 2005. Cardiff Cervical Cytology Study. Enumeration and definition of population and initial acceptance rates. J Epidemiol Community Health 1980; 34:9–13. Benedet JL, Murphy KJ. Cervical cancer screening. Who needs a Pap test? How often? Postgrad Med 1985; 78:69–71. Department of Health. Cervical screening: The facts (2006 edition). London: The Department of Health, 2006. FIGO. Report of the International Federation of Gynaecology and Obstetrics. Stockholm: Radiumhemmet, 1988. Collins CD, Constant O, Fryatt I, et al. Relationship of computed tomography tumour volume to patient survival in carcinoma of the cervix treated by radical radiotherapy. Br J Radiol 1994; 67:252–6. Magee BJ, Logue JP, Swindell R, McHugh D. Tumour size as a prognostic factor in carcinoma of the cervix: assessment by transrectal ultrasound. Br J Radiol 1991; 64:812–15. Shingleton HM, Orr JW. Cancer of the cervix. Edinburgh: Churchill Livingstone, 1995, 191. Inoue T, Morita K. The prognostic significance of number of positive nodes in cervical carcinoma stages IB, IIA and IIB. Cancer 1990; 65:1923. Kamura T, Tsukamoto N, Tsuruchi N, et al. Histopathological prognostic factors in stage IIb cervical carcinoma treated with radical hysterectomy and pelvic node dissection: an analysis with mathematical statistics. Int J Gynecol Cancer 1993; 3:219–25. Delgado G, Bundy B, Zaino R, et al. Prospective surgicalpathological study of disease-free interval in patients with stage Ib squamous cell carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 1990; 38:352–7. Elliot PM, Tattersall MH, Coppleson M, et al. Changing character of cervical cancer in young women. Br Med J 1989; 298:288–90. Dattoli MJ, Gretz HF, Seller U, et al. Analysis of multiple prognostic factors in patients with stage Ib cervical cancer: age as a major determinant. Int J Radiat Oncol Biol Phys 1989; 17:41–7.
References 795
● 27
28
29
30
31
◆ 32
◆ 33
34
35
❉ 36
● 37
38
39
Loundes CM, Gill OM. Cervical cancer, human papillomavirus and vaccination. Br Med J 2005; 331:915–6. Vergote I, Amant F, Berteloot P, van Gramberen M. Laparoscopic lower para-aortic staging lymphadenectomy in stage IB2, II and III cervical cancer. Int J Gynecol Cancer 2002; 12:22–6. Keys HM, Bundy BN, Stehman FB, et al. Cisplatin, radiation and adjuvant hysterectomy compared with radiation and adjuvant hysterectomy for bulky stage Ib cervical carcinoma. N Engl J Med 1989; 340:1154–61. Morris M, Eifel PJ, Lu J, et al. Pelvic radiation with concurrent chemotherapy compared with pelvic and paraaortic radiation for high-risk cervical cancer. N Engl J Med 1999; 340:1137–43. Rose PG, Bundy BN, Watkins EB, et al. Concurrent cisplatinbased radiotherapy and chemotherapy for locally advanced cervical cancer. N Engl J Med 1999; 340:1144–53. Thomas G. Improved treatment for cervical cancer – concurrent chemotherapy and radiotherapy. N Engl J Med 1999; 340:1198–200. Green JA, Kirwan JM, Tierney JF, et al. Systematic review and meta-analysis of randomised trials of concomitant chemotherapy and radiotherapy for cancer of the uterine cervix: better survival and reduced distant recurrence rate. Lancet 2001; 358:781–6. Petereit DG, Sarkaria JN, Chappell R. The adverse effect of treatment prolongation in cervical carcinoma. Int J Radiat Oncol Biol Phys 1995; 32:1301–7. Haie C, Pejovic MH, Gerbaulet A, et al. Is prophylactic paraaortic irradiation worthwhile in the treatment of advanced carcinoma of the cervix? Results of a controlled clinical trial of the EORTC radiotherapy group. Radiother Oncol 1988; 11:101–12. ICRU. Dose and volume specification for reporting intracavitary radiotherapy in gynecology. Report 38. Washington: International Commission on Radiation Units and Measurements, 1984. Brenner DJ, Hall EJ. Fractionated high dose versus low dose rate regimens for intracavitary brachytherapy of the cervix. 1: General considerations based on radiobiology. Br J Radiol 1991; 64:133–41. Fu K, Philips T. High dose-rate versus low dose-rate intracavitary brachytherapy for carcinoma of the cervix. Int J Radiat Biol Oncol Phys 1990; 19:791–6. Patel FD, Sharma SC, Negi PS, et al. Low dose rate vs. high dose rate brachytherapy in the treatment of carcinoma of
40
41
● 42
43
44
45
46
◆ 47
48
49
50
51
the uterine cervix: a clinical trial. Int J Radiat Biol Oncol Phys 1993; 28:335–41. Fyles AW, Milosevic M, Pintilie M, Hill RP. Cervix cancer oxygenation measured following external radiation therapy. Int J Radiat Biol Oncol Phys 1998; 42:751–3. Grogan M, Thomas GM, Melamed I, et al. The importance of haemoglobin levels during radiotherapy for carcinoma of the cervix. Cancer 1999; 86:1528–36. Thigpen T, Vance RE, Balducci L, Blessing J. Chemotherapy in the management of advanced and recurrent cervical and endometrial cancer. Cancer 1981; 48:658–65. Thigpen T, Vance R, Khansur T, Malamud F. The role of paclitaxel in the management of patients with carcinoma of the cervix. Semin Oncol 1997; 24(Suppl. 2):451–6. Tutt ANJ, Lodge N, Blake PR. Palliative chemotherapy in recurrent carcinoma of the cervix: an audit of the use of ifosfamide and review of the literature. Int J Gynecol Cancer 1999; 9:12–17. Sardi JE, Snaanes CE, Giaroli AA, et al. Neoadjuvant chemotherapy in cervical carcinoma stage IIB: a randomised controlled trial. Int J Gynecol Cancer 1998; 8:441–50. Hoskin PJ, Blake PR. Cisplatin, methotrexate and bleomycin (PMB) for carcinoma of the cervix; the influence of presentation and previous treatment upon response. Int J Gynecol Cancer 1991; 1:75–80. Castellsague X, Munoz N. Cofactors in human papilloma virus carcinogenesis: Role of parity, oral contraceptives and tobacco smoking [review]. J Natl Cancer Inst Monogr 2003; 31:20–8. Harris G, Cramp WA, Edwards JC, et al. Radiosensitivity of peripheral blood lymphocytes in autoimmune disease. Int J Radiat Biol 1985; 47:689–99. Denton AS, Bond SJ, Matthews S, et al. Short report: National audit of the management and outcome of carcinoma of the cervix treated with radiotherapy in 1993. Clin Oncol 2000; 12:347–53. Andreyev HJN, Vlavianos P, Blake P, et al. Gastrointestinal symptoms after pelvic radiotherapy: Role for the gastroenterologist? Int J Radiat Biol Oncol Phys 2005; 62:1464–71. Thranov I, Klee M. Sexuality among gynecologic cancer patients – a cross-sectional study. Gynecol Oncol 1994; 52:14–19.
35 Carcinoma of the vagina and vulva SADAF GHAEM-MAGHAMI AND W. PAT SOUTTER
Carcinoma of the vagina Carcinoma of the vulva
796 799
CARCINOMA OF THE VAGINA Invasive vaginal cancer is rare. In 2003, the overall rate for invasive squamous tumours of the vagina was about 0.7 per 100 000 women, the rate being highest in women over the age of 65 (3.9 per 100 000). It comprises 1–2 per cent of all gynaecological malignancies, only 178 women developing this cancer in England and Wales in 2003.1 However, like the cervix, the vagina has a range of pre-malignant lesions, many of which may be previously unrecognized extensions of cervical abnormalities. Coincident with the rise in prevalence of cervical intra-epithelial neoplasia (CIN) is an increase in the frequency with which vaginal intra-epithelial neoplasia (VAIN) is seen.
Aetiology Human papilloma virus (HPV) is found in association with vaginal carcinoma, indicating a role in its aetiology.2 Also seropositivity to HPV 16 increases the risk of developing vaginal carcinoma by a factor of 4.5 and seropositivity for HPV 18 increases the risk of developing VAIN by a factor of 12.3* A field effect in the lower genital tract has been suggested by the observation of multicentric neoplasia involving cervix, vagina and vulva;4 however Sugase et al. found high copy numbers of HPV in vaginal biopsies of patients with VAIN, a large proportion of whom had no associated CIN or vulval intra-epithelial neoplasia (VIN).5 The irritation caused by procidentia and vaginal pessaries has been suggested as an aetiological factor, but this is an infrequent association. Previous radiotherapy for cervical cancer in young women has been suggested as a cause of cancer of the
References
806
vagina. Three small studies suggested that young women treated with radiotherapy for cervical cancer might be at greater risk of developing vaginal cancer.6*,7*,8* However a huge international collaborative study of 25 995 women followed for more than 10 years after treatment with radiotherapy for cervical cancer recorded only 48 cancers of the vulva or vagina. This was not significantly different from the 7 such women out of 5125 treated with surgery. These data suggest that if vaginal cancer is induced by radiotherapy it is a very rare event. For some time the prevalence of clear cell adenocarcinoma of the vagina was thought to be increased by intrauterine exposure to diethylstilbestrol (DES) but with the accrual of more information the risks now seem to be very low and lie between 0.1 and 1.0 per 1000.9*,10*,11 While vaginal adenosis and minor anatomical abnormalities of no significance (e.g. cervical cockscomb) are common following intra-uterine DES exposure, the only lesion of any significance that is seen more commonly is CIN.12*
Anatomy The upper two thirds of the vagina are derived from the Mullerian duct and the lower third from the ectoderm of the cloaca. The vagina is related anteriorly to the bladder above and the urethra below. Posteriorly, the vault of the vagina is covered with the peritoneum of the pouch of Douglas. It thus becomes closely related to loops of small or large bowel. Below this, it is closely related to the anterior wall of the rectum until the perineal body separates it from the anal canal. The ureters run close to the cervix over each side of the vaginal vault to the bladder. Laterally, the vagina is supported by the lower portion of the cardinal ligaments
Carcinoma of the vagina 797
until it reaches the pelvic floor where it is invested by the medial part of the levator ani muscles (pubococcygeus). Lateral to the vagina, most of the tissue is areolar except at the level of the perineal body. The vagina is lined by stratified squamous epithelium. When the transformation zone of the cervix has extended on to the vagina, clefts or glands partly lined by columnar epithelium will be seen deep to the squamous epithelium.
Table 35.1 Modified International Federation of Gynecology and Obstetrics (FIGO) staging of vaginal cancer Stage 0 Stage I Stage IIa Stage IIb
Pathology Vaginal intra-epithelial neoplasia (VAIN) is characterized by loss of stratification and polarity of the cells and by nuclear atypia. The great majority (85–95 per cent) of primary vaginal cancers are squamous. Of these, 60 per cent occur in the upper one third of the vagina, particularly in the posterior fornix. Grossly they present as ulcerative tumours, exophytic masses or infiltrating stenosing lesions. Histologically, squamous cell carcinomas of the vagina range from well-differentiated lesions with abundant keratin formation to poorly differentiated tumours with little squamous differentiation. The majority of lesions are moderately differentiated, non-keratinizing lesions. There appears to be little correlation between histological grade and prognosis. Clear cell adenocarcinomas, malignant melanomas, embryonal rhabdomyosarcomas and endodermal sinus tumours are the most common of the small number of other tumours seen very rarely in the vagina. These are discussed separately.
Natural history Like invasive vaginal carcinoma, VAIN is also seen most commonly in the upper vagina and is unusual in the mid and lower parts. The malignant potential for VAIN is not as well documented as that of CIN, but it is clearly not a completely benign disease. Progression to invasive carcinoma probably only occurs in 5–9 per cent of cases.13* It would therefore seem sensible to treat it with the same respect accorded to CIN. Although the upper vagina is the most common site for invasive disease, about 25–30 per cent are confined to the lower vagina, usually the anterior wall. Squamous vaginal cancer initially spreads by local invasion. Lymphatic spread occurs by tumour embolization to the pelvic nodes from the upper vagina and to both pelvic and inguinal nodes from the lower vagina. Haematogenous spread is unusual.
Clinical staging The modified clinical staging suggested by Perez and colleagues14 has been widely adopted (Table 35.1).
Stage III Stage IVa Stage IVb
Intra-epithelial neoplasia Invasive carcinoma confined to the vaginal mucosa Subvaginal infiltration not extending to the parametrium Parametrial infiltration not extending to the pelvic sidewall Extends to pelvic sidewall Involves mucosa of bladder or rectum Spread beyond the pelvis
Diagnosis and assessment VAIN is usually a cytological or colposcopic diagnosis. Dyskaryosis in cervical or vaginal cytology is an indication for colposcopy to determine the location and nature of the abnormality that is giving rise to the abnormal cells. As a substantial proportion of these women have undergone hysterectomy prior to the detection of VAIN and as the majority of the lesions are in the vaginal vault, straddling the suture line or in the angles of the vault, assessment and biopsy can be difficult.15 General anaesthesia is required for most cases. It should also be remembered that abnormal epithelium or invasive cancer can lie buried behind the sutures closing the vault. Before making a diagnosis of primary vaginal cancer, the following criteria must be satisfied: the primary site of growth must be in the vagina; the uterine cervix must not be involved, and there must be no clinical evidence that the vaginal tumour is metastatic disease.16 The most common presenting symptom is vaginal bleeding (53–65 per cent) with vaginal discharge (11–16 per cent) and pelvic pain (4–11 per cent) being less common.17* Patients with advanced disease may present with bowel or bladder symptoms. The rate of detection of asymptomatic cancer with vaginal cytology varies greatly (10–42 per cent) depending on the patient population studied. Most of the disease thus detected is at an early stage. The most important part of the pre-treatment assessment of invasive cancer of the vagina is a careful examination under anaesthesia. A combined vaginal and rectal examination will help to detect extra-vaginal spread. Cystoscopy, proctosigmoidoscopy and colposcopy may be indicated, the latter to identify co-existing VAIN. A generous, full-thickness biopsy is essential for adequate histological evaluation. Other investigations include a chest X-ray. Magnetic resonance imaging (MRI) with an endovaginal coil can be invaluable in showing local disease spread outside the vagina and lymph node involvement. Transrectal ultrasound or computed tomography (CT) may help to define the size and extent of the lesion.
798 Carcinoma of the vagina and vulva
Treatment and complications – vaginal intra-epithelial neoplasia The treatment of VAIN includes local ablation with the carbon dioxide laser, local excision, partial and total vaginectomy and radiotherapy. Treatment with 5-fluorouracil appears to be associated with high recurrence rates18 and it cannot be recommended. Occasionally, in a postmenopausal woman, the abnormality resolves with oestrogen therapy. Local ablation with the carbon dioxide laser is used for small lesions where there is no suspicion of invasion. Local excision is sufficient for small lesions, otherwise partial or total colpectomy (vaginectomy) is necessary. The vaginal route is preferred because it is usually easier with less morbidity, but it can prove to be extremely taxing in patients with a narrow introitus and no laxity of the vaginal vault. Surgery is the only effective option available to patients previously treated with radiotherapy and is the treatment of choice for women who are post hysterectomy. The reported recurrence rate is 0–5 per cent.18* Surgery can give rise to severe coital problems even when only partial vaginectomy is performed. In the young, sexually active woman these may be overcome by the construction of a neovagina but even this is not immune from later developing cancer.4 Intra-cavitary radiotherapy is an option for lesions in the vaginal vault following a hysterectomy, but this will have the disadvantage that radiotherapy can cause vaginal stenosis or ovarian failure in a pre-menopausal woman.
in 5–6 weeks is given to the pelvis. External radiotherapy is followed by intra-cavitary therapy. This is with an intrauterine tube and ovoids, for lesions limited to the vaginal vault. For lesions below the vault the treatment boost is given with interstitial brachytherapy, which has been found to be more effective than intra-cavitary brachytherapy.19* However, if the tumour has completely regressed following external irradiation, brachytherapy, using a vaginal obturator, will give a sufficient dose to the vaginal mucosa. The total dose from both external radiotherapy and low dose rate brachytherapy is 70–75 Gy to the vaginal mucosa, or its equivalent using a high dose rate system. High dose rate radiotherapy has been reported as being as effective and as well tolerated as low dose rate radiotherapy.20* Complications of radiotherapy Vaginal stenosis may occur and is more likely when advanced tumours are treated. The overall prevalence of vaginal stenosis is around 30 per cent and is an undoubted problem for sexually active patients. This complication can be reduced by the use of vaginal douches during treatment and vaginal dilators afterwards. Mucosal ulceration, either immediate or delayed, can be a distressing complication but conservative therapy is usually effective. Vesicovaginal and rectovaginal fistulas and small bowel complications can occur but are rare. The risk of fistula formation is higher in advanced disease.
SURGERY
Treatment of invasive cancer RADIOTHERAPY
Invasive vaginal cancer is usually treated with radiotherapy, which is given either as a combination of external radiation and brachytherapy (interstitial or intra-cavitary) or by brachytherapy alone. Early cases of vaginal cancer occurring in the lower vagina may be treated entirely with interstitial brachytherapy, usually with iridium-192. A template, for example, the Syed–Neblett applicator, which is afterloaded, allows an even distribution of radiation to the lesion. The objective is to achieve a tumour dose of 70–80 Gy in 2 fractions, 2 weeks apart. The radiation is given at a rate of 10 Gy per day. Tumours of the upper two thirds of the vagina are treated in an identical fashion to cancer of the cervix, either by radiotherapy or by chemo-radiotherapy, usually with cisplatin given concurrently. External radiotherapy to the pelvis includes the parametria and the pelvic nodes, and is followed by brachytherapy. The lower end of the radiation field must cover the lesion in the vagina plus a 2 cm margin. If the lesion extends into the lower third of the vagina, the inguino-femoral nodes must also be included in the treatment volume. A total dose of 45–50 Gy in 25–28 fractions
A stage I lesion in the upper vagina can be adequately treated by radical hysterectomy (if the uterus is still present), radical vaginectomy and pelvic lymphadenectomy.21* A split-thickness skin graft may be used to reconstruct a new vagina. Exenteration is required for more advanced lesions and carries the problems of stomata. However, exenterative surgery may be the treatment of choice for women who have had prior pelvic radiotherapy.
RESULTS
Owing to the rarity of vaginal cancer there is a wide range of reported results. Probably the most reliable figures are from Kucera and Vavra22* who reported on 460 women with cancer of the vagina. They found that 77 per cent of the women with stage I disease were alive at 5 years compared to 45 per cent with stage II, 31 per cent with stage III, and 18 per cent with stage IV. Although survival is predominantly stage related, the site and size of the lesion may be important, with smaller lesions and those in the upper vagina having a better prognosis. Also the ulcerating and deeply invasive cancers have a poorer prognosis than the exophytic lesions.
Carcinoma of the vulva 799
Uncommon vaginal tumours CLEAR CELL ADENOCARCINOMA
The relation of this rare tumour to intra-uterine exposure to DES is discussed under aetiology. The histology is characterized by vacuolated or clear areas in the cytoplasm and a hobnail appearance of the nuclei of cells lining the lumen of glands. Radical surgery or radical radiotherapy is required for invasive lesions. As most are situated in the upper vagina they may be treated as cervical lesions. Lymph node metastases and 5-year survival figures are equivalent to cervical cancer. MALIGNANT MELANOMA
Vaginal melanoma has a 5-year survival rate of less than 10 per cent. Vaginal bleeding and discharge are the most common presenting symptoms. The prognosis depends upon the depth of epithelial invasion. Radical surgery and radiotherapy are of little value if the lesion is deeply invasive because of its propensity to metastasize early by the blood stream. There is at present no effective chemotherapy. RHABDOMYOSARCOMA (SARCOMA BOTRYOIDES)
Ninety per cent of these rare tumours occur in children less than 5 years old. They present with vaginal bleeding and a grape-like mass in the vagina. The appearance of crossstriations in the rhabdomyoblasts is characteristic of this tumour. Treatment is with combination chemotherapy, which is carried out in specialist centres. When chemotherapy produces only a partial response, local surgery is carried out. Occasionally exenteration is required for resistant disease. Radiotherapy is restricted to patients with unresectable tumour in view of the deleterious effects on bone growth. The overall survival rate is over 80 per cent. ENDODERMAL SINUS TUMOURS
These very rare tumours may resemble rhabdomyosarcomas but histology shows a primitive adenocarcinoma. Most occur in infants under the age of 2 years. Treatment with chemotherapy is as for germ cell tumours of the testis or ovary.
CARCINOMA OF THE VULVA Invasive vulval cancer is not a common gynaecological cancer. It represents approximately 4 per cent of all female reproductive organ cancers. There are about 800 new cases each year in England and the annual incidence is approximately 3.5/100 000.1 The majority of these women are elderly – less than 10 per cent are under 55 years of age and 80 per cent are over 65 – and with increased life expectancy this cancer is seen more often.
Cancer of the vulva is unpleasant but potentially curable even in elderly, unfit ladies if referred early and managed correctly from the outset. It is essential that patients are referred to a multidisciplinary team in a cancer centre for treatment as soon as vulval cancer is suspected. Surgery for this condition needs special expertise and should only be undertaken by gynaecological oncologists. Initial, inappropriate surgery leads to a poor outcome.
Aetiology Little is known of the aetiology of vulval cancer. A viral factor has been suggested as DNA from human papilloma virus (HPV) types 16 and 18 has been detected in vulval intraepithelial neoplasia (VIN). The type 2 herpes simplex virus (HSV) may be a co-factor. There are also recognized risks of developing cancer with lichen sclerosus (4 per cent),23 Paget’s disease24 and melanoma in situ.25*
Anatomy The vulva includes the mons pubis, the labia majora and minora, the clitoris, the vestibule of the vagina, the bulb of the vestibule and the greater vestibular glands (Bartholin’s). The mons pubis is a pad of fat anterior to the pubic symphysis and covered by hair-bearing skin. The labia majora extend posteriorly from the mons on either side of the pudendal cleft into which the urethra and vagina open. They merge with one another and the perineal skin anterior to the anus. They consist largely of areolar tissue and fat. On their lateral aspects the skin is pigmented and covered with crisp hairs. On the medial side the skin is smooth and has many sebaceous glands. The labia minora are small folds of skin that lie between the labia majora and divide anteriorly to envelop the clitoris. The medial surfaces contain many sebaceous glands. The clitoris is an erectile structure analogous to the male penis. Partly hidden by the anterior folds of the labia minora, the clitoris consists of a body of two corpora cavernosa lying side by side and connected to the pubic and ischial rami, and a glans of sensitive, spongy erectile tissue. The vestibule is that area between the labia minora into which the urethra and vagina open. The bulbs of the vestibule are elongated masses of erectile tissue lying on either side of the vaginal opening. The greater vestibular glands lie posterior to the bulbs of the vestibule and are connected to the surface by short ducts.
Lymphatic drainage The lymph drains from the vulva to the inguinal and femoral glands in the groin and then to the external iliac glands. Drainage from the perineum and the clitoris is to both groins but some contralateral spread occurs from other sites on the vulva. Direct spread to the pelvic nodes
800 Carcinoma of the vagina and vulva
along the internal pudendal vessels occurs only very rarely and no direct pathway from the clitoris to pelvic nodes has been consistently demonstrated.
Pathology Both squamous vulval intra-epithelial neoplasia (VIN) and Paget’s disease occur on the vulva. The histological features of VIN are analogous to those seen in CIN and VAIN. In the same way, the histological appearance of Paget’s disease is similar to the lesion seen in the breast. In a third of cases of Paget’s disease there is an adenocarcinoma in underlying apocrine glands and these carry an especially poor prognosis. Most invasive cancers (85 per cent) are squamous. Some 5 per cent are melanomas and the remainder is made up of carcinomas of Bartholin’s gland, other adenocarcinomas, basal cell carcinomas, and the very rare verrucous carcinomas, rhabdomyosarcomas and leiomyosarcomas. In squamous cell carcinomas, the presence of infiltrative growth patterns, compared with a pushing pattern, and the presence of lymphatic vascular space invasion (LVSI) is associated with a higher local recurrence rate and poorer prognosis. However, these factors do not necessarily indicate the need for adjuvant treatment.
disease have nodal spread. Spread to the contralateral groin can occur in up to 25 per cent of cases with positive groin nodes. Pelvic node involvement is not common and is usually secondary to groin node involvement but rarely there is direct spread to the pelvic nodes via the internal pudendal vessels. Blood spread to bone or lung is rare. Death can be a long unpleasant process and is often due to sepsis and inanition or haemorrhage. Uraemia from bilateral ureteric obstruction may supervene first. Such is the abject misery of this demise that all patients with resectable vulval lesions should be offered surgery regardless of their age and general condition.
Clinical staging The International Federation of Gynecology and Obstetrics (FIGO) classification is shown in Table 35.2. In spite of the apparent limitations of this classification, it does give a reasonable guide to the prognosis. Formerly, the main drawback was a reliance on clinical palpation of the groin nodes which is notoriously inaccurate.33 Now that the surgical findings are incorporated in the staging evaluation, the prognostic value is greatly improved.
Diagnosis and assessment Natural history A large proportion – 40 per cent – of women with VIN are less than 40 years of age. VIN is histologically very similar to CIN and often occurs in association with it. It used to be said that its malignant potential is less than 5 per cent. However, this opinion is based largely on studies of women who have been treated by excision biopsy or vulvectomy. This may not be true of untreated or inadequately treated patients, as progression to invasive cancer in 2–8 years has been reported.26* The definition of ‘microinvasion’ of the vulva has proved extremely problematical. The purpose is to identify a group of women with invasive carcinoma who could safely be treated with a less mutilating procedure than radical vulvectomy. Although it was initially suggested that up to 5 mm invasion into the stroma might be acceptable,27,28 subsequent reports have suggested lower limits. Some have suggested 2 mm,29 others preferred 1 mm,30 while further reports emphasize the importance of lymphatic or vascular invasion and the degree of differentiation31 or confluence.32 It seems that the safest course to follow is to perform groin node dissection in all cases with more than 1 mm stromal invasion without attempting to differentiate between superficial and deep inguinal nodes.33,34 Invasive disease involves the labia majora in about two thirds of cases and the clitoris, labia minora, or posterior fourchette and perineum in the remainder. The tumour usually spreads slowly, infiltrating local tissue before metastasizing by embolization to the inguinal and femoral nodes. Overall, about 30 per cent of women with operable
Intra-epithelial disease of the vulva often presents as pruritus vulvae but 20–45 per cent are asymptomatic and frequently are found after treatment of other genital tract malignancies, particularly cervical carcinoma. These lesions are often raised above the surrounding skin and have a rough surface and variable colour – white, due to hyperkeratinization; red, due to immaturity of the epithelium; or
Table 35.2 The FIGO staging of vulval cancer (1995) Stage
Definition
Stage Ia
Confined to vulva and/or perineum, 2 cm or less maximum diameter. Groin nodes not palpable. Stromal invasion no greater than 1 mma Stage Ib As for 1a but stromal invasion greater than 1 mm Stage II Confined to vulva and/or perineum, more than 2 cm maximum diameter. Groin nodes not palpable Stage III Extends beyond the vulva, vagina, lower urethra or anus; or unilateral regional lymph node metastasis Stage IVa Involves the mucosa of rectum or bladder; upper urethra; or pelvic bone; and/or bilateral regional lymph node metastases Stage IVb Any distant metastasis including pelvic lymph node a
Depth of invasion is measured from the epithelial stromal junction of the adjacent most superficial dermal papilla to the deepest point of invasion.
Carcinoma of the vulva 801
dark brown, due to increased melanin deposition in the epithelial cells. However, the full extent of the abnormality is often not apparent until 5 per cent acetic acid is applied. After 2 minutes, VIN turns white and mosaic or punctation may be visible. All of these changes are best examined colposcopically. Toluidine blue is also used as a nuclear stain but areas of ulceration give false positive results and hyperkeratinization gives false negatives. Biopsies must be taken from abnormal areas. This can usually be done under local anaesthesia in the out-patient clinic using a disposable 4 mm Stiefel biopsy punch or a Keyes punch. While vulval cancer can be asymptomatic, over 70 per cent of patients with invasive disease complain of irritation, pruritus, pain or soreness and over half note a mass in the vulva or an ulcer. It is usually not until the mass appears that medical advice is sought. Bleeding and discharge are less common presentations. One of the major problems in invasive vulval cancer is the delay between the first appearance of symptoms and referral for a gynaecological opinion. This is only partly due to the patients’ reluctance to attend. In many cases the doctor fails to recognize the gravity of the lesion and prescribes topical therapy, sometimes without examining the woman. Because of the multicentric nature of female lower genital tract cancer the investigation should include inspection of the cervix and cervical cytology. The groin nodes must be palpated carefully and any suspicious nodes may be sampled by fine needle aspiration. A chest X-ray is always required and a CT scan may be helpful. Thorough examination under anaesthesia and a full-thickness biopsy are the most important investigations. The examination under anaesthesia should note particularly the size and distribution of the primary lesion, especially the involvement of the urethra or rectum, and secondary lesions in the vulval or perineal skin must be sought. The groin should be reexamined under general anaesthesia as previously undetected nodes may be palpated at that time.
Treatment of vulval intra-epithelial neoplasia The treatment of VIN is difficult. Uncertainty about the malignant potential, the multifocal nature of the disorder and the discomfort and mutilation resulting from therapy suggest that recommendations should be cautious and conservative in order to avoid making the treatment worse than the disease. The youth of many of these patients is a further, important consideration. Nonetheless, the documented progression of untreated cases to invasive cancer underlines the potential importance of these lesions. If the patient has presented with symptoms, therapy is required. Asymptomatic patients, particularly under the age of 50 years, are probably best observed closely with biopsies repeated if there are any suspicious changes.15 If the lesion is small, an excision biopsy may be both diagnostic and therapeutic.35 If the disease is multifocal or covers a wide area, a skin graft may improve the cosmetic
result of a skinning vulvectomy.36 However, the donor site is often very painful and a satisfactory result can be obtained in most patients without grafting. An alternative approach is to vaporize the abnormal epithelium with the carbon dioxide laser.37 Given the very irregular surface of the vulva, it is very difficult to achieve a uniform depth of destruction. Moreover, the depth of treatment required for VIN is still unclear.38 In some cases hair follicles may be involved for several millimetres below the surface.39 Even with carefully controlled depth of treatment, re-epithelialization of large areas treated with the laser will take several weeks. The other main disadvantage is not having histopathological assessment. The use of 5fluorouracil cream is not widely recommended.40 Five per cent imiquimod cream has been used with modest results in the management of women with VIN: Le et al.41 showed complete response in 9 out of 17 evaluable cases of VIN 2/3 with minor side effects. Results of larger, more definitive studies are awaited.
Treatment of invasive disease Surgery is the mainstay of treatment. The introduction of radical vulvectomy (complete removal of the vulva and bilateral inguino-femoral lymphadenectomy) reduced the mortality from 80 per cent to 40 per cent.42*,43* However, to control lymphatic spread, these techniques removed large areas of normal skin from the groins and primary wound closure was rarely achieved. By using a modified incision, the same objectives can be accomplished without the removal of large areas of normal skin and with the enormous benefit that primary closure can be achieved in nearly all cases.44 A further refinement aimed at reducing still further the problems of wound healing was the use of separate groin incisions for stage I–II cases.45* Studies comparing triple incision with en bloc dissection have not shown any significant difference in either survival or recurrence.46*,47* Using the triple incision technique in 100 women with cancer of the vulva, Grimshaw et al.48 reported a 75 per cent 5-year survival corrected for death from intercurrent disease. The corrected survival for women with stage I disease was 95 per cent. More recently there has been a move to an even more conservative approach. In early-stage disease, in the absence of clinically suspicious or involved groin nodes, the surgery to the primary tumour should be radical to remove the tumour yet ‘the impetus for more conservative approaches stems from the well-recognized psychosexual sequelae’49. The psychological morbidity of radical treatment has been reported by Andersen.50 The management of the vulval lesion and treatment of the groins should be considered separately. EXCISION OF THE VULVAL LESION
If the lesion is less than 2cm in diameter, unifocal, and if the rest of the vulva is healthy, radical local excision is the
802 Carcinoma of the vagina and vulva
treatment of choice.51 In all other cases, a radical removal of the whole vulva is required. The deep and the lateral surgical margins should be no less following radical local excision than after radical vulvectomy. The depth of resection should be to the fascia lata and lateral margins should be at least 10mm to minimize the risk of local recurrence.49 The distal 1cm of the urethra can be excised safely to achieve an adequate margin without risking incontinence. If radical surgery is likely to cause sphincter damage, leading to urinary or faecal incontinence, preoperative radiotherapy or chemoradiotherapy should be considered to shrink the tumour. GROIN NODE DISSECTION
Dissection of the groin nodes is carried out for all cases greater than stage Ia. The groin node dissection includes both the superficial inguinal and deep femoral nodes as superficial groin node dissection alone is associated with a higher risk of groin node recurrence.52** If the nodes are not obviously, clinically involved, separate incisions in the groin can be used.45* In advanced disease, the triple incision technique may be inappropriate and a radical vulvectomy with an en bloc groin node dissection may be required. Because of the extensive crossover of lymphatic channels from the vulva, bilateral groin node dissection is usually performed. However, in lateral tumours where the medial margin of the tumour is at least 2 cm from the midline, an ipsilateral groin node dissection is sufficient. However, if the nodes are found to be positive, the contralateral groin will need to be dissected at a second operation. Preservation of the long saphenous vein may reduce both groin wound and subsequent lower limb problems. There is little value in performing a pelvic node dissection as postoperative radiation therapy to the groins and pelvis gives superior results when more than two groin nodes are involved.53** Sentinel node detection using dye studies and lymphoscintigraphy may be of value in identifying the involved lymph nodes and requiring only limited dissection. However, results from controlled clinical trials are awaited before the use of these techniques becomes more widespread in routine practice. Complications The most common complication is wound breakdown and infection. With the modified surgical techniques referred to above, this is seldom more than a minor problem. Conservative therapy with Eusol and liquid honey packs is all that is required. Lymphocyst formation can be very troublesome. Aspiration or secondary drainage is seldom helpful. Resolution usually occurs spontaneously. Osteitis pubis is a rare but very serious complication that requires intensive and prolonged antibiotic therapy. Thromboembolic disease is always a greatly feared complication of pelvic surgery for malignant disease but the combination of peroperative epidural analgesia to ensure good venous return with
subcutaneous heparin begun postoperatively seems to reduce this risk. Secondary haemorrhage occurs from time to time. Leg oedema may be expected in about 30 per cent of women. Numbness and paraesthesia over the anterior thigh is common due to the division of small cutaneous branches of the femoral nerve. Loss of body image and impaired sexual function undoubtedly occur but the patients’ responses to surgery are enormously variable and probably dependent on the woman’s upbringing and attitudes to life. CHEMOTHERAPY
Chemotherapy has not played a large role in the management of vulval carcinoma because of the advanced age of many of the patients and the frequency of concurrent medical conditions. It may be used as an adjuvant or neoadjuvant treatment or for recurrent disease. Single agents that have been shown to be active include cisplatin, doxorubicin, bleomycin and methotrexate. Concomitant chemo-radiotherapy regimens have been investigated but there are no randomized studies to show whether chemo-radiotherapy is superior to radiotherapy alone.54,55* RADIOTHERAPY
While surgery is the mainstay of treatment for vulval carcinoma, adjuvant radiotherapy has a role in preventing recurrence. In the presence of advanced disease, with bowel or bladder involvement, surgery alone can result in poor function and cosmesis and this has led to renewed interest in the use of non-surgical treatment modalities for this situation.56 Radiotherapy technique External beam radiotherapy has historically been considered too toxic a treatment to be tolerated by the epithelium of the vulva, leading to severe acute moist desquamation and severe late normal tissue damage (vulval fibrosis, atrophy and necrosis, vaginal and urethral stenosis and fistula formation). However, these complications of radiotherapy are avoidable by using doses per fraction of no more than 1.8 Gy, and by limiting the overall total dose. There is general agreement on a threshold dose for improved local control at about 50 Gy. Fraction size is important with 1.7 Gy being close to tolerance. The doses which have been recommended by RCOG49 based on the recommendations of Thomas et al.54*,57 are 55 Gy as a maximum preoperatively [with or without concurrent 5-FU]; 45–50 Gy as adjuvant postoperative treatment; 65 Gy as radical therapy. Higher doses can lead to severe morbidity. Hoffman et al.58* reported radio-necrosis in 6 of 10 patients treated with more than 70 Gy. The optimal fractionation regimen remains to be defined. Thomas et al.54*, using daily fractions of 1.6–1.8 Gy to a total dose of 40–64 Gy to the vulva and 36–59 Gy to the nodes with concomitant 5-FU and mitomycin C chemotherapy in 27 patients, found that 16 needed unplanned treatment breaks for periods of 10–34 days (median 19 days). This
Carcinoma of the vulva 803
delay would only have been partly due to the additional use of the concomitant chemotherapy. The use of twice daily fractionation, giving 1.5 Gy with a 6 hour gap, has been reported from Hammersmith Hospital.59 This regimen consists of 45 Gy in 30 fractions in 3 weeks and in those patients not proceeding to surgery is followed by a planned 2-week gap to allow skin healing before treatment is continued to a total of 64.5 Gy. Experience with this twice daily treatment has shown that unplanned treatment breaks can be eliminated. Such a regimen offers obvious benefits in terms of ensuring overall treatment time is not prolonged. Limitation of the size of the radiation field is also important. When radiotherapy is given with radical intent, it is given in two phases. The first phase treats the primary and nodal sites, using external irradiation, to a dose of 45–50 Gy. For the second phase the tumour is boosted using as small a field as possible. This is achieved by selecting the most appropriate technique such as a direct field to the perineum using electron portals or interstitial radiotherapy or conformal radiotherapy. The total dose from both phases of treatment, as previously stated, is 65 Gy. Postoperative radiotherapy Fifteen to 33 per cent of patients with advanced, operable vulval carcinoma recur after radical surgery – the majority (80–95 per cent) at loco-regional sites in the first instance.60* A number of clinico-pathological variables have been defined which predict local recurrence and overall survival.61*,62* The clinically important features are FIGO stage, presence of clinically involved lymph nodes, tumour size independent of stage, tumour location (midline vs. lateral, with midline tumours having a greater tendency to metastasize to bilateral inguinal nodes), age and performance status. The most significant prognostic features are size of tumour, inguinal node involvement and the width of the vulval tumour-free margin. Almost 1 in 5 patients found to have metastatic involvement of groin nodes removed at surgery will also have disease in the iliac nodes. In regard to tumour margin, there is no risk of recurrence if the tumour-free margin is 8 mm or more but the risk of recurrence is 8 per cent if the margin is 4.8–8 mm rising to 54 per cent if the margin is less than 4.8 mm.62*,63* The criteria for postoperative radiotherapy to the inguinal and pelvic nodes are the presence of a single clinically involved node, more than two histologically involved inguinal nodes or extracapsular spread.57 The dose for postoperative radiotherapy to the groins and pelvic nodes is 45–50 Gy at 1.8–2.0 Gy per fraction. The recommended depth is up to 8 cm, as the depth of the deep nodes will vary from patient to patient. Adjuvant radiotherapy to the vulva should be considered where the disease-free margin is less than 8 mm. This should be given to as small a field as possible to reduce morbidity. The recommended dose is again 45–50 Gy. The role of chemotherapy as a postoperative adjuvant is unclear. The recent Gynecologic Oncology Group (GOG) 185 study, which closed prematurely, had tried to examine
the role of radiation versus chemo/radiation but failed to achieve the recruitment target. A new international study is under development and is proposing the use of chemo/ radiation. If chemotherapy is to be used in the postoperative adjuvant fashion with radiation, it is recommended that cisplatin be used as a single agent, probably at a dose of 40 mg/m2 weekly.49 Radiotherapy as an alternative to surgery for occult nodal disease Since 20–25 per cent of women with clinically stage I cancer of the vulva have occult nodal metastases, some form of local treatment is required to eradicate this potential source of recurrent (and metastatic) disease. The results of adjuvant nodal irradiation53** suggested the hypothesis that nodal irradiation could obviate the need for bilateral inguinal lymph node dissection in low-risk groups. However, a randomized GOG study64** did not confirm this. Fifty two patients with clinically non-suspicious inguinal nodes undergoing radical vulvectomy were randomized to receive either node dissection or inguino-femoral nodal irradiation to a dose of 50 Gy at 2 Gy per fraction dosed at 3 cm depth. The study was closed prematurely when interim analysis revealed a significant advantage in favour of the surgical arm in terms of progression-free interval and survival. A criticism of this study is that treatment to the groins was only given to a depth of 3 cm. However, radiotherapy cannot be recommended at present as an alternative to surgery for occult nodal disease. Preoperative radiotherapy The use of preoperative neoadjuvant therapies has been investigated in an attempt to down-stage tumours and facilitate ‘viscera-preserving’ surgery. Small studies using preoperative radiotherapy have reported encouraging results in patients with advanced stage III and IV disease.65*,66* The largest study consisted of 48 cases, 11 of which had recurrent disease.67* Some of these cases had radiotherapy after surgery but most were treated preoperatively. The projected 5-year survival for the 37 primary cases was 75.6 per cent and for the recurrent cases was 62.6 per cent. Only two patients subsequently underwent an exenterative procedure or stoma formation. The value of preoperative radiotherapy for involved inguinal nodes has also been demonstrated. In recent years attention has focused on the use of neoadjuvant chemo-radiotherapy with the chemotherapy being used as a radiosensitizer rather than as a cytotoxic agent. Most of these studies have been carried out on only small numbers of patients. Koh et al.68* treated 20 patients with bulky cancers of the vulva with radiation doses to a maximum of 70.4 Gy to bulk disease and 54 Gy to areas at risk of microscopic spread combined with 5-FU (six patients also received either cisplatin or mitomycin C). A response rate of 90 per cent was reported with ten pathological complete responses and eight pathological partial responses at viscera-preserving surgery. Gerszten et al.69*
804 Carcinoma of the vagina and vulva
reported treating 18 patients with a twice daily regimen of radiotherapy, cisplatin and 5-FU followed by surgery with a complete response rate of 13 patients. Similarly Geisler et al.70* treated 10 patients with advanced vulval carcinoma involving the anal sphincter and/or urethra with cisplatin and 5-FU followed by radical surgery. They have reported conservation of the anus and the urethra in all the patients with a 100 per cent response rate. By far the largest study is the multi-centre GOG study reported by Moore et al. where a planned split course of concurrent cisplatin/5-FU and twice daily radiotherapy was used prior to surgery in 73 women with stage III or IV vulvar carcinoma. Using this strategy, only 2/71 patients had residual unresectable disease and in only three women was it not possible to preserve urinary and/or gastrointestinal continence.71* It would therefore appear that preoperative radiotherapy, either alone or in combination with chemotherapy, may be useful for advanced tumours to try to downsize the tumour in order to preserve urinary and anal sphincter control. PRIMARY RADICAL NON-SURGICAL THERAPY
The place of radiotherapy alone in the treatment of vulval carcinoma has not been studied systematically. Data accruing from reports of non-surgical management of squamous carcinoma of the anus have led to an evaluation of the possible role of primary radical non-surgical treatment of carcinoma of the vulva based on the treatment strategy for anal carcinoma.72* Thomas et al.54* treated nine patients with radiotherapy to a dose of 45–60 Gy with concomitant 5-FU and mitomycin C. Complete response was achieved in six of the women although there were three local recurrences. After appropriate surgical salvage, seven of the nine patients remained disease free at a median of 20 months follow-up. Two other studies have yielded equivalent results with similar treatment protocols, although cisplatin was substituted for mitomycin C by Berek et al.73* in 12 patients and the majority of patients (14 of 20) received 5-FU alone in the study by Koh et al.68* Based on the success of treating anal carcinoma, definitive radical radiotherapy to the vulva and inguinal nodes is usually given with concurrent 5-FU, the latter as a 4 day infusion in the first and fourth or fifth weeks of radiotherapy. The maximum radiation dose is 65 Gy. CONCLUSIONS
Radiotherapy can be delivered safely to the vulva and regional lymph nodes with or without concomitant chemotherapy. The role of chemotherapy in the treatment of carcinoma of the vulva has not yet been clearly established. The use of radiotherapy for early-stage carcinoma of the vulva should be restricted to adjuvant therapy to the vulva and nodes in patients at high risk of local recurrence after surgery. There is no evidence to support the use of inguinal nodal irradiation as a replacement for bilateral inguinal lymph node dissection.
In advanced FIGO stage III or IV disease the use of preoperative radiotherapy and chemo-radiotherapy in an attempt to facilitate viscera-preserving surgery appears promising.
Results Published 5-year survival rates for invasive cancer vary widely due to the low incidence of this tumour and the advanced age of the majority of these patients. Probably the most reliable information comes from a GOG study74* which defined risk groups for 377 women with cancer of the vulva depending on size of tumour and groin node status, which on multifactorial analysis were the only variables associated with prognosis. The risk was minimal in lesions measuring 2 cm or less and with no groin node involvement. Nearly all these patients were alive at 5 years. Women at low risk were those with lesions larger than 2 cm but less than 8.1 cm with no nodal involvement. They had a survival rate at 5 years of 87 per cent, as did those with lesions less than or equal to 2 cm in size but with a single node involved. Fiveyear survival dropped to 70 per cent for those at intermediate risk which included lesions larger than 8 cm but with no groin node involvement and those with smaller lesions but with one or two nodes positive for metastatic disease. At high risk was any woman whose lesion was greater than 8 cm with two positive nodes, any size lesion with three or more ipsilateral nodes involved or with bilateral node involvement. The survival in this group at high risk was 29 per cent. The effect on 5-year survival rate of lymph node involvement taken as a single prognostic variable showed that a negative node status conferred a survival rate of 92 per cent falling to 75 per cent with ipsilateral node involvement and 30 per cent with bilateral node involvement. If more than two nodes were involved the survival rate was 25 per cent, falling to zero if more than six nodes were positive.
Uncommon tumours of the vulva MELANOMA OF THE VULVA
Approximately 5 per cent of melanomas in women occur on the vulva and it is the second most common carcinoma of the vulva. Melanin production is variable and the lesions range from black to completely amelanotic. The most usual presenting complaint is of a lump or an enlarging mole. Pruritus and bleeding are less common. The prognosis is strongly related to the depth of invasion.75*,76*,77* Because of the absence of a well-defined papillary dermis in much of the vulval skin, levels of invasion as defined by Clark and co-workers76* are unsuitable and measurement of the thickness of the lesion as suggested by Breslow77* is more commonly used. Breslow has five levels of invasion, from the surface epithelium to the point of deepest penetration. Level 1 is 0.76 mm in thickness,
Carcinoma of the vulva 805
level 2 to 1.5 mm, level 3 to 3 mm, level 4 to 4 mm, and level 5 deeper than 4 mm. Local invasion occurs in an outward direction as well as downward so excision margins must be very wide, 3–5 cm being suggested for all but the most superficial lesions. This usually requires a radical vulvectomy without lymphadenectomy unless there is clinical evidence of groin disease. If the groin nodes are removed, the operation should be performed en bloc rather than through separate incisions because of the melanoma’s propensity to spread unseen by lateral intra-dermal infiltration. Other forms of treatment such as radiotherapy, chemotherapy and immunotherapy have had very little impact on this disease although newer immunotherapeutic agents have shown some promise in modifying the disease progress. The prognosis for level 1 melanoma of the vulva is very good as nodal involvement is unlikely. However, approximately one third of patients have groin lymph node metastases at presentation and 2.6 per cent have distant spread. While 5-year survival is approximately 56 per cent if the nodes are negative, the survival rate falls to 14 per cent when the nodes are positive.78 Involvement of urethra or vagina, or the presence of satellite lesions all worsen the prognosis. PAGET’S DISEASE AND APOCRINE ADENOCARCINOMA OF THE VULVA
This is an uncommon condition, similar to that found in the breast. Pruritus is the presenting complaint. The lesion is indistinguishable clinically from squamous intra-epithelial neoplasia and the diagnosis must be made by biopsy. In approximately one third of patients there is an adenocarcinoma in the apocrine glands. This has a poor prognosis if the groin lymph nodes are involved with no survivors at 5 years. The treatment of Paget’s disease is wide local excision, usually involving total vulvectomy because of the propensity of this condition to involve apparently normal skin. In some patients this may entail a simple vulvectomy. The specimen must be examined histologically with great care to exclude an apocrine adenocarcinoma. Excluding underlying adnexal carcinomas, concomitant genital malignancies are found in 15–25 per cent of women with Paget’s disease of the vulva.79 These are most commonly vulval or cervical, but transitional cell carcinoma of the bladder (or kidney), and ovarian, endometrial, vaginal and urethral carcinomas have all been reported.
place of lymphadenectomy is debatable as lymph node metastases are uncommon. Radiotherapy was considered to be contraindicated in verrucous carcinomas as it was thought to result in dedifferentiation of the tumour, but this is no longer considered to be true. BASAL CELL CARCINOMA
This tumour is rarely found on the vulva. Wide local excision gives excellent results in most cases. BARTHOLIN’S GLAND CARCINOMA
Usually an adenocarcinoma, this tumour may be squamous, transitional cell type, or even mixed squamous and adenocarcinoma. It has often spread widely to pelvic and groin nodes before the diagnosis is made. It must be distinguished from adenoid cystic carcinoma which is similar to the tumour found in salivary glands and which seldom gives rise to metastatic disease. The treatment is surgery but, because of its deep origin, part of the vagina, levatores ani and the ischio-rectal fat must be removed. SARCOMAS
These rare tumours of the vulva include leiomyosarcomas, which tend to grow slowly and metastasize late, and rhabdomyosarcomas, which are rapidly growing, aggressive tumours. A radical vulvectomy and groin node dissection is the usual treatment but local recurrence and bloodborne metastases are common. The prognosis for sarcomas of the vulva is the same as for sarcomas in other parts of the body. There are very few long-term survivors.
KEY LEARNING POINTS ●
●
●
●
●
VERRUCOUS CARCINOMA OF THE VULVA
This slowly growing neoplasm is rarely seen on the vulva. Both macroscopically and histologically it resembles condyloma acuminata and the diagnosis can be difficult. Generous biopsies are required to provide sufficient material for the pathologist. The treatment is surgery, usually a radical vulvectomy but very occasionally wide local excision. The
● ●
Vaginal intra-epithelial neoplasia is becoming more common. There is a risk in post-hysterectomy patients of disease hidden above the suture line in the vault. Surgical excision is the most effective treatment. Partial vaginectomy is often required for patients who have had a previous hysterectomy or radiotherapy. Brachytherapy has a place in women who have not had prior radiotherapy but can cause ovarian failure and vaginal stenosis. Invasive vaginal cancer is a rare tumour more often seen in association with an antecedent cervical malignancy. Radiotherapy is the main treatment method. Interstitial therapy alone offers good cure rates in early stages I–IIa occurring in the lower vagina. External radiotherapy is required for more advanced stages. Vaginal cancer occurring
806 Carcinoma of the vagina and vulva
●
●
●
●
●
●
●
at the vault of the vagina is treated in a similar fashion to cancer of the cervix. Vulval intra-epithelial neoplasia is becoming more common, especially in young women. The treatment must be carefully tailored to the individual to avoid mutilating therapy whenever possible. In view of the uncertainty about the malignant potential of these lesions, there may be a place for careful observation. However, it must not be forgotten that some of these patients will develop cancer of the vulva if untreated so the importance of close follow-up must be emphasized to the patient and her general practitioner. The main problems with carcinoma of the vulva are delay in presentation and diagnosis, and inadequate initial therapy. Surgery remains the cornerstone of treatment but, this can be made less extensive than in the past for early-stage disease. Even when radical surgery is necessary, new techniques have reduced the morbidity enormously. Adjuvant radiotherapy is indicated for patients with metastatic groin node disease and to prevent local recurrence when the tumour-free margin is less than 8 mm. Radiotherapy will probably play a larger role in the future management of vulval carcinoma postoperatively to prevent local recurrence and preoperatively to reduce tumour size in patients with advanced disease.
REFERENCES 1 Cancer Registration Statistics. England: Office for National Statistics, 2003. ◆2 Daling J, Sherman KJ. Relationship between human papillomavirus infection and tumours of anogenital sites other than the cervix. IARC Sci Publ 1992; 119:223–41. 3 Bjorge T, Dilner J, Anttila T, et al. Prospective seroepidemiological study of the role of human papillomavirus in non-cervical anogenital cancers. BMJ 1997; 315:646–9. 4 Hopkins MP, Morley GW. Squamous cell carcinoma of the neovagina. Obstet Gynecol 1987; 69:525–7. 5 Sugase M, Matsukura T. Distinct manifestations of human papillomaviruses in the vagina. Int J Cancer 1997; 72:412–5. 6 Futoran RJ, Nolan JF. Stage I carcinoma of the uterine cervix in patients under 40 years of age. Am J Obstet Gynecol 1976; 125:790–7. 7 Choo YC, Anderson DG. Neoplasms of the vagina following cervical carcinoma. Gynecol Oncol 1982; 14:125–32.
8 Barrie JR, Brunschwig A. Late second cancers of the cervix after apparent successful initial radiation therapy. Am J Roentgenol 1970; 108:109–12. 9 McFarlane M, Feinsvein AR, Horwitz R. Diethylstilbestrol and clear cell vaginal carcinoma. Reappraisal of the epidemiological evidence. Am J Radiat Oncol Biol Phys 1986; 81:855–863. 10 Herbst AL, Ulfelder H, Poskanzer DC. Adenocarcinoma of the vagina: association of maternal stilbestrol therapy with tumour appearance in young women. N Engl J Med 1971; 284:878–81. 11 Coppleson M. The DES story. Med J Aust 1984; 141:487–9. 12 Robboy SJ, Noller KL, O’Brien P, et al. Increased incidence of cervical and vaginal dysplasia in 3,980 Diethylstilbestrolexposed young women. JAMA 1984; 252:2979–83. 13 Aho M, Vesterinen E, Meyer B, et al. Natural history of vaginal intraepithelial neoplasia. Cancer 1991; 68:195–197. 14 Perez CA, Arneson AN, Galakatos A, Samanth HK. Malignant tumours of the vagina. Cancer 1973; 31:36–44. 15 Soutter WP. Vaginal intraepithelial neoplasia and colposcopy of the vagina. In: A practical guide to colposcopy. Oxford: Oxford Medical Publications, 1993, 144–60. 16 International Federation of Obstetrics and Gynaecology. Annual report on the results of treatment in gynaecological cancer, volume 24. J Epidemiol Biostat 2001; 6:141. 17 Gallup D, Talledo OE, Shah KJ, Hayes C. Invasive squamous cell carcinoma of the vagina: a 14-year study. Obstet Gynecol 1987; 69:782–5. 18 Dodge JA, Eltabbakh GH, Mount SL, Walker RP, Morgan A. Clinical features and risk of recurrence among patients with vaginal intraepithelial neoplasia. Gynecol Oncol 2001; 83:363–9. 19 Stock R, Mychalczak B, Armstrong JG, et al. The importance of brachytherapy technique in the management of primary carcinoma of the vagina. Int J Radiat Oncol Biol Phys 1992; 24:747–53. 20 Mock UKH, Fellner C, Knocke TH, Potter R. High-dose-rate (HDR) brachytherapy with or without external beam radiotherapy in the treatment of primary vaginal carcinoma: long-term results and side effects. Int J Radiat Oncol Biol Phys 2003; 56(4):950–7. 21 Al-Kurdi M, Monoghan JM. Thirty two years’ experience in the management of primary tumour of the vagina. BJOG 1981; 88:1145–1150. 22 Kucera H, Vavra N. Primary carcinoma of the vagina: clinical and histopathological variables associated with survival. Gynecologica Oncol 1991; 40:12–16. 23 Meffert J, Davis BM, Grimwood RE. Lichen sclerosus. J Am Acad Dermatol 1995; 32:393–416. 24 Fishman D, Chambers SK, Schwartz PE, Kohorn EI, Chambers JT. Extramammary Paget’s disease of the vulva. Gynecol Oncol 1995; 56:266–70. 25 Ragnarsson-Olding B, Johansson H, Rutqvist LE, Ringborg U. Malignant melanoma of the vulva and vagina. Trends in incidence, age distribution, and long-term survival among 245 consecutive cases in Sweden 1960–1984. Cancer 1993; 71:1893–7.
References 807
26 Jones R, Mclean MR. Carcinoma in situ of the vulva: a review of 31 treated and five untreated cases. Obstet Gynecol 1986; 68:499–503. 27 Wharton JT, Gallagher S, Rutledge FN. Microinvasive carcinoma of the vulva. Am J Obstet Gynecol 1974; 118:159–62. 28 Rutledge FN, Smith JP, Franklin EW. Carcinomas of the vulva. Am J Obstet Gynecol 1970; 106:1117–30. 29 Friedrich EG, Wilkinson EJ. The vulva. In: Blaustein A (ed.) Pathology of the Female Genital Tract. New York: SpringerVerlag, 1982, 13–58. 30 Iversen T, Abeler V, Aalders J. Individualised treatment of stage I carcinoma of the vulva. Obstet Gynecol 1981; 57:85–9. 31 Parker RT, Duncan I, Rampone J, Creasman W. Operative management of early invasive epidermoid carcinoma of the vulva. Am J Obstet Gynecol 1975; 123:349–55. 32 Hoffman JS, Kuma NB, Morley GW. Microinvasive squamous carcinoma of the vulva: search for a definition. Obstet Gynecol 1983; 61:615–18. 33 Monoghan JM. Mangement of vulval carcinoma. In: Shepherd JH, Monoghan JM (eds) Clinical Gynaecological Oncology. London: Blackwells, 1985, 133–53. 34 Hacker NF, Berek JS, Lagasse LD, Neiberg RK, Leuchter RS. Individualisation of treatment for stage 1 squamous cell vulvar carcinoma. Obstet Gynecol 1984; 63:155–62. 35 Andreasson B, Bock JE. Intraepithelial neoplasia in the vulval region. Gynecol Oncol 1985; 21:300–5. 36 Caglar H, Delago G, Hreshchyshyn MM. Partial and total skinnning vulvectomy in treatment of carcinoma in situ of the vulva. Obstet and Gynecol 1986; 68:504–7. 37 Townsend DE, Levine RU, Richart RM, Crum CP, Ptrilli ES. Management of vulval intraepithelial neoplasia by carbon dioxide laser. Obstet Gynecol 1982; 60:49–52. 38 Dorsey JH. Skin appendage involvement and vulval intraepithelial neoplasia. In: Sharp F, Jordan JA (eds) Laser Surgery. New York: Perinatology Press, 1986, 193–5. 39 Mene A, Buckley CH. Involvement of the vulval skin appendages by intraepithelial neoplasia. BJOG 1985; 92:634–8. 40 Cavanagh D, Ruffolo EH, Marsden DE. Cancer of the vulva. In: Cavanagh D, Ruffolo EH, Marsden DE (eds) Gynaecological Cancer – a Clinicopathological Approach. Connecticut: Appleton-Century-Crofts, 1985, 1–40. ●41 Le T, Hicks W, Menard C, Hopkins L, Fung MF. Preliminary results of 5% imiquimod cream in the primary treatment of VIN grade 2/3. Am J Obstet Gynecol 2006; 194(2):377–80. ●42 Way S. Carcinoma of the vulva. Am J Obstet Gynecol 1960; 79:692–8. ●43 Taussig F. Cancer of the vulva – an analysis of 155 cases (1911–1940) Am J Obstet Gynecol 1940;40: 764–79. 44 Monaghan JM. Radical surgery for carcinoma of the vulva. In: Monaghan JM (ed.) Bonney’s Gynaecological Surgery. 9th ed. Eastbourne: Bailliere Tindall, 1986, 121–8. ●45 Hacker NF, Leuchter RS, Berek JS, Castaldo TW, Lagasse LD. Radical vulvectomy and bilateral inguinal lymphadenectomy through separate groin incisions. Obstet Gynecol 1981; 58:574–9.
●46
●47
48
❄49
50
51 52
●53
●54
55
◆56
◆57
58
59
60
61
Siller B, Alvarez RD, Connor WD, et al. T2/3 vulva cancer: a case control study of triple incision versus en bloc radical vulvectomy and inguinal lymphadenectomy. Gynecol Oncol 1995; 57(3):335–9. Helm C, Hatch K, Austin JM, et al. A matched comparison of single and triple incision techniques for the surgical treatment of carcinoma of the vulva. Gynecol Oncol 1992; 46:150–6. Grimshaw R, Murdoch, JB, Monaghan JM. Radical vulvectomy and bilateral inguinal-femoral lymphadenectomy through separate incisions: Experience with a hundred cases. Int J Gynecol Cancer 1993; 3:18–23. RCOG. Clinical Recommendations for the Management of Vulval Cancer. London: Royal College of Obstetricians and Gynaecologists, 2006. Andersen B. Predicting sexual and psychological morbidity and improving quality of life for women with gynaecological cancer. Cancer 1993; 71:1678–90. Hacker NF. Current management of early vulvar cancer. Am Acad Med Singapore 1998; 27:688–92. Stehman F, Bundy BN, Dvoretsky PM, Creasman WT. Early stage I carcinoma of the vulva treated with ipsilateral inguinal lymphadenectomy and modified radical hemivulvectomy. A prospective study of the Gynecologic Oncology Group. Obstet Gynecol 1992; 79:490–7. Homesley HD, Bundy BN, Sedlis A, Adcock L. Radiation therapy versus pelvic node resection for carcinoma of the vulva with positive groin nodes. Obstet Gynecol 1986; 68:733–40. Thomas G, Dembo A, DePetrillo A, et al. Concurrent radiation and chemotherapy in vulvar carcinoma. Gynecol Oncol 1989; 34:263–7. Sebag-Montefiore D, McLean C, Arnott SJ, et al. Treatment of advanced carcinoma of the vulva with chemoradiotherapy – can exenterative surgery be avoided? Int J Gynecol Cancer 1994; 4:150–155. Harrington K, Lambert HE. Current issues in the nonsurgical management of primary vulvar squamous cell carcinoma. Clin Oncol 1994; 6:331–6. Thomas G, Dembo AJ, Bryson SCP, et al. Changing concepts in the management of vulvar cancer. Gynecol Oncol 1991; 42:9–21. Hoffman M, Greenberg S, Greenberg H, et al. Interstitial radiotherapy for the treatment of advanced or recurrent vulvar and distal vaginal malignancy. Am J Obstet Gynecol 1990; 162:1278–1282. Soutter W, Lambert HE, McIndoe GAJ. Carcinoma of the vulva and its putative precursors. In: Peckham M, Pinedo H, Veronesi U (eds) Oxford Textbook of Oncology. Oxford University Press: 1995, 1383–94. Bryson S, Dembo AJ, Colgan TJ, et al. Invasive squamous cell carcinoma of the vulva. Defining low and high risk groups for recurrence. Int J Gynecol Cancer 1991; 1:25–31. Hopkins M, Reid GC, Vettrano I, et al. Squamous cell carcinoma of the vulva. Prognostic factors influencing survival. Gynecol Oncol 1991; 43:113–7.
808 Carcinoma of the vagina and vulva
62 Heaps J, Yao SF, Montz FJ, et al. Surgical-Pathologic variables predictive of local recurrence in squamous cell carcinoma of the vulva. Gynecol Oncol 1990; 38:309–14. 63 Hacker N, Van der Velden J. Conservative management of early vulvar cancer. Cancer 1993; 71:1673–77. ●64 Stehman F, Bundy BN, Thomas G, et al. Groin dissection versus groin radiation in carcinoma of the vulva. A Gynecologic Oncology Group study. Int J Radiat Oncol Biol Phys 1992; 24:389–96. 65 Rotmensch J, Rubin SJ, Sutton HG, et al. Preoperative radiotherapy followed by radical vulvectomy with inguinal lymphadenectomy for advanced vulvar carcinomas. Gynecol Oncol 1990; 36:181–4. 66 Hacker NF, Berek JS, Juillard GJF, Lagasse LD. Preoperative radiation therapy for locally advanced vulvar cancer. Cancer 1984; 54:2056–61. 67 Boronow R, Hickman BT, Reagan MT, et al. Combined therapy as an alternative to exenteration for locally advanced vulvovaginal cancer. Am J Clin Oncol 1987; 10:171–81. 68 Koh W, Wallace HJ, Greer BE, et al. Combined radiotherapy and chemotherapy in the management of local-regionally advanced vulvar cancer. Int J Radiat Oncol Biol Phys 1993; 26:809–16. 69 Gerszten K, Selvaraj RN, Kelley J, Faul C. Preoperative chemoradiation for locally advanced carcinoma of the vulva. Gynecol Oncol 2005; 99:640–4. 70 Geisler J, Manahan KJ, Buller RE. Neoadjuvant chemotherapy in vulvar cancer: avoiding primary exenteration. Gynecol Oncol 2006; 100:53–7. ●71 Moore DH, Thomas GM, Montana GS, et al. Preoperative chemoradiation for advanced vulvar cancer: a phase I study
72
73
●74
75
76
77
78
79
of the Gynaecologic Oncology Group. Int J Radiat Oncol Biol Phys 1998; 42:79–85. Nigro N, Vaitkevicius VK, Considine B. Combined therapy for cancer of the anal canal a preliminary report. Dis Colon Rectum 1974; 27:354–6. Berek J, Heaps JM, Fu YS, et al. Concurrent cisplatin and 5-fluorouracil chemotherapy and radiation therapy for advanced-stage squamous carcinoma of the vulva. Gynecol Oncol 1991; 42:197–201. Holmsley H, Bundy BN, Sedlia A, et al. Assessment of current International Federation of Gynecology and Obstetrics staging of vulvar carcinoma relative to prognostic factors for survival (a Gynecological Oncology Group Study). Am J Obstet Gynecol 1991; 164:997–1004. Podratz KC, Gaffey TA, Symmonds RE, Johansen KL, O’Brien PC. Melanoma of the vulva: an update. Gynecol Oncol 1983; 16:153–68. Clark WH, From L, Bernadino EA, Mihm MC. The histogenesis and biologic behaviour of primary human malignant melanomas of the skin. Cancer Res 1969; 29: 705–26. Breslow A. Thickness, cross-sectional areas and depth of invasion in the prognosis of cutaneous melanoma. Ann Surg 1970; 172:902–8. Morrow CP, DiSaia PJ. Malignant melanoma of the female genitalia: a clinical analysis. Obstet Gynecol Surv 1976; 31:233–71. Degefu S, O’Quinn AG, Dhurandhar HN. Paget’s disease of the vulva and urogenital malignancies: a case report and review of the literature. Gynecol Oncol 1986; 25:347–54.
36 Gestational trophoblastic tumours PETER SCHMID AND MICHAEL J. SECKL
Introduction Genetics and pathology Epidemiology and aetiological factors Human chorionic gonadotrophin Clinical features Investigation
809 810 812 814 815 817
INTRODUCTION
Management Prognosis Summary Key references References
820 828 828 829 829
Syncytiotrophoblast Trophoblast Cytotrophoblast
Gestational trophoblastic disease (GTD) was probably first described as ‘dropsy of the uterus’ around 400 BC by Hippocrates and his student, Diocles.1 In 1276 the attendants of Margaret Countess of Henneberg noticed that her abnormal delivery consisted of multiple hydropic vesicles. They probably believed that each vesicle was a separate conception, which led them to christen half John and half Mary. Marie Boivin (1773–1841), who worked as a Parisian midwife, was the first to document the chorionic origin of the hydatids.1 In 1895 Marchand described a malignant uterine disease of syncytial and cytotrophoblastic origin and made the link between hydatidiform mole and other forms of pregnancy.1 Normal gestational trophoblast arises from the peripheral cells of the blastocyst in the first few days after conception. Trophoblastic tissue initially grows rapidly into two layers: 1. an inner cytotrophoblast of mononucleated cells which migrate out and fuse, forming 2. an outer syncytiotrophoblast of large multinucleated cells (Fig. 36.1). The latter subsequently aggressively invades the endometrium and uterine vasculature, generating an intimate connection between the fetus and the mother, known as the placenta. Invasion is, of course, one of the features of malignancy and, indeed, normal trophoblast can even be detected, by polymerase chain reaction, in the maternal circulation.2 Fortunately, complex biological and
Blastocyst
hCG
Figure 36.1 Schematic diagram of an embryo at the blastocyst stage, demonstrating trophoblast development.
immunological mechanisms involved in controlling the relationship between the fetal trophoblast and the maternal host prevent such circulating trophoblast from producing metastases. When GTD arises, the normal regulatory mechanisms controlling trophoblastic tissue are lost. Thus, the excessively proliferating trophoblast may invade through the myometrium, developing a rich maternal blood supply with tumour emboli and haematogenous spread occurring frequently. The World Health Organization has classified GTD into two pre-malignant diseases, termed complete and partial hydatidiform mole (CM and PM), and the four malignant disorders, or gestational trophoblastic tumours (GTT) – invasive mole, gestational choriocarcinoma, placental-site trophoblastic tumour and epithelioid trophoblastic tumour.3 The latter three conditions are also collectively known as gestational trophoblastic neoplasia (GTN). The origins, pathology and clinical behaviour of
810 Gestational trophoblastic tumours
these various forms of GTD are different and will be discussed below. These tumours are important to recognize because they are nearly always curable and, in most cases, fertility can be preserved. This is mainly because: 1. GTTs are exquisitely chemosensitive; 2. they all produce human chorionic gonadotrophin (hCG), a serum tumour marker with a sensitivity and accuracy in screening, monitoring, management and follow-up of patients which is unparalleled in cancer medicine; 3. detailed prognostic scoring has permitted ‘fine-tuning’ of treatment intensity so that each patient only receives the minimum therapy required to eliminate her disease.
GENETICS AND PATHOLOGY Complete hydatidiform mole Genetic analysis of CMs has revealed that they nearly always only contain paternal DNA and are therefore androgenetic.4 This occurs in most cases because a single sperm bearing a 23X set of chromosomes fertilizes an ovum lacking maternal genes and then duplicates to form the homozygote, 46XX4–10 (Fig. 36.2a). However, in up to 25 per cent of CMs fertilization can take place with two spermatozoa, resulting in the heterozygous 46XY or 46XX configuration11,12 (Fig. 36.2b). A 46YY conceptus has not yet been described and is presumably non-viable. Interestingly,
Sperm
23X
Maternal DNA lost from ovum
Sperm
some maternal elements are retained in CM, including the mitochondrial DNA.13 Very rarely, a CM can arise from a fertilized ovum which has retained its maternal nuclear DNA and is therefore biparental in origin.14,15 Macroscopically, the classical CM resembles a bunch of grapes, due to generalized (complete) swelling of chorionic villi. Villous blood vessels are generally lacking and fetal tissues are absent. However, this appearance is only seen in the second trimester, and the diagnosis is usually made earlier these days, when not all the villi are so hydropic. Indeed, in the first trimester the villi microscopically contain little fluid, are branching and consist of hyperplastic syncytio- and cytotrophoblast with many vessels. This characteristic appearance is lost only with increasing gestational age, when the villi lose their vessels and the stromal cisterns fill with fluid. Although it was previously thought that CM produced no fetal tissue, histology from firsttrimester abortions reveals evidence of embryonic elements, including fetal red cells.16–18 This has resulted in many CMs being incorrectly labelled as PMs. Consequently, the reported rate of persistent GTD after PMs has been artificially elevated and is probably less than 0.5 per cent.19 The presence of embryonic tissue from a twin pregnancy comprising a fetus and a CM is a further source of error which can lead to the incorrect diagnosis of PM. Sometimes, there may be diagnostic confusion between CM and a hydropic abortion. Recent work using p57 immunostaining, however, has provided a solution to this difficulty. This molecular marker is imprinted and expressed by the maternal genome. Since CMs are androgenetic, the
23X
Sperm
Maternal DNA lost from ovum
23X
23X
Two paternal genetic contributions
Duplication of haploid sperm
(a)
23X 23X
23X
23X
Maternal and two paternal genetic contributions
46XX
46XX
69XXX
Proliferation of monospermic androgenetic complete HM
Proliferation of dispermic androgenetic complete HM
Proliferation of triploid partial HM
(b)
(c)
Figure 36.2 Schematic diagram showing that the androgenetic diploid complete HM is formed either by duplication of the chromosomes from a single sperm (a) or by two sperm fertilizing the ovum (b) which, in both cases, has lost its own genetic component. The triploid genetic origin of a partial HM is demonstrated in (c).
Genetics and pathology 811
CM intervillous trophoblast is negative whilst the surrounding maternal tissue or villi from a hydropic abortion are positively stained for p57.20
Partial hydatidiform mole Partial hydatidiform moles are genetically nearly all triploid and, rarely, tetraploid, with at least two paternal chromosome sets but also some maternal contribution (Fig. 36.2c). Although triploidy occurs in 1–3 per cent of all recognized conceptions, and in about 20 per cent of spontaneous abortions with abnormal karyotype, triploids due to two sets of maternal chromosomes do not become PMs.21,22 Flow cytometry, which can now be done in formalin-fixed, paraffin-embedded tissues,23 can therefore help in differentiating CM from PM, and PM from diploid non-molar hydropic abortions.24 Although a variety of reports suggested that diploid PM exists, genetic analysis of lesions suspected to be such has not supported this suggestion. In general, a diploid molar gestation is believed to be a complete hydatidiform mole.25 In PMs, swelling tends to be less intense than in CM and affects only some villi (partial). Thus, two populations of villi usually exist in a partial mole. Both swollen and non-swollen villi can have trophoblastic hyperplasia, which is mild and focal.26,27 Nuclear atypia is infrequent. The villi have characteristic indented or scalloped outlines and round inclusions.16,17 An embryo is usually present and can be recognized macroscopically or inferred from the presence of nucleated red cells in villous vasculature. It may survive into the second trimester, but in most cases it dies at about 8–9 weeks’ gestation, and this is followed by loss of vessels and stromal fibrosis. In PMs evacuated early, villus swelling and trophoblastic excess can be so mild and focal that the diagnosis of PM may be missed.19 Indeed, at uterine evacuation for a ‘miscarriage’, it is likely that many PMs are misclassified as products of conception. Fortunately, we only see about one patient per year with persistent GTD related to a previously unrecognized PM. Of the increasing number of PMs that are correctly diagnosed, very few go on to develop persistent GTD. Indeed, in approximately 3000 PMs reviewed and followed at Charing Cross between 1973 and 1997, only 15 (0.5 per cent) have required chemotherapy.28 Consequently, it would be useful to have some way of identifying immediately post evacuation which patients will subsequently require chemotherapy. This would avoid all patients with a PM undergoing hCG follow-up. The potential for new omics technologies such as expression microarrays (genomics) and metabonomics clearly warrants investigation in this arena.
Other pregnancies mistaken for partial hydatidiform mole Over half of first-trimester non-molar abortions are due to trisomy, monosomy, maternally derived triploidy and
translocations. These often develop hydrops, but this is small (3 mm) and PM can be excluded if they are diploid on flow cytometry. Syndromes such as Turner’s, Edwards’ and Beckwith–Wiedemann can also cause histological confusion with PMs.19
Invasive hydatidiform mole This term is applied when a CM or, rarely, a PM invades into the myometrium. Invasive mole is common and is clinically identified by the combination of an abnormal uterine ultrasound (US) and a persistent or rising hCG level following uterine evacuation. Except under unusual circumstances when a curettage specimen contains myometrium with invasive molar villi, the diagnosis of an invasive mole can only be made on a hysterectomy specimen. The pathological confirmation of this condition is however rarely required. Moreover, repeat dilatation and curettage (D&C) is often contraindicated because of the risks of uterine perforation, infection, life-threatening haemorrhage and subsequent hysterectomy. In occasional cases where histology is available, invasive mole can be distinguished from choriocarcinoma by the presence of chorionic villi.
Choriocarcinoma Gestational choriocarcinoma is the most aggressive form of gestational trophoblastic disease. Most choriocarcinomas have been shown to have grossly abnormal karyotypes with diverse ploidies and several chromosome rearrangements, none of which is specific for the disease.29 Studies of the origin of GTTs have confirmed that choriocarcinoma may arise from any type of pregnancy, including a normal term pregnancy,30–33 from a homozygous CM or from a heterozygous CM.33,34 Until recently, it has been thought that PMs cannot give rise to choriocarcinoma. However, work from our group has provided incontrovertible genetic evidence proving that PMs can indeed transform into choriocarcinomas.28 This is important as there are some centres that wrongly believe it is safe to discontinue hCG follow-up following the diagnosis of PMs. Interestingly, choriocarcinoma may not always be due to the antecedent pregnancy.28 A patient with a history of a CM 4 years previously developed a choriocarcinoma following the delivery of a twin pregnancy. Using PCR to amplify short tandem repeat polymorphisms in DNA, this tumour was shown to be genetically identical with the previous CM.35 Like invasive hydatidiform mole (HM), obtaining tissue to make a formal histological diagnosis of choriocarcinoma is often not appropriate and so doubt frequently exists whether patients have one or the other form of GTT. Choriocarcinoma is highly malignant in behaviour, appearing as a soft, purple, largely haemorrhagic mass. Microscopically it mimics an early implanting blastocyst, with central cores of mononuclear cytotrophoblast
812 Gestational trophoblastic tumours
surrounded by a rim of multi-nucleated syncytiotrophoblast and a distinct absence of chorionic villi. Significant cytological atypia is in place, and diffuse hCG immunostaining of the syncytiotrophoblastic cells is characteristic. There are extensive areas of necrosis and haemorrhage and frequent evidence of tumour within venous sinuses. Interestingly, the disease fails to stimulate the connective tissue support normally associated with tumours and induces hypervascularity of the surrounding maternal tissues. This probably accounts for its highly metastatic and haemorrhagic behaviour.
Placental-site trophoblastic tumour Placental-site trophoblastic tumour (PSTT) is a rare form of gestational trophoblastic disease that has been proposed to arise from the intermediate trophoblasts at the implantation site.36 PSTT has been shown to follow term delivery, nonmolar abortion, or CM. A recent case showed furthermore that PSTT can develop after a PM.37 As with choriocarcinoma, the causative pregnancy may not be the immediate antecedent pregnancy.35 Genetic analysis of some PSTTs has demonstrated that they are mostly diploid, originating from either a normal conceptus, and therefore biparental, or androgenetic from a CM.29,38–42 A tetraploid PSTT has been described.40 In the normal placenta, placental-site trophoblast is distinct from villous trophoblast and infiltrates the decidua, myometrium and spiral arteries of the uterine wall. Recapitulating the normal implantation-site trophoblasts, PSTT display a pattern of vascular invasion, characterized by neoplastic cells migrating through, and replacing, vessel walls while maintaining the overall vascular architecture. PSTTs are rare, slow-growing malignant tumours composed mainly of intermediate trophoblast derived from cytotrophoblast, and so produce little hCG. However, they often stain strongly for human placental lactogen (hPL) and 1-glycoprotein. Elevated Ki-67 levels may help in distinguishing PSTT from a regressing placental nodule or exaggerated placental-site reaction.43 Whereas a low level of Ki-67 labelling (1 per cent) is seen in an exaggerated placental site, PSTTs generally show a higher index (10 per cent). In contrast to other forms of GTT, spread tends to occur late by local infiltration and via the lymphatics, although distant metastases can occur. More than 5 mitoses per 10 high-power fields may predict tumours with metastasizing potential.42
Epithelioid trophoblastic tumours Epithelioid trophoblastic tumour (ETT) is a recently described neoplastic proliferation of intermediate trophoblast that is thought by some investigators to be distinct from PSTT and choriocarcinoma. It has been proposed that ETT arises from the intermediate trophoblasts of the
chorionic leaves.36 Histologically, ETT displays a relatively uniform, nodular proliferation of intermediate-sized epithelioid trophoblasts, forming nests and cords. Islands of trophoblast are typically surrounded by areas of hyalinization or eosinophilic debris simulating tumour cell necrosis, resembling keratinous material in a squamous cell carcinoma. ETT can furthermore be associated with focal replacement of the cervical glandular epithelium with stratified neoplastic cells, simulating squamous cervical intraepithelial neoplasia. The cells are positive for cytokeratin, epithelial membrane antigen and inhibin-, whereas trophoblastic markers hPL, hCG and melanoma cell adhesion molecule are only focally expressed.44 Whether ETT is really a distinct disease entity in the GTD spectrum remains a contentious issue. Indeed, we have seen this appearance only in patients who have previously had chemotherapy for invasive mole or choriocarcinoma so this may simply represent the differentiating effects of treatment (as seen with germ cell tumours) rather than a separate type of GTT.
EPIDEMIOLOGY AND AETIOLOGICAL FACTORS Hydatidiform mole INCIDENCE AND ETHNIC ORIGIN
From population-based studies, the incidence of CM in countries of the Western world is approximately 0.5–1/1000 pregnancies. Although the incidence was previously reported to be at least twofold higher in the Far East,45 recent results indicate that this has fallen towards the stable levels found in Europe and North America.46,47 The declining incidence in the Far East may be due to environmental factors, such as dietary change. The incidence of PM has been under-estimated in the past.45,48 More recent studies indicate a much higher rate of PM relative to CM in patients presenting with spontaneous abortions.21 Indeed, the incidence of triploidy amongst firsttrimester spontaneous abortions suggests that as many as 90 per cent of PM could go undiagnosed.49 AGE
A positive relationship has been found between the risk of molar pregnancy and both upper and lower extremes of maternal age (45 years and 15 years, respectively). This association, although present for both CM and PM, is much greater for CM at all maternal ages, and the degree of risk is much greater with older (45 years) rather than younger ( 15 years) maternal age.50 In one study the relative increased risk for developing a CM, compared to the lowest rate between 25 and 29 years, was sixfold in girls under 15 years, threefold between 40 and 45 years, 26-fold between 45 and 49 years and more than 400-fold over 50 years of age.51
Epidemiology and aetiological factors 813
PREVIOUS PREGNANCIES
There is no increased risk for CM associated with increasing gravidity. However, the risk of a subsequent pregnancy being a CM rises from 1 in 1000 to 1 in 76 with one previous CM and to 1 in 6.5 with two previous CMs.51 Similar results have been reported for PM, with the risk for subsequent molar pregnancies rising to 1 in 59 for women with one previous PM.52 Thus, the occurrence of a conception complicated by CM or PM is associated with a 20-fold increase in risk of molar disease in a subsequent pregnancy. Therefore patients with a previous CM must be followed up after each subsequent pregnancy to confirm that their hCG levels return to normal.
Choriocarcinoma The incidence of choriocarcinoma following term delivery without a history of CM is approximately 1:50 000. However, CM is probably the most common antecedent to choriocarcinoma, being 29–83 per cent in various studies across the world.53 Consequently, the overall incidence of choriocarcinoma is much higher. Proof of this is frequently difficult or inadvisable to obtain, but when histology has been available these tumours were identified as choriocarcinoma in 3 per cent and invasive mole in 16 per cent of previous CMs. Rarely, PMs can give rise to choriocarcinoma or PSTT. Unlike HM, choriocarcinoma does not exhibit any clear geographical trends in incidence but the effect of age remains important. Curiously, choriocarcinoma following a full-term pregnancy is more likely to be associated with a very aggressive disease course than after a CM. This may reflect the fact that some of the choriocarcinomas after a CM included cases of invasive mole.
Placental-site trophoblastic tumours There are currently approximately 200 recorded cases of this tumour in the literature and so estimates of its true incidence may well be quite inaccurate (42,54–56). Nevertheless, PSTT is thought to constitute about 1 per cent of all trophoblastic tumours (choriocarcinoma, invasive mole and PSTT).
Other factors with uncertain role Several other factors have been examined for a possible aetiological role in the development of GTD. These include parity, smoking, contraceptive practice, dietary influences, herbicide and radiation exposure. None of these has been shown conclusively to play a role in this group of diseases.57 However, a recent analysis has suggested that oral contraceptive use prior to conception slightly increases the risk of subsequently developing a GTT.58 In contrast,
a meta-analysis failed to show clear evidence for an association between oral contraceptive use during post-molar follow-up period and the incidence of GTT.59
Genetic factors All autosomal genes consist of two alleles (paternal and maternal). Recent work has shown that some alleles are expressed only from one parent and not the other – a phenomenon called genomic imprinting. Interestingly, three closely related genes which are imprinted and located on chromosome 11p15 may be involved in GTT development and in other overgrowth syndromes.60 These are: H19, a putative tumour suppressor gene,61 and p57 kip2, a cyclindependent kinase inhibitor,62 which are both normally expressed by the maternal allele; and the paternally expressed insulin-like growth factor II (IGF-II), a growth factor commonly implicated in tumour proliferation.63 While p57 kip2 showed the expected pattern of expression in CM and choriocarcinoma,64 CM and post-mole tumours were unexpectedly found to express H19,65–67 and some post-term tumours showed bi-allelic expression of both H19 and IGF-II.68,69 This suggests that loss of the normal imprinting patterns of these genes may be an important factor in the development of GTT. The recent identification of rare families in which several sisters have repeat CMs which are biparental in origin14 might help to shed further light on the genes involved in CM formation. Indeed, linkage and homozygosity analysis suggested that in two families there is a defective gene located on chromosome 19q13.3–13.4.70 Interestingly, this is a region which at least one imprinted gene, PEG3,71 has already been mapped. Moreover, further families have been mapped to this area, refining the region of interest on 19q to a gene-rich area of about 1 megabase.72 Exciting recent work has now suggested that at least in two families NALP7 is the gene within this region that carries mutations which may be responsible for the disorder.73 Unfortunately, there is no mouse homologue for this gene so it will not be simple to further investigate the function of NALP7 in animal models. Moreover, genotyping and haplotype analysis in another two families failed to demonstrate mutations on chromosome 19q13.4, providing further evidence to support the hypothesis that familial recurrent hydatidiform moles show genetic heterogeneity.74 Other non-imprinted genes may be involved in the development of GTTs and carry prognostic significance. Patients whose blood group is B or AB, whose partners are either group O or A, may have a worse prognosis.75 The reasons for this are not clear.76,77 Loss of heterozygosity studies have shown that the putative tumour suppressor gene located on chromosome 7 is lost in GTTs.78 However, we have not been able to reproduce these findings (own unpublished observations). The use of SNP arrays, CGH arrays and similar technologies should now rapidly advance our progress in these areas.
814 Gestational trophoblastic tumours
Growth factors There is considerable evidence implicating growth factors as important agents capable of driving tumour proliferation.79 This has been particularly well defined for small-cell lung cancer, where both neuropeptide and polypeptide growth factors may drive cell proliferation in an autocrine and paracrine fashion.79,80 In contrast, the identity and role of growth factors in trophoblastic disease remain poorly understood. It is quite plausible that hCG may function as an autocrine/paracrine growth factor for GTTs. Indeed, recent work looking at the function of hCG suggests that a hyperglycosylated isoform not only triggers invasive/ proliferative behaviour of normal trophoblast during early pregnancy but also promotes the invasion and proliferation of choriocarcinoma cells both in vitro and in vivo.81 In addition, the ligands and/or the receptors for known growth factors, such as hepatocyte growth factor, platelet-derived growth factor, c-fms, c-erbB2, IGF-II and granulocyte colony-stimulating factor (G-CSF), have been demonstrated in trophoblastic tissue or cell lines.82–86 Interestingly, we have frequently administered G-CSF to patients with GTD without any apparent compromise in therapeutic outcome, indicating that clinically important tumour growth is not induced by G-CSF. However, the role of these and other growth factors warrants further investigation in this group of diseases.
Risk of gestational trophoblastic tumours following complete or partial hydatidiform moles The risk of malignant sequelae following evacuation of a PM is probably far less than 1:200, compared to 1:16 for a CM.28,87 Other workers have reported somewhat higher rates of GTT developing after a PM, with values from 2 to 6 per cent.88,89 The differences in these studies may have arisen because the diagnosis was made using morphological criteria without the added benefit of cytogenetics to help discriminate between CM and PM. Since it is not yet possible to predict in advance which patients with a CM or PM will develop persistent GTD, all of them must be registered for hCG monitoring. Following this strict protocol enables the identification of individuals with persistent trophoblastic growth who could benefit from lifesaving chemotherapy.
HUMAN CHORIONIC GONADOTROPHIN Molecular background and function The family of pituitary/placental glycoprotein hormones includes hCG, follicle-stimulating hormone (FSH), luteinizing hormone (LH) and thyroid-stimulating hormone (TSH).
Each hormone comprises an α-subunit, which is common between the family members, and a distinct β-subunit, which is believed to confer receptor specificity. Consequently, assays to measure hCG are directed against the β-subunit. The crystal structures for both the α- and β-subunits of hCG have been solved.90,91 The genes for βhCG and βLH have been mapped to a complex cluster of inverted and tandem genes on chromosome 19, of which six are for βhCG and the terminal seventh for βLH.92,93 However, the precise role of each of the genes, which can all be transcribed, remains unclear.94 Interestingly the gene is highly conserved in nature, being found in prokaryotes.95 So far, only one receptor for βhCG has been identified in mammals, which also serves as the main receptor for LH.96,97 This, of course, raises the question of how hCG might exert specific effects distinct from LH. Furthermore, βhCG is now known to circulate in a variety of forms as a result of post-translational modifications of its structure such as hyperglycosylated hCG (hCG-H), nicked hCG, nicked hyperglycosylated hCG, hCG missing the beta-subunit Cterminal extension, free alpha-subunit, free beta-subunit, free beta-subunit missing the C-terminal extension, hyperglycosylated-free beta-subunit (βhCG-H) and nicked-free beta-subunit.98 It is therefore possible that additional βhCG receptors exist. The function(s) of the intact and the modified forms of βhCG remain obscure. During a normal pregnancy the only forms of hCG found are intact and hCG-H, the latter being present solely during the first trimester. In contrast, in malignancy multiple forms of hCG can be found. Whatever the functions of these different forms of hCG, the fact that hCG production clinically indicates trophoblastic proliferation suggests that this molecule may be a growth factor for normal placenta, GTD and other hCG-producing tumours. Indeed, hCG-H produced during the first trimester of normal pregnancy promotes trophoblast proliferation and invasion into the myometrium and exerts similar effects on choriocarcinoma cells.
hCG Assays The measurement of hCG during pregnancy is relatively straightforward since the molecule is either intact and/or hyperglycosylated so the antibodies used only need to detect these two forms. In contrast, the assays used to detect hCG produced by cancer should ideally be able to recognize intact, all fragments of βhCG and βhCG-H. Moreover, the assay should not produce either false negative or false positive results. Currently, commercial assays used to measure βhCG are mostly designed to recognize intact hCG or βhCG so they are good for pregnancy monitoring and indeed are licensed for this purpose. However, none of these assays is approved for use in cancer and, moreover, most of them fail to recognize all known forms of βhCG. This means that the assays are prone to false negatives. The commercial assays are all based on the sandwich or two-site principle
Clinical features 815
using a capture antibody recognizing one site on hCG and another antibody binding to a second site on the hormone for detection. The mechanism of detection used in these assays is also variable (e.g. complement fixation, enzymeand fluorescence-based) and this, combined with differences in antibody specificity and sensitivity, results in considerable differences in assay results, particularly when the assays are used in cancer patients.98 Added to this is a lack of agreed international standards for some of the hCG isoforms with, for example, no standards for hCG-H. So what does all of this mean? Essentially, it is difficult to ensure direct comparability between different hCG assays in anything other than pregnancy. In addition, it means that there is a clear need for an hCG assay which can be safely used to detect this hormone in cancer patients. At Charing Cross, we have for many years been using a non-commercial competitive one-site radio-immunoassay (RIA). Quite fortuitously, the antibody used in this assay detects a site on βhCG which is common to all the known forms of this hormone and so does not have a false negative problem. However, single-site RIA and the commercial two-site sandwich assays share a problem of false positive detection. This usually occurs as a consequence of heterophile antibodies, for example anti-mouse or anti-rabbit antibodies, produced by the patient. Fortunately, these do not cross into the urine so simple measurement of the urine hCG can confirm whether an elevated serum result is a true positive. In addition, heterophile antibodies, in contrast to true hCG, do not produce results that diminish appropriately when the serum is serially diluted. At Charing Cross, all hCG assays are automatically set up with serial dilutions and all patients have their hCG measured in serum and urine. Consequently, we do not see problems with false positives. This contrasts with many other centres which depend on commercial assays that are frequently not set up in dilution and where urine is not routinely assessed in parallel with serum. Nevertheless, the Charing Cross RIA is slower to generate results, taking 24 hours rather than the commercial assays which generally take just a few hours. Moreover, the RIA requires radioactivity. Whilst speed of results is not usually clinically important, the use of radioactivity is becoming increasingly unacceptable. Therefore, there remains a clear need to develop a new hCG assay specifically for use in cancer and we are currently actively exploring this area.
Use as a tumour marker Human chorionic gonadotrophin, which has a half-life of 24–36 hours, remains the most sensitive and specific marker for trophoblastic tissue. However, hCG production is not confined to pregnancy and GTD. Indeed, hCG is produced by any trophoblastic tissue found, for example, in germ cell tumours and in up to 15 per cent of epithelial malignancies.99 The hCG levels in such cases can be just as high as those seen in GTD. Furthermore, the levels of hCG
produced by GTD are frequently identical to those found in normal pregnancy, although very high levels outside the range for a twin pregnancy may lead to suspicion of a trophoblastic tumour. Consequently, hCG measurements per se do not reliably discriminate between pregnancy, GTD or non-gestational trophoblastic tumours. However, serial measurements of hCG have revolutionized the management of GTD, for several reasons. Thus, the amount of hCG produced correlates with tumour volume, so that a serum hCG of 5 iu/L corresponds to approximately 104–105 viable tumour cells. Consequently, these assays are several orders of magnitude more sensitive than the best imaging modalities available today. In addition, hCG levels can be used to determine prognosis.75 Serial measurements allow progress of the disease or response to therapy to be monitored (Fig. 36.3). Development of drug resistance can be detected at an early stage, which facilitates appropriate management changes. Estimates may be made of the time for which chemotherapy should be continued after hCG levels are undetectable in serum, in order to reduce the tumour volume to zero. For these reasons hCG is the best tumour marker known, and is of wider interest as a model for the way in which tumour markers may be used in other diseases.
CLINICAL FEATURES Complete and partial moles Complete or partial moles most commonly present in the first trimester as a threatened abortion with vaginal bleeding. Patients with CM may rarely notice the passing of grape-like structures (vesicles) and occasionally the entire mole may be spontaneously evacuated. The uterus may be of any size, but is commonly large for gestational age. Patients with marked trophoblastic growth and high hCG levels are particularly prone to hyperemesis, toxaemia and the development of theca lutein cysts, which may sometimes be palpable above the pelvis. Toxaemia was diagnosed in 27 per cent of patients with CM in an old series,100 but is rarely seen today because of early ultrasound diagnosis. Convulsions are similarly rare. The high hCG levels may also produce hyperthyroidism because of cross-reactivity between hCG and TSH at the TSH receptor.101 Although pulmonary, vaginal and cervical metastases can occur, they may spontaneously disappear following removal of the mole. Thus the presence of metastases does not necessarily imply that invasive mole or choriocarcinoma has developed. Patients may present with acute respiratory distress not only because of pulmonary metastases and/or anaemia, but occasionally as a result of tumour embolization. The risk of embolization is reduced by avoiding agents that induce uterine contraction before the cervix has been dilated to enable evacuation of the CM. Patients with PM usually do not exhibit the dramatic clinical features characteristic of CM.102 The uterus is often
816 Gestational trophoblastic tumours
Mole evacuated
Start treatment
Stop treatment
1000 000
Serum hCG (iu/L)
100 000
10 000
1000
Key MTX
100
10
1
0
50
100
150
Days
Figure 36.3 The use of monitoring the serum hCG concentration following evacuation of a hydatidiform mole (HM). In this case, after an initial fall the hCG started to rise, indicating the development of invasive HM or choriocarcinoma, and so the patient was called up for staging. The prognostic score was low risk (see Table 36.5) and the patient was successfully treated with methotrexate (MTX) and folinic acid (see Table 36.6).
not enlarged for gestational age and vaginal bleeding tends to occur later, so that patients most often present in the second trimester with a missed or incomplete abortion. In fact the diagnosis is rarely suspected until the histology of curettings is available. Clues to the diagnosis can occasionally be obtained by ultrasound.103 The pre-evacuation hCG is 100 000 iu/L at diagnosis in over 90 per cent of cases. At present, a proportion of CMs and PMs still go undiagnosed either because of miscarriage at home or because termination centres do not carry out histopathological examination of all abortions.104 This can result in late presentation of disease, sometimes with life-threatening complications. Clearly, there is little that can be done about miscarriages at home. However, for women attending termination centres it may be possible to establish a screening procedure to help prevent subsequent problems from missed diagnosis.
Invasive moles
Twin pregnancies
Choriocarcinoma
Twin pregnancies comprising a normal fetus and a hydatidiform mole are estimated to occur in between 1:20 000 and 1:100 000 pregnancies. Some probably abort in the first trimester and so go undiagnosed. However, some are discovered on ultrasound examination, either routinely or because of complications such as bleeding, excessive uterine size or problems related to a high hCG.105
Choriocarcinoma can present after any form of pregnancy, but most commonly occurs after CM. In the latter situation, it is often not practical to obtain histological proof of choriocarcinoma and so it is impossible to distinguish it from invasive mole. Choriocarcinoma following an apparently normal pregnancy or non-molar abortion usually presents within a year of delivery, but in the Charing Cross
This is usually diagnosed because serial urine or serum hCG measurements reveal a plateaued or rising hCG level in the weeks after evacuation of the mole. Patients may complain of persistent vaginal bleeding or lower abdominal pains and/or swelling. This may occur as a result of haemorrhage from leaking tumour-induced vasculature as the trophoblast invades through the myometrium, or because of vulval, vaginal or intra-abdominal metastases. The tumour may also involve other pelvic structures, including the bladder or rectum, producing haematuria or rectal bleeding, respectively. Enlarging pulmonary metastases or tumour emboli growing in the pulmonary arteries can contribute to life-threatening respiratory complications.106 The risk of these complications is clearly higher in patients where the initial diagnosis of a molar pregnancy was missed and so are not on hCG follow-up.
Investigation 817
series the longest interval to date was 35 years (reviewed in reference 107; unpublished data). The presenting features may be similar to those of HM, with vaginal bleeding, abdominal pain and a pelvic mass. However, one third of all patients with choriocarcinoma present without gynaecological features, and instead have symptoms of distant metastases.107,108 In these cases lives can be saved by remembering to include choriocarcinoma in the differential diagnosis of metastatic malignancy (particularly in lungs, brain or liver) presenting in a woman of childbearing age. Any site may be involved, including skin, producing a purple lesion, cauda equina and the heart. Pulmonary disease may be parenchymal, pleural or may result from tumour embolism and subsequent growth in the pulmonary arteries.106,109 Thus respiratory symptoms and signs can include dyspnoea, haemoptysis and pulmonary artery hypertension. Cerebral metastases may produce focal neurological signs, convulsions, evidence of raised intracranial pressure and intracerebral or subarachnoid haemorrhage. Hepatic metastases may cause local pain or referred pain in the right shoulder. Although none of these presentations is specific to choriocarcinoma, performing a simple pregnancy test or quantitative hCG assay can provide a vital clue to the diagnosis. Other clues may come from features associated with a high circulating hCG level, including occasional thyrotoxicosis and ovarian theca lutein cysts.
Infantile choriocarcinoma Choriocarcinoma in the fetus or newborn is exceptionally rare, with approximately 30 reported cases.110,111 While a primary choriocarcinoma within the infant is possible, in 17 out of 30 cases the mother also had the tumour. This suggests that the disease can cross the placenta, bypass an immature immune system in the fetus and flourish. Interestingly, the diagnosis was often made in the neonate before the mother. In all cases, the infant was anaemic and had a raised hCG, but the site of metastasis was variable, including liver and lung in the majority and brain in 27 per cent of patients. Only a few patients have been treated successfully with platinum-based chemotherapy and surgery, the rest dying within weeks of initial diagnosis, which may have been delayed.112 Consequently, serum or urine hCG levels should be measured in all babies of mothers with choriocarcinoma. As the disease can present up to 6 months after delivery, an argument could be made for serial monitoring of hCG in these infants.
Placental-site trophoblastic tumour The slow growth rate of PSTT means that it can present years after term delivery, non-molar abortion or complete HM. Unlike choriocarcinoma, it tends to metastasize late in its natural history and so patients frequently present with gynaecological symptoms alone.113 In addition to
vaginal bleeding, the production of hPL by the cytotrophoblastic cells may cause hyperprolactinaemia, which can result in amenorrhoea and/or galactorrhoea. In some instances, presumably also as a result of tumour-secreted products, patients can develop nephrotic syndrome or haematuria with deposition of fibrinogen in the glomeruli. Disseminated intra-vascular coagulation has been reported in association with these features.114,115 Metastases may occur in the vagina, extra-uterine pelvic tissues, retroperitoneum, lymph nodes, lungs and brain.42,54,116,117
Epithelioid trophoblastic tumours There are currently only a few recorded cases of this recently described tumour in the literature. Although most patients with ETT are women of childbearing age, a significant percentage of women are peri- or post-menopausal with a remote history of pregnancy.118 Similar to PSTT, patients present with vaginal bleeding and low levels of serum β-hCG (2500 mIU/mL). In half of the reported cases, the tumours arise in the lower uterine segment or cervix and the distinction from a keratinizing squamous cell carcinoma can sometimes be difficult. Extra-uterine locations such as broad ligament as the primary site have also been observed.119 Similar to PSTT, ETT appear to respond relatively poorly to the conventional chemotherapy against GTD, and hysterectomy with close follow-up is appropriate until more experience of this rare lesion becomes available.
INVESTIGATION Human chorionic gonadotrophin As already discussed above, this is an extremely useful marker of trophoblastic disease and its response to therapy.
Plain chest X-ray All patients who are suspected of having gestational trophoblastic tumours should have a chest radiograph. The most common metastatic appearance is of multiple, discrete, rounded lesions (Fig. 36.4), but large solitary lesions or a miliary pattern can occur.120 Furthermore, tumour emboli to the pulmonary arteries can produce an identical picture to venous thromboembolism, with wedge-shaped infarcts and areas of decreased vascular markings. Pulmonary artery hypertension can cause dilatation of the pulmonary arteries. Deposits affecting the pleural surfaces may result in haemorrhagic pleural effusions (Fig. 36.4). These diverse radiological appearances have been reviewed extensively.120 Routine CT scanning of the chest does not add anything to the management of these cases.
818 Gestational trophoblastic tumours
Ultrasonography Ultrasound (US) and colour Doppler imaging is not diagnostic but highly suggestive of persistent GTD when there is a combination of a raised hCG, no pregnancy and a vascular mass within the uterus (Fig. 36.5). Detection rates are higher for complete compared to partial moles, and improve after 14 weeks’ gestation.121 Ultrasound permits accurate delineation of uterine volume, which correlates with the amount of disease and is a prognostic indicator (see GTT scoring
Figure 36.4 Chest radiograph showing multiple lung metastases and a small right pleural effusion in a patient with choriocarcinoma.
Figure 36.5 Ultrasonography with colour Doppler showing persistent gestational trophoblastic disease following a CM within the body and wall of the uterus. A typical vesicular or ‘snow storm’ appearance of residual molar tissue can be seen within the uterus, together with a rich blood supply through the endometrium and myometrium. There is no evidence of a fetus. (See Plate Section.)
system, Table 36.5). In the Charing Cross series, about 75 per cent of patients who require chemotherapy for molar disease have an enlarged uterus (120 mL).122 The tumour is frequently confined to one area of the uterine wall but diffuse uterine disease and extra-uterine deposits may be visualized. Doppler frequently demonstrates a change in the waveform of the uterine arteries (Fig. 36.5). This is attributed to large vascular channels forming in the myometrium, resulting in arteriovenous shunting.123 The uterine artery pulsatility index (UAPI), as an indirect measure of the functional tumour vascularity, has recently been shown to independently predict the response to chemotherapy in GTTs.124 Furthermore, the increased sensitivity of modern colour Doppler US now reveals abnormal blood vessel encroachment through the myometrium into the endometrium (Fig. 36.5). Although this has not been shown to have prognostic significance, the degree of vascular endometrial encroachment may aid in assessing the risk of major haemorrhage in patients with GTT.122 Interestingly, these vascular abnormalities can persist long after the disease has been eradicated with chemotherapy. Patients with repeated vaginal haemorrhage from these vascular malformations may need arteriography and selective embolization (Fig. 36.6). This usually prevents subsequent bleeding but may need to be repeated.125 Importantly, embolization of these abnormal uterine vessels does not appear to affect fertility.122 Pelvic US can also demonstrate ovarian theca lutein cysts and other ovarian masses. Metastatic spread outside the pelvis, for example to the liver or kidneys, can also be identified and shown to have an abnormal Doppler signal. On completion of therapy, the US should be repeated to confirm that the masses have disappeared, or at least reduced in size. This ensures that the masses did not originate from some other pathology.
Investigation of patients with drug-resistant disease When patients develop drug-resistant disease, further investigation is required to more accurately define where the residual tumour is located, as resection can be curative. Whole-body computed tomography (CT) is often helpful (Fig. 36.7) but, where available, magnetic resonance imaging (MRI) of the brain (Fig. 36.8) and pelvis is clearly superior to CT in these areas. Indeed, in our experience MRI of the head has revealed drug-resistant deposits of choriocarcinoma not seen on the CT, thus permitting curative surgery. If the MRI/CT brain is normal, then a lumbar puncture to measure the hCG level in cerebrospinal fluid can be useful to detect disease in the central nervous system. An hCG ratio greater than 1:60 of that found in the serum is highly indicative of the presence of trophoblastic disease.
Experimental imaging techniques Radiolabelled anti-hCG antibodies given intravenously can localize tumours producing hCG where other imaging
Investigation 819
techniques have failed, and has aided in the cure of patients.126 This approach is most effective when there are 106–107 tumour cells, i.e. the serum hCG is 100 iu/L, but even then both false positive and false negative results occur. Thus, anti-hCG scanning should be regarded as complementary to other imaging investigations. More recently, positron emission tomography (PET) has provided a novel approach to image many types of tumours using a variety of labels. Whole-body PET has already been reported to distinguish GTT emboli from blood clot in two
patients with choriocarcinoma.127 Other case reports demonstrate the potential value of PET in the differential diagnosis of undefined pulmonary nodules or in the localization of disease.128,129 However, the role of PET has yet to be confirmed in larger series of patients with GTTs.
Genetic analysis On some occasions it can be helpful to perform a comparative genetic analysis of the patient’s trophoblastic tumour
(a)
Figure 36.6 Arteriographic appearance of a uterine arteriovenous malformation before (left) and after selective embolization (right) in a patient with repeated vaginal haemorrhages following previous curative treatment for invasive HM. The patient’s bleeding subsequently stopped and she had a normal pregnancy.
(b)
Figure 36.7 CT scan demonstrating (a) lung and (b) brain metastases in a patient with choriocarcinoma.
820 Gestational trophoblastic tumours
with their normal tissue and, if available, that of their partner. Thus, if the tumour is suspected of being of nongestational origin, this can be confirmed by the presence of only maternal and the complete absence of paternal DNA. Genetic studies can also determine which of several antecedent pregnancies is the causal pregnancy of the current GTT. This can have an impact on determining appropriate therapy and prognosis.14 In addition, genetic studies can help identify whether a woman with repeated CM pregnancies has the rare biparental variant of CM. The latter condition is rarely compatible with a normal pregnancy and pre-implantation diagnosis is not yet possible. In contrast, a woman with repeated androgenetic CM can be offered in vitro fertilization and pre-implantation selection of unaffected concepti.
MANAGEMENT Molar evacuation Evacuation of the uterine cavity using suction gives the lowest incidence of sequelae. When the molar trophoblast invades the myometrium, it is relatively easy to perforate the uterus if a metal curette is used. Medical induction involving repeated contraction of the uterus induced by oxytocin or prostaglandin, or other surgical approaches including hysterectomy or hysterotomy, increases the risk
of requiring chemotherapy by two- to threefold compared with suction evacuation. This is thought to be because tumour is more likely to be disseminated by uterine contraction and manipulation. For similar reasons, the use of prostanoids to ripen a nulliparous cervix is not recommended. If bleeding is severe immediately after suction evacuation, a single dose of ergometrine to produce one uterine contraction may stem the haemorrhage and does not appear to increase the chance of requiring chemotherapy. In the past it has been common practice for gynaecologists to perform a second and sometimes a third evacuation of the uterine cavity in patients with a molar pregnancy. However, analysis of our data has shown that the chance of requiring chemotherapy after one evacuation is only 2.4 per cent but rises markedly to 18 per cent after two evacuations and 81 per cent after four evacuations (Table 36.1). Patients with an hCG plateau or with rising hCGs are at particularly high risk and 74 per cent of patients require chemotherapy despite undergoing a second evacuation (Table 36.2). Furthermore, high hCG values have been shown to be an indicator for low rates of benefit of a second evacuation with 80 per cent of patients with a pre-interventional hCG of 5000 subsequently requiring chemotherapy.130 Consequently, evacuations are not generally recommended because of the risk of complications and the high chance that the patient will require chemotherapy anyway. However, a second evacuation may be reasonable, if there is a clinical indication such as bleeding or if repeat US shows persisting molar trophoblast within the uterine cavity. The use of US control during this procedure may help to reduce the risk of uterine perforation. Table 36.2 hCG levels at second evacuation and subsequent requirement for chemotherapy hCG at second evacuation
Figure 36.8 The MRI appearances of the brain before (left) and after (right) EMA/CO chemotherapy (see Table 36.7) for metastatic choriocarcinoma.
1500 1500–5000 5000–10 000 10 000–20 000 20 000–50 000 50 000 Total
Patients requiring chemotherapy (%) 50 38 83 75 68 94 74
Table 36.1 Correlation between the number of evacuations performed following a HM and the subsequent requirement for chemotherapy at Charing Cross Hospital (1973–86) Number of evacuations 1 2 3 4
Patients not treated
Patients treated
% Patients treated
4481 1495 106 5
109 267 106 22
2.4 18 50 81
Management 821
Twin pregnancies At Charing Cross Hospital in London, we have seen 77 confirmed cases of CM with a separate normal conceptus.50 Fifty three of these women elected to continue their pregnancy whilst the remainder had a termination. Just over a third of women who continued their pregnancy achieved a live birth, while the remainder had non-viable pregnancies which ended mostly in spontaneous abortions. Interestingly, there was no difference in the risk of developing persistent GTD (a GTT) between women who underwent an early termination and those who elected to continue their pregnancy. Moreover, no woman in this series died either from a pregnancy-related complication such as toxaemia or from GTT. Thus, it appears reasonably safe to allow patients with twin pregnancies in which one of the conceptions is a CM to continue to term provided there are no other complications. This is in line with observations on singleton CM which suggest that later gestational age at termination does not increase the risk of subsequently requiring chemotherapy.104
Registration and follow-up after uterine evacuation The majority of patients require no more treatment after evacuation but 16 per cent of patients with CM and 0.5 per cent with PM develop persistent GTD. This can be invasive mole, choriocarcinoma or, very rarely, PSTT. It is vital that patients with persistent GTD are identified, as virtually all of them can be cured with appropriate therapy. In 1973, under the auspices of the Royal College of Obstetricians and Gynaecologists, a national follow-up service was instituted in the United Kingdom, whereby patients with GTD are registered with one of three laboratories, located in Dundee, Sheffield and London. Approximately 1400 women are registered per annum, and of these we treat 110–120 patients per annum. After registration, the patient’s details and pathology, together with two weekly blood and urine samples, are sent through the post to one of the reference laboratories for confirmation of diagnosis and serial hCG estimations. Following the success of this scheme, other countries have now established, or are attempting to establish, a similar registration programme to reduce their GTT mortality rates. Since a molar pregnancy is a pre-malignant condition, in the majority of cases the molar tissue dies out spontaneously. Consequently, the hCG concentration returns to normal. Once the hCG is within the normal range (4 iu/L), our recommendation is that the patient should not start a further pregnancy until the hCG has been normal for 6 months. We previously used to distinguish between those women whose hCG returned to normal within or beyond 8 weeks of evacuation. The former women had hCG surveillance discontinued after 6 months whilst the latter had to wait 2 years before attempting to have a child.
A review of our data indicated that this distinction was not actually necessary and now, providing the hCG is normal within 6 months of evacuation, women require only a further 6 months of normal values before attempting a new pregnancy. Since patients who have had a previous mole or GTT are more at risk of having a second, all patients should have a further estimation of hCG at 6 and 10 weeks following the completion of each subsequent pregnancy.
Indications for chemotherapy Factors associated with an increased risk of requiring chemotherapy are summarized in Table 36.3. The hormones in the oral contraceptive pill are probably growth factors for trophoblastic tumours and for this reason patients are advised not to use the pill until the hCG levels have returned to normal. The indications for intervention with chemotherapy in patients who have had a CM or PM are shown in Table 36.4. Human chorionic gonadotrophin values 20 000 iu/L 4 weeks after evacuation of a mole, or rising values in this range at an earlier stage, indicate that the patient is at increased risk of severe haemorrhage or uterine perforation with intra-peritoneal bleeding. These complications can be life-threatening and their risk can be reduced by starting chemotherapy. Metastases in the lung, vulva and Table 36.3 Factors increasing the risk of requiring chemotherapy following evacuation of a hydatidiform mole Factor Uterine size gestational age Pre-evacuation serum hCG level 100 000 iu/L Oral contraceptives given before hCG falls to normal Bilateral cystic ovarian enlargement
Reference Curry et al. (131) Berkowitz and Goldstein (100) Stone et al. (132)
Berkowitz and Goldstein (100)
Table 36.4 Indications for chemotherapy 1. Histological evidence of choriocarcinoma 2. Evidence of metastases in brain, liver or gastrointestinal tract, or radiological opacities 2 cm on chest X-ray 3. Pulmonary, vulval or vaginal metastases unless hCG falling 4. Heavy vaginal bleeding or evidence of gastrointestinal or intra-peritoneal haemorrhage 5. Rising hCG after evacuation 6. Serum hCG 20 000 iu/L more than 4 weeks after evacuation, because of the risk of uterine perforation 7. Raised hCG 6 months after evacuation even if still falling Any of the above are indications to treat following the diagnosis of GTD.
822 Gestational trophoblastic tumours
vagina can only be observed if the hCG levels are falling. However, if the hCG levels are not dropping, or the patient has metastases at another site, which can indicate the development of choriocarcinoma, chemotherapy is required.
Prognostic factors/scoring versus FIGO staging The principal prognostic variables for GTTs, which were originally identified by Bagshawe75 and since modified by the WHO and our own experience, are summarized in Table 36.5. Each variable carries a score which, when added together for an individual patient, correlates with the risk of the tumour becoming resistant to single-agent therapy. Thus, the most important prognostic variables carry the highest score and include: 1. the duration of the disease, because drug resistance of GTTs varies inversely with time from the original antecedent pregnancy;
Table 36.5a
2. the serum hCG concentration, which correlates with viable tumour volume in the body; 3. the presence of liver and/or brain metastases. Recent evidence demonstrates that liver metastases correlate with a worse prognosis than brain metastases.133 Consequently, patients with liver involvement now score six points rather than four. Since the ABO blood groups contribute little to the scoring system, and it is difficult to have complete data on both patients and relevant partners, this has now been dropped from the system. Anatomical staging systems such as that of the International Federation of Gynaecology and Obstetrics (FIGO) have been used by several centres managing GTT. Surgery is virtually never indicated in the initial management of this disease and the FIGO system does not appear to add anything in treatment planning to the existing scoring system. Furthermore, the original FIGO staging system did not always correctly predict prognosis, so that some patients were under- or over-treated.134 This problem was overcome by modifying the FIGO system to include some
Scoring system for gestational trophoblastic tumours (Charing Cross Score) Score
Prognostic factor Age (years) Antecedent pregnancy (AP) Interval (end of AP to chemotherapy in months) hCG iu/L Number of metastases Site of metastases Largest tumour mass Prior chemotherapy
0
1
2
6
39 Mole 4 103–104 0 None, lung, vagina – –
39 Abortion or unknown 4–7 103 1–3 Spleen, kidney 3–5 cm –
– Term 7–12 104–105 4–8 GI tract 5 cm Single drug
– – 12 105 8 Brain, liver – 2 drugs
The total score for a patient is obtained by adding the individual scores for each prognostic factor. Low risk, 0–8; high risk, 9. Patients scoring 0–8 currently receive single-agent therapy with methotrexate and folinic acid, while patients scoring 9 receive combination drug therapy with EMA/CO (see Table 36.7). CXH, Charing Cross Hospital; hCG, human chorionic gonadotrophin; GI, gastrointestinal.
Table 36.5b
Scoring system for gestational trophoblastic tumours (WHO/FIGO) Score
Prognostic factor Age (years) Antecedent pregnancy (AP) Interval (end of AP to chemotherapy in months) hCG iu/L Number of metastases Site of metastases Largest tumour mass Prior chemotherapy
0
1
2
4
40 Mole 4 103 0 Lung – –
40 Abortion 4–6 103–104 1–4 Spleen, kidney 3–5 cm –
– Term 7–13 104–105 5–8 GI tract 5 cm Single drug
– – 13 105 8 Brain, Liver
The total score for a patient is obtained by adding the individual scores for each prognostic factor. Low risk, 0–6; high risk, 7.
2 drugs
Management 823
of the WHO variables.135 An international committee was then established and in 2000 recommended a new combined FIGO/scoring system which was internationally accepted in 2002 so that all centres managing this rare group of diseases can more easily compare their results.136 We currently use both our own Charing Cross and the new combined system as outlined in Table 36.5a and b. In both systems, the score rather than the stage determines treatment and thus far there appears to be excellent concordance for assigning patients to either low or high risk groups (see Chemotherapy section below).
Chemotherapy At Charing Cross Hospital, we have used the prognostic scoring system in Table 36.5 to subdivide the patients into three groups, termed low (score 0–4), medium (score 5–8) and high risk (score 8), depending on their overall score. Formerly, each risk group corresponded with a separate treatment regimen and so there were three types of therapy, termed low, medium and high risk. In the early 1990s, we discontinued the medium-risk category and treatment for three reasons: 1. the short- and long-term toxicity of this treatment is probably not significantly different from that of highrisk therapy; 2. some patients treated with medium-risk therapy have developed drug resistance and subsequently required high-risk therapy anyway; and 3. about 30 per cent of medium-risk patients can still be cured on low-risk chemotherapy, which is less toxic than either medium- or high-risk chemotherapy.137 Moreover, there is no evidence that prior treatment failure with methotrexate is an adverse prognostic variable.133,138 Accordingly, patients who score between 5 and 8 are now offered the possibility of receiving low-risk chemotherapy, which was previously only given to those with a score 5. Patients scoring 9 or more are given high-risk treatment. In the new FIGO/scoring system the low-risk group score is up to 6, and those scoring 7 or more are high risk. The details of both low- and high-risk treatment are discussed below. Patients are admitted for the first 1–3 weeks of either therapy principally because the tumours are often highly vascular and may bleed vigorously in this early period of treatment. LOW-RISK PATIENTS
increasing the folinic acid (FA) rescue to 15 mg the incidence of these side effects drops to less than 2 per cent without obviously decreasing efficacy.138 Accordingly, we have for some years routinely used FA at this higher dose rather than the original 7.5 mg. If this approach fails, then giving the FA earlier, at 24 hours after methotrexate, can be helpful. Methotrexate can induce serositis, resulting in pleuritic chest pain or abdominal pain. Myelosuppression is rare, but a full blood count should be obtained before each course of treatment. Liver and renal function should also be monitored regularly. All patients are advised to avoid sun exposure or use complete sun block for 1 year after chemotherapy because the drugs can induce photosensitivity. About 33 per cent of low-risk patients according to the new classification need to change treatment: 2 per cent because of toxicity (usually mucositis, occasionally severe pleuritic pain or drug-induced hepatitis), and 31 per cent as a result of drug resistance, which occurs despite the patients being correctly scored as low risk.138 In the former middle-risk group up to 69 per cent of patients will develop drug resistance.139,140 Thus, careful monitoring for disease response and treatment-induced toxicity is required to ensure that these women achieve complete remission. Survival in these patients is excellent, even though they may need to change treatment. The only deaths in patients treated with this schedule following the introduction of the prognostic scoring system were one from concurrent but not therapy-induced non-Hodgkin’s lymphoma and one from hepatitis.140 Indeed, in a more recent analysis we found no evidence that single-agent therapy with methotrexate increases the risk of developing a second cancer.137 HIGH-RISK PATIENTS
These patients are at risk of developing drug resistance and we have therefore treated them since 1979 with an intensive regimen consisting of etoposide, methotrexate and actinomycin D (EMA), alternating weekly with cyclophosphamide and vincristine, otherwise known as Oncovin\rtm; (CO). Table 36.7 gives details of this therapy, which can be given to most patients with only one overnight stay in hospital every 2 weeks. A patient’s response to this therapy is Table 36.6 Chemotherapy regimen for low-risk and intermediate-risk patients Methotrexate/folinic acid
The regimen used since 1964 at Charing Cross Hospital and widely followed in other centres is shown in Table 36.6. Figure 36.3 illustrates a typical course of a patient who responds to this therapy. This schedule is in general well tolerated, with no alopecia. Some patients develop mucosal ulceration affecting the mouth and, much more rarely, the vaginal and perianal areas. This can be prevented partly by a high fluid intake of 3 L/day while on treatment. However, by
Methotrexate (MTX)
Calcium folinate (folinic acid)
50 mg by intramuscular injection repeated every 48 h for a total of 4 doses 15 mg orally 30 h after each injection of MTX
Courses repeated every two weeks, i.e. days 1, 15, 29, etc.
824 Gestational trophoblastic tumours
shown in Figure 36.9. The regimen is myelosuppressive but prolonged gaps in therapy which can permit tumour regrowth can usually be avoided by the following measures: continuing to treat unless the white cell count is less than 1.5 109/L and/or platelets fall below 60 109/L and/or mucosal ulceration develops. The introduction of Table 36.7 Chemotherapy regimen for high-risk patients EMA Day 1
Etoposide Actinomycin D Methotrexate
Day 2
Etoposide Actinomycin D Folinic acid rescue (starting 24 h after commencing the methotrexate infusion)
CO Day 8
100 mg/m2 by IV infusion over 30 min 0.5 mg IV bolus 300 mg/m2 by IV infusion over 12 h 100 mg/m2 by IV infusion over 30 min 0.5 mg IV bolus 15 mg IM or orally every 12 h for 4 doses
1 mg/m2 IV bolus 1(max. 2 mg) 600 mg/m2 IV infusion over 30 min
Vincristine Cyclophosphamide
EMA alternates with CO every week. To avoid extended intervals between courses caused by myelosuppression, it may occasionally be necessary to reduce the EMA by omitting the Day 2 doses of etoposide and actinomycin D.
G-CSF in patients who have a low neutrophil count also helps to maintain treatment intensity and has reduced the number of neutropenic febrile episodes. The cumulative 5-year survival of patients treated with this schedule is 86 per cent, with no deaths from GTT beyond 2 years after the initiation of chemotherapy.133 While these results are good, significant adverse prognostic variables continue to be a problem in subgroups of patients: presence of liver metastases, the longer the interval from the antecedent pregnancy, brain metastases and term delivery in the antecedent pregnancy. Early deaths accounted for a significant proportion of the overall mortality, the causes being: respiratory failure, cerebral metastases, hepatic failure and pulmonary embolism. None of these women had a hydatidiform mole and so were not registered for follow-up and close monitoring to prevent them presenting with extensive disease. Clearly, it will be difficult to improve the survival of this particular subgroup. However, any woman of childbearing age presenting with widespread malignancy should have an hCG measurement, as very high levels of this hormone are highly suggestive of choriocarcinoma. Earlier recognition of choriocarcinoma will help to reduce disease extent and consequent mortality. Surprisingly, previous chemotherapy was a good prognostic factor in our high-risk patients.133 This can be explained partly by the fact that these patients were all on follow-up and so the disease extent was less than that of many who had not had previous treatment. The long-term risk of chemotherapy-induced second tumours in patients treated for GTTs in our centre has recently been reviewed137 and is discussed below.
Key
100 000 EMA CO
Intrathecal MTX
Serum hCG (iu/L)
10 000
1000
100
10
1
0
50
100
150
Days
Figure 36.9 The fall in hCG and, by inference, the tumour response to EMA/CO chemotherapy in a patient with choriocarcinoma who scored within the high-risk group. The patient has pulmonary metastases and received intrathecal methotrexate with each course of CO (cyclophosphamide and vincristine) until the hCG levels were within the normal range.
Management 825
Management of drug-resistant disease Frequent measurement of the serum hCG is a simple way to detect drug resistance at an early stage as the hormone levels will stop falling and may start to rise long before there are other clinical changes. However, it is important that decisions to alter treatment are not made on the basis of a single hCG result but on a progressive trend over two to three values. In patients receiving methotrexate for low-/intermediate-risk disease, if the hCG is 100–300 iu/L when drug resistance occurs, the disease can often be cured simply by substituting actinomycin D (0.5 mg IV total dose daily for 5 days in every 2 weeks). This drug is more toxic than methotrexate, inducing some hair thinning (occasionally complete alopecia), myelosuppression and more oral ulceration. Low-risk risk patients failing methotrexate whose serum hCG is 100–300 iu/L are now all treated with the EMA/CO regimen outlined above. While this treatment has saved all low-risk patients, some high-risk patients develop drug-resistant disease. Fortunately, 70 per cent of the 47 patients who have failed EMA/CO were still salvaged by further chemotherapy and/or surgery.133 Indeed, the combination of surgical removal of the main site of drug resistance (usually in the uterus, lung or brain), together with chemotherapy, is particularly effective. Preoperative investigations include transvaginal or abdominal US Doppler of the pelvis, plain chest radiography, whole-body CT scan and MRI of the brain, lumbar puncture to measure hCG levels in the cerebrospinal fluid, and experimental imaging techniques such as anti-hCG or 18F-FDG-PET scanning. If all these investigations are negative, hysterectomy should be considered. When multiple possible sites of resistant disease are found, anti-hCG or 18F-FDG-PET imaging can potentially distinguish the biologically active from dead/necrotic lesions and so guide appropriate surgery. Following surgery, or when surgery is not appropriate, we use the cis-platinum-containing regimen, EP (etoposide 150 mg/m2 and cis-platinum 75 mg/m2 with hydration) alternating weekly with EMA (omitting Day 2 except the folinic acid). This is not an easy schedule to administer clinically, owing to both myelosuppression and complications associated with even minor renal impairment. Other options that can be considered include use of some of the new anti-cancer agents such as the taxanes, topotecan, gemcitabine and temozolomide. Paclitaxel has been shown to have activity in patients with germ cell tumours that have failed on prior treatment.141 Several cases of drug-resistant GTT that responded to paclitaxel-based single-agent or combination therapy have been reported.142–145 Current salvage regimens include paclitaxel/etoposide, alternating two weekly with paclitaxel/platinum or gemcitabine/platinum/ taxane combinations. Another approach in patients with refractory disease involves high-dose chemotherapy with autologous bone marrow or peripheral stem-cell transplantation. Patient selection here is probably important in determining
outcome as has been shown for refractory germ cell tumours where patients with drug-sensitive disease are the ones that stay in remission.146,147 In the largest series of high-dose chemotherapy for refractory GTT, 5 out of 11 patients showed a temporary partial or complete response, but all patients progressed eventually.148 High-dose chemotherapy with autologous stem-cell support for GTN remains therefore investigational. Further studies are needed to better define the role of high-dose chemotherapy, in particular in patients with high-risk GTN who fail their first salvage treatment for recurrent disease.
Management of acute disease-induced complications HAEMORRHAGE
Heavy vaginal or intra-peritoneal bleeding is the most frequent immediate threat to life in patients with trophoblastic tumours. The bleeding mostly settles with bed rest and after starting chemotherapy appropriate to the risk group. However, occasionally the bleeding can be torrential, requiring massive transfusion. In this situation if the bleeding is coming from the uterus, it may be necessary to consider a uterine pack or emergency embolization of the tumour vasculature.125 Fortunately, hysterectomy is rarely required. If the bleeding is intra-peritoneal and does not settle with transfusion and chemotherapy, laparotomy may be required. Indeed, patients occasionally present this way.
RESPIRATORY FAILURE
Occasionally patients present with respiratory failure due to multiple pulmonary metastases or, more rarely, as a result of massive tumour embolism to the pulmonary circulation.106,109 Fever with or without purulent sputum may be present and, in these cases, blood and sputum cultures should be obtained and antibiotics started to treat potential infection. If tumour embolus is suspected, a ventilation/perfusion scan, MRI or dynamic CT chest (Fig. 36.10) and an electrocardiogram should be obtained and a heparin infusion started. Pulse oximetry and/or arterial blood gas should be regularly measured to allow appropriate adjustment of oxygen therapy and to monitor any deterioration in pulmonary function that may occur following the start of chemotherapy. The latter occurs probably because of oedema and inflammation around tumour deposits that are becoming necrotic. To prevent this we usually commence therapy with only one or two drugs and introduce the other drugs once pulmonary function is stable. Occasionally patients require masked continuous positive airway pressure ventilation but mechanical ventilation has in our experience only saved one patient. The others have died from intra-pulmonary haemorrhage probably as a result of trauma to the tumour vasculature induced by
826 Gestational trophoblastic tumours
Figure 36.10 Contrast-enhanced CT scan of the thorax at the level of the main pulmonary arteries, showing a filling defect in the right main pulmonary artery (arrow). The patient presented with a brief history of increasing shortness of breath which had suddenly worsened. During the previous 18 months she had suffered from irregular heavy bleeding per vagina, had four separate positive pregnancy tests and two normal pelvic ultrasound investigations. She was successfully treated with EMA/CO chemotherapy, with some resolution of the changes seen on CT and ventilation perfusion scanning. Post-mortem examinations of similar cases have revealed that the filling defect in the main pulmonary artery is tumour embolus and not clot. (Reproduced from Seckl et al., 1991106 with permission from Elsevier Science.)
high positive airway pressures generated on mechanical ventilation. For this reason extra-corporeal oxygenation has been proposed.149
Management of cerebral metastases Involvement of the central nervous system (CNS) by GTT may either be overt and require intensive therapy or occult and need prophylaxis. Any patient with a GTT who has lung metastases is at risk of either having or developing CNS disease. Furthermore, the second most common site of metastases in high-risk patients is the CNS and nearly all these individuals had lung deposits.150 The presence of neurological symptoms and signs may alert the clinician to the presence of brain metastases. However, some high-risk patients do not have either overt pulmonary or CNS disease at presentation but subsequently develop cerebral metastases which are then drug resistant. Consequently, careful investigation of patients at risk of developing brain metastases is warranted so that appropriate CNS-penetrating chemotherapy is given rather than the standard low- or high-risk treatments. Investigations include CT and/or MRI of the brain and, in patients who do not have raised intracranial pressure,
measurement of the hCG levels in cerebrospinal fluid. A cerebrospinal fluid:serum ratio greater than 1/60 suggests the presence of CNS disease. Prophylaxis against possible CNS disease (MRI brain normal) is given to patients from all risk categories with lung metastases and all high-risk patients regardless of the absence or presence of lung deposits. The prophylaxis consists of 12.5 mg methotrexate administered intrathecally, followed 24 hours later by 15 mg of folinic acid orally. This is given with every course of low-risk therapy, or with each CO in the high-risk therapy for three doses. Since the introduction of this policy, the development of brain metastases without evidence of drug resistance elsewhere has been much less frequent.150 Overt CNS disease requires careful management as therapy can induce haemorrhage into the tumour, leading to a rise in intracranial pressure and subsequent loss of life.150 We and others have found that early resection of solitary brain deposits in patients with serious neurological signs can sometimes be life saving.151–153 In this respect, early consideration of the diagnosis may be helpful. Cerebral oedema can be reduced with high-dose corticosteroids and so patients are given 24 mg of dexamethasone in divided doses before starting chemotherapy. The EMA/CO regimen is modified by increasing the dose of methotrexate to 1 g/m2, given as a 24-h infusion on Day 1. The folinic acid rescue is increased to 30 mg given 8 hourly intravenously for 3 days, commencing 32 hours after the start of the methotrexate infusion. Provided there is no evidence of raised intracranial pressure, 12.5 mg of methotrexate is given intrathecally with each CO until the hCG in serum is normal. Modified EMA/CO is then continued for a further 6–8 weeks. Overall, in approximately 80 per cent of patients with cerebral metastases long-term remission can be achieved.154 Patients who survive the first 3 weeks of such treatment have a good prognosis, with an 86–89 per cent chance of cure.150,153,154 Patients who develop cerebral tumour during chemotherapy have a poor prognosis because their disease is almost certainly drug resistant. Nevertheless, a combination of immediate surgery to remove the deposit(s) and modified chemotherapy designed to provide better CNS penetration can be curative in this situation.150,153 Radiotherapy has been advocated as an alternative therapeutic approach. However, it has not been shown to eradicate tumour in its own right, and in combination with chemotherapy has produced less effective results than chemotherapy alone.150,153 Nevertheless, stereotactic radiotherapy probably has a role in the treatment of isolated deep lesions that cannot be removed surgically, especially if still present at the end of chemotherapy.
Management of placental-site trophoblastic tumours Placental-site trophoblastic tumours are biologically quite different from the other forms of GTD, producing little
Management 827
Table 36.8 Follow-up of patients with gestational trophoblastic tumours who have been treated with chemotherapy Low/medium/high risk post-chemotherapy patients, hCG concentration sampling Urine Year 1 Week 1–6 after chemotherapy Months, 2–6 Months, 7–12 Year 2 Year 3 Year 4 Year 5 After Year 5
Blood
Weekly 2-weekly 2-weekly 4-weekly 8-weekly 3-monthly 4-monthly 6-monthly
Weekly 2-weekly – – – – –
hCG, growing slowly, metastasizing late and being relatively resistant to combination chemotherapy regimens. Therefore, hysterectomy remains the treatment of choice, provided the disease is still localized to the uterus.42,54–56 When metastatic disease is present, despite reports to the contrary,38 individual patients can respond and be apparently cured by chemotherapy (EP/EMA), either alone or in combination with surgery.42 Apart from stage and myometrial invasion, the most important prognostic variable in these patients is the interval from the last pregnancy: where this is less than 4 years the prognosis is good, and where greater than 4 years the outlook is poor. Radiotherapy has produced mixed results and has not yet been proven to cure the disease. Because PSTT is so rare, it is unlikely that its treatment will ever be optimized.
deaths.155 Indeed, 71 per cent of women continued their pregnancy to term. Consequently, although we continue to advise women to avoid pregnancy for 1 year after completing chemotherapy, those that do become pregnant can be reassured of a likely favourable outcome. Similar results have been reported by others.156 When a patient becomes pregnant, it is important to confirm by US and other appropriate means that the pregnancy is normal. Followup is then discontinued until 3 weeks after the end of pregnancy when the hCG due to the pregnancy should have returned to normal. Patients who do not require chemotherapy following evacuation of their first mole, although not on life-long follow-up, should have their hCG levels measured 3 weeks and then 3 months after the end of any subsequent pregnancy. This is because they are at increased risk compared to the general population of a further molar pregnancy.52 In addition, with a subsequent pregnancy there is a small risk of reactivation of ‘dormant’ residual molar tissue, even if the pregnancy itself is normal. The follow-up of PSTT differs from that of other GTD variants, since serum hCG is a less reliable tumour marker and late recurrences are probably more common.42,54 Indeed, there are some reports suggesting that these tumours may fail to secrete hCG at relapse despite an extensive tumour burden.117 While this may be possible we have not seen this phenomenon in over 50 cases of PSTT except in a woman who was subsequently shown genetically not to have a gestational tumour. Another explanation for this surprising failure to detect hCG may be that some PSTTs secrete a form of the hormone which is not detected by the particular hCG assay being used by a given treatment centre. Nevertheless, patients in remission from PSTT should probably be followed up both serologically for life and at regular intervals in the clinic for 5 years. There is a clear need to develop an international database of PSTTs to help guide future management.
Patient follow-up after chemotherapy
Contraceptive advice
On completion of their chemotherapy, patients need to be followed up regularly for life, with hCG estimations to confirm that their disease is in remission. Initially the follow-up is with serum and urine samples (Table 36.8), but in due course the follow-up is only on urine samples. In the UK this is computerized and automatic reminders are sent to patients so that they do not get lost to follow-up. Patients are advised not to become pregnant until 12 months after completing their chemotherapy. This minimizes the potential teratogenicity of treatment and avoids confusion between a new pregnancy or relapsed disease as the cause of a rising hCG. Despite this advice, 230 women on follow-up at our centre between 1969 and 1998 have become pregnant during the first year. Fortunately, this did not appear to be associated with an increased risk of relapse or fetal morbidity and there were no maternal
Patients using oral contraceptives before the hCG is normal following evacuation of an HM have an increased incidence of sequelae requiring chemotherapy.132 In a more recent analysis of our patients registered between 1973 and 1989, we have found that the proportion of patients requiring chemotherapy after evacuation of their HM is 30.7 per cent, compared with 8 per cent in the overall population.157 For this reason, patients are advised to avoid the oral contraceptive pill until the hCG has returned to normal after removal of an HM. Patients who have had chemotherapy for their GTT are advised not to use the oral contraceptive pill until their hCG is normal and chemotherapy is completed. More data are needed about the importance of oral contraceptives but, until these are available, we feel that this is reasonable practice.
828 Gestational trophoblastic tumours
Long-term complications of therapy
SUMMARY
Most patients, including those who have received intensive chemotherapy, return to normal activity within a few months, and the majority of the side-effects are reversible. Complete hair regrowth is seen in all patients with chemotherapy-induced alopecia, although sometimes it may be initially curly rather than straight. Late sequelae from chemotherapy have been remarkably rare. We have recently analysed 15 279 patient years of follow-up for the late sequelae from chemotherapy and this confirms that patients treated with methotrexate and folinic acid in the low-risk category have no significant increase in the incidence of second tumours.137 In contrast, 26 patients receiving intensive combination chemotherapy for their gestational trophoblastic tumours developed another cancer, when the expected rate was only 16.5; a significant difference.137 Acute myeloid leukaemia (AML) (relative risk 16.6) accounted for the early tumours up to 5 years after completing chemotherapy. Subsequently, at 5–9 years an increase was seen in colorectal cancer (relative risk 4.6), at 10–14 years in melanomas, and beyond 25 years in breast cancer (relative risk 5.8). The increased risk of AML probably reflects the rising use of etoposide, which is now well recognized to induce this form of leukaemia.137,158–160 Fertility is an important issue in the management of patients with GTTs. Although chemotherapy appears to induce a menopause 3 years earlier than would normally be expected,161 fertility does not otherwise appear to be affected.162 In 392 patients receiving single-agent methotrexate, 327 (83.4 per cent) had successful live births. Interestingly, in the 336 patients receiving multi-agent chemotherapy, including EMA/CO, 280 (83.3 per cent) also succeeded in having normal pregnancies. Importantly, in this and a previous analysis there was no increase in the incidence of congenital malformations compared to the general population.162,163
In the past, many women have died from GTD. However, during the past 50 years we have learnt much about the biology, pathology and natural history of this group of disorders. Furthermore, accurate diagnostic and monitoring methods have been developed, together with effective treatment regimens. As a result, the management of GTD today represents one of the modern success stories in oncology, with few women dying from their trophoblastic tumours.
KEY LEARNING POINTS ●
●
●
●
PROGNOSIS All patients in the low-risk group can be expected to be cured of their GTTs since the introduction of etoposide.51,139 For high-risk patients, survival has progressively improved and is currently 86 per cent.133 The diagnosis of choriocarcinoma is often not suspected until the disease is advanced. As a result, some deaths occur before chemotherapy has a chance to be effective. The number of such patients can be diminished by a greater awareness of the possibility that multiple metastases in a woman of childbearing age may be due to choriocarcinoma. The simple measurement of the hCG level in such individuals is a very strong indicator of choriocarcinoma and could help to hasten referrals for lifesaving chemotherapy. However, patients in the high-risk group still die from drug-resistant disease and there remains a need to develop novel therapeutic approaches.
●
●
●
●
While it is well known that CMs can transform into choriocarcinomas, it had previously been thought that PMs do not do this. Recent work, however, has conclusively shown that PMs can also transform into choriocarcinoma, so all patients with PMs require hCG follow-up. All patients with suspected GTT should be discussed with a gestational trophoblastic disease centre and have their histology centrally reviewed. The use of serum and urine hCG measurements to follow the disease course has been a key feature in the successful management of women with GTT. However, recent work has shown that care is required in the interpretation of hCG values obtained from commercial kit assays. This is because (a) only a single epitope of hCG is detected in most kit assays and this epitope may be lost in some GTTs; and (b) some patients have anti-mouse antibodies which bind to the mouse monoclonal antibody used in kit assays, generating a false-positive hCG. Currently the most reliable hCG assay involves the use of a polyclonal rabbit antisera used in a radio-immunoassay which detects many different epitopes of hCG. Low-risk patients with GTT are initially treated with MTX/FA as this regimen has no long-term sequelae. Patients relapsing on MTX/FA are switched to actinomycin D if their hCG level is 100–300 iu/L or EMA/CO combination chemotherapy if the hCG is 100–300 iu/L. EMA/CO, like MTX/FA, does not appear to impair fertility, but there is a small increased risk of second tumours with EMA/CO. The long-term outcome for patients with GTT is excellent, with all low-risk patients and approximately 90 per cent of high-risk patients being cured.
References 829
KEY REFERENCES Newlands ES, Paradinas FJ, Fisher RA. Recent advances in gestational trophoblastic disease. Hematol Oncol Clin North Am 1998; 13:225–44. Paradinas FJ. The diagnosis and prognosis of molar pregnancy. The experience of the National Referral Centre in London. Int J Gynaecol Obstet 1998; 60:S57–64. Rotmensch S, Cole LA. False diagnosis and needless therapy of presumed malignant disease in women with false positive hCG concentrations. Lancet 2000; 355:712–15. Seckl MJ, Fisher RA, Salerno GA, et al. Choriocarcinoma and partial hydatidiform moles. Lancet 2000; 356:36–9.
REFERENCES 1 Ober WB, Fass RO. The esarly history of choriocarcinoma. Ann N Y Acad Sci 1961; 172:299–426. 2 Mueller UW, Hawes CS, Wright AE, et al. Isolation of fetal trophoblast cells from peripheral blood of pregnant women. Lancet 1990; 336:197–200. 3 Genest DR, Berkowitz RS, Fisher RA, et al. Gestational trophoblastic disease. In: Tavassoli FA, Devilee P, eds. World Health Organization Classification of Tumours. Pathology and Genetics. Tumours of the Breast and Female Genital Organs. Lyon: IARC Press; 2003:250–256 4 Kajii T, Ohama K. Androgenetic origin of hydatidiform mole. Nature 1977; 268:633–4. 5 Vassilakos P, Riotton G, Kajii T Hydatidiform mole: two entities. A morphologic and cytogenic study with some clinical considerations. Am J Obstet Gynaecol 1977; 127:167–70. 6 Lawler SD, Pickthall VG, Fisher RA, Povey S, Wyn Evans M, Szulman AE. Genetic studies of complete and partial hydatidiform mole (letter). Lancet 1979; ii:580. 7 Lawler SD, Povey S, Fisher RA, Pickthall VG. Genetic studies on hydatidiform moles II: the origin of complete moles. Ann Hum Genet 1982; 46:209–22. 8 Jacobs PA, Wilson CM, Sprenkle JA, Rosenshein NB, Migeon BR. Mechanism of origin of complete hydatidiform mole. Nature 1980; 286:714–16. 9 Fisher RA, Sheppard DM, Lawler DW. Twin pregnancy with complete hydatidiform mole (46,XX) and fetus (46,XY): genetic origin proved by analysis of chromosome polymorphisms. BMJ 1982; 1:1218–20. 10 Davis JR, Surwit EA, Garay JP, Fortier KJ. Sex assignment in gestational trophoblastic neoplasia. Am J Obstet Gynecol 1984; 148:722–5. 11 Ohama K, Kajii T, Okamoto E, et al. Dispermic origin of XY hydatidiform mole. Nature 1981; 292:551–2. 12 Fisher RA, Povey S, Jeffreys AJ, Martin CA, Patel I, Lawler SD. Frequency of heterozygous complete hydatidiform moles, estimated by locus-specific minisatellite and Y chromosome-specific probes. Human Genet 1989; 82:259–63.
13 Edwards YH, Jeremiah SJ, McMillan SL, Povey S, Fisher RA, Lawler SD. Complete hydatidiform moles combine maternal mitochondria with paternal nuclear genome. Ann Hum Genet 1984; 48:119–27. 14 Fisher RA, Newlands ES. Gestational trophoblastic disease: molecular and genetic studies. J Reprod Med 1998; 43:81–97. 15 Fisher RA, Hodges MD, Newlands ES. Familial recurrent hydatidiform mole: a review. J Reprod Med 2004; 49(8):595–601. 16 Paradinas F. The histological diagnosis of hydatidiform moles. Curr Diagn Pathol 1994; 1:24–31. 17 Paradinas FJ, Browne P, Fisher RA, Foskett M, Bagshawe KD, Newlands E. A clinical, histopathological and flow cytometric study of 149 complete moles, 146 partial moles and 107 non-molar hydropic abortions. Histopathology 1996; 28:101–10. 18 Paradinas FJ, Fisher RA, Browne P, Newlands ES. Diploid hydatidiform moles with fetal red blood cells in molar villi: I. Pathology, incidence and prognosis. J Pathol 1997; 181:183–8. 19 Paradinas FJ. The diagnosis and prognosis of molar pregnancy. The experience of the National Referral Centre in London. Int J Gynaecol Obstet 1998; 60:S57–64. 20 Sebire NJ, Rees HC, Peston D, Seckl MJ, Newlands ES, Fisher RA. p57(KIP2) immunohistochemical staining of gestational trophoblastic tumours does not identify the type of the causative pregnancy. Histopathology 2004; 45(2):135–41. 21 Jacobs PA, Hunt PA, Matsuuro JS, Wilson CC. Complete and partial hydatidiform mole in Hawaii: cytogenetics, morphology and epidemiology. Br J Obstet Gynaecol 1982; 89:258–66. 22 Lawler SD, Fisher RA, Pickthall VG, Povey S, Wyn Evans M. Genetic studies on hydatidiform moles I: the origin of partial moles. Cancer Genet Cytogenet 1982; 4:309–20. 23 Fisher RA, Lawler SD, Ormerod MG, Imrie PR, Povey S. Flow cytometry used to distinguish between complete and partial hydatidiform moles. Placenta 1987; 8:249–56. 24 Fukunaga M, Katabuchi H, Nagasaka H, et al. Interobserver and intraobserver variability in the diagnosis of hydatidiform mole. Am J Surg Pathol 2005; 29:942–7. 25 Genest DR, Ruiz RE, Weremowicz S, et al. Do nontriploid partial hydatidiform moles exist? A histologic and flow cytometric reevaluation of nontriploid specimens. J Reprod Med 2002; 47:363–8. 26 Szulman A, Surti U. The syndromes of hydatidiform mole. I. Cytogenetic and morphological correlations. Am J Obstet Gynecol 1978; 13:665–71. 27 Szulman, A, Surti, U. The syndromes of hydatidiform mole. II. Morphological evidence of the complete and partial mole. Am J Obstet Gynecol 1978; 132:20–7. 28 Seckl MJ, Fisher RA, Salerno G, et al. Choriocarcinoma and partial hydatidiform moles. N Engl J Med 2000; 356(9223):36–9. 29 Arima T, Imamura T, Amada S, Tsuneyoshi M, Wake N. Genetic origin of malignant trophoblastic neoplasms. Cancer Genet Cytogenet 1994; 73:95–102.
830 Gestational trophoblastic tumours
30 Wake N, Tanaka K-I, Chapman V, Matsui S, Sandberg AA. Chromosomes and cellular origin of choriocarcinoma. Cancer Res 1981; 41:3137–43. 31 Chaganti RSK, Kodura PRK, Chakraborty R, et al. Genetic origin of trophoblastic choriocarcinoma. Cancer Res 1990; 50:6330–3. 32 Osada H, Kawata M, Yamada M, Okumura K, Takamizawa H. Genetic identification of pregnancies responsible for choriocarcinomas after multiple pregnancies by restriction fragment length polymorphism analysis. Am J Obstet Gynecol 1991; 165:682–8. 33 Fisher RA, Newlands ES, Jeffreys AJ, et al. Gestational and non-gestational trophoblastic tumours distinguished by DNA analysis. Cancer 1992; 69:839–45. 34 Fisher RA, Lawler SD, Povey S, Bagshawe KD. Genetically homozygous choriocarcinoma following pregnancy with hydatidiform mole. Br J Cancer 1988; 58:788–892. 35 Fisher RA, Soteriou BA, Meredith L, Paradinas FJ, Newlands ES. Previous hydatidiform mole identified as the causative pregnancy of choriocarcinoma following birth of normal twins. Int J Cancer 1995; 5:64–70. 36 Shih IM, Kurman RJ. The pathology of intermediate trophoblastic tumors and tumor-like lesions. Int J Gynecol Pathol 2001; 20:31–47. 37 Palmieri C, Fisher RA, Sebire NJ, et al. Placental site trophoblastic tumour arising from a partial hydatidiform mole. Lancet 2005; 366:688. 38 Lathrop J, Lauchlan S, Nayak R, Ambler M. Clinical characteristics of placental site trophoblastic tumour (PSTT). Gynaecol Oncol 1988; 31:32–42. 39 Fisher RA, Paradinas FJ, Newlands ES, Boxer GM. Genetic evidence that placental site trophoblastic tumours can originate from a hydatidiform mole or a normal conceptus. Br J Cancer 1992; 65:355–8. 40 Kotylo PK, Michael H, Davis TE, Sutton GP, Mark PR, Roth LM. Flow cytometric DNA analysis of placental-site trophoblastic tumours. Int J Gynaecol Pathol 1992; 11:245–52. 41 Fukunaga M, Ushigome S. Malignant trophoblastic tumours: immunohistochemical and flow cytometric comparison of choriocarcinoma and placental site trophoblastic tumours. Human Pathol 1993; 24:1098–106. 42 Newlands ES, Bower M, Fisher RA, Paradinas FJ. Management of placental site trophoblastic tumours. J Reprod Med 1998; 43:53–9. 43 Shih IM, Kurman RJ. Ki-67 labelling index in the differential diagnosis of exaggerated placental site, placental site trophoblastic tumour, and choriocarcinoma: a double staining technique using Ki-67 and Mel-CAM antibodies. Hum Pathol 1998; 29:27–33. 44 Hui P, Martel M, Parkash V. Gestational trophoblastic diseases. Recent advances in histopathologic diagnosis and related genetic aspects. Adv Anat Pathol 2005; 12:116–25. 45 Matsuura J, Chiu D, Jacobs PA, Szulman AE. Complete hydatidiform mole in Hawaii: an epidemiological study. Genet Epidemiol 1984; 1:271–84.
46 Hando T, Masaguki O, Kurose T. Recent aspects of gestational trophoblastic disease in Japan. Int J Gynaecol Oncol 1998; 60:S71–6. 47 Martin BH, Kim JH. Changing face of gestational trophoblastic disease. Int J Gynaecol Oncol 1998; 60:S111–20. 48 Lawler SD, Fisher RA. Genetic aspects of gestational trophoblastic tumours. In: Ichinoe K (ed.) Trophoblastic Diseases. Tokyo: Igaku-Shoin, 1986, 23–33. 49 Newlands ES, Paradinas FJ, Fisher RA. Recent advances in gestational trophoblastic disease. Hematol Oncol Clin North Am 1998; 13:225–44. 50 Sebire NJ, Foskett M, Fisher RA, Rees H, Seckl MJ. Newlands ES. Risk of partial and complete hydatidiform molar pregnancy in relation to maternal age. BJOG 2002; 109(1):99–102. 51 Bagshawe KD, Dent J, Webb J. Hydatidiform mole in the United Kingdom 1973–1983. Lancet 1986; ii:673. 52 Sebire NJ, Fisher RA, Foskett M, Rees H, Seckl MJ, Newlands ES. Risk of recurrent hydatidiform mole and subsequent pregnancy outcome following complete or partial hydatidiform molar pregnancy. BJOG 2003; 110(1):22–6. 53 WHO. Gestational Trophoblastic Diseases. Technical Report series 692. Geneva: WHO, 1983, 7–81. 54 Papadopoulos AJ, Foskett M, Seckl MJ, McNeish I, Paradinas FJ, Rees H, Newlands ES. Twenty-five years’ clinical experience with placental site trophoblastic tumors. J Reprod Med 2002; 47(6):460–4. 55 Hassadia A, Gillespie A, Tidy J, Everard R, Wells M, Coleman R, Hancock B. Placental site trophoblastic tumour: clinical features and management. Gynecol Oncol 2005: 99(3):603–7. 56 Baergen RN, Rutgers JL, Young RH, Osann K, Scully RE. Placental site trophoblastic tumor: A study of 55 cases and review of the literature emphasizing factors of prognostic significance. Gynecol Oncol 2006; 100(3):511–20. 57 Palmer JR. Advances in the epidemiology of gestational trophoblastic disease. J Reprod Med 1994; 39:155–62. 58 Palmer JR, Driscoll SG, Rosenberg L, et al. Oral contraceptive use and risk of gestational trophoblastic tumors. J Natl Cancer Inst 1999; 91:635–40. 59 Costa HL, Doyle P. Influence of oral contraceptives in the development of post-molar trophoblastic neoplasia – A systematic review. Gynecol Oncol 2006; 100(3):579–85. 60 Li M, Squire JA, Weksberg R. Overgrowth syndromes and genomic imprinting: from mouse to man. Clin Genet 1998; 53:165–70. 61 Hao Y, Crenshaw T, Moulton T, Newcomb E, Tycko B. Tumor suppressor activity of H19 RNA. Nature 1993; 365: 764–7. 62 Matsuoka S, Thompson JS, Edwards MC, et al. Imprinting of the gene encoding a human cyclin-dependent kinase inhibitor, p57kip2, on chromosome 11p15. Proc Natl Acad Sci USA 1996; 93:3026–30. 63 Ogawa O, Eccles MR, Szeto J, et al. Relaxation in insulin-like growth factor II gene imprinting implicated in Wilm’s tumour. Nature 1993; 362:749–51.
References 831
64 Chilosi M, Piazzola E, Lestani M, et al. Differential expression of p57kip2, a maternally imprinted cdk inhibitor, in normal human placenta and gestational trophoblastic disease. Lab Invest 1998; 78(3):269–76. 65 Mutter GL, Stewart CL, Chaponot ML, Pomponio RJ. Oppositely imprinted genes H-19 and insulin-like growth factor 2 are co-expressed in human androgenetic trophoblast. Am J Hum Genet 1993; 53:1096–102. 66 Ariel I, Lustig O, Oyer CE, et al. Relaxation of genomic imprinting in trophoblastic disease. Gynaecol Oncol 1994; 53:212–19. 67 Walsh C, Miller SJ, Flam F, Fisher RA, Ohlsson R. Paternally derived H19 is differentially expressed in malignant and non-malignant trophoblast. Cancer Res 1995; 55:1111–16. 68 Hashimoto K, Azuma C, Koyama M, et al. Loss of imprinting in choriocarcinoma. Nat Genet 1995; 9:109–10. 69 Arima T, Matsuda T, Takagi N, Wake N. Association of IGF2 and H19 imprinting with choriocarcinoma development. Cancer Genet Cytogenet 1997; 93:39–47. 70 Moglabey YB, Kircheisen R, Seoud M, El Mogharbel N, Van den Veyver I, Slim R. Genetic mapping of a maternal locus responsible for familial hydatidiform moles. Hum Mol Genet 1999; 8:667–71. 71 Kim J, Ashworth L, Branscomb E, Stubbs L. The human homolog of a mouse-imprinted gene, Peg3, maps to a zinc finger gene-rich region of human chromosome 19q13.4. Genome Res 1997; 7:532–40. 72 Hodges MD, Rees HC, Seckl MJ, Newlands ES, Fisher RA. Genetic refinement and physical mapping of a biparental complete hydatidiform mole locus on chromosome 19q13.4. J Med Genet 2003; 40(8):e95. 73 Murdoch S, Djuric U, Mazhar B, et al. Mutations in NALP7 cause recurrent hydatidiform moles and reproductive wastage in humans. Nat Genet 2006; 38(3):300–2. 74 Zhao J, Moss J, Sebire NJ, Cui QC, Seckl MJ, Xiang Y, Fisher RA. Analysis of the chromosomal region 19q13.4 in two Chinese families with recurrent hydatidiform mole. Hum Reprod 2006; 21(2):536–41. 75 Bagshawe KD. Risk and prognostic factors in trophoblastic neoplasia. Cancer 1976: 38:1373–85. 76 Lawler SD, Klouda PA, Bagshawe KD. The relationship between HLA antibodies and the causal pregnancy in choriocarcinoma. Br J Obstet Gynaecol 1976; 83:651–5. 77 Yamashita K, Ishikawa M, Shimizu T, Kuroda M. HLA antigens in husband–wife pairs with trophoblastic tumour. Gynecol Oncol 1981; 12:68–74. 78 Matsuda T, Sasaki M, Kato H, et al. Human chromosome 7 carries a putative tumor suppressor gene(s) involved in choriocarcinoma. Oncogene 1997; 15:2773–81. 79 Rozengurt E. Polypeptide and neuropeptide growth factors: signalling pathways and role in cancer. In: Peckham M, Pinedo HM, Veronesi U (eds) Oxford Textbook of Oncology. Oxford: Oxford Medical Publications, 1995, 12–20. 80 Seckl MJ, Rozengurt E. Neuropeptides, signal transduction and small cell lung cancer. In: Martinet Y, Hirsch FR, Martinet N, Vignaud J-M, Mulshine JL (eds) Clinical and Biological
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
Basis of Lung Cancer Prevention. Basel: Birkha¨user Verlag, 1998, 129–42. Cole LA, Butler SA, Khanlian SA, Giddings A, Muller CY, Seckl MJ, Kohorn EI. Gestational trophoblastic diseases: 2. Hyperglycosylated hCG as a reliable marker of active neoplasia. Gynecol Oncol 2006; 10(2):151–9 Uzumaki H, Okabe T, Sasaki N, et al. Identification and characterization of receptors for granulocyte colonystimulating factor on human placenta and trophoblastic cells. Proc Natl Acad Sci USA 1989; 86:9323–6. Wolf HK, Zarnegar R, Oliver L, Michalopoulos GK. Hepatocyte growth factor in human placenta and trophoblastic disease. Am J Pathol 1991; 138:1035–43. Holmgren L, Flam F, Larsson E, Ohlsson R. Successive activation of the platelet-derived growth factor beta receptor and platelet-derived growth factor B genes correlates with the genesis of human choriocarcinoma. Cancer Res 1993; 53:2927–31. Fulop V, Mok SC, Genest DR, Szigetvari I, Cseh I, Berkowitz RS. c-myc, c-erbB2, c-fms and bcl-2 oncoproteins. Expression in normal placenta, partial and complete mole, and choriocarcinoma. J Reprod Med 1998; 43: 101–10. Fulop V, Mok SC, Genest DR, Gati I, Doszpod J, Berkowitz RS. p53, p21, Rb and mdm2 oncoproteins. Expression in normal placenta, partial and complete mole, and choriocarcinoma. J Reprod Med 1998; 43:119–27. Bagshawe KD, Lawler SD, Paradinas FJ, Dent J, Brown P, Boxer GM. Gestational trophoblastic tumours following initial diagnosis of partial hydatidiform mole. Lancet 1990; 334:1074–6. Rice LW, Berkowitz RS, Lage JM, Goldstein DP, Bernstein MR. Persistent gestational trophoblastic tumour after partial hydatidiform mole. Gynaecol Oncol 1990; 36:358–62. Goto S, Yamada A, Ishizuka T, Tomoda Y. Development of postmolar trophoblastic disease after partial molar pregnancy. Gynaecol Oncol 1993; 48:165–70. Lapthorn AJ, Harris DC, Littlejohn A, et al. Crystal structure of human chorionic gonadotropin. Nature 1994; 369:455–61. Tegoni M, Spinelli S, Verhoveyen M, Davis P, Cambillau C. Crystal structure of a ternary complex between human chorionic gonadotropin (hCG) and two Fv fragments specific for the alpha and beta-subunits. J Mol Biol 1999; 289:1375–85. Boorstein WR, Vamkakopoulos NC, Fiddes JC. Human chorionic gonadotropin beta subunit is encoded by at least eight genes arranged in tandem and inverted pairs. Nature 1982; 300:419–22. Policastro PF, Daniels-McQueen S, Carle G, Boime I. A map of the hCG beta-LH beta gene cluster. J Biol Chem 1986; 261:5907–16. Bo M, Boime I. Identification of the transcriptionally active genes of the chorionic gonadotropin beta gene cluster in vivo. J Biol Chem 1992; 267:3179–84. Grover S, Woodward SR, Odell WD. Complete sequence of the gene encoding a chorionic gonadotrophin-like
832 Gestational trophoblastic tumours
96
97 98
99
100 101
102
103
104
105
106
107
108 109 110
111
112
113
114
protein from Xanthomonas maltophilia. Gene 1995; 156:75–8. Tsai-Morris CH, Buczko E, Wang W, Xie XZ, Dufau ML. Structural organization of the rat lutenizing hormone (LH) receptor gene. J Biol Chem 1991; 266:11355–9. Dufau ML. The luteinizing hormone receptor. Annu Rev Physiol 1998; 60:461–96. Cole LA, Sutton JM. Selecting an appropriate hCG test for managing gestational trophoblastic disease and cancer. J Reprod Med 2004; 49(7):545–53. Vaitukaitis JL. Human chorionic gonadotrophin – a hormone secreted for many reasons. N Engl J Med 1979; 301: 324–6. Berkowitz RS, Goldstein DP. Pathogenesis of gestational trophoblastic neoplasms. Pathol Annu 1981; 11:391. Fradken JE, Eastman RC, Lesniak MA, Roth J. Specificity spillover at the hormone receptor – exploring its role in human disease. N Engl J Med 1989; 320:640–5. Goldstein DP, Berkowitz RS. Current management of complete and partial molar pregnancy. J Reprod Med 1994; 39:139–46. Fine C, Bundy AL, Berkowitz RS, Boswell SB, Berezin AF, Doubilet PM. Sonographic diagnosis of partial hydatidiform mole. Obstet Gynaecol 1989; 73:414–8. Seckl MJ, Gillmore R, Foskett M, Sebire NJ, Rees H, Newlands ES. Routine terminations of pregnancy – should we screen for gestational trophoblastic neoplasia? Lancet 2004; 21–27;364(9435):705–7. Sebire NJ, Foskett M, Paradinas FJ, et al. Outcome of twin pregnancies with complete hydatidiform mole and healthy co-twin. Lancet 2002; 359(9324):2165–5. Seckl MJ, Rustin GJS, Newlands ES, Gwyther SJ, Bomanji J. Pulmonary embolism, pulmonary hypertension, and choriocarcinoma. Lancet 1991; 338:1313–5. Tidy JH, Rustin GJS, Newlands ES, et al. Presentation and management of choriocarcinoma after nonmolar pregnancy. Br J Obstet Gynaecol 1995; 102:715–9. McGrath IT, Golding PR, Bagshawe KD. Medical presentations of choriocarcinoma. BMJ 1971; 2:633–7. Savage P, Roddie M, Seckl MJ. A 28-year-old woman with a pulmonary embolus. Lancet 1998; 352:30. Blohm ME, Gobel U. Unexplained anaemia and failure to thrive as initial symptoms of infantile choriocarcinoma: a review. Eur J Pediatr 2004; 163(1):1–6. Sebire NJ, Lindsay I, Fisher RA, Seckl MJ. Intraplacental choriocarcinoma: experience form a tertiary referral center and relationship with infantile choriocarcinoma. Fetal Pediatr Pathol 2005; 24(1):21–9. Johnson EJ, Crofton PM, O’Neill JM, et al. Infantile choriocarcinoma treated with chemotherapy alone. Med Pediatr Oncol 2003; 41(6):550–7. Finkler NJ, Berkowitz RS, Driscoll SG, Goldstein DP, Bernstein MR. Clinical experience with placental site trophoblastic tumours at the New England Trophoblastic Disease Center. Obstet Gynaecol 1988; 71:854–7. Eckstein R, Paradinas F, Bagshawe KD. Placental site trophoblastic tumour (trophoblastic pseudotumour): a
115 116
117
118
119
120 121
122
123
124
125
126
127
128
study of four cases requiring hysterectomy including one fatal case. Histopathology 1982; 6:211–26. Young R, Scully R. Placental site trophoblastic tumour: current status. Clin Obstet Gynaecol 1984; 27:248–58. Samlowski WE, Abbott TM, Kepas DE, Eyre HJ. Placental-site trophoblastic tumor (trophoblastic pseudotumor): case report demonstrating failure of chemotherapy, surgery, and radiotherapy to control metastatic disease. Gynaecol Oncol 1985; 21:111–7. Hopkins MP, Drescher CW, McQuillan A, Keyser J, Schmidt R. Malignant placental site trophoblastic tumour associated with placental abruption, fetal distress, and elevated CA125. Gynecol Oncol 1992; 47:267–71. Coulson LE, Kong CS, Zaloudek C. Epithelioid trophoblastic tumor of the uterus in a postmenopausal woman. Am J Surg Pathol 2000; 24:1558–62. Kuo KT, Chen MJ, Lin MC. Epithelioid trophoblastic tumor of the broad ligament: A case report and review of the literature. Am J Surg Pathol 2004; 28:405–9. Bagshawe KD, Noble MIM. Cardiorespiratory effects of trophoblastic tumours. QJM 1965; 137:39–54. Fowler DJ, Lindsay I, Seckl MJ, Sebire NJ. Routine pre-evacuation ultrasound diagnosis of hydatidiform mole: experience of more than 1000 cases from a regional referral center. Ultrasound Obstet Gynecol 2006; 27(1):56–60. Boultebee JE, Newlands ES. New diagnostic and therapeutic approaches to gestational trophoblastic tumours. In: Bourne TH, Jauniaux E, Jurkovic D (eds) Transvaginal Colour Doppler. The Scientific Basis and Practical Application of Colour Doppler in Gynaecology. Berlin: Springer, 1995, 57–65. Long MG, Boultbee JE, Langley R, Newlands ES, Begent RHJ, Bagshawe KD. Doppler assessment of the uterine circulation and the clinical behaviour of gestational trophoblastic tumours requiring chemotherapy. Br J Cancer 1992; 66:882–7. Agarwal R, Strickland S, McNeish IA, et al. Doppler ultrasonography of the uterine artery and the response to chemotherapy in patients with gestational trophoblastic tumors. Clin Cancer Res 2002; 8(5):1142–7. Lim AK, Agarwal R, Seckl MJ, Newlands ES, Barrett NK, Mitchell AW. Embolization of bleeding residual uterine vascular malformations in patients with treated gestational trophoblastic tumors. Radiology 2002; 222(3):640–4. Begent RHJ, Bagshawe KD, Green AJ, Searle AJ. The clinical value of imaging with antibody to human chorionic gonadotrophin in the detection of residual choriocarcinoma. Br J Cancer 1987; 55:657–60. Hebart H, Erley C, Kaskas B, et al. Positron emission tomography helps to diagnose tumor emboli and residual disease in choriocarcinoma. Ann Oncol 1996; 7:416–8. Shaw SW, Wang CW, Ma SY, Ng KK, Chang TC. Exclusion of lung metastases in placental site trophoblastic tumor using [18F]fluorodeoxyglucose positron emission tomography: a case report. Gynecol Oncol 2005; 99(1):239–42.
References 833
129 Numnum TM, Leath CA, Straughn JM, Conner MG, Barnes MN. Occult choriocarcinoma discovered by positron emission tomography/computed tomography imaging following a successful pregnancy. Gynecol Oncol 2005; 97(2): 713–5. 130 Savage P, Seckl MJ. The role of repeat uterine evacuation in trophoblast disease. Gynecol Oncol 2005; 99(1):251–2. 131 Curry SL, Hammond CB, Tyrey L, Creasman WT, Parker RT. Hydatidiform mole: diagnosis, management, and long-term follow up of 347 patients. Obst Gynaecol 1975; 45(1):1–8. 132 Stone M, Dent J, Kardana A, Bagshawe KD. Relationship of oral contraceptive to development of trophoblastic tumour after evacuation of hydatidiform mole. Br J Obstet Gynaecol 1976; 86:913–6. 133 Bower M, Newlands ES, Holden L, et al. EMA/CO for high-risk gestational trophoblastic tumours: results from a cohort of 272 patients. J Clin Oncol 1997; 15:2636–43. 134 Smith DB, Holden L, Newlands ES, Bagshawe KD. Correlation between clinical staging (FIGO) and prognostic groups in gestational trophoblastic disease. Br J Obstet Gynaecol 1992; 100:157–60. 135 Goldstein DP, Zanten-Przybysz IV, Bernstein MR, Berkowitz RS. Revised FIGO staging system for gestational trophoblastic tumors. J Reprod Med 1998; 43:37–43. 136 Kohorn EI. Negotiating a staging and risk factor scoring system for gestational trophoblastic neoplasia. A progress report. J Reprod Med 2002; 47(6):445–50. 137 Rustin GJS, Newlands ES, Lutz J-M, et al. Combination but not single agent methotrexate chemotherapy for gestational trophoblastic tumours (GTT) increases the incidence of second tumours. J Clin Oncol 1996; 14:2769–73. 138 McNeish IA, Strickland S, Holden L, Rustin GJ, Foskett M, Seckl MJ, Newlands ES. Low-risk persistent gestational trophoblastic disease: outcome after initial treatment with low-dose methotrexate and folinic acid from 1992 to 2000. J Clin Oncol 2002; 20(7):1838–44. 139 Newlands ES, Bagshawe KD, Begent RHJ, Rustin GJS, Holden L, Dent J. Development of chemotherapy for medium- and high-risk patients with gestational trophoblastic tumours (1979–1984). Br J Obstet Gynaecol 1986; 93:63–9. 140 Bagshawe KD, Dent J, Newlands ES, Begent RHJ, Rustin GJS. The role of low dose methotrexate and folinic acid in gestational trophoblastic tumours (GTT). Br J Obstet Gynaecol 1989; 96:795–802. 141 Motzer RJ, Bajorin DF, Schwartz LH, et al. Phase II trial of paclitaxel shows antitumour activity in patients with previously treated germ cell tumours. J Clin Oncol 1994; 12:2277–83. 142 Jones WB, Schneider J, Shapiro F, Lewis JLJ. Treatment of resistant gestational choriocarcinoma with taxol: a report of two cases. Gynaecol Oncol 1996; 61:126–30. 143 Termrungruanglert W, Kudelka AP, Piamsomboon S, et al. Remission of refractory gestational trophoblastic disease
144
145
146
147
148
149
150
151 152
153
154
155
156
157
with high-dose paclitaxel. Anticancer Drugs 1996; 7:503–6. Shorbagi A, Aksoy S, Kilickap S, Guler N. Successful salvage therapy of resistant gestational trophoblastic disease with ifosfamide and paclitaxel. Gynecol Oncol 2005; 97(2):722–3. Osborne R, Covens A, Mirchandani D, Gerulath A. Successful salvage of relapsed high-risk gestational trophoblastic neoplasia patients using a novel paclitaxel-containing doublet. J Reprod Med 2004; 49(8):655–61. Beyer J, Kramar A, Mandanas R, et al. High-dose chemotherapy as salvage treatment in germ cell tumors: a multivariate analysis of prognostic variables. J Clin Oncol 1996; 14:2638–45. Lyttelton MP, Newlands ES, Giles C, et al. High-dose therapy including carboplatin adjusted for renal function in patients with relapsed germ cell tumor: outcome and prognostic factors. Br J Cancer 1998; 77:1672–6. El-Helw LM, Seckl MJ, Haynes R, et al. High-dose chemotherapy and peripheral blood stem cell support in refractory gestational trophoblastic neoplasia. Br J Cancer. 2005; 93(6):620–1. Kelly MP, Rustin GJS, Ivory C, Phillips P, Bagshawe KD. Respiratory failure due to choriocarcinoma: a study of 103 dyspneic patients. Gynaecol Oncol 1990; 38: 149–54. Athanassiou A, Begent RHJ, Newlands ES, Parker D, Rustin GJS, Bagshawe KD. Central nervous system metastases of choriocarcinoma: 23 years’ experience at Charing Cross Hospital. Cancer 1983; 52:1728–35. Ishizuka T. Intracranial metastases of choriocarcinoma: a clinicopathologic study. Cancer 1983; 52:1896–903. Song HZ, Wu PC. Treatment of brain metastases in choriocarcinoma and invasive mole. In: Song HZ, Wu PC (eds) Studies in Trophoblastic Disease in China. Oxford: Pergamon, 1988, 231–7. Rustin GJS, Newlands ES, Begent RHJ, Dent J, Bagshawe KD. Weekly alternating chemotherapy (EMA/CO) for treatment of central nervous systems of choriocarcinoma. J Clin Oncol 1989; 7:900–3. Newlands ES, Holden L, Seckl MJ, McNeish I, Strickland S, Rustin GJ. Management of brain metastases in patients with high-risk gestational trophoblastic tumors. J Reprod Med 2002; 47(6):465–71. Blagden SP, Foskett MA, Fisher RA, Short D, Fuller S, Newlands ES, Seckl MJ. The effect of early pregnancy following chemotherapy on disease relapse and foetal outcome in women treated for gestational trophoblastic tumours. Br J Cancer 2002; 86(1):26–30. Fan X, Yan L, Jia S, Ma A, Qiao C. A study of early pregnancy factor activity in the sera of women with trophoblastic tumor. Am J Reprod Immunol 1999; 41:204–8. Seckl MJ, Newlands ES. Treatment of gestational trophoblastic disease. Gen Diagn Pathol 1997; 143 (2-3):159–71.
834 Gestational trophoblastic tumours
158 Pui CH, Ribeiro RC, Hancock ML, et al. Acute myeloid leukaemia in children treated with epipodophyllotoxins for acute lymphoblastic lymphoma. N Engl J Med 1991; 325:1682–7. 159 Whitlock JA, Greer JP, Lukens JN. Epipodophyllotoxin-related leukaemia. Identification of a new subset of secondary leukaemia. Cancer 1991; 68:600–4. 160 Hawkins MM, Kinnier Wilson LM, Stovall MA, et al. Epipodophyllotoxins, alkylating agents, and radiation and risk of secondary leukaemia after childhood cancer. BMJ 1992; 304:951–8.
161 Bower M, Rustin GJS, Newlands ES, et al. Chemotherapy for gestational trophoblastic tumours hastens menopause by 3 years. Eur J Cancer 1998; 34:1204–7. 162 Woolas RP, Bower M, Newlands ES, Seckl MJ, Short D, Holden L. Influence of chemotherapy for gestational trophoblastic disease on subsequent pregnancy outcome. Br J Obstet Gynaecol 1998; 105(9):1032–5. 163 Rustin GJS, Booth M, Dent J, Salt S, Rustin F, Bagshawe KD. Pregnancy after cytotoxic chemotherapy for gestational trophoblastic tumours. BMJ 1984; 288:103–6.
37 Non-melanoma skin cancer STEPHEN L. MORRIS, SEAN WHITTAKER AND MARGARET SPITTLE
Introduction Aetiology Familial skin cancer syndromes Naevoid basal cell carcinoma syndrome (Gorlin syndrome) Xeroderma pigmentosum Pre-malignant conditions Actinic keratosis and Bowen’s disease Basal cell carcinoma Squamous cell carcinoma
835 836 836 836 838 839 839 840 850
INTRODUCTION Skin cancer is the commonest form of cancer in the UK. Although each cell type in the skin can give rise to a different type of cancer, it is convenient to classify skin cancer broadly into non-melanoma skin cancers (NMSC) and malignant melanoma (MM). Secondary deposits from other cancers can also present in the skin. The total number of skin cancers has doubled in the last 10 years. Official figures show there were 7300 cases of MM and 62 700 cases of NMSC registered in 2001.1 This accounts for 20 per cent of all cancer registrations and is likely to be an underestimate as registration of NMSC in the UK is not standardized or complete. Although common, the vast majority of NMSC is not life threatening and it is commonly treated in primary care.1 Malignant melanoma is not as common but is often fatal. Of 2280 people in the UK who died from skin cancer in 2003, 1770 died from malignant melanoma. Non-malignant skin cancer is most common in the elderly, with 70 per cent of registered cases in people over 65 years and almost half of these will develop another cancer within 5 years. This chapter will discuss the common NMSC basal cell carcinoma (BCC), squamous cell carcinoma (SCC) and also cover
Primary cutaneous lymphoma Primary cutaneous T-cell lymphoma Primary cutaneous B-cell lymphoma Other non-melanoma skin cancers Extra-mammary Paget’s disease Merkel cell carcinoma Langerhans cell histiocytosis Key references References
853 853 864 865 865 866 867 870 870
some of the rarer tumours including primary cutaneous lymphoma, Merkel cell carcinoma, extra-mammary Paget’s disease (EMPD) and Langerhans cell histiocytosis (LCH). Mela-noma is covered in another chapter. There are many other skin tumours including superficial soft tissue tumours that can predominantly involve the skin, such as dermatofibrosarcoma protuberans, vascular tumours such as Kaposi’s sarcoma and angiosarcoma, smooth and skeletal muscle tumours such as leiomyosarcoma and neural tumours, which are covered in other chapters.2 There is also a wide range of rare adnexal malignant skin tumours, which often have benign counterparts. These malignant tumours may derive from different adnexal structures within normal skin such as sebaceous, eccrine, apocrine and follicular epithelium. They may clinically simulate common forms of NMSC such as BCC and SCC but are distinguished by distinct and classical histological and immunophenotypical features. Recognition of these different malignant adnexal tumours is critical, as prognosis and treatment may be different to common forms of NMSC. In addition, the role of sentinel lymph node biopsy for staging the more aggressive malignant adnexal tumours is unclear but currently under investigation. These tumours are referred to in the discussion of differential diagnosis where appropriate.
836 Non-melanoma skin cancer
AETIOLOGY In common with other cancers in the body, carcinogenesis of the skin is thought to be a multi-step process. It is generally agreed that a tumour results from the progeny of a single cell having acquired one or more somatic mutations.3 The likelihood of acquiring these mutant clones depends on a complex interplay between the genetic susceptibility of the individual and a wide range of environmental factors on the one hand and normal function of an intact immune system to rectify them on the other. In the hereditary skin cancers discussed below, individuals are predestined to develop skin cancer because of there genetic constitutions, whereas the greatly increased incidence of skin cancers in the older population is seen as a culmination of oncogenic events and a diminished ability of the body to destroy the transformed cells with age. The most potent environmental agent capable of inducing skin cancer is ultra-violet light (UV) exposure. The incidence of NMSC increases with decline in latitude being highest in Australia with an annual incidence rate per million population of 1372 for men and 702 for women.4,5 Ozone depletion in the atmosphere allowing more harmful radiation to reach human skin is probably partly responsible for the alarming increase in incidence. This is compounded by the popularity of sunny holidays abroad, outdoor recreational activities and the culture of the bronzed body beautiful. Soldiers during the World War II who had high UV light exposure while serving in North Africa have a very high incidence of NMSC and are entitled to compensation from the Ministry of Defence (MOD), which many have successfully claimed. Melanin affords some protection against UV skin damage and explains the high incidence of skin tumours in Caucasian populations, particularly those with Celtic and Caledonian ancestry (skin type I), and the low incidence in the more pigmented ethnic groups (skin types III–IV). Most SCCs occur in sun-exposed areas of the skin and where the skin is more sensitive to UV damage (e.g., albinism) or where the skin is unable to repair UV-induced damage to DNA such as xeroderma pigmentosum, described below. Most actinic keratosis and SCCs contain UV signature mutations of the p53 tumour suppressor gene. P53 controls the G1 cell cycle checkpoint allowing either time for cellular DNA repair or promotion of apoptosis, and, therefore, it contributes to the malignant phenotype if mutated.6 Other environmental factors known to induce skin cancer include arsenic, ionizing radiation, human papilloma virus (HPV) and polycyclic aromatic hydrocarbons. There are still patients alive presenting with NMSC on the head and neck who had low-dose radiation to the scalp for ringworm as a child, and multiple BCCs on the trunk are seen in patients after radiotherapy for ankylosing spondylitis. Chronic wound healing can be associated with the later development of NMSC especially SCC (Marjolin’s ulcer)
as seen in patients with severe dystrophic epidermolysis bullosa in the third and fourth decades. Patients on immunosuppressive therapy regimens for kidney and heart transplants have an increased risk of developing SCC, BCC, MM and Kaposi sarcoma. The risk of post-transplant skin SCC is related to the degree of immunosuppression.7 Such tumours in immunocompromised patients are invariably more aggressive with greater local invasion and earlier metastases than in immunocompetent patients. Human papilloma virus has been advocated as a possible aetiological factor in these patients.8,9 Human papilloma viruses (especially types 16 and 18) have been identified in SCC cells by in-situ hybridization and polymerase chain reaction (PCR) techniques, commonly in lesions affecting the genital, oral and peri-ungual areas. However, no role for HPV has been found in those patients infected with the human immunodeficiency virus (HIV).10 A naturally occurring model for the association of HPV and SCC is seen in epidermodysplasia verruciformis. In this autosomal recessive (AR) condition patients develop widespread and extensive warty lesions identical to those of the common wart, and ultimately develop intra-epithelial neoplasia and SCC following UV exposure. Human papilloma viruses has been identified in these lesions, with types 5 and 8 most commonly associated with malignancy; occasionally types 14, 17, 20 and 47 are implicated.11–13 A study of 252 cases of SCC and 525 of BCC found HPV antibodies more frequently in SCC patients than controls, but no difference in patients with BCC. Seropositivity to HPV types in genus beta, particularly HPV5 was associated with SCC risk. Individuals with SCCs on chronically sun-exposed areas were more likely to be seropositive for beta-HPV than individuals with SCCs at other sites.14
FAMILIAL SKIN CANCER SYNDROMES NAEVOID BASAL CELL CARCINOMA SYNDROME (GORLIN SYNDROME) Naevoid basal cell carcinoma syndrome (NBCCS) or Gorlin syndrome is a rare autosomal dominant disorder with linkage to chromosome 9q22.15 It is characterized by multiple BCCs and developmental defects.16 The sexes are equally affected. Most patients are Caucasian, but cases have been reported in Afro-Caribbeans and Asians.17 The syndrome is caused by mutations in Patched (PTCH), a tumour suppressor gene. A single point mutation in one PTCH allele may be responsible for the malformations found in the syndrome. Inactivation of both PTCH alleles results in the formation of tumours and cysts.17 However, there is poor correlation between identifiable genetic mutations and the resulting clinical phenotype, suggesting phenotypical variability in NBCCS.15 The skin lesions are indistinguishable from BCC but a broader spectrum of histological subtypes is found. Multiple keratinizing odontogenic and epidermal
Naevoid basal cell carcinoma syndrome (Gorlin syndrome) 837
cysts are often seen in the NBCCS and multiple keratinizing cysts within BCCs have been recorded.18
Clinical features There are over 100 recognized features of Gorlin syndrome. The diagnostic criteria are shown in Box 37.1. Skin tumours: Eighty per cent of Caucasian patients have at least one BCC, with the first tumour occurring at a mean age of 23 years. The number of BCCs can range from 1 to more than 1000 (Fig. 37.1). Jaw cysts: These are odontogenic keratocysts (OKCs) and occur in 74 per cent of patients with the first cysts occurring in 80 per cent by the age of 20 years. The number of total jaw cysts can range from 1 to 28. Most sporadically occurring OKCs behave in a benign manner; however, those associated with NBCCS are clinically more aggressive in their behaviour.19 OKCs are often the first signs of NBCCS and can occasionally be detected in patients younger than 10 years of age. It is suggested that early onset of an OKC should prompt investigation for NBCCS.19 Cutaneous cysts may be found with an identical histology to those found in the jaw.20 Palmar and plantar pits: These are seen in 87 per cent of patients, they have vertical sides, are up to several
millimetres in diameter and have an erythematous base. They are said to result from impaired maturation of basal keratinocytes resulting in defective keratin. CNS: Patients with NBCCS have a predisposition to develop medulloblastoma and primitive neuroectodermal tumours of the central nervous system.21 Cardiac tumours: There now appears to be an established association between cardiac tumours and NBCCS and evaluation of cardiac status is recommended in all patients with NBCCS.20 Patients with NBCCS may exhibit coarse facies, cleft lip or palate, relative macrocephaly, hypertelorism, frontal bossing, pectus deformity or Sprengel deformity. Radiologically calcification may be observed in various sites including the falx cerebri and tentorium cerebelli. Other abnormalities include bridged sella, hemivertebrae, fusion of the vertebral bodies and flame-shaped lucencies of the phalanges, metacarpal and carpal bones of the hands.22 There are no routine abnormal laboratory tests apart from, rarely, a raised alkaline phosphatase possibly due to the presence of growing odontogenic keratocysts.23
Box 37.1 Diagnostic criteria for Gorlin’s syndrome Major criteria ● Multiple (2) basal cell carcinomas or one under the age of 30 years or 10 basal cell naevi ● Odontogenic keratocyst (proved on histology) or polyostotic bone cyst ● Palmar or plantar pits (3 or more) ● Ectopic calcification: lamellar or early (20 years) falx calcification ● First-degree relative with Gorlin’s syndrome Minor criteria Congenital skeletal abnormality: bifid, fused, splayed or missing rib or fused vertebrae ● Occipitofrontal circumference (OFC) 97th percentile ● Cardiac or ovarian fibroma ● Medulloblastoma ● Lymphomesenteric cysts ● Congenital malformation: cleft lip/palate, polydactyly, eye anomaly (cataract, coloboma, microphthalmia) ●
A diagnosis can be made when two major or one major and two minor criteria are present.18
Figure 37.1 Naevoid BCC syndrome. Multiple basal cell carcinomas (BCCs) on the back of a 45-year-old woman.
838 Non-melanoma skin cancer
Treatment The first aim of treatment is to educate patients and parents as to the importance of avoiding known carcinogenic factors with sun avoidance being of paramount importance. The BCCs may be more aggressive than their sporadic counterparts and are best treated early by one of the many options for BCCs discussed below. Clinical experience suggests that a proportion of patients with Gorlin syndrome are extremely sensitive to ionizing radiation.24 However, there is little in-vitro evidence that the cells of individuals with Gorlin syndrome systematically demonstrate increased sensitivity to ionizing radiation. Reports are conflicting, with some studies suggesting a decrease in cultured fibroblast survival on exposure to ionizing radiation,25,26 whereas other studies have no alteration.27,28 These conflicting results could be explained, in part, by variable radiation sensitivity of different cell types, and by the use of inappropriate methods of detecting what is likely to be a subtle difference in radiation sensitivity between tissues from normal and Gorlin syndrome patients. An alternative explanation suggests that there is genetic heterogeneity whereby not all mutations in the PTCH gene confer the same degree of radiation sensitivity.29,30 Hence, some families will be more susceptible to the effects of ionizing radiation than others. However, until the molecular basis for ionizing radiation sensitivity is determined, it is sensible to limit the exposure of all patients with Gorlin syndrome to ionizing radiation. The long-term effects of radiation can be significant with skin and soft-tissue atrophy, and telangiectasia as well as induction of malignancy. The latter was demonstrated in a case of patients with Gorlin syndrome who developed a renal-cell carcinoma adjacent to a previous spinal-radiation field.24 Surgical excision remains the most appropriate therapy but sun-protection advice is essential and topical therapies such as PDT (photo dynamic therapy) and imiquimod are possible alternatives. Retinoids have a therapeutic effect on existing BCCs anda prophylactic effect in inhibiting new tumour formation. However, to be of any major benefit, retinoids have to be maintained long term,31 and, in view of the need to start such therapy at an early age, the patient would be subject to long-term rheumatological complications such as diffuse idiopathic skeletal hyperostosis (DISH) and pseudo-coxarthritis.32 Success has also been reported with a combin-ation of topical fluorouracil (5-FU) and topical tretinoin, which may prevent the development of new tumours, inhibit the growth of existing tumours and cause the regression of superficially invasive BCCs.33
XERODERMA PIGMENTOSUM Incidence Xeroderma pigmentosum is found world-wide, in all races with an equal sex incidence. The incidence is said to be
higher in Japan and Libya. Interestingly, the spectrum of disease manifestations in different countries is quite varied.34,35 The incidence in Europe and America is 1:250 000; in Japan 1:40 000; and in Libya 15–20 per million.36
Aetiology Xeroderma pigmentosum represents a group of autosomal recessive disorders, and the genetic basis of all xeroderma pigmentosum variants has recently been elucidated. The manifestations of xeroderma pigmentosum occur as a result of a defect in excision repair of UV-induced pyrimidine dimmers.36,37 Nucleotide excision repair (NER) is one of the best characterized DNA repair systems. The consequences of a defect in one of the NER proteins are manifested in three rare, recessive photosensitive syndromes: xeroderma pigmentosum, Cockayne’s syndrome and a photosensitive form of brittle-hair disorder, trichothiodystrophy (TTD). NUCLEOTIDE EXCISION REPAIR
Nucleotide excision repair is the most flexible of all DNA repair mechanisms because of its ability to eliminate a plethora of structurally unrelated DNA lesions. Cells that are NER-deficient are sensitive to insults that cause a significant DNA helical distortion. The clinically most relevant NER substrates are cis-syn cyclobutane dimers (CPDs) and pyrimidine(6-4)pyrimidone photoproducts. Both are formed between adjacent pyrimidines, and they constitute the two major classes of DNA lesions induced by solar UV light. The NER process involves the action of approximately 30 proteins, which enable damage recognition, local opening of DNA double helix around the injury and incision of the damaged strand on either side of the lesion. After excision of the DNA lesion the resulting gap is closed by DNA repair synthesis, followed by strand ligation.38 Using the unscheduled DNA synthesis (UDS) assay, eight complementation groups of xeroderma pigmentosum have been identified: groups A, B, C, D, E, F, G and V.39 The UDS measures DNA synthesis during the G1 and G2 phases of the cell cycle on cultured fibroblasts.
Clinical features The clinical expression of the disease depends on the complementation group, the ethnic origin of the patient and geographic environment. At birth, the skin is normal. However, freckling (lentigines) and xerosis of light-exposed sites begins between 6 months and 3 years of age. This process may extend to non sun-exposed areas. The skin becomes atrophic and telangiectatic and superficial ulcers may develop, which can leave significant scarring (Fig. 37.2). The development of pre-malignant and malignant skin tumours will depend on the complementation group and
Actinic keratosis and Bowen’s disease 839
and neoplastic lesions are treated as described in the sections on BCC and SCC. Radiotherapy is contraindicated in these patients due to the risk of inducing further malignancy.
PRE-MALIGNANT CONDITIONS Pre-malignant epidermal conditions include solar or actinic keratosis and Bowen’s disease which shows microscopic features of carcinoma in situ confined to the epidermis. The stage at which the lesion becomes a malignant tumour is determined by breach of the basement membrane. As well as actinic keratosis and Bowen’s disease, which are discussed below, skin damaged by ionizing radiation, heat, carcinogens, long standing scars, sinuses and ulcers may be the site of a future cancer.
Figure 37.2 Xeroderma pigmentosum in a young patient with severe freckling and pigmentary skin changes.
the level of UV exposure that the patient is subjected to. Often there are numerous actinic keratoses as well as cutaneous horns and kerato-acanthomas. Both SCC and BCCs may appear from an early age.34,35 Melanomas occur in up to 5 per cent of Caucasian xeroderma pigmentosum patients but studies of patients in Libya and South Africa have not shown a greater incidence of malignant melanoma in Libyans and black South Africans.34,35 The commonest eye manifestation is photophobia with telangiectasia of the conjunctiva. There is gross hyperpigmentation of the eyelids and severe ectropion can result from superficial ulceration of the skin around the eye. Scarring of the cornea and pinguecula-like growths may occur. Patients are particularly prone to SCC of the lip and this has a particularly poor prognosis. Eighteen per cent of xeroderma pigmentosum patients display progressive neurological abnormalities. The most severe are seen in group A patients, with mild disease seen in group D patients. This is caused by primary neuronal degeneration and loss. A mild clinical phenotype, XP variant, is recognized which can be missed clinically but is usually associated with early onset of NMSC and marked photodamage.
Treatment The most important single measure is UV avoidance with the patient forbidden to go out during the peak periods of sun exposure and advised to completely cover up in the sun and to always wear a high-factor sun-block. Occasionally patients have been treated with systemic retinoids with variable responses, but tumours tend to recur when the dose of the drug is reduced.40,41 Patients with dysplastic
ACTINIC KERATOSIS AND BOWEN’S DISEASE Actinic keratosis is a UV-induced pre-malignant condition with a very low malignant potential.42,43 The aetiology of Bowen’s disease is unclear as definite evidence of a pathogenic role for UV radiation is lacking but chronic arsenic ingestion is a risk factor. Although Bowen’s disease is an SCC, in-situ progression to invasive SCC is very rare. It is estimated that 3–5 per cent of Bowen’s disease becomes invasive SCC and tends to be aggressive with a metastases rate of 30 per cent.44 Actinic keratosis are small erythematous superficial keratotic lesions on sun-exposed skin. The surrounding skin shows evidence of sun exposure and damage with telangiectasia, discolouration and non-uniform pigmentation. The scale is adherent and attempts to remove it lead to bleeding. Bowen’s disease may occur anywhere on the skin or mucous membranes and is an erythematous scaly plaque with a sharply defined margin (Fig. 37.3). When SCC in situ is found on the skin of the penis it is called erythroplasia of Queyrat. The risk of progression of genital Bowen’s disease is greater than at other sites, in the order of 10 per cent.45 A biopsy or excision is required for Bowen’s disease or actinic keratosis if there is suspicion of an invasive component suggested by induration followed by development of a nodular lesion with a keratotic surface, which may break down to an ulcer with an ill-defined margin. Clinicopathological variants of actinic keratosis include hypertrophic, atrophic, acantholytic, bowenoid and pigmented. They all show atypical keratinocyte proliferation (dyskeratoses) within the epidermis. Initially, dysplastic changes are confined to small foci in the epidermis in which there are aggregates of atypical pleomorphic keratinocytes at the basal layer. There may be hyperkeratosis and parakeratosis overlying the dysplastic keratinocytes in the epidermis.
840 Non-melanoma skin cancer
The atypical keratinocytes show loss of polarity, nuclear pleomorphism, disordered maturation and increased numbers of mitotic figures. There are many treatments for actinic keratosis and Bowen’s disease. Cryosurgery is the most common treatment. Curettage and cautery may be used, particularly where diagnosis is in doubt to obtain histology. Topical treatment with 5-FU is a convenient and cosmetically acceptable treatment, the most common regimen being daily application of cream to the area for 4 weeks. The skin becomes inflamed and the patient will experience pain, pruritus and burning at the site, which sometimes extends to other sub-clinical areas of actinic damage.46 There are successful reports of new topical targeted treatments, such as imiquimod, which induce an innate immune response via one of the Toll-like receptors.47 For extensive Bowen’s disease not treatable by simple topical treatment and too extensive for surgery, radiotherapy is a very successful treatment with similar regimens to those used for SCC described below.48 Photodynamic therapy is a further option for extensive Bowen’s disease particularly on the lower limbs.
BASAL CELL CARCINOMA Basal cell carcinoma (rodent ulcer or basal cell epithelioma) is a malignancy derived from keratinocytes and the stroma of the pilo-sebaceous follicle49–51 with recent evidence suggesting that the tumour might be derived from follicular keratinocytes.
Incidence Basal cell carcinoma is the most common human cancer affecting an estimated 750 000 Americans per year.52 The age-standardized incidence of BCC in South Wales was estimated at 114.2 per 100 000 population in 1998.53 The incidence of BCC has been shown to double every 10 years with the rapid rise attributed to the current fashion of ‘bronze body beautiful’. Widespread use of sun beds and solariums and the popularity of sun-bathing holidays have resulted in more BCCs and a younger population of affected patients.54 Estimates predict that 28 per cent of Caucasians born after 1994 will develop a BCC in their lifetime.52
Aetiology
Figure 37.3 Bowen’s disease on the lower leg of a patient with a history of arsenic ingestion.
There is compelling epidemiological data implicating UV radiation exposure in BCC tumorigenesis. Sixty-six per cent of BCCs occur on the head and neck. The incidence is much greater in those with fair skin and skin type I, the tumour only very rarely occurring in African-Americans.55 The incidence of BCCs in Caucasian patients increases with decline in latitude, being highest in Australia.56 However, the anatomical distribution of BCCs does not correspond well to the area of maximum exposure to UV. Basal cell carcinoma is very common on the head and neck, consistent with a derivation from follicular keratinocytes, but unusual on other light-exposed areas such as the backs of the hands and forearms, unlike actinic keratoses and SCCs, which occur on all light-exposed areas. The inner canthus and eyelids, which are more shielded from sunlight than other parts of the face, are frequently involved. Basal cell carcinoma commonly affects the trunk and rare cases of vulval BCC also occur. The occurrence of BCCs in relatively sun-protected sites suggests that other co-factors may be important and is consistent with regional concentration of follicular sebaceous units. One series found that 7.3 per cent of 1774 cases of BCC had a previous history of trauma to the site of the BCC.57 Basal cell carcinomas may arise in congenital naevus sebaceus, skin damage by X-irradiation, burns or vaccination scars. Arsenic salts used as tonics in the 1930s were also an important aetiological factor. Arsenic-induced tumours are usually multiple and occur mainly on the trunk and may also cause arsenical keratoses on acral sites.58
Basal cell carcinoma 841
Although exposure to UV radiation is accepted as a critical causative factor in the pathogenesis of BCC, the magnitude of the risk associated with increased exposure seems to be insufficient to explain either why particular people get these tumours whereas others do not, or the considerable phenotypical diversity shown by patients in terms of the number and site of tumour and patterns of presentation.59,60 Susceptibility to BCC seems to be determined by a complex interaction between duration and intensity of exposure to UV radiation and polymorphic genes. Specific and distinct clinical phenotypes include presentation with clusters of BCC (termed multiple presentation phenotype) and development of tumours on the trunk. In particular, patients with truncal BCC have more BCCs, are younger and develop larger clusters of BCC.59,61 Susceptibility genes associated with multiple tumours include cytochrome P450 CYP2D6, glutathione S-transferase GSTT1, vitamin D receptor and the tumour necrosis factor family.59 The molecular basis of the basal cell naevus syndrome has been elucidated and is due to constitutive activation of the Sonic Hedgehog signalling pathway due to germ-line inactivating mutations of the PTCH gene.2 The Patched gene is the human homologue of the Drosophila segment polarity gene, which is also called PTCH. PTCH is now known to act as a tumour suppressor gene, requiring mutations in both inherited alleles in order to be inactivated.62 Studies on the PTCH gene in patients with basal cell naevus syndrome and sporadic BCC have demonstrated abnormalities in both copies of PTCH in the vast majority of tumours. PTCH inhibits binding of a protein known to control growth and patterning, called Sonic Hedgehog (SHH), which in turn binds to its transmembrane receptor, Smoothened. Inactivation of PTCH results in the constitutive activation of the SHH signalling and activation of the transcription factors GLI1/2 involved in stem cell proliferation.63,64 Activating mutations of Smoothened have been identified in some cases of sporadic BCC. Recent studies of transgenic mice overexpressing the SHH signalling pathway have demonstrated that only one mutational event is required to produce BCCs in human xenografts, which may
Sonic Hedgehog
Binding of Sonic Hedgehog to PTCH relieves inhibition of Smoothened
PTCH
explain why BCC is so common.65,66 Therefore spor-adic BCCs may result from either an acquired mutation (67 per cent) in both copies of PTCH or an acquired activation mutation (10–21 per cent) of a single copy of the Smoothened allele (Fig. 37.4). However, the correlation between the tumour biology and the clinical diversity observed in patients is currently unknown. Additional events include inactivation of the tumour suppressor gene p53 which controls the G1 cell cycle checkpoint and thus regulates DNA repair or induces apoptosis of cells irretrievably damaged. Mutations of the p53 gene have been detected in almost all human tumours and in skin tumours, these have been shown to be UV related. Mutation of a single p53 allele leads to the production of mutant p53 protein, which will inactivate wildtype p53, allowing cells that have DNA damage to proliferate. One study demonstrated that phenotypically normal UV-exposed human epidermal cells show islands of keratinocytes that show mutant p53 on immunostaining.66 Such islands suggest that these cells are clonally expanded and therefore selectively out-grow those keratinocytes without mutation for p53.67 Indeed, 30 per cent of sporadic BCCs show UV-type mutations of the p53 gene.67 Polymorphisms in the gene for glutathione reductase, which is important in the detoxification of active oxygen species, have been linked to BCCs as have polymorphisms in the cytochrome P450 genes, which are important in metabolizing environmental carcinogens.68 Chemokines such as CXCR4 may play a critical role in progression and angiogenesis of certain subtypes of BCC with a more aggressive nature, and functional blockade of CXCR4 has been proposed as a potential therapeutic strategy for these tumours.69
Clinical features The majority of BCCs, approximately 80 per cent, occur on the head and neck, especially the upper central portion of the face. The other 20 per cent are mainly on the trunk and
Mutation of one allele promotes transcription and proliferation
Smoothened Suppression
Mutation of both copies of PTCH leads to loss of inhibition of Smoothened
Transcriptional factor – Gli
Cell proliferation
Figure 37.4 Schematic mechanism of interaction between Sonic-Hedgehog, PTCH and Smoothed.
842 Non-melanoma skin cancer
lower limbs, particularly in women. It is a common tumour on the eyelid, the inner canthus of the eye and behind the ear. The palm of the hand, sole of the foot and vermilion of the lip are never involved. Early BCCs are commonly small, translucent or pearly, with raised areas through which dilated vessels may show (telangiectasia). The classic form is the rodent ulcer, which has an indurated edge and ulcerated centre. This tumour is slow growing but, if neglected, can spread deeply to cause considerable morbidity due to tissue destruction, especially around the eye, nose, or ear. It may even extend into the periorbital tissues and bone. The morphoeic type occurs almost exclusively on the face, whilst the superficial type is more common on the trunk.70 An aggressive form may occur on the occiput. Basal cell carcinoma has a tendency to grow along the lines of embryonic fusion, which can lead to invasive BCC migrating along the peri-chondrium, periosteum, fascia, or tarsal plate.71 This type of spread accounts for higher recurrence rates noted in tumours involving the eyelid, nose and scalp. The most susceptible areas include the inner canthus, philtrum, middle to lower chin, nasolabial groove, preauricular area and the retro-auricular sulcus.71 Perineural spread is very uncommon, occurs most often in recurrent BCC,72–76 and may present with paresthesia, pain and weakness or, in some cases, paralysis.76–78 Involvement of the cranial nerves and, in one case, thoracic spine has been reported.73,77–79 Metastatic BCC is rare, with incidence rates varying from 0.0028 per cent to 0.1 per cent.80–82 Metastases, when reported, have involved lung, lymph nodes, oesophagus, oral cavity and skin.82–87 Although long-term survival has been reported, the prognosis for metastatic BCC is generally poor with a median survival of 8–10 months after diagnosis.87 Platinum-based chemotherapy is the treatment of choice for metastatic BCC.88,89 In immunosuppressed patients, tumours may be more aggressive locally90 but metastasis remains extremely rare (0.1 per cent) with spread to regional lymph nodes or lung.91 Haematogenous spread has been only rarely reported. The interval from onset to metastasis ranged in one study from 7 to 34 years, with a median of 9 years; the 5-year survival rate was 10 per cent for metastatic BCC in the context of immunosuppression.92 Death may occur by local invasion of tissues such as major blood vessels and the brain.
tissue, which may show degenerative or metaplastic changes. There may also be an inflammatory infiltration of the stroma by lymphocytes, histiocytes and, occasionally, plasma cells.49 The six clinicopathological subtypes of BCC93 are: ● ● ● ● ● ●
Nodular BCC Pigmented BCC Cytic BCC Morphoeic (sclerosing) BCC Superficial BCC Linear BCC.
NODULAR BASAL CELL CARCINOMA
Clinical features This tumour starts as a small ‘glassy/pearly’ papule, which subsequently becomes nodular and undergoes central ulceration. The margins of the tumour are well defined, slightly raised with a rolled border and with a pearly shiny appearance. Blood vessels traversing over the margin produce a telangiectatic appearance (Fig. 37.5). In larger lesions, ulceration may be a feature but the raised, rolled border is often prominent (Fig. 37.6). Histopathology This tumour is composed of discreet islands of darkly staining cells with uniform nuclei and scant cytoplasm within the dermis. Typically, peripheral palisading is present. A prominent connective tissue stroma is an integral part of the tumour. If nodules measure less than 15 μm, the tumour may be called micronodular. An infiltrative pattern is seen in 15–20 per cent of BCCs and is referred to as a histologically aggressive type.
Clinicopathological subtypes The typical histology of BCC consists of clumps of darkly staining basaloid cells with characteristic peripheral palisading. The tumours tend to infiltrate laterally rather than deeply and this, in conjunction with central ulceration, is responsible for the rolled border seen clinically. In addition to keratinocyte proliferation, the stroma proliferates to a varying degree according to the clinicopathological subtype of the BCC. The stroma appears as fine cellular connective
Figure 37.5 Typical nodular basal cell carcinoma (BCC) on a patient’s face.
Basal cell carcinoma 843
PIGMENTED BASAL CELL CARCINOMA
Histopathology
Clinical features
The small basaloid cells with darkly staining uniform nuclei and scant ill-defined cytoplasm may contain granules of melanin, and in such tumours large collections of melanophages may be present in the stroma.
This type of BCC is clinically similar to the nodular BCC but the rolled margins of the tumour are irregularly pigmented. Such pigmented BCCs must be clinically distinguished from melanoma (Fig. 37.7).
CYSTIC BASAL CELL CARCINOMA
Clinical features This is a well-defined papule that attains a pearly coloured lobulated appearance with a telangiectatic surface. The central part of the tumour may ulcerate later in its evolution. Histopathology The hallmark of this tumour is the presence of cystic foci creating a lace-like pattern of small, darkly staining cells. It is thought that the cystic spaces may be due to the cells outgrowing their nutritional supply with the appearance of necrotic areas in the centre of the epithelial strands. However, some cystic spaces may also represent sweat duct differentiation. If mucinous degeneration of the stroma occurs, as well as degeneration of the basaloid tumour, this may give rise to the appearance of the adenocystic variety of BCC characterized by a large cystic space due to necrosis of tumour cells with a peripheral lace-like pattern caused by mucoid degeneration of the stroma. MORPHOEIC (SCLEROSING) BASAL CELL CARCINOMA
Clinical features These tumours clinically appear as sclerotic, slightly depressed areas of skin with prominent telangiectasia (Fig. 37.8). It is often difficult to clinically determine the margins of the tumour, and this type of BCC often recurs due to inadequate excision. Histopathology Figure 37.6 Large ulcerated basal cell carcinoma (BCC) in an 82-year-old woman who was terrified that she had cancer and did not leave the house for 3 years.
Figure 37.7 Pigmented basal cell carcinoma (BCC) on the chest of a 35-year-old man.
The stromal element predominates and thin epithelial cords and strands of basal cells are seen embedded in a dense, sometimes hyalinized connective tissue stroma.
Figure 37.8 Morphoeic basal cell carcinoma (BCC) showing some central ulceration on the cheek of a 67-year-old man.
844 Non-melanoma skin cancer
SUPERFICIAL BASAL CELL CARCINOMA
Clinical features This often occurs on the trunk or limbs and is characterized by a well-defined, erythematous patch that may be large (5–20 cm or more). The surface is telangiectatic and may become eroded. A thread-like margin of the tumour is present, often with a pearly appearance. Most lesions are solitary and may be pigmented. Over many years, the lesion may thicken and ulcerate. Multiple surface BCCs have been associated with arsenic ingestion.58 Histopathology There are small nests of cells along the dermal/epidermal junction with an overlying atrophic epidermis. It has been demonstrated that tumour nests form direct contact with each other in three-dimensional space. LINEAR BASAL CELL CARCINOMA
Clinical features This is an uncommon variant of BCC and was first described in 1985.94 Clinically it is a linear pearly and telangiectatic lesion and is located most often on the head and neck. This variant belongs to a more aggressive subtype and is more likely to have subclinical spread.95 Various mechanisms have been suggested to account for this morphological variant including a Koebner phenomenon or limited lateral spread of the BCC secondary to dermal fibrosis.96
micrographic surgery because of high rates of local recurrence. Radiotherapy is relatively contraindicated.97 There are four generally accepted methods for obtaining tissue diagnosis. These are: shave biopsy, punch biopsy, cytology or definitive surgical excision. A shave biopsy usually is adequate for raised lesions such as nodular BCC. Punch biopsy is effective for sampling superficial BCC, for which shave or excision would be technically inappropriate. An excision biopsy may be used to sample deep dermal and subcutaneous tissue. A comparison of shave biopsy versus punch biopsy was performed retrospectively in 86 cases showing that the two techniques have equivalent diagnostic accuracy rates. Cytology provides a rapid alternative to either punch biopsy or shave biopsy, and it can yield a diagnosis during the initial out-patient appointment. The accuracy of cytology in eyelid lesions suspected clinically to be BCCs was compared retrospectively in 20 lesions from 17 consecutive patients who underwent cytology followed by excision biopsy.98 The sensitivity of cytology for the diagnosis of BCC was 92 per cent. This was compared with a second group of 26 clinical BCCs from 22 consecutive patients who had an incisional biopsy and histological examination followed by excision with histological confirmation, in which the sensitivity was 100 per cent and the accuracy was 96 per cent. Cytology is sufficiently accurate to plan excision and reconstructive surgery when the diagnosis can be confirmed histologically, but it is not sufficiently sensitive for conservative regimens such as radiotherapy because of the small risk of false-negative diagnosis.98
Histopathlogy Nodular masses of basal neoplastic cells are present in the dermis typical of basal cell epithelioma. In the tumour and in the stroma, deposits of melanin may be found, some of which are seen in melanophages.96
Staging Basal cell carcinoma rarely metastasizes, and so a staging work-up is not necessary. However the extent of recurrent aggressive and neglected BCCs may be delineated by MRI.
Diagnosis Management The diagnosis of BCCs is primarily a clinical diagnosis supported by a histopathological examination. The clinical differential diagnoses of BCC includes squamous cell carcinoma, malignant melanoma (pigmented BCC), melanocytic cystic naevi (pigmented BCC), Bowen’s disease (especially superficial BCC), psoriasis (superficial BCC), sebaceous hyperplasia, molluscum contagiosum, and numerous benign appendigeal tumours especially those of follicular origin. Where clinical doubt exists, or when patients are referred for specialized forms of treatment a preoperative biopsy is recommended. Biopsy will also provide information on the histological subtype of the BCC, which influences management and prognosis. A specific adnexal tumour that needs to be distinguished histologically from BCC is microcystic adnexal carcinoma. This tumour is slow growing but is locally very aggressive and destructive, commonly affecting facial areas and specifically the lip. The treatment of choice is extensive surgical resection and preferably
A number of factors have a direct influence on the prognosis of BCC. Patients can be defined in terms of low- or high-risk categories by considering prognostic factors including tumour size, tumour site, tumour type and definition of margins, growth pattern/histological subtype, failure of previous treatment/recurrent tumours and those in immunocompromised patients.99 ●
●
Low risk BCCs are generally small (2 cm), well defined in a non-critical site with non-aggressive histology. Only 5 per cent of well-defined BCCs 2 cm show subclinical spread beyond 5 mm.99 High risk BCCs are generally large (2 cm), indistinct or morphoeic, in a critical site such as around the eyes, ears, lips, nose and nasolabial folds, show aggressive histology such as morphoeic, infiltrative, micronodular or perineural spread.
Basal cell carcinoma 845
The therapeutic aim is to completely eradicate the BCC with the best possible cosmetic outcome, the least chance of tumour recurrence and the minimum of long-term side effects. Factors considered when determining the treatment strategy for BCC are overall health status, histological tumour type, location, size, primary vs. recurrent tumour and patient choice. Patient quality-of-life issues regarding cosmetic outcome associated with surgical procedures are also important (e.g., location of tumour leading to undesirable cosmetic outcomes or large superficial BCC). The closure of the wound needs to be considered and larger tumours may need plastic surgical reconstruction with skin grafts and flaps, with the additional possible complications of dehiscence, necrosis, haematoma and seroma. Not all BCCs require treatment; aggressive treatment might be inappropriate for patients of advanced age or poor general health, especially for asymptomatic low-risk lesions that are unlikely to cause significant morbidity. Furthermore, some elderly or frail patients with symptomatic or high-risk tumours might prefer less aggressive treatments designed to palliate rather than cure. Local availability of various specialized services, together with the experience and preferences of the dermatologist managing the case are also factors that will influence the selection of therapy. Most BCCs are adequately treated by primary excision with low risks of recurrence and no requirement for further therapy. In one study, 41 consecutive patients with 42 BCCs apparently removed surgically were treated by subsequent micrographic surgery, and blocks of tissue, sectioned consecutively until exhausted, were examined for the presence of residual tumour.100 In 28 of 42 cases (66 per cent), residual cancer was identified. The presence of residual cancer was not related to age, site, histological subtype, or extent of surrounding inflammation. The results indicate that patients with small BCCs that appear to be completely removed by initial biopsy may be at risk for recurrence. Infiltrative and micronodular BCC types are the most likely to be incompletely removed by conventional excision. Rates of incomplete excision vary from 5 to 17 per cent. Incompletely excised infiltrative and micronodular BCCs may recur at rates of 33 to 39 per cent. Recurrences after radiotherapy show a tendency toward infiltrative histology and evidence of squamous transformation, and even recurrent BCC after surgical excision may become metatypical. In general, recurrences are more frequent in BCCs with infiltrative and micronodular histology, when clear margins are less than 0.38 mm, and when the histology suggests the presence of squamous differentiation. MANAGEMENT OPTIONS
Many series have shown that treatment by curettage and cautery, surgical excision, radiotherapy, cryotherapy and Mohs’ micrographic surgery have cure rates of well over
90 per cent. Tumours in certain high-risk sites have a greater risk of recurrence, namely the nasal alae, nasolabial fold, tragus and post-auricular area. This may be due to the fusion of embryonal planes at these sites allowing covert migration of tumour cells.101 Whilst most BCCs are treated by one of three methods – surgery, cryotherapy or radiotherapy – a number of novel treatment modalities are now established. Patients with high-risk tumours are best managed by a multidisciplinary team approach to select the most appropriate treatment.
Surgical treatments CURETTAGE AND ELECTRODESSICATION
The tumour and 2–4 mm of adjacent normal tissue are removed by curettage under local anaesthesia and haemostasis is secured by cautery or electrosurgery (electrodessication or coagulation).99,102 Typically two or three treatment cycles are recommended to completely remove the tumour. If the dermis and fatty layer is penetrated then surgical excision should be performed. The wound is left to heal by secondary intention. Topical antibiotics may reduce infection of the open wound. Most patients develop a white scar, or occasionally a hypertrophic scar. The main indication for this treatment is for selective low-risk lesions (small, well-defined primary lesions with non-aggressive histology in non-critical sites) where 5-year cure rates of up to 97 per cent are possible.99 Recurrence rates are higher for tumours in nasal, paranasal, forehead, and mask areas and for tumours 6 mm on the cheek, forehead, scalp and neck.103 However, a study using micrographic surgical techniques104 found that residual tumour was left in at least 30 per cent of cases following curettage, which is far higher than the actual recurrence rate. It is possible that debulking of the tumour provokes an immune response that clears the residual tumour.105 CRYOSURGERY
Liquid nitrogen is used to produce tissue destruction by reducing the temperature to tumoricidal levels. A fine, intermittent spray is preferred, as a continuous one produces a rapid lateral extension of the ice front rather than penetration to the bottom of the tumour. A safe margin of 3–5 mm is first drawn around the tumour, as its edges are difficult to see once frozen. Satisfactory results may be obtained by spraying the centre of the lesion until the ice front is maintained for 3–4 mm beyond the visible edge of the lesion for 30 seconds. After the tumour is frozen, it is allowed to thaw and then the freeze cycle is repeated at least once, which can cause considerable transient discomfort and swelling.106 Two freeze–thaw cycles are recommended for facial BCCs.107 Cryosurgery is suitable for BCCs with clinically well-defined borders and non-aggressive histology but not for those tumours that are in critical facial sites.99
846 Non-melanoma skin cancer
With selected cases, cure rates of up to 99 per cent can be achieved.108 A systematic review of studies published between 1970 and 1997 indicated a cumulative 5-year recurrence rate of 4–17 per cent for primary BCCs.99 The eardrum may need to be protected using dry cotton wool when the ear is being treated. Cryosurgery is not suitable for lesions bordering the lip or eyelid, as scarring leads to a contraction deformity at these sites, and should be avoided in areas of hair growth (e.g., scalp or beard). Use of cryotherapy is contraindicated in thick lesions as the borders of these are poorly defined and the tumour may be inadequately treated. It is also absolutely contraindicated in patients with abnormal cold tolerance, e.g., cryoglobulinemia. After treatment there is short-term pain, tenderness, bullae or vesicle formation, erythema, sloughing of necrotic tissue, eschar formation and often considerable oedema followed by a serous discharge, which may last 2–3 weeks. The site should be kept clean until the eschar separates. Hyperor hypo-pigmentation can occur at the site of freezing and is more obvious the darker the skin of the patient. Scarring tends to be minimal and this technique has rarely been associated with nerve damage. Cryosurgery has many advantages as it is quick to perform, requires no hospitalization and most patients require less than two out-patient treatment sessions.
CURETTAGE AND CAUTERIZATION
Excellent results have been obtained by careful curettage of BCC followed by cauterization under local anaesthetic. A study of 50 patients with morphoeic-type BCCs, which were 10 mm or larger in diameter, sited either on the nose or perinasally, and treated by curettage and cauterization showed good cosmetic results that were acceptable in all patients, with only one recurrent BCC.109
SURGICAL EXCISION
Surgical treatment of primary BCCs is highly effective and the standard of care for BCC. The major advantage is that the tumour and surgical margins can be assessed histologically. In general, the cosmetic results are excellent but the major problem is defining the margins of the BCC prior to excision. The definition of such a margin is made more difficult as 15 per cent of BCCs show subclinical spread beyond 3 mm and 5 per cent show subclinical spread beyond 4 mm. Morphoeic BCCs are even more difficult to define, with 18 per cent showing subclinical spread beyond 5 mm and 5 per cent showing subclinical spread beyond 13 mm.110 In a retrospective cohort analysis of 1983 BCC cases, significant risk factors for incomplete excision included lesions located on the head and neck (p 0.001), surgeons performing fewer than 51 procedures during the 2-year study period (p 0.001), and patients with aggressive histological BCC subtypes (e.g., morphoeic and infiltrative) (p 0.01). The study further indicated that curettage
before surgical excision of BCC decreased the incomplete excision rate by up to 24 per cent (p 0.03).111 The excision of BCCs needs to take into account possible subclinical spread as well as the limitations of the tumour size and tumour site. Although histologically a BCC may not be completely excised, its actual clinical recurrence is much lower than would be expected. Therefore, it is a matter of controversy as to whether an incompletely excised BCC needs to be re-excised. A large study of re-excisions after incomplete excision of BCCs has revealed that in most cases re-excision will reveal residual tumour,112 and recurrence is most likely when both lateral and deep margins are involved. Therefore, if only a lateral margin is involved in a BCC that has a non-aggressive histology and is away from a critical site, it is reasonable to take a conservative approach.99 However, if the BCC involves a deep margin at a critical site with aggressive histology then re-excision with micrographic surgery, or postoperative radiotherapy would be advised. MOHS’ MICROGRAPHIC SURGERY
In this technique, the tumour is debulked with a scalpel or curette with the scalpel angled at 45 degrees to the skin. A saucer of tissue is then removed. Anatomical orientation is carefully maintained by scoring the remaining skin edge and surface of the tissue removed. The tissue is divided along the scored lines, inverted and the edges marked with coloured stains. The tissue is embedded and horizontal frozen sections cut and stained so that the whole of the under-surface of the removed tissue is examined for residual tumour. The surgeon acts as both surgeon and pathologist. The dermatological surgeon then returns to the patient and removes tissue only in areas where tumour is microscopically persistent. This process is repeated until the tumour is totally removed. This technique means that the normal surrounding tissue can be preserved while the whole of the tumour is removed. The defect can be left to heal by secondary intention or reconstruction can commence immediately.113 The overall risk of complications is 2 per cent, most of which involve bleeding.114 Mohs’ micrographic surgery is regarded as a standard treatment for BCC and can give cure rates of 99 per cent in primary and 96–98 per cent in recurrent BCCs.115 It is, however, expensive in both time and support facilities compared with conventional methods of treatment. The average procedure takes 5 hours. Therefore, the main indications for Mohs’ micrographic surgery are for those BCCs in critical sites such as the eyelids, ears, lips, nose and nasolabial folds, morphoeic or infiltrated histological subtype and patients with recurrent BCC especially after radiotherapy. It may also be used for large tumours in high-risk sites.99,116 LASER SURGERY
High-energy pulsed carbon dioxide (CO2) lasers have been used extensively to resurface wrinkled and photodamaged
Basal cell carcinoma 847
skin with a low risk of scarring. Results of histological studies demonstrate precise ablation depths in treated skin with minimal thermal damage to underlying tissue. The ultrapulse CO2 laser with high energy and short pulses achieves char-free ablation of tissue, bloodless surgical field, minimal non-specific thermal damage, rapid healing and diminished postoperative pain. Pulsed CO2 laser treatment can also be effective in ablating superficial BCC. However, there is little published follow-up data to date. A retrospective review of 61 biopsy proven non-recurrent nodular and superficial BCCs in 23 patients has reported only two recurrences with a mean follow-up of 41.7 months (range 15–84 months).117 CO2 laser surgery has also been successfully combined with a micrographic approach similar to Mohs’ micrographic surgery in order to increase its cure rate in a patient with a large BCC, but as yet there have been no trials comparing this modality with the established ones.118,119 Since scarring is minimized with this technique, it has been particularly advocated in eyelid BCCs although the treatment of choice for this site is micrographic surgery.
●
●
●
●
Relative contraindications: ●
●
● ● ●
●
Non-surgical treatments ●
RADIOTHERAPY
Ionizing radiation, since its discovery in 1895, has played an important role in the treatment of skin disorders, none more so than the BCC. Radiotherapy of skin cancer yields excellent results and the cure rates for primary BCC are above 90 per cent.120 A review of all studies published since 1947 suggested an overall 5-year cure rate of 91.3 per cent following radiotherapy (RT) for primary BCC, and a review of all studies published since 1945 suggested an overall 5-year cure rate of 90.2 per cent following RT for recurrent BCC.120,121 In many cases BCCs can be managed equally effectively by surgical techniques or RT. Radiotherapy allows greater preservation of normal tissue than surgical excision and is useful for areas where tissue cannot be readily sacrificed because of cosmetic or functional importance. Indications for RT include: ●
●
●
●
●
● ●
Large superficial lesions where a better cosmetic result can be obtained with radiotherapy Large lesions where surgery would cause major loss of function such as paralysis, numbness, dribbling or ectropion Extensive lesions where surgery such as nasectomy, ear amputation or eye enucleation may be avoided Older patients where long-term skin atrophy caused by RT is not an issue Multiple superficial lesions where surgery would be onerous for the patient Patients who are unfit for or refuse surgery Useful for selected tumours of the eyelids and canthi of the eyes
Useful for selected tumours on the nose, ears and lips. Larger lesions overlying cartilage are best treated with electrons rather than superficial radiotherapy.122 Large lesions on the cheek often respond with minimum scarring Recurrent lesions after surgery, incomplete excision or perineural invasion Nodal metastatic spread.
●
Younger patients under 45 years: there is potential for deterioration of the cosmetic outcome over time (5–10 years) and risk of second malignancy Large lesions involving cartilage, bone, tendons or joints: the risk of radionecrosis is high and cure rates are lower Lesions where there is uncertainty over the histology Lesions that recur after RT Hair-bearing skin such as scalp, eyebrow and eyelashes: risk of permanent epilation Lesions around the upper eyelid: risk of lachrymal gland dryness and upper lid conjunctival keratinization Medial Inner Canthus lesions: risk of nasolachrymal duct stenosis Lesions on the lower leg and back: Poor healing and radiation sequelae, particularly telangiectasia, pigment changes, ulceration, and atrophic scarring.
These relative contraindications need to be reviewed in each individual case, as alternative treatments may produce even more problems. The lower leg and back present particular problems for all treatment modalities. Lesions involving the mid third of the upper lid are readily curable by RT but mucosal changes such as keratinization and cicatricial xerosis with secondary damage to the cornea is probable.123 If an upper lid tumour is sufficiently extensive to require complicated reconstruction then RT may be an alternative, as there is a chance keratinization may be avoided and if it does occur a contact lens provides a remedy. A study of 850 patients treated with RT for carcinomas of the eyelid reported a corneal complication rate of 2.3 per cent, cataract rate of 2 per cent and serious ocular complication rate (including eye loss) of 1.4 per cent.124 The incidence of corneal complications increased from 1 per cent for the lower lid to 6.7 per cent for the upper lid. Damage to the lachrymal gland may cause keratoconjunctivitis sicca with a dry eye. Lesions of the outer third of the upper eyelid are therefore best avoided. If the dose to the lachrymal gland is kept below 35 Gy then late complications (excluding asymptomatic cataracts and a mild dry eye) can be avoided.125 Basal cell carcinomas and skin cancers can be treated by a variety of RT techniques; superficial RT, electron beam RT, megavoltage external beam RT and interstitial brachytherapy. Most superficial lesions are treated with superficial X-rays at 80–150 kV. The tumour must be carefully assessed
848 Non-melanoma skin cancer
macroscopically for its lateral extent, and particular attention paid to the depth of the tumour in sites such as the inner canthus, nasolabial fold, ala nasi, tragus and post-auricular area where deep infiltration may occur. The visible tumour is drawn on the skin; typical radial margins to create the field to treat a BCC are 6 mm around the visible tumour, but this must be interpreted in the light of the clinicopathological type of BCC, the clinical site, organs at risk and size of the lesion being treated. For example, a morphoeic BCC may require a margin of 1–1.5 cm. Percentage depth doses for different field sizes at different energies are consulted to select a beam energy that will encompass the target volume by the 80 per cent isodose. The treatment volume needs to be larger than the target volume due to dose fall off at the field edges. The surrounding skin and any organs at risk are protected by sheets of lead cut out to surround the treatment field. The thickness of the lead depends on the energy used. A lead shield may be required to protect the eye, which is inserted using local anaesthetic drops, and to protect the nasal mucosa and cartilage when the ala nasi is treated. Larger lesions more than 4 cm in diameter that are not near the eyes or complicated air spaces should be considered for electron beam therapy. Electron beam RT for periocular areas is relatively contraindicated because of lateral scatter of electron beam radiation. Where electrons are needed to treat the periocular skin, for example, when treating extensive cutaneous angiosarcoma, internal electron eye shields can be used. These need to be made with 3–4 mm of lead lined with 2–3 mm of silicon depending on the electron energy used. They are thicker and more uncomfortable for the patient than the thinner eye shield needed when using superficial RT. Electrons are ideal for large flat tumours on the scalp and trunk, and for treating tumours overlying cartilage, such as the nose. The depth dose characteristics of electron beams are such that bolus material is needed to increase the skin surface to 100 per cent dose or compensate for an irregular contour, but there is a sharp fall off in dose past the 90 per cent isodose at a depth determined by the electron beam energy. The effective treatment depth in cm is approximately one third of the beam energy in MeV, depending on the field size. The radiotherapy is planned in a similar way as above but an additional margin must be added for electron beam therapy because of the shape of the isodose curves, and an electron end plate for the linear accelerator (LINAC) needs to be made. Computed tomography (CT) is useful for planning large lesions. Good results have been obtained for BCCs that are difficult to treat by other methods.122 The daily dose and fractionation schedule prescribed for the patient depends on the site and size of the tumour, its relationship to other structures such as the eye, ear and nose, the age of the patient and the ease in getting the patient to the radiotherapy centre. Fractionated dose schedules are preferred to a single large dose as these have been found to give rise to better cure rates and better
Table 37.1 Various radiotherapy dose schedules and fractionations currently in use Dose (Gy)
No. fractions
Time
Lesions 3 cm diameter Superficial radiotherapy (SRT) 80–140 kV 40–45 36 30–32 18–22 Dose (Gy)
9–12 6–8 4 1
22 days 17 days 15 days 1 day
No. fractions
Time
Lesions 3 cm nose/pinna/poorly vascularized skin SRT or consider electron beam* radiotherapy 40–45 50–54 Dose (Gy)
9–12 20
22 days 4 weeks
No. fractions
Time
Lesions 5 cm Electron beam* or megavoltage radiotherapy 50 60
20 30
4 weeks 6 weeks
*Consider Increasing dose by 10% to account for the reduced relative biological dose of electrons.
cosmetic results, since tumour tissue recovers more slowly than normal tissue.126 Many different dose schedules are in use and all have been shown to be effective (Table 37.1). Erythema usually develops in the first week of treatment followed by an exudative reaction. Healing is completed about 3 weeks after the first dose. Comedones may appear several weeks after treatment, but usually resolve spontaneously. Long-term side effects include atrophy, hyper- and hypo-pigmentation, telangiectasia and alopecia (Fig. 37.9). Non-healing skin ulceration, persistent pain and secondary skin cancers are more serious late side effects. The cosmetic acceptability of radiotherapy is high at 84 per cent, with cosmetic acceptability being higher for smaller irradiated fields. The incidence of acute complications is normally low at 2 per cent and that of late complications is even lower at 0.3 per cent.120 As discussed above, incompletely excised deep margin positive BCCs in critical sites and with aggressive histology require further treatment. The risk of recurrence is 41–58 per cent if left untreated.127,128 In a series of 187 incompletely excised BCCs, with 93 per cent occurring on the head and neck, 119 were immediately re-treated with RT, one was excised and 67 were not treated. After a median follow-up period of 2.7 years, statistical analysis suggested a 5-year
Basal cell carcinoma 849
indoor lighting. Treatment with PDT is typically associated with localized adverse events, including stinging or burning, erythema and oedema.108 Treatment with PDT is contraindicated in patients with porphyria, known allergies to porphyrins, and patients with photosensitivity to wave lengths of applied light sources.108 Further development of PDT for BCC will make this a more practical and useful modality as, for example, the current use of methyl-ALA providing a shorter pre-exposed application phase of 3 hours. DRUG TREATMENT
Topical fluorouracil Figure 37.9 Two atrophic pale patches on the forehead of a 70-year-old man who had received radiotherapy for basal cell carcinomas (BCCs) 10 years earlier.
probability of cure in the RT group of 91 per cent, and in the untreated group of 61 per cent.129 PHOTODYNAMIC THERAPY
Photodynamic therapy (PDT) involves the topical application of δ-aminolevulinic acid (ALA) to the skin, which causes an accumulation of haem precursors several hours after the initial application. These precursors act as photosensitizers and, when exposed to oxygen and an appropriate light source, a cytotoxic reaction induced by oxygen radicals occurs in cells containing these precursors. Its use in the UK is still essentially investigational with on-going clinical trials, and it is not widely available. A prospective study comparing PDT with cryosurgery for sBCCs and nBCCs in 88 patients reported a recurrence rate at 1 year of 25 per cent for PDT and 15 per cent for cryosurgery.130 A European study in 118 patients comparing PDT with cryotherapy showed that fewer lesions recurred at 12 months with PDT, 8 per cent vs. 16 per cent.131 A study comparing PDT to surgery in 101 patients reported a clearance rate at 3 months of 91 per cent for PDT and 98 per cent for surgery, and at 12 months 83 per cent and 96 per cent, respectively.132 Long-term follow up data is needed. One advantage of PDT is that multiple BCC tumours can be treated simultaneously, but it is a relatively inconvenient treatment option. Treatment involves a two stage process requiring several office visits. After the topical agent is applied, patients must wait several hours (e.g., 14–18 h for ALA; 3 h for methyl-ALA) before the light-application phase of the treatment can be initiated. This second stage of treatment must be performed within a certain time frame after agent application. Furthermore, because single PDT treatments demonstrate poor efficacy in BCC, multiple visits may be required. Increased photosensitivity during the first stage of treatment provides additional inconvenience because patients must avoid sunlight and bright
Topical fluorouracil (5-FU) has been used extensively for treatment of BCCs particularly superficial BCCs. Topical formulations of 5 per cent 5-FU can be administered once or twice daily depending on tolerability for 6 weeks or more. Therapy may be required for as long as 12 weeks. A study of 113 patients with sBCC treated with topical 5-FU reports 93 per cent clearance.108 The common type of nodular tumour is, however, too deep for 5-FU to penetrate adequately, and results in these tumours are poor. It may, however, be used to palliate tumours where a patient may be too debilitated for other treatments. In trials to improve and standardize topical 5-FU therapy, thin BCCs were treated with 25 per cent 5-FU in petrolatum under occlusion for 3 weeks using weekly dressing changes. Of 44 thin BCCs treated, the 5-year cumulative recurrence rate was 21 per cent.133 In the same study, light curettage preceding the 25 per cent 5-FU treatment in 244 BCCs yielded a better 5-year cumulative recurrence rate of 6 per cent. Cosmetic results were good to excellent in more than 80 per cent in both series.133 Side effects include local skin reactions, including pain and burning, pruritus, irritation, inflammation, swelling, tenderness, hyperpigmentation, and scarring. Imiquimod Since BCC responds, albeit partially, to interferon therapy, treatment with imiquimod, a novel immune-response modifier that activates Toll-like receptors to stimulate the innate immune response (interferon inducer), has been investigated. It is currently licensed for the treatment of external genital and perianal warts, actinic keratoses on the face or scalp and superficial BCCs. Randomized studies using various schedules, once daily treatments, once or twice a day 3 times a week, for 6 and 12 weeks to treat sBCCs have shown histological clearance rates of 70–88 per cent.134,135 Two randomized controlled trials comparing 5 per cent imiquimod cream 5 times a week and 7 times a week have shown histological clearance of 75 and 73 per cent respectively.136 There is correlation between the severity of the local reaction and the histological clearance rate. Imiquimod has also demonstrated efficacy in the treatment of nBCCs using similar schedules with histological clearance rates of 71–76 per cent.137,138 It has also been
850 Non-melanoma skin cancer
studied as an adjunctive therapy before excision with MMS for sBCC and nBCC and been shown to reduce the size of the target tumour and improve the resulting cosmetic defect. Imiquimod promotes an inflammatory reaction and treatment is associated with mild-to-moderate local skin reaction. The most common reactions are erythema, crusting, flaking and erosion. However, these dose-related side effects are generally well tolerated and in general do not cause the patient to discontinue treatment. Further follow-up and long-term data are needed but these studies suggest that imiquimod may be a useful monotherapy for BCCs, and possibly a useful adjunctive treatment.
and tend to occur earlier in males, with advancing age, rates become more equal between the sexes.143
Aetiology The factors involved in the aetiology and pathogenesis of cutaneous SCC are similar to those for BCC, including exposure to UV radiation, genetic mutations, immunosuppression and human papilloma virus (HPV) infections. Development of cutaneous SCC has also been associated with radiation exposure, burn scars, chronic inflammatory dermatoses, chronic ulcers, osteomyelitis and arsenic ingestion (see the discussion earlier in the chapter).
Follow-up Follow-up allows detection of tumour recurrence and early treatment of new lesions. Thirty-six per cent of patients who have a previous BCC will go on to develop a further BCC. Those especially at risk of BCCs are those with skin type I and excess sun exposure. For such high-risk patients. 20 per cent go on to develop multiple BCCs.139 Recurrent BCCs tend to develop within 3 years. Although patients with multiple BCCs should be followed up at least 6-monthly on a long-term basis, it is not economically justifiable to follow-up every BCC patient, especially if it is a single isolated BCC in an older age group. Patient education is perhaps more important.
SQUAMOUS CELL CARCINOMA Primary cutaneous SCC is a malignant tumour arising from epithelial keratinocytes of the epidermis or its appendages. It can grow rapidly, invade locally and has the rare potential to metastasize.
Clinical features Cutaneous SCC appears most frequently on sun-exposed sites but may occur anywhere. It is usually characterized by a rapidly growing indurated erythematous or skin-coloured (non-pigmented) hyperkeratotic nodule or tumour that may ulcerate (Fig. 37.10). An appropriate diagnostic biopsy or surgical excision should be performed on any suspicious lesion.
Natural history and risk of recurrence Primary cutaneous SCC may grow slowly or rapidly and may metastasize, usually to the regional lymph nodes. In a review of 695 cases of cutaneous SCC of the trunk and limbs, metastases occurred in 34 cases, with a metastatic rate of 5 per cent, an overall mortality of 3 per cent, and a mortality in the metastatic group of 70 per cent.144 Squamous cell carcinomas are more likely to spread initially to regional lymph nodes than to distant sites, but visceral metastases have been reported in as many as 5–10 per cent
Incidence Cutaneous SCC is the second most common human cancer after BCC. There are over 60 000 cases of NMSC in the UK each year and in the USA there are approximately 200 000 new cases per annum of cutaneous SCC.140 In the head and neck area, 20 per cent of NMSC is cutaneous SCC, but this rises to 43 per cent in sites such as the pinna.141 The common pre-SCC lesion is the actinic keratoses. It is estimated that 11.5 per cent of all visits to a dermatologist in the USA are for actinic keratoses.142 The incidence of actinic keratoses and SCCs are dependent on cumulative sun exposure and the individual’s sun sensitivity. Both are more common in individuals with skin type I. Although actinic keratoses and SCCs are more prevalent in males
Figure 37.10 patient.
Squamous cell carcinoma on the lower lip of a
Squamous cell carcinoma 851
of metastatic cutaneous SCCs. In addition, head and neck cutaneous SCCs can spread to the CNS haematogenously or via the perineural space. Local recurrence and regional metastasis is dependent on prior treatment, location, size, depth, cellular differentiation, histological evidence of perineural involvement and host immune status.145,146 Cutaneous lesions that recur after previous treatment have a higher risk of local recurrence and metastases. Overall lesions recur locally in 25 per cent, lip lesions in 31.5 per cent and ear lesions in 45 per cent.145 Lesions found on the ears and lip are known to be at higher risk of local recurrence and metastasis. Squamous cell carcinomas presenting on the lip have an especially high local and metastatic rate147 with 8 per cent of patients presenting with clinically positive lymph-node involvement and an overall 5-year mortality rate of 17 per cent.112 Tumours arising on non sun-exposed sites and sites of previous radiation or thermal injury, chronic ulcers or chronic inflammation are also reported to have higher metastatic potential.148 Squamous cell carcinoma arising from scar tissue is rare but has a higher rate of recurrence than SCC arising in UV-damaged skin. At least 25 per cent of such tumours may recur in 30 months. Recurrence rates double in tumours greater than 2 cm in diameter from 7.4 per cent to 15.2 per cent.145 It has also been shown that local recurrence and lymph-node spread is associated with a depth of invasion over 4 mm and penetration of the subcutis.149 Local recurrence and metastatic rate increases from 6.7 per cent to 45.7 per cent for tumours with a depth greater than 4 mm. Poorly differentiated and anaplastic SCCs metastasize more frequently than well-differentiated SCCs, 32.9 per cent vs. 9.2 per cent. Squamous cell carcinomas with perineural involvement recur in almost half of all cases (47.2 per cent) and show a metastatic rate of 47.3 per cent; SCCs on the mid-face and lip are especially prone to neural involvement. Careful follow-up of patients with these high risk features is recommended.
Pathology Pre-malignant SCC in situ without dermal invasion, actinic keratosis and Bowen’s disease, have been discussed above. Invasive SCC consists of irregular dermal nodules, composed of a mixture of anaplastic keratinocytes, budding from the epidermis into the dermis. Characteristically, these are relatively large cells, showing lack of maturation, nuclear atypia and the presence of mitotic figures. In early tumours there may be an associated lymphocytic infiltrate. They can be graded as well-, moderately or poorly differentiated, which has prognostic value. While perineural involvement is of prognostic significance but lymphovascular invasion is not an independent risk factor.148 Special histological variants include the adenoid type (or acantholytic variant), which occurs mainly on the head
in the elderly, the verrucous type, which occurs mainly in the oral mucosa, penis or sole of the foot and the spindle cell type. Certain types of benign skin tumour such as keratoacanthoma (KA) may mimic the appearance of SCC. Keratoacanthomas grow rapidly and consist of a solitary nodular lesion with a central keratin plug, which resolves spontaneously if left for up to 4–6 months; KA has a characteristic histological appearance consisting of an exophytic squamo-proliferative nodule with symmetrical buttressed edges and large tumour cells with pale eosinophilic cytoplasm. Mitoses may be seen, and there is a stromal reaction associated with blunt down-growths of squamous epithelium at the lower border of the tumour. Indeed, some authorities consider the KA to be an immunologically controlled SCC. Keratoacanthomas may be derived from follicular structures, and it has been proposed that their regression may be due to recapitulation of the hair cycle. They are usually treated by surgical techniques including curettage/cautery. A very rare familial autosomal dominant variant characterized by multiple self-healing KA (Ferguson Smith type) has been localized to a currently unidentified gene on chromosome 9p22 very close to the PTCH gene of Gorlin’s syndrome. As with Gorlin’s syndrome radiotherapy is relatively contraindicated. Other adnexal tumours that should be distinguished clinically and histologically from SCC include eccrine porocarcinoma (mainly located on lower limbs in elderly), sebaceous carcinoma (mostly located head and neck region especially the ocular area with significant risk of metastasis 20–25 per cent) and verrucous carcinoma (mucosal, genital and acral sites).
Molecular pathogenesis Squamous cell carcinoma are associated with high rates (28 per cent) of UV signature mutations of key tumour suppressor genes such as p53. In addition, inactivation of ARF (13 per cent) and P16 (13 per cent) also contributes to loss of the G1 cell cycle checkpoint control. Activation of the RAS pathway inducing RAF-MEK-ERK signalling has also been shown to be a critical step in SCC due to HRAS mutations (22 per cent) and increased expression of RASGTP (75 per cent).150,151
Treatment The methods of treatment of SCC are similar to those described for BCC. However, SCC has a much greater metastatic potential and, therefore, it is important to ensure that all of the tumour is removed. High-risk tumours, especially those on the lips and ears, are treated more aggressively as the risk of local recurrence and distant metastases are higher. Like BCCs, the treatments are highly effective and can achieve cure rates of 95 per cent. Patients with high-risk SCC presenting with involved lymph nodes
852 Non-melanoma skin cancer
should be reviewed by a multidisciplinary oncology team including a dermatologist, pathologist, appropriately trained plastic or maxillo-facial surgeon, clinical oncologist and a clinical nurse specialist.148 SURGICAL EXCISION
Margins of 4 mm are adequate for most SCCs. However, for high-risk SCC associated with a greater risk of subclinical tumour extension at least a 6–10 mm margin is recommended.112,152,153 Elective prophylactic lymph node dissection has been proposed for SCC on the lip greater than 6 mm in depth and cutaneous SCC greater than 8 mm in depth, but the evidence for this is weak.148 It is not routinely practised and the risk of morbidity has to be seriously considered. MOHS’ MICROGRAPHIC SURGERY
Mohs’ micrographic surgery offers the best chance of cure of high-risk tumours.145 However, most surgeons are able to estimate the margin of most tumours clinically, and 90 per cent of SCCs are excised during the first excision phase of the Mohs’ procedure. Those lesions that are not removed completely with the initial excision tend to be those located on the periorbital region, forehead and cheeks.154 CURETTAGE AND CAUTERY
This technique is likely to result in inadequate clearance, and it should not be regarded as an adequate form of treatment for SCC,148 although this is often acceptable for keratoacanthomas and occasionally for palliative treatment of SCC in the elderly. CRYOTHERAPY
Cryotherapy may be used for actinic keratosis, but it is not recommended as treatment for SCC. RADIOTHERAPY
Radiotherapy is generally reserved for patients over 45 years of age because of the theoretical risk of inducing further malignancies. It is not suitable for tumours invading underlying cartilage where the risk of radiochondritis is high and cure rates are lower. There is a relative contraindication to treating SCCs in cardiac or renal transplant patients as these patients may be particularly susceptible to further cutaneous malignancies. The 5-year cure rate for treating non-melanoma skin cancer with radiotherapy is as high as 90 per cent, and the cosmetic results have been evaluated as good or acceptable in 84 per cent of treated lesions. The acute complication rate is low and the chronic complication rate is even lower.122 Treatment schedules are similar to those for BCC, with
fractionated dose schedules. There may be a benefit to higher dose schedules, and larger lesions will require more fractionated courses. Electron-beam therapy is very good for lesions in sites difficult to treat by more conventional methods.126 Radiotherapy is often used as an adjuvant modality for high-risk SCC, for example, over 2 cm with perineural invasion in a high-risk site. It has been shown in highrisk cutaneous SCCs of the head and neck region with lymph node spread to reduce the risk of recurrence by 50 per cent.155 Radiotherapy may also be used palliatively for patients with lymph node metastases.
CHEMOTHERAPY
For patients with extensive SCC that has relapsed after conventional surgery or radiotherapy or for metastatic disease, a variety of drugs including bleomycin, methotrexate, actinomycin, vincristine, vinblastine, fluorouracil (5-FU) and hydroxyurea have been used with varying degrees of success.156 Cisplatin has been used as monotherapy and in combination with other cytotoxic drugs in advanced lesions and metastatic cutaneous SCC. Eleven patients with advanced BCC or SCC of the skin were treated with cisplatin (75 mg/m2, IV) plus doxyrubicin (50 mg/m2, IV) at 3-week intervals. Responses were seen after 10–12 courses and 5 of the 11 patients were in remission at the time of report.155 A combination of cisplatin, 5-FU and bleomycin produced partial or complete remission in 11/14 patients with advanced cutaneous SCC.157 The search for more active combinations continues. A phase II study of interferon-α (IFN-α), retinoic acid and cisplatin in 39 patients with advanced SCC of the skin showed an overall response rate of 34 per cent with a median duration of response of 9 months.158
Follow-up Patients with a previous history of SCC of skin are more likely to develop similar lesions. In a follow-up study of 101 SCC patients, 52 per cent subsequently developed non-melanoma skin cancer within 5 years of therapy for the first lesion,159 and 95 per cent of local recurrences and metastases are detected within 5 years.148 It is recommended that patients should be followed up every 3 months for the first year and every 6 months thereafter, for a minimum of 5 years after initial treatment. Patients should be examined for local recurrence, metastatic spread and new tumours. Recurrent lesions are treated where possible by surgical excision.148 Mohs’ surgery is particularly suitable to ensure complete clearance of recurrent tumour. Patients with recurrent lesions should be followed up indefinitely. All patients should be taught to self-examine regularly for local and regional recurrence.
Primary cutaneous T-cell lymphoma 853
PRIMARY CUTANEOUS LYMPHOMA Cutaneous lymphomas are a heterogenous group of disorders characterized by localization of malignant lymphocytes to the skin. The term ‘primary cutaneous lymphoma’ refers to cutaneous T-cell lymphomas (CTCL) and cutaneous B-cell lymphomas (CBCL) that present in the skin with no evidence of extracutaneous disease at the time of diagnosis. Primary cutaneous lymphomas often have a different clinical behaviour, require different treatment and have a different prognosis from histologically similar systemic lymphomas, which may involve the skin secondarily. Two thirds of primary cutaneous lymphomas are T-cell in origin (CTCL) of which mycosis fungoides (MF) accounts for 60 per cent of new cases. Sézary syndrome (SS) accounts for only 5 per cent of CTCL cases and CBCL accounts for 25 per cent of all cutaneous lymphoma cases reported.
Incidence The incidence of cutaneous lymphomas is 0.4 per 100 000 per year.160 Most are low grade with long survival and the overall prevalence is much higher. An analysis of US registries has shown that the incidence of MF has markedly increased from 1973 to 1984.161 The incidence is higher in the elderly, African–Americans and men. It can, however, occur at any age in any race.
Aetiology Although the fundamental aetiology and molecular abnormalities that underlie the pathogenesis of different primary cutaneous lymphomas are still an enigma, there is now a wealth of data confirming a wide range of molecular changes. These are reflected in the recent WHO EORTC classification of primary cutaneous lymphomas, which also provides a molecular basis for distinguishing extranodal primary cutaneous lymphomas from their nodal counterparts with a similar morphology and immunophenotype, and crucially explains the different prognosis for lymphomas arising in specific extranodal sites.162 The challenge is now to identify disease-specific abnormalities. It has been postulated that MF is a disease of antigen persistence within the skin, causing chronic lymphocyte stimulation with the eventual transformation of lymphocytes into a low-grade T-cell malignancy.163 Studies have also demonstrated that the neoplastic cell in MF/SS is a Th2 cell. In the early stages of the disease, reactive T cells with a Thl phenotype predominate as part of an anti-tumour response and with disease progression, the malignant Th2 cells become dominant with loss of epidermotropism.164–166 Migration of T cells to the epidermis is mediated by various chemokines and adhesion molecules, but this is lost with the development of tumours.167 Detailed investigation of
pre-existing allergies, atopy or biological, physical or chemical exposure has failed to demonstrate a correlation between these factors and CTCL.168 An association of CTCL with the human T-cell lymphoma/leukaemia virus I and II (HTLV-1 and 2) has not been proven.169,170 In primary CBCL of marginal-zone origin, there is evidence from some geographic areas of an association with Borrelia burgdorferi infection. A study from Italy has shown that B. burgdorferi-specific sequence was detected in 15 out of 83 skin samples of patients with primary CBCL (18.1 per cent), but in none out of 83 matched healthy controls (p 0.0001).171
Classification Previously cutaneous lymphoma has been classified according to the WHO-EORTC classification for cutaneous lymphomas.172 Table 37.2 shows the classification and Table 37.3 splits the conditions into those with indolent, intermediate and aggressive behaviour, with supporting data from 1905 patients from the Dutch and Austrian registers.172
PRIMARY CUTANEOUS T-CELL LYMPHOMA This section will discuss the features and treatment of the different types of CTCL as classified by the WHO-EORTC.
Initial assessment, staging and diagnosis The most common type of primary CTCL is MF. The onset of MF is often insidious with initial manifestations difficult to distinguish from inflammatory dermatoses. The other types of primary CTCL may also present insidiously or more rapidly, and their clinicopathological features are discussed in more detail below. The clinical staging system for MF/SS is shown in Table 37.4.173 Numerous studies in MF have shown that prognosis is dependant on the magnitude of the cutaneous and systemic tumour burden. Poor prognosis is also associated with lymph node involvement, appearance of clonal T cells in the blood, a reduced proportion of CD8-positive lymphocytes in tumour infiltrates, large cell transformation, follicular mucinosis and age over 60 years at presentation.173 The following are recommendations from the Joint Guidelines of the British Association of Dermatologists (BAD) and the UK cutaneous lymphoma group:160 ●
●
Repeated skin biopsies (ellipse rather than punch) are often required to confirm a diagnosis of CTCL Histology, immunophenotypical and preferably T-cell receptor (TCR) gene analysis should be performed on
854 Non-melanoma skin cancer
Table 37.2 WHO-EORTC classification of cutaneous lymphomas with primary cutaneous manifestations Cutaneous T-cell and NK-cell lymphomas
Cutaneous B-cell lymphomas
Mycosis fungoides
Primary cutaneous marginal zone B-cell lymphoma
Mycosis fungoides variants and subtypes • Folliculotropic MF • Pagetoid reticulosis • Granulomatous slack skin
Primary cutaneous follicle centre lymphoma
Sézary syndrome
Primary cutaneous diffuse large B-cell lymphoma, leg type
Adult T-cell leukaemia/lympyhoma
Primary cutaneous diffuse large B-cell lymphoma, other • Intravascular large B-cell lymphoma
Primary cutaneous CD30 lymphoproliferative disorders • Primary cutaneous anaplastic large cell lymphoma • Lymphomatoid papulosis
Precursor haematological neoplasm CD4/CD56haematological neoplasm (blastic NK cell lymphomab
Subcutaneous panniculitis-like T-cell lymphomaa Extranodal NK/T-cell lymphoma, nasal type Primary cutaneous peripheral T-cell lymphoma, unspecified • Primary cutaneous aggressive epidermotropic CD8 T-cell lymphoma (provisional) • Cutaneous γ/δ T-cell lymphoma (provisional) • Primary cutaneous CD4 small/medium-sized pleomorphic T-cell lymphoma (provisional) a
Restricted to lymphomas of alpha/beta T-cell origin. Based on recent evidence suggesting derivation from a plamsacytoid dendritic cell precursor, this condition has also been designated as early plasmacytoid dendritic cell leukaemia/lymphoma.22 b
●
●
●
●
all tissue samples (ideally molecular studies require fresh tissue) All patients (with the possible exception of early stage mycosis fungoides (stage IA) and lymphomatoid papulosis) should ideally be reviewed by an appropriate multidisciplinary team for confirmation of the diagnosis and to establish a management strategy Initial staging computed tomography (CT) is required in all patients, with the exception of those with early stages of MF (stage IA and IB) and lymphomatoid papulosis At diagnosis, peripheral blood samples should be analysed for total white cell, lymphocyte and Sézary cell counts, serum lactate dehydrogenase (LDH), liver and renal function, lymphocyte subsets, CD4:CD8 ratios, HTLV-1 serology and, preferably, TCR gene analysis Bone marrow aspirate or trephine biopsies are required for CTCL variants (with the exception of lymphomatoid papulosis) and may also be appropriate for those with late stages of MF (stage IIB or above and SS).
A recent study in 13 patients with MF and SS at risk for secondary lymph node involvement has shown that positron emission tomography (PET)/CT is more sensitive in detecting lymph nodes involved by lymphoma compared with CT data alone, and thus may provide more accurate
staging and prognostic information.174 Palpable lymph nodes should have histological confirmation with an excision biopsy, as fine needle aspirates cannot establish accurately histological lymph node involvement in MF. All histology should be reviewed by a central panel, usually within the specialist cancer centre, and it is important that the histological features are correlated with the clinical features and the patients disease classified according to the WHO-EORTC classification. In addition to morphology, immunophenotypical studies and TCR gene analysis should be performed on all tissue samples. Immunophenotypical studies should be performed on paraffin-embedded sections and include the T-cell markers CD2, CD3, CD4, CD8, B-cell marker CD20 and the activation marker CD30. Additional markers such as p53 may have prognostic significance in MF and other CTCL types such as adult T-cell leukaemia/lymphoma. Markers of cytotoxic function such as TIA-I, the monocyte macrophage marker CD68 and natural killer (NK) cell marker CD56 may be useful for specific CTCL variants.
Mycosis fungoides The term ‘mycosis fungoides’ was first used by Alibert, a French dermatologist, when he described a severe disorder
Primary cutaneous T-cell lymphoma 855
Table 37.3 Relative frequency and disease-specific 5-year-survival of 1905 primary cutaneous lymphoma classified according to the WHO-EORTC classification Number
Frequency (%)
Disease-specific 5-year survival (%)
800 86 14 4 146 236 18 39
44 4 1 1 8 12 1 2
88 80 100 100 95 100 82 75
Aggressive clinical behaviour Sézary syndrome Primary cutaneous NK/T-cell lymphoma, nasal-type Primary cutaneous aggressive CD8 T-cell lymphomab Primary cutaneous γ/δ T-cell lymphomab Primary cutaneous peripheral T-cell lymphoma, unspecifiedb
12 7 14 13 47
3 1 1 1 2
24 NR 18 NR 16
Cutaneous B-cell lymphoma Indolent clinical behaviour Primary cutaneous diffuse large B-cell lymphoma, leg type Primary cutaneous diffuse large B-cell lymphoma, other Primary cutaneous intravascular large B-cell lymphoma
85 4 6
4 1 1
WHO-EORTC classification Cutaneous T-cell lymphoma Indolent clinical behaviour Mycosis fungoides Folliculotropic MF Pagetoid reticulosis Granulomatous slack skin Primary cutaneous anaplastic large cell lymphoma Lymphomatoid papulosis Subcutaneous panniculitis-like T-cell lymphoma Primary cutaneous CD4 small/medium pleomorphic T-cell lymphomaa
55 50 65
a Data are based on 1905 patients with a primary cutaneous lymphoma registered at the Dutch and Austrian Cutaneous Lymphoma Group between 1986 and 2002. b Primary cutaneous peripheral T-cell lymphoma, unspecified excluding the three provisional entities indicated with an asterisk (*); NR: not reached.
Table 37.4 Clinical staging for mycosis fungoides (A) Classification T1 T2 T3 T4 N0 N1 N2 N3 M0 M1 B0 B1
Patches, plaques, or both, involving 10% body surface area Patches, plaques, or both, involving 10% body surface area One or more cutaneous tumours Generalized erythroderma Lymph nodes clinically uninvolved Lymph nodes clinically enlarged but not histologically involved Lymph nodes clinically non-palpable but histologically involved Lymph nodes clinically enlarged and histologically involved No visceral metastases Visceral metastases No circulating atypical cells (Sézary cells), 5% Circulating atypical cells (Sézary cells),
(B) Clinical stages IA IB IIA IIB III IVA IVB
TNM T1 N0 M0 T2 N0 M0 T1-2 N1 M0 T3 N0-1 M0 T4 N0-1 M0 T1-4 N2-3 M0 T1-4 N0-3 M1
Sézary syndrome is staged as T4 N1 or N3 M0 B116.
Expected 5-Year survival16 (%) 96–100 73–86 49–73 40–65 40–57 15–40 0–15
856 Non-melanoma skin cancer
in which large necrotic tumours presented on a patient’s skin. CLINICOPATHOLOGICAL FEATURES
evolution of patches and plaques, then another type of CTCL should be considered. The histopathological changes characteristic of MF are an epidermotropic proliferation of small to medium sized
Mycosis fungoides is the classical ‘Alibert–Bazin’ type of CTCL characterized by the slow evolution of patches, plaques and tumours over years or sometimes decades (Figs 37.11–37.13). Erythrodermic MF is characterized by more than 80 per cent skin involvement with diffuse infiltration that can give rise to a leonine facies (Fig. 37.14). When tumours develop in MF they commonly show ulceration but the patient must have typical polymorphic patches and plaques of MF in other sites to confirm a diagnosis of MF. It typically affects older adults but may occur at any age. The initial skin lesions are most often on sunprotected skin around the bathing trunk distribution. In the later stages it may involve the lymph nodes and visceral organs. If only tumours are present without the preceding Figure 37.13 Tumis cutaneous T-cell lymphomas (CTCL) arising from long-standing plaque stage disease on the forehead and supraocular region.
Figure 37.11 Patch stage of cutaneous T-cell lymphomas (CTCL) of the poikiloderma atrophicans type.
Figure 37.12 Plaque stage of cutaneous T-cell lymphomas (CTCL) with well-demarcated papillosquamous lesions on the forearms.
Figure 37.14 Erythrodermic cutaneous T-cell lymphomas (CTCL) in an elderly man with diffuse infiltration of the skin giving rise to leonine facies.
Primary cutaneous T-cell lymphoma 857
T-lymphocytes with cerebriform nuclei that colonize the basal layer of the epidermis as single haloed cells or a linear configuration of cells. Epidermotropism is more pronounced in plaques than patches. Pautrier microabscesses (intra-epidermal collections of atypical cells) are a highly characteristic feature. The dermal infiltrate becomes more diffuse and epidermotropism may be lost with progression to tumour stage. Large cell transformation may develop, and the cells may be negative or positive for CD30. Transformation is often associated with more aggressive disease and a poorer prognosis. Mycosis fungoides cells have a CD3 , CD4, CD8 memory T-cell phenotype. There are rare cases of CD4, CD8 classical MF, which have the same clinical behaviour and prognosis although CD8 MF is more common in childhood MF and in those patients with hypopigmented variants. MOLECULAR PATHOGENESIS
In CTCL no disease-specific balanced translocations have yet been identified but molecular cytogenetic studies do indicate that MF and SS have a closely related pattern of chromosomal abnormalities, suggesting that the two conditions share a similar if not identical pathogenesis.175,176 Numerical rather than structural abnormalities predominate, with losses of 1p, 10q, 13q, 17p and gains of 4, 17q and 18 common.177–179 As for other haematological malignancies, inactivation of cell cycle and apoptosis genes is common in MF/SS and is probably partly due to a ‘mutator phenotype’ associated with microsatellite instability due to hypermethylation of mismatch repair gene promoter sequences.180–184 Promoter hypermethylation and repression of gene transcription appears to be a common mechanism of gene inactivation in CTCL.185 Chromosomal amplification of JUNB at 19p12 has also been detected in MF/SS.186 JunB is part of the AP1 transcription factor complex involved in control of cell proliferation, differentiation and apoptosis and over-expression of JunB is likely to be the explanation for the Th2 cytokine profile commonly seen in CTCL. cDNA array studies in SS have identified a gene signature that appears to confer a worse prognosis and have also shown over-expression of a tyrosine kinase receptor, EphA4.187,188 Similar studies in MF have shown evidence for dysregulation of tumour necrosis factor (TNF) signalling pathways.189 The signal transducers and activators of transcription (STAT) family of transcription factors are critically involved in regulating T-cell activation and, as for other solid and haematological malignancies, constitutive expression of Stat3 protein has been found in MF/SS.190 Interestingly it has been shown that EphA4 activates the Janus kinase (JAK) STAT downstream signalling pathway, which might explain the constitutive expression of Stat3 in CTCL, but this has yet to be confirmed. There is also evidence for expression of a truncated Stat5 protein in SS due to the presence of a serine protease, a situation mirrored in unactivated T-cells.191 Both early and late stages of
T-cell activation may be further perturbed as there is evidence for hypermethylation of the phosphotyrosine phosphatase (SHP-1) promoter.192 These varied and complex abnormalities have a wide variety of potential consequences, but a unifying hypothesis would be that there is dysregulation of T-cell activation in MF/SS, which prevents activation induced T-cell death. THERAPY
Overview The current therapeutic approach in MF consists of skin directed therapies for early stage of the disease. The rationale for this approach is based on the one randomized controlled trial by Kaye et al. that showed that intensive therapy with total skin electron beam therapy (TSEBT) and combination chemotherapy had a higher response rate but did not improve survival when compared to a palliative initial skin directed approach to treatment.193 The complete response rate was 38 per cent versus 18 per cent, but the morbidity was greater on the combination treatment arm. The overall survival was the same in both groups at 75 months. There is a large number of uncontrolled trials with data showing that the disease responds to treatment but relapses and no therapy has been shown to impact on overall survival. Therapy therefore needs to be individualized to the patient and delivered by an experienced multidisciplinary team. Maintenance of quality of life is central to the therapeutic strategy. Psychological issues must be addressed, and issues such as skin pain control are important and often overlooked. The therapeutic options are discussed in detail below. In the majority of patients with early disease and normal life expectancy, toxic and aggressive therapies should be avoided. Patients with more advanced disease should be offered entry into appropriate clinical trials as the therapies discussed below are mostly palliative and do not have a significant impact on disease outcome. Table 37.5 shows the various therapeutic options by stage of disease.160 Topical therapy Topical emollients can be used with aqueous cream combined with menthol and is sometimes helpful for pruritus. Topical corticosteroids can induce complete clearance of disease but this is usually short lived.194 Topical mechlorethamine (nitrogen mustard, NH2) is effective for superficial disease, but in Europe is only available in France since 9/11. There is no consensus on whether it should be applied to individual lesions or to the whole skin, daily or twice daily and the duration of treatment. Hypersensitivity reactions and irritant or allergic dermatitis occurs in 10–67 per cent, but is less likely with the ointment formulation. It should not be used in pregnancy. Topical carmustine (BCNU) is a topical chemotherapeutic agent with response rates of 47–86 per cent dependant on stage. Alternate day or daily treatment with 10 mg of carmustine in 60 ml of
858 Non-melanoma skin cancer
Table 37.5 Treatment of mycosis fungoids/Sézary syndrome Stage
First line
Second line
Experimental
Not suitable
IA
SDT or no therapy
Bexarotene gel
Chemotherapy
IB
SDT
SDT or no therapy IFN-α PUV, TSEB IFN-α PUV, TSEB
Chemotherapy
IIA
SDT
IFN-α PUV, TSEB
IIB
Radiotherapy or TSEB, chemotherapy PUVA IFN-α, ECP IFN-α, methotrexate
IFN-α, denileukin, diftitox, bexarotene TSEB, bexarotene, denileukin diftitox,* chemotherapy, alemtuzumab IFN-α, denileukin diftitox*, alemtubumab Palliative therapy
Denileukin diftitox,* bexarotene Denileukin diftitox,* bexarotene Autologous PBSCT, mini-allograft Autologous PBSCT, mini-allograft
III
IVA IVB
Radiotherapy or TSEB, chemotherapy Radiotherapy, chemotherapy
Autologous PBSCT, mini-allograft Mini-allograft
Chemotherapy Ciclosporin Ciclosporin
Ciclosporin
ECP, extra-corporeal photopheresis; IFN-α, interferon-α; PBSCT, peripheral blood stem cell transplant; PUVA, psoralen ultraviolet A; SDT, skin-directed therapy including topical emollients, steroids, mechlorethamine, carmustine, bexarotene gel, UVB/PUVA, superficial radiotherapy; TSEB, total skin electron beam. Stage III includes Sézary syndrome, although some cases of Sézary syndrome will be stage IVA. ECP ideal for those patients with peripheral blood involvement. *Not yet licensed in Europe.
dilute alcohol (95 per cent) or a 20–40 per cent ointment is used. Hypersensitivity reactions are seen in 5–10 per cent of cases. Regular blood counts should be performed to monitor for bone marrow suppression, and for this reason maintenance therapy is not advised.173 Bexarotene gel is a new topical retinoid (‘rexinoid’) that has been approved by the US Food and Drug Administration (FDA) for topical therapy of early stage I MF, but currently does not have a licence in Europe. A phase I/II study in 67 patients reported a 63 per cent response rate with 21 per cent complete responses. The median time to response was 20 weeks with a median duration of 99 weeks. It is generally well tolerated with irritant side effects restricted to the site of application.173 Phototherapy The standard for early stage MF is PUVA, which consists of oral psoralen (8-methoxypsoralen; 8-MOP; methoxsalen) plus ultraviolet A phototherapy.160,173 Psoralens form bifunctional and monofunctional DNA adducts when photoactivated. PUVA produces response rates of 79–88 per cent in stage IA and 52–59 per cent in stage IB. It is ideal for patients with stage IB/IIA who are intolerant or fail to respond to topical therapy, although both therapies can be complementary. There is no significant response in stage IIB tumour stage MF, but PUVA can occasionally be effective for erythrodermic stage III disease, if tolerated, and combined PUVA and IFN-α has also been used for stage III disease with some benefit. There is an ongoing EORTC phase III trial of PUVA plus bexarotene compared with PUVA alone in early stage IB-IIA disease.
Treatment schedules can vary but are usually two to three times weekly treatment until the disease is cleared or best partial response attained. Efforts are made to restrict the total cumulative dose of PUVA to 1200 J/cm2, or less than 200 treatment sessions, to reduce the risk of non-melanoma skin cancers. Nausea occurs rarely but can be avoided by using 5-methoxypsoralen (5-MOP) instead of 8-MOP. PUVA is one of the most effective therapies for early stage MF but there is no data to establish whether it improves overall survival. Broadband and narrowband UVB and high-dose UVA1 phototherapy can also achieve good responses in early stages of MF but may not be as effective as PUVA for those with extensive plaques. UVB may have a lower risk of cutaneous carcinogenesis than PUVA but there have been no adequate comparative studies in MF as yet. Radiotherapy Radiotherapy (RT) is a very effective single agent for the treatment of MF. Dramatic responses to low doses of X-ray therapy were first reported in 1902.195 Radiotherapy is now widely used to treat individual thick plaques, eroded plaques or tumours, and the whole skin can be successfully treated with TSEBT. Mycosis fungoides is a very radio-sensitive disease. Several clinical studies have evaluated the dose–response relationship in MF. There is a direct correlation between disease-free interval and dose, and split dose experiments have shown no difference in tumour regrowth at intervals of 1 or 7 days suggesting little recovery between fractions enabling protracted fractionation regimens.196
Primary cutaneous T-cell lymphoma 859
A study of fractionated RT for individual plaques and tumours has shown a response rate of 89–96 per cent with the likelihood of recurrence inversely related to dose. Local recurrence occurred in 42 per cent with doses less than 10 Gy, 32 per cent for doses of 10–20 Gy, 21 per cent for doses 20–30 Gy and no recurrences with doses over 30 Gy.197 A similar dose control relationship has been shown for TSEBT. The Stanford data has shown the complete remission (CR) rate to be 18 per cent for less than 10 Gy, 55 per cent for 10–20 Gy, 75 per cent for doses 25–30 Gy and 94 per cent for doses over 30 Gy, with improved durability of response over 25 Gy.198 Current standard therapy consists of low-dose superficial RT to treat individual thick or eroded plaques and tumours initially. This can be combined with other modalities such as PUVA. Large tumours may be treated with electrons dependant on the tumour position, size and thickness. Doses of 8 Gy in two fractions, 12 Gy in three fractions or 15 Gy in five fractions are used. Closely adjacent and overlapping fields can often be retreated successfully and the low doses allow the safe use of TSEBT later in the patient’s disease. The possible exception to this is unilesional presentation of MF, which is rare, but local RT may be curative in this setting.198 Disease-free survival following RT alone is reported as 85 per cent at 10 years using fractionated treatment to a total dose of 20–30 Gy.198 Whole-body TSEBT therapy has been evaluated extensively in CTCL although it is not widely available. It was first reported for MF in 1953 by Trump et al. and informed the design of the linear accelerator (LINAC). The most widely used technique is that reported by Stanford in 1960, and the first clinical results were published by the St John’s Institute of Dermatology in 1962.199,200 The Stanford technique involves the patient standing in six different positions to allow maximum unfolding of the skin in areas such as the axillae. Sites such as the perineum, soles of the feet and scalp need extra patch treatments with RT. The RT is delivered by a LINAC in high dose-rate electron mode with the beam attenuated so that the effective dose at the patient surface is approximately 4 MeV. It is a technically challenging treatment and careful attention to dosimetry is needed. Consensus guidelines for TSEBT have been published by the EORTC and make recommendations on the technique, dosimetry, patch and boost treatments, radiation prescription dose, schedule, fractionation and shielding.201 The 80 per cent isodose must be at least 4 mm deep with respect to the skin surface along the primary axis of the beam, and the dose at 20 mm should be less than 20 per cent of the maximum skin-surface dose. Photon contamination must be reduced to less than 0.7 Gy for the full course to limit the dose to the bone marrow. Areas that are inherently shielded may be patch treated, and there is retrospective data that suggests this may improve outcome.201 Large tumours should be debulked with palliative short courses of RT prior to TSEBT to improve response and to start the healing of large tumours, which helps the
patient tolerate treatment. Other techniques such as the McGill rotating technique are used by some centres and meet the EORTC recommendations.202 The EORTC consensus recommends at least 26 Gy to a depth of 4 mm in the skin and 31–36 Gy to the truncal surface with similar patch treatments to inherently shielded areas such as the soles of the feet. Fractionation to 30–36 fractions over 6–10 weeks is recommended, and keeping the dose per fraction below 2–2.5 Gy reduces the long-term side effects. The Stanford group currently deliver 36 Gy in 36 fractions over 9 weeks, and this is used in other US centres. Availability of TSEBT is limited to a few large cancer centres in the UK. The schedule that has been used for many years for patients under the St John’s Institute of Dermatology is 30 Gy in 20 fractions over 5 weeks, and acute and long-term side effects are similar to those seen with more protracted schedules. A systematic review of open uncontrolled and mostly retrospective studies of TSEBT as monotherapy in 952 patients with CTCL has established that responses are stage-dependent.161 It has a role in every stage of the disease. In stage IA, complete responses of 96 per cent with 10-year recurrence-free survival (RFS) of 50 per cent are reported.203,204 However this is a controversial first-line treatment for such early stage disease where long-term survival is similar to healthy controls; TSEBT is therefore reserved for patients refractory to standard first-line skindirected therapy.205 In stage IB/IIA disease CR rates of 76–90 per cent are reported but with a 10-year RFS of only 10 per cent with TSEBT alone. This indicated that TSEBT is not curative, even in early disease, but significant disease control can be achieved and hopefully maintained with adjuvant therapy.201,206–208 Retrospective data from Yale has shown that adjuvant PUVA improves the 5-year DFS from 50 per cent to 85 per cent, and data from Stanford shows that adjuvant use of topical mechlorethamine improves 10-year RFS from 10 per cent to 40 per cent. In stage IIB disease, complete responses are less common (36–54 per cent) with a 5-year RFS of 30 per cent with TSEBT alone.160,203,207 Adjuvant mechlorethamine improves the 5-year RFS to 55 per cent, and retrospective data from Yale has shown a 5-year RFS of 100 per cent in patients treated with adjuvant photophoresis following TSEBT. Other adjuvant options include IFN-α and bexarotene. The newer single-agent chemotherapy drugs can be used to debulk extensive tumours prior to TSEBT. For erythrodermic (stage III) disease TSEBT is very effective and can produce rapid and sustained responses. A retrospective study of erythrodermic disease has also shown 60 per cent CR with 26 per cent progression-free at 5-years.208 In this study the overall median survival was 3.4 years with a median dose of 32 Gy given as five fractions per week over 6 to 9 weeks. Patients with stage III disease did best compared with those with significant nodal or haematological (IVA and IVB) disease. The duration of response was also longer for those who received more than 20 Gy using 4–9 MeV. Adjuvant photophoresis may be of
860 Non-melanoma skin cancer
benefit and retrospective data from Yale shows an improvement in 2-year cause-specific survival from 69 per cent to 100 per cent.208 In stage IVa disease, TSEBT may be used palliatively with CR in the order of 70 per cent, but this is short lived.201 Acute adverse effects of TSEB are usually minor with modern techniques and attention to care of the patients skin.209 They include fatigue, temporary alopecia, nail loss, leg swelling and blisters, minor nose bleeds (1 in 30), reduced sweating (1 in 30) minor parotitis (1 in 30) gynaecomastia (1 in 30) and skin infection, which is rare but must be treated aggressively (1 in 100). Late effects are skin atrophy, hypothyroidism, nail and finger changes, sun sensitivity and infertility in men. Combined with previous SDT and PUVA, TSEBT adds to the patients risk of other cutaneous malignancies. Although TSEBT is usually only given once in a lifetime, several reports have documented patients who have received two or three courses; however, the total doses tolerated and the duration of response have been lower with subsequent courses.160,201
but their role remains to be established. Ciclosporin is contraindicated as it may cause disease progression.217–218 Alemtuzumab is a humanized recombinant IgG monoclonal antibody specific for the CD52 cell surface glycoprotein found on malignant B and T cells, but opportunistic infections such as cytomegalovirus (CMV) reactivation are a concern.173 Overall, RR of 55 per cent with 32 per cent CR in advanced stages of MF/SS have been reported, with a 12 month median duration of response. Denileukin difitox is a recombinant fusion protein comprising diphtheria toxin fragments and IL-2 sequences.173 Completed studies have shown an overall response rate of 30 per cent and CR rate of 10 per cent with a median duration of 6.9 months. Steroid pre-treatment has been shown to increase the overall response rate (ORR) to 60 per cent and reduce adverse events. However, side effects include fever, chills, myalgia, nausea and vomiting and approximately one quarter of patients developed a ‘vascular leak’ syndrome characterized by the presence of two or more of the following: hypotension, oedema, hypoalbuminemia.173 Denileukin difitox may be useful for advanced disease, but further experience in Europe is required.
Immunotherapy
Chemotherapy
Immunotherapy has been used in CTCL to enhance the anti-tumour host responses by promoting the generation of cytotoxic T-cells and Th1 cytokine responses. The most commonly used immunotherapy is IFN-α, which binds to the type 1 interferon receptor expressed by CTCL. It acts via its effects on cell cycle regulation, oncogene suppression and modulation of cell adhesion.210 Studies have shown overall response rates of 45–74 per cent with CR rates of 10–27 per cent.211–213 The dosage schedule most commonly used is 3 MU three times a week, increased if tolerated to a maximum of 36 MU per week, but in practice it is rare for patients to tolerate more than 15 MU per week. Response rates are higher with higher doses and higher in earlier compared with later stages of disease.212,213 Side effects include dose-related ‘flu’-like symptoms, elevated transaminases, leukopenia and thrombocytopenia. Interferon-α has been used in combination therapy. Adding PUVA to IFN-α has been shown to improve the CR rate from 38 per cent to 70 per cent, but there is no data on duration of response.214 Uncontrolled studies of PUVA and IFN-α in MF and SS have shown overall remission rate (RR) of 100 per cent with 62 per cent CR rate, and combination therapy is useful for patients with resistant earlystage disease such as those with thick plaques and folliculotropic disease.215 A randomized study in earlystage disease comparing PUVA with PUVA and IFN-α has shown similar RR but the cumulative dose of PUVA was lower in the combination arm. Interferon-α has been used in combination with retinoids but no improvement in RR has been shown.216 Small cohort studies have reported that interleukin-12 (IL-12) and interferon-γ can produce clinical responses
Mycosis fungoides is relatively chemo-resistant and responses are usually short lived as shown by the randomized controlled trial by Kaye et al.174 This may reflect the low proliferative rate and a high prevalence of inactivating p53 mutations producing a relative resistance to tumour cell apoptosis. Systematic reviews of chemotherapy have shown CR rates of 33 per cent for single-agent chemotherapy with median duration of 2–3 months and CR rates of 38 per cent for combination chemotherapy with median duration of 5–41 months. Patients with CTCL are prone to infection, and septicaemia is a common pre-terminal event, and is a problem for patients receiving chemotherapy. There are now newer single-agent chemotherapy drugs with higher response rates and a safer toxicity profile, but the duration of response remains short and patients should be offered controlled clinical trials to find more effective chemotherapeutic treatments for this disease. Chemotherapy should only be used for patients with stage IIB to IVB disease. Combination chemotherapy with CHOP (cyclophosphamide, doxorubicin, vincristine and prednisolone) has a high response rate, but with significant toxicity and should be reserved for patients with good performance status and stage IIB to IVB disease as initial treatment if the patient is to be considered for a stem cell transplant, and for similarly good performance patients with effaced lymph node disease or visceral and critical organ disease where a rapid palliative response is needed. Other patients should be considered for clinical trials or treatment with newer single agent chemotherapy drugs to reduce the toxicity of treatment and maintain the patient’s quality of life. Single agent chemotherapy that has been widely used includes oral chlorambucil, oral methotrexate, oral etoposide,
Primary cutaneous T-cell lymphoma 861
and the intravenous use of purine analogues pentostatin (2-deoxycoformycin) and fludarabine phosphate. Open studies of pentostatin have shown RR of 35–71 per cent with CR rates of 10–33 per cent, and it is especially useful for erythrodermic disease and SS.220,221 It is however nephrotoxic and causes T-cell depletion increasing the risk of opportunistic infections requiring prophylaxis with co-trimoxazole (Septrin), aciclovir and antifungals. Methotrexate produces CR of 41 per cent in patients with erythrodermic disease given as a single weekly dose ranging of 5–125 mg. Fludarabine in combination with cyclophosphamide has been studied in erythrodermic MF and SS but responses are short lived and bone marrow toxicity is common.222 Other single-agent chemotherapy agents liposomal doxorubicin hydrochloride and gemcitabine have been shown to produce high response rates with a more favourable toxicity profile. A phase II study of gemcitabine monotherapy achieved an ORR of 70 per cent (10 per cent CR and 60 per cent PR) in 30 previously treated patients with T3 or T4 disease. The median durations of CR and PR were 15 months (range: 6–22 months) and 10 months (range: 2–15 months), respectively. Treatment was well tolerated; haematological toxicity was mild, and no nausea/vomiting or organ toxicity was recorded.223 The same group previously also reported 32 untreated patients (26 MF; 5 PTCLU, 1 SS) showing a CR rate 22 per cent and PR rate of 53 per cent with a median duration of response of 10 months (range: 4–22 months) with only mild toxicity.224 A UK multicentre trial of gemcitabine in combination with bexarotene is due to start recruiting in 2006, and it is hoped that the bexarotene will improve the duration of response. Liposomal doxorubicin has shown good efficacy as a single agent for MF. Pegylated liposomes are stable long-circulating carriers useful for delivering doxorubicin to tumour sites with a lower toxicity profile than the free drug. A phase II study of pegylated liposimal doxorubicin as second-line therapy in 34 patients with CTCL showed an overall response rate of 88.2 per cent. Fifteen patients achieved a CR and 15 patients a PR with an overall median survival of 17.8 ( 10.5) months and event free survival of 12 ( 9.5) months.225 Adverse effects were seen in 14 patients and were temporary and generally mild. Only 6 patients had Grade 3 or 4 adverse effects. A retrospective study of liposomal doxorubicin for patients with stage IVB MF has shown 30 per cent with PR and 20 per cent with stable disease (SD) in 10 patients with limited toxicity.226 An EORTC phase II trial of liposomal doxorubicin for stage IIB–IVA MF is still recruiting. Allogeneic stem cell or bone marrow transplantation has only been used in a few patients with encouraging results.227,228 A recent report of reduced intensity allogeneic stem cell transplant in 8 patients with advanced MF and SS reported that 6 remain alive after a median follow up of 56 months with no evidence of lymphoma. Two patients died from transplant-related complications.229
The graft-versus-lymphoma effect may be important, and further assessment of such reduced intensity allografts for CTCL is under way. All patients with a poor prognosis and good performance status should be selected for existing clinical trials. Novel retinoids Bexarotene is highly selective for the retinoid X receptor, and was the first ‘rexinoid’ to undergo clinical development.230,231 The drug has received EMEA approval in Europe for the treatment of skin manifestations in advanced CTCL.232 Although the precise mechanisms are unknown, in-vitro studies have shown that bexarotene can inhibit growth in tumour cell lines and cause in-vivo tumour regression in animal models: the drug also stimulates apoptosis.233 In phase II and III studies of 152 patients with CTCL response rates from 20 to 67 per cent have been reported.231,234 Bexarotene is usually administered at 300 mg/m2/day and treatment is continued indefinitely in patients who respond.160,235 Bexarotene causes severe central hypothyroidism with high frequency, associated with marked reductions in serum concentrations of thyroidstimulating hormone and thyroxine. During treatment, patients should be monitored for thyroid function and for hyper-triglyceridemia.159 Most patients will require concomitant treatment with a lipid-lowering agent and thyroxine.160,235 Gemfibrozil is contraindicated in this regard because it increases plasma concentrations of bexarotene, presumably due to inhibition of cytochrome P450 3A4, which, in turn, results in a paradoxical elevation of triglycerides.160,235 Extra-corporeal photophoresis The use of extra-corporeal photophoresis (ECP) was first reported in 1987 by Edelson and colleagues who showed a 73 per cent RR in patients with erythrodermic CTCL.234 Response rates were lower in earlier stages of disease (38 per cent). In this procedure, peripheral blood leukocytes are harvested, mixed with 8-MOP, exposed to UV radiation, and then returned to the patient.236 This is usually performed on two successive days every 4 weeks.237 The schedule is generally continued for up to 6 months in order to assess response: maintenance therapy is tailored according to disease course. In general, ECP is well tolerated, although patients with a history of heart disease require careful monitoring due to changing fluid volumes,236 and ECP is considered a first-line option for stage III erythrodermic disease and SS. PROGNOSIS
A patient’s life expectancy is not adversely affected in stage IA disease.160 Patients with stage IB and IIA and IIB disease have a 73–86 per cent, 49–73 per cent and 40–65 per cent overall 5-year survival respectively.160 Patients with erythrodermic stage III disease have a 5-year survival
862 Non-melanoma skin cancer
of 45–57 per cent, and patients with stage IVA and IVB have a 5-year survival of 15–40 per cent and 0–15 per cent respectively.160 Studies also suggest that stage 1A and 1B patients with thick plaques may have a worse prognosis but this depends on a histological assessment of plaque thickness, which can be difficult to reproduce. Similarly folliculotropic disease appears to have a poor prognosis compared with other patients with stage IB disease. The presence of a peripheral blood T-cell clone may indicate which patients with early-stage disease are likely to develop disease progression.159 Sézary syndrome patients, by definition, are staged as T4 N1–3 M0 B1 and have a poor prognosis with an overall median survival of 32 months from diagnosis.160 Recent studies of erythrodermic CTCL have shown that the presence of peripheral nodal disease is the most important prognostic factor in a multivariate analysis, although the degree of haematological involvement was also close to significance.160
strikingly epidermotropic proliferation of atypical T cells, which may have an aberrant CD4 CD8 phenotype. This localized type is also known as Woringer–Kolopp disease and the prognosis is excellent (Fig. 37.15). It should not be confused with the disseminated type known as Ketron–Goodman disease that has a worse prognosis and requires aggressive treatment. Extracutaneous dissemination or disease-related deaths have not been reported. The preferred treatment is radiotherapy or surgical excision. Granulomatous slack skin disease This is an extremely rare subtype characterized by the slow development of folds of lax skin in flexural sites, especially the axilla. The histology is characterized by a granulomatous infiltrate with elastolysis. Discrete clonal T-cell populations are present. In most patients it has an indolent course. Rapid recurrences after surgery are reported, and radiotherapy may be effective but experience is limited.
VARIANTS AND SUBTYPES OF MYCOSIS FUNGOIDES
There are a number of clinical variants of MF discussed below.172 Although there have been no specific therapeutic trials in these variants, these clinical variants appear to have a good prognosis and are often responsive to skin-directed therapies such as radiotherapy. The exception is folliculotropic MF, which may have a worse prognosis. Clinical variants such as bullous and hyper- or hypopigmented MF have a clinical behaviour similar to classical MF and are not considered separately. Folliculotropic mycosis fungoides Patients present with groups of follicular papules, indurated plaques and sometimes tumours, often associated with alopecia. Infiltrated plaques in the eyebrows with alopecia are a common finding. Pruritus can be severe. This variant is characterized by the presence of folliculotropic infiltrates, often sparing the epidermis, and preferential involvement of the head and neck area. Most cases show mucinous degeneration of the hair follicles (follicular mucinosis). The most relevant feature is the deep localization of the neoplastic infiltrate that makes it less accessible to skin-directed therapies. Folliculotropic MF is less responsive to PUVA and topical therapies. Radiotherapy is an effective treatment for individual lesions, and patients with extensive disease may benefit from TSEBT. In combination with retinoids or IFN-α, PUVA can also be useful. The prognosis is similar to that of classical tumour stage MF with a 5-year DS state of 70–80 per cent.
Sézary syndrome Sézary syndrome is defined clinically by the presence of erythroderma, generalized lymphadenopathy and peripheral blood neoplastic T cells (Sézary cells). It occurs exclusively in adults and, in its most advanced cases, may present with a severe infiltrated erythroderma with leonine facies (see Fig. 37.14) and hyperkeratosis, lichenification, oedema and fissuring of the palms and soles associated with severe pruritus. Lymphadenopathy, ectropion, alopecia and onychodystrophy are common. The histological features are similar to MF but the cellular infiltrate is often more monotonous and epidermotropism may be absent. Overall, only 30 per cent of skin biopsies are diagnostic while 30 per cent are consistent and 30 per cent are not diagnostic for lymphoma. Bone marrow may be involved but the infiltrates are often sparse
Pagetoid reticulosis This variant of MF is characterized by the presence of localized hyperkeratotic or scaly psoriasiform patches or plaques often affecting acral sites. The histology shows a
Figure 37.15 Woringer–Kolopp disease presenting as an isolated plaque on the hand of a 30-year-old man.
Primary cutaneous T-cell lymphoma 863
and interstitial. It has been proposed that the minimal criteria for diagnosis of SS should be erythroderma, compatible skin pathology, more than 5 per cent circulating atypical mononuclear cells, an aberrant T-cell phenotype and evidence of a peripheral blood T-cell clone detected by PCR for the T-cell receptor or cytogenetics.238 Sézary syndrome is an aggressive disease and the median survival is less than 3 years.160 Most patients die of opportunistic infections due to immunosuppression. The first-line treatment options include ECP, which has an overall response rate of 75 per cent239 and can be used alone or in combination with other modalities such as IFN-α and/or bexarotene. Chlorambucil plus prednisolone can be of benefit but CR is uncommon. Skin-directed therapy such as radiotherapy can be used palliatively for individual tumours or lymph node regions. Second-line treatment options include single-agent chemotherapy with pentostatin or oral methotrexate and clinical trials. Combination therapy regimens are currently being evaluated.
Adult T-cell leukaemia/lymphoma Adult T-cell leukaemia/lymphoma (ATLL) is a T-cell neoplasm aetiologically associated with human T-cell leukaemia virus 1 (HTLV1) infection, which can present in acute, leukaemic, lymphoma, chronic and ‘smouldering’ forms.240 Adult T-cell leukaemia/lymphoma is endemic in regions such as southwest Japan, the Caribbean, South America and Central Africa with a high prevalence of HTLV1. It develops in only 1–5 per cent of seropositive individuals after more than two decades of viral persistence. The acute forms present with leukaemia, lymphadenopathy, organomegaly, hypercalcaemia and skin lesions in about 50 per cent of cases. The skin lesions can be nodules, tumours, generalized papules or plaques. Chronic and smouldering variants frequently present with cutaneous involvement alone and may resemble MF with few if any circulating neoplastic T cells. The histological picture may be indistinguishable from MF, but clonally integrated HTLV1 proviral DNA can be found in all cases. The clinical subtype is the main prognostic factor. The acute and lymphomatous types are aggressive with poor survival and initial therapy consists of antivirals, interferon and multi-agent chemotherapy. The chronic and smouldering forms have a less aggressive course and longer survival but may transform to an acute subtype. In the chronic and smouldering forms, skin-directed therapies such as those used in MF may be used.172
Primary cutaneous CD30 lymphoproliferative disorders This is the second most common group, accounting for almost 30 per cent of CTCL cases. It includes primary
cutaneous CD30anaplastic large cell lymphoma (C-ALCL), lymphomatoid papulosis (LyP) and borderline cases. It is a spectrum of disease and the clinical appearance and course are used to differentiate between primary C-ALCL and LyP. PRIMARY CUTANEOUS CD30 ANAPLASTIC LARGE CELL LYMPHOMA
Primary cutaneous CD30 anaplastic large cell lymphoma is a primary cutaneous large cell lymphoma that presents with no previous history of MF or other CTCL variant. It mainly affects adults and presents as a solitary tumour or localized nodules and plaques that may show ulceration. Twenty per cent of patients have multifocal lesions. In some patients, C-ALCL may show partial spontaneous regression. Frequently, relapses occur in the skin but extra-cutaneous dissemination is rare and only 10 per cent of cases develop regional lymph node disease. The histology shows cohesive sheets of large tumour cells with a characteristic anaplastic morphology expressing CD30 in over 75 per cent of the tumour cells. Unlike systemic nodal CD30 lymphomas, C-ALCL expresses the cutaneous lymphocyte antigen (CLA) but does not express EMA or ALK1 (anaplastic lymphoma kinase), which is due to the (2;5) translocation characteristic of systemic nodal ALCL. Unlike Hodgkin’s lymphoma, staining for CD15 is generally negative. Rare cases express CD56 but this does not appear to be associated with an unfavourable prognosis. The prognosis is excellent with 10-year disease-specific survival exceeding 90 per cent. Aggressive loco-regional disease may indicate a poor prognosis but this requires confirmation. The treatment of choice for localized tumours is surgical excision or RT. Multifocal skin lesions can be treated with RT if they can be encompassed by an RT field, otherwise low-dose methotrexate is an option. The RT dose and schedule should be a fractionated course, such as 30–36 Gy in 18 fractions, to limit the long-term side effects and minimize the chance of recurrence. The rare patient who develops extracutaneous disease, or rapidly progressive disease should be treated with combination chemotherapy such as CHOP.
LYMPHOMATOID PAPULOSIS
Lymphomatoid papulosis (LyP) is defined as a chronic, recurrent, self healing papulonecrotic or papulonodular skin eruption with histological features suggestive of a CD30 lymphoma. It generally occurs in adults but may be seen in children. Individual lesions disappear within 3–12 weeks to leave superficial varioliform scars. Up to 20 per cent of cases are associated with another type of cutaneous lymphoma such as MF, primary cutaneous CD30 ALCL or Hodgkin’s lymphoma. Lymphomatoid papulosis has an excellent prognosis, and in a study of 118 patients only five
864 Non-melanoma skin cancer
developed a systemic lymphoma, and only two died of systemic disease over 77 months’ follow-up.241 No treatment affects the course of the disease and, therefore, the shortterm merits of treatment must be weighed against the potential side effects. Low-dose oral methotrexate is effective, and both PUVA and topical chemotherapy can be beneficial, but the disease recurs on discontinuation of treatment. Superficial radiotherapy is appropriate for localized or extensive regional disease.
Subcutaneous panniculitis-like T-cell lymphoma Subcutaneous panniculitis-like T-cell lymphoma (SPTL) is a cytotoxic T-cell lymphoma that is clinically characterized by deep indurated erythematous plaques often on the limbs, which may mimic cellulitis. There are two groups of SPTL with different histology, phenotype and prognosis. Cases with an α/β T-cell phenotype are usually CD8, restricted to subcutaneous tissue with no dermal or epidermal involvement and run an indolent clinical course. In contrast in cases with a γ/δ T-cell phenotype, which are often CD4 and CD8 but occasionally CD56, the neoplastic infiltrates may involve the epidermis and dermis, and patients invariably have a poor prognosis. Cases of α/β SPTL have a 5-year survival of 80 per cent.172 Treatment of aggressive cases involves CHOP chemotherapy and radiotherapy, but recent studies suggest patients with indolent SPTL can be controlled for long periods on systemic corticosteroids.172
Primary cutaneous peripheral T-cell lymphoma – unspecified This is a heterogeneous group, which includes all CTCL variants that do not fit into any of the other well-defined categories.175 Provisional entities such as primary cutaneous aggressive CD8 cytotoxic T-cell lymphoma (Berti’s lymphoma) and cutaneous γ/δ T-cell lymphoma both have an aggressive course and multi-agent systemic chemotherapy is used but the results are disappointing. Primary cutaneous small/medium pleomorphic CD4 CTCL is a separate provisional category, and presents with solitary lesions that may be amenable to surgical excision or radiotherapy, but the optimum treatment remains to be defined. In all cases, MF must be excluded by careful clinical examination and an accurate history to identify typical subtle and polymorphic patches and plaques of MF.
PRIMARY CUTANEOUS B-CELL LYMPHOMA Primary cutaneous B-cell lymphomas are defined as B-cell lymphomas originating in the skin without evidence of
extracutaneous disease at presentation and for 6 months after diagnosis as assessed by adequate staging procedures. They represent only 25 per cent of all cutaneous lymphomas. In recent years there has been considerable debate over the difference between the previous EORTC and WHO classifications of primary cutaneous B-cell lymphoma.172 Under the previous EORTC classification, primary cutaneous follicle centre cell lymphoma (PCFCCL) defined a group of cutaneous lymphomas highly responsive to radiotherapy with an excellent prognosis. The same type of lymphoma with a diffuse growth pattern of large cells was classified under the previous WHO classification as a diffuse large B-cell lymphoma, leading to over treatment with multi-agent chemotherapy. The other main group debated was primary cutaneous large B-cell lymphoma of the leg (PCLBCL-leg) that was recognized under the EORTC classification as a separate subgroup reflecting its somewhat different histology and more unfavourable prognosis. Recent clinicopathological and genetic studies support the contention that PCFCCL and PCLBCL-leg are distinct groups, and a consensus was agreed in the new WHO-EORTC classification with PCFCCL being defined as primary cutaneous follicle centre lymphoma distinct from nodal follicular lymphoma and PCLBCL-leg as primary cutaneous diffuse large B-cell lymphoma, leg type (See Tables 37.2 and 37.3). All patients with PCBCL should be staged with a CT scan of the neck, chest, abdomen and pelvis and a bone marrow aspirate and trephine to exclude systemic disease. It is likely that PET/CT will supersede CT alone for staging in the future.
Primary cutaneous marginal zone B-cell lymphoma Primary cutaneous marginal zone B-cell lymphoma (PCMZL) is considered part of the broad spectrum of extranodal marginal zone B-cell lymphomas commonly involving mucosal sites, so-called MALT (mucosal associated lymphoid tissue). It is an indolent lymphoma composed of small cells including marginal zone cells, lymphoplasmacytoid cells and plasma cells. It includes cases previously designated primary cutaneous immunocytoma and exceptional cases of primary cutaneous plasmacytoma without multiple myeloma. Most cases present with red to violaceous papules, plaques or nodules preferentially on the trunk and extremities.173 Multi-focal lesions are frequent, ulceration is uncommon and extracutaneous dissemination is very rare. Occasional spontaneous remission is observed and the development of anetoderma in spontaneously resolving lesions is reported. Associated autoimmune disease is rare in PCMZL and suggests systemic lymphoma. The prognosis is excellent with a 5-year survival of 99 per cent.173 Patients with solitary lesions can be treated by surgical excision or radiotherapy. Retrospective series
Extra-mammary Paget’s disease 865
have shown that radiotherapy schedules delivering a total dose from 20 to 54 Gy, and radiation portals with a margin of at least 2.0–3.0 cm are successful with a low risk of recurrence.242–244 The recurrence rate is higher with smaller margins.245 The St John’s Institute of Dermatology schedule is 15 Gy 5 fractions over 1 week. An association with B. burgdorferi infection has been reported from some, but not all, geographical areas, and systemic antibiotics should be used in those patients with serological evidence of B. burgdorferi infection. For patients with multi-focal skin lesions chlorambucil, IFN-α and rituximab have all produced very good results.172 An expectant strategy similar to that used in other indolent B-cell lymphomas can be adopted.
Primary cutaneous follicle centre lymphoma Primary cutaneous follicle centre lymphoma (PCFCL) is derived from follicle centre cells, with a follicular, follicular and diffuse or diffuse growth pattern. The infiltrate consists of cleaved centrocyte-like cells with only a few large centroblasts. It must be differentiated from lymphomas with a diffuse growth pattern and monotonous proliferation that are classified as PCLBCL-leg or other. Immunophenotype studies can help in the diagnosis. Primary cutaneous follicle centre lymphoma consistently expresses Bcl6, but, unlike nodal and secondary cutaneous follicular lymphomas, PCFCL rarely express Bcl2 and does not show the t(14:18) characteristically found in systemic follicular lymphomas.172 It characteristically presents as a solitary or group of plaques and tumours, usually on the scalp, forehead or trunk and rarely on the legs. The tumours may be surrounded by erythematous papules and indurated plaques that precede the development of the tumour. Multi-focal presentation is seen in a minority and not associated with a worse prognosis. If untreated the skin lesions gradually increase in size but dissemination is uncommon.172 The prognosis is excellent with a 5-year survival of over 95 per cent.172 Radiotherapy is the preferred mode of treatment and is the same as that used for PCMZL detailed above. Cutaneous relapses occur in and out of the previous treatment fields in approximately 20 per cent of patients, and respond to treatment or retreatment with radiotherapy. Chemotherapy is occasionally needed for patients with extensive cutaneous disease or extracutaneous disease, and the regimens used are similar to that for systemic B-cell lymphoma.
Primary cutaneous diffuse large B-cell lymphoma, leg type Primary cutaneous diffuse large B-cell lymphoma, leg type is a lymphoma characteristically presenting with skin tumours
on the lower leg, but can arise at other sites. It mainly affects elderly patients, particularly females. It presents as a rapidly growing large red or bluish tumour on the lower leg, and dissemination to extracutaneous sites is common reflecting the more unfavourable prognosis when compared with PCFCL.172 The histology shows a diffuse infiltrate of large centroblasts often extending to subcutaneous tissue with a prominent stromal reaction. The neoplastic cells express CD20 and, in contrast to PCFCL, shows strong Bcl2 expression and express Mum1. The t(14:18) is not found but there is evidence for chromosomal amplification of the BCL2 gene locus. The 5-year survival is 55 per cent in the Dutch and Austrian registries (see Table 37.3). The presence of multiple skin lesions is an adverse factor. The treatment is that for systemic diffuse large B-cell lymphoma with anthracyclinecontaining combination chemotherapy such as CHOP plus rituximab, followed by involved field radiotherapy to the leg, but very elderly patients may require palliative radiotherapy only.172
Primary cutaneous diffuse large B-cell lymphoma – other This rare group includes morphological variants of primary cutaneous diffuse large B-cell lymphoma, anaplastic or plasmablastic subtypes, or T-cell rich/histiocyte-rich large B-cell lymphomas, but prognosis and treatment is similar to that for PCLBCL.172
Primary cutaneous intravascular large B-cell lymphoma This is a rare but well-defined subtype that preferentially affects the CNS, lungs and skin and is generally associated with a poor prognosis. Cases with only skin involvement have a better survival that other presentations (3-year overall survival: 56 vs. 22 per cent). Multi-agent chemotherapy is the preferred treatment, even for disease limited to the skin.172
OTHER NON-MELANOMA SKIN CANCERS EXTRA-MAMMARY PAGET’S DISEASE Sir James Paget first described the intra-epithelial carcinoma, which now bears his name, in the breast in 1874 and suggested that the disease might also be found in other areas.246 The first case of extra-mammary Paget’s disease (EMPD) was described by Crocker in 1888247 and peri-anal Paget’s disease (PPD) was first reported in 1893.248
866 Non-melanoma skin cancer
Incidence and aetiology Extra-mammary Paget’s disease is a rare intra-epithelial adenocarcinoma, characterized by the presence of malignant Paget cells lying within the epidermis of the skin. The aetiology is unknown. It is usually a primary tumour but may be associated with an underlying adjacent or distant invasive carcinoma. Approximately 25 per cent of cases have an underlying cutaneous adnexal carcinoma and 10–15 per cent have an subjacent or distant internal carcinoma.249 In the later stages, the epidermal pagetoid cells are thought to represent metastases and the prognosis is associated with the underlying malignancy with a mortality rate of 50 per cent or higher.
Radiotherapy is an effective treatment but has mainly been reserved for patients unfit for surgery, and for those with recurrent disease or those who decline surgery.249 A minimum dose of 50 Gy using a single appositional field is recommended.249,253
MERKEL CELL CARCINOMA Merkel cell carcinoma (MCC) is tumour of neuroendocrine cell origin. Studies have shown that MCC shares pathogenetic mechanisms with other neoplasms of neural crest derivation, such as melanoma and neuroblastoma.254
Incidence Clinical features Extra mammary Paget’s disease begins as an erythematous eczematous, slowly spreading plaque, which can bleed easily and may ulcerate. It usually affects sites with a high density of apocrine glands (e.g., vulval, perineal and peri-anal regions). The penis, axillae, umbilicus, eyelids, and external auditory meatus are less common sites. Investigations should always include a diagnostic biopsy and exclusion of an associated underlying carcinoma.
Pathology Within the epidermis, characteristic Paget cells are dispersed between keratinocytes. These cells have clear abundant cytoplasm and do not establish intercellular bridges with the adjacent keratinocytes. In EMPD, there is a tendency for the cells to accumulate in the basal area, especially in the interpapillary ridges. They stain positively for acid as well as neutral mucopolysaccharides and carcinoembryonic antigen.
Treatment Surgery is the treatment of choice if possible.249,250 Wide local excision is required, and extensive disease requires plastic surgical reconstruction and, in the peri-anal area, abdominoperineal resection may be required. Local recurrence is a major problem because of the difficulty in accurately identifying the extent of occult disease in a multi-centric process within the skin, as well as accurately identifying spread into the anal canal. Local recurrence rates of 31–61 per cent for surgery have been reported.251 Mohs’ micrographic surgery has lower recurrence rates of 23–33 per cent.251 Topical fluorouracil prior to and in addition to surgery has been advocated. Imiquimod has been used in a similar way as monotherapy and in combination with surgery to try to reduce the extent of excision.252
It is a rare tumour with 470 new cases in the USA each year, and most often occurs in Caucasian men between the seventh and ninth decade of life.255
Aetiology Merkel cells are specialized sensory cells present in the basal or supra-basal layers of the epidermis. They are believed to be neurosensory cells that are derived from the amine precursor uptake and decarboxylation system (APUD). They function as slowly adapting type 1 mechanoreceptors, with a higher density in sun exposed areas. Forty-seven per cent of cases occur on the head and neck, 40 per cent on the extremities and 8 per cent on the trunk.256 Risk factors include UV light and immunosuppression.257–260 Rapid progression of MCC occurs with immunosuppressive therapy after organ transplantation.261 A similar pattern of genomic changes has been detected in MCC and neuroendocrine tumours such as melanoma.254 In metastatic MCC Bcl2 is strongly expressed. In a SCID mouse xeno-transplantation model for human MCC, administration of BCL2 antisense oligonucleotide results in either a dramatic reduction of tumour growth or complete remission.262
Clinical features Merkel cell carcinomas present as solitary pigmented or nonpigmented nodular tumours on the head and neck regions, but may be seen on the trunk and extremities. Multiple tumours at presentation have been reported. Lymph node areas must be examined and staged with a CT scan of the neck, chest, abdomen and pelvis. Positron emission tomography has been shown to be useful in staging MCC and also for assessing response to treatment; it can however be negative in MCC with low proliferative rates.263 A common clinical staging system is shown in Box 37.2.
Langerhans cell histiocytosis 867
Box 37.2 Common staging system based upon clinical presentation264 Stage IA: Primary tumour 2 cm with no evidence of lymph nodes or distant spread Stage IB: Primary tumour 2 cm with no evidence of lymph nodes or distant spread Stage II: Regional node involvement but no distant spread Stage III: Systemic metastases beyond regional lymph nodes
Pathology Merkel cell carcinomas consist of dermal nodules with a clear Grenz zone separating the tumour from the epidermis, and sheets of small, undifferentiated and tightly packed rounded cells that possess only scanty cytoplasm. It may be histologically difficult to differentiate from other small cell neoplasms. Tumour cells stain positive for neurofilaments, cytokeratin (CAM5.4), neuron-specific enolase, and epithelial membrane antigen. Leukocyte common antigen, S100, and chromogranin are negative. Karyotyping of neoplastic cells shows loss of chromosome Y, the significance of which is unknown.265
Treatment Aggressive therapy is needed as MCC has a high propensity for local recurrence (20–75 per cent), regional node metastases (31–80 per cent) and distant metastases (26–75 per cent), and approximately one third of patients eventually die of the disease.256,261 Treatment for localized disease without nodal involvement is surgical excision of the primary tumour with an adequate margin of clearance of 2.5–3.0 cm. Elective lymphnode dissection improves the length of the disease-free period but does not improve overall survival.266 Recurrence rates with surgery alone are reported to be between 8 per cent and 100 per cent. Post-excision radiotherapy to the site of the original tumour and the primary draining lymph nodes reduce recurrence rates and in some series improves disease-free survival.256,267 Fractionated doses of 50–60 Gy given over 4–6 weeks are recommended.268 A study of 86 patients in Sydney showed that the combination of surgery and RT improved the median disease-free survival compared with surgery alone.269 In contrast, a study from the Memorial Sloane Kettering of 251 patients showed very low recurrence rates of 8 per cent after nodenegative excision and 8–15 per cent after operative nodal staging but did not find a benefit for adjuvant radiotherapy.270 However, only a minority of patients (17 per cent) received RT in the latter study. It is clear that complete surgical excision with wide clearance of the primary and complete lymph node dissection for local or loco-regional disease is needed and sentinel lymph node biopsy may be
of benefit for staging patients.271 Therefore, adjuvant RT should be considered in all stages. In early stage IA disease with a low risk of recurrence after complete resection with negative staging and no lymph node disease, RT may be omitted after discussion of the risks with the patient. Chemo-radiotherapy has been investigated in a phase II trial undertaken by the Trans Tasman Radiation Oncology Group (TROG).272 Patients received concurrent and adjuvant combination chemotherapy (etoposide/carboplatin) and radiotherapy (50 Gy). Eligible patients needed to have one or more unfavourable features (10 mm primary, nodal disease, residual disease post surgery, surgical recurrence, or recurrence outside an irradiated field). The study reported an impressive 3-year overall survival, locoregional control and distant control rates of 76 per cent, 75 per cent, and 76 per cent, respectively. Only 17 per cent experienced loco-regional recurrence. These findings in patients with poor prognostic features, particularly the presence of nodal metastases in 33 (62 per cent stage III) patients, strongly suggest a benefit for the addition of combination chemotherapy in patients with unfavourable features. Merkel cell carcinomas are chemosensitive but rarely curable in patients with metastasis or locally advanced tumours. A high incidence of mortality directly related to chemotherapy has been reported.273 However, orally administered etoposide has been used with great success in advanced MCC, even in patients previously treated intravenously with the same drug.274
LANGERHANS CELL HISTIOCYTOSIS Langerhans cell histiocytosis (previously known as histiocytosis X, eosinophilic granuloma, Hand–Schiller–Christian disease or Letterer–Siwe disease) is characterized by a neoplastic proliferation of epidermal Langerhans cells in various organs with damage to the respective organ and, subsequently, dysfunction. Although most patients with single system disease do well without the need for treatment, patients with multi-system disease may develop fulminant disease, which may be fatal.
Incidence Langerhans cell histiocytosis is an uncommon disease, with an estimated 50 new cases a year in England and Wales.275 In childhood disease there is a male-to-female ratio of 2:1, but in adults the sex incidence is equal.
Aetiology The cause of LCH is unknown. Several theories including viral, immune defects, dysregulation of cytokine gene and
868 Non-melanoma skin cancer
oncogene expression have been suggested. The findings of Birbeck granules in the cytoplasm of cells involved in LCH under transmission electron microscopy276 and the presence of a membrane bound glycoprotein, CD1a complex,277,278 confirm an origin from epidermal Langerhans cells. It appears that clonal expansion of Langerhans cells is the first biological event in LCH, with further biological events determining the course of the disease.
Clinical features The natural history of LCH is extremely variable. In singlesystem disease, particularly of bone, LCH may be an incidental radiological finding that resolves without causing any disability to the patient. At the other extreme, multi-system disease presenting in early infancy may rapidly progress to death, whilst intermediate cases may continue to progress over decades with substantial morbidity or may rapidly burn themselves out without residual defect. The number of systems involved by LCH is the most reliable prognostic factor. When either one or two systems are involved, mortality is rare; when three or more systems are involved, mortality rates rise to 30 per cent or higher. Other poor prognostic factors include early age of disease onset before 2 years of age, internal organ involvement and male gender. The most aggressive form of LCH presents in the first few months of life with multi-system involvement and behaves like an aggressive malignancy; prognosis is extremely poor.
Pathology Langerhans cell histiocytosis consists of a proliferation of large cells with indented reniform nuclei, which are positive for non-specific esterase, surface ATP-ase and α-D-mannosidase, and negative for alkaline phosphatase. The cells characteristically express the S100 protein and produce dense cell surface and paranuclear staining with peanut agglutinin. Immunohistochemistry shows that the LCH cells are CD1a and CD4 positive279 and exhibit the phenotype of a Langerhans cell that has been arrested at an early stage of maturation.280 On electron microscopy Birbeck granules identical to those described in Langerhans cells are seen. There is no histological difference between lesions in multi-system compared with localized disease. Organs commonly affected by LCH, in descending order of frequency, are: the skin (including external ear and gum), cortical bone, lymph nodes, central nervous system, liver/spleen, bone marrow, gastrointestinal tract and thymus. Single cases of LCH involvement of the eye and pancreas have been reported. Organs not involved include the heart and kidneys. Langerhans cell histiocytosis has protean manifestations and patients may present to many different medical
Figure 37.16 Langerhans cell histiocytosis (LCH) of the skin in a 13-month-old boy showing typical petechial inter-trigenous involvement.
and surgical specialities. Lesions in bone may be asymptomatic or cause pathological feature and are seen as lucent areas on radiography. Skin rashes may be initially mistaken for seborrhoeic dermatitis, particularly in the scalp and napkin areas (Fig. 37.16), Pituitary involvement initially causes diabetes insipidus but may progress to anterior pituitary failure. Lung involvement produces diffuse upper zone shadowing on X-ray. The liver and spleen may be involved, causing organ enlargement, and bone marrow infiltration may produce anaemia, leukopenia or thrombocytopenia. Intermittent pyrexia may occur in patients and may be due to increased levels of interleukin-1 (IL-1) and tumour necrosis factor-α (TNF-α) production rather than secondary infection. A diagnosis of LCH can only be made on tissue biopsy of any clinically suspicious lesions. The total number of systems involved, age of disease onset and the presence of ‘vital organ’ involvement are the important prognostic factors in LCH. The Lahey scoring system279 where one point is allocated for each organ system involved by clinical assessment or simple laboratory investigation, is still the most widely used (Box 37.3). A CD1a radiolabelled technique has been evaluated as an adjunct to staging and diagnosis.
Langerhans cell histiocytosis 869
Box 37.3 Lahey scoring system for Langerhans cell histiocytosis (LCH) One point is awarded for evidence of involvement of the following: ● Skeleton ● Skin ● Liver ● Spleen ● Lung ● Pituitary ● Haemoglogin (Hb) 10.5 g/dL ● White blood cells (WBC) 3000/mm3 or 14 000/mm3 ● Platelets 2 000 000/mm3
More recently the Histiocyte Society has simplified the staging of patients with LCH to three categories: ● ● ●
Single-system disease Multi-system disease Multi-system disease with evidence of organ dysfunction.
reducing to 1mg/kg for the second and tailing off over the third. Although this is effective in children, adults with LCH do not respond as well to systemic steroids and frequently develop side effects from the treatment. Cytotoxic drugs carry the risks of myelosuppression, immunosuppression and organ-specific toxicity. The vinca alkaloids, methotrexate and mercaptopurine are effective agents but the epidophyllotoxin, etoposide (VPI6) may be the most effective single agent. Some patients who are resistant to other forms of chemotherapy have responded to etoposide whilst others, with steroid-resistant disease, have become steroid sensitive after etoposide (2–6 courses of 150 mg/m2 intravenously for 3 days or 300 mg/m2 orally for 3 days).284 However, some patients relapse after an initial successful response to etoposide. Maintenance treatment with azathioprine or mercaptopurine with or without weekly methotrexate has proved very effective in maintaining disease remission and, after 6–12 months, treatment can be stopped without recurrence of disease. A study of 17 patients treated with injections of calf thymic extract reported responses at least as good as that of historical controls treated with more conventional chemotherapy,285 but subsequent clinical trials have failed to substantiate these findings. Immunotherapy has no proven value. As the disease is uncommon and the manifestations variable, careful multicentre trials are needed to assess alternative approaches to treatment.
Organ dysfunction is taken as abnormal liver function tests, abnormal blood film, failure to thrive or abnormal lung function tests. Prognosis decreases with more advanced disease.
KEY LEARNING POINTS Treatment In each patient, the likely natural history of the disease should be assessed so the benefits of treatment can be weighed against the risks. For treatment purposes, LCH can be divided into single- and multi-system disease. In single-system disease, there is an excellent prognosis so local measures only are indicated. Intra-lesional steroids279 are effective in bone disease. In isolated skin disease, topical nitrogen mustard or PUVA are the treatments of choice.282 Steroid injection into bony lesions may be useful and low-dose RT to the pituitary can reverse early diabetes insipidus.283 This study was, however, based on clinical rather than laboratory findings and children with LCH have been observed to have episodes of polyuria and polydipsia without biochemical evidence of diabetes insipidus.284 There is much controversy over how aggressive chemotherapy should be. There have been no trials that show combination chemotherapy to be more effective than single-agent treatment, but toxicity is certainly greater. A recent study, however, suggests that the incidence of diabetes insipidus may be reduced by combination treatment. A conservative regimen275 involves the use of prednisolone for 2–3 months, starting at 2 mg/kg for the first month then
●
●
●
●
●
Skin cancer is the commonest malignancy presenting to the physician in the white population. The most important environmental agent capable of inducing skin cancer is exposure to ultraviolet light. Selection of the treatment modality for non-melanoma skin cancers (NMSC) depends on a number of different factors, such as health status, site, size, histological type and whether one is dealing with a primary or recurrent lesion. Recurrent and high risk NMSC should be managed by a multidisciplinary team to select the best treatment option for the patient. Primary cutaneous lymphoma is classified according to the WHO-EORTC (2005) classification. Two thirds of cases are primary cutaneous T-cell, and one quarter are primary cutaneous B-cell lymphomas. Primary cutaneous T-cell lymphoma treatment needs to be delivered by an experienced multidisciplinary team, with quality of life central to the therapeutic strategy.
870 Non-melanoma skin cancer
KEY REFERENCES Ziegler A, Jonason AS, Leffell DJ, et al. Sunburn and p53 in the onset of skin cancer. Nature 1994; 372: 773–6. Telfer NR, Colver G, Bowers PW. Guidelines for the management of basal cell carcinoma. Br J Dermatol 1999; 141:415–23. Motley R, Kersey P, Lawrence C, et al. Multiprofessional guidelines for the management of the patient with primary cutaneous squamous cell carcinoma. Br J Dermatol 2002; 146:18–25. Willemze R, Jaffe ES, Burg G, et al. WHO-EORTC classification for cutaneous lymphomas. Blood 2005; 105:3768–85. Whittaker SJ, Marsden JR, Spittle M, et al. Joint British Association of Dermatologists and UK Cutaneous Lymphoma Group guidelines for the management of primary cutaneous T-cell lymphomas. Br J Dermatol 2003; 149:1095–107.
REFERENCES 1 Cancer Research UK, CancerStats: Incidence UK. 2004 www.cancerresearchuk.org/cancerstats 2 LeBoit P, Burg G, Weedon D, Sarasin A (eds) WHO classification of tumours: pathology and genetics of skin tumours. Lyon: IARC press, 2006. 3 Knudson A. Hereditary cancer, oncogenes and anti oncogenes. Cancer Res 1985; 45:1437–43. 4 Giles GG, Mark R, Foley P. Incidence of non-melanocytic skin cancer treated in Australia. Br Med J 1988; 296:13–17. 5 Scotto J, Cotton G, Urbach F, Berger D, Fears T. Biologically effective ultraviolet radiation: surface measurements in the United States, 1974 to 1985. Science 1988; 239:762–4 6 Ziegler A, Jonason AS, Leffell DJ, et al. Sunburn and p53 in the onset of skin cancer. Nature 1994; 372:773–6. 7 Jensen P, Hansen S, Moller B, et al. Skin cancer in kidney and heart transplant recipients and different long term immunosuppressive therapy regimens. Am Acad Dermatol 1999; 40:177–86. 8 Pieceall WE, Goldberg LH, Ananthaswarry HN. Presence of human papilloma virus type 16 DNA sequences in human non-melanoma skin cancers. Invest Dermatol 1991; 97:880–4. 9 Shamanin V, Zur Hausen H, Lavergne D, et al. Human papillomavirus infections in non-melanoma skin cancers from renal transplant recipients and non-immunosuppressed patients. Nat Cancer Inst 1996; 88:802–11. 10 Maurer TA, Christian KV, Kerschmann RL, et al. Cutaneous squamous cell carcinoma in human immunodeficiency virus-infected patients. A study of epidemiologic risk factors, human papillomavirus, and p53 expression. Arch Dermatol 1997; 133:577–83. 11 Ostrow RR, Manias D, Mitchell AJ, et al. Epidermodysplasia verruciformis. A case associated with primary lymphatic dysplasia, depressed cell-mediated immunity and Bowen’s disease containing human papillomavirus 16 DNA. Arch Dermatol 1987; 123:1511–16. 12 Pfister H. Human papilloma virus and impaired immunity vs. epidermodysplasia verruciform. Arch Dermatol 1987; 123:1469–70.
13 Yutsudo M, Tangigaki T, Kanda R, et al. Involvement of human papilloma virus type 20 in epidermodysplasia verruciformis skin carcinogenesis. J Clin Microbiol 1994; 32:1076–8. 14 Karagas MR, Nelson HH, Sehr P, et al. Human papillomavirus infection and incidence of squamous cell and basal cell carcinomas of the skin. J Natl Cancer Inst 2006; 98:389–95. 15 Wicking C, Shanley S, Smyth I. Most germ-line mutations in the nevoid basal cell carcinoma syndrome lead to a premature termination of the PATCHED protein, and no genotypephenotype correlations are evident. Am J Hum Genet 1997; 60:21–6. 16 Levanat S, Mubrin MK, Crnic I, et al. Variable expression of Gorlin syndrome may reflect complexity of the signalling pathway. Pflugers Arch 2000; 439:31–3. 17 Cohen MM Jr. Nevoid basal cell carcinoma syndrome: molecular biology and new hypotheses. Int Oral Maxillofac Surg 1999; 28:216–23. 18 Lindeberg H, Jepsen FL. The nevoid basal cell carcinoma syndrome. Histopathology of the basal cell tumors. J Cutan Pathol 1983; 10:68–72. 19 Lo Muzio L, Staibano S, Pannone G, et al. Expression of cell cycle and apoptosis-related proteins in sporadic odontogenic keratocysts and odontogenic keratocysts associated with the nevoid basal cell carcinoma syndrome. J Dent Res 1999; 78:1345–53. 20 Barr RJ, Headley JL, Jensen JL, Howell JB. Cutaneous keratocysts of nevoid basal cell carcinoma syndrome. J Am Acad Dermatol 1986; 14:572–6. 21 Wolter M, Reifenberger J, Sommer C, et al. Mutations in the human homologue of the Drosophila segment polarity gene patched (PTCH) in sporadic basal cell carcinomas of the skin and primitive neuroectodermal tumors of the central nervous system. Cancer Res 1997; 57:2581–5. 22 Kimonis VE, Goldstein AM, Pastakia B, et al. Clinical manifestations in 105 persons with nevoid basal cell carcinoma syndrome. Am J Med Genet 1997; 31:299–308. 23 Lindeberg H, Halaburt H, Larsen PO. The naevoid basal cell carcinoma syndrome. Clinical, biochemical and radiological aspects. J Maxillofa Surg 1982; 10:246–9. 24 Mitchell G, Farndon PA, Braydon P, Murcky PA, Eeles RA. Genetic predisposition to cancer: The consequences of delayed diagnosis of Gorlin syndrome. Clinical Oncology (2005) 17:650–54 25 Nagasawa H, Little FF, Burke MJ, et al. Study of basal cell nevus syndrome fibroblasts after treatment with DNA-damaging agents. Basic Life Sci 1984; 29:775–85. 26 Chan GL, Little JB. Cultured diploid fibroblasts from patients with the nevoid basal cell carcinoma syndrome are hypersensitive to killing by ionizing radiation. Am J Pathol 1983; 111:50–55. 27 Newton JA, Black AK, Arlett CF, et al. Radiobiological studies in the naevoid basal cell carcinoma syndrome. Br J Dermatol 1990; 123:573–80. 28 Featherstone T, Taylor AM, Harnden DG. Studies on the radiosensitivity of cells from patients with basal cell naevus syndrome. Am J Hum Genet 1983; 35:58–66.
References 871
29 Arlett CF, Priestley A. Deficient recovery from potentially lethal damage in some gamma-irradiated human fibroblast cell strains. Br J Cancer Suppl 1984; 6:227–32. 30 Little JB, Nichols WW, Troilo P, et al. Radiation sensitivity of cell strains from families with genetic disorders predisposing to radiation induced cancer. Cancer Res 1989; 49:4705–14. 31 Craven NM, Griffiths CEM. Retinoids in the management of non-melanoma skin cancer and melanoma. Skin Cancer 1996; 26:275–6. 32 Theiler R, Hubscher E, Wagenhauser FJ, Panizzon R, Michel B. Diffuse idiopathic skeletal hyperostosis (DISH) and pseudocoxarthritis following long-term etretinate therapy. Schweiz Med Wochenschr 1993; 123:649–53. 33 Strange PR, Lang PG Jr. long-term management of basal cell nevus syndrome with topical tretinoin and 5-fluorouracil. J Am Acad Dermatol 1992; 27:842–5. 34 Jacyk WK. Xeroderma pigmentosum in black South Africans. Int J Dermatol 1999; 39:511–14. 35 Khatri ML, Bemghazil M, Shafi M, et al. Xeroderma pigmentosum in Libya. Int J Dermatol 1999; 38:520–4. 36 Clever J. Defective repair replication of DNA in xeroderma pigmentosum. Nature 1968; 218:652–6. 37 Reed W, Landing B, Sugarman G, et al. Xeroderma pigmentosum: clinical and laboratory investigations of its basic defect. J Am Med Assoc 1969; 207:2073–9. 38 de Boer J, Hoeijmakers JH. Nucleotide excision repair and the human syndrome. Carcinogenesis 2000; 21:453–60. 39 Lambert WC, Kuo HR, Lambert MW. Xeroderma pigmentosum. In Chu AC, Edelson, RL (eds) Malignant tumours of the skin. London: Arnold, 1999, 119–37. 40 Guillot B, Favoer C, Guilhou JJ, et al. Xeroderma pigmentosum. Un case traite par I’association betacarotene – canthaxanthine et retinoide aromatique. Arch Dermatol Venereol 1984; 111:65–6. 41 Somos S, Farkas B, Schneider I. Cancer protection in xeroderma pigmentosum variant (XP-V). Anticancer Res 1999; 3:2195–9. 42 Mittelbronn MA, Mullins DL, Ramos-Caro FA, et al. Frequency of pre-existing actinic keratosis in cutaneous squamous cell carcinoma. Int J Dermatol 1998; 37:677–81. 43 Montgomery H, Dorffel J. Verruca senilis und keratoma senile. Arch Dermatol Syph 1932; 166:286–9. 44 Kao GF. Carcinoma arising in Bowen’s disease. Arch Dermatol 1986; 122:1124. 45 Mikhail GR. Cancers, precancers, and pseudo cancers on the male genitalia. J Dermatol Surg Oncol 1980; 6:1027. 46 Dinehart SM. The treatment of actinic keratoses. J Am Acad Dermatol 2000; 42:25–8. 47 Mackenzie-Wood A, de Kossard S, Launey J, Wilkinson B, Owens MI. Imiquimod 5% cream in the treatment of Bowen’s disease. J Am Acad Dermatol 2001; 44:462–70. 48 Lukas Van der Spek LA, Pond GR, Wells W, Tsang RW. Radiation therapy for Bowen’s disease of the skin. Int J Radiat Oncol Biol Phys. 2005; 63:505–10. 49 Milne JA. An introduction to the diagnostic histopathology of the skin. London: Edward Arnold, 1972, 261–2.
50 Jih DM, Lyle S, Elenitsan R, et al. Cytokine 15 expression in trichoepitheliomas and a subset of basal cell carcinomas suggests they originate from hair follicle stem cells. J Cutan Pathol 1999; 26:113–18. 51 Kruger K, Blume-Peytavi U, Orfanos CE. Basal cell carcinoma possibly originates from the outer root sheath and/or the bulge region of the vellus hair follicle. Arch Dermatol Res 1999; 29:253–9. 52 Miller DL, Weinstock MA. Non-melanoma skin cancer in the United States: incidence. J Am Acad Dermatol 1994; 30:774–8. 53 Holmes SA, Malinovszky K, Roberts DL. Changing trends in non-melanoma skin cancer in South Wales, 1988–1998. Br J Dermatol 2000; 143:1224–9. 54 Wallberg P, Skog E. The incidence of basal cell carcinoma in an area of Stockholm county during the period 1971–1980. Acta Derm Venereol (Stockh) 1991; 71:134–7. 55 Aszterbaum M, Beech J, Ervin H, et al. Ultraviolet radiation mutagenesis of hedgehog pathway genes in basal cell carcinomas. J Invest Dermatol Symp Proc 1999; 4:41–5. 56 Staples M, Marks R, Giles G. Trends in the incidence of nonmelanocytic skin cancer (NMSC) treated in Australia 1985–1995: are primary prevention programs starting to have an effect? Int J Cancer 1998; 78:144–8. 57 Noodleman FR, Pollack SV. Trauma as a possible etiologic factor in basal cell carcinoma. J Dermatol Surg Oncol 1986; 12:841–6. 58 Yeh S. Skin cancer in chronic aresenicism. Hum Pathol 1973; 4:469–85. 59 Ramachandran S, Fryer AA, Lovatt T, Lear JT, Smith AG, Strange RC. Susceptibility and modifier genes in cutaneous basal cell carcinomas and their associations with clinical phenotype. J Photochem Photobiol 2001; 63:1–7. 60 Lear JT, Heagerty AHM, Smith A, et al. Truncal site and detoxifying enzyme polymorphisms significantly reduce time to presentation of next cutaneous basal cell carcinoma. Carcinogenesis 1997; 18:1499–503. 61 Ramachandran S, Fryer AA, Smith AG, et al. Cutaneous basal cell carcinomas: distinct host factors are associated with the development of tumors on the trunk and on the head and neck. Cancer 2001; 92:354–8. 62 Johnson RL, Rothman AL, Xie J, et al. Human homolog of patched, a candidate gene for the basal cell nevus syndrome. Science 1996; 14:1668–71. 63 Riddle RO, Johnson RL, Laufer E, et al. Sonic hedgehog mediates the polarising activity of the ZPA. Cell 1993; 75:1401–16. 64 Roelink H, Porter JA, Chiang C, et al. Floor plate and motor neuron induction by different concentrations of the aminoterminal cleavage product of sonic hedgehog autoproteolysis. Cell 1995; 81:445–55. 65 Xie J, Murone M, Luoh SM, et al. Activating smoothened mutations in sporadic basal-cell carcinoma. Nature 1998; 391:90–2. 66 Fan H, Khavari PA. Sonic Hedgehog apposes epithelial cell cycle arrest. J Cell Biol 1999; 147:71–76.
872 Non-melanoma skin cancer
67 Johason AS, Kunala S, Price GJ, et al. Frequent clones of P53-mutated keratinocytes in normal human skin. Proc Natl Acad Sci USA 1996; 93:14025–9. 68 Lear JT, Smith AG, Heagerty AHM, et al. Truncal site and detoxifying enzyme polymorphisms significantly reduce time to presentation of further primary cutaneous basal cell carcinoma. Carcinogenesis 1997; 18:1499–503. 69 Chen GS, Yu HS, Lan CC, et al. CXC chemokine receptor CXCR4 expression enhances tumorigenesis and angiogenesis of basal cell carcinoma. Br J Dermatol 2006; 154:910–8. 70 MacKie R. Epidermal skin tumours. In: Champion, RH, Burton RH, Burns DA, Breathnach SM (eds) Textbook of dermatology, 6th edn. Oxford: Blackwell Scientific, 1998, 1651–93. 71 Miller SJ. Biology of basal cell carcinoma (part I). J Am Acad Dermatol 1991; 24:1–13. 72 Medenhall WM, Amdur RJ, Williams LS, et al. Carcinoma of the skin of the head and neck with perineural invasion. Head Neck 2002; 24:78–83. 73 Di Gregorio C, Florena AM, Gebbia V, et al. Mental nerve invasion by basal cell carcinoma of the chin: a case report. Anticancer Res 1998; 18:4723–6. 74 Terashi H, Kurata S, Tadokoro T. Perineural and neural involvement in skin cancers. Dermatol Surg 1997; 23:259–64. 75 Carlson KC, Roenigk RK. Know your anatomy: perineural involvement of basal and squamous cell carcinoma on the face. J Dermatol Surg Oncol 1990; 16:827–33. 76 Niazi ZB, Lamberty BG. Perineural infiltration in basal cell carcinomas. Br J Plast Surg 1993; 46:156–7. 77 Ten Hove MW, Glaser JS, Schatz NJ. Occult perineural tumor infiltration of the trigeminal nerve. Diagnostic considerations. J Neuroophthalmol 1997; 17:176–7. 78 Morselli P, Tosti A, Guerra L, et al. Recurrent basal cell carcinoma of the back infiltrating the spine. J Dermatol Surg Oncol 1993; 19:917–22. 79 Silbert PL, Kelsall GR, Shepherd JM, et al. Enigmatic trigeminal sensory neuropathy diagnosed by facial skin biopsy. Clin Exp Neurol 1992; 29:234–8. 80 Berti JJ, Sharata HH. Metastatic basal cell carcinoma to the lung. Cutis 1999; 63:165. 81 Christian MM, Murphy CM, Wagner RF Jr. Metastatic basal cell carcinoma presenting as unilateral lymphedema. Dermatol Surg 1998; 24:1151–3. 82 Berardi RS, Korba J, Mellow J, et al. Pulmonary metastasis in nevoid basal cell carcinoma syndrome. Int Surg 1991; 76:64–6. 83 Degner RA, Kerley SW, McGregor DH, et al. Metastatic basal cell carcinoma: report of a case presenting with respiratory failure. Am J Med Sci 1991; 301:395–7. 84 Blinder D, Taicher S. Metastatic basal cell carcinoma presenting in the oral cavity and auditory meatus. A case report and review of literature. Int J Oral Maxillofac Surg 1992; 21:31–4. 85 Cruse CW, O’Neill W, Rayhack J. Metastatic basal cell carcinoma of the upper extremity. J Hand Surg [Am] 1992; 17:1093–4. 86 Mizushima J, Ohara K. Basal cell carcinoma of the vulva with lymph node and skin metastasis – report of a case and review of 20 Japanese cases. J Dermatol 1995; 22:36–42.
87 Raszewski RL, Guyuron B. Long-term survival following nodal metastases from basal cell carcinoma. Ann Plast Surg 1990; 24:170–5. 88 Pfeiffer P, Hansen O, Rose C. Systemic cytotoxic therapy of basal cell carcinoma. A review of the literature. Eur J Cancer 1990; 26:73–7. 89 Moeholt K, Aagaard H, Pfeiffer P, et al. Platinum-based cytotoxic therapy in basal cell carcinoma – a review of the literature. Acta Oncol 1996; 35:677–82. 90 Gormley DP, Hirsch P. Aggressive basal cell carcinoma of the scalp. Arch Dermatol 1978; 114:782–3. 91 Cotran R. Metastasizing basal cell carcinomas. Cancer 1961; 14:1036–40. 92 Wong CSM, Strange RC, Lear JT. Basal cell carcinoma. Br Med J 2003; 327:794–798. 93 du Vivier A. Atlas of Clinical Dermatology, 2nd edn. New York: MosbyWolfe, 1993, 9.17–9.22. 94 Lewis JE. Linear basal cell epithelioma. Int J Dermatol 1985; 24:124–5. 95 Lim KK, Randle HW, Roenig RK, et al. Linear basal cell carcinoma: report of seventeen cases and review of the presentation and treatment. Dermatol Surg 1999; 25:63–7. 96 Chopra KF, Cohen PR. Linear basal cell carcinomas: report of multiple sequential tumors localized to a radiotherapy port and review of the literature. Tex Med 1997; 93:57–9. 97 Stein JM, Ormsby A, Esclamado R, Bailin P. The effect of radiation therapy on microcystic adnexal carcinoma: a case report. Head Neck 2003; 25:251–4. 98 Barton K, Curling OM, Paridaens AD, et al. The role of cytology in the diagnosis of periocular basal cell carcinomas. Ophthal Plast Reconstr Surg 1996; 12:190–4, discussion 195. 99 Telfer NR, Colver G, Bowers PW. Guidelines for the management of basal cell carcinoma. Br J Dermatol 1999; 141:415–23. 100 Holmkvist KA, Rogers GS, Dahl PR. Incidence of residual basal cell carcinoma in patients who appear tumor free after biopsy. J Am Acad Dermatol 1999; 41:600–5. 101 Albright S. Treatment of skin cancer using multiple modalities. J Am Acad Dermatol 1982; 7: 143–71. 102 Spiller WE, Spiller RE. Treatment of basal cell epithelioma by curettage and electro desiccation. J Am Acad Dermatol 1984; 11:808–14. 103 Salasche SJ. Curettage and electrodessication in the treatment of primary basal cell epithelioma. J Am Acad Dermatol 1984; 10:285–7. 104 Edens RL, Bartlow GA, Haghighi P, et al. Effectiveness of curettage and electrodessication in the removal of basal cell carcinoma. J Am Acad. Dermatol 1983; 9:383. 105 Nouri K, Spencer JM, Taylor JR, et al. Does wound healing contribute to the eradication of basal cell carcinoma following curettage and electrodessication? Dermatol Surg 1999; 25:183–7, discussion 187–8. 106 McLean D, Haynes HA, McCarthy PI, et al. Cryotherapy of basal cell carcinoma by a simple method of standardized freeze-thaw cycles. J Dermatol Surg Oncol 1978; 4:175–7.
References 873
107 Ceilley RI, Del Rosso JQ. Current modalities and new advances in the treatment of basal cell carcinoma. Int J Derm 2006; 45:489–98. 108 Kuflik EG, Gage AA. The five-year cure rate achieved by cryosurgery for skin cancer. J Am Acad Dermatol 1991; 24:1002–4. 109 Nordin P, Larko O, Stenquist B. Five-year results of curettage-cryosurgery of selected large primary basal cell carcinomas on the nose: an alternative treatment in a geographical area underserved by Mohs’ surgery. Br J Dermatol 1997; 136:180–3. 110 Breuninger H, Gutknecht M, Dietz K, et al. Locally infiltrative growth of squamous cell carcinoma of the skin and treatment guidelines resulting from it. Hautarzt 1991; 42:559–63. 111 Chiller K, Passaro D, McCalmont T, et al. Efficacy of curettage before excision in clearing surgical margins of nonmelanoma skin cancer. Arch Dermatol 2000; 136:1327–1332. 112 Griffiths RW. Audit of histologically incompletely excised basal cell carcinomas: recommendations for management by re-excision. Br J Plast Surg 1999; 52:24–8. 113 Mohs F. Chemosurgery: a microscopically controlled method of cancer excision. Arch Surg 1991; 42:279–95. 114 Cook JL, Perone JB. A prospective evaluation of the incidence of complications associated with Mohs’ micrographic surgery. Arch Dermatol 2003; 139:143–152. 115 Cottell WI, Proper S. Mohs’ surgery, fresh tissue technique: our technique with a review. J Dermatol Surg Oncol 1982; 8:576–87. 116 Smith SP, Grande DJ. Basal cell carcinoma recurring after radiotherapy: a unique, difficult treatment subclass of recurrent basal cell carcinoma. J Dermatol Surg Oncol 1991; 17:26–30. 117 Iyer S, Bowes L, Kricorian G, et al. Treatment of basal cell carcinoma with the pulsed carbon dioxide laser: a retrospective analysis. Dermatol Surg 2004; 30:1214–8. 118 Humphreys TR, Malhotra R, Scharf MJ, et al. Treatment of superficial basal cell carcinoma and squamous cell carcinoma in situ with a high-energy pulsed carbon dioxide laser. Arch Dermatol 1998; 134:1247–52. 119 Krunic AL, Viehman GE, Madani, S, et al. Microscopically controlled surgical excision combined with ultrapulse CO2 vaporization in the management of a patient with the nevoid basal cell carcinoma syndrome. J Dermatol 1998; 25:10–12. 120 Caccialanza C Piccinno R, Beretta M. Results and side effects of dermatologic radiotherapy: a retrospective study of irradiated cutaneous epithelial neoplasms. Am Acad Dermatol 1999; 41:589–94. 121 Rowe DE, Carroll RJ, Day CL Jr. Mohs’ surgery is the treatment of choice for recurrent (previously treated) basal cell carcinoma. J Dermatol Surg Oncol 1989; 15:424–31. 122 Miller RA, Spittle MF. Electron beam therapy for difficult cutaneous basal and squamous cell carcinomas. Br J Dermatol 1982; 106:429–36. 123 Lederman M. Radiation treatment of cancer of the eyelids. Br J Ophthal 1976; 60:794–805.
124 Schlienger P, Brunin F, Desjardin L, et al. External radiotherapy for carcinoma of the eyelid: Report of 850 cases treated. Int J Radiation Oncology Biol Phys 1996; 34:277–87. 125 Stafford SL, Kozelsky TF, Garrity JA, et al. Orbital Lymphoma: radiotherapy outcome and complications. Radiotherapy and Oncology 2001; 59:139–44. 126 Reisner K, Haase W. Electron beam therapy of primary tumors of the skin. Radiol Med (Torino) 1990; 80:114–15. 127 DeSilva SP, Dellon AL. Recurrence rate of positive margin basal cell carcinoma results of a five-year prospective study. J Surg Oncol 1985; 28:72–4. 128 Richmond JD, Davie RM. The significance of incomplete excision in patients with basal cell carcinoma. Br J Plast Surg 1987; 40:63–7. 129 Liu FF, Maki R, Warde P, Payne D, Fitzpatrick P. A management approach to incompletely excised basal cell carcinomas of skin. Int J Radiat Oncol Biol Phys 1991; 20:423–8. 130 Wang I, Bendsoe N, Klinteberg CA, et al. Photodynamic therapy vs. cryosurgery of basal cell carcinomas: results of a phase III clinical trial. Br J Dermatol 2001; 144:832–40. 131 Basset-Seguin N, Ibbotson S, Emtestam L. Photodynamic therapy using Metvix™ is as efficacious as cryotherapy in BCC with better cosmetic results. J Eur Acad Dermatol Venereol 2001 15:226. Cited in: Foley P. Clinical efficacy of methylaminolevulinate (metvix) photodynamic therapy. J Dermatol Treat 2003; 14:15–22. 132 Rhodes LE, de Rie M, Enstrom Y, et al. Photodynamic therapy using topical methyl aminolevulinate vs. surgery for nodular basal cell carcinoma: results of a multicenter randomized prospective trial. Arch Dermatol 2004; 140:17–23. 133 Epstein E. Fluorouracil paste treatment of thin basal cell carcinomas. Arch Dermatol 1985; 121:207–13. 134 Marks R, Gebauer K, Shumack S, et al. Imiquimod 5% cream in the treatment of superficial basal cell carcinoma: results of a multicenter 6-week dose–response trial. J Am Acad Dermatol 2001; 44:807–13. 135 Geisse JK, Rich P, Pandya A, et al. Imiquimod 5% cream for the treatment of superficial basal cell carcinoma: a doubleblind, randomized, vehicle-controlled study. J Am Acad Dermatol 2002; 47: 390–8. 136 Geisse J, Caro I, Lindholm J, et al. Imiquimod 5% cream for the treatment of superficial basal cell carcinoma: results from two phase III, randomized, vehicle-controlled studies. J Am Acad Dermatol 2004; 50:722–33. 137 Shumack S, Robinson J, Kossard S, et al. Efficacy of topical 5% imiquimod cream for the treatment of nodular basal cell carcinoma: comparison of dosing regimens. Arch Dermatol 2002; 138:1165–71. 138 Sterry W, Ruzicka T, Herrera E, et al. Imiquimod 5% cream for the treatment of superficial and nodular basal cell carcinoma: randomized studies comparing low-frequency dosing with and without occlusion. Br J Dermatol 2002; 147:1227–36. 139 Robinson JK. Risk of developing another basal cell carcinoma. Cancer 1987; 60:118–20.
874 Non-melanoma skin cancer
140 Salasche SJ. Epidemiology of actinic keratoses and squamous cell carcinoma. J Am Acad Dermatol 2000; 42:4–7. 141 Ahmed I, Das Gupta AR. Epidemiology of basal cell carcinoma and squamous cell carcinoma of the Pinna. J Laryngol Otol 2001; 115:85–6. 142 Feldman SR, Fleischer AB Jr, McConnell C, et al. Most common dermatologic problems identified by internists 1990–1994. Arch Intern Med 1998; 158:726–30. 143 Holman CD, Armstrong DK, Evans PR, et al. Relationship of solar keratosis and history of skin cancer to objective measures of actinic skin damage. Br J Dermatol 1984; 110:129–38. 144 Joseph MG, Zulueta WP, Kennedy PJ. Squamous cell carcinoma of the skin of the trunk and limbs: the incidence of metastases and their outcome. Aust NZ J Surg 1992; 62:697–701. 145 Rowe DE, Carroll RJ, Day CL Jr. Prognostic factors for local recurrence, metastasis, and survival rates in squamous cell carcinoma of the skin, ear, and lip. Implications for treatment modality selection. J Am Acad Dermatol 1992; 26:976–90. 146 Cherpelis BS, Marcusen C, Lang PG. Prognostic factors for metastasis in squamous cell carcinoma of the skin. Dermatol Surg 2002; 28:268–73. 147 Antoniades DZ, Styanidis K, Papanayotou P, et al. Squamous cell carcinoma of the lips in a northern Greek population. Evaluation of prognostic factors on 5-year survival rate. Eur J Cancer 1995; 318:333–9. 148 Motley R, Kersey P, Lawrence C, et al. Multiprofessional guidelines for the management of the patient with primary cutaneous squamous cell carcinoma. Br J Dermatol 2002; 146:18–25. 149 Friedman HI, Cooper PH, Wanebo HJ. Prognostic and therapeutic use of microstaging of cutaneous squamous cell carcinoma of the trunk and extremities. Cancer 1985; 56:1099–105. 150 Lazarov M, Kubo Y, Cai T, et al. CDK4 Co-expression with Ras generates malignant human epidermal tumorigenesis. Nat Med 2002; 8:1105–14. 151 Dajee M, Lazarov M, Zhang J, et al. NF-kappa Beta blockade and oncogenic Ras trigger invasive human epidermal neoplasia. Nature 2003; 421:639–43. 152 Brodland DG, Zitelli JA. Surgical margins for excision of primary cutaneous squamous cell carcinoma. J Am Acad Dermatol 1992; 27:241–8. 153 Lawrence N, Cottel WI. Squamous cell carcinoma of skin with perineural invasion. J Am Acad Dermatol 1994; 31:30–3. 154 Ghauri RR, Gunter AA, Weber RA. Frozen section analysis in the management of skin cancers. Ann Plast Surg 1999; 43:156–60. 155 Veness MJ, Morgan GJ, Palme CE, Gebski V. Surgery and adjuvant radiotherapy in patients with cutaneous head and neck squamous cell carcinoma metastatic to lymph nodes: combined treatment should be considered best practice. Laryngoscope 2005; 115:870–5.
156 Stephens FO, Harker GJS. The use of intra-arterial chemotherapy for treatment of malignant skin neoplasms. Australas J Dermatol 1979; 20:99–107. 157 Sadek H, Azli N, Wendling JL, et al. Treatment of advanced squamous cell carcinoma of the skin with cisplatin 5-fluorouracil and bleomycin. Cancer 1990; 66:1692–6. 158 Shin DM, Glisson BS, Khuri FR, et al. Phase II and biologic study of interferon alfa, retinoic acid, and cisplatin in advanced squamous skin cancer. J Clin Oncol 2002; 20:364–70. 159 Frankel D, Hanusa BH, Zitalli JA. New primary non-melanoma skin cancer patients with a history of squamous cell carcinoma of the skin. Implications and recommendations for follow-up. J Am Acad Dermatol 1992; 26:720–6. 160 Whittaker SJ, Marsden JR, Spittle M, et al. Joint British Association of Dermatologists and UK Cutaneous Lymphoma Group guidelines for the management of primary cutaneous T-cell lymphomas. Br J Dermatol 2003; 149:1095–107. 161 Weinstock MA, Gardstein B. Twenty-year trends in the reported incidence of mycosis fungoides and associated mortality. Am J Public Health 1999; 89:1240–4. 162 Willemze R, Jaffe ES, Burg G, et al. WHO-EORTC classification for cutaneous lymphomas. Blood 2005; 105: 3768–85. 163 Tan R, Butterworth CM, Mclaughin H, et al. Mycosis fungoides – a disease of antigen persistence. Br J Dermatol 1974; 91:607–16. 164 Vowels BR, Cassin M, Vonderheid EC, et al. Aberrant cytokine production by Sèzary syndrome patients: cytokine secretion pattern resembles murine Th2 cells. J Invest Dermatol 1992; 99: 90–8. 165 Vowels BR, lessin SR, Cassin M, et al. Th2 cytokine mRNA expression in skin in cutaneous T cell lymphoma. J Invest Dermatol 1994; 103:29–32. 166 Saed G, Fivenson DP, Naidu T, et al. Mycosis fungoides exhibits a Th1-type cell mediated cytokine profile, whereas, Sèzary syndrome expresses a Th1-type profile. J Invest Dermatol 1994; 103:29–33. 167 Sugerman G, Bigby M. Preliminary function analysis of human epidermal T-cells. Arch Dermatol Res 2000; 292:9–15. 168 Howard K, Charif M, Martin A, et al. Epidemiology and clinical manifestations of cutaneous T-cell lymphoma. Hematol Oncol Clin North Am 1995; 9:5–11. 169 Lisby G, Reitz MS, Vejlsgaard GL, Reitz MS Jr. No detection of HTLV-l DNA in punch biopsies from patients with cutaneous T-cell lymphoma by the polymerase chain reaction. J Invest Dermatol 1992; 98:417–20. 170 Pancake RA, Zucker-Franklin D, Coutavas EE. The cutaneous T-cell lymphoma, mycosis fungoides, is a human T-cell lymphotrophic virus-associated disease. J Clin Invest 1995; 95:547–54. 171 Tothova SM, Bonin S, Trevisan G, Stanta G. Mycosis fungoides: is it a Borrelia burgdorferi-associated disease? Br J Cancer. 2006; 94:879–83. 172 Willemze R, Jaffe ES, Burg G, et al. WHO-EORTC classification for cutaneous lymphomas. Blood 2005; 105:3768–85.
References 875
173 Trautinger F, Knobler R, Willemze R, et al. EORTC consensus recommendations for the treatment of mycosis fungoides/Sèzary syndrome. Eur J Cancer 2006; 42:1014–30. 174 Tsai EY, Taur A, Espinosa L, et al. Staging accuracy in mycosis fungoides and Sèzary syndrome using integrated positron emission tomography and computed tomography. Arch Dermatol 2006; 142:577–84. 175 Thangarelu M, Finn WG, Yelavarthi KK, et al. Recurring structural chromosomal abnormalities in peripheral blood lymphocytes of patients with mycosis fungoides/Sèzary syndrome. Blood 1997; 89:3371–7. 176 Mao X, Lillington DM, Czepulkowski B, et al. Molecular cytogenetic characterization of Sèzary syndrome. Genes Chromosomes. Cancer 2003; 36:250–60. 177 Karenko L, Hyytinen E, Sarna S, Ranki A. Chromosomal abnormalities in cutaneous T-cell lymphoma and in its premalignant conditions as detected by G-banding and interphase cytogenetic methods. J Invest Dermatol 1997; 108:22–9. 178 Mao X, Lillington D, Scarisbrick JJ, et al. Molecular cytogenetic analysis of cutaneous T-cell lymphomas: identification of common genetic alterations in Sèzary syndrome and mycosis fungoides. Br J Dermatol 2002; 147:464–75. 179 Fischer T, Gellrich S, Muche M, et al. Genomic aberrations and survival in cutaneous T-cell lymphomas. J Invest Dermatol 2004; 122:579. 180 McGregor J, Crook T, Fraser-Andrews E, et al. Spectrum of p53 mutations suggests a possible role for ultraviolet radiation in the pathogenesis of advanced cutaneous lymphomas. J Invest Dermatol 1999; 112:317–21. 181 Navas IC, Ortiz-Romero PL, Villuendas R, et al. p16(INK4a) gene alterations are frequent in lesions of mycosis fungoides. Am J Pathol 2000; 156:1565–72. 182 Scarisbrick JJ, Woolford AJ, Calonje E, et al. Frequent abnormalities of the p15 and p16 genes in mycosis fungoides and Sèzary syndrome. J Invest Dermatol 2002; 118:493–9. 183 Scarisbrick JJ, Mitchell TJ, Calonje E, et al. Microsatellite instability is associated with hypermethylation of the hMLH1 gene and reduced gene expression in mycosis fungoides. J Invest Dermatol 2003; 121:894–901. 184 Rubben A, Kempf W, Kadin ME, et al. Multilineage progression of genetically unstable tumor subclones in cutaneous T-cell lymphoma. Exp Dermatol 2004; 13:472–83. 185 Van Doorn R, Zoutman W, Dijkman, et al. Epigenetic profiling of cutaneous T-cell lymphoma: Promotor hypermethylation of multiple tumour suppressor genes including BCL7a, PTPRG and p73. J Clin Oncol 2005; 23:3886–96. 186 Mao X, Orchard G, Lillington DM, et al. Amplification and overexpression of JUNB is associated with primary cutaneous T-cell lymphomas. Blood 2003; 101:1513–9. 187 Van Doorn R, Dijkman R, Vermeer MH, et al. Aberrant expression of the tyrosine kinase receptor EphA4 and the transcription factor twist in Sézary syndrome identified by gene expression analysis. Cancer Res 2004; 64:5578–86.
188 Kari L, Loboda A, Nebozhyn M, et al. Classification and prediction of survival in patients with the leukemic phase of cutaneous T cell lymphoma. J Exp Med 2003; 197:1477–88. 189 Tracey L, Villuendas R, Dotor A, et al Mycosis fungoides shows concurrent deregulation of multiple genes involved in the TNF signaling pathway: an expression profile study. Blood 2003; 102:1042–50. 190 Sommer VH, Clemmensen OJ, Nielsen O, et al. In vivo activation of STAT3 in cutaneous T-cell lymphoma. Evidence for an antiapoptotic function of STAT3. Leukemia 2004; 18:1288–95. 191 Mitchell TJ, Whittaker SJ, John S. Dysregulated expression of COOH-terminally truncated Stat5 and loss of IL2-inducible Stat5-dependent gene expression in Sèzary Syndrome. Cancer Res 2003; 63:9048–54. 192 Zhang Q, Raghunath PN, Vonderheid E, et al. Lack of phosphotyrosine phosphatase SHP-1 expression in malignant T-cell lymphoma cells results from methylation of the SHP-1 promoter. Am J Pathol 2000; 157:1137–46. 193 Kaye FJ, Bunn PA Jr, Steinberg SM, et al. A randomized trial comparing combination electron beam radiation and chemotherapy with topical therapy in the initial treatment of mycosis fungoides. N Engl J Med 1989; 321:1748–90. 194 Zackheim H, Kashani-Sabet M, Amin S. Topical corticosteroids for mycosis fungoides. Arch Dermatol 1998; 134:949–54. 195 Scholtz W. Ueber den einfluss der rontgenstrahlen auf die haut in gesunden und krankem zustande. Arch Dermat U Syph 1902: 59:421. 196 Kim JH, Nisce LZ, D’Anglo GJ. Dose-time fractionation study in patients with mycosis fungoides and lymphoma cutis. Radiology 1976: 119:439–42. 197 Cotter GW, Baglan RJ, Wasserman TH, Mill W. Palliative radiation treatment of cutaneous mycosis fungoides – a dose–response. Int J Radiat Oncol Biol Phys 1983: 9:1477–80. 198 Hoppe RT. Mycosis fungoides: radiation therapy. Dermatol Ther 16; 2003: 347–54. 199 Karzmark CJ, Loevinger R, Steele RE, Weissbluth M. A technique for large-field, superficial electron therapy. Radiology 1960: 74:633–44. 200 Szur L, Silvester JA, Bewley DK: Treatment of the whole body surface with electrons. Lancet 1:1373–7. 201 Jones GW, Kacinski BM, Wilson LD, et al. Total skin electron radiation in the management of mycosis fungoides: consensus of the European Organization for Research and Treatment of Cancer (EORTC) Cutaneous Lymphoma Project Group. J Am Acad Dermatol 2002; 47:364–70. 202 Kim TH, Pla C, Pla M, Podgorsak EB. Clinical aspects of a rotational total skin electron beam irradiation. Br J Radiol 1984; 57:501–6. 203 Jones GW, Hoppe RT, Glastein E. Electron Beam treatment for CTCL. Hematol Oncol Clin North Am 1995; 9:1057–76. 204 Kim YH, Jensen RA, Watanabe GL, et al. Clinical stage IA mycosis fungoides. A long term outcome analysis. Arch Dermatol 1996; 132:1309–13.
876 Non-melanoma skin cancer
205 Jones G, Wilson LD, Fox-Goguen L. Total skin electron beam radiotherapy for patients who have mycosis fungoides. Hematol Oncol Clin North Am 2003; 17:1421–34. 206 Kim YH, Martinez G, Varghese A, et al. Topical nitrogen mustard in the management of mycosis fungoides: update of the Stanford experience. Arch Dermatol 2003; 139:165–73. 207 Chinn DM, Chow S, Kim YH, et al. Total skin electron beam therapy with or without adjuvant topical nitrogen mustard or nitrogen mustard alone as initial treatment of T2 and T3 mycosis fungoides. Int J Radiat Oncol Biol Phys 1999; 43:951–8. 208 Jones GW, Rosenthal D, Wilson LD. Total skin electron beam radiation for patients with erythrodermic cutaneous T-cell lymphoma (mycosis fungoides and the Sèzary syndrome). Cancer 1999; 85:1985–95. 209 Wilson LD. Delivery and Sequelae of Total Skin Electron Beam Therapy. Arch Derm 2003; 139:812–13. 210 Sun WH, Pabon C, Alsayed Y, et al. Interferon-alpha resistance in a cutaneous T-cell lymphoma cell line is associated with lack of STAT1 expression. Blood 1998; 91:570–6. 211 Bunn PA Jr, Ihde DC, Foon KA. The role of recombinant interferon alpha-2a in the therapy of cutaneous T-cell lymphomas. Cancer 1986; 57:1689–95. 212 Olsen EA, Rosen ST, Vollmer RT, et al. Interferon alfa-2a in the treatment of cutaneous T-cell lymphoma. J Am Acad Dermatol 1989; 20:395–407. 213 Papa G, Tura S, Mandelli F, et al. Is interferon alpha in cutaneous T-cell lymphoma a treatment of choice? Br J Haematol 1991; 79:48–51. 214 Stadler R, Otte HG, Luger T, et al. Prospective randomized multicentre clinical trial on the use of interferon alpha-2a plus acitretin versus interferon alpha-2a plus PUVA in patients with cutaneous T-cell lymphoma stages I and II. Blood 1998; 10:3578–81. 215 Kuzel TM, Roenigk HH Jr, Samuelson E, et al. Effectiveness of interferon alfa-2a combined with phototherapy for mycosis fungoides and the Sézary syndrome. J Clin Oncol 1995; 13:257–63. 216 Dreno B, Claudy A, Meynadier J, et al. The treatment of 45 patients with cutaneous T-cell lymphoma with low doses of interferon- alpha 2a and etretinate. Br J Dermatol 1991; 125:456–9. 217 Rook AH, Wood GS, Yoo EK, et al. Interleukin-12 therapy of cutaneous T-cell lymphoma induces lesion regression and cytotoxic T-cell responses. Blood 1999; 94:902–8. 218 Kaplan EH, Rosen ST, Norris DB, et al. Phase II study of recombinant interferon gamma for treatment of cutaneous T-cell lymphoma. J Nat Cancer Inst 1990; 82:208–12. 219 Cooper DL, Braverman IM, Sarris AH, et al. Cyclosporine treatment of refractory T-cell lymphomas. Cancer 1993; 71:2335–41. 220 Kurzrock R, Pilat S, Duvic M. Pentostatin therapy of T-cell lymphomas with cutaneous manifestations. J Clin Oncol 1999; 17:3117–21.
221 Dearden C, Matutes E, Catovsky D. Pentostatin treatment of cutaneous T-cell lymphoma. Oncology 2000; 14:37–40. 222 Scarisbrick JJ, Child FJ, Clift A, et al. A trial of fludarabine and cyclophosphamide combination chemotherapy in the treatment of advanced refractory primary cutaneous T-cell lymphoma. Br J Dermatol 2001; 144:1010–15. 223 Zinzani PL, Baliva G, Magagnoli M, et al. Gemcitabine treatment in pretreated cutaneous T-cell lymphoma: experience in 44 patients. J Clin Oncol 2000; 18:2603–6. 224 Marchi E, Alinari L, Tani M, et al. Gemcitabine as frontline treatment for cutaneous T-cell lymphoma: phase II study of 32 patients. Cancer 2005; 104:2437–41. 225 Wollina U, Dummer R, Brockmeyer NH, et al. Multicenter study of pegylated liposomal doxorubicin in patients with cutaneous T-cell lymphoma. Cancer 2003; 98:993–1001. 226 Di Lorenzo G, Di Trolio R, Delfino M, De Placido S. Pegylated liposomal doxorubicin in stage IVB mycosis fungoides. Br J Dermatol 2005; 153:183–5. 227 Burt RK, Guitart J, Traynor A, et al. Allogeneic hematopoietic stem cell transplantation for advanced mycosis fungoides: evidence of a graft-versus-tumour effect. Bone Marrow Transplant 2000; 25:111–13. 228 Molina A, Nademanee A, Arber DA, Forman SJ. Remission of refractory Sèzary syndrome after bone marrow transplantation from a matched unrelated donor. Biol Blood Marrow Transplant 1999; 5:400–4. 229 Molina A, Zain J, Arber DA, et al. Durable clinical, cytogenetic, and molecular remissions after allogeneic hematopoietic cell transplantation for refractory Sèzary syndrome and mycosis fungoides. J Clin Oncol 2005; 23:6163–71. 230 Boehm MF, Zhang L, Badea BA, et al. Synthesis and structure–activity relationships of novel retinoid X receptorselective retinoids. J Med Chem 1994; 37:2930–41. 231 Duvic M, Martin AG, Kim Y, et al. Phase 2 and 3 clinical trial of oral bexarotene (Targretin capsules) for the treatment of refractory or persistent early-stage cutaneous T-cell lymphoma. Arch Dermatol 2001; 137:581–93. 232 Committee for proprietary medicinal products. European public assessment report (EPAR): Targretin. European Agency for the Evaluation of Medicinal Products, London; 2001. 233 Zhang C, Hazarika P, Ni X, et al. Induction of apoptosis by bexarotene in cutaneous T-cell lymphoma cells: relevance to mechanism of therapeutic action. Clin Cancer Res 2002; 8:1234–40. 234 Edelson R, Berger C, Gasparro F, et al. Treatment of cutaneous T-cell lymphoma by extracorporeal photochemotherapy. Preliminary results. N Engl J Med 1987; 316:297–303. 235 Talpur R, Ward S, Apisarnthanarax N, et al. Optimizing bexarotene therapy for cutaneous T-cell lymphoma. J Am Acad Dermatol 2002; 47:672–84. 236 Knobler R, Jantschitsch C. Extracorporeal photochemoimmunotherapy in cutaneous T-cell lymphoma. Transfus Apheresis Sci 2003; 28:81–9. 237 Vonderheid EC, Bernengo MG, Burg G, et al. Update on erythrodermic cutaneous T-cell lymphoma: report of the International Society for Cutaneous Lymphomas. J Am Acad Dermatol 2002; 46:95–106.
References 877
238 Russell Jones RR, Whittaker S. T-cell receptor gene analysis in the diagnosis of Sèzary syndrome. J Am Acad Dermatol 1999; 41:254–9. 239 Suchin KR, Cucchiara AJ, Gottleib SL, et al. Treatment of cutaneous T-cell lymphoma with combined immunomodulatory therapy: a 14-year experience at a single institution. Arch Dermatol 2002; 138:1054–60. 240 Bazarbachi A, Ghez D, Lepelletier Y, et al. New therapeutic approaches for adult T-cell leukaemia. The Lancet Oncol 2004; 5:664–72. 241 Bekkenk M, Geelen FAMJ, van Voorst, et al. Primary and secondary cutaneous CD30-positive lymphoproliferative disorders: long term follow-up data of 219 patients and guidelines for diagnosis and treatment. A report from the Dutch Cutaneous Lymphoma Group. Blood 2000; 95:3653–61. 242 Eich HT, Eich D, Micke O, et al. Long term efficacy, curative potential and prognostic factors of radiotherapy in primary cutaneous B cell lymphoma. Int J Radiat Oncol Biol Phys 2003; 55:899–906. 243 Smith BD, Glusac EJ, McNiff JM, et al. Primary cutaneous B cell lymphoma treated with radiotherapy: a comparison of the EORTC and WHO classification systems. J Clin Oncol 2004; 15:634–9. 244 Rijlaarsdam JU, Toomska J, Meijer OW, et al. Treatment of primary cutaneous B cell lymphoma of follicle centre cell origin: a clinical follow up study of 55 patients treated with radiotherapy or polychemotherapy. J Clin Oncol 1996; 14:549–55. 245 Piccinno R, Caccialanza M, Berti E, et al. Radiotherapy of cutaneous B cell lymphomas: our experience of 31 cases. Int J Radiat Oncol Biol Phys 1993; 27:385–9. 246 Paget J. On disease of the mammary areola preceding cancer of the mammary gland. St Barthol Hosp Report 1874; 10:87–89. 247 Crocker HR. Paget’s disease affecting the scrotum and penis. Trans Path Soc 1888; 40:187–191. 248 Darier J, Coulillaud P. Sur un cas de malade de Paget de la region perineo-anale et scrotale. Ann Dermatol Syph 1893; 4:25–33. 249 Brown RSD, Lankester KJ, McCormak M, Power DA, Spittle MF. Radiotherapy for perianal Paget’s Disease. Clin Oncol 2002; 14:272–84. 250 Coldiron BM, Goldsmith BA, Robinson JK. Surgical treatment of extramammary Paget’s disease. A report of six cases and a re-examination of Mohs micrographic surgery compared with conventional surgical excision. Cancer 1991; 67:933–8. 251 Mohs FE, Blanchard L. Microscopically controlled surgery for extramammary Paget’s disease. Arch Dermatol 1979; 115:706–8. 252 Mirer E, El Sayed F, Ammoury A, Lamant L, Messer L, Bazex J. Treatment of mammary and extramammary Paget’s skin disease with topical imiquimod. J Dermatol Treat 2006; 17:167–71. 253 Besa P, Rich TA, Delclos L, Edwards CL, Ota DM, Wharton JT. Extramammary Paget’s disease of the perineal skin: role of radiotherapy. Int J Radiat Oncol Biol Phys 1992; 24:73–78.
254 Vortmeyer AO, Merino MJ, Boni R, et al. Genetic changes associated with primary Merkel cell carcinoma. Am J Clin Pathol 1998; 109: 565–70. 255 Haag ML, Glass LF, Fenske NA. Merkel cell carcinoma. Diagnosis and treatment. Dermatol Surg 1995; 21:669–83. 256 Akhtar S, Oza KK, Wright J. Merkel cell carcinoma: report of 10 cases and review of the literature. J Am Acad Dermatol 2000; 43:755–67. 257 Iacocca MV, Abernethy JL, Stefanato CM, et al. Mixed Merkel cell carcinoma and squamous cell carcinoma of the skin. J Am Acad Dermatol 1998; 39:882–7. 258 Simstein NL, Sduggs NK. Merkel cell tumor: two cases. Int Surg 1998; 83:60–2. 259 Lunder EJ, Stern RS. Merkel cell carcinoma in patients treated with methoxsalen and ultraviolet radiation. N Engl J Med 1998; 339:1247–8. 260 Williams RH, Morgan MB, Mathieson IM, et al. Merkel cell carcinoma in a renal transplant patient: increased incidence? Transplantation 1998; 65:1396–7. 261 Ott MJ, Tanabe KK, Gadd MA, et al. Multimodality management of Merkel cell carcinoma. Arch Surg 1999; 134:388–92. 262 Schlagbauer-Wadl H, Klosner G, Heere-Ress E. BcI-2 antisense oligonucleotides (G3139) inhibit Merkel cell carcinoma growth in SCID mice. J Invest Dermatol 2000; 114:725–30. 263 Belhocine T, Pierard GE, Fruhling J, et al. Clinical addedvalue of 18FDG PET in neuroendocrine-merkel cell carcinoma. Oncol Rep 2006; 16:347–52. 264 Yiengpruksawan A, Coit DG, Thaler HT, et al. Merkel cell carcinoma, prognosis and management. Arch Surg 1991; 126:1513–9. 265 Tope WD, Sangueza OP. Merkel cell carcinoma. Histopathology, immunohistochemistry, and cytogenetic analysis. J Dermatol Surg Oncol 1994; 20:648–52, quiz 653–4. 266 Allen PJ, Zhang ZF, Coit DG. Surgical management of Merkel cell carcinoma. Ann Surg 1999; 229:97–105. 267 Eich HT, Eich D, Staar S, et al. Role of postoperative radiotherapy in the management of Merkel cell carcinoma. Am J Clin Oncol 2002; 25:50–6. 268 Ratner D, Nelson BR, Brown MD, Johnson TM. Merkel cell carcinoma. J Am Acad Dermatol 1993; 29:143–56. 269 Veness MJ, Morgan GJ, Palme CE, Gebski V. Surgery and adjuvant radiotherapy in patients with cutaneous head and neck squamous cell carcinoma metastatic to lymph nodes: combined treatment should be considered best practice. Laryngoscope 2005; 115:870–5. 270 Allen PJ, Bowne WB, Jacques DP, et al. Merkel Cell Carcinoma: Prognosis and Treatment of Patients From a Single Institution J Clin Oncol 2005; 23:2300–9. 271 Hill ADK, Brady MS, Coit DG. Intraoperative lymphatic mapping and sentinel lymph node biopsy for Merkel cell carcinoma. Br J Surg 1999; 86:518. 272 Poulsen M, Rischin D, Walpole E, et al. Analysis of toxicity of Merkel cell carcinoma of the skin treated with synchronous carboplatin/etoposide and radiation: A Trans-Tasman Radiation Oncology Group study. Int J Radiat Oncol Biol Phys 2003; 51:156–163.
878 Non-melanoma skin cancer
273 Voog E, Biron P, Martin JP, et al. Chemotherapy for patients with locally advanced or metastatic Merkel cell carcinoma. Cancer 1999; 85:2589–95. 274 Fenig E, Brenner B, Njuguna E, et al. Oral etoposide for Merkel cell carcinoma in patients previously treated with intravenous etoposide. Am J Clin Oncol 2000; 23:65–7. 275 Broadbent V, Pritchard J. Histiocytosis X- current controversies. Arch Dis Child 1985; 60:605–7. 276 Basset F, Turiaf J. Identification par la microscopic electronique de particules de nature probablement virale dans les lesions granulomateuses d’une hystiocytose X pulmonaire CR. Acad Sci (Paris) 1965; 261:3701–3. 277 Beckstead J, Wood GS, Turner RR. Histiocytosis X cells and Langerhans cells: enzyme histochemical and immunological similarities. Hum Pathol 1984; 15:826–33. 278 Groh V, Gadner H, Radaszkiewicz T, et al. The phenotypic spectrum of histiocytosis X cells. J Invest Dermatol 1988; 90:441–7.
279 Harrist T, Bahn AK, Murphy GF, et al. Histiocytosis X in situ characterization of cutaneous infiltrates with monoclonal antibodies. Am J Clin Pathol 1983; 79:294–300. 280 Chu AC, Jaffe N. The normal Langerhans cell and the LCH cell. Br J Cancer 1994; 70:S4–10. 281 Lahey M. Prognosis in reticuloendotheliosis in children. J Pediatr 1962; 60:664–71. 282 Munn S, Chu AC. Langerhans cell histiocytosis of the skin. Hematol Oncol Clin North Am 1998; 12:269–86. 283 Greenberger J, Cassady JR, Jaffe N, et al. Radiation therapy in patients with histiocytosis: management of diabetes insipidus and bone lesions. Int J Radiat Oncol Bio Phys 1979; 5:1749–55. 284 McLelland J, Pritchard J, Chu AC. Current controversies in histiocytosis X. Hematol Oncol Clin North Am 1987; 1:147–62. 285 Osband M, Lipton JM, Lavin P, et al. Histiocytosis X. Demonstration of abnormal immunity, T-cell histamine H2receptor deficiency and successful treatment with thymic extract. N Engl J Med 1981; 304:146–53.
38 Malignant melanoma SUSANA BANERJEE AND MARTIN E. GORE
Introduction Pathogenesis Aetiology Primary prevention Diagnosis Pathology Staging and prognosis
879 879 879 880 881 881 882
Initial staging investigations Treatment Novel therapies Intra-ocular melanoma Conclusions References
884 885 894 895 896 897
INTRODUCTION
PATHOGENESIS
Malignant melanoma arises from melanocytes and is the most serious form of skin cancer. Eight thousand new cases are diagnosed in the UK each year and this contributes to 3 per cent of all cancers.1 The lifetime risk is in excess of 1 in 200. Melanoma affects young individuals and is second only to adult leukaemia in terms of loss of years of potential life. The incidence of melanoma is rising by 5 per cent per year in white populations, with the greatest increase in men over 65 years of age.2 This rate of increase is more rapid than all other forms of cancer.1 Suggested reasons for the rapid rise in incidence are multi-factorial and probably are a result of a combination of increased sun exposure, increased amounts of UVB irradiation reaching the earth surface due to ozone depletion and earlier detection of melanoma. The majority of patients present with early-stage disease and 90 per cent of these patients are cured by surgical excision. The outlook for advanced disease is very poor. There were 1817 deaths from melanoma in the UK during 2005, accounting for 1 per cent of all cancer deaths.1 Five-year all-stage survival for patients diagnosed in the UK between 1996 and 1999 was 80 per cent. The median survival for metastatic melanoma was 8.1 months in a meta-analysis including 5392 patients.3 There is no treatment that effectively impacts on survival in patients with metastatic melanoma. There is an urgent need for more effective systemic therapies in patients with metastatic melanoma and this is an area of ongoing research.
Ultraviolet solar radiation is thought to promote malignant change by having direct mutagenic effects on DNA, by stimulating the production of growth factors, by reducing cutaneous immune defences, and by promoting reactive oxygen species of melanin that cause DNA damage and suppress apoptosis.4 Signals from growth factor receptor interactions result in activation of the Ras/Raf pathway triggering the transcription of genes involved in cellular proliferation and migration. BRAF mutations are the most common oncogene mutation in melanoma, occurring in 10–30 per cent of primary melanoma. This pathway is believed to play a role in melanocyte biology. However the development of melanoma is complicated and other pathways are likely to be involved as well.
AETIOLOGY The risk factors for developing melanoma are environmental and genetic. Known risk factors include a past history or family history of melanoma, fair complexion, freckling tendency, propensity to burn, duration of sun exposure between age 10–24 years, presence of an atypical mole, congenital melanocytic naevus, presence of a high number of common melanocytic naevi greater than 2 mm and immunosuppression.5,6
880 Malignant melanoma
Sun exposure Exposure to sunlight is believed to be the major causative factor in melanoma. Evidence for the role of sunlight comes mainly from epidemiological studies.7 Melanoma is 12 times more common in Caucasians and 6 times more common in those of Hispanic origin than in black-skinned races living the same lifestyle. Those of European descent living in Australia have a significantly greater risk of developing melanoma than those remaining in northern Europe. In addition, residence during early life in more equatorial latitudes is associated with an increased risk of melanoma.8 Several observations demonstrate that the relationship between sun exposure and melanoma is complex. The incidence of melanoma does not consistently increase across geographical areas with increasing sunlight. Melanoma is relatively common in young people without a long cumulative sun exposure and is more common in indoor than outdoor workers. Melanoma can occur in relatively unexposed sites such as the palms, soles and mucous membranes. An overview of 29 epidemiological studies suggests an association between melanoma and intermittent sun exposure.9 Possible theories are that brief, intense exposure may initiate carcinogenesis or that UVB irradiation may cause generalized cutaneous immunosuppression allowing melanoma to develop on areas of the body not necessarily exposed to high levels of sunlight. There are epidemiological data on the possible contribution of tanning beds/lamps that emit UVA to melanoma risk. Exposures from 106 379 women aged 30 to 50 enrolled in the Women’s Lifestyle and Health Cohort Study in Sweden and Norway between 1991 and 1992 were linked to national cancer registry data. Women who were exposed to such devices one or more times per month had a significantly higher risk of melanoma compared to those who did not use them (relative risk (RR) 1.55), and the risk was greatest for women exposed in their 20s.10 Results form a metaanalysis indicated a significantly increased risk of melanoma subsequent to tanning beds/lamps exposure.11 The role of sunburn remains unclear. It may be that sunburn is a marker of both intermittent, intense sun exposure and genetic susceptibility.12 The role of sunscreens as a preventative measure is uncertain. There is randomized evidence that sunscreen use increases time spent in the sun.13 Randomized trials aiming at establishing the link between sunscreen use and melanoma prevention are difficult to conduct because melanoma is a rare disease and sunscreen use is highly prevalent in many populations. There is currently no evidence of the effectiveness of sunscreens; sun exposure avoidance remains the most effective preventative measure and is discussed later.14
PUVA therapy Exposure to oral psoralen and ultraviolet A radiation (PUVA) is associated with a delayed increase in the risk of
melanoma. In one multi-centre series of 1380 patients with severe psoriasis who were first treated with PUVA in 1975 and 1976, the incidence rate ratio for invasive or in situ cutaneous melanomas was not elevated in the first 15 years following treatment. However, the incidence rate ratio for all melanomas increased to 5.4 by years 16 to 20, and to 9.3 beyond 20 years of follow-up.15
Genetics There is a strong family history in 5–7 per cent of melanoma patients.16 Many of these patients have atypical mole syndrome (AMS). There is considerable genetic heterogeneity among these different families, suggesting that multiple genes contribute to melanoma predisposition.17 The proportion of all melanomas that is due to these genes is estimated to be less than 2 per cent. The major gene resides on chromosome 9p and encodes the tumour suppressor gene CDKN2A that is critical in cell cycle regulation. Carriers of mutations in CDKN2A have a risk of developing melanoma of 50–90 per cent by age 80 years.18 Inherited mutations of this gene are present in 30–40 per cent of families with three or more members with melanoma.19 Inherited variants of low penetrance melanoma susceptibility genes may modify the risk of developing melanoma. Inheritance of melanocortin-1 receptor gene variants gives rise to increased UV sensitivity and an increased melanoma risk.20 The BRAF and NRAS genes are commonly mutated as somatic events resulting in deregulating melanocyte proliferation but are not associated with inherited melanoma susceptibility.21
PRIMARY PREVENTION The majority of melanomas may be attributed to excessive sun exposure and therefore minimizing sun exposure is a suitable method of primary prevention. The most effective way of reducing sun exposure is achieved by shade, wearing sun-protective clothing and avoiding peak hours of sun intensity. This is especially important in childhood.22 The aim of screening for melanoma is to reduce mortality and morbidity. Intervention at an earlier stage in the natural history results in a better outcome for the patient. Methods range from self-examination to examination by a skin cancer specialist. Evidence for screening in melanoma is weak; there are no results available from randomized trials or cohort studies. A quantitative review that included 18 studies examining sunscreen use and melanoma failed to find any association between the use of sunscreens and the incidence of this disease.23 However it may take decades to detect a protective association between melanoma and use of the newer formulations of sunscreens. The Anti-Cancer Council SunSmart Campaign (Australia) implemented the ‘Slip! Slop! Slap!’ programme (slip on a shirt, slop on sunscreen, slap on a hat) and achieved substantial attitude
Pathology 881
shifts with a decrease in the proportion of sunburning, as well as improvements in hat wearing and sunscreen use.24 On the basis that there is a lack of empirical evidence for screening, the Australian and UK guidelines do not recommend this. The risk of malignant progression in an individual mole is very low and there is no role for prophylactic excision of benign naevi. The UK guidelines recommend that individuals with a previous primary melanoma or large numbers of moles, some of which are clinically atypical, should receive education about melanoma risk and be taught to self-examine for changing naevi.25 People with a strong family history of melanoma (greater than three cases in the extended family, two cases with AMS or multiple primary tumours in an individual) should be referred to a cancer genetics clinic for counselling and investigation. Those with giant hairy pigmented naevi require lifetime follow-up. High-risk individuals should be advised on the specific changes that suggest melanoma and encouraged to undertake monthly self-examination. Photography of lesions can be useful in helping to detect melanoma early. Chemoprevention is a strategy recently being explored in melanoma.26 Lipid-lowering drugs such as statins and fibrates have been under investigation. A meta-analysis failed to identify an effect of lipid-lowering drugs on the incidence of melanoma.27 In murine models of skin carcinogenesis, dietary administration of indometacin, piroxicam or celecoxib has been shown to prevent up to 85 per cent of UV-induced skin cancer.28 Polyphenols that are contained in green tea appear to provide photoprotection from ultraviolet light in the skin of humans and could represent a possible primary prevention strategy.29
DIAGNOSIS Early diagnosis of melanoma allows treatment to be undertaken at a stage when cure is likely to be achieved. The ‘ABCD’ system is widely used in public education and describes the four major clinical features of melanoma; Asymmetry, Border irregularity, Colour variation and Diameter greater than 6 mm. Melanoma can be located anywhere on the body. It most commonly occurs on the lower limbs in women and on the trunk in men. The incidence of non-pigmented melanoma is reported to be between 1.6 per cent and 10 per cent.30 The UK guidelines have adopted a seven-point scoring system to help non-dermatologists recognize early melanoma. Major features are: ● ● ●
change in size, irregular shape, irregular colour.
Minor features are: ● ●
largest diameter 7 mm or more, inflammation,
● ●
oozing, change in sensation.
Lesions with any of the major features or three minor ones are suspicious of melanoma and should be seen by specialists who are routinely treating large numbers of patients with pigmented lesions.25 Skin surface microscopy, dermatoscopy and dermoscopy are terms for a technique that utilizes the principle of epiluminescence to differentiate between benign and malignant skin lesions. The use of the handheld lens in combination with oil immersion allows visualization of morphological features that are not seen with the naked eye. The accuracy of diagnosing melanoma is increased if skin microscopy is combined with clinical examination.31,32 There have been efforts to improve the diagnostic accuracy by using computer-aided instrumentation, e.g. image analysis and spectrophotometric analysis.33 At this present time, computer-aided diagnosis of pigmented lesions is not robust enough for routine clinical application. Any significant change in a pigmented lesion is considered to be an indication for excision biopsy. The entire lesion (full-thickness) with a 2–5 mm clearance margin is excised for histological examination. Punch or incisional biopsies are not recommended because they may not provide accurate staging of tumour thickness. If performed, they should only be carried out by specialists with a particular expertise in melanoma. Shave biopsy and curettage are contraindicated.
PATHOLOGY Melanomas can be categorized on the basis of their macroscopic appearance: superficial spreading, nodular, acral lentiginous and lentigo maligna melanomas. The superficial spreading type makes up 70 per cent of all melanomas and often arises as a dark area within a pre-existing junctional naevus. Nodular melanomas (20 per cent of all melanomas) usually arise from apparently normal skin and are aggressive. These are usually darker than superficial spreading melanoma, uniform in colour and often domeshaped. Rarely, nodular melanomas are amelanotic. Acral lentiginous melanoma accounts for 5 per cent of melanomas and affects the palms, soles and nail beds. They are more common in dark-skinned older patients. Lentigo maligna melanoma (5 per cent of melanomas) tends to occur on the face or neck of older patients and overall survival is higher than other forms of melanoma. A detailed description of the histopathology of melanoma is beyond the scope of this discussion. The pathology report of an excised primary lesion should include margins of excision, Breslow thickness, Clark’s level, mitotic count, growth pattern, presence or absence of ulceration, tumour regression, lymphocytic inflammatory infiltrate and lymphatic or vascular invasion. Immunohistochemical stains widely used in melanoma include S-100, Melan-A and HMB45.
882 Malignant melanoma
Haematoxylin and eosin (H&E) staining is the gold standard for pathological analysis of the sentinel lymph node. Examination of nodal metastases by H&E alone can be difficult to interpret due to hypercellularity of the lymph node. Immunohistochemical staining for melanoma-associated tumour markers (e.g. S-100, MART-1) aids the detection of metastatic melanoma cells. In a retrospective analysis of 235 H&E negative sentinel lymph node biopsies from 94 patients with melanoma, deeper serial sections and immunohistochemical stains detected microscopic metastases in approximately 12 per cent of cases that would have been reported as negative for metastasis by H&E alone.34 The lymphatic drainage enters the node at the level of the subcapsular sinus, and the majority of nodal metastasis therefore involve the subcapsular area. In a study of 281 sentinel lymph node biopsies, 86 per cent involved at least the subcapsular region and 11 per cent were purely restricted to the parenchyma.35
STAGING AND PROGNOSIS Tumour stage at presentation is the most significant determinant of prognosis. The identification of increasingly
powerful prognostic factors has led to sequential modifications of the cutaneous melanoma staging system. The current system in use is the American Joint Committee on Cancer (AJCC) staging classification revised in 2002 (Table 38.1). This updated system takes into account the presence of ulceration and the implications of sentinel node technology. The correlation between Breslow thickness (total vertical height of the invasive lesion) and survival is better than Clark’s level (histological depth of dermal invasion) (Fig. 38.1). Level of invasion appears to be significant only in patients with tumor thickness less than 1 mm.36 Ulceration is a significant prognostic indicator. Stage I and II disease with and without ulceration is associated with 50 per cent and 79 per cent 10-year survival respectively.37 Other important adverse prognostic markers in localized melanoma are the presence of vascular invasion or lymphocytic infiltration, increased age and male sex. The growth pattern and anatomical site of the primary (head, neck and trunk worse than limb) also influence the prognosis. Other prognostic markers under investigation include Ki67,38 circulating melanoma cells,39 and serum S-100 protein.40 The number of positive lymph nodes is the most powerful prognostic factor in patients with nodal metastases.41
Table 38.1 American Joint Committee on Cancer (AJCC) staging (adapted from ref. 36) Stage 0 IA IB IIA IIB IIC IIIA IIIB
IIIC
IV
TNM
Features
Tis N0 M0 T1a N0 M0 T1b N0 M0 T2a N0 M0 T2b N0 M0 T3a N0 M0 T3b N0 M0 T4a N0 M0 T4b N0 M0 T1–4a N1a M0 T1–4a N2a M0 T1–4b N1a M0 T1–4b N2a M0 T1–4a N1b M0 T1–4a N2b M0 T1–4a or b N2c M0 T1–4b N1b M0 T1–4b N2b M0 Any T N3 M0
In situ 1.0 mm or less without ulceration 1.0 mm or less with ulceration 1.01–2.0 mm without ulceration 1.01–2.0 mm with ulceration 2.01–4 mm without ulceration 2.01–4 mm with ulceration 4 mm without ulceration 4 mm with ulceration 1 micrometastatic node, non-ulcerated primary 2–3 micrometastatic nodes, non-ulcerated primary 1 micrometastatic node, ulcerated primary 2–3 micrometastatic nodes, ulcerated primary 1 macrometastatic node, non-ulcerated primary 2–3 macrometastatic nodes, non-ulcerated primary In transit or satellite met(s) 1 macrometastatic node, ulcerated primary 2–3 macrometastatic nodes, ulcerated primary 4 or more metastatic nodes or in transit/satellite metastases with metastatic nodes Distant skin, subcutaneous or nodal metastases, normal LDH Lung metastases, normal LDH All other visceral metastases with normal LDH or any distant metastasis with elevated LDH
Any T Any N M1a Any T Any N M1b Any T Any N M1c
T – primary lesion ‘a’ without ulceration; ‘b’ with ulceration. N – nodal disease ‘a’ micrometastatic’; ‘b’ macrometastatic. M – distant metastases ‘a’ distant skin, subcutaneous or nodal; ‘b’ lung; ‘c’ all other visceral sites or elevated LDH. LDH – lactose dehydrogenase.
Staging and prognosis 883
Survival correlates with the number of metastatic nodes involved regardless of whether the metastatic nodes are clinically apparent (macrometastases) or clinically occult (micrometastases) (Fig. 38.2). The presence of intralymphatic metastases as indicated by the presence of satellite and/or in-transit metastases reflects poor outcome.41 The presence of melanoma cells in the sentinel node in patients with clinically negative nodes is an important prognostic factor for recurrence.42 The rationale for sentinel node biopsy (SNB) is that within a given lymph node basin, one or more nodes are the first to be contaminated with metastatic disease, and reflect the status of the entire basin. Intraoperative intra-dermal injection of blue dye and radiolabelled colloid allow the tracing of lymphatic channels to the sentinel lymph node(s). Sentinel nodes are successfully identified in 90–100 per cent of patients using this combined 100
95
91
89 77
% 5 year survival
technique. Metastases in the sentinel node are detected in 4–35 per cent of patients with stage I or II melanoma,42,43 (Fig. 38.3). Reported recurrence (false negative) rates in sentinel node-negative lymph node basins have ranged from 4 to 9 per cent with median follow-up durations from 23 to 72 months.42,43 Studies have suggested that the microanatomic location of the metastatic deposit within a sentinel lymph node may predict the likelihood of involved non-sentinel lymph nodes. The microanatomic location of metastatic melanoma in sentinel lymph nodes of 146 patients was analysed and correlated with the presence of involved non-sentinel nodes in the complete lymph node dissection (CLND) samples. Patients with isolated subcapsular micrometastases had a significantly lower rate of involvement in the complete dissection specimen than those with micrometastases in other locations. These results suggest
79
40 63
34.4
67 30 45
50
19.2
20 10 0
1
1.01–2
2.01–4
0
Breslow thickness Non-ulcerated primary
1*
Ulcerated primary
1.5
1.51–4
4
Breslow thickness
Figure 38.1 Five-year overall survival from AJCC Melanoma Staging Database36 comparing different Breslow thickness (mm) for primary melanoma (stage IA–IIC).
Figure 38.3 The incidence of positive sentinel lymph nodes (SLNs) according to Breslow thickness (mm) for primary melanoma. (Adapted from ref. 42) *Ulceration or Clark level IV.
100
100 70 63 53
50
50
0
% 5 year survival
% 5 year survival
4.8
4.7
4
59 46
50 29
24
27
0 1 2–3 No. of micrometastatic nodes (a)
1 2–3 Other No. of macrometastatic nodes (b)
Non-ulcerated primary Ulcerated primary 4 metastatic nodes, matted nodes or intransit metastases/satellite(s) with metastatic node(s)
Figure 38.2 Five-year overall survival from AJCC Melanoma Staging Database36 comparing micrometastatic (a) and macrometastatic (b) nodal melanoma (stage IIIA–IIIC).
884 Malignant melanoma
100
% survival
that for patients with only subcapsular deposits in the sentinel node, CLND is not necessary.44 Reverse transcriptase polymerase chain reaction (RTPCR) assay for molecular detection of melanoma-specific tumour markers such as tyrosinase messenger RNA in lymph node tissue may be more sensitive for the detection of occult metastases than routine histology. However, benign subcapsular naevi can cause false positives with this technique. The prognostic value of sentinel lymph node analysis by RT-PCR for tyrosinase has been investigated. In a study of 114 patients who underwent sentinel node biopsy, 91 patients had histologically negative nodes; of these, 47 patients were RT-PCR positive for tyrosinase m-RNA. In the group of patients with histologically negative sentinel nodes, RT-PCR positivity was associated with a significantly higher risk of recurrence than RT-PCR negativity (13 per cent vs. 2 per cent). The recurrence rate was much higher (61 per cent) in patients who were positive on both histological and RT-PCR examinations.45 However, another study failed to demonstrate that detection of microscopic disease by tyrosinase RT-PCR defined a subgroup with higher recurrence risk when compared with patients with RT-PCR-negative sentinel lymph nodes.46 Five-year disease-specific survival for patients with negative and positive sentinel nodes is 88 per cent and 71 per cent respectively.47 Accurate staging by determining pathological nodal status in patients without clinically obvious lymphatic spread may identify patients with primary lesions who have a poor prognosis. These patients may be eligible for adjuvant therapy in the context of clinical trials. The Multicenter Selective Lymphadenectomy Trial (MSLT-1) addressed the impact of sentinel node biopsy on survival. Patients with primary melanoma were randomly assigned to wide local excision and postoperative observation or wide local excision and sentinel node biopsy. Patients with a positive biopsy underwent immediate lymphadenectomy. This trial confirmed the prognostic role of sentinel node biopsy but failed to show an overall survival benefit for the sentinel node biopsy procedure.47 Patients that undergo sentinel node biopsy are still at risk of recurrence and require careful follow-up. The psychological impact of a positive sentinel node biopsy on an individual patient needs to be considered. The message a patient will take from the consultation is that a positive biopsy result means the likelihood of a melanoma recurrence at 5 years is 30 per cent and there is, at present, no treatment that can alter this outcome. The present role of the sentinel node biopsy thus remains as a staging procedure in clinical trials as there is still no proven overall survival benefit for any adjuvant therapy. MSLT-II will examine the value of CLND versus sentinel node biopsy alone on sentinel node-positive patients. The Sunbelt Melanoma Trial will be determining the place of molecular staging (RT-PCR) in patients who have had sentinel node biopsy and the role of adjuvant interferon in patients with one microscopically positive lymph node.
59
57
50
41 19 10
7 0 Mla Mlb 1 year overall survival
Mlc 5 year overall survival
Figure 38.4 One-year and 5-year overall survival from AJCC Melanoma Staging Database36 comparing different M categories (stage IV). (Refer to Table 38.1 for definitions.)
Other methods of identifying sentinel node status are under investigation. For instance, the use of ultrasound and fine needle aspiration (FNA) to screen lymph node basins for non-palpable metastatic disease may identify those patients with sentinel node involvement and thus a worse prognosis.48,49 Prognostic features in stage IV melanoma are the site of metastases, number of metastases and elevated serum lactic dehydrogenase (LDH). LDH represents the most powerful prognostic factor in patients with disseminated melanoma50 (Fig. 38.4).
INITIAL STAGING INVESTIGATIONS Melanoma is an unpredictable disease and can metastasize to a number of sites throughout the body. Imaging methods are very important for the staging and follow-up of patients with this disease. There is little evidence to support the use of radiological investigation in stage I and IIA melanoma. Patients with moderate or high risk of recurrent disease (AJCC stage IIB and III) should have baseline blood tests (full blood count, liver function tests, LDH), chest X-ray and liver ultrasound or computed tomography (CT) of the chest, abdomen / pelvis (CT neck if head and neck primary). Further investigations such as bone scan or magnetic resonance imaging (MRI) of the brain should be undertaken if clinically indicated. In patients with stage IV disease, the likelihood of detecting further metastases is significant and therefore the investigations listed above for stage III disease should be performed. In the UK, imaging of the brain is not part of the routine initial staging investigations, although some centres recommend this. There is an increased interest in applying positron emission tomography (PET) scans to the clinical management of melanoma. However, an established role for PET scanning in melanoma is unclear at the present time. The lack of sensitivity of PET scanning for lesions smaller than 1 cm
Treatment of primary melanoma 885
limits its use in local (stage I or II) disease. PET scanning is more sensitive but as specific as conventional imaging modalities such as CT and MRI in advanced melanoma.51,52 Some have suggested that PET scanning has a role in staging patients with melanoma and it may alter the therapy plan in a proportion of cases. In a series of 100 patients with stage IV melanoma that underwent a range of imaging modalities to detect metastatic lesions, PET scanning showed a sensitivity rate of 93 per cent. In 20 patients, PET detected 24 metastases up to 6 months earlier than conventional imaging or physical examination.53 Furthermore, the selection of surgical and medical treatment was said to have been influenced by the PET findings in 22 of these patients. PET scanning may be helpful in patients with an apparently isolated pulmonary metastasis prior to planned resection of the metastasis.54 If the PET study shows additional sites of disease, local resection may not be indicated. PET may also be useful to clarify the nature of a suspicious lesion identified on CT scan. However, the use of PET scanning in known metastatic melanoma should be restricted because effective therapeutic options available for stage IV disease are very limited.
TREATMENT Primary melanoma The definitive surgical treatment for primary cutaneous melanoma is a wide local excision. The recommended excisional margin is dependent upon the thickness of the melanoma. There are four randomized prospective trials that document equivalent rates of freedom from local recurrence with narrow (1–2 cm) versus wider (3–5 cm) excisional margins in thin (2 mm) and intermediate melanomas (0.75–4 mm).55–58 These results support the view that for thin melanomas 2 cm margins are recommended. The American Guidelines have recommended that margins of 1 cm are adequate for thin melanomas. This is based on the results of the WHO trial57 which assigned patients with thin melanomas to 3 cm or 1 cm excision margin. No overall survival difference was seen at 12 years. Despite the lack of a survival difference, the WHO recommended that the use of 1 cm margins should be limited for those patients with melanomas 1 mm thick. The situation for melanomas between 2 and 4 mm in thickness is less clear. The Melanoma Intergroup Trial randomly assigned 468 patients with intermediate thickness (1–4 mm) melanomas to 2 cm or 4 cm excision margins.56 At a median of 8-year follow-up, the overall survival (p 0.07) and recurrence rates were similar in the two groups. In contrast, the UK Melanoma Study Group/British Association of Plastic Surgeons Trial reported higher local recurrence rates but equivalent overall survival with 1 cm versus 3 cm margins in patients with melanomas 2 mm in thickness.59 This study included a subset of patients with thick melanoma (4 mm). The only other published study
Table 38.2 Recommended surgical excision margins UK25 BT Margins (mm) (cm) In situ 1 1–2 2.1–4.0 4
0.2–0.5 1 1–2 2 2–3
USA61 BT Margins (mm) (mm)
Australia62 BT Margins (cm) (mm)
In situ 2 2
In situ 0–1.5 1.54–4.0 4
0.5 1 2
0.5 1 1–2 2–3
BT, Breslow thickness.
for patients with thick primary melanomas is a retrospective review of 278 patients with a median tumor thickness of 6 mm.60 No correlation between the width of resection (2 cm vs 2 cm) and local recurrence or overall survival was seen. There has not been a study directly comparing 2 cm with 3 cm margins for patients with thick primary melanomas. The United Kingdom Study Group recommends a 2–3 cm margin with 2 cm being the preferred option.25 The American, Australian and UK guidelines are based on the interpretation of the current evidence and there are some differences on the recommended minimum safe margin of excision reflecting different international interpretations of the data.25,61,62 However, the general guidelines are similar and are summarized in Table 38.2. Special consideration is required for primary melanoma of certain sites. For example, in many head and neck primaries, the morbidity of surgery associated with achieving adequate clearance has dictated a more conservative approach. Many centres treat such cases with radiotherapy. Superficial lentigo maligna (confined to the epidermis) and lentigo maligna melanoma (invasive into the dermis) predominantly occur in the elderly and on the face. They tend to have a slow growth rate and low metastatic potential and have been treated successfully with radiotherapy.63,64
Regional nodes THERAPEUTIC LYMPH NODE DISSECTION (TLND)
The presence of palpable nodes, usually confirmed on fine needle aspiration biopsy, is an indication for node clearance within that basin. The aims of TLND are local control and possible cure. Outcome is influenced by the number of lymph nodes involved and evidence of extracapsular spread.65 There is controversy over whether patients with clinically involved superficial groin nodes should undergo a superficial dissection alone, or an en bloc dissection which includes external iliac and obturator nodes.66 A conservative approach involves a superficial inguinal dissection followed by iliac node dissection if Cloquet’s node (junction
886 Malignant melanoma
of deep and superficial groups) is histologically positive. This strategy is criticized by some who feel that all patients with clinically involved groin nodes should undergo deep as well as superficial dissection and that the presence of disease in Cloquet’s node is unreliable as an indicator for iliac node involvement. In patients with axillary node involvement, complete dissection to include level III nodes medial to pectoralis minor is indicated. The main complication of lymphadenectomy is limb oedema. This is more common in the lower limb, especially in patients who have had an ilio-inguinal block dissection. ELECTIVE LYMPH NODE DISSECTION (ELND)
The rationale for removal of subclinically involved lymph nodes by ELND is that the majority of melanomas develop secondary spread via the lymphatics and the most common site of first relapse is the loco-regional lymph node basin. ELND for clinically node-negative (stages I and II) patients was controversial. The low yield of metastases in these patients and the substantial morbidity of radical regional node dissection limited the potential value of this approach. Prospective randomized studies, however, have failed to support a survival benefit from ELND and it is now accepted that there is no role for ELND.67,68 The accepted surgical treatment approach for patients with a positive sentinel node biopsy is radical dissection of the involved nodal basin.
Adjuvant therapy Patients with thick primary lesions or lymph node involvement are at high risk of relapse. The survival of patients with metastatic melanoma is poor. Therefore there is considerable interest in developing adjuvant therapies. LOCAL TREATMENT
There are no randomized trials showing an overall survival benefit using regional therapies such as elective lymph node dissection, isolated limb perfusion or radiotherapy.69 Uncontrolled series have suggested a benefit for adjuvant radiotherapy in decreasing the likelihood of local recurrence after complete resection, but this does not impact on survival, in part because of the high rate of systemic relapse in these patients.70,71 Treatment-related limb lymphoedema is common following radiotherapy.72 CYTOTOXIC CHEMOTHERAPY
Randomized trials of cytotoxic chemotherapy using dacarbazine (DTIC), nitrosoureas, vinca alkaloids, cisplatin and paclitaxel have failed to show any consistent survival benefit. Neoadjuvant chemo-immunotherapy does have potential
benefit for shrinking the primary tumour prior to surgery and for assessing in vivo sensitivity. This type of neoadjuvant approach may be applicable to use in the setting of bulky nodal disease. Good response rates have been shown in a phase II study of preoperative cisplatin, vinblastine, DTIC with IL-2 and interferon-alpha (IFN-α).73 INTERFERON-ALPHA (IFN-α)
The most mature and promising results for adjuvant therapy involve IFN-α. IFN-α is a cytokine and has direct anti-proliferative activity, indirect anti-tumour effects mediated through immune response modulation and anti-angiogenic effects. The mechanisms of action include activation of natural killer (NK) cells, cytotoxic T-lymphocytes (CTL) and macrophages and the up-regulation of antigens, in particular major histocompatibility complexes (MHC). The effects depend on the concentration of IFN-α. There is, however, no consistent evidence that any dosing schedule of IFN-α has an impact on overall survival74–85 (Table 38.3). There are only two trials that show a significant impact on disease-free survival (DFS) and overall survival (OS) at 5-year follow-up.74,76 In both trials, however, the impact on OS is lost with longer follow-up. The Eastern Co-operative Oncology Group (ECOG) trial 168476 compared high-dose interferon (HDI) to observation. A total of 287 patients with node-positive disease or primaries thicker than 4 mm were randomized to treatment with intravenous IFN-α (20 MU/m2 daily) for 1 month followed by maintenance IFN subcutaneously for 1 year (10 MU/m2 3 times per week) or observation after surgery. A significant improvement in relapse-free survival (1.0–1.7 years, p 0.002) and overall survival (2.8–3.8 years, p 0.02) were shown. However, re-analysis with longer follow-up failed to demonstrate an overall survival benefit although the improvement in relapse-free survival remained (p 0.02).77 Significant toxicity was experienced, requiring dose reduction in the majority, with two treatment-related deaths. The follow-up trial, E1690, failed to demonstrate an overall survival benefit for HDI.74 It has been suggested that this may be because patients on the observation arm were salvaged at relapse by HDI. However, only a minority of patients crossed over to HDI suggesting a bias in this patient population. Another Intergroup trial (E1694) randomized patients to HDI or the ganglioside GM2 vaccine in resected stage IIB and III melanoma.75 Both disease-free (HR 1.47) and overall survival (HR 1.52) were significantly superior in the IFN-a group. The trial was unblinded and reported early at a median follow-up of 16 months. The updated analysis of this trial at median follow-up of 2.1 years confirmed the superiority of HDI to the ganglioside vaccine in terms of relapse-free survival (HR 1.33, p 0.006) and OS (HR 1.32, p 0.04). A pooled analysis of the ECOG and Intergroup trials of adjuvant HDI confirmed the beneficial effects of HDI improving relapse-free survival compared with no treatment.77 Similarly, the
Treatment of primary melanoma 887
Table 38.3 Randomized trials of adjuvant interferon versus observation following surgery alone. Trial reference
AJCC stage
Number of patients
ECOG 168476,77 ECOG 169074,77 Intergroup trial E169475,77 NCCTG78 Austrian trial79 EORTC 1895280,86 Scottish trial81 French trial82 WHO trial 1683 UK-MCG AIM-HIGH84 EORTC 18871/DKG85
IIB, III IIB, III IIB, III IIA, IIB, III IIA, B IIB, III IIA, B IIA, B III IIB, III IIA, IIB, III
262 311 1418 95 499 424 674 830
EORTC 1899188
III
1256
280 642 880
Treatment
Result
HDI IV then MD SC 1 yr HDI IV then MD SC 1 yr vs. LDI SC 2 yr HDI IV then MD SC vs. *GM2 vaccine
DFS p 0.02 No difference HDI superior DFS p 0.006 OS p 0.04 No difference DFS p 0.02 No difference No difference DFS p 0.04 No difference No difference No difference
HDI IM 3 months LDI SC 1 yr IDI 10 MU 1 yr vs. 5 MU 2 yr LDI SC 6 months LDI SC 18 months LDI SC 3 yr LDI SC 2 yr LDI SC 1 yr vs. IFN-γ SC; 1 yr vs. Iscador M PEG-IFN SC 5 yr
RFS p 0.01 In patients with microscopic nodal involvement, RFS p 0.02; DMFS p 0.03
*interferon vs. non-observation arm; HDI, high-dose interferon; MD, medium-dose interferon; IDI, intermediate-dose interferon; LDI, low-dose interferon; IV, intravenous; SC, subcutaneous; OS, overall survival; DFS, disease-free survival; RFS, relapse-free survival.
pooled analysis did not show an overall survival benefit with HDI. There is no consistent impact on overall survival demonstrated for low-dose interferon (LDI) although trials have demonstrated a transient beneficial effect on disease-free survival that is lost when treatment is stopped. The French Cooperative Group Trial investigated 18 months of LDI (3 μg subcutaneously three times a week) versus observation after surgery. The significant effect on overall survival seen at 5 years was lost at 8 years’ follow-up. The EORTC trial 18952 evaluated intermediate doses of IFN- and early analysis indicated that duration of treatment was an important factor. Treatment with 10 MU interferon for 1 year had no effect on distant-metastasis-free survival (DMFI) whilst the lower dose of 5 μg for 2 years showed a significant impact on DMFI (p 0.0145).80 However, this benefit was lost with longer follow-up.86 A meta-analysis of 12 trials confirmed that adjuvant IFN- reduces recurrence of melanoma in high-risk patients but does not impact on overall survival. This beneficial effect on recurrence-free survival occurs at all doses of IFN- except for the very low dose treatments.87 The EORTC trial 18991 evaluated the impact of 5 years of pegylated interferon (PEG-IFN alpha-2b) as long-term maintenance therapy in stage III melanoma.88 PEG-IFN alpha-2b had a significant impact on relapse-free survival (HR 0.82, p 0.01) but not on distant metastasis-free survival or overall survival.88 Importantly, for patients with only microscopic nodal involvement, PEG-IFN improved both distant metastasisfree survival (HR 0.75 p 0.03) and relapse-free survival
(HR 0.73 p 0.02). No such benefit was seen in patients with palpable nodes. Prospective clinical trials in patients with microscopic disease are needed to confirm this observation. The optimal dose and duration of adjuvant IFN-α remains unknown. High-dose IFN- has become approved as standard therapy in the USA. Most groups in Europe do not use adjuvant therapy with IFN- outside of the clinical trial setting. Ongoing trials include the ECOG trial 1697, which evaluates the effect of high dose IFN- for 4 weeks. VACCINATION APPROACHES
There is evidence that the immune system plays a major role in the natural history of melanoma. Stimulation of the immune system by vaccination is a potential effective adjuvant therapy with minimal morbidity. Current vaccine strategies exert their effects by increasing cellular and humoral host responses, enhancing the antigen-presenting ability of dendritic cells and promoting resistance to immunosuppressive factors secreted in melanoma. A response to a tumour vaccine takes longer to develop than to chemotherapy (4–8 weeks) but may be more durable. Toxicity is low and is usually confined to local irritation at the injection site. The development of effective vaccines has been hindered by a combination of poor immunogenicity of relevant antigens, tumour-reinforced tolerance, antigen shedding, and tumour heterogeneity. Cell-derived vaccines may be allogeneic or autologous. Allogenec vaccines are prepared from cultured cell lines
888 Malignant melanoma
and are therefore readily available and can be used for patients with low disease burden. Examples of allogenec vaccines are Canvaxin (CancerVax), Melacine, shed antigens, vaccinia oncolysates and dendritic cell vaccines. Autologous vaccines are made up of material from the patient’s own tumour, which is treated to improve immunogenicity. This method ensures that the target tumour expresses the antigens present in the vaccine. A significant tumour volume is required in order to prepare the vaccine and therefore this technique is limited to treating patients with bulky disease amenable to resection. Examples of autologous vaccines include hapten-conjugated (DNP) vaccines and heat shock protein vaccines. Several antigens associated with melanocytic lineage have been identified and have allowed defined antigen vaccines to be developed, e.g. ganglioside and peptide vaccines. Ganglioside GM2 is expressed on most melanoma cells and has been used as a melanoma vaccine. The importance of ganglioside as a target for immunotherapy is illustrated by the ability to induce antibody responses and the association between the de novo presence of antibody to GM2 ganglioside and prolonged survival in patients with melanoma.89 The GM2 antigen has been shown to induce antibody production more effectively when linked to the carrier protein keyhole limpet haemocyanin (KLH). Peptide melanoma antigens recognized by T-cells have been identified (MAGE, BAGE, CAGE proteins, tyrosinase, gp100 and MART-1) and can produce clinical responses when processed by dendritic cells. Peptide vaccines are delivered
with an immune adjuvant such as BCG or cytokines such as IL-12 or GM-CSF to stimulate the immune response. The results of eight randomized trials on the adjuvant use of melanoma vaccines have shown no significant improvement in survival77,90–95 (Table 38.4). There is encouraging evidence of a survival benefit in a subset of patients including those that develop a host response. The Melacine vaccine consists of a lysate of two homogenized melanoma cell lines in combination with the adjuvant Detox. The Southwest Oncology Group conducted a phase III trial (SWOG 9035) of the Melacine vaccine in 689 patients with stage II disease after surgery.94 Vaccine therapy was not associated with a significant benefit in DFS or overall survival. Further analyses of the results suggest that the Melacine vaccine might have a preferential impact in patients who were HLA-A2 or -C3 (5-year recurrencefree survival for vaccinated and observed patients 77 vs. 64 per cent, respectively, p 0.004).97 Results from a trial of the melanoma lysate vaccine plus low-dose IFN-α compared with high-dose IFN-α in resected stage III melanoma demonstrated no survival difference.98 Canvaxin consists of three irradiated melanoma cell lines that contain tumour-associated antigens. The immune response following Canvaxin vaccine therapy correlated with outcome.99,100 The vaccine induced TA90 antigen– antibody immune complex in melanoma patients with subclinical disease. TA90 seroconversion was associated with improved disease-free and overall survival. Recent preliminary results from a multi-centre randomized trial of
Table 38.4 Randomized trials of vaccines (against no active treatment) used as adjuvant treatment in melanoma Reference
Vaccine
Morton et al95
Melanoma cell
Livingston et al90
GM2/BCG
Bystryn et al91
Shed antigen
Wallack et al93 Kirkwood et al77
Vaccinia melanoma oncolysate GM2
Hersey et al96
Vaccinia lysate
Sondak et al94
Melacine
Morton92
Canavaxin BCG
AJCC stage and trial design
Number of patients
Survival
III BCG vaccine vs. vaccine alone III BCG Cy vs vaccine Cy III placebo vs. vaccine III vaccinia alone vs. vaccine IIB, III HDI vs. vaccine IIB, III observation vs. vaccine IIA observation vs. vaccine IIB, III HDI vs. BCG alone vs. vaccine
134
No difference
122
In patients that developed GM2 antibodies DFS p 0.02
38
No difference
217
No difference
880
Significant difference in favour of HDI DFS p 0.006 OS p 0.04 No difference
700 689 Trial discontinued
HDI, high-dose interferon; DFS, disease-free survival; OS, overall survival; Cy, cyclophosphamide.
In HLA-A2 or HLA-C3 positive patients DFS p 0.04 No difference
Treatment of primary melanoma 889
Canvaxin plus BCG (bacillus Calmette–Guérin) versus placebo plus BCG in patients without evidence of residual disease following resection of stage III or stage IV melanoma have shown no survival difference.92 The study was terminated after the interim analysis. In another double-blind trial, 38 patients with stage III disease were randomly assigned to a polyvalent shed-antigen melanoma vaccine or to a placebo vaccine.91 Although the melanoma vaccine was associated with a significant improvement in time to disease progression (1.6 vs. 0.6 years), there was no difference in survival. Two trials of viral (vaccinia) oncolysate vaccines failed to demonstrate a survival advantage.93,96 Modification of an autologous vaccine with the hapten dinitrophenyl (DNP) increases immunogenicity. A trial of this vaccine (M-Vax) in 214 patients with stage III melanoma (post lymphadenectomy) showed that patients who developed delayed-type hypersensitivity (DTH) had a significantly longer relapse-free survival (RFS) and overall survival (OS) than DTH-negative patients (OS, 59.3 per cent vs. 29.3 per cent respectively; p 0.001).101 A trial with ganglioside GM2 demonstrated a benefit in a subset analysis of stage III patients who were seronegative for ganglioside antibodies prior to study entry.90 This randomized phase III trial involved 122 patients who had recently undergone resection of regional node metastases and compared the GM2/BCG vaccine to BCG alone. Eightyfive per cent of patients in the GM2/BCG arm developed IgM antibodies to the ganglioside, while 6 patients (5 per cent) (5 from the BCG alone arm) had pre-existing antiGM2 antibodies. If these 6 patients were excluded from the analysis, administration of the GM2/BCG vaccine was associated with a significant improvement in both RFS (23 per cent, p 0.02) and a non-significant 14 per cent improvement in overall survival. The E1694 randomized trial compared the effects of HDI to GM2 vaccine in resected stage IIB and III melanoma.75 These results have been discussed earlier and showed superiority in the IFN-α group. Trials in progress include the EORTC trial 18961 that compares the effects of the GM2 vaccine with untreated controls in stage II patients. The Canvaxin plus BCG vaccine is being compared to HDI and BCG alone. The phase III trial ECOG E4697 is evaluating a multiple antigen peptide vaccine (MART-1, tyrosinase, gp100) alone or administered with GM-CSF compared to placebo or GM-CSF alone in high-risk stage III and resected stage IV patients. It is hoped that these trials will help clarify the benefit of these vaccine preparations.
Loco-regional recurrent disease Recurrent local disease represents any tumour occurring within 5 cm of a previous primary closure or skin graft. In-transit metastases represent lymphatic permeation towards the regional lymph nodes and develop in the subcutis or in the skin between the primary and the regional lymph
basin. The majority of local recurrences occur within 5 years. The risk of recurrence is dependent on the thickness of the primary lesion – low for tumours less than 0.75 mm thick (0.2 per cent) and significantly higher for primary tumours greater than 4 mm (13 per cent). Loco-regional disease control is an important quality of life issue because of pain and wound complications. Several palliative management options are available but there has been no randomized comparison of these alternatives. Surgery is the treatment of choice for loco-regional recurrences. Margins of 2 cm are recommended to achieve clear margins but this is not always practical.102 Isolated limb perfusion (ILP) is the preferred treatment when the disease becomes beyond surgical control, i.e. multiple, frequent or large recurrences. ILP is a technique using an extra-corporeal bypass circuit via the main artery and vein to the target limb and allows delivery of higher doses of antineoplastic agents than are tolerated systemically. The treatment period is about 90 minutes and is combined with limb hyperthermia (38–40°C). Good response rates are achievable with ILP (54–80 per cent).103,104 Melphalan is the most commonly used agent; cisplatin and dactinomycin have also been used. DTIC is not useful in this situation, as it requires hepatic conversion to its active metabolite. In a study of 114 patients with locally recurrent or metastatic limb melanoma, treatment with hyperthermic ILP with melphalan resulted in a complete and partial response in 73 per cent and 13 per cent, respectively. Complete response was continued in 37 of 81 cases at a median follow-up of 33 months.103 Combinations of melphalan, TNF-α and IFN-α have been investigated.105,106 A phase II trial of ILP in patients with stage III melanoma and intransit metastases treated with melphalan and TNF-α, with or without IFN-γ, showed no significant difference in response rate or survival.106 The data from this study were compared to historical controls treated with melphalan alone and this comparison suggested that ILP treatment with the combination of TNF-α and melphalan might be superior to melphalan alone. A retrospective analysis of 130 patients treated with melphalan alone or with melphalan and TNF-α demonstrated no additional benefit with the addition of TNF-α.105 The phase III ACOSOG trial Z0020 in which patients with locally advanced extremity melanoma were randomly assigned to ILP with melphalan or melphalan and TNF-, demonstrated no additional benefit with the addition of TNF-.107 Treatment complications can be severe and include acute respiratory distress syndrome and acute renal failure due to 5 per cent of the perfusate escaping into the systemic circulation. Response durations are typically short and patients typically develop distant metastases regardless of whether local control is achieved. ILP is therefore given with palliative rather than curative intent. ILP is a complex technique and therefore other, simpler techniques to treat locally recurrent melanoma have been investigated. Isolated limb infusion (ILI) describes intraarterial infusion of a cytotoxic agent into the affected limb
890 Malignant melanoma
via a percutaneous catheter. ILI is a less invasive procedure than ILP. Response rates of up to 85 per cent with ILI and median response durations comparable to that achieved by ILP have been reported.108,109 It has been suggested that ILI is a therapeutic option for patients above 70 years who might not be suitable for ILP.109 Laser ablation (CO2 laser) is useful for treating multiple small (1.5 cm) lesions.110 Recovery following treatment is much quicker than with ILP. The recurrence rate at treated sites is 2 per cent. Further crops of lesions are likely and can be re-treated with the laser. Melanoma has been historically considered a ‘radioresistant’ disease. However, radiation can be very effective in some clinical situations. A large dose per fraction (4Gy) may be more effective for superficial skin metastases. Radiotherapy is an alternative to surgery for managing symptomatic skin metastases.
Distant metastatic disease Patients with metastatic melanoma have a poor prognosis with a median survival of 6–10 months and less than 5 per cent of patients survive for more than 5 years.3 There is no consistent randomized evidence for a significant improvement in overall survival with any treatment in patients with metastatic melanoma. However, response to treatment can result in good palliation and therefore the focus of patient care is disease palliation rather than cure. There is an urgent need for more effective systemic therapies in patients with metastatic melanoma and patients should be offered participation in clinical trials where possible. Withholding treatment is a valid option in the presence of slow-growing asymptomatic metastases because deferment of treatment may allow quality of life to be maintained. In patients with poor prognosis and widespread disease, treatment is unlikely to be of benefit. SURGERY AND RADIOTHERAPY
Surgery and radiotherapy offer valuable palliation in certain circumstances, for example bone metastases. Palliation of painful skeletal metastases can be obtained using radiotherapy in 50–86 per cent of patients.111,112 No particular fractionation technique has proven more effective. Surgery is a valuable option for those with surgically resectable disease in up to three visceral sites. Following complete resection of pulmonary metastases, 5-year survival rates of 20 per cent have been reported113 and 28–41 per cent after complete resection of gastrointestinal metastases.114,115 However, these patients are highly selected and these data need to be interpreted with great caution. There are no randomized data to address the effectiveness of such interventions. The roles of surgery and radiotherapy in the management of cerebral metastases will be discussed later.
CHEMOTHERAPY
No significant improvement in overall survival with chemotherapy has ever been demonstrated. Some cytotoxic drugs, however, do have activity in metastatic melanoma and these include dacarbazine (DTIC), nitrosoureas, platinum compounds, vinca alkaloids, taxanes, temozolomide and fotemustine. Response rates of 10–15 per cent are observed with nitrosoureas, platinum compounds, vinca alkaloids and taxanes. The most commonly used single agent for metastatic melanoma is the alkylating agent, DTIC, with best response rates in the range 15–20 per cent in phase II studies and a median duration of response of up to 6 months. Patients with skin, soft tissue and lung metastases are more likely to achieve a response than those with other visceral sites of disease. Complete responses are uncommon. DTIC is ineffective in cerebral metastases because it does not cross the blood–brain barrier. DTIC is well tolerated with minimal toxicity when given with 5-hydroxytryptamine antagonists. Principal side effects are nausea, vomiting, flulike symptoms, fever, moderate myelosuppression and occasional hepatotoxicity. DTIC can usually be administered as an outpatient treatment and is typically given as a single dose of 850 mg/m2 intravenously over 1 hour every 3 weeks. Currently, DTIC remains the standard option of care for patients with metastatic melanoma. Temozolomide is another alkylating agent. It is a prodrug of monomethyl-5-triazenoimidazole carboxamide (MTIC), the active metabolite of DTIC. Temozolomide is given orally, has 100 per cent bioavailability and has good central nervous system penetration. The first phase II study demonstrated 3 complete responses and 9 partial responses in 49 evaluable patients, resulting in an overall response of 21 per cent.116 A randomized phase III trial of temozolomide versus DTIC in 305 patients with metastatic melanoma (brain metastases excluded) reported response rates of 12.1 per cent and 13.5 per cent with temozolomide and DTIC respectively. This study showed a significant improvement in progression-free survival (1.9 months vs. 1.5 months, p 0.012) in favour of the temozolomide group but no improvement in overall survival. It was therefore concluded that temozolomide is at least as effective as DTIC.117 Quality of life was improved in the temozolomide group. Temozolomide is well tolerated with nausea, vomiting, constipation and mild myelosuppression being the most frequently observed toxicities. Given that temozolomide can be administered orally, it is a convenient drug for palliative treatment of patients with metastatic melanoma. The use of temozolomide and fotemustine in CNS disease will be discussed later. In order to improve the effectiveness of treatments, combination chemotherapy has been studied with agents including cisplatin, vinblastine and DTIC. Promising activity with response rates of 30–40 per cent have been reported in single institution phase II trials but subsequently not confirmed in multi-centre trials. The Dartmouth regimen is an example of combination chemotherapy using DTIC, BCNU
Treatment of metastatic melanoma 891
(carmustine), cisplatin and tamoxifen. Phase II studies of this combination range from 20 to 60 per cent.118–120 A phase III study of 240 patients randomized between DTIC alone and the Dartmouth regimen demonstrated no significant difference in response rates (10.2 vs. 18.5 per cent, p 0.09), or overall survival (6.3 vs. 7.7 months). Toxicities were more common and severe in the combination arm. A significant difference was seen in favour of combination chemotherapy for soft tissue disease (32 per cent vs. 14 per cent, p 0.05).121 High-dose chemotherapy has been explored using single agents and combination. Toxicity was significant and treatment-related deaths occurred in up to one third of patients in the very high-dose protocols. Therefore highdose chemotherapy remains experimental in the absence of randomized evidence of superiority over conventional doses. In the absence of evidence on improved quality of life or survival, chemotherapy should be restricted to symptomatic metastatic disease and patients with impending symptoms due to significant tumour bulk. Symptomatic improvement and overall quality of life should be the main criteria to assess chemotherapy effect. Second-line chemotherapy should be confined to the context of clinical trials because no agent has shown consistent activity following failure of first-line therapy with DTIC. IMMUNOTHERAPY
The host immune response is thought to play an important role in melanoma. Therefore there has been considerable interest in the use of biological response modifiers IFN-α and interleukin-2 (IL-2) and vaccines. Interferon-α An overview of 11 trials involving 315 patients using IFNα as a single agent demonstrated a response rate of 15 per cent.122 Responses were seen in previously treated and untreated patients and a third of responses were complete. The median survival was 8 months. Toxicity is dose dependent and is usually limited to nausea, flu-like symptoms and lethargy. IFN-α has similar response rates to chemotherapy in patients with metastatic melanoma. There have been no randomized comparisons with DTIC or best supportive care. The role of IFN-α alone in the treatment of metastatic melanoma should be limited to clinical trials. Interleukin-2 IL-2 is an immunostimulatory cytokine exerting its antitumour effects by activating NK cells and CTLs. The largest study of single-agent IL-2 using high-dose bolus therapy involved 134 patients with metastatic melanoma and the overall response rate was 17 per cent. Seven per cent of patients demonstrated a complete response, which was maintained for a longer duration than seen with chemotherapy.123 In contrast to chemotherapy, the activity of IL-2 does not appear to be site dependent with similar
response rates in visceral and non-visceral metastases. A retrospective analysis of 270 patients treated with highdose IL-2 demonstrated an objective tumour response in 16 per cent of patients (6 per cent complete response, 10 per cent partial response).124 Responses were less frequent in patients with poor performance status or those who had received prior systemic therapy but, in contrast to IFN-α, were seen with equal frequency in patients with visceral metastases and/or large tumour burdens. The median duration of the response is yet to be reached among those with a complete response, and disease progression has not been noted in any patient with a response lasting more than 30 months.125 Based on the durable responses in a proportion of patients with metastatic melanoma, high-dose bolus IL-2 received FDA approval in 1998 and is considered as standard of care for appropriately selected patients in the USA. However, toxicities of high-dose IL-2 are significant and include sepsis, hypotension, capillary leak syndrome, renal impairment and myelosuppression. Its use is limited to carefully selected patients treated in specialist centres. Combination immunotherapy Pre-clinical evidence suggested synergy between interferon and IL-2. Despite encouraging phase II results, the only randomized trial of high-dose IL-2 with or without IFN-α showed no survival advantage in adding IFN-α (10.2 months IL-2 vs. 9.7 months combination). Response rates were disappointing (5 per cent IL-2 vs. 10 per cent combination) and the trial was terminated early.126 Biochemotherapy Biochemotherapy refers to the combination of cytotoxic chemotherapy with IFN-α and/or IL-2. Cisplatin-containing regimens appear to be the most effective and phase II studies have shown response rates of 40–50 per cent127,128 and a complete remission rate of 13 per cent.127 Differential responses have been observed with biochemotherapy for metastatic melanoma.129–136 Table 38.5 summarizes the randomized trials of biochemotherapy in metastatic melanoma. A randomized study from the European Organization for the Research and Treatment of Cancer (EORTC) confirmed the benefits of the addition of cisplatin to IL-2 and IFN-α. One hundred and thirty-eight patients were treated with IFN-α plus IL-2 with or without cisplatin. There was a significant improvement in response with the addition of cisplatin (33 per cent vs. 18 per cent, p 0.04). There was no significant difference in overall survival with a 9 month median survival for all patients but there was a significant improvement in progression-free survival (92 days vs. 53 days, p 0.02).129 In a subsequent study, 65 patients were randomized to the Dartmouth regimen or combined with IL-2 and IFN-α.130 No differences in response rate, progression-free survival or overall survival were observed between the two treatment groups. Response rates were relatively low in both arms and may be due to the low doses of immunotherapy used in an attempt to reduce toxicity. Another trial randomized 102 patients to
892 Malignant melanoma
Table 38.5 Randomized trials of biochemotherapy in metastatic melanoma Reference Keilholz et al.129 Johnston et al.130 Rosenberg et al.131 Eton et al.132 Ridolfi et al.133 Atkins et al.134 Del Vecchio et al.135 Keilholz et al.136
Regimen
No. of patients
Response rate %
Survival
IL-2/IFN C BCDT IL-2/IFN CDT IL-2/IFN CVD IL-2/IFN CD IL-2/IFN CVD IL-2/IFN CVD IL-2/IFN CD/IFN IL-2
138 65 102 190 176 405 145 363
33 vs. 18 (p 0.04) 23 vs. 27 44 vs. 27 48 vs. 25 (p 0.001) 25 vs. 20 17 vs. 11 33 vs. 22 21 vs. 23
No difference No difference No difference DFS advantage (p 0.0007) No difference No difference No difference No difference
IL-2, interleukin-2; IFN, interferon; B, carmustine; C, cisplatin; D, dacarbazine; T, tamoxifen; V, vinblastine; DFS, disease-free survival.
cisplatin, DTIC and tamoxifen alone or combined with IFN-α and IL-2.131 This study demonstrated a nonsignificant improvement in response to biochemotherapy (44 per cent vs. 27 per cent, p 0.07) and no survival difference. A comparison of cisplatin, vinblastine and DTIC alone or in combination with IL-2 and IFN-α demonstrated a significantly superior response rate for biochemotherapy (48 per cent vs. 25 per cent for chemotherapy alone, p 0.001).132 Median time to progression (p 0.008) but not overall survival was superior in the biochemotherapy group. The Eastern Cooperative Oncology Group phase III study (cisplatin, vinblastine, DTIC with or without IFN-α) and the EORTC 18951 phase III study (cisplatin, DTIC and IFN-α with or without IL-2) showed no survival benefit with biochemotherapy compared with chemotherapy or immunotherapy alone.136 Toxicity from biochemotherapy is significant and patients require hospitalization. It should not be considered in the elderly or frail. Routine use of these more toxic combinations is not recommended and is limited to clinical trials that aim to identify subgroups of patients who may benefit. Vaccine therapy The principles of vaccine therapy have been discussed earlier. Patients with metastatic melanoma have a poor prognosis and thus may not be the most likely group to benefit from vaccine treatment. Phase II studies of Canvaxin in stage IV melanoma demonstrated a survival advantage over historical controls.137 The phase III trial of Canvaxin in stage IV patients did not show any benefit and was closed early.138 Response rates of 12 per cent have been observed in a phase II trial of Melacine in patients with metastatic melanoma.139 A multi-centre phase III comparison of low-dose cyclophosphamide plus Melacine versus DTIC, cisplatin, carmustine and tamoxifen showed no difference in response rates and survival and it may be that both treatments are equally effective.140 The use of Melacine in stage IV melanomas is licensed in Canada.
A study of the autologous vaccine modified with DNP in metastatic melanoma produced five responses in 40 patients with a median duration of 10 months. Anti-tumour responses were strongly associated with DTH.141 Early clinical phase I/II studies reported the ability of vaccination with autologous dendritic cells to induce clinical regression.142 A randomized phase III study comparing the effects of autologous peptide-pulsed dendritic cells versus DTIC was stopped prematurely. There was no statistically significant difference for response (3.8 per cent vs. 5.5 per cent respectively), overall or progression-free survival between both study arms.143 Another approach to vaccine therapy involves the injection of autologous HSP–peptide complexes. These have been shown to stimulate a T-cell response against melanomaassociated antigens in vivo. The HSP70–gp96 peptide complex (HSPPC96, Oncophage, Vitespen) vaccine induced a clinical response in a minority of patients with metastatic melanoma.144 Preliminary results of the phase III trial of HSPPC96 vaccine compared to IL-2/ DTIC/temozolomide and/or complete resection in stage IV patients failed to reveal benefit for the vaccine-treated group.145 Recombinant vaccines against MART-1, gp100 and tyrosinase are being investigated. A modified gp100 peptide co-administered with IL-2 produced responses in 42 per cent of patients with metastatic melanoma.146 The combination of IL-2 with recombinant tyrosinase vaccines did not demonstrate additional clinical activity than would be expected with IL-2 alone.147 There has been interest in the development of antibodies to CTLA4 (cytotoxic T-cell lymphocyte A4). CTLA4 can inhibit T-cell response and hence the ability of the tumour to elicit an immune response. Treatment with a monoclonal antibody against CTLA4 has showed increased tumour immunogenicity in previously vaccinated stage IV patients.148 A phase I trial of the anti-CTLA4 antibody, CP675,206, has shown anti-tumour activity in melanoma.149 There is considerable interest in combining anti-CTLA4 antibodies to block inhibitory activity with other cancer immunotherapies to improve clinical outcome. A study of
Treatment of metastatic melanoma 893
56 patients with stage IV melanoma treated with the anti CTLA4 antibody, MDX-010, and concomitant gp100 vaccine showed a correlation between grade III/IV autoimmune toxicity and clinical response. The overall response rate was 13 per cent and durable responses were achieved.150 A study evaluating the combination of CTLA4 blockade and IL-2 did not support a synergistic effect.151 There is an ongoing randomized phase III trial evaluating the effectiveness of MDX-010 alone or in combination with a gp100 peptide vaccine in stage III or IV patients. Studies have shown that a small proportion of patients with advanced melanoma respond to vaccines. Current work concentrates on optimization of vaccine strategies. It may be the case that melanoma vaccines are proven to be effective in a subset of patients and therefore identification of these patients is essential.
Management of brain metastases Melanoma has a high propensity to metastasize to the brain and brain metastases are responsible for 20–54 per cent of reported deaths in patients with melanoma.152 All patients with melanoma are at risk of brain metastases and certain characteristics have been consistently shown to correlate with brain metastasis. These include male gender, primary lesions on the skin of the trunk, head, or neck and mucosal surfaces, acral lentiginous, nodular, thick or ulcerated lesions, lymph node or visceral metastasis at the time of diagnosis. In a series of 686 patients with brain metastases treated with modern techniques, median survival from the time of diagnosis of cerebral metastasis was 4.1 months.153 There are, however, some patients who survive more than 3 years. Median survival is dependent on treatment, which in turn is dependent on patient selection. Significant factors associated with improved survival are surgical treatment, no concurrent extra-cerebral metastases, younger age and longer disease-free interval.153 Three major treatment modalities are used to treat brain metastases from melanoma: surgery, radiation therapy (both whole brain and stereotactic irradiation), and chemotherapy. SURGERY
Complete surgical resection is considered for patients with a solitary brain metastasis.154 Patients with multiple accessible lesions in whom all lesions can be removed may have a prognosis similar to those with a solitary metastasis and therefore surgery is a consideration.154,155 Surgery may also be considered to remove a large symptomatic or life-threatening lesion in patients with multiple other surgically inaccessible brain metastases who have limited or controlled systemic disease. A retrospective analysis of 686 patients with cerebral melanoma metastases153 suggested a benefit from surgery. The median survival for those patients undergoing surgery with or without postoperative
radiotherapy (8.9 and 8.7 months, respectively) was significantly greater than that of patients treated with radiotherapy alone or supportive care (3.4 and 2.1 months, respectively). RADIOTHERAPY
There are limited data on the efficacy of whole brain radiotherapy (WBRT) after surgical removal compared to surgery alone.152,156 In a series of 34 patients with solitary cerebral metastases treated with postoperative adjuvant cranial irradiation, radiotherapy had an impact on survival (18 months in irradiated group vs. 6 months in surgery alone group). Another study evaluated 35 patients who underwent resection of a single melanoma brain metastasis, 19 of whom received postoperative radiation therapy.156 Radiotherapy extended the median time to central nervous system (CNS) recurrence from 6 to 27 months and reduced the chance of dying from CNS disease from 84 per cent to 24 per cent. Median survival was not affected, illustrating the importance of controlling systemic disease. The decision to treat with postoperative radiotherapy is an important issue in view of the possible late neurotoxic effects of whole brain radiotherapy. The potential onset of dementia following radiotherapy is particularly relevant for the group of patients with a solitary cerebral metastasis because this group has the greatest potential of long-term survival. Based on the available evidence, for patients with solitary brain metastasis who may have a prolonged survival following surgery, a detailed assessment of the risks and benefits of postoperative radiotherapy is necessary on an individual basis. Experience with stereotactic irradiation in patients with melanoma brain metastases is increasing. The term stereotactic radiosurgery refers to dose delivery by the Gamma Knife technique in a single fraction. The term stereotactic radiotherapy is used to describe dose delivery as a series of fractions. Stereotactic radiosurgery can be given as an outpatient procedure and does not require general anaesthesia. Several lesions can be treated at the same procedure. The maximum size of lesions treated with stereotactic techniques is 3 cm. In a study of 118 melanoma brain metastases treated with stereotactic radiosurgery, freedom from progression was achieved in 90–95 per cent of lesions, with reduction in size occurring in 55 per cent.157 In a retrospective analysis of 26 patients with intracranial melanoma, the median survival was 6 months after Gammaknife radiosurgery.158 The question of whether stereotactic radiosurgery alone is sufficient to treat brain metastases is being addressed by the ECOG E6397 trial. This is a phase II study of radiosurgery for 1–3 newly diagnosed brain metastases from melanoma, renal cell carcinoma and sarcoma. Preliminary results have shown that median survival was equivalent to published surgical series and better than WBRT alone.159 A prognostic index that predicts outcome following palliative whole brain radiotherapy for melanoma has been identified.160 The RTOG recursive partitioning analysis (RPA) classification for brain metastases was used to define
894 Malignant melanoma
prognostic groups (class 1: Karnofsky Performance Score (KPS) 70 per cent, age 65 years with no extracranial metastasis; class 3: KPS 70 per cent; class 2: all others). RPA class 3 patients, those with leptomeningeal disease (median survival 3 weeks) and those with extra-cranial metastases (median survival 39 days) were found to have the worst prognosis. This group should be treated with best supportive care alone. Patients in RPA class 1 and 2 should be considered for active treatment. The prognostic index identified a subgroup of RPA class 2 patients (leptomeningeal disease, five extracranial sites, or three extracranial sites and progressive brain disease prior to radiotherapy) that have a median survival of 21 days and are unlikely to benefit from whole brain radiotherapy. CHEMOTHERAPY
Treatment with chemotherapy for brain metastases from melanoma has been disappointing. Fotemustine has been shown to have some activity and is used in this setting. Toxicities include neutropenia and thrombocytopenia. A phase III trial of fotemustine plus WBRT versus fotemustine alone in patients with cerebral metastases demonstrated a longer time to progression of cerebral metastases in the combination arm. However, no differences in objective response or overall survival were shown.161 A phase III study of fotemustine compared with DTIC as first-line chemotherapy in patients with disseminated malignant melanoma (without brain metastases) showed a trend in favour of fotemustine for time to occurrence of brain metastases (22.7 vs 7.2 months, p 0.059).162 Temozolomide is a well-tolerated oral alkylating agent with activity in the CNS. A phase II study of temozolomide in patients with brain metastases showed that temozolomide has modest activity in previously untreated patients (objective response 7 per cent, stable disease 29 per cent, median survival 3.5 months).163 The combination of thalidomide and temozolomide, both of which cross the blood–brain barrier, has shown some activity in the treatment of brain metastases from melanoma. In a phase II study of 26 patients, three had a complete or partial response, and seven had a minor response or stable disease in the brain.164 Five of these 10 patients had disease progression at extra-cranial sites, and a median survival of 5 months. A preliminary report of a study by the Cytokine Working Group, in which 40 patients with brain metastases from melanoma received whole brain radiotherapy with thalidomide and temozolomide showed only 3 objective responses in the brain, no systemic responses and a median survival of 4 months.165
NOVEL THERAPIES Up to 90 per cent of patients with metastatic melanoma do not have significant benefit from standard treatments currently available and therefore novel therapies are under investigation. Advances in our knowledge of tumour
molecular biology allow the identification of molecular targets for anti-cancer drug development. These novel therapies are in early development and examples including gene therapy are outlined below.
Signalling inhibitors The RAS-RAF-MEK-ERK signalling pathway is activated in the majority of melanomas. Activation of this pathway can occur via BRAF mutations and these have been found in 69 per cent of melanoma cell lines. Up-regulation of this signalling pathway has been shown in chemotherapyresistant cell lines, as well as the over-expression of the angiogenic factors interleukin-8 (IL-8) and vascular endothelial growth factor (VEGF). Sorafenib (BAY 439006) is a potent inhibitor of Raf kinase and thereby inhibits ERK phosphorylation in cancer cell lines. It also inhibits angiogenesis by blocking the receptor tyrosine kinases VEGFR-2 and PDGFR-β. Anti-tumour activity has been demonstrated in pre-clinical xenograft models, and results from phase I data were encouraging. In a phase I dose-escalation trial of sorafenib combined with DTIC in metastatic melanoma, preliminary anti-tumour activity was seen.166 Sorafenib was well tolerated, skin toxicity being the commonest side effect. A phase II discontinuation trial of sorafenib demonstrated limited anti-tumour activity as a single agent in patients with stage IV melanoma.167 However, a phase II study of sorafenib and DTIC as first-line therapy in advanced melanoma demonstrated encouraging progression-free survival and overall survival rates.168 A phase III study of paclitaxel plus carboplatin with or without sorafenib as second-line treatment reported no improvement in survival or response rate with the addition of sorafenib in advanced melanoma.169 However, these patients had progressed through previous treatments and the utility of the combination of chemotherapy with sorafenib in chemotherapy-naive advanced melanoma patients is currently being addressed in a phase III trial (E2603). A phase II study combining sorafenib with temozolamide has shown early promising results.170 Efforts to demonstrate a correlation between tumour response and the presence of the activating mutations in melanoma cells have so far failed ton show such a relationship.171,172 There has been considerable interest in the use of angiogenesis inhibitors to target melanoma. 17-AAG (17-N allylamino-17-demethoxygeldanamycin), AG-013736 and bevacizumab target angiogenesis and are currently being tested in phase II trials of metastatic melanoma.
Adoptive immunotherapy Pre-clinical data indicated that the activity of IL-2 was optimal when high doses were combined with IL-2 activated peripheral blood lymphocytes (LAK cells). Adoptive
Intra-ocular melanoma 895
cell transfer (ACT) immunotherapy is based on the ex vivo selection of tumour-reactive lymphocytes and their activation and numerical expansion before re-infusion into the autologous tumour-bearing host.173 This approach can be optimized with non-myeloablative but lymphodepleting chemotherapy to promote engraftment of tumour-reactive T-cells. This approach was studied in 35 patients with refractory metastatic melanoma. Patients underwent lymphodepleting conditioning with 2 days of cyclophosphamide (60 mg/kg) followed by 5 days of fludarabine (25 mg/m2).174 On the day following the final dose of fludarabine, all patients received cell infusion with autologous tumourreactive, rapidly expanded tumour-infiltrating lymphocyte cultures and high-dose IL-2 therapy. Eighteen (51 per cent) patients (three complete responses, 15 partial responses) had objective regression of metastases, including those in lung, liver and brain with a mean duration of 11.5 months. Autoimmunity such as vitiligo and anterior uveitis was seen in some patients who experienced an objective response. An association between successful immunotherapy and the onset of autoimmunity has been previously noted and suggests a link between the mechanism of tumour regression and autoimmune attack on normal tissues.
Gene therapy Gene therapy approaches in melanoma have focused on targeting either melanoma cells or the host’s immune cells. Results are available from clinical trials of autologous cell lines transduced with genes coding for IL-2,175,176 IL-7,177 IL-12,178 IFN-γ179 or GM-CSF180 and allogeneic lines transduced with genes coding for IL-2181 or IL-4.182 These have shown good tolerability but limited clinical efficacy. Twenty to thirty per cent of vaccinated patients showed an increase in T-cell immune reactivity. Autologous cells from 12 patients with metastatic melanoma were genetically engineered using recombinant retroviruses to secrete IL-2. No clinical responses were seen but DTH and CTL responses were seen in 5 patients and the median diseasefree survival in this group was significantly improved (7 months vs. 1 month, p 0.005).175 There is an ongoing phase II trial of tumour-infiltrating lymphocytes transduced with the IL-2 gene in metastatic melanoma. Clinical responses to vaccination with granulocyte macrophage colony-stimulated factor (GM-CSF) transfected melanoma cells have been reported.180,183,184 A phase I clinical trial testing the biological activity of vaccination with irradiated, autologous melanoma cells engineered to secrete GM-CSF by adenoviral-mediated gene transfer showed brisk or focal T-lymphocyte and plasma cell infiltrates with tumour necrosis in the resected metastatic lesions of 10 of 16 patients. One complete, one partial, and one mixed response were noted. Twenty-six patients had assessable disease and of those 10 patients (29 per cent) are alive, with a minimum follow-up of 36 months; 4 of these patients have no evidence of disease at the time of the report.180
Dendritic cells are antigen-presenting cells that are very effective at generating T-cell immunity. Ongoing and future clinical trials aim to demonstrate clinical activity of dendritic cells transduced with genes coding for tumour-associated antigens. There is a phase I/II trial of vaccination with MART-1 adenovirus transduced dendritic cells in metastatic melanoma in progress. Anti-sense gene therapy targets specific messenger RNA molecules to inhibit their translation into protein. Alterations of the anti-apoptotic protein bcl-2 are common in cancer and bcl-2 is a target for anti-sense oligonucleotides such as oblimersen (Genasense, G3139). This drug sensitized melanoma cells to chemotherapy in pre-clinical studies.185 A randomized phase III trial of 771 patients with advanced melanoma, comparing DTIC with or without bcl-2 antisense (oblimersen), showed that combined therapy was associated with a significantly higher response rate (13.5 vs. 7.5 per cent) and progression-free survival (2.6 vs. 1.6 months).186 Prodrug gene therapy is another novel approach being tested in a variety of tumours. This involves a prodrug metabolizing gene such as herpes simplex thymidine kinase used with systemic administration of the prodrug ganciclovir, which is converted by the enzyme into a toxic metabolite, ganciclovir triphosphate. This toxic metabolite can diffuse across cell membranes and exert cytotoxic effect on surrounding tumour cells. This method has been shown to suppress tumour growth in pre-clinical models of human cancer cells, established murine melanoma tumours and human melanoma xenografts.187,188 A phase I/II study of retrovirus-mediated herpes simplex virus type 1 thymidine kinase (HSV-tk) gene therapy and ganciclovir treatment in nodules of metastatic melanoma demonstrated limited clinical response although tumour necrosis was detected. This lack of clinical activity may be related to poor gene transfer efficiency.189 Replacement of p53 represents an important target for gene replacement therapy. DL-1520 (ONYX-015) is a mutant adenovirus that replicates in p53 deficient cells but not in normal cells and therefore has the potential to be cancer specific. Clinical trials using ONYX-15 alone and in combination with chemotherapy are ongoing. Other trials in progress are a phase II study of intratumoral INGN 241 in patients with in-transit melanoma. INGN 241 is an adenoviral vector carrying the MDA-7 gene, which is a tumour suppressor gene. Developments in vector technology and an increased understanding of the cellular and molecular mechanisms elicited by gene therapy mean that this approach may have better clinical activity in the near future.
INTRA-OCULAR MELANOMA Intra-ocular melanoma is a rare malignancy arising from uveal melanocytes. The prognosis of metastatic intra-ocular melanoma is very poor with median survival in the range of 2–5 months. Management is dependent on tumour size
896 Malignant melanoma
and a variety of modalities have been evaluated in the treatment of intra-ocular melanoma. They can be treated locally with laser photocoagulation, enucleation or radiotherapy using either charged particles or radioactive plaques (brachytherapy). The role of enucleation of the eye has been a controversial issue since survival outcomes do not appear to differ from those achieved with radiotherapy.190–192 The Collaborative Ocular Melanoma Study (COMS) randomly assigned 1317 patients with medium-sized choroidal melanomas to enucleation versus 125I brachytherapy. Neither the unadjusted estimated 5-year survival rates (81 per cent vs. 82 per cent) nor the 5-year rates of death with histopathologically confirmed metastatic melanoma (11 per cent vs. 9 per cent) were significantly different following enucleation and brachytherapy, respectively.192 Results from the Collaborative Ocular Melanoma Study (COMS) randomized trial of pre-enucleation radiation of large choroidal melanoma showed that there is no benefit for pre-enucleation radiotherapy.193 Enucleation should be considered in selected patients, such as patients with large tumours, tumours with large extra-scleral extensions or neovascular glaucoma in a painful eye. There are no effective treatments for the majority of patients with metastatic ocular melanoma. In the majority of cases, intra-ocular melanoma metastasizes to the liver. Liver metastases are particularly refractory to treatment. In selected patients, surgical resection of metastatic uveal melanoma may provide long-term survival.194,195 In one series, 24 of 78 patients with metastatic liver involvement underwent resection.195 The median survival and 5-year survival rate for patients treated with surgery were 38 months and 39 per cent respectively, versus 9 months and 0 per cent for patients undergoing non-surgical therapies. Chemotherapy and/or immunotherapies have been used in the treatment of metastatic melanoma with limited efficacy. Better responses have been described with regional therapeutic approaches for patients with hepatic metastases. An example of this approach is the intra-hepatic arterial administration of fotemustine.196,197 In a multi-centre study of 66 patients, 22 patients (36 per cent) achieved an objective response with seven complete and 15 partial remissions; the median overall survival was 14.8 months, and 27 per cent survived 1 year.197 In another study, hyperthermic isolated hepatic perfusion with melphalan alone was administered to 29 patients with liver metastasis secondary to ocular melanoma.198 There were 18 objective responses, three of which were complete. The median progression-free and overall survival rates were 8 and 12.1 months, respectively. Transient grade 3 or 4 hepatotoxicity occurred in 65 per cent. A randomized trial by the EORTC is underway comparing intravenous and intra-hepatic chemotherapy for hepatic involvement in metastatic ocular melanoma. The development of extra-hepatic disease is an important limitation to all regional treatment approaches. Although regional therapy can be considered for patients with isolated hepatic metastases, better systemic therapies are necessary.
CONCLUSIONS The incidence of malignant melanoma continues to rise and primary prevention of malignant melanoma is highly important. For primary melanoma, surgery remains the main treatment. The current role of sentinel node biopsy is as a staging procedure for clinical trials of adjuvant therapy. Adjuvant interferon provides clinical benefit in terms of relapse-free survival. Identifying the optimal dose and duration of interferon as well as the relevance of microscopic nodal involvement are areas of ongoing research. So far, results from clinical trials do not support the use of vaccines in the adjuvant setting. Metastatic melanoma remains an incurable disease despite many treatment strategies having shown potential in phase II trials. The main treatment options for advanced melanoma are agents in clinical trials. Signalling inhibitors, adoptive immuno-therapy and gene therapy are examples of promising approaches.
KEY LEARNING POINTS ●
●
●
●
●
●
●
Sunlight is a major aetiological factor in melanoma. Protective measures to reduce sun exposure, particularly during childhood, are recommended. Patients usually present with early-stage disease and the majority of these are cured by surgical excision. The outlook for advanced disease is very poor. The UK guidelines recommend the following excision margins: 1 cm for primary tumours less than 1 mm, 1–2 cm for lesions between 1 and 2 mm, and 2 cm for lesions greater than 2 mm. The sentinel lymph node status is an important prognostic factor for recurrence. There is no overall survival benefit for sentinel lymph node biopsy. Sentinel lymph node biopsy is an accurate staging method and helps select patients for entry to trials of adjuvant therapy. The roles of IFN-α and vaccines in the adjuvant setting require further randomized trials. There is no consistent randomized evidence in metastatic disease for an improvement in overall survival with any therapy. The focus of patient care is palliation. DTIC has an established palliative role. Advances in our knowledge of molecular biology have identified molecular targets for anti-cancer drug development. Novel therapies include signalling inhibitors and gene therapy. Treatment with passive immunotherapy using anti-CTLA4 antibodies and adoptive immunotherapy look promising but require further clinical investigation.
References 897
REFERENCES 1 Cancer Research UK, CancerStats, Malignant Melanoma 2005. http://info.cancerresearchuk.org/cancerstats/types/ melanoma/?a5441 2 Rees JL. The melanoma epidemic: reality and artefact. BMJ 1996; 312(7024):137–8. 3 Lee ML, Tomsu K, Von Eschen KB. Duration of survival for disseminated malignant melanoma: results of a metaanalysis. Melanoma Res 2000; 10(1):81–92. 4 Meyskens FL, Jr., Farmer PJ, Anton-Culver H. Etiologic pathogenesis of melanoma: a unifying hypothesis for the missing attributable risk. Clin Cancer Res 2004; 10(8):2581–3. 5 Marks R, Dorevitch AP, Mason G. Do all melanomas come from ‘moles’? A study of the histological association between melanocytic naevi and melanoma. Australas J Dermatol 1990; 31(2):77–80. 6 Bataille V, Grulich A, Sasieni P, et al. The association between naevi and melanoma in populations with different levels of sun exposure: a joint case-control study of melanoma in the UK and Australia. Br J Cancer 1998; 77(3):505–10. 7 Boyle P, Maisonneuve P, Dore JF. Epidemiology of malignant melanoma. Br Med Bull 1995; 51(3):523–47. 8 Weinstock MA, Colditz GA, Willett WC, et al. Nonfamilial cutaneous melanoma incidence in women associated with sun exposure before 20 years of age. Pediatrics 1989; 84(2):199–204. 9 Elwood JM, Jopson J. Melanoma and sun exposure: an overview of published studies. Int J Cancer 1997; 73(2):198–203. 10 Veierod MB, Weiderpass E, Thorn M, et al. A prospective study of pigmentation, sun exposure, and risk of cutaneous malignant melanoma in women. J Natl Cancer Inst 2003; 95(20):1530–8. 11 Gallagher RP, Spinelli JJ, Lee TK. Tanning beds, sunlamps, and risk of cutaneous malignant melanoma. Cancer Epidemiol Biomarkers Prev 2005; 14(3):562–6. 12 Finkel E. Sorting the hype from the facts in melanoma. Lancet 1998; 351(9119):1866. 13 Autier P, Dore JF, Negrier S, et al. Sunscreen use and duration of sun exposure: a double-blind, randomized trial. J Natl Cancer Inst 1999; 91(15):1304–9. 14 Autier P. Cutaneous malignant melanoma: facts about sunbeds and sunscreen. Expert Rev Anticancer Ther 2005; 5(5):821–33. 15 Stern RS. The risk of melanoma in association with long-term exposure to PUVA. J Am Acad Dermatol 2001; 44(5):755–61. 16 Lindor NM, Greene MH. The concise handbook of family cancer syndromes. Mayo Familial Cancer Program. J Natl Cancer Inst 1998; 90(14):1039–71. 17 Goldstein AM, Dracopoli NC, Engelstein M, Fraser MC, Clark WH, Jr., Tucker MA. Linkage of cutaneous malignant melanoma/dysplastic nevi to chromosome 9p, and evidence for genetic heterogeneity. Am J Hum Genet 1994; 54(3):489–96.
18 Bishop DT, Demenais F, Goldstein AM, et al. Geographical variation in the penetrance of CDKN2A mutations for melanoma. J Natl Cancer Inst 2002; 94(12):894–903. 19 Kefford RF, Mann GJ. Is there a role for genetic testing in patients with melanoma? Curr Opin Oncol 2003; 15(2):157–61. 20 Hayward NK. Genetics of melanoma predisposition. Oncogene 2003; 22(20):3053–62. 21 Casula M, Colombino M, Satta MP, et al. BRAF gene is somatically mutated but does not make a major contribution to malignant melanoma susceptibility: the Italian Melanoma Intergroup Study. J Clin Oncol 2004; 22(2):286–92. 22 Donawho C, Wolf P. Sunburn, sunscreen, and melanoma. Curr Opin Oncol 1996; 8(2):159–66. 23 Dennis LK, Beane Freeman LE, VanBeek MJ. Sunscreen use and the risk for melanoma: a quantitative review. Ann Intern Med 2003; 139(12):966–78. 24 Hill D, Dixon H. Promoting sun protection in children: rationale and challenges. Health Educ Behav 1999; 26(3):409–17. 25 Bishop JA, Corrie PG, Evans J, et al. UK guidelines for the management of cutaneous melanoma. Br J Plast Surg 2002; 55(1):46–54. 26 Demierre MF, Nathanson L. Chemoprevention of melanoma: an unexplored strategy. J Clin Oncol 2003; 21(1):158–65. 27 Freeman SR, Drake AL, Heilig LF, Graber M, McNealy K, Schilling LM, Dellavalle RP. Statins, fibrates, and melanoma risk: a systematic review and meta-analysis. J Natl Cancer Inst 2006; 98(21):1538–46. 28 Fischer SM. Is cyclooxygenase-2 important in skin carcinogenesis? J Environ Pathol Toxicol Oncol 2002; 21(2):183–91. 29 Elmets CA, Singh D, Tubesing K, Matsui M, Katiyar S, Mukhtar H. Cutaneous photoprotection from ultraviolet injury by green tea polyphenols. J Am Acad Dermatol 2001; 44(3):425–32. 30 Giuliano AE, Cochran AJ, Morton DL. Melanoma from unknown primary site and amelanotic melanoma. Semin Oncol 1982; 9(4):442–7. 31 Westerhoff K, McCarthy WH, Menzies SW. Increase in the sensitivity for melanoma diagnosis by primary care physicians using skin surface microscopy. Br J Dermatol 2000; 143(5):1016–20. 32 Kittler H, Pehamberger H, Wolff K, Binder M. Diagnostic accuracy of dermoscopy. Lancet Oncol 2002; 3(3): 159–65. 33 Bono A, Tomatis S, Bartoli C, et al. The invisible colours of melanoma. A telespectrophotometric diagnostic approach on pigmented skin lesions. Eur J Cancer 1996; 32A(4):727–9. 34 Yu LL, Flotte TJ, Tanabe KK, et al. Detection of microscopic melanoma metastases in sentinel lymph nodes. Cancer 1999; 86(4):617–27. 35 Murray CA, Leong WL, McCready DR, Ghazarian DM. Histopathological patterns of melanoma metastases in sentinel lymph nodes. J Clin Pathol 2004; 57(1):64–7.
898 Malignant melanoma
36 Balch CM, Soong SJ, Gershenwald JE, et al. Prognostic factors analysis of 17,600 melanoma patients: validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol 2001; 19(16):3622–34. 37 Balch CM. Cutaneous melanoma: prognosis and treatment results worldwide. Semin Surg Oncol 1992; 8(6):400–14. 38 Gimotty PA, Van Belle P, Elder DE, et al. Biologic and prognostic significance of dermal Ki67 expression, mitoses, and tumorigenicity in thin invasive cutaneous melanoma. J Clin Oncol 2005; 23(31):8048–56. 39 Koyanagi K, O’Day SJ, Gonzalez R, et al. Serial monitoring of circulating melanoma cells during neoadjuvant biochemotherapy for stage III melanoma: outcome prediction in a multicenter trial. J Clin Oncol 2005; 23(31):8057–64. 40 Mårtenson ED, Hansson LO, Nilsson B, von Schoultz E, Månsson Brahme E, Ringborg U, Hansson J. Serum S-100b protein as a prognostic marker in malignant cutaneous melanoma. J Clin Oncol 2001; 19(3):824–31. 41 Buzaid AC, Ross MI, Balch CM, et al. Critical analysis of the current American Joint Committee on Cancer staging system for cutaneous melanoma and proposal of a new staging system. J Clin Oncol 1997; 15(3):1039–51. 42 Gershenwald JE, Thompson W, Mansfield PF, et al. Multiinstitutional melanoma lymphatic mapping experience: the prognostic value of sentinel lymph node status in 612 stage I or II melanoma patients. J Clin Oncol 1999; 17(3):976–83. 43 Vuylsteke RJ, van Leeuwen PA, Statius Muller MG, Gietema HA, Kragt DR, Meijer S. Clinical outcome of stage I/II melanoma patients after selective sentinel lymph node dissection: long-term follow-up results. J Clin Oncol 2003; 21(6):1057–65. 44 Dewar DJ, Newell B, Green MA, Topping AP, Powell BW, Cook MG. The microanatomic location of metastatic melanoma in sentinel lymph nodes predicts nonsentinel lymph node involvement. J Clin Oncol 2004; 22(16): 3345–9. 45 Shivers SC, Wang X, Li W, et al. Molecular staging of malignant melanoma: correlation with clinical outcome. JAMA 1998; 280(16):1410–5. 46 Kammula US, Ghossein R, Bhattacharya S, Coit DG. Serial follow-up and the prognostic significance of reverse transcriptase-polymerase chain reaction – staged sentinel lymph nodes from melanoma patients. J Clin Oncol 2004; 22(19):3989–96. 47 Morton DL, Thompson JF, Cochran AJ, Essner R, Elashoff R, Multicenter Selective Lymphadenectomy Trial Group. Interim results of the Multicenter Selective Lymphadenectomy Trial (MSLT-1) in clinical stage I melanoma. J Clin Oncol 2005 ASCO Annual Meeting Proceedings 2005; 23(16S):7500. 48 Starritt EC, Uren RF, Scolyer RA, Quinn MJ, Thompson JF. Ultrasound examination of sentinel nodes in the initial assessment of patients with primary cutaneous melanoma. Ann Surg Oncol 2005; 12(1):18–23. 49 Rossi CR, Mocellin S, Scagnet B, et al. The role of preoperative ultrasound scan in detecting lymph node
50
51
52
53
54
55
56
57
58
59
60
61
62
63
metastasis before sentinel node biopsy in melanoma patients. J Surg Oncol 2003; 83(2):80–4. Eton O, Legha SS, Moon TE, et al. Prognostic factors for survival of patients treated systemically for disseminated melanoma. J Clin Oncol 1998; 16(3):1103–11. Swetter SM, Carroll LA, Johnson DL, Segall GM. Positron emission tomography is superior to computed tomography for metastatic detection in melanoma patients. Ann Surg Oncol 2002; 9(7):646–53. Holder WD, Jr., White RL, Jr., Zuger JH, Easton EJ, Jr., Greene FL. Effectiveness of positron emission tomography for the detection of melanoma metastases. Ann Surg 1998; 227(5):764–9; discussion 769–71. Damian DL, Fulham MJ, Thompson E, Thompson JF. Positron emission tomography in the detection and management of metastatic melanoma. Melanoma Res 1996; 6(4):325–9. Dalrymple-Hay MJ, Rome PD, Kennedy C, Fulham M, McCaughan BC. Pulmonary metastatic melanoma – the survival benefit associated with positron emission tomography scanning. Eur J Cardiothorac Surg 2002; 21(4):611–4; discussion 614–5. Cohn-Cedermark G, Rutqvist LE, Andersson R, et al. Long term results of a randomized study by the Swedish Melanoma Study Group on 2-cm versus 5-cm resection margins for patients with cutaneous melanoma with a tumor thickness of 0.8–2.0 mm. Cancer 2000; 89(7):1495–501. Balch CM, Soong S, Ross MI, et al. Long-term results of a multi-institutional randomized trial comparing prognostic factors and surgical results for intermediate thickness melanomas (1.0 to 4.0 mm). Intergroup Melanoma Surgical Trial. Ann Surg Oncol 2000; 7(2):87–97. Veronesi U, Cascinelli N, Adamus J, et al. Thin stage I primary cutaneous malignant melanoma. Comparison of excision with margins of 1 or 3 cm. N Engl J Med 1988; 318(18):1159–62. Khayat D, Rixe O, Martin G, et al. Surgical margins in cutaneous melanoma (2 cm versus 5 cm for lesions measuring less than 2.1-mm thick). Cancer 2003; 97(8):1941–6. Thomas JM, Newton-Bishop J, A’Hern R, et al. Excision margins in high-risk malignant melanoma. N Engl J Med 2004; 350(8):757–66. Heaton KM, Sussman JJ, Gershenwald JE, et al. Surgical margins and prognostic factors in patients with thick (4 mm) primary melanoma. Ann Surg Oncol 1998; 5(4):322–8. Sober AJ, Chuang TY, Duvic M, et al. Guidelines of care for primary cutaneous melanoma. J Am Acad Dermatol 2001; 45(4):579–86. Clinical Practice Guidelines for the Management of Cutaneous Melanoma (Australia). National Health and Medical Research Council 1999. www.nhmrc.gov.au/ publications/synopses/cp68syn.htm Harwood AR. Conventional fractionated radiotherapy for 51 patients with lentigo maligna and lentigo maligna melanoma. Int J Radiat Oncol Biol Phys 1983; 9(7):1019–21.
References 899
64 Schmid-Wendtner MH, Brunner B, Konz B, et al. Fractionated radiotherapy of lentigo maligna and lentigo maligna melanoma in 64 patients. J Am Acad Dermatol 2000; 43(3):477–82. 65 Kissin MW, Simpson DA, Easton D, White H, Westbury G. Prognostic factors related to survival and groin recurrence following therapeutic lymph node dissection for lower limb malignant melanoma. Br J Surg 1987; 74(11):1023–6. 66 Fraker DL. Surgical issues in the management of melanoma. Curr Opin Oncol 1997; 9(2):183–8. 67 Balch CM, Soong SJ, Bartolucci AA, et al. Efficacy of an elective regional lymph node dissection of 1 to 4 mm thick melanomas for patients 60 years of age and younger. Ann Surg 1996; 224(3):255–63; discussion 263–6. 68 Cascinelli N, Morabito A, Santinami M, MacKie RM, Belli F. Immediate or delayed dissection of regional nodes in patients with melanoma of the trunk: a randomised trial. WHO Melanoma Programme. Lancet 1998; 351(9105):793–6. 69 Koops HS, Vaglini M, Suciu S, et al. Prophylactic isolated limb perfusion for localized, high-risk limb melanoma: results of a multicenter randomized phase III trial. European Organization for Research and Treatment of Cancer Malignant Melanoma Cooperative Group Protocol 18832, the World Health Organization Melanoma Program Trial 15, and the North American Perfusion Group Southwest Oncology Group-8593. J Clin Oncol 1998; 16(9):2906–12. 70 Owens JM, Roberts DB, Myers JN. The role of postoperative adjuvant radiation therapy in the treatment of mucosal melanomas of the head and neck region. Arch Otolaryngol Head Neck Surg 2003; 129(8):864–8. 71 Temam S, Mamelle G, Marandas P, et al. Postoperative radiotherapy for primary mucosal melanoma of the head and neck. Cancer 2005; 103(2):313–9. 72 Ballo MT, Zagars GK, Gershenwald JE, et al. A critical assessment of adjuvant radiotherapy for inguinal lymph node metastases from melanoma. Ann Surg Oncol 2004; 11(12):1079–84. 73 Buzaid AC, Colome M, Bedikian A, et al. Phase II study of neoadjuvant concurrent biochemotherapy in melanoma patients with local-regional metastases. Melanoma Res 1998; 8(6):549–56. 74 Kirkwood JM, Ibrahim JG, Sondak VK, et al. High- and lowdose interferon alfa-2b in high-risk melanoma: first analysis of intergroup trial E1690/S9111/C9190. J Clin Oncol 2000; 18(12):2444–58. 75 Kirkwood JM, Ibrahim JG, Sosman JA, et al. High-dose interferon alfa-2b significantly prolongs relapse-free and overall survival compared with the GM2-KLH/QS-21 vaccine in patients with resected stage IIB–III melanoma: results of intergroup trial E1694/S9512/C509801. J Clin Oncol 2001; 19(9):2370–80. 76 Kirkwood JM, Strawderman MH, Ernstoff MS, Smith TJ, Borden EC, Blum RH. Interferon alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: the Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 1996; 14(1):7–17. 77 Kirkwood JM, Manola J, Ibrahim J, Sondak V, Ernstoff MS, Rao U. A pooled analysis of Eastern Cooperative Oncology Group
78
79
80
81
82
83
84
85
86
87
88
and intergroup trials of adjuvant high-dose interferon for melanoma. Clin Cancer Res 2004; 10(5):1670–7. Creagan ET, Dalton RJ, Ahmann DL, et al. Randomized, surgical adjuvant clinical trial of recombinant interferon alfa2a in selected patients with malignant melanoma. J Clin Oncol 1995; 13(11):2776–83. Pehamberger H, Soyer HP, Steiner A, et al. Adjuvant interferon alfa-2a treatment in resected primary stage II cutaneous melanoma. Austrian Malignant Melanoma Cooperative Group. J Clin Oncol 1998; 16(4):1425–9. Eggermont AMM, Kleeberg UR, Ruiter DJ, Suciu S. The European Organization for Research and Treatment of Cancer Melanoma Group trial experience with more than 2000 patients, evaluating adjuvant therapy treatment with low or intermediate doses of interferon alpha-2b. In: Perry MC (ed) American Society of Clinical Oncology Educational Book. Alexandria, Virginia: 2001, 88. Cameron DA, Cornbleet MC, Mackie RM, et al. Adjuvant interferon alpha 2b in high risk melanoma – the Scottish study. Br J Cancer 2001; 84(9):1146–9. Grob JJ, Dreno B, de la Salmoniere P, et al. Randomised trial of interferon alpha-2a as adjuvant therapy in resected primary melanoma thicker than 1.5 mm without clinically detectable node metastases. French Cooperative Group on Melanoma. Lancet 1998; 351(9120):1905–10. Cascinelli N, Belli F, MacKie RM, Santinami M, Bufalino R, Morabito A. Effect of long-term adjuvant therapy with interferon alpha-2a in patients with regional node metastases from cutaneous melanoma: a randomised trial. Lancet 2001; 358(9285):866–9. Hancock BW, Wheatley K, Harris S, et al. Adjuvant interferon in high-risk melanoma: the AIM HIGH Study – United Kingdom Coordinating Committee on Cancer Research randomized study of adjuvant low-dose extended-duration interferon Alfa-2a in high-risk resected malignant melanoma. J Clin Oncol 2004; 22(1):53–61. Kleeberg UR, Suciu S, Brocker EB, et al. Final results of the EORTC 18871/DKG 80-1 randomised phase III trial. rIFNalpha2b versus rIFN-gamma versus ISCADOR M versus observation after surgery in melanoma patients with either high-risk primary (thickness 3 mm) or regional lymph node metastasis. Eur J Cancer 2004; 40(3):390–402. Eggermont AM, Suciu S, MacKie R, et al. Post-surgery adjuvant therapy with intermediate doses of interferon alfa 2b versus observation in patients with stage IIb/III melanoma (EORTC 18952): randomised controlled trial. Lancet 2005; 366(9492):1189–96. Wheatley K, Ives N, Hancock B, Gore M, Eggermont A, Suciu S. Does adjuvant interferon-alpha for high-risk melanoma provide a worthwhile benefit? A meta-analysis of the randomised trials. Cancer Treat Rev 2003; 29(4):241–52. Eggermont AM, Suciu S, Santinami M, et al., EORTC Melanoma Group. EORTC 18991: Long-term adjuvant pegylated interferon-alpha2b (PEG-IFN) compared to observation in resected stage III melanoma, final results of a randomized phase III trial. J Clin Oncol ASCO Annual Meeting Proceedings Part I. 2007; 25(18S):8504.
900 Malignant melanoma
89 Jones PC, Sze LL, Liu PY, Morton DL, Irie RF. Prolonged survival for melanoma patients with elevated IgM antibody to oncofetal antigen. J Natl Cancer Inst 1981; 66(2):249–54. 90 Livingston PO, Wong GY, Adluri S, et al. Improved survival in stage III melanoma patients with GM2 antibodies: a randomized trial of adjuvant vaccination with GM2 ganglioside. J Clin Oncol 1994; 12(5):1036–44. 91 Bystryn JC, Zeleniuch-Jacquotte A, Oratz R, Shapiro RL, Harris MN, Roses DF. Double-blind trial of a polyvalent, shed-antigen, melanoma vaccine. Clin Cancer Res 2001; 7(7):1882–7. 92 Morton DL, Mozzillo N, Thompson JF, et al., MMAIT Clinical Trials Group. An international, randomized, phase III trial of bacillus Calmette-Guerin (BCG) plus allogeneic melanoma vaccine (MCV) or placebo after complete resection of melanoma metastatic to regional or distant sites. J Clin Oncol ASCO Annual Meeting Proceedings Part I. 2007; 25(18S):8508. 93 Wallack MK, Sivanandham M, Balch CM, et al. Surgical adjuvant active specific immunotherapy for patients with stage III melanoma: the final analysis of data from a phase III, randomized, double-blind, multicenter vaccinia melanoma oncolysate trial. J Am Coll Surg 1998; 187(1):69–77; discussion 77–9. 94 Sondak VK, Liu PY, Tuthill RJ, et al. Adjuvant immunotherapy of resected, intermediate-thickness, node-negative melanoma with an allogeneic tumor vaccine: overall results of a randomized trial of the Southwest Oncology Group. J Clin Oncol 2002; 20(8):2058–66. 95 Morton DL, Eilber FR, Holmes EC, Ramming KP. Preliminary results of a randomized trial of adjuvant immunotherapy in patients with malignant melanoma who have lymph node metastases. Aust N Z J Surg 1978; 48(1):49–52. 96 Hersey P, Coates AS, McCarthy WH, et al. Adjuvant immunotherapy of patients with high-risk melanoma using vaccinia viral lysates of melanoma: results of a randomized trial. J Clin Oncol 2002; 20(20):4181–90. 97 Sosman JA, Unger JM, Liu PY, et al. Adjuvant immunotherapy of resected, intermediate-thickness, nodenegative melanoma with an allogeneic tumor vaccine: impact of HLA class I antigen expression on outcome. J Clin Oncol 2002; 20(8):2067–75. 98 Mitchell MS, Abrams J, Thompson JA, et al. Randomized trial of an allogeneic melanoma lysate vaccine with low-dose interferon alfa-2b compared with high-dose interferon alfa2b for resected stage III cutaneous melanoma. J Clin Oncol 2007; 25(15):2078–85. 99 DiFronzo LA, Gupta RK, Essner R, et al. Enhanced humoral immune response correlates with improved disease-free and overall survival in American Joint Committee on Cancer stage II melanoma patients receiving adjuvant polyvalent vaccine. J Clin Oncol 2002; 20(15):3242–8. 100 Tsioulias GJ, Gupta RK, Tisman G, et al. Serum TA90 antigenantibody complex as a surrogate marker for the efficacy of a polyvalent allogeneic whole-cell vaccine (CancerVax) in melanoma. Ann Surg Oncol 2001; 8(3):198–203.
101 Berd D, Sato T, Maguire HC, Jr., Kairys J, Mastrangelo MJ. Immunopharmacologic analysis of an autologous, haptenmodified human melanoma vaccine. J Clin Oncol 2004; 22(3):403–15. 102 Brown CD, Zitelli JA. The prognosis and treatment of true local cutaneous recurrent malignant melanoma. Dermatol Surg 1995; 21(4):285–90. 103 Thompson JF, Hunt JA, Shannon KF, Kam PC. Frequency and duration of remission after isolated limb perfusion for melanoma. Arch Surg 1997; 132(8):903–7. 104 Klaase JM, Kroon BB, van Geel AN, Eggermont AM, Franklin HR, Hart AA. Prognostic factors for tumor response and limb recurrence-free interval in patients with advanced melanoma of the limbs treated with regional isolated perfusion with melphalan. Surgery 1994; 115(1):39–45. 105 Noorda EM, Vrouenraets BC, Nieweg OE, van Geel BN, Eggermont AM, Kroon BB. Isolated limb perfusion for unresectable melanoma of the extremities. Arch Surg 2004; 139(11):1237–42. 106 Lienard D, Eggermont AM, Koops HS, et al. Isolated limb perfusion with tumour necrosis factor-alpha and melphalan with or without interferon-gamma for the treatment of intransit melanoma metastases: a multicentre randomized phase II study. Melanoma Res 1999; 9(5):491–502. 107 Cornett WR, McCall LM, Petersen RP, et al. American College of Surgeons Oncology Group Trial Z0020. Randomized multicenter trial of hyperthermic isolated limb perfusion with melphalan alone compared with melphalan plus tumor necrosis factor: American College of Surgeons Oncology Group Trial Z0020 J Clin Oncol 2006; 24(25):4196–201. 108 Eton O, East M, Legha SS, et al. Pilot study of intra-arterial cisplatin and intravenous vinblastine and dacarbazine in patients with melanoma in-transit metastases. Melanoma Res 1999; 9(5):483–9. 109 Lindner P, Doubrovsky A, Kam PC, Thompson JF. Prognostic factors after isolated limb infusion with cytotoxic agents for melanoma. Ann Surg Oncol 2002; 9(2):127–36. 110 Hill S, Thomas JM. Treatment of cutaneous metastases from malignant melanoma using the carbon-dioxide laser. Eur J Surg Oncol 1993; 19(2):173–7. 111 Konefal JB, Emami B, Pilepich MV. Analysis of dose fractionation in the palliation of metastases from malignant melanoma. Cancer 1988; 61(2):243–6. 112 Rate WR, Solin LJ, Turrisi AT. Palliative radiotherapy for metastatic malignant melanoma: brain metastases, bone metastases, and spinal cord compression. Int J Radiat Oncol Biol Phys 1988; 15(4):859–64. 113 Harpole DH, Jr., Johnson CM, Wolfe WG, George SL, Seigler HF. Analysis of 945 cases of pulmonary metastatic melanoma. J Thorac Cardiovasc Surg 1992; 103(4):743–8; discussion 748–50. 114 Krige JE, Nel PN, Hudson DA. Surgical treatment of metastatic melanoma of the small bowel. Am Surg 1996; 62(8):658–63. 115 Ricaniadis N, Konstadoulakis MM, Walsh D, Karakousis CP. Gastrointestinal metastases from malignant melanoma. Surg Oncol 1995; 4(2):105–10.
References 901
116 Bleehen NM, Newlands ES, Lee SM, et al. Cancer Research Campaign phase II trial of temozolomide in metastatic melanoma. J Clin Oncol 1995; 13(4):910–3. 117 Middleton MR, Grob JJ, Aaronson N, et al. Randomized phase III study of temozolomide versus dacarbazine in the treatment of patients with advanced metastatic malignant melanoma. J Clin Oncol 2000; 18(1):158–66. 118 Tan EH, Ang PT. Combination chemotherapy (dacarbazine, carmustine, cisplatin, and tamoxifen) in advanced melanoma. Singapore Med J 1996; 37(2):165–7. 119 Fierro MT, Bertero M, Novelli M, et al. Therapy for metastatic melanoma: effective combination of dacarbazine, carmustine, cisplatin and tamoxifen. Melanoma Res 1993; 3(2):127–31. 120 Del Prete SA, Maurer LH, O’Donnell J, Forcier RJ, LeMarbre P. Combination chemotherapy with cisplatin, carmustine, dacarbazine, and tamoxifen in metastatic melanoma. Cancer Treat Rep 1984; 68(11):1403–5. 121 Chapman PB, Einhorn LH, Meyers ML, et al. Phase III multicenter randomized trial of the Dartmouth regimen versus dacarbazine in patients with metastatic melanoma. J Clin Oncol 1999; 17(9):2745–51. 122 Legha SS. The role of interferon alfa in the treatment of metastatic melanoma. Semin Oncol 1997; 24(1 Suppl 4):S24–31. 123 Rosenberg SA, Yang JC, Topalian SL, et al. Treatment of 283 consecutive patients with metastatic melanoma or renal cell cancer using high-dose bolus interleukin 2 JAMA 1994; 271(12):907–13. 124 Atkins MB, Lotze MT, Dutcher JP, et al. High-dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J Clin Oncol 1999; 17(7):2105–16. 125 Atkins MB, Kunkel L, Sznol M, Rosenberg SA. High-dose recombinant interleukin-2 therapy in patients with metastatic melanoma: long-term survival update. Cancer J Sci Am 2000; 6 Suppl 1:S11–4. 126 Sparano JA, Fisher RI, Sunderland M, et al. Randomized phase III trial of treatment with high-dose interleukin-2 either alone or in combination with interferon alfa-2a in patients with advanced melanoma. J Clin Oncol 1993; 11(10):1969–77. 127 Khayat D, Antoine E, Rixe O, et al. Chemoimmunotherapy of metastatic malignant melanoma. The Salpetriere Hospital (SOMPS) experience. Eur J Cancer 1993; 29A Suppl 5:S2–5. 128 Legha SS, Ring S, Bedikian A, et al. Treatment of metastatic melanoma with combined chemotherapy containing cisplatin, vinblastine and dacarbazine (CVD) and biotherapy using interleukin-2 and interferon-alpha. Ann Oncol 1996; 7(8):827–35. 129 Keilholz U, Goey SH, Punt CJ, et al. Interferon alfa-2a and interleukin-2 with or without cisplatin in metastatic melanoma: a randomized trial of the European Organization for Research and Treatment of Cancer Melanoma Cooperative Group. J Clin Oncol 1997; 15(7):2579–88. 130 Johnston SR, Constenla DO, Moore J, et al. Randomized phase II trial of BCDT [carmustine (BCNU), cisplatin,
131
132
133
134
135
136
137
138
139
140
141
dacarbazine (DTIC) and tamoxifen] with or without interferon alpha (IFN-alpha) and interleukin (IL-2) in patients with metastatic melanoma. Br J Cancer 1998; 77(8):1280–6. Rosenberg SA, Yang JC, Schwartzentruber DJ, et al. Prospective randomized trial of the treatment of patients with metastatic melanoma using chemotherapy with cisplatin, dacarbazine, and tamoxifen alone or in combination with interleukin-2 and interferon alfa-2b. J Clin Oncol 1999; 17(3):968–75. Eton O, Legha SS, Bedikian AY, et al. Sequential biochemotherapy versus chemotherapy for metastatic melanoma: results from a phase III randomized trial. J Clin Oncol 2002; 20(8):2045–52. Ridolfi R, Chiarion-Sileni V, Guida M, et al. Cisplatin, dacarbazine with or without subcutaneous interleukin-2, and interferon alpha-2b in advanced melanoma outpatients: results from an Italian multicenter phase III randomized clinical trial. J Clin Oncol 2002; 20(6):1600–7. Atkins MB, Lee S, Flaherty LE, Sosman JA, Sondak VK, Kirkwood JM. A prospective randomized phase III trial of concurrent biochemotherapy (BCT) with cisplatin, vinblastine, dacarbazine (CVD), IL-2 and interferon alpha-2b (IFN) versus CVD alone in patients with metastatic melanoma (E3695): An ECOG- coordinated intergroup trial. Proc Am Soc Clin Oncol 2003; 22:708 (abstr 2847). Del Vecchio M, Bajetta E, Vitali M, et al. Multicenter phase III randomiaed trial of cisplatin, vindesine and dacarbazine (CVD) versus CVD plus subcutaneous (sc) interleukin-2 (IL-2) and interferon-alpha-2b (IFN) in metastatic melanoma patients (pts). Proc Am Soc Clin Oncol 2003; 22:709 (abstr 2849). Keilholz U, Punt CJ, Gore M, et al. Dacarbazine, cisplatin, and interferon-alfa-2b with or without interleukin-2 in metastatic melanoma: a randomized phase III trial (18951) of the European Organisation for Research and Treatment of Cancer Melanoma Group. J Clin Oncol 2005; 23(27):6747–55. Morton DL, Foshag LJ, Hoon DS, et al. Prolongation of survival in metastatic melanoma after active specific immunotherapy with a new polyvalent melanoma vaccine. Ann Surg 1992; 216(4):463–82. Serono Corporation. Serono announces discontinuation of onercept in moderate-to-severe psoriasis and Canvaxin in stage IV melanoma. In: www.serono.com press release; April 2005. Elliott GT, McLeod RA, Perez J, Von Eschen KB. Interim results of a phase II multicenter clinical trial evaluating the activity of a therapeutic allogeneic melanoma vaccine (theraccine) in the treatment of disseminated malignant melanoma. Semin Surg Oncol 1993; 9(3):264–72. Mitchell MS. Perspective on allogeneic melanoma lysates in active specific immunotherapy. Semin Oncol 1998; 25(6):623–35. Berd D, Maguire HC, Jr., McCue P, Mastrangelo MJ. Treatment of metastatic melanoma with an autologous tumor-cell vaccine: clinical and immunologic results in 64 patients. J Clin Oncol 1990; 8(11):1858–67.
902 Malignant melanoma
142 Nestle FO, Alijagic S, Gilliet M, et al. Vaccination of melanoma patients with peptide- or tumorlysate-pulsed dendritic cells. Nat Med 1998; 4(3):328–32. 143 Schadendorf D, Nestle FO, Broecker EB, et al. Dacarbazine (DTIC) versus vaccination with autologous peptide-pulsed dendritic cells (DC) in first-line treatment of patients with metastatic melanoma: a randomized phase III trial of the DC study group of the DeCOG. Ann Oncol 2006; 17(4):563–70. 144 Belli F, Testori A, Rivoltini L, et al. Vaccination of metastatic melanoma patients with autologous tumor-derived heat shock protein gp96-peptide complexes: clinical and immunologic findings. J Clin Oncol 2002; 20(20):4169–80. 145 Richards J, Testori A, Whitman E, et al. Autologous tumorderived HSPPC-96 vs. physician’s choice (PC) in a randomized phase III trial in stage IV melanoma. J Clin Oncol ASCO Annual Meeting Proceedings Part I. 2006; 24(18S):8002. 146 Rosenberg SA, Yang JC, Schwartzentruber DJ, et al. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma. Nat Med 1998; 4(3):321–7. 147 Lindsey KR, Gritz L, Sherry R, et al. Evaluation of prime/boost regimens using recombinant poxvirus/tyrosinase vaccines for the treatment of patients with metastatic melanoma. Clin Cancer Res 2006; 12(8):2526–37 148 Hodi FS, Mihm MC, Soiffer RJ, et al. Biologic activity of cytotoxic T lymphocyte-associated antigen 4 antibody blockade in previously vaccinated metastatic melanoma and ovarian carcinoma patients. Proc Natl Acad Sci USA 2003; 100(8):4712–7. 149 Ribas A, Camacho LH, Lopez-Berestein G, et al. Antitumor activity in melanoma and anti-self responses in a phase I trial with the anti-cytotoxic T lymphocyte-associated antigen 4 monoclonal antibody CP-675,206 J Clin Oncol 2005; 23(35):8968–77. 150 Attia P, Phan GQ, Maker AV, et al. Autoimmunity correlates with tumor regression in patients with metastatic melanoma treated with anti-cytotoxic T-lymphocyte antigen-4. J Clin Oncol 2005; 23(25):6043–53. 151 Maker AV, Phan GQ, Attia P, et al. Tumor regression and autoimmunity in patients treated with cytotoxic T lymphocyte-associated antigen 4 blockade and interleukin 2: a phase I/II study. Ann Surg Oncol 2005; 12(12):1005–16. 152 Skibber JM, Soong SJ, Austin L, Balch CM, Sawaya RE. Cranial irradiation after surgical excision of brain metastases in melanoma patients. Ann Surg Oncol 1996; 3(2):118–23. 153 Fife KM, Colman MH, Stevens GN, et al. Determinants of outcome in melanoma patients with cerebral metastases. J Clin Oncol 2004; 22(7):1293–300. 154 Sampson JH, Carter JH, Jr., Friedman AH, Seigler HF. Demographics, prognosis, and therapy in 702 patients with brain metastases from malignant melanoma. J Neurosurg 1998; 88(1):11–20. 155 Konstadoulakis MM, Messaris E, Zografos G, Androulakis G, Karakousis C. Prognostic factors in malignant melanoma
156
157
158
159
160
161
162
163
164
165
166
167
168
patients with solitary or multiple brain metastases. Is there a role for surgery? J Neurosurg Sci 2000; 44(4):211–8; discussion 219. Hagen NA, Cirrincione C, Thaler HT, DeAngelis LM. The role of radiation therapy following resection of single brain metastasis from melanoma. Neurology 1990; 40(1): 158–60. Mori Y, Kondziolka D, Flickinger JC, Kirkwood JM, Agarwala S, Lunsford LD. Stereotactic radiosurgery for cerebral metastatic melanoma: factors affecting local disease control and survival. Int J Radiat Oncol Biol Phys 1998; 42(3):581–9. Koc M, McGregor J, Grecula J, Bauer CJ, Gupta N, Gahbauer RA. Gamma Knife radiosurgery for intracranial metastatic melanoma: an analysis of survival and prognostic factors. J Neurooncol 2005; 71(3):307–13. Manon RR, Oneill A, Mehta M, et al. Phase II trial of radiosurgery (RS) for 1 to 3 newly diagnosed brain metastases from renal, melanoma and sarcoma: An Eastern Cooperative Oncology group Study (E6397). ASCO Annual Meeting Proceedings 2004; 22 (14S: 1507). Morris SL, Low SH, A’Hern RP, et al. A prognostic index that predicts outcome following palliative whole brain radiotherapy for patients with metastatic malignant melanoma. Br J Cancer 2004; 91(5):829–33. Mornex F, Thomas L, Mohr P, et al. A prospective randomized multicentre phase III trial of fotemustine plus whole brain irradiation versus fotemustine alone in cerebral metastases of malignant melanoma. Melanoma Res 2003; 13(1):97–103. Avril MF, Aamdal S, Grob JJ, et al. Fotemustine compared with dacarbazine in patients with disseminated malignant melanoma: a phase III study. J Clin Oncol 2004; 22(6):1118–25. Agarwala SS, Kirkwood JM, Gore M, et al. Temozolomide for the treatment of brain metastases associated with metastatic melanoma: a phase II study. J Clin Oncol 2004; 22(11):2101–7. Hwu WJ, Lis E, Menell JH, et al. Temozolomide plus thalidomide in patients with brain metastases from melanoma: a phase II study. Cancer 2005; 103(12):2590–7. Atkins MB, Sosman J, Agarwala S, et al. A Cytokine Working Group phase II study of temozolamide (TMZ), thalidomide (THAL) and whole brain radiotherapy (WBRT) for patients with brain metastases from melanoma. ASCO Annual Meeting Proceedings 2005; 23 (16S: 7552). Eisen T, Ahmad T, Gore ME, et al. Phase I trial of BAY 43-9006 (sorafenib) combined with dacarbazine (DTIC) in metastatic melanoma patients. ASCO Annual Meeting Proceedings 2005; 23 (16S: 7508). Eisen T, Ahmad T, Flaherty KT, et al. Sorafenib in advanced melanoma: a Phase II randomised discontinuation trial analysis. Br J Cancer 2006; 95(5):581–6. Eisen T, Marais R, Affolter A, et al., for the 11538 study investigators. An open-label phase II study of sorafenib and dacarbazine as first-line therapy in patients with advanced
References 903
169
170
171
172
173
174
175
176
177
178
179
180
melanoma. J Clin Oncol 2007 ASCO Annual Meeting Proceedings Part I. 2007; 25(18S):8529. Agarwala SS, Keilholz U, Hogg D, et al. Randomized phase III study of paclitaxel plus carboplatin with or without sorafenib as second-line treatment in patients with advanced melanoma. J Clin Oncol 2007 ASCO Annual Meeting Proceedings Part I. 2007; 25(18S):8510. Amaravadi R, Schuchter LM, McDermott DF, et al. Updated results of a randomized phase II study comparing two schedules of temozolomide in combination with sorafenib in patients with advanced melanoma. J Clin Oncol 2007 ASCO Annual Meeting Proceedings Part I. 2007; 25(18S):8527. Ahmad T, Marais R, Pyle L, James M, Schwartz B, Gore M, Eisen T. BAY 43-9006 in patients with advanced melanoma: The Royal Marsden experience. J Clin Oncol 2004 ASCO Annual Meeting Proceedings. 2004; 22(14S):7506. Hardin E, Pavlick AC, Liebes L, et al. A phase II trial of BAY 43-9006 in metastatic melanoma with molecularly characterized B-Raf status. J Clin Oncol 2006 ASCO Annual Meeting Proceedings Part I. 2006; 24(18S):8046. Dudley ME, Rosenberg SA. Adoptive-cell-transfer therapy for the treatment of patients with cancer. Nat Rev Cancer 2003; 3(9):666–75. Dudley ME, Wunderlich JR, Yang JC, et al. Adoptive cell transfer therapy following non-myeloablative but lymphodepleting chemotherapy for the treatment of patients with refractory metastatic melanoma. J Clin Oncol 2005; 23(10):2346–57. Palmer K, Moore J, Everard M, et al. Gene therapy with autologous, interleukin 2-secreting tumor cells in patients with malignant melanoma. Hum Gene Ther 1999; 10(8):1261–8. Schreiber S, Kampgen E, Wagner E, et al. Immunotherapy of metastatic malignant melanoma by a vaccine consisting of autologous interleukin 2-transfected cancer cells: outcome of a phase I study. Hum Gene Ther 1999; 10(6):983–93. Moller P, Sun Y, Dorbic T, et al. Vaccination with IL-7 genemodified autologous melanoma cells can enhance the antimelanoma lytic activity in peripheral blood of patients with a good clinical performance status: a clinical phase I study. Br J Cancer 1998; 77(11):1907–16. Sun Y, Jurgovsky K, Moller P, et al. Vaccination with IL-12 gene-modified autologous melanoma cells: preclinical results and a first clinical phase I study. Gene Ther 1998; 5(4):481–90. Nemunaitis J, Bohart C, Fong T, et al. Phase I trial of retroviral vector-mediated interferon (IFN)-gamma gene transfer into autologous tumor cells in patients with metastatic melanoma. Cancer Gene Ther 1998; 5(5):292–300. Soiffer R, Hodi FS, Haluska F, et al. Vaccination with irradiated, autologous melanoma cells engineered to secrete granulocyte-macrophage colony-stimulating factor by adenoviral-mediated gene transfer augments antitumor immunity in patients with metastatic melanoma. J Clin Oncol 2003; 21(17):3343–50.
181 Belli F, Arienti F, Sule-Suso J, et al. Active immunization of metastatic melanoma patients with interleukin-2transduced allogeneic melanoma cells: evaluation of efficacy and tolerability. Cancer Immunol Immunother 1997; 44(4):197–203. 182 Arienti F, Belli F, Napolitano F, et al. Vaccination of melanoma patients with interleukin 4 gene-transduced allogeneic melanoma cells. Hum Gene Ther 1999; 10(18):2907–16. 183 Mastrangelo MJ, Maguire HC, Jr., Eisenlohr LC, et al. Intratumoral recombinant GM-CSF-encoding virus as gene therapy in patients with cutaneous melanoma. Cancer Gene Ther 1999; 6(5):409–22. 184 Chang AE, Li Q, Bishop DK, Normolle DP, Redman BD, Nickoloff BJ. Immunogenetic therapy of human melanoma utilizing autologous tumor cells transduced to secrete granulocyte-macrophage colony-stimulating factor. Hum Gene Ther 2000; 11(6):839–50. 185 Jansen B, Schlagbauer-Wadl H, Brown BD, et al. bcl-2 antisense therapy chemosensitizes human melanoma in SCID mice. Nat Med 1998; 4(2):232–4. 186 Bedikian AY, Millward M, Pehamberger H, et al.; Oblimersen Melanoma Study Group. Bcl-2 antisense (oblimersen sodium) plus dacarbazine in patients with advanced melanoma: the Oblimersen Melanoma Study Group. J Clin Oncol 2006; 24(29):4738–45. 187 Shashkova EV, Cherenova LV, Kazansky DB, Doronin K. Avian adenovirus vector CELO-TK displays anticancer activity in human cancer cells and suppresses established murine melanoma tumors. Cancer Gene Ther 2005; 12(7): 617–26. 188 Wang S, Qi J, Smith M, Link CJ. Antitumor effects on human melanoma xenografts of an amplicon vector transducing the herpes thymidine kinase gene followed by ganciclovir. Cancer Gene Ther 2002; 9(1):1–8. 189 Klatzmann D, Cherin P, Bensimon G, et al. A phase I/II doseescalation study of herpes simplex virus type 1 thymidine kinase ‘suicide’ gene therapy for metastatic melanoma. Study Group on Gene Therapy of Metastatic Melanoma. Hum Gene Ther 1998; 9(17):2585–94. 190 Seddon JM, Gragoudas ES, Egan KM, et al. Relative survival rates after alternative therapies for uveal melanoma. Ophthalmology 1990; 97(6):769–77. 191 Augsburger JJ, Correa ZM, Freire J, Brady LW. Long-term survival in choroidal and ciliary body melanoma after enucleation versus plaque radiation therapy. Ophthalmology 1998; 105(9):1670–8. 192 Diener-West M, Earle JD, Fine SL, et al. The COMS randomized trial of iodine 125 brachytherapy for choroidal melanoma, III: initial mortality findings. COMS Report No. 18. Arch Ophthalmol 2001; 119(7):969–82. 193 Hawkins BS. The Collaborative Ocular Melanoma Study (COMS) randomized trial of pre-enucleation radiation of large choroidal melanoma: IV. Ten-year mortality findings and prognostic factors. COMS report number 24. Am J Ophthalmol 2004; 138(6):936–51.
904 Malignant melanoma
194 Aoyama T, Mastrangelo MJ, Berd D, et al. Protracted survival after resection of metastatic uveal melanoma. Cancer 2000; 89(7):1561–8. 195 Hsueh EC, Essner R, Foshag LJ, Ye X, Wang HJ, Morton DL. Prolonged survival after complete resection of metastases from intraocular melanoma. Cancer 2004; 100(1):122–9. 196 Leyvraz S, Spataro V, Bauer J, et al. Treatment of ocular melanoma metastatic to the liver by hepatic arterial chemotherapy. J Clin Oncol 1997; 15(7):2589–95.
197 Leyvraz S, Bosshard W, Salmon R, Fiorentini G, Keilholz U, Gillet M. Prolonged survival of patients with liver metastases from ocular melanoma: multicentric experience with fotemustine hepatic arterial infusion. Proc Am Soc Clin Oncol 2002; 21:341a. 198 Alexander HR, Jr., Libutti SK, Pingpank JF, et al. Hyperthermic isolated hepatic perfusion using melphalan for patients with ocular melanoma metastatic to liver. Clin Cancer Res 2003; 9(17):6343–9.
39 Bone JEREMY S. WHELAN, ROB C. POLLOCK AND ANNA M. CASSONI
Introduction Incidence and aetiology General principles of management Osteosarcoma Ewing’s sarcoma family of tumours
905 905 906 911 917
INTRODUCTION Cancers which arise in bone are exceptionally uncommon. Several discrete diseases can be identified through their differing clinico-pathological features, but for all bone tumours pain is the most common presenting symptom. As this is a feature of a vast array of musculoskeletal disorders, delay in diagnosis is a well-recognized problem. Principles of management, particularly surgical, are broadly applicable across the different histological types. Osteosarcoma and Ewing’s tumours together constitute the greatest numbers of cases. Occurring most often in teenagers and young adults, both diseases are now curable in a significant proportion of cases with modern multimodality treatment. For chondrosarcoma, a locally aggressive tumour of adults, there have been fewer advances in either our knowledge of the underlying biology or treatment of recurrent disease. Advances in surgical techniques have been substantial. Thus amputation is now uncommonly performed for extremity tumours. As improvements in systemic therapy have increased the number of survivors, so attention has turned to improving long-term functional outcomes. Of critical importance in the care of these diseases is close co-operation between experienced surgeons, radiologists and oncologists. Appropriate supportive care facilities and experienced nursing staff are necessary for those receiving intensive chemotherapy and, as many are children and teenagers, a multidisciplinary approach is required to meet the additional needs which arise, including educational provision and
Malignant fibrous histiocytoma of bone Primary non-Hodgkin’s lymphoma of bone (PLB) Chondrosarcoma Key references References
920 920 921 921 922
psychosocial support for both patient and family. All those with suspected primary bone tumours should therefore be referred quickly to a recognized specialist centre.
INCIDENCE AND AETIOLOGY A classification of primary bone tumours is shown in Table 39.1. These diseases do not appear to share a common aetiology, although several aetiological factors are apparent, particularly for osteosarcoma. Accurate data for the incidence of these diseases are not readily available. Cancer registry data may not always contain accurate histological information, particularly for tumours occurring in adults. On the other hand, specialist bone tumour registers may not give a true reflection of population incidence, but rather of referral patterns, especially when a registry is located in a specialist centre. The patterns of incidence differ between the diseases shown in Table 39.1. Both osteosarcoma and Ewing’s tumour have a peak incidence early in adolescence. This occurs slightly earlier in girls, and the association with a pubertal growth spurt, plus the common location of tumours around the knee and in the proximal humerus, indicate an aetiological association with rapid bone growth. While Ewing’s tumours occur very uncommonly after the age of 40 years, there is a low incidence of osteosarcoma throughout adulthood and, indeed, a second peak of incidence in the elderly that is explained in part by an association with Paget’s disease. Both osteosarcoma and Ewing’s tumours tend to have a male preponderance when occurring early in
906 Bone
Table 39.1 Classification of malignant primary bone tumours A Osteosarcoma 1. High-grade central Mixed, fibroblastic, osteoblastic, chondroblastic, osteoclast-rich, small cell 2. Low-grade central 3. Surface High-grade surface, periosteal, parosteal B Ewing’s family of tumours C Chondrosarcoma 1. Chondrosarcoma (grades 1–3) 2. Dedifferentiated chondrosarcoma 3. Mesenchymal chondrosarcoma D Malignant fibrous histiocytoma of bone E Other spindle cell tumours of bone Fibrosarcoma, leiomyosarcoma, liposarcoma, haemangiopericytoma, haemangioendothelioma F
Primary bone lymphoma
G Post-radiation sarcoma H Paget’s sarcoma Table 39.2 Aetiology of malignant primary bone tumours A
Genetic syndromes Familial retinoblastoma Li–Fraumeni syndrome Multiple enchondromata (diaphyseal achalasia)
B
Radiation
C
Paget’s disease
D
Miscellaneous, e.g. polyostotic fibrous dysplasia
life, and Ewing’s tumours have an unexplained racial pattern, being exceptionally uncommon in Africans and AfricanAmericans. In the majority, the occurrence of any of the tumours shown in Table 39.1 is a sporadic event. However, there are some well-recognized aetiological factors, of which the most important is radiation. More recently, a number of genetic abnormalities associated with the development of bone tumours have been identified. These factors are summarized in Table 39.2. Radiation was identified as an important causative factor for bone tumours from observations of luminous dial painters after the First World War. Large-scale production of instrument dials and watches took place in Canada and the USA. The dials were hand painted, most often by young female workers using paint containing radium and mesothorium mixed with zinc sulphide. The practice of pointing the paintbrushes in the workers’ mouths led to widespread ingestion of radium. Radionecrosis, primarily of the jaw but also occurring in other bones, resulted and over the next 40 years osteosarcomas frequently developed, those ingesting more than 700 mCi having a cumulative incidence as high as 70 per cent during this period.1
Sarcomas are a recognized late complication of therapeutic radiation. The most common histological subtype is osteosarcoma, and tumours arise after a latent period averaging between 8 and 20 years. Of greatest concern is the development of a radiation-induced sarcoma as a consequence of successful treatment of childhood malignancy. Although the overall incidence is low, it has become an important factor to consider in the development of new treatments which now attempt to limit use of radiation. Exposure to alkylating agents may add to this risk of sarcoma development. Radiation used in the treatment of breast and gynaecological cancers in adults is also associated with the development of sarcomas, although the risk appears to be less than that in children. Such tumours often pose considerable management problems. Survivors of retinoblastoma have a striking vulnerability to subsequent development of osteosarcoma. The excess risk (variously estimated as between 150- and 400fold) is principally in those with familial retinoblastoma and is especially associated with prior treatment with radiotherapy, with most, but not all, these secondary osteosarcomas arising within a field of previous irradiation .2 The identification of the retinoblastoma gene as a tumour suppressor gene altered in a wide range of cancers had led to insights into osteosarcoma pathogenesis. Loss of heterozygosity for Rb is a frequent finding, possibly associated with an adverse prognosis. A further key genetic discovery was made by Li and Fraumeni, who discovered families with an inherited predisposition to specific cancers, of which sarcomas in childhood and adolescence are particularly notable. This pre disposition arises as a consequence of a germ-line mutation in another tumour suppressor gene, p53.3 Tumours arising in Li–Fraumeni families account for a small but significant proportion of osteosarcoma.4 Sporadic mutations in p53 are thought to account for approximately 3 per cent of osteosarcomas. Thus, for most osteosarcomas the underlying pathogenesis remains obscure. The discovery of a chromosomal translocation specific to Ewing’s sarcoma (t11;22) has provided important insights, particularly in identifying primitive neuro-ectodermal tumours which share this translocation as having a common lineage. These related diseases are now usefully referred to as Ewing’s sarcoma family of tumours (ESFT). Although presumed to be neural, the identification of an originating cell for these tumours remains elusive. Other factors associated with the development of bone tumours include Paget’s disease.5 and some rare familial syndromes. Malignant fibrous histiocytoma of bone often arises in an area of pre-existing abnormal bone, such as an infarct.
GENERAL PRINCIPLES OF MANAGEMENT A broad framework of management, from initial evaluation to treatment, is appropriate for all patients with primary
General principles of management 907
bone tumours. At all stages, specialist experience and liaison between disciplines is essential.
Initial evaluation The most common presenting symptom is pain, often characterized by a gradual increase in intensity. An important distinguishing feature is that of night pain of sufficient severity to disturb sleep. This symptom should always be taken seriously and investigated appropriately. Swelling often accompanies pain, and again is a feature that demands rapid and careful investigation. Most patients, even those with high-grade tumours, have no systemic symptoms. When these occur, the manifestations are most often fever, weight loss and even sweats, usually as a consequence of metastatic ESFT. Like soft tissue sarcomas, the principal site of metastatic spread is to the lung, but this is only present in some 10–15 per cent of patients at diagnosis, and symptoms from lung metastases are much less frequent. Plain radiographs in two planes are the essential first investigation. As described in the section on individual tumour types, radiological features are often characteristic. Further evaluation of a primary tumour can then be undertaken with magnetic resonance imaging (MRI) and, particularly for lesions on the surface of the bone, computed tomography (CT). Radiological evaluation is mandatory before diagnostic biopsy is undertaken.
Biopsy Biopsy of bone tumours is an essential but potentially hazardous investigation. Several factors need to be taken into consideration. The biopsy must be sited appropriately since the biopsy tract is presumed to be contaminated and must be excised at the time of definitive surgery. Wrongly placed biopsy tracts and those that become infected or develop a haematoma may compromise future surgery.6 An adequate volume of representative tissue must be obtained and appropriately prepared. Ideally the specimen is sent fresh to the pathologist. Bone tumours are often heterogeneous, so the placement of the biopsy should take into account radiological changes that may indicate areas of higher or lower grade of malignancy within the same tumour. Nowadays most bone biopsies are CT guided and this has been shown to improve diagnostic accuracy.7 If a lesion is small, conveniently located, appears relatively benign radiologically and excision biopsy is considered to be technically straightforward, this should be the first surgical procedure. In all other cases the lesion should be biopsied before any treatment is undertaken. For diagnostic purposes, a core biopsy, using a disposable bone marrow biopsy trephine, provides adequate tissue for the diagnosis of all musculoskeletal lesions.8
Very hard lesions, such as periosteal osteosarcomas, may require a fine-toothed trephine, and intra-osseous lesions may require the proximal cortex of bone to be drilled. Incisional biopsy, however, has significant disadvantages in musculoskeletal oncology. Movement of tissue planes hampers subsequent effective excision of the biopsy scar and its contaminated field. Subsequent haematoma formation and infection can also occur. Expert interpretation of the biopsy is essential as most pathologists will be unfamiliar with primary bone tumours and the appearances are often misleading, for instance when distinguishing reactive from tumour bone in boneforming tumours such as osteosarcoma. Conventional histopathological techniques, together with immunocytochemistry, are now sufficiently advanced to provide a reliable diagnosis. Touch preparations are helpful for staining for alkaline phosphatase. The presence of this enzyme in malignant cells is a hallmark of osteosarcoma. In cases where Ewing’s tumour is suspected, molecular genetic studies are performed to confirm the presence of the t11;22 translocation which is diagnostic.
Staging The staging system recognized throughout the world is that published by Enneking and his associates in 1980.9 Two grades of tumour are recognized and two extents, within and without the cortex. As defined by these workers, this implied that almost all osteosarcomas are stage IIb as they almost invariably spread on both sides of the bone. Despite this ‘extra compartmental extent’, at operation these lesions are very well defined by pseudo-periosteum. We would therefore re-interpret Enneking’s definition of the compartment to be that bounded by the pseudo-periosteum as demonstrated on initial MRI scan and often more obvious after two courses of chemotherapy. Interpreted in this way, the great majority of osteosarcomas would be stage II and have much less risk of local recurrence than the very occasional tumour which does burst out and frankly invade muscle, unaided by biopsy. A practical staging notation used within the London Bone and Soft Tissue Tumour Service is found in Table 39.3. The patterns of spread of primary bone tumours are well characterized. Appropriate staging investigations are shown in Table 39.4. For those tumours with metastatic potential, the lungs and other sites in the skeleton are most often affected. Thus, CT of the chest and an isotope bone scan are necessary. Bone marrow biopsy is required for the assessment of Ewing’s tumours, when approximately 10 per cent will have evidence of bone marrow involvement.
Treatment The integration of local and systemic therapies is an important feature of the management of high-grade tumours,
908 Bone
Table 39.3 Staging notation Stage
Grade
Site
Treatment
Example
Ia Ib IIa IIb
Low Low High High
Within pseudo-periosteum Invading muscle Within pseudo-periosteum Invading muscle
Low-grade central osteosarcoma Parosteal osteosarcoma, chondrosarcoma Most high-grade osteosarcoma Ewing’s tumour
IIIa
High
Oligometastatic
IIIb
High
Polymetastatic
Wide local excision Wide local excision Chemotherapy, local excision Chemotherapy, local excision or radiotherapy if inaccessible Chemotherapy, local excision, metastatectomy Palliative chemotherapy and radiotherapy
Any sarcoma with few metastases Inoperable metastatic sarcoma
Table 39.4 Staging investigations
Osteosarcoma Ewing’s tumours Chondrosarcoma MFH Bone lymphoma
Plain radiographs
MRI of primary
Isotope bone scan
Thorax CT
Bone marrow biopsy
Others
Y Y Y Y Y
Y Y Y Y Y
Y Y Y Y Y
Y Y Y Y Y
N Y N N Y
Abdomen CT, LDH
MFH, malignant fibrous histiocytoma; CT, computed tomography; LDH, lactate dehydrogenase.
where most will begin treatment with chemotherapy, local therapy being carried out anywhere between 6 and 20 weeks later. Other tumours may be managed by definitive surgery alone, after biopsy and staging as described above.
Local therapy – surgery The guiding principles in sarcoma surgery are: 1. High-grade sarcoma is usually systemic at presentation. Surgery is therefore aimed at local disease control. 2. The surgical margins must be free of tumour. 3. The safe margin of resection includes a biologically competent barrier. 4. A tissue plane that has normal mobility is not crossed by the tumour. 5. Neoadjuvant chemotherapy may improve the local grade. MARGINS
The safe margin of resection should be decided on clinical grounds. This may be a layer of fascia, or epineurium. Within a compartment such as a bone or muscle, this margin is less obvious. There remains a spread of opinion across the world in this regard, with the Japanese Musculoskeletal Oncology Society recommending 5 cm of normal tissue, at one extreme, and the British Orthopaedic Oncology Society using 2 cm as the margin of error. MRI helps the surgeon decide this margin within the bone or muscle,
while the mobility of tissues is a better guide to involvement by tumour across adventitial layers. The surgeon is therefore the one who makes this decision, by observing that the structure is uninvolved, thus frequently saving a vessel or nerve which is displaced by the tumour rather than encased by it. This is at present a superior strategy to simple acceptance of MRI scanning information. The partial volume effects at the junction of different tissue types are indistinguishable from tumour involvement at our current level of understanding. LIMB SALVAGE VERSUS AMPUTATION
At present, there is no agreement regarding the place of limb salvage in paediatric and adolescent sarcoma surgery. Massive replacement, which is the mainstay of reconstructive surgery, has a significant re-operation rate, with a significant human and financial cost, although the costs of amputation are also considerable.10 It remains difficult to justify this on quality-of-life terms alone,11 but more and more surgeons do attempt to save limbs rather than amputate, even in the young. Amputation is only reserved for those cases in which there is segmental involvement of the neuro-vascular bundle as well as bone and muscle. LOCAL RECURRENCE
Local recurrence is often cited as a very major cause of relapse, with the implication that inadequate surgery is a cause of death. There are no controlled data to confirm
General principles of management 909
Table 39.5 Criteria for deciding operative margin Intra-osseous/intra-muscular extent Soft tissue extent Margin of excision
MRI: T1 weighting Examination, and operative findings Grade determined by biopsy
this, only the observation that the mortality rate in those centres with very aggressive surgeons is not significantly better than those with more conservative ones. In multivariate analysis, it is the size of the tumour and associated chemosensitivity that dictate this, not the type of surgery. Skip lesions may commonly be the cause of local recurrence, having developed early in the disease process as a result of a failure of cell-mediated control mechanisms. Thus local recurrence is usually a marker of disease activity, and a failure of host tumour control mechanisms at a cellular level, rather than a cause of it.
Table 39.6 Surgical grade of tumour Histological grade Low Intermediate with no adjuvants Intermediate with adjuvants available High with no preoperative treatment High with response to chemotherapy
SURGICAL GRADE
The surgical grade of a lesion will determine how much margin the surgeon should use. In Table 39.6, the margin is seen to be as wide in an intermediate-grade lesion such as parosteal osteosarcoma as in a high-grade lesion. With effective neoadjuvant chemotherapy, however, a high-grade Ewing’s sarcoma, for instance, may be treated with significantly smaller margins than a chondrosarcoma of much lower grade for which no adjuvant therapies are effective. RECONSTRUCTIVE STRATEGY
At present, reconstructive options world-wide vary according to availability of technology, surgical tradition, and the expectations of the patient group. A significant number of these patients are skeletally immature and predicted growth of the limb needs to be taken into account at the time of surgery. In general the excised segment of bone can be replaced with a metal endoprosthesis or a bone graft. Table 39.7 shows how such options are used in Europe today. Extendible endoprostheses are now available which can be lengthened in the out-patient setting without the
Narrow Wide Narrow: may save vessels and nerves Wide Intermediate or narrow in axial plane Still wide within bone or muscle
Table 39.7 The indications for different reconstructive methods Reconstruction method Massive prosthetic replacement
RESECTION STRATEGY
The surgical strategy combines both the resection and the reconstruction. The surgeon will plan his resection according to the examination and imaging, and the reconstruction according to the tools he has available. Resection of a bone sarcoma in the limbs depends upon the extent of the tumour, in both the longitudinal and axial directions, and the response to neoadjuvant treatment. Neoadjuvant chemotherapy may allow the surgeon to save structures that would have been sacrificed to avoid doubt. The local extent is decided by imaging studies and clinical findings, as shown in Table 39.5.
Required surgical margins
Expanding massive prosthetic replacement
Indications When a joint is involved When immediate strength is desirable If growth in that limb segment is anticipated
Biological replacement Diaphyseal resection close Pedicled graft, e.g. fibula to donor site Free vascularized graft, e.g. fibula, Diaphyseal resection with iliac crest, rib no local donor site 10 cm metadiaphyseal resection, with well preserved bone at site of lesion Allograft bone 10 cm metadiaphyseal resection, with bone destruction Combination graft (e.g. irradiated 10 cm metadiaphyseal autograft and vascularized fibula) resection, allowing joint sparing solution Exogenous grafts Irradiated/pasteurized resection specimen
need for multiple surgical procedures. This is a major technological advance and nowadays is often the reconstruction of choice in children. AXIAL TUMOURS
The surgery of axial tumours follows the same strategy as that for limb tumours. The surgeon and the patient have to make even harder choices regarding the sacrifice of function. Once again, the aggressive surgical choices of resection and reconstruction are often reserved for the intermediategrade lesion with longer life expectancy but no effective adjuvant: a chordoma or chondrosarcoma should be considered for wide local surgery even if mutilatory, while
910 Bone
Table 39.8 Principles of radiotherapy for bone tumours Action Plan with MRI and CT and after discussion with musculoskeletal radiologist and, if after surgery, the surgeon GTV based on largest tumour extent (except where it originally protruded into body cavity and has regressed) Margin along relevant tissue planes
Include scar and all prosthetic material, but as this may considerably extend volume, consider risk of contamination in individual case Spare a ‘corridor’ for lymphatic drainage Without compromising on target volume: Spare as much of joint as possible Spare epiphyses if possible Avoid gonads
Comment Definition of extent difficult, especially where there is oedema Risk of microscopic contamination after surgery or chemotherapyinduced regression Risk of spread along marrow cavity and especially in soft tissues along muscle planes. Intact fascia forms an effective barrier to spread There is a risk of microscopic disease here. However the size of this risk with chemotherapy and good surgical technique is not known High doses are used for these tumours. Failure to provide for lymphatic drainage will produce very severe effects The more spared, the less the risk of stiffness. However disability can be minimized by exercise 10 Gy will probably stop growth Careful technique may sometimes allow this
MRI, magnetic resonance imaging; CT, computed tomography; GTV, gross tumour volume.
a high-grade Ewing’s tumour or osteosarcoma would be re-staged after chemotherapy, and limb salvage might be applicable. In these circumstances, it may be appropriate to give all the chemotherapy preoperatively to minimize the impact of surgery, by improving the margins.
Local therapy – radiotherapy INDICATIONS
There are indications for radiotherapy that are common to several tumour types. Surgery is the local treatment of choice in most primary bone tumours except lymphoma. However, the decision on what is and what is not operable will depend on the likely success of the alternative modality and, particularly where survival is likely to be poor and unaffected by local issues, what level of morbidity is acceptable. Factors such as the skeletal maturity and the consequences on growth and function of surgery and radiotherapy, success of prosthetic replacement at each site and extent of soft tissue involvement must be considered in every case (Table 39.8). Residual microscopic disease at the site of radical surgery is a risk factor for local recurrence and is an indication for radiotherapy. Exceptions to this are where the tumour is extremely indolent and/or at a site where recurrences would be amenable to further surgery, or where the risk of radiation-induced damage is considered inappropriate to the eventual risk of recurrence. Open biopsy carries the risk of implantation along the biopsy track. Where this cannot be excised during definitive surgery postoperative radiotherapy is indicated. Displaced pathological fracture before surgery carries an increased risk of recurrence and may also be considered an indication for postoperative radiotherapy.
PRINCIPLES OF PLANNING
Radiotherapy for bone tumours is based on the same principles as soft tissue sarcoma and should be performed by clinical oncologists experienced in such tumours. Whilst the principles are clear, their application to individual sites and patients may be complex. Protocol deviations have been associated with reduced control, while centralization of treatment planning in the Cooperative Ewing’s Sarcoma Studies (CESS) improved local control.12 Meticulous immobilization, CT-based volume definition and optimization of dose distribution by compensators or in-field boost techniques should be standard. The presence of a metal-based limb prosthesis will perturb the dose distribution, producing areas of dose reduction along the axis of the incident beam.13 This effect may be minimized, in some circumstances, by using in-field boosts. In the treatment of limbs a strip of unirradiated skin and subcutaneous tissue sufficient to allow lymphatic drainage must be left. Where possible, the spared strip should be medial and joint spaces excluded or shielded. The traditional recommendation is that the whole scar and the full length of the prosthetic material should be included in the target volume. However the length of scars often reflects the requirements of inserting the prosthesis and may be difficult to cover without compromising function. The radiation target volume includes all tissues initially involved by tumour and is therefore based on the prechemotherapy extent, with margins along relevant tissue planes depending on the histology of the tumour and the precise anatomy of the part. This may be modified, for example, when a bulky tumour protrudes into a body cavity but has subsequently regressed. As much bowel as possible should be excluded from the treatment volume by positioning, treating with a full bladder if appropriate or,
Osteosarcoma 911
on occasion, displacement with a gel-filled prosthesis or an omental sling. Gonads may be repositioned in both females and young males,14,15 Radiotherapy for ESFT will be given after or with chemotherapy agents. These drugs may increase the risk of gut, renal, bladder and spinal cord toxicity where these are in the high-dose area. Radical treatments may be given in two phases: the first volume includes a margin of 3–5 cm or the tissues initially involved on imaging and in the second phase this is reduced to 2 cm. In the pelvis or spine, smaller margins are sometimes necessary to minimize toxicity. Where radiotherapy is adjuvant to surgery an initial wide volume is based on the above guidelines or inclusion of the whole operative bed, whichever is larger. For doses of over 45 Gy the volume may be reduced to the former only, excluding the full length of the operative field.
ADVERSE CONSEQUENCES OF RADIOTHERAPY
On the whole, acute complications are mild to moderate with conventional fractionation. However, when it is necessary to include perineum or significant amounts of bowel within the field these may produce severe acute toxicity. Of greater concern are the long-term effects. These relate primarily to abnormalities of growth and limb function and to induction of second malignancies. The risk of impaired function of organs such as kidney or heart from the radiotherapy must be added to the organ-specific toxic effects of chemotherapy used in multi-modality treatment. High-dose radiotherapy to long bones is associated with a higher frequency of fracture, especially where the full circumference must be irradiated, or where there has been extensive cortical bone destruction or biopsy. Effective chemotherapy used before radiotherapy may allow considerable resolution of tumour and healing of bone before radiotherapy is introduced, and it is anticipated that this will result in a reduced risk of fracture. Reports of effects on late function vary considerably in their frequency and severity. They are related to technique, volume, dose, patient age, pre-treatment function and the use of physiotherapy. While older series report complications in up to 30% of cases, more recent studies suggest that 80% of patients have normal or minimally affected limb function.16 Active physiotherapy, particularly aimed towards developing and maintaining joint extension and flexion, should be instituted early and continued for 1–2 years during which time the maximum radiation-induced fibrosis develops. Induction of second malignancy, primarily solid tumours, is an important consideration for radiotherapy of bone tumours. Several studies show an increased risk quantifying this at 2.7 times greater after radiotherapy and a 33% cumulative risk at 10 years,17,18 Risks are higher at doses of 60 Gy and when alkylating agents are also used. The risk persists even in the megavoltage era but appears to be less in more recently treated patients.
Table 39.9 Advantages of neoadjuvant chemotherapy for bone tumours 1. 2. 3. 4. 5.
Rapid improvement in symptoms Early treatment of micro-metastatic disease Facilitation of resection in responding tumours Time to manufacture customized endoprosthesis Prognostic information from assessment of histological response
Chemotherapy Osteosarcoma and Ewing’s tumours are fatal as a consequence of metastatic spread, principally to the lungs. Advances in chemotherapy over the past 30 years are largely responsible for the dramatic improvements in survival from these tumours. Significant proportions of patients remain incurable, often those who have advanced disease at presentation. Furthermore, current chemotherapy regimens are associated with considerable morbidity, both in the short and long term. As many survivors of bone tumours are young, late toxic effects, particularly cardiotoxicity and infertility, are of great concern to investigators planning effective treatments. Chemotherapy is often given after biopsy has established an appropriate diagnosis and before definitive local therapy, so-called neoadjuvant therapy. This practice began in the 1970s as orthopaedic techniques to avoid amputation improved. While rates of conversion of tumours from only suitable for amputation to suitable for limb salvage surgery may be low (approximately 10 per cent), operability is certainly facilitated in many tumours. Additionally, the prognostic information gained from histological scoring of chemotherapy response in the resected tumour remains a very valuable clinical tool. Other advantages of neoadjuvant chemotherapy are shown in Table 39.9. The role of chemotherapy in the management of less common high-grade primary bone tumours is not clearly defined and is discussed as appropriate in specific sections below.
OSTEOSARCOMA This is the most common high-grade primary tumour of bone. Survival has improved dramatically since the introduction of chemotherapy. Many of the principles underpinning the treatment of osteosarcoma are applicable to other bone tumours.
Subtypes of osteosarcoma HIGH-GRADE CENTRAL OSTEOSARCOMA
This variant accounts for most cases. Histologically there are areas of spindle-cell formation and variable amounts of
912 Bone
tumour bone formation. In other areas of the tumour there may be malignant cartilage or fibroblastic differentiation. Occasionally there is dense osteoblast activity giving rise to a great deal of new bone formation and radiological appearances of dense tumour bone. The malignant osteoblasts make alkaline phosphatase and this is a useful cytological stain for confirmation that the tumour is, in fact, an osteosarcoma. An uncommon variant is an aggressive form of osteosarcoma known as telangiectatic osteosarcoma. There is a lytic process radiologically, with little evidence of new bone formation. Histologically there are vascular spaces and the tumour can be mistaken for an aneurysmal bone cyst. Typically high-grade tumours start at the epiphysis. They expand within the bone and break through the cortex to lift the periosteum. The elevated periosteum is associated with new bone formation, giving rise to the typical Codman’s triangle. The tumour finally erupts through the periosteal boundary and impinges on the adjacent soft tissues. When very large, the tumour extends to the skin and incorporates the neurovascular bundle behind the knee or in the axilla. At this stage endoprosthetic replacement may be extremely difficult. LOW-GRADE CENTRAL OSTEOSARCOMA
usual analgesics but not relieved by rest. After a few weeks the pain intensifies and becomes constant. A characteristic feature, and one which would lead to earlier diagnosis if it were better appreciated, is that patients complain of intensification of pain at night time. After several weeks the second symptom appears, which is swelling. At first this may not be obvious since the soft tissues of the muscle and subcutaneous fat conceal the swelling. At this point the patients or parents become alarmed and usually insist on X-rays if these have not already been arranged. Sadly, several months often elapse before the diagnosis of a malignant bone tumour is made. This is due to the rarity of the tumours and because in the young, musculoskeletal pains, trivial injuries and strains are common. Very often the patients will have had physiotherapy for some weeks without avail and, indeed, physiotherapy may make symptoms worse. A long period of delay in diagnosis may create great tension in the family and sometimes considerable difficulties for the relationship between the patient and the family practitioner. Occasionally patients with osteosarcoma may have a very aggressive tumour that is accompanied by fever and constitutional malaise, anaemia and weight loss. This always implies an extremely bad prognosis and is usually associated with multiple bone or lung metastases.
This tumour has a slower rate of progression and is better demarcated radiologically. Histologically, it can be difficult to distinguish from fibrous dysplasia, but the cellular pleomorphism will usually give the diagnosis. It is rare and has a good prognosis when treated by surgery alone. SURFACE OSTEOSARCOMAS
Some osteosarcomas arise on the surface of the bone rather than centrally. Occasionally these are high-grade tumours but two other variants are more commonly recognized. Periosteal osteosarcoma appears to carry an intermediate course between high-grade tumours and the low-grade surface tumour, parosteal osteosarcoma. Although the prognosis is better than with classic central osteosarcoma, both parosteal and periosteal osteosarcomas may contain areas of high-grade tumour within them. They tend to arise in a more diaphyseal site than the central osteosarcomas, and the medulla of the bone is not involved until late. In parosteal tumours there is intense formation of trabecular bone and the tumour becomes heavily calcified with a lobulated appearance. The periosteal osteosarcoma contains more cartilage and the calcification is more punctate. Histologically, tumour cells which are making tumour bone and which have alkaline phosphatase in the cytoplasm help to differentiate the tumour from chondrosarcoma.
Clinical presentation The two major clinical features are pain and swelling. Typically pain precedes the swelling by weeks or months. It starts as an intermittent pain, partially relieved by the
Figure 39.1 High-grade osteosarcoma of lower femur. A diffuse permeating lesion is seen with elevation of the periosteum (Codman’s triangle).
Osteosarcoma 913
Osteosarcoma nearly always metastasizes to the lung in the first instance. However, bone metastases also occur and, with increasing experience of combination chemotherapy, recurrence at other sites such as skin, brain, and intraabdominally, is increasingly recognized. In the lung, metastases are usually asymptomatic. They are typically situated subpleurally where they may give rise to pneumothorax. If the metastasis involves the pleural space a pleural effusion develops. Centrally located metastases may compress one or other main bronchus and give rise to breathlessness, chest pain or haemoptysis.
Investigations Plain X-rays are essential. The characteristic features of a malignant bone tumour are a permeating, lytic lesion without any clear dividing boundary. When the tumour has broken through the cortex of the bone the periosteum is lifted and there may be formation of a Codman’s triangle
(a)
(Fig. 39.1). In osteosarcoma there may be varying degrees of tumour bone formation leading to dense sclerosis (Fig. 39.2a) or small areas of spiculation of bone, sometimes arranged at right angles to the long axis of the bone (Fig. 39.2b – ‘sunray spiculation’). Surface osteosarcomas are associated with the typical radiological appearances shown in Figure 39.3a and b. An isotopic bone scan will typically show an area of increased uptake of isotope at the site of the lesion, and may show bone metastases if present. A CT scan of the chest is essential and may show pulmonary metastases (Fig. 39.4). MRI is used to determine both the extent of the intra-medullary component of the tumour and the soft tissue extension of the mass (Fig. 39.5). Accurate assessment of audiological, cardiac, renal glomerular and tubule function should be carried out in all newly diagnosed patients with osteosarcoma. Periodic re-assessment during therapy is indicated, as nephrotoxicity due to agents such as cisplatin and ifosfamide or anthracycline-induced cardiotoxicity may require adjustments in treatment to minimize the risks of permanent damage.
(b)
Figure 39.2 (a) Plain X-ray of osteoblastic osteosarcoma, showing dense tumour bone surrounding the metaphysis of the upper humerus. (b) Plain X-ray of osteosarcoma of the lower third of the femur, showing ‘sunray spiculation’.
914 Bone
(b)
(a)
Figure 39.3 (a) Periosteal osteosarcoma, showing a permeating lesion in the upper tibia cortex. (b) X-ray of upper humerus, showing dense calcification in a parosteal osteosarcoma.
Figure 39.4 CT scan of the chest in a patient with osteosarcoma and a normal chest X-ray. A single subpleural metastasis is seen on the left. This is a typical position for bone sarcoma metastases.
Figure 39.5 MRI scan showing osteosarcoma of the left lower femur. The tumour has extended into the soft tissues. Typically, it has stopped at the level of the unfused epiphysis.
Osteosarcoma 915
Surgery Osteosarcoma is usually a tumour involving the long bones, so the lesion is always close to blood vessels. Using MRI scanning to define the intra-osseous extent, it is only necessary to excise sufficient normal bone to account for the surgical error in measuring the levels of bony resection. The vascular bundles close to the tumour may need to be excised with the tumour and replaced, or they can be dissected clear of the tumour. In making this decision preoperatively, it is often wise to wait for a second MRI scan after the preoperative chemotherapy cycles, since the situation may change. This second scan will often show the tumour itself to be separate from the main vessels. If there is definite involvement of the vascular bundle, this will need to be sacrificed. Sarcomas rarely invade peripheral nerves and excision is not usually needed. Epineural dissection of the nerves may be undertaken instead, although this means a marginal excision of the tumour. Reconstruction is planned after using MRI scans to determine the safe level of bone section and soft tissue dissection. Occasionally amputation may be the only possible procedure because of extensive soft tissue and skin involvement. However, the limb can usually be saved by use of a massive endoprosthesis. Because the tumour usually arises close to joints, the joint must often be excised and an artificial joint on the end of the titanium shaft will be substituted for the bone segment. The use of allograft bone as an alternative has not provided the long-term stability obtained with endoprostheses, and so is less favoured at the present time, although it is still an option. When circumstances allow, the joint closest to the tumour may be saved and a so-called ‘intercalary’ prosthesis used to take the place of the bone segment that has been excised. Many osteosarcoma patients are skeletally immature and the physis of the affected bone often has to be resected with the tumour in order to achieve a safe margin. Consequently growth of the affected limb will be reduced compared to the normal side. Modern extendable endoprostheses allowing limb lengthening in an out-patient setting are ideal for this group of patients. Limb segments may be reconstituted using vascularized fibular grafts or by means of bone transport, but both of these techniques require very extensive operative intervention. In these patients with high-grade tumours, the rapid rehabilitation associated with the use of massive prostheses is advantageous. This allows chemotherapy to begin again within 2–3 weeks of the surgery.
Chemotherap Before the 1970s osteosarcoma was a devastating disease, treatable only by amputation, with more than 80 per cent of patients succumbing rapidly due to lung metastases. Subsequent developments have transformed this appalling outlook. First, cytotoxic agents were identified which induced responses in advanced osteosarcoma. The most
important of these were methotrexate given at very high doses, doxorubicin and cisplatin. Second was the observation that resection of pulmonary metastases was both technically possible and profitable in terms of survival. Finally, improving surgical techniques allowed limb preservation in some cases. This last development led to the concept of preoperative chemotherapy, response to which seemed to facilitate ‘limb salvage’ surgery and which allowed time to manufacture custom-measured metallic endoprostheses. After 1975, evidence of a survival advantage for adjuvant chemotherapy grew, in particular through a series of uncontrolled studies conducted at the Memorial Sloan Kettering Hospital of combination chemotherapy given after surgery. These drug regimens used doxorubicin, high-dose methotrexate, and bleomycin, cyclophosphamide and actinomycin D – a combination known as BCD. These programmes evolved to produce a regimen called T10, for which impressive 2-year survival rates were claimed.19 At the time, these studies were difficult to interpret because of the lack of any concomitant control and the very short follow-up period. Alterations in selection criteria which had occurred during the 1970s – particularly the introduction of CT scanning – made firm assessment of the results difficult. The doubts were allayed to a considerable extent by two randomized studies. Link and co-workers compared T10-based chemotherapy with no chemotherapy. There was a clear difference in relapse-free survival in favour of chemotherapy at diagnosis.20 A further study from Eilber and colleagues randomized patients to receive no further chemotherapy following initial intra-arterial doxorubicin and tumour irradiation. The other arm of the randomization was to receive postoperative treatment with BCD and high-dose methotrexate. There was a significant improvement in relapse-free survival and overall survival in the small number of patients who were randomized.21 Since then, adjuvant chemotherapy has been accepted as part of standard management for high-grade osteosarcoma. Chemotherapy given prior to surgery for the primary has also become accepted, offering the advantages outlined in Table 39.9. Assessment of the histological response to chemotherapy has been identified as the most powerful prognostic factor in this disease; those patients who experience very extensive necrosis (in excess of 90 per cent) have an overall survival in excess of 70 per cent at 5 years, while those in whom the degree of necrosis falls short of this have a significantly inferior survival. There are theoretical disadvantages associated with preoperative chemotherapy, namely that poorly or non-responsive tumours may grow during treatment, increasing the risk of amputation and of distant spread. The single randomized trial to address this issue showed that those receiving preoperative chemotherapy enjoyed a significantly higher rate of limb salvage surgery without any survival disadvantage.22 The most active cytotoxic agents include methotrexate, doxorubicin, cisplatin, ifosfamide and etoposide. There have been no important recent additions to this list and clinical
916 Bone
studies have been aimed at definition of optimal treatment schedules. Intra-arterial chemotherapy has been shown to produce high rates of histological necrosis but does not improve survival and is not routinely indicated. There are relatively few randomized trials of chemotherapy for osteosarcoma that have sufficient statistical power to provide firm evidence for best practice. The most recent completed published study was undertaken by the Childrens Oncology Group in the USA. In a 2 2 factorial design, this study investigated the addition of ifosfamide to a standard arm of cisplatin, doxorubicin and methotrexate. A second randomization compared the addition of the biological agent MTPPE to chemotherapy. An unequivocal advantage for these interventions was not demonstrated although the results have attracted considerable attention.23 Important studies were carried out by the European Osteosarcoma Intergroup (EOI), the first of which aimed to define the importance of high-dose methotrexate, and the second of which compared a prolonged multi-drug regimen, similar to T10, to a two-drug combination given for just six cycles.24,25 The third study then compared the standard two-drug regimen of cisplatin and doxorubicin with a doseintensive schedule using granulocyte colony-stimulating factors. No advantage for dose intensity was seen and the overall survival in these studies falls short of that reported by groups using three or four drugs as standard.26 The EOI studies are summarized in Table 39.10. The studies described above have all taken many years to complete. Individual study groups have now joined together in recognition of both the need to improve survival and the advantages of international co-operation. The first combined European and American randomized study in osteosarcoma, EURAMOS 1, is now under way.27 Adaptation of therapy according to risk factors remains insufficiently developed for patients with extremity osteosarcoma. This is because the most powerful indicator of outcome, histological response, is only available well into any programme of systemic therapy. Furthermore, no study has clearly identified that alteration of therapy on the basis of histological response can improve survival for those with poor rates of necrosis. It is hoped that newer biological markers, such as c-erbB2 expression, may provide earlier indications of outcome that can be used to plan individualized
treatment. This is anticipated to be an important focus for clinical research in the future.
Radiotherapy The role of radiotherapy in osteosarcoma is confined largely to the treatment of inoperable tumours, where it may extend the period of local control but is unlikely to produce local cure. This relatively radioresistant tumour does have a degree of response to radiotherapy if high doses are given. Moderate term control doses of 70 Gy or more have been used.28 This produced, in unresected tumours, a reported local control rate of 20–25 per cent, and reports of tumour sterilization in amputation specimens of 33 per cent. Small, more recent studies, have also suggested a modest control rate with high doses. High-dose radiotherapy is required to control local symptoms for the remainder of the patient’s life. Symptomatic improvement is produced in approximately 50 per cent of patients. There is some evidence of a degree of benefit from adjuvant lung radiotherapy in microscopic disease but this role has been supplanted by combination chemotherapy.29
Treatment of metastases Presentation with pulmonary metastasis is a grave prognostic sign and almost no patients are cured if they present with bone metastases. Patients who present with one or two pulmonary metastases should be treated with chemotherapy. If they respond completely, a very close surveillance policy must be followed and resectable metastases should be removed if they re-appear on CT scanning. Patients presenting with multiple pulmonary metastases will almost certainly not be cured even if there is complete response to chemotherapy. Nevertheless treatment is worthwhile and very durable responses may be obtained. The more common situation is for metastases to appear after chemotherapy has been completed. Some patients are certainly cured by thoracotomy. Good prognostic factors for the success of thoracotomy are few metastases, unilateral rather than bilateral disease, a long period of freedom
Table 39.10 European Osteosarcoma Intergroup randomized studies of chemotherapy for extremity osteosarcoma Study
No. of patients
Arms
Outcome
Reference
1 (accrual 1983–86)
179
Doxorubicin cisplatin, vs. Doxorubicin cisplatin methotrexate
Equivalent overall survival
Bramwell et al. 199224
2 (1988–92)
391
Doxorubicin cisplatin vs. ‘T10’
Equivalent overall survival
Souhami et al. 199725
3 (1993–2002)
497
Doxorubicin cisplatin, 3 weekly vs. Doxorubicin cisplatin, 2 weekly with GCSF
Equivalent overall survival
Lewis et al. 200326
GCSF, granulocyte colony-stimulating factor.
Ewing’s sarcoma family of tumours 917
from relapse following the cessation of chemotherapy, and the peripheral location of the metastases.30 Patients who develop pulmonary metastases while on chemotherapy have a very bad prognosis indeed. Although the introduction of new agents may produce responses, and is worthwhile, the chances of cure are very small. Similarly, relapse immediately, or a few months after, the cessation of chemotherapy is also very adverse. Pleural effusion is an extremely adverse sign and nearly always indicates intra-pleural spread of the tumour, which is incurable. There are many questions still to be asked about optimum management of pulmonary metastases. It is not clear how often repeated thoracotomies are successful. Some patients have multiple thoracotomies, each of which appears justified but which fails to produce lasting benefit. Occasionally freedom from disease may be achieved, especially when the intervals between surgery are relatively long. Prophylactic pulmonary irradiation after metastatectomy has never been investigated appropriately to determine whether it might confer a lasting benefit.
Variants of osteosarcoma
tumours, primitive neuro-ectodermal tumours of bone and soft tissue which share the same cytogenetic abnormality but may express more neural markers.
Clinical features and diagnosis The presentation is as for osteosarcoma. The main symptoms are pain and swelling, and the location of the tumour is usually diaphyseal rather than epiphyseal. Unlike osteosarcoma, ESFTs are much more commonly found in the pelvis, the ribs and the axial skeleton. The most common sites are femur, humerus, ilium, other regions of the pelvis, ribs, skull and jaw, and small bones of the hands and feet. The tumour is typically permeating, spreading widely within the medulla and the vascular spaces of the bone cortex. It causes necrosis of bone, and the tumour itself may be necrotic. New bone is often laid down around the site of the tumour, giving rise, radiologically, to the typical ‘onion skin’ appearance. There is often thickening and sclerosis of the cortex of the bone, although occasionally Ewing’s tumours are purely destructive with widespread bone lysis (Fig. 39.6).
Small-cell osteosarcoma and telangiectatic osteosarcoma are high-grade tumours which should be treated as for conventional osteosarcoma. Parosteal osteosarcoma is of low-grade malignancy and is not usually treated with chemotherapy unless there is clear evidence of high-grade dedifferentiation within the tumour.31 Periosteal osteosarcoma is of higher grade than parosteal osteosarcoma but still with a lower risk of metastasis than conventional osteosarcoma.32,33 There is little definite evidence that chemotherapy benefits these patients, but it is usually given, especially if there are extensive areas of high-grade tumour. Post-irradiation and Paget’s osteosarcoma are always of high grade histologically and may have an appearance more like malignant fibrous histiocytoma. The main treatment is surgical resection, but this is often very difficult in view of their location. There are no survivors of Paget’s sarcoma but other secondary osteosarcomas are frequently cured if complete excision is possible.
EWING’S SARCOMA FAMILY OF TUMOURS James Ewing’s original descriptions of this disease remain pertinent today. It is a member of the group of small, round, blue cell tumours of childhood, with neuroblastoma, rhabdomyosarcoma and some lymphomas, distinguishable only by immunocytochemistry or biological markers. Its aetiology and cell of origin remain obscure despite the demonstration of characteristic chromosomal translocations, t(11;22) and, less commonly, t(21;22). These translocations have illuminated the close relationship between classical Ewing’s tumour and other related
Figure 39.6 X-ray of Ewing’s sarcoma of the upper femur. Dense cortical thickening is present with ‘onion-skin’ periosteal reaction.
918 Bone
Diagnosis is by biopsy, and it is important that the specimen is examined by a skilled pathologist in conjunction with the X-rays. The differential diagnosis will include nonmalignant conditions such as osteomyelitis. An appropriate panel of immunocytochemical markers must be used to exclude lymphomas and carcinomas. Various degrees of neural differentiation will be indicated by markers such as S-100 and chromogranin, and expression of CD99, although not specific for ESFT, is usually strongly positive. Access to molecular diagnostics to detect chromosomal translocations is now essential and should be regarded as the gold standard for diagnostics.
Investigations Investigations should include plain radiographs of the affected bone. An isotope bone scan will confirm the increased uptake in the presence of the primary tumour and may show other bone metastases. A CT scan of the lungs is essential to confirm or exclude visible pulmonary metastases and an MRI of the affected bone is useful both in assessing tumour size (and thus prognosis, see below) and in planning primary treatment. Staging is completed by examination of a bone marrow aspirate and trephine biopsy. Investigations of organ function should be carried out as for osteosarcoma.
Management When a patient with Ewing’s sarcoma first presents, it is essential that a thorough discussion takes place about the way in which management is to be conducted. This discussion must involve medical oncologists, radiotherapists and surgeons. The first decision to be made will concern the local treatment. Essentially there will be two choices, either that the lesion can be removed surgically or that the patient will have to receive radiation. Occasionally it will not be possible to decide definitively at the outset between these two alternatives, but the response to initial chemotherapy will have to be awaited in order to determine whether surgical resection is feasible. The decision will be based on several factors: ●
● ● ● ●
●
Is the lesion likely to be resectable in its entirety with uninvolved resection margins? Will radiotherapy thereby be avoided? Will the functional results of surgery be acceptable? Will the functional results of radiotherapy be acceptable? What is likely to be the long-term local control rate with either surgery or radiation? Will a synchronous combined programme of chemotherapy and radiation be feasible over the size of the planned radiation field?
The answers to these questions are extremely difficult and require great experience in the management of the tumours. On the one hand, small expendable bones can be
easily resected and the rate of local recurrence will be low. On the other hand, massive tumours in the pelvis which cannot be resected have to be treated with radiation, with a high risk of local recurrence because of the tumour size. In between, there will be highly complex decisions where the balance of advantages must be carefully assessed. The problem with radiation is that local control may not always be achieved, and that the risk of second cancer in the radiated bone is significant. In the case of surgery, functional results may be really quite poor and, if the resection margins are involved, the need for radiation will not have been avoided. On the other hand, the local recurrence rates with surgical resection with or without radiation are probably lower than with radiation alone. Finally, it should be noted that isolated local relapse is relatively uncommon in ESFT, with most patients having distant metastases at relapse.
Chemotherapy Alkylating agents were among the first drugs used in Ewing’s sarcoma, and cyclophosphamide was the standard agent for many years. Early combination therapy comprised vincristine, actinomycin D and cyclophosphamide (VAC). A randomized study carried out by the American co-operative group (IESS) demonstrated that the addition of doxorubicin to this combination led to a significant survival advantage, superior to that provided by pulmonary irradiation. More recently, ifosfamide has been shown to be a highly active drug, and many current regimens use ifosfamide as the alkylating agent of choice. Combination chemotherapy regimens are based on combinations of vincristine, actinomycin and cyclophosphamide, or vincristine, doxorubicin and cyclophosphamide. More recently, the combinations have substituted ifosfamide for cyclophosphamide and there has been increased interest in the use of etoposide as part of more intensive programmes. A study between the UK and the CESS studies from West Germany compared predominantly ifosfamide-based chemotherapy with chemotherapy based on ifosfamide and cyclophosphamide in low-risk Ewing’s sarcomas. In another group of patients, the ifosfamide-based chemotherapy was compared with the same chemotherapy with the addition of etoposide. Follow-up in this study is still short but there appears to be no clear advantage in favour of etoposide in this schedule. Overall survival is similar to earlier studies and there remains considerable room for improvement, particularly in patients presenting with adverse features.34 Prognosis in this disease can be related to several features. The most important of these is the presence of metastatic disease. It is evident that a proportion of patients with metastatic disease confined to the lungs will be cured by conventional chemotherapy, but there are virtually no survivors if the bone marrow or other bones are involved. Tumour volume can also be used to stratify patients, and those with tumours larger than 100–200 mL fare less well. Finally, in patients who undergo surgery, the response to
Ewing’s sarcoma family of tumours 919
preoperative chemotherapy appears to be an important factor, as in osteosarcoma. There has been considerable interest in dose-escalated therapy using peripheral stem-cell rescue for those with an adverse prognosis. Although studies so far are small and uncontrolled, the evidence is such that the value of this approach is now being addressed in a large randomized study.
Surgery The surgery of Ewing’s sarcoma is much more demanding than that of osteosarcoma because the tumour is almost always stage IIb rather than IIa (using the modified staging notation described for osteosarcoma). The margins of resection are much more difficult to define. Surgery is reserved for those cases in whom cure is a probability. In the limbs this will often involve the resection of the greater part of the long bone. Reconstruction of the limb is performed in the same way as for osteosarcoma. In the pelvis and axial skeleton, these lesions can only rarely be excised with confidence about the resection margins, and are sometimes too extensive for reasonable functional reconstruction. However, it is in just these sites that radiation has proved less efficient in providing local control of a very large lesion. Increasingly, combined surgical and radiation approaches are being used for large pelvic Ewing’s tumours. While it is sometimes possible to avoid postoperative radiotherapy by careful planning, the tumour is highly permeative and presents considerable technical problems. The aim, where possible, is to avoid the combined effects of surgery and radiation.
Radiotherapy The aim of local treatment is to achieve control while preserving function. Reported local control produced by radiotherapy alone ranged from 50 to 75 per cent. Multi-agent chemotherapy, in addition to improving the survival rate, contributed to an improvement in local control rate which was achieved in approximately 85–90 per cent,35 with single institution studies generally reporting better results than multi-centre groups. The observation of lower survival in patients treated with radiotherapy rather than surgery, the finding of unexpectedly high local recurrence rates in autopsy series, and the risk of second malignancy has led to the increasing use of surgery. The majority of studies show that improved local control is associated with the use of surgery and this is generally the local modality of choice, except where it would produce significant morbidity.36–38 INDICATIONS FOR RADIOTHERAPY
Radiotherapy is used as sole local therapy in inoperable tumours, and as adjuvant to surgery where there is risk of
residual microscopic disease. Preoperative radiotherapy is associated with an increased complete resection rate and manageable postoperative toxicity. This strategy may be less valuable with greater intensification of chemotherapy before surgery. Radiotherapy has been added when margins of excision are narrow and/or when response to chemotherapy has been poor. Recent evidence has suggested an improvement with radiotherapy even when wide margins were achieved if chemotherapy response has been poor.39 Clear evidence on what constitutes a good margin, however, is lacking. Consideration of the nature of the margin as well as its size is important. One millimetre of dense fascia can be considered an effective margin whilst 3 mm of loose soft tissue may not, especially for a diffusely infiltrative lesion in soft tissue, as these tumours may be. As a guide, margins of 3–5 mm may be considered adequate. TIMING OF RADIOTHERAPY
Neoadjuvant chemotherapy is universal in Ewing’s protocols. It increases resectability, increases the proportion with clear margins and reduces the need for radiotherapy. Local therapy is given after four to six cycles of chemotherapy. Radio-therapy given as sole local therapy or as adjuvant after surgery is administered concurrently with chemotherapy. In order to limit toxicity, therefore, the chemotherapy schedule may need to be modified. There is concern about delaying definitive radiotherapy until the end of a prolonged course of chemotherapy. Timing of adjuvant radiotherapy, however, does not appear to affect outcome.40 TARGET DEFINITION
Data on target definition are lacking. Where detail of local failure is available it is mostly in-field rather than edge recurrence. However, protocol violations are associated with lower control rates.37 It is not necessary to treat the full length of the bone, a 5 cm margin having been shown to be sufficient. In modern protocols, margins of at least 2–3 cm along relevant tissue planes are generally recommended. The clinical target volume includes all tissues initially involved by tumour, within the bone and in soft tissues. There is no necessity to cover areas where tumour covered by intact pleura or peritoneum bulged into a body cavity and has regressed back after chemotherapy. DOSE
There are no good data on a dose response for radiotherapy alone above 55 Gy, but where doses have varied, higher doses have been used for bulkier tumours and numbers in each dose group are small. For inoperable tumours 55 Gy is recommended, 45–55 Gy as adjuvant.41 The precise dose depends on the nature of the margin and the degree of chemotherapy-induced necrosis.
920 Bone
FRACTIONATION
Conventional fractionation (1.8–2 Gy per fraction) is usual. Hyperfractionated regimens have been used. These have not been shown to be associated with better control or survival.36 but the use of lower doses per fraction may facilitate integration with chemotherapy and reduce long-term toxicity, including pathological fracture.
SPECIAL SITES
Whole lung radiotherapy This has been associated with reduced development of pulmonary relapses but was less effective than the addition of doxorubicin to a three-drug regimen so was largely abandoned. In recent European protocols it has been associated with reduced relapse rates, but has not been tested in a randomized trial.29 Chest wall Hemithorax irradiation has been advocated for chest wall tumours with pleural effusion, pleural infiltration and intraoperative contamination and may improve event-free survival.42 Doses of 14–20 Gy are used depending on patient age. Pelvis This is the site where radiotherapy is most commonly used alone. These tumours are most often associated with poor prognostic signs such as tumour bulk and metastases at diagnosis. The overall survival of those treated with radiotherapy has been worse than those treated surgically in most studies, although the local recurrence rate has not always been materially different. This outcome has altered little over the last decade. This suggests that where prognosis is poor any impact produced by small differences in local control may not be relevant or detectable. Even with both local modalities used together systematically, overall survival remains at 45 per cent, the same as a group treated predominantly with radiotherapy, although local relapses are reduced from 25 per cent to 5 per cent. Surgical revision is high, although quality of life is good.43
Results and prognosis Approximately 60 per cent of all patients with ESFT will be cured of their disease. However, this overall figure is a simplification, since patients with small tumours have a better prognosis than those with very large lesions.44 The outlook for ESFT of the small bones of the hands, feet and jaw is already excellent. On the other hand, the prognosis in the pelvis, treated with combination chemotherapy and radiation, has remained unsatisfactory. At this site, with these very large tumours, local control with surgery or radiation is difficult to achieve.
MALIGNANT FIBROUS HISTIOCYTOMA OF BONE Malignant fibrous histiocytoma (MFH) is a distinct clinicopathological entity which was first described by Feldman and Norman.45 It accounts for approximately 5 per cent of primary malignant bone tumours. The cell of origin is unclear. The peak incidence is in middle age, although it tends to occur slightly later in women. In approximately 20 per cent of patients MFH arises in an area where the bone has been previously abnormal, such as fibrous dysplasia, bone infarction, or Paget’s disease or radiation. The primary site is usually the femur or tibia, although the humerus and pelvis are sometimes affected. Until the past decade, treatment for MFH of bone has been with surgery, sometimes combined with local radiation. However, it is now clear that only 30 per cent of patients will survive 5 years if local treatment is given alone. More recently, adjuvant chemotherapy has been used, either preoperatively, as in osteosarcoma, or after the primary surgical excision. No randomized trials of chemotherapy have been conducted, but it is clear that the tumour is chemosensitive, as judged by both clinical response and by histopathological evidence of tumour necrosis in the resection specimen. The chemotherapeutic agents which should be used have not been defined systematically but, in general, treatment programmes have followed those for osteosarcoma .46 Cisplatin, doxorubicin and ifosfamide have been the agents used in these studies. However, large-scale studies, where patients are followed over a long period of time, are necessary to define the current cure rate in this condition.
PRIMARY NON-HODGKIN’S LYMPHOMA OF BONE (PLB) Primary non-Hodgkin’s lymphoma accounts for about 4 per cent of primary bone tumours. Symptoms at presentation are frequently similar to those of other primary bone tumours. The single most common type is diffuse B-cell lymphoma. Staging investigations fail to reveal generalized spread of the disease in the marrow or distant lymph node sites. The long bones of the lower limb are the most common site of PLB, but it also occurs in flat bones, the spine and pelvis. Radiotherapy produces good in-field control but distant relapse occurs in up to 50 per cent of patients. No randomized trial has addressed the optimal treatment for PLB, but recent series using chemotherapy and radiotherapy report 5-year disease free survival rate of 80 per cent or more.47 In the paediatric age group, good long-term survival has been reported using chemotherapy alone.48 but this has not been reported in the adult practice where radiotherapy remains standard practice. In weight-bearing bones, especially with extensive destruction, orthopaedic assessment of stability is required and endoprosthetic replacement may be required if healing is limited.
Key references 921
Primary non-Hodgkin’s lymphoma of bone has been considered an indication for central nervous system prophylaxis (as is the case in some protocols for marrow infiltration). This has not, however, been usual in the reported series and no evidence supports its use in stage I disease. The suitability of the strategy used in stage I–II nodal disease of three cycles of CHOP plus involved field radiotherapy has not been reported in PLB where tumours are often bulky. Standard treatment at present, therefore, in adults with stage I–IIE diffuse B-cell lymphoma of bone is six cycles of combination chemotherapy followed by involved field radiotherapy to 40 Gy. Patients presenting with multi-focal primary bone lymphoma are treated with intensive combination chemotherapy as in advanced non-Hodgkin’s lymphoma.
CHONDROSARCOMA
Prognosis The prognosis of chondrosarcoma depends on its grade and site.49 The resection margin depends entirely on the size and site of the lesion and the adequacy of preoperative planning. With meticulous planning of approach and planes of dissection, even very large tumours in the pelvis may be removed with minimal risk of local contamination. Curettage of such lesions is never curative and prevents secondary surgery from ever being successful. In the elderly and debilitated, intra-lesional surgery may be contemplated for palliation where the physical cost of curative surgery is too great. In the young, however, where cure is essential, ablative surgery with reasonable margins must be undertaken to prevent a distressing prolonged illness consisting of progressive locally recurrent disease.
Pathology Chondrosarcomas are malignant tumours of cartilage. Histological grading based on cellularity and grading is reflected by clinical behaviour. Frank dedifferentiation to a high-grade tumour is also recognized and such tumours have a high metastatic potential and a poor overall survival. The tumour is rare before the third decade and more common in the fourth and fifth decades. It has a predilection for the girdles and proximal long bones, but may arise at any site. The tumours are usually slow growing and, particularly in the pelvis, may reach a great size before detection. The principal presentation is with a painful lump.
Investigation The radiographic features are usually diagnostic. However, thorough preoperative assessment is essential. A CT and MR scan of the lesion should be performed and subsequent needle biopsy of the most malignant-looking area should be undertaken. This will prevent a high-grade tumour which has developed in an osteochondroma from being inadequately treated.
KEY LEARNING POINTS ●
●
●
●
●
●
Surgical management Chondrosarcoma does not respond to chemotherapy or radiotherapy. These tumours should be resected after meticulous imaging to determine the levels and planes of resection. Wide local excision should be performed. Like other musculoskeletal sarcomas, these tumours invade neither peripheral nerves nor dura and so they may be resected without causing serious mutilation, even when awkwardly sited. However, vascular grafts might be necessary. Reconstruction of the limb segments that have been removed is as for osteosarcoma. In dedifferentiated chondrosarcoma chemotherapy (using cisplatin/doxorubicin/ifosfamide combinations) may produce responses.
Primary tumours of bone are uncommon but share similar clinical features of pain and swelling. These are often ignored for a considerable time. Initial evaluation should be conducted in specialist centres. Appropriate imaging should be carried out before planned biopsy. Diagnosis can usually be made by core needle biopsy. Placement of the biopsy track should be determined after consideration of future surgery. Histological interpretation requires special expertise and should take account of clinical and radiological features. Many primary bone tumours, particularly osteosarcoma and Ewing’s sarcoma, may be cured by appropriate multi-modality therapy. Such treatment is intensive and complex. It requires close communication between surgical and nonsurgical oncology teams. Further advances in treatment are likely to come through a growing understanding of the unusual biological features of these diseases.
KEY REFERENCES Campanacci M. Bone and Soft Tissue Tumours. Vienna: Springer Verlag, 1986. Mirra JM, Picci P, Gold RH. Bone Tumours. Clinical, Radiologic and Pathologic Correlations. Philadelphia: Lea and Febiger, 1989. Whelan JS. Osteosarcoma. Eur J Cancer 1997; 33:1611–9.
922 Bone
REFERENCES 1 Polednak AP. Bone cancer among female radium dial workers. Latency periods and incidence rates by time of exposure. J Natl Cancer Inst 1978; 60:77–82. 2 Draper GJ, Sanders BM, Kingston JE. Second primary neoplasms in patients with retinoblastoma. Br J Cancer 1986; 53:661–71. 3 Li FP. Molecular epidemiology studies of cancer in families. Br J Cancer 1993; 68(2):217–9. 4 Porter DE, Holden ST, Steel CM, Cohen BB, Wallace MR, Reid R. A significant proportion of patients with osteosarcoma may belong to Li-Fraumeni cancer families. J Bone Joint Surg Br 1992; 74(6):883–6. 5 Price CHG, Goldie W. Paget’s sarcoma of bone. A study of 80 cases from the Bristol and Leeds Bone Tumour Registries. J Bone Joint Surg Br 1969; 51B:205–24. 6 Mankin HJ, Mankin CJ, Simon MA. The hazards of the biopsy, revisited. Members of the Musculoskeletal Tumor Society. J Bone Joint Surg Am 1996; 78(5):656–63. 7 Hau A, Kim I, Kattapuram S, et al. Accuracy of CT-guided biopsies in 359 patients with musculoskeletal lesions. Skeletal Radiol 2002; 31(6):349–53. 8 Stoker DJ, Cobb JP, Pringle JA. Needle biopsy of musculoskeletal lesions. A review of 208 procedures. J Bone Joint Surg Br 1991; 73(3):498–500. 9 Enneking WF, Spanier SS, Goodman MA. A system for the surgical staging of musculoskeletal sarcoma. Clin Orthop Relat Res 1980; 153:106–20. 10 Grimer RJ, Carter SR, Pynsent PB. The cost-effectiveness of limb salvage for bone tumours. J Bone Joint Surg Br 1997; 79(4):558–61. 11 Postma A, Kingma A, De Ruiter JH, et al. Quality of life in bone tumor patients comparing limb salvage and amputation of the lower extremity. J Surg Oncol 1992; 51(1):47–51. 12 Dunst J, Sauer R, Burgers JM, et al. Radiation therapy as local treatment in Ewing’s sarcoma. Results of the Cooperative Ewing’s Sarcoma Studies CESS 81 and CESS 86. Cancer 1991; 67(11):2818–25. 13 Hudson FR, Crawley MT, Samarasekera M. Radiotherapy treatment planning for patients fitted with prostheses. Br J Radiol 1984; 57(679):603–8. 14 Okoye BO, Spooner D, Townley JF, Gornall P. Radioprotective reverse orchidopexy. J Pediatr Surg 2002; 37(2):236–9. 15 Visvanathan DK, Cutner AS. A new technique of laparoscopic ovariopexy before irradiation. Fertil Steril 2003; 79(5):1204–6. 16 Jentzsch K, Binder H, Cramer H, et al. Leg function after radiotherapy for Ewing’s sarcoma. Cancer 1981; 47(6):1267–78. 17 Tucker MA, D’Angio GJ, Boice JD, et al. Bone sarcomas linked to radiotherapy and chemotherapy in children. N Engl J Med 1987; 317:588–93. 18 Hawkins MM, Wilson LM, Burton HS, et al. Radiotherapy, alkylating agents, and risk of bone cancer after childhood cancer. J Natl Cancer Inst 1996; 88(5):270–8.
19 Rosen G, Caparros B, Huvos AG, et al. Preoperative chemotherapy for osteogenic sarcoma: selection of postoperative adjuvant chemotherapy based on the response of the primary tumor to preoperative chemotherapy. Cancer 1982; 49(6):1221–30. 20 Link MP, Goorin AM, Miser MD, et al. The effect of adjuvant chemotherapy on relapse-free survival in patients with osteosarcoma of the extremity. N Engl J Med 1986; 314:1600–6. 21 Eilber F, Giuliano A, Eckardt J, Patterson K, Moseley S, Goodnight J. Adjuvant chemotherapy for osteosarcoma: a randomised prospective trial. J Clin Oncol 1987; 5:21–6. 22 Goorin AM, Schwartzentruber DJ, Devidas M, et al. Presurgical chemotherapy compared with immediate surgery and adjuvant chemotherapy for nonmetastatic osteosarcoma: Pediatric Oncology Group Study POG-8651. J Clin Oncol 2003; 21(8):1574–80. 23 Meyers PA, Schwartz CL, Krailo M, et al. Osteosarcoma: a randomized, prospective trial of the addition of ifosfamide and/or muramyl tripeptide to cisplatin, doxorubicin, and high-dose methotrexate. J Clin Oncol 2005; 23(9):2004–11. 24 Bramwell VH, Burgers M, Sneath R, et al. A comparison of two short intensive adjuvant chemotherapy regimens in operable osteosarcoma of limbs in children and young adults: the first study of the European Osteosarcoma Intergroup. J Clin Oncol 1992; 10(10):1579–91. 25 Souhami RL, Craft AW, Van der Eijken JW, et al. Randomised trial of two regimens of chemotherapy in operable osteosarcoma: a study of the European Osteosarcoma Intergroup. Lancet 1997; 350(9082):911–7. 26 Lewis IJ, Nooij M. Chemotherapy at standard or increased dose intensity in patients with operable osteosarcoma of the extremity; a randomised controlled trial conducted by the European Osteosarcoma Intergroup (ISRCTN 86294690). Proc ASCO 2003; 22:3281. 27 www.euramos.org. 28 Cade S. Osteogenic sarcoma: a study based on 133 patients. J R Coll Surg Edin 1955; 1:79–111. 29 Whelan JS, Burcombe RJ, Janinis J, Baldelli AM, Cassoni AM. A systematic review of the role of pulmonary irradiation in the management of primary bone tumours. Ann Oncol 2002; 13(1):23–30. 30 Ward WG, Mikaelian K, Dorey F, et al. Pulmonary metastases of stage IIB extremity osteosarcoma and subsequent pulmonary metastases. J Clin Oncol 1994; 12(9):1849–58. 31 Sheth DS, Yasko AW, Raymond AK, et al. Conventional and dedifferentiated parosteal osteosarcoma. Diagnosis, treatment, and outcome. Cancer 1996; 78(10):2136–45. 32 Rose PS, Dickey ID, Wenger DE, Unni KK, Sim FH. Periosteal osteosarcoma: long-term outcome and risk of late recurrence. Clin Orthop Relat Res 2006; 453:314–7. 33 Grimer RJ, Bielack S, Flege S, et al. Periosteal osteosarcoma – a European review of outcome. Eur J Cancer 2005; 41(18):2806–11. 34 Paulussen M, Craft A, Lewis I, et al. Ewing tumor of bone – updated report of the European Intergroup Cooperative Ewing’s Sarcoma Study EICESS 92. Proc ASCO 2002:1568.
References 923
35 Vietti TJ, Cangir A, Tefft M, et al. Multimodal therapy for the management of primary, nonmetastatic Ewing’s sarcoma of bone: a long term follow-up of the first Intergroup study. J Clin Oncol 1990; 8:1664–74. 36 Dunst J, Jurgens H, Sauer R, et al. Radiation therapy in Ewing’s sarcoma: an update of the CESS 86 trial. Int J Radiat Oncol Biol Phys 1995; 32(4):919–30. 37 Donaldson SS, Torrey M, Link MP, et al. A multidisciplinary study investigating radiotherapy in Ewing’s sarcoma: end results of POG #8346. Pediatric Oncology Group. Int J Radiat Oncol Biol Phys 1998; 42(1):125–35. 38 Carrie C, Mascard E, Gomez F, et al. Nonmetastatic pelvic Ewing sarcoma: report of the French Society of Pediatric Oncology. Med Pediatr Oncol 1999; 33(5):444–9. 39 Schuck A, Ahrens S, Paulussen M, et al. Local therapy in localized Ewing tumors: results of 1058 patients treated in the CESS 81, CESS 86, and EICESS 92 trials. Int J Radiat Oncol Biol Phys 2003; 55(1):168–77. 40 Schuck A, Rube C, Konemann S, et al. Postoperative radiotherapy in the treatment of Ewing tumors: influence of the interval between surgery and radiotherapy. Strahlenther Onkol 2002; 178(1):25–31. 41 Donaldson SS. Ewing sarcoma: radiation dose and target volume. Pediatr Blood Cancer 2004; 42(5):471–6. 42 Schuck A, Ahrens S, Konarzewska A, et al. Hemithorax irradiation for Ewing tumors of the chest wall. Int J Radiat Oncol Biol Phys 2002; 54(3):830–8.
43 Rodl RW, Hoffmann C, Gosheger G, Leidinger B, Jurgens H, Winkelmann W. Ewing’s sarcoma of the pelvis: combined surgery and radiotherapy treatment. J Surg Oncol 2003; 83(3):154–60. 44 Cotterill SJ, Ahrens S, Paulussen M, et al. Prognostic factors in Ewing’s tumor of bone: analysis of 975 patients from the European Intergroup Cooperative Ewing’s Sarcoma Study Group. J Clin Oncol 2000; 18(17):3108–14. 45 Feldman F, Norman D. Intra- and extraosseous malignant histiocytoma (malignant fibrous xanthoma). Radiology 1972; 104(3):497–508. 46 Bramwell VH, Steward WP, Nooij M, et al. Neoadjuvant chemotherapy with doxorubicin and cisplatin in malignant fibrous histiocytoma of bone: A European Osteosarcoma Intergroup study. J Clin Oncol 1999; 17(10):3260–9. 47 Barbieri E, Cammelli S, Mauro F, et al. Primary non-Hodgkin’s lymphoma of the bone: treatment and analysis of prognostic factors for Stage I and Stage II. Int J Radiat Oncol Biol Phys 2004; 59(3):760–4. 48 Suryanarayan K, Shuster JJ, Donaldson SS, Hutchison RE, Murphy SB, Link MP. Treatment of localized primary nonHodgkin’s lymphoma of bone in children: a Pediatric Oncology Group study. J Clin Oncol 1999; 17(2):456–9. 49 Lee FY, Mankin HJ, Fondren G, et al. Chondrosarcoma of bone: an assessment of outcome. J Bone Joint Surg Am 1999; 81(3):326–38.
40 Soft tissue sarcomas THOMAS F. DELANEY, ANDREW E. ROSENBERG, DAVID C. HARMON, FRANCES J. HORNICEK, SAM YOON, DAVID G. KIRSCH, HENRY J. MANKIN AND DANIEL ROSENTHAL
Introduction Aetiology Clinical evaluation Pathology Staging Management of the extremity, trunk, or head and neck primary tumour Adjuvant chemotherapy Functional outcome Treatment of local recurrence
924 926 929 933 937 938 953 956 956
INTRODUCTION Sarcomas are malignant tumours that arise from skeletal and extra-skeletal connective tissues, including the peripheral nervous system. The term ‘sarcomas of soft tissues’ embraces all of the malignant tumours which arise from the mesenchymal tissues excluding bone, i.e. malignant fibrous histiocytoma, liposarcoma, leiomyosarcoma, synovial sarcoma, rhabdomyosarcoma, epithelioid sarcoma, angiosarcoma, fibrosarcoma, etc. In addition, malignant tumours of peripheral nerve sheaths are included despite being ectodermal in origin, as their clinical behaviour is not measurably different from the other sarcomas. Gastrointestinal stromal tumours are derived from cells (interstitial cells of Cajal) with neural and smooth muscle features and are also considered soft tissue sarcomas. The relative frequency of different histological subtypes of sarcomas is shown in Table 40.1. This chapter discusses current practice in the management of patients with soft tissue sarcoma (other than rhabdomyosarcoma of the paediatric age group) as well as selected benign, infiltrative soft tissue tumours such as desmoids whose management is similar. Soft tissue sarcomas are rare with an estimated incidence in the USA of approximately 8680 diagnosed annually, representing less than one per cent of all newly diagnosed malignant tumours.1 Approximately 37 per cent of these
Radiotherapy alone Treatment of unresectable or locally advanced soft tissue sarcoma Management of the retroperitoneal primary lesion Management of gastrointestinal stromal tumours Treatment of metastatic disease Desmoid tumours References
956 957 957 959 959 961 966
are expected to die of this disease. Review of the statistics of recent years suggests an increase in the incidence of soft tissue sarcomas, although it is not clear whether this represents a true increase or merely reflects more accurate diagnosis Table 40.1 Frequency of histopathological types of soft tissue sarcoma in large series of patients with extremity and trunk soft tissue sarcomas MGH Milan 1994190 2005207
Lund MSKCC 2004441 1996204
Number of patients
738
911
298
1041
MFH Liposarcoma Fibrosarcoma Leiomyosarcoma Sarcoma, NOS Synovial sarcoma MPNST Rhabdomyosarcoma Vascular sarcomas Other
22% 16% 11% 10% 9% 8% 10% 3% NR 11%
8% 31% NR 14% NR 15% 10% NR 5% 16%
13% 14% 13% 33% 9% 6% 7% NR 1% 4%
25% 29% 10% 8% NR 12% 5% NR NR 12%
MFH, malignant fibrous histiocytoma; MPNST, malignant peripheral nerve sheath tumour; NR, not reported; MGH, Massachusetts General Hospital; MSKCC, Memorial Sloan-Kettering Cancer Center.
Introduction 925
and increasing interest in these tumours.2 Although the malignant tumours of soft tissue are rare, benign tumours are common. It is estimated that the frequency of benign tumours is 100 times that for the malignant lesions.2 To appreciate the rarity of the sarcomas, note that during the same period there are expected to be 211 240 newly diagnosed cases of breast cancer in women in the USA as compared with only 3890 soft tissue sarcomas in women. Thus, women are approximately 54 times more likely to develop a carcinoma of the breast than a sarcoma. The sites of appearance of soft tissue sarcoma (STS) in order of frequency (Table 40.2) are: lower extremity (46 per cent), torso (19 per cent), upper extremity (14 per cent), retroperitoneum (13 per cent), head/neck (8 per cent).3–6 The small number of cases seen and the great diversity in anatomical site, histopathology and biology complicate understanding of the natural history of these tumours and their response to diverse therapies. A large proportion of the sarcoma patients are referred to major centres with subspecialist teams because of the following factors: ● ● ● ●
the relative rarity; appearance at all body sites; occurrence at all ages; and broad spectrum of histological types.
This has resulted in a rather greater clinical and laboratory research activity than might be expected for tumours of this low frequency. This is reflected in the large number of papers (more than 2400 in 2005) being published each year in the general area of soft tissue sarcoma. Although these lesions are uncommon, there has been a considerable increase in general medical interest over recent years in their management. This is due to rapidly changing management strategies and improvements in clinical results. In particular, solid evidence has accumulated that, in addition to surgery, there are important roles for radiation therapy and chemotherapy in the management of some groups of these patients.
Table 40.2 Distribution (%) of sarcoma of soft tissue according to anatomical site in three US surveys
Lower extremity Upper extremity Head and neck Torso Retroperitoneum Other
MGHa,443 (n 788)
Russell166 (n 1215)
Lawrence6 (n 4550)
44 21 12 10 6 7
40 13 15 18 13 1
46 13 8 18 12 1
a All patients seen 1971–1993 and treated by radiation and surgery for stage M0 disease. MGH, Massachusetts General Hospital.
When treating soft tissue sarcomas, the major therapeutic goals are survival, local tumour control, optimal function and minimal morbidity. Surgical resection of the primary tumour is an essential component of treatment for virtually all patients, with the exception of a small percentage of selected patients with radiosensitive histologies such as extra-osseous Ewing’s sarcoma or presentations in anatomical locations where resection would have unacceptable functional or cosmetic consequences. However, local control by surgery alone can be poor in selected anatomical locations if the surgical intent is also to preserve function and good cosmesis. For example, local control is suboptimal in the majority of patients with large extremity lesions unless the procedure removes large volumes of grossly normal tissue, i.e. widely negative margins are necessary because sarcomas tend to infiltrate normal tissue adjacent to the evident lesion. Consequently, removal of the gross lesion by a simple excision alone (only a narrow margin) is followed by local recurrence in 60–90 per cent of patients. Radical resections are associated with reduction in the local recurrence rate to 10–30 per cent, but may compromise limb function. Local control is also poor for retroperitoneal sarcomas when managed with surgery alone, where local recurrence rates are reported to be as high as 70 per cent, because of the difficulty in obtaining widely negative margins. Thus, the combination of surgery and radiation achieves better outcomes than either treatment alone for most patients with soft tissue sarcomas. Chemotherapy appears to be import-ant for selected patients with large, high-grade sarcomas and specific histologies such as extraosseous Ewing’s sarcoma, rhabdomyosarcoma and gastrointestinal stromal tumours (GISTs). The excellent results currently being obtained by more conservative treatment strategies are through a multidisciplinary approach to the overall management of these patients (diagnostic evaluation, biopsy, treatment, rehabilitation and follow-up). The multidisciplinary team includes not only the surgeon but also the radiation, medical and paediatric oncologists, pathologists and diagnostic radiologists. The pathologist is an extremely valuable member of the team because the staging system now employed (for other than childhood rhabdomyosarcoma) utilizes histopathological grade as the principal determinant of stage. The pathological diagnosis is now based on much more objective criteria with the introduction of immunohistochemistry and, to a lesser extent, cytogenetics. Assessment of the local extent of the sarcoma and the search for evidence of distant metastatic tumour is conducted with the diagnostic radiologist. Modern imaging techniques, computed tomography (CT) and magnetic resonance imaging (MRI), have made enormous improvements in the ability to define the size, precise anatomical site, local invasion and the pattern of spread of the sarcoma. Positron emission tomography (PET) scanning and more recently PET/CT are now being evaluated in several centres with expected gains in the assessment of the magnitude of response to treatment.
926 Soft tissue sarcomas
The strategy for the control of the primary/regional lesion is, in most patients, combined modality and this is achieving greatly enhanced functional and cosmetic outcomes. These gains have stemmed largely from the integration of radiation with surgery. Despite positive findings in some phase III trials, the efficacy of chemotherapy to prolong survival has not been as clearly and uniformly shown with the drug protocols tested to date. A meta-analysis of the phase III trials suggests an improvement in disease-free survival and a small but not statistically significant survival gain with the use of adjuvant chemotherapy in resectable, high-grade soft tissue sarcoma in adults.7*** A randomized trial not completed for that meta-analysis has since shown an overall survival advantage at 4 years.8** There are several important developments being evaluated in the laboratory and the clinic, which offer the potential of further gains.
AETIOLOGY Genetic Involvement of genetic factors in the genesis of soft tissue sarcoma is manifest by the strong hereditary tendency for certain sarcomas.9–12 Gardner’s syndrome is a hereditary disease, one feature of which is a liability to desmoid tumours (also termed fibromatosis).13 Neurofibromatosis I also features tumours of the soft tissues: neurofibromas and malignant peripheral nerve sheath tumours (MPNST). A significant proportion of these patients ultimately exhibit transformation of the neurofibroma into MPNST.9,14*,15 This disorder is associated with a mutation in the NF1 gene. It has been proposed that malignant degeneration reflects the two-hit hypothesis in which one allele is constitutionally inactivated in the germ-line while the other allele undergoes somatic inactivation (the second hit).16 Sarcomas of soft tissue and bone, particularly osteosarcoma, have been observed later in life in surviving patients with familial or bilateral retinoblastoma.17*,18* In one study of 1604 patients with retinoblastoma, the cumulative incidence of a second cancer at 50 years after diagnosis was 51 per cent for hereditary retinoblastoma (which is associated with mutations in the retinoblastoma tumour suppressor gene, see below), and 5 per cent for non-hereditary (sporadic) retinoblastoma.18* More than 60 per cent of the cancers were a form of sarcoma. Malignant schwannoma may complicate the multiple endocrine neoplasia syndrome.19 Patients with Li–Fraumeni syndrome often develop sarcomas.12,20* The Li–Fraumeni syndrome is inherited as an autosomal recessive trait and is primarily characterized by soft tissue and bone sarcomas and breast cancer; other features include brain tumour, leukaemia and adrenocortical cancer occurring before the age of 45.12,20 Some patients develop multiple malignancies.21 A germ-line mutation in the p53 tumour suppressor gene is found in most affected families (see p53 gene below).22–24 In addition, germ-line mutations in hCHK2, a protein kinase that regulates p53 in response to
DNA damage, has also been found in a subset of families with Li–Fraumeni syndrome.25 In one series of 151 children with soft tissue sarcoma, 5 of the families (3.3 per cent) manifested the classic Li–Fraumeni familial cancer syndrome, another 10 (6.6 per cent) had features consistent with the syndrome, and 16 (10.5 per cent) had one parent with a possible hereditary cancer syndrome or with cancer before the age of 60.26* In an analysis of 754 first-degree relatives of 177 children with sarcomas of soft tissue, Birch et al. found an increased incidence of malignant tumours in relatives of patients who were male, 2 years of age, and with an embryonal rhabdomyosarcoma.27*
Cytogenetics In recent years, it has become evident that the genetics of sarcomas segregate into two major types. One type has specific genetic alterations and usually simple karyotypes, including fusion genes due to reciprocal translocations (e.g. PAX3–FKHR in alveolar rhabdomyosarcomas) or specific point mutations (e.g. c-KIT mutations in gastrointestinal stromal tumours). A second type has non-specific genetic alterations and complex, unbalanced karyotypes, reflected by numerous genetic losses and gains (e.g. osteosarcoma, malignant fibrous histiocytoma, liposarcomas other than the myxoid type, angiosarcoma, leiomyosarcoma).28 Sarcomas with recurrent chromosome translocations account for approximately one third of all sarcomas.29 In many cases, the aberrant protein product of the fused gene acts as an abnormal transcriptional regulator, thus providing the molecular basis for oncogenesis. These non-random chromosomal aberrations occur in specific types of sarcomas and are being increasingly utilized in the definitive diagnosis of these sarcomas. Furthermore, these chromosomal abnormalities have been characterized at the molecular level and many of the chimeric genes have been identified, providing clues to the molecular alterations that are fundamental for the development of soft tissue sarcomas. The most well recognized and frequently employed are mentioned here briefly. Most synovial cell sarcomas are characterized by the translocation t(x;18)(p11.2;q11.2).30 The breakpoint of this translocation fuses the SYT gene from chromosome 18 to one of two homologous genes, SSX1 or SSX2 on the X chromosome.31,32 The SYT–SSX gene is thought to function as an aberrant transcriptional regulator. The nature of the chimaeric gene appears to have prognostic and pathogenetic importance, as metastasis-free survival may be higher with SYT–SSX2 compared to SYT–SSX1.33 SYT–SSX1 is associated with biphasic tumours (glandular epithelial differentiation on a background of spindle tumour cells), while SYT–SSX2 is associated with monophasic tumours that lack glandular epithelial differentiation. Other chromosomal changes characteristic of specific sarcoma type include the reciprocal exchange t(11:22)(q24;ql2) seen in approximately 85–90 per cent of Ewing’s sarcoma
Aetiology 927
and PNET.34–36 In this translocation, the EWS gene from chromosome 22q12 is covalently linked to the ETS family member, FLI-1 to form the EWS–FLI-1 fusion gene.37 The chimeric proteins that result from this translocation may alter transcription of an unidentified gene on chromosome 22.37,38 A less common translocation t(21;22)(q22;q12) has also been identified and links EWS to a different ETS family member, ERG.39 Myxoid and round cell subtypes of liposarcomas display a reciprocal translocation t(12;16) (q13;p11).40,41 In this translocation, the CHOP (induced by DNA damage) gene is inserted adjacent to a novel gene called TLS. The fusion gene, called TLS–CHOP shows sequence homology to the Ewing’s fusion gene.40–43 It fails to induce G1/S arrest, which is one of the functions of the nononcogenic form of CHOP (GADD153).44 Identification of the fusion gene has been used as a diagnostic aid for these subtypes of liposarcoma.42 Alveolar rhabdomyosarcomas show a translocation at t(2;13)(q35;q14) or less often t(1;13)(p36;q14); the chimeric genes have been cloned and have been termed PAX3–FKHR and PAX7–FKHR, respectively.45,46 These translocations are associated with over-expression of the fusion product.46 PAX7–FKHR tumours more often present with extremity lesions, are more likely to be localized, and are less likely to metastasize widely than PAX3–FKHR tumours.47 A downstream target of PAX3–FKHR may be MET, which encodes a receptor involved in growth and motility signalling.48 Molecular determination of minimal residual disease in alveolar rhabdomyosarcoma is possible but the clinical significance of this finding is uncertain.49 Clear cell sarcoma is usually classified as a malignant melanoma, although cytogenetically the tumours are distinct. Clear cell sarcomas often exhibit a translocation at t(12;22)(q13–14;q12), which is not seen in malignant melanoma.50 Trisomy of chromosome 8 is also observed in clear cell sarcoma.51 Alveolar soft part sarcoma is another tumour sometimes associated with melanoma that also exhibits a specific chromosomal translocation t(X;17) (p11;q25), which creates the fusion product TFE–ASPL.52 In contrast to the sarcomas that have recurrent, nonrandom chromosome translocations and relatively simple karyotypes, the molecular pathogenesis of sarcomas with non-specific genetic alterations and complex karyotypes has not been definitively proven. However, inactivation of the p53 pathway appears to be a common feature in these tumours, and it has been proposed that p53 pathway inactivation may differentiate these tumours from those with simple genetic alterations.53 Among the mechanisms for p53 pathway inactivation are p53 point mutation and deletion, homozygous deletion in CDKN2A, and MDM2 amplification.28 It should be noted, however, that inactivation of the p53 pathway has been identified in a subset of tumours with specific chromosomal translocations. For example, up to 25 per cent of Ewing’s sarcomas show p53 mutation or loss of CDKN2A.54,55 Importantly, patients with Ewing’s sarcomas with inactivation of the p53 pathway showed significantly worse survival.
Molecular biology The importance of inactivating the p53 pathway to sarcoma development can be understood by exploring the molecular biology of the p53 tumour suppressor gene. Somatic mutations in the p53 gene are the most frequently detected molecular alteration in sporadic soft tissue sarcoma. These mutations have been detected in a variety of soft tissue sarcomas including malignant fibrous histiocytoma (MFH), leiomyosarcoma, liposarcoma and rhabdomyosarcoma.56–67 Germ-line mutations of p53 are also found in most families with the Li–Fraumeni syndrome.68 Germ-line mutations in this gene also may occur in other patients with soft tissue sarcoma, particularly those with other cancers that are not considered indicative of the Li–Fraumeni syndrome.56,57,68 Moreover, in families with Li–Fraumeni syndrome where no p53 mutations can be identified, heterozygous mutations in hCHK2, which encodes a protein kinase that phosphorylates p53, may be identified.25 The p53 protein is a transcriptional activator that plays a key role in the integration of signals inducing cell division, arrest of DNA synthesis following DNA damage, and programmed cell death (apoptosis). DNA damage results in increased levels of p53 protein. p53 functions to increase the expression of a number of genes including: 1) MDM2, which inhibits p53 function; 2) p21, a cyclin-dependent kinase inhibitor that causes cell cycle arrest; and 3) genes that initiate apoptosis.69–71 The wild-type p53 in normal tissue has a short half-life and is not detectable by immunohistochemical methods; in comparison, mutations of the gene result in a stabilized p53 protein that accumulates in the cell and often becomes detectable by immunohistochemistry. Gross rearrangements and non-sense mutations, however, may show no immunostaining for p53, indicating that this technique fails to identify a significant proportion of tumours with p53 alterations.72 Animal models are compatible with p53 defects having a pathogenetic role in sarcoma development. Mice deficient for p53 also develop a variety of neoplasms including bone and soft tissue sarcomas.73 Similarly, irradiated transgenic mice harbouring mutant p53 show higher frequencies of sarcomas.74 On the other hand, transduction of wild-type p53 genes into soft tissue sarcomas bearing mutated p53 genes restores enhanced cell cycle control and suppresses sarcoma growth.75 p53 may also be activated in response to the expression of an activated oncogene. For example, expression of the EWS–FLI-1 fusion gene in human fibroblasts causes activation of p53 and growth arrest.76 Activation of p53 by EWS–FLI-1 appears to occur by up-regulation of p14Arf encoded by the CDKN2A gene.77 p14Arf in turn inhibits MDM2. Since the activation of p53 is usually held in check by MDM2 through a negative feedback loop, up-regulation of p14Arf removes this check, allowing p53 activation to proceed and growth arrest to ensue. Therefore, in some settings sarcoma formation may be blocked by p53 activation and sarcoma development may only occur when cooperating mutations occur in p53, CDKN2A that expresses p14Arf, or
928 Soft tissue sarcomas
by MDM2 over-expression. Indeed, the MDM2 gene, located at 12q134, is over-expressed in a variety of human tumours including soft tissue sarcomas.78–81 In addition to p53 pathway inactivation, mutation of the retinoblastoma gene Rb frequently occurs in soft tissue sarcomas. In fact, the first demonstration of a specific gene abnormality associated with malignant transformation in man was the loss of the Rb gene, a tumour suppressor gene, in retinoblastoma.82,83 Even before the Rb gene had been identified, it was recognized that some sporadic sarcomas had deletions on chromosome 13 similar to those observed in some patients with retinoblastoma. Deletions or mutations of the tumour suppressor retinoblastoma (Rb) gene are critical in the pathogenesis of retinoblastoma and a variety of solid tumours. Alterations in the Rb gene are common in soft tissue sarcoma, occurring in up to 70 per cent of tumours.84–86 It has been proposed that Rb alterations are primary events in human sarcomas and may be involved in tumorigenesis or the early phases of tumour progression.85 The Rb gene is critical for proper entry and transition through the cell cycle because the protein encoded by Rb (pRb) controls the expression of other genes necessary for G1–S cell cycle progression. Normally, this cell cycle progression occurs when cyclin-dependent kinase 4 (cdk4) phosphorylates and inhibits pRb function. As mentioned above, the p53 regulated cyclin-dependent kinase inhibitor, p21, can inhibit cdk4 and prevent cell cycle progression by preventing phosphorylation of pRb. However, when Rb is mutated, the normal cell cycle is perturbed. This change has been shown to be associated with infrequent osteogenic or other sarcomas as second malignant neoplasms in patients with hereditary retinoblastoma.83 Further, the absence of detectable Rb protein is associated with a small proportion of apparently sporadic sarcomas.87 In an examination of 43 sarcomas of bone and soft tissues, Wunder et al. found alterations in the Rb gene in approximately 40 per cent of the tumours.86 Other genetic changes have also been identified in sarcomas. Duda et al. reported that approximately 40 per cent of sarcomas were found to have an increased expression of c-erbB-2 oncogene and EGFr.88 Additional gene mutations or other alterations in sarcomas include: CDK4,89 ras,90, 91 myc,92 and SAS.93,94 Reports regularly appear with accounts of other genetic abnormalities in sarcomas: for a recent review, see that by Helman and Meltzer.28
Gene expression profiling Gene expression profiling by means of DNA microarrays is the newest approach to cancer classification95 and is now being applied to soft tissue sarcomas. Gene chip microarrays are able to give a global snapshot of gene expression in tumours. This technology has been increasingly used in sarcomas to differentiate between histological subtypes,96 better classify equivocal histological subtypes,97 and determine prognosis and response to chemotherapy.98 Interestingly, in
one analysis of global gene expression patterns for clear cell sarcomas compared to other sarcomas and malignant melanomas, clear cell sarcomas clustered with malignant melanomas and not with other sarcomas.99 In one report, transcriptional profiling for 5520 different known genes was used to separate 41 soft tissue sarcomas into five distinct groups:96* ●
●
● ●
●
Gastrointestinal stromal tumours (GIST) highly expressed a cluster of 125 genes, which separated them from other sarcomas. Synovial sarcomas expressed a specific cluster of 104 genes, among them molecules involved in retinoic acid pathways, and the epidermal growth factor receptor. Neural tumours (e.g. peripheral nerve sheath tumours). One half of all leiomyosarcomas, in which 24 specific genes were highly expressed. A broad group, containing all of the liposarcomas, the malignant fibrous histiocytomas, and the remainder of the leiomyosarcomas, in which molecular profiles were not predicted by histological features or immunohistochemistry.
The clinical utility of such a classification system has been realized only for the GISTs, in which tyrosine kinaseactivating c-KIT mutations predict responsiveness to targeted therapy. However, the potential for future benefit of improved classification of soft tissue tumours includes better prediction of natural history, and patient-tailored therapy based upon the specific targets identified by microarray analysis.96
Environmental factors Radiation is recognized as capable of inducing sarcoma of bone and soft tissue. The frequency increases with radiation dose and with the post-radiation observation period.100–102 The most frequent histopathological type of radiationinduced sarcoma arising in soft tissues is malignant fibrous histiocytoma (approximately 70 per cent).103 Taghian et al. studied 7620 patients treated for carcinoma of the breast; the rate of induced tumours at 10 years was 0.2 per cent.104* Karlsson et al. have presented the findings of an analysis of 13 patients who developed soft tissue sarcoma out of a total of 13 490 women treated by surgery with or without radiation in the West of Sweden Health Care Region.105 Although seen rarely after low doses (40 Gy), this is predominantly a complication of high-dose treatment. The actuarial frequency at 15–20 years is approximately 0.5 per cent for radiation of normal bone and soft tissue in the adult treated with radiation alone to full dose. Previous studies showed that children treated for soft tissue sarcoma have increased risk for developing second cancers, and although further analysis confirms that multiple malignancies could in some cases reflect a genetic syndrome, such as neurofibromatosis type 1 or the Li–Fraumeni
Clinical evaluation 929
syndrome, the chemotherapy and radiation patients receive also contributes to their risk.106* Among 1499 children reported to the Surveillance, Epidemiology, and End Results (SEER) population-based cancer registries 28 subsequent primary malignancies developed, compared with 4.5 expected malignancies based on general population rates (observed-to-expected [O/E] ratio 6.3 (95 per cent confidence interval [95 per cent CI], 4.2–9.1). Initial therapy with radiation and chemotherapeutic agents was associated with a higher risk of second malignancies compared with surgery alone (O/E ratio 15.2 vs. 1.4; p 0.0001). Elevated risks were observed for acute myeloid leukaemia, cutaneous melanoma, breast cancer and sarcomas of bone and soft tissue, with generally higher risks among patients who received combined modality therapy. Excess cancers of the oral cavity were prominent among long-term survivors. This is further supported by the analysis of 1458 patients followed after treatment for retinoblastoma for 17 years. The cumulative mortality from second primary neoplasms at 40 years was 26 per cent and 1.5 per cent among patients with bilateral and unilateral disease respectively.107* Within the bilateral population, the figures were 30 per cent and 6.4 per cent for irradiated and non-irradiated patients. For sarcomas (including those which arose within and without the irradiated field), the observed/expected ratios were 61 and 22 per cent for patients with bilateral disease. The comparable figures for patients with unilateral disease were 5 and 2 per cent. There was a difference in the use of chemotherapy in patients treated for bilateral and unilateral retinoblastoma, viz. 48 per cent and 13 per cent, respectively. The UK Registry of Childhood Tumours has investigated for secondary bone neoplasms108*a cohort of 13 175 paediatric age patients who had survived for 3 years. These patients had been seen over the time period 1940–1983. A total of 55 patients had developed secondary bone tumours. Among these 3 years survivors, 0.9 per cent developed bone tumours within 20 years. There was an increasing frequency with radiation dose. In addition, there was also an increasing frequency with dose of alkylating agent. In an analysis at Sloan-Kettering Cancer Center, New York, of 130 long-term survivors after treatment for rhabdomyosarcoma (2 years off chemotherapy) and a median followup period of 9 years, three patients developed leukaemia and four developed a solid tumour.109 Of the latter, three were in the radiation field; the fourth appeared in a nonirradiated subject. This experience is rather more favourable than that with retinoblastoma. Chemotherapeutic agents are likewise associated with risks of sarcoma induction. For example, there are two reports describing the appearance of osteosarcoma in children treated for leukaemia by drugs without radiation therapy.110 Tucker et al. analysed the late sequelae in 9170 long-time survivors of childhood cancer.111* Chemotherapy alone was concluded to be an independent risk factor. Exposure to a few selected industrial chemicals including vinyl chloride, phenoxyacetic acid, arsenic, and phenoxy herbicides may be followed by the appearance of sarcomas.
There are, however, a number of inherent problems in occupational epidemiology with small numbers of patients in any given series and the difficulty in isolating a single agent.112 For these reasons, few associations can be considered established and causal.113 For example, there is a clear association between vinyl chloride and hepatic angiosarcomas.114 Phenoxyacetic acid115 and arsenic116 are also implicated as inducing agents for hepatic sarcomas in humans. Wingren et al. reported an increased incidence of soft tissue sarcomas in gardeners (phenoxy herbicides), railroad workers, construction workers exposed to impregnating agents or asbestos, and unspecified chemical workers.117* The association between exposure to phenoxy herbicides and soft tissue sarcoma has been corroborated.118 The last risk may be greater with exposure to phenoxy herbicides contaminated with 2,3,7,8-tetrachlorodibenzo-dioxin (TCDD) or higher chlorinated dioxins.119,120 A role for dioxin per se is controversial. A population-based case control study, however, found no increased risk for soft tissue sarcoma among Vietnam veterans, including those exposed to Agent Orange, which contains dioxin.121* High-intensity chlorophenol exposure in jobs involving wood preservation or machinists who use cutting fluids may increase the risk of soft tissue sarcoma, independent of phenoxy herbicides.117,122 However, some studies have not confirmed this association.115 Chronic oedema and trauma may also be contributing factors to the malignant transformation. Sarcomas of soft tissue (primarily lymphangiosarcomas) may be observed following massive and quite protracted oedema. Classically, this has been seen in the post-mastectomy, lymphoedematous arm (Stewart–Treves syndrome).123,124 It has also been described following chronic lymphoedema due to filarial infection.125 Chronic irritation secondary to foreign bodies may be a factor in the induction of sarcomas. Trauma is rarely a factor in the development of these tumours with the exception of desmoid tumours. In an occasional patient, there is a history of major trauma to the affected site many months prior to the appearance of local symptoms of tumour. The usual history is of a traumatic incident occurring shortly prior to the awareness of the mass, suggesting that the trauma merely brought the patient’s attention to the presence of the mass.
CLINICAL EVALUATION Clinical history The most frequent initial complaint is that of a painless lump of a few weeks to several months’ duration. Occasionally, pain or tenderness precedes the detection of a lump. With progressive growth of tumour, symptoms appear which are secondary to infiltration of or pressure on adjacent structures (e.g. tendons, muscles, nerves) or organs. Occasionally symptoms secondary to the metabolic effects of the tumour products are seen, e.g. fever, anaemia, lethargy, weight loss, histamine-like reactions. These are not rare in patients with
930 Soft tissue sarcomas
malignant fibrous histiocytoma.2 To accrue clinical genetic data in sarcoma patients, the history should include details of the cause of death and history of malignant disease in siblings, parents, grandparents and progeny. ANATOMICAL SITE, SEX AND AGE
Sites of appearance of soft tissue sarcoma in order of frequency are listed above in the introduction and are shown in Table 40.2. There is a very slight preponderance of soft tissue sarcomas in males. They are more common in older people, with 40 per cent in persons 55 years of age, but the median age is 50, which is younger than that of carcinomas. It is important to emphasize that they can occur in all age groups and 15 per cent appear in patients 15 years of age.2 Rhabdomyosarcomas almost always arise in children, synovial sarcomas develop in late adolescence and young adulthood, and liposarcoma and malignant fibrous histiocytoma usually occur during mid and late adulthood.
Physical examination There must be a complete physical examination with particular attention paid to the region of the primary lesion: definition of size, site of origin (superficial or deep, attached to or fixed to deep structures), solitary or multi-nodular, involvement or discolouration of overlying skin, functional status of vessels and nerves, mass effect on adjacent organs, presence of distal oedema, muscular strength, range of motion of affected part, etc. (Fig. 40.1). If the patient has had prior excision, the operated site should be examined for presence of ecchymosis, status of wound healing, palpable evidence of residual tumour, and location of drain site. The regional and distant lymph node groups need to be examined with care in all patients, especially those with large grade 2 and 3 sarcomas. In an analysis of the experience at Massachusetts General Hospital, no patient
Figure 40.1 A 28-year-old man with large, T2b grade 2/3 myxoid liposarcoma of the right thigh.
with grade I sarcoma developed regional node involvement. The incidence in patients with grades 2 and 3 sarcomas was 2 per cent and 12 per cent, respectively. Of the patients with grade 3 sarcoma the incidences were 3 per cent and 15 per cent for lesions 5 cm and 5 cm, respectively.126* Fong et al. found a slightly lower frequency of metastasis to regional nodes in the Sloan-Kettering Cancer Center, New York, series.127 Involvement of regional nodes is relatively frequent in patients with rhabdomyosarcoma and epithelioid sarcoma, can occur in patients with angiosarcoma, synovial sarcoma, and clear cell sarcoma, but is uncommon in patients with fibrosarcoma and malignant fibrous histiocytoma. Myxoid liposarcomas frequently metastasize first to bone or soft tissue. For intra-abdominal, retroperitoneal and pelvic tumours, one should perform a careful abdominal and rectovaginal exam.
Laboratory investigations Laboratory studies for patients with extremity, truncal and head/neck sarcomas need not go beyond a complete blood count and blood urea nitrogen and serum creatinine (if intravenous contrast is to be administered for the chest CT or for radiation planning CT scan) unless the patient is to receive chemotherapy, in which case liver function tests should be performed as well. For patients with retroperitoneal sarcomas, blood work should include complete blood count, electrolytes, renal function and liver function tests. For extraskeletal Ewing’s sarcoma, lactate dehydrogenase (LDH, which has been shown to have prognostic value in Ewing’s sarcoma) should be obtained.128
Radiographic evaluation For the primary site, the radiographic evaluation should include plain films and MRI scanning. The most useful radiological study to evaluate the primary site for extremity, head and neck and truncal sarcomas is MRI (Fig. 40.2);129 CT can be useful to evaluate the relatively uncommon situation where there is suspected bony involvement. CT scan is also the procedure of choice for evaluating intraabdominal and retroperitoneal sarcomas. In these patients, CT scans should be carefully reviewed to assess for invasion of adjacent organs, vessels and nerves. Abdominal MRI is sometimes helpful in examining the relationship of tumours to major vessels and nerves. Plain radiographs are helpful in evaluation of soft tissue tumours by demonstrating bone involvement and soft tissue masses arising from bone tumours. However, unlike bone tumours, imaging studies cannot be used to assess the biological behaviour of soft tissue tumours.130 Indeed, specific diagnosis remains impossible for many soft tissue lesions regardless of the choice of imaging.131 Imaging studies alone cannot definitively distinguish malignant from benign soft tissue lesions. MRI may prove of clinical value in
Clinical evaluation 931
planning the biopsy site, using either needle or incisional techniques. Furthermore, the demonstration of necrotic regions appears to be of diagnostic importance in discrimination between benign and malignant tumours.132 PET scanning (Fig. 40.3) has been shown to be useful in discriminating between benign and high-grade lesions although it is unsuitable for distinguishing between benign and low to intermediate grade lesions.133 PET may be of substantial value in defining response to preoperative therapy.134,135 MRI studies should always include T2-weighted sequences as these provide the optimum contrast between lesion and muscle. Contrast enhanced T1-weighted images (especially with fat suppression techniques) are also helpful. MRI often provides a clearer demonstration of the anatomical location of the lesion and the pattern of local extension. For extremity, head and neck and truncal tumours, the findings from these scans should be correlated with those of the physical examination to assess the details of the anatomical site of the tumour. It should be determined whether the lesion is in the subcutaneous tissue, transgressing the fascia, intermuscular or intramuscular, displacing or enveloping major vessels or nerves, abutting or invading bone, etc. Depending upon the pattern of presentation and the nature of any planned surgery, an arteriogram or CT angiogram may be of value. For rhabdomyosarcoma, epithelioid sarcoma, high-grade synovial and unclassified sarcomas, PET, PET/CT or MRI evaluation of the regional nodes should be obtained. Bone scans need not be performed unless specifically indicated. We do not consider a positive bone scan near or adjacent to a soft tissue sarcoma to be proof of invasion of periosteum or bone. For a diagnosis of invasion of bone there must be clear radiographic evidence of destruction of cortical bone. The single most important examination for
distant metastasis is whole lung CT; this should be obtained in all patients with intermediate or high-grade tumours. This has been extensively confirmed by the study of Peuchot and Libshitz who reported that, of the nodules detected by CT but not by chest X-ray and those biopsied, 94 per cent were metastatic tumour.136 For intra-abdominal, retroperitoneal and pelvic tumours, the liver should be examined by CT scan to exclude hepatic metastases. Imaging of the response to treatment with CT and MRI has been generally disappointing up to the present. Decrease in tumour size may occur, but does not correlate well with successful radiation or chemotherapy.137 Furthermore, although tumour volume can be approximated from twodimensional images (CT or plain films), reliable algorithms to distinguish objectively tumour from surrounding tissues are not available, and thus three-dimensional imaging techniques have, as yet, not found a role. Changes in MRI signal characteristics have been unreliable. Absence of high signal intensity on T2-weighted images has been shown to indicate freedom from tumour. However, residual high signal may be due to tumour, oedema or fibrosis.138 MRI spectroscopy has been used to detect high-energy phosphate metabolism in the lesions. This has helped in the distinction between malignant and benign tumours.139 Several studies with phosphorus-31 magnetic resonance spectroscopy have shown changes in high-energy phosphate metabolism after effective chemotherapy, but the range of variation in sarcoma is large, and because of limitations in spatial resolution the procedure cannot be reliably done unless a large soft tissue mass is present.140–142 Schuetze and colleagues evaluated 46 patients with high-grade sarcoma who received neoadjuvant chemotherapy and reported that reductions in standardized uptake value of FDG (SUVmax)
(a)
(b)
Figure 40.2 T1-weighted coronal (a) and T2-weighted axial (b) MRI scan of the right thigh in patient shown in Figure 40.1.
932 Soft tissue sarcomas
Figure 40.3 PET-CT scan of patient shown in Figure 40.1, obtained after completion of three cycles of interdigitated neoadjuvant MAID (mesna, doxorubicin, ifosfamide and dacarbazine) chemotherapy and 44 Gy of radiation therapy (See Plate Section.).
correlated with the degree of necrosis in the resection specimen.135 They also noted that tumours with a baseline SUVmax of 6 had a higher risk of developing subsequent metastatic disease. Patients with a 40 per cent decrease in tumour SUVmax had a significantly lower risk of disease recurrence and of metastasis. All four patients with local disease recurrence as the initial event had a 40 per cent reduction in tumour SUVmax after chemotherapy. A reduction in the pre-surgery sarcoma SUVmax of 40 per cent relative to the baseline SUVmax correlated with overall survival. A metabolic response as determined by FDG-PET scans (as seen in Fig. 40.3) correlated with a pathologic response to preoperative doxorubicin-based chemotherapy but was a stronger determinant of distant recurrence of disease than the pathologic response. PET scans have been able to document response in patients with gastrointestinal stromal (GIST) tumours to imatinib within a week of starting therapy.143
Biopsy The optimal treatment strategy is based on the correct diagnosis. An adequate biopsy must be obtained in order that
the best feasible histopathological diagnosis as to tumour type and grade can be made. Core needle biopsy is the preferred initial biopsy technique. For palpable lesions with a superficial component that is away from the neurovascular bundle, a Tru-Cut needle biopsy can often be obtained in the clinic without image guidance. For non-palpable, deeper lesions, and those near the neurovascular bundle, core biopsies can be obtained under CT or ultrasound guidance. There is a recent body of literature that supports the use of core needle biopsy for definitive diagnosis of bone and soft tissue sarcomas.144–146 The accuracy of obtaining diagnostic tissue by needle biopsy is reported to be 96 per cent,147 with a 94 per cent accuracy in diagnosing the tumour as malignant, an 85 per cent accuracy of subtyping the sarcoma, and an 88 per cent accuracy of grading the sarcoma.148 When one is able to successfully obtain a diagnosis in this fashion, it is preferred over the open biopsy. It is faster, simpler, and presents less risk to the patient of infection and haematoma that can delay or compromise treatment. The core biopsy is reported to have accuracy, sensitivity and specificity that is not substantially different from open biopsy.149–151 It can be difficult to distinguish lipomas from well-differentiated liposarcomas on core biopsies but this
Pathology 933
may also be the case with open biopsy, and definitive diagnosis in these cases may not be made until definitive resection. Several comments are pertinent with regard to the use of the core biopsy technique. The needle path for biopsy should allow excision of the biopsy track at the time of definitive surgery to avoid potential needle track recurrence.152 Careful attention to haemostasis and sterility are important to avoid haematoma or infection. Image guidance with ultrasound or CT can help direct and document placement of the biopsy needle. Availability of the pathologist at the time of the procedure can help confirm that diagnostic tissue was indeed obtained with the needle. Access to relevant clinical history and radiographic studies may help the pathologist is making the diagnosis from the available biopsy material. As an increasing number of protocols are seeking to gather molecular profiles on patients accrued to studies, it is also important to note that sufficient material for cDNA arrays can be obtained by core biopsies.153 Open biopsies can be reserved for patients in whom diagnostic material cannot be obtained by needle core biopsy. There has also been recent discussion as to whether fine needle aspiration cytology in lieu of biopsy can suffice for the initial diagnostic procedure. Because of the range of histological types of soft tissue sarcomas, fine needle aspiration (FNA) may be insufficient to adequately characterize and subclassify the histological type of tumour. It is, however, already accepted as an important and widely used diagnostic procedure for the evaluation and documentation of local recurrences, as well as metastases from previously diagnosed soft tissue sarcomas and for diagnosing benign nonneoplastic lesions and metastatic carcinomas in soft tissue.154 Its role as the primary modality for the initial diagnosis of soft tissue sarcomas is not as widely accepted by all clinicians and pathologists. Its main limitation is in the evaluation of architecture, which is available on core biopsies, but not on FNA. A recent report from an experienced Swedish group where both FNA and core biopsies were able to be done at the same procedure noted a high degree of diagnostic accuracy and reported that the clinicians managing the patients rapidly perceived the advantages of this double-diagnostic approach. They indicated that the FNA cytology smears allowed for excellent evaluation of cytomorphology, sampling of different parts of large tumours and were well suited for rapid staining, which allowed for immediate diagnosis in many cases on the day of the clinic visit. The core biopsies gave architectural information and also provided sufficient material for ancillary studies (immunocytochemistry, electron microscopy and cytogenetic and molecular genetic analyses) permitting high degrees of diagnostic accuracy.155* This may be a model that is worth evaluating at other centres. When core biopsy is not diagnostic, open biopsy must be performed. Under optimal conditions, the open biopsy procedure should be performed by an experienced surgeon who is part of the multidisciplinary care-taking team and will be responsible for the definitive surgery. An excisional biopsy is generally performed for superficial tumours less
than 3–5 cm in size and an incisional biopsy is generally performed for deeper tumours greater than 3–5 cm in size. The surgeon should adhere to the same principles as for the definitive surgery (see below) if complications are to be avoided. Prior to the biopsy, the imaging studies should be carefully studied to ascertain the most logical approach to the lesion, with explicit consideration of the regions to be traversed in subsequent surgical procedures (including marginal or wide resection or amputation) so that the biopsy track will not interfere with either surgery or radiation field.156 When incisional biopsies are performed, the incision for the biopsy of lesions on an extremity should be longitudinal (there is almost never a reason for a transverse incision on an extremity). The incision should be as short as possible yet long enough to avoid excessive retraction of tissue or to make dissection and haemostasis difficult. The biopsy track should go through a muscle belly rather than along fascial planes (the former tends to keep the tumour ‘spill’ within an anatomical compartment while the latter allows transgression of two or more compartments) and careful attention should be paid to achieve haemostasis in order to avoid ecchymosis or a haematoma. The wound should be closed in layers with a narrow skin closure; as a rule, drains should not be utilized (the tract of the drain is considered to be contaminated with tumour and may greatly extend the planes of subsequent surgery or the radiation treatment volume). For the occasional small lesion in a readily accessible site (e.g. wrist or ankle), an excisional biopsy may be the approach of choice. The surgeon should adhere to the same principles as for the definitive surgery (see below) if complications are to be avoided. The biopsy specimen needs to be of sufficient volume to be certain that it is representative. A pathological assessment of a frozen section is useful in assuring that the tissue obtained is from the lesion and is adequate for the diagnostic evaluation. Cultures should always be obtained. Specimens are processed for haematoxylin and eosin staining and various immunohistochemical stains considered necessary to aid in the diagnosis. A small portion of the tissue is set aside for electron microscopy, cytogenetics and increasingly, for gene arrays.
PATHOLOGY Practical comments The pathologist needs to aware of the clinical and radiographic findings and the diagnostic considerations of each case. In this way the pathologist will be best prepared to choose the appropriate methodology needed to make a complete and accurate assessment of tissue specimens. THE BIOPSY
Ideally, the tissue should be in the fresh state when received by the pathologist so that a portion can be used to perform
934 Soft tissue sarcomas
a frozen section. Frozen section analysis can determine if there is adequate material for diagnosis and may even permit the rendering of a specific diagnosis. Depending on the results of the frozen section, the pathologist can then triage the tissue as necessary and submit samples for electron microscopy, DNA flow cytometry, cytogenetics, FISH (fluorescence in situ hybridization) or keep tissue frozen for immunoperoxidase or other molecular studies. If a core needle biopsy is performed, then three cores of tumourbearing tissue are usually required; if an open biopsy is performed, 0.5 cm3 of tumour is sufficient to perform the necessary studies. All tissue not used for special studies should be examined by light microscopy. THE RESECTION SPECIMEN
At the Massachusetts General Hospital, the surgeon takes the operative specimen to the operating room pathology suite. At this time the surgeon orients the specimen for the pathologist and discusses potentially close or positive margins. Once the specimen is in the pathology laboratory, it is carefully dissected to determine the size of the sarcoma, its gross characteristics, its relationship to normal structures, pattern of invasion, and the adequacy of excision. If necessary, a frozen section to confirm the status of a margin is performed. All portions of the sarcoma are thoroughly sampled for study by light microscopy. The final analysis includes classification and grading, determination of vascular invasion, status of the margins, proximity to normal structures, and cytotoxic therapy effect as seen by percentage necrosis.
Histological classification The rationale for developing a well-defined, comprehensive and flexible classification system of soft tissue tumours is to provide morphological guidelines which expand our understanding of neoplasia, predict biological behaviour, and facilitate the development of more effective treatment. Originally, classification schemes were descriptive in nature and based on tumour cell configuration. Subsequently they have evolved through the concept of histogenesis or ‘cell of origin’ to the current belief that a primitive or stem-like mesenchymal cell undergoes neoplastic transformation and, depending on the genetic code translated, differentiates along one or multiple cell lines. Light microscopy in most instances is the modality of choice for determining whether a soft tissue tumour is benign or malignant, subtyping, grading, assessing margins, and determining the presence or absence of vascular invasion. However, electron microscopy, immunohistochemistry, DNA flow cytometry, cytogenetics and molecular analysis can provide valuable information that substantiates the histological interpretation. Cytogenetic analysis is now being performed more frequently on sarcomas. Molecular biology will play a more important role in evaluating soft
tissue sarcomas in the near future as the technology becomes more widely available. The identification of specific DNA and RNA gene sequences and oncogenes will help to diagnose and predict the biological behaviour of sarcomas. For example, the expression of the gene product MYO D1 has already proved to be helpful in recognizing tumours showing skeletal muscle differentiation.157,158 Currently the most widely used classification system is the Enzinger and Weiss modification of the World Health Organization formulation.2 In this system, soft tissue tumours, including non-neoplastic tumour-like lesions, are categorized into three broad groups: 1. tumours which differentiate along cell or tissue lines that have normal counterparts, i.e. fibrous tissue, fat, vessels, smooth muscle, skeletal muscle, nerve, ganglia, synovium, bone and cartilage; 2. tumours whose lines of differentiation have no normal counterpart but which are consistent and recognized by a distinctive morphology, i.e. myxoma, epithelioid sarcoma, and alveolar soft part sarcoma; 3. tumours which are so poorly differentiated and morphologically unique that they defy classification. The vast majority of tumours fall into the first two groups. Overall, there are approximately 200 different entities, of which 80 are malignant. As intensive study of these tumours is rapidly expanding and new diagnostic procedures are increasingly employed, there is inevitably some flux in the diagnostic criteria for the diverse groups of soft tissue tumours. Examples include: 1. the reclassification of most adult pleomorphic rhabdomyosarcomas and many pleomorphic liposarcomas to malignant fibrous histiocytoma; 2. the recognition that a granular cell tumour is a schwann cell neoplasm and that clear cell sarcoma is a malignant melanoma primary to the soft tissues; 3. extra-skeletal Ewing’s sarcoma is a primitive neuroectodermal tumour.
Grading The histological typing of soft tissue tumours does not per se provide sufficient information on which to base therapeutic decisions. Tumour grading is based on the concept that morphology reflects biological behaviour. The specific microscopic characteristics of soft tissue tumours that best predict their aggressiveness, i.e. the potential for regional and distant metastasis, can be identified, integrated and represented by grade. The AJC and UICC staging systems for sarcoma of soft tissue are based upon classification of the tumours into lowand high-grade tumours. The WHO employs a four-tiered grading system: G1 well differentiated, G2 moderately differentiated, G3 poorly differentiated, and G4 undifferentiated,
Pathology 935
(a)
(c)
with the G1 and G2 lesions considered low grade and G3 and G4 considered high grade. Some institutions employ a three-step grading, i.e. low, intermediate and high grade neoplasms, with the intermediate and high-grade designations considered to be high grade for staging purposes. The designation of grade is based upon a consideration and integration of each of these morphological features: degree of cellular differentiation, extent of necrosis, number of mitoses, cellularity, pleomorphism or anaplasia, quantity of matrix, vascularity, haemorrhage, vascular invasion and encapsulation.159–165 Among these variables, necrosis, mitoses and degree of differentiation appear to be the best predictors of outcome. Despite some lack of agreement on the number of grades employed and the significance of individual morphological parameters (there is inevitably a subjective component in assigning grade and only a part of the tumour is examined), grading, more than any clinical and pathological parameter available, is the most important prognosticator.166* Highpower views of a soft tissue sarcoma, grades 1, 2 and 3 (on a three-grade scale) are shown in Figure 40.4a–c. The problems with current grading systems are that their criteria are not precisely defined, application and interpretation is
(b)
Figure 40.4 Photomicrographs showing (a) low, (b) intermediate, and (c) high grade soft tissue myxoid liposarcoma. (See Plate Section.)
subjective and implementation is complex. Consistent grading requires adequate tissue and experienced pathologists.
The pathologist’s armamentarium in diagnosing and grading sarcomas Light microscopy is fundamental in diagnosing soft tissue sarcoma. However, electron microscopy, immunohistochemistry, DNA flow cytometry, cytogenetics and molecular analysis can provide valuable information that substantiates the histological interpretation. Electron microscopy (EM) is particularly helpful for identifying poorly differentiated tumours, the various spindle cell sarcomas and malignant round cell tumours. In some cases the presence of single ultrastructural features such as Z-bands in rhabdomyosarcomas may be diagnostic. EM, however, is frequently not reliable in distinguishing benign from malignant soft tissue tumours.157 Immunohistochemistry has become increasingly import-ant in diagnosing sarcomas.167 This technique is based on the utilization of visually tagged monoclonal or polyclonal antibodies directed against proteins found in specific cell types. If a tumour
936 Soft tissue sarcomas
Table 40.3 Immunohistochemistry in soft tissue sarcomas Vimentin Keratin Desmin Glial fibrillary acidic protein Neurofilament Leucocyte common antigen S-100 protein Myo D1 Myoglobin Factor VIII related antigen CD34 Smooth muscle actin EMA Leu 7 CD99 (MIC2 gene product) CD117 (c-kit)
Almost all sarcomas and some carcinomas Almost all carcinomas and some sarcomas (epithelioid sarcoma, synovial sarcoma) Leiomyosarcoma, rhabdomyosarcoma and occasionally MFH Some schwannomas Primitive neuroectodermal tumour, neuroblastoma Lymphoma Malignant schwannoma, melanoma, clear cell (melanoma of soft parts), chondrosarcoma, leiomyosarcoma, rhabdomyosarcoma, liposarcoma Rhabdomyosarcoma Rhabdomyosarcoma Angiosarcoma, Kaposi’s sarcoma Angiosarcoma, epithelioid sarcoma, MFH, haemangiopericytoma, solitary fibrous tumour Leiomyosarcoma Carcinomas, synovial sarcoma, meningioma Malignant schwannoma, leiomyosarcoma, synovial sarcoma, rhabdomyosarcoma Ewing’s sarcoma/PNET, some rhabdomyosarcomas, some synovial sarcomas Gastrointestinal stromal tumour
MFH, malignant fibrous histiocytoma; PNET, primitive peripheral neuro-ectodermal tumour.
expresses certain proteins as detected by this method, then its phenotype or line of differentiation can be identified (Table 40.3). The more important antibodies include those that recognize some of the different intermediate filaments (vimentin, keratin, desmin, neurofilament), leucocyte common antigen, S-100 protein, myoglobin and factor VIII-related antigen. For some antigens (e.g. neurofilament and surface antigens to lymphocytes), frozen tissue is clearly superior to formalin-fixed tissue.168 For other antigens, such as S-100 protein, peptide hormones, neuronspecific enolase and α1-antitrypsin, formalin fixation is optimal.168 It is important to remember that there is no antibody that distinguishes a benign from a malignant soft tissue tumour and that there are a multitude of exceptions to the expected distribution of antigens. Consequently there can be significant overlap in the immunohistochemical profile of various sarcomas.169 In addition to identifying antigens associated with a specific phenotype, immunohistochemistry has been used to evaluate the drug resistance and proliferative rate of sarcomas and their expression of growth factors. P-glycoprotein is a plasma membrane glycoprotein which appears to play a role in multi-drug chemotherapy resistance170 and can be detected by immunohistochemistry. Investigations have tried to determine whether the level of P-glycoprotein expression predicts the type of response to chemotherapy. High levels of P-glycoprotein expression may adversely affect the chemosensitivity of sarcomas, which correlates with a worse prognosis.171 Similarly, the absence of P-glycoprotein in some sarcomas portends a favourable outcome.171,172 Ki-67, a nuclear antigen expressed in all phases of cell proliferation except the resting phase,173 is a recognized marker of cell proliferation. Demonstration of Ki-67 appears to measure the number of proliferating cells in
a tumour more accurately than does DNA flow cytometry.173 In malignant fibrous histiocytoma the expression of Ki-67, as determined by immunohistochemistry, has shown that proliferative activity is independent of grade.174 Some studies have demonstrated a correlation between Ki-67 and prognosis in general,175 but further investigation is warranted before firm conclusions can be established.176 Mutations of p53, a tumour suppressor gene, have been shown to play an important role in the development of a variety of neoplasms including soft tissue tumours. The abnormal accumulation of p53 indicates that a mutation has occurred. p53 can be visualized by immunohistochemistry and it has been detected both in benign and malignant soft tissue neoplasms.177 In sarcomas p53 has been associated with specific subtypes, a high histological grade and poor prognosis.33 Finally, several growth factors including platelet-derived growth factor,178 insulin growth factor179 and epidermal growth factor88 have been identified in a variety of benign and malignant soft tissue tumours. The significance of their presence is unclear but may relate to both cell differentiation and tumour growth. DNA flow cytometry can play a role in determining the prognosis of sarcomas; however, it still cannot replace diagnosis by light microscopy. DNA flow cytometry measures the amount of DNA in cells. Generally, the cells in reactive processes, benign tumours and low-grade sarcomas have normal amounts of DNA and are termed diploid. In contrast, the neoplastic cells in high-grade sarcomas have a tendency to contain abnormal amounts of DNA and are termed aneuploid. Some studies have shown a statistically significant relationship between aneuploidy of sarcomas and poor clinical outcome.180 There also appears to be a relationship in some cases between tumour grade and ploidy level; higher proportions of aneuploid tumours are found
Staging 937
amongst the higher-grade sarcomas.181 Besides ‘ploidy’ of a tumour, flow cytometry can perform cell cycle analysis and determine the number of cells that are synthesizing DNA in preparation for cell division. In general, malignant tumours have a higher proportion of cells that are undergoing mitosis than benign tumours.182 Similarly, high-grade sarcomas usually have a greater number of cycling cells than low-grade sarcomas.182 The problem with flow cytometry is that there are many exceptions to the rules: benign tumours may be aneuploid and actively growing; high-grade sarcomas may be diploid with a relatively low proportion of cycling cells. Cytogenetic analysis is now being performed more frequently on sarcomas. It is a diagnostically useful technique because some sarcomas have specific cytogenetic alterations which appear to be pathognomonic. For instance, 83 per cent of Ewing’s sarcoma/primitive neuroectodermal tumours have a characteristic t(11;22) translocation and 50 per cent of alveolar rhabdomyosarcomas show a t(2;13) translocation.36,183,184 Myxoid liposarcomas have been found to have a t(12;16) translocation, clear cell sarcoma a t(12; 22) translocation; extra-skeletal myxoid chondrosarcoma a t(9;22) translocation; and synovial sarcomas a t(x;18) translocation.157 Characteristic cytogenetic changes are also found in infantile fibrosarcomas and dermatofibrosarcoma protuberans. Structural abnormalities of chromosome 1p have been found in 70–80 per cent of neuroblastomas. They are often associated with double minute chromosomes and/or homogeneously staining regions (sites of amplification of complex rearranged genes), including a core of multiple copies of N-myc oncogene. Aside from being helpful in diagnosis, the findings of these 1p rearrangements and double minute chromosomes further denote a poor clinical outcome.185 Molecular biology will play a more important role in evaluating soft tissue sarcomas in the near future as the technology becomes more available. The identification of specific DNA and RNA gene sequences and oncogenes will help to diagnose and predict the biological behaviour of sarcomas.
STAGING The Task Force on Soft Tissue Sarcomas of the American Joint Committee on Cancer (AJCC) Staging and End Result Reporting has established a staging system for soft tissue sarcomas that is an extension of the tumour/node/ metastasis (TNM) system to include G for histological grade. Grade, size, depth and presence of nodal or distant metastases are the determinants of stage. This staging system is applied to all sarcomas of soft tissue except rhabdomyosarcoma (for which there is a special staging system), Kaposi’s sarcoma, dermatofibrosarcoma, desmoid, and sarcoma arising from the dura mater, brain, parenchymatous organs or hollow viscera. The staging system was revised in 1998 with the addition of subgroupings of the T stage to designate
Table 40.4 2002 American Joint Committee on the staging of cancer staging system for soft tissue sarcomas442 Primary tumour (T) TX T0 T1
T2
Primary tumour cannot be assessed No evidence of primary tumour Tumour 5 cm or less in greatest dimension T1a superficial tumour* T1b deep tumour* *Superficial tumour is located exclusively above the superficial fascia without invasion of the fascia; deep tumour is located either exclusively beneath the superficial fascia, superficial to the fascia with invasion of or through the fascia, or both superficial yet beneath the fascia. Tumour more than 5 cm in greatest dimension T2a superficial* T2b deep*
Regional lymph nodes (N) N0 N1
No regional lymph node metastasis Regional lymph node metastasis
Distant metastases (M) M0 M1
No distant metastasis Distant metastasis
Histological grade (G) G1 G2 G3 G4
Well differentiated Moderately differentiated Poorly differentiated Undifferentiated
Stage grouping I II III IV
Low grade, small or large, superficial or deep: G1–2, T1a–2b, N0, M0 High grade, small, superficial or deep or large superficial: G3–4, T1a–b, T2a N0, M0 High grade, large, deep: G3–4, T2b, N0, M0 Any metastasis, any G or T, at least N1 or M1 or both
This staging system is not to be used for Kaposi’s sarcoma, dermatofibrosarcoma (protuberans), fibrosarcoma grade I (desmoid tumour) and sarcoma arising from the dura mater, brain, parenchymatous organs or hollow viscera.
superficial and deep lesions and the assignment of patients with nodal involvement to stage IV. Superficial lesions do not involve the superficial investing fascia in extremity lesions. The current system, last updated in 2002, is outlined in Table 40.4. Grade of sarcoma is determined on the basis of the histological features of the individual tumour. T stage is determined on the basis of size and depth. As evidence for the importance of size as a determinant of frequency of distant metastasis, we present in Table 40.5 an analysis of distant metastasis versus tumour size among patients who have achieved local control. For patients with
938 Soft tissue sarcomas
Table 40.5 Five-year actuarial distant metastasis (DM) probability in 501 consecutive local control patients as a function of tumour size for grades 2 and 3 in series from Massachusetts General Hospital (treatment by radiation and surgery)190 Size (mm)
No. of patients
DM (%)
25 26–50 51–100 101–150 151–200 201 plus
58 128 177 68 49 21
3 22 34 43 58 57
Total
501
35
grade 1 sarcomas, distant metastases are quite uncommon. The pooled data for grade 2 and grade 3 lesions show regular increases in distant metastases with tumour size. From these data there clearly is importance in stratification of patients according to grade and size in attempts to compare efficacy of different modes of treatment, defining the natural history of various histological types, or assessing the role of site, patient age, sex, etc.
MANAGEMENT OF THE EXTREMITY, TRUNK, OR HEAD AND NECK PRIMARY TUMOUR The intent of treatment for extremity sarcomas is to eradicate tumour while optimizing limb function. For truncal and head and neck sarcomas, minimizing the functional and cosmetic defects associated with treatment is also extremely important.
Overview Because sarcomas tend to infiltrate normal tissue adjacent to the evident lesion, simple excision alone is followed by local recurrence (LR) in 60–90 per cent of patients.186* Radical resection of a wider margin of apparently normal tissue around the tumour reduces the local failure rate to approximately 25–30 per cent.187* More recently, with the advent of compartmental resections, the local failure rate has fallen to 10–20 per cent with surgery alone.187 One study that reported a zero local failure rate derives from the amputation arm of the National Cancer Institute trial comparing amputation to limb salvage treatment.188** The combination of surgery and radiation achieves better outcomes than either treatment alone for nearly all soft tissue sarcomas.159,189 The rationale for combining radiation with surgery is to avoid the functional and cosmetic deformity associated with radical resection, and the late consequences of high radiation doses to large volumes of normal tissue in patients treated with primary radiation alone.
Radiation at moderate dose levels (60–65 Gy) is as effective as radical resection in eradicating the microscopic extensions beyond the gross lesion, resulting in similar high rates of local control. This has allowed maximization of functional and cancer-related outcome without the significant morbidity of radical surgery. Most centres report local control rates of approximately 90 per cent for high-grade extremity STS and 90–100 per cent for low-grade STS depending upon the size.4*,188*,190*,191*,192*,193*,194* In addition to its benefit in improving local control rates, adjunctive radiotherapy has also had a significant impact on limb salvage for extremity sarcomas. As an example, in the 1970s, 50 per cent of patients with extremity sarcoma underwent amputation; those patients treated by wide excision alone with limb preservation experienced a 30 per cent rate of local recurrence (LR). With the subsequent application of radiotherapy and advanced reconstructive techniques, the rate of amputation at major centres has been reduced to less than 10 per cent, and the incidence of LR with limb preservation has been reduced to 10–15 per cent without any measurable fall in overall survival (OS).188*,195*,196*,197* A single, prospective randomized trial showed similar rates of disease-free survival (DFS) and OS for patients treated with amputation or the combination of limb-sparing surgery and radiotherapy for extremity STS.188** The success of a conservative surgical approach has, as mentioned above, resulted in an amputation rate at major centres of only 5 per cent in patients with extremity STS. The current indications for amputation include: massive disease such that a functional limb is not achievable, as well as severely compromised normal tissues due to age, peripheral vascular disease and other co-morbidities. The functional and cosmetic results of conservative procedures are dependent upon the size and anatomical location of the tumour, the magnitude of the surgical procedure, extent to which muscles, tendons or nerves must be sacrificed, volume of tissues irradiated and the radiation dose administered. The combined modality strategy of surgery and radiation therapy with or without chemotherapy has been less successful in obtaining control of retroperitoneal sarcomas, where local failure rates are reported to be in the range of 70 per cent.198*
Surgical considerations Until the 1950s, treatment protocols for soft tissue tumours often consisted of marginal local surgical extirpation of the lesion (‘shelling out’) or, for larger or more extensive tumours, amputation. The former procedures were done without a clear understanding of the nature of the composition of the compressed ‘pseudocapsule’ or recognition of the presence of daughter nodules within or just outside this region, so that the local recurrence rates for such limited procedures were extraordinarily high. In the 1960s many authors were able to demonstrate improved results by a more radical resection in which the surgical resection was
Management of the extremity, trunk, or head and neck primary tumour 939
through normal tissue in three dimensions; the local recurrence rate was drastically reduced from 75–90 per cent to 10–30 per cent.199*,200* By the early 1980s better imaging led to improved staging methodology and a clearer understanding of the pattern of local extension of sarcoma by invasion through the pseudo-capsule201 and most surgeons advocated wide or even radical margins for the management of soft tissue tumours with vast improvement in local control.186 The technique utilized to achieve a wide or radical extirpation of the tumour will vary considerably with anatomical site and anatomical details of prior surgical procedures (especially the biopsy). The surgery is heavily dependent on the knowledge of anatomy of the part and careful study of the imaging studies is an essential part of the preparation for the procedure. The surgery, in theory at least, if it is to be successful in avoiding a local recurrence, must include not only the complete tumour and the track of the prior biopsy, but as wide a cuff of normal tissue as is possible without compromising function so severely as to make the procedure of limited value as limb-sparing surgery. Thus, if one is treating a tumour of the vastus lateralis by surgery alone, not only should that muscle be at least partially resected but prudence dictates that the underlying vastus intermedius, the lateral intermuscular septum and the adjacent portions of the biceps femoris and rectus femoris should be included in the specimen. Provided the surgical margins are proved to be clear, the procedure is usually successful (a local recurrence rate of 5 per cent). Despite the mass of muscle removed from that site, the patient demonstrates only modest disability in normal activities. The surgical procedure as described above is an effective solution to the problem in sites such as the fleshy parts of the thigh or the calf, or even some parts of the arm in selected patients, but is usually unsatisfactory and more complex when the tumour lies close to vital structures or in an anatomical site where a wide margin is difficult to obtain. Tumours that lie in the popliteal or antecubital fossae are particularly problematic as are those lesions arising in the soft tissues of the volar aspects of the forearm, hands and feet. In general the results of surgery alone, even with the best techniques, have a local failure rate of 10–20 per cent when all patients on whom the surgery is performed, including those with unsatisfactory margins, are considered.186*,201* The local control rate after surgery alone was 67 per cent in a recent phase III trial of surgery versus surgery and brachytherapy.193** A local failure not only enormously compounds the management problems at the site of the original lesion but also increases the risk of distant metastases. This has been a problem especially for sarcomas of the head and neck region. SURGICAL MARGINS
The most important surgical variable that influences local control is the presence or absence of tumour cells at the surgical margins.188*,195*,196*,202*,203*,204*,205*,206*,207* In series
Table 40.6 Five-year actuarial local control (LC) results and margin status after preoperative radiotherapy for extremity soft tissue sarcoma (Massachusetts General Hospital) Margin 1 mm 1 mm Undefined No tumour*
No. of patients
LC (%)
27 106 27 36 21 22
81 97 96 97 94 100
*Specimen negative for tumour.
that report radical resection with clear margins, such as the Scandinavian Sarcoma Group, the local failure rates are quite low (8 per cent).208* By contrast, in a second study of 559 patients who were treated with surgery alone from the same group, an inadequate surgical margin led to a 2.9fold greater risk of LR than did clear surgical margins.205* Distant metastases are extremely uncommon for low-grade lesions, but occur with high-grade lesions with a frequency that is influenced by the size of the lesion4 and whether local control is achieved. In the Scandinavian Sarcoma Group experience, local recurrence was identified as a risk factor for distant metastasis (4.4-fold higher).208* The status of the surgical margins also influences the local recurrence rate in patients treated with combined surgery and radiation.196*,202*,206* In one review of 132 consecutive patients with STS of the extremities who were treated with preoperative radiation therapy followed by resection, the 5-year actuarial local control rates were 97 and 81 per cent for patients with negative and positive margins respectively (Table 40.6).202 Local control was not a function of sarcoma size in patients with negative surgical margins. In a second series of 225 patients, all of whom received combined surgery and radiotherapy (either preoperative, postoperative or both), local control rates at 5 years were 88, 76 and 64 per cent for patients with negative, uncertain and positive margins, respectively.206* Fagundes et al.209* reported that patients treated by resection and postoperative radiation for malignant fibrous histiocytoma failed locally in nine of 23 (39 per cent) and one of 11 instances for positive and negative margins, respectively. Among 26 patients treated with resection, doxorubicin and postoperative radiation (51 Gy [3 Gy 17]) for positive margins,210* local failure was observed in six (23 per cent). Herbert et al.211* observed local control in 50 per cent of patients with positive margins compared with 100 per cent in those with negative margins. These reports indicate that positive margins are associated with a reduced local control rate. The experience at the Princess Margaret Hospital196* strongly supports efforts to achieve negative margins. Local relapse rates at 5 years were 13 per cent, 24 per cent and 77 per cent for negative margins, micropositive margins and gross residual disease in the wound, respectively. Our policy
940 Soft tissue sarcomas
Table 40.7 Local control in selected patients treated by conservative surgery with wide margins, generally including fascia and margins 1 cm Centre Lund DFCI/Brigham MDAH (5 cm) Goteborg Roswell Park University of Chicago
No. of patients
Local failure
Reference
56 74 43 63 116 35
7% 7% 17% 8% 10% 0%
Rydholm et al. 1991216 Baldini et al. 1999217 Fleming et al. 1999450 Markhede et al. 1982186 Karakousis et al. 1995451 Gibbs et al. 1997452
DFCI, Dana Farber Cancer Institute; MDAH, MD Anderson Hospital.
is to make serious efforts to achieve negative margins, although the majority of patients, even with a positive margin, will achieve local control, and conservative procedures in this setting are generally warranted, with the caveat that additional radiotherapy dose should be delivered to the area of positive margin.202 The exact size of the negative margin that is optimal for local control is not known. In one study, the local control rate did not differ in patients with a negative margin 1 or 1 mm (local control 96 versus 97 per cent).202* Most clinicians recommend that if surgery is used as the sole modality of treatment, the margin should be at least 1 cm in all directions212 or, if less, include a supervening fascial barrier. If surgery is combined with radiation therapy, the surgical margin can probably be safely reduced to 0.5 cm without compromising the rate of local control.202 The guiding principle of surgery is total en bloc excision of the primary tumour without cutting into tumour tissue. Tissues should be cut outside of the tumour pseudo-capsule, if one exists, through normal uninvolved tissue. Violation of the tumour results in a higher local failure rate. In one report, for example, the local control rate in 95 patients with extremity STS was 47 per cent if tumour violation occurred compared to 87 per cent without violation.203* The majority of STS do not involve bone; as a result, it is seldom necessary to resect adjacent bone. It is also rarely necessary to resect a major nerve unless the tumour is a neurogenic sarcoma. Non-amputative surgery is now accomplished in more than 90 per cent of patients. In planning primary therapy for a patient who has had a suboptimal resection by a non-oncologic surgeon and/or insufficient imaging with preoperative CT or MRI, it is important to consider re-resection. Approximately 37–68 per cent of such patients will have residual tumour in a re-resection specimen.197*,213*,214* A partial excision of the tumour before referral to a tertiary centre does not appear to compromise limb preservation, local control or survival rates in such patients,197 although the re-resection may entail a larger procedure than a de novo procedure and impact upon the functional result. In one series of 295 patients who underwent re-resection at a single institution (final resection margins negative in 87 per cent), local control rates at 5, 10 and 15 years were 85, 85 and 82 per cent; the corresponding values for those who did not undergo re-resection were 78,
73 and 73 per cent, respectively.214* A similar degree of benefit for re-resection was apparent for metastasis-free and disease-specific survival. Some patients, for medical, functional or cosmetic reasons may not be candidates for re-resection; appropriate doses of radiation therapy can help ensure local control in a substantial majority of these patients, 86 per cent in one recent series.215* SELECTION OF PATIENTS FOR TREATMENT WITH CONSERVATIVE SURGERY ALONE
Because of potential acute and late morbidity from radiation therapy, it is important to select patients who may be effectively treated with conservative surgery alone. Several series have evaluated wide-excision, limb-sparing surgery alone (Table 40.7). In one report, 119 selected patients with extremity STS were grouped according to anatomical location as subcutaneous (n 40), intra-muscular (n 30) or extra-muscular (n 49).216* The 70 patients with subcutaneous and intra-muscular tumours were all treated by local surgery, and a wide margin, requiring a cuff of fat tissue around the tumour and inclusion of the deep fascia beneath the tumour, was obtained in 56. These patients were followed without postoperative radiation. During a median follow-up of 5 years (range, 3.5–10 years), only four had a local recurrence, despite the fact that 84 per cent had high-grade tumours. The authors concluded that postoperative radiation may not be necessary in this subgroup. In another study, 74 patients with localized STS of the extremity or trunk underwent function-sparing surgery without radiation.217* The overall 10-year actuarial local control rate was 93 per cent, and was dependent on the adequacy of surgical margins (87 versus 100 per cent for patients with margins of 1 cm and 1 cm, respectively). The 10-year survival rate was 73 per cent. This approach may be appropriate for carefully selected patients with small (5 cm), superficial tumours that can be resected with all margins 1 cm.
Combining surgery with radiation therapy Although wide resection or amputation is often effective in achieving local control, there is usually a significant
Management of the extremity, trunk, or head and neck primary tumour 941
1.0
Cumulative frequency
0.8
0.6
0.4 U. Chicago – STS (15) MGH – STS (6)
0.2
U. Chicago – SCC (20)
0.0 0.0
0.2
0.4
0.6
0.8
1.0
SF2
Figure 40.5 Cumulative distribution of measured SF2 values for cell lines derived from human soft tissue sarcoma (STS) and squamous cell carcinoma (SCC) of head and neck region. SF2, proportion of cells which survive a dose of 2 Gy; MGH, Massachusetts General Hospital.
functional and cosmetic loss. Despite careful preoperative evaluation, a portion of the resected specimen will be found to have unsatisfactory margins and mean that additional treatment is essential. The consequence is that those patients experience both radical surgery and radiation. Until the early 1960s the prevailing medical opinion was that the sarcomas were radiation resistant and that radiation had little to offer these patients.218 This opinion was based upon the observation that patients treated by radiation during the 1930s and 1940s fared poorly. This was due to the fact that treatment was by radiation alone, tumours were large, radiation doses were low due to the availability of X-rays of energy of 300 kVp, and the clinicians expected a prompt regression. More recent clinical experience and laboratory research has shown without ambiguity that radiotherapy combined with surgery is a highly effective modality (see below). Further, radiation sensitivity of cell lines from human sarcomas is comparable to that of epithelial tumours as determined by in vitro assays using colony-forming ability as the end point for cell viability. This is illustrated by Figure 40.5 which shows the cumulative distribution of the SF2 (proportion of cells which survive a dose of 2 Gy) for sarcoma cell lines to be no less favourable than that for squamous cell carcinoma cell lines. The available evidence indicates that the tumour control probability for a soft tissue sarcoma and an epithelial tumour of the same size and treated to the same radiation dose are similar. For example, the 5-year local control results of treatment of a patient with carcinoma of the breast or soft tissue sarcoma by resection with negative margins and radiation
are essentially the same (approximately 0.9), despite the latter tumour being much the larger. There is experience with treatment with radiation therapy alone for unresected sarcoma, discussed below, that indicates that radiation is able to effectively control a proportion of these lesions. In animal models, a significantly lower radiation dose is required to achieve local control when radiation is combined with simple excision as compared to radiation alone, which provides the laboratory basis for the combination of conservative resection and radiation in the clinic.219 Preoperative radiotherapy in the treatment of patients with sarcoma of soft tissue is not new. Atkinson et al. reported in 1963 on 15 patients with operable sarcoma of soft tissue who were treated by preoperative radiotherapy (45 Gy in 4–5 weeks) followed 4–6 weeks later by ‘block resection’.220* The tumours were 4–20 cm in size; 10 were recurrent lesions. There had been one local failure and no distant metastases in these patients. Median follow-up was 3 years and 5 months. In the same clinic there had been 40 local recurrences following block resections alone performed on 54 patients with ‘comparable’ lesions. Barkley et al. have reported on 110 patients with locally advanced lesions treated by preoperative irradiation and resection; their local control rate was 90 per cent.221* The rationale for utilizing radiotherapy with a conservative surgical procedure is that radiotherapy in less than radical doses eradicates the small number of tumour cells remaining after a less extensive excision, i.e. those which would be removed in the grossly normal tissue included in the radical surgical specimen but not in that from the
942 Soft tissue sarcomas
Table 40.8 Local control results in patients treated by surgery and radiation Centre
No. of patients
Local failure %
Reference
176 300 89 53 128 23 35 64 67 50 88 818
14 22 14 14 10 9 14 8 13 24 8 18
Spiro et al. 1997443 Lindberg et al. 1981189 Abbatucci et al. 1986444 Karakousis et al. 1986192 Potter et al. 19854 Wilson et al. 1994445 Pao and Pilepich 1990243 Keus et al. 1994446 Fein et al. 1995241 Mundt et al. 1995237 O’Sullivan et al. 2002225 Zagars et al. 2003206
181 110 39 58 94 286
10 10 3 9 7 15
Spiro et al. 1997443 Barkley et al. 1988221 Wilson et al. 1999447 Brant et al. 1990191 O’Sullivan et al. 2002225 Zagars et al. 2003206
55 63 371 55
18 8* 10 2
Harrison et al. 1993193 Schray et al. 1990222 Eilber et al. 1993448 Wanebo et al. 1995255
POSTOP RADIATION Massachusetts General Hospital MD Anderson Hospital Institut Gustav Roussy Roswell Park Memorial Institute US National Cancer Institute Toronto St. Louis Amsterdam Penn/Fox Chase University of Chicago NCI Canada MD Anderson Hospital PREOP RADIATION Massachusetts General Hospital MD Anderson Hospital Toronto University of Florida NCI Canada MD Anderson Hospital INTRA-OP BRT Memorial Mayo IA/IV ADRIA RADIATION UCLA University of Virginia * Mean follow-up was 20 months.
simple excision. Thus, moderate radiation doses can be expected to provide the gain over simple surgery that has been shown for radical surgery. An attractive feature of radiotherapy is the relative ease with which the treatment volume can be designed to include tissues suspected of involvement without concern for the position of nerves, vessels and tendons. Results from several clinical studies demonstrate that this approach is clinically practical, and has achieved local control and survival results equal to those obtained by radical surgery with the important gain in cosmetic and functional results (see below). The 5-year results of treatment of patients with soft tissue sarcoma of the extremities by radiotherapy and conservative surgery are presented in Table 40.8. These results document that long-term local control is achieved in 76–98 per cent of patients. This is the equivalent of the local control rates following modern, radical surgical procedures. For example, in the phase III trial of brachytherapy at the Memorial Hospital the local control result in the surgery alone group of 70 patients was 67 per cent.193** That interstitial therapy is an effective modality in the treatment of these patients is shown by that trial. Five-year actuarial local control rates were 82 per cent and 67 per cent for the surgery brachytherapy and the surgery alone arms
respectively. Likewise, high local control rates for brachytherapy have been obtained at the Mayo Clinic.222* Patients whose sarcomas are so extensive that even a conservative resection would result in a minimally useful limb would be best served, in most instances, by a prompt amputation. For example, the elderly patient with a large sarcoma on the ankle and who has poor vascular supply is a poor candidate for an attempted limb salvage procedure. At the opposite extreme are patients with small sarcomas (5 cm) of the subcutaneous tissue or other anatomical sites which can be resected with wide margins and negligible functional consequences. They need only surgery as noted above. For the more frequent deeply sited lesions there is usually a close margin at some point on the resected specimen and a combined radiation and surgical approach is required to realize a very low likelihood of local failure. Hence, the recommended treatment for patients with extremity, truncal and head and neck soft tissue sarcomas who are medically and technically operable is the combination of function-preserving surgery and radiation, with the exception of that minority of patients with small, superficial lesions that can be widely excised with secure margins and good functional result. In most instances, the probability of tumour control and the late functional and
Management of the extremity, trunk, or head and neck primary tumour 943
cosmetic result is clearly superior following this combined modality approach. The impact of combined modality treatment that includes external beam radiation (EBRT) on both local control and survival has been evaluated in only one prospective randomized trial. In this study, 91 patients with high-grade lesions were randomly assigned to surgery plus postoperative chemotherapy with or without postoperative adjuvant EBRT, and 50 with low-grade lesions were randomized to surgery plus adjuvant EBRT or surgery alone.223** In the patients with high-grade lesions, there were no LRs in the patients randomized to EBRT, while the patients receiving only adjuvant chemotherapy had a 22 per cent actuarial local failure rate at 10 years. In patients with low-grade sarcoma, the LR rates were 4 versus 33 per cent in the postoperative EBRT and surgery alone groups, respectively. There was no influence of postoperative radiation on OS for either high- or low-grade tumours.
there is a modest increase in the acute reaction as manifest by a delay in the healing of the surgical wound. Radiation doses are lower for preoperative radiation therapy (50 Gy pre-op vs. 60–66 Gy post-op), which can reduce late effects. Preoperative radiotherapy facilitates the development of an overall treatment plan by the surgeon, radiation therapist and medical/paediatric oncologist before any therapeutic manoeuvre is implemented. Initiation of radiotherapy is not delayed when given preoperatively. Where radiotherapy is given postoperatively, delays of 10–14 days or even longer may occur before the treatment is started. This means that residual tumour cells are in a tumour bed flooded with growth factors and, hence, have an opportunity to increase in number. Where wound healing delays treatment further, recurrent tumour may be grossly evident at the time of initiation of irradiation (especially where surgery was not complete and the lesion was high grade).
Preoperative (neoadjuvant) versus postoperative (adjuvant) radiotherapy
POSTOPERATIVE RADIATION THERAPY
There are potential advantages to both preoperative and postoperative administration of radiation.
1. Histopathological diagnosis/grade is made from tissue samples taken throughout the entire tumour rather than an incisional biopsy specimen. 2. Surgery is immediate. This is important to some patients. 3. There is no delay in wound healing caused by prior radiation therapy. 4. Initial surgery is the only feasible sequence for quite small lesions removed by excisional biopsy, often judged to be benign. This has rarely led to problems other than the occasionally poorly placed incision by a non-oncological surgeon.
PREOPERATIVE RADIATION THERAPY
Preoperative radiation therapy might be expected to reduce tumour burden prior to resection, theoretically allowing more conservative surgical therapy. At the time of surgery, the tumour is usually smaller and surrounded by a relatively dense pseudo-capsule if radiotherapy has been given preoperatively. This means that the margins can be reduced and, hence, a more conservative approach is feasible. This should be associated with lesser disability. In addition, a lesion previously considered inoperable may regress to such an extent that it becomes operable. Radiation given preoperatively reduces the number of viable tumour cells to such small absolute levels that the likelihood of autotransplantation in the surgical bed is virtually eliminated. Preoperative radiation fields can be limited to the known gross tumour and adjacent tissues at risk for microscopic infiltration. In contrast, postoperative irradiation fields must include not only the site of the tumour but also all tissues handled during the surgical procedure including the stab wound for the drain tube(s). Therefore, the treatment volume for postoperative radiotherapy will usually be larger than for preoperative irradiation. Nielsen et al.224* reported a prospective study of radiation field size in a series of 26 patients who were planned for radiation to be given preoperatively and then following surgery were re-planned: the field sizes were 241 cm2 and 391 cm2 respectively. The consequence of the smaller treatment volumes means that late radiation reactions are expected to be less severe in patients who receive radiation preoperatively. However,
Postoperative radiotherapy has the following advantages:
There is one randomized controlled study comparing preoperative and postoperative radiotherapy. This study was designed to evaluate the incidence of acute wound healing complications in patients with potentially curable extremity soft tissue sarcoma.225** In this Canadian trial, 190 patients were randomly assigned to either preoperative (50 Gy pre-op for all 94 patients randomized to this arm with 16–20 Gy post-op boost reserved for the 14 patients in this arm with a positive margin) or postoperative (50 Gy initial field 16–20 Gy boost field for all patients) radiotherapy. Complications were defined as secondary wound surgery, hospital admission for wound care, or the need for deep packing or prolonged wound dressings within 120 days of tumour resection. The study was terminated when a highly significant result was obtained at the time of a planned interim analysis. With a median follow-up of 3.3 years, a significantly higher percentage of preoperatively treated patients had acute wound complications (35 versus 17 per cent). Other factors associated with wound complications were the volume of resected tissue and lower limb location of the tumour
944 Soft tissue sarcomas
Table 40.9 Risk of acute wound healing complications in patients with extremity sarcomas randomized to receive preoperative or postoperative radiation therapy in NCI Canada study Wound complication by anatomical site Upper arm
Preoperative XRT Postoperative XRT (n 88) (n 94) No 9 (90%) Yes 1 (10%) No 8 (100%) Yes 0 No 24 (55%) Yes 20 (45%) No 16 (62%) Yes 10 (38%)
Lower arm Upper leg Lower leg
11 (100%) 0 8 (100%) 0 39 (72%) 15 (28%) 20 (95%) 1 (5%)
Adapted from O’Sullivan et al. 2002.225
Table 40.10 Risk of late treatment-related complications in patients with extremity sarcomas randomized to receive preoperative or postoperative radiation therapy in NCI Canada study Late radiation toxicity by treatment arm
Preoperative radiotherapy (n 73)
Postoperative radiotherapy (n 56)
was significantly more common in the postoperative group, 36 per cent vs. 23 per cent, p 0.02.227 There is thus a difference in the morbidity profile between preoperative and postoperative radiotherapy, with a higher rate of generally reversible acute wound-healing complications in the patients receiving preoperative treatment, offset by a higher rate of generally irreversible late complications, including grade 3–4 fibrosis, in those patients receiving postoperative radiation therapy. Because very few acute wound-healing complications occurred in either group when the tumour was in the upper extremity, it would seem prudent to treat these patients with preoperative radiation therapy. We have also favoured preoperative radiation for the majority of lower extremity patients, because acute wound complications can usually be managed and will go on to heal, whereas the late treatment effects are generally irreversible. For patients with lower extremity lesions, this study, however, makes it clear that new strategies are needed: first, to reduce the risk of acute wound healing problems when patients receive preoperative radiation therapy and, second, to reduce the risk of late treatmentinduced effects when higher-dose, larger-field postoperative radiation is given.
Brachytherapy Subcutaneous fibrosis Joint stiffness Oedema
grade 2 grade 2 grade 2 grade 2 grade 2 grade 2
68.5% 31.5% 82.2% 17.8% 84.9% 15.1%
51.8% 48.2% 76.8% 23.2% 76.8% 23.2%
Adapted from Davis et al. 2005.449
(Table 40.9). Late morbidity was initially not reported. Because the radiation therapy fields for the postoperative radiation therapy were larger and the dose delivered for most patients was higher, the authors indicated that more follow-up would be needed to assess whether these larger radiation volumes and higher radiation doses would lead to more late treatment effects in these patients. In a later publication, the local recurrence rate, regional or distant failure rate, progression-free survival and functional outcome did not differ between the groups.226 These data have now been updated with a median follow-up of 6.9 years. There remain no differences in local control between the patients in the two arms of the study with over 90 per cent local control. The regional and distant failure rates as well as the progression-free and overall survival rates are also no different between the two arms of the study. The postoperative patients, however, have now developed more grade 2–4 late toxicity (86 per cent) when compared to the preoperative patients (68 per cent), p 0.02 (Table 40.10). Notably, grade 3 (severe induration and loss of subcutaneous tissue or field contracture 10 per cent linear measurement) or grade 4 (necrosis) subcutaneous fibrosis
Compared to EBRT, brachytherapy (BRT) minimizes the radiation dose to surrounding normal tissues, maximizes the radiation dose delivered to the tumour, and shortens treatment times. In the usual dosage schedule, treatment is completed within 6 days and requires only one hospitalization. Afterloading catheters are placed in a target area of the tumour operative bed, defined by the surgeon, and spaced at 1 cm intervals to cover the entire area of risk (Fig. 40.6). Brachytherapy can also be used for delivery of a boost to the tumour bed in conjunction with external beam radiation therapy.222 A phase III trial of postoperative BRT versus no BRT was conducted in 126 patients who had complete resection of either extremity or superficial trunk STS.193** The BRT dose was 45 Gy. Five-year local control rates were 82 and 67 per cent for the brachytherapy and surgery alone groups, respectively. The advantage of BRT was seen only in the high-grade sarcomas.193,228 It was limited to local control, since there was no difference between the groups in distant metastasis or disease-specific survival.193 Although it is unclear if brachytherapy is associated with a higher risk of wound complications (see Wound healing after surgery and radiation, below), the rate of wound re-operation may be higher.229 BRT has been combined with free flap construction as a means of enhancing primary healing in difficult anatomical situations without an increase in the incidence of wound breakdown.230 There have been no randomized comparisons of the relative efficacy or morbidity of external beam radiation therapy compared with brachytherapy.
Management of the extremity, trunk, or head and neck primary tumour 945
(a)
(b)
Figure 40.6 This boy with a recurrent epithelioid sarcoma on the hypothenar eminence of his left hand following an excisional biopsy one year earlier (lesion mistakenly called a benign fibrous histiocytoma at that time by an inexperienced pathologist) was treated at age 11 with a combination of excision (Figure 6a), 45 Gy of low-dose rate iridium-192 brachytherapy (Figure 6b), and 20 Gy delivered following the implant with 6 MeV electrons. He is free of any evidence of disease with normal hand function at 5 years.
Brachytherapy for sarcomas has traditionally been given by low dose rate radiation. There is some preliminary information on the use of fractionated high-dose rate schedules.231*,232*,233* High dose rate brachytherapy has used in conjunction with external beam radiotherapy for the tumour bed boost in doses of 15–24 Gy, often hyperfractionated at 2.3–4 Gy bid.232 One report using high dose rate brachytherapy alone in doses of 40 Gy at 2.3–3 Gy bid unfortunately reported poor local control of only 20 per cent, in contrast with 100 per cent when brachytherapy was combined with external beam radiation.233*
External beam radiation treatment planning The radiation treatment technique should be carefully planned so that the tissues being irradiated are only those judged to be at risk. In order to utilize smaller treatment volumes, the part to be irradiated must be securely and reproducibly immobilized. We have special immobilization devices prepared for the individual patient. This may require casting, especially for hand, foot or elbow sites (Fig. 40.7). For some sites the part is placed in standard plastic supports and the extremity fastened tightly in place using a Velcro fastener. Others describe their experience with casts and polyurethane foam systems.234 The principal tasks involved in the development of a treatment plan are: ●
●
Design an immobilization device and a means to assure that the target is on the beam. It is important that the patient be comfortable in this position so that it can be maintained for the treatment delivery sessions. CT scan (or where MRI simulators are available, perform MRI scan) of the immobilized, affected extremity
Figure 40.7 Customized immobilization device for use in a patient with a soft tissue tumour involving the left foot.
for radiotherapy planning. This is facilitated by the availability of a large-bore scanner that allows maximum flexibility in arranging the limb such that the contralateral extremity and the trunk will be out of the beam.
946 Soft tissue sarcomas
(a)
(b)
(c)
Figure 40.8 Axial (a), coronal (b), and sagittal (c) radiation treatment plans in a patient undergoing preoperative radiation therapy for treatment of a large, T2b, grade 3/3 malignant fibrous histiocytoma of the left leg. (Courtesy of Brian Napolitano, Massachusetts General Hospital.) ●
●
●
●
●
●
Define the target volume(s) in three dimensions on the CT/MRI of the affected region. Review of these studies with the diagnostic radiologist and surgeon, as well as review of any surgical procedures performed with the surgeon, can be extremely important in defining the appropriate target volume(s). Image fusion of MRI scans with the treatment planning CT scan may facilitate target definition. Define non-target critical structures in the treatment volume and specify dose constraints for each such structure. Estimate the distribution of number of tumour clonogens/ unit volume of tissue throughout the target volume. Define a series of target volumes to realize the appropriate dose distribution using ‘shrinking treatment volume methods’. Design treatment techniques that achieve the closest feasible conformation of treatment to target volume. This may require complex field arrangements, treatment angles, gapped fields, wedge filters, tissue compensators, bolus, various radiation modalities (i.e. electrons,
● ●
●
●
protons, intensity-modulated radiotherapy [IMRT], brachytherapy, intra-operative radiation therapy). Avoid inclusion of an entire joint space. Avoid full-dose irradiation of adjacent bone to reduce the risk of pathological fracture. Utilize wedges and tissue compensators as needed to account for tissue heterogeneities and minimize dose inhomogeneity. Review the treatment plan at multiple levels along the extremity to assess dose homogeneity to the target and normal tissues (Fig. 40.8).
A valuable point to keep in mind in planning treatment is that there may be a failure (recurrence or necrosis) and the plan should, if feasible, allow for potential sites for flaps, surgical incision, etc., if salvage surgery is necessary.
Radiation treatment volumes and dose The extent of normal tissue to be irradiated adjacent to the tumour bed in the case of preoperative radiation therapy
Management of the extremity, trunk, or head and neck primary tumour 947
and adjacent to the surgical bed in the case of postoperative radiation therapy is not definitively known. Few patterns of failure studies to relate the extent of the radiation therapy field to the site of local tumour recurrence have been reported. Because sarcomas are judged to infiltrate along rather than through tissue planes, longitudinal margins proximally and distally have traditionally been considerably more generous than radial margins. Historically, fields that extended from the muscle origin to insertion235 or provided generous proximal/distal margins on the tumour were employed.236 In some centres from the 1970s through the mid 1990s, 5–10 cm proximal and distal block margins were used for large grade 1 and small grade 2 lesions and more generous fields with 10–15 cm margins were encompassed for large grade 2–3 lesions.236 The advent of improved MRI delineation of tumour extent and subsequent surgical experience with high rates of local control when surgical margins 1 cm could be obtained prompted radiation oncologists to employ 5 cm proximal/distal margins for small grade 1 lesions and 5–7 cm proximal/distal margins for larger, higher-grade lesions. Newer three-dimensional (3D) treatment planning systems appear to allow smaller and more accurate treatment volumes in patients with extremity STS. There are very few studies in the literature looking at the target volume used when planning radiotherapy. This tends to be poorly reported. One group found a remarkable difference in 5-year local control where the margin was 5 cm (30 per cent) or 5 cm (93 per cent).237* This conflicts with the brachytherapy data where acceptable results are achieved using 4 cm margins longitudinally and 2 cm laterally. A recent publication from the Royal Marsden Hospital has suggested that as in other tumour sites the great majority of local recurrences occur within the highdose volume.238* This raises the question as to whether the large-volume phase 1 is necessary or whether the boost is necessary where an adequate surgical margin has been achieved. This question is particularly relevant now because of the advent of techniques such as IMRT and protons that allow us to selectively spare normal tissues. It will be important to determine the volumes that can safely be spared before implementing these techniques. Such studies are currently being considered by cooperative groups in the USA and Europe. One recent paper presented some very provocative findings that will need to be very carefully considered in the discussion of radiation therapy volumes. This study, performed by investigators in Toronto, provided a histological assessment of peritumoural oedema as demonstrated by increased T2-weighted signal intensity on MRI scans performed preoperatively on 15 patients with high-grade extremity or truncal sarcomas ranging in size from 3.1 to 30.1 cm (mean, 13.8 cm) who did not undergo any neoadjuvant therapy prior to resection.239* The extent of peritumoural T2-weighted signal intensity changes beyond the tumour ranged from 0 to 7.1 cm (mean, 2.5 cm); contrast enhancement ranged from 0 to 5.3 cm (mean 1.1 cm).
Tumour cells were identified histologically in the tissues beyond the tumour in 10 of 15 cases. In six cases, the tumour cells were located within 1 cm of the tumour margin, and in four cases, malignant cells were found at a distance 1 cm and up to 4 cm. The location of the tumour cells did not correlate with tumour size or extent of peritumoural changes on the MRI scans. In nine of 10 cases, however, the tumour cells were identified histologically in areas with corresponding high T2-weighted signal changes on MRI. These T2weighted signal changes were more prominent in the proximal–distal dimensions than in the radial direction, consistent with radiation treatment guidelines that have used larger radiation field margins in the proximal–distal direction than the radial margins. With ever-increasing ability of the available radiation oncology technology to conform the radiation dose to the target, this study has very significant implications for radiotherapy target design and must be considered in future studies of radiotherapy volumes in this disease. The radial margins should be viewed with respect to the direction of most likely spread; the beam edges can be designed to deliver the prescription dose to a clinical target volume consisting of a margin of 1 cm radial to the gross tumour volume with approximately 0.5 cm for daily set-up variation. This 1 cm radial margin is rationally derived from the surgical experience with high rates of local control with 1 cm margins. Because of the penumbra of the beam, this usually means that the radial block edge is approximately 2 cm from the edge of gross tumour. Where there is intervening bone, interosseous membrane or major fascial planes, and these planes are intact in the imaging studies, the full prescription dose can be delivered to the surface of these structures which approximate the tumour, again allowing for daily set-up variation. When a fascial plane has been violated, wider margins are appropriate to cover areas of potential contamination by tumour. The use of 5 cm proximal and distal block margins for the first 50 Gy (on the tumour for preoperative radiotherapy and the surgical bed for postoperative radiotherapy) and 2 cm radial block margins provided very high rates of local control in the randomized NCI Canada trial discussed above. For patients receiving preoperative radiation, 50 Gy is administered over 5 weeks, followed 3–5 weeks later by a conservative resection. With negative surgical margins and no other unfavourable prognostic features such as tumour cut-through or satellite lesions after prior surgical interventions, 50 Gy of preoperative radiation therapy appears sufficient to provide local control in a very high proportion of patients. Sadoski et al.202 analysed 132 consecutive patients with soft tissue sarcoma of the extremities treated with preoperative radiotherapy and resectional surgery and found that: 1. The 5-year actuarial local control rates were 97 per cent and 81 per cent for patients with negative margins and positive margins, respectively (this difference is highly significant).
948 Soft tissue sarcomas
2. There was no difference between the various subcategories of negative margins: negative at 1 mm (96 per cent); negative at 1 mm (97 per cent); not measured (94 per cent); and no tumour in the specimen (100 per cent). 3. There was no difference in local control for treatment of primary and locally recurrent lesions (after previous surgery alone) when the tumours were stratified for margin status. 4. For the patients with negative margins, local control was not a function of sarcoma size. For patients with positive margins following preoperative radiation therapy, it is recommended to use a ‘shrinking treatment volume technique’ with delivery of either brachytherapy or a postoperative external beam radiation boost dose of 16–18 Gy to the tumour bed once the surgical wound has healed. A boost dose to 66–68 Gy is given postoperatively or intra-operatively for microscopically positive margins, and to 75 Gy if there is gross residual disease. In patients with frozen section evidence of close or positive margins, an intra-operative boost dose can be administered by brachytherapy (BRT) or electron beam. BRT has the advantage of fractionation and one can use a low dose rate of 16 Gy for microscopically positive margins (or more recently, high dose rate of 12–16 Gy given at 3–4 Gy bid), and 25 Gy for gross residual tumour. For patients undergoing postoperative radiation therapy, irradiation usually begins 14–20 days following surgery, once the wound is healed. Following resection of large tumours, it may be necessary to wait 3–4 weeks to allow resorption of the seroma. The initial radiation target volume must include all tissues handled during the surgical procedure, including the drain site, often encompassing the surgical bed with 5 cm proximal/distal block margins and 2 cm radial block margins. The dose to this initial volume is 50 Gy and then, through shrinking treatment volumes to encompass the tumour bed, the final dose is 60 Gy for negative margins, 66–68 Gy for positive margins or locally recurrent disease,240 and 75 Gy for gross residual sarcoma. The available information on a dose–response relationship for the local control of sarcomas treated with surgery and postoperative radiation therapy is somewhat conflicting. Mundt et al. reported that local control was dose dependent. While postoperative patients receiving 60 Gy had lower local control than those receiving 60 Gy, no difference was seen in local control between patients receiving 60–63.9 Gy (74.4 per cent) vs. those receiving 64–66 (87.0 per cent) (p 0.5). Severe late sequelae were more frequent in patients treated with doses 63 Gy compared to patients treated with lower doses (23.1 per cent vs. 0 per cent).237* Fein et al. noted that patients receiving 62.5 Gy had 5-year local control of 78 per cent vs. 95 per cent where the dose was 62.5 Gy.241* In a multivariate analysis of patients undergoing postoperative radiotherapy, Zagars and
Ballo identified dose as an independent variable for local control.240* Doses of 64 Gy correlated with improved local control. Recognizing that the effectiveness of a particular dose was also related to other factors influencing local control such as margin status, anatomical site and locally recurrent presentation, they recommended postoperative doses of 60 Gy for patients with negative margins and otherwise favourable prognostic features, while suggesting increasing doses for less favourable presentations, up to doses of 68 Gy for positive margins. In contrast, other investigators have not been able to demonstrate a clear dose–response relationship in their reviews of their patients undergoing postoperative radiation therapy.196*,242*,243*,244* In practice, most centres give 60 Gy of postoperative radiotherapy for patients with negative margins and 66–68 Gy for positive margins, using shrinking fields as described above. With regards to preoperative radiation therapy, Robinson et al. failed to demonstrate a variation in local control according to dose although the response rate to preoperative radiotherapy was clearly dose dependent.245* The accepted dose for preoperative radiotherapy is 50 Gy. Radiation doses may need to be modified for patients with diabetes or connective tissue diseases, and in those receiving chemotherapy. For treatment of an extremity lesion, a good functional result demands that only a portion of the cross-section of the extremity be irradiated to any worthwhile dose level.246 Thus, some tissue should not be irradiated to provide for lymphatic drainage. For very large tumours that are treated with wide resection, there may be persistent leg oedema, requiring the use of a pressure-type stocking, even though the radiation treatment volume is less than circumferential. This can be a problem for patients with large (10 cm) sarcomas of the medial thigh. When postoperative radiation is combined with adjuvant chemotherapy, radiation daily dose has generally been reduced by 10 per cent from 200 to 180 cGy; radiation is not given concurrently with doxorubicin. Instead, 2–3 days are allowed between the doxorubicin and radiation. Some preoperative protocols have interdigitated chemotherapy and radiation therapy (see below); total preoperative radiation has been reduced (i.e. 44 Gy) in this setting.
Intensity-modulated photon radiation therapy (IMRT) The purpose of radiation therapy is to maximize the dose delivered to the tumour while minimizing the exposure of dose-sensitive critical structures to high dose. This has been achieved traditionally by shaping the spatial distribution of the high radiation dose to conform the target volume (hence, 3D conformal radiation therapy), thereby reducing the dose to the non-target structures. Although this approach is satisfactory in the treatment of targets that are roughly convex in shape, it is less than optimal for
Management of the extremity, trunk, or head and neck primary tumour 949
(a)
(b)
Figure 40.9 Axial (a), coronal (b), and sagittal (c) radiation treatment plans in a patient undergoing neoadjuvant, interdigitated chemotherapy and preoperative IMRT radiation therapy for treatment of a 25 cm, T2b, grade 3/3 malignant fibrous histiocytoma of the left thigh. IMRT was used to minimize dose to the perineal tissues in this patient with a large, proximal thigh lesion. (Courtesy of David Gierga, Ph.D., Massachusetts General Hospital.)
targets that contain complex concavities or that wrap around critical structures.247 Growing experience suggests that IMRT plans produce superior dose distribution to the patient as compared to 3D conformal plans, both in terms of dose conformity in the tumour and dose reduction to the specified critical normal structures, albeit at the cost of increased integral dose to the normal tissues (Fig. 40.9). Recent dosimetric studies comparing IMRT and conformal radiotherapy for soft tissue sarcoma have been reported. When evaluating sarcomas arising in the extremities, pelvis, trunk and paranasal sinuses, IMRT plans were more conformal. In the extremities, bone and subcutaneous doses were reduced by up to 20 per cent. A conformalIMRT comparative planning study has been reported for a large extra-skeletal chondrosarcoma of the extremity.248 Not surprisingly, IMRT produced excellent conformal treatment plans for this complex target volume, with a reduction of the maximum dose to the bone as compared to the 3D-photon plan. Hong et al. performed treatment planning comparisons of IMRT and 3D conformal radiotherapy for 10 patients with soft tissue sarcoma of the thigh.249* They were able to document a reduction in femur dose without compromise in tumour coverage. In addition, IMRT reduced hot spots in the surrounding soft tissues and skin. It is worth noting, however, that IMRT treatment plans often have localized areas within the high-dose volumes where dose inhomogeneities can be in the range of 10–15 per cent above the prescription dose. Because there can also be dose inhomogeneities in the range of 5 per cent below the target dose, treatment plans may be normalized to the 95 per cent isodose line, meaning that selected areas of the treatment volume are receiving daily fractions and total doses of 15–20 per cent above the target dose. Depending upon the location of these ‘hot spots’, there can be unanticipated acute normal tissue toxicity.250 Because of the
multiple field angles employed with IMRT, integral doses to the extremity will probably also be higher with IMRT, meaning that more of the extremity will see some radiation dose, albeit relatively lower levels. Whether there are late effects attributable to these focal areas of high dose or the higher integral dose remains to be seen.
Proton beam radiotherapy The rationale for the use of protons (or other charged particles) rather than photons (i.e. X-rays, which have traditionally been used for radiotherapy) is the superior dose distribution which can be achieved with protons. Protons and other charged particles deposit little energy in tissue until near the end of the proton range where the residual energy is lost over a short distance, resulting in a steep rise in the absorbed dose, known as the Bragg peak (Fig. 40.10).251,252 The Bragg peak is too narrow for practical clinical applications, so for the irradiation of most tumours, the beam energy is modulated by superimposing several Bragg peaks of descending energies (ranges) and weights to create a region of uniform dose over the depth of the target; these extended regions of uniform dose are called ‘spread-out Bragg peaks’ (SOBP). Although the beam modulation to spread out the Bragg peaks does increase the entrance dose, the proton dose distribution is still characterized by a lower-dose region in normal tissue proximal to the tumour, a uniform high-dose region in the tumour, and zero dose beyond the tumour. Protons have been extensively employed for bony sarcomas of the skull base and spine; more recently they have been used for treatment of soft tissue tumour of the paraspinal tissues and there are clearly opportunities to employ protons with very significant sparing of normal tissues in selected patients with extremity, truncal and
950 Soft tissue sarcomas
100
SOBP
Dose / %
80 10 MV Photon 60 Pristine Peak 40
20
50
100
150
Depth / mm
Figure 40.10 Depth–dose distributions for a spread-out Bragg peak (SOBP, red), its constituent pristine Bragg peaks (blue), and a 10 MV photon beam (black). The SOBP dose distribution is created by adding the contributions of individually modulated pristine Bragg peaks. The penetration depth, or range, measured as the depth of the distal 90% of plateau dose of the SOBP dose distribution, is determined by the range of the most distal pristine peak (labelled ‘Pristine Peak’). The dashed lines (black) indicate the clinically acceptable variation in the plateau dose of 2%. The dot-dashed lines (green) indicate the 90% dose and spatial, range and modulation width, intervals. The SOBP dose distribution of even a single field can provide complete target volume coverage in depth and lateral dimensions, in sharp contrast to a single photon dose distribution; only a composite set of photon fields can deliver a clinical target dose distribution. Note the absence of dose beyond the distal fall-off edge of the SOBP. (Courtesy of Hanne Kooy, PhD., Massachusetts General Hospital.)
(a)
(b)
Figure 40.11 Axial (a) and sagittal (b) proton dose distribution for a 37 year-old female with a high-grade, T2b sarcoma of the left proximal, posterior thigh, managed with three cycles of neoadjuvant, interdigitated MAID chemotherapy and radiotherapy. The use of protons allowed sparing of the femur and soft tissues of the perineum which were not in the clinical target volume (CTV). The use of IMRT would have markedly increased the integral dose received by the patient. (Courtesy of Judy Adams, C.M.D., Massachusetts General Hospital.)
retroperitoneal sarcomas.253 Large, medial proximal thigh lesions can be effectively treated with sparing of the femur, hip joint, genitalia and anorectal tissue (Fig. 40.11). Lesions around the shoulder can be treated without irradiating the
lung apex, and avoiding the shoulder. With the recent or anticipated completion of proton beam facilities in major sarcoma centres in the USA (Massachusetts General Hospital, MD Anderson Cancer Centre, and University of Florida),
Management of the extremity, trunk, or head and neck primary tumour 951
it is anticipated that a larger proportion of these patients will be treated with protons.
Neoadjuvant doxorubicin-based chemotherapy plus radiation therapy The UCLA group popularized preoperative regional chemotherapy and radiation therapy followed by limb salvage surgery in patients with high-grade sarcomas.254* They were able to achieve a high rate of primary limb salvage, low rate of local recurrence (approximately 9 per cent) and long-term survival in 65 per cent of patients. The current regimen consists of doxorubicin (30 mg per day for 3 days) followed by radiation given at 28 Gy in 8 fractions. A number of other groups have utilized this regimen, also obtaining low rates of local recurrence with varying degrees of toxicity. As an example, the Southeastern Cancer Study Group evaluated this protocol in 66 patients with nonmetastatic high-grade extremity sarcoma who received intra-arterial doxorubicin infused directly into the vessel feeding the tumour (30 mg per 24 hours for 3 days).255* Concurrent radiation therapy was administered (30 Gy in 10 fractions, 35 Gy in 10 fractions, or 46 Gy in 23–25 fractions). Limb-sparing surgery was possible in 60 of 66 patients; an additional two patients required amputation due to wound healing complications. Five-year survival and disease-free survival were 59 and 44 per cent, respectively. The local failure rate was 1.5 per cent. It is not clear that intra-arterial administration provides added benefit to intravenous doxorubicin. One study compared these two methods of administration.256* The intraarterial route was thought to be associated with a higher incidence of complications and no improvement in survival or function. Another report evaluated two separate protocols utilizing preoperative treatment with intravenous doxorubicin and ifosfamide with or without intraarterial cisplatin; the histological response and local failure rate following surgery were better with the all intravenous regimen.257* Combination chemotherapy regimens such as MAID (mesna, doxorubicin, ifosfamide and dacarbazine) may provide better anti-tumour activity than single agent doxorubicin. Interesting results have been noted with neoadjuvant MAID plus radiation therapy.258*,259* The experience with preoperative MAID chemotherapy interdigitated with 44 Gy radiation, and followed by surgery, postoperative MAID and further radiation (16 Gy) for those with positive margins was reported in a series of 48 patients with highgrade extremity sarcomas 8 cm.258 Despite the low objective response rate to preoperative therapy (partial response in 11 per cent and stable disease in 77 per cent), all patients were able to undergo limb-sparing surgery initially, with 15 per cent having positive margins. Median tumour necrosis was 95 per cent, suggesting that conventional imaging in this setting may underestimate the degree of response to
therapy. Twenty-five per cent of patients required hospitalization for febrile neutropenia at some time during treatment. Wound healing complications occurred in 14 of 48 MAID patients (29 per cent). One MAID patient developed late fatal myelodysplasia. The 5-year rates of local control (92 vs. 86 per cent), freedom from distant metastases (75 vs. 44 per cent), disease-free survival (70 vs. 42 per cent), and overall survival (87 vs. 58 per cent) all compared favourably with the outcomes of a cohort of historical control patients who were matched for tumour size, grade, patient age, and era of treatment. Similar results were noted when this regimen was utilized in a multi-centre United States cooperative group trial conducted by the Radiation Therapy Oncology Group, in which 66 patients with primary high-grade extremity or truncal soft tissue sarcomas 8 cm in diameter received a modified MAID regimen plus granulocyte colonystimulating factor and radiation preoperatively, followed by resection and postoperative chemotherapy.259* In this report, although preoperative chemotherapy and radiation was successfully completed by 79 and 89 per cent of patients respectively, grade 4 haematological and non-haematological toxicity was experienced by 80 and 23 per cent of patients. Delayed wound healing was noted in 31 per cent. With a median follow-up of 2.75 years, the estimated 3-year survival, disease-free survival and local control rates were 75, 55 and 79 per cent, respectively. Two patients developed late myelodysplasia. It remains to be confirmed in randomized studies if these aggressive interdigitated approaches offer benefit to the subgroup of patients with large, high-grade sarcomas who are at highest risk of treatment failure.
Isolated limb perfusion with tumour necrosis factor-alpha Hyperthermic isolated limb perfusion with chemotherapeutic agents has been tried as another way to control very large tumours that would otherwise require amputation because of proximity to nerve or blood vessels. Most such regimens do not appear superior to surgery plus radiotherapy. However, hyperthermic TNF-α plus melphalan has achieved complete response rates of approximately 30 per cent and partial remission rates of 50 per cent with overall limb-salvage rates over 80 per cent.260* This work was initiated by Lejeune et al. and has been extended to several other centres.261* TNF-α has long been known to be a potent non-specific tumour cell cytotoxic agent with demonstrated high efficacy in treatment of tumour-bearing rodents. Application in man has been severely restricted due to systemic toxicity. However, by the use of the isolated limb perfusion technique, the systemic toxicity has been effectively bypassed. Eggermont et al. reported the results from a multi-centre study of isolated limb perfusion with TNF-α, interferon-γ (IFN) and melphalan for extremity sarcomas.262* The complete response (CR) and
952 Soft tissue sarcomas
(a)
(b)
Figure 40.12 A 51-year-old right-handed composer and pianist with low-grade spindle cell neoplasm with features most suggestive of fibrosarcoma, initially excised with a positive margin, subsequently treated with tumour bed re-excision and low-dose rate Ir-192 brachytherapy (a) of 45 Gy over 4.5 days. She returned to playing the piano approximately 2 months later. Photos taken 5 months after her brachytherapy document excellent functional and cosmetic results, as seen in (b).
partial response (PR) rates were 18 per cent and 64 per cent; limb salvage was achieved in 84 per cent of patients. Of 39 patients who also had conservative resection, there have been five local failures. Vaglini et al. reported from Milan a high response rate using the same protocol.263* Olieman et al. reported the results of a series of 34 patients treated by this isolated limb perfusion with TNF-α, IFN and melphalan followed by resection and radiation in 15 patients.264 Limb salvage was realized in 85 per cent. Local recurrence developed in 26 per cent of patients treated by perfusion alone and in none of the perfusion–radiation patients. The two groups sustained approximately equivalent treatment-related morbidity. There is one report indicating an increase in severe morbidity when radiation is combined with the perfusion.265 The mechanism of action of the TNF-α is not fully understood. Its efficacy is reported to be reduced in hypoxic cells.266 Gutman et al. (1997) report a 37 per cent CR rate using TNF-α and melphalan.267* Even in previously radiated limbs this technique has been used to avoid amputation for local recurrence.268*
Soft tissue sarcomas of the hand and foot *The 5-year survival rate for sarcoma of the hand and foot is approximately 80 per cent, better than that usually given for extremity soft tissue sarcomas.269–271 This is likely related to the smaller size of these lesions at presentation. With surgical excision and the use of adjunctive radiotherapy when the minimum surgical margin is narrow (less than 2 mm), limb amputation can be avoided as primary therapy in most patients, and up to two thirds of patients can retain a normal or fairly normal extremity (Fig. 40.12).
Wound healing after surgery and radiation In general, the use of adjunctive radiation is associated with a higher frequency of wound complications. Quantitation of the impact of radiation on wound healing is difficult because of the significant complications that can be seen with surgery alone. In addition, there is much heterogeneity among patients with soft tissue sarcomas with respect to anatomical site, histological type, lesion size, prior surgery, medical status and age. The use of adjunctive radiotherapy can also be associated with joint stiffness, oedema, and decreased range of motion.223**,272** In one trial, extremity radiation resulted in significantly worse limb strength, oedema and range of motion compared to surgery alone for extremity STS, but the symptoms were transient and did not affect global quality of life.223
Preoperative radiation and wound complications Preoperative radiation is associated with a higher incidence of acute wound complications.225** In the randomized study of preoperative versus postoperative radiation therapy, a significantly higher percentage of preoperatively treated patients had acute wound complications (35 vs. 17 per cent). The difference in wound healing complications was confined to patients with lower extremity lesions. Wound healing complications were also more common in patients with tumours 10 cm. In another series of 202 patients undergoing preoperative radiation therapy, the overall wound complication rate was 37 per cent.272* One patient died with necrotizing fasciitis, and 33 patients (17 per cent) required secondary surgery including six
Adjuvant chemotheraphy 953
(3 per cent) who required amputation. In this series, multivariate analyses of the data showed that the following factors were significantly associated with wound morbidity: tumour in the lower extremity (p 0.001), increasing age (p 0.004) and postoperative boost with interstitial implant (p 0.016). Accelerated fractionation (twice daily) reached borderline statistical significance (p 0.074). Two other factors showed association with wound morbidity by univariate analysis but not in the multivariate model: high pathological grade (p 0.02) and estimated volume of resected specimen 200 cm3 (p 0.065). This was observed to be a problem, especially for the elderly obese patients with large sarcomas in the proximal thigh. In another report, wound morbidity was 25 per cent (four of 16) in patients treated with preoperative EBRT plus brachytherapy at the time of surgery but only 5 per cent (two of 40) in those treated postoperatively with EBRT plus BRT.222*
Brachytherapy and wound complications The use of perioperative brachytherapy (BRT) may increase the incidence of wound complications. In one study of 105 patients with extremity and truncal sarcomas, major wound complications occurred in nine of 41 (22 per cent) cases treated with BRT compared to two of 64 (3 per cent) non-BRT-treated patients.273* Patients treated with BRT also had a higher total number of complications as well as a higher combined frequency of major and moderate wound complications (44 vs. 14 per cent), and a longer time to wound healing (189 days vs. 49 days). However, the findings were different in a randomized trial of adjuvant BRT versus no BRT in 164 patients with resected extremity or truncal soft tissue sarcoma. The incidence of serious wound complications was not significantly increased in the group receiving BRT (24 vs. 14 per cent, p 0.133), but the incidence of wound re-operation was increased (6 vs. 0 per cent).229**
Strategies to reduce wound morbidity Based upon published experience, the following strategies are suggested to reduce acute wound morbidity in patients being treated with preoperative radiation272 or perioperative brachytherapy:229 ● ●
● ●
●
● ●
gentle handling of tissue during surgery; meticulous attention to achieving haemostasis before wound closure; avoidance of closure under tension; elimination of all wound dead space, using a rotated flap to fill the space, if necessary; wound drainage with tubes remaining in place until drainage is decreasing in a satisfactory manner; use of compression dressings; immobilization of the affected part for approximately 7 days;
●
●
leaving external sutures in place in radiated skin for at least 2–3 weeks; delineation of a subgroup of patients where postoperative boost dose can be omitted, which would include patients with negative margins and no tumour cut-through, complete tumour necrosis, or absence of tumour in the resection specimen.
ADJUVANT CHEMOTHERAPY Although surgery and radiotherapy achieve control of the primary tumour and cure most adult patients with soft tissue sarcomas, many patients, especially those with large grade 2 or 3 primaries, die of metastatic disease not evident at diagnosis. Several studies have shown chemotherapy to be effective against clinically evident metastases.274*,275*,276*,277*,278*,279* This has suggested to many that chemotherapy used as an adjuvant therapy (in high-risk stage M0 patients) would inactivate micrometastases and increase long-term survival. The first prospective randomized trial of adjuvant chemotherapy for this group of tumours was reported from the M.D. Anderson Hospital;189** they failed to demonstrate an advantage of vincristine, cyclophosphamide, adriamycin and dactinomycin chemotherapy over untreated controls. Similarly, a randomized Mayo Clinic series280** found a slight extension of disease-free survival with no increase in overall survival for patients receiving vincristine, cyclophosphamide and dactinomycin alternating with vincristine, doxorubicin and dacarbazine. However, using adriamycin (530 mg/m2) and cyclophophamide (500–700 mg/m2) followed by high-dose methotrexate, Rosenberg et al.281** demonstrated a significantly improved disease-free survival for patients with extremity lesions randomized to adjuvant chemotherapy (92 per cent vs. 60 per cent at 3 years with a survival advantage of 95 per cent vs. 74 per cent). Some serious cardiotoxicity was encountered. With longer follow-up in this small randomized trial, the survival advantage was no longer significant.282 Single agent doxorubicin has failed to show an improvement in disease-free or overall survival in patients receiving postoperative chemotherapy compared with surgery alone.283** Studies from Dana Farber Cancer Institute, Massachusetts General Hospital and ECOG284** and from UCLA285** and from the Scandinavian Sarcoma Group208** showed similar results. The very large EORTC study of adjuvant chemotherapy has employed CYVADIC consisting of cyclophosphamide, vincristine, adriamycin and DTIC. Results reported by Bramwell et al. in 1994 demonstrated a significant delay in the appearance of relapse: in 317 patients the relapsefree survival rates at 7 years were 56 per cent for the treated group and 43 per cent for controls (p 0.007).286** However, survival rates were not significantly different at 63 per cent and 56 per cent. There were fewer local failures among the CYVADIC treated head, neck and trunk
Table 40.11
Randomized adjuvant chemotherapy trials in soft tissue sarcomas Accrual period
Drugs Dox
Dox dose (mg/m2 total per cycle)
GOG354 DFCI/MGH284
1973–82 1978–83
None None
480 60 450 90
ECOG453
1978–82
None
450 90
SSG208
1981–86
None
540 60
Rizzoli283 IGSC454
1981–86 1983–86
None None
450 75 420 70
MDA455 Mayo NCI 4, 6188,282 NCI456,457
1973–76 1975–81 1977–81 1977–89
Cyclo, DAC, VCR VCR, cyclo, DAC, dacarbazine Cyclo, MTX Cyclo, MTX
420 60 200 50 500–550 50–70 500–550 50–70
EORTC286 Bergonie457
1977–88 1981–88
Cyclo, VCR, dacarbazine Cyclo, VCR, dacarbazine
400 50 400–500 50
SAKK287 IRCT8 Siena290
1987–90 1992–96 1985–96
Ifosfamide Ifosfamide, epirubicin /ifosfamide
550 50–90 600 120 (epirubicin) 300 75 (epirubicin)
Study
Sites Uterus Extremities, trunk, head, neck, retroperitoneum Extremities, trunk, head, neck, retroperitoneum Extremities, trunk, head, neck, breast, thorax, abdomen Extremities Extremities, trunk, head, neck, retroperitoneum Extremities, trunk Extremities, trunk Extremities Trunk, head, neck, breast, retroperitoneum Extremities, trunk, head, neck Extremities, trunk, head, neck, retroperitoneum, pelvis Extremities, trunk Extremities Extremities, trunk, retroperitoneum, abdomen
Number of patients
Chemotherapy
No chemotherapy
DFS %
OS%
DFS%
OS%
225 46
59 67
68 71
47 68
65 72
47
62
62
52
57
240
46
53
42
52
77 92
73* 65
91* 65
45 46
70 53
59 76 67 80
77 67 54* 42
58 71 65 42
83 63 28 41
71 46 44
468 65
56* 61
63 70
43 27
56 44
29 104 88
67 50* 69*
64 69* 72
67 37 44
80 50 47
GOG, Gynecologic Oncology Group; DFCI/MGH, Dana-Farber Cancer Institute/Massachusetts General Hospital; ECOG, Eastern Cooperative Oncology Group; SSG, Scandinavian Sarcoma Group; Rizzoli, Istituti Ortepedici Rizzoli; IGSC, Intergroup Sarcoma Committee; MDA, M D ANDERSON CANCER CENTER; Mayo, Mayo Clinic; NCI, National Cancer Institute; EORTC, EUROPEAN ORGANISATION FOR RESEARCH AND TREATMENT OF CANCER; Bergonie, Institut Bergonie; SAKK, Swiss Group for Clinical Cancer Research; IRCT, Italian Randomized Cooperative Trial; Dox, doxorubicin; Cyclo, cyclophosphamide; DAC, actinomycin; VCR, vincristine; MTX, methotrexate, *significant difference.
Adjuvant chemotheraphy 955
patients (p 0.002) but not among those with extremity tumours. There was no difference in distant metastasis (32 per cent vs. 36 per cent, p 0.42). A summary of randomized trials is presented in Table 40.11. A meta-analysis of 14 randomized trials of adjuvant chemotherapy versus control in soft tissue sarcomas demonstrated that adriamycin-based chemotherapy yielded an absolute gain in overall recurrence-free survival of 10 per cent from 45 per cent to 55 per cent (p 0.0001) and a trend for improvement of overall survival of 4 per cent from 50 to 54 per cent, (p 0.12). For local control the gain was 6 per cent at 10 years, viz. 75 per cent to 81 per cent, p 0.016. The most clear evidence of a gain in survival was obtained for patients 31–60 years old, recurrent lesions, extremity and high grade.287*** Encouraging results with more intense adjuvant therapy were reported by investigators at Massachusetts General Hospital who evaluated doxorubicin, ifosfamide and dacarbazine chemotherapy and radiation in adult patients with intermediate- or highgrade soft tissue sarcomas of the extremity 8 cm.258,288* Moreover, a newer randomized trial of more intense adjuvant chemotherapy with epirubicin, ifosfamide, mesna and G-CSF also seems to show a possible further advantage in disease-free and overall survival.8** In this trial, in which two thirds of the enrolled patients had either synovial sarcoma or liposarcoma (two particularly chemosensitive STS histologies), 144 patients with high-grade large (5 cm) or recurrent spindle cell sarcomas involving the extremities or girdles were randomly assigned to no postoperative therapy, or to five cycles of a dose-intensive epirubicin/ ifosfamide combination (epirubicin 60 mg/m2 on days 1 and 2 plus ifosfamide 1.8 g/m2 on days 1 to 5) with mesna and granulocyte colony-stimulating factor support.8 Accrual was prematurely discontinued at two years, when a significant difference in the cumulative incidence of distant metastasis was found (45 vs. 28 per cent, favouring the chemotherapy group); the overall survival difference (85 vs. 72 per cent), also favouring the chemotherapy group, did not reach the level of statistical significance. When the trial was reported, the 4-year overall survival rates were significantly higher in favour of chemotherapy (69 vs. 50 percent, p 0.04), although the distant relapse rates were by then similar in both groups (44 and 45 per cent for the chemotherapy and control groups, respectively). It is difficult to interpret these results, since the main benefit of adjuvant systemic chemotherapy is expected to be in reducing the rate of distant relapse. There was a trend towards improved local control in the adjuvant chemotherapy group, with a 17 per cent local failure rate at 4 years in the control group compared to 6 per cent in the adjuvantly treated group, p 0.07. Two additional randomized trials have explored the benefit of doxorubicin or ifosfamide-based chemotherapy in extremity,289**,290** one of which showed a survival benefit for adjuvant chemotherapy.290 The second trial, also from Italy, randomly assigned 88 patients with high risk STS to surgery with or without RT (n 43), or to surgery plus
chemotherapy (n 45, 26 with epirubicin alone, and 19 to epirubicin plus ifosfamide) with or without RT.290** The 5-year survival rate of patients treated with chemotherapy was significantly higher than that of patients who did not receive chemotherapy (72 vs. 47 per cent). However, the large number of treatment variables and the small number of studied patients makes interpretation of this result problematic. The importance of long-term follow-up in assessing benefit from chemotherapy was shown in a report of the combined experience of two major cancer centres (Memorial Sloan-Kettering and MD Anderson) that included 674 consecutive adults undergoing resection of a stage III extremity soft tissue sarcoma between 1984 and 1999.291* Adjuvant doxorubicin-based chemotherapy was administered to 336 (50 per cent), while the remainder received local therapy only. Although not a randomized trial, there were no significant differences between the chemotherapy and non-chemotherapy groups with respect to tumour size, anatomic site, histopathological subtype, or resection margin status. With a median follow-up of 6.1 years, the effect of chemotherapy appeared to vary over time. During the first year, the hazard ratio (HR) for disease-specific survival for chemotherapy versus no chemotherapy was 0.37 (95 per cent CI 0.20 to 0.69); thereafter, the HR was 1.36 (95 per cent CI 1.02 to 1.81). Another recent study from Memorial Sloan-Kettering Cancer Centre retrospectively analysed the relationship between neoadjuvant chemotherapy (NAC) and outcome in patients with high-grade, deep, 5 cm extremity sarcomas. Patients diagnosed between 1990 and 2001 were treated with surgery only (n 282) or NAC containing doxorubicin/ ifosfamide/mesna (AIM) (n 74). NAC was associated with improved disease-specific survival for this cohort of patients (p 0.02). This overall improvement appears to be driven by the benefit of NAC on disease-specific survival for patient with tumours 10 cm. The 3-year disease-specific survival for tumours 10 cm was 0.62 (95 per cent CI: 0.53–0.71) for patients not receiving NAC and 0.83 (95 per cent CI: 0.72–0.95) for patients receiving NAC.292* A study of concurrent, interdigitated neoadjuvant MAID chemotherapy and radiation therapy in patients with extremity sarcomas 8 cm showed higher disease-free and overall survival when compared to a historical matched patient cohort.258* These data emphasize the need for further prospective clinical studies of neoadjuvant or adjuvant chemotherapy for patients with large high-grade extremity sarcomas. Despite many randomized trials, the role of adjuvant chemotherapy remains uncertain and cannot be adopted as the standard of practice for all extremity sarcomas. The positive Italian trial is the only study that enrolled predominantly patients with chemosensitive types of STS (i.e. liposarcomas and synovial sarcomas).8** However, the lack of a difference in the distant metastatic rate at 4 years (the time point at which the survival benefit was most pronounced) calls into question the validity of the conclusion that adjuvant chemotherapy improves survival in patients with high-grade or recurrent sarcomas. If there is a survival
956 Soft tissue sarcomas
benefit for adjuvant chemotherapy, it is probably small (no more than 4–8 per cent absolute increase in survival at 5–10 years). Although it has been proposed that patients be selected for adjuvant chemotherapy based upon poor prognostic tumour characteristics such as histology (i.e. synovial sarcoma293), grade or size, there is no evidence to date that this approach leads to improved outcome in any subset of patients.294 The absolute benefit, patient selection, and optimal regimen remain to be defined. A large, randomized study of postoperative adjuvant adriamycin and ifosfamide was recently completed by the European Oncology Research and Treatment Cooperative (EORTC) Group. The results of this study, which are pending, may help to better define the role for adjuvant chemotherapy in these patients.
FUNCTIONAL OUTCOME There are increasing data available on the functional outcome of patients undergoing limb salvage procedures.196*,246*,296,297 The majority of patients have good or excellent functional outcome. In one series of 88 patients treated with surgery and either preoperative or postoperative radiation therapy, 68 had acceptable functional results and 61 returned to work.196* Large tumours, neural sacrifice, proximal thigh tumours and postoperative complications were associated with poor outcome. Subcutaneous tumours have a more favourable functional outcome.295 In a single institution series of 145 patients who underwent limb-sparing surgery plus radiation therapy, long-term treatment complications included bone fracture in 6 per cent, contracture in 20 per cent, significant oedema in 19 per cent, moderate to severe decrease in muscle strength in 20 per cent, requirement for a cane or crutch in 7 per cent and tissue induration in 57 per cent.246* Three patients (2 per cent) required amputations for treatment-related complications. The percentage of patients ambulating without assistive devices and with mild or no pain was 84 per cent. Higher doses of radiation therapy, a long radiation portal and irradiation of more than 75 per cent of the extremity diameter were associated with increased complications. Another study examined issues related to quality of life in patients with STS of the lower limb.296 Although radiation therapy was associated with reduced muscle power and range of motion, compared to the use of surgery alone, most patients retained good to excellent limb function and quality of life. The functional outcome is often not as good in patients requiring amputation. In a matched case-control study of patients with lower extremity sarcoma undergoing amputation (n 12) or a limb-sparing approach (n 24), there was a trend toward increased disability and handicap for those in the amputation group.297* Seven of the 12 amputees reported ongoing problems with the soft tissue overlying the stump. A few studies have assessed quality of life issues in amputees who had been treated for STS with amputation and chemotherapy compared with patients who underwent limb salvage with radiotherapy and chemotherapy.298*,299*
Contrary to expectations, there were no significant differences in measures of psychological outcome. Thus, a psychological advantage of limb-salvage surgery compared to amputation has yet to be demonstrated.
TREATMENT OF LOCAL RECURRENCE Approximately 10–15 per cent of patients with extremity STS who are treated with complete resection and adjuvant radiation will develop a local tumour failure, the majority within the first two years.196*,197*,202*,206* The approach to the patient with an isolated local recurrence is similar to that for primary disease with some modifications. As with primary treatment, the goal is to provide limb salvage with conservative resection. However, approximately 10–25 percent of patients with locally recurrent disease will require amputation.300*,301*,302*,303* Surgery is an important component of successful salvage therapy.303* For patients whose primary treatment was surgery alone, re-excision combined with adjuvant radiation is the treatment of choice. If radiation therapy was used in primary treatment, further radiation may not be possible because the maximal tolerance for adjacent normal tissues would have to be exceeded, resulting in problems in wound healing and radiation fibrosis, although additional radiation given by brachytherapy (mean dose 47.2 Gy) has been employed in these cases with a 52 per cent local control and a 33 per cent disease-free survival in one series of 26 patients.304* Optimal treatment for a local recurrence may require both surgery and radiation. This was illustrated in one report of salvage therapy using surgery alone or surgery plus reirradiation for 25 patients with locally recurrent extremity STS.303* Eighteen patients underwent surgery alone, 11 treated by a conservative procedure and seven requiring amputation. Seven of these 18 relapsed. Of the 10 treated with surgery plus radiation, none experienced relapse with a median follow-up of 24 months. Six (60 per cent) experienced significant wound healing complications, but three recovered completely.
RADIOTHERAPY ALONE It is worth also reviewing the experience with radiation alone for patients with sarcomas. Radiotherapy alone is used for those patients who, for reasons of anatomical location, medical inoperability, or refusal of any surgery, are not candidates even for a conservative surgical procedure. In 1951 Cade described 22 patients who, for a variety of reasons, were treated by radiation alone.305* Six survived for 5–26 years. Windeyer et al. described results of radiotherapy alone in fibrosarcoma; of eight patients available for 5-year follow-up, four were free of local disease, although one had required surgery for a recurrence.306* McNeer et al.307* reported their analysis of the results of treatment of 653 patients with sarcomas of soft tissue treated at the Memorial Hospital. Twenty-five of the 653 patients were treated by
Management of the retroperitoneal primary lesion 957
radiotherapy alone; 15 of the 25 were surviving at 5 years, and eight of 20 were surviving at 10 years. Local control was achieved in 14 of the 25 patients. For small sarcomas, good local control rates can be achieved by radiation alone. However, local control probabilities of 90 per cent for tumours of estimated volume 15–65 mL (approximately a sphere of 3–5 cm in diameter) require high radiation doses (75 Gy).215*,308* For unresected sarcomas, higher rates of local control were seen in patients treated to higher dose; with the median dose of 63 Gy chosen as a cutoff, local control was significantly higher with doses above 63 Gy.215* Smaller tumours are significantly more likely to be controlled than larger lesions; for lesions 5 cm, local control of 72 per cent was reported for doses above 63 Gy.215* Current recommendations for gross disease would be to deliver doses in the range of 75 Gy with shrinking field and highly conformal techniques if possible. If the volume of gross residual disease is large, the late normal tissue changes resulting from these dose levels may be important.
TREATMENT OF UNRESECTABLE OR LOCALLY ADVANCED SOFT TISSUE SARCOMA In patients with advanced STS in whom the tumour has progressed beyond surgical resectability, treatment options depend upon the site of tumour involvement. For patients with unresectable disease limited to the extremity, isolated limb perfusion (ILP) protocols have been applied with striking early success. Selected patients can have their tumours, especially when small, controlled with radiation therapy with or without chemotherapy. For patients treated to a dose of 6400 cGy or greater, Tepper and Suit noted control of unresected soft tissue sarcomas in 87.5 per cent of cases where tumours were less than 5 cm in diameter.308* Treatment was less effective for larger tumours, with local control falling to 53 per cent for lesions 5–10 cm diameter (53 per cent) and 30 per cent for those greater than 10 cm. Kepka and colleagues recently updated and expanded this experience, reporting on the efficacy of radiation on 112 patients with unresected soft tissue sarcomas. For patients receiving 63 Gy, local control at 5 years was 72.4 per cent in patients with lesions 5 cm, 42.4 per cent for lesions 5–10 cm in size, and 25.4 per cent for lesions 10 cm.215 Several centres have reported local control rates of approximately 50 per cent with fast neutron irradiation of inoperable STS.309*,310*,311* In addition, several radiation sensitizers have been used to treat patients with extensive sarcoma with promising early preliminary results.312*,313*
MANAGEMENT OF THE RETROPERITONEAL PRIMARY LESION Approximately 10–15 per cent of soft tissue sarcomas arise in the retroperitoneum.198,314 These tumours are often
Figure 40.13 CT scan of large left retroperitoneal liposarcoma. Well-differentiated tumour appears as abnormal fat while de-differentiated tumour (DD) appears as more solid masses.
asymptomatic and identified on imaging studies for unrelated complaints. In other cases, patients may present with a palpable abdominal mass or with symptoms such as abdominal pain or lower extremity neurological symptoms.198,315 About 10–20 per cent of patients are found to have distant metastases on initial presentation.316* Since the retroperitoneum can often accommodate large tumours without symptoms, the average size of tumours in large series is often greater than 10 cm.315,316 Upon histological examination, about two thirds of tumours are either liposarcomas or leiomyosarcomas, with the remaining tumours distributed among a large variety of histological subtypes.315 Retroperitoneal liposarcomas are further classified into well-differentiated, de-differentiated, and myxoid–round cell subtypes. In the largest series of retroperitoneal liposarcomas (n 177), 56 per cent were well differentiated, 37 per cent de-differentiated, and 7 per cent myxoid-round cell.317* About 40 per cent of tumours are low grade and the remaining 60 per cent are intermediate or high grade.314* Low-grade tumours uncommonly metastasize, while intermediate and high-grade tumours can metastasize to the lung and liver. Most unifocal tumours in the retroperitoneum that do not arise from adjacent organs will be either benign soft tissue tumours (e.g. schwannomas) or sarcomas. Other malignancies in the differential diagnosis include primary germ cell tumour, metastatic testicular cancer and lymphoma. Following a careful history and physical examination, radiological assessment of these tumours is usually performed with an abdomen and pelvic CT scan. Liposarcomas often have a characteristic appearance, with large areas of abnormal fat (well-differentiated liposarcoma) often containing higher density nodules (de-differentiated liposarcoma) (Fig. 40.13). Leiomyosarcomas appear as heterogeneous solid tumours (Fig. 40.14). MRI scans may be helpful in delineating tumours from adjacent soft tissues, nerves or major blood vessels, but are usually not necessary. To evaluate for
958 Soft tissue sarcomas
Figure 40.14 Three-dimensional CT reconstruction image of inferior vena cava leiomyosarcoma.
metastatic disease, the abdomen CT is adequate in evaluating the liver. Patients with intermediate and high-grade tumours should have a chest CT to evaluate for lung metastases. A chest X-ray is obtained for low-grade tumours. The primary treatment for the local control of these tumours is surgical resection. The optimal goal of surgical resection is complete gross resection with microscopically negative margins. However, even complete gross resection can be difficult to obtain, and complete gross resection rates in large series are between 54 per cent and 67 per cent.315*,316*,318* In about three quarters of cases, complete gross resection requires resection of adjacent viscera.315* Of note, for tumours abutting the kidney, the renal capsule can often be resected rather than formal nephrectomy, given that 75 per cent of kidneys resected in one series showed no renal capsule, parenchyma or vessel invasion.319 The ability to obtain negative microscopic margins for large retroperitoneal tumours is also difficult. These tumours are surrounded by a pseudo-capsule that often contains microscopic disease, and dissection with a normal tissue margin away from the pseudo-capsule is difficult, especially along the posterior aspect of the tumour where it abuts the retroperitoneal fat and musculature. Positive or negative microscopic margin was not a prognostic factor for local recurrence in the largest series of retroperitoneal sarcomas from Memorial Sloan-Kettering Cancer Centre.315* It is likely that many large retroperitoneal tumours thought to have negative microscopic margins in fact have positive margins in areas not specifically examined under the microscope. Prognostic factors for local recurrence in that series were incomplete gross resection, high grade and liposarcoma histology. Controversy exists as to the optimal role of radiation therapy for local control of retroperitoneal sarcomas. The American College of Surgeons Oncology Group was not able to complete a recent phase III randomized trial of preoperative radiation and surgery versus surgery alone for retroperitoneal sarcomas. Those who advocate radiation
therapy usually prefer that radiation be delivered preoperatively.198* With the tumour still in place, the margin around the tumour at risk of local recurrence is more clearly defined, and the effective radiation dose to control microscopic disease is probably lower. Petersen et al. reviewed the Mayo experience with 43 patients who received external beam radiation (usually preoperatively) and IORT.320* Local control was 75 per cent at 5 years. About one fifth of patients had IORT toxicity, including gastrointestinal fistula, neuropathy and ureteral injury. The University of Toronto Sarcoma Group treated 55 patients with retroperitoneal sarcomas with either preoperative external beam radiation therapy or brachytherapy and found that external beam radiation therapy (median dose 45 Gy) was well tolerated while brachytherapy to the upper abdomen was associated with significant toxicity.321* In a report from our institution, 29 patients were treated with preoperative radiation to a median dose of 45 Gy and then underwent complete gross resection.322* Intraoperative radiation 10–20 Gy was delivered to 16 patients and no IORT was delivered to 13 patients. Local control at 5 years was 83 per cent for patients who received both preoperative radiation and IORT and 61 per cent for those who received only preoperative radiation. Significant toxicity from IORT occurred in 4 patients and included neuropathy, ureteral stricture and vaginal fistula. More recently, we have incorporated the use of preoperative proton beam radiation therapy along with resection and IORT for retroperitoneal tumours and have found these techniques allow dose escalation to areas at risk while decreasing dose to adjacent organs at risk.253 In the extremity, the local control of sarcomas treated with radiation therapy and total gross resection with positive microscopic margins is about 75 per cent.196*,202*,203* Typically, positive microscopic margins are treated with a boost of postoperative radiation to a total dose of about 66–68 Gy. It seems reasonable to assume that total gross resection of retroperitoneal tumours along with adequate doses of radiation could achieve local control rates similar to those seen for extremity tumours resected with positive microscopic margins. The availability of intensity modulated radiation therapy techniques, proton beam radiation and IORT may facilitate the efficacy and minimize morbidity of adjuvant radiation therapy for the retroperitoneal tumours and translate into improved local control. The addition of chemotherapy to current radiation therapy strategies may also improve the local control of retroperitoneal sarcomas and also may have some impact on distant disease. A few small series have been reported. In one study, 16 patients were treated with iododeoxyuridine and radiation therapy with the only grade 4 toxicity being vomiting in three patients.323* Eleven patients went on to resection, and four patients had a local recurrence. Pisters et al. at MD Anderson Cancer Center treated 35 patients in a phase I trial of doxorubicin and escalating doses of EBRT. Two of 11 patients had grade 3 or 4 toxicity at the highest dose of EBRT (50.4 Gy). IORT was delivered
Treatment of metastatic disease 959
to 22 of 29 patients who subsequently underwent surgery. Local control results have not yet been published.324*
MANAGEMENT OF GASTROINTESTINAL STROMAL TUMOURS Gastrointestinal stromal tumours (GISTs) are the most common mesenchymal tumour of the gastrointestinal tract and are thought to originate from the interstitial cells of Cajal.325 These cells are thought to act as pacemaker cells, and have immunohistochemical and electron microscopy features of both neuronal and smooth muscle cells. Interstitial cells of Cajal and the vast majority of GISTs express cKIT, which is a 145 kD trans-membrane glycoprotein that acts at the receptor for stem cell factor (SCF). Prior to the identification and availability of immunohistochemistry for c-KIT, the majority of GISTs were thought to be of smooth muscle origin and termed leiomyomas, leiomyosarcomas and leiomyoblastomas. Ninety-five per cent of GISTs stain positive on immunohistochemistry for c-KIT, and the majority of GISTs have a mutation in the c-KIT gene. Unlike most other cancers, GISTs seem highly dependent on this single pathway for neoplastic growth. Gastrointestinal stromal tumours can occur at nearly all ages, with the median age being around 60.326 The incidence is roughly equal in men and women. The tumours are located most commonly in the stomach (60 per cent) followed by the small intestine (30 per cent), colon/rectum (5 per cent), and oesophagus (5 per cent).327 Many tumours are found on upper endoscopy, where they appear as submucosal lesions. However, even large GISTs of the stomach may not be seen on endoscopy if they are pedunculated or exophytic. Endoscopic biopsy can establish the diagnosis, especially following staining for c-KIT expression. For small bowel tumours, CT-guided biopsy poses the risk of inadequate tissue and spillage of tumour cells and is probably not necessary given that most isolated small bowel tumours require resection. Staging workup should include abdomen and pelvis CT scan to rule out intra-abdominal or liver metastases. GISTs are usually strongly positive on PET scans, and thus PET scans can be a valuable study for excluding metastatic disease as well as evaluating response to medical therapy such as imatinib. Surgical resections of GISTs give tumours that can range from small, pedunculated lesions to large lesions with adherence or invasion of surrounding tissues and organs. Thus the surgical approach can be quite varied but certain principles should be followed. Upon surgical exploration, the liver and peritoneal cavity should be examined for possible metastatic disease. Gastrointestinal stromal tumours uncommonly metastasize to lymph nodes so a regional lymphadenectomy is not required. Thus, for a pedunculated gastric tumour, wedge resection of the gastric wall along with resection of the tumour is adequate. Some gastric tumours encompass a large portion of the stomach and a formal distal, subtotal or even total gastrectomy may be
required. For small bowel and colon tumours, segmental bowel resection can be performed. Small rectal GISTs can be removed through transanal procedures while larger tumours may require low anterior resection or even abdominal-perineal resection. Some large GISTs with significant necrosis are susceptible to tumour rupture and spillage, which can lead to intra-peritoneal spread of disease, so tumours should be manipulated carefully. Risk of recurrence is most closely related to size and mitotic count. DeMatteo et al. found in 80 patients with resected GISTs that size was the most important predictor of overall survival, with 5-year overall survival being 60 per cent for tumours 5 cm, 45 per cent for tumours 5–10 cm, and 20 per cent for tumours 10 cm.326* The National Institutes of Health (NIH) convened a GIST Workshop in 2001 and stratified GIST from very low risk to high risk based on size and mitotic count.328 The most common sites of metastases are the liver and peritoneal cavity. Given the efficacy of imatinib in treating metastatic GISTs, there has been significant interest in the use of imatinib in the adjuvant setting.327* The American College of Surgeons Oncology Group (ACOSOG) recently closed a phase II study of adjuvant imatinib for one year for patients with resected high-risk tumours (size 10 cm, tumour rupture), and results are pending. Another ACOSOG trial is currently randomizing patients with resected GISTs 3 cm to imatinib for one year or placebo, and similar adjuvant imatinib trials are under way in Europe. Further trials are currently examining the use of imatinib in the neoaduvant setting for locally advanced or marginally resectable tumours.
TREATMENT OF METASTATIC DISEASE Overview Metastatic disease rarely occurs in patients with low-grade sarcomas but occurs at an appreciable frequency in patients with high-grade sarcomas that is related to grade and size.294 With intermediate- or high-grade sarcomas, this risk may exceed 50 per cent when the tumour is larger than 10 cm.204,205,207 For extremity sarcomas, the lung is the most common site of metastatic disease.4,329 Some histologies, notably myxoid liposarcoma, which can metastasize to abdominal sites and bone,330,331 and epithelioid sarcomas, which manifest regional nodal failure, are exceptions to the more general pattern. While most patients with metastatic sarcoma will ultimately die from their tumour, a modest proportion of patients will be long-term survivors after management with surgery and or chemotherapy.
Resection of pulmonary metastases The median survival of patients with pulmonary metastases is in the range of 15 months. Patients whose lung
960 Soft tissue sarcomas
metastases can be resected fare better than those with unresectable metastases. In one series, patients treated with resection had a median survival after complete resection of 33 months.332* Their 3-year actuarial survival rate was 46 per cent, with a 5-year actuarial survival rate of 37 per cent. The patients who did not undergo resection had a median survival of 11 months and a 3-year actuarial survival rate of 17 per cent. Patients to be considered for pulmonary resection are medically fit patients with controlled primary tumours without pleural effusion or hilar disease. The procedure generally involves wedge resections of the nodules. Patients with a limited number of nodules fare better, but there is no consensus on the upper limit of nodules that should be considered for resection. The role for additional adjuvant chemotherapy after resection is not settled. A subset of patients may benefit from repeat thoracotomy for recurrent disease in the chest.333 There are reports of resection of isolated metastatic disease in liver and other extra-pulmonic sites as well.334
Chemotherapy for metastatic disease For most patients with metastatic disease that is not resectable, treatment with chemotherapy is likely to be palliative in outcome. A small number of patients will be longterm survivors. An analysis of 1888 patients treated on studies organized by the Soft Tissue and Bone Sarcoma Group of the European Organization for Research and Treatment of Cancer (EORTC) reported 88 5-year survivors, which translates to an 8 per cent 5-year survival rate.335* A brief discussion of chemotherapy for metastatic disease will be presented here; readers can find more complete discussion in recent reviews on this subject.336***,337*** Doxorubicin and ifosfamide have been demonstrated to be the most active chemotherapy agents in widely disseminated soft tissue sarcoma.338 For doxorubicin, objective response rates between 20 and 40 per cent for the single agent have been reported; few are complete responses and response duration averages 8 months. A steep dose–response curve for objective responses was described by O’Bryan et al.339*. There is also a dose-response for ifosfamide.340*,341* Dacarbazine (DTIC) by itself has a modest response rate around 16 per cent.342* Cyclophosphamide appears less active in adults than in children and less active than the related compound ifosfamide.343** Gemcitabine with or without a taxane is active in a subset of patients with sarcomas.344* Angiosarcoma of the scalp and face may respond to paclitaxel.345* A pegylated liposomal formulation of doxorubicin has activity against sarcomas.346 Gemcitabine has been shown to have modest single-agent activity in previously treated patients,347* as does vinorelbine.348* Studies have shown a 24 per cent 6-month disease progression control rate using trabectidin, ecteinascidin-743, in patients with advanced pre-treated sarcoma.349* Other less active agents include methotrexate350*,351* and cisplatin.352*,353*
Many combination chemotherapy regimens for metastatic disease have been studied in phase II trials. Most of these trials include doxorubicin (or epirubicin) and an alkylating agent. Adding DTIC to doxorubicin improved the response rate to 41 per cent as described by Gottlieb et al.,274* but the response rate has decreased over time.342* Randomized trials354**,355** found some gain for the combination. A SWOG phase III trial compared bolus versus infusional administration of doxorubicin plus dacarbazine and reported no differences in overall response (17 per cent in both arms) or complete responses (5 per cent in both arms). Additionally, there was no difference in the median survival, 10.5 months in both groups.356** Adding cyclophosphamide to the basic duo was reported to raise the response rate to 56 per cent357* and this was confirmed by a ** randomized trial.358* Comparisons have shown that the addition of less active drugs necessitates lower doses of doxorubicin and, accordingly, reduces overall effectiveness. Adding ifosfamide seems to be clearly beneficial as reported by Blum et al.361** and Schutte et al.362**. The Eastern Cooperative Group (ECOG) conducted a three-arm trial comparing doxorubicin alone, doxorubicin plus ifosfamide, and mitomycin plus doxorubicin plus cisplatin. Objective tumour regression occurred more frequently in the combination arms than in the single-agent arm (20 per cent with doxorubicin alone, 34 per cent in doxorubicin plus ifosfamide, and 32 per cent in the mitomycin plus doxorubicin plus cisplatin arm). However, the combination regimens resulted in significantly greater myelosuppression, e.g. 80 per cent of the doxorubicin/ifosfamide group had grade 3 or greater myelosuppression. Most notably, no significant survival differences were observed between the three treatment regimens.363** *The extensively utilized CYVADIC regimen has evolved from the sequential trials sponsored by SWOG.363a Greenall et al. reviewed studies reporting response rates of 15–60 per cent, the average being 41 per cent.364* It is noteworthy that up to 15 per cent of the responses have been scored as complete. The median response was longer (13 months), and 21 per cent of complete responders described by Yap et al.365 achieved 5-year disease-free status. Pinedo and Verweij351* have also reviewed alternative regimens. A popular regimen adds ifosfamide to adriamycin and DTIC.366 A combination of ifosfamide with mesna, doxorubicin and dacarbazine has resulted in response rates in measurable metastatic sarcomas as high as 47 per cent with complete response rates as high as 10 per cent.367*,368* Another protocol describes activity with DTIC and cisplatin in pretreated patients.369* Uterine leiomyosarcoma may respond to highdose gemcitabine with docetaxel.344* In the absence of any clear benefit for the combination regimens, some clinicians favour initiation of chemotherapy with single-agent adriamycin.336 Dose intensity may be important as described by Zaninelli et al.370. Very high dose ifosfamide had been used by Rosen et al.371* with high response rates despite some toxicity. Higher-dose therapy with standard agents372* may require
Desmoid tumours 961
special supportive care such as bone marrow transplantation,373* but may offer a chance for higher complete response rates and longer response duration. Considerable interest has been focused on maintaining dose intensity of chemotherapy using colony-stimulating factors to alleviate myelosuppression. Granulocyte–macrophage colony-stimulating factor (GM-CSF) or granulocyte colony-stimulating factor G-CSF has been used with a variety of regimens to help maintain dose intensification. In a few studies this has resulted in improved response rates.374* Even higher-dose chemotherapy, as used at MD Anderson, seems to result in higher response rates (59–69 per cent).375*,376* Attempts to intensify treatment by increasing the dose of doxorubicin in combination with ifosfamide, although promising in phase II studies,377* were not confirmed in a subsequent randomized trial compared with the standard doses.378** High-dose therapy with growth factor support has been evaluated in several investigational studies but the data to date demonstrate increased toxicity with clear evidence of therapeutic gain, so this is still considered investigational treatment.379*,380*
The incidence of desmoids is higher in patients with familial adenomatous polyposis (FAP), affecting from 4 to 20 per cent of patients.384 The simultaneous appearance of FAP and desmoid tumours was described by Gardner in 1951, and is now designated as Gardner’s syndrome. Until early elective colectomy became routine in patients with FAP, the dominant cause of death in these patients was carcinoma of the colon. With the increasing use of prophylactic colectomy, desmoid tumours have become an important cause of morbidity385 and, in some instances, mortality.386
Pathology Desmoids tend to be large bulky tumours that lack pseudoencapsulation and locally infiltrate adjacent tissue structures. Histologically, they are characterized by small bundles of spindle cells in an abundant fibrous stroma. The fibroblasts have a propensity to concentrate at the periphery of the lesion, and the cellularity is low. There are usually few mitotic figures and necrosis is absent.
Targeted agents Mutations in c-KIT in gastrointestinal stromal tumours create an ideal molecular target, which has allowed imatinib to have marked success in the treatment of that rare subtype of sarcoma.381* Since imatinib also inhibits the receptor of platelet-derived growth factor beta, it can be effective in the treatment of dermatofibrosarcoma protuberans as well.382*
DESMOID TUMOURS Introduction Desmoid tumours are deep-seated, benign, slowly growing fibroblastic neoplasms that arise from musculoaponeurotic stromal elements. Although they are locally aggressive, desmoids do not have the capacity to establish metastatic lesions. Nevertheless, tumour-related destruction of vital structures and/or organs can be fatal.
Epidemiology Desmoid tumours are uncommon. The estimated incidence in the general population is 2–4 per million inhabitants per year, which in the USA translates into approximately 900 new tumours annually.383 Individuals between the ages of 15 and 60 are most often affected; desmoids are rare in the young and in the elderly. They are slightly more common in women than in men, and there is no significant racial or ethnic distribution.
Aetiology and pathogenesis The aetiology of desmoid tumours is unknown. However, the identification of clonal chromosomal changes in a significant fraction of cases supports the neoplastic nature of these tumours387 and emerging evidence implicates dysregulated wound healing in the pathogenesis of these and other fibroblastic lesions. Non-random clonal chromosomal changes, particularly trisomy 8 and/or 20, occur in one third or more of sporadic desmoid tumours.388 Similar non-random genetic aberrations have been found in benign fibrous bone lesions (such as fibrous dysplasia), suggesting a similar pathogenesis.389 Although the clinical relevance of these genetic abnormalities is unclear, their presence appears to be associated with a higher risk of recurrence.388 As an example, in one report, trisomy 8 and/or trisomy 20 were observed in cells cultured from 6 of 13 desmoid tumours.388 Fluorescence in situ hybridization (FISH) analysis performed on the nuclei from 25 desmoid tumours from paraffin blocks or frozen tissue indicated that local recurrence was more likely in desmoids with trisomy. Among patients followed for more than one year, local recurrence was more frequent in tumours with trisomy 8 (4 of 6, as compared to 2 of 17 recurrences in trisomy 8-negative tumours). The finding of individual trisomies and their association in the same cell is rare in solid tumours, particularly mesenchymal tumours.389 However, these aberrations are known to occur in related benign, fibrous lesions arising in both soft tissue and bone tumours (e.g. Dupuytren’s contracture, plantar fibrosis, Peyronie’s disease, carpal tunnel syndrome and infantile fibrosarcoma). Trisomy 8 is also a frequent finding in
962 Soft tissue sarcomas
haematological malignancies, but is remarkably infrequent in non-fibrous solid tumours. GARDNER’S SYNDROME
Gardner’s syndrome is a variant of FAP that is distinguished by the presence of prominent extra-intestinal lesions, such as desmoid tumours, osteomas and cysts. When these occur in any member of an FAP family, the family has traditionally been said to have Gardner’s syndrome rather than FAP, since all members of the family segregate the same mutation in the adenomatous polyposis coli (APC) gene. Desmoids may be the first manifestation of Gardner’s syndrome. Families have also been reported that exhibit desmoids as the only manifestation of an APC mutation. The estimated risk of developing a desmoid tumour in patients with FAP is between 4 and 20 per cent.390 They have a particular predilection for surgical sites (e.g. the mesentery or abdominal wall following colectomy, the site of an ileal pouch–anal anastomosis). In one series, prior abdominal surgery had been performed in 68 per cent of patients with FAP and abdominal desmoid tumours; lesions develop within five years after surgery in approximately one half.391 One patient dramatically demonstrated this association with surgery. She had been treated, apparently successfully, with chemotherapy for an intra-abdominal desmoid tumour. To confirm the response, laparoscopy was performed. Within months, desmoid tumours began developing in each of the three trocar sites; the tumours became massive and inoperable and led to the death of the patient. APC MUTATIONS AND BETA-CATENIN
Mutations of the APC gene on chromosome 5q are responsible for FAP. More than 300 mutations have been described, most of which lead to frame shifts or premature stop codons, resulting in a truncated APC gene product. Several studies have attempted to correlate specific APC mutations with the clinical phenotype. As a general rule, mutations between codons 169 and 1393 are associated with the classic form of FAP while mutations that are more 3 or 5 are associated with the attenuated form of FAP. The site of the germ-line mutation in patients with FAP may be important for the risk of developing a desmoid tumour. Mutations between codons 1445 and 1578 have been associated with desmoid tumours in some reports,392 although others have not found this association.393 How the abnormal gene promotes the formation of tumours such as desmoids is incompletely understood. However, increasing evidence points to involvement of germ-line APC mutations in the molecular pathogenesis of desmoids in patients with Gardner’s syndrome.394 The normal APC protein prevents the accumulation of betacatenin, a cytosolic and nuclear protein, by mediating its phosphorylation and resultant degradation. The loss of the beta-catenin regulatory domain in the truncated protein
allows beta-catenin to accumulate, bind to and activate the transcription factor Tcf-4.395 It is hoped that the elucidation of the central role of beta-catenin in the pathogenesis of desmoid tumours will lead to future therapeutic advances targeting this molecule. TRAUMA AND PREGNANCY
Up to 30 per cent of patients with desmoid tumours have a history that involves antecedent trauma,396 particularly surgical trauma, in patients with FAP (see above). A similar relationship has been observed in some sporadically occurring desmoid tumours. In one series, an antecedent history of trauma at the tumour site was elicited in 28 per cent of 32 primary desmoid tumours.397 Abdominal desmoids tend to occur in women during or following pregnancy. The classic presentation is that of an abdominal mass that is separate from the uterus.398 Trauma related to pregnancy and exposure to elevated hormone levels may both be contributory. As an example, one case report describes a desmoid tumour that developed at the site of a prior caesarean section scar during a subsequent pregnancy.399 Subsequent pregnancy is not necessarily a risk factor for recurrence or development of new disease in a woman who develops a pregnancy-related desmoid.400
Clinical presentation Most desmoid tumours present as a painless or minimally painful mass with a history of slow growth. Intra-abdominal desmoids can be associated with intestinal obstruction, mucosal ischaemia or functional deterioration in an ileoanal anastomosis (typically in a patient who has undergone colectomy for FAP). Desmoid tumours can develop at virtually any body site; they most commonly arise in the torso (shoulder girdle and hip–buttock region) and the extremities. The location is usually deep in the muscles or along fascial planes.396 Desmoid tumours may be multi-focal on an extremity, but they rarely occur in different regions in the same patient.
Diagnosis Ultrasound is often the first method of examination of a soft tissue lesion on the torso or extremity. If the mass is solid, CT or MRI is needed to determine adherence to adjacent structures and resectability. Although desmoids can be adequately evaluated by CT, MRI is preferred for definition of the pattern and extent of involvement. There are no radiographic characteristics that can reliably distinguish desmoids from malignant soft tissue tumours. The diagnosis of a desmoid tumour can only be established by histological examination of a biopsy specimen. Incisional biopsy is often preferred over a needle biopsy because of the need to distinguish between a benign and malignant
Desmoid tumours 963
process with very high confidence. There is no accepted staging system for desmoid tumours.
Natural history Although benign, desmoid tumours are locally infiltrative and can be fatal by causing destruction of adjacent vital structures and organs. In a report of 138 patients managed at one institution between 1965 and 1984, 11 died of their disease.401 Factors associated with a poor outcome in this study were: age 18–30 years, presentation with locally recurrent disease, incomplete excision, and no postoperative radiation therapy (RT). However, this series may not be representative since many patients had advanced disease at the time of diagnosis. Other centres report an overall mortality rate 1 per cent in patients with desmoid tumours at other than intra-abdominal sites.402 Thus, even though desmoids are benign in the sense that they cannot produce distant metastases, the disease process may be devastating, and occasionally fatal. Fortunately, the pace of progression is usually relatively slow, with periods of comparative stability or even temporary regression. Re-growth is not an inevitable consequence following grossly incomplete surgical resection.403
Treatment Treatment of desmoid tumours is indicated when they cause symptoms, if there is imminent risk to adjacent structures or if they create cosmetic concerns. SURGERY
Because of their locally infiltrative nature, desmoid tumours are treated by surgical resection with a wide margin when medically and technically feasible.402 Since these are benign tumours, the treatment strategy should be to obtain tumour-free margins using function-preserving approaches to minimize major morbidity (functional and/or cosmetic). Extra-abdominal and abdominal wall desmoids are more often resectable than are intra-abdominal tumours. Surgery may be difficult and even impossible for intra-abdominal desmoids, although it may be an important option for selected patients.404 Medical therapy may be considered as a first-line option, particularly for tumours that involve the mesentery or encase vessels and organs. Despite their benign character, desmoid tumours have a high rate of recurrence with surgery alone. Even patients who undergo aggressive resection with wide margins have recurrence rates of 23–39 per cent.402* When they recur, salvage therapy with RT and/or repeat excision is usually successful. Intra-abdominal desmoids that arise in the setting of Gardner’s syndrome have a greater tendency for local recurrence and multiple lesions. Moreover, they may be relatively refractory to RT and may be rendered more
aggressive after surgical intervention.405* This has led some clinicians to advocate conservative management with noncytotoxic therapy as opposed to resection and/or RT in these patients,406* although this is controversial. IMPORTANCE OF RESECTION MARGINS
The available data are conflicting with regard to the importance of complete resection. Some authors report that the risk of recurrence is independent of margin status,407*,408*,409* while others demonstrate higher recurrence rates with close or positive resection margins.401,402 As an example, in one series of 203 patients undergoing surgery for either primary or recurrent desmoid tumours over a 35-year period, margins were microscopically positive in 57 and negative in 146.408* As expected, the disease-free survival rate was significantly better in patients with primary disease (76 vs. 59 per cent at 10 years), but it was not significantly worse for those with microscopically positive versus negative margins at primary surgery (5-year disease-free survival rate for those with positive and negative margins, 79 vs. 82 per cent; at 10 years, 74 vs. 77 per cent). These data have led some to conclude that aggressive attempts to achieve negative resection margins are not warranted if they result in excessive morbidity.407* Moreover, uncertainty as to the importance of positive resection margins has led to controversy with regard to the utility of postoperative RT for patients with incompletely resected disease. RADIATION THERAPY
Radiation therapy is an effective primary therapeutic option for patients who are not good surgical candidates, those who decline surgery and those for whom surgical morbidity would be excessive. The time to regression after RT alone is often quite long and several years may elapse before regression is complete.410* In a number of reports, RT alone (50–60 Gy) or combined with surgery in patients with positive resection margins achieves longterm control in approximately 70–80 per cent of desmoids.403*,409*,410*,411*,412*,413*,414* The volume of disease does not appear to affect the probability of local control. In a comparative review of published experience with treatment of desmoid tumours, local control rates after surgery alone (n 381) were 61 per cent overall, and were 72 and 41 per cent for those with negative or positive margins, respectively.412*** For patients undergoing surgery plus RT (n 297), the local control rate was 75 per cent overall, and it was 94 and 75 per cent for cases with negative and positive margins, respectively. The overall local control rate for RT alone (n 102) was 78 per cent, and was 83 and 73 per cent for those treated for primary or recurrent tumours, respectively. The recommended dose of RT for definitive therapy is 50–60 Gy in 6–7 weeks at 1.8–2 Gy per fraction. Local recurrence rates do not appear to be reduced by the use of higher doses. In one study, for example, 23 patients were
964 Soft tissue sarcomas
treated with RT for unresectable disease; the relapse rate at 5 years was 31 per cent.413* Radiation doses above 56 Gy did not improve outcome, and were associated with more complications (30 vs. 5 per cent with lower doses at 15 years). Positive resection margins were not an adverse prognostic factor in this report. POSTOPERATIVE RADIATION THERAPY
Postoperative RT is generally not recommended for patients with uninvolved resection margins.403* The benefit of postoperative RT for those with positive resection margins is controversial. As noted previously, the status of the resection margins has not been shown to significantly increase the risk of recurrence in several retrospective series.407–409 Furthermore, even in those series that did show a higher rate of recurrence in patients with positive margins in the absence of RT, successful salvage therapy at the time of recurrence has been possible in the majority of patients.401*,402*,412* From these data, many clinicians conclude that local failure is not inevitable if residual tumour is left in situ. Deferring radiation is an acceptable option for patients with microscopically positive margins as long as local progression, if it occurred, would not risk significant morbidity. Postoperative irradiation is generally recommended for microscopically positive margins after resection of recurrent disease and for patients with gross or macroscopic residual disease,403* although some others recommend earlier use of radiation for incompletely resected disease.412* NEOADJUVANT THERAPY
A novel approach to reducing the rate of local recurrence is the use of neoadjuvant (preoperative) chemo-radiotherapy. In one small series, 13 patients with potentially resectable desmoid tumours received doxorubicin 30 mg by continuous infusion daily for three days concurrent with RT (10 3 Gy), with resection performed 4–6 weeks later.415* With a median follow-up of 71 months, there were only two local recurrences (15 per cent). While these results are promising, confirmation with larger, prospective, randomized trials is needed. SYSTEMIC THERAPY
Patients with desmoid tumours have been treated with a variety of agents, including NSAIDs, hormone manipulation, interferon, cytotoxic chemotherapy and pirfenidone (5-methyl-1-phenyl-2-(1H)-pyridone), an anti-fibroblast agent. Most of the reported data come from isolated case reports, limiting the conclusions that can be drawn as to the relative effectiveness of these agents in the treatment of desmoid tumours. Non-cytotoxic systemic therapy Clinical and experimental evidence suggests the hormone dependency of desmoid growth. Particularly in patients
who do not have good options for surgery or radiation, therapy is often begun with a hormonal agent such as tamoxifen, the anti-oestrogen toremifene, raloxifene, or pro-gestational agents.416*,417*,418*,419*,420*,421* An objective response is reported in approximately 50 per cent of patients, mostly partial rather than complete. Response durations range from 7 months to 12 years.421* The mechanism underlying benefit is unclear since responses to tamoxifen have been seen in desmoids that do not express hormone receptors.422 There are also documented responses to NSAIDs (most often sulindac), both alone and in combination with tamoxifen.416*,419*,422*,423*,424* Although response rates as high as 70 per cent are reported with combined therapy,422* regression is usually partial and may take many months after an initial period of tumour enlargement. In one report, 10 of 13 patients with FAP-associated desmoids responded to daily tamoxifen (120 mg daily) and sulindac (300 mg daily), as did two of eight recurrent sporadic tumours.422* The contribution of tamoxifen to these results is unclear. At least one report documents the resolution of a desmoid tumour being treated with indometacin and ascorbic acid for 14 months.425* NSAIDs alone represent a potentially attractive treatment option, particularly since they also appear to protect against colon cancer. Several case reports describe objective response or prolonged periods of disease stabilization with interferonalpha,426*,427* in some cases following failure of sulindac and tamoxifen.410*,428* At least one case report suggests clinical and radiographic benefit from the tyrosine kinase inhibitor imatinib (Gleevec\rtm;), an effect that is presumably due to tumour expression of activated receptor tyrosine kinases c-kit and/or platelet-derived growth factor receptor (PDGFR).429* A study of imatinib in 50 patients with recurrent desmoid tumours was completed, and preliminary data were presented at the 2004 meeting of the American Society of Clinical Oncology.430 The 2- and 4-month progressionfree survival rates were 91 and 78 per cent, respectively. Polymorphisms and mutations were found in exon 18 of PDGFR-α in some patients. Maturation of these data will be needed to determine the true clinical efficacy of imatinib in patients with desmoid tumours. Many of the above clinical studies using non-cytotoxic agents pre-date the widespread use of objective response criteria such as the response criteria for solid tumours (RECIST) criteria. Most experts treating large numbers of these recurrent tumours find that the true objective regression rate with these agents is in the range of 10–15 per cent with another 25 per cent of patients experiencing minor shrinkage or tumour stabilization (clinical benefit rate 50 per cent).431* Chemotherapy Chemotherapy may be effective in patients with unresectable tumours that are refractory to tamoxifen and/or sulindac. Low doses of methotrexate and vinblastine produce worthwhile response rates, particularly in children.406*,432*,433*
Desmoid tumours 965
In one report of 30 patients (age range 4–68, median 27), weekly methotrexate (30 mg/m2) and vinblastine (6 mg/m2) for between 4 and 20 months resulted in a partial response in 12 (40 per cent), and 18 others had stable disease or a minor response with symptomatic relief.406* The 10-year progression-free survival was 67 per cent. The combination of methotrexate and vinorelbine may produce a similar clinical benefit with less neurotoxicity.434* Liposomal doxorubicin may also be a useful agent.435* More aggressive doxorubicin- or ifosfamide-based regimens are more active, but also more toxic; thus, they are usually reserved for cases that are refractory to other therapies.436*,437* In one of the largest published series, 11 patients received doxorubicin (60–90 mg/m2) plus dacarbazine (750–1000 mg/m2) for a median of five cycles.436* Six of nine evaluable patients had an objective response (two complete, four partial). One complete responder died while the remainder were alive and progression-free. Others have demonstrated activity in patients with FAPassociated desmoids using 4–5 cycles of an infusional regimen of doxorubicin (20 mg/m2 daily) plus dacarbazine (150 mg/m2 daily), both for 96 hours every 28 days, followed sequentially by daily administration of an NSAID (meloxicam).438* Three of seven patients with hormone non-responsive desmoid tumours had a complete response, while four others had a partial response. The median progression-free survival was 74 months. In a report of eight patients with desmoid tumours and Gardner’s syndrome, alternating doxorubicin/ dacarbazine and carboplatin/dacarbazine resulted in two complete and four partial remissions, some durable.439* Since randomized trials are not available, the optimal regimen is uncertain. In general, an approach which matches the expected toxicity of systemic therapy with the degree of symptoms caused by the tumour or the rate of tumour growth would appear to be the most reasonable. INTESTINAL TRANSPLANTATION
●
●
●
●
●
Intestinal transplantation has been successful in a few cases of unresectable intra-abdominal desmoids.387* PERCUTANEOUS CHEMICAL ABLATION
At least one report suggests that percutaneous injection of acetic acid into the desmoid can achieve objective tumour shrinkage and prolonged periods of disease stabilization.440* Further experience with this approach is needed.
KEY LEARNING POINTS ●
Treatment for STS requires individual tailoring of the approach because of the wide variety of clinical situations that can arise from a tumour that involves a variety of anatomic sites with a
●
●
range of histologies of variable grade and size. Nevertheless, the following suggestions can serve as useful guides. Surgery is always indicated but the use of adjuvant therapy can vary according to the anatomic site, size and histological grade. In general, patients with extremity, superficial low-grade tumours that are less than 5 cm in diameter can be treated with surgical excision alone, and can expect excellent local control and survival rates approximating 90 per cent. In patients with intermediate-grade lesions, surgical excision with negative margins in combination with radiotherapy has achieved excellent local control with overall survival rates approximating 80 per cent. For larger, deepseated tumours, preoperative radiation therapy may facilitate resection and permits smaller fields and lower doses than postoperative radiation. Acute wound healing complications are higher with preoperative radiation therapy for lower extremity lesions, but generally irreversible late complications, including grade 3–4 fibrosis, are more common in those patients receiving postoperative radiation therapy. In patients with high-grade STS greater than 5 cm, excellent local control can be achieved with surgery and radiotherapy, but at least 50 per cent of these patients will develop metastatic disease. In this setting, the use of adjuvant chemotherapy may benefit some and should be considered, where possible in the context of a clinical trial, to be combined with surgery and adjuvant or neoadjuvant radiotherapy. Brachytherapy can provide excellent local control and functional result in appropriately selected patients. Complete surgical resection is the most important component of treatment for retroperitoneal sarcomas. Nevertheless, local recurrences occur in a substantial proportion of patients after surgery. Adjuvant radiotherapy may be of benefit; if employed, preoperative radiotherapy is preferred. The management of gastrointestinal stromal tumours is rapidly evolving with the development of targeted chemotherapy agents. Desmoid tumours are locally infiltrative, benign fibrous neoplasms that are most commonly managed initially with surgery if a functionsparing procedure can be performed. Radiation therapy can be used to manage recurrent lesions, often in conjunction with additional surgery. Non-cytoxic and cytotoxic systemic chemotherapy is available to manage recurrent desmoid tumours.
966 Soft tissue sarcomas
REFERENCES 1 Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55:10–30. 2 Weiss SW, Goldblum JR. Enzinger and Weiss’s Soft Tissue Tumors, 4th ed. Philadelphia: Mosby, 2001. 3 Abbas JS, Holyoke ED, Moore R, Karakousis CP. The surgical treatment and outcome of soft-tissue sarcoma. Arch Surg 1981; 116:765–9. 4 Potter DA, Glenn J, Kinsella T, et al. Patterns of recurrence in patients with high-grade soft tissue sarcomas. J Clin Oncol 1985; 3:353–66. 5 Torosian MH, Friedrich C, Godbold J, Hajdu SI, Brennan MF. Soft-tissue sarcoma: initial characteristics and prognostic factors in patients with and without metastatic disease. Semin Surg Oncol 1988; 4(1):13–9. 6 Lawrence W Jr, Donegan WL, Natarajan N, Mettlin C, Beart R, Winchester D. Adult soft tissue sarcomas. A pattern of care survey of the American College of Surgeons. Ann Surg 1987; 205(4):349–59. ●7 Sarcoma Meta-analysis Collaboration. Adjuvant chemotherapy for localised resectable soft-tissue sarcoma of adults: meta-analysis of individual data. Lancet 1997; 350:1647–54. ●8 Frustaci S, Gherlinzoni F, DePaoli A, et al. Adjuvant chemotherapy for adult soft tissue sarcomas of the extremities and girdles: results of the Italian Cooperative Trial. J Clin Oncol 2001; 19:1238–47. 9 Strong LC. Genetic considerations in pedi-oncology. In: Sutow WW (ed.) Clinical Pediatric Oncology. St Louis: CV Mosby, 1977, 16–32. 10 Rowley JD. Seminars from the University of Minnesota. Chromosome translocations: dangerous liaisons. J Lab Clin Med 1998; 132(4):244–50. 11 Littlefield JW. Genes, chromosomes and cancer. J Pediat 1984; 104:489–94. ●12 Li FP, Fraumeni JF Jr, Mulvihill JJ, et al. A cancer family syndrome in twenty-four kindreds. Cancer Res 1988; 48:5358–62. 13 McAdam WAF, Goligher JC. The occurrence of desmoids in patients with familial polyposis coli. Br J Surg 1970; 57:618–31. 14 Fraumeni JF. Genetic factors in the etiology of cancer. In: Holland JF, Frei EM (eds) Cancer Medicine. Philadelphia: Lea and Febiger, 1973, 7–15. 15 Zoller ME, Rembeck B, Oden A, Samuelsson M, Angervall L. Malignant and benign tumors in patients with neurofibromatosis type 1 in a defined Swedish population. Cancer 1997; 79:2125–31. 16 Colman SD, Williams CA, Wallace MR. Benign neurofibromas in type 1 neurofibromatosis (NF1) show somatic deletions of the NF1 gene. Nat Genet 1995; 11:90–2. 17 Derkinderen DJ, Koten JW, Nagelkerke NJ, Tan KE, Beemer FA. Non-ocular cancer in patients with hereditary retinoblastoma. Int J Cancer 1988; 41:499–504.
18 Wong FL, Boice JD Jr, Abramson DH, et al. Cancer incidence after retinoblastoma. Radiation dose and sarcoma risk. JAMA 1997; 278:1262–7. 19 Pizzo PA, Miser JS, Cassady JR, Filler RM. Solid tumors in childhood. In: DeVita VT, Hellman S, Rosenberg SA (eds) Cancer: Principles and Practice of Oncology. Philadelphia: JB Lippincott, 1985, 1511–89. 20 Li FP, Fraumeni JF Jr. Soft tissue sarcomas, breast cancer, and other neoplasms. A familial syndrome? Ann Intern Med 1969; 71:747–52. 21 Hisada M, Garber JE, Fung CY, Fraumeni JF Jr. Multiple primary cancers in families with Li–Fraumeni syndrome. J Natl Cancer Inst 1998; 90:606–11. 22 Evans SC, Lozano G. The Li–Fraumeni syndrome: An inherited susceptibility to cancer. Mol Med Today 1997; 13:390–5. 23 Varley JM, McGown G, Thorncroft M, et al. Germ-line mutations of TP53 in Li–Fraumeni families: An extended study of 39 families. Cancer Res 1997; 57:3245–52. 24 Malkin D. p53 and the Li–Fraumeni syndrome. Cancer Genet Cytogenet 1993; 66:83–92. 25 Bell DW, Varley JM, Szydlo TE, et al. Heterozygous germ line hCHK2 mutations in Li–Fraumeni syndrome. Science 1999; 286(5449):2528–31. 26 Hartley AL, Birch JM, Blair V, Kelsey AM, Harris M, Jones PH. Patterns of cancer in the families of children with soft tissue sarcoma. Cancer 1993; 72:923–30. 27 Birch JM, Hartley AL, Blair V, et al. Cancer in the families of children with soft tissue sarcoma. Cancer 1990; 66:2239–48. ◆28 Helman LJ, Meltzer P. Mechanisms of sarcoma development. Nat Rev Cancer 2003; 3(9):685–94. 29 Borden EC, Baker LH, Bell RS, et al. Soft tissue sarcomas of adults: state of the translational science. Clin Cancer Res 2003; 9(6):1941–56. 30 Turc-Carel C, Dalcin P, Limon J. Translocation of x;18 in synovial sarcoma. Cancer Genet Cytogenet 1986; 22:93–4. 31 Clark J, Rocques PJ, Crew AJ, et al. Identification of novel genes, SYT and SSX, involved in the t(X;18)(p11.2;q11.2) translocation found in human synovial sarcoma. Nat Genet 1994; 7:502–8. 32 de Alava E, Kawai A, Healey JH, et al. EWS–FLI1 fusion transcript structure is an independent determinant of prognosis in Ewing’s sarcoma. J Clin Oncol 1998; 16:1248–55. 33 Kawai A, Noguchi M, Beppu Y. Nuclear immunoreaction of p53 protein in soft tissue sarcomas. A possible prognostic factor. Cancer 1994; 73:2499–505. 34 Turc-Carel C, Aurias A, Mugneret F, et al. Chromosomes in Ewing’s sarcoma. I. An evaluation of 85 cases of remarkable consistency of t(11;22)(q24;ql2). Cancer Genet Cytogenet 1988; 32:229–38. 35 Ladanyi M, Heinemann FS, Huvos AG, Rao PH, Chen OG, Jhanawar SC. Neural differentiation in small round cell tumors of bone and soft tissue with the translocation
References 967
36
37
38
39
40
41
42
43
44
45
46
47
48
49
t(11;22)(q24;q12): An immunohistochemical study of 11 cases. Hum Pathol 1990; 21:1245–51. Bailly RA, Bosselut R, Zucman J, et al. DNA-binding and transcriptional activation properties of the EWS–FLI-1 fusion protein resulting from the t(11;22) translocation in Ewing sarcoma. Mol Cell Biol 1994; 14(5):3230–41. May WA, Gishizky ML, Lessnick SL, et al. Ewing sarcoma 11;22 translocation produces a chimeric transcription factor that requires the DNA-binding domain encoded by FLI1 for transformation. Proc Natl Acad Sci USA 1993; 90(12):5752–6. Delattre O, Zucman J, Plougastel B, et al. Gene fusion with an ETS DNA-binding domain caused by chromosome translocation in human tumors. Nature 1992; 359:162–5. Sorensen PH, Lessnick SL, Lopez-Terrada D, Liu XF, Triche TJ, Denny CT. A second Ewing’s sarcoma translocation, t(21;22), fusion the EWS gene to another ETS-family transcription factor, ERG. Nat Genet 1994; 6:146–51. Crozat A, Aman P, Mandahl N, Ron D. Fusion of CHOP to a novel RNA binding protein in human myxoid liposarcoma. Nature 1993; 363:640–4. Rabbitts TH, Forster A, Larson R, Nathan P. Fusion of the dominant negative transcription regulator CHOP with a novel gene FUS by translocation t(12;16) in malignant liposarcoma. Nat Genet 1993; 4:175–80. Hisaoka M, Tsuji S, Morimitsu Y, Shimajiri S, Komiya S, Ushijima M. Detection of TLS/FUS–CHOP fusion transcripts in myxoid and round cell liposarcomas by nested reverse transcription-polymerase chain reaction using archival paraffin-embedded tissues. Diagn Mol Pathol 1998; 7:96–101. Aman P, Ron D, Mandahl N, et al. Rearrangement of the transcription factor gene CHOP in myxoid liposarcomas with t(12;16)(q13;p11). Genes Chromosomes Cancer 1992; 5:278–85. Barone MV, Crozat A, Tabaee A, Philipson L, Ron D. CHOP (GADD153) and its oncogenic variant, TLS–CHOP, have opposing effects on the induction of G1/S arrest. Genes Dev 1994; 8:453–64. Barr FG, Nauta LE, Hollows JC. Structural analysis of PAX3 genomic rearrangements in alveolar rhabdomyosarcoma. Cancer Genet Cytogenet 1998; 102:32–9. Davis RJ, Barr FG. Fusion genes resulting from alternative chromosomal translocations are overexpressed by genespecific mechanisms in alveolar rhabdomyosarcoma. Proc Natl Acad Sci USA 1997; 94:8047–51. Kelly KM, Womer RB, Sorensen PH, Xiong QB, Barr FG. Common and variant gene fusions predict distinct clinical phenotypes in rhabdomyosarcoma. J Clin Oncol 1996; 15:1831–6. Ginsberg JP, Davis RJ, Bennicelli JL, Nauta LE, Barr FG. Up-regulation of MET but not neural adhesion molecule expression by the PAX3–FKHR fusion protein in alveolar rhabdomyosarcoma. Cancer Res 1998; 58:3542–6. Kelly KM, Womer RB, Barr FG. Minimal disease detection in patients with alveolar rhabdomyosarcoma using a
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
reverse transcriptase polymerase chain reaction method. Cancer 1996; 78:1320–7. Fletcher J. Translocation (12;22)(q13-14;q12) is a nonrandom aberration in soft-tissue clear-cell sarcoma. Genes Chromosomes Cancer 1992; 5:184. Travis JA, Bridge JA. Significance of both numerical and structural chromosomal abnormalities in clear cell sarcoma. Cancer Genet Cytogenet 1992; 64:104–6. Ladanyi M, Lui MY, Antonescu CR, et al. The der(17)t(X;17)(p11;q25) of human alveolar soft part sarcoma fuses the TFE3 transcription factor gene to ASPL, a novel gene at 17q25. Oncogene 2001; 20(1):48–57. Ladanyi M, Bridge JA. Contribution of molecular genetic data to the classification of sarcomas. Hum Pathol 2000; 31(5):532–8. de Alava E, Antonescu CR, Panizo A, et al. Prognostic impact of P53 status in Ewing sarcoma. Cancer 2000; 89(4):783–92. Wei G, Antonescu CR, de Alava E, et al. Prognostic impact of INK4A deletion in Ewing sarcoma. Cancer 2000; 89(4):793–9. McIntyre JF, Smith-Sorensen B, Friend SH, et al. Germline mutations of the p53 tumor suppressor gene in children with osteosarcoma. J Clin Oncol 1994; 12:925–30. Toguchida J, Yamaguchi T, Dayton SH, et al. Prevalence and spectrum of germline mutations of the p53 gene among patients with sarcoma [see comments]. N Engl J Med 1992; 326(20):1301–8. Wadayama B, Toguchida J, Yamaguchi T, Sasaki MS, Yamamuro T. p53 expression and its relationship to DNA alterations in bone and soft tissue sarcomas. Br J Cancer 1993; 68(6):1134–9. Porter PL, Gown AM, Kramp SG, Coltrera MD. Widespread p53 overexpression in human malignant tumors. An immunohistochemical study using methacarn-fixed, embedded tissue. Am J Pathol 1992; 140:145–53. Soini Y, Vahakangas K, Nuorva K, Kamel D, Lane DP, Paakko P. p53 immunohistochemistry in malignant fibrous histiocytomas and other mesenchymal tumours. J Pathol 1992; 168:29–33. Stratton MR, Moss S, Warren W, et al. Mutation of the p53 gene in human soft tissue sarcomas: Association with abnormalities of the RB1 gene. Oncogene 1990; 5:1297–301. Mulligan LM, Matlashewski GJ, Scrable HJ, Cavenee WK. Mechanisms of p53 loss in human sarcomas. Proc Natl Acad Sci USA 1990; 87:5863–7. Patterson H, Gill S, Fisher C, et al. Abnormalities of the p53, MDM2, and DCC genes in human leiomyosarcomas. Br J Cancer 1994; 69:1052–58. Andreassen A, Oyjord T, Hovig E, et al. p53 abnormalities in different subtypes of human sarcomas. Cancer Res 1993; 53:468–71. Blom R, Guerrieri C, Sta°l O, Malmstrom H, Simonsen E. Leiomyosarcoma of the uterus: A clinicopathologic, DNA flow cytometric, p53 and mdm-2 analysis of 49 cases. Gynecol Oncol 1998; 68:54–61.
968 Soft tissue sarcomas
66 Simms WW, Ordonez NG, Johnston D, Ayala AG, Czerniak B. p53 expression in dedifferentiated chondrosarcoma. Cancer 1995; 76:223–7. 67 Pollock RE, Lang A, Luo J, El-Naggar AK, Yu D. Soft tissue sarcoma metastasis from clonal expansion of p53 mutated tumor cells. Oncogene 1996; 12:2035–9. 68 Malkin D, Folly KW, Barbier N, et al. Germline mutations of the p53 tumor suppressor gene in children and young adults with second malignant neoplasms. N Engl J Med 1992; 326:1309–15. 69 El-Deiry WS, Tokino T, Velculesco VE. WAF1, a potential mediator of p53 suppression. Cell 1993; 75:817–25. 70 Lane DP. P53, guardian of the genome. Nature 1992; 358:15–16. 71 Levine AJ, Pery ME, Chang A, et al. The 1993 Walter Hubert Lecture: The role of the p53 tumour-suppressor gene in tumorigenesis. Br J Cancer 1994; 69:409–16. 72 Toguchida J, Yamaguchi T, Ritchie B, et al. Mutation spectrum of the p53 gene in bone and soft tissue sarcomas. Cancer Res 1992; 52(22):6194–9. 73 Donehower LA, Harvey M, Slagle BL, et al. Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours. Nature 1992; 356:215–21. 74 Lee JM, Abrahamson JL, Kandel R, Donehower LA, Bernstein A. Susceptibility to radiation carcinogenesis and accumulation of chromosomal breakage in p53 deficient mice. Oncogene 1994; 9:3731–6. 75 Pollock R, Lang A, Ge T, Sun D, Tan M, Yu D. Wild-type p53 and a p53 temperature-sensitive mutant suppress human soft tissue sarcoma by enhancing cell cycle control. Clin Cancer Res 1998; 4:1985–94. 76 Lessnick SL, Dacwag CS, Golub TR. The Ewing’s sarcoma oncoprotein EWS/FLI induces a p53-dependent growth arrest in primary human fibroblasts. Cancer Cell 2002; 1(4):393–401. 77 Rorie CJ, Weissman BE. The Ews/Fli-1 fusion gene changes the status of p53 in neuroblastoma tumor cell lines. Cancer Res 2004; 64(20):7288–95. 78 Oliner JD, Kinzler KW, Meltzer PS, George DL, Vogelstein B. Amplification of a gene encoding a p53-associated protein in human sarcomas. Nature 1992; 358:80–3. 79 Florenes VA, Maelandsmo GM, Forus A, Andreassen A, Myklebost O, Fodstad O. MDM2 gene amplification and transcript levels in human sarcomas: Relationship to TP53 gene status. J Natl Cancer lnst 1994; 86:1297–302. 80 Khatib ZA, Matsushime H, Valentine M, Shapiro DN, Sherr CJ, Look AT. Coamplification of the CDK4 gene with MDM2 and GLI in human sarcomas. Cancer Res 1993; 53:5535–41. 81 Nilbert M, Rydholm A, Willen H, Mitelman F, Mandahl N. MDM2 gene amplification correlates with ring chromosomes in soft tissue tumors. Genes Chromosomes Cancer 1994; 9:261–5. ●82 Cavenee WK, Hansen MF, Nordenskjold M, et al. Genetic origin of mutations predisposing to retinoblastoma. Science 1985; 228:501–3.
83 Friend SH, Horowitz JM, Gerber MR, et al. Deletions of a DNA sequence in retinoblastomas and mesenchymal tumors: organization of the sequence and its encoded protein. Proc Natl Acad Sci USA 1987; 84:9059–63. 84 Cance WG, Brennan MF, Dudas M, Huang CM, Cordon-Cardo C. Altered expression of the retinoblastoma gene product in human sarcomas. N Engl J Med 1990; 323:1457–62. 85 Karpeh MS, Brennan MF, Cance WG, et al. Altered patterns of retinoblastoma gene product expression in adult softtissue sarcomas. Br J Cancer 1995; 72:986–91. 86 Wunder JS, Czitrom AA, Kandel R, Andrulis IL. Analysis of alterations in the retinoblastoma gene and tumor grade in bone and soft-tissue sarcomas. J Natl Cancer Inst 1991; 83:194–200. 87 Shew JY, Ling N, Yang X, Fodstad O, Lee WH. Antibodies detecting abnormalities of the retinoblastoma susceptibility gene product (pplIORB) in osteosarcomas and synovial sarcomas. Oncogene Res 1989; 1:205–14. 88 Duda RB, Cundiff D, August CZ, Wagman ED, Bauer KD. Growth factor receptor and related oncogene determinations in mesenchymal tumors. Cancer 1993; 71:3526–30. 89 Kanoe H, Nakayama T, Murakami H, et al. Amplification of the CDK4 gene in sarcomas: tumor specificity and relationship with the RB gene mutation. Anticancer Res 1998; 18(4A):2317–21. 90 Brown R, Marshall CJ, Pennie SG, Hall A. Mechanism of activation of an N-ras gene in the human fibrosarcoma cell line HT1080. EMBO J 1984; 3:1321–6. 91 Chardin P, Yermian P, Madaule P, Tavitian A. N-ras gene activation in the RD human rhabdomyosarcoma cell line. Int J Cancer 1985; 35:647–52. 92 Kato M, Toguchida J, Honda K, et al. Elevated frequency of a specific allele of the L-myc gene in male patients with bone and soft-tissue sarcomas. Int J Cancer 1990; 45(1):47–9. 93 Jankowski SA, Mitchell DS, Smith SH, Trent JM, Meltzer PS. SAS, a gene amplified in human sarcomas, encodes a new member of the transmembrane 4 superfamily of proteins. Oncogene 1994; 9:1205–11. 94 Smith SH, Weiss SW, Jankowski SA, Coccia MA, Meltzer PS. SAS amplification in soft tissue sarcomas. Cancer Res 1992; 52:3746–9. 95 Golub TR, Slonim DK, Tamayo P, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999; 286(5439):531–7. ●96 Nielsen TO, West RB, Linn SC, et al. Molecular characterisation of soft tissue tumours: a gene expression study. Lancet 2002; 359(9314):1301–7. 97 Segal NH, Pavlidis P, Antonescu CR, et al. Classification and subtype prediction of adult soft tissue sarcoma by functional genomics. Am J Pathol 2003; 163:691–700. 98 Ohali A, Avigad S, Zaizov R, et al. Prediction of high risk Ewing’s sarcoma by gene expression profiling. Oncogene 2004; 23(55):8997–9006.
References 969
99 Segal NH, Pavlidis P, Noble WS, et al. Classification of clear-cell sarcoma as a subtype of melanoma by genomic profiling. J Clin Oncol 2003; 21(9):1775–81. 100 Kim JH. Radiation induced soft tissue sarcoma and bone sarcoma. Radiology 1978; 129:501–8. 101 Sadove AM, Block M, Rossof AH. Radiation carcinogenesis in man: new primary neoplasms in fields of prior therapeutic radiation. Cancer 1981; 48:1139–43. 102 Robinson E, Neugut A, Wylie P. Clinical aspects of post-irradiation sarcomas. J Natl Cancer Inst 1988; 80:233–40. 103 Laskin T. Postradiation soft tissue sarcomas, an analysis of 53 cases. Cancer 1988; 62:2330–40. 104 Taghian A, deVathaire F, Terrier P, et al. Long-term risk of sarcoma following radiation treatment for breast cancer. Int J Radiat Oncol Biol Phys 1991; 21:361–7. 105 Karlsson P, Holmberg E, Johansson KA, Kindblom LG, Carstensen J, Wallgren A. Soft tissue sarcoma after treatment for breast cancer. Radiother Oncol 1996; 38:25–31. 106 Cohen RJ, Curtis RE, Inskip PD, Fraumeni JF Jr. The risk of developing second cancers among survivors of childhood soft tissue sarcoma. Cancer 2005; 103(11):2391–6. 107 Eng C, Li FP, Abramson DH, Ellsworth RM, et al. Mortality from second tumors among long-term survivors of retinoblastoma. J Natl Cancer Inst 1993; 85:1121–1128. 108 Hawkins MM, Wilson LM, Burton HS, et al. Radiotherapy, alkylating agents, and risk of bone cancer after childhood cancer. J Natl Cancer Inst 1996; 88:270–8. 109 Scaradavou A, Heller G, Sklar CA, Ren L, Ghavimi F. Second malignant neoplasms in long-term survivors of childhood rhabdomyosarcoma. Cancer 1995; 76:1860–67. 110 Shaw PJ, Bergin M, Stevens M. Osteogenic sarcoma following acute lymphoblastic leukemia. Am J Pediat Hemat Oncol 1988; 10:81–7. 111 Tucker MA, D’Angio GJ, Boice JD, et al. Bone sarcomas linked to radiotherapy and chemotherapy in children. N Engl J Med 1987; 317:588–93. 112 Sathiakumar N, Delzell E. A review of epidemiologic studies of triazine herbicides and cancer. Crit Rev Toxicol 1997; 27:599–612. 113 Dich J, Zah SH, Hanberg A, Adami HO. Pesticides and cancer. Cancer Causes Control 1997; 8:420–43. 114 Lee FI, Smith PM, Bennett B, Williams DM. Occupationally related angiosarcoma of the liver in the United Kingdom 1972–1994. Gut 1996; 39:312–8. 115 Hardell L, Eriksson M. The association between soft tissue sarcoma and exposure to phenoxyacetic acids. Cancer 1988; 62:652–6. 116 Lander JJ, Stanley RJ, Sumner HW, Boswell DC, Aach RD. Angiosarcoma of the liver associated with Fowler’s solution (potassium arsenite). Gastroenterol 1975; 68:1582–6. 117 Wingren G, Fredrikson M, Brage N, Nordenskjold B, Axelson O. Soft tissue sarcoma and occupational exposures. Cancer 1990; 66:806–11.
118 Vineis P, Faggiano F, Tedeschi M, Ciccone G. Incidence rates of lymphomas and soft tissue sarcomas and environmental measurements of phenoxy herbicides. J Natl Cancer Inst 1991; 83:362–3. ●119 Kogevinas M, Becher H, Benn T, et al. Cancer mortality in workers exposed to phenoxy herbicides, chlorophenols and dioxins. An expanded and updated international cohort study. Am J Epidemiol 1997; 145:1061–75. 120 Fingerhut MA, Halperin WE, Marlow DA, et al. Cancer mortality in workers exposed to 2,3,7,8tetrachlorodibenzo-p-dioxin. N Engl J Med 1991; 324:212–8. ●121 Group TSCCS. The association of selected cancers with service in the US military in Vietnam. II. Soft tissue and other sarcomas. Arch Intern Med 1990; 150:2485–92. 122 Hoppin JA, Tolbert PE, Herrick RF, et al. Occupational chlorophenol exposure and soft tissue sarcoma risk among men aged 30–60 years. Am J Epidemiol 1998; 148:693–703. 123 Stewart FW, Treves N. Lymphangiosarcoma in postmastectomy lymphedema: A report of six cases in elephantiasis chirurgica. Cancer 1948; 1:64–81. 124 Tomita K, Yokogawa A, Oda Y, Terahata S. Lymphangiosarcoma in postmastectomy lymphedema (Stewart–Treves syndrome): Ultrastructural and immunohistologic characteristics. J Surg Oncol 1988; 38:275–82. 125 Muller R, Hajdu SI, Brennan MF. Lymphangiosarcoma associated with chronic filarial lymphedema. Cancer 1987; 59:179–83. 126 Mazeron JJ, Suit HD. Lymph nodes as sites of metastasis from sarcomas of soft tissue. Cancer 1987; 60:1800–8. 127 Fong Y, Coal DG, Woodruff JM, Brennan MF. Lymph node metastasis from soft tissue sarcoma in adults. Analysis of data from a prospective data base of 1772 sarcoma patients. Ann Surg 1993; 217:72–7. 128 Bacci G, Capanna R, Orlandi M, et al. Prognostic significance of serum lactic acid dehydrogenase in Ewing’s tumor of bone. Ric Clin Lab 1985; 15:89–96. 129 Bland KI, McCoy DM, Kinard RE, Copeland EM 3rd. Application of magnetic resonance imaging and computerized tomography as an adjunct to the surgical management of soft tissue sarcomas. Ann Surg 1987; 205:473–81. 130 Kransdorf MJ, Jelinek JS, Moser RP. Imaging of soft tissue tumors. Radiol Clin North Am 1993; 31:359–71. 131 O’Keefe F, Lorigan JG, Wallace S. Radiological features of extraskeletal Ewing sarcoma. Br J Radiol 1990; 63:456–60. 132 Gustafson P, Herrlin K, Biling L, Willen H, Rydholm A. Necrosis observed on CT enhancement is of prognostic value in soft tissue sarcoma. Acta Radiol 1992; 33:474–76. 133 Nieweg OE, Pruim J, van Ginkel RJ, et al. Fluorine-18fluorodeoxyglucose PET imaging of soft-tissue sarcoma. J Nucl Med 1996; 37:257–61. 134 Prosnitz LR, Maguire P, Anderson JM, et al. The treatment of high-grade soft tissue sarcomas with preoperative thermoradiotherapy. Int J Radiat Oncol Biol Phys 1999; 45(4):941–9.
970 Soft tissue sarcomas
●135
136
137
138
139
140
141
142
●143
●144
145
146
147 148 149
150
Schuetze SM, Rubin BP, Vernon C, et al. Use of positron emission tomoography in localized extremity soft tissue sarcoma treated with neoadjuvant chemotherapy. Cancer 2005; 103(2):339–48. Peuchot M, Libshitz HI. Pulmonary metastatic disease: radiologic-surgical correlation. Radiology 1987; 164:719–22. Sanchez RB, Quinn SF, Walling A, Estrada J, Greenberg H. Musculoskeletal neoplasms after intraarterial chemotherapy: correlation of MR images with pathologic specimens. Radiology 1990; 174:237–40. Vanel D, Lacombe MJ, Couanet D, Kalifa C, Spielmann M, Genin J. Musculoskeletal tumors follow-up with MR imaging after treatment with surgery and radiation therapy. Radiology 1987; 164:243–5. Negendank WG, Crowley MG, Ryan JR, Keller NA, Evelhoch JL. Bone and soft tissue lesions: diagnosis with combined H-1 MR imaging and P-31 spectroscopy. Radiology 1989; 173:181–7. Dewhirst MW, Sostman HD, Leopold KA, et al. Soft-tissue sarcomas: MR imaging and MR spectrosopy for prognosis and therapy monitoring. Radiology 1990; 174:847–53. Koutcher JA, Ballon D, Graham M, et al. 31P NMR spectra of extremity sarcomas: diversity of metabolic profiles and changes in reponse to chemotherapy. Magn Reson Med 1990; 16:19–34. Redmond OM, Bell E, Stack JP, et al. Tissue characterization and assessment of preoperative chemotherapeutic response in musculoskeletal tumors by in vivo 31P magnetic resonance spectroscopy. Magn Reson Med 1992; 27:226–37. Gayed I, Vu T, Iyer R, et al. The role of 18F-FDG PET in staging and early prediction of response to therapy of recurrent gastrointestinal stromal tumors. J Nucl Med 2004; 45(1):17–21. Barth RJ Jr, Merino MJ, Solomon D, Yang JC, Baker AR. A prospective study of the value of core needle biopsy and fine needle aspiration in the diagnosis of soft tissue masses. Surgery 1992; 112(3):536–43. Heslin MJ, Lewis JJ, Woodruff JM, Brennan MF. Core needle biopsy for diagnosis of extremity soft tissue sarcoma. Ann Surg Oncol 1997; 4(5):425–31. Hau A, Kim I, Kattapuram S, et al. Accuracy of CT-guided biopsies in 359 patients with musculoskeletal lesions. Skeletal Radiol 2002; 31(6):349–53. Epub 2002 Mar 7. Shives TC. Biopsy of soft-tissue tumors. Clin Orthop Related Res 1993; 289:32–5. Ball ABS, Fisher C, Pittam M. Diagnosis of soft tissue tumours by Tru-Cut biopsy. Br J Surg 1990; 77:756–8. Welker JA, Henshaw RM, Jelinek JS, Shmookler BM, Malawer MM. The percutaneous needle biopsy is safe and recommended in the diagnosis of musculoskeletal masses. Cancer 2000; 89:2677–86. Hoeber I, Spillane AJ, Fisher C, Thomas JM. Accuracy of biopsy techniques for limb and limb girdle soft tissue tumors. Ann Surg Oncol 2001; 8(1):80–7.
151 Ray-Coquard I, Ranchere-Vince D, Thiesse P, et al. Evaluation of core needle biopsy as a substitute to open biopsy in the diagnosis of soft-tissue masses. Eur J Cancer. 2003; 39(14):2021–5. 152 Cannon SR, Dyson PHP. Relationship of the site of open biopsy of malignant bone tumours to local recurrence following resection and prosthetic replacement. J Bone Joint Surg 1987; 69-B:492. 153 Sotiriou C, Khanna C, Jazaeri AA, Petersen D, Liu ET. Core biopsies can be used to distinguish differences in expression profiling by cDNA microarrays. J Mol Diagn 2002; 4(1):30–6. 154 Singh HK, Kilpatrick SE, Silverman JF. Fine needle aspiration biopsy of soft tissue sarcomas: utility and diagnostic challenges. Adv Anat Pathol 2004; 11(1):24–37. 155 Domanski HA, Akerman M, Carlen B, et al. Core-needle biopsy performed by the cytopathologist: a technique to complement fine-needle aspiration of soft tissue and bone lesions. Cancer 2005; 105(4):229–39. ●156 Mankin HJ, Lange TA, Spanier SS. The hazards of biopsy in patients with malignant primary bone and soft-tissue tumors. J Bone Joint Surg Am 1982; 64:1121–7. 157 Angervall L, Kindblom LG. Principles for pathologicanatomic diagnosis and classification of soft-tissue sarcomas. Clin Orthop 1993; (289):9–18. 158 Dias P, Dealing M, Houghton P. The molecular basis of skeletal muscle differentiation. Semin Diagn Pathol 1994; 11:3–14. 159 Suit HD, Russell WO, Martin RG. Sarcoma of soft tissue: clinical and histopathologic parameters and response to treatment. Cancer 1975; 35:1478–83. 160 Myhre-Jensen, Kaae S, Madsen EH, Sneppen O. Histopathological grading in soft tissue tumors. Acta Path Microbiol Immunol Scand Sect. A 1975; 91:145–50. 161 Rydholm A, Berg NO, Gullber B, Thorngren KG, Persson BM. Epidemiology of soft tissue sarcoma in the locomotor system. Acta Pathol Microbiol Immunol Scand Sect. A 1984; 92:363–74. ●162 Costa J, Wesley RA, Glatstein E, Rosenberg SA. The grading of soft tissue sarcomas. Results of a clinicohistopathologic correlation in a series of 163 cases. Cancer 1984; 53:530–41. 163 Trojani M. Staging system for soft tissue and bone. Int J Cancer 1984; 33:37–42. 164 Lack EE, Steinberg SM, White DE, et al. Extremity soft tissue sarcomas: analysis of prognostic variables in 300 cases and evaluation of tumor necrosis as a factor in stratifying higher grade sarcomas. J Surg Oncol 1989; 41:263–73. 165 Kulander BG, Polissar L, Yang CY, Woods JS. Grading of soft tissue sarcomas: necrosis as a determinate of survival. Mod Pathol 1989; 2:205–8. 166 Russell WO, Cohen J, Enzinger FM, et al. A clinical and pathological staging system for soft tissue sarcomas. Cancer 1977; 40:1562–70.
References 971
167 Ushigome S, Shimoda T, Nikaido T, Takasaki S. Histopathologic diagnostic and histogenetic problems in malignant soft tissue tumors. Reassessment of malignant fibrous histiocytoma, epithelioid sarcoma, malignant rhabdoid tumor, and neuroectodermal tumor. Japanese Society of Pathology 1992; 42:691–706. 168 Miettinen M. Immunohistochemistry of soft-tissue tumors. Possibilities and limitations in surgical pathology. Pathol Annu 1988; 25:1–36. 169 Wick MR, Swanson PE, Manivel JC. Immunohistochemical analysis of soft tissue sarcomas. Appl Pathol 1988; 6:169–96. 170 Weinstein RS, Kuszak JR, Kluskens LF, Coon JS. P-glycoproteins in pathology: the multidrug resistance gene family in humans. Hum Pathol 1990; 21:34–48. 171 Elias AD. Chemotherapy for soft-tissue sarcomas. Clin Orthop Relat Res 1993; 289:94–105. 172 Chan HSL, Thorner PS, Haddad G, Ling V. Immunohistochemical detection of P-glycoprotein: prognostic correlation in soft tissue sarcoma of childhood. J Clin Oncol 1990; 8:689–704. 173 Sahin AA, Ro JY, El-Naggar AK, et al. Tumor proliferative fraction in solid malignant neoplasms. A comparative study of Ki-67 immunostaining and flow cytometric determinations. Am J Clin Pathol 1991; 96:512–9. 174 Swanson SA, Brooks IJ. Proliferation markers Ki-67 and p105 in soft-tissue lesions. Correlation with DNA flow cytometric characteristics. Am J Pathol 1990; 137:1491–500. 175 Ueda T, Aozasa K, Tsujimoto M, et al. Prognostic significance of Ki-67 reactivity in soft tissue sarcomas. Cancer 1989; 63:1607–11. 176 Zehr RJ, Bauer TW, Marks KE, Weltevreden A. Ki-67 and grading of malignant fibrous histiocytomas. Cancer 1990; 66:1984–90. 177 Dei Tos AP, Doglioni C, Laurino L, Barbareschi M, Fletcher CD. p53 protein expression in non-neoplastic lesions and benign and malignant neoplasms of soft tissue. Histopathology 1993; 22:45–50. 178 Palman C, Bowen-Pope DF, Brooks JJ. Platelet-derived growth factor receptor (P-subunit) immunoreactivity in soft tissue tumors. Lab Invest 1992; 66:108–15. 179 Roholl PJM, Skottner A, Prinsen I, Lips CJ, Den Otter W, Van Unnik JA. Expression of insulin-like growth factor I in sarcomas. Histopathology 1990; 16:455–60. 180 Agarwal V, Greenebaum E, Wersto R. DNA ploidy of spindle cell soft-tissue tumors and its relationship to histology and clinical outcome. Arch Pathol Lab Med 1991; 115:558–62. 181 Kroese MCS, Rutgers DH, Wils IS, Van Unnik JA, Roholl PJ. The relevance of the DNA index and proliferation rate in the grading of benign and malignant soft tissue tumors. Cancer 1990; 65:1782–8. 182 Bodensteiner D, Reidinger D, Rosenfeld C, Neff JR, Lin F. Flow cytometry of needle aspirates from bone and soft tissue tumors. Southern Med J 1991; 84:1451–4. 183 Parham DM. The molecular biology of childhood rhabdomyosarcoma. Semin Diagn Pathol 1994; 11(1):39–46.
184 DeChiara A, T’Ang A, Triche TJ. Expression of the retinoblastoma susceptibility gene in childhood rhabdomyosarcomas. J Natl Cancer Inst 1993; 85:152–7. 185 Sainati L, Stella M, Montaldi A, et al. Value of cytogenetics in the differential diagnosis of the small round cell tumors of childhood. Med Pediatr Oncol 1992; 20:130–5. 186 Markhede G, Angervall L, Stener B. A multivariate analysis of the prognosis after surgical treatment of malignant soft tissue tumors. Cancer 1982; 40:1721–33. 187 Simon MA, Enneking WF. The management of soft-tissue sarcomas of the extremities. J Bone Joint Surg Am 1976; 58(3):317–27. ●188 Rosenberg SA, Tepper J, Glatstein E, et al. The treatment of soft-tissue sarcomas of the extremities: prospective randomized evaluations of (1) limb-sparing surgery plus radiation therapy compared with amputation and (2) the role of adjuvant chemotherapy. Ann Surg 1982; 196(3):305–15. 189 Lindberg RD, Martin RG, Romsdahl MM, Barkley HT Jr. Conservative surgery and postoperative radiotherapy in 300 adults with soft-tissue sarcomas. Cancer 1981; 47:2391–97. 190 DeLaney TF, Rosenberg AE, Harmon DC, et al. Soft tissue sarcomas. In: Price PM, Sikora K (eds) Treatment of Cancer, 4th ed. London: Arnold, 2003, 869–907. 191 Brant TA, Parsons JT, Marcus RB, et al. Preoperative irradiation for soft tissue sarcomas of the trunk and extremities in adults. Int J Radiat Oncol Biol Phys 1990; 19:899–906. 192 Karakousis CP, Emrich LJ, Rao U, Krishnamsetty RM. Feasibility of limb salvage and survival in soft tissue sarcomas. Cancer 1986; 57:484–91. ●193 Harrison LB, Franzese F, Gaynor JJ, Brennan MF. Long-term results of a prospective randomized trial of adjuvant brachytherapy in the management of completely resected soft tissue sarcomas of the extremity and superficial trunk. Int J Radiat Oncol Biol Phys 1993; 27:259–65. ●194 Eilber FC, Rosen G, Nelson SD, et al. High-grade extremity soft tissue sarcomas: factors predictive of local recurrence and its effect on morbidity and mortality. Ann Surg 2003; 237(2):218–26. 195 Williard WC, Hajdu SI, Casper ES, Brennan MF. Comparison of amputation with limb-sparing operations for adult soft tissue sarcoma of the extremity. Ann Surg 1992; 215:389–96. 196 LeVay J, O’Sullivan B, Catton C, et al. Outcome and prognostic factors in soft tissue sarcoma in the adult. Int J Radiat Oncol Biol Phys 1993; 27:1091–9. 197 Karakousis CP, Driscoll DL. Treatment and local control of primary extremity soft tissue sarcomas. J Surg Oncol 1999; 71(3):155–61. ✾198 Pisters PW, O’Sullivan B. Retroperitoneal sarcomas: combined modality treatment approaches. Curr Opin Oncol 2002; 14(4):400–5. 199 Cantin J, McNeer GP, Chu FC, Booher RJ. The problem of local recurrence after treatment of soft tissue sarcoma. Ann Surg 1968; 168:47–53.
972 Soft tissue sarcomas
200 Martin RG, Butler JJ, Albores-Saavedra J. Soft tissue tumors – surgical treatment and results. In: Tumors of Bone and Soft Tissue. Chicago: Year Book Medical Publishers, 1965, 333–48. 201 Enneking WF, Spanier SS, Malawar MD. The effect of anatomic setting on the results of surgical procedures for soft parts sarcoma of the thigh. Cancer 1981; 47:1005–22. ●202 Sadoski C, Suit HD, Rosenberg A, Mankin H, Efird J. Preoperative radiation, surgical margins, and local control of extremity sarcomas of soft tissues. J Surg Oncol 1993; 52(4):223–30. ●203 Tanabe KK, Pollock RE, Ellis LM, Murphy A, Sherman NE, Romsdahl MM. Influence of surgical margins on outcome in patients with preoperatively irradiated extremity soft tissue sarcomas. Cancer 1994; 73:1652–9. ●204 Pisters PW, Leung DH, Woodruff J, Shi W, Brennan MF. Analysis of prognostic factors in 1041 patients with localized soft tissue sarcomas of the extremities. J Clin Oncol 1996; 14:1679–89. 205 Trovik CS, Bauer HC, Alvegard TA, et al. Surgical margins, local recurrence and metastasis in soft tissue sarcomas: 559 surgically-treated patients from the Scandinavian Sarcoma Group Register. Eur J Cancer 2000; 36:710–6. 206 Zagars GK, Ballo MT, Pisters PW, et al. Prognostic factors for patients with localized soft-tissue sarcoma treated with conservation surgery and radiation therapy: an analysis of 225 patients. Cancer 2003; 97(10):2530–43. 207 Gronchi A, Casali PG, Mariani L, et al. Status of surgical margins and prognosis in adult soft tissue sarcomas of the extremities: a series of patients treated at a single institution. J Clin Oncol 2005; 23(1):96–104. 208 Alvegard TA, Sigurdsson H, Mouridsen H, et al. Adjuvant chemotherapy with doxorubicin in high-grade soft tissue sarcoma: a randomized trial of the Scandinavian Sarcoma Group. J Clin Oncol 1989; 7(10):1504–13. 209 Fagundes HM, Lai PP, Dehner LP, et al. Postoperative radiotherapy for malignant fibrous histiocytoma. Int J Radiat Oncol Biol Phys 1992; 23:615–9. 210 Wiklund TA, Alvegard TA, Mouridsen HT, Rydholm A, Blomqvist CP. Marginal surgery and postoperative radiotherapy in soft tissue sarcomas. The Scandinavia Sarcoma Group experience. Eur J Cancer 1993; 29:306–9. 211 Herbert SH, Corn BW, Solin LJ, Lanciano RM, Schultz DJ, McKenna WG. Limb-preserving treatment for soft tissue sarcomas of the extremities. Cancer 1993; 72:1230–8. ✾212 Eilber FR, Eckardt J. Surgical management of soft tissue sarcomas. Semin Oncol 1997; 24(5):526–33. 213 Noria S, Davis A, Kandel R, et al. Residual disease following unplanned excision of soft-tissue sarcoma of an extremity. J Bone Joint Surg Am 1996; 78(5):650–5. 214 Zagars GK, Ballo MT, Pisters PW, Pollock RE, Patel SR, Benjamin RS. Surgical margins and reresection in the management of patients with soft tissue sarcoma using conservative surgery and radiation therapy. Cancer 2003; 97(10):2544–53.
●215
●216
●217
218
●219
220
221
222
●223
●224
●225
●226
227
●228
229
Kepka L, DeLaney TF, Goldberg SI, Suit HD. Results of radiation therapy for unresected soft tissue sarcomas. Int J Radiat Oncol Biol Phys 2005; 63:852–9. Rydholm A, Gustafson P, Rooser B, et al. Limb-sparing surgery without radiotherapy based on anatomic location of soft tissue sarcoma. J Clin Oncol 1991; 9:1757–65. Baldini EH, Goldberg J, Jenner C, et al. Long-term outcomes after functon-sparing surgery without radiotherapy for soft tissue sarcoma of the extremities and trunk. J Clin Oncol 1999; 17:3252–9. Suit HD, Mankin HJ, Wood WC, Proppe KH. Radiation and surgery in the treatment of primary sarcoma of soft tissue: pre-operative, intra-operative and post-operative. Cancer 1985; 55:2659–67. Todoroki T, Suit HD. Therapeutic advantage in preoperative single-dose radiation combined with conservative and radical surgery in different-size murine fibrosarcomas. J Surg Oncol 1985; 29(4):207–15. Atkinson L, Garvan JM, Newton NC. Behavior and management of soft connective tissue sarcomas. Cancer 1963; 16:1552–62. Barkley HT, Martin RG, Romsdahl MM, Lindberg R, Zagars GK. Treatment of soft tissue sarcomas by preoperative irradiation and conservative surgical resection. Int J Radiat Oncol Biol Phys 1988; 14:693–99. Schray MF, Gunderson LL, Sim FH, Pritchard DJ, Shives TC, Yeakel PD. Soft tissue sarcomas. Integration of brachytherapy, resection, and external irradiation. Cancer 1990; 66:451–6. Yang JC, Chang AE, Baker AR, et al. Randomized prospective study of the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcomas of the extremity. J Clin Oncol 1998; 16(1):197–203. Nielsen OS, Cummings B, O’Sullivan B, Catton C, Bell RS, Fornasier VL. Preoperative and postoperative irradiation of soft tissue sarcomas: effect on radiation field size. Int J Radiat Oncol Biol Phys 1991; 21:1595–9. O’Sullivan B, Davis AM, Turcotte R, et al. Preoperative versus postoperative radiotherapy in soft-tissue sarcoma of the limbs: a randomized trial. Lancet 2002; 359:2235–41. Davis AM, O’Sullivan B, Bell RS, et al. Function and health status outcomes in a randomized trial comparing preoperative and postoperative radiotherapy in extremity soft tissue sarcoma. J Clin Oncol 2002; 20(22):4472–7. O’Sullivan B, Davis AM, Turcotte R, et al. Five-year results of a randomized phase III trial of pre-operative vs. post-operative radiotherapy in extremity soft tissue sarcoma. Proc Am Soc Clin Oncol 2004; 23:815. Pisters PW, Harrison LB, Leung DH, Woodruff JM, Casper ES, Brennan MF. Long-term results of a prospective randomized trial of adjuvant brachytherapy in soft tissue sarcoma. J Clin Oncol 1996; 14(3):859–68. Alektiar KM, Zelefsky MJ, Brennan MF. Morbidity of adjuvant brachytherapy in soft tissue sarcoma of the extremity and superficial trunk. Int J Radiat Oncol Biol Phys 2000; 47(5):1273–9.
References 973
230 Panagopoulos I, Hoglund M, Mertens F, Mandahl N, Mitelman F, Aman P. Fusion of EWS and CHOP genes in myxoid liposarcoma. Oncogene 1996; 12:489–94. 231 Kretzler A, Molls M, Gradinger R, et al. Intraoperative radiotherapy of soft tissue sarcoma of the extremity. Perioperative fractionated high-dose rate brachytherapy in the treatment of soft tissue sarcomas. Strahlenther Onkol 2004; 180(6):365–70. ◆232 Nag S, Shasha D, Janjan N, Petersen I, Zaider M. The American Brachytherapy Society recommendations for brachytherapy of soft tissue sarcomas. Int J Radiat Oncol Biol Phys 2001; 49(4):1033–43. 233 Petera J, Neumanova R, Odrazka K, et al. Perioperative fractionated high-dose rate brachytherapy in the treatment of soft tissue sarcomas. Neoplasma 2004; 51(1):59–63. 234 Niewald M, Berberich W, Schnabel K, Lehmann W, Pastyr O. A simple method for positioning and fixing the extremities during the radiotherapy of soft-tissue sarcomas. Strahlenther Onkol 1990; 166(4):295–6. 235 Tepper J, Rosenberg SA, Glatstein E. Radiation therapy technique in soft tissue sarcomas of the extremity – policies of treatment at the National Cancer Institute. Int J Radiat Oncol Biol Phys 1982; 8(2):263–73. 236 Suit HD, Mankin HJ, Wood WC, et al. Treatment of the patient with stage M0 sarcoma of soft tissue. J Clin Oncol 1988; 6:854–62. 237 Mundt AJ, Awan A, Sibley GS, et al. Conservative surgery and adjuvant radiation therapy in the management of adult soft tissue sarcoma of the extremities: clinical and radiobiological results. Int J Radiat Oncol Biol Phys 1995; 32:977–85. 238 Cleator SJ, Cottril C, Harmer C. Pattern of local recurrence after conservative surgery and radiotherapy for soft tissue sarcoma. Sarcoma 2001; 5:83–8. ●239 White LM, Wunder JS, Bell RS, et al. Histologic assessment of peritumoral edema in soft tissue sarcoma. Int J Radiat Oncol Biol Phys 2005; 61(5):1439–45. ●240 Zagars GK, Ballo MT. Significance of dose in postoperative radiotherapy for soft tissue sarcoma. Int J Radiat Oncol Biol Phys 2003; 56(2):473–81. 241 Fein DA, Lee WR, Lanciano RM, et al. Management of extremity soft tissue sarcomas with limb-sparing surgery and postoperative irradiation: do total dose, overall treatment time, and the surgery-radiotherapy interval impact on local control? Int J Radiat Oncol Biol Phys 1995; 32:969–76. 242 Bell RS, O’Sullivan B, Liu FF, et al. The surgical margin in soft tissue sarcoma. J Bone Joint Surg 1989; 71:370–5. 243 Pao WJ, Pilepich MV. Postoperative radiotherapy in the treatment of extremity soft tissue sarcomas. Int J Radiat Oncol Biol Phys 1990; 19:907–11. 244 Robinson M, Barr L, Fisher C, et al. Treatment of extremity soft tissue sarcomas with surgery and radiotherapy. Radiother Oncol 1990; 18:221–33. 245 Robinson MH, Ball AB, Schofield J, Fisher C, Harmer CL, Thomas JM. Preoperative radiotherapy for initially
●246
✾247
248
●249
250
●251
252
253
254
255
●256
257
●258
●259
inoperable extremity soft tissue sarcomas. Clin Oncol (R Coll Radiol) 1992; 4:36–43. Stinson SF, DeLaney TF, Greenberg J, et al. Acute and long term effects on limb function of combined modality limb sparing therapy for extremity soft tissue sarcomas. Int J Radiat Oncol Biol Phys 1991; 21:1493–99. Verhey LJ. Comparison of three-dimensional conformal radiation therapy and intensity-modulated radiation therapy systems. Semin Radiat Oncol 1999; 9(1):78–98. Chan MF, Chui CS, Schupak K, Amols H, Burman C, Ling CC. The treatment of large extraskeletal chondrosarcoma of the leg: comparison of IMRT and conformal radiotherapy techniques. J Appl Clin Med Phys 2001; 2:3–8. Hong L, Alektiar KM, Hunt M, Venkatraman E, Leibel SA. Intensity-modulated radiotherapy for soft tissue sarcoma of the thigh. Int J Radiat Oncol Biol Phys 2004; 59:752–9. Lee N, Chuang C, Quivey JM, et al. Skin toxicity due to intensity-modulated radiotherapy for head-and-neck carcinoma. Int J Radiat Oncol Biol Phys 2002; 53(3):630–7. Wilson RR. Radiological uses of fast protons. Radiology 1946; 47:487–91. Austin-Seymour M, Munzenrider J, Goitein M, et al. Fractionated proton radiation therapy of chordoma and low grade chondrosarcoma of the base of skull. J Neurosurg 1989; 70:13–17. DeLaney TF, Chung CS, Trofimov A, et al. Comparison of intensity modulated photon and proton radiation therapy plans for treatment of retroperitoneal sarcoma. Proceedings of the Proton Therapy Cooperative Oncology Group, December 2005. Eilber FC, Rosen G, Eckardt J, et al. Treatment-induced pathologic necrosis: a predictor of local recurrence and survival in patients receiving neoadjuvant therapy for high-grade extremity soft tissue sarcomas. J Clin Oncol 2001; 19(13):3203–9. Wanebo HJ, Temple WJ, Popp MB, Constable W, Aron B, Cunningham SL. Preoperative regional therapy for extremity sarcoma. A tricenter update. Cancer 1995; 75:2299–306. Eilber F, Eckardt J, Rosen G, Forscher C, Selch M, Fu YS. Preoperative therapy for soft tissue sarcoma. Hematol Oncol Clin North Am 1995; 9(4):817–23. Merimsky O, Meller I, Issakov J, et al. Adriamycinifosfamide induction chemotherapy for extremity soft tissue sarcoma: comparison of two non-randomized protocols. Oncol Rep 1999; 6(4):913–20. DeLaney TF, Spiro IJ, Suit HD, et al. Neoadjuvant chemotherapy and radiotherapy for large extremity softtissue sarcomas. Int J Radiat Oncol Biol Phys 2003; 56:1117–27. Kraybill WG, Harris J, Spiro IJ, et al. Phase II study of neoadjuvant chemotherapy and radiation therapy in the management of high-risk, high-grade, soft tissue sarcomas of the extremities and body wall: Radiation Therapy Oncology Group Trial 9514. J Clin Oncol 2006; 24(4):619–25.
974 Soft tissue sarcomas
●260
261
262
263
264
265
266
267
268
269
●270
271
●272
Schraffordt Koops H, Eggermont AM, Lienard D, et al. Hyperthermic isolated limb perfusion for the treatment of soft tissue sarcomas. Semin Surg Oncol 1998; 14:210–4. Lejeune FJ, Lienard D, el Douaihy M, Seyedi JV, Ewalenko P. Results of 206 isolated limb perfusions for malignant melanoma. Eur J Surg Oncol 1989; 15:510–9. Eggermont AM, Koops HS, Lienard D, et al. Isolated limb perfusion with high-dose tumor necrosis factor-alpha in combination with interferon-gamma and melphalan for nonresectable extremity soft tissue sarcomas: A multicenter trial. J Clin Oncol 1996; 14:2653–65. Vaglini M, Belli F, Ammatuna M, et al. Treatment of primary or relapsing limb cancer by isolation perfusion with high-dose alpha-tumor necrosis factor, gammainterferon, and melphalan. Cancer 1994; 73:483–92. Olieman AF, Pras E, van Ginkel RJ, Molenaar WM, Schraffordt Koops H, Hoekstra HJ. Feasibility and efficacy of external beam radiotherapy after hyperthermic isolated limb perfusion with TNF-alpha and melphalan for limb-saving treatment in locally advanced extremity soft-tissue sarcoma. Int J Radiat Oncol Biol Phys 1998; 40:807–14. Vrouenraets BC, Keus RB, Nieweg OE, Kroon BB. Complications of combined radiotherapy and isolated limb perfusion with tumor necrosis factor alpha interferon gamma and melphalan in patients with irresectable soft tissue tumors. J Surg Oncol 1997; 65:88–94. Lynch EM, Sampson LE, Khalil AA, Horsman MR, Chaplin DJ. Cytotoxic effect of tumour necrosis factor-alpha on sarcoma F cells at tumour relevant oxygen tensions. Acta Oncol 1995; 34:423–7. Gutman M, Inbar M, Lev-Shiush D, et al. High dose tumor necrosis factor alpha and melphalan administered via isolated limb perfusion for advanced limb soft tissue sarcoma results in a 90 per cent response rate and limb preservation. Cancer 1997; 79:1129–37. Lans TE, Grunhagen DJ, de Wilt JH, van Geel AN, Eggermont AM. Isolated limb perfusions with tumor necrosis factor and melphalan for locally recurrent soft tissue sarcoma in previously irradiated limbs. Ann Surg Oncol 2005; 12(5):406–11. Talbert ML, Zagars GK, Sherman NE, Romsdahl MM. Conservative surgery and radiation therapy for soft tissue sarcoma of the wrist, hand, ankle, and foot. Cancer 1990; 66(12):2482–91. Johnstone PAS, Wexler LH, Venzon DJ, et al. Sarcomas of the hand and foot: analysis of local control and functional result with combined modality therapy in extremity preservation. Int J Radiat Oncol Biol Phys 1994; 29:735–45. Karakousis CP, De Young C, Driscoll DL. Soft tissue sarcomas of the hand and foot: management and survival. Ann Surg Oncol 1998; 5(3):238–40. Bujko K, Suit HD, Springfield DS, Convery K. Wound healing after surgery and preoperative radiation for sarcoma of soft tissues. Surg Gynecol Obstet 1992; 176:124–34.
273 Arbeit JM, Hilaris B, Brennan MF. Wound complications in the multimodality treatment of extremity and superficial truncal sarcomas. J Clin Oncol 1987; 5:480–8. 274 Gottlieb JA, Baker LH, Quagliana JM, et al. Chemotherapy of sarcomas with a combination of adriamycin and dimethyl triazeno imidazole carboxamide. Cancer 1972; 30:1632–8. 275 Townsend CM, Eilber FR, Morton DL. Skeletal and soft tissue sarcomas. results of surgical adjuvant chemotherapy. Proc Am Ass Cancer Res 1976; 17:265. 276 Sordillo PP, Magill GB, Shiu MH, Lesser M, Hajdu SI, Golbery RB. Adjuvant chemotherapy of adult soft-part sarcomas with ALOMAD (S4). J Surg Oncol 1981; 19:345–53. ●277 Rosenberg SA, Kent H, Costa J, et al. Prospective randomised evaluation of the role of limb-sparing surgery, radiation therapy and adjuvant chemoimmunotherapy in the treatment of adult soft-tissue sarcomas. Surgery 1978; 84:62–9. 278 das Gupta TK, Patel MK, Chaudhuri PK, Briele HA. The role of chemotherapy as an adjuvant to surgery in the initial treatment of primary soft tissue sarcomas in adults. J Surg Oncol 1982; 19:139–44. 279 Antman KH. Survival of patients with localized high-grade soft tissue sarcoma with multimodality therapy. A matched control study. Cancer 1983; 51:396–401. ●280 Edmonson JH, Fleming TR, Ivins JC, et al. Randomized study of systemic chemotherapy following complete excision of nonosseous sarcomas. J Clin Oncol 1984; 2:1390–6. ●281 Rosenberg SA, Tepper J, Glatstein E, et al. Prospective randomized evaluation of adjuvant chemotherapy in adults with soft tissue sarcoma of the extremities. Cancer 1983; 52:424–34. ●282 Chang AE, Kinsella T, Glatstein E, et al. Adjuvant chemotherapy for patients with high-grade soft-tissue sarcomas of the extremity. J Clin Oncol 1988; 6:1491–500. ●283 Picci P, Bacci G, Gherlinzoni F, et al. Results of a randomized trial for the treatment of localized soft tissue sarcoma (STS) of the extremities in adult patients. In: Ryan JR, Baker LO (eds) Recent Concepts in Sarcoma Treatment. Boston, MA: Kluwer Academic Publishers, 1988, 144–8. ●284 Wilson RE, Wood WC, Lerner HL, et al. Doxorubicin chemotherapy in the treatment of soft tissue sarcoma. Combined results of two randomized trials. Arch Surg 1986; 121:1354–9. ●285 Eilber FG, Giuliano A, Huth J, Mirra JJ, Rosen G, Morton DL. Neoadjuvant chemotherapy, radiation, and limited surgery for high grade soft tissue sarcoma of the extremity. Recent concepts in sarcoma treatment. In: Ryan JR, Baker LH (eds) Recent Concepts in Sarcoma Treatment. Dordrecht, The Netherlands: Kluwer Academic Publishers, 1988, 115–22. ●286 Bramwell V, Rouesse J, Steward W, et al. Adjuvant CYVADIC chemotherapy for adult soft tissue sarcoma – reduced local recurrence but no improvement in survival: a study of the European Organization for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group. J Clin Oncol 1994; 12:1137–49.
References 975
●287
288
●289
●290
291
292
293
294
295
●296
●297
298
299
300
301
302
Sarcoma Meta-analysis Collaboration Adjuvant chemotherapy for localised resectable soft-tissue sarcoma of adults: meta-analysis of individual data. Lancet 1997; 350:1647–54. Spiro IJ, Suit H, Gebhardt M, et al. Neoadjuvant chemotherapy and radiotherapy for large soft tissue sarcomas. Proc Am Soc Clin Oncol 1996; 15:524. Brodowicz T, Schwameis E, Widder J, et al. Intensified adjuvant IFADIC chemotherapy for adult soft tissue sarcoma. A prospective randomized feasibility trial. Sarcoma 2000; 4:151. Petrioli R, Coratti A, Correale P, et al. Adjuvant epirubicin with or without ifosfamide for adult soft-tissue sarcoma. Am J Clin Oncol 2002; 25:468–73. Cormier JN, Huang X, Xing Y, et al. Cohort analysis of patients with localized, high-risk, extremity soft tissue sarcoma treated at two cancer centers: chemotherapyassociated outcomes. J Clin Oncol 2004; 22:4567–74. Grobmyer SR, Maki RG, Demetri GD, et al. Neo-adjuvant chemotherapy for primary high-grade extremity soft tissue sarcoma. Ann Oncol 2004; 15:1667–72. Ferrari A, Gronchi A, Casanova M, et al. Synovial sarcoma: a retrospective analysis of 271 patients of all ages treated at a single institution. Cancer 2004; 101:627–34. Coindre JM, Terrier P, Bui NB, et al. Prognostic factors in adult patients with locally controlled soft tissue sarcoma. A study of 546 patients from the French Federation of Cancer Centers Sarcoma Group. J Clin Oncol 1996; 14(3):869–77. Gerrand CH, Wunder JS, Kandel RA, et al. The influence of anatomic location on functional outcome in lowerextremity soft-tissue sarcoma. Ann Surg Oncol 2004; 11(5):476–82. Epub 2004 Apr 12. Robinson MH, Spruce L, Eeles R, et al. Limb function following conservation treatment of adult soft tissue sarcoma. Eur J Cancer 1991; 27:1567–74. Davis AM, Devlin M, Griffin AM, Wunder JS, Bell RS. Functional outcome in amputation versus limb sparing of patients with lower extremity sarcoma: a matched casecontrol study. Arch Phys Med Rehabil 1999; 80(6):615–8. Weddington WW Jr, Segraves KB, Simon MA. Psychological outcome of extremity sarcoma survivors undergoing amputation or limb salvage. J Clin Oncol 1985; 3(10):1393–9. Sugarbaker PH, Barofsky I, Rosenberg SA, Gianola FJ. Quality of life assessment of patients in extremity sarcoma clinical trials. Surgery 1982; 91:17–23. Catton C, Davis A, Bell R, et al. Soft tissue sarcoma of the extremity. Limb salvage after failure of combined conservative therapy. Radiother Oncol 1996; 41(3):209–14. Karakousis CP, Proimakis C, Rao U, Velez AF, Driscoll DL. Local recurrence and survival in soft-tissue sarcomas. Ann Surg Oncol 1996; 3(3):255–60. Ueda T, Yoshikawa H, Mori S, et al. Influence of local recurrence on the prognosis of soft-tissue sarcomas. J Bone Joint Surg Br 1997; 79(4):553–7.
303 Catton C, Davis A, Bell R, et al. Soft tissue sarcoma of the extremity. Limb salvage after failure of combined conservative therapy. Radiother Oncol 1996; 41(3):209–14. 304 Pearlstone DB, Janjan NA, Feig BW, et al. Re-resection with brachytherapy for locally recurrent soft tissue sarcoma arising in a previously radiated field [see comments]. Cancer J Sci Am 1999; 5(1):26–33. 305 Cade S. Soft tissue tumours: their natural history and treatment. Proc R Soc Med 1951; 44:19–36. 306 Windeyer B, Dische S, Mansfield CM. The place of radiotherapy in the management of fibrosarcoma of the soft tissues. Clin Radiol 1966; 17:32–40. 307 McNeer GP, Cantin J, Chu F, Nickerson JJ. Effectiveness of radiation therapy in management of sarcoma of soft somatic tissues. Cancer 1968; 22:391–7. ●308 Tepper JE, Suit HD. Radiation therapy alone for sarcoma of soft tissue. Cancer 1985; 56:475–9. 309 Pickering DG, Stewart JS, Rampling R, Errington RD, Stamp G, Chia Y. Fast neutron therapy for soft tissue sarcoma. Int J Radiat Oncol Biol Phys 1987; 13(10):1489–95. 310 Schmitt G, Pape H, Zamboglou N. Long term results of neutron- and neutron-boost irradiation of soft tissue sarcomas. Strahlenther Onkol 1990; 166(1):61–2. 311 Steingraber M, Lessel A, Jahn U. Fast neutron therapy in treatment of soft tissue sarcoma – the Berlin-Buch study. Bull Cancer Radiother 1996; 83(Suppl):122s–4s. ●312 Goffman T, Tochner Z, Glatstein E. Primary treatment of large and massive adult sarcomas with iododeoxyuridine and aggressive hyperfractionated irradiation. Cancer 1991; 67(3):572–6. 313 Rhomberg W, Hassenstein EO, Gefeller D. Radiotherapy vs. radiotherapy and razoxane in the treatment of soft tissue sarcomas: final results of a randomized study. Int J Radiat Oncol Biol Phys 1996; 36(5):1077–84. 314 Mendenhall WM, Zlotecki RA, Hochwald SN, Hemming AW, Grobmyer SR, Cance WG. Retroperitoneal soft tissue sarcoma. Cancer 2005; 104(4):669–75. ●315 Lewis JJ, Leung D, Woodruff JM, Brennan MF. Retroperitoneal soft-tissue sarcoma: analysis of 500 patients treated and followed at a single institution. Ann Surg 1998; 228(3):355–65. 316 Stoeckle E, Coindre JM, Bonvalot S, et al. Prognostic factors in retroperitoneal sarcoma: a multivariate analysis of a series of 165 patients of the French Cancer Center Federation Sarcoma Group. Cancer 2001; 92(2):359–68. 317 Singer S, Antonescu CR, Riedel E, Brennan MF. Histologic subtype and margin of resection predict pattern of recurrence and survival for retroperitoneal liposarcoma. Ann Surg 2003; 238(3):358–70. 318 van Dalen T, Hoekstra HJ, van Geel AN, et al. Locoregional recurrence of retroperitoneal soft tissue sarcoma: second chance of cure for selected patients. Eur J Surg Oncol 2001; 27(6):564–8. 319 Russo P, Kim Y, Ravindran S, Huang W, Brennan MF. Nephrectomy during operative management of
976 Soft tissue sarcomas
●320
321
●322
323
324
✾325
326
327 328
329
330
331
●332
333
334
retroperitoneal sarcoma. Ann Surg Oncol 1997; 4(5):421–424. Petersen IA, Haddock MG, Donohue JH, et al. Use of intraoperative electron beam radiotherapy in the management of retroperitoneal soft tissue sarcomas. Int J Radiat Oncol Biol Phys 2002; 52(2):469–75. Jones JJ, Catton CN, O’Sullivan B, et al. Initial results of a trial of preoperative external-beam radiation therapy and postoperative brachytherapy for retroperitoneal sarcoma. Ann Surg Oncol 2002; 9(4):346–54. Gieschen HL, Spiro IJ, Suit HD, et al. Long-term results of intraoperative electron beam radiotherapy for primary and recurrent retroperitoneal soft tissue sarcoma. Int J Radiat Oncol Biol Phys 2001; 50(1):127–31. Robertson JM, Sondak VK, Weiss SA, Sussman JJ, Chang AE, Lawrence TS. Preoperative radiation therapy and iododeoxyuridine for large retroperitoneal sarcomas. Int J Radiat Oncol Biol Phys 1995; 31(1):87–92. Pisters PW, Ballo MT, Fenstermacher MJ, et al. Phase I trial of preoperative concurrent doxorubicin and radiation therapy, surgical resection, and intraoperative electronbeam radiation therapy for patients with localized retroperitoneal sarcoma. J Clin Oncol 2003; 21(16):3092–7. Corless CL, Fletcher JA, Heinrich MC. Biology of gastrointestinal stromal tumors. J Clin Oncol 2004; 22(18):3813–25. DeMatteo RP, Lewis JJ, Leung D, Mudan SS, Woodruff JM, Brennan MF. Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg 2000; 231(1):51–8. van der Zwan SM, DeMatteo RP. Gastrointestinal stromal tumor: 5 years later. Cancer 2005; 104(9):1781–8. Fletcher CD, Berman JJ, Corless C, et al. Diagnosis of gastrointestinal stromal tumors: a consensus approach. Int J Surg Pathol 2002; 10(2):81–9. Gadd MA, Casper ES, Woodruff J, McCormack PM, Brennan MF. Development and treatment of pulmonary metastases in adult patients with extremity soft tissue sarcoma. Ann Surg 1993; 218:705–12. Pearlstone DB, Pisters PW, Bold RJ, et al. Patterns of recurrence in extremity liposarcoma: Implications for staging and follow-up. Cancer 1999; 85:85–92. Spillane AJ, Fisher C, Thomas JM. Myxoid liposarcoma – the frequency and the natural history of nonpulmonary soft tissue metastases. Ann Surg Oncol 1999; 6:389–94. Billingsley KG, Lewis JJ, Leung DHY, Casper ES, Woodruff JM, Brennan MF. Multifactorial analysis of the survival of patients with distant metastasis arising from primary extremity sarcoma. Cancer 1999; 85:389–95. Pogrebniak HW, Roth JA, Steinberg SM, Rosenberg SA, Pass HI. Reoperative pulmonary resection in patients with metastatic soft tissue sarcoma. Ann Thorac Surg 1991; 52:197–203. Lang H, Nussbaum KT, Kaudel P, Fruhauf N, Flemming P, Raab R. Hepatic metastases from leiomyosarcoma: a
●335
336
◆337
●338
339
340
●341
342
●343
344
345
346
347
single-center experience with 34 liver resections during a 15-year period. Ann Surg 2000; 231:500–5. Blay J-Y, van Glabbeke M, Verweij J, et al. Advanced softtissue sarcoma: a disease that is potentially curable for a subset of patients treated with chemotherapy. Eur J Cancer 2003; 39:64–9. Judson I. Systemic therapy of soft tissue sarcoma: an improvement in outcome. Ann Oncol 2004; 15(Suppl 4):iv193–6. Maki RG. Role of chemotherapy in patients with soft tissue sarcomas. Expert Rev Anticancer Ther 2004; 4:229–36. Santoro A, Tursz T, Mouridsen H, et al. Doxorubicin versus CYVADIC versus doxorubicin plus ifosfamide in first-line treatment of advanced soft tissue sarcomas: a randomized study of the European Organization for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group. J Clin Oncol 1995; 13:1537–45. O’Bryan RM, Baker LH, Gottlieb JE, et al. Dose response evaluation of adriamycin in human neoplasia. Cancer 1977; 39:1940–8. Christman KL, Casper ES, Schwartz GK. High-intensity scheduling of ifosfamide in adult patients with soft-tissue sarcoma. Proc Ann Meet Am Soc Clin Oncol 1993; 12:A1642. van Oosterom AT, Mouridsen HT, Nielsen OS, et al. EORTC SoftTissue and Bone Sarcoma Group. Results of randomised studies of the EORTC Soft Tissue and Bone Sarcoma Group (STBSG) with two different ifosfamide regimens in first-and second-line chemotherapy in advanced soft tissue sarcoma patients. Eur J Cancer 2002; 38:2397–406. Gottlieb JA, Benjamin RS, Baker LH, et al. Role of DTIC (NSC45388) in the chemotherapy of sarcomas. Cancer Treat Rep 1976; 60:199–203. Bramwell VH, Mouridsen HT, Santoro A, et al. Cyclophosphamide versus ifosfamide: a randomized Phase II trial in adult soft-tissue sarcomas. The European Organization for Research and Treatment of Cancer [EORTC], Soft Tissue and Bone Sarcoma Group. Cancer Chemother Pharmacol 1993; 31(Suppl 2):S180–4. Hensley ML, Maki RG, Venkatraman E, et al. Gemcitabine and docetaxel in patients with unresectable leiomyosarcoma: results of a phase II trial. J Clin Oncol 2002; 20:2824–31. Fata F, O’Reilly E, Ilson D, et al. Paclitaxel in the treatment of patients with angiosarcoma of the scalp or face. Cancer 1999; 86:2034–7. Judson I, Radford JA, Harris M, et al. Randomised phase II trial of pegylated liposomal doxorubicin (DOXIL/CAELYX) versus doxorubicin in the treatment of advanced or metastatic soft tissue sarcoma: a study by the EORTC Soft Tissue and Bone Sarcoma Group. Eur J Cancer 2001; 37:870–7. Merimsky O, Meller I, Flusser G, et al. Gemcitabine in soft tissue or bone sarcoma resistant to standard chemotherapy: a phase II study. Cancer Chemother Pharmacol 2000; 45:177–81.
References 977
348 Fidias P, Demetri G, Harmon D. Navelbine shows activity in previously treated sarcoma patients: Phase II results from MGH/Dana Farber/Partner’s Cancer Care Study. Proc Am Soc Clin Oncol 1998; 17:513a (abstr). 349 Yovine A, Riofrio M, Blay JY, et al. Phase II study of ecteinascidin-743 in advanced pretreated soft tissue sarcoma patients. J Clin Oncol 2004; 22:890–9. 350 Subramanian S, Wiltshaw E. Chemotherapy of sarcoma. Lancet 1978; 1:683–6. 351 Pinedo HM, Verweij J. The treatment of soft tissue sarcomas with a focus on chemotherapy: a review. Radiother Oncol 1986; 6:193–205. 352 Karakousis CP, Holtermann OA, Holyoke ED. Cis-dichlorodiammineplatinum (II) in metastatic soft tissue sarcomas. Cancer Treat Rep 1979; 63:2071–5. 353 Grabois M, Frappaz D, Bouffet E. High-dose VP-16 cisplatinum in soft tissue sarcoma of children. Cancer Chemother Pharmacol 1994; 33:355–7. 354 Omura GA, Major FJ, Blessing JA, et al. A randomized trial of adriamycin with and without dimethyl triazenoimidazole carboxamide in advanced uterine sarcoma. Cancer 1983; 52:626–32. 355 Borden EC, Amato DA, Rosenbaum C, et al. Randomized comparison of three adriamycin regimens for metastatic soft tissue sarcomas. J Clin Oncol 1987; 5:840–50. 356 Zalupski M, Metch B, Balcerzak S, et al. Phase III comparison of doxorubicin and dacarbazine given by bolus versus infusion in patients with soft-tissue sarcomas: a Southwest Oncology Group study. J Natl Cancer Inst 1991; 83(13):926–32. 357 Blum RH, Corson JM, Wilson RE, Greenberger JS, Canellos GP, Frei E 3rd. Successful treatment of metastatic sarcomas with cyclophosphamide, adriamycin, and DTIC (CAD). Cancer 1980; 46:1722–6. 358 Baker LH, Frank J, Fine G, et al. Combination chemotherapy using adriamycin, DTIC, cyclophosphamide, and actinomycin D for advanced soft tissue sarcomas: a randomized comparative trial. A phase III, Southwest Oncology Group Study (7613). J Clin Oncol 1987; 5(6):851–61. 359 Cruz AB Jr, Thames EA Jr, Aust JB, et al. Combination chemotherapy for soft tissue sarcomas: a phase III study. J Surg Oncol 1979; 11:313–23. 360 Schoenfeld DA, Rosenbaum C, Horton J, Wolter JM, Falkson G, DeConti RC. A comparison of adriamycin versus vincristine and adriamycin, and cyclophosphamide versus vincristine, actinomycin-D, and cyclophosphamide for advanced sarcoma. Cancer 1982; 50:2757–62. 361 Blum RH, Edmonson J, Ryan L, Pelletier L. Efficacy of ifosfamide in combination with doxorubicin for the treatment of metastatic soft-tissue sarcoma. The Eastern Cooperative Oncology Group. Cancer Chemother Pharmacol 1993; 31(Suppl 2):S238–40. 362 Schutte J, Mouridsen HT, Steward W, et al. Ifosfamide plus doxorubicin in previously untreated patients with advanced soft-tissue sarcoma. Cancer Chemother Pharmacol 1993; 31(Suppl 2):S204–9.
●363
363a
364
365
366
367
368
369
370
371
372
373
374 375
376
Edmonson JH, Ryan LM, Blum RH, et al. Randomized comparison of doxorubicin alone versus ifosfamide plus doxorubicin or mitomycin, doxorubicin, and cisplatin against advanced soft tissue sarcomas. J Clin Oncol 1993; 11:1269–75. Gottlieb JA, Baker LJ, O’Bryan RM, et al. Adriamycin (NCS-123127) used alone and in combination for soft tissue and bone sarcomas. Cancer Chemother Rep 1975; 6:271–82. Greenall NJ, Magill GB, DeCosse JJ, Brennan MF. Chemotherapy for soft tissue sarcoma. Surg Gynecol Obstet 1986; 162:193–8. Yap BS, Sincovics JG, Burgess MA, Benjamim RS, Bodey GP. The curability of advanced soft tissue sarcomas in adults with chemotherapy. Proc Am Soc Clin Oncol 1983; 2:239. Elias AD, Antman KH. Doxorubicin, ifosfamide, and dacarbazine (AID) with mesna uroprotection for advanced untreated sarcoma: a phase I study. Cancer Treat Rep 1986; 70:827–33. Chang AE, Rosenberg SA, Glatstein EJ, Antman KH. Sarcomas of soft tissues. In: DeVita VT, Hellman S, Rosenberg SA (eds) Cancer: Principles and Practice of Oncology, 3rd ed. Philadelphia: JB Lippincott, 1989, 1345. Elias A, Ryan L, Sulkes A, Collins J, Aisner J, Antman KH. Response to mesna, doxorubicin, ifosfamide and dacarbazine in 108 patients with metastatic or unresectable sarcoma and no prior chemotherapy. J Clin Oncol 1989; 7:1208–16. Piver MS, Lele SB, Patsner B. Diamminedichloroplatinum plus dimethyltriazenoimidazole carboximide as second and third-line chemotherapy for sarcomas of the female pelvis. Gynecol Oncol 1986; 23:371–5. Zaninelli M, Pasini F, Pancheri F. Dose intensity influences the response rate in metastatic or locally advanced unresectable soft-tissue sarcoma (STS). Cancer 1993; 73:1644–51. Rosen G, Forscher C, Lowenbraun S, et al. Synovial sarcoma. Uniform response of metastases to high dose ifosfamide. Cancer 1994; 73:3506–11. Bodey GP. Protected environment prophylactic antibiotic program for malignant sarcoma: randomized trial during remission induction chemotherapy. Cancer 1981; 47:2422–9. Kessinger A, Petersen K, Bishop M, Schmit Pokorny K. High dose therapy (HDT) with autologous hematopoietic stem cell rescue (HSCR) for patients with metastatic soft tissue sarcoma. Proc ASCO 1994; 13:1674. Mertens WC, Bramwell VH. Soft tissue sarcoma in adults. Curr Opin Oncol 1993; 5:678–85. Patel S, Vadhan-Raj S, Burgess MA, et al. Results of two consecutive trials of dose-intensive chemotherapy with doxorubicin and ifosfamide in patients with sarcomas. Am J Clin Oncol 1998; 21:317–21. Pisters PW, Patel SR, Varma DG, et al. Preoperative chemotherapy for stage IIIB extremity soft tissue sarcoma: long-term results from a single institution. J Clin Oncol 1997; 15:3481–7.
978 Soft tissue sarcomas
377 Steward WP, Verweij J, Somers R, et al. Granulocytemacrophage colony-stimulating factor allows safe escalation of dose-intensity of chemotherapy in metastatic adult soft tissue sarcomas: a study of the European Organization for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group. J Clin Oncol 1993; 11:15–21. 378 Le Cesne A, Judson I, Crowther D, et al. Randomized phase III study comparing conventional-dose doxorubicin plus ifosfamide versus high-dose doxorubicin plus ifosfamide plus recombinant human granulocyte-macrophage colonystimulating factor in advanced soft tissue sarcomas: a trial of the European Organization for Research and Treatment of Cancer/Soft Tissue and Bone Sarcoma Group. J Clin Oncol 2000; 18:2676–84. 379 Dumontet C, Biron P, Bouffet E, et al. High dose chemotherapy with ABMT in soft tissue sarcomas: a report of 22 cases. Bone Marrow Transplant 1992; 10:405–8. 380 Schwella N, Rick O, Meyer O, et al. Mobilization of peripheral blood progenitor cells by disease-specific chemotherapy in patients with soft tissue sarcoma. Bone Marrow Transplant 1998; 21:863–8. ●381 Demetri GD, von Mehren M, Blanke CD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med 2002; 347:472–80. 382 Maki RG, Awan RA, Dixon RH, Jhanwar S, Antonescu CR. Differential sensitivity to imatinib of 2 patients with metastatic sarcoma arising from dermatofibrosarcoma protuberans. Int J Cancer 2002; 100:623–6. 383 Reitamo JJ, Hayry P, Nykyri E, Saxen E. The desmoid tumor. I. Incidence, sex-, age- and anatomical distribution in the Finnish population. Am J Clin Pathol 1982; 77:665–73. 384 Jarvinen H. Desmoid disease as a part of the familial adenomatous polyposis coli. Acta Chir Scand 1987; 153:379–83. 385 Church JM. Mucosal ischemia caused by desmoid tumors in patients with familial adenomatous polyposis: report of four cases. Dis Colon Rectum 1998; 41:661–3. 386 Arvanitis ML, Jagelman DG, Fazio VW, Lavery IC, McGannon E. Mortality in patients with familial adenomatous polyposis. Dis Colon Rectum 1990; 33:639–42. 387 De Wever I, Dal Cin P, Fletcher CD, et al. Cytogenetic, clinical, and morphologic correlations in 78 cases of fibromatosis: a report from the CHAMP Study Group. Chromosomes And Morphology. Mod Pathol 2000; 13:1080–5. 388 Fletcher JA, Naeem R, Xiao S, Corson JM. Chromosome aberrations in desmoid tumors. Trisomy 8 may be a predictor of recurrence [see comments]. Cancer Genet Cytogenet 1995; 79:139–43. 389 Bridge JA, Swarts SJ, Buresh C, et al. Trisomies 8 and 20 characterize a subgroup of benign fibrous lesions arising in both soft tissue and bone. Am J Pathol 1999; 154:729–33. 390 Hizawa K, Iida M, Mibu R, Aoyagi K, Yao T, Fujishima M. Desmoid tumors in familial adenomatous polyposis/Gardner’s syndrome. J Clin Gastroenterol 1997; 25:334–7.
391 Lynch HT, Fitzgibbons RJ. Surgery, desmoid tumors, and familial adenomatous polyposis: case report and literature review. Am J Gastroenterol 1996; 91:2598–601. 392 Bertario L, Russo A, Sala P, et al. Genetype and phenotype factors as determinants of desmoid tumors in patients with familial adenomatous polyposis. Int J Cancer 2001; 95:102–7. 393 Giardiello FM, Petersen GM, Piantadosi S, et al. APC gene mutations and extraintestinal phenotype of familial adenomatous polyposis. Gut 1997; 40:521–5. 394 Halling KC, Lazzaro CR, Honchel R, et al. Hereditary desmoid disease in a family with a germline Alu I repeat mutation of the APC gene. Hum Hered 1999; 49:97–102. 395 Li CK, Bapat B, Alman BA. Adenomatous polyposis gene coli gene mutation alters proliferation through its betacatenin-regulatory function in aggressive fibromatosis (desmoid tumor). Am J Pathol 1998; 153:709–14. 396 Schlemmer M. Desmoid tumors and deep fibromatoses. Hematol Oncol Clin North Am 2005; 19:565–71. 397 Enzinger FM, Weiss SW. Soft tissue tumors, 3rd ed. St. Louis: Mosby, 1995. 398 Gansar GF, Markowitz IP, Cerise EJ. Thirty years of experience with desmoid tumors at Charity Hospital. Am Surg 1987; 53:318–9. 399 De Cian F, Delay E, Rudigoz RC, Ranchere D, Rivoire M. Desmoid tumor arising in a cesarean section scar during pregnancy: monitoring and management. Gynecol Oncol 1999; 75:145–8. 400 Way JC, Culham BA. Desmoid tumour. The risk of recurrent or new disease with subsequent pregnancy: a case report. Can J Surg 1999; 42:51–4. ●401 Posner MC, Shiu MH, Newsome JL, Hajdu SI, Gaynor JJ, Brennan MF. The desmoid tumor. Not a benign disease. Arch Surg 1989; 124:191–6. ●402 Ballo MT, Zagars GK, Pollack A, Pisters PW, Pollack RA. Desmoid tumor: prognostic factors and outcome after surgery, radiation therapy, or combined surgery and radiation therapy. J Clin Oncol 1999; 17:158–67. ●403 Spear MA, Jennings LC, Mankin HJ, et al. Individualizing management of aggressive fibromatoses. Int J Radiat Oncol Biol Phys 1998; 40:637–45. 404 Rodriguez-Bigas MA, Mahoney MC, Karakousis CP, Petrelli NJ. Desmoid tumors in patients with familial adenomatous polyposis. Cancer 1994; 74:1270–4. 405 Jones IT, Jagelman DG, Fazio VW, Lavery IC, Weakley FL, McGannon E. Desmoid tumors in familial polyposis coli. Ann Surg 1986; 204:94–7. 406 Azzarelli A, Gronchi A, Bertulli R, et al. Low-dose chemotherapy with methotrexate and vinblastine for patients with advanced aggressive fibromatosis. Cancer 2001; 92:1259–64. 407 Merchant NB, Lewis JJ, Woodruff JM, Leung DH, Brennan MF. Extremity and trunk desmoid tumors: a multifactorial analysis of outcome. Cancer 1999; 86:2045–52. ●408 Gronchi A, Casali PG, Mariani L, et al. Quality of surgery and outcome in extra-abdominal aggressive fibromatosis: a
References 979
409
410 411
✾412
413
414
415
416
417 418
419
420
421
422
423
424
425
series of patients surgically treated at a single institution. J Clin Oncol 2003; 21:1390–7. Miralbell R, Suit HD, Mankin HJ, Zuckerberg LR, Stracher MA, Rosenberg AE. Fibromatoses: from postsurgical surveillance to combined surgery and radiation therapy. Int J Radiat Oncol Biol Phys 1990; 18:535–40. Acker JC, Bossen EH, Halperin EC. The management of desmoid tumors. Int J Radiat Oncol Biol Phys 1993; 26:851–8. Goy BW, Lee SP, Eilber F, et al. The role of adjuvant radiotherapy in the treatment of resectable desmoid tumors. Int J Radiat Oncol Biol Phys 1997; 39:659–65. Nuyttens JJ, Rust PF, Thomas CR Jr, Turrisi AT 3rd. Surgery versus radiation therapy for patients with aggressive fibromatosis or desmoid tumors. A comparative review of 22 articles. Cancer 2000; 88:1517–23. Ballo MT, Zagars GK, Pollack A. Radiation therapy in the management of desmoid tumors. Int J Radiat Oncol Biol Phys 1998; 42:1007–14. Micke O, Seegenschmiedt MH. Radiation therapy for aggressive fibromatosis (desmoid tumors): results of a national Patterns of Care Study. Int J Radiat Oncol Biol Phys 2005; 61:882–91. Baliski CR, Temple WJ, Arthur K, Schachar NS. Desmoid tumors: a novel approach for local control. J Surg Oncol 2002; 80:96–9. Tsukada K, Church JM, Jagelman DG, Fazio VW, Lavery IC. Noncytotoxic drug therapy for intra-abdominal desmoid tumor in patients with familial adenomatous polyposis. Dis Colon Rectum 1992; 35:29–33. Wilcken N, Tattersall MH. Endocrine therapy for desmoid tumors. Cancer 1991; 68:1384–8. Thomas S, Datta-Gupta S, Kapur BM. Treatment of recurrent desmoid tumor with tamoxifen. Aust N Z J Surg 1990; 60:919–21. Izes JK, Zinman LN, Larsen CR. Regression of large pelvic desmoid tumor by tamoxifen and sulindac. Urology 1996; 47:756–9. Tonelli F, Ficari F, Valanzano R, Brandi ML. Treatment of desmoids and mesenteric fibromatosis in familial adenomatous polyposis with raloxifene. Tumori 2003; 89:391–6. Janinis J, Patriki M, Vini L, Aravantinos G, Whelan JS. The pharmacological treatment of aggressive fibromatosis: a systematic review. Ann Oncol 2003; 14:181–90. Hansmann A, Adolph C, Vogel T, Unger A, Moeslein G. High-dose tamoxifen and sulindac as first-line treatment for desmoid tumors. Cancer 2004; 100:612–20. Lackner H, Urban C, Kerbl R, Schwinger W, Beham A. Noncytotoxic drug therapy in children with unresectable desmoid tumors. Cancer 1997; 80:334–40. Waddell WR, Gerner RE, Reich MP. Nonsteroid antiinflammatory drugs and tamoxifen for desmoid tumors and carcinoma of the stomach. J Surg Oncol 1983; 22:197–211. Brooks MD, Ebbs SR, Colletta AA, Baum M. Desmoid tumours treated with triphenylethylenes. Eur J Cancer 1992; 281:1014–8.
426 Raney RB, Asmar L, Vassilopoulou-Sellin R, et al. Late complications of therapy in 213 children with localized, nonorbital soft-tissue sarcoma of the head and neck: A descriptive report from the Intergroup Rhabdomyosarcoma Studies (IRS)- II and -III. IRS Group of the Children’s Cancer Group and the Pediatric Oncology Group. Med Pediatr Oncol 1999; 33(4):362–71. 427 Fernberg JO, Brosjo O, Larsson O, Soderlund V, Strander H. Interferon-induced remission in aggressive fibromatosis of the lower extremity. Acta Oncol 1999; 38:971–2. 428 Geurs F, Kok TC. Regression of a great abdominal desmoid tumor by interferon alpha. J Clin Gastroenterol 1993; 16:264–5. 429 Mace J, Sybil Biermann J, Sondak V, et al. Response of extraabdominal desmoid tumors to therapy with imatinib mesylate. Cancer 2002; 95:2373–9. 430 Baker LH, Wathen K, Chugh R, et al. Activity of imatinib mesylate in desmoid tumors: interim analysis of a Sarcoma Alliance for Research through Collaboration (SARC) phase II trial (abstract). Proc Am Soc Clin Oncol 2004; 22:821s. ◆431 Patel SR, Benjamin RS. Desmoid tumors respond to chemotherapy: defying the dogma in oncology. J Clin Oncol 2006; 24:11–2. 432 Weiss AJ, Lackman RD. Low-dose chemotherapy of desmoid tumors. Cancer 1989; 64:1192–4. 433 Skapek SX, Hawk BJ, Hoffer FA, et al. Combination chemotherapy using vinblastine and methotrexate for the treatment of progressive desmoid tumor in children. J Clin Oncol 1998; 16:3021–7. 434 Weiss AJ, Horowitz S, Lackman RD. Therapy of desmoid tumors and fibromatosis using vinorelbine. Am J Clin Oncol 1999; 22:193–5. 435 Wehl G, Rossler J, Otten JE, Boehm N, Uhl M, Kontny Y, et al. Response of progressive fibromatosis to therapy with liposomal doxorubicin. Onkologie 2004; 27:552–6. 436 Patel S, Evans H, Benjamin R. Combination chemotherapy in adult desmoid tumors. Cancer 1993; 72:3244–7. 437 Okuno SH, Edmonson JH. Combination chemotherapy for desmoid tumors. Cancer 2003; 97:1134–5. 438 Gega M, Yanagi H, Yoshikawa R, et al. Successful chemotherapeutic modality of doxorubicin plus dacarbazine for the treatment of desmoid tumors in association with familial adenomatous polyposis. J Clin Oncol 2006; 24:102–5. 439 Poritz LS, Blackstein M, Berk T, Gallinger S, McLeod RS, Cohen Z. Extended follow-up of patients treated with cytotoxic chemotherapy for intra-abdominal desmoid tumors. Dis Colon Rectum 2001; 44:1268–73. 440 Clark TW. Percutaneous chemical ablation of desmoid tumors. J Vasc Interv Radiol 2003; 14:629–34. 441 Engellau J. Prognostic factors in soft tissue sarcoma: Tissue microarray for immunostaining, the importance of whole-tumor sections and time-dependence. Acta Orthop Scand (Suppl 314) 2004; 75:1–52. 442 Cancer AJCo. AJCC Cancer Staging Manual, 6th ed. New York: Springer, 2002.
980 Soft tissue sarcomas
443 Spiro IJ, Gebhardt MC, Jennings LC, Mankin HJ, Harmon DC, Suit HD. Prognostic factors for local control of sarcomas of the soft tissues managed by radiation and surgery. Semin Oncol 1997; 24(5):540–6. 444 Abbatucci JS, Boulier N, de Ranieri J, et al. Local control and survival in soft tissue sarcomas of the limbs, trunk walls and head and neck: A study of 113 cases. Int J Radiat Oncol Biol Phys 1986; 12:579–86. 445 Wilson AN, Davis A, Bell RS, et al. Local control of soft tissue sarcoma of the extremity: The experience of a multidisciplinary sarcoma group with definitive surgery and radiotherapy. Eur J Cancer 1994; 30A:746–51. 446 Keus RB, Rutgers EJ, Ho GH, Gortzak E, Albus-Lutter CE, Hart AAM. Limb-sparing therapy of extremity soft tissue sarcomas: Treatment outcome and long-term functional results. Eur J Cancer 1994; 30:1459–63. 447 Wilson RB, Crowe PJ, Fisher R, Hook C, Donnellan MJ. Extremity soft tissue sarcoma: factors predictive of local recurrence and survival. Aust N Z J Surg 1999; 69(5):344–9. 448 Eilber FR, Eckardt JJ, Rosen G, Fu YS, Seeger LL, Selch MD. Neoadjuvant chemotherapy and radiotherapy in the multidisciplinary management of soft tissue sarcomas of the extremity. Surg Oncol Clin NA 1993; 2:611–20. 449 Davis AM, O’Sullivan B, Turcotte R, et al. Late radiation morbidity following randomization to preoperative versus postoperative radiotherapy in extremity soft tissue sarcoma. Radiother Oncol 2005; 75(1):48–53. 450 Fleming JB, Berman RS, Cheng SC, et al. Long-term outcome of patients with American Joint Committee on Cancer stage IIB extremity soft tissue sarcomas. J Clin Oncol 1999; 17(9):2772–80. 451 Gibbs CP, Peabody TD, Mundt AJ, Montag AG, Simon MA. Oncological outcomes of operative treatment of
452
453
454
455
456
457
458
subcutaneous soft-tissue sarcomas of the extremities. J Bone Joint Surg Am 1997; 79(6):888–97. Karakousis CP, Proimakis C, Walsh DL. Primary soft tissue sarcoma of the extremities in adults. Br J Surg 1995; 82(9):1208–12. Lerner HJ, Amato DA, Savlov E, et al. Eastern Cooperative Oncology Group: a comparison of adjuvant doxorubicin and observation for patients with localized soft tissue sarcoma. J Clin Oncol 1987; 5:613–17. Antman K, Ryan L, Borden E, et al. Pooled results from three randomized adjuvant studies of doxorubicin versus observation in soft tissue sarcoma: 10 year results and review of the literature. In: Salmon SE, ed. Adjuvant therapy of cancer VI. Philadelphia: WB Saunders, 1990: 529–43. Benjamin RS, Terjanian TO, Fenoglio CJ, et al. The importance of combination chemotherapy for adjuvant treatment of high-risk patients with soft-tissue sarcomas of the extremities. In: Salmon SE, ed. Adjuvant therapy of cancer V. Orlando: Grune and Stratton, 1987: 735–44. Glenn J, Kinsella T, Glatstein E, et al. A randomized, prospective trial of adjuvant chemotherapy in adults with soft tissue sarcomas of the head and neck, breast, and trunk. Cancer 1985; 55:1206–14. Glenn J, Sindelar W, Kinsella T, et al. Results of multimodality therapy of resectable soft-tissue sarcomas of the retroperitoneum. Surgery 1985; 97:316–25. Ravaud A, Bui NB, Coindre JM, et al. Adjuvant chemotherapy with cyvadic in high risk soft tissue sarcoma: a randomized prospective trial. In: Salmon SE, ed. Adjuvant therapy of cancer VI. Philadelphia: WB Saunders, 1990: 556–66.
41 Leukaemias TARIQ I. MUGHAL AND JOHN M. GOLDMAN
Introduction Acute leukaemias Acute myeloid leukaemia Acute lymphoblastic leukaemia Chronic leukaemias
981 981 981 990 996
INTRODUCTION Leukaemias comprise a heterogeneous group of clonal disorders of haemopoiesis, affecting both haemopoietic stem cells and progenitor cells within the myeloid and lymphocytic lineages. They are characterized by an excessive accumulation of abnormal leukaemia cells in the bone marrow and peripheral blood. Historically, leukaemias are often considered as ‘acute’ and ‘chronic’. Acute leukaemias are typically of rapid onset, whereas chronic leukaemias more usually appear to have evolved slowly before diagnosis; neither term refers to the severity of the disease. Myelodysplastic syndromes (MDS) are considered pre-leukaemic conditions that may progress to acute myeloid leukaemia (AML).1 The leukaemias are classified in accordance with salient pathological features of the abnormal excessive haemopoietic cells. The principal purpose is an attempt to group together conditions that share molecular features and so reduce the overall heterogeneity of the leukaemias. The acute leukaemias are broadly divided into AML and acute lymphoblastic leukaemia (ALL) and the chronic leukaemias into chronic myeloid leukaemia (CML) and chronic lymphocytic leukaemia (CLL). Undoubtedly as our understanding of the molecular pathogenesis of leukaemias and MDS improves, more homogeneous subgroups will be recognized .2 Many of the genes implicated are now well characterized.3,4 Progress is being made in elucidating the genetic features of these diseases that appear to inactivate the tumour suppressor genes required for normal haemopoiesis and the identification of mutations in the haemopoietic stem cells and progenitor cells.5
Chronic myeloid leukaemia Chronic lymphocytic leukaemia Other lymphoid leukaemias Myelodysplastic syndromes References
996 1005 1009 1011 1014
The recent World Health Organization (WHO) classification of leukaemias and MDS accommodates some, but not all, of the molecular information accrued in recent years.2 This new classification points the way to a better and sometimes molecularly targeted treatment for the individual patients. However, the older classification proposed by the French–American–British (FAB) group, which was based entirely on morphology and cell surface membrane markers, remains in general use;6 it has been revised and modified and is therefore still valuable.7–9 In this chapter we summarize current understanding of the molecular pathogenesis, prognosis and treatment of leukaemias, and emphasize molecularly targeted advances in therapy.
ACUTE LEUKAEMIAS ACUTE MYELOID LEUKAEMIA Acute myeloid leukaemia is characterized by the malignant transformation of myeloid stem cells in the bone marrow, which are then incapable of normal differentiation and maturation. Since normal haemopoiesis is organized hierarchically, the malignant transformation can occur at different levels and AML may arise in a stem cell capable of differentiating into cells of erythroid, granulocytic, monocytic and megakaryocytic lineages, or in a lineage-restricted stem cell. The transformation may occur in a stem cell capable of differentiating into both lymphoid and myeloid
982 Leukaemias
lineages but only on rare occasions does AML appear as a hybrid or biphenotypic leukaemia, in which the predominating leukaemia cell expresses both lymphoid and myeloid markers.
Epidemiology Acute myeloid leukaemia accounts for nearly 80 per cent of all adult acute leukaemias with an annual incidence of approximately 3.4 per 100 000 adults; its incidence increases progressively with age to greater than 20 per 100 000 adults aged 80 years or older.10 In contrast, AML accounts for 10–15 per cent of childhood leukaemias with an annual incidence of less than 1 per 100 000 children. Overall, the median age at diagnosis in the western world is about 70 years. In most cases the aetiology is not obvious, but some constitutional and acquired disorders do predispose to AML. For example, children with Down syndrome, Bloom’s syndrome or Fanconi anaemia have an increased risk of developing acute leukaemia. Acute myeloid leukaemia in older patients sometimes follows an earlier diagnosis of MDS or a myeloproliferative disorder, such as polycythemia vera or, more rarely, essential thrombocythemia. Patients with such ‘secondary’ AMLs often have unfavourable karyotypes and expression of multidrug resistance proteins, both associated with a relatively poor prognosis. Many studies of causality suggest that exposure to ionizing radiation or certain aromatic hydrocarbons, in particular benzene, increases the risk of developing leukaemia to a significant but minor degree.11,12 Importantly the serious risk of therapy-related acute leukaemia, especially in patients previously exposed to chemotherapeutic agents such as alkylating agents and topoisomerase II inhibitors, is well recognized. With alkylating agents the overall risk appears to be 3–10 per cent and peaks 5–10 years after the start of therapy. Some of these patients initially develop MDS with deletions of chromosomes 5 and 7.13,14 The therapy-related AML that occurs as a complication of drugs that inhibit topoisomerase II, mainly the epipodophyllotoxins, differs from the alkylating agent-related AML: the former develops after a relatively short latent period of 2–3 years, is usually not preceded by MDS, and is often associated with balanced aberrations with a rearrangement, but no gain or loss of chromosomal material.15,16
bone marrow or peripheral blood (Fig. 41.1). Conventional cytochemical investigations show that the blasts cells contain cytoplasmic myeloperoxidase and the non-specific esterases, i.e., α-naphthyl acetate and naphthyl butyrate. These may allow distinction of AML from ALL and MDS and a preliminary classification, but further characterization requires study of surface membrane markers, cytogenetics and/or direct molecular analysis of the leukaemic blasts.17
Classifications WHO CLASSIFICATION
The WHO classification of AML was introduced in 1999 and is based on the cytogenetic and molecular genetic abnormalities as well as clinical, morphological and surface membrane abnormalities.18 It recognizes four major categories: (i) AML with recurring genetic abnormalities; (ii) AML with multi-lineage dysplasia; (iii) AML that is therapy-related;
Clinical Most patients with AML present with signs and symptoms arising from bone marrow failure and organ infiltration by leukaemic cells. Pallor, lethargy, dyspnoea, infections and bleeding manifestations are all common. Occasionally patients present as a consequence of hyperleukostasis. The diagnosis of AML is made when more than 20 per cent leukaemic blast cells (also known as myeloblasts) (in accordance to the WHO classification, see below) are found in the
Figure 41.1 A peripheral blood film of a patient with acute myeloid leukaemia, FAB subtype M1, showing Type I and Type II blasts ( 960). Kindly provided by Professor Barbara Bain, Imperial College, London.
Acute myeloid leukaemia 983
and (iv) AML not otherwise categorized. The WHO classification is contrasted with the FAB classification in Box 41.1; it includes 17 subtypes and the pertinent features include a lowering of the threshold for the diagnosis of AML to 20 per cent of blasts compared with 30 per cent in the FAB classification, abolition of the MDS category of ‘refractory anaemia with excess blasts in transformation’ and allowing a diagnosis of AML to be made regardless of the percentage of bone marrow blasts in patients with abnormal haemopoiesis and characteristic clonal structural cytogenetic abnormalities. FAB CLASSIFICATION
The FAB classification of acute leukaemias described first in 1976 relies solely on morphological studies to identify the lineage of the leukaemic cells and the degree of differentiation present.6 The annotated version, shown in Box 41.1, divides AML into ten distinct subtypes that differ morphologically and histochemically; immunological and genetic differences are also used to define some subtypes. The latest version of the FAB classification remains popular amongst many haematologists. The most common subtypes are
Box 41.1 The French–American–British (FAB) and World Health Organization (WHO) classifications of acute myeloid leukaemias (AML)
M2 and M4, followed by M1, and then by M5 and M4Eo; M3, M6 and M7 are relatively rare. The subtype M0 accounts for approximately 5 per cent of all patients with AML and denotes minimally differentiated blast cells; the myeloid nature of the blast cells must still be demonstrated in order to differentiate the leukaemia from ALL. In the subtype M1 the blasts, which account for over 90 per cent of non-erythroid cells, show minimal evidence of differentiation. Two types of M1 blasts have been sub-categorized: Type I lack granules and Type II contain granules and have a somewhat lower nuclear:cytoplasmic ratio. In the subtype M2 the blasts account for 30–90 per cent of non-erythroid cells, and there is evidence of maturation to the promyelocyte stage or beyond; Auer rods are common. Subtype M3 cells are heavily granulated promyelocytes, sometimes with bilobed nuclei and often with bundles of Auer rods in the cytoplasm (Fig. 41.2); M3v cells define a hypogranular variant. Disseminated intravascular coagulation (DIC) is common at presentation in both varieties of M3. Subtype M4 is characterized by myelomonocytic morphology and monocytosis is common; a variant of M4, M4Eo is recognized when dysplastic eosinophils constitute 5 per cent or more of
Box 41.2 Prognostic factors – molecular markers ●
FAB classification of AML M0: Minimally differentiated M1: Myeloblastic leukaemia without maturation M2: Myeloblastic leukaemia with maturation M3: Hypergranular promyelocytic leukaemia M3v: Microgranular promyelocytic leukaemia M4: Myelomonocytic leukaemia M4Eo: variant, increase in marrow eosinophils M5: Monocytic leukaemia M6: Erythroleukaemia M7: Megakaryoblastic leukaemia WHO classification of AML AML with recurrent cytogenetic abnormalities: ● AML with t(8;21)(q22;q22); (AML1/ETO) ● Acute promyelocytic leukaemia: AML with t(15;17)(q22;q12) and variants ● AML with abnormal bone marrow eosinophils inv(16)(p13q22) and t(16;16) (p13;q22); (CBFβ/MYH11) ● AML with 11q23 (MLL) abnormalities AML with multi-lineage dysplasia, with or without prior MDS AML with MDS, therapy related AML not otherwise categorized: ● FAB M0, M1, M4, M5, M6, M7 and acute basophilic leukaemia
● ●
●
Transmembrane transporter protein that confers multidrug resistance (MDR1) Mutations in, or overexpression of, specific genes: Unfavourable prognosis: WT1, BAX, BCL2/BAX, BAALC, EVI1, KIT, FLT3, MLL, ERG Favourable prognosis: C/EBPα, NPM1
Expressed in cells from patients with normal karyotype.
Figure 41.2 A peripheral blood film from a patient with acute myeloid leukaemia, FAB subtype M3, ( 960). Kindly provided by Professor Barbara Bain, Imperial College, London.
984 Leukaemias
nucleated cells. Two variants of subtype M5 are recognized: M5a, in which more than 80 per cent of the non-erythroid cells are monoblasts, and M5b, in which fewer than 80 per cent of the non-erythroid cells are monoblasts and the rest are pro-monocytes and monocytes. Infiltration of tissues, in particular gums, perianal area and skin, with blasts is frequent in subtypes M4 and M5. In the subtypes M6 and M7, the malignant transformation occurs at the level of stem cells committed to the erythroid and megakaryocytic lineages, respectively. In the subtype M6, over 30 per cent of non-erythroid cells are Type I or II blasts and over 50 per cent of all marrow cells are erythroblasts. Subtype M7 is characterized by marrow fibrosis, and large polymorphic blasts with cytoplasmic blebs are often seen, which are CD41-positive.
Molecular pathogenesis Over the past two decades, many efforts directed to assessing the genetic abnormalities in AML have led to a better understanding of the underlying molecular pathogenesis. This means that AML patients can now be classified into groups with different prognoses based on cytogenetic and molecular analyses (Box 41.2).19,20 For example, leukaemia cells with t(8;12), inv(16) or t(15;17) indicate good-risk AML, while cells with translocations involving the MLL gene at 11q23, t(6;9) or chromosome 7q changes indicate a poorrisk AML. Acute promyelocytic leukaemia (APL; FAB subtype M3) is an example of a leukaemia with an apparent differentiation blockade. Patients with APL have a balanced reciprocal translocation resulting in the fusion of portions of the promyelocytic leukaemia gene (PML) on chromosome 15 with the retinoic acid receptor alpha gene (RARα) on chromosome 17.21 This ‘chimeric’ or ‘fusion’ gene encodes a PML-RARα protein, which retains the retinoic acid ligandbinding domain and may account for the unique propensity of APL cells to differentiate in the presence of retinoids such as all trans-retinoic acid (ATRA).22 There are at least three further breakpoints in the RARα gene (17q21) associated with APL, although the position of the breakpoint in the PML gene is relatively consistent; the precise make-up of the fusion protein determines the degree of responsiveness to ATRA and potential for long-term remission.23 As well as ATRA, arsenic-based compounds can also induce APL cells to differentiate and thus reverse the leukaemic phenotype.24 Arsenic trioxide also induces apoptosis with caspase activation in leukaemic cells. Other well-characterized genetic abnormalities involve the acute myeloid leukaemia 1 gene (AML1), the mixedlineage leukaemia (MLL) gene, the FMS-like tyrosine kinase-3 gene (FLT3) and others. AML1 is involved in three different chromosomal translocations, which share a breakpoint at its site on chromosome 21q22: t(8;21), t(3;21) and t(12;21). AML1 encodes the α-subunit of core-binding factor (CBF), a multi-component transcription factor
complex that regulates a number of haemopoiesis-specific genes and is essential for normal haemopoiesis. Core binding factor abnormalities are often found in the t(8;21) translocation, which is present in 40 per cent of patients with AML, FAB subtype M2. They are also associated to a lesser degree with t(3;21), which is found in some patients with MDS and also in myeloid blast transformation of CML, and with t(12;21), which is found in childhood ALL.25 The fusion proteins show a common theme of transcription-factor domain exchanges whereby the normal function of AML1 is disrupted. Further insights into AML1 function have been gained by examining its binding partners in the CBF, namely CBFβ, which is targeted in AML, associated chromosomal rearrangement inv(16) or its variant t(16;16).26 Collectively these data suggest the AML1 pathway as a common target for leukaemia-specific translocations. MLL gene rearrangements occur in structural abnormalities of chromosome 11q23, which are found in about 10 per cent of patients with AML, FAB subtypes M4 or M5, and about 85 per cent of patients with secondary AML that develops after exposure to topoisomerase II inhibitors.27 Abnormalities of the FLT3 gene, which seem to confer a poor prognosis, are found in over one third of all AML patients.28,29 Approximately 25–30 per cent of patients have internal tandem duplication (ITD) in the FLT3 gene and about 5–7 per cent have mutations in the tyrosine kinase domain of the receptor encoded by FLT3. Other genetic alterations associated with an unfavourable outcome include MLL-PTD (partial tandem duplication of the myeloid/lymphoid or mixed-lineage leukaemia) gene, overexpression of the ERG gene or overexpression of the BAALC gene. In contrast, mutations involving the transcription factor gene CEBPA, also known as the CCAAT enhancer binding protein alpha (C/EBPα), or NPM1 gene often tend to have a good prognosis (Fig 41.3).30 Other examples of molecular abnormalities include mutations in the tyrosine kinase receptor gene KIT and the GTPase genes N-RAS and K-RAS. Mutations involving KIT are often noted in patients with inv(16) and t(8;21) and are associated with an increased relapse rate.31,32 Many efforts are currently being made to develop treatments that target these abnormal proteins, especially those encoded by FLT3 mutations (see below).33 Increasing use of gene expression profiling should facilitate these efforts.34
Prognostic factors A number of clinical, morphological, immunological and cytogenetic features of AML have allowed patients to be stratified into risk categories based on prognosis, which is largely dependent on being able to achieve a complete remission (CR) and prolong overall survival. Currently the karyo-type at diagnosis appears to be the most important prognostic factor.19 Patients are often divided into a good, intermediate or poor-risk category based on the karyotype of the leukaemic cell. The speed of the initial response
Acute myeloid leukaemia 985
CEBPA+/MLL-PTD+ 1.2 FLT3-TKD+/MLL-PTD+ 0.8 FLT3-TKD+ 2.4 FLT3-ITD+/CEBPA+ 1.6 FLT3-ITD+/MLL-PTD+ 2.4
CEBPA+ 5.7
MLL-PTD+ 4.5
WT 23.6
FLT3=ITD+ 8.1 NPM1+/FLT3-TKD+/CEBPA+MLL-PTD+ 0.4 NPM1+/FLT3-ITD+/MLL=PTD+ 0.4 NPM1=/FLT3-ITD+/CEBPA+ 1.2 NPM1+/FLT3-ITD+/FLT3-TKD+ 1.2 NPM1+/CEBPA+ 1.2
NPM1+/CEBPA+ 3.7 NPM1+/FLT3-TKD+ 4.1
NPM1+ 19.5 NPM1+/FLT3-ITD+ 17.9
Figure 41.3 Presence or Absence of Mutation (percentages) by Gene in AML. (Courtesy of Professor Martin Tallman, Northwest Cancer Center, Chicago)
to induction therapy and the initial leukaemia cell burden (white-cell count of greater than 20 109/l) are also useful in assessing risk of relapse. The age of the patient at diagnosis has a significant impact on the biology of AML. Older patients, particularly those over the age of 60 years, often present with poor-risk karyotypes, more primitive phenotype, often express P-glycoprotein and have an increased inherent resistance to many commonly used cytotoxic agents. The good risk category includes patients with t(15;17), t(8;21), or inv(16) and FAB subtypes M1 or M2. These patients have the best prognosis after treatment with an anthracycline and standard-dose cytarabine with 70–90 per cent achieving a CR. They have a 30–40 per cent risk of relapse and 40–50 per cent event-free survival at 5 years.19 The poor risk category includes patients with abnormalities involving chromosome 5 (5 or 5q), chromosome 7 (7 or 7q) or chromosome 3. They have a poor outcome with 30–50 per cent achieving CR and fewer than 5 per cent remaining in CR at 5 years. Patients who require more than one course of chemotherapy to induce CR, who have an elevated serum lactate dehydrogenase or who have secondary AML (related to an antecedent haematological disorder, such as MDS) also belong to this poor risk category with a 30–50 per cent rate of CR and less than 20 per cent event-free survival at 5 years. The intermediate-risk category includes patients with a normal karyotype. They have a 50–85 per cent rate of CR and about 30 per cent event-free survival at 5 years.
Treatment GENERAL PRINCIPLES
For patients with AML there are two sequential objectives of current standard treatment. The first is to restore normal haemopoiesis and the second is to prevent relapse. Classically the term CR is defined as a state in which a patient has achieved fewer than 5 per cent blasts in the bone marrow, all of which lack a leukaemia immunophenotype, in conjunction with complete recovery of the peripheral blood counts. However, molecular genetic markers are now being used more often and the definition of CR has become more complex. The conventional treatment plan for all patients consists of a remission induction followed by a post-remission, or ‘consolidation’, therapy. Currently there is no role for ‘maintenance’ therapy using lower doses of cytotoxic drugs for longer periods of time. The objective of the remission induction therapy is to achieve a CR and that of the post-remission therapy is to eradicate ‘residual’ leukaemia, allowing for a possible long-term remission. Patients who do not receive post-remission therapy usually relapse within 6–9 months. Patients with AML should not be considered as a single entity for the post-remission therapy, but rather such therapy should be tailored according to the individual risk category, preferably defined by the WHO classification. It is also appropriate to consider separately treatment for patients with APL, which is now the
986 Leukaemias
most curable subtype of AML, and for all patients over the age of 60 years (often referred to as older adults). REMISSION INDUCTION FOR PATIENTS UNDER THE AGE OF 60 YEARS
Substantial improvements have been made during the past two decades with regard to the proportion of patients under the age of 60 years who are able to achieve CR.35,36 This progress appears to be due largely to better supportive care, which has enabled the safer delivery of more intensive remission induction, particularly with high doses of cytarabine.37 A variety of induction regimens followed by consolidation treatment with or without prolonged maintenance has been extensively investigated and the results suggest that about 70 per cent of patients achieve a CR and with post-remission therapy and 25–30 per cent will remain in CR at 5 years. For patients who receive less intensive postremission therapy, only 10–15 per cent will remain in CR at 5 years.38 The major unresolved issue for patients under the age of 60 years is how best to ensure continuing CR. The most common remission induction consists of two courses of an anthracycline (non cell cycle specific), usually daunorubicin or doxorubicin at 40–60 mg/m2 per day for 3 days, and a continuous infusion of the cell cycle-specific drug, cytarabine (cytosine arabinoside), at 100–200 mg/m2 per day for 7 days. Several recent randomized trials of alternatives to daunorubicin or doxorubicin suggested that idarubicin and mitoxantrone are more effective, although both result in longer periods of pancytopenia.39,40 There is, however, some debate with regard to the equivalent doses and further studies are in progress. In the British Medical Research Council’s (MRC) AML-10 trial, 1820 patients under the age of 56 years were randomized to daunorubicin, cytarabine and 6-thioguanine or daunorubicin, cytarabine and etoposide chemotherapy as induction treatment for two courses.41 After the two induction courses, patients in CR received a third induction course consisting of amsacrine, cytarabine and etoposide following which bone marrow stem cells were harvested and patients who lacked an HLA-matched sibling donor or did not wish to receive an allograft were randomized to receive a further course of induction treatment consisting of mitoxantrone and cytarabine (at 1 g/m2, every 12 h for 3 days) followed by either autologous stem cell transplant (SCT) or no further treatment. In this trial, 80 per cent of patients achieved CR and the overall survival at 7 years was 40 per cent. In an attempt to improve the overall survival, several studies have incorporated high-dose cytarabine (3 g/m2), additional conventional cytotoxic drugs such as etoposide, the purine analogue fludarabine or the camptothecin topotecan. Moreover haemopoietic growth factors such as granulocyte colony-stimulating factor (G-CSF) had also been used either as haematological support or to encourage or ‘prime’ the leukaemic cells to enter the cell cycle to increase their susceptibility to chemotherapy in the induction phase.42,43
Sadly, for the most part, there appears to be no major benefit compared with the standard remission induction treatment. Selected groups appear to have achieved some benefit from certain strategies. For example, patients in the intermediaterisk category appear to benefit from G-CSF priming and those in a poor risk-category may fare better with the highdose cytarabine treatment. This latter strategy was assessed in the MRC AML-12 trial, which recruited 2400 patients.44 This study showed some improvement in the overall survival, but at the expense of considerably greater toxicity, particularly neurotoxicity. POST-REMISSION THERAPY FOR PATIENTS UNDER THE AGE OF 60 YEARS
All patients under the age of 60 years who achieve a CR following induction therapy require further treatment in order to improve their chance of remaining in continuous CR. Currently there are three well defined and intensively investigated options: an allograft from an HLA-matched donor (sibling or unrelated), an autograft, or further intensive chemotherapy. The degree of benefit is related to the karyotype at diagnosis. For patients with good-risk cytogenetics, 3 to 4 cycles of high-dose cytarabine results in the best longterm results. It is possible that 1 to 2 cycles of high-dose cytarabine followed by an autograft is a reasonable option for this cohort but firm data to support this are not currently available. For patients with intermediate-risk cytogenetics, it appears reasonable to consider 3 to 4 cycles of high-dose cytarabine or an allograft if a suitable donor is available. For patients with poor-risk cytogenetics, it is best to consider an allograft using a suitable matched sibling or an alternative donor, which we discuss below. TREATMENT OF PATIENTS OLDER THAN 60 YEARS OF AGE WITH ACUTE MYELOID LEUKAEMIA
Patients over the age of 60 years account for about 75 per cent of all those newly diagnosed AML, the majority of whom are in the poor-risk group.45 Furthermore older patients are less able to tolerate the rigours associated with the intensive induction regimens, and certainly not a conventional conditioning allogeneic stem cell transplant. The incidence of significant co-morbid medical conditions is also high in this cohort. A nihilistic approach unfortunately results in a poor quality of life and very poor survival. Patients who are younger than 80 years of age, have no significant medical history and have a good performance status have an approximately 50 per cent chance of achieving CR with conventional induction treatment and an event-free survival of about 20 per cent at 2 years.46 A minority may benefit from low-dose maintenance therapy, in particular from a quality of life perspective. Many efforts are being devoted to improving the supportive care for these patients and to use haemopoietic growth factors, in particular G-CSF, as an adjunct to induction therapy.47 The current MRC AML-14 trial evaluates two treatment strategies in patients
Acute myeloid leukaemia 987
aged 60 years or older with AML or high-risk myelodysplastic syndrome. The trial has an ‘intensive’ and a ‘nonintensive’ approach; where there is uncertainty, patients are randomized between the ‘intensive’ and ‘non-intensive’ treatments. The ‘intensive’ treatment arm compares two doses of daunorubicin, the addition of the resistance modulator PSC833, two doses of cytarabine, and three versus four courses of treatment in total. The ‘non-intensive’ chemotherapy treatment arm compares a hydroxyurea control versus low dose cytarabine, with or without ATRA. Cloretazine, a novel DNA alkylating agent that selectively targets the O-6 position in guanine, has recently entered clinical trials for elderly AML and high-risk MDS patients. Preliminary results from 105 patients with de novo AML demonstrate a 27 per cent leukaemia-free survival (LFS) and 22 per cent overall survival at 12 months of follow-up.48 The drug is administered as a 30 to 60 minute intravenous infusion and appears to be well tolerated. Clofarabine is another novel agent being tested, which we discuss below. In older patients with APL, it may be worthwhile considering less toxic consolidation treatment such as liposomal ATRA or anti-CD33 monoclonal antibody-based strategies.49,50 INVESTIGATIONAL APPROACHES
The recent improvements in the understanding of the molecular pathogenesis of AML have led to a number of rationally designed agents to enter clinical trials. Many have been developed to target a specific molecular abnormality. Table 41.1 shows some of these agents that are in clinical trial.
Table 41.1 Agents in clinical trial targeting specific molecular anomalies in acute myeloid leukaemia (AML) Class Antibodies MDR inhibitors FT inhibitors FLT3 inhibitors HDAC inhibitors
Agent
Target
Gemtuzumab PSC833, zosuquidar Tipifarnib PKC-412, CEP-701, MLN518, SU11248 Valproic acid, SAHA, depsipeptide Bevacizumab Oblimersen Clofarabine
CD33 P-gp Lamin A, HJJ-2 FLT3/ITD
Anti-angio agents Apoptosis inhibitors Deoxyadenosine analogues Hypomethylating agents Decitabine Antisense AEG35156
HDAC VEGF BCL2 DNA DNA XIAP
FLT3, FMS-like tyrosine kinase 3; FT, farnesyl transferase; HDAC, histone deacetylase; ITD, internal tandem duplication; MDR, multidrug resistance; P-gp, P glycoprotein; SAHA, suberoylanilide hydroxamic acid; VEGF, vascular endothelial growth factor; XIAP, X-chromosome-linked inhibitor of apoptosis protein.
Gemtuzumab ozogamicin Several studies suggest the value of adding a specific antibody to standard chemotherapy. Gemtuzumab ozogamicin (Mylotarg) is a calicheamicin conjugated monoclonal antibody directed against the CD33 antigen that is expressed on most AML blast cells. Its use in adults under the age of 60 years is associated with a higher CR rate (91 per cent).51,52 The results of using this drug as a single agent in patients over the age of 60 years showed a modest benefit and led to regulatory approval in the USA and Japan, but not in Europe, for patients over the age of 60 years in first relapse. Rare patients developed a syndrome resembling venoocclusive disease of the liver. Multidrug resistance inhibitors Studies of the mechanisms underlying increased inherent resistance to cytotoxic drugs in patients over the age of 60 years and in those with relapsed and refractory AML have identified an increased expression of P-glycoprotein, a cellular membrane protein encoded by the gene MDR1. Though attempts to inhibit P-glycoprotein with first generation agents such as cyclosporine and PSC-833 were largely negative, preliminary results with second generation agents such as zosuqidar (LY335979) appear encouraging.53 The results of a completed prospective randomized study of induction with daunorubicin and cytarabine with or without zosuqidar should be reported in the near future. Farnesyl transferase inhibitors Following the observation of mutation of the RAS gene and a number of farnesylated proteins, such as GTPases Rho B and the centromere proteins CENP-E and CENP-F, being associated with AML, efforts were directed towards the development of farnesyl transferase inhibitors (FTIs). Preliminary results with tipifarnib (Zarnestra) as a single agent in the treatment of AML in patients over the age of 60 years suggest a CR rate of 18 per cent.54 Trials assessing the role of the drug as maintenance are also in progress. Histone deacetylase and proteasome inhibitors Histone acetylation is required for chromatin remodelling, a step which appears important in leukaemogenesis. Histone deacetylases (HDACs) promote an opposite effect and restore chromatin conformation. Inhibitors of HDACs, such as suberoylanilide hydroxamic acid (SAHA), valproic acid, depsipeptide and MS-275, all induce differentiation of leukaemic cells and are currently in clinical trials. The proteasome inhibitor bortezomib (Velcade) has in vitro synergistic activity with the HDACs and is being assessed in a phase I dose escalation study in combination with idarubicin and cytarabine induction therapy.55 Antiangiogenesis agents Increased levels of vascular endothelial growth factor (VEGF) have been associated with a poor survival in AML patients. VEGF has also been shown to stimulate growth
988 Leukaemias
and proliferation of leukaemic cells. Preliminary results with bevacizumab, a VEGF antibody, and SU5416, a small molecule that inhibits phosphorylation of VEGF receptors 1 and 2, appear encouraging. For example, in patients with refractory/relapsed AML, bevacizumab given after the completion of conventional chemotherapy on day 8 resulted in a CR rate of 33 per cent.56 FLT3 and other tyrosine kinase inhibitors Following the observations that abnormalities, including certain mutations, in the FLT3 receptor tyrosine kinase occur in over 30 per cent of patients with AML and often confer a poor prognosis, efforts were directed towards the development of FLT3 inhibitors. Currently several FLT3 inhibitors are being studied: PKC-412 (Novartis), CEP-701 (Cephalon), MLN518 (Millennium) and SU11248 (Sugen). All appear to be tolerated well after oral administration and inhibit both the unmutated and mutated FLT3 targets which, interestingly, translates to a very modest clinical efficacy so far.57 Imatinib has some activity in patients whose leukaemia cells are KIT positive. These and other similar trials are being continued.58 Apoptosis inhibitors High levels of the apoptosis inhibitor protein Bcl-2 expression in AML are associated with a poor prognosis. A phase I trial of Bcl-2 antisense oligonucleotide (oblimersen sodium, Genasense) together with conventional chemotherapy in previously untreated patients over the age of 60 years resulted in a CR rate of 45 per cent.59 A randomized phase III study assessing the role of Bcl-2 antisense oligonucleotide both in remission induction and post remission therapy is currently in progress. Deoxynucleoside analogues Phase II studies assessing the role of clofarabine (2-chloro2fluoro-deoxy-9-β-D-arabinofuranosyladenine), a novel nucleoside analogue developed from fludarabine and cladribine, administered in combination with cytarabine for relapsed and refractory AML patients resulted in a CR rate of 22 per cent.60 Two recent trials assessing clofarabine as a single agent for remission induction therapy in previously untreated patients over the age of 60 years, who were considered ‘unfit’ for conventional therapy, showed a CR rate of 60 per cent.61,62 In another trial, the BIOV-121, 66 patients aged over 65 years with untreated AML were enrolled to receive single agent clofarabine and the preliminary analysis suggests results superior to low-dose cytarabine in both unfavourable cytogenetic AML and all AML patients over the age of 70 years.63 A similar drug, troxacitabine, has just entered clinical trials. TREATMENT OF ACUTE PROMYELOCYTIC LEUKAEMIA
The successful outcomes for the majority of patients with APL come largely from the rapid incorporation into their
therapy of appropriate novel agents. The standard remission induction therapy for patients with APL now comprises ATRA with an anthracycline (daunorubicin or idarubicin).64,65 Cytarabine can probably be omitted from the induction therapy, though this issue is not completely resolved.66 A recent study showed that patients with APL fare just as well without cytarabine as part of their induction or consolidation treatment.67 The European APL group is addressing this issue further with patients being randomized to ATRA plus daunorubicin or to ATRA plus daunorubicin and cytarabine. There have been no randomized trials addressing the choice of anthracyclines so far, though there is agreement about the dose. Recently there has been much interest in the notion of using arsenic trioxide (ATO), either as a monotherapy or in combination with ATRA, which appear to have synergy, for induction treatment.68 This study led by the Shanghai group, who had previously conducted the landmark ATRA trial in 1988, randomized patients to receive ATRA, ATO or the combination of ATRA and ATO. The CR rates were more than 90 per cent for all the cohorts, but the combination therapy resulted in a significantly shorter time to achieve a molecular CR. More recently, groups from India and Iran, involving a smaller number of patients reported similar CR rates with single agent ATO as induction therapy.69,70 The results suggest that the majority of patients were able to achieve a CR, though the early mortality of about 10 per cent persisted. These results suggest that it is possible to omit not only cytarabine, but also anthracyclines, from the remission induction regimens, at least for the low- and intermediate-risk patients. Gemtuzumab ozagamicin (GO) is also effective in APL and various combinations of this with ATRA and ATO are being studied.71,72 Though these results are clearly very promising, until the longer term results involving larger number of patients are available, the standard induction approach should remain ATRA plus chemotherapy.73 From a therapeutic perspective, patients with APL can be stratified as low-, intermediate- or high-risk for relapse based on an initial white blood cell (WBC) count less than or greater than 10.0 109/l and platelet count greater or less than 40 109/l.74 It is reasonable to commence induction with ATRA as a single agent for 2 to 4 days, provided the WBC count is not high, as it is remarkably effective in controlling the characteristic coagulopathy, which can often be life threatening, and thereafter to commence anthracycline treatment. Earlier studies suggested CR rates of 80–90 per cent with ATRA alone, but the remissions were shorter than those produced by ATRA followed by chemotherapy or chemotherapy alone.75 Despite the qualified success of ATRA, the mortality associated with the induction therapy remains about 10 per cent and an acquired retinoid resistance contributes to relapse in about 20–30 per cent of patients. ATRA does have the serious side effect of capillary leak syndrome often associated with leukocytosis (referred to as the ‘ATRA syndrome’). Ongoing studies should show how best to integrate this novel molecular targeting and conventional chemotherapy.
Acute myeloid leukaemia 989
For patients with APL who enter CR, anthracyclineconsolidation therapy is required, though there is some uncertainty with regard to the number of cycles needed. Studies have also suggested that all patients should receive maintenance therapy with ATRA, with or without lowdose chemotherapy.76 There is some concern with regard to the possible increased neurotoxicity associated with ATRA maintenance therapy, particularly in children.77 Therapy-related MDS and secondary AML is also being increasingly recognized in APL patients and better risk stratification and monitoring of residual disease by polymerase chain reaction (PCR) should facilitate better individualisation of therapies.78 ATO is considered very effective treatment for those patients who relapse or are refractory to ATRA-based therapy.79,80 It is now accepted by most, but not all, specialists that patients with APL who achieve molecular negativity (by realtime quantitative PCR, RQ-PCR) for PML-RARα fusion transcripts after post-remission therapy do not require any further therapy.76 Patients who are not in a molecular remission after completing post-remission therapy should be considered as induction failures. Standard maintenance therapy consists of 6 months of low-dose chemotherapy, such as methotrexate and 6-mercaptopurine, perhaps with ATRA. ALLOGENEIC STEM CELL TRANSPLANTATION
Allogeneic SCT (allo-SCT) from an HLA-matched sibling donor has been standard treatment for AML in first CR since the 1980s.81,82 It offers a 50–60 per cent chance of long-term remission. The risk of relapse following an allograft is approximately 20 per cent. This relatively low risk of relapse is largely due to the immunological attack of donor cells against the leukaemia, usually referred to as the ‘graft-versus-leukaemia’ (GvL) effect. Conversely 20–30 per cent of patients experience severe toxicity or even mortality attributable to transplant-related complications, such as immunosuppression and graft-versus-host disease (GvHD). This high incidence of complications must therefore be weighed against the anticipated benefits of an allo-SCT. There have been no prospective randomized trials of allografting, but studies comparing patients in first remission with suitable donors with those without such donors have confirmed that patients who received allografts survived longer than controls.83,84 Recent reports suggest that transplant-related mortality has gradually decreased over the past decade or so, largely due to improved supportive care.85 Allogeneic stem cell transplantation in first complete remission Patients with good-risk AML in first CR are not usually offered conventional allo-SCT in first remission, since their probability of relapse is less than 35 per cent. If however they do relapse, they may be offered transplants in second remission. In contrast, patients with poor-risk AML should be offered an allo-SCT if a suitable donor can be identified. Patients with intermediate-risk disease should
be assessed on an individual basis, but most will probably not receive an allo-SCT in first CR. It may be possible to improve the outcome further by adding novel agents, such as targeted radioimmunotherapy delivered by 131Ianti-CD45 antibody combined with dose-adjusted busulphan and cyclophosphamide.86 Allogeneic stem cell transplantation for advanced acute myeloid leukaemia Allo-SCT is often the preferred treatment for any relapsed AML patient who has a suitable donor and wishes to be transplanted. The outcome of SCT for advanced disease is, however, inferior to that of SCT in first CR, due to an increased risk of transplant-related mortality (about 25– 30 per cent) and relapse (40–45 per cent). The results are even poorer for those who might have been subjected to a prior SCT. Allogeneic stem cell transplantation for remission induction failure Patients who fail to achieve CR following after 1 or 2 courses aimed at remission induction have a dismal prognosis. Small series of patients subjected to allo-SCT for remission induction failure suggest that about a third of all patients might achieve a CR following an allo-SCT. Older patients and those with the poor-risk cytogenetics tend to fare very poorly and probably should not be allografted. Reduced-intensity allogeneic stem cell transplantation Much interest has focused on reduced intensity conditioning regimens prior to allo-SCT over the past decade. This strategy is based on an attempt to maximize the contribution of donor-derived lymphoid cells in mediating a GvL effect while at the same time minimising the use of cytotoxic drugs designed to kill leukaemia cells. The mainstay of such regimens is typically fludarabine, a potent immunosuppressive agent. Studies have confirmed that reducing the intensity of the conditioning does indeed reduce the 100-day transplant-related morbidity and mortality but at the expense of an increased relapse rate.87,88 This has led to the search for more effective anti-leukaemic regimens, which would avoid the toxicity of the conventional high-dose regimens. Two such approaches have yielded promising early results, the addition of gemtuzumab ozagamicin or radiolabelled monoclonal antibodies to the conditioning regimens and the use of ‘moderate’ dose busulphan or melphalan. Currently reduced-intensity alloSCT for AML should probably be offered only in the context of clinical trials. Allogeneic stem cell transplantation for acute promyelocytic leukaemia Allo-SCT has no role in the management of APL in first remission since the results from current induction treatments are excellent. There are, however, no studies
990 Leukaemias
that have compared the results of SCT in relation to the induction therapy containing ATRA. Follow-up of 362 patients with APL from the database maintained by the European Group for Blood and Marrow Transplantation (EBMT) who received SCT in first CR in the pre-ATRA era suggests that 45 per cent had long-term remissions and were possibly cured.89 Allo-SCT may be useful for patients who relapse following an ATO-induced second CR. Impressive results have also been observed in patients with relapsed disease who received autografts using stem cells in which no PML-RARα transcripts could be detected.90
following a remission of at least 12 months, the survival rate following subsequent chemotherapy treatment is about 20 per cent.96 For these reasons every effort is directed to prevent relapse. For children and adults younger than 40 years of age, it is reasonable to proceed with an SCT, either an autograft or an allograft using an HLA-matched sibling donor. It is uncertain whether these patients should first receive induction therapy or proceed directly with transplantation. An earlier study from the Seattle group assessed these aspects and showed that survival with both approaches was equivalent.97 Current EBMT registry data suggest a survival rate of about 30 per cent for patients with AML in first relapse or second CR treated by an autologous or allogeneic SCT using an HLA-identical sibling donor.98
AUTOLOGOUS STEM CELL TRANSPLANTATION
Autologous SCT has also been widely used for the past two decades and data from the EBMT suggest a long-term survival rate of 45 to 55 per cent.91 Relapse is the most common reason for failure, presumably due either to residual disease in the patient or to absence of a putative GvL effect in the autograft procedure or to a combination of both causes. This is offset by the appreciably lower risk of transplant related mortality, although 5 to 8 per cent of autografted patients die in CR, often because of poor engraftment. There is no evidence that in-vitro ‘purging’ of the harvested cells in an attempt to kill residual leukaemia cells confers additional benefit.92 The MRC AML-10 trial assessed the value of adding autologous SCT for a cohort of patients who had already received intensive post-remission therapy. This trial showed a substantial reduction in risk of relapse with autologous SCT (37 per cent vs. 58 per cent; p 0.0007) resulting in a superior 7 year event-free survival (53 per cent vs. 40 per cent; p 0.04) despite a higher mortality rate in the transplantation group. Parenthetically, the event-free survival advantage was only in patients who lived beyond 2 years of the transplant. Several other European comparative studies reveal similar findings.93 A US Intergroup trial comparing high-dose cytarabine (3 g/m2 every 12 h for 12 doses) with autologous or allogeneic SCT during first remission of AML showed an equivalent event-free survival and a better overall survival with chemotherapy than autologous or allogeneic SCT.94 Another study showed a doseresponse effect of cytarabine in patients with AML, including those in the poor risk group.95 These studies have introduced an element of uncertainty as to how best to treat patients under the age of 60 years in first CR. Current comparative trials are assessing the potential improvements in autologous SCT by better engraftment with the use of peripheral blood stem cells and it should be possible to better identify patients who can benefit from salvage transplantation.
TREATMENT OF ACUTE MYELOID LEUKAEMIA IN RELAPSE
For patients who relapse with AML, the treatment options are few. For patients in the good-risk group who relapse
Conclusions Considerable advances have been made in the molecular understanding of AML which have led in some cases to studies that target leukaemia cells with molecular and immunologic strategies. The molecular pathogenesis of AML is quite complex and it is likely that multiple targeted approaches should be used in concert, in contrast to CML in which the targeting of single molecular abnormality has been successful.99 Similar success has been seen in patients with APL being treated with ATRA or arsenic oxide. It is remarkable that over a relatively short period of time, focus on the management of APL has moved to strategies that have permitted conventional cytotoxic drugs to be omitted entirely.100 Currently a myriad of novel agents designed to inhibit important and diverse molecular abnormalities, ranging from activated tyrosine kinases such as Flt3 and essential components of signal transduction pathways such as RAS, are in clinical trial and some of these are already in routine use. These molecularly targeted therapies may improve current treatments, both remission induction and post remission, and should also reduce toxicity.101 Further improvements in peripheral blood stem cell transplant technology and the newer emphasis on stem cell transplant as potential immunotherapeutic approach using methods designed to augment the GvL effect may improve the cure rates for patients under 60 years of age, whilst new approaches are needed for treating patients over 60 years of age.
ACUTE LYMPHOBLASTIC LEUKAEMIA Acute lymphoblastic leukaemia (ALL) most commonly affects children, particularly those between 2 and 10 years of age. It accounts for 80 per cent of all childhood leukaemias and is the most common type of cancer in children.102 It also affects adults, mainly those between 30 and 50 years of age, accounting for 20 per cent of all adult acute leukaemias. Approximately 80–85 per cent of ALL arises from a precursor B-cell phenotype.
Acute lymphoblastic leukaemia 991
Epidemiology The annual world-wide incidence of ALL is about 2 per 100 000 children and about 0.7 per 100 000 adults. The peak incidence in the UK and the US appears to be between the ages of 3 and 5 years. In some countries, such as Turkey, ALL may be less prevalent than AML in children. The disease appears to afflict more males than females and there appears to be a higher prevalence in Caucasians. ALL has been linked more often than AML to a possible infectious aetiology, in particular Epstein–Barr virus (EBV) and malaria, but no firm proof has emerged.16 There is, however, a firm association between one of the several known strains of the human T-cell lymphotrophic virus (HTLV), HTLV-1, and adult T-cell leukaemia/lymphoma (ATL). Epidemiological studies have suggested that a subtype of childhood ALL, common ALL, may arise as a consequence of rare and immunologically abnormal response to a common infection in genetically susceptible children.103 There is also evidence for a genetic aetiology in some other cases.
Box 41.3 The French–American–British (FAB) and World Health Organization (WHO) classification of acute lymphoblastic leukaemias (ALL) FAB classification of ALL L1: Small homogeneous, high nuclear:cytoplasmic ratio, small nucleoli L2: Larger, pleomorphic, low nuclear:cytoplasmic ratio, prominent nucleoli L3a: Larger, vacuolated basophilic cytoplasm, large vesicular nucleus, large nucleoli; resemble Burkitt’s lymphoma cells WHO classification of ALL Precursor B-lymphoblastic leukaemia/lymphoblastic lymphomab (precursor B-cell acute lymphoblastic leukaemia) (equivalent to B-cell ALL, FAB L1 and L2) Precursor T-lymphoblastic leukaemia/lymphoblastic lymphomab (precursor T-cell acute lymphoblastic leukaemia) (equivalent to T-cell ALL, FAB L1 and L2) a
Clinical Most patients with ALL present with features of bone marrow failure and symptoms resulting from organ infiltration by leukaemic lymphoblasts, in particular bone pain and arthralgias. Extramedullary involvement is not uncommon, with central nervous system, lymph nodes, liver, spleen, skin and gonads affected. Rarely patients present with respiratory difficulties as a consequence of a mediastinal mass. The diagnosis of ALL is based on a combination of morphological features, cytochemical reactions and immunological markers. The disease is biologically and clinically heterogeneous. In childhood ALL, the overall prognosis is excellent, while in adults with ALL the prognosis is generally poor. This difference appears to be due to a much higher frequency of poor risk factors and the likelihood of an intrinsically different biological disease in older patients, since age appears to be an independent prognostic factor.
Classifications
The FAB ALL L3 category has been put into the Burkitt lymphoma/leukaemia group by the WHO. b The WHO felt that, because of the biological unity of ALL and Lymphoblastic lymphoma, the use of one or the other term is arbitrary. When the disease process is confined to a mass lesion with no or minimal evidence of blood and less than 25 per cent marrow involvement, the diagnosis is lymphoblastic lymphoma; with blood and greater than 25 per cent marrow involvement, ALL is the appropriate term.
when there is a significant blood and over 25 per cent marrow involvement, the disease is designated ALL. Patients are classified as having a precursor B-cell or T-cell neoplasm. Precursor B lymphoblastic leukaemia (B-ALL)/lymphoblastic lymphoma (B-LBL) is defined as a neoplasm of lymphoblasts committed to the B-cell lineage, typically composed of small- to medium-sized blast cells (Fig. 41.3), involving bone marrow and blood (B-ALL) and sometimes nodal and extranodal sites (B-LBL). Precursor T-lymphoblastic leukaemia (T-ALL)/lymphoblastic lymphoma (T-LBL) is a neoplasm of lymphoblasts committed to T-lineage, also composed of small to medium-sized blasts (Fig. 41.4), involving bone marrow and blood (T-ALL) and occasionally nodal and extranodal sites (T-LBL).
WHO CLASSIFICATION
The WHO classification of ALL categorizes the lymphoblast by analogy to its normal counterpart in the B- or T-lymphoid lineages, and recognizes the overlap between ALL and lymphoblastic lymphoma (LBL) (Box 41.3).104 It considers ALL and LBL as a single biological entity and the use of one or the other term depends on the clinical presentation. When the disease process is confined to a mass lesion without a significant blood or less than 25 per cent bone marrow involvement, the preferred diagnosis is LBL;
FAB CLASSIFICATION
The FAB classification recognizes three distinct subtypes solely on morphological criteria: L1, L2 and L3 (see Box 41.3).105 In the subtype L1, the lymphoblast is relatively small with scanty cytoplasm and inconspicuous nucleoli. In the subtype L2, the lymphoblast is larger with more abundant cytoplasm. In the subtype L3, the lymphoblast is mature and resembles the Burkitt’s lymphoma cell (Fig. 41.5). Since there is no clear relationship between
992 Leukaemias
Figure 41.4 A peripheral blood film from a patient with acute lymphoblastic leukaemia, FAB subtype L3, showing vacuolated lymphoblasts ( 960). Kindly provided by Professor Barbara Bain, Imperial College School of Medicine, London.
the FAB subtypes and the immunological surface markers, a parallel immunological classification attempted to categorize the lymphoblast by analogy with its normal counterpart in the B- or T-lymphoid lineages. Moreover, in contrast to AML, ALL lymphoblasts lack significant cytochemical features, emphasising the diagnostic role of immunophenotyping. Cytogenetic analysis of the lymphoblasts has revealed an abnormal karyotype in more than 90 per cent of cases studied and led to the adoption of an integrated classification based on FAB subtype, immunophenotype and cytogenetic category. This modified classification appears to be much more clinically useful and also facilitates the addition of new categories as they are recognized. Approximately 20 per cent of all ALL lymphoblasts are T cell in origin, 75 per cent are precursor B cell, and 5 per cent are more mature B cells (usually FAB subtype L3). Acute undifferentiated leukaemia (AUL) is uncommon. In 25 per cent of children and about 35 per cent of adults with ALL, the lymphoblasts express both lymphoid and myeloid antigens. This feature was previously associated with a poorer prognosis, but it is now considered to have no prognostic or therapeutic implication.106
Molecular pathogenesis The recognition of genetic abnormalities in the majority of lymphoblasts has contributed enormously to understanding the molecular pathogenesis and prognosis of ALL.107,108 The cytogenetic information collated can be used to stratify patients according to whether they would benefit from more intensive approaches. More recently, a number of mutations in specific signalling molecules have been described, which could be targeted therapeutically in the near future.109 In B-ALL/B-LBL, the various cytogenetic abnormalities can conveniently be divided into several groups: hypodiploid, hyperdiploid with either less than or more
than 50 chromosomes, translocations and pseudodiploid (Box 41.4). Hyperdiploidy with more than 50 chromosomes is found in up to 25 per cent of children and 6 per cent of adults with ALL and is associated with an excellent prognosis since the lymphoblasts demonstrate unique susceptibility to antimetabolites.110,111 The observation that about half of these patients develop additional cytogenetic abnormalities, in particular duplications of chromosome 1q and isochromosome of 17q, has led to the hypothesis of a probable ‘two-hit’ genetic event resulting in a transformed phenotype that may not respond to therapy as well. Approximately 25 per cent of adults and 4 per cent of children have a Philadelphia (Ph) chromosome [t(9;22)(q34;q11)]. One third of ALL patients have BCR-ABL transcripts indistinguishable from those found in CML. The remaining two thirds have a breakpoint in the first intron of the BCR gene (between e1 and e2) in an area designated the minor breakpoint cluster region (m-bcr); this is transcribed as an e1a2 mRNA, which encodes a 190 kD protein (p190BCR-ABL), in contrast to the p210BCR-ABL typical of CML.112 About a fifth of all children with ALL have a t(12;21) (p12;q22), which results in the formation of a chimeric gene consisting of the 5 portion of the TEL (also known as ETV6) gene linked to an almost complete AML1 gene (also known as RUNX1). The resulting chimeric oncoprotein acts as a putative transcriptional repressor of the interleukin-3 (IL-3) promoter, thought to be central to leukaemogenesis.113 An elegant analysis of neonatal Guthrie blood-spot cards prepared at the time of birth in children who subsequently developed ALL between 2 and 5 years revealed evidence that the TEL-AML1 fusion gene must have been present already in utero. This means that this subtype of childhood ALL may have its inception early in fetal life.114 Studies in identical twins show however that an
Box 41.4 Prognostic groups with current treatment of acute lymphoblastic leukaemias Good prognostic group Hyperdiploidy between 51 and 65 chromosomes t(12;21)(p21;q22) Intermediate prognostic group del(6q) del(9p) del(12p) Hyperdiploidy less than 51 Poor prognostic group t(9;22) t(4;11) t(1;19) Hypodiploidy
Acute lymphoblastic leukaemia 993
additional postnatal genetic event is also required for clinical evidence of the leukaemia.115 Structural abnormalities involving chromosome 11q23, with fusion of the MLL gene, account for approximately 6 per cent of children and adults with ALL but are the most frequent genetic abnormalities observed in infants. The most frequent 11q23 abnormality, the t(4;11), is associated with mixed-lineage disease and usually carries a dire prognosis. Interestingly, as many as 20 per cent of MLLrearranged ALL patients also have various mutations involving the receptor tyrosine kinase FLT3 gene. Translocations involving t(1;19), found in 5 per cent of children and 3 per cent of adults with ALL, result in fusion of the gene encoding the E2A transcription factor on chromosome 19 with that encoding the Pbx1 transcription factor on chromosome 1. The resulting chimeric protein leads to leukaemias in mice that are characterized by a block in differentiation and growth factor dependence.116 An uncommon translocation, t(11;17) results in the fusion gene E2A-HLF, which is known to inhibit apoptosis.117 In patients with B-ALL/B-LBL, with Burkitt morphology, it is not uncommon to find a t(8;14) translocation or its variant t(2;8) or t(8;22). As a consequence of these rearrangements, the MYC gene is dysregulated and in the presence of other co-operating mutations, such as K-RAS and N-RAS, this results in the malignant process.118 Genes affected in a similar manner in T-ALL/T-LBL include SCL (TAL-1), LMO1 (formerly known as RBTN1 or TTG-1), LMO2 ((formerly known as RBTN2 or TTG-2), and the cytoplasmic tyrosine kinase gene LCK (on 1p34.3–35), which are essential for the development of all haemopoietic lineages.119 In the translocation t(1;14), which is present in about 5 per cent of T-ALL, SCL is rearranged into the TCRδ gene and carries a poor prognosis.120 The translocation t(11;14)(p15;q11) results in rearrangement of LMO1 into the TCRα/δ locus, whereas the t(11;14)(p13;q11) results in the rearrangement of LMO2 into this locus. Several other translocations involving a host of other developmental genes have also been described. For example, t(10;14) and t(7;10) involve the fusion of HOX11 and tend to be associated with a good prognosis in children treated with intensive therapies; t(5;14) involves the fusion of HOX11L2 to the TCRδ locus. Rarely, NOTCH1 gene mutations are found in T-ALL involving t(7;9).121,122
Prognostic factors Patients are stratified into various risk groups in accordance with well described clinical, biological, morphological and, in particular, molecular criteria at diagnosis. Good prognostic groups include those with hyperdiploidy between 51 and 65 chromosomes and those with t(12;21). The intermediate prognostic group includes del(6q), del(9p), del(12p), hyperdiploidy less than 51, near triploidy and near tetraploidy. The poor prognostic groups include the presence of t(9;22), t(4;11), t(1;19) and hypodiploidy.107,123
Adults appear to have a relatively high frequency of unfavourable molecular changes and their prognosis is considerably worse than that of children with equivalent molecular changes and immunophenotypes.124
Treatment GENERAL PRINCIPLES
The conventional approach for all patients with ALL, with the exception of some with certain ALL variants, such as the Ph-positive ALL, is to offer an intensive induction treatment. The benefits of an intensive consolidation followed by a maintenance treatment plan are unclear in adults.123 In general, the results are better in children, in whom a maintenance treatment also appears to accord some benefit. All patients with ALL who achieve CR should receive some form of central nervous system (CNS) prophylaxis. The treatment of the Ph-positive ALL will be addressed separately. Another area of debate is the role of each individual drug in the various treatment combinations that have evolved over the past few decades.124 In children, for example, dexamethasone is superior to prednisone, both in terms of LFS and reducing the risk of relapse in the CNS.125 Moreover dose-intensification of the various non-myelosuppresive drugs, such as vincristine, steroids and L-asparaginase, results in an improved overall survival in children.126 REMISSION INDUCTION
The principal aim is to achieve CR with restoration of normal haemopoiesis. Better supportive care has made it safer for patients with standard and poor-risk ALL to receive more intensive induction regimens with four or more drugs and over 98 per cent of children and 80 per cent of adults should achieve a CR. With consolidation, about 80 per cent of children and 30 to 40 per cent of adults will remain in CR at 5 years.127,128 Universally induction therapy includes a glucocorticoid, prednisolone or dexamethasone, vincristine and L-asparaginase. The Italian GIMEMA group demonstrated the value of a 7-day prednisolone pre-induction therapy for predicting the outcome in adult ALL patients.129 The realization that the rapid reduction of lymphoblasts following the initial therapy suggested that the intensity of this phase of the treatment plan is important has led to efforts to intensify it. For this reason, an anthracycline, usually daunorubicin or doxorubicin, and sometimes also cyclophosphamide were added, in particular for adult patients. Likewise the subsequent phases of the treatment plan were also made more intensive. It is possible that dexamethasone may provide better protection than prednisolone against a CNS relapse since it more readily crosses the blood–brain barrier.125,130 Nevertheless all patients with ALL who achieve CR should
994 Leukaemias
receive some form of prophylaxis directed specifically at the CNS. The standard approach has been to administer fractionated cranial irradiation totalling 1800 cGy in conjunction with intrathecal methotrexate. However, it seems now that the dose of cranial irradiation can safely be reduced to 1200 cGy, even in patients at high risk,131 because this lower dose of irradiation may reduce the risk of late cerebral toxicity. Intrathecal methotrexate without irradiation has resulted in comparatively high rates of CNS relapse, and it is possible that methotrexate in combination with cytarabine and dexamethasone (triple intrathecal therapy) might be better. It could be that the need for additional irradiation has actually diminished,132 since most induction regimens nowadays include drugs that are able to cross the blood–brain barrier. Recently, however, the results of the Dana-Farber Cancer Institute (DFCI) ALL consortium protocol at a median follow-up of 9.2 years suggested that eliminating cranial irradiation without a concomitant increased intensity of systemic or intrathecal therapy results in a marked increase in CNS relapse in standard-risk male children, but interestingly not female children.133 Two forms of L-asparaginase, each with a different pharmokinetic profile, are currently available. In a single randomized study, patients treated with L-asparaginase derived from Escherichia coli fared better than those treated with that derived from Erwinia carotovora. Studies have also compared the efficacy of mitoxantrone and daunorubicin and found them to be equivalent. Parenthetically, it should be emphasized that, as attempts have been made to increase the intensity of the remission induction, complex regimens using multiple drugs have entered the clinic at frequent intervals without being tested rigorously in randomized trials. This has made it difficult to assess with any precision the results of the more intensive treatments, nor is it possible to draw firm conclusions on the contribution of each individual drug. These points notwithstanding, most current treatment protocols are well received by the specialists in view of the improving outcomes of the patients. More intensive treatment strategies, based on experiences from the treatment of children and young adults with Burkitt’s lymphoma, have resulted in the inclusion of fractionated high-dose cyclophosphamide, high-dose methotrexate and cytarabine (standard dose). These modifications led to the CR rates improving from about 35 per cent to 75 per cent and the event-free survival to around 50 per cent. These improvements have been ascribed to the high-dose components of the protocols, although it is difficult to assess which of the high-dose combinations is more important and the optimal doses remain debatable. Further refinements have occurred since, and the currently reported CR rate is 85 per cent.134,135 For the moment the favoured dose of cyclophosphamide is 8 g/m2 and that of methotrexate is 5 g/m2. Once a CR is achieved, it is now customary to administer intensive consolidation for about 8 months. Since the high-dose drugs used are active in CNS disease, patients usually receive additional intrathecal methotrexate but not cranial irradiation.
The best responses, in particular for patients with Burkitt-type ALL (FAB L3), have been reported with the hyper CVAD regimen, which includes hyperfractionated cyclophosphamide, vincristine, doxorubicin and dexamethasone alternating with high-dose methotrexate and cytarabine.136,137 In a study of 288 patients with adult ALL investigators in Houston reported a CR of 92 per cent. The 5-year overall survival was 38 per cent. The DFCI ALL Consortium is testing the feasibility and efficacy of an age-unrestricted approach in the treatment of ALL in children and adults. The consortium had previously demonstrated a 5-year LFS of 77 per cent for adolescents aged 15–18 years treated between 1991 and 2000.138 Since this result was similar to the 5-year LFS of 83 per cent observed in children and adolescents aged 1–18 years, the consortium elected to enrol all patients aged 18–50 years on a similar intensive protocol (no. 00–01). The preliminary results based on 34 adults, 31 per cent with a T-cell phenotype and 18 per cent with Ph-positive ALL, showed a CR of 79 per cent and, importantly, confirmed the feasibility of a dose-intensification protocol in adults with highrisk ALL.139 The Dutch group have reported similar results.140 POST-REMISSION THERAPY
Post-remission therapy is designed to eradicate minimal residual disease (MRD). Once remission is achieved and normal haemopoiesis restored, it is unclear how intensive the subsequent treatment should be. Historically, low-risk patients were offered a less myelosuppressive treatment and high-risk patients were subjected to myelosuppressive therapy, including the use of SCT.141 Several recent studies have confirmed that intensified post-remission therapy was not beneficial.142 Studies have also compared 3 months and a single month of intensive consolidation and found no difference. Most less myelosuppressive schedules incorporate the use of daily 6-mercaptopurine and weekly oral methotrexate concomitantly with a once monthly dose of alternating daunorubicin, vincristine and cyclophosphamide or a similar alternative schedule. Following the completion of consolidation therapy, the optimal duration of which remains uncertain, most patients in continuous CR, with the exception of those with mature B-cell phenotype, are offered maintenance treatment for a few years. The value of this approach is unclear but most specialists believe that long-term multidrug exposure may aid in eliminating the residual lymphoblasts. A recent meta-analysis of trials assessing the value of maintenance therapy concluded that it was probably useful, particularly in children, but not for longer than 3 years.143 Ph-POSITIVE ACUTE LYMPHOBLASTIC LEUKAEMIA
Prior to the introduction of the original Abl tyrosine kinase inhibitor, imatinib mesylate (IM), most patients with
Acute lymphoblastic leukaemia 995
Ph-positive ALL tended to have a very poor prognosis and few achieved durable CR. A small trial was carried out soon after IM became available in which 10 patients with Ph-positive ALL who had failed conventional induction therapy or had relapsed after such therapy, were treated with IM in a dose-escalation manner.144 The overall response rate was 70 per cent, with 20 per cent of patients achieving a CR, though the median duration of remission was only 58 days. A subsequent trial confirmed similar responses, which again were short-lived.145 Thereafter, a number of studies have assessed the potential of combining IM with intensive chemotherapy and, in general, the results were better.146,147 The notion was to achieve a rapid response and thereafter proceed to an allo-SCT consolidation. In a CALGB study (10001), patients received either allogeneic or autologous SCT as consolidation.148 Preliminary analysis of results from 7 of the 18 patients in this study (allo-SCT, n 3 and autoSCT, n 4) demonstrated complete molecular remissions in 2 of the 4 who received an autologous SCT. The peripheral blood stem cells harvested were negative for BCR-ABL transcripts by RQ-PCR in 3 of the 4 patients autografted. So far the allo-SCT cohort have fared very badly with 2 of the 3 patients who were transplanted dying and the third relapsing on day 113. Studies are currently also assessing the benefit of maintenance IM following the SCT. Studies are in progress to assess the role of the second generation ABL kinase inhibitors, dasatinib (BMS354825) and nilotinib (AMN107). The START-L study is an openlabel phase II study of dasatinib in Ph-positive ALL and CML in lymphoid blast crisis who are resistant or refractory to imatinib. Of the 36 patients with Ph-positive ALL, there were 31 per cent complete cytogenetic responses (CCyR) and an additional 27 per cent of partial cytogenetic responses (PCyR).149 Preliminary results of nilotinib in a study of patients with relapsed/refractory Ph-positive ALL demonstrates encouraging results.150 TREATMENT OF RELAPSED ACUTE LYMPHOBLASTIC LEUKAEMIA
Despite the significant improvement in the therapy of ALL, in particular in children, about 25 per cent of children and 50 to 60 per cent of adults relapse. Second remissions can be achieved in most of these patients, but in many cases they are not sustained. Factors associated with a poor outcome after relapse include a shorter length of first remission, bone marrow as the initial site of relapse, older age, T-ALL and male sex. The MRC UKALL-R1 study showed that patients who had a bone marrow relapse within 24 months of diagnosis fare worst.151 This study showed that patients with an isolated extramedullary relapse have a better prognosis, even though many such patients also show evidence of bone marrow relapse at the molecular level.152 The MRC UKALL X trial showed that only 3 per cent of patients with a bone marrow relapse are alive in second remission at 5 years irrespective of the type of their second treatment.
There is considerable uncertainty with regard to the appropriate further treatment once a second remission is established. There have been no prospective randomized trials to compare chemotherapy and SCT, and the little evidence available is largely based on retrospective patient series. For patients who relapse more than 2 years after achieving CR, remission may be re-induced with the same drugs that induced the first remission. For primary resistant disease, most specialists use high-dose cytarabine either alone or in combination with an anthracycline. New agents, such as nelarabine (506U78), a prodrug that is rapidly demethylated to AraG, have substantial clinical activity in T-ALL in first relapse.153,154 Studies exploring its role in newly diagnosed patients with T-ALL are now being planned. Another new drug, clofarabine, a purine analogue discussed above, has recently been approved in the US for the treatment of relapsed ALL in children. ALLOGENEIC STEM CELL TRANSPLANTATION
Allo-SCT using an HLA matched donor, preferably genotypically matched, is often offered to patients with ALL who have failed remission induction or have achieved a second CR.155 Transplantation in first CR as consolidation therapy is considered unproved at present, but it may well be the desired treatment for those patients who fall in the poor-risk category.156 In patients with Ph-positive ALL who achieve a CR, allo-SCT is the preferred mode of consolidation.157 In patients who are in their second CR, it is probably reasonable to offer an allo-SCT if a suitable donor is available and the patient wishes to be transplanted, though the results are relatively poor, with a 3-year LFS of 12–23 per cent. AUTOLOGOUS STEM CELL TRANSPLANTATION
The precise role of autologous SCT remains unclear, despite encouraging preliminary results. Most comparative studies show no major survival difference compared with chemotherapy.158–160 A small Italian study suggested a possible role for autologous SCT in patients who sustain an early CNS relapse, but in general at present it is probably best to offer this treatment as part of a clinical trial.161 A prospective randomized study comparing allo-SCT with autologous SCT as consolidation for patients in first CR was carried out over a decade ago.162 Following a median follow-up of 30 months from CR, the 3-year leukaemia-free survival was significantly better for the allo-SCT cohort than for the autologous SCT recipients (68 per cent vs. 26 per cent, respectively). The study was, however, criticized since the outcome for patients who received autologous SCT was considerably poorer than that of historical studies.
Conclusions Though the results of intensive chemotherapy in children with good- and standard-risk ALL are relatively good, the
996 Leukaemias
treatment of the remaining children and most adults remains difficult with long-term survival rates around 25 to 40 per cent. Substantial efforts are being made to improve these outcomes, both in terms of understanding better the molecular pathogenesis the roles of new therapeutic agents and SCT.163 Allo-SCT in first remission is used in children and adults with high-risk ALL, such as Ph-positive ALL; SCTs, both allo-SCT and autologous SCT, are also useful to treat relapse after chemotherapy. Targeted immunotherapy strategies are also being explored in an attempt to induce immune response against the lymphoblast. The CD20 monoclonal antibody, rituximab, in combination with hyper-CVAD and the CD52 antibody, alemtuzumab, are being tested in ALL in first remission. Other new approaches include immunotoxins, measures to enhance immune modulation particularly following allogeneic SCT in an attempt to reduce the risk of relapse and gene therapy.164,165
CHRONIC LEUKAEMIAS CHRONIC MYELOID LEUKAEMIA Chronic myeloid leukaemia is a clonal disease that results from an acquired molecular change in a pluripotential haemopoietic stem cell. The leukaemia cells usually have a consistent cytogenetic abnormality, the Ph chromosome, which carries a BCR-ABL fusion gene. This gene encodes an oncoprotein, p210BCR-ABL, which has dysregulated tyrosine kinase activity and is believed to be the initiating event in the chronic phase CML.166–168 The discovery in the late 1990s that this kinase activity could be inhibited in a highly specific manner revolutionized the management of patients with CML following the successful introduction of the first generation of tyrosine kinase inhibitor (TKI), imatinib mesylate (IM, Gleevec; Glivec).169 Since then IM has become well established as optimal initial therapy for the majority of patients with newly diagnosed CML.170,171 Importantly, this ‘molecularly targeted therapy’ also marked the beginning of a new era in which relatively non-specific and often toxic drugs for treating malignant disease should gradually be replaced by safer and better tolerated agents whose precise mechanism of action is much better defined than that of their predecessors (Fig. 41.5). IM reduces substantially the number of leukaemia cells in a patient’s body and seems certain to prolong very considerably the median survival for chronic phase patients. However, complete molecular responses are still relatively rare, and allogeneic SCT is still regarded as the only modality that can in many cases eradicate all residual leukemia stem cells.172,173
Epidemiology The incidence of CML is about 1.5 per 100 000 of population per annum. It represents approximately 15 to 20 per
cent of all adult leukaemias and less than 5 per cent of all childhood leukaemias. The median age of onset in the Western world is 55 years, and there is slight male excess. Although in most cases there are no known predisposing factors, there is a marginally increased risk of developing CML following exposure to high doses of irradiation, as occurred in survivors of Hiroshima and Nagasaki atomic bombs in 1945.14 A small number of families with high incidence of the disease have been reported, and relapse of CML originating in donor cells following related donor allogeneic SCT has been recorded.174 Nevertheless it is extremely difficult to incriminate any single aetiological factor in individual patients with CML.
Clinical Characteristically, CML is a biphasic or triphasic disease that is usually diagnosed in the initial chronic phase (CP). This CP used to last 3 to 6 years but is today very significantly longer for patients treated de novo with imatinib or imatinibcontaining combinations. Figure 41.6 shows a peripheral blood film from a patient with typical CML in CP. About one third of CML patients are diagnosed following a routine blood test and the remainder present with signs and symptoms related to anaemia, spontaneous haemorrhage or splenomegaly. Before the advent of imatinib the disease progressed inexorably to blast transformation; about 70–80 per cent of patients entered a myeloid blast transformation and their survival was usually between 2 to 6 months, while those who entered a lymphoid blast transformation had a slightly longer survival. About half of the patients in the CP transformed directly into blast transformation and the remainder did so following a period of accelerated phase.175 It is likely that imatinib has prolonged survival in CML very considerably but the pattern of progression for patients in the ‘imatinib era’ cannot yet be reliably summarized.
Molecular pathogenesis The Ph chromosome is an acquired cytogenetic abnormality present in all leukaemic cells of the myeloid lineage and in some B cells and T cells in CML patients.176,177 It is formed as a result of a reciprocal translocation of genetic material of chromosomes 9 and 22 expressed as t(9;22)(q34;q11) (Fig. 41.7). The classical Ph chromosome is easily identified in about 90 per cent of CML patients. A further 5 per cent of patients have variant translocations which may be ‘simple’ involving chromosome 22 and a chromosome other than chromosome 9, or ‘complex’, where chromosome 9, 22 and other additional chromosomes are involved. About 5 per cent of patients with clinical and haematological features typical of CML lack the Ph chromosome and are referred to as ‘Ph-negative’ CML.
Chronic myeloid leukaemia 997
Curative intent
Palliative therapy Arsenic Spleen irradiation Busulfan
Hydroxycarbamide
Interferon alpha
Imatinib Second generation TKIs 1845
1903
1953
1964
1980
1983
1998
2008
Figure 41.5 Historical evolution of treatment of CML
Figure 41.6 A peripheral blood film from a patient with chronic myeloid leukaemia in the chronic phase ( 960). Kindly provided by Professor Barbara Bain, Imperial College, London. Figure 41.7 Partial karyotype of the Philadelphia chromosome translocation t(9;22)(q34;q11) showing the breakpoints on chromosomes 9 and 22.
About half of these patients have a BCR-ABL chimeric gene and are referred to as Ph-negative, BCR-ABL-positive cases; the remainder are BCR-ABL-negative and some of these have mutations in the RAS gene. The Ph-negative BCR-ABL-positive patients have a disease clinically indistinguishable from Ph-positive CML. In contrast the BCRABL-negative patients have a more aggressive clinical course. Some patients acquire additional clonal cytogenetic abnormalities, in particular 8, Ph, iso17q and 19, as their disease progresses. The emergence of such clones may herald the onset of blastic transformation. Various genetic events have now been elucidated and the chimeric BCR-ABL gene is believed to play a central role in the pathogenesis of CML, though the precise mechanism(s) are still not fully understood.178 Three distinct breakpoint locations in the BCR gene in chromosome 22
have been identified (Fig. 41.8). The break in the major breakpoint cluster region (M-BCR) occurs in the intron between exons e13 and e14 or in the intron between exons e14 and e15 (toward the telomere). By contrast, the position of the breakpoint in the ABL gene on chromosome 9 is highly variable and may occur at almost any position upstream of exon a2. The Ph translocation results in the juxtaposition of 5 sequences from the BCR gene with 3 sequences from the ABL gene. This event results in the generation of the chimeric BCR-ABL fusion gene transcribed as an 8.5 kb mRNA, which encodes a protein of 210 kD (p210BCR-ABL) that has a greatly enhanced tyrosine kinase activity compared with the normal Abl protein. The different breakpoints in the M-BCR result in two slightly
998 Leukaemias
Figure 41.8 A schematic representation of the various breakpoints in the ABL and BCR genes and the proteins encoded in BCR-ABL positive leukaemias.
different chimeric BCR-ABL genes, that in turn result in either an e13a2 or e14a2 transcript. The type of BCR-ABL transcript has no important prognostic significance. The second breakpoint location in the BCR gene occurs between exons e1 and e2 in an area designated the minor breakpoint cluster region (m-bcr) and forms a smaller BCR-ABL fusion gene. This is transcribed as an e1a2 mRNA, which encodes a p190BCR-ABL oncoprotein.179 This protein characterizes about two thirds of patients with Ph-positive ALL. A third breakpoint location is found in patients with the very rare Ph-positive chronic neutrophilic leukaemia.180 This breakpoint occurs on an area designated the micro breakpoint cluster region (μ-bcr) and results in e19a2 mRNA, which encodes a larger protein of 230 kD (p230BCR-ABL). The identification of several features in the Bcr-Abl oncoprotein that are essential for cellular transformation led to the characterization of signal transduction pathways activated in BCR-ABL-positive cells (Fig. 41.9). Much attention has since focused on determining the precise role played by the various Bcr-Abl proteins in the pathogenesis of CML.181 A number of possible mechanisms of BCR-ABLmediated malignant transformation have been implicated, which are not necessarily mutually exclusive. These include constitutive activation of mitogenic signalling, reduced apoptosis, impaired adhesion of cells to the stroma and extracellular matrix, and proteasome-mediated degradation
of ABL inhibitory proteins. The deregulation of the ABL tyrosine kinase facilitates autophosphorylation, resulting in a marked increase of phosphotyrosine on Bcr-Abl itself, which creates binding sites for the SH2 domains of other proteins. A variety of such substrates, which can be tyrosine phosphorylated, have now been identified. Although much is known of the abnormal interactions between the Bcr-Abl oncoprotein and other cytoplasmic molecules, the finer details of the pathways through which the ‘rogue’ proliferative signal is mediated, such as the Ras-Map kinase, the Jak-Stat, and the PI3 kinase pathways, are incomplete and their relative contributions to the leukaemic ‘phenotype’ are still unknown. Moreover, the multiple signals initiated by the Bcr-Abl have both proliferative and anti-apoptotic qualities, which are often difficult to separate. Moreover much remains to be learned about the significance of tyrosine phosphatases in the transformation process. It is generally believed that the some CML stem cells, at a cytokinetic level, are in a quiescent or dormant (G0) phase.182 These quiescent CML cells appear to be able to exchange between a quiescent and a cycling status, allowing them to proliferate under certain circumstances.183 This provides some rationale for autografting as treatment for CML. There is also evidence that some Phpositive cells are quiescent and cannot be eradicated by cycle-dependent cytotoxic drugs, even at high doses, or indeed by imatinib.184
Chronic myeloid leukaemia 999
Figure 41.9 Signal transduction pathways activated in BCR-ABL positive cells. The P symbol indicates molecules that appear to be constitutively activated in the presence of p210BCR-ABC
The Ph-positive cell is prone to acquire additional chromosomal changes, presumably as a result of acquired ‘genetic instability’, and this presumably underlies the progression to advanced phases of the disease. The average length of chromosomal telomeres in the Ph-positive cells is generally less than that in corresponding normal cells and the enzyme telomerase, which is required to maintain the length of telomere, is upregulated as the patient’s disease enters the advanced phases. There is no consistent pattern of molecular abnormalities in patients whose disease has progressed from chronic phase. About 25 per cent of patients with CML in myeloid transformation have point mutations or deletions in the p53 gene and about half of all patients in lymphoid transformation show homozygous deletion in the p16 gene. There is also evidence supporting the role of the RB (retinoblastoma) and the MYC genes in disease progression.
formula that takes account of the patient’s age, spleen size, blast cell count and platelet count at diagnosis.185 Stratifying patients into good, intermediate and poor risk categories may assist in the decision-making process regarding appropriate treatment options. The Euro or Hasford system is an updated Sokal index that incorporates the effects of increased basophil and eosinophil numbers.186 Another possible prognostic factor is the presence or absence of deletions in the derivative 9q chromosome, though its significance in the imatinib era is now controversial. Other factors may help to define prognosis after a patient has started treatment. For example, the rate of shortening of telomeres in the leukaemia clone may be informative. More recently, the response to imatinib may predict for survival without progression to advanced phase.187 Where allo-SCT is being considered, the EBMT has produced a risk-score based upon pre-transplant variables, which may predict the risk of mortality and relapse for patients treated by allo-SCT.188
Prognostic and predictive factors Treatment Efforts have been made to establish criteria definable at diagnosis that may help to predict survival for individual patients. Historically, the first useful method was that devised by Sokal and colleagues, whereby patients were divided into various risk categories based on a mathematical
GENERAL PRINCIPLES
It is now generally accepted that IM is indeed the best initial treatment for the vast majority of newly diagnosed patients
1000 Leukaemias
with CML in chronic phase (CP). IM induces complete haematologic remissions (CHR) in almost all such patients and a complete cytogenetic response (CCyR) in the vast majority. It is very likely that IM will prolong median survival by 10 or more years. Current experience suggests that about 2 per cent of all chronic phase patients progress to advanced phase disease each year, which contrasts with estimated annual progression rates of more than 15 per cent for patients treated with hydroxyurea and about 8–10 per cent for patients receiving interferon alpha (IFN-α), either with or without cytarabine. Many patients still show evidence of residual leukaemia at low level and IM will probably not eradicate residual CML in the vast majority of patients. A current central issue is therefore whether total eradication of all residual leukaemia stem cells is actually necessary, since the survival of small numbers of residual leukaemia stem cells might be compatable with long-term survival in an individual patient. This would be tantamount to cure at an operational level, as may well be the case after allogeneic stem cell transplantation.
Imatinib mesylate is a 2-phenylaminopyrimidine compound that inhibits the enzymatic action of the activated Bcr-Abl tyrosine kinase. It acts partly by occupying the ATP-binding pocket of the Abl kinase component of the Bcr-Abl oncoprotein and thereby blocking the capacity of the enzyme to phosphorylate downstream effector molecules (Fig. 41.10). It also binds to an adjacent part of the kinase domain in a manner than holds the Abl component of the Bcr-Abl oncoprotein molecule in an inactive configuration.190,191 The drug rapidly reverses the clinical and haematologic abnormalities of CML and induces CHR in over 95 per cent of previously untreated CP patients. It is usually administered orally at a dose of 400 mg/day, though one
important study suggests that initial treatment with 600 mg or 800 mg daily may give better results.192 Side effects include nausea, headache, rashes, bone pains, a spectrum of skin reactions and fluid retention. Significant cytopenias and hepatotoxicity occur less commonly. Very rare cases of severe or fatal cerebral oedema have been reported.193 The toxicity in general seems to be appreciably less than that associated with IFN-α.194 A multicentre randomized phase III trial (IRIS) designed to compare IM as a single agent with the combination of IFN-α with cytarabine in previously untreated CP patients started in June 2000 and by January 2001 1106 patients with untreated CML in CP had been recruited from 16 countries. The results following a 5-year follow-up were published in December 2006 and showed the estimated cumulative best rates of CHR and complete cytogenetic responses (CCyR) were 98 per cent and 87 per cent respectively (Fig. 41.11).189 These estimates include a small minority of patients who did achieve such responses but subsequently lost them or died of unrelated causes. The actual number of patients who were known to still be on IM at 5 years was about 68 per cent (Fig. 41.12).195 Survival without progression to advanced phase disease was significantly better for patients who received IM as initial therapy than for those who started treatment with IFN-α/Ara-C (97.2 per cent vs. 90.3 per cent; p 0.001), despite the fact that many patients in the control arm actually crossed over to receive IM as secondary therapy. In December 2007, 6-year follow-up results were presented and noted to show a CCyR of 83%. Results also suggest that less than 2 per cent of patients enter the advanced phases of the disease in each year on treatment but this percentage seems to be falling with each year of treatment. Furthermore, comparing survival in patients treated with IM with historical controls treated with IFN-α or IFN-α plus cytarabine provides strong support for the notion that the survival benefit is directly attributable to the improved cytogenetic response and is likely to be appreciable with IM.196
Figure 41.10 Schematic representation of the mode of action of STI571 (imatinib mesylate). Adapted from Goldman and Mughal 2001.
Figure 41.11 Cumulative best response at 12 and 60 months on first-line imatinib mesylate for patients with CML in CP enrolled onto the IRIS trial. Adapted, with permission from Druker et al, N Engl J Med, Dec 2006.
IMATINIB MESYLATE
Chronic myeloid leukaemia 1001
Patients responding to treatment with IM are best monitored by using RQ-PCR to measure BCR-ABL transcripts, which may show a steady decline to very low levels, although not all patients achieve a status where transcripts are undetectable.197 Patients who have achieved a CCyR or near-CCyR state should probably continue IM indefinitely, as most, though not all, patients appear to relapse once IM is discontinued. About 25 per cent of patients with CML in early chronic phase eventually become resistant to IM.198 Resistance may be primary or secondary and is more common in patients who start IM in late chronic phase. It is seen in about 70 per cent of those treated in myeloid blast crisis and all of those in lymphoid blast crisis. Primary resistance whereby a patient never achieves haematologic or cytogenetic remission is relatively uncommon but secondary resistance is well recognized and appears to result from a variety of diverse mechanisms, including amplification of the BCR-ABL fusion gene, relative overexpression of the Bcr-Abl protein, overexpression of P-glycoprotein that enhances the cellular removal of IM and expansion of a pre-existing sub-clone of Ph-positive cells with point mutations in the BCR-ABL kinase domain.199,200 Structural studies suggest that not all mutations, of which about 70 different mutations have been identified so far, are equivalent: T315I, E255K and some (but not all) P-loop mutations are often associated with poor prognosis, possibly because they interfere with IM-binding to the Abl kinase domain; in contrast mutations such as M244V are associated with very variable or no resistance.201 Thus currently there is considerable debate about the significance of these mutations in CML, particularly since some mutations have been identified in newly-diagnosed patients and probably reflect the natural evolution of the CML stem cells.184 The emergence of a mutation, particularly in patients who appear to be responding to IM, should not therefore automatically be considered as a failure of treatment. COMBINATIONS CONTAINING IMATINIB MESYLATE
For patients who develop resistance to IM, a number of newer inhibitors of Bcr-Abl have been developed and three
of these, dasatinib (Sprycel, Bristol-Myers Squibb), nilotinib (Tasigna, Novartis) and bosutinib (Wyeth) are currently in the clinic. There are a number of clinical trials which are assessing the benefits of adding other treatments to IM, following an initial trial of an increased dose of IM (up to 800 mg/day), based on in vitro studies suggesting potential synergism between IM and a number of conventional as well as investigational agents. Preliminary results from trials assessing combinations of IM and IFN-α are reasonably encouraging, but only relatively small doses of IFN-α can be tolerated in combination with full dose IM. More recent efforts are being directed towards use of the long-acting form of IFN-α, pegylated IFN-α, which was introduced a decade ago in the hope of reducing many of the IFN-α related toxic effects. An Italian study confirmed CCyR in 63 per cent of the cohort, but 58 per cent of the patients experienced toxicity, both haematological (50 per cent) and non-haematological (50 per cent).202 Small series of patients subjected to combinations of IM and low-dose cytarabine suggest a CCyR in 57 per cent but an excessive amount of side effects, both haematological and non-haematological, were noted.203 Over 50 per cent of all patients discontinued therapy at 9 months due to toxicity. Recently there has been considerable interest in combining IM with hypomethylating agents, such as 5-azacytidine (azacytidine; Vidaza) and 5-aza-2-deoxycytidine (decitabine), both of which disrupt the chromatin by inhibiting DNA methylation and thus lead to gene silencing.204 Both agents are active when used independently in small pilot studies, and preclinical studies show synergy when IM is added, particularly with decitabine. Other investigational approaches include combining IM with a farnesyl transferase inhibitor (FTI), such as lonafarnib (SCH66336) or tipifarnib (R115777) or BMS-214662.205,206 Homoharringtonine (HHT), a semi-synthetic plant alkaloid that enhances apoptosis of CML cells, is active in combination with IM in IM-resistant/refractory patients.207 Pilot studies using other agents, such as arsenic trioxide and bortezomib, in combination with IM are also in progress; preliminary results are encouraging.
Figure 41.12 Relative rates of Complete cytogenetic remissions (CCyRs) in previously untreated patients who started imatinib as initial therapy at time of various landmark analyses (adapted, with permission, from Golman J, Blood, 2007).
1002 Leukaemias
SECOND GENERATION TYROSINE KINASE INHIBITORS
Dasatinib (previously known as BMS-354825) is a thiazolecarboxamide structurally unrelated to IM. It binds to the Abl kinase domain with the activation loop in both active (open) and inactive (closed) conformations and also inhibits some of the SRC family kinases. Preclinical studies confirmed that dasatinib was 300-fold more potent than IM and is active against most known IM-resistant mutants, with the notable exception of the T315I mutant.208 In phase II studies, CHR was achieved in 84 per cent of patients resistant to IM, and in 100 per cent of those intolerant of IM. The corresponding major cytogenetic responses were 35 per cent and 50 per cent.209,210 No doselimiting toxicity was identified. Dasatinib is also active in patients with CML in chronic phase who were resistant/refractory to both IM and nilotinib.211 Common non-haematological side-effects included diarrhoea, nausea and fluid retention including development of pleural and very rarely pericardial effusions; a minority of patients developed severe neutropenia and thrombocytopenia requiring dose reductions. Dasatinib targets an earlier progenitor population than IM, but does not eliminate the quiescent stem cell population.212 Nilotinib (previously known as AMN 107) is an aminopyrimidine that is a structural derivative of IM. Compared with IM, it has a 25-fold increased potency to bind the Abl kinase domain in the inactive conformation. Pre-clinical studies confirmed its efficacy against all known IM-resistant mutants, other than the T315I mutant.213 Current phase I studies in patients with IM-resistant accelerated and blastic phases of CML suggest a CHR of around 50 per cent with CCyR of 14 per cent.81 Phase II studies suggest major CyR in 19 per cent of IM-resistant/refractory patients.214 Sideeffects were similar to IM and dasatinib. Both drugs have now received regulatory approval in several countries for use in patients who are in failure. Table 41.2 depicts the various characteristics of the two drugs. INTERFERON ALPHA
Until the early 1980s, the standard treatment for patients with CML in chronic phase was hydroxyurea, which largely replaced busulphan as standard therapy.215 By the mid 1990s, it was clear that IFN-α was better than either of these cytotoxic drugs; patients who achieved complete cytogenetic responses obtained maximal benefit.216,217 A metaanalysis confirmed the superiority of IFN-α over both busulphan and hydroxyurea.218 The analysis found a 5-year survival rate of 57 per cent for IFN-α treated patients compared with 42 per cent for the chemotherapy treated cohort; the absolute difference in 5-year survival for IFN-α against hydroxyurea and busulphan was 12 per cent and 20 per cent, respectively. The maximal cytogenetic response usually occurred by 12–18 months following IFN-α therapy.219 Side-effects, in particular fatigue, flu-like symptoms and weight-loss, were frequently observed with IFN-α.
Table 41.2 Dasatinib and nilotinib in chronic myeloid leukaemia (CML) Parameter Potency vs. IM Target BCR-ABL binding Resistant mutation MTD (mg bid) DLT c-Kit (vs. IM) PDGFRβ Clinical activity
Dasatinib
Nilotinib
325 Dual SRC & ABL Inactive active T315I 70–100 BM ↓ effusions 8 50 Highly activea
20 BCR-ABL Inactive T315I; Y253H 600 BM ↓; ↑ ind. bili Highly activea
DLT, direct lymphocytotoxicity test; IM, imatinibmesylate; MTD, maximum tolerated dose; PDGFRβ, platelet-derived growth factor receptor, beta subunit. a Both drugs are highly active; not necessarily exclusive.
IMMUNOTHERAPY
Following the realization that a molecular remission and ‘cure’ might not be possible with IM alone, many efforts were directed to exploring the potential of developing an active specific immunotherapy strategy for patients with CML by inducing an immune response to a tumourspecific antigen.220,221 The principle involves generating an immune response to the unique amino-acid sequence of p210 at the fusion point. Clinical responses to the Bcr-Abl peptide vaccination, including CCyRs, have been reported in a small series.222 In contrast to previous earlier unsuccessful attempts, the current series included administration of GM-CSF as an immune adjuvant and patients were only enrolled if they had measurable residual disease and HLA alleles to which the selected fusion peptides were predicted to bind avidly. If these results can be confirmed, vaccine development against Bcr-Abl and other CML-specific antigens could become an attractive treatment for patients who have achieved a minimal residual disease status with IM. Other targets for vaccine therapy now being studied include peptides derived from the Wilms’ tumour-1 protein, PRAME, proteinase-3 (Pr1) and elastase (ELA1), all of which are overexpressed in CML cells.223,224 Another vaccine strategy that may prove useful for patients who do not achieve a CCyR to IM is use of the K562 CML cell line engineered to produce GM-CSF. OTHER INVESTIGATIONAL APPROACHES
A third generation tyrosone kinase inhibitor bosutinib (previously known as SK1-606), which like dasatinib, targets the SRC and ABL kinases, is now in phase 2 studies. Current results of treating patients with CML in CP and advanced phase appear encouraging and the toxicity profile appears reasonable with the principal dose limiting toxicity being diarrhoea and skin rash.225 Since the drug does not target the PDGFR, fluid retention is uncommon.
Chronic myeloid leukaemia 1003
Numerous other studies have also tested other specific inhibitors of signal transduction pathways downstream of Bcr-Abl alone and in combination with IM. Many of these agents have either not entered or are just entering formal clinical trials. Histone deacetylase inhibitors, such as LAQ824, reduce Bcr-Abl expression and induce apoptosis in CML cell lines derived from patients with advanced phase disease.226 Another agent, 17-allylaminogeldanamycin (17-AAG), which degrades the Bcr-Abl oncoprotein by inhibiting heat shock protein 90 (hsp90), a molecular chaperone required for stabilization of Bcr-Abl, has just entered phase I studies;227 17-AAG appears to have activity in patients with the E255K and T315I mutations. It also down-regulates BCR-ABL mRNA, though the precise mechanism remains unclear. Another novel tyrosine kinase inhibitor, PD166326, also appears to have significant activity in patients with the H396P and M351T mutants.228 This agent also appears to be superior to IM in murine models. Other potential agents include rapamycin (sirolimus), an mTOR inhibitor, and wortmannin, which is a PI3K inhibitor but not currently available in a formulation suitable for clinical use.229 Rapamycin is synergistic with IM in inhibiting Bcr-Abl-transformed cells, including those that are IM-resistant. Recently another small molecule, INNO-406 (previously known as NS-187; Innovine) has been reported as active in 17 of 18 Bcr-Abl mutants tested.230 This compound is a dual Abl-Lyn kinase inhibitor and about 25–55 times more potent than IM. Phase I results have confirmed activity in patients refractory to IM, dasatinib and nilotinib. Another interesting drug is the Aurora kinase inhibitor MK-0457, which blocks the activity of various IM-resistant mutant forms, including the T315I mutant, which is resistant to both dasatinib and nilotinib.231 Efforts are also being directed to down-regulating survivin, a key member of the inhibitor of apoptosis family of proteins, in both IM-sensitive and IM-resistant CML patients.232
Figure 41.13 Survival 10 years after allo-SCT for CML in 1st CP. (Results from the Hammersmith Hospital at Imperial College, London; October 2006).
the original donor. Such donor lymphocyte infusions (DLI) reflect the capacity of lymphoid cells collected from the original transplant donor to mediate a GvL effect even though they may have failed to eradicate the leukaemia at the time of the original transplant.235 Current results from the German group who analysed the results of patients with IM-resistant CML in both chronic and advanced phases are encouraging.236 Currently, most clinicians recommend initial treatment with IM for all newly diagnosed patients who would have been eligible for early transplant in the pre-IM era and to reserve transplant for those who fail IM. However, there are conflicting data on a possible adverse effect of prior IM on the results of delayed transplant, and there is very little information on children. Some clinicians feel that adult patients who are classified as ‘poor-risk’ by the Sokal criteria and ‘good-risk’ by the EBMT risk stratification score and all children should still be considered for an allogeneic SCT as a first line therapy, provided that they have a suitable donor and indeed wish to be transplanted following an informed discussion.237 AUTOLOGOUS STEM CELL TRANSPLANT
ALLOGENEIC STEM CELL TRANSPLANT
For patients with CML in CP treated by allo-SCT with marrow from HLA-identical siblings or a matched unrelated donor, the overall LFS at 5 years is now 80 per cent and 60 per cent, respectively.233 The transplant-related mortality is about 20 per cent and the chance of relapse is about 15 per cent. Most, but not all, patients who are negative for BCR-ABL transcripts at 5 years following the SCT, remain negative for long periods and will probably never relapse (Fig. 41.13).202 About 10–30 per cent of patients submitted to allo-SCT relapse within the first three years post transplant.234 Rare patients in cytogenetic remission relapse directly to advanced phase disease without any identified intervening period of chronic phase disease. There are various options for the management of relapse to chronic phase disease, including use of IM, IFN-α, a second transplant using the same or another donor or infusion of lymphocytes from
Because only a minority of patients are eligible for allogeneic SCT, much interest focused until the advent of IM on the possibility that life might be prolonged and some ‘cures’ effected by autografting CML patients still in chronic phase.238 It is possible that the pool of leukemic stem cells can be substantially reduced by an autograft procedure, and autografting may confer a short-term proliferative advantage on Ph-negative (presumably normal) stem cells. Some studies have been reported in which patients have been autografted with Ph-negative stem cells collected from the peripheral blood in the recovery phase following high-dose combination chemotherapy; some such patients achieved durable Ph-negativity.239,240 Currently Ph-negative CD34 cells have been harvested from a number of patients induced to Ph-negativity with IM, but few patients if any have been autografted with these cells.241 At present autografting should be considered investigational.
1004 Leukaemias
TREATMENT OF PATIENTS WITH ADVANCED PHASES
Accelerated phase disease It is difficult to make general statements about the optimal management of patients in accelerated phase disease, partly because the definition of this phase is not universally agreed. Patients who have not previously been treated with IM may obtain benefit from the use of this agent. For patients who progressed to accelerated phase while on IM, it is best to discontinue IM and consider alternative strategies. Patients whose disease seems to be moving towards overt blastic transformation may benefit from cytotoxic drug combinations appropriate for de novo AML or ALL. Allogeneic SCT should certainly be considered for younger patients if suitable donors can be identified. Reduced intensity conditioning allografts are probably not indicated since the efficacy of the GvL effect in advanced phase CML is not established. Clinical trials exploring the use of either dasatinib or nilotinib are available for those who wish to enrol into a clinical study and the preliminary results, discussed above, are encouraging.242 Blastic phase disease Patients in blastic transformation may be treated with combinations of cytotoxic drug combinations analogous to those used for de novo AML or ALL, in the hope of prolonging life, but cure can no longer be a realistic objective. Patients in lymphoid transformation tend to fare slightly better in the short term than those in myeloid transformation.243 If intensive therapy is not deemed appropriate, one can offer a relatively innocuous drug such a hydroxyurea at higher than usual dosage; the blast cell numbers will be reduced substantially in most cases but their numbers usually increase again within 3 to 6 weeks. Combination chemotherapy may restore 20 per cent of patients to a situation resembling chronic phase disease and this benefit may last for 3 to 6 months. A very small minority, probably less than 10 per cent, may achieve substantial degrees of Ph-negative haemopoiesis. This is most likely in patients who entered blastic transformation very soon after diagnosis. IM can be remarkably effective in controlling the clinical and haematological features of CML in advanced phases in the very short term.244 In some patients in established myeloid blastic transformation who received 600 mg daily, massive splenomegaly was entirely reversed and blast cells were eliminated from the blood and marrow but such responses are almost always short-lived. Therefore IM should be incorporated into a programme of therapy that involves also use of conventional cytotoxic drugs and possibly also allo-SCT. As in the case of accelerated phase disease, it is useful to consider patients who enter blastic phase whilst on IM for clinical trails using either dasatinib or nilotinib.245,246 Allo-SCT using HLA-matched sibling donors can be performed in accelerated phase; the probability of leukaemiafree survival at 5 years is 30–50 per cent. Stem cell transplantation performed in overt blastic transformation is
nearly always unsuccessful. The mortality resulting from GvHD is extremely high, and the probability of relapse in those who survive the transplant procedure is very considerable. The probability of survival at 5 years is consequently 0–10 per cent. TREATMENT FOR RELAPSE OF CHRONIC MYELOID LEUKAEMIA POST ALLOGENEIC STEM CELL TRANSPLANT
In the 10–20 per cent of patients who relapse post allo-SCT for CML, this occurs in the first 3 years. This relapse tends to follow an orderly progression with the patient initially demonstrating evidence of a molecular relapse with increasing positivity of BCR-ABL transcripts by PCR, followed by a cytogenetic relapse when the Ph chromosome is found and then haematological and clinical relapse. Molecular monitoring of all SCT recipients is therefore valuable.202 For patients with molecular relapse, remission can be re-induced simply by withdrawal of immunosuppression or by the transfusion of donor lymphocytes, providing additional evidence of the potent role of GvL in CML.235 Donor lymphocytic infusions (DLI) can induce remissions in 60–80 per cent of patients with molecular or cytogenetic relapse. The potential benefit of adding IFN-α to DLI is currently being assessed. Patients who fail to enter remission with DLI may be candidates for a second alloSCT, but the risk of transplant-related mortality (TRM) is relatively high.
Conclusions The substantial understanding of the molecular features and pathogenesis of CML has provided important insights into targeting the treatment to specific molecular defects. The successful introduction of IM as targeted therapy for CML has made the approach to management of the newly diagnosed patient fairly complex. IM has unequivocally established the principle that molecularly targeted treatment can work. The second generation of tyrosine kinase inhibitors, dasatinib and nilotinib, have already been shown to have significant activity in selected patients, in both CP and the more advanced phases of the disease, who are resistant to IM, and may in some cases replace it. The notion that the GvL effect is the principal reason for success in patients with CML subjected to an allograft has renewed interest in immunotherapy and the use of kinase inhibitors in conjunction with various immunotherapeutic strategies is now being studied. For the moment, the majority of patients who present with CML in CP should receive an initial trial of IM. All patients must be monitored by bone marrow cytogenetics at diagnosis and at three month intervals until achieving a CCyR. Thereafter marrow cytogenetics need only be performed annually, but is not mandatory. RQ-PCR for BCRABL transcripts should be measured at three months
Chronic lymphocytic leukaemia 1005
intervals after starting IM; in patients with increasing BCRABL transcript numbers, the frequency of monitoring should be increased. For patients who fail IM, the best available definitions of IM-failure are those recommended by the European LeukemiaNet panel,187 and are ineligible for a SCT should be offered dasatinib or nilotinib. For those patients with a matched donor, no firm recommendations can be made at present, though, in practice, it might be reasonable to offer them a second generation TKI and proceed with an SCT in the absence of a durable response after a finite period of, say 6–9 months. Based on current experience, both dasatinib and nilotinib seem to be equally active in patients deemed to have failed IM. Neither agent is recommended for the patient with a T315I mutant clone predominating. Recently the National Cancer Institute and the National Cancer Center Network issued papers that reflect some of these guidelines.247,248,249
CHRONIC LYMPHOCYTIC LEUKAEMIA Chronic lymphocytic leukaemia (CLL) is a heterogeneous disease comprising of a clonal expansion of mature B lymphocytes that typically express high levels of the antiapoptotic proteins Bcl-2, Bcl-xd, and low levels of the proapoptotic protein Bax.250 The CLL cells are therefore resistant to apoptosis and most are in the G0/G1 phase. Most specialists consider that CLL results primarily from accumulation of abnormal lymphocytes in the early phases of the cell cycle, rather than from their excessive proliferation. It is the most common type of leukaemia in the Western world accounting for more than 30 per cent of all leukaemias. It affects predominantly older people. It is more prevalent in males, with a male:female ratio of 2:1. The median age of onset is 60 years, and the incidence increases steadily with age, reaching 40 per 100 000 of adults in the eighth decade. In the US, over 10 000 new cases are diagnosed each year. Based on the increasing number of asymptomatic individuals who are noted to have an absolute lymphocytosis, it is likely that the true incidence of CLL is considerably higher.251 Very rarely the disease affects T lymphocytes.
Epidemiology Unlike the other leukaemias, CLL is not known to be induced by exposure to any known chemicals or irradiation. There is, however, a single study, which remains unconfirmed, suggesting an increase in CLL among persons chronically exposed to electromagnetic fields.250 This, of course, if confirmed, could have broad public health implications for technologies such as mobile telephones. There have also been unconfirmed suggestions of a possible causal association with infection by the human T-cell lymphotrophic virus-1 (HTLV-1).252 A second related retrovirus, HTLV-11, has been identified in some cases of the
related diseases: T-cell prolymphocytic leukaemia (PLL) and T-cell hairy cell leukaemia (HCL).253,254 Family studies, involving small number of patients, suggest that CLLphenotype cells can be detected in up to 13.5 per cent of families with a positive history.255
Clinical About half of all patients with CLL are diagnosed by a routine blood test for unrelated reasons, which shows an absolute lymphocytosis (Fig. 41.14). The remainder present with fatigue, fever, weight loss, lymphadenopathy, hepatosplenomegaly or infections. The disease follows a variable course, with survival ranging from a few years to a few decades. The characterisation of the CLL cell is based on cell morphology and immunological markers as proposed by the FAB group.8 Immunophenotypic features show a dominance of CD19, CD20 and CD23 (B-cell markers) and CD5 (T-cell marker).The B cells express either κ or λ light chains and surface immunoglobulins (sIg). These features often suffice for a firm diagnosis and help differentiate from related diseases, such as PLL, HCL, mantle-cell lymphoma and the leukaemic phase of other lymphomas. Initial guidelines for the diagnosis were proposed in 1989 by a joint National Cancer Institute (US) and the International Workshop on Chronic Lymphoid Leukaemia group.256 These guidelines were revised in 1996 and minimal diagnostic criteria were suggested, namely a blood lymphocytosis (absolute lymphocyte count of at least 5.0 109/l) with mature-appearing lymphocytes and a characteristic immunophenotype of monoclonal B cells, as described above.257 Chronic lymphocytic leukaemia cells are typically negative for surface CD22 and FMC7; the cells are in the G0 phase of the cell cycle and unresponsive to conventional mitogenic stimuli.
Molecular pathogenesis Though the molecular pathogenesis of CLL is still relatively obscure, much has been learned over the past decade.258 Chronic lymphocytic leukaemia cells express very low levels of the B-cell receptor (B-CR), a multimeric complex formed by sIg and CD79a/CD79b, which plays an important role in orchestrating the signal transduction pathways involved in eliciting cellular responses. This probably results in most CLL cells being anergic. Chronic lymphocytic leukaemia cells have a mantle zone-like phenotype (CD5 and IgM/IgD positivity) of naive B cells, indicating that the cell of origin has passed through the germinal centre, but many harbour mutated immunoglobulin (Ig) genes suggesting that the cells mature in a lymphoid follicle.259 The mutational profile of the Ig genes and the expression of CD38 are firmly associated with the prognosis of the disease.260 Patients with unmutated V genes (both VH and VL) displayed higher percentages of
1006 Leukaemias
Figure 41.14 A peripheral blood film from a patient with CLL ( 960). Kindly provided by Professor Daniel Catovsky, Royal Marsden Hospital, London.
CD38 cells and were frequently noted to have an advanced stage and significantly poorer survi-val.261,262 The CD38 expression can, however, vary during the course of the disease, contributing to the heterogeneity of the disease.263 Another important observation has been the high levels of expression of zeta-associated protein 70 (ZAP-70), a receptor-associated protein tyrosine kinase expressed in cells from most patients with unmutated Ig V genes.264 ZAP-70 is usually found in T and NK cells, but not normally in B cells. Cytogenetic abnormalities, typically deletions, trisomies and mutations, occur in about 80 per cent of all patients with CLL. This is in contrast to most other B-cell lymphoproliferative diseases, where chromosomal translocations are found. Moreover, the mutational profiles of Ig genes appear to be associated with specific genetic abnormalities, suggesting different underlying molecular pathogenesis.265 The most frequent abnormality is the 13q deletion found in 55 per cent of patients, followed by trisomy 12 and the 11q deletion; the 17p deletion involving the p53 protein is seen less frequently. It is of interest that the 13q deletion is associated with good prognosis in contrast to deletions in 11q and 17p, which are found in poor prognostic disease being associated with unmutated Ig genes. There is contradictory evidence, at present, whether trisomy 12 is associated with poor prognosis. There is some controversy with regard to the gene expression profiling studies of B-CLL cells, which suggest a common gene expression signature that appears to be independent of Ig mutations, suggesting, perhaps, a common molecular mechanism for the origin of all CLL cells.266 Interestingly, the monotonous morphology and the phenotypic under-expression of the B-CR in the CLL cells would support this notion.
STAGING SYSTEMS AND PROGNOSTIC FACTORS
The historical staging systems, Rai and Binet, introduced in 1975, and based on Dameshek’s proposed model of orderly
disease progression in CLL has facilitated the division of patients with CLL into three prognostic groups, which enabled clinicians to offer appropriate treatments to those with advanced/poor prognostic disease (Table 41.3).267–269 Typically, untreated patients with good prognosis, who comprise about 60 per cent of all patients, have a 10-year overall survival of over 50 per cent, which approached the survival of people matched for sex and age who do not have CLL; those with intermediate prognosis have an overall survival of about 2–4 years; and those with poor prognosis disease about 1.0–1.5 years. For most of the patients in the good prognostic group, neither system is able to predict the patients who are more likely to progress and therefore could be offered treatment earlier. Neither system, of course, takes account of the recently identified markers, such as the V gene mutation status, expression of CD38, expression of ZAP-70, cytogenetic abnormalities and other, currently unconfirmed markers such as the beta-2-microglobulin, lymphocyte doubling time, the ratio of lipoprotein lipase and ADAM29 mRNA quantifications and levels of thymidine kinase.270 The degree of heterogeneity in the expression of the immunophenotypic markers within the specific disease category can also be used for risk stratification.271 Investigators in London examined these aspects and proposed a scoring system for the diagnosis of CLL based on reactivity with CD5, CD23 and FMC7 and the intensity of expression of membrane immunoglobulins and CD22. Each marker is assigned a value of 0 or 1 according to whether it is typical or not for CLL. Scores range from 5 (typical CLL) to 0 (atypical CLL). This method has been validated and is useful in the classification of CLL. Clearly a practical new staging system is now due.
Treatment GENERAL PRINCIPLES
Since the overall survival of many patients with goodprognosis CLL approaches that of the normal population, they may not require any specific therapy since this would not prolong survival.272 For the small minority with good prognosis who are destined to progress, we need to have a better validated staging system that will help to identify those who should receive treatment earlier. Most specialists offer treatment when patients present with symptoms related to disease and advanced disease (intermediate- and poorprognosis Rai and Binet groups). It is now almost a decade since the National Cancer Institute Working Group (NCIWG) published revised criteria for treatment initiation, particularly for those with good-prognosis disease. These guidelines include patients with ‘B’ symptoms, worsening anaemia and/or thrombocytopenia, autoimmune cytopenias, progressive splenomegaly, progressive lymphadenopathy and lymphocyte doubling time of 6 months or less.273
Chronic lymphocytic leukaemia 1007
Table 41.3 Clinical staging systems for chronic lymphocytic leukaemia Rai staging system
Stage 0 1 2 3 4
Features
Level of risk
Median survival (years)
Lymphocytosis Lymphocytosis and lymphadenopathy Lymphocytosis, lymphadenopathy and spleen and/or liver enlargement Lymphocytosis and anemia (Hb 11 g/dL) Lymphocytosis and thrombocytopenia (platelets 100 109/L)
Low Intermediate
10 7
High
1.5
Binet staging system Stage A B C
Lymphoid involvement 0, 1 or 2 areas 3 or more areas
Hb (g/dL)
Platelets (ⴛ109/L)
Survival (years)
10 10 10 and/or
100 100 100
12 5 2
The NCIWG has proposed response criteria which also require updating. Complete remission by NCI criteria stipulates less than 30 per cent marrow lymphocytes by morphological examination, a situation that often reveals significant residual disease when assessed by other technologies, such as sensitive flow cytometry or molecular studies. Currently there is no firm consensus with regards to the optimal techniques for monitoring responding patients. Most specialists define minimal residual disease (MRD) as the presence of small numbers of cells with the characteristics of CLL identified by flow cytometry or amplification of the Ig gene rearrangement characteristic of CLL by PCR. Complete remissions that are MRDnegative are associated with prolonged time to progression and probable improved survival.274 For patients being selected for potentially aggressive treatments, the objective should be to achieve a MRD-negative CR. The NCIWG has published a document on the use of the new diagnostic tools in the treatment decisions of patients with CLL.275 The consensus was that most, if not all, of the new prognostic markers have yet to be studied in large prospective studies and their routine use cannot be recommended at present, except, of course, in the context of a clinical trial. First-line treatment The choice of first-line treatment depends very much on the presence or absence of adverse prognostic factors in the individual patients, though at present there is paucity of
firm clinical data supporting this approach and consideration for an appropriate clinical trial might represent the best treatment. Outside of clinical trials, chlorambucil, either alone or with a glucocorticoid steroid, usually prednisolone, remains the most frequently used first-line drug globally, as it has for over four decades. Cyclophosphamide orally is an equivalent drug. However, increasing numbers of haematologists regard fludarabine as the optimal choice if treatment is started with a single agent. In the US, fludarabine-based therapy is now considered the first-line treatment of choice. For patients presenting with more advanced disease, and for younger patients, combination chemotherapy is often resorted to. The most common regimens are cyclophosphamide, vincristine and prednisolone; sometimes an anthracycline is also added. Although these latter regimens have resulted in higher responses, there are no major durable responses or survival advantages compared with chlorambucil. A meta-analysis on data collated by the CLL Trialists’ Collaborative Group validated these observations and confirmed no survival advantages for immediate versus deferred chemotherapy in patients with early-stage CLL.275 It is, however, important to note that this analysis is applicable to treatment with alkylating agents only and may not be useful for drugs such as fludarabine. Fludarabine (9-β-D-arabinofuranosyl 2-fluroadenine monophosphate) is a purine analogue, which has been found to be a remarkably effective treatment for CLL patients
1008 Leukaemias
resistant to chlorambucil.276 The drug is myelosuppressive and may be associated with an increased risk of viral infections. A US randomized trial comparing fludarabine with chlorambucil in newly diagnosed patients with CLL confirmed a higher incidence of CR (27 per cent vs. 3 per cent), overall responses (70 per cent vs. 43 per cent) and duration of these responses (2.75 years vs. 1.4 years) in the fludarabinetreated cohorts, but no evidence of prolongation of survival.277 The long-term results of this and another similar study confirmed these early results, which were quite durable.274,278 A smaller French randomized study of fludarabine versus CAP (cyclophosphamide, adriamycin [doxorubicin HCl] and prednisolone) revealed a similar finding.279 Recently the preliminary results from the UK LRF CLL4 trial were presented, in which about 66 per cent of all patients received an oral preparation of fludarabine, unlike the US cohorts who all received intravenous fludarabine; these results showed similar efficacy and toxicity profile.280 Adverse effects associated with fludarabine and other purine analogues in clinical trials at present, such as pentostatin (2-deoxycoformycin, 2-DCF), cladrabine (2-chlorodeoxyadenosine, 2-CdA) and clofarabine, are quite frequent and occasionally significant.281,282 Typically these include myelosuppression, particularly neutropenia and lymphopenia, the latter increasing the risk of infections with viruses, fungi and protozoa such as Pneumocystis carinii (now known as pneumocystis jerovecii). Occasionally autoimmune phenomena, such as haemolytic anaemia, are noted. Early studies suggested a frequent association with autoimmune haemolytic anaemia but current studies suggest it to be uncommon and the prevalence is considerably lower than that with chlorambucil.283 Currently the standard-of-care for previously untreated patients with CLL favours fludarabine either as a single agent or in combination with rituximab with or without the addition of cyclophosphamide (FCR). The incidence of CR is about 70 per cent, many of which are MRD-negative and the LFS and overall survival appears to be longer with the addition of rituximab and/or cyclophosphamide.284,285 The combination is also remarkably effective in patients with relapsed disease. Efforts are now being directed to intensify therapies in an attempt to improve the MRDnegative CR rates by adding alemtuzumab (Campath-1H), which has demonstrated significant activity as second-line therapy, to first-line therapy in combination with fludarabine alone or in combination with both fludarabine and rituximab. This is discussed below. Second-line treatments Alemtuzumab, a fully humanized rat antibody directed against CD52 and the heavy and light chain constant regions of human IgG1 and kappa, respectively, was recently approved in the US for the treatment of relapsed CLL.286 Monotherapy with alemtuzumab produced responses ranging from 33 to 53 per cent, with a median duration of response ranging from 8.7 to 15.4 months, in chemotherapy-refractory patients with CLL.287 Responses were seen in patients with 11q and 17p deletions. Based on
the observation of significant complete responses, with 38 per cent molecular complete responses, studies are now assessing the role of alemtuzumab as first-line therapy compared with chlorambucil.288 Preliminary analysis of a phase III study assessing alemtuzumab versus chlorambucil in therapy-naive patients with progressive B-cell CLL demonstrated overall response rates of 82.6 per cent for alemtuzumab, compared with 54.7 per cent for chlorambucil; the corres-ponding CRs were 22.1 per cent and 2.0 per cent.289 Safety analysis confirmed 34.7 per cent of the alemtuzumab-treated cohort and 19.7 per cent of the chlorambucil-treated cohort to have had side-effects attributed to therapy. Infections, particularly cytomegalovirus, were seen more commonly in the alemtuzumab-treated cohort; there was treatment-related death in the chlorambucil arm. Based on in-vitro synergism, a combination of fludarabine and alemtuzumab also appears attractive. A phase II study in relapsed patients confirmed overall responses of 83 per cent, including complete responses in 30 per cent of patients.290 A phase III study of the combination versus fludarabine alone in newly diagnosed patients is currently being undertaken. The combination of alemtuzumab and rituximab has also been investigated in relapsed/refractory patients with CLL, resulting in an overall response rate of 52 per cent, including complete responses in 8 per cent.291 Conventionally, alemtuzumab has been administered intravenously, but recent studies confirm its efficacy when administered subcutaneously, though local reactions, such as erythema and pain at site of injection, are more prominent.292 Other adverse effects include fevers, chills and profound immunosuppression. Several trials are also assessing the potential role of thalidomide and its analogue, lenalidomide, immunomodulatory and anti-angiogenesis agents, based on the recognition of the bone marrow micro-environment and angiogenesis in the pathogenesis of CLL. Both drugs are currently being used in patients with transfusion-dependent anaemia due to low- or intermediate-risk MDS associated with a chromosome 5q deletion with or without additional cytogenetic abnormalities and plasma cell myeloma.293 Observations confirming thalidomide’s clinical activity in patients with CLL led to the recent phase II study investigating the role of lenalidomide in patients with relapsed/refractory CLL.294,295 Preliminary results from 19 of 29 patients who had had at least 2 months of therapy suggest 68 per cent objective responses with 15 per cent CR. The side effects associated with lenalidomide were largely haematological, neutropenia and thrombocytopenia; 8 per cent of the patients experienced thromboembolic phenomena. Further accrual continues. Investigational treatments A novel agent, denileukin deftitox, a fusion of an interleukin-2 and diphtheria toxin fusion protein targeting CD25, which was recently approved in the US for patients with cutaneous T-cell lymphoma, has been investigated in patients with fludarabine-refractory CLL.296 It has resulted in over 90 per
Other lymphoid leukaemias 1009
cent responses and further studies are currently in progress. Targeting BCL-2 with an antisense oligonucleotide, such as oblimersen sodium or SPC2996, is another potentially useful treatment for patients with CLL. Recently a small molecule, obatoclax (GX15-070), which is a pan Bcl-2 family inhibitor has resulting in encouraging phase I results and is now being pursued further.297 Other investigational approaches include other immunotherapies, such as lumiliximab (anti-CD23), epratuzumab (anti-CD22), apolizumab (anti-HLA-DR) and an anti-CD80 (IDEC-114) antibody.298 Flavopiridol has also demonstrated activity in fludarabine-refractory patients with CLL and further accrual continues.299 A number of vaccine strategies, such as vaccination with modified autologous leukaemic cells and the use of antigenpulsed dendritic cells, are also being investigated.300 Gene transfer of CD40-ligand induces autologous immune recognition of CLL cells and is being developed further for the treatment of CLL.301 STEM CELL TRANSPLANTATION
Although the median age of onset for CLL is 65 years, about 40 per cent of patients are under the age of 60 years and about 10 per cent are younger than 50 years of age at diagnosis. Many of these younger patients have high-risk disease and become chemotherapy refractory. Such patients have been offered SCT over the past decade, but, at present, there are no studies comparing conventional chemotherapy with SCT other than a small retrospective matched-pair analysis that suggested a survival advantage for autologous SCT compared with chemotherapy.302 A small UK study in which 65 patients received an autologous SCT showed a 5 year LFS of 51.5 per cent and an overall survival of 77.5 per cent, but 8 per cent of these patients developed MDS/AML.303 Long-term follow-up using PCR amplification and sequencing of the rearranged IgH gene from the original malignant clone show that though some patients have no residual disease detectable, a significant majority relapse. Most studies suggest a 6-year LFS of about 30 per cent with no evidence of a plateau.304 In contrast, there is some evidence that some patients may have prolonged LFS and possibly overall survival with alloSCT, but the mortality and morbidity associated are quite high. The EBMT registry data suggest a TRM of about 46 per cent, much of which is from GvHD and infections.305 Some small single centre studies suggest a 5-year LFS of 78 per cent for previously untreated patients and 31 per cent for chemotherapy-refractory patients, and a TRM of 11 per cent.306 Currently there are no formal direct comparative data of allo-SCT versus autologous SCT.307 Single-centre nonrandomized comparisons suggest an improved outcome with some durable remissions following an allo-SCT compared with autologous SCT, but there was no difference in the overall survival.304,308 The primary reason to consider the use of allo-SCT is the potential for a GvL effect, which appears to be important for the success of the SCT in patients with CLL.309 In an attempt to reduce the excessive TRM associated with allo-SCT, the use of reduced intensity conditioning regimens
has been explored. Results collated by the EBMT registry from 73 patients with CLL subjected to a reduced intensity SCT were compared with those from 82 matched patients who had been subjected to conventional allo-SCT.310 There were no significant differences in LFS or overall survival but there was a significant reduction in the TRM for the reduced intensity cohort. These latter patients had a relatively high rate of relapse rate but a few were able to achieve CR following DLI. These preliminary experiences are encouraging and further studies are in progress. The concept of fludarabinebased non-myeloablative conditioning regimens being able to enhance the GvL effect and decrease toxicity attributed to conditioning has also been demonstrated, though the TRM remains high.311 This latter experience suggests that standard dose fludarabine-based chemotherapy is sufficiently immunosuppressive to allow engraftment and subsequent DLI should enhance the GvL effect further. Results of another similar phase II study using alemtuzumab-conditioning in the reduced intensity allo-SCT of fludarabine-refractory CLL patients revealed a 48 per cent overall 3-year survival for related and 68 per cent for unrelated recipients; the corresponding figures for 3-year leukaemia-free survival was 37 per cent and 51 per cent.312 The 3-year non-relapse mortality was similar for both groups. The trend towards a better leukaemia-free survival for the unrelated recipients is noteworthy.
Conclusions Much has been learned about the molecular pathogenesis and potential molecular prognostic variables in CLL. Treatment decisions clearly need to be based on individual stage and disease progression and the current staging systems are outdated. Newer systems incorporating the validated molecular prognostic features are being developed. In the US and other parts of the world, a fludarabine-based treatment, either as monotherapy or in combination with cyclophosphamide and/or rituximab is used as first-line therapy; in general chlorambucil is still the most popular world-wide first-line agent. Alemtuzumab is increasingly used as the treatment of choice for second-line therapy and its potential place as a first-line therapy is being investigated. High-dose therapy and allo-SCT are indeed feasible, and even potentially curative, for younger patients with poor prognostic disease, but the use of this treatment should still be considered investigational at present. Sadly it results in a TRM of 11 to 27 per cent. One needs to balance carefully the excessive TRM with the potential to cure the individual patients.
OTHER LYMPHOID LEUKAEMIAS T-cell chronic lymphocytic leukaemia Epidemiology studies from the Far East suggest that T-cell CLL, which represents less than 3 per cent of all European
1010 Leukaemias
and US CLL patients, is the most common form of CLL in Asia. Morphological and immunophenotyping studies show this disease to be a T-prolymphocytic leukaemia (PLL) subtype. The latest WHO classification of hematopoietic malignancies does indeed describe T-CLL as T-PLL. Patients often present with pancytopenia and prominent splenomegaly and the lack of lymphadenopathy is noteworthy. The general prognosis is similar to poor-risk CLL and most patients tend to be refractory to fludarabine.
Prolymphocytic leukaemia Prolymphocytic leukaemia can be de novo or arise as a consequence of CLL transformation. Prolymphocytic leukaemia cells are larger and less homogeneous than CLL cells, and have a clear and more abundant cytoplasm, clumped nuclear chromatin, and a single prominent nucleolus.313 The immunophenotype of a PLL cell is FMC7, CD22 and surface IgM. Cytogenetic analysis often reveals 14q and t(11; 14) (q13; q32). Most patients have a clinical course similar to poor-risk CLL and do not respond well to fludarabine. A minority of patients have an indolent course.
Large granular lymphocytic leukaemia Large granular lymphocytes (LGL) are a heterogeneous group of large lymphocytes, which express CD8 and CD56 (previously known as natural killer or NK) surface antigens and are associated with antibody-dependent cell-mediated cytotoxicity. Large granular lymphocytic leukaemia also described as Tγ lymphoproliferative disorder has a varied natural history, ranging from an indolent condition to that of ALL. Most patients present with recurrent infections, anaemia, splenomegaly and, usually, no significant lymphadenopathy. The majority of patients with LGL leukaemia do not require any specific treatment at presentation and various strategies have been offered on an ad hoc basis, including prednisolone, cyclophosphamide (low dose), IFN-α, splenectomy and intensive chemotherapy for the more aggressive forms. The prognostic factors associated with poor prognosis are fever, low CD56 expression and low granular lymphocyte counts.
Hairy cell leukaemia Hairy cell leukaemia (HCL) is an uncommon mature B-lymphoproliferative disorder characterized by distinct clinical, morphological and histological features.314 It accounts for about 2 per cent of all leukaemias, with approximately 700 to 800 new patients in the US per annum. The median age of occurrence is 52 years and the disease has a strong predilection for men. It is characterized by splenomegaly and marrow failure in conjunction with infiltration of myeloid tissues by a lymphoid cell with villous
processes (resembling hairs to some degree) that project from the cytoplasm of the cell. Clinically HCL should probably be considered as an indolent non-Hodgkin’s lymphoma, rather than a chronic leukaemia. Immunophenotyping is useful to distinguish HCL from B-cell disorders with hairy or villous lymphocytes, such as the variant form of HCL or splenic lymphoma with villous lymphocytes (SLVL). Typically HCL cells express the following surface markers: CD11c, CD25, HC2 and B-ly-7. The Royal Marsden group in London devised a scoring system based on these four markers that offers a reliable diagnosis of HCL.315 The identification and precise diagnosis of these disorders is important because of the prognostic and therapeutic implications. Whereas patients with typical HCL (Catovsky Score 4) respond to IFN-α and pentostatin (2DCF), these drugs are less useful in the variant form of HCL and splenectomy is the treatment of choice for SLVL. Currently, if treatment is indicated, most specialists consider cladribine (2-CdA) or pentostatin as first-line treatment. Cladribine is probably slightly preferred, simply because of a greater experience and the association of complete remissions of over 90 per cent, most of which continue beyond 4 years.316 Since very few patients relapse, there is no firm consensus as to how they should be managed. Novel therapies include combinations of various chemotherapeutic agents with rituximab. Rituximab is also being investigated in combination with cladribine as first-line therapy and preliminary observations suggest eradication of minimal residual disease in most patients.317
Adult T-cell leukaemia/lymphoma Adult T-cell leukaemia/lymphoma (ATL) is a distinct form of leukaemia/lymphoma first described on the island of Kyushu in south Japan and subsequently found to occur in the Caribbean, US and other countries.318 Epidemiological studies have described the causal association between ATL and HTLV-1.14 The WHO classification defines ATL as a peripheral T-cell neoplasm associated with infection by HTLV-1. It has a median age of occurrence of 40 years. These cells typically have the phenotype of mature helper T cells and express CD2, CD3 and CD4 surface antigens. They also express CD25 and clonal rearrangement of the T-cell receptor β-chain is often present. Cytogenetic abnormalities are common and include trisomy 3q or 6q, 14q and inv.14 Adult T-cell leukaemia/lymphoma is characterized by lymphadenopathy and hepatosplenomegaly. There is a high incidence of cutaneous involvement, which may take a variety of forms. Lytic bone lesions and hypercalcaemia occur commonly. Standard therapy is similar to that employed in the management of poor prognosis high-grade non-Hodgkin’s lymphoma, but the outcome is generally unsatisfactory. Allogeneic SCT could be contemplated if a suitable donor is available.
Myelodysplastic syndromes 1011
MYELODYSPLASTIC SYNDROMES The myelodysplastic syndromes (MDS) are clonal disorders characterized clinically and morphologically by ineffective haemopoiesis leading to bone marrow failure and a high probability of malignant transformation at the level of a myeloid stem cell resulting in AML.319 The cytogenetic events found in association with MDS are heterogeneous but certain specific abnormalities tend to prevail, in particular loss or gain of all or parts of chromosomes 5, 7, 8 and 20.320 Myelodysplastic syndromes generally affect about 1 in 500 people over the age of 60 years; although an increasing number of younger adults are being reported, mostly with therapy-associated MDS (t-MDS), particularly after having been subjected to high-dose chemotherapy, alkylating agents and radiotherapy. Childhood MDS is rare and found in constitutional conditions such as Down syndrome. Patients with the isolated (del)5q cytogenetic abnormality tend to have a rather benign clinical course and only about 25 per cent will transform into AML. In contrast to other MDS subtypes, mostly women are affected.
Clinical Although the common presenting features are those due to haemopoietic failure, many patients are diagnosed as a consequence of pancytopenia on a routine examination. The diagnosis of MDS is often suspected on morphological grounds in a patient with cytopenias. The bone marrow cellularity is either normal or increased. Characteristic morphological changes include mature hypogranular and hypolobulated (pseudo-Pelger–Huët) granulocytes, micromegakaryocytes and a variety of red-cell precursor abnormalities, ranging from ringed sideroblasts to megaloblastic changes. Dysplastic abnormalities in all cell lineages are prominent.321 Cytogenetic analysis of the bone marrow is useful for confirming the diagnosis. Cytokinetic studies suggest a substantially higher rate of cell division and apoptosis is prominently increased.
Molecular pathogenesis The precise molecular changes leading to AML have not been fully elucidated but sequential mutations are probably required. A number of potential candidate genes have now been identified and include the N-RAS, the p53, the IRF-1, the BCL-2, the p15ink and the MLL genes.320 A recent report suggests the notion of the ribosomal protein encoding RP514 gene causing the 5q-syndrome.322 A plausible model of the molecular pathogenesis of MDS has been proposed whereby a normal haemopoietic stem cell acquires successive genetic abnormalities that, in a minority, leads to a malignant transformation and clonal expansion. Early mutations in stem cells result in
differentiation arrest resulting in dysplasia, and additional genetic abnormalities thereafter affect myeloid cell proliferation, which ultimately leads to frank AML, with greater than 20 per cent myeloblasts in the bone marrow. At present, the precise primary event that leads to the early mutations and dysplasia is not defined. Myeloproliferative disorders (MPD) are haemopoietic stem cell disorders characterized by uncontrolled cellular growth resulting in clonal expansion of the affected lineages. Chronic myeloid leukaemia is, of course, well characterized by the activated Bcr-Abl kinase (discussed above). Recently a specific activating mutation in the tyrosine kinase gene JAK2 has been reported in three distinct forms of MPD: polycythaemia vera, essential thrombocythemia and primary myelofibrosis.323,324 Following these and other observations, such as the often shared chromosomal regions of deletions between MRD and MDS, many efforts have been directed to the study of tyrosine kinase oncogenes in MDS but currently no specific mutations noted.
Classifications FAB CLASSIFICATION
The FAB classification proposed in 1982, almost 30 years following the landmark description of MDS (when the term ‘pre-leukaemia’ was designated) by Matthew Block and colleagues, is intended for diagnostic and prognostic purposes (Box 41.5).325,326 The classification has several important limitations. Firstly, the term ‘refractory anaemia’ is imprecise and cannot be identified morphologically. Secondly, chronic myelomonocytic leukaemia (CMML) is perhaps more closely related to the myeloproliferative disorders than to the other subtypes of MDS. Thirdly, the use of an arbitrary criterion, for example the percentage of marrow blasts, to distinguish between an advanced MDS and a frank AML is often misleading. WHO CLASSIFICATION
The WHO proposed a classification in 2001 (Box 41.5), which addresses some, but not all, of the limitations of the FAB classification. Further revisions based on cytogenetic findings with important prognostic and therapeutic implications, such as normal karyotypes versus balanced chromosomal aberrations versus complex karyotypes, can be anticipated. Less emphasis is placed on the dogma of the precise number of myeloblasts present in the marrow, and the threshold of 20 per cent blasts in the blood or marrow is not mandatory for treatment decisions; the treatment decision should be made on a constellation of findings based upon age, prior history of MDS and clinical findings. The FAB category of CMML was placed in a new category: ‘MDS-MPD’ since it shares features of both disease categories. Myelodysplastic syndrome with isolated (del)5q was recognized as a separate category with a good prognosis.
1012 Leukaemias
Box 41.5 The French–American–British (FAB) and World Health Organization (WHO) classifications of myelodysplastic syndromes (MDS)
diagnosed and being seen at a specialist centre and the considerable heterogeneity between the ‘low-risk’ and ‘intermediate subgroup 1 risk’.
Treatment FAB classification of MDS ● Refractory anaemia (RA) ● Refractory anaemia with ringed sideroblasts (RARS) ● Refractory anaemia with excess blasts (RAEB) ● Refractory anaemia with excess blasts in transformation (RAEB-T) ● Chronic myelomonocytic leukaemia (CMML) WHO classification of MDS Refractory anaemia without ringed sideroblasts (equivalent to FAB: RA) ● Refractory anaemia with ringed sideroblasts (equivalent to FAB: RARS) ● Refractory cytopenia with multi-lineage dysplasia (RCMD) (new) ● Refractory cytopenia with multi-lineage dysplasia and ringed sideroblasts (RCMD-RS) (new) ● Refractory anaemia with excess blasts 1 (equivalent to FAB: RAEB) ● Refractory anaemia with excess blasts 2 (equivalent to FAB: RAEB) ● MDS associated with isolated del(5q) (new) ● MDS-U unclassifiable (new) ●
The major changes from the FAB classification of MDS are: ●
●
●
The RAEB-T of the FAB is now considered to be acute leukaemia, not an MDS type CMML of the FAB is now in a new category MPD/MDS devised by the WHO New groups have been created by the WHO to allow for a better stratification of these patients, i.e., RCMD, MDS with isolated del(5q) and MDS-U.
Risk assessment The FAB and the newer WHO classifications, in their current formats, are not helpful from the prognostic aspects. The International Myelodysplastic Syndrome Risk Analysis Workshop has proposed a prognostic scoring system, termed the international prognostic scoring system (IPSS) which has substantially improved the classifications.327 Patients are assigned a score based upon the cytogenetic abnormalities, the percentage of blasts in the marrow and the number of lineages affected in the cytopenia, and stratified into four prognostic groups: low risk (median survival 5.7 years), intermediate risk subgroup 1 (median survival 3.5 years), intermediate risk subgroup 2 (median survival 1.2 years) and high risk (median survival 0.4 year). The principal limitations of the IPSS are that it does not account for the lag time between the patient having been
GENERAL PRINCIPLES
The treatments for patients with MDS can conveniently be considered as those designed to reduce morbidity from ineffective haemopoiesis and improve symptoms and quality of life, and those which improve the overall survival and decrease the risk of progressing to frank AML. From a practical viewpoint, most specialists base treatment decisions on the age of the patient, the performance status (PS) and the IPSS-risk category.328 In general, patients who are younger than 60 years of age and selected older patients (who do not have major co-morbid conditions), have a good PS and an intermediate risk subgroup 2 or higher IPSS risk score, are considered for AML-type therapy, including allogeneic SCT. Patients who are in the low and intermediate subgroup 1 IPSS risk category, irrespective of age, tend to be offered supportive treatments, such as haemopoietic growth factors, in particular G-CSF, exogenous erythropoietin (EPO) and blood transfusions. Patients who require frequent red cell transfusions have significant morbidity and some mortality from iron overload. The role of iron chelators, in particular the newer oral preparations such as desferiprone, remains under investigation in adults with iron overload.329 Immunosuppression, either with anti-thymocyte globulins or cyclosporine A, though useful in some young patients who have a short history of being transfusion dependent, has a limited role in the modern management of MDS.330–332 Differentiating agents, such as low-dose cytarabine, retinoids and cholecalciferols have been assessed extensively based on in-vitro observations that the myeloblasts in MDS might be induced to differentiate and that remission might be achieved without a marrow aplasia, and in general not found to be clinically helpful in most patients.333 HAEMOPOIETIC GROWTH FACTORS
Several trials assessing the role of G-CSF and EPO, and indeed GM-CSF, have been completed over the past decade, many antedating the introduction of the IPSS. Exogenous EPO alone appears to improve anaemia in about 25 per cent of low-risk patients who have a low endogenous EPO level.334 The addition of G-CSF appears to exert synergism with EPO with the best responses (around 75 per cent) noted in patients with the RARS subtype of MDS.335 A recent study of 129 patients demonstrated a median duration of response to G-CSF and EPO of 23 months for all ‘low-risk’ patients.336 This study also addressed the impact of this treatment on survival and
Myelodysplastic syndromes 1013
leukaemic transformation and found none. Several ongoing studies are addressing these issues further and are also addressing the role of the longer acting EPO, darbepoietin alfa. A pilot study using darbepoietin alfa alone suggests a response rate of 60 per cent with additional responses when G-CSF is added.337
IMMUNOMODULATORY AGENTS
Based on the notion of an abnormal marrow microenvironment, such as enhanced tumour necrosis factor alpha (TNF-α) levels, and abnormal angiogenesis in MDS, efforts were directed towards assessing the role of immunomodulatory and anti-angiogenesis agents in patients with MDS. In small phase II studies, thalidomide was noted to reduce the transfusion requirements of about 20 per cent of low-risk patients, some of whom were refractory to EPO.338 Thalidomide was associated with significant side effects, such as neuropathy, somnolence and constipation and is teratogenic. Lenalidomide (CC5013; revlimid), a thalidomide analogue, is more potent than thalidomide and has far fewer side-effects. It does not appear to be teratogenic. Current trials have confirmed its activity in a patients with one particular subtype of MDS, namely MDS associated with (del)5q also known as 5q, with about 66 per cent patients achieving an erythropoietic response.293,339 The major side effects were neutropenia and thrombocytopenia. The dose of lenalidomide used is 10 mg/d, compared with the dose of 25 mg/d typically used in patients with plasma cell myeloma. Following the approval of lenalidomide by the FDA in 2006, it is now the preferred treatment for patients with transfusion-dependent anaemia due to low- or intermediate-1-risk MDS associated with a chromosome 5q deletion with or without additional cytogenetic abnormalities and this is reflected in the most recent version of the NCCN Practice Guidelines (version 4.2006).340 Its place in other subtypes of MDS is being investigated at present.
HYPOMETHYLATING AGENTS
Cellular biologists have known for some time that DNA methylation can result in silencing of gene expression. Following the observation suggesting a link between the DNA methylation of the p15ink4b gene promoter and transformation to AML in MDS patients, there has been much interest in epigenetic modulation of genetic function.341 The DNA hypomethylating pyrimidine analogues 5-azacytidine (azacytidine; Vidaza) and 5-aza-2-deoxycytidine (decitabine), also discussed under AML (above), reduce hypermethylation and induce re-expression of relevant tumour suppressor genes in patients with MDS. Azacytidine was studied in a phase III randomized study which showed a 60 per cent overall response and a longer time to progression to AML for the study arm, compared with 5 per cent for the control (supportive care) cohort;342
this led to the drug’s approval by the FDA for patients with high risk MDS. In August 2007 Pharmion, the makers of this drug, issued a press statement that there was a significant survival advantage at 2 years for the cohort receiving azacytidine, but the actual data are unpublished at this time. Decitabine has also now been approved in the US for MDS patients. This was based on a similar study of 170 patients, which showed that AML-free survival, but not overall survival was better in the decitabine-treated cohort.343 Histone acetylation is important in gene transcription regulation and is closely linked to DNA methylation. This has led investigators to explore inhibition of histone deacetylation in patients with MDS and other haematological malignancies. Current candidate agents include several HDACs, such as depsipeptide, SAHA and MS275. A recent small study also assessed valproic acid in combination with ATRA.344
OTHER INVESTIGATIONAL AGENTS
Farnesyl tarnsferase inhibitors, based on the notion that N-RAS mutations are found in about 10 per cent of patients with MDS, are also being studied. The oral FTI, tipifarnib recently resulted in a response in 3 of 27 high-risk patients studied.345 Arsenic trioxide appears to have moderate activity in all types of MDS and further investigations continue.346 Very rarely the reciprocal chromosomal translocation including 5q33 generates a fusion protein involving the PDGFB receptor and the fusion protein has been implicated in variant of the subtype of MDS, CMML with eosinophilia.347 Five patients with CMML and eosinophilia were treated with IM and all achieved a rapid and durable response.
STEM CELL TRANSPLANTATION
Allo-SCT, using an HLA-matched sibling donor offers the only chance of achieving a cure for a small proportion of patients with MDS. There is, however, considerable debate with regard to the optimum timing, best conditioning strategy and, indeed, conventional allo-SCT versus reduced-intensity SCT.348 Both a high rate of relapse post SCT and GvHD remain major problems. In an attempt to reduce relapse post SCT, efforts aimed at reducing the ‘disease-bulk’ pre-SCT have been undertaken. A French study observed that patients with t-MDS who achieved CR with pre-transplant chemotherapy had a significantly better LFS post allo-SCT than patients who did not achieve a CR with chemotherapy.349 More recently, these issues were addressed in a retrospective study in Seattle, which suggested that pre-transplant chemotherapy reduced the risk of post-transplant relapse but failed to show a LFS advantage.350 These observations support the notion that pre-transplant chemotherapy may select for patients who might have fared better without any previous
1014 Leukaemias
treatments prior to being subjected to a SCT. This is currently an unanswered challenging question that is being addressed prospectively by the EBMT group. In general, it appears that patients with less advanced/ early transformation disease fare better with a 3 year LFS of about 65 to 75 per cent compared with those transplanted who have more advanced/frank AML.351,352 Similar outcomes have been reported with reduced-intensity SCT. A recent Seattle study in which targeted oral busulphan was used for conditioning showed a non-transplant mortality (NRM) at 100 days of 12 per cent and a 1 year LFS of 56 per cent.353 The King’s College London group used an alemtuzumab-based conditioning and reported an NRM of 0 per cent for matched and 11 per cent for unrelated donors; the 1 year LFS for the less advanced cohort was 86 per cent, in contrast to 33 per cent for the more advanced patients.354 The Dana Farber group compared the results for conventional allo-SCT and reduced-intensity SCT and observed no important long-term LFS, though there were considerable differences in the causes of failure.355 Randomized trials between using these different SCTs are required to address these issues further. Patients with secondary MDS appear to fare the worst. It is of interest that patients who developed t-MDS after being subjected to an autologous SCT for non-Hodgkin’s lymphoma (NHL), the NRM at 100 days was 49 per cent and the 2 year LFS 28 per cent.356 This is particularly worrisome, since this modality is increasingly being used for NHL with aggressive histologies in first relapse and the prevalence of tMDS appears to be about 10 per cent in this group. Autologous SCT should also be useful since most patients with MDS appear to have the ability to regenerate cytogenetically normal haemopoiesis following myeloablative therapy. Currently 2-year LFS rates of about 30 per cent are being reported, though the relapse rates continue to be high for the moment.357
KEY LEARNING POINTS Biology ●
●
●
●
●
●
Treatment ●
●
●
●
●
●
Conclusions Although much progress has been made in the understanding of the molecular pathogenesis of MDS, much work remains, particularly in t-MDS. The role of epigenetic modulation resulting in the US-approval of two novel agents, azacytidine and decitibine, and the targeting of the marrow micro-environment and angiogenesis with the third US-approved agent, lenalidomide, clearly epitomizes this. Further improvements in the risk assessment and incorporation of cytogenetic abnormalities should improve the classifications. Currently SCT remains the sole modality offering the chance of long-term remission, and probable cure, to a small minority of patients with MDS. Much work with regard to GvHD strategies, optimal timing and type of transplant is still required. The use of the newer agents will also have an impact on SCT in general and strategies to incorporate the newer agents for MRD also need to be addressed.
Cytogenetics is valuable for classifying all forms of leukaemia and myelodysplastic syndrome (MDS). The AML-1, MLLV and PML-RARα genes play crucial roles in pathogenesis of acute myeloid leukaemia (AML). The Bcr-Abl fusion protein plays a central role in the pathogenesis of all Philadelphia-positive leukaemias. Abnormalities in chromosome 13q14 are common in chronic lymphocytic leukaemia (CLL). Some types of therapy-related acute leukaemia involve specifically chromosome 11q23. Several candidate genes are implicated in the transformation of MDS to AML.
●
●
●
Molecular and immunological strategies to target leukaemic cells now in clinical trials. Imatinib mesylate, the original Abl tyrosine kinase inhibitor, established the paradigm for successful targeted therapy for CML and a number of the cancers. All trans-retinoic acid (ATRA) and arsenic trioxide are now part of standard treatment for acute promyelocytic leukaemia (APL). Numerous small and larger molecules are now being used with success in the treatment of leukaemias and MDS. Allogeneic stem cell transplantation (allo-SCT) is established as ‘post-remission’ or ‘consolidation’ therapy for selected patients with acute leukaemia. Non-myeloablative allogeneic SCT is useful in the treatment of poor-risk leukaemias, particularly CLL. Donor lymphocyte infusions (DLI) have established the efficacy of a graft-versusleukaemia (GvL) effect in patients with CML who relapse after an allo-SCT. Remissions in MDS are difficult to achieve; lenalidomide is useful in the 5q syndrome. First-line therapy for progressive CLL is now a fludarabine-based treatment, but long-term remissions remain difficult for patients with poor-risk disease.
REFERENCES 1 Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 1982; 51:189–99.
References 1015
2 Lowenberg B, Downing R, Burnett A. Acute Myeloid Leukemia. N Engl J Med 1999; 341:1051–62. 3 Fourth International Workshop on Chromosomes in Leukemia, 1982 Clinical significance of chromosome abnormalities in acute nonlymphoblastic leukemia. Cancer Genetics and Cytogenetics 1984; 11:332–50. 4 Rowley JD. Chromosome changes in acute leukaemia. Br J Haematol 1980; 44:339–46. 5 Golub TR, Slonim DK, Tamayo P, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999; 286:531–7. 6 Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of the acute leukaemias. Br J Haematol 1976; 33:451–8. 7 Idem. Proposed revised criteria for the classification of acute myeloid leukemia: a report of the French–American–British Co-operative Group. Ann Intern Med 1985; 103:620–5. 8 Bloomfield CD, Brunning RD. The revised French–American– British classification of acute myeloid leukemia: is new better? Ann Intern Med 1985; 103:614–6. 9 Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of chronic (mature) B and T lymphoid leukaemias. J Clin Path 1989; 42:567–84. 10 US Government Statistics, Surveillance, epidemiology, and end-results (SEER) of the National Cancer Institute 2004. www.seer.cancer.gov 11 Austin A, Delzell E, Cole P. Benzene and leukemia: a review of the literature and risk assessment. Am J Epidemiol 1988; 137:419–39. 12 Ichimaru M, Ishimaru T, Belsky JL. Incidence of leukemia in atomic bomb survivors belonging to a fixed cohort in Hiroshima and Nagasaki, 1950–1971: radiation dose, years after exposure, age at exposure, and type of leukemia. J Radiat Res 1978; 19:262–82. 13 Kaldor JM, Day NE, Clarke EA, et al. Leukemia Following Hodgkin’s disease. N Engl J Med 1990; 322:7–13. 14 van Leeuwen FE. Risk of acute myelogenous leukaemia and myelodysplasia following cancer treatment. Bailliéres Clin Haematol 1996; 9:57–8. 15 Pedersen-Bjergaard J. Insights into leukemogenesis from therapy-related leukemia. N Engl J Med 2005; 352:1591–4. 16 Greaves MF. Aetiology of acute leukaemia. Lancet 1997; 349:344–9. 17 Bennett JM, Catovsky D, Daniel MT, et al. Proposal for the recognition of minimally differentiated acute myeloid leukaemia (AML-MO). Br J Haemotol 1991; 78:325–9. 18 Lowenburg B, Downing JR, Burnett A. Acute myeloid leukaemia. N Engl J Med 1999; 341:1051–62. 19 Rowley JD. Recurring chromosome abnormalities in leukemia and lymphoma. Semin Hematol 1990; 27:122–36. 20 Keating MJ, Smith TL, Kanatarjian H, et al. Cytogenetic pattern in acute myelogenous leukemia: a major reproducible determinant of outcome. Leukemia 1998; 2:403–12. 21 Grimwade D, Solomon E. Characterisation of the PML/RAR alpha rearrangement associated with t(15:17) acute promyelocytic leukaemia. Curr Top Microbiol Immunol 1997; 220:81–112.
22 Lin RJ, Nagy L, Inoue S, et al. Role of the histone deacetylase complex in acute promyeloctic leukaemia. Nature 1998; 391:811–4. 23 Grigani F, De Matteis S, Nervi C, et al. Fusion proteins of the retinoic acid receptor-alpha recruit histone deacetylase in promyelocytic leukaemia. Nature 1998; 391:815–8. 24 Soignet S, Maslak P, Wang Z, et al. Complete remission after treatment of acute promyelocytic leukemia with arsenic trioxide. N Eng J Med 1998; 339:1341–48. 25 Sawyers CL. Molecular genetics of acute leukaemia. Lancet 1997; 349:196–200. 26 Liu P, Tarle SA, Hajra A, et al. Fusion between transcription factor CBFβ/PEBP2β and a myosin heavy chain in acute myeloid leukemia. Science 1993; 261:1041–4. 27 Waring PM, Cleary ML. Disruption of a homolog of trithorax by 11q23 translocations: leukemogenic and transcriptional implications. Curr Top Microbiol Immunol 1997; 220:1–23. 28 Gilliland DG, Griffin JD. The roles of FLT3 in hemopoiesis and leukaemia, Blood 2002; 100:1532–42. 29 Levis M, Small D. FLT3: IT Does matter in leukaemia. Leukemia 2003; 17:1738–52. 30 Mrózek K, Bloomfield CD. Chromosome Aberrations, Gene Mutations and Expression Changes, and Prognosis in Adult Acute Myeloid Leukemia. American Society of Hematology Education Program Book 2006, 169–77. 31 Care RS, Valk PJ, Goodeve AC, et al. Incidence and prognosis of c-KIT and FLT3 mutations in core binding factor (CBF) acute myeloid leukemias. Br J Haematol 2003; 121:775–7. 32 Shimada A, Taki T, Tabuchi K, et al. KIT mutations, and not FLT3 internal tandem duplication, are associated with a poor prognosis in pediatric acute myeloid leukaemia with t(8;21): a study of the Japanese Childhood AML Cooperative Study Group. Blood 2006; 107:1806–9. 33 Rubnitz JE, Behm FG, Downing JR. Rearrangements in acute leukemia. Leukemia 1996; 10:74–82. 34 Bullinger L and Valk PJM. Gene expression Profiling in Acute Myeloid Leukemia, J Clin Oncol 2005; 23:6269–305. 35 Rees JKH, Gray RG, Wheatley K. Dose intensity in acute myeloid leukemia: greater effectiveness at lower cost. Principal report of the MRC’s AML 9 Study. Br J Haematol 1996; 94:89–98. 36 Mayer RJ, Davis RB, Schiffer CA, et al. Intensive postremission chemotherapy in adults with acute myeloid leukemia. N Engl J Med 1994; 331:896–942. 37 Bishop JF, Matthews JP, Young GA, et al. A randomized study of high-dose cytarabine in induction in acute myeloid leukemia. Blood 1999; 87:1710–17. 38 Buchner T, Urbanitz D, Hiddemann W, et al. Intensive induction and consolidation with or without maintenance chemotherapy for acute myeloid leukemia (AML): two multicenter studies of German AML Co-operative Group. J Clin Oncol 1985; 3:1583–9. 39 Berman E, Heller G, Santorsa J, et al. Results of a randomized trial comparing idarubicin and cytosine arabinoside with daunorubicin and cytosine arabinoside in adult patients with newly diagnosed acute myelogenous leukemia. Blood 1991; 77:1666–74.
1016 Leukaemias
40 Arlin Z, Case DC Jr, Moore J, et al. Randomized multicenter trial of cytosine arabinoside with mitoxantrone or daunorubicin in previously untreated adult patients with acute nonlymphocytic leukemia (ANLL). Leukemia 1990; 4:177–83. 41 Burnett AK, Goldstone AH, Stevens RMF, et al. Randomised comparison of addition of autologous bone-marrow transplantation to intensive chemotherapy for acute myeloid leukaemia in first remission: results of MRC AML 10 trial. Lancet 1998; 351:700–8. 42 Weick JK, Kopecky KJ, Appelbaum FR, et al. A randomized investigation of high-dose versus standard cytosine arabinoside with daunorubicin in patients with previously untreated acute myeloid leukemia: A Southwest Oncology Group Study. Blood 1996; 88:2841–51. 43 Lowenberg B, van Putten W, Theobald M, et al. Effect of priming with granulocyte colony-stimulating factor on the outcome of chemotherapy for acute myeloid leukemia. N Engl J Med 2003; 349:727–9. 44 Burnett AK, Goldstone AH, Milligan DW. (1999). Daunorubicin versus Mitoxantrone as induction for AML in younger adults given intensive chemotherapy: preliminary results of MRC AML 12 Trial. Br J Haematol 1999; 105 (Suppl 1): 67a. 45 Lowenberg B. Treatment of the elderly patient with acute leukaemia. Baillieres Clin Haematol 1996; 9:147–59. 46 Lowenberg B, Zittoun R, Kerkhofs H, et al. On the value of intensive remission-induction chemotherapy in elderly patients of 65 years with acute myeloid leukemia: a randomised phase III study of the European Organisation for Research and Treatment of Cancer Leukemia Group. J Clin Oncol 1989; 7:1268–74. 47 Dombret H, Chastang C, Fenaux P, et al. A controlled study of recombinant human granulocyte colony-stimulating factor in elderly patients after treatment for acute myelogenous leukemia. N Engl J Med 1995; 332:1678–83. 48 Karp JE, Rizzieri D, Vey N, et al. Cloretazine is an effective therapy in elderly patients with poor-risk de novo AML. J Clin Oncol 2006; 24:18s, abstract 6512 49 Latagliata R, Avvisati G, Lo Coco F. The role of all-transretinoic acid (ATRA) treatment in newly-diagnosed acute promeylocytic leukemia patients aged 60 years. Ann Oncol 1997; 8:1273–5. 50 Jurcic JG, DeBlasio T, Dumont L, et al. Molecular remission induction with retinoic acid and anti-CD33 monoclonal antibody HuM195 in acute promyelocytic leukemia. Clin Cancer Res 2000; 6:372–80. 51 Sievers EL, Appelbaum FR, Speilberger RT, et al. Selective ablation of acute myeloid leukemia using antibody-targeted chemotherapy: A phase I study of an anti-CD33 caliceamicin immunoconjugate. Blood 1999; 93:3678–84. 52 DeAngelo D, Stone R, Durant S, et al. Gemtuzumab ozogamicin (Mylotarg) in combination with induction chemotherapy for treatment of patients with de novo acute myeloid leukemia: Two age-specific phase 2 trials. Blood 2003; 102: abstract 100a. 53 Cripe L, Tallman M, Karanes C, et al. A phase II trial Zosuqidar (LY335979), a modulator of P-glcoprotein (P-gp), plus
54
55
56
57
58
59
60
61
62
63
64
65
66
daunorubicin and high-dose cytarabine in patients with newly-diagnosed secondary acute myeloid leukaemia (AML) refractory anemia with excess blasts in transformation (RAEBt) or relapsed (refractory) AML. Blood 2001; 98: abstract 595a. Lancet J, Gotlib J, Gojo I, et al. Tipifarnib (Zarnestra) in previously untreated poor-risk AML of the elderly: updated results of a phase 2 trial. Blood 2004; 104: abstract 249a. Attar EC, DeAngelo DJ, Ballen KK, et al. Phase I dose escalating trial of bortezomib (velcade) in combination with idarubicin and cytarabine in patients with acute myeloid leukemia. Blood 2004; 104: abstract 498a. Karp JE, Gojo I, Pili R, et al. Targeting vascular endothelial growth factor for relapsed and refractory adult acute myelogenous leukemias: therapy with sequential 1-beta-darabinofuranosylcytosine, mitoxantrone, and bevacizumab. Clin Cancer Res 2004; 10:3577–85. Stone RM, DeAngelo DJ, Klimek V, et al. Patients with acute myeloid leukemia and a deteriorating mutation in FLT3 response to a small FLT3 tyrosine kinase inhibitor, PKC412. Blood 2005; 105:54–60. Kindler T, Breitenbuecher F, Marx A, et al. Efficacy and safety of imatinib in adult patients with c-kit-positive acute myeloid leukemia. Blood 2004; 101:425–32. Marcucci G, Stock W, Zwiebe J, et al. Clinical activity of Genasense (GNS, Oblimersen Sodium), in combination with daunorubicin and cytarabine: a phase I study in previously untreated elderly acute myeloid leukemia. Blood 2003; 102:abstract 385a. Faderl S, Ghandi V, O’Brien S, et al. Results of a phase 1–2 study of clofarabine in combination with cytarabine (ara-C) in relapsed and refractory acute leukemias. Blood 2005; 105:940–7. Burnett AK, Russell, Kell JW, et al. A Phase 2 evaluation of single agent clofarabine as first line treatment of older patients with AML who are not considered fit for intensive chemotherapy. Blood 2004; 104:abstract 248a. Faderl S, Ghandi V, Verstovsek S, et al. Clofarabine plus cytarabine (ARA-C) combination is active in newly diagnosed patients (pts) age 50 with acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS). Blood 2004; 104:abstract 250a. Burnett AK, Baccarani M, Johnson P, et al. Clofarabine in previously untreated (65 yrs) AML patients with an unfavorable cytogenetic profile are considered unfit for standard intensive chemotherapy. J Clin Oncol 2006; 24:18s, abstract #6513. Tallman MS, Andersen JW, Schiffer CA, et al. All-transretenoic acid in acute promyelocytic leukemia. N Engl J Med 1997; 337:1021–8 Tallman M, Nabhan C, Feusner JH, and Rowe JM. Acute promyleocytic leukemia: evolving therapeutic strategies. Blood 2002; 99:759–67. Sanz MA, Martin G, Rayon C, et al. A modified AIDA protocol with anthracycline-based consolidation results in high antileukemic efficacy and reduced toxicity in newly diagnosed PML/RAR-alpha-positive acute promyelocytic leukemia. PETHEMA group. Blood 1999; 94:3015–21.
References 1017
67 Fenaux P, Castaigne S, Dombret H, et al. All-transretinoic acid followed by intensive chemotherapy gives a high complete remission rate and may prolong remissions in newly diagnosed acute promyelocytic leukemia: a pilot study on 26 cases. Blood 1992; 80:2176–81. 68 Shen Zx, Shu ZZ, Fang J, et al. All-trans retinoic acid/As2O yields a high quality remission and survival in newly diagnosed acute promyelocytic leukemia. Proc Natl Acad Sci USA 2004; 101:5328–35. 69 Mathews V, George B, Kakshmi KV, et al. Single-agent arsenic trioxide in the treatment of newly diagnosed promyelocytic leukemia: durable remissions with minimal toxicity. Blood 2006; 107:2627–32. 70 Ghavamzadeh A, Alimoghaddam K, Ghaffari SH, et al. Treatment of acute promyelocytic leukaemia with arsenic trioxide without ATRA and/or chemotherapy. Ann Oncol 2006; 17:131–4. 71 Lo Coco F, Cimino G, Breccia M, et al. Gemtuzumab ozogamicin (Mylotarg) as a single agent for molecularly relapsed acute promyelocytic leukemia. Blood 2004; 104:1995–9. 72 Estey E, Garcia-Manero G, Ferrajoli A, et al. Use of all-trans retinoic acid plus arsenic trioxide as an alternative to chemotherapy in untreated acute promyelocytic leukemia. Blood 2006; 107:3469–73. 73 Khanna-Gupta A, Berliner N. Finding targeted APL therapy targets. Blood 2007; 110:476–7. 74 Burnett AK, Grimwade D, Solomon E, et al. Presenting white blood cell count and kinetics of molecular remission predict prognosis in acute promyelocytic leukemia treated with alltransretenoic acid: results of Randomized MRC Trial. Blood 1999; 94:3015–21. 75 Fenaux P, Chastang C, Chevret S, et al. A randomized comparison of all-trans-retenoic acid (ATRA) followed by chemotherapy and ATRA plus chemotherapy and the role of maintenance therapy in newly diagnosed acute promyelocytic leukemia. The European APL group. Blood 1999; 94:1192–1200. 76 Ohno R, Asou N. The recent JALSG study for newly diagnosed patients with acute promyelocytic leukaemia. Ann Hematol 2004; 83:S77–78. 77 Mahmoud HH, Hurwitz CA, Roberts WM, et al. Tretinoin toxicity in children with acute promyelocytic leukaemia. Lancet 1993; 342:1394–5. 78 Latagliata R, Petti MC, Fenu S, et al. Therapy-related myelodysplastic syndrome-acute myelogenous leukemia in patients treated for acute promyelocytic leukemia: an emerging problem. Blood 2002; 99:822–4. 79 Soignet SL, Maslak P, Wang ZG, et al. Complete remission after treatment of acute prolmyelocytic leukemia with arsenic trioxide. N Engl J Med 1998; 339:1341–8. 80 Jing Y, Wang R, Xai L, et al. Combined effects of alltransretinoic acid and arsenic trioxide in acute promyelocytic leukemia cells in vitro and in vivo. Blood 2001; 97:264–9. 81 Thomas ED, Buckner CD, Clift RA, et al. Marrow transplantation for acute nonlymphoblastic leukemia in first remission. N Engl J Med 1979; 301:597–9.
82 Appelbaum FR, Dahlberg S, Thomas ED, et al. Bone marrow transplantation or chemotherapy after remission induction for adults with acute nonlymphoblastic leukemia: a prospective comparison. Ann Intern Med 1984; 101:581–8. 83 Keating S, de Witte T, Suciu S, et al. The influence of HLAmatched sibling donor availability on treatment outcome for patients with AML: an analysis of the AML 8A study of the EORTC Leukemia Co-operative Group and GIMEMA. Br J Haematol 1998; 102:1344–53. 84 Zittoun RA, Mandelli F, Willemze R, et al. Autologous or allogeneic bone marrow transplantation compared with intensive chemotherapy in acute myelogenous leukemia. N Engl J Med 1995; 332:217–23. 85 Appelbaum FR. Dose intensity and the toxicity and efficacy of allogeneic hematopoietic cell transplantation. Leukemia 2005; 19:171–5. 86 Pagel JM, Appelbaum FR, Eary JF, et al. 131I-anti-CD45 antibody plus busulfan and cyclephosphamide before allogeneic hematopoietic cell transplantation for treatment of acute myeloid leukemia in first remission. Blood 2006; 107:2184–91. 87 Maris MB, Niederwieser D, Sandmaier BM, et al. HLAmatched unrelated donor haematopoietic cell transplantation after non-myeloablative conditioning for patients with hematologic malignancies. Blood 2003; 102:2021–30. 88 Mohty M, De Lavallade H, Ladaique P, et al. The role of reduced intensity conditioning allogeneic stem cell transplantation in patients with acute myeloid leukaemia: a donor vs. no donor comparison. Leukemia 2005; 19:916–20. 89 Mandelli F, Labopin M, Granena A, et al. European survey of bone marrow transplantation in acute promyelocytic leukemia (M3). Working Party on Acute Leukemia of the European Cooperative Group for Bone Marrow Transplantation (EMBT). Bone Marrow Transplant 1994; 14:293–98. 90 Meloni G, Diverio D, Vignetti M, et al. Autologous bone marrow transplantation for acute promyelocytic leukemia in second remission: prognostic relevance of pre-transplant minimal residual disease assessment by reversetranscription polymerase chain reaction of the PML/RARα fusion in gene. Blood 1997; 90:1321–5. 91 Gorin NC. Autologous stem cell transplantation in acute myelocytic leukemia. Blood 1998; 92:1073–90. 92 Gorin NC, Labopin M, Meloni G, et al. Autologous bone marrow transplantation for acute myeloblastic leukemia in Europe: further evidence of the role of marrow purging by mafosfamide. Leukemia 1991; 5:896–904. 93 Harousseau JL, Cahn JY, Pignon B, et al. Comparison of autologous bone marrow transplantation and intensive chemotherapy as postremission therapy in adult acute myeloid leukemia. Blood 1997; 90:2978–86. 94 Cassileth PA, Harrington DP, Appelbaum FR, et al. Chemotherapy compared with autologous or allogeneic bone marrow transplantation in the management of acute myeloid leukemia in first remission. N Engl J Med 1998; 339:1649–56.
1018 Leukaemias
95 Kern W, Aul C, Maschmeyer G, et al. Superiority of highdose over intermediate-dose cytosine arabinoside in the treatment of patients with high-risk acute myeloid leukemia: results of an age-adjusted prospective randomised comparison. Leukemia 1998; 12:1049–55. 96 Grimwade D, Walker H, Oliver F, et al. The importance of diagnostic cytogenetics on outcome in AML: analysis of 1,612 patients entered into the MRC AML 10 Trial. Blood 1998; 92:2322–33. 97 Petersen FB, Lynch MHE, Clift RA, et al. Autologous marrow transplantation for patients with acute myeloid leukemia in untreated first relapse or in second complete remission. J Clin Oncol 1993; 11:353–60. 98 Allogeneic bone marrow transplantation for leukemia in Europe: report from the Working Party on Leukemia, European Group for Bone Marrow Transplantation. Lancet 1998; 1:1379–82. 99 Druker BJ, Lydon NB. Lessons learned from the development of an abl tyrosine kinase inhibitor for chronic myelogenous leukemia. J Clin Invest 2000; 105:3–7. 100 Tallman M. From rags to riches. Blood 2006; 107:3425. 101 Tallman MS, Gilliland DG, Rowe JM. Drug therapy for acute myeloid leukemia. Blood 2005; 106:1154–61. 102 Cortes JE, Kantarjian HM. Acute lymphoblastic leukemia: a comprehensive review with emphasis on biology and therapy. Cancer 1995; 76:2393–417. 103 Kinlen LJ. Epidemiological evidence for an infective basis in childhood leukaemia. Br J Cancer 1995; 71:1–5. 104 Jaffe E, Harris NL, Stein H, Vardiman JW (Eds) 2001 World Health Organization Classification of Tumors. Pathology and Genetics of Tumors of Haematopoietic and Lymphoid Tissues. IARC Press, Lyon, France. 105 First MIC Co-operative Study Group Morphologic, immunologic and cytogenetics (MIC) working classification of acute lymphoblastic leukaemias. Cancer Genetics and Cytogenetics 1986; 23:189–97. 106 Uckun FM, Sather HN, Gaynon PS, et al. Clinical features and treatment outcome of children with myeloid antigen positive acute lymphoblastic leukemia: a report from the Children’s Cancer Group. Blood 1997; 90:28–35. 107 Armstrong SA, Look AT. Molecular genetics of acute lymphoblastic leukemia. J Clin Oncol 2005; 23:6306–15. 108 Rowley JD. The critical role of chromosome translocations in human leukemias. Annu Rev Genet 1998; 32:495–519. 109 Ferrando AA, Neuberg DS, Staunton J, et al. Gene expression signatures define novel oncogenic pathways in T cell acute lymphoblastic leukaemia. Cancer Cell 2002; 1:75–87. 110 Armstrong SA, Kung AL, Mabon ME, et al. Inhibition of FLT3 in MLL. Validation of a therapeutic target identified by gene expression based classification. Cancer Cell 2003; 3:173–83. 111 Kaspers GJ, Smets LA, Pieters R, et al. Favorable prognosis of hyperdiploid common acute lymphoblastic leukemia may be explained by sensitivity to antimetabolites and other drugs: results of an in vitro study. Blood 1995; 85:751–6. 112 Melo JV. The diversity of the BCR-ABL fusion proteins and their relationship to leukemia phenotype. Blood 1996; 88:2375–84.
113 Uchida H, Downing JR, Miyazaki Y, Frank R, et al. Three distinct domains in TEL-AMLI are required for transcriptional repression of the IL-3 promoter. Oncogene 1999; 18:1015–22. 114 Wiemels JL, Cazzaniga G, Daniotti M, et al. Prenatal origin of acute lymphoblastic leukaemia in children. Lancet 1999; 354:1499–1503. 115 Ford AM, Bennett CA, Price CM, et al. Fetal origins of the TEL-AML1 fusion gene in identical twins with leukemia. Proc Natl Acad Sci USA 1998; 95:4584–8. 116 Dedre DA, Waller EK, LeBrun DP, et al. Chimeric homeobox gene E2A-PBX1 induces proliferation, apoptosis, and malignant lymphomas in transgenic mice. Cell 1993; 74:833–43. 117 Inaba T, Inukai T, Yoshihara T, et al. Reversal of apoptosis by the leukaemia-associated E2A-HLF chimaeric transcription factor. Nature 1996; 382:541–4. 118 Harris AW, Pinkert CA, Crawford M, et al. The Emu-myc transgenic mouse: A model for high-incidence spontaneous lymphoma and leukemia of early B cells. J Exp Med 1988; 167:353–71. 119 Porcher C, Wojciech S, Rockwell K, et al. The T cell leukemia oncoprotein SCI/tal-1 is essential for development of all hematopoietic lineages. Cell 1996; 86:47–57. 120 Group Francais de Cytogenetique Hematologique. Cytogenetic abnormalities in adult acute lymphoblastic leukemia: correlations with hematologic findings and outcome: a collaborative study of the Groupe Francais de Cytogenetique Hematologique. Blood 1996; 87:3135–42. [Erratum, Blood 1996; 88:2818.] 121 Hoeman CD, Beaulieu N, Girard L, et al. Two distinct Notch 1 mutant alleles are involved in the induction of T-cell leukemia in c-myc transgenic mice. Mol Cell Biol 2000; 20:3831–42. 122 Nachman J. Clinical characteristics, biologic features and outcome for young patients with acute lymphoblastic leukaemia. Br J Haematol 2005; 130:166-73. 123 Chessells JM, Hall E, Prentice HG, et al. The impact of age on outcome in lymphoblastic leukaemia: MRC UKALL X and XA compared: a report from the MRC Paediatric and Adult Working Parties. Leukemia 1998; 12:463–73. 124 Larson RA, Dodge RK, Burns CP, et al. A five-drug remission induction regimen with intensive consolidation for adults with acute lymphoblastic leukemia: Cancer and Leukemia Group B Study 8811. Blood 1995; 85:2025–37. 125 Bostrum BC, Sensel MR, Sather HN, et al. Dexamethasone versus prednisone and daily oral versus weekly intravenous mercaptopurine for patients with standard-risk acute lymphoblastic leukaemia: a report from the Children’s Cancer Group. Blood 2003; 101:3809–17. 126 Schwartz CL, Thompson EB, Gelber RD, et al. Improved response with higher corticosteroid dose in children with acute lymphoblastic leukaemia. J Clin Oncol 2001; 19:1040–6 127 Durrant IJ, Richards SM, Prentice HG, Goldstone AH. The Medical Research Council trials in adult lymphoblastic leukemia. Hematol Oncol Clin North Am 2000; 14:1327–52. 128 Laport GF, Larson RA. Treatment of adult lymphoblastic leukemia. Semin Oncol 1997; 24:70–82.
References 1019
129 Anino L, Vegna ML, Camera A, et al. Treatment of adult acute lymphoblastic leukemia (ALL): long-term follow-up of the GIMEMA ALL 028 randomized study. Blood 2002; 99:863–71. 130 Conter V, Schrappe M, Arico M, et al. Role of cranial radiotherapy for childhood T-cell acute lymphoblastic leukemia with high WBC count and good response to prednisolone. J Clin Oncol 1997; 15:2786–91. 131 Cortes J, O’Brien S, Pierce S, et al. The value of high-dose systemic chemotherapy and intrathecal therapy for central nervous system prophylaxis in different risk groups of adult acute lymphoblastic leukemia. Blood 1995; 86:2091–7. 132 Schrappe M, Reiter A, Ludwig WD, et al. Improved outcome in childhood acute lymphoblastic leukaemia despite reduced use of anthracyclines and cranial radiotherapy: results of trial ALL-BFM 90. German–Austrian–Swiss ALL-BFM Study Group. Blood 2000; 95:3310–22. 133 LeClerc JM, Billett AL, Gelber RD, et al. Treatment of childhood acute lymphoblastic leukemia: Results of DanaFarber ALL consortium protocol 87-01. J Clin Oncol 2002; 20:237–46. 134 Soussain C, Patte C, Ostronoff M, et al. Small noncleaved cell lymphoma and leukemia in adults: a retrospective study of 65 adults treated with the LMB pediatric protocols. Blood 1995; 85:664–74. 135 Hoelzer D, Ludwig WD, Thiel E, et al. Improved outcome in adult B-cell acute lymphoblastic leukemia. Blood 1996; 87:495–508. 136 Kantarjian H, Thomas D, O’Brien S, et al. Long-term followup results of hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone (Hyper-CVAD), a dose-intensive regimen, in adult acute lymphocytic leukaemia. Cancer 2004; 101:2788–801 137 Thomas DA, O’Brien S, Cortes J, et al. Outcome with the hyper-CVAD regimens in lymphoblastic lymphoma. Blood 2004; 104:1624–30. 138 Silverman LB, Gelber RD, Dalton VK, et al. Improved outcome for children with acute lymphoblastic leukaemia: results of Dana-Farber Consortium Protocol 91-01. Blood 2001; 97:1211–18. 139 DeAngelo D. (2005) The treatment of adolescents and young adults with acute lymphoblastic leukemia. American Society of Hematology, Education Program Book, 123–130. 140 de Bont JM, Holt B, Dekker AW, et al. Significant difference in outcome for adolescents with acute lymphoblastic leukaemia treated on pediatric versus adult protocols in the Netherlands. Leukemia 2004; 18:2032–35. 141 Chessells JM, Bailey C, Richards SM. Intensification of treatment and survival in all children with lymphoblastic leukaemia: results of UK Medical Research Council trial UKALL X. Lancet 1995; 345:143–8. 142 Pui CH, Evans WE. Acute lymphoblastic leukemia. N Engl J Med 1998; 339:605–15. 143 Childhood ALL Collaborative Group. Duration and intensity of maintenance chemotherapy in acute lymphoblastic leukaemia: overview of 42 trials involving 12,000 randomised children. Lancet 1996; 347:1783–8.
144 Druker BJ, Sawyers CL, Kantarjian H, et al. Activity of a specific inhibitor of the BCR-ABL tyrosine kinase in blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N Engl J Med 2001; 344:1038–42. 145 Ottman OG, Druker BJ, Sawyers CL, et al. A phase 2 study of imatinib in patients with relapsed or refractory Philadelphia-chromosome positive acute lymphoblastic leukemias. Blood 2002; 100:1965–71. 146 Wassman B, Pfeifer H, Scheuring UJ, et al. Early prediction of response in patients with relapsed or refractory Philadelphia chromosome positive acute lymphoblastic leukemia (PhALL) treated with imatinib mesylate (Glivec). Blood 2004; 103:1495–8. 147 Towatari M, Yanada M, Usui N, et al. Combination of intensive chemotherapy and imatinib can rapidly induce high-quality complete remission for a majority of patients with newly diagnosed BCR-ABL-positive acute lymphoblastic leukemia. Blood 2004; 104:3507–35. 148 Wetzler M, Stock W, Owzar K, et al. Sequential imatinib and chemotherapy yield reverse-transcriptase polymersae chain reaction (RT-PCR)-negative peripheral stem cell collections in Philadelphia (Ph) chromosome positive acute lymphoblastic leukemia (ALL) – Preliminary results of CALGB 10001. J Clin Oncol 2006; 24:18s, abstract 6549. 149 Coutre S, Martinelli G, Dombret A, et al. Dasatinib (D) in patients (Pts) with chronic myelogenous leukemia (CML) in lymphoid blast crisis (LB-CML) or Philadelphia-chromosome positive acute lymphoblastic leukemia (PhALL) who are imatinib (IM)-resistant or intolerant (IM-I): The CA 180015 ‘START’-L study. J Clin Oncol 2006; 24:18s, abstract 6528. 150 Giles FJ, Larson RA, LeCoutre P, et al. A phase II study of AMN107, a novel inhibitor of Bcr-Abl, administered to imatinib-resistant or intolerant patients (pts) with Ph chronic myelogenous leukemia (CML) in blast crisis (BC) or relapsed/refractory Ph acute lymphoblastic leukemia (ALL). J Clin Oncol 2006; 24:18s, abstract 6536. 151 Wheeler K, Richards S, Bailey C, et al. Comparison of Bone Marrow Transplant and chemotherapy for relapsed childhood acute lymphoblastic leukaemia – the MRC UK ALL X experience. Br J Haematol 1998; 101:94–103. 152 Lawson SE, Harrison G, Richards S, et al. The UK experience in treating relapsed childhood acute lymphoblastic leukaemia: a report on the Medical Research Council UKALLR1 study. Br J Haematol 2000; 108:531–43. 153 Kurtzberg J, Ernst TJ, Keating M, et al. Phase I study of 506U administered on a consecutive 5-day schedule in children and adults with refractory hematologic malignancies. J Clin Oncol 2005; 23:3396–403. 154 Berg SL, Blaney SM, Devidas M, et al. Phase II study of nelarabine (compound 506U78) in children and young adults with refractory T-cell malignancies: a report from the Children’s Oncology Group. J Clin Oncol 2005; 23:3376–82. 155 Appelbaum FR. Allogeneic hematopoietic stem cell transplantation for acute leukemia. Semin Oncol 1997; 24:114–23.
1020 Leukaemias
156 Sebban C, Lepage E, Vernant JP, et al. Allogeneic bone marrow transplantation in adult acute lymphoblastic leukemia in first complete remission: a comparative study. J Clin Oncol 1994; 12:2580–7. 157 Garcia-Manero G, Thomas DA. Salvage therapy for refractory or relapsed acute lymphoblastic leukemia. Hematol/Oncol Clin North Am 2001; 15: 163–205. 158 Billett AL, Kornmehl E, Tarbell NJ, et al. Autologous bone marrow transplantation after a long first remission for children with recurrent acute lymphoblastic leukemia. Blood 1993; 81:1651–7. 159 Borgmann A, Schmid H, Hartmann R, et al. Autologous marrow transplants compared with chemotherapy for children with acute lymphoblastic leukaemia in a second remission: a matched pair analysis. Lancet 1995; 346:873–6. 160 Weisdorf DJ, Billett AL, Hannan P, et al. Autologous versus unrelated donor allogeneic marrow transplantation for acute lymphoblastic leukaemia. Blood 1997; 90:2962–8. 161 Messina C, Valsecchi MG, Arico M, et al. Autologous bone marrow transplantation for treatment of isolated central nervous system relapse of childhood acute lymphoblastic leukemia. Bone Marrow Transplant 1998; 21:9–14. 162 Attal M, Blaise D, Marit G, et al. Consolidation treatment of adult acute lymphoblastic leukemia: A prospective randomized trial comparing allogeneic versus autologous bone marrow transplantation and testing the impact of recombinant interleukin-2 after autologous bone marrow transplant. BGMT Group. Blood 1995; 86:1619–28. 163 Braun SE, Chen K, Battiwalla M, et al. Gene therapy strategies for leukemia. Mol Med Today 1997; 3:39–46. 164 Ek O, Gaynon P, Zeren T, et al. Treatment of human B-cell precursor leukemia in SCID mice by using a combination of the anti-CD immunotoxin B43-PAP with the standard chemotherapeutic drugs vincristine, methylprednisolone, and L-asparaginase. Leuk Lymphoma 1998; 31:143–9. 165 Locatelli F. The role of repeat of transplantation of haemopoietic stem cells and adoptive immunotherapy in treatment of leukaemia relapsing following allogeneic transplantation. Br J Haematol 1998; 102:633–8. 166 Nowell PC, Hungerford DA. A minute chromosome in human chronic granulocytic leukemia. Science 1960; 132:1497. 167 Rowley JD. A new consistent chromosomal abnormality in chronic myelogenous leukaemia identified by quinacrine fluorescence and Giemsa staining. Nature 1973; 243: 290–3. 168 Groffen J, Stephenson JR, Heisterkamp N, et al. Philadelphia chromosomal breakpoints are clustered within a limited region, bcr, on chromosome 22. Cell 1984; 36:93–9. 169 Druker BJ, Tamura S, Buchdunger E, et al. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of the BCR-ABL positive cells. Nature Med 1996; 2:561–6. 170 O’Brien SG, Guilhot F, Larson RA, et al. Imatinib compared with interferon and low dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med 2003; 348:994–1004. 171 Mughal TI, Goldman JM. Chronic myeloid leukaemia: A therapeutic challenge. Ann Oncol 1995; 6:637–44.
172 Goldman JM, Apperley JF, Jones LM, et al. Bone marrow transplantation for patients with chronic myeloid leukemia. N Engl J Med 1986; 314:202–7. 173 Mughal TI, Yong A, Szydlo R, et al. The probability of longterm leukaemia free survival for patients in molecular remission 5 years after allogeneic stem cell transplantation for chronic myeloid leukaemia in chronic phase. Br J Haematol 2001; 115:569–74. 174 Marmont A, Frassoni F, Bacigalupo A, et al. Recurrence of Ph’-leukemia in donor cells after marrow transplantation for chronic myeloid leukemia. N Engl J Med 1984; 310:903. 175 Sawyers CL. Chronic myeloid leukemia. N Engl J Med 1999; 340:1330–8. 176 Kurzrock R, Gutterman JU, Talpaz M. The molecular genetics of Philadelphia chromosome positive leukemias. N Engl J Med 1988; 31:990. 177 Groffen J, Heisterkamp N. The chimeric BCR-ABL gene. Bailliere’s Clin Haematol 1997; 10:187–201. 178 Ahuja H, Bar-Eli M, Arlin Z, et al. The spectrum of molecular alterations in the evolution of chronic myeloid leukemia. J Clin Invest 1991; 87:2042–46. 179 Goldman JM, Melo JV. Chronic myeloid leukemia – advances in biology and new approaches to treatment. N Engl J Med 2003; 349:1451–64. 180 Pane F, Frigeri F, Sindona M, et al. Neutrophilic-chronic myeloid leukemia: a distinct disease with a specific molecular marker. Blood 1996; 88:2410–4. [Erratum, Blood 1997; 89:4244.] 181 Deininger MW, Goldman JM, Melo JV. The molecular biology of chronic myeloid leukemia. Blood 2000; 96:3343–56. 182 Epstein FH. The biology of chronic myeloid leukemia. N Engl J Med 1999; 341:164–72. 183 Holyoake T, Jiang Xiaoyan, Eaves C and Eaves Allen. Isolation of a highly quiescent subpopulation of primitive leukemic cells in chronic myeloid leukemia. Blood 1999; 94: 2056–64. 184 Goldman JM, Gordon MY. Why do chronic myelogenous leukemia stem cells survive allogeniec stem cell transplantation or imatinib: does it really matter? Leukemia & Lymphoma 2006; 47:1–7. 185 Sokal JE, Cox EB, Baccarani M et al. Prognostic discrimination in “good-risk” chronic granulocytic leukemia. Blood 1984; 63:789–99. 186 Hasford J, Pfirrmann M, Hehlmann R et al. A new prognostic score for survival of patients with chronic myeloid leukaemia treated with interferon alpha. J Natl Cancer Inst 1998; 90:850–8. 187 Baccarani M, Saglio G, Goldman JM, et al. Evolving concepts in the management of chronic myeloid leukemia. Recommendations from an expert panel on behalf of the European Leukemia-net. Blood 2006; 108:1809–20. 188 Gratwohl A, Hermans J, Goldman J, et al. Risk assessment for patients with chronic myeloid leukaemia before allogeneic blood or marrow transplantation. Lancet 1998; 352:1078–92. 189 Schiffer CA. BCR-ABL tyrosine kinase inhibitors for chronic myelogenous leukemia. N Engl J Med 2007; 357:258–65.
References 1021
190 Druker BJ, Talpaz M, Resta D, et al. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001; 344:1031–7. 191 Goldman JM, Melo JV. Targeting the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001; 344:1084–6. 192 Mughal TI, Goldman JM. Chronic Myeloid leukemia: Current Status and Controversies. Oncology 2004; 18:837–47. 193 Ebonether M, Stentoft J, Ford J, et al. Cerebral edema as a possible complication of treatment with imatinib. Lancet 2002; 359:1751–2. 194 Druker B, Guilhot F, O’Brien S, et al. Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia in chronic phase. N Engl J Med 2006; 355:2408–17. 195 Goldman JM. How I treat chronic myeloid leukemia in the imatinib era. Blood 2007; 110:2828–2837. 196 Kantarjian HM, Talpaz M, O’Brien S, et al. Survival benefit with imatinib mesylate versus interferon alpha-based regimens in newly diagnosed chronic phase chronic myelogenous leukemia. Blood 2006; 108:1835–40. 197 Kaeda J, O’Shea D, Szydlo RM, et al. Serial measurements of BCR-ABL transcripts in the peripheral blood after allogeneic stem cell transplantation for chronic myeloid leukemia: an attempt to define patients who may not require further therapy. Blood 2006; 107:4171–6. 198 Gorre ME, Mohammed M, Ellwood K, et al. Clinical resistance to STI571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science 2001; 293:876–80. 199 Mughal TI, Cortes J, Cross NCP, et al. Chronic myeloid leukemia – some topical issues. Leukemia 2007; 21:1347–52. 200 Deininger M, Buchdunger E, Druker BJ. The development of imatinib as a therapeutic agent for chronic myeloid leukemia. Blood 2005; 105:2640–53. 201 Mughal TI. Do kinase domain mutations in CML matter? Clin Leuk, 2007; 207. 202 Baccarani M, Martinelli G, Rosti G, et al. Imatinib and pegylated human recombinant interferon-alpha2b in early chronic-phase chronic myeloid leukemia. Blood 2004; 104:4245–51. 203 Cortes J, Kantarjian H. New targeted approaches in chronic myeloid leukemia. J Clin Oncol 2005; 23:6316–24. 204 Issa J-P, Gharibyan V, Cortes J, et al. Phase II study of lowdose decitabine in patients with chronic myelogenous leukemia resistant to imatinib mesylate. J Clin Oncol 2005; 23:3948–56. 205 Cortes J, Albitar M, Thomas D, et al. Efficacy of the farnesyl transferase inhibitor R115777 in chronic myeloid leukemia and other hematologic malignancies. Blood 2003; 101:1692–7. 206 Cortes J, O’Brien S, Verstovsek S, et al. Phase I study of lonafarnib (SCH66336) in combination with imatinib for patients (pts) with chronic myeloid leukemia (CML) after failure to imatinib. Blood 2004; 104:288a, abstract 1009. 207 O’Brien S, Kantarjian H, Keating M, et al. Homoharringtonine therapy induces responses in patients with chronic myelogenous leukemia in late chronic phase. Blood 1995; 86:3322–6.
208 Mughal TI, Goldman JM. Molecularly targeted treatment of chronic myeloid leukemia: Beyond the imatinib era. Frontiers in Bioscience 2006; 1:209–20. 209 Talpaz M, Shah NP, Kantarjian H, et al. Dasatinib in imatinib-resistant Philadelphia chromosome-positive leukemias. N Engl J Med 2006; 354:2531–41. 210 Guilhot F, Apperley J, Kim DW, et al. Dasatinib induces significant hematologic and cytogenetic responses in patients with imatinib-resistant or -intolerant chronic myeloid leukemia in accelerated phase. Blood 2007; 109:4143–50. 211 Cortes J, Rousselot P, Kim DW, et al. Dasatinib induces complete hematologic and cytogenetic responses in patients with imatinib-resistant or -intolerant chronic myeloid leukemia in blast crisis. Blood 2007; 109:3207–13. 212 Copland M, Hamilton A, Elrick LJ, et al. Dasatinib (BMS354825) targets an earlier progenitor population than imatinib in primary CML, but does not eliminate the quiescent fraction. Blood, 2006: 107:4532–4539. 213 Kantarjian H, Giles F, Wunderle L, et al. A phase II study of nilotinib in imatinib-resistant CML and Philadelphia-positive ALL. N Engl J Med 2006; 354:2542–51. 214 Rosti G, le Coutre P. Bhalla K, et al. A phase II study of nilotinib, a novel tyrosine kinase inhibitor administered to imarinib-resistant and -intolerant patients with Phialdelphia-poistive chronic myelogenous leukemia in chrnic phase. J Clin Oncol 2007; 25(Suppl):18S [abstract]. 215 Bolin RW, Robinson WA, Sutherland J, et al. Busulfan versus hydroxyurea in the long-term therapy of chronic myelogenous leukemia. Cancer 1982; 50:1683–7. 216 Allan NC, Richards SM, Shepherd PCA. UK Medical Research Council randomised multicentre trial of interferon-αn1 for chronic myeloid leukaemia: improved survival irrespective of cytogenetic response. Lancet 1995; 345:1392–7. 217 The Chronic Myeloid Leukemia Trialists’ Collaborative Group. Interferon alpha versus chemotherapy for chronic myeloid leukemia: a meta-analysis of seven randomized trials. J Natl Cancer Inst 1997; 89:1616–20. 218 Silver RT, Woolf SH, Hehlmann R, et al. An evidence-based analysis of the effect of busulfan, hydroxyurea, interferon, and allogeneic bone marrow transplantation in treating the chronic phase of chronic myeloid leukemia: Development for the American Society of Hematology. Blood 1999; 94:1517–36. 219 Guilhot F, Chastang C, Michallet M et al. Interferon alpha-2b combined with cytarabine versus interferon alone in chronic myelogenous leukemia. N Engl J Med 1997; 337:223–9. 220 Srivastava PK. Immunotherapy of human cancer: lessons from mice. Nat Immunol 2000; 1:363–6. 221 Molldrem JJ, Lee PP, Wang C, et al. Evidence that specific T lymphocytes may participate in the elimination of chronic myelogenous leukemia. Nat Med 2000; 6:1018–23. 222 Bocchia M, Gentili S, Abruzzese E, et al. Effect of a p210 multipeptide vaccine associated with imatinib or interferon in patients with chronic myeloid leukaemia and persistent residual disease: a multicentre observational trial. Lancet 2005; 365:657–9.
1022 Leukaemias
223 Zeng Y, Graner MW, Thompson S, et al. Induction of BCR-ABL-specific immunity following vaccination with chaperone rich lysates from BCR-ABL tumor cells. Blood 2005; 105:2016–22. 224 Oka Y, Tsuboi A, Taguchi T, et al. Induction of WTI (Wilms tumor gene)-specific cytotoxic T lymphocytes by WTI peptide vaccine and the resultant cancer regression. PNAS 2004; 101:13885–90. 225 Quintás-Cardama A, Kantarjian H, Cortes J. Tyrosine kinase inhibitors for chronic myeloid leukemia. N Engl J Med 2007; 15:1557–1558. 226 Nimmanapalli R, Fuino L, Bali P, et al. Histone deacetylase inhibitor LAQ824 both lowers expression and promotes proteasomal degradation of Bcr-Abl and induces apoptosis of imatinib mesylate-sensitive or –refractory chronic myelogenous leukemia-blast crisis cells. Cancer Res 2003; 63:5126–35. 227 Nimmanapalli R, O’Bryan E, Bhalla K. Geldanamycin and its analogue 17-allylamino-17-demethoxygeldanamycin lowers Bcr-Abl levels and induces apoptosis and differentiation of Bcr-Abl-positive human leukemic blasts. Cancer Res 2001; 61:1799–804. 228 Wolff NC, Veach DR, Tong WP, et al. PD166326, a novel tyrosine kinase inhibitor, has greater antileukemic activity than imatinib mesylate in a murine model of chronic myelogenous leukemia. Blood 2005; 105:3995–4003. 229 Mayerhofer M, Aichberger KJ, Florian S, et al. Identification of mTOR as a novel bifunctional target in chronic myeloid leukemia: dissection of growth-inhibitory and VEGFsuppressive effects of rapamycin in leukemic cells. FASEB J 2005; 19:960–2. 230 Jabbour E, Cortes JE, Ghanem H, et al. Targeted therapy in chronic myeloid leukemia. Expert Rev Anticancer Ther 2008; 1:99–100. 231 Giles FJ, Cortes J, Jones D, et al. MK-0457, a novel kinase inhibitor, is active in patients with chronic myeloid leukemia or acute lymphocytic leukemia with the T3151 BCR-ABL mutation. Blood 2007; 109:500–2. 232 Mughal TI, Goldman JM. Treatment strategies for T3151 CML. Clin Lymphoma 2007; suppl S81–84. 233 Carter BZ, Mak DH, Schober WD, et al. Regulation of surviving expression through Bcr-Abl/MAPK cascade: targeting surviving overcomes imatinib resistance and increases imatinib sensitivity in imatinib-responsive CML cells. Blood 2006; 107:1555–63. 234 Wiesdorf DJ, Anasetti C, Antin JH, et al. Allogeneic bone marrow transplantation for chronic myelogenous leukemia: comparative analysis of unrelated versus matched sibling donor transplantation. Blood 2002; 99:1971–7. 235 Gratwohl A, Brand R, Apperley J, et al. Allogeneic hematopoietic stem cell transplantation for chronic myeloid leukemia in Europe 2006: transplant activity, long-term data and current results. An analysis by the Chronic Leukemia Working Party of the European Group for Blood and Marrow Transplantation (EBMT). Haematologica 2006; 91:513–21 236 Dazzi F, Szydlo RM, Cross NC, et al. Durability of responses following donor lymphocytes infusions for patients who
237
238
239
240
241
242
243
244
245
246
247 248
249
250
251
relapse after allogeneic stem cell transplantation for chronic myeloid leukemia. Blood 2000; 96:2712–16. Bornhäuser M, Kröger N, Schwerdtfeger R, et al. Allogeneic haematopoietic cell transplantation for chronic myelogenous leukaemia in the era of imatinib: a retrospective multicente study. Eur J Haematol 2006; 76:9–17. Carella AM, Lerma E, Corsetti MT, et al. Autografting with Philadelphia chromosome negative mobilized hematopoietic progenitor cells in chronic myelogenous leukemia. Blood 1999; 83:1534–1539. Apperley JF, Boque C, Carella A, et al. Autografting in chronic myeloid leukaemia: a meta-analysis of six randomised trials. Bone Marrow Transplantation 2004; 33 (Suppl 1): S28. Mughal TI, Hoyle C, Goldman JM. Autografting for patients with chronic myeloid leukemia – the Hammersmith Experience. Stem Cell 2004; 11:20–22. Perseghin P, Gambacorti-Passerini C, Tornaghi L, et al. Peripheral blood progenitor cell collection in chronic myeloid leukemia patients with complete cytogenetic response after treatment with imatinib mesylate. Transfusion 2005; 45:1214–1220. Le Coutre PD, Ottmann O, Gatterman N, et al. A phase II study of AMN107, a novel inhibitor of Bcr-Abl administered to imatinib-resistant or intolerant patients (pts) with chronic myelogenous leukemia (CML) in accelerated phase (AP). J Clin Oncol 2006; 24:18s abstract #6531. Mughal TI, Goldman JM. Chronic myeloid leukemia: Why does it evolve from chronic phase to blast transformation? Frontiers in Bioscience 2006; 1:198–208. Kantarjian HM, Cortes J, O’Brien S, et al. Imatinib mesylate (STI571) therapy for Philadelphia chromosome-positive chronic myelogenous leukaemia in blast phase. Blood 2002; 99:3547–53. Cortes J, Rousselot P, Kim DW, et al. Dasatinib induces complete hematologic and cytogenetic responses in patients with imatinib-resistant or -intolerant chronic myeloid leukemia in blast crisis. Blood 2007; 109:3207–13. Mughal TI. Nilotinib in the post-imatinib era for the treatment of Philadelphia-positive leukemia. Clin Leuk 2006; 1:84–6. Mughal TI, Cortes J, Cross NCP. Chronic myeloid leukemia: some topical challenges. Leukemia 2007; 21:1347–52. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: chronic myelogenous leukemia. Version 2.2007. (Accessed June 21, 2007, at http://www.nccn.org/professionals/physician_gls/PDF/cml.pdf Chronic myelogenous leukemia (PDQ): treatment. Bethesda, MD: National Cancer Institute (Accessed June 21, 2007, at http://www.cancer.gov/cancertaopics/pdq/treatment/CML/ HealthProfessional.) Wierda W, O’Brien S, Wang X, et al. Prognostic nomogram and index for overall survival in previously untreated patients with chronic lymphocytic leukemia. Blood 2007; 109:4679–85. Zent CS, Kyasa MJ, Evans R, Schichman SA. Chronic lymphocytic leukaemia incidence is substantially higher
References 1023
252
253
254
255
256
257
258 259
260
261
262
263 264
265
266
267 268
than estimated from tumour registry data. Cancer 2001; 92:1325–30. Feychting B, Forssen U, Floderus B. Occupational and residential magnetic field exposure and leukemia and central nervous system tumors. Epidemiology 1997; 8:384–89. Mann DL, DeSantis P, Mark G. HTLV-1 associated B-cell CLL: indirect role for retrovirus in leukemogenesis. Science 1987; 263:1103–6. Kalyanaraman VS, Sarnagadharan MG, Robert-Guroff M, et al. A new subtype of human T-cell leukemia virus (HTLV-2) associated with a T-cell variant of hairy cell leukemia. Science 1982; 218:571–3. Rawstron AC, Yuille MR, Fuller J, et al. Inherited predisposition to CLL is detectable as subclinical monoclonal B-lymphocyte expansion. Blood 2002; 100:2289-90. International workshop on Chronic Lymphocytic Leukemia. Chronic lymphocytic leukemia: recommendations for diagnosis, staging and response criteria. Ann Intern Med 1989; 110:236–8. Cheson BD, Bennett JM, Grever M, et al. National Cancer Institute-Sponsored working Group guidelines for chronic lymphocytic leukemia: Revised guidelines for diagnosis and treatment. Blood 1996; 87:4990–97. Chiorazzi N, Rai K, Ferrarini M. Chronic Lymphocytic Leukemia. N Engl J Med 2005; 352:804–15. Matutes E, Owusu-Ankomah K, Morilla R, et al. The immunological profile of B-cell disorders and proposal of a scoring system for the diagnosis of CLL. Leukemia 1994; 8:1640–5. Gaidano G, Newcomb EW, Gong JZ, et al. Analysis of alterations of oncogenes and tumor suppressor genes in chronic lymphocytic leukemia. Am J Path 1994; 144:1312–9. Kalachikov S, Migliazza A, Cayanis E, et al. Cloning and gene mapping of the chromosomes 13q14 region detected in chronic lymphocytic leukemia. Genomics 1997; 42:369–77. Kienle D, Benner A, Kröber A, et al. Distinct gene expression patterns in chronic lymphocytic leukaemia defined by usage of specific VH genes. Blood 2006; 107:2090–3. Caligaris-Cappio F B-chronic lymphocytic leukemia: A malignancy of anti-self B cells. Blood 1996; 87:2615–20. Crespo M, Bosch F, Villamor N, et al. ZAP-70 expression as a surrogate for immunoglobulin-varaible-region mutations in chronic lymphocytic leukaemia. N Engl J Med 2003; 348:1764–75. Hamblin TJ, Orchard JA, Ibbotson RE, et al. CD38 expression and immunoglobulin variable region mutatations are independent prognostic variables in chronic lymphocytic leukemia, but CD38 expression may vary during the course of the disease. Blood 2002; 99:1023–9. Krober A, Seller T, Benner A, et al. V(H) mutation status, CD38 expression level, genomic aberrations, and survival in chronic lymphocytic leukaemia. Blood 2002; 100:1410–16. Rai KR, Sawitsky A, Cronkite EP, et al. Clinical staging of chronic lymphocytic leukemia. Blood 1975; 46:219–34. Binet JL, Augier A, Dighiero G, et al. A new prognostic classification of chronic lymphocytic leukemia derived from a multivariate survival analysis. Cancer 1981; 48:198–206.
269 Rai KR, Montserrat E. Prognostic factors in chronic lymphocytic leukemia. Semin Hematol 1987; 4:252–6. 270 Montserrat E. New prognostic markers in CLL. Hematology 2006; 1:279–84. 271 Chronic Lymphocytic Leukemia Trialists’ Collaborative Group. Chemotherapeutic options in chronic lymphocytic leukemia: a meta-analysis of the randomized trials. J Natl Cancer Inst 1999; 91:861–8. 272 Cheson BD, Bennett JM, Grever M, et al. National Cancer Institute-Sponsored Working Group guidelines for chronic lymphocytic leukaemia: revised guidelines for diagnosis and treatment. Blood 1996; 87:4990–7. 273 Moreno C, Villamor N, Colomer D, et al. Clinical significance of minimal residual disease, as assessed by different techniques, after stem cell transplantation for chronic lymphocytic leukemia. Blood 2006; 106:4563–9. 274 CLL Trialists’ Collaborative Group. Chemotherapeutic options in chronic lymphocytic leukaemia: a meta-analysis of the randomized trials. J Natl Cancer Inst 1999; 91:1848–54. 275 Binet J-L, Caligaris-Cappio, Catovsky D, et al. Perspectives on the use of new diagnostic tools in the treatment of chronic lymphocytic leukemia. Blood 2007; 107:859–61. 276 Keating MJ. Fludarabine phosphate in the treatment of chronic lymphocytic leukemia. Semin Oncol 1990; 17:49–62. 277 Rai KR, Peterson B, Elias I, et al. A randomised comparison of fludarabine and chlorambucil for patients with previously untreated chronic lymphocytic leukemia. A CALBG, SWOG, CTG/NCI-C and ECOG Intergroup study. Blood 1996; 88 (Suppl 1): 141a. 278 Rai KR, Peterson B, Appelbaum FR, et al. Fludarabine compared with chorambucil as primary therapy for chronic lymphocytic leukemia. N Engl J Med 2000; 343:1750–7. 279 The French Cooperative Group on CLL, Johnson S, Smith AG, et al. Multicentre prospective randomized trial of fludarabine versus cyclophosphamide, doxorubicin, and prednislone (CAP) for treatment of advanced stage chronic lymphocytic leukaemia. Lancet 1996; 347:1432–8 280 Catovsky D, Richards S, Hillmen P, et al. Early results from LRF CLL4: a UK multicentre randomized trial. Blood 2005; 106:212a, abstract 716. 281 Byrd JC, Peterson B, Piro L, et al. A phase II study of cladribine treatment for fludarabine refractory B cell chronic lymphocytic leukemia: results from CALGB study 9211. Leukemia 2003; 17:323–77. 282 Lindelmalm S, Liliemark J, Gruber A, et al. Comparison of cytotxicity of 2-chloro-2’-arabino-fluoro-2’-deoxyadenosine (clofarabine) with cladrabine in mononuclear cells from patients with acute myeloid and chronic lymphocytic leukemia. Haematologica 2003; 88:324–32. 283 Catovsky D, Richards S. Incidence of hemolytic anemia after chemotherapy in the CLL4 trial: a possible protective role for fludarabine plus cyclophosphamide. Blood 2004; 101:3413–3135. 284 Byrd JC, Peterson BL, Morrison VA, et al. Randomized phase 2 study of fludarabine with concurrent versus sequential treatment with rituximab in symptomatic, untreated patients with B-cell chronic lymphocytic leukemia: results
1024 Leukaemias
285
286
287
288
289
290
291
292
293 294
295
296
297
298
from Cancer and Leukemia Group B 9712 (CALGB 9712). Blood 2003; 1:6–14. Byrd JC, Rai K, Peterson BL, et al. Addition of rituximab to fludarabine may prolong progressive-free survival and overall survival in patients with previously untreated chronic lymphocytic leukemia: an updated retrospective comparative analysis of CALGB 9712 and CALGB 9011. Blood 2005; 105:49–53. Keating M, Coutre S, Rai KR, et al. Management guidelines for use of alemtuzumab in B-cell chronic lymphocytic leukemia. Clin Lymphoma 2004; 4:220–7. Rai KR, Freter CE, Mercier RJ, et al. Alemtuzumab in previously treated chronic lymphocytic leukemia patients who also had received fludarabine. J Clin Oncol 2002; 20:3891–7. Moreton P, Kennedy B, Lucas G, et al. Eradication of minimal residual disease in B-cell chronic lymphocytic leukemia after alemtuzumab therapy is associated with prolonged survival. J Clin Oncol 2005; 23:4079–88. Nabhan C, Coutre S, Hillmen P. Minimal residual disease in chronic lymphocytic leukaemia: is it ready for primetime? Br J Haematol 2007; 3:379–92. Elter T, Borchmann P, Schulz H, et al. 2005 Fludarabine in combination with alemtuzumab is effective and feasible in patients with relapsed or refractory B-cell chronic lymphocytic leukemia: results of a phase II trial. J Clin Oncol on-line publication ahead of edition September 6. Faderl S, Thomas DA, O’Brien S, et al. Experience with alemtuzumab plus rituximab in patients with relapsed and refractory lymphoid malignancies. Blood 2003; 101:3413–15. Hale G, Rebello P, Brettman LR, et al. Blood concentrations of alemtuzumab and antiglobulin responses in patients with chronic lymphocytic leukemia following intravenous or subcutaneous routes of administration. Blood 2004; 104:948–55. List A, Kurtin S, Roe DJ, et al. Efficacy of lenalidomide in myelodysplastic syndromes. N Engl J Med 2005; 352:549–57. Chanan-Khan A, Miller KC, Takeshita K, et al. Results of a phase I clinical trial of thalidomide in combination with fludarabine as initial therapy for patients with treatmentrequiring chronic lymphocytic leukemia (CLL). Blood 2005; 106:3348–52. Chanan-Khan A, Miller KC, Musial L, et al. Clinical efficacy of lenalidomide in patients with relapsed or refractory chronic lymphocytic leukemia: results of a phase II study. J Clin Oncol 2006; 34:5343–9. Frankel AE, Fleming DR, Hall DR, et al. A phase II study of DT fusion protein denileukin diftitox in patients with fludarabine-refractory chronic lymphocytic leukemia. Clin Cancer Res 2003; 9:3555–61. O’Brien S, Kipps T, Faderl S, et al. A phase I trial of the small molecule pan Bcl-2 family inhibitor GX15-070 administered intravenously (IV) every 3 weeks to patients previously treated chronic lymphocytic leukemia (CLL). Blood 2005; 106:135a, abstract 446. Kater AP, van Oers MHJ, Kipps TJ. Cellular immune therapy of chronic lymphocytic leukemia. Blood prepublished July 16, 2007; DOI 10.1182/blood-2007-01-068932.
299 Byrd JP, Lin TS, Dalton JT, et al. Flavopiridol administered using a pharmacologically derived schedule is associated with marked clinical efficacy in refractory, genetically high-risk chronic lymphocytic leukemia. Blood 2007; 109:399–404. 300 Orsini E, Pasquale A, Maggio R, et al. Phenotypic and functional characterization of monocyte-derived dendritic cells in chronic lymphocytic leukemia patients: Influence of neoplastic CD19 cells in vivo and in vitro. Br J Haematol 2004; 125:720–8. 301 Kato K, Cantwell MJ, Sharma S, et al. Gene transfer of CD40-ligand induces autologous immune recognition of chronic lymphocytic leukemia B cell. J Clin Invest 1998; 101:1133–41. 302 Khouri I, Przepiorka D, Besien K, et al. Allogeneic blood or marrow transplantation for chronic lymphocytic leukaemia: timing of transplantation and potential effect of fludarabine on acute graft-versus-host disease. Br J Haematol 1997; 97:466–73. 303 Milligan DW, Fernandes S, Dasgupta R, et al. Results of the MRC pilot study show autografting for younger patients with chronic lymphocytic leukemia is safe and achieves a high percentage of molecular responses. Blood 2005; 105:397–404. 304 Gribben JG, Zahrieh D, Stephans K, et al. Autologous and allogeneic stem cell transplantation for poor risk chronic lymphocytic leukemia. Blood 2005; 106:4389–96. 305 Michallet M, Archimbaud E, Bandini G, et al. HLA-identical sibling bone marrow transplantation in younger patients with chronic lymphocytic leukemia. European Group for Blood and Marrow Transplantation and the International Bone Marrow Transplant Registry. Ann Intern Med 1996; 124:311–15. 306 Khouri IF, Keating MJ, Saliba RM, et al. Long-term follow-up of patients with CLL treated with allogeneic hematopoietic transplantation. Cytotherapy 2002; 4:217–21. new 2007 307 Esteve J, Villamor N, Colomer D, et al. Stem cell transplantation for chronic lymphocytic leukemia: different outcome after autologous and allogeneic transplantation and correlation with minimal residual disease status. Leukemia 2001; 15:445–51. 308 Khouri IF, Keating MJ, Vriesendorp HM, et al. Autologous and allogeneic bone marrow transplantation for chronic lymphocytic leukaemia: preliminary results. J Clin Oncol 1994;12:748–58. 309 Ritgen M, Stilgenbauer S, von Neuhoff N, et al. Graftversus-leukemia activity may overcome therapeutic resistance of chronic lymphocytic leukemia with unmutated immunoglobulin variable heavy-chain gene status: implications if minimal residual disease measurement with quantitavtive PCR. Blood 2004; 104:2600–2. 310 Dreger R, Brand R, Milligan D, et al. Reduced-intensity conditioning lowers treatment-related mortality of allogeneic stem cell transplantation for chronic lymphocytic leukemia: a population-matched analysis. Leukemia 2005;19:1029–33. 311 Delgado J, Thomson K, Russell N, et al. Results of alemtuzumab-based reduced-intensity allogeneic transplantation for chronic lymphocytic leukaemia: a British Society of Blood and Marrow Transplantation Study. Blood 2006; 107:1724–30.
References 1025
312 Sorror M, Sandmaier BM, Storer B, et al. Impacts of comorbidities on outcome of patients diagnosed with B-cell malignancies and treated with allogeneic hematopoietic stem cell transplantation using non-myeloablative vs. myeloablative conditioning. Blood 2006; 108:550. 313 Melo JV, Brito-Babapulle V, Pombo de Oliverial MS, et al. The relationship between chronic lymphocytic leukemia and prolymphocytic leukemia. In: Gale RP, Rai KR, eds. Chronic Lymphocytic leukemia 1987. Recent progress and future direction New York: Alan R Liss. 205–14. 314 Mulligan S, Matutes E, Deardren C, et al. Splenic lymphoma with villous lymphocytes: natural history and response to therapy in 50 cases. Br J Haemotol 1991; 78:206–9. 315 Matutes E, Morilla R, Owusu-Ankomah K, et al. The immunophenotype of hairy cell leukemia (HCL). Proposal for a scoring system to distinguish HCL from B-cell disorders with hairy or villous lymphocytes. Leuk Lymphoma 1994; 14(Suppl 1):57–61. 316 Goodman GR, Burian C, Koziol JA, Saven A. Extended followup of patients with hairy cell leukaemia after treatment with cladrabine. J Clin Oncol 2003; 21:891–96. 317 Ravandi F, Jorgensen J, O’Brien S, et al. Eradication of minimal residual disease (MRD) in hairy cell leukaemia (HCL). Blood 2006; 107:4658–62. 318 Wong-Staal F, Gallo RC. The family of human T-lymphotropic leukemia/lymphoma virus: HTLV-I as the cause of adult T cell leukemia and HTLV-III as the cause of acquired immuno deficiency syndrome. Blood 1985; 65:253–63. 319 Koeffler HP, Golde DW. Human preleukemia. Ann Intern Med 1980; 93:347–53. 320 Pederson-Bjergaard J, Rowley JD. The balanced and the unbalanced chromosome aberrations of acute myeloid leukemia may develop in different ways and may contribute differently to malignant transformation. Blood 1994; 83:2780–6. 321 Davis RE, Greenberg PL. Bcl-2 expression by myeloid precursors in myeloid precursors in myelodysplastic syndromes: relation to disease progression. Leuk Res 1998; 22:767–77. 322 Ebert B, Pretz J, Bosco J, et al. Identification of RSP14 as the 5q-syndrome gene by RNA Interference screen. Blood 2007; 110:abstract 1. 323 Levine RL, Wadleigh M, Cools J, et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell 2005; 7:387–97. 324 Goldman JM. A unifying mutation in chronic myeloproliferative disorders. N Engl J Med 2005, 352:1744–6. 325 Hofmann WK, Ottomann OG, Ganser A, et al. Myelodysplastic syndromes: clinical features. Semin Hematol 1996; 33:177–85. 326 Block M, Jacobson LO, Bethard WF. Preleukemic acute human leukemia. JAMA 1953; 152:361–5. 327 Greenberg P, Cox C, LeBeau MM, et al. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 1997; 89:2079–88. [Erratum, Blood 1998; 91:1100.] 328 Estey E, Thall P, Beran M, et al. Effect of diagnosis (refractory anemia with excess blasts, refractory anemia with excess blasts in transformation, or acute myeloid
329 330 331
332
333 334
335
336
337
338
339
340
341 342
343
344
345
leukemia [AML]) on outcome of AML-type chemotherapy. Blood 1997; 90:2969–77. Hoffbrand VA. Deferiprone therapy for transfusional iron overload. Best Pract Res Clin Haematol 2005; 18:299–317. Matthews DC. Immunotherapy in acute myelogenous leukemia and myelodysplastic syndrome. Leukemia 1998; 100:304–9. Saunthararajah Y, Nakamura R, Nam JM, et al. HLA-DR15 (DR2) is overrepresented in myelodysplastic syndrome and aplastic anemia and predicts a response to immunosuppression in myelodysplastic syndrome. Blood 2002; 100:1570–4. Shimamoto T, Tohyama K, Okamoto T, et al. Cyclosporine A therapy for patients with myelodysplastic syndrome: multicentre pilot studies in Japan. Leuk Res 2003; 27:783–8. Morosetti R, Koeffler HP. Differentiation therapy in myelodysplastic syndromes. Semin Hematol 1996; 33:236–45. Italian Cooperative Study Group for rHuEpo in Myelosysplastic Syndromes. A randomized double-blind placebo-contolled study with subcutaneous recombinant human erythropoietin in patients with low-risk myelodysplastic syndromes. Br J Haematol 1998; 103:1070–4. Jadersten M, Montgomery SM, Dybedal I, et al. Long-term outcome of treatment of anemia in MDS with erythropoietin and G-CSF. Blood 2005; 106:803–11. Musto P, Lanza F, Balleari E, et al. Darbepoetin alpha for the treatment of anaemia in low-intermediate risk myelodysplastic syndromes. Br J Haematol 2005;128:204–9. Vadhan-Raj S, Keating M, LeMaistre A, et al. Effects of recombinant human granulocyte-macrophage colony stimulating factor in patients with myelodysplastic syndrome. Blood 1987; 73:31–7. Musto P, falcone A, sanpaolo G, et al. Thalidomide abolishes transfusion-dependence in selected patients with myelodysplastic syndromes. Haematologica 2002; 87: 884–6. List A, Dewald G, Bennett J, et al. Lenalidomide in the myelodysplastic syndrome with chromosome 5q deletion. N Engl J Med 2006; 355:1456–65. National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines in Oncology v4.2006. Myelodysplastic syndromes. 2006 National Comprehensive Cancer Network. Available at http://www.nccn.org. Accessed August 8, 2006. Issa JP. CpG island methylator phenotype in cancer. Nat Rev Cancer 2004; 4:988–93. Silverman JR, Demakos EP, Peterson BL, et al. Randomized controlled trial of azacytidine in patients with the myelodysplastic syndrome: a study of the cancer and leukaemia group B. J Clin Oncol 2002; 20:2429–40. Saba HI, Rosenfeld CS, Issa J, et al. Clinical benefit and survival endpoints from a phase III trial comparing decitabine (DAC) vs. supportive care (SC) in patients with advanced myelodysplastic syndrome. Proc Am Soc Clin Oncol 2005; 23: abstact. Kuendgen A, Strupp C, Aivado M, et al. Treatment of myelodysplastic syndrome with valproic acid alone or in combination with all-trans retinoic acid. Blood 2004; 104:1266–9. Kurzrock R, Albitar M, Cortes J, et al. Phase II study of R115777, a farnesyl transferase inhibitor, in myelodysplastic syndrome. J Clin Oncol 2004; 22:1287–92.
1026 Leukaemias
346 Schiller GJ, Slack J, Hainsworth JD, et al. Phase II multicenter study of arsenic trioxide in patients with myelodysplastic syndromes. J Clin Oncol 2006; 24: 2456–64. 347 Cools J, Stover EH, Gilliland DG. Detection of the FIP1L1PDGFRA fusion in idiopathic hypereosinophitic syndrome and chronic eosiniphilic leukemia. Methods Mol Med 2006; 125:177–187. 348 Cutler CS, Lee SJ, Greenberg P, et al. A decision analysis of allogeneic bone marrow transplantation for the myelodysplastic syndromes: delayed transplantation for lowrisk myelodysplastic syndromes: delayed transplantation for low-risk myelodysplasia is associated with improved outcome. Blood 2004; 104:579–85. 349 Yokoub-Agha I, de La Salmoniére P, Ribaud P, et al. Allogeneic bone marrow transplantation for therapy-related myelosysplastic syndrome and acute myeloid leukemia: a long-term study of 70 patients – report of the French Society of Bone Marrow Transplantation. J Clin Oncol 2000; 18:963–71. 350 Scott BL, Storer B, Loken M, et al. Pretransplantation induction chemotherapy and posttransplantation relapse in patients with advanced myelodysplastic syndrome. Biol Blood Marrow Transplant 2005; 11:65–73. 351 Appelbaum FR, Anderson J. Allogeneic bone marrow transplantation for myelodysplastic syndromes: outcome analysis according to IPSS score. Leukemia 1998; 12(Suppl 1):S25–S29.
352 Anderson JE, Anasetti C, Appelbaum FR. et al. Unrelated donor marrow transplantation for myelodysplasia (MDS) and MDS-related acute myeloid leukaemia. Br J Haematol 1996; 93:59–67. 353 Chang CK, Storer BE, Scott BL, et al. Hematopoietic cell transplantation in patients with myelodysplastic syndrome or acute myeloid leukemia arising from myelodysplastic syndrome: similar outcomes in patients with de novo disease and disease following prior therapy or antecedent hematologic disorders. Blood 2007; 110:1379–87. 354 Ho AY, Pagliuca A, Kenyon M, et al. Reduced-intensity allogeneic hematopoietic stem cell transplantation for myelodysplastic syndrome and acute myeloid leukemia with multilineage dysplasia using fludarabine, buslphan and alemtuzumab (FBC) conditioning. Blood 2004; 104:1616–23. 355 Alyea EP, Kim HT, Ho V, et al. Comparative outcome of nonmyeloablative allogeneic hematopoietic cell transplantation for patients older than 50 years of age. Blood 2005; 105:1810–14. 356 Brown JR, Yeckes H, Friedberg JW, et al. Increasing incidence of late second malignancies after conditioning with cyclophosphamide and total-body irradiation and autologous bone marrow transplantation for non-Hodgkin’s lymphoma. J Clin Oncol 2005; 23:2208–14. 357 De Witte T, Van Biezen A, Hermans J, et al. Autologous bone marrow transplantation for patients with myelodysplastic syndrome (MDS) or acute myeloid leukemia following MDS. Blood 1997; 90:3853–7.
42 Hodgkin’s lymphoma BEATE KLIMM AND ANDREAS ENGERT
Introduction Epidemiology and aetiology Pathophysiology Diagnosis, staging and choice of treatment First-line treatment Relapsed and refractory Hodgkin’s lymphoma
1027 1028 1028 1029 1031 1036
INTRODUCTION Hodgkin’s lymphoma (HL) is one of the few curable cancers in adults. Thomas Hodgkin first described the disease in 1832 in his historic paper entitled On Some Morbid Appearances of the Absorbent Glands and Spleen.1 About 70 years later Carl Sternberg (1898) and Dorothy Reed (1902) contributed the first definitive microscopic descriptions of the pathognomonic Hodgkin and Reed–Sternberg (H-RS) cells.2,3 At that time Dorothy Reed wrote ‘the treatment for this disease is dismal. All patients die within 3–4 years. Even if you resect the tumor totally, it will recur and grow even faster than before…’. Hodgkin had already assumed that it was an autonomous lymphatic process rather than an inflammatory condition or an infectious disease like tuberculosis. Despite fragments of evidence for the malignant nature of HL for a very long period of time, the malignant clonal origin of H-RS cells from germinal centre-derived B-lymphocytes was demonstrated only very recently.4,5 Since these first descriptions, therapeutic strategies for HL have developed remarkably from surgery, herbs and arsenic acid to sophisticated stage- and risk-adapted treatment regimens including modern polychemotherapy and radiotherapy. To date, about 80 per cent of patients achieve long-term disease-free survival, rendering this entity one of the most curable human cancers. It is therefore of pivotal importance to maintain the high standard of cure rates reached over all stages and at the same time to reduce the toxicity of treatment.
Special considerations Long-term management Experimental therapies Acknowledgements References
1038 1039 1039 1040 1040
Current first-line treatment strategies aim at further improving outcome and thereby trying to minimize or prevent therapy-induced complications, such as infertility, cardiopulmonary toxicity and secondary malignancies. Ongoing trials for patients in early stages aim at finding the minimal necessary curative therapy with the least acute and long-term toxicity. In the last years, there has been a trend to combine chemo- and radiotherapy. Recent studies have predominantly investigated lower radiation doses and smaller radiation fields and a possible reduction of chemotherapy in terms of drugs or cycles given. For patients in advanced stages, new schedules of established drug combinations with higher dose density and intensity have been developed and are currently being evaluated in clinical trials. Furthermore, ongoing studies try to utilize positron emission tomography (PET) with 218F, Fluoro-2 deoxy-D-glucose (FDG) as a new diagnostic tool to possibly detect early response during chemotherapy or a satisfactory response after chemotherapy, eventually rendering consolidating radiation unnecessary. Approaches for relapsed HL consist of radiotherapy, chemotherapy and high-dose chemotherapy followed by autologous stem cell transplantation. In recent years, the introduction of effective salvage high-dose therapy and a better understanding of prognostic factors have remarkably improved the management of relapsed HL. For multiply pre-treated patients, radioimmunoconjugates, monoclonal antibodies and more recently small molecules targeting signal transcription pathways have demonstrated some clinical efficacy; these efforts, however, are still experimental.
1028 Hodgkin’s lymphoma
EPIDEMIOLOGY AND AETIOLOGY Incidence, age and histological distribution In Europe and the United States, the annual incidence of HL is about 2–3 per 100 000 persons at risk, and has remained remarkably constant over the last decades.6 As a result of clinical progress in recent years, the mortality rate has dropped, particularly in the 1990s, from earlier rates above 2 to a current mortality rate of about 0.5 per 100 000.7 Slightly more men than women develop HL with a ratio of 1.4:1. Four of five males and three of four females develop HL prior to the age of 60, which is very early compared to most other malignancies.8 In industrialized countries, the age at onset has historically shown two peaks, one in the third decade and a second peak for patients older than 50 years. However, in more recent analyses the second peak seems to disappear because non-Hodgkin’s lymphomas were misclassified mainly as lymphocyte-depleted HL in the past.9 There is a noteworthy difference in the onset of HL between developing and industrialized countries: in developing countries, the disorder usually appears during childhood and the incidence decreases with age, whereas in industrialized countries the first peak is seen in young adulthood. Furthermore, in economically developed countries the early occurrence of HL is often related to better maternal education, early birth order, low number of siblings and playmates and single family dwellings.10,11 The incidence of HL by age also depends upon histological subtype. Among young adults, the most common subtype is nodular sclerosing (NS) HL, occurring at a higher frequency than the mixed cellularity (MC) subtype. The frequency of MC increases with age, while NS subtypes reach a plateau in the group 30 years of age. According to the World Health Organization classification (WHO), the subtypes of lymphocyte-rich classical HL (LRCHL), nodular lymphocyte-predominant HL (LPHL), and lymphocyte-depleted HL (LDHL) are less commonly diagnosed (Table 42.1).12,13
Role of Epstein – Barr virus and genetic factors An involvement of viral infections (e.g. EBV) in the pathogenesis of HL was suggested by several studies. Patients with a medical history of EBV-related mononucleosis have a two- to threefold increased risk of developing HL.14 In about 50 per cent of the cases of classical HL in Western countries, EBV DNA is present in the H-RS cells, predominantly in the mixed cellularity subtype.15 In contrast, patients with low socio-economic status as well as patients from developing countries show EBV-positive H-RS cells in about 90 per cent of cases.16–18 However, since EBV is not present in the tumour cells of a substantial proportion of patients in the Western world, other viruses might be involved in the transformation
Table 42.1 World Health Organization classification of Hodgkin’s lymphoma. Extracted and modified from Harris et al, 199913 Classification Classical Hodgkin’s lymphoma Nodular sclerosis Hodgkin’s lymphoma (grades 1 and 2) Mixed cellularity Hodgkin’s lymphoma Lymphocyte-rich classical Hodgkin’s lymphoma Lymphocyte-depleted Hodgkin’s lymphoma Nodular lymphocyte-predominant Hodgkin’s lymphoma
Frequency
60–70% 20–30% 3–5% 0.8–1% 3–5%
process of HL in these cases. To date, the role of other viruses for the pathogenesis of HL is uncertain and other transforming mechanisms must also be taken into account. Genetic components seem to contribute to the appearance of this malignancy. Family members of patients affected by HL are at a three- to ninefold increased risk of developing the same disease.19,20 Furthermore, analysis of monozygotic twin pairs and the remarkable proportion in which both twins are affected strongly supports the idea of HL as a genetically imbalanced disorder.21 However, no specific mechanism of inheritance or evidence for a genetic translocation unique to cases of familial HL has been identified so far and familial HL only appears to play a role in a small subset of HL patients.
PATHOPHYSIOLOGY For a long time HL was considered an infectious disease, as indicated by its former designation ‘lymphogranulomatosis’. The giant mono- and multinucleated H-RS cells typically account for 1 per cent of the affected tissue in classical HL, which made systematic analyses difficult in the past. The detection of their malignant clonal origin in microdissected cells by polymerase chain reaction was demonstrated only recently.4 H-RS cells are derived from germinal centre B-cells in more than 90 per cent of cases;5 however, in a small group of patients the H-RS cells exhibit T-cell characteristics.22 Hodgkin’s lymphoma can basically be distinguished from other types of malignant lymphoma by the presence of characteristic types of tumour cells, called Hodgkin and Reed–Sternberg (H-RS) cells, in a background of non-neoplastic cells such as lymphocytes, histiocytes, neutrophils, eosinophils and monocytes. The histological subclassification of HL (see Table 42.1) considers both the morphology and immunophenotype of the H-RS cells and the composition of the cellular background. The WHO differentiates between the classical form of HL with CD30-positive H-RS cells and the nodular lymphocyte-predominant form of HL (LPHL) with CD20-positive lymphocytic and histiocytic
Diagnosis, staging and choice of treatment 1029
(L&H) cells. In classical HL, immunophenotyping demonstrated that H-RS cells stain positive for CD15 in about 80 per cent and for CD30 in about 90 per cent of cases. The activation of B-cell antigens has only been reported in few cases.23 In contrast, in NLPHL, the L&H cells are scattered in the nodular structures and are usually CD45 positive. They express B-cell associated antigens such as CD20 in 98 per cent of the cases, but also express CD19, CD22, CD79a, and EMA; however they lack CD15 and CD30 expression. Despite enormous efforts and progress in basic research, many key questions concerning transforming events and pathways, oncogenic viruses and the exact mechanism(s) by which H-RS cells proliferate and resist apoptosis in the germinal centre still remain unanswered. Some reports suggest that NFκB is a central effector of malignant transformation in classical HL by down-regulation of an anti-apoptotic signalling network.24 Also, LMP1 as an EBV-encoded gene may induce tumorigenesis by triggering NFκB-activation.25
DIAGNOSIS, STAGING AND CHOICE OF TREATMENT Clinical presentation Typically, swollen indolent lymph nodes localized in the cervical or supraclavicular region (60–70 per cent) are noticed (Fig. 42.1), but axillary or inguinal lymph nodes can also be affected. Almost two thirds of patients with newly diagnosed classical HL have radiographic evidence of intra-thoracic involvement (Fig. 42.2). Symptoms caused by a large mediastinal mass include a feeling of pressure, cough, venous congestion, or even dyspnea owing to tracheal compression or pericardial or pleural effusions. Hepato- or splenomegaly might indicate hepatic or splenic involvement, but affected organs can also present with normal size. In advanced stages, adjacent regions such as lung, pericardium, chest wall or bone are affected and patients sometimes suffer from osseous pain, neurological or endocrinological symptoms. Compared with NHLs, bulky infradiaphragmatic lesions with obstructive symptoms are rare in HL. Bone marrow involvement occurs in less than 10 per cent of newly diagnosed patients. Details of organ involvement at diagnosis are given in Table 42.2. About 40 per cent of patients, especially those with initial abdominal involvement or advanced-stage disease, report systemic symptoms, so called ‘B-symptoms’ which are defined as fever 38°C, drenching night sweats, or weight loss 10 per cent within the previous 6 months (not from other causes). Other symptoms comprise pain at the site of nodal involvement shortly after drinking alcohol, pruritus or fatigue. Compared with classical HL, nodular LPHL usually begins as a localized, slow-growing and rather benign entity with participation of only one peripheral nodal region, mostly a cervical or axillary or inguinal lymph node.
Figure 42.1
Clinical presentation of HL
Initial diagnostics An excisional biopsy of a suspicious lymph node should be performed to confirm the initial diagnosis of Hodgkin’s lymphoma. Enough material should be extracted, however ‘spacious debulking’ to reduce tumour mass does not improve the prognosis of HL. The assessment of bone marrow is important for disease staging and for evaluation of normal bone marrow function prior to therapy. Accurate staging procedures and risk factor assessment are indispensable for adequate allocation to treatment groups. Clinical staging (CS) methods have become less invasive in recent years. They usually include chest X-ray, abdominal sonography, CT scans of the neck, thorax, abdomen and pelvis, bone marrow biopsy and bone marrow or skeletal radionuclide imaging. In some cases, additional procedures including magnetic resonance imaging (MRI), PET or a liver biopsy may be indicated. Pathological staging (PS) procedures such as laparotomy or splenectomy to assess occult infradiaphragmatic disease are no longer used routinely. They are associated with possible side effects such as the OPSI (overwhelming post-splenectomy infection) syndrome. Furthermore, better imaging techniques and the
1030 Hodgkin’s lymphoma
Differential diagnosis Differential diagnosis for this disease includes all types of benign or malignant lymph node swelling owing to infectious or reactive disease or to other types of lymphoma or solid tumours. Infectious lymphadenopathy can be of bacterial (e.g. purulent or tuberculous), viral (e.g. EBV, HIV, CMV), fungal (e.g. coccidioidomycosis) or parasitic (e.g. toxoplasmosis) origin. Reactive lymphadenopathy can be associated with sarcoidosis as well as other diseases of the soft tissues or the skin or can be drug induced (e.g. diphenylhydantoin). Malignant causes include metastases of other solid tumours, leukaemias or NHL. The differential diagnosis between certain types of HL and NHL can be very challenging and should be performed by an experienced haematopathologist. Occasionally, a composite lymphoma, consisting of HL and NHL parts, is diagnosed.
Staging classification
Figure 42.2 Intrathoracia HL Table 42.2 Anatomic sites of disease involved in untreated patients with Hodgkin’s lymphoma; modified from: Gupta RK, Gospodarowicz MK, Lister TA. Clinical evaluation and staging of Hodgkin’s disease. In: Mauch PM, Armitage JO, Diehl V, Hoppe RT, Weiss LM (eds) Hodgkin’s Disease, Chapter 15, Lippincott Williams & Wilkins, Philadelphia, 1999. Anatomic site Waldeyer’s ring Cervical nodes Axillary nodes Mediastinum Hilar nodes Para-aortic nodes Iliac nodes Mesenteric nodes Inguinal nodes Spleen Liver Bone marrow Total extranodal
Involvement (%) 1–2 60–70 30–35 50–60 15–35 30–40 15–20 1–4 8–15 30–35 2–6 1–4 10–15
introduction of systemic chemotherapy in the majority of patients with early-stage disease have restricted invasive measures to a very few patients for whom the initial diagnostics give conflicting or unresolved results.
Hodgkin’s lymphoma patients are usually treated according to stage and risk factors. The histological subtype – except the lymphocyte-predominant type – does not influence the treatment decision. The stage of HL at diagnosis is determined according to the Ann Arbor classification.26 The presence (B) or absence (A) of systemic symptoms further characterizes severity of disease. Clinical, biological and serological (age, performance status, high ESR) risk factors further influence the choice of treatment. The Cotswolds classification is a modification of the Ann Arbor Classification using information from staging and treatment over 20 years. This classification was proposed in 1989 during a meeting held in the Cotswolds, UK.27 Information about prognostic factors such as mediastinal mass, other bulky nodal disease, extra-nodal extension of disease, and the extent of subdiaphragmatic disease is included in this classification (Table 42.3).
Prognostic factors and choice of treatment Prognostic factors define the likely outcome of the disease for an individual patient at diagnosis. Such analysis may be used to inform a patient, or in the context of clinical trials to describe the study population or adjust the data analysis. For the clinician, the most important use of prognostic factors is to select appropriate treatment strategies. There are two major determinants for dividing HL patients according to a risk- or prognosis-adapted therapeutic approach: stage, and systemic symptoms, including fever, night sweats and weight loss. A third clinically relevant factor meets trans-Atlantic acceptance: massive local tumour burden, e.g. bulk 10 cm or large mediastinal mass more than one third of the thoracic diameter. Generally, patients with clinical stage I and II without risk factors are allocated to the early-stage favourable group,
First-line treatment 1031
Table 42.3 The Cotswolds Staging Classification for Hodgkin’s lymphoma; extracted and modified from Lister et al. 198927 Cotswold staging classification Stage I
Involvement of a single lymph node region or lymphoid structure (e.g., spleen, thymus, Waldeyer’s ring) or involvement of a single extra-lymphatic site (IE)
Stage II
Involvement of two or more lymph node regions on the same side of the diaphragm; localized contiguous involvement of only one extra-nodal organ or site and lymph node region(s) on the same side of the diaphragm (IIE). The number of anatomic regions involved should be indicated by a subscript (e.g. II3)
Stage III
Involvement of lymph node regions on both sides of the diaphragm (III), which may also be accompanied by involvement of the spleen (IIIS) or by localized contiguous involvement of only one extra-nodal organ site (IIIE) or both (IIISE) With or without involvement of splenic, hilar, celiac or portal nodes With involvement of para-aortic, iliac and mesenteric nodes
III1 III2 Stage IV
Diffuse or disseminated involvement of one or more extra-nodal organs or tissues, with or without associated lymph node involvement
Designations applicable to any disease stage A: No symptoms B: Fever (temperature, 38°C), drenching night sweats, unexplained loss of 10 per cent of body weight within the preceding 6 months X: Bulky disease (a widening of the mediastinum by more than one third of the presence of a nodal mass with a maximal dimension greater than 10 cm) E: Involvement of a single extra-nodal site that is contiguous or proximal to the known nodal site CS
Clinical stage
PS:
Pathological stage (as determined by laparotomy)
those with risk factors to the early-stage unfavourable group. Patients in stage III and IV disease are assigned to the advanced-stage risk group. Besides stage and B-symptoms, most groups have implemented larger tumour burden as a relevant prognostic factor including bulky disease 10 cm or a large mediastinal mass more than one third of thoracic diameter. However, there are still small differences in the definition of risk factors used and the classification of certain subgroups of HL patients among the different study groups in Europe and the USA. In the USA, patients are usually allocated to either early or advanced stages. This results in more patients, even with little tumour burden, being included in the advanced stage group. These patients then receive more therapy than in other groups, which must be considered when comparing the data (Table 42.4).28 In an attempt to more precisely define the risk for patients with advanced HL, a variety of clinical and laboratory parameters were analysed to construct a prognostic index. The International Prognostic Score (IPS) consists of seven factors that were significantly related to an unfavourable prognosis when present at initial diagnosis of HL: serum albumin 4 g/dL, haemoglobin 10.5 g/dL, male sex, age 45 years, stage IV disease, leukocytosis 15 000/mm3, lymphocytopenia 600/mm3 and/or 8 per cent of white cells.29
FIRST-LINE TREATMENT Early-stage favourable Hodgkin’s lymphoma In the treatment of early-stage disease, extended field radiotherapy (EF-RT) has been considered the standard treatment modality for a long time. The EF strategy delivers radiation to all initially involved and adjacent lymph node regions, leading to large irradiation fields compared with involved field radiotherapy (IF-RT), which is restricted to initially involved lymph node regions only. Together with the successful introduction of MOPP30 and ABVD31 chemotherapy for advanced stages in the 1980s, the paradigm shift from radiation alone to additional chemotherapy in early stages was accelerated by realization of the long-term toxicity and mortality related to large radiation fields and radiotherapy doses. Longer follow-up of patients who underwent EF-RT disclosed severe late effects as competing causes of death, including heart failure,32 pulmonary dysfunction33 and secondary malignancies.34–36 Furthermore, although complete remission was generally achieved, there was a high risk of relapse from first-line treatment when EF-RT alone was administered.37 Two different strategies were explored to prevent these relapses: either applying even more intensive radiotherapy or adding
1032 Hodgkin’s lymphoma
Table 42.4 Definition of treatment groups according to the EORTC/GELA, GHSG, and NCIC/ ECOG Treatment group
EORTC/GELA
GHSG
NCIC/ECOG
Early-stage favourable
CS I–II without risk factors (supradiaphragmatic)
CS I–II without risk factors
Standard risk group: favorable CS I–II (without risk factors)
Early-stage unfavourable (intermediate)
CS I–II with 1 risk factors (supradiaphragmatic)
CS I, CSIIA 1 risk factors; CS IIB with C/D but without A/B
Standard risk group: unfavorable CS I–II (at least one risk factor)
Advanced-stage
CS III–IV
CS IIB with A/B; CS III–IV
High risk group: CS I or II with bulky disease; intra-abdominal disease; CS III, IV
Risk factors (RF)
A large mediastinal mass B age 50 years C elevated ESRa D 4 involved regions
A large mediastinal mass B extra-nodal disease C elevated ESRa D 3 involved areas
A 40 years B not NLPHL or NS histology C ESR 50 mm/h D 4 involved nodal regions
GHSG, German Hodgkin Lymphoma Study Group; EORTC, European Organization for Research and treatment of Cancer; GELA, Groupe d’Etude des Lymphomes de l’Adulte; ECOG, Eastern Cooperative Oncology Group; NCIC, National Cancer Institute of Canada. a ESR, erythrocyte sedimentation rate (50 mm/h without or 30 mm/h with B-symptoms).
chemotherapy for early favourable stages to control occult lesions.38 The latter strategy produced better outcomes and at the same time enabled reduction of radiotherapy to the involved fields for this risk group. Most centres and groups in Europe and the USA have now accepted combined modality treatment, consisting of two to four cycles of ABVD, followed by 20–30 Gy IF-RT as the standard of care for early favourable stage disease. Several randomized studies confirmed the superiority of combined modality treatment over radiotherapy alone. Other trials were conducted to investigate and reduce radiation fields and dose and, likewise, to decrease chemotherapy drug combinations and duration of treatment. A selection of recent and ongoing trials is listed in Table 42.5. The Southwest Oncology Group (SWOG) demonstrated that patients treated with combined modality therapy consisting of three cycles of doxorubicin and vinblastine followed by subtotal lymphoid irradiation (STLI) had a markedly superior outcome in terms of freedom from treatment failure (FFTF) than those receiving STLI alone.39 Studies from Milan and Stanford revealed that STLI can be effectively replaced by IF-RT after short-duration chemotherapy such as ABVD or Stanford V (8 weeks), while maintaining progression-free and overall survival.40,41 The European Organization for Research and Treatment of Cancer (EORTC) and the Groupe d’Etude des Lymphomes de l’Adulte (GELA) demonstrated that combined modality with either six courses of EBVP (H7F trial) or three of MOPP/ABV (H8F trial) followed by IF-RT yielded a significantly better event-free survival than was achieved by subtotal nodal irradiation alone.42,43 The aim of their H9F trial was to evaluate a possible dose reduction of radiotherapy (36 Gy or 20 Gy or no radiotherapy) after administering
six cycles of EBVP. However, the arm without radiotherapy was closed prematurely due to a higher relapse rate than expected.44,45 Thus, the use of chemotherapy alone in early stages should still be regarded as experimental. A recent randomized trial for non-bulky, asymptomatic stage I–III HL failed to demonstrate any superiority of ABVD RT over ABVD alone; however, the total number of patients was small and all patients received six cycles of ABVD, even those in clinical stage I and II without risk factors.46 A combined modality approach was also established in the HD7 trial by the German Hodgkin Study Group (GHSG). In this trial two cycles of ABVD plus EF-RT were shown to be superior to EF-RT alone in terms of FFTF. Overall survival (OS) was equal in both arms due to effective salvage treatment.47 Further improvement of treatment, with respect to the excellent long-term survival rates, seems difficult. Thus, strategies to reduce drug dose and toxicity while maintaining efficacy are being pursued. In the subsequent HD10 trial of the GHSG, a possible reduction in chemotherapy from four to two cycles of ABVD and/or IF-RT from 30 to 20 Gy was evaluated. After a median observation time of 2 years, FFTF and OS rates were 96.6 per cent and 98.5 per cent. So far, no significant differences in FFTF and OS have been detected between recipients of four cycles of ABVD and those receiving two cycles of ABVD or between patients receiving different doses of radiotherapy (30 Gy vs. 20 Gy).48 The aim of the ongoing GHSG HD13 trial is to omit the presumably less effective drugs, bleomycin or dacarbazine, from the ABVD regimens. Patients are thus randomized between two cycles of ABVD, ABV, AVD or AV followed by 30 Gy IF-RT. Patients with first diagnosis of the nodular lymphocytepredominant subtype of HL in clinical stage IA without
First-line treatment 1033
Table 42.5 Selected trials for early-stage favourable Hodgkin’s lymphoma Therapy regimen
No. of patients
Outcome
Ref.
SWOG #9133
A. 3 (dox. vinbl.) STLI (36–40 Gy) B. STLI (36–40 Gy)
165 161
94% (FFTF); 98% (SV); 81% (FFTF); 96% (SV); [3 years]
39
Milan 1990–97
A. 4 ABVD STLI B. 4 ABVD IF-RT
65 68
97% (FFP); 93% (SV) 97% (FFP); 93% (SV); [5 years]
40
Stanford V (CSI-IIA)
8 weeks of Stanford V modified IF-RT (30 Gy)
65
94.6% (FFP); 96.6% (SV) [16 months; estimated for 3 years]
41
EORTC/ GELA H7F
A. 6 EBVP IF-RT (36 Gy) B. STNI
168 165
90% (RFS); 98% (SV) 81% (RFS); 95% (SV); [5 years]
42
EORTC/ GELA H8F
A. 3 MOPP/ABV IF-RT (36 Gy) B. STNI
271 272
99% (RFS); 99% (SV) 80% (RFS); 95% (SV); [4 years]
43
EORTC/ GELA H9F
A. 6 EBVP IF-RT (36 Gy) B. 6 EBVP IF-RT (20 Gy) C. 6 EBVP
783
87% (EFS); 98% (SV) 84% (EFS); 98% (SV); [4 years] C closed because of high relapse rate
44, 45
GHSG HD7
A. EF-RT 30 Gy (40 Gy IF) B. 2 ABVD EF-RT 30 Gy (40 Gy IF)
305 312
75% (FFTF); 94% (SV); 91% (FFTF); 94% (SV); [5 years]
47
GHSG HD10
A. 4 ABVD IF-RT (30 Gy) B. 4 ABVD IF-RT (20 Gy) C. 2 ABVD IF-RT (30 Gy) D. 2 ABVD IF-RT (20 Gy)
847
Interim analysis [2 years] all patients: 96.6% (FFTF) 98.5% (SV)
48
GHSG HD13
A. 2 ABVD IF-RT (30 Gy) B. 2 ABV IF-RT (30 Gy) C. 2 AVD IF-RT (30 Gy) D. 2 AV IF-RT (30 Gy)
Trial
Ongoing trial
Abbreviations: SWOG, Southwest Oncology Group; EORTC, European Organization for Research and Treatment of Cancer; GELA, Groupe d’Etude des Lymphomes de l’Adulte; GHSG, German Hodgkin Lymphoma Study Group; EF/IF-RT, extended/involved field radiotherapy; STLI, subtotal lymphoid irradiation; STNI, subtotal nodal irradiation; FFTF, freedom from treatment failure; RFS, relapse-free survival; FFP, freedom from progression; EFS, event-free survival; SV, overall survival.
risk factors are usually not included in the ongoing trials. On the basis of the very favourable prognosis of this subtype, the EORTC and the GHSG currently recommend treatment with 30 Gy IF radiation only. While being less toxic, this strategy seems to produce similar responses for LPHL IA patients compared with combined modality treatment.49 Experimental approaches for these patients focus on the humanized monoclonal anti-CD20 antibody rituximab which has given impressive results in relapsed LPHL50,51 and will shortly be evaluated in a GHSG phase II study in selected IA LPHL patients. Compared with LPHL IA patients, advanced LPHL stages at initial diagnosis have less favourable outcomes and are thus treated according to protocols used for classical HL.
Early-stage unfavourable (intermediate) HL Patients with early-stage unfavourable (intermediate) HL generally qualify for combined modality treatment. However, the ideal chemotherapy and radiation regimens are not yet
clearly defined and there is an ongoing desire to optimize therapy in this risk group. This is being attempted by reducing radiation doses and field sizes in a similar manner to that for early favourable stages. Several trials seem to indicate that the reduction of field size does not compromise the efficacy of treatment. A co-operative study comparing six cycles of MOPP sandwiched around 40 Gy of radiotherapy, applied either to an involved or extended field, indicated no difference in terms of disease-free survival or OS.52 Another trial from Italy comparing STLI with IF-RT after four cycles of ABVD in patients with early favourable (Table 42.5) and unfavourable stages reported a similar treatment outcome in both arms.40 In the H8U trial, the EORTC randomized patients between six cycles of MOPP/ABV 36 Gy IF-RT, four cycles of MOPP/ABV 36 Gy IF-RT, and four cycles MOPP/ ABV STLI. There was no difference between the arms in terms of response rates, failure-free survival, or overall survival.53 The largest trial investigating radiotherapy reduction was conducted by the GHSG: in the HD8 trial, patients were randomized to two alternating cycles of COPP/ABVD
1034 Hodgkin’s lymphoma
Table 42.6 Selected trials for early-stage unfavourable Hodgkin’s lymphoma Trial
Therapy regimen
No. of patients
Outcome
Ref.
EORTC/GELA H8U
A. 6 MOPP/ABV IF-RT (36 Gy) B. 4 MOPP/ABV IF-RT (36 Gy) C. 4 MOPP/ABV STNI
335 333 327
94% (RFS); 90% (SV) 95% (RFS); 95% (SV) 96% (RFS); 93% (SV); [4 years]
53
GHSG HD8
A. 2 COPP ABVD EF RT (30 Gy) Bulk (10 Gy) B. 2 COPP ABVD IF-RT (30 Gy) Bulk (10 Gy)
532 532
86% (FFTF); 91% (SV) 84% (FFTF); 92% (SV); [5 years]
54
SWOG/ECOG #2496
A. 6 ABVD IF-RT (36 Gy) to bulk (5 cm) B. 12 weeks Stanford V IF-RT (36 Gy) to bulk (5 cm)
EORTC/GELA H9U
A. 6 ABVD IF-RT B. 4 ABVD IF-RT C. 4 BEACOPP bas. IF-RT
GHSG HD11
A. 4 ABVD IF-RT (30 Gy) B. 4 ABVD IF-RT (20 Gy) C. 4 BEACOPP bas. IF-RT (30 Gy) D. 4 BEACOPP bas. IF-RT (20 Gy)
GHSG HD14
A. 4 ABVD IF-RT (30 Gy) B. 2 BEACOPP esc. 2 ABVD IF-RT (30 Gy)
Ongoing trial
808
1047
94% (EFS); 96% (SV) 89% (EFS); 95% (SV) 91% (EFS); 93% (SV); [4 years]
45
Interim analysis [2 years] all patients: 97.4% (FFTF) 89.9% (SV)
57
Ongoing trial
Abbreviations: SWOG, Southwest Oncology Group; EORTC, European Organization for Research and Treatment of Cancer; GELA, Groupe d’Etude des Lymphomes de l’Adulte; GHSG, German Hodgkin Lymphoma Study Group; ECOG, Eastern Cooperative Oncology Group; EF/IF-RT, extended/involved-field radiotherapy; STNI, subtotal nodal irradiation; FFTF, freedom from treatment failure; RFS, relapse-free survival; EFS, event-free survival; SV, overall survival.
plus radiotherapy to either extended (arm A) or involved field (arm B). Final results at 5 years did not disclose significant differences between the two arms in terms of FFTF and overall survival, however, more toxicity was reported in the patients who were treated with EF-RT (Table 42.6).54 A NCIC/ECOG trial argues in favour of combined modality treatment over ABVD alone in unfavourable non-bulky stage IA/IIA HL.55 Furthermore, a recent retrospective analysis supports the use of approximately 30 Gy IF-RT after a good response to ABVD,56 a strategy that has been adopted in the ongoing GHSG and EORTC trials. Efforts were also made to improve the efficacy of chemotherapy by altering drugs and schedules as well as the number of cycles. In the past, alternation or hybridization of a MOPP-like regimen with ABVD did not produce better outcomes when compared with ABVD alone. Furthermore, studies in advanced-stage HL indicated that ABVD alone is equally effective and less myelotoxic compared with alternating MOPP/ABVD, and both are superior to MOPP alone.58 Thus, a combined modality treatment consisting of four courses of ABVD followed by 30 Gy IF-RT is considered standard treatment for patients with early-stage unfavourable HL. Despite the excellent initial remission rates obtained with ABVD and radiotherapy, approximately 15 per cent of patients in early unfavourable stages
relapse within 5 years and about another 5 per cent suffer from primary progressive disease. These outcome rates are rather similar to those in patients in advanced stages, when treated with more intense regimens. Thus, study groups are currently evaluating different regimens for the early unfavourable group that were previously pioneered for the treatment of advanced stages (Table 42.6): in their ongoing intergroup trial #2496, the ECOG and SWOG are assessing whether the Stanford V regimen (12 weeks) is superior to six cycles of ABVD. In another approach, four cycles of ABVD and 4 cycles of BEACOPP-baseline were compared by the EORTC-GELA (H9U trial) and by the GHSG (HD11 trial). In addition, two large trials analysed whether four cycles of combined modality treatment are equally effective as six cycles (EORTC: H8U and H9U trial). In the H9U trial that was recently presented by the EORTC and GELA, patients were randomly assigned to six cycles of ABVD or four cycles of ABVD or four cycles of BEACOPP-baseline, followed by 30 Gy of IF-RT in all arms. After a median follow-up of 4 years, no significant difference was observed between the three different treatment arms with respect to event-free survival (EFS) or OS.45 Interim results of the GHSG trial HD11 at two years demonstrate an FFTF of 89.9 per cent and an OS of 97.4 per cent for all patients. At that interim point, there was no
First-line treatment 1035
difference with respect to outcome, either between the ABVD and BEACOPP arms or between the 30 Gy and 20 Gy IF-RT.57 Although it should be taken into account that these are relatively early data, there is, nevertheless, no evidence for changing treatment from four to six cycles of ABVD or for recommending four cycles of BEACOPPbaseline in this group of patients. However, the low FFTF in this risk group led the GHSG to a further intensification of treatment. In the ongoing HD14 trial for early unfavourable stages, the BEACOPP escalated regimen was introduced, which had shown high efficacy in the treatment of advanced HL.59 Patients are currently being randomized to two courses of escalated BEACOPP plus two courses of ABVD or four courses of ABVD followed by 30 Gy IF-RT in both treatment arms.
Advanced-stage Hodgkin’s lymphoma Before the introduction of combination chemotherapy, more than 95 per cent of patients with advanced HL succumbed to their disease within 5 years. Thus, the remission rates in excess of 50 per cent achieved with MOPP were a major breakthrough in oncology. MOPP was successfully used for many years for advanced-stage disease, resulting in long-term remission rates of nearly 50 per cent.30,60 The regimen was then replaced by ABVD after a series of large multicentre trials had proven the superiority of ABVD and alternating MOPP/ABVD over MOPP alone.61,62 Hybrid regimens such as MOPP/ABV were only equally effective when compared with alternating MOPP/ABVD and even rapidly alternating multi-drug regimens such as COPP/ ABV/IMEP did not result in better outcome.63,64 However, more acute toxicity and a higher incidence of leukaemia were reported after MOPP/ABV hybrid as compared with ABVD.61 Thus far, ABVD is regarded as the current standard regimen against which all new combinations have to be tested in the future. However, a long-term follow-up report of 123 patients that were previously treated with ABVD for advanced HL revealed a failure-free survival of only 47 per cent and an overall survival of 59 per cent after 14.1 years.65 Different study groups tried to improve these rates by developing new regimens with additional drugs and by increasing dose intensity and dose density with the support of colony-stimulating factors and modern antibiotics. These new approaches include multi-drug regimens such as Stanford V, MEC, VAPEC-B, ChlVPP/EVA and BEACOPP (Table 42.7).59,66–68 Stanford V seemed to be a promising strategy when used at a single centre. However, a prospectively randomized multi-centre comparison with MEC and ABVD showed that Stanford V was clearly inferior.67 These conflicting results might be partially explained by the use of less radiotherapy in the randomized setting and the better treatment quality of single-centre reports. The GHSG HD9 trial compared COPP/ABVD, BEACOPP-baseline and BEACOPPescalated. Results from 1195 randomized patients showed
a clear superiority of escalated BEACOPP over BEACOPPbaseline and COPP/ABVD at 5 years.59 The follow-up data at 7 years confirm these results: with a median follow-up of 82 months, the FFTF and OS rates were 67 per cent and 79 per cent in the COPP/ABVD group, 75 per cent and 84 per cent in the BEACOPP-baseline group, and 85 per cent and 90 per cent in the BEACOPP-escalated group.69 The subsequent GHSG HD12 trial aimed at de-escalating chemotherapy and radiotherapy by comparing four courses of BEACOPP-escalated with four courses of escalated and four courses of baseline BEACOPP, with or without consolidating radiation to initial bulky and residual disease. In the latest interim analysis of HD12 at a median followup of 30 months, the FFTF was 88 per cent and OS 94 per cent for the whole cohort. So far, there was no significant difference between the different arms.69 In the ongoing HD15 trial, patients are randomized between eight courses of BEACOPP-escalated, six courses of BEACOPPescalated, or eight courses of BEACOPP-14, which is a time-intensified variant of BEACOPP-baseline.70 Additional radiotherapy is only applied to residual lesions 2.5 cm positive by PET. Whether escalated BEACOPP is superior to ABVD alone in a randomized setting is currently being evaluated in an intergroup trial initiated by the EORTC (#20012).44 Here, eight cycles of ABVD are being compared with four cycles of BEACOPP-escalated plus four cycles of BEACOPP-baseline. Further intensification of first-line therapy in high-risk patients by directly administering high-dose chemotherapy and autologous stem cell transplantation after four instead of eight cycles of ABVD did not improve outcome compared with conventional treatment.71 BEACOPP chemotherapy is generally associated with more haematological toxicity, sterility and secondary leukaemia when compared with ABVD. Nevertheless, cardiotoxicity and pulmonary side effects are similar with both regimens, especially when combined with radiotherapy. A combination of gemcitabine and bleomycin in a BEACOPP variant (BAGCOPP) resulted in severe pulmonary toxicity and should be avoided.72 The role of consolidating radiotherapy after effective chemotherapy in the treatment of patients with advanced HL is still subject to clinical research. A meta-analysis comparing combined modality approaches and chemotherapy alone reported equal tumour control and even better overall survival in patients treated with chemotherapy alone.73 Therefore, randomized trials are currently evaluating the impact of radiotherapy after effective chemotherapy for advanced HL. A study conducted by the EORTC indicated that consolidating IF-RT did not result in better outcome in patients who had already achieved a complete remission after six to eight cycles of MOPP/ABV, although it may be beneficial to patients with partial remissions.74 Longer follow-up of the recently terminated GHSG HD12 trial and the ongoing HD15 trial may help to define the role of radiotherapy for residual disease. In ongoing studies, such as the HD15 trial, PET scan is not only utilized as a tool to
1036 Hodgkin’s lymphoma
Table 42.7 Selected trials for advanced Hodgkin’s lymphoma Trial
Therapy regimen
No. of patients
Outcome
Ref.
95% (OS) 83% (FFP) [CS III/IV; 12 years]
66
Stanford
Stanford V (12 weeks) (RT to initial mediastinal bulk hilar supraclavicular nodes)
108
Intergroup Italy
A. ABVD (6 cycles) B. Stanford V (12 weeks) C. MEC hybrid (6 courses) (RT initial bulk/residual mass)
98 89 88
83% (FFS); 86% (FFP); 90% (OS) 67% (FFS); 76% (FFP); 83% (OS) 85% (FFS); 93% (FFP); 90% (OS) [5 years]
67
Intergroup GB and Italy
A. ChlVPP/EVA hybrid (6 cycles) B. VAPEC-B (11 weeks) ( RT initial bulk/ residual mass)
144 138
82% (FFP); 78% (EFS); 89% (OS) 62% (FFP); 58% (EFS); 79% (OS) [5 years]
68
GHSG HD9
A. COPP/ABVD (4 cycles) B. BEACOPP-baseline (8 cycles) C. BEACOPP-escalated (8 cycles)
260 469 466
67% (FFTF); 79% (OS) 75% (FFTF) 84% (OS) 84% (FFTF) 90% (OS) [7 years]
69
GHSG HD12
A. 8 BEA esc. B. 8 BEA esc. C. 4 BEA esc. 4 BEA baseline D. 4 BEA esc. 4 BEA baseline (A. C. RT bulk/residual mass)
348 345 351 352
Fourth interim analysis [2 years] all patients: 88% (FFTF); 94% (OS)
69
GHSG HD15
A. 8 BEA esc. B. 6 BEA esc. C. 8 BEA-14 (RT to PET residual mass 2.5 cm)
Ongoing trial
Intergroup #20012 EORTC
8 ABVD 4 BEA esc. 4 BEA baseline
Ongoing trial
Abbreviations: SWOG, Southwest Oncology Group; EORTC, European Organization for Research and Treatment of Cancer; GELA, Groupe d’Etude des Lymphomes de l’Adulte; GHSG, German Hodgkin Lymphoma Study Group; ECOG, Eastern Cooperative Oncology Group; EF/IF-RT, extended/involved-field radiotherapy; STNI, subtotal nodal irradiation; FFS, failure-free survival; FFP, freedom from progression; FFTF, freedom from treatment failure; RFS, relapsefree survival; EFS, event-free survival; OS, overall survival.
analyse tumour activity in residual masses after chemotherapy: there are some data suggesting that early PET scan during chemotherapy may discriminate between responders and non-responders and thus have a potential role for use in response-adapted strategies.75,76
RELAPSED AND REFRACTORY HODGKIN’S LYMPHOMA The majority of patients achieve complete remission (CR) with current first-line treatment. However, those patients relapsing still have a chance of being cured with adequate salvage treatment. Patients with relapsed and refractory HL have various treatment options depending on first-line therapy. Conventional chemotherapy is usually the treatment of choice for patients relapsing after initial radiotherapy only. In contrast, options for those who relapse after prior chemotherapy include salvage radiotherapy for strictly localized relapse in previously non-irradiated areas,
salvage chemotherapy, or high-dose chemotherapy (HDCT) followed by autologous stem cell transplantation (SCT). Other experimental options such as allogeneic SCT and monoclonal antibodies are being evaluated for multiply pre-treated patients. Depending on the duration of remission after first-line treatment, most study groups categorize failures into three subgroups: early and late relapses of HL and primary progressive HL.
Salvage radiotherapy and salvage polychemotherapy Salvage radiotherapy alone offers an effective treatment option for a selected subset of patients with relapsed HL. This applies to patients with localized relapses in previously non-irradiated areas. In a retrospective analysis from the GHSG database including 624 relapsed or refractory HL patients, 100 patients were eligible to receive salvage radiotherapy alone: the 5-year freedom from second failure
Relapsed and refractory hodgkin’s lymphoma 1037
(FF2F) and OS rates were 28 per cent and 51 per cent, respectively. Prognostic factors for OS were B-symptoms, stage at relapse, performance status and duration of first remission in limited-stage relapses.77 Conventional chemotherapy is the treatment of choice for patients who relapse after initial radiotherapy for earlystage disease. The survival of these patients is equal to that of patients with advanced-stage HL initially treated with chemotherapy.78 The best treatment for recurrent HL after primary chemotherapy is high-dose chemotherapy. A number of conventional salvage protocols have been developed during the last decade. However, long-term follow-up data are scarce since most of the patients achieving complete (CR) and partial remission (PR) immediately proceeded to HDCT plus autologous SCT. Overall response rates with conventional salvage therapy ranged between 60 and 80 per cent, but less than 30 per cent of patients achieve a lasting remission. Patients relapsing after more than two cycles of polychemotherapy should be treated with HDCT at relapse. The best treatment for those relapsing after two cycles of treatment plus IF radiotherapy remains to be defined.
High-dose chemotherapy followed by autologous stem cell transplantation Younger patients relapsing after initial chemotherapy are usually treated with HDCT and peripheral blood SCT. This strategy has been shown to produce 30–65 per cent long-term disease-free survival in selected patients with refractory and relapsed HL.79,80 The reduction of early transplant-related mortality to less than 5 per cent has led to widespread acceptance. Thus far, two randomized trials have demonstrated the superiority of HDCT followed by autologous SCT over conventional chemotherapy. The British National Lymphoma Investigation (BNLI) reported that patients with relapsed or refractory HL receiving high-dose BEAM with autologous SCT fared significantly better than those treated with conventional dose mini-BEAM, resulting in a 3-year event-free survival of 53 per cent versus 10 per cent.81 In the HD-R1-trial of the GHSG, chemosensitive patients relapsing after initial chemotherapy were randomized between four cycles of Dexa-BEAM and two cycles of Dexa-BEAM followed by BEAM and autologous SCT. The final results demonstrated a higher FFTF in the transplanted group than in the group receiving conventional salvage chemotherapy (55 per cent vs. 34 per cent).82 Even in the subgroup of patients with late relapse the FFTF was significantly better (75 per cent and 44 per cent). The FFTF for patients with early relapse was 41 per cent and 12 per cent, respectively. The OS did not differ significantly between the two treatment arms. This might be due, at least in part, to the fact that many patients relapsing after conventional salvage therapy received autologous SCT upon subsequent relapse. The follow-up data after 7 years confirm these results.83
The success of HDCT followed by autologous SCT does not only depend on obvious factors such as tumour burden or chemosensitivity. A prognostic score based on treatment outcome of patients with relapsed HL also identified the time to relapse, the clinical stage at relapse and the presence of anaemia as independent risk factors.84 The reduction of tumour volume prior to HDCT followed by autologous SCT is an important variable affecting outcome in relapsed and refractory HL. A brief tumourreducing programme with two cycles of DHAP given in short intervals supported by G-CSF was shown to be both effective and well tolerated in patients with relapsed and refractory disease.85 Therefore, this regimen was chosen for the HD-R2 study instead of previously used regimens such as Dexa-BEAM which were associated with severe treatmentrelated toxicity and mortality before the dose reduction of etoposide. Furthermore, the DHAP regimen can be used to successfully collect stem cells in most HL patients.86 The strategy of sequential HDCT follows the Norton– Simon hypothesis: after initial cytoreduction, few noncross-resistant agents are given at short intervals. The transplantation of peripheral blood stem cells and the use of growth factors allow the application of the most effective drugs at the highest possible doses at small intervals of one to three weeks. The HDR2 pilot study conducted by the GHSG evaluated the feasibility and efficacy of this novel approach including a high-dose sequential chemotherapy programme and a final myeloablative course in 102 patients with relapsed or refractory HL. Treatment consisted of two cycles of DHAP (dexamethasone, ara-C, cisplatin) followed by sequential high-dose chemotherapy with cyclophosphamide, methotrexate plus vincristine, and etoposide. The final myeloablative course was BEAM followed by peripheral blood SCT. With a median follow-up of 30 months, FF2F and OS rates were 59 per cent and 78 per cent, respectively, for all patients. FF2F and OS for patients with early relapse were 62 per cent and 81 per cent, for those with late relapse 65 per cent and 81 per cent; for those with progressive disease 41 per cent and 48 per cent, and for those with multiple relapse 39 per cent and 48 per cent, respectively. In multivariate analysis, response after DHAP and duration of first remission were prognostic factors for FF2F and OS.87 Based on the promising results of this study, the GHSG started a prospective European intergroup trial together with the EORTC, the GELTAMO and the EBMT (HD-R2) that is still ongoing. The rationale is to compare the effectiveness of two courses of DHAP followed by BEAM with the intensified sequential strategy in a randomized setting. Patients with histologically confirmed early relapsed HL, and patients in second relapse with no prior HDCT are included.88
Primary progressive and refractory Hodgkin’s lymphoma For patients with primary progressive disease during induction treatment or within 3 months after the end of
1038 Hodgkin’s lymphoma
first-line therapy, conventional salvage chemotherapy has given disappointing results in the vast majority of patients. No response at all or only a very short response to salvage treatment resulted in a very poor 8-year OS ranging between 0 per cent and 8 per cent.89,90 To determine prognostic factors and treatment outcome of patients with primary progressive HL, the GHSG retrospectively analysed 206 patients with progressive disease. The five year FFTF and OS for all patients was 17% and 26%, whereas the corresponding figures for patients treated with HDCT were 31 per cent and 43 per cent, respectively. The low percentage (only 33 per cent) of patients who received HDCT was due to rapidly fatal disease or life-threatening severe toxicity after salvage therapy. Other reasons for not proceeding to HDCT were insufficient stem cell harvest, poor performance status and older age. In multi-variate analysis, low Karnofsky performance score at the time of progression, age above 50 years, and failure to attain a temporary remission on firstline treatment were significant adverse prognostic factors for OS.91 The effectiveness of HDCT and autologous SCT for patients with biopsy-proven primary refractory HL was shown by the Memorial Sloan Kettering Cancer Center, New York, in a study of 75 consecutive patients who were treated with HDCT and autologous SCT. At a median follow-up of 10 years for surviving patients, the event-free survival, progression-free survival, and overall survival rates were 45 per cent, 49 per cent and 48 per cent respectively. Chemosensitivity to standard-dose second-line chemotherapy was predictive of a better survival.92
Allogeneic stem cell transplantation Allogeneic SCT cannot yet be considered an alternative standard treatment in patients with relapsed HL. So far, the advantages of a potential graft-versus-lymphoma effect have been offset by a substantial transplant-related mortality (TRM) of more than 50 per cent. Furthermore, donor availability and age constraints have limited a broader application of allogeneic SCT in HL patients. As shown by a recently published matched-pair analysis, TRM might be significantly reduced by employing reduced-intensity conditioning (RIC).93 Furthermore, another recent study using RIC in 49 HL patients indicates the potential for durable responses in patients who have previously had substantial treatment for Hodgkin’s lymphoma. The low non-relapserelated mortality suggests that allogeneic trans-plants should be considered earlier in the course of the disease.94 Allogeneic SCT following RIC might thus become an appropriate strategy in selected subgroups of young poorrisk patients, e.g. for those failing autologous transplantation, or patients with early relapse and further risk factors who are chemotherapy sensitive.95 However, to date the number of patients treated is small, requiring further clinical studies and information in order to define clear indications.
SPECIAL CONSIDERATIONS Elderly patients Although there is great variety in the health and physical condition of elderly patients with HL, the age at diagnosis remains an unfavourable risk factor, particularly in patients with advanced stages. In most groups patients are considered ‘elderly’ if they are older than 60 years. Factors such as more aggressive disease, more frequent diagnosis of advanced stage, co-morbidity, poor tolerance of treatment, failure to maintain dose intensity, shorter survival after relapse, and death due to other causes contribute to the poorer outcome of elderly patients. A retrospective analysis of GHSG trials showed that elderly patients have a poorer risk profile, more treatment-associated toxicity, a lower dose intensity and higher mortality as major factors for poorer outcome.96 In the HD9elderly trial of the GHSG, patients between 66 and 75 years with advanced-stage HL were treated with either COPP/ABVD or BEACOPP baseline. Tumour control appeared to be better with the BEACOPP regimen, but toxicity was higher, resulting in no differences in FFTF or OS.97 In phase I/II trials of the GHSG, two new regimens are currently being evaluated for elderly patients: PVAG (prednisone, vinblastine, doxorubicin and gemcitabine) and BACOPP (bleomycin, doxorubicin, cyclophosphamide, vincristine, procarbazine and prednisone).
Fertility Based on the substantially improved long-term survival in young patients with HL undergoing chemotherapy, preservation of fertility becomes an increasingly important issue. Regimens such as COPP, MOPP, or BEACOPP containing alkylating agents such as procarbazine or cyclophosphamide often lead to therapy-induced infertility, whereas ABVD is less gonadotoxic. For men, cryopreservation of semen prior to therapy is possible; however, strategies for women are less clear. A first birth after ovarian cryopreservation and re-implantation has been reported previously.98 Nevertheless, reproductive technologies including ovarian cryopreservation with auto- or xenotransplantation as well as in vitro maturation of thawed primordial follicles followed by fertilization and embryo transfer are not yet successfully established. Data on co-treatment preserving ovarian function are still scarce. A retrospective analysis of the GHSG demonstrates that the rate of therapy-induced amenorrhoea is higher in women receiving eight cycles of escalated BEACOPP compared to women treated with ABVD alone, COPP/ABVD or standard BEACOPP. Moreover, amenorrhoea after therapy was most pronounced in women with advanced-stage HL, in those older than 30 years at treatment, and in women who did not take oral contraceptives during chemotherapy.99 Administration of oral contraceptives or gonadotropin-releasing hormone agonists
Experimental therapies 1039
(goserelin) during chemotherapy may achieve lower rates of ovarian failure.100,101 An ongoing GHSG phase II study (PROFE) for the first time compares both strategies in a randomized setting for females (18–40 years) with advanced HL.
LONG-TERM MANAGEMENT Palliative treatment Depending on age, number and character of relapses, previous therapies and presence of concomitant diseases, doctors should carefully evaluate whether a curative or a palliative approach is chosen. A palliative regimen can achieve satisfactory pain control, improve general condition, and lead to partial, sometimes long-lasting, remissions. Drugs such as gemcitabine, vinorelbine, vinblastine, idarubicin or etoposide are mostly given as monotherapy which can be combined with corticosteroids. The most promising alternative is gemcitabine, which proved to be a suitable and well-tolerated substance, even for patients with multiple relapses of HL in a phase II study.102 Clinical trials are underway to incorporate Gemcitabine into the treatment of untreated Hodgkin lymphoma in elderly patients who tolerate ABVD less well.
Follow-up and sequelae of treatment During the follow-up period of HL patients, attention should be paid to some crucial points. First, more than two thirds of relapses occur within 2.5 years and more than
90 per cent within 5 years after initial treatment. Thus, the patient should obtain a schedule for follow-up visits and examinations (Table 42.8). Second, a number of long-term toxic effects related to treatment of HL can occur. They include minor disorders such as endocrine dysfunction, long-term immunosuppression and viral infections. Serious impairments consist of lung fibrosis from bleomycin and/ or irradiation,33,103,104 myocardial damage from anthracyclines and/or irradiation,32,105 sterility,106,107 growth abnormalities in children, opportunistic infections, psychological and psychosocial problems and fatigue.108 Potentially fatal effects include the overwhelming post-splenectomy infection syndrome after splenectomy or spleen irradiation and secondary neoplasms. Acute myeloid leukaemia/myelodysplastic syndrome is mostly observed within the first 3–5 years and secondary non-Hodgkin’s lymphoma mainly at 5–15 years after initial treatment. Solid tumours, e.g. lung or breast cancer, can also occur decades after initial treatment and sometimes even as multiple tumours.34–36,109
EXPERIMENTAL THERAPIES Data from HL and non-Hodgkin’s lymphoma (NHL) suggest that small numbers of residual tumour cells remaining after first-line treatment can give rise to late relapses. Thus, eliminating residual malignant lymphoma cells after first-line treatment might further improve outcome in these diseases. In addition, combined immuno-chemotherapy might help to reduce the amount of cytotoxic drugs needed, resulting in less toxicity of the currently used regimens. Experimental strategies in the treatment of HL include passive immunotherapy based on monoclonal antibodies
Table 42.8 Follow-up examinations (extracted and modified from the current GHSG trial protocol HD 15 for treatment of advanced-stage HL) First year Examination time point Anamnesis Physical examination Laboratory tests: Blood count and differential distribution ESR, CRP TSH Computed tomographya (if PR) Chest X-ray (if no CT) Lung function Abdominal ultrasound a
Second to fourth year
Fifth year onwards
Month 3
Month 6
Month 12
Every 6 months
Annually
X X
X X
X X
X X
X X
X
X
X
X
X
X X Xa X
X X
X X X X X X
X X
X X
Xb
X
X
X
X
Further CT is recommended according to findings in final re-staging and follow-up. Imaging examinations annually. ESR, erythrocyte sedimentation rate; CRP, C-reactive protein; TSH, thyroid-stimulating hormone; CT, computed tomography; PR, partial remission.
b
1040 Hodgkin’s lymphoma
to specifically target malignant cells and active immunotherapy with modulation of cellular response by cytokines, tumour vaccines or gene transfer. Hodgkin’s lymphoma seems to be an ideal target for antibody-based therapeutic approaches since H-RS cells express specific surface antigens such as CD25 and CD30 in large amounts. Approaches involving antibody-based agents have given promising results in experimental HL models and have demonstrated some clinical efficacy in patients with advanced refractory HL.110,111 Clinical phase I/II trials are ongoing with unmodified humanized or human monoclonal antibodies (MAbs), which either induce target cell death by direct interaction or antibodydependent cellular or complement-dependent cytotoxicity. Different approaches evaluated clinically include bispecific immunotoxins and radioimmunoconjugates. Fully human antibodies against CD30 have given promising results more recently.112,113 However, it seems unlikely that resistant patients with larger tumour masses can be cured by either of these approaches. Future strategies aim at combining conventional chemo- and/or radiotherapy for debulking, and experimental therapies with biological agents to kill residual H-RS cells and thus prevent relapses.
ACKNOWLEDGEMENTS Our research is supported in parts by German Cancer Aid (Deutsche Krebshilfe), the German Federal Ministry of Education and Research (Bundesministerium für Bildung und Forschung) and the Competence Network Malignant Lymphoma (Kompetenznetz Maligne Lymphome).
REFERENCES ●1
2
3
4
5
KEY LEARNING POINTS ●
●
●
●
●
●
●
Hodgkin’s lymphoma (HL) should always be proven by excisional biopsy sent to an expert haematopathologist for initial diagnosis. Accurate staging procedures and risk factor assessment as described are indispensable for adequate stage- and risk-adapted therapy. Patients with Hodgkin’s lymphoma should preferably be enrolled into clinical trials. In most centres patients with early-stage favourable or early-stage unfavourable HL are treated with combined modality strategies including 2–6 cycles of polychemotherapy (e.g. ABVD) followed by radiotherapy to the involved field. The treatment for advanced stages mostly consists of more cycles or of a more intensive regimen, plus radiotherapy to possible residual masses. For patients relapsing after combined modality treatment or patients with primary progressive disease, high-dose chemotherapy plus autologous stem cell transplantation still offers a chance of cure. With regard to the excellent cure rates, modern treatment strategies also aim at reducing therapy-induced acute and long-term toxicities without loss of efficacy. Potential future treatments may combine chemotherapy with experimental strategies.
6
7
8
9
10 11
12
*13
Hodgkin T. On some morbid appearances of the absorbent glands and spleen. Medico-Chirurgical Trans 1832; 17:68–97. Sternberg C. Über eine eigenartige unter dem Bilde der Pseudoleukamie verlaufende Tuberkulose des lymphatischen Apparates. Zeitschr Heilk 1898; 19:21–90. Reed DM. On the pathological changes in Hodgkin’s disease, with special reference to its relation to tuberculosis. John Hopkins Hosp Rep 1902; 10:133–96. Kuppers R, Rajewsky K. The origin of Hodgkin and Reed/Sternberg cells in Hodgkin’s disease. Annu Rev Immunol 1998; 16:471–93. Braeuninger A, Kuppers R, Strickler JG, et al. Hodgkin and Reed–Sternberg cells in lymphocyte predominant Hodgkin disease represent clonal populations of germinal centrederived tumor B cells. Proc Natl Acad Sci U S A 1997; 94:9337–42. Parkin DM, Muir CS. Cancer incidence in five continents. Comparability and quality of data. IARC Sci Publ 1992; 120:45–173. EUCAN online database. Incidence and mortality data on Hodgkin’s disease: 1998 estimates, version 5.0. www.dep.iarc.fr/eucan/eucan.htm, 1998. Arbeitsgemeinschaft Bevölkerungsbezogener Krebsregister in Deutschland, in Zusammenarbeit mit dem Robert-KochInstitut. Krebs in Deutschland, Häufigkeiten und Trends. 4. überarbeitete, aktualisierte Ausgabe. Saarbrücken: 2004. Miller T, LeBlanc M, Braziel R, et al. Was the bimodal age incidence of Hodgkin’s lymphoma a result of mistaken diagnoses of non-Hodgkin’s lymphoma? Annual Meeting of the American Society of Hematology. Blood 2002; 100:3048a. Gutensohn N, Shapiro D. Social class risk factors among children with Hodgkin’s disease. Int J Cancer 1982; 30:433–5. Chang ET, Zheng T, Lennette ET, et al. Heterogeneity of risk factors and antibody profiles in Epstein–Barr virus genomepositive and -negative Hodgkin lymphoma. J Infect Dis 2004; 189:2271–81. Crowther D, Bonadonna G. Hodgkin’s disease in adults. In: Peckham M PH, Veronesi U (eds) Oxford Textbook of Oncology. New York: Oxford University Press; 1995: 1720–5. Harris NL, Jaffe ES, Diebold J, et al. The World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues. Report of the Clinical Advisory Committee meeting, Airlie House, Virginia, November, 1997. Ann Oncol 1999; 10:1419–32.
References 1041
14 Mueller N, Evans A, Harris NL, et al. Hodgkin’s disease and Epstein–Barr virus. Altered antibody pattern before diagnosis. N Engl J Med 1989; 320:689–95. 15 Weiss LM, Strickler JG, Warnke RA, et al. Epstein–Barr viral DNA in tissues of Hodgkin’s disease. Am J Pathol 1987; 129:86–91. 16 Glaser S, Lin R, Stewart S, et al. Epstein–Barr virusassociated Hodgkin’s disease: epidemiologic characteristics in international data. Int J Cancer 1997; 70: 375–382. 17 Zarate-Osorno A, Roman L, Kingma D, et al. Hodgkin’s disease in Mexico. Prevalence of Epstein–Barr virus sequences and correlations with histologic subtype. Cancer 1995; 75:1360–6. 18 Jarrett RF, MacKenzie J. Epstein–Barr virus and other candidate viruses in the pathogenesis of Hodgkin’s disease. Semin Hematol 1999; 36:260–9. 19 Haim N, Cohen Y, Robinson E. Malignant lymphoma in firstdegree blood relatives. Cancer 1982; 49:2197–200. 20 Lindelof B, Eklund G. Analysis of hereditary component of cancer by use of a familial index by site. Lancet 2001; 358:1696–8. 21 Mack T, Cozen W, Shibata D, et al. Concordance for Hodgkin’s disease in identical twins suggesting genetic susceptibility to the young-adult form of the disease. N Engl J Med 1995; 332:413–8. 22 Muschen M, Rajewsky K, Brauninger A, et al. Rare occurrence of classical Hodgkin’s disease as a T cell lymphoma. J Exp Med 2000; 191:387–94. 23 Falini B, Stein H, Pileri S, et al. Expression of lymphoidassociated antigens on Hodgkin’s and Reed–Sternberg cells of Hodgkin’s disease. An immunocytochemical study on lymph node cytospins using monoclonal antibodies. Histopathology 1987; 11:1229–42. 24 Bargou RC, Emmerich F, Krappmann D, et al. Constitutive nuclear factor-kappaB-RelA activation is required for proliferation and survival of Hodgkin’s disease tumor cells. J Clin Invest 1997; 100:2961–9. 25 Gires O, Zimber-Strobl U, Gonnella R, et al. Latent membrane protein 1 of Epstein–Barr virus mimics a constitutively active receptor molecule. EMBO J 1997; 16:6131–40. 26 Rosenberg SA, Boiron M, DeVita VT Jr, et al. Report of the Committee on Hodgkin’s Disease Staging Procedures. Cancer Res 1971; 31:1862–3. *27 Lister TA, Crowther D, Sutcliffe SB, et al. Report of a committee convened to discuss the evaluation and staging of patients with Hodgkin’s disease: Cotswolds meeting. J Clin Oncol 1989; 7:1630–6. ◆28 Diehl V, Stein H, Hummel M, et al. Hodgkin’s lymphoma: biology and treatment strategies for primary, refractory, and relapsed disease. Hematology (Am Soc Hematol Educ Program) 2003; 2003:225–472. ●29 Hasenclever D, Diehl V. A prognostic score for advanced Hodgkin’s disease. International Prognostic Factors Project on Advanced Hodgkin’s Disease. N Engl J Med 1998; 339:1506–14.
30 Longo DL, Young RC, Wesley M, et al. Twenty years of MOPP chemotherapy for Hodgkin’s disease. J Clin Oncol 1986; 4:1295–306. ●31 Bonadonna G, Zucali R, Monfardini S, et al. Combination chemotherapy of Hodgkin’s disease with adriamycin, bleomycin, vinblastine, and imidazole carboximide versus MOPP. Cancer 1975; 36:252–9. 32 Hancock SL, Tucker MA, Hoppe RT. Factors affecting late mortality from heart disease after treatment of Hodgkin’s disease. JAMA 1993; 270:1949–55. 33 Dubray B, Henry-Amar M, Meerwaldt JH, et al. Radiationinduced lung damage after thoracic irradiation for Hodgkin’s disease: the role of fractionation. Radiother Oncol 1995; 36:211–7. 34 Ng AK, Bernardo MV, Weller E, et al. Second malignancy after Hodgkin disease treated with radiation therapy with or without chemotherapy: long-term risks and risk factors. Blood 2002; 100:1989–96. 35 Bhatia S, Yasui Y, Robison LL, et al. High risk of subsequent neoplasms continues with extended follow-up of childhood Hodgkin’s disease: report from the Late Effects Study Group. J Clin Oncol 2003; 21:4386–94. ●36 van Leeuwen FE, Klokman WJ, Stovall M, et al. Roles of radiation dose, chemotherapy, and hormonal factors in breast cancer following Hodgkin’s disease. J Natl Cancer Inst 2003; 95:971–80. 37 Horwich A, Specht L, Ashley S. Survival analysis of patients with clinical stages I or II Hodgkin’s disease who have relapsed after initial treatment with radiotherapy alone. Eur J Cancer 1997; 33:848–53. 38 Specht L, Gray RG, Clarke MJ, et al. Influence of more extensive radiotherapy and adjuvant chemotherapy on long-term outcome of early-stage Hodgkin’s disease: a meta-analysis of 23 randomized trials involving 3,888 patients. International Hodgkin’s Disease Collaborative Group. J Clin Oncol 1998; 16:830–43. 39 Press OW, LeBlanc M, Lichter AS, et al. Phase III randomized intergroup trial of subtotal lymphoid irradiation versus doxorubicin, vinblastine, and subtotal lymphoid irradiation for stage IA to IIA Hodgkin’s disease. J Clin Oncol 2001; 19:4238–44. 40 Bonadonna G, Bonfante V, Viviani S, et al. ABVD plus subtotal nodal versus involved-field radiotherapy in earlystage Hodgkin’s disease: long-term results. J Clin Oncol 2004; 22:2835–41. 41 Horning SJ, Hoppe RT, Breslin S, et al. Very brief (8 week) chemotherapy (CT) and low dose (30 Gy) radiotherapy (RT) for limited stage Hodgkin’s disease (HD): preliminary results of the Stanford–Kaiser G4 Study of Stanford V RT. Blood 1999; 94:1717a. 42 Carde P, Noordijk E, Hagenbeek A, et al. Superiority of EBVP chemotherapy in combination with involved field irradiation over subtotal nodal irradiation in favorable clinical stage I–II Hodgkin’s disease: The EORTC-GPMC H7F randomized trial. Proc ASCO 1997; 16:13. 43 Hagenbeek A, Eghbali H, Fermé C, et al. Three cycles of MOPP/ABV hybrid and involved-field irradiation is more
1042 Hodgkin’s lymphoma
◆44
45
46
47
48
49
50
51
52
53
●54
effective than subtotal nodal irradiation in favorable supradiaphragmatic clinical stages I–II Hodgkin’s disease: Preliminary results of the EORTC-GELA H8F randomized trial in 543 patients. Blood 2000; 96:575a. Raemaekers J, Kluin-Nelemans H, Teodorovic I, et al. The achievements of the EORTC Lymphoma Group. European Organisation for Research and Treatment of Cancer. Eur J Cancer 2002; 38:S107–13. Noordijk EM, Thomas J, Fermé C, et al. First results of the EORTC-GELA H9 randomized trials: the H9-F trial (comparing 3 radiation dose levels) and H9-U trial (comparing 3 chemotherapy schemes) in patients with favorable or unfavorable early stage Hodgkin’s lymphoma (HL). J Clin Oncol 2005; 23:6505a. Straus DJ, Portlock CS, Qin J, et al. Results of a prospective randomized clinical trial of doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) followed by radiation therapy (RT) versus ABVD alone for stages I, II, and IIIA nonbulky Hodgkin disease. Blood 2004; 104:3483–9. Sieber M, Franklin J, Tesch H, et al. Two cycles ABVD plus extended field radiotherapy is superior to radiotherapy alone in early stage Hodgkin’s disease: results of the German Hodgkin’s Lymphoma Study Group (GHSG) trial HD7. Blood 2002; 100:341a. Diehl V, Brillant C, Engert A, et al. HD10: Investigating reduction of combined modality treatment intensity in early stage Hodgkin’s lymphoma. Interim analysis of a randomized trial of the German Hodgkin Study Group (GHSG). J Clin Oncol 2005; 23:6506a. Nogova L, Reineke T, Eich HT, et al. Extended field radiotherapy, combined modality treatment or involved field radiotherapy for patients with stage IA lymphocytepredominant Hodgkin’s lymphoma: a retrospective analysis from the German Hodgkin Study Group (GHSG). Ann Oncol 2005; 16:1683–7. Rehwald U, Schulz H, Reiser M, et al. Treatment of relapsed CD20 Hodgkin lymphoma with the monoclonal antibody rituximab is effective and well tolerated: results of a phase 2 trial of the German Hodgkin Lymphoma Study Group. Blood 2003; 101:420–4. Ekstrand BC, Lucas JB, Horwitz SM, et al. Rituximab in lymphocyte-predominant Hodgkin disease: results of a phase 2 trial. Blood 2003; 101:4285–9. Zittoun R, Audebert A, Hoerni B, et al. Extended versus involved fields irradiation combined with MOPP chemotherapy in early clinical stages of Hodgkin’s disease. J Clin Oncol 1985; 3:207–14. Ferme C, Eghbali H, Habenbeek A, et al. MOPP/ABV (M/A) hybrid and irradiation in unfavorable supradiaphragmatic clinical stages I–II HD: Comparison of three treatment modalities, preliminary results of the EORTC-GELA H8-U randomized trial in 995 patients. Blood 2000; 96:576a. Engert A, Schiller P, Josting A, et al. Involved-field radiotherapy is equally effective and less toxic compared with extended-field radiotherapy after four cycles of chemotherapy in patients with early-stage unfavorable Hodgkin’s Lymphoma: Results of the HD8 trial of the
55
56
57
●58
●59
60
61
62
63
64
65 ●66
●67
German Hodgkin’s Lymphoma Study Group. J Clin Oncol 2003; 21:3601–8. Meyer RM, Gospodarowicz MK, Connors JM, et al. Randomized comparison of ABVD chemotherapy with a strategy that includes radiation therapy in patients with limited-stage Hodgkin’s lymphoma: National Cancer Institute of Canada Clinical Trials Group and the Eastern Cooperative Oncology Group. J Clin Oncol 2005; 23:4634–42. Vassilakopoulos TP, Angelopoulou MK, Siakantaris MP, et al. Combination chemotherapy plus low-dose involvedfield radiotherapy for early clinical stage Hodgkin’s lymphoma. Int J Radiat Oncol Biol Phys 2004; 59:765–81. Klimm B, Engert A, Brillant C, et al. Comparison of BEACOPP and ABVD chemotherapy in intermediate stage Hodgkin’s lymphoma: results of the fourth interim analysis of the HD 11 trial of the GHSG. J Clin Oncol 2005; 23:6507a. Canellos GP, Anderson JR, Propert KJ, et al. Chemotherapy of advanced Hodgkin’s disease with MOPP, ABVD, or MOPP alternating with ABVD. N Engl J Med 1992; 327:1478–84. Diehl V, Franklin J, Pfreundschuh M, et al. Standard and increased-dose BEACOPP chemotherapy compared with COPP-ABVD for advanced Hodgkin’s disease. N Engl J Med 2003; 348:2386–95. Bonadonna G, Valagussa P, Santoro A. Alternating non-cross-resistant combination chemotherapy or MOPP in stage IV Hodgkin’s disease. A report of 8-year results. Ann Intern Med 1986; 104:739–46. Duggan DB, Petroni GR, Johnson JL, et al. Randomized comparison of ABVD and MOPP/ABV hybrid for the treatment of advanced Hodgkin’s disease: report of an intergroup trial. J Clin Oncol 2003; 21:607–14. Santoro A, Bonadonna G, Valagussa P, et al. Long-term results of combined chemotherapy–radiotherapy approach in Hodgkin’s disease: superiority of ABVD plus radiotherapy versus MOPP plus radiotherapy. J Clin Oncol 1987; 5:27–37. Connors JM, Klimo P, Adams G, et al. Treatment of advanced Hodgkin’s disease with chemotherapy – comparison of MOPP/ABV hybrid regimen with alternating courses of MOPP and ABVD: A report from the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 1997; 15:1638–45. Sieber M, Tesch H, Pfistner B, et al. Treatment of advanced Hodgkin’s disease with COPP/ABV/IMEP versus COPP/ABVD and consolidating radiotherapy: final results of the German Hodgkin’s Lymphoma Study Group HD6 trial. Ann Oncol 2004; 15:276–82. Canellos GP, Niedzwiecki D. Long-term follow-up of Hodgkin’s disease trial. N Engl J Med 2002; 346:1417–18. Horning SJ, Hoppe RT, Breslin S, et al. Stanford V and radiotherapy for locally extensive and advanced Hodgkin’s disease: mature results of a prospective clinical trial. J Clin Oncol 2002; 20:630–7. Gobbi PG, Levis A, Chisesi T, et al. ABVD versus modified Stanford V versus MOPP/EBVCAD with optional and limited radiotherapy in intermediate- and advanced-stage Hodgkin’s lymphoma: final results of a multicenter
References 1043
●68
69
70
71
72
73
●74
75
76
77
78
79
randomized trial by the Intergruppo Italiano Linfomi. J Clin Oncol 2005; 23:9198–207. Radford JA, Rohatiner AZ, Ryder WD, et al. ChlVPP/EVA hybrid versus the weekly VAPEC-B regimen for previously untreated Hodgkin’s disease. J Clin Oncol 2002; 20:2988–94. Diehl V, Brillant C, Franklin J, et al. BEACOPP chemotherapy for advanced Hodgkin’s disease: results of further analyses of the HD9- and HD12- trials of the German Hodgkin Study Group (GHSG). Blood 2004; 104:307a. Sieber M, Bredenfeld H, Josting A, et al. 14-day variant of the bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone regimen in advanced-stage Hodgkin’s lymphoma: results of a pilot study of the German Hodgkin’s Lymphoma Study Group. J Clin Oncol 2003; 21:1734–9. Federico M, Bellei M, Brice P, et al. High-dose therapy and autologous stem-cell transplantation versus conventional therapy for patients with advanced Hodgkin’s lymphoma responding to front-line therapy. J Clin Oncol 2003; 21:2320–5. Bredenfeld H, Franklin J, Nogova L, et al. Severe pulmonary toxicity in patients with advanced-stage Hodgkin’s disease treated with a modified bleomycin, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisone, and gemcitabine (BEACOPP) regimen is probably related to the combination of gemcitabine and bleomycin: a report of the German Hodgkin’s Lymphoma Study Group. J Clin Oncol 2004; 22:2424–9. Loeffler M, Brosteanu O, Hasenclever D, et al. Meta-analysis of chemotherapy versus combined modality treatment trials in Hodgkin’s disease. International Database on Hodgkin’s Disease Overview Study Group. J Clin Oncol 1998; 16:818–29. Aleman BM, Raemaekers JM, Tirelli U, et al. Involved-field radiotherapy for advanced Hodgkin’s lymphoma. N Engl J Med 2003; 348:2396–406. Hutchings M, Mikhaeel NG, Fields PA, et al. Prognostic value of interim FDG-PET after two or three cycles of chemotherapy in Hodgkin lymphoma. Ann Oncol 2005; 16:1160–8. Hutchings M, Loft A, Hansen MT, et al. FDG-PET after two cycles of chemotherapy predicts treatment failure and progression-free survival in Hodgkin lymphoma. Blood 2006; 107:52–9. Josting A, Nogova L, Franklin J, et al. Salvage radiotherapy in patients with relapsed and refractory Hodgkin’s lymphoma: a retrospective analysis from the German Hodgkin Lymphoma Study Group. J Clin Oncol 2005; 23:1522–9. Santoro A, Viviani S, Villarreal CJ, et al. Salvage chemotherapy in Hodgkin’s disease irradiation failures: superiority of doxorubicin-containing regimens over MOPP. Cancer Treat Rep 1986; 70: 343–51. Bierman PJ, Bagin RG, Jagannath S, et al. High dose chemotherapy followed by autologous hematopoietic rescue in Hodgkin’s disease: long-term follow-up in 128 patients. Ann Oncol 1993; 4:767–73.
80 Reece DE, Connors JM, Spinelli JJ, et al. Intensive therapy with cyclophosphamide, carmustine, etoposide cisplatin, and autologous bone marrow transplantation for Hodgkin’s disease in first relapse after combination chemotherapy. Blood 1994; 83:1193–9. ●81 Linch DC, Winfield D, Goldstone AH, et al. Dose intensification with autologous bone-marrow transplantation in relapsed and resistant Hodgkin’s disease: results of a BNLI randomised trial. Lancet 1993; 341:1051–4. ●82 Schmitz N, Pfistner B, Sextro M, et al. Aggressive conventional chemotherapy compared with high-dose chemotherapy with autologous haemopoietic stem-cell transplantation for relapsed chemosensitive Hodgkin’s disease: a randomised trial. Lancet 2002; 359:2065–71. 83 Schmitz N, Haverkamp H, Josting A, et al. Long term followup in relapsed Hodgkin’s disease (HD): updated results of the HD-R1 study comparing conventional chemotherapy (cCT) to high-dose chemotherapy (HDCT) with autologous stem cell transplantation (ASCT) of the GHSG and the working Party Lymphoma of the EBMT. J Clin Oncol 2005; 23:6508a. 84 Josting A, Franklin J, May M, et al. New prognostic score based on treatment outcome of patients with relapsed Hodgkin’s lymphoma registered in the database of the German Hodgkin’s lymphoma study group. J Clin Oncol 2002; 20:221–30. 85 Josting A, Rudolph C, Reiser M, et al. Time-intensified Dexamethasone/cisplatin/cytarabine: an effective salvage therapy with low toxicity in patients with relapsed and refractory Hodgkin’s disease. Ann Oncol 2002; 13:1628–35. 86 Smardova L, Engert A, Haverkamp H, et al. Successful mobilization of PBSC with the DHAP (dexamethasone, ara-c, cisplatinum) regimen plus G-CSF in patients with relapsed Hodgkin’s disease. Leukemia lymphoma 2005; 46:1017–22. 87 Josting A, Rudolph C, Mapara M, et al. Cologne high-dose sequential chemotherapy in relapsed and refractory Hodgkin lymphoma: results of a large multicenter study of the German Hodgkin Lymphoma Study Group (GHSG). Ann Oncol 2005; 16:116–23. 88 Glossmann JP, Josting A, Pfistner B, et al. A randomized trial of chemotherapy with carmustine, etoposide, cytarabine, and melphalan (BEAM) plus peripheral stem cell transplantation (PBSCT) vs single-agent high-dose chemotherapy followed by BEAM plus PBSCT in patients with relapsed Hodgkin’s disease (HD-R2). Ann Hematol 2002; 81:424–9. 89 Longo DL, Duffey PL, Young RC, et al. Conventional-dose salvage combination chemotherapy in patients relapsing with Hodgkin’s disease after combination chemotherapy: the low probability for cure. J Clin Oncol 1992; 10:210–8. 90 Bonfante V, Santoro A, Viviani S, et al. Outcome of patients with Hodgkin’s disease failing after primary MOPP/ABVD. J Clin Oncol 1997; 15:528–34. 91 Josting A, Rueffer U, Franklin J, et al. Prognostic factors and treatment outcome in primary progressive Hodgkin lymphoma: a report from the German Hodgkin Lymphoma Study Group. Blood 2000; 96:1280–6.
1044 Hodgkin’s lymphoma
92 Moskowitz CH, Kewalramani T, Nimer SD, et al. Effectiveness of high dose chemoradiotherapy and autologous stem cell transplantation for patients with biopsy-proven primary refractory Hodgkin’s disease. Br J Haematol 2004; 124:645–52. 93 Sureda A, Robinson S, De Elvira CR, et al. Allogeneic stem cell transplantation significantly reduces transplant related mortality in comparison with conventional allogeneic transplantation in relapsed or refractory Hodgkin’s disease: Results of the European Group for Blood and Marrow Transplantation. Blood 2003; 102:692a. ●94 Peggs KS, Hunter A, Chopra R, et al. Clinical evidence of a graft-versus-Hodgkin’s lymphoma effect after reducedintensity allogeneic transplantation. Lancet 2005; 365:1934–41. 95 Schmitz N, Sureda A, Robinson S. Allogeneic transplantation of hematopoietic stem cells after nonmyeloablative conditioning for Hodgkin’s disease: indications and results. Semin Oncol 2004; 31:27–32. ●96 Engert A, Ballova V, Haverkamp H, et al. Hodgkin’s lymphoma in elderly patients: a comprehensive retrospective analysis from the German Hodgkin’s Study Group. J Clin Oncol 2005; 23:5052–60. 97 Ballova V, Ruffer JU, Haverkamp H, et al. A prospectively randomized trial carried out by the German Hodgkin Study Group (GHSG) for elderly patients with advanced Hodgkin’s disease comparing BEACOPP baseline and COPP-ABVD (study HD9 elderly). Ann Oncol 2005; 16:124–31. ●98 Donnez J, Dolmans MM, Demylle D, et al. Livebirth after orthotopic transplantation of cryopreserved ovarian tissue. Lancet 2004; 364:1405–10. 99 Behringer K, Breuer K, Reineke T, et al. Secondary amenorrhea after Hodgkin’s lymphoma is influenced by age at treatment, stage of disease, chemotherapy regimen, and the use of oral contraceptives during therapy: a report from the German Hodgkin’s Lymphoma Study Group. J Clin Oncol 2005; 23:7555–64. 100 Kreuser ED, Felsenberg D, Behles C, et al. Long-term gonadal dysfunction and its impact on bone mineralization in patients following COPP/ABVD chemotherapy for Hodgkin’s disease. Ann Oncol 1992; 3 Suppl 4:105–10. 101 Blumenfeld Z, Dann E, Avivi I, et al. Fertility after treatment for Hodgkin’s disease. Ann Oncol 2002; 13 Suppl 1:138–47.
102 Santoro A, Bredenfeld H, Devizzi L, et al. Gemcitabine in the treatment of refractory Hodgkin’s disease: results of a multicenter phase II study. J Clin Oncol 2000; 18:2615–9. 103 Hirsch A, Vander Els N, Straus DJ, et al. Effect of ABVD chemotherapy with and without mantle or mediastinal irradiation on pulmonary function and symptoms in earlystage Hodgkin’s disease. J Clin Oncol 1996; 14:1297–305. 104 Horning SJ, Adhikari A, Rizk N, et al. Effect of treatment for Hodgkin’s disease on pulmonary function: results of a prospective study. J Clin Oncol 1994; 12:297–305. 105 Hull MC, Morris CG, Pepine CJ, et al. Valvular dysfunction and carotid, subclavian, and coronary artery disease in survivors of Hodgkin lymphoma treated with radiation therapy. JAMA 2003; 290:2831–7. 106 Viviani S, Ragni G, Santoro A, et al. Testicular dysfunction in Hodgkin’s disease before and after treatment. Eur J Cancer 1991, 27:1389–92. 107 Familiari G, Caggiati A, Nottola SA, et al. Ultrastructure of human ovarian primordial follicles after combination chemotherapy for Hodgkin’s disease. Hum Reprod 1993; 8:2080–7. 108 Ruffer JU, Flechtner H, Tralls P, et al. Fatigue in long-term survivors of Hodgkin’s lymphoma; a report from the German Hodgkin Lymphoma Study Group (GHSG). Eur J Cancer 2003; 39:2179–86. 109 Travis LB, Gospodarowicz M, Curtis RE, et al. Lung cancer following chemotherapy and radiotherapy for Hodgkin’s disease. J Natl Cancer Inst 2002; 94:182–92. 110 Borchmann P, Schnell R, Fuss I, et al. Phase 1 trial of the novel bispecific molecule H22xKi-4 in patients with refractory Hodgkin lymphoma. Blood 2002; 100:3101–7. 111 Schnell R, Dietlein M, Staak JO et al. Treatment of refractory Hodgkin’s lymphoma patients with an iodine131-labeled murine anti-CD30 monoclonal antibody. J Clin Oncol 2005; 23:4669–78. 112 Borchmann P, Treml JF, Hansen H, et al. The human antiCD30 antibody 5F11 shows in vitro and in vivo activity against malignant lymphoma. Blood 2003; 102:3737–42. ◆113 Borchmann P, Schnell R, Schulz H, et al. Monoclonal antibody-based immunotherapy of Hodgkin’s lymphoma.Curr Opin Investig Drugs 2004; 5:1262–7.
43 Non-Hodgkin’s lymphoma ADRIAN J.C. BLOOR AND DAVID C. LINCH
Epidemiology, classification and staging of non-Hodgkin’s lymphoma Aggressive non-Hodgkin’s lymphoma
1045 1048
EPIDEMIOLOGY, CLASSIFICATION AND STAGING OF NON-HODGKIN’S LYMPHOMA Epidemiology Non-Hodgkin’s lymphoma (NHL) encompasses a heterogeneous group of tumours of the lymphoid system. The worldwide incidence is approximately 10 cases per 100 000 person years, although this is increased in Western countries with the highest figures reported in the United States (Fig. 43.1). Data from the USA and Europe have indicated that the overall incidence of NHL rose by 3–4 per cent per year between the 1970s and 1990s although the rate of rise appears to have slowed during the last decade.1,2 Despite the identification of many risk factors for the development of NHL, the reason for this changing pattern is unclear. Lymphomas of all types are more common in the elderly (Fig. 43.2) making this an increasing health-care problem in an ageing population, and they have an uneven sex and racial distribution, occurring more frequently in men and people with a black African ancestry.2 Other than demographic variations, three other major groups of risk factors have been identified that are implicated in the pathogenesis of NHL – immunodeficiency, infection and environmental factors. Immunodeficiency, either congenital or secondary to infection (in particular human immunodeficiency virus – HIV), drugs, or autoimmune disorders, predisposes to several types of NHL. The pathogenesis of lymphoma in this group is incompletely understood but is related to chronic antigenic stimulation and impaired immune surveillance of oncogenic viruses and host cell changes. A number of infections are associated with
Indolent non-Hodgkin’s lymphoma Future developments References
1056 1062 1063
the development of NHL. Epstein–Barr virus (EBV) is a major co-factor in many types of lymphomas; infection causes B-cell proliferation that may predispose to the formation of malignant lymphoid clones, especially in immunocompromised hosts. Other infectious agents (notably Helicobacter pylori and HTLV-1) that are associated with the development of specific lymphoma subtypes are discussed elsewhere. Finally, an increased incidence of NHL has been associated with a wide range of environmental factors including diet, radiation, pesticides, chemicals and drugs, although the causal link between most of these and lymphoma development is questionable.2
Classification NHL classification has undergone several changes over the last 30 years. In the earliest systems (for example Rappaport), lymphomas were classified entirely on morphological features. Later systems (Kiel and Lukes–Collins) attempted to correlate lymphomas to their normal counterparts, and all three systems were in use into the 1980s. The Working Formulation was introduced in 1982 in an attempt to standardized NHL classification. However, the utility of this system was hampered by making no distinction between B-cell and T-cell lineage tumours and as the understanding of lymphoma biology progressed, an increasing number of subtypes became unclassifiable, leading to the development of the REAL classification in 1994 incorporating morphological, immunological and clinical information. The current WHO classification introduced in 20013 is built on this foundation. It divides NHL primarily according to B-cell
1046 Non-Hodgkin’s lymphoma
Figure 43.1 Estimated 2002 world-wide age-adjusted incidence of non-Hodgkin’s lymphoma (males, all lymphoma subtypes) per 100 000.233
Male
Rate/cases per 100 000
25
Female 20
15
01
99
20
97
19
95
19
93
19
91
19
89
19
87
19
85
19
83
19
81
19
79
19
77
19
75
19
19
19
73
10
Year 120 Female
80 60 40 20
5– 10 9 –1 15 4 –1 20 9 –2 25 4 –2 30 9 –3 35 4 –3 40 9 –4 45 4 –4 50 9 –5 55 4 –5 60 9 –6 65 4 –6 70 9 –7 75 4 –7 80 9 –8 4 84 +
4
0
1–
Rate/cases per 100 000
Male 100
Age/year
Figure 43.2 Incidence of non-Hodgkin’s lymphoma by year of diagnosis and patient age. Upper panel shows crude incidence of NHL by year of diagnosis. Lower panel shows age-adjusted incidence of NHL in male and female patients by age at diagnosis. Data are for all NHL subtypes in the USA.234
Epidemiology, classification and staging of non-Hodgkin’s lymphoma 1047
or T-cell lineage, further subdivided into tumours with an immature or blastic appearance versus those developing from more mature stages of lymphoid development (Table 43.1). The majority of lymphomas are of B-cell lineage with over half of the cases of NHL representing cases of diffuse
large B-cell or follicular lymphoma (Fig. 43.3). Although the WHO system represents the current standard world-wide for NHL classification, some of the features from earlier systems remain of use for clinical practice, and in particular the division into high-grade and low-grade disease remains relevant
Table 43.1 WHO classification of lymphoid neoplasms B-cell lymphoma (85%)
T-cell lymphoma (15%)
Precursor B-cell Precursor B-lymphoblastic leukaemia/lymphoma
Precursor T-cell Precursor T-lymphoblastic leukaemia/lymphoma
Mature (peripheral) B-cell B-cell chronic lymphocytic leukaemia/small lymphocytic lymphoma B-cell prolymphocytic leukemia Lymphoplasmacytic lymphoma Splenic marginal zone B-cell lymphoma (with or without villous lymphocytes) Hairy cell leukemia Plasma cell myeloma/plasmacytoma Extra-nodal marginal zone B-cell lymphoma (with or without monocytoid B cells) Nodal marginal zone B-cell lymphoma (with or without monocytoid B cells) Follicular lymphoma Mantle cell lymphoma
Mature (peripheral) T-cell T-cell prolymphocytic leukemia
Diffuse large B-cell lymphoma Burkitt lymphoma/Burkitt cell leukemia
T-cell granular lymphocytic leukemia Aggressive NK-cell leukemia Adult T-cell lymphoma/leukemia (HTLV-1 positive) Extra-nodal NK/T-cell lymphoma, nasal type Enteropathy-type T-cell lymphoma Hepatosplenic γδ T-cell lymphoma Subcutaneous panniculitis-like T-cell lymphoma Mycosis fungoides/Sézary syndrome Primary cutaneous CD30-positive T-cell lymphoproliferative disorders Blastic NK-cell lymphoma Peripheral T-cell lymphoma, unspecified Angioimmunoblastic T-cell lymphoma Anaplastic large cell lymphoma
35
25 20 15 10 5
O th er
iff u B- se l ce ar ll ge Fo llic ul ar M ar gi na lz on Pe e rip he ra lT Sm -c al el ll l ym p C h LL o cy tic / M an tle An ce ap ll la s ce tic ll lar ge an B d u TBu rk ce rk itt ll itt ly lik m e ph ob Ly la m st ph ic op la sm ac yt ic
0
D
Percentage of total cases
30
Figure 43.3 Incidence of non-Hodgkin’s lymphoma by histological subtype according to the WHO classification.3
1048 Non-Hodgkin’s lymphoma
as a starting position in the approach to treatment. This chapter covers the management of both common and uncommon types of NHL, although the management of two subtypes that occur relatively frequently in the WHO classification (small lymphocytic lymphoma and lymphoblastic lymphoma) are not considered as they are biologically best considered as variants of chronic lymphocytic or acute lymphoblastic leukaemia respectively, and should be treated accordingly.
Ann Arbor staging system Although originally developed for staging in patients with Hodgkin lymphoma, the modified Ann Arbor system provides a basis for staging in NHL (Table 43.2).4 It is limited by the high frequency of extra-nodal disease in NHL compared to Hodgkin lymphoma, resulting in a disproportionate Table 43.2 Ann Arbor staging system for lymphoma with Cotswold Modifications Stage
Feature
I
Involvement of a single lymph node region or lymphoid structure (e.g. spleen, thymus, Waldeyer’s ring) Involvement of two or more lymph node regions on the same side of the diaphragm Involvement of lymph regions or structures on both sides of the diaphragm Involvement of extra-nodal site(s) beyond that designated E. Any involvement of liver, lung or bone marrow
II
III IV
Suffix A B E (stages I–III only)
Cotswold modifications
No symptoms Fever (38°C), drenching sweats, weight loss (10% body weight) over 6 months Involvement of a single, extra-nodal site contiguous or proximal to nodal site of disease Massive mediastinal disease defined as transverse mass diameter: internal thoracic diameter ratio 33% The number of lymph node fields involved are indicated by a subscript (e.g., II3) Stage III may be subdivided into stage III1, (splenic, hilar, coeliac, or portal nodes) and stage III2 (para-aortic, iliac or mesenteric nodes) Staging should be identified as clinical stage (CS) or pathologic stage (PS) Response category (CRu) defines persistent radiological abnormalities of uncertain significance
number of patients having stage IV disease due to bone marrow involvement. Despite these shortcomings, however, it is the most frequently used means of staging in adult patients with NHL and with few exceptions has been universally adopted. Current practice is that all patients should be staged by a combination of physical examination, whole body CT scanning, and bone marrow trephine biopsy. There has been much recent interest in the role of positron emission tomography (PET) scanning in staging (discussed in Chapter 6) although this is not yet performed routinely world-wide. The Ann Arbor system divides patients into four groups based on the anatomical extent of their disease, with the suffix B used to indicate the presence of unexplained fever, drenching night sweats or weight loss of greater than 10 per cent in the previous 6 months. Staging is performed with regard to groups of nodes in anatomically defined fields rather than individual nodes and unlike Hodgkin lymphoma it is not uncommon for NHL to involve non-contiguous nodal regions. Localized extra-nodal disease is designated with the suffix E, and involvement of the spleen with the suffix S, whilst involvement of the bone marrow, liver or lung is always stage IV.
AGGRESSIVE NON-HODGKIN’S LYMPHOMA Diffuse large B-cell lymphoma EPIDEMIOLOGY AND HISTOLOGICAL SUBTYPES
Diffuse large B-cell lymphoma (DLBCL) accounts for 30–40 per cent of all cases of NHL making it the most common histological subtype.1,5 The overall incidence is 2–8 cases per 100 000 person-years1,5 and although the median age at diagnosis is 64 years it occurs over a broad age range and may occur in children.6 The incidence is slightly increased in males (male:female relative risk 1.5–1.8) and Caucasians (white:black relative risk 1.5–1.8), and is significantly increased in the elderly, rising to over 50 per 100 000 in patients over the age of 75.1 In common with other types of NHL, the incidence of DLBCL rose in patients of all ages between the 1970s and 1990s.2 However, more recent epidemiological data suggest that during the last decade, although the incidence continues to rise in the elderly, it appears to be declining in younger individuals.1 Histologically, DLBCL displays marked heterogeneity. The WHO classification recognizes six morphological variants (centroblastic, immunoblastic, anaplastic, T-cell/histiocyte rich, plasmablastic and ALK positive anaplastic), and two subtypes (mediastinal/thymic and intravascular). Amongst the morphological variants, those displaying plasmablastic or immunoblastic morphology are thought by some to pursue an aggressive clinical course and respond poorly to treatment,7–9 and it is unclear whether the other histologies are of prognostic significance.10,11 The DLBCL subtypes may be distinguished by their clinical presentation, morphology and genetics. Mediastinal
Aggressive non-Hodgkin’s lymphoma 1049
large B-cell lymphoma (MBL) predominantly affects young women with a median age of presentation of 30–35. Clinically, it typically presents with a rapidly growing mediastinal mass with frequent extension into local structures, although more distant spread is uncommon.12 Histologically, it is believed to arise from thymic medullary B cells13 with a diffuse growth pattern with variable degrees of sclerosis. MBL also has a distinct immunophenotype and genetic expression profile. Tumour specimens typically express CD23, consistent with its thymic B-cell origin, with a molecular signature that is distinct from DLBCL but overlaps with that of nodular sclerosing classical Hodgkin lymphoma.14–16 The intravascular variant of DLBCL is very uncommon and characterized by the presence of lymphoma cells only in the lumina of small blood vessels. Two patterns of disease have been described; the first occurring mainly in the elderly with disseminated extra-nodal disease predominantly affecting the skin and central nervous system, and a second pattern most frequently described in Japanese patients, characterized by involvement of the bone marrow, liver and spleen and a haemophagocytic syndrome (the so-called Asian variant).17,18 With the exception of disease confined to the skin, both patterns are associated with a poor prognosis. PROGNOSTIC FACTORS
The International Prognostic Index (IPI) has been established for more than a decade as the most widely used prognostic system for patients with aggressive lymphoma (Table 43.3).19 It is based on five clinical variables – age, Ann Arbor stage, serum lactate dehydrogenase (LDH), performance status and the number of extra-nodal sites involved – allowing subdivision of patients according to the number of prognostic factors into low risk (none or one factor), low-intermediate risk (two factors), highintermediate risk (three factors), or high risk (four or five factors) with predicted 5-year survival of 73 per cent, 51 per cent, 43 per cent, and 26 per cent respectively. Initially developed with reference to intermediate grade disease within the working formulation, and for all stages of disease, it has subsequently been validated in patients with DLBCL.20 An age-adjusted IPI is also widely used to take into account the different outcomes following treatment of younger and older patients,19 and a stage modified IPI has been proposed more recently for patients with limited stage disease.21 Although the IPI is a useful means of stratifying patients according to their prognosis, there is a marked variability of response to treatment within the IPI subgroups. Moreover, treatment intensification for patients with an adverse prognosis based on high IPI score does not appear to result in improved outcome.22 One explanation for these observations is that the IPI does not reflect the heterogeneity of the underlying disease, prompting attempts to identify biological determinants that will better determine the response to specific types of treatment.23,24 Numerous studies have attempted to associate the patient survival with expression
of individual biomarkers associated with cell cycle regulation, apoptosis, differentiation, angiogenesis and cell–cell interactions and although each has contributed to the understanding of the pathobiology of the disease, none has yet emerged as a reliable prognostic marker.23,24 Genome-wide expression profiling using microarray technology represents a means of analysing the expression of mRNAs within tumour samples to develop a molecular fingerprint that is independent of aberrant expression of individual proteins. When this approach was applied to DLBCL, two disease subgroups were initially identified with gene expression patterns characteristic of either germinal centre (GC) cells or activated B-cells (ABC), with the subsequent identification of a third molecular signature characteristic of primary mediastinal disease.15,16,25,26 Patients with a GC phenotype had a significantly improved 5-year survival compared to those with an ABC phenotype (76 per cent vs. 26 per cent in the original study) that was independent of their IPI.25,26 The reason for the different prognosis between these disease groups is unclear and although gene expression profiling defines biologically distinct subtypes of DLBCL, it does not identify the contribution of individual gene products to the tumour phenotype; furthermore it is not yet a suitable tool for routine diagnostic use. Analysing the expression of the products of a limited range of genes identified from the original microarray
Table 43.3 International Prognostic Index (IPI) Risk factors
CR (%)
Five-year PFS (%)
Five-year OS (%)
*All patients Low Low-intermediate High-intermediate High
0,1 2 3 4,5
87 67 55 44
70 50 49 40
73 51 43 26
§Age-adjusted (age ⬍60) Low Low-intermediate High-intermediate High
0 1 2 3
92 78 57 46
86 66 53 58
83 69 46 32
†Stage-adjusted (stage I/II) Low Intermediate High
0,1 2 3,4
– – –
77 60 34
82 71 48
*Risk factors: Age 60, raised LDH, performance status 2, stage III, IV disease, more than 1 extranodal site of disease. §Risk factors: Raised LDH, performance status 2, stage III, IV disease. †Risk factors: Age 60, raised LDH, performance status 2, stage II disease. Abbreviations: CR, complete remission; PFS, progression-free survival; OS, overall survival.
1050 Non-Hodgkin’s lymphoma
analysis is one means by which the relative contribution of each to overall prognosis can be determined.26–28 However, this approach is not part of routine clinical practice and it is as yet unclear how this information should be used to inform decisions regarding treatment and whether it will translate into improved outcome. TREATMENT OF NEWLY DIAGNOSED ADVANCED-STAGE DISEASE
Following its description in the mid-1970s,29 treatment with 6–8 cycles of CHOP chemotherapy remained the standard of care for patients with advanced DLBCL (stage IB, bulky stage II, stage IIB and stages III–IV) for the next 25 years. Although such treatment is potentially curative, this is only achieved in a minority of patients with a favourable prognosis. The overall complete remission rate is only 35–50 per cent with a 5-year projected survival of less than 40 per cent.6,30 A number of newer ‘third generation’ regimens developed in the 1980s showed promising results in single-centre phase II studies. However, despite the suggestion of an advantage for MACOP-B type chemotherapy in the treatment of the mediastinal/thymic variant or the addition of etoposide to CHOP (CHOEP) for treating patients under 60 with DLBCL31–33 none has been convincingly shown to be superior to CHOP, indicating the need for alternative treatment strategies.30 Rituximab is a humanized mouse monoclonal antibody directed against CD20, a transmembrane protein expressed on more than 90 per cent of B-cell lymphomas, and has multiple anti-lymphoma activities including complement and antibody-dependent cytotoxicity, inhibition of proliferation and induction of apoptosis.34 When used alone in the treatment of relapsed DLBCL, the effects of rituximab are fairly modest.35 However, the results of a pivotal study using rituximab-based immunochemotherapy (rituximab plus CHOP – R-CHOP) for the treatment of newly diagnosed patients aged 60–80 demonstrated a significant improvement in complete response rate (76 per cent vs 63 per cent)36 and 5-year overall survival (58 per cent vs 45 per cent), compared to CHOP alone.37 This benefit was seen in both good and poor risk disease, however the greatest benefit was seen in the low IPI better prognosis group, and importantly, there was no increase in the treatment related toxicity with the addition of rituximab. A similar benefit was shown subsequently for treating younger patients with good prognosis disease (MiNT study).38 Dose intensification by means of either increased drug dose (dose escalation), early autologous stem cell transplantation (ASCT) or reduced treatment interval (dose density) has been investigated as potential means to improve the treatment outcome in DLBCL. Dose escalation is feasible although it typically results in increased treatment related toxicity.39 A recent large randomized study comparing the intensive ACVBP regimen with CHOP in poor-prognosis patients showed a survival advantage in the ACVBP arm,40 although other studies have
failed to show any advantage for alternative dose-escalated regimens.39 With the exception of DLBCL arising from transformed low-grade disease (see Follicular lymphoma below), conflicting results have also been obtained using ASCT to consolidate first remission. ASCT following attenuated induction chemotherapy appears equivalent at best compared to conventional chemotherapy alone,41–43 and there does not appear to be any advantage in using stem cell transplantation to consolidate first remission following standard chemotherapy in patients with high-risk disease.44–46 The results of a study from the mid-1990s demonstrating improved remission rates and event-free survival using intensive induction chemotherapy followed by stem cell transplantation compared with standard chemotherapy for treating patients with newly diagnosed high-grade NHL have prompted more recent interest in this approach.47 Three subsequent studies have also demonstrated a benefit of ASCT after dose-escalated induction chemotherapy in patients under 60, suggesting an advantage of early dose intensity,48–50 although other studies failed to show a similar advantage.51 Moreover, whether the possible advantage would still be apparent if rituximab was added to the induction chemotherapy is unknown. The mixed outcome following using high-dose therapies has focused more recent attention on the use of dosedense regimens. A large study of more than 1500 patients demonstrated that CHOP could be safely delivered on a 14-day schedule (CHOP-14) with growth factor support in patients up to 75 years old with improvements in the remission rate in the elderly and survival in patients of all ages.33,52 The addition of rituximab to dose-dense CHOP (R-CHOP-14) further improves the time to treatment failure in elderly patients.53 However, it is not yet known whether the long-term outcome will be better than that following R-CHOP-21, and several large studies are currently underway to investigate the use of dose-dense CHOP combined with rituximab in patients of all ages. Based on the current evidence, combination therapy with 6–8 cycles of R-CHOP is the standard treatment for advanced-stage DLBCL in patients up to 75 years. Provisional results in the elderly indicate that the outcome may be improved by using dose-dense regimens although this has yet to be validated in the rituximab era or in younger patients. For young patients with poor-risk disease, there is a possible advantage of using intensive treatment with or without consolidation ASCT. However, until the results of studies combing such dose-escalated chemotherapy with rituximab are available, this type of treatment should only be considered as part of a clinical trial (Fig. 43.4). TREATMENT OF NEWLY DIAGNOSED LIMITED-STAGE DISEASE
For patients with non-bulky limited-stage disease, standard treatment for much of the last two decades has been
Aggressive non-Hodgkin’s lymphoma 1051
Diagnosis of diffuse large B-cell lymphoma
Advanced stage Bulky stage I–II§, B–symptoms, stage III–IV
Limited-stage disease
Fit for anthracycline-based chemotherapy?
Fit for anthracycline-based chemotherapy?
• Performance status <2 • Adequate organ function • (Age <75)
• Performance status <2 • Adequate organ function • (Age <75)
Yes
No
Yes
No
R-CHOP 6–8 Consider • Clinical trial • Dose-dense R-CHOP • Dose-intense regimen plus autologous stem cell transplant <60 years • PET scanning – interval or at completion of treatment
Patient dependent Consider • Clinical trial • R-CVP 6–8 or similar
CHOP 4 plus involved field radiotherapy (RT) Consider • Clinical trial • RCHOP minus RT in lowrisk disease • Dose-intense regimen minus RT if <60 years • PET scanning – interval or at completion of treatment
Patient dependent Consider • Clinical trial • CVP 4 plus involved field RT or similar • Radiotherapy alone
PR/refractory
CR
Relapse
Observe
Salvage chemotherapy and autologous stem cell transplant if chemosensitive disease, age <70 and adequate organ function Consider • Gemcitabine, clinical trial or palliative care if not suitable for stem cell transplant • Allogeneic transplant if <35
PR/refractory
Patient dependent Consider • Gemcitabine • Clinical trial • Palliative care • Allogeneic transplant
PR/refractory/ relapse
CR
Treat as for advanced disease
PR/refractory/ relapse
Treat as for advanced
Observe
Key §
Bulky disease
Relapse CR, PR
Maximum nodal diameter greater than 10 cm, mediastinal ratio >0.33. Radiotherapy should be considered in addition to above to sites of bulky disease Complete remission, partial remission
Figure 43.4 Suggested treatment algorithm for the management of diffuse large B-cell lymphoma.
attenuated CHOP chemotherapy (3–4 cycles) plus involved field therapy.54 This results in long-term survival of around 80 per cent, and is superior to that observed in historical studies using radiotherapy alone.55 However, evidence is emerging to suggest that radiotherapy may not be necessary in all patients. First, in two large randomized
studies, the improved disease control obtained when radiotherapy was added to either three or eight cycles of CHOP did not translate into a long-term survival advantage.21,56,57 Second, in a smaller study of elderly patients with low-risk disease, radiotherapy did not improve the outcome following attenuated CHOP chemotherapy.58 Third, in younger
1052 Non-Hodgkin’s lymphoma
patients, the outcome following aggressive chemotherapy using ABCVP plus sequential high-dose treatment was superior to combined modality with CHOP and radiotherapy, although this may reflect the inclusion of high-risk patients with bulky disease in this study.59 It is difficult to draw firm conclusions from these data, however they suggest a possible risk-stratified approach to treatment of limited-stage DLBCL (Fig. 43.4). Patients with bulky disease (maximum diameter 10 cm) or B symptoms (irrespective of disease stage) should be regarded as having advanced disease and treated with a minimum of six cycles of R-CHOP (21-day cycle), and consideration given to the addition of consolidation radiotherapy to any sites of disease bulk. In elderly patients with early-stage disease and no other adverse prognostic factors, attenuated chemotherapy using CHOP and involved field radiotherapy should still be regarded as the standard of care as it offers good local disease control with minimal systemic toxicity. In younger patients, more aggressive chemotherapy may offer a longer-term prospect of cure although further studies are needed to confirm this. In very low-risk patients of all ages, it is possible that radiotherapy is unnecessary, and the use of more effective systemic therapies should be explored in this group, perhaps with the use of rituximab. However, although rituximab appears to be effective in limited-stage DLBCL,60 mirroring its effect in advanced disease, there is a paucity of data available to support its routine use. TREATMENT OF DLBCL IN THE ELDERLY
More than half of the patients presenting with DLBCL are over 60 years old and have a worse outcome following chemotherapy than younger individuals.19 The reason for this difference is likely to be multifactorial, related to chronological age, co-morbidity, performance status, achievable dose intensity and physiological reserve.61 However, chronological age per se only becomes the major determinant of outcome above 75 years,62 suggesting that patients between 60 and 75 years with good performance status and adequate organ function should be treated as younger individuals, borne out by the successful use of standard and dose-dense R-CHOP in this age range.37,52,53 For elderly patients unfit for CHOP type chemotherapy, there is little consensus on the optimal treatment. CVP/COP (CHOP with omission of the doxorubicin) is generally well tolerated and is used in this patient group although a wide range of other regimens have also been described.61 It is likely that some of the decreased efficacy of CVP compared to CHOP63 may be offset by the addition of rituximab, although there are currently no published data to support this. CENTRAL NERVOUS SYSTEM (CNS) LYMPHOMA AND PROPHYLACTIC CNS CHEMOTHERAPY
Primary CNS lymphoma (PCNSL) is rare, accounting for only 1–2 per cent of malignant lymphoma. It is more
common in the elderly and in younger patients with HIV/AIDS, and the large majority of PCNSL represent cases of DLBCL.64 PCNSL has a poor prognosis and the optimal treatment is unclear. Historically, the standard approach has been radiotherapy, although this results in a median survival of less than 1 year and is associated with a significant incidence of permanent neuro-cognitive dysfunction, particularly in elderly patients.65 High-dose methotrexate-based chemotherapy in combination with radiotherapy is more effective than radiotherapy alone, although it is also associated with significant neurological toxicity,64 prompting investigations of the use of chemotherapy alone. Methotrexate at doses of 3–8 g/m2 is well tolerated as a single agent although mixed results have been reported concerning its efficacy, and the optimal dose and scheduling to maximize the response also have not been clearly defined.66–68 More recently, promising results have been described using combinations of CNS-penetrating chemotherapy drugs, however the use of these regimens may be limited by systemic toxicity, especially in the elderly.69,70 Secondary CNS disease has been described in association with most lymphoma subtypes. It is most frequently observed in B-cell NHL with an incidence that varies by histological grade, ranging from under 3 per cent in lowgrade disease to 25 per cent in Burkitt’s lymphoma, and complicates around 5 per cent of cases of DLBCL.71,72 Patients presenting with CNS involvement have a poor prognosis, and there is no consensus on treatment although this normally includes CNS-directed chemotherapy with or without radiotherapy.73 The results of treatment with autologous stem cell transplant are more encouraging although this is only possible in a selected group of patients.73 With regard to strategies used to prevent CNS relapse, there is universal agreement that lymphoblastic lymphoma and Burkitt’s lymphoma require CNS-directed therapy using both a combination of systemic and intrathecal chemotherapy, however there is less consensus regarding management in DLBCL.72,74 The occurrence of CNS relapse in DLBCL varies according to the presence of risk factors, including disease site (increased risk with extra-nodal disease – paranasal sinuses, epidural space, testicles, breast and bone marrow), serum LDH and IPI score.71,72 CNS prophylaxis is typically reserved for at-risk patients and is most commonly administered as intermittent intrathecal injections of methotrexate,74,75 although whether this is effective is unclear.76,77 An alternative strategy is to use chemotherapy regimens incorporating CNSpenetrating drugs such as the ACVBP plus sequential methotrexate, ifosfamide and Ara-C (cytarabine) (MIA) intensification regimen developed by the Groupe d’Etudes des Lymphomes de l’Adulte (GELA).78 A comparison of the outcome of treatment of high-risk DLBCL using either CHOP or ACVBP showed a significantly reduced rate of CNS relapse in the ACVBP arm,40 although this advantage was offset by increased regimen-related toxicity, suggesting that this approach may only be appropriate for a subset of high-risk patients.
Aggressive non-Hodgkin’s lymphoma 1053
TREATMENT OF RELAPSED DISEASE
The Parma study in the mid-1990s established autologous stem cell transplantation as the standard treatment for relapsed chemotherapy sensitive DLBCL, resulting in longterm disease-free survival in 45 per cent of patients,79 with the degree of response to salvage chemotherapy and duration of first remission (more or less than 1 year) being predictive of outcome. A number of salvage chemotherapy regimens have been used prior to ASCT, resulting in response rates of 43–63 per cent (Table 43.4).44,80–83 The role of rituximab in salvage chemotherapy is unclear, in particular whether there is any benefit in re-treatment or whether it improves the outcome following ASCT. The addition of rituximab to ICE chemotherapy prior to ASCT results in improved complete remission rate compared to historical controls although this does not translate into a long-term survival benefit.84 It is hoped that the results of current phase III studies comparing the outcome of different salvage regimes incorporating rituximab will better define the role of immuno-chemotherapy as salvage therapy. For patients refractory to salvage chemotherapy or relapsed post ASCT, there is little benefit in re-treatment with an alternative conventional regimen,85 and the prognosis in this group is poor. Possible therapeutic options that have shown promise in early-phase trials include nucleoside analogues (gemcitabine or clofarabine), radioimmunotherapy using 90Y-ibritumomab tiuxetan, and inhibitors of histone deacetylase or angiogenesis although their use is currently experimental. Allogeneic transplantation is possible if a further remission can be achieved using either conventional chemotherapy agents and is discussed below.
Burkitt lymphoma Burkitt lymphoma (BL) is a highly aggressive variant of B-cell NHL accounting for 1–2 per cent of cases in Europe and the USA.3 It typically presents as a rapidly growing tumour with frequent involvement of extra-nodal sites, CNS involvement being particularly common. Three clinical
Table 43.4 Salvage chemotherapy regimens used in high-grade lymphoma Regimen ESHAP DHAP miniBEAM IVE ICE
Reference
Number of patients
Response rate
Velasquez et al.81 Philip et al.79 Girouard et al.83 Proctor et al.82 Moskowitz et al.80
122 215 103 61 163
64% 48% 43% 43% 63%
variants are recognized. The sporadic variant occurs most frequently in children and young adults and has a worldwide distribution, whereas endemic BL disease occurs in equatorial Africa and Borneo, predominantly in children, and in association with chronic infection (viral – notably EBV and HIV, bacterial and parasitic). A third variant occurs in immunosuppressed individuals, primarily in association with HIV infection.3 The disease shows significant morphological heterogeneity that overlaps with the appearances of DLBCL; however, consistent findings are a proliferation fraction of greater than 99 per cent, absent expression of TdT and BCL2, and chromosomal rearrangements involving the c-myc locus on chromosome 8, with a t(8;14) translocation present in over 80 per cent of cases.3,86 Because of the high frequency of extra-nodal disease, it is often difficult to stage BL accurately according to the Ann Arbor scheme. Although this remains widely used, particularly in adult patients, an alternative staging system has been proposed87 and is recommended for use within the WHO classification.3 However, it should be noted that this revised system was developed for use in children, in an era when surgical resection of abdominal disease was commonplace, and its utility with more modern treatment protocols is now questionable. With regard to prognosis, there is no universally accepted prognostic scoring system. Advanced stage (Ann Arbor III–IV), raised LDH, bulky disease (maximum diameter 10 cm), older age, CNS and bone marrow involvement have been associated with a poor prognosis88–90 although there is a lack of consensus regarding which of these factors are of greatest significance. The IPI has not been validated in BL although there is a trend towards a worse outcome in patients with a high IPI score,91 and an alternative system based on four risk factors (LDH, Ann Arbor stage, performance status and tumour bulk) defining high- and low-risk groups has been adopted for use in some studies.91,92 Historically, the outcome of treating BL, predominantly using regimens designed for treatment of acute lymphoblastic leukaemia, was poor. However, more recent treatment strategies using fewer cycles of high-intensity treatment incorporating CNS-directed therapy have transformed the prognosis. In adults with advanced disease, the outcome following the CODOX-M/IVAC and HyperCVAD protocols developed in the USA,90–92 or European LMB 81–89 and BFM 84–86 regimens93,94 is similar, resulting in remission rates of 70–90 per cent and overall survival of 50–70 per cent.86 All are associated with significant treatment-related toxicity; tumour lysis may complicate the early stages of treatment and grade 3–4 cytopenias occur in the majority of patients, factors that may limit the use of this type of treatment in older patients. There are also no data available directly comparing any of these regimens, and differences between the ages and disease status of the patients in each study make it difficult to infer a superiority of one regimen over another. The outcome in limited-stage disease is better, with 2–5-year survival rates
1054 Non-Hodgkin’s lymphoma
of 80–95 per cent,89,92 and for patients with early-stage disease with no adverse risk factors, excellent results can be obtained using attenuated treatment regimens suggesting that more intensive therapy is unnecessary in this group.91,92 Whether the outcome in BL of all stages can be further improved by the addition of rituximab to chemotherapy is unknown. The majority of cases express CD203 and there are preliminary data to suggest that rituximab may improve the outcome following treatment with HyperCVAD chemotherapy, although the follow-up period is currently relatively short.95 Several studies are ongoing to investigate this approach using other treatment protocols.
HIV-associated lymphoma Non-Hodgkin’s lymphoma is a late manifestation of infection with HIV. The vast majority of cases are high-grade B-cell lymphomas, mainly DLBCL or Burkitt lymphoma, whose incidence in AIDS patients is increased 150-fold and 650-fold respectively.96 HIV-associated NHL typically occurs with peripheral blood CD4 counts of less than 0.2 109/L, with the exception of Burkitt lymphoma that is frequently seen with CD4 counts above this level.97 There is also an increased incidence of rare lymphomas seen infrequently outside the HIV-positive population.98 Primary effusion lymphoma presents with neoplastic effusions in pleural, pericardial and peritoneal cavities typically without a contiguous tumour mass,3 and is universally associated with human herpes virus 8 (HHV-8) infection.99 Plasmablastic lymphoma was originally described in HIV-positive patients with a limited distribution, affecting the oral cavity and jaw,100 although more recently cases have been reported in HIV-negative individuals and with extra-nodal disease at other sites.101 Both of these uncommon lymphomas are highly aggressive, responding poorly to chemotherapy, and have a poor prognosis.102 For the more commonly occurring histologies, the advent of highly active anti-retroviral therapy (HAART) has been marked by both a significant fall in the incidence and an improvement in the prognosis of HIV-associated NHL, although the optimal treatment has not yet been determined.102 Historically, the outcome following chemotherapy was poor and associated with significant treatmentrelated toxicity, prompting the use of low-intensity regimens which, although better tolerated, resulted in a short overall survival.103 However, since the introduction of HAART, the results of using standard chemotherapy have improved markedly.102 Standard dose CHOP chemotherapy combined with HAART and haemopoietic growth factor support is feasible, and results not dissimilar to those obtained in HIVnegative patients have been reported.104 The typically aggressive histologies seen in HIV-associated lymphomas have also prompted the use of more intensive chemotherapy regimens. In a small series there were no differences in survival between HIV positive and negative patients
treated according to the CODOX-M/IVAC protocol for Burkitt’s lymphoma,105 and the results of dose-adjusted infusional chemotherapy (EPOCH) for HIV-associated DLBCL are impressive, with a remission rate and 5-year survival of 74 per cent and 60 per cent respectively.106 Whether all patients with HIV-related lymphoma will benefit from aggressive chemotherapy remains undetermined, and a number of lines of evidence suggest that the HIV disease status and response to HAART may also be important determinants of outcome. First, the response to antiviral therapy appears to predict the response to CHOPlike chemotherapy,107,108 although this relationship is inconsistent and not observed in a more recent study.109 Second, if patients are risk stratified according to an HIV score (CD4 count 100 109/L, performance status 2, or previous AIDS) there is no benefit in aggressive over low-dose chemotherapy for poor-risk patients irrespective of the use of HAART.110 Third, the outcome following EPOCH chemotherapy was significantly worse in patients with a CD4 count 100 109/L,106 due in part to patients in this study not receiving antiviral treatment until completion of chemotherapy.106 Finally, the addition of rituximab to chemotherapy for HIV-associated lymphoma, based on its success in non-HIV-associated disease, results in significantly greater toxicity,111,112 mainly due to infections in patients with a CD4 count under 50 109/L, that offsets any advantage gained in improved remission rates.112 In conclusion, although intensive combination chemotherapy is feasible since the introduction of HAART and effective in many patients, it should be used with caution in those refractory to antiviral treatment or with advanced HIV disease.
Aggressive peripheral T-cell lymphomas Peripheral T-cell lymphomas (PTCL) account for around 10 per cent of cases of NHL.6 They are biologically diverse, making classification difficult, and in the WHO classification they are classified into three main groups according to the predominant site of disease: leukaemic, nodal and extra-nodal.3 Despite their heterogeneity, they share some common features: having a post thymic origin, being relatively uncommon, histologically aggressive, and with few exceptions having a worse prognosis than similar B-cell diseases.113–115 Identification of reliable prognostic factors in PTCL has been hampered by the rarity of many of the subtypes and difficulties in disease classification. The IPI cannot reliably be used to predict outcome in PTCL as a whole, although it has been validated in specific subtypes including anaplastic large cell lymphoma (ALCL) and PTCL-unspecified.116–118 As with DLBCL, attempts have also been made to identify biomarkers with prognostic significance, amongst which expression of the anaplastic lymphoma kinase (ALK) gene product in ALCL is best characterized. Gene rearrangements of the ALK locus on chromosome 2, most commonly as a
Aggressive non-Hodgkin’s lymphoma 1055
result of a t(2;5) translocation,119 result in ALK overexpression being observed in up to half of cases of ALCL in children and young adults, although it occurs much less commonly in the elderly, and it is associated with a significantly better prognosis than ALK-negative cases.120 More recently the results of gene expression profiling suggest that PTCL may be classified according to groups with a different molecular signature,121–123 although whether this information will identify molecular targets for treatment and translate into improved survival is unknown. ANAPLASTIC LARGE CELL LYMPHOMA
ALCL accounts for around 3 per cent of cases of NHL and has a bimodal incidence being seen most frequently in children and young adults under 30 with a second smaller peak in the elderly.3 Based on clinical and pathological features, it may be classified into three groups. Cutaneous ALCL is a more indolent variant predominantly seen in the elderly and described elsewhere. ALK-positive ALCL occurs predominantly in children and young adults with frequent extra-nodal involvement,118,124 and has a good response to treatment whereas ALK-negative ALCL occurs in older adults with less frequent extra-nodal disease and has a poor prognosis.118 The mainstay of treatment of adults with systemic ALCL is CHOP-type chemotherapy and although this remains widely used, the overall long-term survival is poor at less than 40 per cent.118,120,125 If patients are grouped according to ALK expression status, the outcome is better in the ALK-positive group with 5-year overall survival figures of 60–90 per cent, amongst whom those with a low IPI have a particularly favourable prognosis.118,120,125 More intensive chemotherapy regimens have been used to try and improve on the results obtained using CHOP, especially in ALK-negative patients, although none of these appears to result in a survival advantage.126 In children, the majority of tumours are ALK positive and the overall prognosis is better with a 5-year survival of 80–85 per cent and in contrast to adult patients, the results following chemotherapy are improved by using intensive chemotherapy regimens.124,127 The outcome of autologous stem cell transplantation in ALCL has been reported in several series. The long-term overall survival following ASCT in PTCL is 35–55 per cent128–130 with most series reporting improved outcomes in patients with ALCL.128–131 It is, however, difficult to compare the results of these studies: all are fairly small, few differentiate between ALK positive and negative cases and there is wide variation in the timing of transplants reported (first remission or beyond). The best results described have, perhaps unsurprisingly, been obtained in young patients, resulting in long-term remission rates of over 80 per cent.130,131 It is, however, difficult to recommend early transplant in this patient group given the good results following chemotherapy. In adult patients, the overall results of ASCT appear superior to chemotherapy alone, although it is unclear whether it should be performed in
first remission or reserved for salvage therapy.128,130,131 The outcome in the worst-risk patients (refractory disease, ALK-negative and high IPI) remains poor irrespective of whether they are transplanted, highlighting the need for novel treatments in this group. ANGIOIMMUNOBLASTIC T-CELL LYMPHOMA
Angioimmunoblastic T-cell lymphoma (AITL) is a clinical syndrome characterized by generalized lymphadenopathy, hepatosplenomegaly, and skin rashes often with pruritus, typically presenting in the middle-aged and elderly.3 The histological appearance and laboratory findings (including raised immunoglobulins, circulating autoantibodies and immune complexes, cold agglutinins and autoimmune haemolytic anaemia) resulted in this originally being considered as a pre-malignant reactive disorder; however, the recognition that the majority of cases contained clonal T-cell populations led to its re-classification as a lymphoma.132 Patients with AITL are typically immunosuppressed as a consequence of the disease, resulting in significant morbidity and mortality secondary to infection133,134 and less commonly the development of secondary EBV-driven B-cell lymphomas.135,136 As with the majority of PTCL, treatment is unsatisfactory and the median survival following combination chemotherapy (CHOP or other) is around 3 years.133,134 Nucleoside analogues (fludarabine and 2-chlordeoxyadenosine) may also be effective although their use has only been reported in a handful of cases.132 The results of ASCT have been reported in small numbers of patients although the results of the largest series are favourable with a durable remission rate of 40 per cent that was improved if the transplant was carried out in first remission.137 Finally, the use of immunomodulatory therapies has also been investigated due to the immune dysregulation and secondary B-cell proliferation that accompanies AITL. Thalidomide, interferon alpha, ciclosporin and rituximab have all been used with variable degrees of success132,138,139 and further studies are currently in progress to better determine their efficacy and role in treatment. HTLV-1 ASSOCIATED ADULT T-CELL LEUKAEMIA/ LYMPHOMA
Adult T-cell leukaemia/lymphoma (ATLL) is caused by the human T-cell leukaemia virus type 1 (HTLV-1) and occurs principally in areas of endemic infection (Japan, the Caribbean and parts of Africa). HTLV-1 is principally acquired in childhood via vertical transmission in breast milk although it is also transmitted via blood; around 3 per cent of infected individuals will develop ATLL after a long latency.140 The disease has a broad clinical spectrum and four variants have been described: acute, lymphomatous, chronic and smouldering, with projected 4-year survival figures of 5 per cent, 5.7 per cent, 26.9 per cent and 62.8 per cent respectively.141 The most common acute variant is characterized by circulating lymphoma cells, skin rash,
1056 Non-Hodgkin’s lymphoma
widespread lymphadenopathy, organomegaly and hypercalcaemia. Patients in this group are immunosuppressed and opportunistic infections are common. The lymphomatous variant is characterized by extensive lymphadenopathy, but involvement of the peripheral blood and hypercalcaemia are less common. The chronic variant typically presents principally with skin lesions, and the smouldering variant is characterized by low levels of circulating lymphoma cells together with lymphomatous deposits in the skin and lungs. Progression from the chronic and smouldering disease to more aggressive variants occurs in around 25 per cent of cases after a latency period often of several years. Following the original description of the disease in the late 1970s, the prognosis for the majority of patients with ATLL remained dismal for much of the next two decades, with a median survival of 6 months.142,143 Conventional chemotherapy with a CHOP-type regimen or HTLV-1 directed treatment using interferon alpha and zidovudine results in typically short-lived remissions in around 60 per cent of patients.144–147 The results of combination treatment using CHOP for remission induction and then maintenance antiviral therapy are better, resulting in remission durations of around 12–18 months.148,149 However, despite these improvements, the vast majority of patients with ATLL are destined to die from disease. Allogeneic stem cell transplantation is feasible in a minority of patients and is potentially curative, although the number of successfully treated patients is extremely small.150 Given the overall response to treatment, preventative strategies to reduce the incidence of ATLL may be more effective. Avoidance of breast feeding by infected mothers would reduced the risk of transmission by 80 per cent although the impact of this would not be seen for many years.143 OTHER PERIPHERAL T-CELL LYMPHOMAS
Peripheral T-cell lymphoma, unspecified (PTCLUS), accounts for around half of the cases of PTCL in Western countries.114,117,118 It is a heterogeneous group of diseases, mainly affecting adults, and has a predominantly nodal distribution.3 Different prognostic groups have been identified on the basis of clinical and laboratory variables together with chemokine expression patterns, although this does not currently influence treatment.116,151,152 The outcome following CHOP-type chemotherapy is variable reflecting the biological heterogeneity of the underlying disorder, ranging from under 20 per cent in the worst-risk patients to around 60 per cent in patients without adverse risk factors.116 ASCT also appears effective in PTCLUS, resulting in 5-year overall survival figures ranging from 35 to 58 per cent.128–130 However, the number of patients in each study was relatively small making it difficult to exclude selection bias and there is no consensus on the optimal timing of transplant (first remission or beyond) or which patients (low vs. high risk) benefit most from highdose therapy.128–130
Enteropathy associated T-cell lymphoma (EATL) is an uncommon lymphoma most commonly arising in lymphoid tissue associated with the small intestine. There is a clear association with coeliac disease,153 although there is often little clinical evidence of this prior to the lymphoma diagnosis. The pathogenesis of EATL is incompletely understood, however it is postulated that chronic antigenic stimulation by gliadin in the context of the susceptible HLA-DQA1*0501, HLA-DQB1*0201 haplotype seen in most patients with coeliac disease leads to the development of T-cell clones that predispose to lymphoma following additional genetic mutations.154–157 Patients often have a poor performance status at presentation, malnutrition and intestinal perforation are common, and the outcome following conventional chemotherapy is poor, with longterm remissions only achieved in 10–30 per cent.158–160 There are preliminary data suggesting that better results may be obtained from the use of more intensive chemotherapy followed by ASCT,161 and the results of larger studies to investigate this approach are awaited. Extra-nodal NK/T-cell lymphomas are uncommon in Western populations, being most frequently observed in Asia, Mexico and Central and South America. It is more common in men and frequently presents with localized extra-nodal disease typically affecting the nasopharynx.162,163 The tumours are highly sensitive to radiotherapy and relatively resistant to chemotherapy in part due to high p-glycoprotein expression.164 High remission rates may be obtained in patients with localized disease (stage IE) without extensive local tissue destruction using radiotherapy with or without chemotherapy, although relapse at distant sites occurs relatively frequently.162,163 For patients with more advanced disease, the prognosis is poor with conventional treatment although there may be a role for early allogeneic transplantation in younger patients with an HLA-matched donor.165 Hepatosplenic T-cell lymphoma (HSTL) is a recently described rare subtype of PTCL typically presenting in young male patients with hepatosplenomegaly and bone marrow infiltration. Most cases have a γδ T-cell phenotype although cases with αβ T-cell receptor rearrangements have also been described. A response to CHOP-type chemotherapy is observed in the majority of patients although this is typically short lived and there are very few reports of long-term remission.166 Allogeneic stem cell transplantation has been used successfully to treat HSTL although only reported in a handful of cases.167
INDOLENT NON-HODGKIN’S LYMPHOMA Follicular lymphoma EPIDEMIOLOGY, HISTOLOGICAL GRADING AND PROGNOSTIC FACTORS
Follicular lymphoma (FL) accounts for 20–35 per cent of all cases of NHL, and for the majority of low-grade
Indolent non-Hodgkin’s lymphoma 1057
disease.1,5 The overall incidence is 2–4 cases per 100 000 person-years and in common with other types of NHL it is somewhat more common in men.1,5 It is predominantly a disease of older adults with a median age at diagnosis of 59 years6 and although cases have been uncommonly described in children, FL in this age group is best regarded as a distinct disease entity.168 Cases of FL can be histologically graded1,2,3a,3b according to the proportion of large transformed cells (centroblasts) present in the predominant background population of smaller untransformed cells (centrocytes). Grade 1 or 2 disease is typically clinically indolent, although transformation to high-grade lymphoma (DLCBL) may occur.3 The prognosis and optimal treatment of grade 3 disease is controversial, in part due to its relative rarity and a lack of consensus on histological criteria for diagnosis in older classifications. Grade 3b disease is more aggressive; retrospective studies have generally shown that it has more in common with high-grade disease169 and current practice is that it is treated as DLBCL. There is less consensus regarding the treatment of grade 3a disease although with regard to treatment, this is most often grouped with grade 1 or 2 disease. The prognosis in FL may be predicted by a number of clinical, genetic and molecular factors. Analogous to the IPI in high-grade disease, prognostic systems have been developed for use in follicular lymphoma, of which the follicular lymphoma IPI (FLIPI) is the most recently described.170 This stratifies patients into three risk groups based on five variables: age, Ann Arbor stage, haemoglobin, LDH and the number of sites of nodal disease, with projected 5-year overall survival in the low-risk (0–1 factors), intermediate-risk (2 factors) and high-risk (3–5 factors) groups of 90.6 per cent, 77.6 per cent and 52.5 per cent respectively. Most cases of FL harbour multiple cytogenetic abnormalities, the hallmark of which is t(14;18), seen in up
to 95 per cent of cases, leading to over-expression of the anti-apoptotic protein bcl-2. The position of the bcl-2 breakpoint and blc-2 protein expression level have been correlated with outcome,171,172 whilst other chromosomal abnormalities may confer either a favourable (7, 8) or unfavourable (1q, 6q, 17p) prognosis.173 The outcome in FL may also be predicted by expression of individual molecular markers or gene expression profiling although, as with other types of NHL, it is not yet clear how this information should be applied to decisions regarding treatment.170,174 TREATMENT OF NEWLY DIAGNOSED STAGE II–IV DISEASE (GRADE 1, 2, 3A)
The majority of patients with FL present with advanced-stage (III or IV) disease and despite remission being achievable in the majority using single alkylating agents, combination chemotherapy or combined modality chemo-radiotherapy, a continuous rate of relapse is observed. For asymptomatic patients with no adverse risk factors, immediate treatment confers no advantage over conservative management in which treatment is deferred until disease progression (watchful waiting).175–177 The 5-year survival using either approach is 60–80 per cent with a median survival of 8–10 years. The last decade has been marked by considerable interest in the role of rituximab in the management of FL following the pivotal study demonstrating its activity in around 50 per cent of patients with relapsed disease.178 For patients with newly diagnosed advanced disease, its use has been explored in three main areas; initial monotherapy, in combination with chemotherapy, and as maintenance treatment (Table 43.5). Rituximab monotherapy administered as a standard schedule of 4–8 weekly doses of 375 mg/m2 results in remission rates of up to 70 per cent
Table 43.5 Rituximab monotherapy for the initial treatment of advanced follicular lymphoma Reference Induction treatment Colombat et al.179 Witzig et al.180 Piro et al.181 Induction treatment ⫹ maintenance Hainsworth et al.172 Ghielmini et al.182 Gordan et al.183
Regimen
Response rate (%)
Median TTP/EFS (months)
Weekly 4 Weekly 4 Weekly 8
73 72 60
12 26 NR (19)
Weekly 4, repeated every 6 months to 2 years Weekly 4, then single dose every 2 months 4 Weekly 4, then single dose every 3–4 months to 2 years; target level 25 μg/mL
73
34
63 (chemo naive) 46 (prior chemo) 63
19 (chemo naive) 36 (prior chemo) NR (35)
Abbreviations: NR, not reached; TTP, time to progression; EFS, event-free survival. Single dose in all regimens is 375 mg/m2.
1058 Non-Hodgkin’s lymphoma
and median progression-free survival of 1–3 years.172,179–181 Higher remission rates have been described in low-risk patients (stage I and low LDH) and also using an extended dosing schedule.180,182 Treatment is generally well tolerated, however no studies have demonstrated a survival advantage compared to the historical results of watch and wait or standard chemotherapy. With regard to treatment in combination with chemotherapy, the results of five randomized phase III studies have now been published (Table 43.6). Each demonstrated a highly significant improvement in remission rate and progression-free survival for the addition of rituximab to chemotherapy184–188 with an additional overall survival benefit in two of the studies,185,186 suggesting that immunochemotherapy is superior to conventional therapy in patients requiring treatment. Finally, a number of studies have also investigated the use of rituximab maintenance treatment either as monotherapy or following chemotherapy. When used as a single agent, maintenance treatment following initial therapy with either a single infusion every 2–3 months over a period of 8–12 months183,190 or four infusions every 6 months over 2 years,172 results in improved response rate and progression-free survival. However, whether the long-term results of this approach are superior to watch and wait or conventional chemotherapy is unknown. Rituximab maintenance is also effective when given to consolidate prior chemotherapy. The results of maintenance rituximab (given according to the Hainsworth schedule172) following CVP chemotherapy appear similar to those obtained using rituximab concurrently with chemotherapy,189 although at a significant financial cost disadvantage incurred by doubling the number of doses of rituximab administered. Whether the results of immunochemotherapy could be further improved by using rituximab as part of initial chemotherapy and as maintenance is currently unknown and forms the basis of ongoing studies.
Two other treatment modalities that have been investigated in newly diagnosed advanced FL are radioimmunotherapy and autologous transplantation. The radiosensitivity and expression of suitable target antigens make FL an attractive target for treatment with radiolabelled immunoconjugates. Most clinical experience has been derived from the use of 90Y-ibritumomab (Zevalin) and 131I-tositumomab (Bexxar), both of which produce response rates of 60–80 per cent when used as monotherapy in patients with refractory or relapsed disease.191 Recent studies using 131I-tositumomab alone or in combination with chemotherapy as initial therapy of FL have demonstrated impressive response rate and low incidence of early disease relapse suggesting a possible role for these agents in the early stages of disease treatment (see Chapter 10 for more details).192–194 The use of high-dose chemotherapy and ASCT to consolidate first remission is contentious. Several studies have demonstrated improved progressionfree survival although none has shown improved overall survival and several have resulted in significant rates of secondary malignancies consequent on radiation-based transplant conditioning.195 Overall there does not appear to be any consistent benefit obtained from early autografting with the exception of patients with high-grade transformation of FL in whom the outcome following ASCT appears to be superior to that from chemotherapy alone.195 How best to interpret these data? For patients with asymptomatic low-risk disease, watchful waiting is still an appropriate initial strategy, with the possible exception of those under 40 in whom an early allogeneic transplant may be considered (discussed below). Whether the use of early rituximab will translate into improvements in survival or quality of life is unknown but may be determined by the results of a current European phase III trial. There are currently too few data available to recommend radioimmunoconjugates for initial treatment, and cost and
Table 43.6 Immunochemotherapy using rituximab for the initial treatment of advanced follicular lymphoma Reference Marcus et al184 Hiddemann et al185 Herold et al186,187
Salles et al188 Hochster et al189
Regimen
Response rate (%)
Significance (p-value)
Median TTP/EFS (months)
Significance (p-value)
CVP R-CVP CHOP R-CHOP MCP R-MCP
57% 81% 90% 96% 75% 92%
0.0001
7 27 31 NR 19 NR
0.0001
CHVPinterferon-α R-CHVPinterferon-α CVP CVPR maintenance
85% 94% Response rate to CVP 80%
0.011
0.0001 0.0001
Abbreviations: NR, not reached; TTP, time to progression; EFS, event-free survival; N/A, not applicable. *Studies also showing a statistically significant overall survival benefit for immunochemotherapy.
NR NR 50 18
0.0001*
0.0001* N/A 0.0001*
Indolent non-Hodgkin’s lymphoma 1059
availability currently limit their use in many centres. For patients requiring treatment, immunochemotherapy is emerging as the optimal first-line therapy with the majority of the published data concerning the use of rituximab. However, despite its undoubted efficacy in the treatment of FL, several questions regarding its use remain unanswered. First, do all patients benefit from rituximab? Apart from clinical and laboratory variables, a number of molecular markers including bcl-2 expression, Fc-γIII receptor phenotype and gene expression profile have been associated with rituximab sensitivity suggesting the possibility of a targeted approach to treatment.172,190,196 Second, should patients who require treatment be treated with rituximab monotherapy or in combination with conventional chemotherapy and is a risk stratified approach justified? Finally, for patients fit to receive chemotherapy, what is the optimal regimen to use and which rituximab schedule should be chosen – up front, maintenance or both? The answers to the latter questions will hopefully be answered as the results of currently recruiting trials emerge; however, until these results are available, a suggested treatment algorithm based on the available data is shown in Figure 43.5. TREATMENT OF NEWLY DIAGNOSED STAGE I DISEASE (GRADE 1, 2, 3A)
The standard treatment for patients with localized disease has, for many years, been involved field radiotherapy, resulting in 10-year overall (progression-free) survival figures of 60–80 per cent (30–50 per cent).197 The radiation dose used has been variable, typically between 30 and 55 Gy, although more recent data in relapsed disease suggest that equivalent results may be obtained using 24 Gy or even as low as 4 Gy.198,199 It is unclear which factors most strongly predict outcome due to conflicting reports from different studies, although older age and bulky disease have most consistently been associated with reduced overall survival.197 Whether the outcome can be improved with adjuvant chemotherapy is unclear. A number of older randomized trials indicated that there was no benefit to the addition of low-intensity chlorambucil or multi-agent chemotherapy although the number of patients in the majority of the studies was very small.197,200 However, larger retrospective studies have shown a possible survival advantage to combined modality therapy using combination chemotherapy and radiotherapy201 and a randomized trial is currently underway to further investigate this. The long survival and prospects of cure using radiationbased treatment for patients with limited (particularly stage I) disease have resulted in relatively little interest being paid to a watchful waiting strategy. However, in a recent single centre retrospective analysis of patients with newly diagnosed stage I/II follicular lymphoma, the median survival in patients receiving no initial therapy appeared comparable to that of those treated with radiotherapy or combined modality therapy.202 On the basis of
the currently available evidence it would be difficult to recommend a watchful waiting strategy for all patients with limited-stage FL, although it may be appropriate for some selected patients with good prognosis in whom there is little benefit from early intervention. TREATMENT OF RELAPSED DISEASE
Irrespective of the initial mode of treatment, the majority of patients with FL are destined to relapse and require further therapy. Following relapse, a further remission is normally achievable with chemotherapy,203 with the choice of regimen dependent on the previous treatment received, the duration of the first remission, the fitness of the patient and whether consolidation with high-dose therapy is being considered. If an autologous transplantation is planned then salvage regimens as described for DLBCL are effective and facilitate stem cell mobilization. For patients unsuitable for high-dose therapy, re-treatment using front-line therapy (if no anthracyclines were included) may be considered if the remission duration was greater than 1 year, or alternative regimens chosen. Nucleoside analogues such as fludarabine, either alone or in combination, are highly effective in FL but not usually recommended prior to ASCT due to concerns about the impact on stem cell mobilization. A suggested treatment algorithm is shown in Figure 43.5. Whether relapse treatment should include rituximab is debatable. It has been demonstrated that re-treatment is safe and preliminary data suggest that this is an effective strategy for patients with an initial remission duration of greater than 6 months,204,205 although whether this is superior to standard relapse treatment or cost effective is yet to be determined. Promising results have also been obtained using radioimmunoconjugates for the treatment of relapsed disease including cases refractory to rituximab (see Chapter 10 for more details).191,203 Although further remissions are achievable with chemotherapy, historical data suggest that they are likely to be of a shorter duration than the first. A number of phase II studies suggested that disease-free survival could be improved by autologous transplantation in second remission,195 confirmed more recently in the randomized phase III CUP (Chemotherapy (C), high-dose therapy followed by autologous unpurged (U) or purged (P) stem-cell transplantation) study, that demonstrated an additional overall survival benefit for ASCT over chemotherapy alone.206 The results of the CUP study have led to ASCT being widely adopted as the standard treatment for patients under the age of 65–70 in second remission and fit for a transplant, although a number of lines of evidence suggest that this should be re-evaluated. First, the CUP study included only 89 patients, leading to possible selection bias. Second, the patients in the study all received TBI (total body irradiation) based conditioning and whether a similar outcome can be derived from widely used chemotherapy based conditioning regimens is unproven. Third, this was performed in the ‘prerituximab’ era and whether the results remain applicable is
1060 Non-Hodgkin’s lymphoma
Diagnosis of follicular lymphoma grade 1, 2, 3a
Stage I
Grade 3b
Treat as diffuse large B-cell lymphoma
Radiotherapy Consider • Clinical trial • Watchful waiting
Stage II–IV
Symptomatic and/or bulky§
Asymptomatic and non-bulky§
R+chemo1*
Watchful waiting
R-monotherapy
Progression
CR or PR
Observe Consider • Clinical trial • R-maintenance • Autologous stem cell transplant • Allogeneic stem cell transplant if <40
Relapse
Late (>6–12 months)
Refractory
Late (<6–12 months)
R+chemo1* if no anthracycline used Consider • Clinical trial * • R+chemo2
Chemo2*
CR or PR
Refractory
Key §
Bulky disease
Maximum nodal diameter greater than 7 cm or more than 3 nodal masses greater than 3 cm. Radiotherapy should be considered in addition to chemotherapy to sites of bulky disease
R, CR, PR
Rituximab, complete remission, partial remission
*Chemo1
Combination chemotherapy (e.g. CVP or CHOP) or if unfit single agent treatment (e.g. chlorambucil)
*Chemo2
Salvage chemotherapy (as for DLBCL. e.g. ESHAP) if stem cell transplant planned. Otherwise consider CHOP (if no anthracyclines), fludarabine-based regimen or alternative single agent.
*Chemo3
Consider fludarabine-based regimen (if not previously used), radio immunotherapy, clinical trial, or palliative care.
Autologous stem cell transplant Consider • Clinical trial • Reduced intensity Allogeneic transplant • Allogeneic transplant if <40
Figure 43.5 Suggested treatment algorithm for the management of follicular lymphoma.
Relapse
Chemo3*
Indolent non-Hodgkin’s lymphoma 1061
unknown. Fourth, the emerging role of reduced-intensity allogeneic transplantation as an effective treatment for FL suggests that this may become an alternative and potentially superior treatment to ASCT (discussed below).
Mantle cell lymphoma Mantle cell lymphoma (MCL) accounts for around 8 per cent of cases of NHL, typically presenting as advanced disease with a predominantly nodal distribution in older adults. Pathologically it has a distinct phenotype and is defined by nuclear over-expression of cyclin D1, most commonly as a result of a t(11;14) leading to rearrangement of the gene locus at 11q13.3 Although cyclin D1 expression has been described less commonly in other lymphomas it is present in almost all cases of MCL.207 The majority of cases of MCL are histologically indolent although two aggressive blastoid variants are recognized, associated with additional genetic abnormalities and a worse prognosis.208 Many patients with MCL present with clinical poor prognostic factors such as hepatomegaly, and whilst many initially respond to chemotherapy, the remission duration is typically short with an overall survival typically no better than 3–4 years. A minority of patients present with a lower tumour burden, typically with splenomegaly, and involvement of the peripheral blood. In these patients the disease may pursue a more indolent course although the reasons for this are unknown.209 Although many different chemotherapy regimens have been utilized, they may be broadly categorized into three groups: first, standard-dose chemotherapy with or without rituximab; second, intensive chemotherapy with or without rituximab; third, autologous transplantation following either standard or intensive induction treatment. Remission rates of 50–75 per cent are achieved using standard chemotherapy regimens (typically CHOP or combination chemotherapy including fludarabine) although the median progression-free survival is only 12–18 months. When using CHOP chemotherapy, the remission rate and progression-free survival may be improved with the addition of rituximab, although the magnitude of the improvement is significantly less than that observed in follicular lymphoma and there is no benefit in terms of overall survival.187,210,211 If a fludarabine-based regimen is used for initial treatment then preliminary data suggest that the addition of rituximab confers no benefit,212 although larger studies are planned to confirm this observation. In patients under 65, more intensive induction chemotherapy such as alternating courses of hyperCVAD and high-dose methotrexate/cytarabine together with rituximab produces significantly better remission rates (95 per cent) with median time to progression of over 4 years, although at the cost of a 5 per cent treatment-related mortality and worrying incidence of secondary myelodysplasia.213 As is the case for the majority of subtypes of NHL, particularly since the introduction of rituximab, the role of early ASCT in MCL is
debatable. When using conventional-dose chemotherapy, a single randomized study showed that TBI conditioned ASCT in first remission after CHOP chemotherapy resulted in an improved median progression-free survival compared to CHOP chemotherapy plus interferon maintenance (46 months vs. 23 months) although there was no change in overall survival.214 When using high-dose initial chemotherapy, the results of hyperCVAD chemotherapy followed by autologous transplantation are not dissimilar to those described above for hyperCVAD plus rituximab.213,215 The optimal treatment for MCL has not been defined. The outcome of either standard-dose chemotherapy (typically CHOP) followed by autologous transplantation or hyperCVAD combined with rituximab appears similar and both could be considered in fit patients under 65. For older patients, treatment is determined by their performance status with the principal treatment options being either anthracycline or fludarabine based chemotherapy. The improved remission rate and time to progression suggest that all patients should receive rituximab as part of initial chemotherapy (although perhaps only for CHOP-type regimens), however the argument for this is less compelling than with other subtypes of NHL. As with follicular lymphoma, rituximab maintenance treatment and radioimmunoconjugates have been investigated for the treatment of MCL although a established role for these as part of initial therapy has not yet been defined.216 For patients relapsing beyond first-line therapy, the prognosis is poor although a number of agents that have shown promise in early-stage trials (for example thalidomide, proteosome inhibitors, radioimmunoconjugates) could be considered.216
Marginal zone lymphoma Marginal zone lymphomas comprise the relatively common extra-nodal mucosa-associated lymphoid tissue (MALT) type and less common nodal and splenic variants. MALT-type lymphomas are predominantly indolent, with 5-year survival typically 80 per cent or better. They occur mainly in adults, and frequently on a background of a chronic inflammatory disorder, either chronic infections or autoimmune conditions.3 More than half of patients have limited-stage disease at presentation and the most common anatomical site of involvement is the gastrointestinal tract, particularly the stomach, with the lung, head and neck including ocular adnexa, skin, thyroid and breast representing other common sites of involvement.3 The pathogenesis of these lymphomas is complex; however, in the early stages antigen expression and T-cell stimulation due to the underlying infection or autoimmune condition lead to B-cell proliferation, with secondary clonal evolution and lymphoma development as a result of acquired genetic abnormalities.217 Early-stage gastric lymphomas associated with Helicobacter pylori can be successfully treated by eradication of the underlying infection although the therapeutic response may take up to 1 year to achieve.
1062 Non-Hodgkin’s lymphoma
The evidence for an association between infections and MALT lymphomas at other sites (Chlamydia psittaci and ocular adnexal tumours or Borrelia burgdorferi and skin tumours) is less convincing and although it provides a rationale for antibiotic treatment, this approach should currently be regarded as investigational.218 For patients with localized disease, not responding to antibiotics, excellent results can be obtained with moderate doses of radiotherapy,219 whereas systemic disease is usually well controlled using oral alkylating agents (chlorambucil or cyclophosphamide).218 Nucleoside analogues (fludarabine or cladribine), combination chemotherapy, or rituximab are also effective but are more toxic without a convincing improvement in response or survival. Nodal marginal zone lymphoma is uncommon, typically presenting with minimally symptomatic lymphadenopathy with the neck being the most common site of involvement. The literature concerning the clinical course and treatment is sparse, although they appear to behave in an indolent manner and can successfully be treated with chemotherapy as outlined for MALT lymphomas although early relapse is common.3,218 Splenic marginal zone lymphoma (SMZL – previously referred to as splenic lymphoma with villous lymphocytes) is uncommon, most frequently presenting with massive splenomegaly and bone marrow infiltration in middleaged or elderly adult patients. This disease is typically indolent and treatment can usually be deferred until the onset of symptoms (splenic pain or constitutional such as weight loss) or cytopenias. Splenectomy is the most common initial treatment and may result in a durable remission; chemotherapy along the lines described for MALT lymphomas is also effective, but usually reserved for patients unfit for a splenectomy or with progressive disease.218 Although the overall 5-year survival for SMZL is up to 80 per cent, a subset of patients have more aggressive disease, and recent data suggest that they may be identified by a number of clinical and laboratory variables,220 however whether those with an adverse prognosis will benefit from more intensive treatment is currently unknown.
FUTURE DEVELOPMENTS Allogeneic transplantation Allogeneic stem cell transplantation is a potentially curative therapy for NHL although its role in the treatment of this group of diseases remains unclear. Several studies have demonstrated a long-term survival benefit of allogeneic over autologous transplantation221,222 in the treatment of lymphoma although the reason for this advantage is uncertain. Part of this benefit stems from the use of an uncontaminated graft, supported by the reduced relapse rate observed in patients with NHL treated with syngeneic transplantation compared to autologous procedures.223 A second mechanism for this advantage is a graft-versus-tumour (GVT)
effect. The anti-tumour capacity of allogeneic T-lymphocytes has long been recognized in recipients of allogeneic transplantation,224 most notably in the treatment of chronic myeloid leukaemia.225 The evidence for a GVT effect in NHL is, however, less robust. When the outcomes of syngeneic and allogeneic transplants for the treatment of NHL were compared, the relapse rate in both groups was not significantly different, and there was no difference in the relapse rates between recipients of allogeneic transplants with or without graft-versus-host disease (GvHD).223 These results have been interpreted as demonstrating a lack of GVT effect in lymphoma223 although this conclusion should be interpreted with caution due to the heterogeneity of the patients studied.226 In contrast, the demonstration of significant responses in patients with NHL following withdrawal of immunosuppression or infusions of donor lymphocytes in a number of smaller series226–229 is consistent with a clinically significant GVT effect in these diseases. The utility of allogeneic transplantation is however limited by the procedure-related toxicity, and with a median age at diagnosis in the mid-60s most patients with NHL are ineligible for this treatment. The development of reducedintensity conditioning (RIC) approaches has allowed application of allogeneic therapies to a broader range of patients. Whilst RIC allows durable engraftment with reduced toxicity, the conditioning employed would be expected to have less tumour activity and with many protocols GvHD remains a significant cause of morbidity and mortality.230 When the results of allogeneic transplantation are analysed for the two major subtypes, the outcome of conventional allogeneic transplantation in DLBCL obtained from registry data is no better than prospective data for autologous transplantation with long-term survival of around 40 per cent.222 A single prospective series suggests that better results may be obtained in young patients (under 35 years) who have achieved complete remission and not received a prior autograft.231 The outcome following RIC is small and generally disappointing due to high procedure-related mortality and early relapse rate.232 For follicular lymphoma, an analysis of registry data indicates that the relapse rate following allogeneic transplantation is significantly reduced compared to ASCT and as for DLBCL, the outcome was improved in younger patients (under 40) with chemosensitive disease at the time of transplant.222,223 Data concerning RIC transplantation in FL are also encouraging with 60–70 per cent 3-year survival and procedurerelated mortality of 10–20 per cent.232 These data suggest that for patients with DLBCL, given the efficacy of ASCT, allogeneic transplantation should only be considered in young patients with very poor prognosis disease, ideally as part of a trial, and whether RIC has any role in this disease is questionable. For patients under 40 years with FL and a sibling donor, allogeneic transplantation should be considered in first remission, whilst for older patients the encouraging results of RIC protocols suggest that this may be a more effective alternative to ASCT as salvage treatment and a trial comparing these treatments is needed.
References 1063
The role of allogeneic transplantation for other lymphoma subtypes is less clear, due to their rarity and the small number of transplants performed. Based on the available evidence it is, however, possible to make some generalizations. First, for all indications, the outcome in patients with refractory disease is very poor and with a significant procedure-related morbidity/mortality it should only be considered in patients with chemotherapy sensitive disease. Second, the mortality following full-intensity conditioning rises rapidly above the age of 40 and with the exception of subtypes that respond badly to all conventional treatments (for example γδ-hepatosplenic lymphoma) full-intensity transplantation is rarely justified above this age. Third, the current literature concerning the use of RIC transplants in NHL is very small and although it is feasible up to the age of 70 the available evidence points to this being principally effective in low-grade disease.232
KEY LEARNING POINTS ●
●
●
●
●
●
The incidence of NHL has risen over the last three decades. This has been most marked in the developed world. Taken together with an increased prevalence in the elderly, it is becoming an increasing health-care problem in an ageing population. The classification of NHL has become increasingly refined; however, a division into low-grade and high-grade histologies of B-cell or T-cell origin still remains a useful starting point to base therapeutic decisions. The advent of monoclonal antibody therapy over the last decade has resulted in a significant improvement in the outcome following treatment for common lymphoma subtypes (diffuse large B-cell and follicular lymphoma) with rituximab based immuno-chemotherapy emerging as the first-line treatment of choice for patients requiring therapy. The treatment of rarer subtypes including many T-cell lymphomas remains unsatisfactory and further clinical studies are warranted to investigate the role of new treatments such as novel agents and intensive treatment regimens. PET scanning is emerging as a useful tool in staging and response assessment, particularly in patients with high-grade histologies The role of allogeneic transplantation in the management of NHL remains unclear, although the most encouraging results have been obtained in patients with indolent disease. The development of reduced-intensity conditioning protocols has extended the eligibility for this procedure to older patients.
REFERENCES 1 Morton LM, Wang SS, Devesa SS, et al. Lymphoma incidence patterns by WHO subtype in the United States, 1992–2001. Blood 2006; 107:265–76. 2 Müller AMS, Ihorst G, Mertelsmann R, Engelhardt M. Epidemiology of non-Hodgkin’s lymphoma (NHL): trends, geographic distribution, and etiology. Ann Hematol 2005; 84:1–12. ◆3 Jaffe ES, Harris NL, Stein H, Vardiman JW (eds). Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon: IARC Press, 2001. 4 Lister TA, Crowther D, Sutcliffe SB, et al. Report of a committee convened to discuss the evaluation and staging of patients with Hodgkin’s disease: Cotswolds meeting. J Clin Oncol 1989; 7:1630–6. 5 Groves FD, Linet MS, Travis LB, Devesa SS. Cancer Surveillance Series: Non-Hodgkin’s lymphoma incidence by histologic subtype in the United States from 1978 through 1995. J Natl Cancer Inst 2000; 92:1240–51. ◆6 The Non-Hodgkin’s Lymphoma Classification Project. A clinical evaluation of the International Lymphoma Study Group classification of non-Hodgkin’s lymphoma. Blood 1997; 89:3909–18. 7 Engelhard M, Brittinger G, Huhn D, et al. Subclassification of diffuse large B-cell lymphomas according to the Kiel classification: distinction of centroblastic and immunoblastic lymphomas is a significant prognostic factor. Blood 1997; 89:2291–7. 8 Simonitsch-Klupp I, Hauser I, Ott G, et al. Diffuse large B-cell lymphomas with plasmablastic/plasmacytoid features are associated with TP53 deletions and a poor clinical outcome. Leukemia 2004; 18:146–55. 9 Gascoyne RD, Lamant L, Martin-Subero JI, et al. ALKpositive diffuse large B-cell lymphoma is a associated with Clathrin-ALK rearrangements: report of 6 cases. Blood 2003; 102:568–73. 10 Haralambieva E, Pulford KAF, Lamant L, et al. Anaplastic large-cell lymphomas of B-cell phenotype are anaplastic lymphoma kinase (ALK) negative and belong to the spectrum of diffuse large B-cell lymphomas. Br J Haematol 2000; 109:584–91. 11 Bouabdallah R, Mounier N, Guettier C, et al. T-cell/ histiocyte-rich large B-cell lymphomas and classical diffuse large B-cell lymphomas have similar outcome after chemotherapy: a matched-control analysis. J Clin Oncol 2003; 21:1271–7. 12 Barth TF, Leithauser F, Joos S, et al. Mediastinal (thymic) large B-cell lymphoma: where do we stand? Lancet Oncol 2002; 3:229–34. 13 Addis BJ, Isaacson PG. Large cell lymphoma of the mediastinum: a B-cell tumour of probable thymic origin. Histopathology 1986; 10:379–90. 14 Calaminici M, Piper K, Lee AM, Norton AJ. CD23 expression in mediastinal large B-cell lymphomas. Histopathology 2004; 45:619–24.
1064 Non-Hodgkin’s lymphoma
●15
16
17
18
●19
20
●21
22
◆23
◆24
●25
26
27
28
●29
Savage KJ, Monti S, Kutok JL, et al. The molecular signature of mediastinal large B-cell lymphoma differs from that of other diffuse large B-cell lymphomas and shares features with classical Hodgkin lymphoma. Blood 2003; 102:3871–9. Rosenwald A, Wright G, Leroy K, et al. Molecular diagnosis of primary mediastinal B cell lymphoma identifies a clinically favorable subgroup of diffuse large B cell lymphoma related to Hodgkin lymphoma. J Exp Med 2003; 198:851–62. Ferreri AJ, Campo E, Seymour JF, et al. Intravascular lymphoma: clinical presentation, natural history, management and prognostic factors in a series of 38 cases, with special emphasis on the ‘cutaneous variant’. Br J Haematol 2004; 127:173–83. Murase T, Nakamura S, Kawauchi K, et al. An Asian variant of intravascular large B-cell lymphoma: clinical, pathological and cytogenetic approaches to diffuse large Bcell lymphoma associated with haemophagocytic syndrome. Br J Haematol 2000; 111: 826–4. International Non-Hodgkin’s Lymphoma Prognostic Factors Project. A predictive model for aggressive non-Hodgkin’s lymphoma. N Engl J Med 1993; 329:987–4. Wilder RB, Rodriguez MA, Medeiros LJ, et al. International prognostic index-based outcomes for diffuse large B-cell lymphomas. Cancer 2002; 94:3083–8. Miller TP, Dahlberg S, Cassady JR, et al. Chemotherapy alone compared with chemotherapy plus radiotherapy for localized intermediate- and high-grade non-Hodgkin’s lymphoma. N Engl J Med 1998; 339:21–6. Strehl J, Mey U, Glasmacher A, et al. High-dose chemotherapy followed by autologous stem cell transplantation as first-line therapy in aggressive nonHodgkin’s lymphoma: a meta-analysis. Haematologica 2003; 88:1304–15. Wu G, Keating A. Biomarkers of potential prognostic significance in diffuse large B-cell lymphoma. Cancer 2006; 106: 247–57. Lossos IS, Morgensztern D. Prognostic biomarkers in diffuse large B-cell lymphoma. J Clin Oncol 2006; 24:1–12. Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000; 403:503–11. Rosenwald A, Wright G, Chan WC, et al. The use of molecular profiling to predict survival after chemotherapy for diffuse large B-cell lymphoma. N Engl J Med 2002; 346:1937–47. Shipp MA, Ross KN, Tamayo P, et al. Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine learning. Nat Med 2002; 8: 68–74. Wright G, Tan B, Rosenwald A, et al. A gene expressionbased method to diagnose clinically distinct subgroups of diffuse large B cell lymphoma. Proc Natl Acad Sci U S A 2003; 100:9991–6. McKelvey EM, Gottlieb JA, Wilson HE, et al. Hydroxyldaunomycin (Adriamycin) combination chemotherapy in malignant lymphoma. Cancer 1976; 38:1484–93.
●30
31
32
●33
●34
●35
●36
37
38
◆39
40
41
42
Fisher RI, Gaynor ER, Dahlberg S, et al. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin’s lymphoma. N Engl J Med 1993; 328:1002–6. Todeschini G, Secchi S, Morra E, et al. Primary mediastinal large B-cell lymphoma (PMLBCL): long-term results from a retrospective multicentre Italian experience in 138 patients treated with CHOP or MACOP-B/VACOP-B. Br J Cancer 2004; 90:372–6. Savage KJ, Al-Rajhi N, Voss N, et al. Favorable outcome of primary mediastinal large B-cell lymphoma in a single institution: the British Columbia experience. Ann Oncol 2006; 17:123–30. Pfreundschuh M, Trumper L, Kloess M, et al. Two-weekly or 3-weekly CHOP chemotherapy with or without etoposide for the treatment of young patients with good-prognosis (normal LDH) aggressive lymphomas: results of the NHL-B1 trial of the DSHNHL. Blood 2004; 104:626–33. Maloney DG, Liles TM, Czerwinski DK, et al. Phase I clinical trial using escalating single-dose infusion of chimeric antiCD20 monoclonal antibody (IDEC-C2B8) in patients with recurrent B-cell lymphoma. Blood 1994; 84: 2457–66. Coiffier B, Haioun C, Ketterer N, et al. Rituximab (anti-CD20 monoclonal antibody) for the treatment of patients with relapsing or refractory aggressive lymphoma: a multicenter phase II study. Blood 1998; 92:1927–32. Coiffier B, Lepage E, Briere J, et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 2002; 346:235–42. Feugier P, Van Hoof A, Sebban C, et al. Long-term results of the R-CHOP study in the treatment of elderly patients with diffuse large B-cell lymphoma: a study by the Groupe d’Etude des Lymphomes de l’Adulte. J Clin Oncol 2005; 23:4117–26. Pfreundschuh MG, Trümper L, Ma D, et al. Randomized intergroup trial of first line treatment for patients 60 years with diffuse large B-cell non-Hodgkin’s lymphoma (DLBCL) with a CHOP-like regimen with or without the antiCD20 antibody rituximab – early stopping after the first interim analysis. ASCO 2004; 556a. Gregory SA, Trümper L. Chemotherapy dose intensity in non-Hodgkin’s lymphoma: is dose intensity an emerging paradigm for better outcomes? Ann Oncol 2005; 16:1413–24. Tilly H, Lepage E, Coiffier B, et al. Intensive conventional chemotherapy (ACVBP regimen) compared with standard CHOP for poor-prognosis aggressive non-Hodgkin lymphoma. Blood 2003; 102:4284–9. Gisselbrecht C, Lepage E, Molina T, et al. Shortened firstline high-dose chemotherapy for patients with poorprognosis aggressive lymphoma. J Clin Oncol 2002; 20:2472–9. Kaiser U, Uebelacker I, Abel U, et al. Randomized study to evaluate the use of high-dose therapy as part of primary treatment for ‘aggressive’ lymphoma. J Clin Oncol 2002; 20:4413–9.
References 1065
43 Martelli M, Gherlinzoni F, De Renzo A, et al. Early autologous stem-cell transplantation versus conventional chemotherapy as front-line therapy in high-risk, aggressive non-Hodgkin’s lymphoma: an Italian multicenter randomized trial. J Clin Oncol 2003; 21:1255–62. 44 Haioun C, Lepage E, Gisselbrecht C, et al. Benefit of autologous bone marrow transplantation over sequential chemotherapy in poor-risk aggressive non-Hodgkin’s lymphoma: updated results of the prospective study LNH87-2. Groupe d’Etude des Lymphomes de l’Adulte. J Clin Oncol 1997; 15:1131–7. 45 Santini G, Salvagno L, Leoni P, et al. VACOP-B versus VACOP-B plus autologous bone marrow transplantation for advanced diffuse non-Hodgkin’s lymphoma: results of a prospective randomized trial by the non-Hodgkin’s Lymphoma Cooperative Study Group. J Clin Oncol 1998; 16:2796–802. 46 Kluin-Nelemans HC, Zagonel V, Anastasopoulou A, et al. Standard chemotherapy with or without high-dose chemotherapy for aggressive non-Hodgkin’s lymphoma: randomized phase III EORTC study. J Natl Cancer Inst 2001; 93:22–30. ●47 Gianni AM, Bregni M, Siena S, et al. High-dose chemotherapy and autologous bone marrow transplantation compared with MACOP-B in aggressive B-cell lymphoma. N Engl J Med 1997; 336:1290–7. 48 Milpied N, Deconinck E, Gaillard F, et al. Initial treatment of aggressive lymphoma with high-dose chemotherapy and autologous stem-cell support. N Engl J Med 2004; 350:1287–95. 49 van Imhoff GW, van der Holt B, Mackenzie MA, et al. Impact of three courses of intensified CHOP prior to highdose sequential therapy followed by autologous stem-cell transplantation as first-line treatment in poor-risk, aggressive non-Hodgkin’s lymphoma: comparative analysis of Dutch-Belgian Hemato-Oncology Cooperative Group Studies 27 and 40. J Clin Oncol 2005; 23:3793–801. 50 Schmitz N, Kloess M, Reiser M, et al. Four versus six courses of dose-escalated cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) regimen plus etoposide (megaCHOEP) and autologous stem cell transplantation. Cancer 2006; 106:136–45. 51 Betticher DC, von Rohr T, Kovacsovics T, et al. Sequential high-dose chemotherapy (SHDC) with autologous stem cell transplantation (ASCT) vs standard CHOP regimen for pts with untreated aggressive NHL. Result of the MISTRAL trial from the Swiss Group for Clinical Cancer Research. Ann Oncol 2005; 16 suppl 5:217a. ●52 Pfreundschuh M, Trumper L, Kloess M, et al. Two-weekly or 3-weekly CHOP chemotherapy with or without etoposide for the treatment of elderly patients with aggressive lymphomas: results of the NHL-B2 trial of the DSHNHL. Blood 2004; 104:634–41. 53 Pfreundschuh M, Kloess M, Schmits R, et al. Six, not eight cycles of bi-weekly CHOP with rituximab (R-CHOP-14) is the preferred treatment for elderly patients with diffuse large B-cell lymphoma (DLBCL): results of the RICOVER-60
●54
55
56
57
58
59
60
61
62
●63
◆64
65 66
67
trial of the German High-Grade Non-Hodgkin Lymphoma Study Group (DSHNHL). Blood 2005; 106:13a. Connors JM, Klimo P, Fairey RN, Voss N. Brief chemotherapy and involved field radiation therapy for limited-stage, histologically aggressive lymphoma. Ann Intern Med 1987; 107:25–30. Spicer J, Smith P, Maclennan K, et al. Long-term follow-up of patients treated with radiotherapy alone for early-stage histologically aggressive non-Hodgkin’s lymphoma. Br J Cancer 2004; 90:1151–5. Miller TP, LeBlanc M, Spier CM, et al. CHOP alone compared to CHOP plus radiotherapy for early stage aggressive non-Hodgkin’s lymphomas: update of the Southwest Oncology Group (SWOG) randomised trial. Blood 2001; 98:724a. Horning SJ, Weller E, Kim K, et al. Chemotherapy with or without radiotherapy in limited-stage diffuse aggressive non-Hodgkin’s lymphoma: Eastern Cooperative Oncology Group study 1484. J Clin Oncol 2004; 22:3032–8. Bonnet C, Fillet G, Mounier N, et al. Radiotherapy is unnecessary in elderly patients with localized aggressive non Hodgkin lymphoma: results of the LNH 93-4 study. Ann Oncol 2002; 13 Suppl 2:79a. Reyes F, Lepage E, Ganem G, et al. ACVBP versus CHOP plus radiotherapy for localized aggressive lymphoma. N Engl J Med 2005; 352:1197–205. Miller TP, Unger JM, Spier CM, et al. Effect of adding rituximab to three cycles of CHOP plus involved-field radiotherapy for limited-stage aggressive diffuse B-cell lymphoma (SWOG-0014). Blood 2004; 104:158a. Mainwaring PN, Cunningham D, Gregory W, et al. Mitoxantrone is superior to doxorubicin in a multiagent weekly regimen for patients older than 60 with high-grade lymphoma: results of a BNLI randomized trial of PAdriaCEBO versus PMitCEBO. Blood 2001; 97:2991–7. Maartense E, Kluin-Nelermans HC, le Cessie S, et al. Different age limits for elderly patients with indolent and aggressive non-Hodgkin lymphoma and the role of relative survival with increasing age. Cancer 2000; 89:2667–76. Jones SE, Grozea PN, Metz EN, et al. Superiority of adriamycin-containing combination chemotherapy in the treatment of diffuse lymphoma: a Southwest Oncology Group study. Cancer 1979; 43:417–25. Hoang-Xuan K, Camilleri-Broet S, Soussain C. Recent advances in primary CNS lymphoma. Curr Opin Oncol 2004; 16:601–6. New P. Radiation injury to the nervous system. Curr Opin Neurol 2001; 14:725–34. Guha-Thakurta N, Damek D, Pollack C, Hochberg FH. Intravenous methotrexate as initial treatment for primary central nervous system lymphoma: response to therapy and quality of life of patients. J Neurooncol 1999; 43:259–68. Herrlinger U, Schabet M, Brugger W, et al. German Cancer Society Neuro-Oncology Working Group NOA-03
1066 Non-Hodgkin’s lymphoma
68
69
70
71
◆72
73
74
75
76
77
78
●79
80
multicenter trial of single-agent high-dose methotrexate for primary central nervous system lymphoma. Ann Neurol 2002; 51:247–52. Batchelor T, Carson K, O’Neill A, et al. Treatment of primary CNS lymphoma with methotrexate and deferred radiotherapy: a report of NABTT 96-07. J Clin Oncol 2003; 21:1044–9. Pels H, Schmidt-Wolf IG, Glasmacher A, et al. Primary central nervous system lymphoma: results of a pilot and phase II study of systemic and intraventricular chemotherapy with deferred radiotherapy. J Clin Oncol 2003; 21:4489–95. Moreton P, Morgan GJ, Gilson D, et al. The development of targeted chemotherapy for CNS lymphoma – a pilot study of the IDARAM regimen. Cancer Chemother Pharmacol 2004; 53:324–8. Hollender A, Kvaloy S, Nome O, et al. Central nervous system involvement following diagnosis of non-Hodgkin’s lymphoma: a risk model. Ann Oncol 2002; 13:1099–107. McMillan A. Central nervous system-directed preventative therapy in adults with lymphoma. Br J Haematol 2005; 131:13–21. Jahnke K, Thiel E, Martus P, et al. Retrospective study of prognostic factors in non-Hodgkin lymphoma secondarily involving the central nervous system. Ann Hematol 2006; 85:45–50. Cheung CW, Burton C, Smith P, et al. Central nervous system chemoprophylaxis in non-Hodgkin lymphoma: current practice in the UK. Br J Haematol 2005; 131:193–200. Buckstein R, Lim W, Franssen E, Imrie KL. CNS prophylaxis and treatment in non-Hodgkin’s lymphoma: variation in practice and lessons from the literature. Leuk Lymphoma 2003; 44:955–62. Recht L, Straus DJ, Cirrincione C, et al. Central nervous system metastases from non-Hodgkin’s lymphoma: treatment and prophylaxis. Am J Med 1988; 84:425–35. Chua SL, Seymour JF, Streater J, et al. Intrathecal chemotherapy alone is inadequate central nervous system prophylaxis in patients with intermediate-grade non-Hodgkin’s lymphoma. Leuk Lymphoma 2002; 43:1783–8. Haioun C, Besson C, Lepage E, et al. Incidence and risk factors of central nervous system relapse in histologically aggressive non-Hodgkin’s lymphoma uniformly treated and receiving intrathecal central nervous system prophylaxis: a GELA study on 974 patients. Groupe d’Etudes des Lymphomes de l’Adulte. Ann Oncol 2000; 11:685–90. Philip T, Guglielmi C, Hagenbeek A, et al. Autologous bone marrow transplantation as compared with salvage chemotherapy in relapses of chemotherapy-sensitive nonHodgkin’s lymphoma. N Engl J Med 1995; 333:1540–5. Moskowitz CH, Bertino JR, Glassman JR, et al. Ifosfamide, carboplatin, and etoposide: a highly effective cytoreduction and peripheral-blood progenitor-cell mobilization regimen for transplant-eligible patients with non-Hodgkin’s lymphoma. J Clin Oncol 1999; 17:3776–85.
81 Velasquez WS, McLaughlin P, Tucker S, et al. ESHAP – an effective chemotherapy regimen in refractory and relapsing lymphoma: a 4-year follow-up study. J Clin Oncol 1994; 12:1169–76. 82 Proctor SJ, Taylor PR, Angus B, et al. High-dose ifosfamide in combination with etoposide and epirubicin (IVE) in the treatment of relapsed/refractory Hodgkin’s disease and nonHodgkin’s lymphoma: a report on toxicity and efficacy. Eur J Haematol Suppl 2001; 64:28–32. 83 Girouard C, Dufresne J, Imrie K, et al. Salvage chemotherapy with mini-BEAM for relapsed or refractory non-Hodgkin’s lymphoma prior to autologous bone marrow transplantation. Ann Oncol 1997; 8:675–80. 84 Kewalramani T, Zelenetz AD, Nimer SD, et al. Rituximab and ICE as second-line therapy before autologous stem cell transplantation for relapsed or primary refractory diffuse large B-cell lymphoma. Blood 2004; 103: 3684–8. ●85 Ardeshna KM, Kakouros N, Qian W, et al. Conventional second-line salvage chemotherapy regimens are not warranted in patients with malignant lymphomas who have progressive disease after first-line salvage therapy regimens. Br J Haematol 2005; 130:363–72. ◆86 Blum KA, Lozanski G, Byrd JC. Adult Burkitt leukemia and lymphoma. Blood 2004; 104:3009–20. 87 Murphy SB, Hustu HO. A randomized trial of combined modality therapy of childhood non-Hodgkin’s lymphoma. Cancer 1980; 45:630–7. 88 Divine M, Lepage E, Briere J, et al. Is the small non-cleavedcell lymphoma histologic subtype a poor prognostic factor in adult patients? A case-controlled analysis. The Groupe d’Etude des Lymphomes de l’Adulte. J Clin Oncol 1996; 14:240–8. 89 Lopez TM, Hagemeister FB, McLaughlin P, et al. Small noncleaved cell lymphoma in adults: superior results for stages I–III disease. J Clin Oncol 1990; 8:615–22. 90 Thomas DA, Cortes J, O’Brien S, et al. Hyper-CVAD program in Burkitt’s-type adult lymphoblastic leukemia. J Clin Oncol 1999; 17:2461–70. 91 Mead GM, Sydes MR, Walewski J, et al. An international evaluation of CODOX-M and CODOX-M alternating with IVAC in adult Burkitt’s lymphoma: results of a United Kingdom Lymphoma Group LY06 study. Ann Oncol 2002; 13:1264–74. 92 Magrath IT, Adde M, Shad A, et al. Adults and children with small non-cleaved-cell lymphoma have similar excellent outcome when treated with the same chemotherapy regime. J Clin Oncol 1996; 14:925–34. 93 Hoelzer D, Ludwig DA, Thiel E, et al. Improved outcome in adult B-cell acute lymphoblastic leukemia. Blood 1996; 87:495–508. 94 Soussain C, Patte C, Ostronoff M, et al. Small noncleaved cell lymphoma and leukemia in adults, a retrospective study of 65 adults treated with the LMB pediatric protocols. Blood 1995; 85:664–74. 95 Thomas DA, Faderl S, O’Brien S, et al. Chemoimmunotherapy with Hyper-CVAD plus rituximab for adult
References 1067
96
97
98
●99
●100
101
◆102
103
104
105
●106
107
108
Burkitt’s and Burkitt’s type lymphoma (BL) or acute lymphoblastic leukemia (B-ALL). Blood 2005; 106:149a. Cote TR, Biggar RJ, Rosenberg PS, et al. Non-Hodgkin’s lymphoma among people with AIDS: incidence, presentation and public health burden. AIDS/Cancer Study Group. Int J Cancer 1997; 73:645–50. Davi F, Delecluse HJ, Guiet P, et al. Burkitt-like lymphomas in AIDS patients: characterization within a series of 103 human immunodeficiency virus-associated non-Hodgkin’s lymphomas. Burkitt’s Lymphoma Study Group. J Clin Oncol 1998; 16:3788–95. Carbone A, Gloghini A. AIDS-related lymphomas: from pathogenesis to pathology. Br J Haematol 2005; 130:662–70. Cesarman E, Chang Y, Moore PS, et al. Kaposi’s sarcomaassociated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas. N Engl J Med 1995; 332:1186–91. Delecluse HJ, Anagnostopoulos I, Dallenbach F, et al. Plasmablastic lymphomas of the oral cavity: a new entity associated with the human immunodeficiency virus infection. Blood 1997; 89:1413–20. Dong HY, Scadden DT, de Leval L, et al. Plasmablastic lymphoma in HIV-positive patients: an aggressive EpsteinBarr virus-associated extramedullary plasmacytic neoplasm. Am J Surg Pathol 2005; 29:1633–41. Navarro WH, Kaplan LD. AIDS-related lymphoproliferative disease. Blood 2006; 107:13–20. Kaplan LD, Straus DJ, Testa MA, et al. Low-dose compared with standard-dose m-BACOD chemotherapy for nonHodgkin’s lymphoma associated with human immunodeficiency virus infection. National Institute of Allergy and Infectious Diseases AIDS Clinical Trials Group. N Engl J Med 1997; 336:1641–8. Navarro JT, Lloveras N, Ribera JM, et al. The prognosis of HIV-infected patients with diffuse large B-cell lymphoma treated with chemotherapy and highly active antiretroviral therapy is similar to that of HIV-negative patients receiving chemotherapy. Haematologica 2005; 90:704–6. Wang ES, Straus DJ, Teruya-Feldstein J, et al. Intensive chemotherapy with cyclophosphamide, doxorubicin, highdose methotrexate/ifosfamide, etoposide, and high-dose cytarabine (CODOX-M/IVAC) for human immunodeficiency virus-associated Burkitt lymphoma. Cancer 2003; 98:1196–205. Little RF, Pittaluga S, Grant N, et al. Highly effective treatment of acquired immunodeficiency syndrome-related lymphoma with dose-adjusted EPOCH: impact of antiretroviral therapy suspension and tumor biology. Blood 2003; 101:4653–9. Antinori A, Cingolani A, Alba L, et al. Better response to chemotherapy and prolonged survival in AIDS-related lymphomas responding to highly active antiretroviral therapy. AIDS 2001; 15:1483–91. Hoffmann C, Wolf E, Fatkenheuer G, et al. Response to highly active antiretroviral therapy strongly predicts
109
110
111
112
113
114
115
116
117
118
●119
120
outcome in patients with AIDS-related lymphoma. AIDS 2003; 17:1521–9. Levine AM, Tulpule A, Espina B, et al. Liposomeencapsulated doxorubicin in combination with standard agents (cyclophosphamide, vincristine, prednisone) in patients with newly diagnosed AIDS-related non-Hodgkin’s lymphoma: results of therapy and correlates of response. J Clin Oncol 2004; 22:2662–70. Mounier N, Spina M, Gabarre J, et al. AIDS-related nonHodgkin’s lymphoma: final analysis of 485 patients treated with risk-adapted intensive chemotherapy. Blood 2006: 107:3832–40. Kaplan LD, Lee JY, Ambinder RF, et al. Rituximab does not improve clinical outcome in a randomized phase 3 trial of CHOP with or without rituximab in patients with HIVassociated non-Hodgkin lymphoma: AIDS-Malignancies Consortium Trial 010. Blood 2005; 106:1538–43. Spina M, Jaeger U, Sparano JA, et al. Rituximab plus infusional cyclophosphamide, doxorubicin, and etoposide in HIV-associated non-Hodgkin lymphoma: pooled results from 3 phase 2 trials. Blood 2005; 105:1891–7. Morabito F, Gallamini A, Stelitano C, et al. Clinical relevance of immunophenotype in a retrospective comparative study of 297 peripheral T-cell lymphomas, unspecified, and 496 diffuse large B-cell lymphomas: experience of the Intergruppo Italiano Linformi. Cancer 2004; 101:1601–8. Gisselbrecht C, Gaulard P, Lepage E, et al. Prognostic significance of T-cell phenotype in aggressive nonHodgkin’s lymphomas. Groupe d’Etudes des Lymphomes de l’Adulte (GELA). Blood 1998; 92:76–82. Melnyk A, Rodriguez A, Pugh WC, Cabannillas F. Evaluation of the Revised European-American Lymphoma classification confirms the clinical relevance of immunophenotype in 560 cases of aggressive nonHodgkin’s lymphoma. Blood 1997; 89:4514–20. Gallamini A, Stelitano C, Calvi R, et al. Peripheral T-cell lymphoma unspecified (PTCL-U): a new prognostic model from a retrospective multicentric clinical study. Blood 2004; 103:2474–9. Lopez-Guillermo A, Cid J, Salar A, et al. Peripheral T-cell lymphomas: initial features, natural history, and prognostic factors in a series of 174 patients diagnosed according to the R.E.A.L. Classification. Ann Oncol 1998; 9:849–55. Savage KJ, Chhanabhai M, Gascoyne RD, Connors JM. Characterization of peripheral T-cell lymphomas in a single North American institution by the WHO classification. Ann Oncol 2004; 15:1467–75. Rimokh R, Magaud JP, Berger F, et al. A translocation involving a specific breakpoint (q35) on chromosome 5 is characteristic of anaplastic large cell lymphoma (‘Ki-1 lymphoma’). Br J Haematol 1989; 71:31–6. Gascoyne RD, Aoun P, Wu D, et al. Prognostic significance of anaplastic lymphoma kinase (ALK) protein expression in adults with anaplastic large cell lymphoma. Blood 1999; 93:3913–21.
1068 Non-Hodgkin’s lymphoma
121 Martinez-Delgado B, Melendez B, Cuadros M, et al. Expression profiling of T-cell lymphomas differentiates peripheral and lymphoblastic lymphomas and defines survival related genes. Clin Cancer Res 2004; 10:4971–82. 122 Piccaluga PP, Agostinelli C, Zinzani PL, et al. Expression of platelet-derived growth factor receptor alpha in peripheral T-cell lymphoma not otherwise specified. Lancet Oncol 2005; 6:440. 123 Ballester B, Ramuz O, Gisselbrecht C, et al. Gene expression profiling identifies molecular subgroups among nodal peripheral T-cell lymphomas. Oncogene 2006: 25:1560–70. 124 Brugieres L, Deley MC, Pacquement H, et al. CD30() anaplastic large-cell lymphoma in children: analysis of 82 patients enrolled in two consecutive studies of the French Society of Pediatric Oncology. Blood 1998; 92:3591–8. ◆125 Falini B, Pileri S, Zinzani PL, et al. ALK lymphoma: clinicopathological findings and outcome. Blood 1999; 93:2697–706. 126 Escalon MP, Liu NS, Yang Y, et al. Prognostic factors and treatment of patients with T-cell non-Hodgkin lymphoma: the M. D. Anderson Cancer Center experience. Cancer 2005; 103:2091–8. 127 Williams DM, Hobson R, Imeson J, et al. Anaplastic large cell lymphoma in childhood: analysis of 72 patients treated on The United Kingdom Children’s Cancer Study Group chemotherapy regimens. Br J Haematol 2002; 117:812–20. 128 Blystad AK, Enblad G, Kvaloy S, et al. High-dose therapy with autologous stem cell transplantation in patients with peripheral T cell lymphomas. Bone Marrow Transplant 2001; 27:711–6. 129 Song KW, Mollee P, Keating A, Crump M. Autologous stem cell transplant for relapsed and refractory peripheral T-cell lymphoma: variable outcome according to pathological subtype. Br J Haematol 2003; 120:978–85. 130 Rodriguez J, Caballero MD, Gutierrez A, et al. High-dose chemotherapy and autologous stem cell transplantation in peripheral T-cell lymphoma: the GEL-TAMO experience. Ann Oncol 2003; 14:1768–75. 131 Fanin R, Ruiz de Elvira MC, Sperotto A, et al. Autologous stem cell transplantation for T and null cell CD30-positive anaplastic large cell lymphoma: analysis of 64 adult and paediatric cases reported to the European Group for Blood and Marrow Transplantation (EBMT). Bone Marrow Transplant 1999; 23:437–42. 132 Dogan A, Attygale AD, Kyriakou C. Angioimmunoblastic Tcell lymphoma. Br J Haematol 2003; 121:681–91. 133 Pautier P, Devidas A, Delmer A, et al. Angioimmunoblasticlike T-cell non Hodgkin’s lymphoma: outcome after chemotherapy in 33 patients and review of the literature. Leuk Lymphoma 1999; 32:545–52. 134 Siegert W, Nerl C, Agthe A, et al. Angioimmunoblastic lymphadenopathy (AILD)-type-T-cell lymphoma: prognostic impact of clinical observations and laboratory findings at presentation. The Kiel Lymphoma Study Group. Ann Oncol 1995; 6:659–64. 135 Zettl A, Lee S-S, Rudinger T, et al. Epstein-Barr virusassociated B-cell lymphoproliferative disorders in
angioimmunoblastic T-cell lymphoma and peripheral T-cell lymphoma, unspecified. Am J Clin Pathol 2002; 117:368–79. 136 Attygale AD, Diss TC, Bacon CM, et al. Natural history of angioimmunoblastic T-cell lymphoma in sequential biopsies. USCAP 2004; 1005a. 137 Schetelig J, Fetscher S, Reichle A, et al. Long-term diseasefree survival in patients with angioimmunoblastic T-cell lymphoma after high-dose chemotherapy and autologous stem cell transplantation. Haematologica 2003; 88: 1272–8. 138 Dogan A, Ngu LS, Ng SH, Cervi PL. Pathology and clinical features of angioimmunoblastic T-cell lymphoma after successful treatment with thalidomide. Leukemia 2005; 19: 873–5. 139 Joly B, Frenkel V, Gaulard P, et al. Rituximab in combination with CHOP regimen in angioimmunoblastic T-cell lymphoma (AITL). Preliminary results in 9 patients treated in a single institution. Blood 2005; 106:2686a. 140 Nicot C. Current views in HTLV-I-associated adult T-cell leukemia/lymphoma. Am J Hematol 2005; 78:232–9. 141 Shimoyama M. Diagnostic criteria and classification of clinical subtypes of adult T-cell leukaemia-lymphoma. A report from the Lymphoma Study Group (1984–87). Br J Haematol 1991; 79: 428–37. 142 Uchiyama T, Yodoi J, Sagawa K, et al. Adult T-cell leukemia: clinical and hematologic features of 16 cases. Blood 1977; 50:481–92. ◆143 Taylor GP, Matsuoka M. Natural history of adult T-cell leukemia/lymphoma and approaches to therapy. Oncogene 2005; 24:6047–57. 144 Pawson R, Richardson DS, Pagliuca A, et al. Adult T-cell leukemia/lymphoma in London: clinical experience of 21 cases. Leuk Lymphoma 1998; 31:177–85. 145 Tsukasaki K, Ikeda S, Murata K, et al. Characteristics of chemotherapy-induced clinical remission in long survivors with aggressive adult T-cell leukemia/lymphoma. Leuk Res 1993; 17:157–66. 146 Shimoyama M, Ota K, Kikuchi M, et al. Major prognostic factors of adult patients with advanced T-cell lymphoma/leukemia. J Clin Oncol 1988; 6:1088–97. ●147 Gill PS, Harrington W Jr, Kaplan MH, et al. Treatment of adult T-cell leukemia-lymphoma with a combination of interferon alfa and zidovudine. N Engl J Med 1995; 332:1744–8. 148 Matutes E, Taylor GP, Cavenagh J, et al. Interferon alpha and zidovudine therapy in adult T-cell leukaemia lymphoma: response and outcome in 15 patients. Br J Haematol 2001; 113:779–84. 149 Besson C, Panelatti G, Delaunay C, et al. Treatment of adult T-cell leukemia-lymphoma by CHOP followed by therapy with antinucleosides, alpha interferon and oral etoposide. Leuk Lymphoma 2002; 43:2275–9. 150 Fukushima T, Miyazaki Y, Honda S, et al. Allogeneic hematopoietic stem cell transplantation provides sustained long-term survival for patients with adult T-cell leukemia/lymphoma. Leukemia 2005; 19:829–34.
References 1069
151 Ishida T, Inagaki H, Utsunomiya A, et al. CXC chemokine receptor 3 and CC chemokine receptor 4 expression in T-cell and NK-cell lymphomas with special reference to clinicopathological significance for peripheral T-cell lymphoma, unspecified. Clin Cancer Res 2004; 10:5494–500. 152 Tsuchiya T, Ohshima K, Karube K, et al. Th1, Th2 and activated T-cell marker and clinical prognosis in peripheral T-cell lymphoma, unspecified: comparison with AILD, ALCL lymphoblastic lymphoma and ATLL. Blood 2004; 103:236–41. 153 Catassi C, Fabiani E, Corrao G, et al. Risk of non-Hodgkin lymphoma in celiac disease. JAMA 2002; 287: 1413–9. 154 Howell WM, Leung ST, Jones DB, et al. HLA-DRB, -DQA, and -DQB polymorphism in celiac disease and enteropathyassociated T-cell lymphoma. Common features and additional risk factors for malignancy. Hum Immunol 1995; 43:29–37. 155 Cellier C, Delabesse E, Helmer C, et al. Refractory sprue, coeliac disease, and enteropathy-associated T-cell lymphoma. French Coeliac Disease Study Group. Lancet 2000; 356:203–8. 156 Zettl A, Ott G, Makulik A, et al. Chromosomal gains at 9q characterize enteropathy-type T-cell lymphoma. Am J Pathol 2002; 161:1635–45. 157 Obermann EC, Diss TC, Hamoudi RA, et al. Loss of heterozygosity at chromosome 9p21 is a frequent finding in enteropathy-type T-cell lymphoma. J Pathol 2004; 202:252–62. 158 Egan LJ, Walsh SV, Stevens FM, et al. Celiac-associated lymphoma. A single institution experience of 30 cases in the combination chemotherapy era. J Clin Gastroenterol 1995; 21:123–9. 159 Gale J, Simmonds PD, Mead GM, et al. Enteropathy-type intestinal T-cell lymphoma: clinical features and treatment of 31 patients in a single center. J Clin Oncol 2000; 18:795–803. 160 Daum S, Ullrich R, Heise W, et al. Intestinal non-Hodgkin’s lymphoma: a multicenter prospective clinical study from the German Study Group on Intestinal non-Hodgkin’s Lymphoma. J Clin Oncol 2003; 21:2740–6. 161 Lennard AL, White J, Tiplady C. Prospective evaluation of a novel treatment protocol for intestinal T-cell NHL. Ann Oncol 2002; 13 suppl 2:43a. 162 You JY, Chi KH, Yang MH, et al. Radiation therapy versus chemotherapy as initial treatment for localized nasal natural killer (NK)/T-cell lymphoma: a single institute survey in Taiwan. Ann Oncol 2004; 15:618–25. 163 Chim CS, Ma SY, Au WY, et al. Primary nasal natural killer cell lymphoma: long-term treatment outcome and relationship with the International Prognostic Index. Blood 2004; 103:216–21. 164 Yamaguchi M, Kita K, Miwa H, et al. Frequent expression of P-glycoprotein/MDR1 by nasal T-cell lymphoma cells. Cancer 1995; 76:2351–6. 165 Murashige N, Kami M, Kishi Y, et al. Allogeneic haemopoietic stem cell transplantation as a promising
166
167
168
169
170
171
●172
173
174
175
176
◆177
●178
179
treatment for natural killer-cell neoplasms. Br J Haematol 2005; 130:561–7. Belhadj K, Reyes F, Farcet JP, et al. Hepatosplenic gammadelta T-cell lymphoma is a rare clinicopathologic entity with poor outcome: report on a series of 21 patients. Blood 2003; 102:4261–9. Mansour MR, Dogan A, Morris EC, et al. Allogeneic transplantation for hepatosplenic alphabeta T-cell lymphoma. Bone Marrow Transplant 2005; 35: 931–4. Lorsbach RB, Shay-Seymore D, Moore J, et al. Clinicopathologic analysis of follicular lymphoma occurring in children. Blood 2002; 99:1959–64. Rodriguez J, McLaughlin P, Hagemeister FB, et al. Follicular large cell lymphoma: an aggressive lymphoma that often presents with favorable prognostic features. Blood 1999; 93:2202–7. Perea G, Altes A, Montoto S, et al. Prognostic indexes in follicular lymphoma: a comparison of different prognostic systems. Ann Oncol 2005; 16:1508–13. Lopez-Guillermo A, Cabanillas F, McDonnell TI, et al. Correlation of bcl-2 rearrangement with clinical characteristics and outcome in indolent follicular lymphoma. Blood 1999; 93:3081–7. Hainsworth JD, Litchy S, Burris 3rd HA, et al. Rituximab as first-line and maintenance therapy for patients with indolent non-hodgkin’s lymphoma. J Clin Oncol 2002; 20:4261–7. de Jong D. Molecular pathogenesis of follicular lymphoma: a cross talk of genetic and immunologic factors. J Clin Oncol 2005; 23:6358–63. Glas AM, Kersten MJ, Delahaye LJ, et al. Gene expression profiling in follicular lymphoma to assess clinical aggressiveness and to guide the choice of treatment. Blood 2005; 105:301–7. Young RC, Longo DL, Glatstein E, et al. The treatment of indolent lymphomas: watchful waiting v aggressive combined modality treatment. Semin Hematol 1988; 25:11–6. Brice P, Bastion Y, Lepage E, et al. Comparison in lowtumor-burden follicular lymphomas between an initial notreatment policy, prednimustine, or interferon alfa: a randomized study from the Groupe d’Etude des Lymphomes Folliculaires. Groupe d’Etude des Lymphomes de l’Adulte. J Clin Oncol 1997; 15:1110–7. Ardeshna KM, Smith P, Norton A, et al. Long-term effect of a watch and wait policy versus immediate systemic treatment for asymptomatic advanced-stage non-Hodgkin lymphoma: a randomised controlled trial. Lancet 2003; 362:516–22. Maloney DG, Grillo-Lopez AJ, White CA, et al. IDEC-C2B8 (Rituximab) anti-CD20 monoclonal antibody therapy in patients with relapsed low-grade non-Hodgkin’s lymphoma. Blood 1997; 90:2188–95. Colombat P, Salles G, Brousse N, et al. Rituximab (antiCD20 monoclonal antibody) as single first-line therapy for patients with follicular lymphoma with a low tumor
1070 Non-Hodgkin’s lymphoma
180
181
182
183
●184
●185
●186
187
●188
●189
190
burden: clinical and molecular evaluation. Blood 2001; 97:101–6. Witzig TE, Vukov AM, Habermann TM, et al. Rituximab therapy for patients with newly diagnosed, advancedstage, follicular grade I non-Hodgkin’s lymphoma: a phase II trial in the North Central Cancer Treatment Group. J Clin Oncol 2005; 23:1103–8. Piro LD, White CA, Grillo-Lopez A, et al. Extended rituximab (anti-CD20 monoclonal antibody) therapy for relapsed or refractory low-grade or follicular non-Hodgkin’s lymphoma. Ann Oncol 1999; 10:655–61. Ghielmini M, Schmitz SF, Cogliatti SB, et al. Prolonged treatment with rituximab in patients with follicular lymphoma significantly increases event-free survival and response duration compared with the standard weekly 4 schedule. Blood 2004; 103:4416–23. Gordan LN, Grow WB, Pusateri A, et al. Phase II trial of individualized rituximab dosing for patients with CD20positive lymphoproliferative disorders. J Clin Oncol 2005; 23:1096–102. Marcus R, Imrie K, Belch A, et al. CVP chemotherapy plus rituximab compared with CVP as first-line treatment for advanced follicular lymphoma. Blood 2005; 105:1417–23. Hiddemann W, Kneba M, Dreyling M, et al. Frontline therapy with rituximab added to the combination of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) significantly improves the outcome for patients with advanced-stage follicular lymphoma compared with therapy with CHOP alone: results of a prospective randomized study of the German Low-Grade Lymphoma Study Group. Blood 2005; 106:3725–32. Herold M, Pasold R, Srock S, et al. Rituximab plus mitoxantrone, chlorambucil, prednisolone (R-MCP) is superior to MCP alone in advanced indolent and follicular lymphoma – results of a phase III study (OSHO39). Ann Oncol 2005; 16 suppl 5:60a. Herold M, Pasold R, Srock S, et al. Results of a prospective randomised open label phase III study comparing rituximab plus mitoxantrone, chlorambucil, prednisolone chemotherapy (R-MCP) versus MCP alone in untreated advanced indolent non-Hodgkin’s lymphoma (NHL) and mantle-cell-lymphoma (MCL). Blood 2004; 104:584a. Salles GA, Foussard C, Nicolas M, et al. Rituximab added alphaIFNCHVP improves the outcome of follicular lymphoma patients with a high tumor burden: to first analysis of the GELA-GOELAMS FL-2000 randomized trial in 359 patients. Blood 2004; 104:160a. Hochster HS, Weller E, Gascoyne RD, et al. Maintenance rituximab after CVP results in superior clinical outcome in advanced follicular lymphoma (FL): results of the E1496 phase III trial from the Eastern Cooperative Oncology Group and the Cancer and Leukemia Group B. Blood 2005; 106:349a. Ghielmini M, Rufibach K, Salles G, et al. Single agent rituximab in patients with follicular or mantle cell
◆191
192
●193
194
◆195
196
197
198
199
200
201
202
203
204
lymphoma: clinical and biological factors that are predictive of response and event-free survival as well as the effect of rituximab on the immune system: a study of the Swiss Group for Clinical Cancer Research (SAKK). Ann Oncol 2005; 16:1675–82. Hiddemann W, Buske C, Dreyling M, et al. Treatment strategies in follicular lymphomas: current status and future perspectives. J Clin Oncol 2005; 23:6394–9. Press OW, Unger JM, Braziel RM, et al. A phase 2 trial of CHOP chemotherapy followed by tositumomab/iodine I-131 tositumomab for previously untreated follicular nonHodgkin lymphoma: Southwest Oncology Group Protocol S9911. Blood 2003; 102:1606–12. Kaminski MS, Tuck M, Estes J, et al. 131-I-tositumomab therapy as initial treatment for follicular lymphoma. N Engl J Med 2005; 352:441–9. Leonard JP, Coleman M, Kostakoglu L, et al. Abbreviated chemotherapy with fludarabine followed by tositumomab and iodine I-131 tositumomab for untreated follicular lymphoma. J Clin Oncol 2005; 23:5696–704. Tse WW, Lazarus HM, Van Besien K. Stem cell transplantation in follicular lymphoma: progress at last? Bone Marrow Transplant 2004; 34:929–38. Bohen SP, Troyanskaya OG, Alter O, et al. Variation in gene expression patterns in follicular lymphoma and the response to rituximab. Proc Natl Acad Sci U S A 2003; 100:1926–30. Guadagnolo BA, Li S, Neuberg D, et al. Long-term outcome and mortality trends in early-stage, Grade 1–2 follicular lymphoma treated with radiation therapy. Int J Radiat Oncol Biol Phys 2006; 64: 928–34. Haas RLM, Poortmans P, de Jong D, et al. High response rates and lasting remissions after low-dose involved radiotherapy in indolent lymphomas. J Clin Oncol 2003; 21:2474–80. Hoskin P, et al. Radiation dose in non-Hodgkin’s lymphoma: preliminary results of UK NCRN randomised trial. Ann Oncol 2005; 16 suppl 5:51a. Kelsey SM, Newland AC, Hudson GV, Jelliffe AM. A British National Lymphoma Investigation randomised trial of single agent chlorambucil plus radiotherapy versus radiotherapy alone in low grade, localised non-Hodgkins lymphoma. Med Oncol 1994; 11:19–25. Seymour JF, Pro B, Fuller LM, et al. Long-term follow-up of a prospective study of combined modality therapy for stage I-II indolent non-Hodgkin’s lymphoma. J Clin Oncol 2003; 21:2115–22. Advani R, Rosenberg SA, Horning SJ. Stage I and II follicular non-Hodgkin’s lymphoma: long-term follow-up of no initial therapy. J Clin Oncol 2004; 22:1454–9. Freedman AS. Non-transplant-related treatment options in follicular lymphoma. Biol Blood Marrow Transplant 2006; 12:53–8. Davis TA, Grillo-Lopez AJ, White CA, et al. Rituximab antiCD20 monoclonal antibody therapy in non-Hodgkin’s lymphoma: safety and efficacy of re-treatment. J Clin Oncol 2000; 18:3135–43.
References 1071
205 Hainsworth JD, Litchy S, Shaffer DW, et al. Maximizing therapeutic benefit of rituximab: maintenance therapy versus re-treatment at progression in patients with indolent non-Hodgkin’s lymphoma-a randomized phase II trial of the Minnie Pearl Cancer Research Network. J Clin Oncol 2005; 23:1088–95. ●206 Schouten HC, Qian W, Kvaloy S, et al. High-dose therapy improves progression-free survival and survival in relapsed follicular non-Hodgkin’s lymphoma: results from the randomized European CUP trial. J Clin Oncol 2003; 21:3918–27. ●207 de Boer CJ, van Krieken JH, Kluin-Nelemans HC, et al. Cyclin D1 messenger RNA overexpression as a marker for mantle cell lymphoma. Oncogene 1995; 10:1833–40. 208 Raty R, Franssila K, Jansson SE, et al. Predictive factors for blastoid transformation in the common variant of mantle cell lymphoma. Eur J Cancer 2003; 39:321–9. 209 Angelopoulou MK, Siakantariz MP, Vassilakapoulos TP, et al. The splenic form of mantle cell lymphoma. Eur J Haematol 2002; 68:12–21. 210 Forstpointner R, Dreyling M, Repp R, et al. The addition of rituximab to a combination of fludarabine, cyclophosphamide, mitoxantrone (FCM) significantly increases the response rate and prolongs survival as compared with FCM alone in patients with relapsed and refractory follicular and mantle cell lymphomas: results of a prospective randomized study of the German Low-Grade Lymphoma Study Group. Blood 2004; 104:3064–71. 211 Lenz G, Dreyling M, Hoster E, et al. Immunochemotherapy with rituximab and cyclophosphamide, doxorubicin, vincristine, and prednisone significantly improves response and time to treatment failure, but not long-term outcome in patients with previously untreated mantle cell lymphoma: results of a prospective randomized trial of the German Low Grade Lymphoma Study Group (GLSG). J Clin Oncol 2005; 23:1984–92. 212 Rule S, Burton C, Walseski J, et al. A randomised phase II study of fludarabine/cyclophosphamide / Rituximab in patients with untreated mantle cell lymphoma. Ann Oncol 2005; 16 suppl 5:198a. 213 Romaguera JE, Fayad L, Rodriguez MA, et al. High rate of durable remissions after treatment of newly diagnosed aggressive mantle-cell lymphoma with rituximab plus hyper-CVAD alternating with rituximab plus high-dose methotrexate and cytarabine. J Clin Oncol 2005; 23:7013–23. 214 Lenz G, Dreyling M, Schiegnitz E, et al. Myeloablative radiochemotherapy followed by autologous stem cell transplantation in first remission prolongs progression-free survival in follicular lymphoma: results of a prospective, randomized trial of the German Low-Grade Lymphoma Study Group. Blood 2004; 104:2667–74. 215 Khouri IF, Romaguera J, Kantarjian H, et al. Hyper-CVAD and high-dose methotrexate/cytarabine followed by stem-cell transplantation: an active regimen for
◆216
217
◆218
219
220
221
222
223
●224
225
226
227
228
aggressive mantle-cell lymphoma. J Clin Oncol 1998; 16:3803–9. Witzig TE. Current treatment approaches for mantle-cell lymphoma. J Clin Oncol 2005; 23:6409–14. Farinha P, Gascoyne RD. Molecular pathogenesis of mucosa-associated lymphoid tissue lymphoma. J Clin Oncol 2005; 23:6370–8. Bertoni F, Zucca E. State-of-the-art therapeutics: marginal-zone lymphoma. J Clin Oncol 2005; 23:6415–20. Tsang RW, Gospodarowicz MK, Pintilie M, et al. Localized mucosa-associated lymphoid tissue lymphoma treated with radiation therapy has excellent clinical outcome. J Clin Oncol 2003; 21:4157–64. Arcaini L, Lazzarino M, Colombo N, et al. Splenic marginal zone lymphoma: a prognostic model for clinical use. Blood 2006; 107:4643–49. Schimmer AD, Jamal S, Messner H, et al. Allogeneic or autologous bone marrow transplantation (BMT) for nonHodgkin’s lymphoma (NHL): results of a provincial strategy. Ontario BMT Network, Canada. Bone Marrow Transplant 2000; 26:859–64. Peniket AJ, Ruiz de Elvira MC, Taghipour G, et al. An EBMT registry matched study of allogeneic stem cell transplants for lymphoma: allogeneic transplantation is associated with a lower relapse rate but a higher procedure-related mortality rate than autologous transplantation. Bone Marrow Transplant 2003; 31:667–78. Bierman PJ, Sweetenham JW, Loberiza FR Jr, et al. Syngeneic hematopoietic stem-cell transplantation for non-Hodgkin’s lymphoma: a comparison with allogeneic and autologous transplantation – The Lymphoma Working Committee of the International Bone Marrow Transplant Registry and the European Group for Blood and Marrow Transplantation. J Clin Oncol 2003; 21: 3744–53. Weiden PL, Flournoy N, Thomas ED, et al. Antileukemic effect of graft-versus-host disease in human recipients of allogeneic marrow grafts. N Engl J Med 1979; 300:1068–73. Kolb HJ, Mittermuller J, Clemm C, et al. Donor leukocyte transfusions for treatment of recurrent chronic myelogenous leukemia in marrow transplant patients. Blood 1990; 15:2462–5. Butcher BW, Collins Jr RH. The graft-versus-lymphoma effect: clinical review and future opportunities. Bone Marrow Transplant 2005; 36:1–17. Peggs KS, Thomson K, Hart DP, et al. Dose-escalated donor lymphocyte infusions following reduced intensity transplantation: toxicity, chimerism, and disease responses. Blood 2004; 103:1548–56. Russell NH, Byrne JL, Faulkner RD, et al. Donor lymphocyte infusions can result in sustained remissions in patients with residual or relapsed lymphoid malignancy following allogeneic haemopoietic stem cell transplantation. Bone Marrow Transplant 2005; 36: 437–41.
1072 Non-Hodgkin’s lymphoma
229 Grigg A, Ritchie D. Graft-versus-lymphoma effects: clinical review, policy proposals, and immunobiology. Biol Blood Marrow Transplant 2004; 10:579–90. 230 Mielcarek M, Martin PJ, Leisenring W, et al. Graft-versushost disease after nonmyeloablative versus conventional hematopoietic stem cell transplantation. Blood 2003; 102:756–62. 231 Dhedin N, Giraudier S, Gaulard P, et al. Allogeneic bone marrow transplantation in aggressive non-Hodgkin’s lymphoma (excluding Burkitt and lymphoblastic lymphoma): a series of 73 patients from the SFGM database. Societ
Francaise de Greffe de Moelle. Br J Haematol 1999; 107:154–61. ◆232 Peggs KS, Mackinnon S, Linch DC. The role of allogeneic transplantation in non-Hodgkin’s lymphoma. Br J Haematol 2005; 128:153–68. 233 Ferlay J, Bray F, Pisani P, Parkin DM. GLOBOCAN 2002: Cancer Incidence, Mortality and Prevalence Worldwide. Lyon: LARC CancerBase No. 5 version 2.0; 2004. 234 Ries LAG, Eisner MP, Kosary CL, et al. SEER Cancer Statistics Review 1975–2002: National Cancer Institute, Bethesda MD; 2005.
44 Multiple myeloma JENNIFER M. BIRD AND CATHY WILLIAMS
Introduction Incidence and aetiology Pathogenesis and biology Clinical features and complications Investigation, diagnosis and staging Prognostic factors and staging Course of the disease When to start treatment and criteria for response General aspects of care Chemotherapy
1073 1073 1074 1076 1078 1081 1081 1082 1082 1083
INTRODUCTION Multiple myeloma (MM) is a malignant disease of plasma cells in the bone marrow and forms part of a spectrum of plasma cell disorders that also includes monoclonal gammopathy of undetermined significance (MGUS), solitary plasmacytomas (bone or extramedullary) and primary amyloidosis. Myeloma is characterized by the production of a monoclonal immunoglobulin (Ig) molecule and is associated with clinical symptoms relating to bone destruction, disturbed bone marrow function, reduction of normal immune function and renal failure. In 80 per cent of patients there is a paraprotein in the serum, usually of IgG or IgA subtype. The abnormal cells may also produce free light chains, which pass into the urine as Bence Jones protein (BJP), frequently causing tubular damage. In 20 per cent of patients, free light chains only are produced (Bence Jones only myeloma) and there is no paraprotein in the serum. The number of plasma cells in the bone marrow is increased and osteoclasts are activated in the region of plasma cell foci, causing bone resorption. The residual normal B-cell population is suppressed, leading to a reduction in polyclonal immunoglobulins (immune paresis) and increased susceptibility to infection. The median survival is 3–4 years, although survival varies widely according to a number of prognostic factors.
Radiotherapy High-dose therapy and stem-cell transplantation Maintenance therapy Bisphosphonates Refractory and relapsed disease New approaches to treatment Other related conditions Key references References
1085 1086 1088 1089 1089 1090 1090 1092 1092
Updated guidelines on the diagnosis and management of myeloma have recently been produced by the UK Myeloma Forum and Nordic Myeloma Study Group on behalf of the British Committee for Standards in Haematology.1
INCIDENCE AND AETIOLOGY Incidence Myeloma accounts for about 1 per cent of all cancers and 10 per cent of haematological malignancies. Internationally, the reported incidence of myeloma varies substantially. Multiple myeloma has an annual incidence in the UK and North America of approximately 40 per million and a median age at presentation of about 70 years.2,3 There are approximately 2500 new cases each year in the UK. There is some evidence to suggest that the incidence of myeloma has been increasing in recent years.4,5 The incidence increases with age; approximately 15 per cent of patients are aged less than 60 years and a further 15 per cent are aged between 60 and 65 years. Fewer than 2 per cent of myeloma patients are under 40 years of age at diagnosis. It is more common in males, with a male:female ratio of around 1.5:1. Myeloma has a higher incidence in Afro-Caribbean ethnic groups compared to Caucasians.
1074 Multiple myeloma
In the USA, African-Americans have a reported incidence 1.5 to 2 times that of Caucasian Americans. Conversely, Asians living in the USA appear to have low incidence rates. In the UK, the incidence in immigrants from West Africa and the Caribbean is two to four times that in native Britons.
Epidemiological risk factors Most cases of myeloma present de novo; a minority evolve from previously diagnosed monoclonal MGUS, an asymptomatic disorder involving clonal proliferation of plasma cells and paraprotein production. However, a far greater number may arise from previously undiagnosed MGUS.
Radiation exposure The only clearly defined risk factor for the development of myeloma is exposure to ionizing radiation, as in Japanese survivors of the atomic bombs dropped on Hiroshima and Nagasaki in 1945, and in radiologists exposed to relatively large doses of irradiation over many years. There have been numerous studies of radiation-exposed workers at nuclear plants in the USA, UK and other countries, which have generally shown an increased incidence of myeloma compared to the unexposed population, and also a direct relationship between radiation dose and mortality.6 Recently, a study of Chernobyl clean-up workers showed the incidence of myeloma to be double that expected in the general population.7 Some authors have observed a modest increase in myeloma incidence in patients who have received therapeutic irradiation or have been exposed to numerous diagnostic X-ray procedures,8 but evidence from a study in the USA suggested that exposure to diagnostic X-rays does not increase the risk of developing MM.9
Other epidemiological risk factors Numerous studies have been published looking at exposure to chemical and physical agents, including solvents, pesticides and metals.6,10 The incidence of myeloma appears to be increased in persons exposed to benzene, petroleum refinery waste or pesticides. The use of pesticides may partly account for the increased risk of myeloma which has been observed in agricultural workers. The use of hair dyes has been reported in some (but not all) studies to be associated with an increased risk of myeloma, although the incidence in hairdressers is not increased.6 The possibility that chronic immune stimulation could give rise to an increased risk of myeloma has been investigated. Epidemiological studies in patients with allergic conditions and those undergoing immunizations do not suggest that immune stimulation increases myeloma risk. Similarly, there is no clear evidence for an increased risk in patients with autoimmune disease.
Monoclonal gammopathy of undetermined significance MGUS is a condition where a patient is found by chance to have a paraprotein in the serum but no clinical features of myeloma. There are fewer than 10 per cent plasma cells in the marrow, no bone lesions, and normal haemoglobin and renal function (see Table 44.2). Although many patients with MGUS remain stable throughout their lifetime, longterm follow-up has shown that the annual risk of transformation to myeloma is around 1 per cent.11 Epidemiological studies have identified similar risk factors for MGUS and myeloma. The incidence of both conditions increases with age, is higher among males than females, and in AfricanAmericans than in Caucasians, while both have been reported in the same family. The association of MGUS and myeloma is consistent with the concept of multi-step carcinogenesis in the development of myeloma, with a first oncogenic event causing MGUS and one or more subsequent events leading to malignant transformation.
Familial myeloma There have been a small number of reports of myeloma and MGUS developing in two or more members of one family, but it is unclear whether this reflects genetic or environmental factors.3,12,13
PATHOGENESIS AND BIOLOGY There have been substantial advances in our understanding of the biology of myeloma and this has led to the development of a range of new agents used for treatment, many targeting interactions and pathways between myeloma cells and the bone marrow microenvironment. Multiple myeloma is thought to arise frequently on a background of previous MGUS, a premalignant condition. In at least half of MGUS patients, clonal plasma cell proliferation is thought to be initiated by primary translocations involving the immunoglobulin heavy chain (IgH) locus on chromosome 14q32.14 Not all patients with MGUS develop myeloma and the precise causes of this transformation remain unclear. Additional genetic abnormalities such as ras/p53 mutations or secondary translocations and complex interactions with the bone marrow microenvironment, particularly bone marrow stromal cells and other normal cellular constituents of the bone marrow, are all thought to be important. These help to promote the ability of myeloma cells to survive and resist conventional proapoptotic stimuli, particularly conventional chemotherapy. Increased osteoclast activity and corresponding reduced osteoblast activity, mediated by increased levels of receptor activator of nuclear factor-kB ligand (RANKL) and decreased levels of osteoprotegerin (the decoy receptor for RANKL), result in lytic bone lesions and osteoporosis.
Pathogenesis and biology 1075
Cell of origin Myeloma is a tumour of plasma cells in the bone marrow, but the precise nature of the myeloma stem cell has not been fully elucidated. The cell of origin is likely to be a postgerminal centre plasma cell that has already undergone somatic mutation (see later) and subsequently homes to the bone marrow. The diversity of the human antibody repertoire derives largely from the fact that immunoglobulin heavy and light chain genes are assembled somatically from multiple component germ-line gene segments,15 a process known as variable region recombination. During B-cell maturation, immunoglobulin heavy chain (IgH) genes undergo a further series of changes involving DNA breaks and recombination. Somatic hypermutation resulting from antigen exposure is also a significant factor in contributing to antibody diversity. Isotype switching is the process by which a given Ig variable region can be associated with different heavy chain isotype. This means that antibodies produced from a given lymphocyte retain their original specificity but are able to perform different effector functions. In myeloma, the majority of translocations involve the IgH locus on chromosome 14q32, the region in which normal switch recombinations occur. Studies of the IgH gene in myeloma marrow have shown that the cell of origin is a post-germinal-centre B-cell. There is a consistent, unique IgH rearrangement in myeloma cells from individual patients and the complementarity-determining regions of the IgH gene sequence show mutations from the germ-line sequence which are identical in all the cells of the clone.16,17 This indicates that the malignant transformation must have occurred in a B-cell which has already undergone somatic mutation, i.e. has been exposed to antigen in the germinal centre of the lymph node. IgH gene rearrangement studies have also shown that peripheral-blood B lymphocytes contain some cells belonging to the myeloma clone.18 It is considered likely that malignant transformation occurs in a post-germinal-centre B-cell in the lymph node and that the progeny of this cell migrate to the marrow via the peripheral blood and develop there into plasma cell tumours. Expression of adhesion molecules, including syndecan 1 (CD138), neural cell adhesion molecule (N-CAM; CD56), and platelet endothelial cell adhesion molecule (PECAM 1; CD31) by myeloma plasma cells is thought to facilitate homing to the marrow.19
Myeloma and the BM micro-environment There is a close relationship between myeloma cells and the marrow micro-environment, which promotes tumour growth and also leads to the development of myeloma bone disease. Many of these interactions and pathways can be targeted for the development of new therapies.20
INTERACTIONS OF MYELOMA CELLS WITH BONE MARROW (BM) STROMAL CELLS, CYTOKINES AND SIGNALLING CASCADES
Interactions between myeloma cells and the bone marrow microenvironment including extracellular proteins, stromal cells (a heterogenous group of cells with fibroblast-like morphology), osteoblasts and osteoclasts, are mediated by cell-to-cell contact, adhesion molecules and cytokines.21 Adherence of myeloma cells to stromal cells induces cytokine production, in both an autocrine (from the myeloma cells) and paracrine (from the BM stromal cells) fashion and also stimulates signaling cascades within myeloma cells. There are numerous cytokines important in the pathophysiology of myeloma. These factors, produced in the BM microenvironment, activate a number of signalling pathways that in turn trigger downstream events. These include anti-apoptotic mechanisms caused by up-regulation of anti-apoptotic factors (e.g. Mcl-1, bcl-2) and degradation both of pro-apoptotic mediators and negative cell cycle regulators. Cytokines known to be important include interleukin-6, insulin growth factor-1 (IGF-1), IL-1α and β, VEGF, TNF-α, stroma derived factor-1 and hepatocyte growth factor. These pathways can also be activated by direct cell adhesion with similar results. One of the most important cytokines in myeloma pathogenesis is IL-6, a major growth and survival factor for myeloma plasma cells, which themselves express specific receptors for IL-6.22 Adhesion of myeloma cells to bone marrow stromal cells induces IL-6 production.23 IL-6 triggers myeloma cell proliferation through a range of signalling cascades including activation of the PI3/Akt and JAK/STAT3 pathways. This results in protection against apoptosis by dexamethasone. In addition, IL-6 induces expression and secretion of vascular endothelial growth factor (VEGF) by myeloma cells, stimulating angiogenesis, which promotes tumour growth. Both IL-6 and VEGF inhibit the antigen presenting function of dendritic cells, contributing to the reduced level of immunity seen in myeloma patients. Clinically, raised serum IL-6 levels in myeloma patients have been identified as a poor prognostic factor.24 IL-6 and its receptor are potential targets for therapeutic intervention, as are intracellular pathways that are triggered by IL-6 activation. Disease responses have also been reported in patients treated with monoclonal antibody to IL-6. Plasma cells from patients with myeloma express and secrete vascular endothelial growth factor (VEGF), and the expression of VEGF is upregulated by adherence to marrow stroma.25,26 In addition to VEGF, other angiogenic factors have been identified in myeloma including basic FGF and tissue metalloproteinases.25 The finding of increased angiogenesis in MM provides a rationale for the use of antiangiogenic therapy, such as thalidomide (see later). MYELOMA BONE DISEASE
The main pathological event in myeloma is uncoupling of bone resorption from new bone formation resulting in a
1076 Multiple myeloma
net loss of bone. This process is mediated by increased numbers and activity of osteoclasts and decreased activity of osteoblasts and results from complex interactions between tumour cells and BM stromal cells (reviewed in reference 27). The central pathway in this process involves RANK, the receptor activator of nuclear factor-kappa B, RANKL – the ligand for RANK, and osteoprotegerin (OPG) – a decoy receptor for RANKL. These molecules, are part of the TNF superfamily and are the most important regulators of normal osteoclast activity. Both RANKL and OPG are produced by stromal cells in the marrow, including osteoblasts. RANKL binds to its receptor (RANK) on osteoclast precursors to promote differentiation, and on osteoclasts to stimulate resorption. OPG, the decoy receptor for RANKL, blocks the RANKL-RANK interaction and inhibits osteoclast function, thus preventing bone resportion. The balance between expression of RANKL and OPG determines osteoclast activity and the extent of bone resorption (reviewed in reference 28). The expression of RANKL by stromal cells in myeloma is increased,29 while the expression of OPG is down-regulated.30 Many osteoclast activating factors that are released by myeloma cells directly or by normal marrow cells in response to myeloma cells have been identified. These include IL6, IL-1β, TNFα, hepatocyte growth factor and macrophage-inhibitory protein 1α.31 It is now recognized that most of these factors stimulate osteoclasts by enhancing RANKL expression and that the fundamental abnormality in myeloma bone disease is an altered balance between RANKL and OPG.32 Bone destruction is associated with further cytokine release that perpetuates a cycle of myeloma cell growth and further bone damage. The importance of RANKL and OPG as mediators of bone disease in myeloma is shown by experiments with transgenic mice. Mice that lack RANK or RANKL or that over-express OPG develop osteopetrosis because of decreased osteoclast activity.33 OPG knockout mice develop osteoporosis and fractures because of the lack of inhibition of RANKL activity.34 The importance of RANKL in MM bone disease has been confirmed in animal models, where infusion of RANKL inhibitors including recombinant OPG has been shown to prevent bone resorption.35 In humans, however, despite clinical responses, treatment with recombinant OPG has resulted in the development of anti-OPG antibodies. Recent studies in post-menopausal women with osteoporosis have confirmed safety and efficacy of a single dose of a human monoclonal antibody to RANKL36 and this compound is currently undergoing further clinical trials.
Cytogenetic abnormalities The chromosomal abnormalities described in myeloma are highly complex and reflect a high degree of karyotypic instability. Conventional cytogenetic analysis is often unsuccessful because of the low mitotic rate of plasma cells. Because of
this difficulty and the diversity of chromosomal changes seen, establishing correlates between abnormalities and clinical outcome has been difficult. However, with the use of alternative techniques such as fluorescence in situ hybridization (FISH) and comparative genomic hybridization (CGH), it has become clear that cytogenetic abnormalities are present in at least 75 per cent of cases and that these abnormalities are often complex.37,38 In cases with an abnormal karyotype, translocations occur in 50 per cent, and these usually involve chromosome 14q, the site of the IgH gene.38 Several partner chromosomes have been identified, including 11q, 4p and 8q. The breakpoints in these chromosomes are frequently near the site of cellular protooncogenes, e.g. cyclin D1 on 11q13, FGFR3 (fibroblast growth factor receptor 3) on 4p16 and c-myc on 8q24. These translocations may lead to overexpression of important genes, such as the upregulation of the FGFR3 receptor, which occurs in cases with the t (4;14) translocation. Multiple trisomies are also common, as are losses of all or part of a chromosome. Of particular importance is 13q-, involving loss of the Rb gene, which occurs in at least 35 per cent of cases and which is associated with a poor prognosis.39,40 Other abnormalities reported include point mutations of ras, p53 and Fas. Whether these cytogenetic abnormalities play a role in the origin or growth of the myeloma is as yet unclear. However, cytogenetic abnormalities are also common in MGUS. It appears that translocations of chromosome 14 are as common as in myeloma but chromosome 13 deletions are less frequent in MGUS and may predict for transformation to overt myeloma.41 These observations are consistent with a multi-step pathogenesis for MM in which early dysregulation of oncogenes leads to immortalization of plasma cells and subsequent cytogenetic changes such as loss of Rb or activating ras mutations lead to tumour progression. Gene expression profiling, along with standard cytogenetic and molecular techniques, may lead to a molecular classification of myeloma that may identify specific targets for therapy.42
CLINICAL FEATURES AND COMPLICATIONS The clinical presentation of myeloma is varied. Presenting features include symptoms of bone disease, impaired renal function, anaemia, hypercalcaemia, recurrent or persistent bacterial infection, symptoms of hyperviscosity or of spinal cord or nerve root compression. AL amyloidosis is also associated with myeloma and other plasma cell disorders and some patients with myeloma may present with symptoms of associated nephrotic syndrome or cardiac failure. These symptoms or abnormal test results found when investigating a patient with newly diagnosed myeloma closely reflect the definitions of myeloma-related organ or tissue impairment developed by an international working group in 2003 and which form part of the basis for a new
Clinical features and complications 1077
classification of MGUS and myeloma43 described in more detail in the next section.
Bone disease and hypercalcaemia Bone disease is a major feature of multiple myeloma. Pain arising from the skeletal complications of myeloma is the commonest presenting symptom and these complications, especially vertebral compression/collapse, occurs in the majority of patients with myeloma and up to 60 per cent at diagnosis.44 The destruction of bone in myeloma results from an increased activity of osteoclasts, which is not accompanied by a comparable increase of osteoblast function. X-rays may show lytic lesions and/or generalized osteoporosis. Some patients have osteoporosis without lytic lesions. Persistent, unexplained back pain is the commonest presenting symptom of myeloma bone disease. The ribs are also a common site of pain. Vertebral collapse is frequent leading to back pain, kyphosis and loss of height, and occasionally resulting in cord compression. Pathological fracture of long bones may also occur. Bone destruction can give rise to hypercalcaemia, although serum alkaline phosphatase (ALP) is usually not increased as this bone enzyme reflects osteoblast activity and in myeloma the osteoblasts are inhibited. For the same reason, radionuclide bone scans may be negative in myeloma and a radiological skeletal survey is recommended for assessing the site and extent of bone damage.
Renal failure Renal impairment is also common in myeloma patients. About one third of patients present with significant renal impairment45 and it will affect approximately half at some point during the course of the illness.46 A small number of patients (3–12 per cent) present with renal failure severe enough to require dialysis.47 The main cause of renal failure in myeloma is deposition of Bence Jones protein in the renal tubules. The classic histological features of ‘myeloma kidney’ are of fractured distal tubular casts with a surrounding chronic inflammatory infiltrate including giant cells. The physicochemical properties of individual immunoglobulin light chains result in differing degrees of renal impairment. Less frequently, light-chain deposition may produce a form of glomerulonephritis (light-chain deposition disease). Many other factors may contribute to renal failure and these include hypercalcaemia, infection, dehydration, hyperuricaemia, amyloid deposition and non-steroidalanti-inflammatory drugs. Renal failure is most common in patients with a high tumour burden, which is in itself an adverse prognostic factor in myeloma.46 Renal failure has an adverse effect on prognosis because of increased numbers of early deaths. One study reported a 30 per cent early death rate,48 a figure
roughly three times that expected overall in myeloma.49 So, although impaired renal function does reduce life expectancy to a small extent the outlook is now much more closely related to that of the underlying myeloma.50 Renal failure is reversible in up to 50 per cent of cases51 and should not be a contraindication to active management. The chance of recovery of renal function is influenced by the severity of the renal impairment and whether reversible precipitating factors are present. The median time to renal recovery is 2 months.48
Other clinical features Haematopoiesis is frequently impaired in myeloma. A normochromic, normocytic anaemia is common, with a haemoglobin below 120 g/L seen in 60 per cent of patients. Severe anaemia, neutropenia and thrombocytopenia are, however, rare at presentation. The mechanisms causing anaemia are not fully known but it appears to be mediated by cytokines such as IL-1 rather than being directly due to bone marrow infiltration. In some patients, renal failure may also contribute to the anaemia. There is a complex multifactorial immune deficiency leading to impairment both of humoral and cell-mediated immunity and an increased susceptibility to infection, both bacterial and viral. An important contributory factor is the suppression of normal polyclonal immunoglobulins but other factors include neutropenia, reduced T-lymphocyte function and natural killer cell activity. Chest infections are particularly common. About 10 per cent of patients with myeloma develop AL (primary) amyloidosis. In this disorder, monoclonal immunoglobulin light chains or light chain fragments are linked together and form amyloid fibrils, which can then be deposited in almost all organs. The kidney is frequently affected, with deposition of amyloid in the glomeruli leading to generalized proteinuria and nephrotic syndrome. Peripheral neuropathy (particularly carpal tunnel syndrome), congestive cardiac failure and involvement of skin, muscle and joints may also occur. The diagnosis is usually made by biopsy of an affected organ and staining with Congo Red but if this is not practicable, rectal biopsy (including submucosal tissue) gives a positive result with a sensitivity of about 80 per cent.52 Peripheral neuropathy may also occur in myeloma patients without amyloidosis.53 Very high immunoglobulin levels, particularly of IgA, may result in hyperviscosity syndrome, with headaches, visual disturbance and loss of concentration.
Asymptomatic patients Myeloma is increasingly detected incidentally by the finding of a persistently abnormal erythrocyte sedimentation rate
1078 Multiple myeloma
(ESR), plasma viscosity or serum protein electropheresis on routine screening or when being investigated for an unrelated problem.
INVESTIGATION, DIAGNOSIS AND STAGING Investigation Initial investigation of a patient with suspected myeloma should include the screening tests indicated in Table 44.1, followed by further tests to confirm the diagnosis. The urgency with which these should be carried out depends on the method of presentation. Symptomatic patients with suspected myeloma require urgent specialist referral. Spinal cord compression, hypercalcaemia and renal failure are medical emergencies requiring immediate admission to hospital. Patients with a paraprotein found on routine testing and who have no clinical symptoms and no anaemia, hypercalcaemia or renal impairment do not necessarily require urgent referral, but specialist advice should be sought. Appropriate initial investigations are summarized in Table 44.1 (permission granted from BCSH).
Haematological investigations The most common abnormality seen on a full blood count is a normochromic, normocytic anaemia. The blood film often shows rouleaux formation due to the presence of a paraprotein. Plasma cells are not often seen in the peripheral blood film but may be present in patients with advanced or aggressive disease. When circulating plasma cells are seen at a level of over 2 109/L, a diagnosis of plasma cell leukaemia can be made. The ESR is usually raised in myeloma as the result of the high serum globulin levels and this is typically accompanied by a low albumin. However, the ESR is frequently normal in patients with Bence Jones only myeloma. A bone marrow examination is an essential diagnostic investigation. A marrow aspirate may be sufficient to confirm the diagnosis, but because the plasma infiltrate may be patchy, a trephine biopsy is likely to give a more reliable estimate of plasma cell numbers. Immunohistology or flow cytometry to demonstrate monoclonality is useful where there is only a modest increase in plasma cell numbers. Results of cytogenetic analysis and FISH may provide useful prognostic information. Their role in routine clinical practice is currently being evaluated in clinical trials.
Biochemical investigations Biochemical screening may show abnormalities of urea, creatinine, calcium, uric acid, total protein and/or albumin. Serum ALP is usually normal but is commonly raised after a fracture. Electrophoresis of serum and concentrated urine should be performed to look for the presence of a paraprotein. It is important to test the urine in all patients, since
patients with Bence Jones only myeloma have no paraprotein in the serum and patients with a serum paraprotein are at greater risk of renal failure if they also produce BJP. The paraprotein in the serum and/or urine is then confirmed and typed by immunofixation. The paraprotein is of IgG subtype in 55–60 per cent of cases and IgA in 20–25 per cent, of whom two thirds also have BJP in the urine. Rarely, the paraprotein may be of IgD, IgE or IgM type. Overall two thirds of paraproteins have kappa light chains and one third lambda. Twenty per cent of cases are Bence Jones only myeloma. In less than 1 per cent of patients, no paraprotein is detectable in either serum or urine (non-secretory myeloma). Quantification of the serum paraprotein and 24-hour urinary excretion of BJP should be measured as a baseline against which to monitor response to treatment and future progress of the disease. Levels of non-isotypic immunoglobulins should also be measured to assess immune paresis. Serum B2-microglobulin should be measured in all patients at diagnosis because of its prognostic significance. C-reactive protein (CRP) and lactate dehydrogenase (LDH) may also give useful prognostic information (see later). The serum free light chain assay is a relatively new test that can be used as an alternative to the measurement of urinary free light chains or where this is difficult. This test is particularly useful for diagnosis and monitoring free light chain only myeloma54 and patients in whom the serum and urine is negative on immunofixation (non-secretory myeloma).55
Radiology and imaging A skeletal survey should be performed in all patients. It is important to evaluate areas of bone pain and to document areas of bone destruction which could lead to pathological fracture. It should include all sites of active marrow in the adult, i.e. the entire axial skeleton, humeri and femora.1 Any other symptomatic areas should also be X-rayed. The most common radiographic findings are a combination of osteoporosis and lytic lesions. Pathological fractures may occur and commonly affect vertebral bodies and ribs. Vertebral compression fractures often lead to long-term pain and deformity and may be complicated by cord compression. Soft tissue extension may sometimes be seen on plain X-rays. Osteoporosis without lytic lesions occurs in 5–10 per cent of patients. Sclerotic lesions are very rare, except in the polyneuropathy (P), organomegaly (O), endocrinopathy (E) and skin changes (S) (POEMS) syndrome (see later). The presence of lytic bone lesions is a major diagnostic criterion for diagnosis but it is recognized that plain radiography has limited sensitivity and may fail to detect early bone disease. Computed tomography (CT) scanning is useful for the visualization of soft tissue involvement and for more detailed evaluation of bones in myeloma. It can also be useful in guiding needle biopsy for histological diagnosis and in the planning of radiotherapy and surgery.56 Magnetic resonance imaging (MRI) scanning is also useful for the assessment of soft tissue disease and is the investigation of
Table 44.1 Initial investigations in patients with myeloma Tests to establish diagnosis
Tests to estimate tumour burden and prognosis
Tests to assess myeloma-related organ impairment (ROTI)
Special tests indicated in some patients
Full blood count (FBC), ESR or plasma viscosity
Bone marrow aspirate / trephine biopsy
Bone marrow cytogenetics or fluorescence in situ hybridization (FISH) analysis
FBC (anaemia)
Bone marrow immunohistology or flow cytometry
Serum or plasma electrolytes, urea, creatinine, calcium, albumin and uric acid
Immunofixation of serum and urine
Serum or plasma urea and creatinine
Vitamin B12 and folate assay
Electrophoresis of serum and concentrated urine
Quantification of monoclonal protein in serum and urine
Creatinine clearance (measured or calculated)
Magnetic resonance imaging (MRI), computerized tomography (CT) scan
Quantification of non-isotypic immunoglobulins
Calcium albumin β2 microglobulin
Calcium Albumin Lactate dehydrogenase (LDH) C-related protein (CRP)
X-ray of symptomatic areas
Skeletal survey†
Screening tests
Skeletal survey
Quantification of non-isotypic immunoglobulins Skeletal survey
FBC, full blood count; ESR, erythrocyte sedimentation rate; FISH, fluorescent in situ hybridization. * Where there is macrocytosis (not uncommon in myeloma); †Recommendations for skeletal survey are given in Section 3.1
1080 Multiple myeloma
Table 44.2 Diagnostic criteria for MGUS, asymptomatic multiple myeloma and symptomatic Multiple myeloma (International Myeloma Working Group, 2003) Asymptomatic multiple myeloma
Symptomatic multiple myeloma‡
M-protein in serum 30 g/L Bone marrow clonal plasma cells 10% and low level of plasma cell infiltration in a trephine biopsy (if done)
M-protein in serum 30 g/L and/or Bone marrow clonal plasma cells 10% proven plasmacytoma
M-protein in serum and/or urine† Bone marrow (clonal) plasma cells or biopsy
No myeloma-related organ or tissue impairment (including bone lesions) or symptoms.
No myeloma-related organ or tissue impairment (including bone lesions) or symptoms
Myeloma-related organ or tissue impairment (including bone lesions)
MGUS
No evidence of other B-cell proliferative disorders or light-chain associated amyloidosis or other light-chain, heavy-chain or immunoglobulin-associated tissue damage* * AL amyloid and the IgM paraprotein-related neurological syndromes would be instances of monoclonal gammopathy associated with specific syndromes. † No specific level required for diagnosis. A small percentage of patients have no detectable M-protein in serum or urine but do have myeloma-related organ impairment (ROTI) and increased bone marrow plasma cells (non-secretory myeloma); ‡ Patients without symptoms but with significant myelomarelated organ damage are grouped with symptomatic myeloma because of the need for treatment.
choice for patients with suspected cord compression, where there may be involvement at several vertebral levels. MRI differs from other radiological techniques in that it visualizes the marrow directly and does not visualize the cortical bone. Myelomatous deposits have a decreased signal intensity on T1-weighted images compared to the normal marrow, and an increased signal on T2 images.57 Marked enhancement is seen with gadolinium diethylenetriaminepentaacetic acid (DTPA). MRI is the most sensitive imaging method routinely available giving excellent imaging of the axial skeleton. Its prognostic significance lies in predicting which patients with asymptomatic myeloma are likely to progress in the near future.58 Radionuclide bone scans are less useful than X-rays for detecting sites of disease in myeloma and generally have no place in the routine investigation of patients with myeloma. The role of fluorodeoxyglucose positron emission tomography (PET) scanning is being evaluated. Several preliminary reports suggest that PET scanning can be useful in detecting occult disease in patients with myeloma59 and solitary plasmacytoma.60
Diagnostic criteria and differential diagnosis The diagnosis of myeloma is usually made by the demonstration of a monoclonal protein (M-protein/paraprotein) in the serum or urine and/or lytic lesions on X-ray together with an increased number of plasma cells in the bone marrow. There are many other conditions in which an M-protein may be present and these include: ● ● ●
MGUS AL amyloidosis Solitary plasmacytoma (skeletal or extramedullary)
●
● ●
B-cell non-Hodgkin’s lymphoma (including Waldenstrom’s macroglobulinaemia) Chronic lymphocytic leukaemia Connective tissue disorders
MGUS has a high prevalence in the population (up to 2 per cent in people over 50 years old and 3 per cent in those over 70 years old). The increasing use of routine screening blood tests for a variety of disorders has led to an increase in the number of people shown to have MGUS. It is important to note that most patients presenting with an M-protein in the serum will have MGUS rather than myeloma and that the criteria for distinguishing these conditions are therefore extremely important. An international working group has recently recommended a new classification of MGUS and myeloma based on the level/concentration of serum M-protein, percentage of bone marrow plasma cells and the presence or absence of ROTI.43 The classification defines criteria for MGUS, asymptomatic myeloma and symptomatic myeloma (Table 44.2). The distinction between symptomatic and asymptomatic myeloma depends on the presence or absence of myeloma-related organ or tissue impairment (ROTI) (Table 44.3). Asymptomatic myeloma in this classification is largely equivalent to the categories previously termed equivocal, indolent or smouldering myeloma. However, some patients without clinical symptoms may fall into the ‘symptomatic’ group because of organ impairment and require immediate therapy. Patients with asymptomatic myeloma do not require immediate treatment but do require careful follow-up. It should also be noted that low level Bence Jones proteinuria and a reduced level of polyclonal immunoglobulins may be observed in some patients with MGUS and are not in themselves indications of malignancy.11
Course of the disease 1081
Table 44.3 Myeloma-related organ or tissue impairment (ROTI)* (International Myeloma Working Group, 2003) ●
●
Calcium levels increased: corrected serum calcium 0.25 mmol/L above the upper limit of normal or 2.75 mmol/L Renal insufficiency attributable to myeloma (see section 6.2 – Early management of renal failure) ● Anaemia: haemoglobin 2 g/dL (1.2 mmol/L) below the lower limit of normal or haemoglobin 10 g/dL (100 g/L or 6.2 mmol/L) ● Bone lesions: lytic lesions or osteoporosis with compression fractures: (MRI or CT may clarify) ● Other: symptomatic hyperviscosity, amyloidosis, recurrent bacterial infections (2 episodes in 12 months)
*Where there is uncertainty as to whether or not organ or tissue impairment is attributable to myeloma the percentage bone marrow plasma cells should be 30%.
Healthy individuals with a small M-protein should not be overinvestigated. The extent of diagnostic procedures in asymptomatic patients should take into consideration the age of the patient, presence of other disease and levels of M-protein. The investigation and diagnosis of AL amyloidosis and of solitary plasmacytoma have been reviewed in recent UKMF/BCSH guidelines.52,61
PROGNOSTIC FACTORS AND STAGING The outcome of multiple myeloma is variable with survival times ranging from a few weeks to 20 years. The analysis of factors affecting the likely prognosis is essential to compare outcomes within and between clinical trials. For individual patients the best staging systems can predict survival outcome with a sensitivity and specificity of around 70 per cent. It is possible that in the future, staging systems may determine choice of therapy but this is as yet unproven. Many prognostic factors have been identified. In the past, these have focused mainly on clinical parameters but more recently biological variables, conventional cytogenetics and molecular markers have been shown to give increasingly useful prognostic information. A high level of serum β2-microglobulin is known to be an independent marker of poor prognosis.62 Anaemia and hypercalcaemia are both adverse risk factors as are raised LDH and C-reactive protein and a low albumin level, the latter two factors both reflecting raised IL-6 levels. Atypical plasma cell morphology and high proliferative activity also indicate a bad prognosis.63 In recent years, cytogenetic abnormalities have been found to have strong prognostic significance. However, because the abnormalities found in myeloma are frequent and highly complex, establishing correlations between these abnormalities and clinical outcome has been difficult.38 Deletions/monosomy of chromosome 13, non-hyperdiploidy, loss of 17p13 by FISH and certain specific IgH
Table 44.4 New International Staging System (Greipp et al., 2003) Stage I
II III
Criteria Serum β2 microglobulin 3.5 mg/L (296 nmol/L) and serum albumin 3.5 g/dL (532 μmol/L) Neither I or III* Serum β2 microglobulin 5.5 mg/L (465 nmol/L)
Median survival (months) 62 months
45 months 29 months
* There are two sub-categories: serum β2 microglobulin 3.5 mg/L, but serum albumin 3.5 g/dL, or serum β2 microglobulin 3.5 – 5.5 mg/L irrespective of the serum albumin level.
translocations such as t(4;14), t(14;16), have a strong negative impact on prognosis.42 Gene expression profiling using microarray techniques is likely to allow molecular classification and more accurate prognostic information in the future. At present, however, there is little evidence to support the use of prognostic factors in determining treatment for individual patients. Many prognostic factors have been combined to create a number of staging systems. The best known and most widely used of these has been the Durie/Salmon staging system.64 The advantage of this system is that the parameters are simple and widely available at diagnosis in all patients. However, attempts have been made to improve on this system as it does not include either albumin or serum β2-microglobulin concentrations. When serum β2-microglobulin is combined with other biological markers such as C-reactive protein or low serum level albumin, there is clear correlation with survival.65,66 A working group has recently proposed an International Prognostic Index based on serum levels of β2microglobulin and albumin that separates patients into three prognostic groups irrespective of type of therapy67 and is shown in Table 44.4. Future incorporation of cytogenetic or molecular information into this model may further improve risk stratification for patients.
COURSE OF THE DISEASE There continues to be major challenges in the management of myeloma. Despite many new approaches to treatment and the introduction of a variety of new agents, MM at present remains essentially incurable for the vast majority of patients with a median survival of 3–4 years. The first effective treatment described for myeloma was melphalan and prednisolone in the early 1960s. Various combinations of alkylating agents have since been used for the treatment of myeloma but there is no evidence that they are superior to melphalan and prednisolone.68 Treatment with standard-dose chemotherapy produces a response in approximately two thirds of patients, with a fall in
1082 Multiple myeloma
paraprotein and improvement or resolution of clinical symptoms. However, complete remission (CR) or disappearance of the paraprotein is unusual with conventional-dose chemotherapy. In most patients the paraprotein falls but reaches a plateau after a few months of treatment, the disease is said to be in ‘plateau phase’ and treatment is stopped. One of the most significant advances of the last 20–30 years has been the introduction of high-dose therapy supported by autologous stem cell transplantation. Several trials have demonstrated the superiority of high-dose over standard chemotherapy in prolonging disease control and survival in patients up to the age of 60–65 and possibly beyond this. Whether or not the patient receives initial treatment with standard- or high-dose chemotherapy, eventually the paraprotein will start to rise again and/or new symptoms may develop, indicating relapse. Further treatment at this stage, perhaps with a different drug or drug combination may again produce a response, but the duration of response is usually shorter than that of the initial remission. Eventually the disease becomes refractory to treatment and the patient succumbs to infection, renal failure or other disease complication.
WHEN TO START TREATMENT AND CRITERIA FOR RESPONSE
in the absence of any new myeloma-related organ or tissue damage. Uniform criteria developed by an international working group73 have recently been expanded and updated74 to include stricter definitions for complete remission, incorporation of the serum free light chain assay for patients with non-secretory disease and clarification of previous criteria, particularly with respect to practical details and the definitions of disease progression and relapse.
GENERAL ASPECTS OF CARE Renal failure Some degree of renal impairment is common in myeloma, the incidence varying according to the definition used. Renal failure can be prevented or reversed by prompt and effective correction of contributory factors such as dehydration, hypercalcaemia and infection and by the withdrawal of nonsteroidal anti-inflammatory drugs. Prompt hydration and adequate chemotherapy are essential. Patients with established renal failure may need dialysis. In some patients renal function recovers as the disease responds to treatment, but in other patients long-term dialysis may be needed. Early plasmapheresis may improve the chance of recovery of renal function.75
Chemotherapy is indicated for the treatment of symptomatic myeloma and asymptomatic myeloma with myelomarelated organ damage. Early intervention in other asymptomatic myeloma has shown no benefit in two randomized, controlled trials.69,70 These patients can remain stable for a long period without treatment and should not be treated at this stage. They should be followed carefully, and this should include regular clinical assessment, measurement of serum and urinary paraprotein and periodic bone marrow examinations and skeletal X-rays, since the first sign of progression may be the development of bone disease. Patients who are asymptomatic, but in whom X-rays show evidence of bone disease, should be treated at diagnosis as they are classed as having symptomatic myeloma and are likely to progress within 1 year.71 Two studies have shown that patients with normal X-rays but abnormal marrow appearance on MRI are at higher risk of disease progression than those with normal MRI scans.58,72 If in doubt as to whether to start treatment, it is better to defer therapy until there is clear evidence of disease progression or myeloma-related organ impairment.
Hypercalcaemia
Response criteria
Bone disease and pain control
It is important to have uniform, reproducible response criteria to facilitate comparisons between treatment strategies. Evaluation of response to therapy is based on changes in serum paraprotein levels and/or urinary light chain excretion
Local radiotherapy is the treatment of choice for the relief of pain due to localized painful bone lesions and may also palliate soft tissue disease. Relatively low radiation doses (usually 8 Gy as a single fraction) are usually needed to
Hypercalcaemia occurs in up to 30 per cent of myeloma patients and usually occurs in the presence of active disease, e.g. at presentation or relapse. Prompt recognition and treatment will minimize renal damage. Management requires vigorous rehydration with close monitoring of fluid balance and renal function combined with IV intra-venous bisphosphonate therapy, usually pamidronate or zoledronate. However, renal function must be closely monitored as nephrotoxicity from pamidronate and zoledronic acid has been reported and is most likely to occur if the recommended dose or rate of infusion is exceeded. Randomized trials have also shown that long-term use of a bisphosphonate, either IV or oral, can reduce the rate of progression of myeloma bone disease (see later). IV diuretic therapy (e.g. furosemide) increases renal clearance of calcium and may also be required to maintain fluid balance. Additional therapy with IV corticosteroids is required in patients with persistent hypercalcaemia and intramuscular calcitonin can also be used in refractory patients.
Chemotherapy 1083
relieve pain. This has been shown to be as effective as fractionated radiotherapy for pain relief of skeletal metastases from a range of cancers, including myeloma, in six randomized, controlled trials.76 Significant side effects are uncommon and the possibility of re-irradiation for local recurrence of symptoms remains, if required. Surgical fixation and subsequent radiotherapy may be needed to treat or prevent pathological fractures of long bones. Radiotherapy should be given postoperatively. Analgesia is clearly important but non-steroidal anti-inflammatory drugs should be avoided if possible because of the risk of renal toxicity. Oral analgesia is preferable and side effects, when they occur, should be managed actively. Patients with severe or progressive bone disease often require opiate analgesia. Additional adjuvant drugs that are not primarily analgesics may also be helpful in certain circumstances. Amitriptyline, carbamazepine or gabapentin may help in neuropathic pain. Corticosteroids, particularly dexamethasone, may be used to relieve bone pain in late stage disease. In cases of persistent/refractory pain due to vertebral collapse, the newer techniques of vertebroplasty and kyphoplasty are alternative options for controlling pain. No randomized studies of either technique have been carried out in myeloma but reported case series have shown significant improvements in pain scores.77,78
Anaemia Anaemia is present in two thirds of patients at presentation. It usually improves with response to treatment. Red cell transfusions may be needed to manage symptomatic anaemia but must be given with caution in patients with high paraprotein levels because of the risk of exacerbating hyperviscosity. Several randomized, double-blind placebo-controlled trials of recombinant human erythropoietin (EPO) that included myeloma patients have shown a significant reduction in transfusion requirements, rise in haemoglobin levels and improved quality of life in patients receiving EPO.79–82 Current guidelines recommend that in newly diagnosed patients, EPO should usually not be considered before response to chemotherapy has been assessed.1 EPO is, however, indicated for the treatment of anaemia in patients with myeloma and chronic renal failure.
Infection Patients with myeloma are at high risk of infection for a variety of reasons including immune paresis, neutropenia and reduced mobility because of skeletal disease. In a large study of early mortality after diagnosis of multiple myeloma,49 infection and renal failure were found to be the main causes of early death. Patients should be advised what to do in the event of symptoms of infection and arrangements should be in
place for access to 24-hour specialist team advice. Any infection must be treated promptly and vigorously and admission for IV antibiotic therapy is usually needed for severe systemic infection. Patients with recurrent infections may benefit from immunoglobulin replacement. All patients should receive influenza vaccine annually.
CHEMOTHERAPY There is no evidence that early treatment of patients with asymptomatic myeloma prolongs survival compared with therapy at the time of symptoms but clinical trials are currently underway to investigate whether new agents can delay progression.
Choice of initial chemotherapy The choice of initial therapy is determined mainly by the patient’s age, performance status and renal function. These factors are important in assessing the likelihood of future high-dose therapy and autologous stem cell transplantation (ASCT). For patients who are eligible for ASCT, it is vital to avoid the use of stem-cell damaging drugs, in particular alkylating agents such as melphalan, which can interfere with adequate stem cell mobilization. Patients up to the age of at least 65 years should be considered as candidates for future autografting, even if performance status is initially poor or if they present with acute renal failure. VAD (vincristine/adriamycin/dexamethasone) or a VAD-type regimen has been the standard approach for induction prior to stem cell transplantation but much of its effectiveness is due to dexamethasone. Autografting can be safely performed in older patients with good performance status, but for most patients over 65 years, the choice lies between simple oral chemotherapy and standard combination regimens. Overall there is little difference in long-term outcome between these options and simple oral treatment therefore seems the most appropriate choice for older patients.1 For patients presenting with renal failure, VAD (or an oral equivalent) or dexamethasone alone is commonly used. Melphalan and cyclophosphamide are both difficult to use in patients with renal impairment, since myelosuppression is unpredictable even when doses are reduced. Steroids alone, e.g. pulsed high-dose dexamethasone, are useful initial treatment in patients where there is an urgency to treat but in whom it is difficult to decide immediately on appropriate chemotherapy, e.g. those presenting with pancytopenia or patients with renal failure. Chemotherapy is usually continued until the patient has achieved at least a partial remission and is in ‘plateau phase’, i.e. disease markers have been stable for a period of 3 months, since there is no advantage to prolonging therapy once plateau has been reached.83 In patients who are proceeding to high-dose therapy and transplant as consolidation therapy, there is probably no
1084 Multiple myeloma
advantage to continuing induction chemotherapy beyond the point of maximum response; this is usually no longer than 3–4 months when VAD-type regimens are used. In the last decade steady progress has been made in the treatment of myeloma. Several promising new agents that are non-cytotoxic biological drugs and exert their effect by targeting pathways integral to myeloma cell survival and function have been developed. Many of these have been shown to have proven benefit in the relapse setting and are now in clinical trials to assess their impact as part of induction therapy. These include the immunomodulatory drugs, thalidomide, lenalidomide and bortezomib.
Initial chemotherapy prior to high-dose therapy For patients where high-dose therapy (HDT) is planned, the aim of initial treatment is to achieve rapid cytoreduction without impairing stem cell mobilization. VAD, a combination of vincristine, adriamycin and dexamethasone or similar infusional regimens such as VAMP (where methyl prednisolone replaces dexamethasone) or C-VAMP (addition of cyclophosphamide) have been most widely used. These regimens are associated with a high response rate and a CR rate of 10–25 per cent (based on negative routine electrophoresis).84,85 Although higher CR rates have been claimed for C-VAMP,86 there have been no randomized trials comparing these three regimens. Unfortunately these remissions are not durable, lasting an average of 18 months, even in patients who achieve CR. However, the rate of response is rapid, with most patients achieving 90 per cent of maximal response within 6 weeks, making it ideal initial cytoreductive therapy for patients in whom ASCT is planned. These regimens also have particular advantages in certain clinical situations. As none of the drugs are renally excreted, they can be given without dosage modification in patients in severe renal failure, including those on dialysis. They produce little myelosuppression and so are particularly useful in patients presenting with cytopenias. Their rapid action is also advantageous in patients who require rapid tumour reduction, e.g. those with rapidly progressive bone disease or incipient renal failure.The significant disadvantage of these regimens is the need for a central line, which leads to an appreciable incidence of line-related infection and thrombosis. The introduction of oral idarubicin allowed the development of an oral regimen combining this agent with dexamethasone (Z-Dex), which appears to be similar to VAD in terms of inducing responses and avoids the need for a central line.87 A randomized study demonstrated that Z-Dex is a suitable oral alternative to VAD for treating newly diagnosed myeloma patients but did not provide definitive evidence for equivalence.88 As with VAD, it is likely that much of the efficacy of Z-Dex is due to the dexamethasone, and a Nordic Myeloma Study Group (NMSG) study of single agent idarubicin in relapsed/refractory patients showed little effect.89 It is
therefore recognized that high-dose dexamethasone (HDD) is responsible for much of the effect of the VAD and similar regimens, and dexamethasone alone has also been shown to be an effective agent for inducing remission.90 It is widely used in the USA for initial therapy prior to ASCT. It therefore appears that VAMP, C-VAMP, Z-Dex and HDD give response rates similar to VAD, but in the absence of large randomized trials no firm conclusion can be drawn about equivalence. Combinations of thalidomide and dexamethasone, or thalidomide with other chemotherapy regimens, are currently under evaluation as initial therapy prior to ASCT. Early phase II data indicated that the addition of thalidomide increases toxicity, particularly the risk of venous thromboembolism91,92 but a retrospective analysis also confirmed a higher response rate than VAD93 without any negative impact on ability to collect stem cells. Preliminary data from a phase III randomized comparison of thalidomide and HDT vs. HDT alone has confirmed this higher response rate (80 per cent vs. 53 per cent; P 0.0023). Combinations of other agents with thalidomide have been tested to increased response rates. The combination of oral cyclophosphamide with thalidomide and dexamethasone (CTD) has shown promising results in relapsed and refractory patients94,95 and is now being tested in newly diagnosed patients in the UK MRC Myeloma IX trial. Duration of treatment prior to HDT is generally 4–6 months, which achieves a maximal response in most patients. Although CR prior to HDT is a good prognostic factor, there is no evidence that prolonged induction in the hope of reaching CR will improve outcome.
Initial chemotherapy where HDT is not planned – ‘conventional therapy’ ALKYLATING AGENTS
The aim of therapy in patients not considered candidates for HDT and ASCT on the basis of age, performance status and comorbidity is to achieve a response with minimal treatment-related toxicity. An oral regimen is therefore preferable. Intermittent melphalan (M) and prednis(ol)one (P) has been the standard therapy for MM for many years,96 although there has no clear long-term survival advantage noted for the addition of steroid. M and P are usually given for 4 days every 4–6 weeks, typical doses being melphalan 7–8 mg/m2/day and prednisolone 40–60 mg/day. A randomized study of M and P vs. M and dexamethasone did not show dexamethasone to be superior but was associated with more toxicity.97 Melphalan has also been compared with cyclophosphamide, another orally active alkylating agent, and no differences in outcome were noted. A frequently used schedule is weekly IV cyclophosphamide (150–300 mg/m2) with alternate-day prednisolone. An oral equivalent of the IV cyclophosphamide weekly schedule is cyclophosphamide 200–400 mg/m2 weekly. Weekly cyclophosphamide is
Radiotherapy 1085
100 90 80 Estimated percentage still alive
considerably less myelotoxic than melphalan, but causes more alopecia. This type of alkylating agent therapy will cause a response in 50–60 per cent of patients, usually over a period of 3–6 months, and will reach a stable plateau phase, during which the paraprotein level does not continue to fall but remains stable. Complete remission is rare. It is recommended that treatment should be continued to maximum response plus 3 months1 as giving further chemotherapy does not prolong the duration of the remission and may favour the development of drug resistance. Randomized trials have shown no advantage of continuing chemotherapy beyond this time.83,98 The median duration of remission is around 18 months, with a median survival of around 3 years in most series.
70 60 50 40 30
24.4%
20
23.0%
19.4%
Combination chemotherapy
– allocated CCT (%) 10
A number of combination chemotherapy regimens have been developed with the hope of improving response rate and survival. Vincristine, adriamycin and the nitrosoureas, especially carmustine (BCNU)(B), together with melphalan and cyclophosphamide have been the mainstay of combination chemotherapy schedules. Several regimens have been developed that incorporate some or all of these agents, usually together with prednis(ol)one. The most widely used have been the VBMCP and VMCP/VBAP protocols developed in the USA. In the UK, the ABCM regimen (adriamycin, BCNU, cyclophosphamide and melphalan) has been widely used. There is little difference in long-term outcome between any of these regimens, and little evidence that they are more effective than single-agent therapy. Most of the published studies comparing combination chemotherapy regimens with oral melphalan and prednisolone (MP) have shown an improved response rate but no significant improvement in survival. A recent metaanalysis reviewing the results of 27 published studies concluded that overall there was no difference in survival between combination chemotherapy and MP68 (Fig. 44.1) and the current consensus is that combination chemotherapy has little advantage over simpler regimens. The use of thalidomide in combination with MP is currently being explored. Early data suggested an improved response rate but an increased incidence of side effects, particularly venous thromboembolism.99,100 Important data from an ongoing IFM trial show that oral M and P plus thalidomide (MP-T) resulted in statistically significant improvements in overall survival and progression-free survival when compared with MP and tandem autologous transplantation (using melphalan 100 mg/m2) in newly diagnosed elderly patients.101 However, this was associated with an increase in side effects compared with patients treated with MP, including infections, cardiac events, neuropathy and constipation. Grade 3 or 4 adverse events (including DVT and PE) were reported in 12 per cent of patients receiving MP-T compared with 4 per cent of patients treated with MP.
1.4% 18.0%
– allocated MP (%)
0 0
1
2
3 4 5 Time (years)
6
Deaths/person-years: CCT 642/1999 392/1456 305/1044 196/724 133/506 255/1130 MP 576/1968 407/1423 294/983 194/652 130/444 215/839
Figure 44.1 Overall survival of 6633 patients in 27 randomized trials comparing melphalan plus prednisolone (MP) and combination chemotherapy (CCT). Median survival in both groups was 29 months. (Reproduced with permission from Myeloma Trialists’ Collaborative Group (1998). Combination chemotherapy versus melphalan plus prednisone as treatment for multiple myeloma: an overview of 6633 patients from 27 randomized trials. J Clin Oncol 16:3832–42.)
RADIOTHERAPY Myeloma is a very radiosensitive tumour and the limitation of radiotherapy as a treatment modality is myelosuppression. Local radiotherapy is a very effective means of relieving bone pain, with improvement usually seen within a few days of starting treatment. This is more rapid than can be achieved with any form of chemotherapy. Treatment may be given as a single fraction (usually 8 Gy) or as a fractionated course (usually 15–20 Gy) in 7–10 fractions or 30–35 Gy in 10–15 fractions.102 A dose of 8–10 Gy is usually sufficient to provide pain relief, and above this dose there is no evidence of a dose–response curve in relation to quality and duration of symptomatic control. However, higher doses (30–35 Gy) may be needed for a tumoricidal effect and may be used in the treatment of a solitary plasmacytoma or cord compression.103 In cases of fracture or impending fracture of a long bone, surgical fixation is required but radiation may be given immediately as wound healing is complete. A randomized study of single-fraction vs. fractionated radiotherapy for painful lesions showed no difference in rapidity of onset or duration of pain relief. However,
1086 Multiple myeloma
Table 44.5 Randomised Trials Comparing Conventional Chemotherapy and High Dose Therapy (HDT) Median EFS (months)
CR rate (%) Trial IFM90 MAG91 Pethema MRC7 US Intergroup
Median OS (months)
No. Pts
Age (years)
CC
HDT
CC
HDT
CC
HDT
200 190 164 401 516
65 55-65 Median 56 65 70
5* NE 11* 8* 17
22* NE 30* 44* 15
18* 19* 33 20* NR
28* 24* 42 32* NR
44* 50 64 42* NR
57* 55 72 54* NR
EFS, event-free survival; OS, overall survival; CC, conventional chemotherapy; HDT, high dose therapy; NR, Not reported; * Significant p value
fractionated radiotherapy is preferable for the relief of spinal cord compression (30–35 Gy) because there is less risk of radiation-induced oedema, and dexamethasone should also be given (e.g. 4 mg four times daily) and continued to cover the period of radiation.
Autologous transplantation As conventional chemotherapy has not resulted in longterm remission there has been increasing interest in highdose therapy with stem cell support. It is now evident that autologous transplantation, using peripheral blood progenitor cells (PBPCs) with or without post-transplant growth factors, is an extremely safe procedure for patients up to at least the age of 65 years. However, it is also evident that at present no patients are cured by autologous transplantation. The median duration of event-free survival (EFS) is around 30 months in most series and all patients ultimately relapse, although those who achieve CR posttransplant may have very long survival. Since autologous transplantation is not curative, it is important to establish whether or not it prolongs remission and/or survival as compared with conventional chemotherapy. Five prospective randomized trials have been published comparing standard therapy with a high dose approach as first-line therapy for newly diagnosed Durie and Salmon stage II/III myeloma patients up to age 65 years (Table 44.5). In most of these studies, maintenance interferon (IFN) was given in both arms. Two studies showed a survival advantage for HD therapy. The first of these, comparing autologous transplantation with conventional chemotherapy, was carried out by the InterGroupe Francophone du Myelome (IFM).104 A total of 200 newly diagnosed patients were randomized to receive either two courses of combination chemotherapy (VMCP/VBAP) followed by autograft, or 10 courses of VMCP/VBAP. EFS and overall survival (OS) were significantly longer in the HDT arm. Updated results at a median
High dose
75 Overall survival (%)
HIGH-DOSE THERAPY AND STEM-CELL TRANSPLANTATION
100
50 Conventional dose
25
0 0 Conventional dose High dose
15
30 Months 63 (53–73) 69 (58–78)
45 35 (22–50) 61 (50–71)
60 12 (1–40) 52 (36–67)
Figure 44.2 Overall survival of newly diagnosed patients randomized to combination chemotherapy or autologous transplantation. (Reproduced with permission from Attal et al (1996) Autologous bone marrow transplantation versus conventional chemotherapy in multiple myeloma: a prospective randomized trial. N Engl J Med 335:91–7. Copyright 1996 Massachusetts Medical Society. All rights reserved.
follow-up of 7 years continued to show significantly better EFS (16 per cent vs. 8 per cent; P 0.01) and OS (43 per cent vs. 25 per cent; P 0.03). The MRC VII trial105 comparing ABCM with C-VAMP followed by HDT also showed significantly longer EFS and OS in the HDT arm (Fig. 44.2). Three studies have shown no significant survival advantage. The French MAG91 study106 showed a significant prolongation of EFS but not of OS, possibly because of the
High-dose therapy and stem-cell transplantation 1087
use of HDT at the time of relapse in those initially treated with chemotherapy, while in the Spanish PETHEMA trial107 there was no significant difference in EFS or OS. The USA Intergroup Trial108 compared the outcome of patients receiving VAD and followed by either VBMCP or a melphalan/TBI autotransplant. No significant difference was seen in response rates or survival, and the 7-year EFS and OS estimates of the two arms were 14 per cent and 38 per cent after VBMCP and 17 per cent and 38 per cent after HDT, respectively. As a result of these combined trials and current evidence it is generally agreed that high-dose therapy prolongs remission and survival and should form part of the standard approach to initial treatment in younger patients.1
randomized study (IFM 94) comparing one high-dose procedure with two. Initially, after 3 years, no survival benefit was demonstrated, but after a follow-up of 75 months, a significant difference has been seen in favour of the double transplants. EFS was 25 vs. 30 months (P 0.03) and OS was 48 vs. 58 months, (P 0.01).117 Further analysis has shown that the subgroup of patients who have a 90 per cent response to the first transplant are most likely to benefit from a second transplant. Interim analysis of three other randomized studies from France, Italy and the Netherlands has not shown an overall survival benefit for tandem transplants to date, although this may alter with longer follow-up. At present this approach cannot be recommended as standard therapy, however may be considered in patients who have a 90 per cent response to their initial transplant.
Conditioning Allogeneic stem cell transplants The most widely accepted conditioning regimen for highdose therapy in the treatment of myeloma is melphalan 200 mg/m2. Initially, a combination of melphalan 140 mg/m2 and total body irradiation (TBI) 8 Gy was used but a randomized study from the IFM comparing these regimens109 showed that the melphalan/TBI combination was associated with increased toxicity. Although the EFS was comparable in the two arms, the OS was shorter in the TBI arm. Doses of melphalan above 200 mg/m2 have been explored in small studies,110 but have been associated with increased nonhaematological toxicity, particularly gastrointestinal. A retrospective analysis of 821 patients from the Spanish registry suggested that improved results might be obtained with a combination of busulphan and melphalan (BuM)111 and results from a small randomized study of high-dose melphalan (HDM) vs. BuM112 showed a trend in EFS in favour of BuM, but as yet the data do not clearly indicate that any other regimen is superior to melphalan 200 mg/m2. Patients with renal failure, a common complication of multiple myeloma, may also undergo high-dose therapy and stem cell transplantation but may require dose reduction of the melphalan to 100 or 140 mg/m2.113,114 Older or less fit patients may also benefit from a modified approach as shown in a randomized study115 comparing conventional chemotherapy to intermediate-dose melphalan (IDM), consisting of two courses of melphalan 100 mg/m2 with stem cell support. Response rates and survival were significantly improved with the IDM, especially in patients aged 65–70.
Double autologous transplantation A double (tandem) autograft represents a means of attempting to deliver more effective eradication of endogenous disease. This approach has been pioneered by the Little Rock group.116 They have shown that the results are superior to standard therapy, but have not compared their results with those of single autografts either in a randomized trial or in a case–control study. The French IFM group has carried out a
Until recent years, allogeneic stem cell transplants (SCTs) used fully myeloablative conditioning consisting of TBI at doses of 10–12 Gy combined with high-dose chemotherapy such as cyclophosphamide or melphalan. This procedure, using an HLA-matched sibling donor can result in longterm survival and may have a role in a few selected patients. However, one of the main problems is the high transplantrelated mortality (TRM) despite evidence from both the EBMT and individual centre studies that this has improved over the last 10 years. The 2-year TRM has fallen from 46 per cent before 1994 to 30 per cent since that date.118,119 This may reflect transplantation earlier in the course of the disease, improved supportive care and careful patient selection. Allografts should be done in first remission as several studies show little benefit from allografts in the progressive disease/relapse situation.120,121 Patients transplanted in first remission have a about a 60 per cent chance of entering CR and one third of these patients are in a persistent molecular remission with a very low risk of relapse.122 Overall, there is about a 30 per cent chance of long-term disease-free remission. The potential benefit of this outcome may justify the risks of allogeneic SCT in patients up to 50 years of age, particularly for patients early in their disease. The choice between allogeneic and autologous transplant in those patients who have the option is a difficult one and depends on assessment of the risks and benefits on an individual basis. Matched unrelated donor transplants using full myeloablative conditioning have a significantly higher TRM than matched sibling SCTs and are not recommended.1
Graft vs. myeloma It is now apparent that donor lymphocyte infusions (DLIs) given to patients not in remission after allogeneic SCT can produce a disease response, clearly demonstrating a graftvs.-myeloma (GvM) effect. Patients relapsing after an allogeneic SCT have been shown to respond to DLI with
1088 Multiple myeloma
response rates of over 50 per cent and CR rates of 17 per cent.123 However, response is closely associated with the development of graft-vs.-host disease and fatal marrow aplasia may also occur. These risks may be reduced by using gradually escalated doses of T-cells. There is also evidence that immunomodulation using interferon124 may induce a CR in patients who still have evidence of residual disease 3–6 months post-transplant.
Reduced intensity conditioning allografts The aim of reduced intensity conditioning (RIC) allografts is to induce a GvM effect while reducing the toxicity of conditioning. RIC allografts are clearly associated with a lower toxicity and lower TRM ranging between 10–17 per cent125,126 but long-term outcome is still to be evaluated. This type of conditioning may increase the numbers of patients eligible for allogeneic SCT and patients up to 65 years of age are included in many ongoing studies. Various different conditioning regimens are in use, some containing low-dose irradiation, some using chemotherapy alone127,128 and others including Campath-1H (alemtuzumab).129 No specific recommendation can be made at present regarding these various types of conditioning. DLI is as effective in the treatment of residual or relapsed disease as after conventional allografts. RIC allografting can be used alone or following an autologous transplant, i.e. as part of a tandem procedure.128 As with conventional conditioning, RIC allografts should be done in the earlier phase of the disease and the best results are seen in chemo-responsive patients.127,128 RIC MUD transplantation may have a lower TRM than conventional MUD transplants, with recent studies showing TRM of 20 per cent.130 Response rates of 90 per cent have been reported with an overall survival at 2 years of 53 per cent,127 but these are still only recommended within a clinical trial.1
High-dose therapy: summary and future directions Autologous stem cell transplantation following induction therapy has now become the standard of care for the majority of patients with multiple myeloma. Suitability for the procedure will depend on a combination of age and performance status and currently patients up to at least 70 years of age should be considered. For older patients, intermediate dose melphalan with stem cell support is increasingly a viable and effective option. The use of double autologous transplants is not recommended as routine practice but can be considered if the response to the initial transplant is 90 per cent. A second transplant at relapse is a possible treatment in patients who have had a good response to their initial PBSCT and have sufficient stem cells available. Fully myeloablative allogeneic transplants are now rarely done in
light of the high TRM and the continued risk of relapse, although RIC transplants may have an increasing role. Currently used preparative regimens do not eradicate disease in most patients and future developments within transplantation will aim to achieve cure. These include the possibility of increasing doses of conditioning chemotherapy in the autologous setting with the advent of novel endothelial growth factors such as keratinocyte growth factor (KGF), which may abrogate non-haematological toxicity. Other approaches such as the use of radiolabelled monoclonal antibodies directed against haemopoietic cells are being explored and may deliver high doses of radiotherapy targeted to the myeloma cells without systemic effects. The use of tandem transplant procedures using a combination of autologous followed by RIC allogeneic transplants also continues to be investigated and, combined with immunological modalities such as DLI or vaccines to exploit the graft myeloma effect, may prove effective.
MAINTENANCE THERAPY The aim of maintenance therapy, either post-chemotherapy or post-transplant is to maintain and prolong the response achieved. Ideally, it may also increase depth of response.
Interferon The first drug investigated and used for this purpose was IFN as it in known to have anti-myeloma activity. There have also been numerous trials addressing the issue of whether maintenance therapy with IFN prolongs remission and/or survival. Some trials (but not all) have shown a benefit in terms of remission duration, but a survival benefit has been difficult to demonstrate. A meta-analysis of over 4000 patients in 24 different trials suggested that IFN either during induction or maintenance prolongs remission by about 6 months and survival by about 3 months.131 IFN maintenance also appears to prolong remission after autologous transplantation.132 Overall, whilst it appears that interferon confers a small benefit, when weighed up against the cost and side effects it is no longer recommended as maintenance therapy.
Thalidomide Thalidomide is known to be active against myeloma and, at low doses, is now being considered as maintenance therapy either post chemotherapy or post transplant. Early pilot studies suggest that low dose maintenance (50 mg/day) is safe with few side effects. This is now being trialed as part of the current UK MRC Myeloma IX study. A recent randomized study133 showed thalidomide maintenance after high-dose therapy to significantly increase event-free and overall
Refractory and relapsed disease 1089
survival compared to no maintenance or pamidronate maintenance. Clearly, there is a risk that the known side effects of thalidomide such as peripheral neuropathy and thrombosis, although less frequent at low doses, may become problematic with long-term use. Also the risk of thalidomide-resistant relapse needs to be addressed. Therefore, although promising, the use of thalidomide maintenance should, at this stage, be confined to clinical trials.
Other agents Newer agents such as bortezomib and lenalidomide, may offer further options for maintenance therapy and are currently being investigated in this setting in several clinical trials. Whether they will provide a survival benefit is as yet unknown.
BISPHOSPHONATES Bisphosphonates, which act by inhibiting osteoclast-mediated bone resorption, are the treatment of choice for hypercalcaemia persisting after rehydration. The question of whether long-term use of bisphosphonates can arrest progression of bone disease in patients with MM has been addressed in a number of trials. Both oral clodronate and IV pamidronate were found, in different studies, to reduce bone pain and the incidence of fractures.134–136 Zoledronate has been shown to be as effective as pamidronate in reducing skeletal-related events in patients with bone disease due to MM or breast cancer137 and is more convenient to administer. The benefit in the above studies was not confined to patients with bone disease at the start of treatment and it is currently recommended that all patients who require treatment for their myeloma should receive a bisphosphonate.1 At present there is no evidence to support the use of bisphosphonates in asymptomatic patients. Side effects are few but there is now increasing evidence that prolonged administration of zoledronate may be associated with the development of osteonecrosis of the jaw (ONJ). In a prospective trial of over 200 patients receiving bisphosphonates,138 the incidence of ONJ in patients receiving zoledronate was over 10 per cent. Risk factors included over 36 months’ treatment and previous dental surgery. Patients with dental problems should probably receive an alternate bisphosphonate. For all bisphosphonates, care should be taken in patients with renal dysfunction and dose modification may be required. There are no data comparing the efficacy of clodronate with that of pamidronate or zoledronate; the absorption of bisphosphonates is poor and variable but there are practical advantages of prescribing an oral drug. It has been suggested that the use of bisphosphonates long term may also improve survival,135 but a meta-analysis of published data found no significant survival benefit.139
REFRACTORY AND RELAPSED DISEASE Refractory disease Some patients will exhibit resistance to primary treatment, although most will respond to a thalidomide-containing regimen. For those patients not started on this as initial therapy and who prove refractory to other regimens, such as VAD or MP, CTD or MPT should be considered. New drugs, such as bortezomib and lenalidomide are now being trialed in patients with primary refractory disease and may prove to be a useful option in thalidomide-refractory patients.
Relapsed myeloma All patients with myeloma will relapse, and at this stage the disease may respond to the same chemotherapy as that used previously, to a different regimen, or prove refractory to treatment. Very few clinical trials have been done in relapsed myeloma and the development of treatment pathways has developed over the years largely on the basis of clinical experience. Because myeloma is an incurable disease, treatments must aim to control the disease and preserve quality of life. This is particularly true at relapse. Several different treatments are available; including alkylating agents, thalidomide, steroids and, more recently, newer drugs such as bortezomib and lenalidomide. The optimal choice of drug therapy at relapse is influenced by several factors including initial therapy, severity and subtype of disease, and performance status of the patient. For many patients who achieve a response of 1 year or more, it may be worth trying the same regimen that induced the initial remission. Until recently these comprised mainly MP or VAD. Over 50 per cent of patients who initially responded to MP will respond again to MP at relapse.83 Patients who are refractory to MP may still respond to cyclophosphamide. In younger patients, VAD can be given again at relapse, and if they have had a disease response of over 12 months following high-dose melphalan and PBSCT, this can also be repeated if stem cells are available. A retrospective study140 showed this to significantly improve survival, but no prospective, randomized trials have been carried out to date. Dexamethasone has a long history of being used at relapse141 and is particularly useful in patients with renal failure or pancytopenia. It can either be used alone, or more recently, in combination with other drugs, such as thalidomide. The advent of thalidomide in the late 1990s in the treatment of relapsed myeloma provided a powerful alternative therapy.142 Thalidomide alone produces a response in 30 per cent of relapsed/refractory patients143,144 and when combined with dexamethasone the response rate rises to over 50 per cent.145 Combinations such as cyclophosphamide, thalidomide and dexamethasone (CTD) have been shown to be even more effective with response rates of
1090 Multiple myeloma
up to 80 per cent.146 The use of these types of regimens has become clinically recognized as the mainstay of treatment for relapsed disease. However, now many patients are receiving thalidomide-containing regimes as induction therapy and/or maintenance, there is a question as to whether resistance will develop. This will need to be examined through studies and clinical trials. However, a suggested approach would be that if a patient has had a good initial response to thalidomide of over 12 months, there is often room for re-treatment with both thalidomide and dexamethasone at higher doses or an alternate thalidomide-containing regimen. The increased understanding of the biology of myeloma cells has led to the development of new drugs such as bortezomib, a proteasome inhibitor which inhibits cancer cell growth and increases apoptosis. It is now licensed for use at first relapse and beyond in multiple myeloma. It is one of the few drugs that have been tested through randomized trials at relapse and the Phase III trial compared bortezomib and high-dose dexamethasone. This trial included a total of 669 patients and response rates were 38 per cent with a median time to progression of over 6 months in all patients receiving bortezomib.147 This was significantly better than in the dexamethasone arm. Responses increase to 50 per cent when combined with dexamethasone. Another novel drug recently in development is lenalidomide, which is a member of the family of immunomodulatory drugs that are analogues of thalidomide. Lenalidomide has been shown to inhibit TNF-α, inhibit angiogenesis and have a co-stimulatory effect on T-cells, effects that are shared with thalidomide, but has fewer side effects than the parent compound.148 Two randomized phase III trials have been conducted comparing the combination of lenalidomide/dexamethasone to dexamethasone alone in myeloma patients with relapsed or refractory disease. Results have demonstrated overall response rates in excess of 58 per cent,149 with median time to progression of 11.1 months and 13.3 months, respectively, vs. 4.7 months and 5.1 months for dexamethasone alone, a significant difference. Both bortezomib and, in the future, lenalidomide, may be useful drugs to use at first relapse, either alone or in combination therapies. The choice of which to use, and where thalidomide would then fit, is yet to be fully established. Ultimately, many patients may receive a selection, if not all of these drugs, at some stage in their disease, with different regimens and combinations being used at sequential relapses. This sequential approach should give patients the best chance of longterm disease control with the various different mechanisms of action of these compounds and lack of cross-reactivity.
modes of action such as bortezomib, a proteasome inhibitor, and lenalidomide, an immunomodulatory drug, are now being widely used in clinical trials and moving rapidly into routine clinical practice. These drugs are discussed in more detail earlier in the section on treatment of relapsed disease. New drugs currently in development and entering clinical trials include inhibitors of receptor tyrosine kinases, farnesyl transferase inhibitors, and heat shock protein (HSP90) inhibitors. Immunological approaches such as monoclonal antibodies are also now in development. Treatment with murine monoclonal antibody to IL-6 has produced transient responses, and trials with chimeric human–murine IL-6 antibodies, antibodies to the IL-6 receptor and new IL-6 inhibitors are under way. Another approach is to use monoclonal antibodies directed against plasma cell antigens such as CD138 to deliver either an immunotoxin or a radioactive agent to the tumour cells. There is considerable interest in vaccination strategies using either the idiotypic protein or the idiotypic DNA to induce immune responses. Such strategies would be most applicable to patients in remission after highdose therapy with a low level of residual tumour.
OTHER RELATED CONDITIONS Monoclonal gammopathy of undetermined significance Although the overall incidence of monoclonal gammopathy of undetermined significance (MGUS) progression to malignancy is 1 per cent per year, it is now possible to more accurately predict the risk of progression based on a new risk-stratification model.151 Despite this, patients with MGUS need indefinite follow-up and should be monitored regularly. Paraprotein levels should be measured every 3 months initially and then in 6 months and if stable, annually thereafter. Patients who meet the criteria for low-risk MGUS (serum M protein 15 g/L; IgG subtype and normal free-light chain ratio can be followed less often).152 Patients should always be instructed to contact their specialist if there is any change in their clinical condition. When patients are reviewed, clinical and laboratory features should be re-evaluated to determine if there is any evidence of progression to myeloma or other related disorder. Bone marrow aspiration and radiological examination is indicated in any patient who develops bone pain, unexplained anemia, hypercalcaemia, or renal failure. Treatment is not indicated in the absence of any features of progression and chemopreventative intervention is not used.
NEW APPROACHES TO TREATMENT The increased understanding of the biology of myeloma and the interaction between myeloma cells and the microenvironment has led to the recent development of a wide variety of new drugs aimed at targeting these interactions or specific intracellular signalling pathways.150 Drugs with novel
Solitary plasmacytoma of bone Solitary plasmacytoma of bone (SPB) may involve any bone but is most common in the axial skeleton. Pain and nerve root or cord compression are the most common
Other related conditions 1091
modes of presentation. Biopsy of the lesion shows a monoclonal population of plasma cells. A serum and/or urinary paraprotein has been reported in approximately 60 per cent of patients but is usually at a low level. The bone marrow and radiological skeletal survey are normal and specific imaging of the spine is required and is best achieved by MRI, possibly combined with CT.61 About 30 per cent of patients presenting with SBP will be found on MRI to have disease at other sites and should be considered as multiple myeloma and treated accordingly. Recommended treatment for SBP is radical local radiotherapy (40–50 Gy in 20–25 fractions) and this achieves initial disease control in the majority of patients.153 Patients not responding to radiotherapy should be treated with chemotherapy as used in myeloma. However, patients should continue to be monitored carefully, because the risk of recurrence at a single site is around 25 per cent, and that of developing MM is over 50 per cent with a median time to progression of 2–4 years.154 Where a paraprotein is present in SPB, persistence for more than 1 year after radiotherapy, or reappearance an after initial fall, almost always indicates disease elsewhere.155 It is therefore evident that the disease is not truly localized at diagnosis in many patients but, in the past, this situation has not been found at staging. This should change as MRI becomes an established part of the staging criteria and patients with early multiple myeloma are identified and treated as such.
Treatment is difficult, but local radiotherapy for localized plasmacytomas and chemotherapy for generalized disease is recommended. There are now a number of reports of neurological improvement or stabilization following high-dose therapy and stem cell transplantation.156 This appears to be an effective therapy for POEMS syndrome but may also be associated with significant morbidity.
KEY LEARNING POINTS ● ●
●
●
●
●
Extramedullary plasmacytoma In contrast to SPB, extramedullary plasmacytoma (EMP) appears to be truly localized in the majority of cases and rarely progresses to MM.153 It most commonly arises in the upper respiratory passages, but a wide variety of different organs may occasionally be involved. As with SPB, biopsy of the lesion shows a monoclonal population of plasma cells, and a serum and/or urinary paraprotein may be present, but the bone marrow and radiological skeletal survey are normal. Local radiotherapy (40–50 Gy over 3–5 weeks), rather than surgical removal, is the treatment of choice. The paraprotein should be monitored following completion of radiotherapy and will normally disappear within 6 months.
●
●
●
●
POEMS syndrome ●
This is a rare syndrome in which a serum paraprotein (M-component) is associated with polyneuropathy (P), organomegaly (O), endocrinopathy (E) and skin changes (S). Approximately 50 per cent of cases are associated with multiple myeloma and the remainder with solitary plasmacytomas or more subtle plasma cell dyscrasias. The cardinal feature is a severe sensorimotor neuropathy with osteosclerotic bone lesions, hepatomegaly and lymphadenopathy, hormonal abnormalities and skin hyperpigmentation. The mechanism for these multisystem changes is unknown.
●
●
When evaluating a pregnant woman with Myeloma is due to proliferation of a single clone of plasma cells, which originate from a postgerminal-centre B lymphocyte. Growth of myeloma in the bone marrow and consequent bone destruction is dependent on interactions between myeloma plasma cells and marrow stromal cells, including osteoclasts. The classical presenting features are bone pain, anaemia and renal impairment. A serum paraprotein is present in 80 per cent of cases, but in 20 per cent free light chains only are produced and there is no paraprotein in the serum. In 1 per cent of cases there is no paraprotein in serum or in urine. The most important prognostic factors are serum b2-microglobulin and the presence of deletion of chromosome 13q. Other factors are serum albumin, serum calcium, haemoglobin and renal function. The combination of melphalan and prednisolone results in a 50 per cent response rate, a median response duration of 18 months and a median survival of 3–4 years. Combination chemotherapy regimens have not proven more effective. All patients with symptomatic myeloma should receive a bisphosphonate. Autologous transplantation results in a higher frequency of response, a longer duration of response and a median survival of around 5 years. Patients who are potential candidates for autologous transplantation should receive induction chemotherapy that does not damage normal marrow stem cells. Allogeneic transplantation has a high mortality rate but may be curative. This effect is mediated at least partly by a GVM effect. The role of maintenance therapy is still being tested in clinical trials but low-dose thalidomide has been shown to prolong event-free survival. The choice of treatment of relapsed disease depends on initial therapy. Thalidomide or newer agents such as bortezomib have been demonstrated to be effective.
1092 Multiple myeloma
KEY REFERENCES Attal M, Harousseau J. Randomized trial experience of the InterGroupe Francophone du Myelome. Semin Hematol 2001; 38:226–230. Dalton WS, Bergsagel PL, Kuehl WM, Anderson KC, Harousseau JL. Multiple myeloma. Hematology (Am Soc Hematol Educ Program) 2001; Jan157–77. (available on line) Fonseca R, Barlogie B, Bataille R, et al. Genetics and cytogenetics of multiple myeloma: a workshop report. Cancer Research 2004; 64: 1546–58. Hideshima T, Bergsagel PL, Kuehl WM, Anderson KC. Advances in biology of multiple myeloma: clinical applications. Blood 2004; 104(3):607–18. Mitsiades CS, Mitsiades NS, Munshi NC, Richardson PG, Anderson KC. The role of the bone microenvironment in the pathophysiology and therapeutic management of multiple myeloma: interplay of growth factors, their receptors and stromal interactions. Eur J Cancer 2006; 42(11):1564–73. Smith A, Wisloff F, Samson D; UK Myeloma Forum; Nordic Myeloma Study Group; British Committee for Standards in Haematology. Guidelines on the diagnosis and management of multiple myeloma. Br J Haematol 2005; 132(4):410–51. Terpos E, Dimopoulos MA. Myeloma bone disease: pathophysiology and management. Ann Oncol 2005; 16(8):1223–31.
REFERENCES 1 Smith A, Wisloff F, Samson D; UK Myeloma Forum; Nordic Myeloma Study Group; British Committee for Standards in Haematology. Guidelines on the diagnosis and management of multiple myeloma Br J Haematol 2005; 132(4):410–51. 2 Office of National Statistics. Cancer Trends in England and Wales 1950–1999. http://www.statistics.gov.uk 2001. 3 Phekoo KJ, Schey SA, Richards MA, Bevan DH, Bell S, Gillett D, Moller H on behalf of Consultant Haematologists, South Thames Haematology Specialist Committee. A population study to define the incidence and survival of multiple myeloma in a National Health Service Region in UK. Br J Haematol 2004; 127:299–304. 4 Bray I, Brennan P, Boffetta P. Current trends and future projections of lymphoid neoplasms – an age-period-cohort analysis. Cancer Causes Control 2001; 12:813–20. 5 Altieri A, Chen B, Bermejo JL, Castro F, Hemminki K. Familial risks and temporal incidence trends of multiple myeloma. Eur J Cancer 2006; 42(11):1661–70. 6 Herrinton LJ, Weiss, NS, Olshan AF. Epidemiology of myeloma. In: Malpas JS, Bergsagel DE, Kyle RA, Anderson KC. (eds) Myeloma: Biology and Management. Oxford: Oxford University Press, 1998, 150–86. 7 Gluzman D, Imamura N, Sklyarenko L, Nadgornaya V, Zavelevich M, Machilo V. Patterns of hematological malignancies in Chernobyl clean-up workers (1996–2005). Exp Oncol 2006; 28(1):60–3. 8 Boice JD, Morin MM, Glass AG, et al. Diagnostic X-ray procedures and the risk of leukaemia, lymphoma and multiple myeloma. JAMA 1991; 265:1290–4.
9 Hatcher JL, Baris D, Olshan AF, et al. Diagnostic radiation and the risk of multiple myeloma ( United States) Cancer Causes Control 2001; 12:755–761. 10 Speer SA, Semenza JC, Kurosaki T, Anton-Culver H Risk factors for acute myeloid leukaemia and multiple myeloma: a combination of GIS and case-control studies. J Environ Health 2002; 64:9–16. 11 Kyle PA, Therneau TM, Rajkumar SV, et al. A long-term study of prognosis in monoclonal gammopathy of undetermined significance. N Engl J Med 2002; 346:564–569. 12 Grosbois B, Jego P, Attal M, et al. Familial multiple myeloma: report of fifteen families. Br J Haematol 1999; 105:768–70. 13 Lynch HT, Sanger WG, Pirruccello S, Quinn-Laquer B, Weisenburger DD. Familial multiple myeloma: a family study and review of the literature. J Natl Cancer Inst 2001; 93:1479–83. 14 Fonseca R, Bailey RJ, Ahmann GJ, et al. Genomic abnormalities in monoclonal gammopathy of undetermined significance. Blood 2002; 100(4):1417–24. 15 Tonegawa S. Somatic generation of antibody diversity. Nature 1983; 302:575–581. 16 Bakkus MH, Heirman C, van Riet I, van Camp B. Evidence that the clonogenic cell in multiple myeloma originates from a pre-switched but somatically mutated B cell. Blood 1994; 80:2326–3235. 17 Vescio RA, Cao J, Hong CH, et al. Myeloma Ig heavy chain V region sequences reveal prior antigenic selection and marked somatic mutation but no intraclonal diversity. J Immunol 1995; 155:2487–97. 18 Billadeau D, Quam L, Thomas W, et al. Detection and quantitation of malignant cells in the peripheral blood of multiple myeloma patients. Blood 1992; 80:1818–4. 19 Van Riet I, Vanderkerken K, de Greef C, Van Camp B Homing behaviour of the malignant cell clone in multiple myeloma. Med Oncol 1998; 15:154–64. 20 Hideshima T, Bergsagel PL, Kuehl WM, Anderson KC. Advances in biology of multiple myeloma: clinical applications. Blood 2004; 104(3):607–12. 21 Mitsiades CS, Mitsiades NS, Munshi NC, Richardson PG, Anderson KC. The role of the bone microenvironment in the pathophysiology and therapeutic management of multiple myeloma: interplay of growth factors, their receptors and stromal interactions. Eur J Cancer 2006; 42(11):1564–73. 22 Klein B, Zhang XG, Lu ZY, Bataille R. Interleukin-6 in human multiple myeloma. Blood 1995; 85:863–72. 23 Uchiyama H, Barut BA, Mohrbacher AF, Chauhan D, Anderson KC. Adhesion of human myeloma-derived cell lines to bone marrow stromal cells stimulates interleukin-6 secretion. Blood 1993; 82(12):3712–20. 24 Bataille R, Jourdan M, Zhang XG, Klein B. Serum levels of interleukin 6, a potent myeloma cell growth factor, as a reflection of disease severity in plasma cell dyscrasia. J Clin Invest 1989; 84:2008–11. 25 Bellamy WT. Expression of vascular endothelial growth factor and its receptors in multiple myeloma and other hematopoietic malignancies. Semin Oncol 2001; 28:551–559. 26 Gupta D, Treon SP, Shima Y, et al. Adherence of multiple myeloma cells to bone marrow stromal cells upregulates
References 1093
27
28 29
30
31 32
33
34
35
36
37
38
39
40
41
42
43
vascular endothelial growth factor secretion: therapeutic applications. Leukemia 2001; 15:1950–61. Terpos E, Dimopoulos MA. Myeloma bone disease: pathophysiology and management. Ann Oncol 2005; 16(8):1223–31. Boyle WJ, Simonet WS, Lacey DL Osteoclast differentiation and activation. Nature 2003; 423:337–342. Roux S, Meignin V, Quillard J (2002) RANK (receptor activator of nuclear factor-kappa B) and RANKL expression in multiple myeloma. Br J Haematol 117:86–92. Giuliani N, Bataille R, Mancini C, Lazzaretti M, Barille S. Myeloma cells induce imbalance in the osteoprotegerin/osteoprotegerin ligand system in the human bone marrow environment. Blood 2001; 98:3527–33. Callander NS, Roodman GD. Myeloma bone disease. Semin Hematol 2001; 38(3):276–85. Croucher PI, Shipman CM, Lippitt J, et al. Osteoprotegerin inhibits the development of osteolytic bone disease in multiple myeloma. Blood 2001; 98:3534–40. Kong YY, Yoshida H, Sarosi I, et al. OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 1999; 397:315–23. Bucay N, Sarosi I, Dunstan CR, et al. Osteoprotegerindeficient mice develop early onset osteoporosis and arterial calcification. Genes Dev 1998; 12(9):1260–8. Vanderkerken K, De Leenheer E, Shipman C, et al. Recombinant osteoprotegerin decreases tumor burden and increases survival in a murine model of multiple myeloma. Cancer Res 2003; 63(2):287–9. McClung MR, Lewiecki EM, Cohen SB, et al; AMG 162 Bone Loss Study Group. Denosumab in postmenopausal women with low bone mineral density. N Engl J Med 2006; 354:821–31. Zandecki M, Lai JL, Facon T. Multiple myeloma: almost all patients are cytogenetically abnormal. Br J Haematol 1996; 94(2):217–27. Fonseca R, Barlogie B, Bataille R, et al. Genetics and cytogenetics of multiple myeloma: a workshop report. Cancer Res 2004; 64:1546–58. Tricot G, Barlogie B, Jagganath S, et al. Poor prognosis in multiple myeloma is associated only with partial or complete deletion of chromosome 13 or abnormalities involving 11q and not with other karyotypic abnormalities. Blood 1995; 86:4250–6. Fonseca R, Harrington D, Oken MM, et al. Biological and prognostic significance of interphase fluorescence in situ hybridization detection of chromosome 13 abnormalities (delta13) in multiple myeloma: an Eastern Cooperative Oncology Group study. Cancer Res 2002; 62:715–20. Avet-Loiseau H, Li JY, Morineau N, et al. Monosomy 13 is associated with the transition of monoclonal gammopathy of undetermined significance to multiple myeloma. Blood 1999; 94:2583–9. Stewart AK, Fonseca R. (2005) Prognostic and therapeutic significance of myeloma genetics and gene expression profiling. J Clin Oncol 2005; 23(26):6339–44. International Myeloma Working Group. Criteria for the classification of monoclonal gammopathies, multiple
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
myeloma and related disorders: a report of the International Myeloma Working Group. Br J Haematol 2003; 121:749–57. Kyle RA, Gertz MA, Witzig TE, et al. Review of 1027 patients with newly diagnosed multiple myeloma. Mayo Clin Proc 2003; 78(1):21–33. Knudsen LM, Hippe E, Hjorth M, Holmberg E, Westin J. Renal function in newly diagnosed multiple myeloma-a demographic study of 1353 patients. The Nordic Myeloma Study Group. Eur J Haematol 1994; 53:207–12. Alexanian R, Barlogie B, Dixon D. Renal failure in multiple myeloma. Pathogenesis and prognostic implications. Arch Intern Med 1990; 150:1693–95. Clark AD, Shetty A, Soutar R. Renal failure and multiple myeloma: pathogenesis and treatment of renal failure and management of underlying myeloma. Blood Rev 1999; 13:79–90. Blade J, Fernandez-Llama P, Bosch F, et al. Renal failure in multiple myeloma: presenting features and predictors of outcome in 94 patients from a single institution. Arch Intern Med 1998; 158(17):1889–93. Augustson BM, Begum G, Dunn JA, et al. Early mortality after diagnosis of multiple myeloma: analysis of patients entered onto the United Kingdom Medical Research Council trials between 1980 and 2002 – Medical Research Council Adult Leukaemia Working Party. J Clin Oncol 2005; 23:9219–26. Iggo N. Management of renal complications. In: Malpas JS, Bergsagel DE, Kyle RA, Anderson KC (eds) Myeloma: Biology and Management. Oxford: Oxford University Press, 1998, 381–401. Knudsen LM, Hjorth M, Hippe E. Renal failure in multiple myeloma: reversibility and impact on the prognosis. Nordic Myeloma Study Group. Eur J Haematol 2000; 65:175–81. UK Myeloma Forum; British Committee for Standards in Haematology, British Society for Haematology. Guidelines on the diagnosis and management of AL amyloidosis. Br J Haematol 2004; 125:681–700. Gawler J. Neurological manifestations of myeloma and their management. In: Malpas JS, Bergsagel DE, Kyle RA, Anderson KC. (eds) Myeloma: biology and management. Oxford: Oxford University Press, 1998, 402–38. Bradwell AR, Carr-Smith HD, Mead GP, Harvey TC, Drayson MT. Serum test for assessing patients with Bence Jones myeloma. Lancet 2003; 361:489–91. Drayson M, Tang LX, Drew R, Mead GP, Carr-Smith H, Bradwell AR. Serum free light chain measurements for identifying and monitoring patients with nonsecretory multiple myeloma. Blood 2001; 97:2900–2. Walker MP, Yaszemski MJ, Kim CW, Talac R, Bradford LC. Metastatic disease of the spine: evaluation and treatment. Clin Orthopaed Related Res 2003; 415:S165–175. Healy JC, Armstrong P. Radiological features of multiple myeloma. In: Malpas JS, Bergsagel DE, Kyle RA, Anderson KC. (eds) Myeloma: Biology and Management. Oxford: Oxford University Press, 1998, 222–250. Mariette X, Zagdanski AM, Guermazi A, et al. Prognostic value of vertebral lesions detected by magnetic resonance imaging in patients with stage I multiple myeloma. Br J Haematol 1999; 104:723–9.
1094 Multiple myeloma
59 Orchard K, Barrington S, Buscombe J, Hilson A, Prentice HG, Mehta A. Fluoro-deoxy glucose positron emission tomography imaging for the detection of occult disease in multiple myeloma. Br J Haematol 2002; 117:133–137. 60 Schirrmeister H, Buck AK, Bergmann L, Reske SN, Bommer M. Positron emission tomography (PET) for staging of solitary plasmacytoma. Cancer Biotherapeutics and Radiopharmacology 2003; 18:841–5. 61 Soutar R, Lucraft H, Jackson G, Reece A, Bird J, Low E, Samson D. Guidelines Working Group of the UK Myeloma Forum; British Committee for Standards in Haematology; British Society for Haematology. Guidelines on the diagnosis and management of solitary plasmacytoma of bone and solitary extramedullary plasmacytoma. Br J Haematol 2004; 124:717–26. 62 Norfolk D, Child JA, Cooper EH, Kerruish S, Ward AM. Serum beta 2-microglobulin in myelomatosis: potential value in stratification and monitoring. Br J Cancer 1980; 42(4):510–5. 63 Greipp PR, Lust JA, O’Fallon WM, Katzmann JA, Witzig TE, Kyle RA. Plasma cell labeling index and beta 2-microglobulin predict survival independent of thymidine kinase and Creactive protein in multiple myeloma. Blood 1993; 81(12):3382–7. 64 Durie BGM, Salmon SE. A clinical staging system for multiple myeloma. Cancer 1975; 36:842–54. 65 Bataille R, Boccadoro M, Klein B, Durie B, Pileri A. C-reactive protein and beta-2 microglobulin produce a simple and powerful myeloma staging system. Blood 1992; 80:733–7. 66 Jacobson JL, Hussein MA, Barlogie B, Durie BG, Crowley JJ. A new staging system for multiple myeloma patients based on the Southwest Oncology Group (SWOG) experience. Br J Haematol 2003; 122:441–50. 67 Greipp PR, San Miguel J, Fonseca R, et al. Development of an international prognostic index (IPI) for myeloma: report of the International Myeloma Working Group. JCO 2003; 2006. 68 Myeloma Triallists’ Collaborative Group. Combination chemotherapy versus melphalan plus prednisone as treatment for multiple myeloma: an overview of 6,633 patients from 27 randomized trials. J Clin Oncol 1998; 16:3832–42. 69 Hjorth M, Hellquist L, Holmberg E, Magnusson B, Rodjer S, Westin J. Initial versus deferred melphalan-prednisone therapy for asymptomatic multiple myeloma stage I – a randomized study. Myeloma Group of Western Sweden. Eur J Haematol 1993; 50:95–102. 70 Riccardi A, Mora O, Tinelli C, et al. Long-term survival of stage I multiple myeloma given chemotherapy just after diagnosis or at progression of the disease: a multicentre randomized study. Cooperative Group of Study and Treatment of Multiple Myeloma. Br J Cancer 2000; 82:1254–60. 71 Dimopoulos MA, Moulopoulos LA, Smith T, et al. Risk of disease progression in asymptomatic multiple myeloma. Am J Med 1993; 94:57–61. 72 Weber DM, Dimopoulos MA, Moulopoulos LA, Delasalle KB, Smith T, Alexanian R. Prognostic features of asymptomatic multiple myeloma. Br J Haematol 1997; 97:810–4. 73 Blade J, Samson D, Reece D, et al. Criteria for evaluating disease response and progression in patients with multiple
74
75
76
77
78
79
80
81
82
83
84
85
86
myeloma treated by high-dose therapy and haemopoietic stem cell transplantation. Myeloma Subcommittee of the EBMT. European Group for Blood and Marrow Transplant. Br J Haematol 1998; 102(5):1115–23. Durie BG, Harousseau JL, Miguel JS, et al. International uniform response criteria for multiple myeloma. Leukemia 2006; 20(12):2220. Johnson WJ, Kyle RA, Pineda AA, O’Brien PC, Holley KE. (1990) Treatment of renal failure associated with multiple myeloma. Plasmapheresis, hemodialysis and chemotherapy. Arch Int Med 1990; 150:863–9. Hoskin PJ, Yarnold JR, Roos DR, Bentzen S. Second Workshop on Palliative Radiotherapy and Symptom Control. Radiotherapy for Bone Metastases.Clin Oncol 2001; 13:88–90. Diamond TH, Hartwell T, Clarke W, Manoharan A Percutaneous vertebroplasty for acute vertebral body fracture and deformity in multiple myeloma: a short report. Br J Haematol 2004; 124(4):485–7. Fourney DR, Schomer DF, Nader R, et al. (2003) Percutaneous vertebroplasty and kyphoplasty for painful vertebral body fractures in cancer patients. J Neurosurg Spine 2003; 98:21–30. Dammacco F, Castoldi G, Rodjer S (2001) Efficay of epoietin alfa in the treatment of anaemia of multiple myeloma. Br J Haematol 113:172–17. Littlewood TJ, Bajetta E, Nortier JW, Vercammen E, Rapoport B, and Epoetin Alfa Study Group (2001) Effects of epoetin alfa on hematologic parameters and quality of life in cancer patients receiving nonplatinum chemotherapy: results of a randomized, double-blind, placebo-controlled trial. J Clin Oncol 2001; 19:2865–74. Osterborg A, Brandberg Y, Molostova V, Iosava G, Abdulkadyrov K, Hedenus M, Messinger D; Epoetin Beta Hematology Study Group. Randomized, double-blind, placebo-controlled trial of recombinant human erythropoietin, epoetin beta, in hematologic malignancies. J Clin Oncol 2002; 20:2486–94. Hedenus M, Adriansson M, San Miguel J, et al. Efficacy and safety of darbepoetin alfa in anaemic patients with lymphoproliferative malignancies: a randomized, doubleblind, placebo-controlled study. Br J Haematol 2003; 122:394–403. Belch A, Shelley W, Bergsagel D, et al. A randomized trial of maintenance versus no maintenance melphalan and prednisone in responding multiple myeloma patients. Br J Cancer 1988; 57:94–9. Samson D, Gaminara E, Newland A, et al. Infusion of vincristine and doxorubicin with oral dexamethasone as first-line therapy for multiple myeloma. Lancet 1989; 2:82–5. Alexanian R, Barlogie B, Tucker S. VAD-based regimens as primary treatment for multiple myeloma. Am J Hematol 1990b; 33:86–89. Raje N, Powles R, Kulkarni S, et al. A comparison of vincristine and doxorubicin infusional chemotherapy with methylprednisolone (VAMP) with the addition of weekly cyclophosphamide (C-VAMP) as induction treatment followed
References 1095
87
88
89
90
91
92
93
94
95
96 97
98
99
100
by autografting in previously untreated myeloma. Br J Haematol 1997; 97:153–160. Cook G, Sharp RA, Tansey P, Franklin IM. (1996) A phase I/II trial of Z-Dex (oral idarubicin and dexamethasone), an oral equivalent of VAD, as initial therapy at diagnosis or progression in multiple myeloma. Br J Haematol 1996; 93:931–4. Cook G, Clark RE, Morris TC, et al. A randomized study (WOS MM1) comparing the oral regime Z-Dex (idarubicin and dexamethasone) with vincristine, adriamycin and dexamethasone as induction therapy for newly diagnosed patients with multiple myeloma. Br J Haematol 2004; 126:792–8. Rodjer S, Nilsson B, Westin J; Nordic Myeloma Study Group. Do anthracyclines have a role in the therapy of multiple myeloma? Hematol J 2000; 1:422–6. Alexanian R, Dimopoulos MA, Delasalle K, Barlogie B. Primary dexamethasone treatment of multiple myeloma. Blood 1992; 80:887–90. Cavo M, Zamagni E, Tosi P, et al. First-line therapy with thalidomide and dexamethasone in preparation for autologous stem cell transplantation for multiple myeloma. Haematologica 2004; 89:26–31. Zangari M, Anaissie E, Barlogie B, et al (2001) Increased risk of deep-vein thrombosis in patients with multiple myeloma receiving thalidomide and chemotherapy. Blood 2001; 98:1614–1615. Cavo M, Zamagni E, Tosi P, et al. Superiority of thalidomide and dexamethasone over vincristinedoxorubicin-dexamethasone (VAD) as primary therapy in preparation for autologous transplantation for multiple myeloma. Blood 2005; 106(1):35–9. Dimopoulos MA, Hamilos G, Zomas A, et al. Pulsed cyclophosphamide, thalidomide and dexamethasone: an oral regimen for previously treated patients with multiple myeloma. Hematol J 2004; 5:112–7. Kyriakou C, Thomson K, D’Sa S, Flory A, Hanslip J, Goldstone AH, Yong KL. Low-dose thalidomide in combination with oral weekly cyclophosphamide and pulsed dexamethasone is a well tolerated and effective regimen in patients with relapsed and refractory multiple myeloma. Br J Haematol 2005; 129(6):763–70. Alexanian R, Dimopoulos M. The treatment of multiple myeloma. N Engl J Med 1994; 330:484–9. Hernandez JM, Garcia-Sanz R, Golvano E, et al. Randomized comparison of dexamethasone combined with melphalan versus melphalan with prednisone in the treatment of elderly patients with multiple myeloma. Br J Haematol 2004; 127:159–64. Medical Research Council’s Working Party on Leukaemia in Adults (1980) Report on the second myelomatosis trial after five years of follow-up. Br J Cancer 42:813–22. Oakervee H, Baugh H, Boots M, et al. A study of the safety and efficacy of oral melphalan, prednisolone and thalidomide (MPT) in the treatment of multiple myeloma: a UK Myeloma Forum Pilot Study. Br J Haematol 2004b; 117(Suppl. 1):66. Palumbo A, Bertola A, Musto P, et al. A prospective randomized trial of oral melphalan, prednisone, thalidomide
101 102
103 104
105
106
107
108
109
110
111
112
113
114
(MPT) vs oral melphalan, prednisone (MP): An interim analysis. Blood 2004b; 104:63a. ASCO abstract 06 MP vs MPT. Rowell NP, Tobias JS. The role of radiotherapy in the management of multiple myeloma. Blood Rev 1991; 5:801–4. Norin T. Roentgen treatment of myeloma with special consideration to the dosage. Acta Radiol 1957; 47:46–54. Attal M., Harousseau JL, Stoppa AM, et al. (1996) Autologous bone marrow transplantation versus conventional chemotherapy in multiple myeloma: a prospective, randomized trial. N Engl J Med 1996; 335:91–7. Child JA, Morgan GJ, Davies FE, et al. High-dose chemotherapy with hematopoietic stem-cell rescue for multiple myeloma. N Engl J Med 2003; 348:1875–83. Fermand JP, Ravaud P, Katsahian S, et al. High dose therapy and autologous blood stem cell transplantation versus conventional treatment in multiple myeloma: results of a randomized trial in 190 patients 55 to 65 years of age. Blood 1999; 94(Suppl 1): 396a (abstract). Blade J, San Miguel JF, Fontanillas M, et al. (2001) Increased conventional chemotherapy does not improve survival in multiple myeloma: long-term results of two PETHEMA trials including 914 patients. Hematol J 2001; 2:272–8. Barlogie B, Kyle RA, Anderson KC, et al. Standard chemotherapy compared with high-dose chemoradiotherapy for multiple myeloma: final results of phase III US Intergroup Trial S9321. J Clin Oncol 2006; 24(6):929–36. Moreau P, Facon T, et al (2002) Comparison of 200 mg/m2 melphalan and 8 Gy total body irradiation plus 140 mg/m2 melphalan as conditioning regimens for peripheral blood stem cell transplantation in patients with newly diagnosed multiple myeloma: final analysis of the InterGroupe Francophone du Myelome 9502 randomized trial. Blood 99:731–5. Moreau P, Milpied N, Mahe B, et al. Melphalan 220 mg/m2 followed by peripheral blood stem cell transplantation in 27 patients with advanced multiple myeloma. Bone Marrow Transplant 1999; 23:1003–6. Lahuerta JJ, Grande C, Blade J, et al. Spanish Multiple Myeloma Group. Myeloablative treatments for multiple myeloma: update of a comparative study of different regimens used in patients from the Spanish registry for transplantation in multiple myeloma. Leukemia and Lymphoma 2002; 43:67–74. Ria R, Falzetti F, Ballanti S, et al. Melphalan versus melphalan plus busulphan in conditioning to autologous stem cell transplantation for low-risk multiple myeloma. Hematol J 2004; 5:118–22. Badros A, Barlogie B, Siegel E, et al. (2001b) Results of autologous stem cell transplant in multiple myeloma with renal failure. Br J Haematol 2001b; 114:822–829. Bird JM, Fuge R, Sirohi B, et al. British Society of Blood and Marrow Transplantation. The clinical outcome and toxicity of high-dose chemotherapy and autologous stem cell transplantation in patients with myeloma or amyloid and severe renal impairment: a British Society of Blood and
1096 Multiple myeloma
115
116
117
118
119
120
121
122
123
124
125
126
127
Marrow Transplantation study. Br J Haematol 2006; 134(4):385–90. Palumbo A, Bringhen S, Petrucci MT,et al. Intermediate-dose melphalan improves survival of myeloma patients aged 50 to 70: results of a randomized controlled trial. Blood 2004; 104(10):3052–7. Barlogie B, Jagannath S, Desikan KR, et al. Total therapy with tandem transplants for newly diagnosed multiple myeloma. Blood 1999; 93:55–65. Attal M, Harousseau JL, Facon T, et al. InterGroupe Francophone du Myelome. Single versus double autologous stem-cell transplantation for multiple myeloma. N Engl J Med 2003; 349(26):2495–502. Russell N, Bessell E, Stainer C, Haynes A, Das-Gupta E, Byrne J. Allogeneic haemopoietic stem cell transplantation for multiple myeloma or plasma cell leukaemia using fractionated total body radiation and high-dose melphalan conditioning. Acta Oncologica 2000; 39:837–41. Gahrton G, Svensson H, Cavo M, et al. Progress in allogeneic bone marrow and peripheral blood stem cell transplantation for multiple myeloma: a comparison between transplants performed 1983–93 and 1994–8 at European Group for Blood and Marrow Transplantation centres. Br J Haematol 2001; 113:209–16. Einsele H, Schafer HJ, Hebart H, et al. Follow-up of patients with progressive multiple myeloma undergoing allografts after reduced-intensity conditioning. Br J Haematol 2003; 121:411–8. Kroger N, Perez-Simon JA, Myint H, et al. Relapse to prior autograft and chronic graft-versus-host disease are the strongest prognostic factors for outcome of melphalan/fludarabine-based dose-reduced allogeneic stem cell transplantation in patients with multiple myeloma. Biol Blood Marrow Trans 2004; 10:698–708. Corradini P, Cavo M, Lokhorst H, et al. Molecular remission after myeloablative allogeneic stem cell transplantation predicts a better relapse-free survival in patients with multiple myeloma. Blood 2003; 102:1927–9. Lokhorst HM, Wu K, Verdonck LF, et al. The occurrence of graft versus host disease is the major predictive factor for response to donor lymphocyte infusions in multiple myeloma. Blood 2004; 103:4362–4. Byrne JL, Carter GI, Bienz N, Haynes AP, Russell NH. Adjuvant alpha-interferon improves complete remission rates following allogeneic transplantation for multiple myeloma. Bone Marrow Trans 1998; 22:639–43. Badros A, Barlogie B, Morris C, et al. High response rate in refractory and poor-risk multiple myeloma after allotransplantation using a nonmyeloablative conditioning regimen and donor lymphocyte infusions. Blood 2001b; 97:2574–9. Giralt S, Aleman A, Anagnostopoulos A, et al. Fludarabine/melphalan conditioning for allogeneic transplantation in patients with multiple myeloma. Bone Marrow Trans 2002; 30:367–73. Kroger N, Sayer HG, Schwerdtfeger R, et al. Unrelated stem cell transplantation in multiple myeloma after a
128
129
130
131
132
133
134
135
136
137
138
139
140
reduced-intensity conditioning with pretransplantation antithymocyte globulin is highly effective with low transplantation-related mortality. Blood 2002; 100:3919–24. Maloney DG, Molina AJ, Sahebi F, et al. Allografting with non-myeloablative conditioning following cytoreductive autografts for the treatment of patients with multiple myeloma. Blood 2003; 102:3447–54. D’Sa S, Peggs K, Pizzey A, et al. T- and B-cell immune reconstitution and clinical outcome in patients with multiple myeloma receiving T-cell-depleted, reducedintensity allogeneic stem cell transplantation with an alemtuzumab-containing conditioning regimen followed by escalated donor lymphocyte infusions. Br J Haematology 2003; 123:309–22. Shaw BE, Peggs K, Bird JM, et al. The outcome of unrelated donor stem cell transplantation for patients with multiple myeloma. Br J Haematol 2003; 123:886–95. Myeloma Trialists’ Collaborative Group. Interferon as therapy for multiple myeloma: an individual patient data overview of 24 randomised trials and 4012 patients. Br J Haematol 2001; 113:1020–34. Bjorkstrand B, Svensson H, Goldschmidt H, et al. Alphainterferon maintenance treatment is associated with improved survival after high-dose treatment and autologous stem cell transplantation in patients with multiple myeloma: a retrospective registry study from the European Group for Blood and Marrow Transplantation (EBMT). Bone Marrow Trans 2001; 27:511–15. Attal M, Harousseau JL, Leyvraz S, et al. Maintenance therapy with thalidomide improves survival in multiple myeloma patients. Blood 2006; 108(10):3289–94. Lahtinen R, Laakso M, Palva I, et al. Randomised placebocontrolled multicentre trial of clodronate in multiple myeloma. Lancet 1992; 340:1049–52. Berenson JR, Lichtenstein A, Porter L, et al. (1998) Longterm pamidronate treatment of advanced multiple myeloma patients reduces skeletal events. J Clin Oncol 1998; 16:593–602. McCloskey EV, Maclennan ICM, Drayson MT, et al. A randomized trial of the effect of clodronate on skeletal morbidity in multiple myeloma. Br J Haematol 1998; 103:902–10. Berenson JR, Rosen LS, Howell A, et al. Zoledronic acid reduces skeletal-related events in patients with osteolytic metastases. Cancer 2001; 91:1191–200. Dimopoulos MA, Kastritis E, Anagnostopoulos A, et al. Osteonecrosis of the jaw in patients with multiple myeloma treated with bisphosphonates: evidence of increased risk after treatment with zoledronic acid. Haematologica 2006; 91(7):968–71. Djulbegovic B, Wheatley K, Ross J, et al. Bisphosphonates in multiple myeloma (Cochrane Review). Cochrane Database Syst Rev 2001; 4:CD003188. Tricot G, Jagannath S, Vesole DH, Crowley J, Barlogie B. Relapse of multiple myeloma after autologous transplantation: survival after salvage therapy. Bone Marrow Trans 1995; 16:7–11.
References 1097
141 Alexanian R, Barlogie B, Dixon D. High-dose glucocorticoid treatment of resistant myeloma. Annal Int Med 1986; 105(1):8–11. 142 Singhal S, Mehta J, Desikan R, et al. Antitumor activity of thalidomide in refractory myeloma. N Engl J Med 1999; 341:1565–71. 143 Juliusson G, Celsing F, Turesson I, Lenhoff S, Andriansson M, Malm C. Frequent good partial remissions with thalidomide including best response ever in patients with advanced refractory and relapsed myeloma. Br J Haematol 2000; 109:89–96. 144 Barlogie B, Zangari M, Spencer T, et al. Thalidomide in the management of multiple myeloma. Semin Hematol 2001; 38:250–259. 145 Palumbo A, Giaccone L, Bertola A, et al. Low-dose thalidomide plus dexamethasone is an effective salvage therapy for advanced myeloma. Haematologica 2001; 86:399–403. 146 Sidra G, Williams CD, Russell NH, et al. Combination chemotherapy with cyclophosphamide, thalidomide and dexamethasone for patients with refractory, newly diagnosed or relapsed myeloma. Haematologica 2006; 91:862–863. 147 Richardson PG, Sonneveld P, Schuster M, et al. Bortezomib or high-dose dexamethasone for relapsed multiple myeloma. N Engl J Med 2005; 353:2487–98. 148 Knight R. IMiDs: a novel class of immunomodulators. Semin Oncol 2005; 32(suppl 5):S24–S30. 149 Richardson PG, Blood E, Mitsiades CS, et al. A randomized phase 2 study of lenalidomide therapy for patients with
150
151
152
153
154
155
156
relapsed or relapsed and refractory multiple myeloma. Blood 2006; 108(10):3458–64. Yasui H, Hideshima T, Richardson PG, Anderson KC. Novel therapeutic strategies targeting growth factor signalling cascades in multiple myeloma. Roy J Haematol 2006; 132(4):385–97. Kyle RA, Rajkumar SV. Monoclonal gammopathy of undetermined significance. Br J Haematol 2006; 134(6):573–89. Rajkumar SV, Kyle RA, Therneau TM, et al. Serum free light chain ratio is an independent risk factor for progression in monoclonal gammopathy of undetermined significance. Blood 2005; 106(3):812–7. Greenberg P, Parker RG, Fu YS, Abemayor E. The treatment of solitary plasmacytoma of bone and extramedullary plasmacytoma. Am J Clin Oncol 1987; 10:199–204. Dimopoulos MA, Moulopoulos LA, Maniatis A, Alexanian R. Solitary plasmacytoma of bone and asymptomatic multiple myeloma. Blood 2000; 96(6):2037–44. Wilder RB, Ha CS, Cox JD, Weber D, Delasalle K, Alexanian R. Persistence of myeloma protein for more than one year after radiotherapy is an adverse prognostic factor in solitary plasmacytoma of bone. Cancer 2002; 94:1323–1537. Dispenzieri A, Moreno-Aspitia A, Suarez GA, et al. Peripheral blood stem cell transplantation in 16 patients with POEMS syndrome, and a review of the literature. Blood 2004; 104:3400–7.
45 Introduction to haematopoietic stem cell transplantation EFFIE LIAKOPOULOU AND DAVID I. MARKS
Introduction Overview of haematopoietic SCT Biology of HSCT – determinants of engraftment Indications for SCT/patient selection Principles of conditioning/preparative regimens for HSCT Monitoring of chimaerism and treatment of relapse post SCT
1098 1098 1100 1101 1102 1104
INTRODUCTION In the late 1950s it became possible to rescue human beings from the hazardous consequences of lethal irradiation by infusion of bone marrow cells.1,2 In over 4 decades since then, with the advances of technology and knowledge developed in this period, haematopoietic stem cell transplantation (SCT) has become common practice and a potential curative option for haematological malignancies including lymphomas, solid tumours and other non-malignant disorders.3 The success of haematopoietic SCT is based on the unique properties of the haematopoietic stem cells for selfrenewal and their capacity to differentiate into all lymphohaematopoietic lineages. The curative potential of SCT is based on: 1. eradication of disease with escalated doses of chemotherapy and radiotherapy; and 2. immunotherapeutic potential of the allogeneic graft: graft vs. leukaemia (GVL)/graft vs. tumour effect (GVT). The development of new conditioning regimens, specialized stem cell transplant units, the improvement of tissue typing techniques and the use of alternative sources of stem cells, made SCT safer and available to a wider spectrum of patient and age groups. This is reflected in the
Complications of SCT Infectious complications of SCT Graft versus host disease Transplant programme requirements and accreditation Conclusion References
1104 1107 1109 1111 1112 1112
increased numbers of transplant procedures performed in Europe and worldwide in recent years (Fig. 45.1). The immunotherapeutic effects of SCT, donor lymphocyte infusions (DLI), stem cell plasticity, availability of alternative sources of stem cells and development of targeted treatments against infectious diseases hold great promise for further development in the field.
OVERVIEW OF HAEMATOPOIETIC SCT There are distinct types of stem cell transplants according to the: (1) source of stem cells, (2) donor availability, (3) conditioning regimen, and (4) T-cell content of the graft:
Source of stem cells Haematopoietic stem cells (HSCs) can be obtained for transplantation from various sources. The bone marrow ‘home’ of haematopoietic stem cells is the richest source. Bone marrow stem cells are obtained by repeat aspirations from the posterior iliac crests performed under general anaesthesia. The problems of accessing this organ to obtain adequate amount of HSCs required for transplantation has been overcome with the use of haematopoietic growth factors for mobilization of HSC from the marrow to the peripheral
Overview of haematopoietic SCT 1099
14000
EBMT activity survey on HSCT 1990–2004: autologous
6000
EBMT activity survey on HSCT 1990–2004: allogeneic
HSCT
HSCT
4000 7000
2000
0
0 1990 1992 1994 1996 1998 2000 2002 2004
1990 1992 1994 1996 1998 2000 2002 2004 Lymphoproliferative disorders
Leukaemias
Solid tumours
Non-malignant
Figure 45.1 EBMT Haematopoietic Stem Cell Transplant Activity between 1990 and 2004 by disease group (With kind permission of Prof A Gratwohl for the EBMT Registry) A. Autologous SCTs; B. Allogeneic SCTs.
blood as peripheral blood stem cells (PBSCs). Their use is associated with rapid engraftment and haematopoietic reconstitution but increased incidence of acute and chronic graft vs. host disease (GVHD) in allogeneic transplants.4–7 The cord blood of the human fetus has been identified as a very rich source of HSCs. Therefore, compatible cord blood collections have been more recently used as potential source of stem cells for SCT in children and adults lacking other potential donors/sources of stem cells.8,9
Donor availability Autologous transplants involve infusion of stem cells collected from the recipient themselves at a time of a good partial or complete remission from their disease. The cells are stored and infused back to the recipient after completion of high-dose chemo- and/or radiotherapy, to enable haematopoietic rescue. Contamination of the autologous collection by the disease and difficulty mobilizing HSC particularly in heavily pre-treated individuals are frequent problems in autologous SCT. Allogeneic SCTs are performed by using HSC donations from a human leukocyte antigen (HLA)-matched sibling of the recipient. The donor can be HLA matched or partially mismatched to the recipient. Thirty per cent of the patient population eligible for an allogeneic transplant have an HLA-matched sibling donor, leaving the majority of patients who could potentially benefit from transplantation without a suitable sibling donor. Haematopoietic stem cell donations from volunteer unrelated donors (VUDs) (HLA matched or mismatched) are used when an allogeneic transplant is indicated in
high-risk recurrent diseases for patients with no suitable sibling donor available. Despite the successful development and expansion of several volunteer donor registries, it is not possible to identify suitable HLA-compatible donors for all patients requiring HSCT. The suitability of alternative sources of stem cells for HSCTs has been explored. Umbilical cord blood transplants have been successful in children8,10,11 and results in adults using two units are comparable to the results of other unrelated donor transplants. Cord blood is collected and stored frozen immediately after birth. The product can be produced easily and safely when transplantation is urgent. The use of umbilical cord blood as a source of stem cells in adults is limited by the dose of stem cells available with each unit of cord blood. The dose of HSCs directly affects the time to engraftment and therefore the length of the post-conditioning neutropenia and effectively the transplant-related mortality. This limitation seems to have been overcome with the growth of cord blood banks and the improvement in collection techniques, which increased the yield of nucleated cells in each cord blood unit. Cord blood transplants performed with the combination of two cords enables timely engraftment.9,12 Ex vivo expansion of cord blood stem cells is under investigation.13 Haploidentical donors serve as potential alternative source of haematopoietic stem cells in absence of other sources. The genetic disparity between donor and recipient had been associated with limited success due to high rates of engraftment failure and GVHD. Improved outcomes particularly in children have been achieved with aggressive and more effective T-cell depletion and maximal escalation of the dose of stem cells.14–16
1100 Introduction to haematopoietic stem cell transplantation
Intensity of conditioning–preparative regimen Full-intensity conditioning regimen or myeloablative transplants are performed by using myeloablative doses of chemotherapy and radiotherapy to maximize the therapeutic and immunosuppressive effects achieved by such doses. The most common myeloablative conditioning regimens used in HSCT are the combinations of cyclophosphamide, either with fractionated total body irradiation (TBI) or busulfan. Non-myeloablative but immunosuppressive doses of chemotherapy and or radiotherapy are used for conditioning of patients receiving reduced intensity transplants. The therapeutic potential of these transplants mainly relies on the graft vs. malignancy effect of the graft.
T-cell content of the graft Apart from the HSCs found within the CD34-positive inoculum, the graft contains a repertoire of other immunopotent and auxiliary cells. It is clear that donor T-cells have an active role in engraftment and the development of GVH/GVT effect. Technology enabling the characterization of cell populations in the graft has provided the ability for the manipulation of its cell component and the potential to control the incidence of acute and chronic GVHD and the potentiation of GVL effect. T-cells could be removed prior to (ex vivo) or with their infusion to the recipient (in vivo). There are different T-cell depletion techniques available based mainly on negative and positive selection. Positive selection is achieved with the use of anti-CD34-specific antibodies conjugated with magnetic beads. Immunological methods are mainly used for T-cell depletion in vivo. The use of alemtuzumab (Campath-1H) anti-CD52 monoclonal antibody (20–100 mg) is common practice in several centres for the effective reduction of the incidence of acute GVHD post-transplant.17 The optimal dose of the antibody required remains to be established18 (Campath De-escalation Multicentre Trial). High doses of the antibody administered pre-transplant may be associated with an increased risk of relapse.
BIOLOGY OF HSCT – DETERMINANTS OF ENGRAFTMENT HSCT differs from solid organ transplants in that the immune system of the recipient is mostly destroyed with the preparative (conditioning) for the transplant regimen and is substituted by the immune system of the donor of which mature elements are infused or regenerate from the graft. In HSCT therefore clinical concerns involve the occurrence and the management of GVHD, in addition to the potential of graft rejection. In the autologous setting the
main clinical problems arise from the toxicity of the high doses of chemotherapeutic agents involving vital organs and the intensity of myelosuppression. In the allogeneic transplant setting the degree of genetic compatibility between the donor and recipient in regard to the HLA class I and II antigens in particular is fundamental for the successful outcome of the HSCT.19–21 Host immunosupression is vital for the ‘acceptance’ of the graft and is achieved by the administration of the conditioning regimen and immunosuppressive drugs as proengraftment agents. These immunosuppressive agents continue further post engraftment mainly for prevention of GVHD, until immunological tolerance is eventually established. Restoration of haematopoiesis post myeloablative therapy is dependent on the transplantation of the subset of HSCs characterized as the long-term reconstituting cells (LTCRs). These cells have the unique properties of self-renewal and the capacity to differentiate into all different haematopoietic lineages. In fact a single stem cell can restore the entire lymphohaematopoietic system of a lethally irradiated animal.22 The cell surface glycoprotein CD34 has been widely used as a marker to characterize the HSCs-LTRCs for clinical transplantation. As this marker is expressed also on committed haematopoietic progenitors, the HSC count is only part of the total CD34-positive cell inoculum of the graft. In clinical transplantation engraftment (reconstitution of haematopoiesis by the graft) is normally the point in which the neutrophil count reaches 0.5 109/L. Time to engraftment is dependent on the dose of transplanted stem cells. Although the minimum safe number of CD34positive cells needed for engraftment of all lineages is not known, the lowest acceptable doses of cells for transplantation for most centres are 2 106 CD34-positive cells/kg of peripheral blood stem cells (PBSCs) and 2 108 MNC/kg if bone marrow is used. Due to the effect of the dose of stem cells in reducing the transplant-related mortality but also to the increased incidence of GVHD with stem cell doses over 8 106 CD34 cells/kg, most transplant programmes aim for doses of CD34-positive cells between 4–8 106 CD34-positive cells/kg.4,23 Durable engraftment is facilitated by the presence of donor CD8-positive cells in the graft as they help to overcome the last points of resistance triggered by the post conditioning remnants of the hosts’ T-cells.24–27
HLA compatibility Histocompatibility between donor and recipient is one of the major determinants of the outcome of SCT as it reduces the incidence and the severity of undesirable alloimmune reactions such as graft rejection and GVHD. Alloreactivity can also be potentiated by exposure of the immunocompetent host (patient and donor) to alloantigens. It is often
Indications for SCT/patient selection 1101
due either to transfusion of blood products (diminished with leukodepletion) or less frequently gestation in females. The improved outcome of unrelated donor SCTs has followed closely discoveries about the elements and function of the HLA system in humans. Significant technical improvements on the histocompatibility front in the last decade optimized donor selection and consequently have had an enormous effect on the reduction of transplant-related mortality due to graft rejection and GVHD. Genes encoding histocompatibility antigens (HLAs) are located within the class I and II regions of the major histocompatibility complex (MHC). There are three HLA class I genes, HLA-A, B and C, which are widely expressed on most nucleated cells of the body including platelets. HLA class II genes are somewhat structurally different to those of class I. There are also three classical class II molecules – HLA-DR, DP and DQ – which are transplantation antigens. Their expression is limited on haematopoietic cells such as dendritic cells, macrophages and B-cells.28,29 HLA compatibility is the major determinant for donor selection. Choosing a suitable related or more importantly unrelated donor a set of 10 or 12 antigens comprising both classes of the HLA complex is assessed. HLA-matched siblings are identical for all genes within the MHC as they share the same two extended haplotypes. Allele typing and matching for the classical HLA genetic loci is undertaken for unrelated donors to determine their suitability.30
Other parameters such as donor’s gender, CMV status and blood group play a role in the donor selection process when more than one potential donor is available.31–33 Clearly multiple disparities in class I or II genes between donor and recipient are associated with poor outcome although information for the significance of specific class mismatches is incomplete.19,21,30 Haploidentical transplants are performed with the use of a parent sibling or a child with only one identical haplotype. This type of transplantation involves a different alloreactive mechanism based on natural killer cells expressing combinations of activating and inhibitory killer cell immunoglobulin-like receptors that interact with class I HLA epitopes. Alloreactivity in this setting improves the chances of engraftment and reduces the risk of GVHD.14 More than 50 volunteer donor registries around the world serve as the source of HLA-matched donors. In addition to these there are several cord blood banks in many countries. Between these sources there are more than 8.5 million potential donor/cord blood units registered worldwide. Haematopoietic SC donors provide PBSCs or marrow for more than 15 000 allogeneic HSCTs annually. The donation ratio between related and unrelated donors is 3:1.34
INDICATIONS FOR SCT/PATIENT SELECTION HSCT has become common practice and established for the treatment for haematological malignancies (Table 45.1). The list of recommendations for HSCT is expanding and
Table 45.1 Current indications for SCT in haematological disorders35–37
Disease AML 1st CR Good risk cytogenetics Standard risk cytogenetics Poor risk cytogenetics AML 2nd CR ALL first CR (normal cytogenetics) ALL 1st CR (9;22) ALL 2nd CR CML 1st CP MDS Multiple myeloma HD 1st CR HD relapsed NHL DLBCL 1st CR NHL DLBCL relapse NHL follicular Aplastic anaemia Haemoglobinopathies
Autologous SCT
Allogeneic SCT Sibling SCT
VUD SCTs
No Yes Yes Yes D Yes Yes No No Yes No Yes D Yes D No No
No Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes
No No Yes Yes No Yes Yes Yes* Yes D No D D D D D D
D, developmental; *Post trial with imatinib; CR, complete remission.
1102 Introduction to haematopoietic stem cell transplantation
currently includes non-haematological malignancies, immmunodeficiencies, metabolic and autoimmune diseases.35–40 The plasticity of the HSC indicating its ability to differentiate to other than haematopoietic tissues in the influence of specific environmental stimuli, may extrapolate the use of SCT for the treatment of non-haematopoietic organ disorders in the future. Early transplantation is critical in patients with haematological cancers. The appropriate time to perform the transplant is often subject of controversy and in general HSCT is broadly underused.41 Although SCT may be indicated for the treatment and potential cure of a specific disease, a number of factors should be considered prior to offering the procedure to a patient. It is recommended that the decision to transplant should be made by an expert in the field on the basis that the benefits should outweigh the risks. The patient’s age has long been a significant limiting factor for the wider application of SCT. The toxic effects of allogeneic SCT increase with age and generally the procedure precludes individuals older than 65 years. There has been a continuous expansion of the upper age limit of transplant recipients, mainly due to the development of less intensive preparative for transplant regimens and specialist care in purpose-built transplant units. Pre-transplant work-up consists of a number of investigations to ensure fitness of the vital organs of the recipient for the particular kind of transplant being considered. This consists of routine haematology, biochemistry, reassessment of stage of disease, creatinine clearance/ glomerular filtration rate (GFR), pulmonary function tests and echocardiogram. Physical, psychological fitness and the ability of the transplant candidate to comply with the treatment short and long term need to be carefully assessed to minimize the risks of the transplant procedure.
PRINCIPLES OF CONDITIONING/PREPARATIVE REGIMENS FOR HSCT The success of HSCT is based on overcoming two important but opposing immunological barriers: the graft vs. host and the host vs. graft responses. This is more relevant in the allogeneic setting and is overcome by administration of supralethal doses of chemo- and radiotherapy, infusion of stem cells for rescue and post-transplant immunosuppression to suppress GVHD and induce tolerance. Preparation of the recipient’s system to accept the new organ is termed the ‘conditioning or preparative regimen’ and usually consists of a combination of chemotherapeutic drugs with or without TBI. Agents frequently used in conditioning regimens and their organ-limiting toxicities are described on Table 45.2. Common conditioning regimes used in autologous and allogeneic HSCT are shown on Tables 45.3 and 45.4.
Table 45.2 Agents used in SCT conditioning regimens and their dose-limiting toxicities Agent
Dose
Toxicity/ies
TBI
10–16 Gy
Cyclophosphamide
120 mg/kg
Busulphan
14–16 Gy/kg
Melphalan Etoposide/VP16 Cytarabine Carmustine
100–200 Gy 300 mg/m2/60 mg/kg 800–1600 mg/m2 300–600 mg/m2
Pulmonary, GI/liver Cardiac, renal, bladder Pulmonary, liver/gut GI, renal GI CNS Pulmonary/hepatic
CNS, central nervous system; GI, gastrointestinal system.
Table 45.3 Commonly used conditioning regimens in SCT Autologous transplants
Allogeneic transplants
Myeloma 1st and 2nd Melphalan 200 mg/m2 Melphalan/TBI
Myeloma Cy/TBI, Mel/TBI
Lymphoma BEAM CVB Cy/TBI
Lymphoma BEAM and Maxi BEAM Cy/TBI
Acute myeloid leukaemia Bu/Cy, Cy/TBI
Acute myeloid leukaemia Bu/Cy, Cy/TBI
Acute lymphoblastic leukaemia Cy/TBI
Acute lymphoblastic leukaemia Etoposide (VP16)/TBI, Cy/TBI
BEAM: carmustine (BCNU), etoposide, cytarabine (AraC), melphalan; CVB: cyclophosphamide, etoposide, BCNU; TBI, total body irradiation; Bu, busulphan; Cy, cyclophosphamide.
Table 45.4 The evolution of conditioning regimens and transplant biology 1970s Addition of cyclophosphamide to TBI 1980s Dose intensification Alternatives to cyclophosphamide Use of volunteer donors T-cell depletion 1990s Non-myeloablative conditioning Reduction of TRM and morbidity Improvement of QoL The role of GvL and DLI Knowledge of the requirements for stable engraftment TRM, transplant-related mortality; QoL, quality of life; GvL, graft vs. leukaemia effect; DLI, donor lymphocyte infusion.
Principles of conditioning/preparative regimens for HSCT 1103
The dose intensity of the conditioning regimen may determine the clinical outcome of the transplant (morbidity/ mortality, disease-free survival). As high-dose chemo-radiotherapy increases cell destruction and stem cell support made dose escalation possible, conditioning regimens were originally designed to eliminate the diseased haematopoietic system without causing irreversible vital organ damage. The main objectives of the preparative regimen apart from the disease eradication, is ‘space making’ and immunomodulation/ suppression. TBI has been used as part of some conditioning regimens for its myeloablative and immunosuppressive properties (Fig. 45.2).In addition TBI is a powerful agent against a variety of malignancies, even those resistant to chemotherapy and it has the ability to penetrate sanctuary sites (CNS and testis). It is administered via a linear accelerator and is prescribed as a single dose, fractionated or, (6 fractions over 3 days, 10–12 Gy) or hyper fractionated (8 fractions over 4 days, 14–15 Gy). Fractionation of TBI reduces the incidence and the severity of acute and late complications.42–44 Immunosuppressive low-dose TBI 2–4 Gy has been used alone or in combination with fludarabine as part of reduced intensity conditioning regimens as described in this chapter.
Myeloablative conditioning regimens Conditioning regimens for autologous SCTs are mostly myeloablative usually consisting of disease-specific chemotherapeutic agents with or without TBI. The degree of disease response to previous combination chemotherapy, disease behaviour and the results of the pre-transplant work-up are the main reasons for modification of the specific conditioning in order to meet the individual patient’s requirements. In the allogeneic setting myeloablative conditioning regimes are used for disease eradication and immunomodulation. They consist of high doses of chemo-radiotherapy.
Conditioning regimen
Immunosuppression
Reduced intensity
Myeloablative
Flu/TBI 2Gy Flu/TBI 2Gy
MF 140 FC
Bu/F/Auto
Cy/TBI Bu16/Cy
TBI 2GY FLAG IDA
Myelosuppression CLL
CML
MM
AML
Figure 45.2 Conditioning reimens in Allogeneic Stem Cell Transplantation.
The most commonly used combinations are busulfan with cyclophosphamide, cyclophosphamide with TBI and etoposide plus TBI. Increasing the dose of TBI decreases probability of disease relapse although without improvement in survival as there is an increase in TRM related to TBI dose escalation.43,44 Two randomized trials have shown equivalency of Cy/TBI and Bu/Cy in patients with chronic phase CML receiving HLA-matched allogeneic transplants.45,46 In patients receiving such transplants for acute myeloid leukaemia Cy/TBI was shown to be superior in one study due to lower relapse rate.47 Despite the technological and therapeutic advances disease relapse remains relatively high in patients receiving SCT. This dictates the necessity of novel conditioning/treatment regimens and randomized trials to establish the efficacy of specific regimens for specific groups of patients and diseases and compare them with already established conditioning regimens. Further development of cytoprotective strategies may reduce toxicity of the conditioning regimens and consequently the morbidity/mortality related to them.
Reduced intensity conditioning regimens The major disadvantages of myeloablative SCT have already been alluded to and are the morbidity and mortality associated with the toxicity of the conditioning regimen and the occurrence of GVHD.48 These are also the main limiting factors of the application of allogeneic SCT in broader patient and disease groups. In parallel there has been clear evidence for the curative potential of graft vs. leukaemia/ graft vs. tumour effect and the potential to induce durable remissions with DLIs to treat disease relapse after HSCT.49 The introduction of the reduced intensity conditioning regimens in the mid-1990s aimed to offer the solution to the toxicity-related mortality and to potentiate the GVL effects.50,51 A variety of conditioning regimens are designed to immunosuppress the host, and allowing timely engraftment, to control the disease by augmentation of the graft vs. tumour effect, with overall reduced toxicity. Host immunosuppression is achieved by antimetabolite chemotherapeutic agents such as purine analogues, (e.g. fludarabine), alone or in combination with alkylating agents (busulfan, melphalan), a low dose of TBI (2–4 Gy), or alemtuzumab (Campath 1H). Additional immunosuppression of the host is achieved with cyclosporin A alone or a combination of mycophenolate mofetil (MMF) depending on the specific regimen. The engraftment kinetics, the onset of acute chronic GVHD are altered in this transplant setting and are related to the type of conditioning regimen and the length of the post transplant immunosuppression. Once established the grafts would ‘make space’ in the marrow through GVH reaction. Control of GVH and HVG alloimmune reactions results in graft vs. host tolerance expressed by mixed donor chimerism/persistence of recipient in haematopoietic or lymphoid lineages.
1104 Introduction to haematopoietic stem cell transplantation
The monitoring of chimaerism and the persistence of mixed donor chimaerism serves as a basis for adoptive immunotherapy/GVT effect with DLI. The application of reduced intensity stem cell transplant is indicated in lymphoid malignancies, multiple myeloma, lymphoma and Hodgkin’s disease and potentially in allogeneic SCT for solid tumours (renal cell carcinoma).52,53 Their effect in acute myeloid leukaemia has been assessed in patients not eligible for the preferable curative full intensity transplant.52,54–56 Infections related to the degree and length of immunosuppression and T-cell depletion are of clinical significance in this setting, particularly in alemtuzumab (Campath 1H) containing regimens. In addition to their low incidence of treatment-related toxicity reduced intensity SCT can be offered in the ambulatory care setting.57
GVHD prophylaxis in allogeneic SCT Immunosuppressive agents are used in addition to the conditioning regimens to ensure engraftment, prevent graft failure and prevent GVHD. In conventional myeloblative SCTs the prevention of GVHD is achieved by the combination of cyclosporin A (CsA) and IV methotrexate infusions on days 1, 3, 6 and 11. In vivo T-cell depletion with alemtuzumab, Campath 1H (anti-CD52 antibody) is also used for the prevention of GVHD in the non-myeloablative setting.17 Recipients of mismatched grafts may merit more intense GVHD prophylaxis overall, although patients with high-risk diseases may benefit from the potentiation of graft vs. host/GVL effect. In general the GVHD prophylaxis strategy has to be adapted according to the individual case requirements considering the major causes of treatment failure, which are: the risk of relapse, incompatibility between recipient and donor, and the recipient’s previous treatment.
MONITORING OF CHIMAERISM AND TREATMENT OF RELAPSE POST SCT The engrafted donor cells can be traced throughout the recipient’s life post-transplant by using specific genetic markers. Evaluation of such markers prior to SCT in the donor and the recipient allows the identification of informative loci, which can be tested for post-transplant. Conventional cytogenetics on metaphase chromosomes is a well-established method to monitor chimaerism. Sex chromosomes are used for this purpose in cases of gender disparity between donor and recipient. In situ hybridization of sex chromosome specific probes and short tandem repeat (STR) polymorphisms or variable number of tandem repeat (VNTR) by DNA amplification are widely applied techniques offering high sensitivity, specificity and quantitative results in chimaerism monitoring. Chimaerism is helpful
in monitoring engraftment and early disease relapse, particularly in myeloid diseases with sensitivity 1–5 per cent.58,59 The recognition of the therapeutic potential of the GVT/GVL effects has revolutionized the treatment of the disease relapse post SCT. There is no standard treatment but the situation should be evaluated depending on the type (molecular, cytogenetic, morphological, extramedullary) and time of relapse post SCT. There are various options that may be considered such as cessation of immunosupression, donor lymphocyte infusion (with or without re-induction of chemotherapy), interferon, second transplants, targeted therapy (Imatinib mesylate), irradiation and palliation.58,60*,61
COMPLICATIONS OF SCT Most of the early complications of SCT are related to the conditioning regimen and are associated with the combination of drugs used for conditioning (Table 45.3). The complications of the SCT together with the relapse of primary disease remain the main causes of morbidity and mortality of this treatment (Fig. 45.3). Complications can occur early (up to day 100) and up to several years post transplant (late effects).
Renal complications and haemorrhagic cystitis Renal dysfunction is frequent in stem cell transplant recipients. Apart from the predisposing factors such as the original diagnosis (e.g. multiple myeloma) and previous nephrotoxic treatments, the combination of drugs used during the peri-transplant period can precipitate renal dysfunction. Common nephrotoxins are cyclosporin A, aminoglycoside antibiotics, vancomycin and antifungal agents (amphotericin B and derivatives). Cyclosporine toxicity can manifest as a microangiopathic haemolytic anaemia (MAHA) and cyclosporine-related thrombotic thrombocytopenic purpura (TTP).62 Such manifestations of CsA toxicity can be life threatening and upon suspicion withdrawal of the drug and the use of alternative GVHD prophylaxis is essential. Haemorrhagic cystitis is usually due to the high-dose chemotherapy of the conditioning regimen, particularly high-dose cyclophosphamide or ifosphamide. Viruses such as BK. JC. adenovirus are also causes for haemorrhagic cystitis.63 Hyperhydration and infusion of MESNA is used for the prevention of this complication. Treatment with forced hydration, platelet support, bladder irrigation and additional specific measures is dictated by the severity of this complication and its response to the initial management.
Mucositis and other gastrointestinal complications Mucositis occurs secondary to desquamation of the gastrointestinal epithelium caused by the conditioning regimen.
Complications of SCT 1105
AUTO
Relapse (75%)
Organ toxicity (8%) IPn (2%) Infection (6%)
HLA-ID SIB GVHD (14%)
Other (9%)
Other (13%)
Relapse (38%)
UNRELATED Relapse (32%) GVHD (14%) Infection (17%) IPn (5%)
Organ toxicity (13%)
Organ toxicity (11%) Other (17%) IPn (7%) Infection (19%)
Figure 45.3 Causes of death following transplants done between 1998–2002.
It is an invariable complication related to the intensity of the conditioning regimen and remains an important clinical problem in SCT affecting the patients’ quality of life and performance status. There are several scoring systems for grading oral mucositis. The most commonly used one is the WHO scoring system (Table 45.5). Patients with grade III or IV mucositis usually require topical and systemic opiate analgesia and nutritional support (enteral or parenteral nutrition). Apart from occasional modification of treatment and the conservative measures focussing in the amelioration of symptoms related to mucositis, there was until recently no way of preventing this complication. Recombinant human keratinocyte growth factor (KGF) given prior and after completion of the conditioning regimen reduces the severity and the length of oral mucositis in autologous and allogeneic stem cell transplants.64**,65 This agent can be used for the prevention and treatment of mucositis and improvement of quality of life (QOL) in patients receiving SCTs. Nausea, vomiting, anorexia and weight loss are common problems following administration of chemotherapeutic regimens. The presence of these symptoms early
Table 45.5 Grading of mucositis by WHO criteria Grade 0 Grade 1 Grade 2 Grade 3 Grade 4
No objective findings: function irrelevant Erythema plus soreness: function irrelevant Ulceration: ability to eat solids Ulceration: oral intake limited to liquids Ulceration: unable to tolerate solids or liquids
post transplant (until day 20) is usually due to the intensity of the conditioning regimen. Presence of these symptoms beyond this period of 20 days indicates other causes involved such as infections (herpes viruses), drugs, or acute GVHD. Prophylaxis with serotonin receptor antagonists with or without dexamethasone is commonly used to control conditioning-induced nausea and vomiting.66 Diarrhoea is also mostly related to the conditioning regimen and reflects the underlying mucosal damage caused by it. Diarrhoea is more common in allogeneic SCT recipients. Causes of persistent diarrhoeal symptoms post SCT are acute GVHD, enteric infection by viruses, and bacteria and drugs67 (Table 45.6).
1106 Introduction to haematopoietic stem cell transplantation
Table 45.6 Causes of diarrhoea in stem cell transplantation Chemotherapy-induced diarrhoea/conditioning regimens (days 0–20) A. Infectious diarrhoea (days 20–100) Astrovirus Coxsackie Clostridium difficile Echo (infrequent) Adenovirus Herpes simplex Aeromonas Rotavirus Cytomegalovirus Mixed organisms Others (Pseudomonas sp., Citrobacter, Clostridium sp., fungal overgrowth) B. Non-infectious diarrhoea Diarrhoea-causing IV broad-spectrum antibiotics medications Po non-absorbable antibiotics Magnesium salts Tacrolimus Promoting agents: metaclopramide, erythromycin C. Mixed infective and GVHD
Sinusoidal obstruction syndrome/ veno-occlusive disease of the liver The development of jaundice post SCT is a challenging diagnostic problem usually attributed to multiple factors.68 Sinusoidal obstruction syndrome (SOS) or VOD, is the clinical syndrome characterized by jaundice, painful hepatomegaly, fluid retention, and weight gain (2–5 per cent of basal weight) that presents early post high-dose myeloblative therapy. SOS can be mild, moderate or severe depending on the severity of the presenting features. Obliterative fibrosis of the small hepatic venules is observed by light microscopy of hepatic biopsies.68,69 A Risk factor for the development of SOS/VOD is conditioning with cyclophosphamide, busulfan, TBI, and gemtuzimab (anti-CD33 monoclonal antibody). Predisposing factors include chronic hepatitis, hepatic fibrosis, systemic bacterial or viral infection prior to the initiation of conditioning, previous haemopoietic stem cell transplant or treatment with alemtuzumab, previous hepatic irradiation, and hepatotoxic drugs.70–72 The differential diagnosis of SOS/VOD includes drug toxicity, particularly cyclosporine, azoles, methotrexate and TPN, acute GVHD of the liver and infections (viral, fungal, hepatitis), renal failure, fluid overload and constrictive pericarditis. The incidence of this condition ranges from 0–50 per cent depending on the conditioning regimen. The quoted mortality of the syndrome is 1–3 per cent for all stem cell transplants.71 The diagnosis is based on clinical criteria but
imaging of the liver with Doppler ultrasound could be supportive for the diagnosis. Prevention with careful consideration of the risk factors, alteration of the conditioning regimen when possible, careful observation of the fluid balance and the use of diuretics are important. Targeted chemotherapy (IV busulfan or treosulfan) may help to reduce the incidence of the syndrome post SCT. Defibrotide has anti-ischaemic antithrombotic and thrombolytic properties. Defibrotide infusions have been shown to induce complete resolution of moderate or severe SOS in 35–60 per cent of patients suffering with it.73
Pulmonary complications and intensive care requirements Of stem cell transplant recipients, 40–60 per cent develop pulmonary complications.74 These complications have a higher mortality rate than any other organ. The time of occurrence of pulmonary complications post stem cell transplant is associated with the effects of the conditioning regimen on vital organs, often the kinetics of the graft, the degree and time of immune reconstitution. Early pulmonary complications are usually due to infections, cardiac dysfunction, fluid overload, SOS/ VOD and engraftment syndrome. Later (post day 30), cytomegalovirus, other human herpes viruses, Pneumocystis carinii (PCP), atypical bacteria and Aspergillus are the most common infective pulmonary complications along with noninfective idiopathic pneumonia syndrome and diffuse alveolar haemorrhage. Post day 100, bacterial, fungal (filamentous fungi), Nocardia, viruses, mycobacteria, PCP, obstructive pulmonary disease, bronchiolitis obliterans and relapsed disease are the main causes of pulmonary complications in this period.77 Radiological appearances may be helpful but are rarely pathognomonic. Bronchoalveolar lavage (BAL) is the investigation of choice for the establishment of the diagnosis for diffuse pulmonary infiltrates. Microbiological testing on BAL allows for rapid diagnosis (does not necessarily exclude a cause) and stratification of patient’s management. Cardiogenic and non-cardiogenic pulmonary oedema occur early post transplant and is associated with the direct toxic effects of the conditioning regimen or previous treatments mainly to the heart or other organs. Iatrogenic hypervolaemia is often a precipitating factor.77 Owing to pulmonary and other complications many transplant recipients (7–40 per cent) require HDU or ITU support.77 The wide variation in this number is due to the type of the transplant/conditioning regimen, the patients’ age and the underlying diagnosis. Although the overall ITU mortality rate in cancer patients is about 42 per cent this increases to over 70 per cent in patients requiring mechanical ventilation.74 Prognostic factors influencing the outcome in this population is the cause of severe lung injury, hypotension requiring vasopressor support for more than 4 hours and combined hepatic and
Infectious complications of SCT 1107
renal dysfunction. The occurrence of the latter in mechanically ventilated patients increases the mortality to nearly 100 per cent.78,79 This data should be interpreted with caution as mortality models apply to patient aggregates and not to individual cases. The disease status at transplantation, the outlook of the procedure, the chance of cure with transplantation, patient’s previous history and the degree of vital organ failure should be considered for stratifying the patient’s treatment. Patients post SCT requiring HDU or ITU support are usually treated aggressively for a few days as those that will improve will show such signs within this period. A multidisciplinary team approach should be followed under the care of the treating physician. The patient’s family should be kept well informed prior to transplant and during the period of the critical illness in preparation for possible end-of-life decisions.
Host immune system defect
Device risk Allogeneic patients
INFECTIOUS COMPLICATIONS OF SCT During and after SCT the compromised host defences increase the susceptibility to bacterial and other infections. Neutropenia, damage of skin barrier and the mucosal membranes, decreased cellular and humoral immunity and immunosuppressive therapy are some of the defects of the host’s immunity predisposing to infections during and post transplant. Bacterial, viral and fungal infections significantly contribute to the transplant morbidity and mortality in the short and long term. The occurrence of specific type of infections is associated with the type of the transplant, the length of the neutropenic period, the degree of immunosuppression and the presence of GVHD. The incidence of pathogens is associated with the time post transplantation (Fig. 45.4). Prophylactic treatment and infection control policies have an important role in the
Phase I, Pre-engraftment, 30 days
Phase II, Postengraftment, 30–100 days
Phase III, Late phase, 100 days
Neutropenia, mucositis and acute graft-versushost disease
Impaired cellular immunity and acute and chronic graftversus-host disease
Impaired cellular and humoral immunity and chronic graft-versushost disease
Central line Respiratory and enteric viruses Herpes simplex virus*†
Cytomegalovirus*† Varicella-zoster virus† Epstein-Barr virus lymphoproliferative disease†
Facultative Gram-negative bacilli Staphylococcus epidermidis Encapsulated bacteria (e.g., pneumococcus)
Gastrointestinal tract Streptococcus species All Candida species Aspergillus species
Aspergillus species Pneumocystis carinii Toxoplasma gondii Strongyloides stercoralis
0
30
Days after transplant
*Without standard prophylaxis †Primarily
among persons who are seropositive before transplant
100
360 High incidence (10%) Low incidence (10%) Episodic and endemic Continuous risk
Figure 45.4 Phases of opportunistic infections amongst allogeneic HSCT recipients.135
1108 Introduction to haematopoietic stem cell transplantation
reduction of such life-threatening infections. Prophylaxis consists of specific supporting measures, dedicated transplant units with prophylactic isolation in single and HEPA (High Efficiency Particulate Air) filtered rooms preferably for allogeneic transplant recipients. Chemical prophylaxis consists of mouthwashes, antifungal, antiviral and possibly antibacterial prophylactic agents. Specific strict infection control measures, visiting regulations, screening protocols prior during and post transplant apply but may differ depending on the local infection control policies.
Bacterial infections Bacterial pathogens are involved in the majority of infections during the neutropenic period. Bacterial infections caused by Gram-negative bacteria are predominantly relevant early post transplant during the period of aplasia. Central IV catheter (CVC) infections are normally caused by Gram-positive bacteria and account for the increased incidence of Gram-positive infections in recent years. The most common infection is with a coagulasenegative Staphylococcus (CNS). Resistant Gram-negative pathogens are an increasing problem. Infections caused by encapsulated bacteria (Streptococcus sp., H. influenzae) and Pneumocystis carinii are usually relevant in the later posttransplant period.80 Infective complications are less frequent in the autologous transplant setting, when they occur mostly during the phase of aplasia. In the reduced intensity transplant setting the timing of infection differs with low incidence of early infections but similar or increased risk of late viral and fungal infections related to immunosuppression and GVHD.80 There should be increased awareness of the subtle signs of infection in the immunocompromised patient and their life-threatening potential. With the onset of symptoms or signs or even the suspicion of an evolving episode of fever the patient should undergo screening with blood cultures and other investigations relevant to their signs. Broad-spectrum antibiotics according to individual units’ protocols and patients’ requirements should be initiated empirically.81 The morbidity and mortality associated with these infections has been improving in the last few years as our knowledge of their pathogenesis, epidemiology and effective treatment has advanced significantly.82
Viral infections With the increasing frequency of graft manipulation/T-cell depletion and longer term immunosupression of the recipient, viral infections have become a considerable and increasingly important problem in the post-transplant period. Viruses like cytomegalovirus (CMV), Epstein–Barr virus (EBV) and herpes simplex virus (HSV) are significant causes of morbidity post SCT.82–85 Owing to advances of
the treatment of complications cased by these viruses, the use of PCR for detection and to the increased rates of T-cell depletion other viruses like adenovirus, respiratory viruses, entero and parvo viruses are becoming of increasing importance.86–88 CMV used to be one of the major infectious complications in the post-transplant period and indirectly affects the outcome of a transplant. In the recent years the use of polymerase chain reaction (PCR)-based pre-emptive treatment and the development of specific anti-viral agents have significantly reduced the incidence of CMV-related disease and CMV-related mortality.89 There are three main mechanisms of acquisition of CMV: (1) primary infection by donor cells or blood products in a previously negative recipient, (2) reactivation of endogenous virus, and (3) re-infection with a different viral strain of a previous seropositive patient. CMV prevention is ensured with the administration of selected appropriate blood products. CMV infection could result in asymptomatic viremia, infection or in CMV organ-specific disease. Organs commonly affected by CMV disease are the lungs (pneumonitis), GI tract (ulceration and hepatitis) and the eyes (retinitis).90 Pneumonitis is the most common and serious manifestation of CMV infection, with a mortality rate of about 50 per cent. CMV reactivation occurs in 40–80 per cent of the patients receiving SCT depending on the risk factors.91 In the autologous setting reactivation occurs in 25–45 per cent of seropositive patients.92 Risk stratified directed PCR surveillance and pre-emptive approach to treatment is recommended in allogeneic SCTs and high-risk autologous patients.89 Treatment of CMV reactivation or disease is with ganciclovir 5 mg/kg BD or Foscarnet 60 mg tds with or without immunoglobulins (which can be CMV specific). Ganciclovir is myelotoxic and nephrotoxic. Foscarnet is particularly nephrotoxic. Treatment should continue until achievement of consecutive negative results or for at least 7 days followed by maintenance the length of which depends on the patient’s immunosuppression and other risk factors.89 The use of adoptive cellular immunotherapy with cytotoxic lymphocytes (CTLs) reactive to specific CMV proteins is feasible in allogeneic transplant recipients.93–95
Fungal infections The increasing frequency of fungal infections post allogeneic SCT over the years is due to many factors, mainly thought to be immunosuppression and T-cell depletion. Common pathogens are Aspergillus and Candida species. Invasive fungal infections (IFIs) due to Candida spp. occur post allogeneic and autologous transplants. Their incidence ranges between 5–9 per cent and 2–6 per cent and the mortality 87 per cent and 8–60 per cent, respectively.96 Such infections due to Aspergillus constitute the most common cause of infectious death post allogeneic transplantation with the lung being the most commonly affected organ.97,98
Graft versus host disease 1109
Risk factors for the development of fungal infections post transplant include neutropenia and lymphopenia, disease at diagnosis, previous fungal infection, GVHD and its treatment.97,98 The symptoms of infection are not specific to fungal infection. Sinusitis, pneumonia with pleuritic chest pain and haemoptysis should raise the suspicion of fungal infection. Radiological findings are also variable with frequent absence of ‘classical’ radiological signs. Consequently the diagnosis of fungal infection is mainly based on the clinical presentation and radiological findings.99 Aspergillus PCR and Galactomanan antigenemia in the blood and urine with enzyme-linked immunosorbent assay (ELISA) may be helpful for the early detection of infection but serial sampling is recommended for confirmation of probable or possible infections.100–103 The evolution and use of novel agents for anti-fungal prophylaxis and treatment have changed the incidence and the mortality related to these infections post transplant. Chemoprophylaxis with itraconazole starting just prior to the conditioning regimen is considered standard practice in many centres as it is superior to prophylaxis with fluconazole against Aspergillus.100,104 An ongoing multi-centre Phase III randomized trial comparing itraconazole and voriconazole prophylaxis is likely to improve prophylactic options against fungal infections in allogeneic stem cell transplant recipients. The development of new antifungal agents such as echinocandins, and new members of the azole family such as voriconazole and posaconazole appear promising in the treatment of fungal infections in high-risk patients.105–107 Cellular therapy with prophylactic granulocyte transfusions is shown to be beneficial in high-risk patients undergoing allogeneic SCT but further randomized trials are required.108
GRAFT VERSUS HOST DISEASE This represents the organ damage caused by the interaction of mature donor T-cells with recipient’s antigen representing cells (alloreactivity) and occurs as the result of histocompatibility differences between the host and the donor. The risk increases with recipient and donor age, the use of PBSCs and grafts from unrelated donors. Clinical manifestations of GVHD involve the skin, mucosal membranes, intestine, liver and lungs. This syndrome is a major complication of allogeneic stem cell transplant although it rarely occurs in the autologous setting.109 It is also one of the main limitations for the application of SCT in wider patient and age groups. Although acute GVHD constitutes a distinct clinical and physiological entity, chronic GVHD is defined in relation to the day of infusion of stem cells. Day 100 post stem cell infusion is conventionally used for the distinction between acute and chronic GVHD. Chronic GVHD simulates autoimmune disorders and is often considered an autoimmune disease.110
Acute GVHD is described to occur in three phases where infections, mucosal damage, host antigen-presenting cells, donor T-cells, cytotoxic lymphocytes and NK cells, and cytokine release contribute to the cascade leading to the development of the inflammatory manifestations recognized clinically as acute GVHD.110,111 The risk factors predisposing to the development of acute GVHD could be related to the donor or host, the source of the stem cells and the dose of stem cells112 (Table 45.7). According to the most commonly used grading system criteria (Glucksberg) there are four grades of acute GVHD (Table 45.8), which are directly associated to the clinical course. Grades II–III, usually require local or systemic treatment. Grades III and IV are potentially life threatening112,113 (Grade II can be life threatening via the effects of systemic treatment). Biopsies of the organs involved help to confirm the diagnosis and establish the severity.114 There is inconsistency in the grading of acute GVHD mainly due to the difficulty confirming the diagnosis and approaches to diagnostic investigations, procedures and interpretations. Revised criteria have been proposed in order to include important clinical parameters such as progression of disease through different lines of treatment, performance status, and the degree of immunosuppression required to control the course of disease.115,116 Strategies for GVHD prophylaxis with single or a combination immunosuppressants112 have been described briefly under the conditioning regimens in this chapter. In general a combination of daily IV cyclosporine A with IV methotrexate on days 1, 3, 6 and 11 decreases the incidence of GVHD and improves survival.117 No other regimen has subsequently shown superiority to this combination. In the treatment of acute GVHD glucocorticosteroids have been the most effective initial treatment. Methylprednisolone 2 mg/kg IV in two divided doses is the recommended initial dosing regimen.112,118,119
Table 45.7 Risk factors for the development of GVHD Donor/host HLA disparity between SC donor and recipient (major and minor HLA) Sex mismatching and donor parity Age (increased age of recipient and/or donor) Source of stem cells Fresh bone marrow PBSC Cord blood (less immunologically active) Stem cell dose 6–10 106/kg CD34 PBSC Immunomodulation Increased dose of TBI Seropositivity for herpes viruses (?)
1110 Introduction to haematopoietic stem cell transplantation
GVHD and its management cause profound immunodeficiency predisposing the patients to fatal infections during their course. Chronic GVHD can be limited or extensive. Usual sites are the skin, eyes, hepatobiliary system, gut and lung causing rash and sclerodermatous changes, keratoconjunctivitis, cholestasis, malabsorption, bronchiolitis and cytopenias, respectively.112 It resembles autoimmune disorders and is associated with loss of self-tolerance and causes
generalized immunosuppression.120 The treatment of chronic GVHD requires a prolonged (months to years) course of steroids.7 The development of severe acute and/or chronic acute GVHD significantly contributes to the morbidity, mortality and the cost of the stem cell transplant.121 There is interest in developing different treatments for GVHD with modification of donor T-cells and utilization of proteins responsible for the induction of lymphocyte
Table 45.8 Grading of acute graft vs. host disease (a) Grading system: stage for each organ Stage
Skin/maculo-papular rash
Liver/bilirubin
GI/diarrhoea
*
25% of body surface
35–50 mmol/L
500 mL
**
25–50% of body surface
51–102 mmol/L
1000 mL
***
Generalized erythroderma
103–255 mmol/L
1500 mL
****
Generalized erythroderma with bullous formation and desquamation
255 mmol/L
Severe abdominal pain with or without ileus
(b) Overall grading system (Glucksberg) Grade of AGvHD
Degree of organ involvement
I
Skin to
II
Skin to Gut and/or liver Mild decrease in clinical performance
III
Skin to Gut and/or liver to Marked decrease in clinical performance
IV
Skin to Gut and/or liver to Extreme decrease in clinical performance
(c) Consensus conference on AG-v-HD grading Stage 1 2 3 4 Functional Grading I II III–IV
Skin rash
Liver (Bilirubin)
Gut (diarrhoea/nausea)
25% 25–50% 50% Erythroderma
34–50 mmol/L 51–102 mmol/L 103–255 mmol/L 255 mmol/L
500 mL or nausea 1000 mL 1500 mL Pain/ileus
Skin
Liver
Gut
Stage 1–2 Stage 3 or Stage 4 or
0 Stage 1 or Stage 2–4 or
0 Stage 1 Stage 2–4
Transplant programme requirements and accreditation 1111
apoptosis. Transduced donors’ T-cells express viral genes responsible for the conversion of drugs into cytotoxic products (suicide genes). These cells can be eliminated upon development of severe GVHD.122,123
The cumulative incidence of solid tumours post SCT is 5 per cent in 10 years but may be higher thereafter.131,132 Secondary malignancies can develop post autologous and allogeneic SCTs and include melanoma, PTLD and rarely donor leukaemia.
Late complications Owing to the increasing number of transplants and improved post-transplant care there has been an increasing population of SCT survivors. Consequently there is a significant risk for the development of long-term complications related to SCT in this patient group.124 Such complications of a malignant or non-malignant nature can seriously impact on the recipient’s quality of life (Table 45.9). Late complications also develop in recipients of highdose therapy with or without transplantation. It is not possible to distinguish the influence of the transplant-related factors from the genetic predisposition, the original disease or the effects of previous treatment in the development of late complications. The potential development of late effects related to SCT should be discussed with the patient at their initial counselling for transplant. Long-term follow-up of transplant patients aims for the prevention and early management of such complications. Secondary myelodysplastic syndrome or acute myeloid leukaemia occur post treatment (chemotherapy with or without radiotherapy) for non-Hodgkin’s lymphoma, Hodgkin’s disease and solid tumours.125,126 They can also occur after autologous transplantation.127,128 The treatment of such conditions is difficult as chemotherapy is not well tolerated and remissions when achieved are short lived.129 Exposure to alkylating agents or topoisomerase inhibitors and radiotherapy for primary treatment or stem cell mobilization should be minimized and the duration of pre-transplant treatment should be limited accordingly in candidates for such procedures.130 Table 45.9 Non-malignant late complications of allogeneic SCT Organ affected Eye Skin
Lung Heart Bone Endocrine
Common complications Cataracts; Sicca syndrome Contractures; Sclerodermatous changes; Discoloration COPD; Restrictive lung disease CCF; Cardiomyopathy Osteoporosis; Avascular necrosis Hypothyroidism; Gonadal failure; Infertility; Growth deficiency
TRANSPLANT PROGRAMME REQUIREMENTS AND ACCREDITATION Consideration for SCT requires careful clinical evaluation and counselling by those dedicated and experienced in the field multidisciplinary team. Information from the referring team should include the diagnosis, diagnostic studies and their results, preceding treatments and response, prior course and psychological background. Prior to the initial visit to the transplant centre and meeting the medical team the transplant candidates should have received written information on the principles, rationale, complications and outcome of the transplant. The candidates should be offered the opportunity to meet and speak to previous transplant recipients. The initial visit has two main purposes: 1. examination, clinical and social data collection to identify patients as potential candidates for SCT. 2. to explain to the patient and the relevant family members the rationale and biological basis of the transplant, potential complications and outcomes. Alternative therapies and their possible outcome should be discussed.133–135 A subsequent second meeting should take place later in time to allow the transplant candidate and their family time for discussion, debate and informed decision. Factors influencing the outcome are considered. Disease status at transplant, performance status, co-morbid factors (cardiac, pulmonary, renal or hepatic problems), donor/recipient compatibility, age of patient/donor and previous infections are amongst the main ones. Supportive care, admission in specialized transplant units with strict infection control protocols and the development of new conditioning regimens, have significantly contributed to the reduction of transplant-related mortality. Awareness of symptoms related to possible complications at any stage during the post-transplant period are fundamental for the care of this patient group. SCT is a specialist area of haematological oncology therefore additional training of physicians and nurses is required to optimize outcome.136 The management of the transplant programme is as complicated as the care of its patients.121 The programme consists of stem cell mobilization and collection facilities, stem cell processing/storage laboratory and clinical programme/facilities. Interaction and communication between the different parts of the programme and their members should be optimal to ensure seamless and integral patient care and success of the programme.
1112 Introduction to haematopoietic stem cell transplantation
Quality control systems have been applied to the support of the transplant programme. There are increasingly more regulatory bodies involved. Until recently these regulatory bodies offered accreditation for individual parts of the programme. The Joint Accreditation Committee for ISCT and EBMT (JACIE) aims to offer accreditation for all three parts of the programme in units that comply with SCT standards. This accreditation aims to apply quality control and unify the transplant patient’s care across Europe.137
●
●
CONCLUSION Over the last few decades SCT has followed a distinguished route of continuous development and improving results. The ability to obtain and transplant grafts of different composition, the role of cytokines in mobilization expansion and differentiation of stem cells, the improvements in HLA typing, new anti-infective agents and the introduction of reduced intensity conditioning regimens are some of these developments. A continued improved understanding of cell growth and differentiation, stem cell plasticity, cell and gene regulation of immune recognition and immunoregulation to enhance GVT effect will drive the field forward in the next few decades.
●
●
KEY LEARNING POINTS ●
●
●
●
Haematopoietic stem cell transplantation (HSCT) is an established curative treatment option for a number of haematological and lymphoid malignancies and other disorders. HSCT results in more cures and remissions than alternative treatments in expense of greater morbidity and mortality. The curative potential of HSCT is based on the ability to eradicate disease by administration of high doses of chemo-/radiotherapy and the use of the immunotherapeutic potential of the allogeneic graft through the graft vs. tumour effect. Different kinds of transplants are applicable to different diseases and recipient groups. Transplants are identified according to the source of stem cells, the availability of stem cell donors and the conditioning regime. Cure, mortality and disease relapse rates vary with the type of transplant, the status of disease at transplant and the patient’s eligibility. The objective of the preparative regimen prior to HSCT is both to eradicate the disease and to induce immune suppression to allow engraftment and augmentation of anti-tumour immune response. Efforts are being made to understand the factors enhancing the graft vs.
tumour effect and the effective use of adaptive immunotherapy. The composition of the grafts is being manipulated to accentuate this effect and minimize toxicity. Transplants with lower intensity have been developed. Although they have lower mortality, long-term data regarding their efficacy are still awaited. Complications of transplant (early and late) acco-unt for the procedure’s morbidity and mortality. Most of these complications are related to the toxicity of the conditioning regimen, infections and GVHD. Viral and fungal infections are major stumbling blocks but early diagnosis, newer agents and efforts to improve immune reconstitution are starting to impact on outcome. Despite the improved outcomes of the recent years these continue to constitute the greatest restrictions for the wider application of HSCT. Improved therapy and prevention of GVHD and more emphasis on quality of life will result in better application of HSCT. Understanding of stem cell growth, differentiation and plasticity, cell and gene regulation of immune recognition and immunoregulation to enhance GVT with targeted donor lymphocyte infusions and the potential utilization of embryonic stem cells as source for HSCT hold a great promise for the future of the field.
REFERENCES ●1
●2
◆3
◆4
◆5
Kurnick NB, Montano A, Gerdes JC, Feder BH. Preliminary observations on the treatment of postirradiation hematopoietic depression in man by the infusion of stored autogenous bone marrow. Ann Intern Med 1958; 49:973–986. Thomas ED, Lochte HL Jr, Cannon JH, Sahler OD, Ferrebee JW. Supralethal whole body irradiation and isologous marrow transplantation in man. J Clin Invest 1959; 38:1709–1716. Copelan EA. Hematopoietic stem-cell transplantation. N Engl J Med 2006; 354:1813–1826. Anderson D, DeFor T, Burns L, et al. A comparison of related donor peripheral blood and bone marrow transplants: importance of late-onset chronic graft-versus-host disease and infections. Biol Blood Marrow Transplant 2003; 9:52–59. Pavletic SZ, Smith LM, Bishop MR, et al. Prognostic factors of chronic graft-versus-host disease after allogeneic blood stem-cell transplantation. Am J Hematol 2005; 78:265–274.
References 1113
6 Cutler C, Giri S, Jeyapalan S, Paniagua D, Viswanathan A, Antin JH. Acute and chronic graft-versus-host disease after allogeneic peripheral-blood stem-cell and bone marrow transplantation: a meta-analysis. J Clin Oncol 2001; 19:3685–3691. 7 Stewart BL, Storer B, Storek J, et al. Duration of immunosuppressive treatment for chronic graft-versus-host disease. Blood 2004; 104:3501–3506. ●8 Gluckman E, Rocha V, Boyer-Chammard A, et al. Outcome of cord-blood transplantation from related and unrelated donors. Eurocord Transplant Group and the European Blood and Marrow Transplantation Group. N Engl J Med 1997; 337:373–381. ●9 Rocha V, Labopin M, Sanz G, et al. Transplants of umbilicalcord blood or bone marrow from unrelated donors in adults with acute leukemia. N Engl J Med 2004; 351:2276–2285. 10 Rocha V, Wagner JE Jr, Sobocinski KA, et al. Graft-versus-host disease in children who have received a cord-blood or bone marrow transplant from an HLA-identical sibling. Eurocord and International Bone Marrow Transplant Registry Working Committee on Alternative Donor and Stem Cell Sources. N Engl J Med 2000; 342:1846–1854. 11 Wagner JE, Rosenthal J, Sweetman R, et al. Successful transplantation of HLA-matched and HLA-mismatched umbilical cord blood from unrelated donors: analysis of engraftment and acute graft-versus-host disease. Blood 1996; 88:795–802. ◆12 Petropoulos D, Chan KW. Umbilical cord blood transplantation. Curr Oncol Rep 2005; 7:406–409. ❊13 Jaroscak J, Goltry K, Smith A, et al. Augmentation of umbilical cord blood (UCB) transplantation with ex vivoexpanded UCB cells: results of a phase 1 trial using the AastromReplicell System. Blood 2003; 101:5061–5067. ●14 Ruggeri L, Capanni M, Urbani E, et al. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 2002; 295:2097–2100. 15 Ruggeri L, Capanni M, Mancusi A, et al. Natural killer cell alloreactivity in haploidentical hematopoietic stem cell transplantation. Int J Hematol 2005; 81:13–17. 16 Huang XJ, Liu DH, Liu KY, et al. Haploidentical hematopoietic stem cell transplantation without in vitro T-cell depletion for the treatment of hematological malignancies. Bone Marrow Transplant 2006; 38:291–297. 17 Kottaridis PD, Milligan DW, Chopra R, et al. In vivo CAMPATH-1H prevents graft-versus-host disease following nonmyeloablative stem cell transplantation. Blood 2000; 96:2419–2425. 18 Tholouli E, Liakopoulou E, Greenfield HM, et al. Outcome following 50 mg versus 100 mg alemtuzumab in reducedintensity conditioning stem cell transplants for AML and poor risk MDS. EBMT 2006, Hamburg, Germany, March 19–22. Abstracts. Bone Marrow Transplant 2006; 37 (Suppl 1):S48. 2006. 19 Petersdorf EW, Gooley TA, Anasetti C, et al. Optimizing outcome after unrelated marrow transplantation by comprehensive matching of HLA class I and II alleles in the donor and recipient. Blood 1998; 92:3515–3520.
◆20
❊21
●22
◆23
◆24
25
26
27
◆28
◆29
❊30
❊31
◆32
❊33
34
Petersdorf EW, Anasetti C, Martin PJ, Hansen JA. Tissue typing in support of unrelated hematopoietic cell transplantation. Tissue Antigens 2003; 61:1–11. Morishima Y, Sasazuki T, Inoko H, et al. The clinical significance of human leukocyte antigen (HLA) allele compatibility in patients receiving a marrow transplant from serologically HLA-A, HLA-B, and HLA-DR matched unrelated donors. Blood 2002; 99:4200–4206. Osawa M, Hanada K, Hamada H, Nakauchi H. Long-term lymphohematopoietic reconstitution by a single CD34low/negative hematopoietic stem cell. Science 1996; 273:242–245. Serke S, Johnsen HE. A European reference protocol for quality assessment and clinical validation of autologous haematopoietic blood progenitor and stem cell grafts. Bone Marrow Transplant 2001; 27:463–470. Soiffer RJ. T-cell depletion to prevent graft-vs.-host disease. In: Blume KG, Appelbaum FR, Forman SJ, Thomas ED (eds) Thomas’ Hematopoietic Cell Transplantation. Malden, MA; Oxford: Blackwell Publishing, 2004, 221–233. Voogt PJ, Fibbe WE, Marijt WA, et al. Rejection of bonemarrow graft by recipient-derived cytotoxic T lymphocytes against minor histocompatibility antigens. Lancet 1990; 335:131–134. Fleischhauer K, Kernan NA, O’Reilly RJ, Dupont B, Yang SY. Bone marrow-allograft rejection by T lymphocytes recognizing a single amino acid difference in HLA-B44. N Engl J Med 1990; 323:1818–1822. Donohue J, Homge M, Kernan NA. Characterization of cells emerging at the time of graft failure after bone marrow transplantation from an unrelated marrow donor. Blood 1993; 82:1023–1029. Klein J, Sato A. The HLA system. First of two parts. N Engl J Med 2000; 343:702–709. Marsh SG, Albert ED, Bodmer WF, et al. Nomenclature for factors of the HLA system, 2002. Tissue Antigens 2002; 60:407–464. Tiercy JM, Villard J, Roosnek E. Selection of unrelated bone marrow donors by serology, molecular typing and cellular assays. Transpl Immunol 2002; 10:215–221. Speiser DE, Tiercy JM, Rufer N, et al. High resolution HLA matching associated with decreased mortality after unrelated bone marrow transplantation. Blood 1996; 87:4455–4462. Hansen JA, Yamamoto K, Petersdorf E, Sasazuki T. The role of HLA matching in hematopoietic cell transplantation. Rev Immunogenet 1999; 1:359–373. Duarte RF, Pamphilon D, Cornish J, C et al. Topical issues in unrelated donor haematopoietic stem cell transplants: a report from a workshop convened by the Anthony Nolan Trust in London – 2005. Bone Marrow Transplant 2006; 37:901–908. Horowitz MMEd. State of the Art in Transplantation. International Bone Marrow Transplantation Registry/Autologous Blood and Marrow Transplant Registry. IBMTR/ABMTR slide set. 2002.
1114 Introduction to haematopoietic stem cell transplantation
❊35
❊36
❊37
❊38
❊39
❊40
41
42
43
44
45
46
Frassoni F. Stem cell transplantation in adults – Acute leukaemia. In: Apperley JF, Carreras E, Gluckman E, Gratwohl A, Masszi T (eds) The EBMT Handbook: Haemopoietic Stem Cell Transplantation. Genoa: European School of Haematology, Forum Service Editore, 2004: 238–255. Niederwieser D. Stem cell transplantation in adults – Chronic disorders. In: Apperley JF, Carreras E, Gluckman E, Gratwohl A, Masszi T, editors. The EBMT Handbook: Haemopoietic Stem Cell Transplantation. Genoa: European School of Haematology, Forum Service Editore, 2004: 256–267. Schmitz N. Stem cell transplantation in adults – Lymphomas. In: Apperley JF, Carreras E, Gluckman E, Gratwohl A, Masszi T, editors. The EBMT Handbook: Haemopoietic Stem Cell Transplantation. Genoa: European School of Haematology, Forum Service Editore, 2004: 268–275. Demirer T. Stem cell transplantation in adults – Solid tumours. In: Apperley JF, Carreras E, Gluckman E, Gratwohl A, Masszi T, editors. The EBMT Handbook: Haemopoietic Stem Cell Transplantation. Genoa: European School of Haematology, Forum Service Editore, 2004: 276–279. Tyndall A. Stem cell transplantation in adults – Autoimmune diseases. In: Apperley JF, Carreras E, Gluckman E, Gratwohl A, Masszi T (eds) The EBMT Handbook: Haemopoietic Stem Cell Transplantation. Genoa: European School of Haematology, Forum Service Editore, 2004, 280–285. Schrezenmeier H, Passweg J, Bacigalupo A. Stem cell transplantation in adults – Chronic disorders. In: Apperley JF, Carreras E, Gluckman E, Gratwohl A, Masszi T (eds) The EBMT Handbook: Haemopoietic Stem Cell Transplantation. Genoa: European School of Haematology, Forum Service Editore, 2004, 286–293. Paivanas T. New center provides resources for large transplant-related studies. Oncology News International 2005; 14:51–52, 73. Deeg HJ, Sullivan KM, Buckner CD, et al. Marrow transplantation for acute nonlymphoblastic leukemia in first remission: toxicity and long-term follow-up of patients conditioned with single dose or fractionated total body irradiation. Bone Marrow Transplant 1986; 1:151–157. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with acute myeloid leukemia in first remission: a randomized trial of two irradiation regimens. Blood 1990; 76:1867–1871. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with chronic myeloid leukemia in the chronic phase: a randomized trial of two irradiation regimens. Blood 1991; 77:1660–1665. Clift RA, Buckner CD, Thomas ED, et al. Marrow transplantation for chronic myeloid leukemia: a randomized study comparing cyclophosphamide and total body irradiation with busulfan and cyclophosphamide. Blood 1994; 84:2036–2043. Devergie A, Blaise D, Attal M, et al. Allogeneic bone marrow transplantation for chronic myeloid leukemia in first chronic phase: a randomized trial of busulfan-cytoxan versus
47
◆48
49
50
51
◆52
53
54
55
❊56
57
58
cytoxan-total body irradiation as preparative regimen: a report from the French Society of Bone Marrow Graft (SFGM). Blood 1995; 85:2263–2268. Blaise D, Maraninchi D, Archimbaud E, et al. Allogeneic bone marrow transplantation for acute myeloid leukemia in first remission: a randomized trial of a busulfan-Cytoxan versus Cytoxan-total body irradiation as preparative regimen: a report from the Group d’Etudes de la Greffe de Moelle Osseuse. Blood 1992; 79:2578–2582. Banna GL, Aversa S, Sileni VC, Favaretto A, Ghiotto C, Monfardini S. Nonmyeloablative allogeneic stem cell transplantation (NST) after truly nonmyeloablative and reduced intensity conditioning regimens. Crit Rev Oncol Hematol 2004; 51:171–189. Collins RH Jr, Shpilberg O, Drobyski WR, et al. Donor leukocyte infusions in 140 patients with relapsed malignancy after allogeneic bone marrow transplantation. J Clin Oncol 1997; 15:433–444. Giralt S, Estey E, Albitar M, et al. Engraftment of allogeneic hematopoietic progenitor cells with purine analog-containing chemotherapy: harnessing graft-versus-leukemia without myeloablative therapy. Blood 1997; 89:4531–4536. Slavin S, Nagler A, Naparstek E, et al. Nonmyeloablative stem cell transplantation and cell therapy as an alternative to conventional bone marrow transplantation with lethal cytoreduction for the treatment of malignant and nonmalignant hematologic diseases. Blood 1998; 91:756–763. Haddad N, Rowe JM. Current indications for reducedintensity allogeneic stem cell transplantation. Best Pract Res Clin Haematol 2004; 17:377–86. Ueno NT, Cheng YC, Rondon G, et al. Rapid induction of complete donor chimerism by the use of a reduced-intensity conditioning regimen composed of fludarabine and melphalan in allogeneic stem cell transplantation for metastatic solid tumors. Blood 2003; 102:3829–36. Sayer HG, Kroger M, Beyer J, et al. Reduced intensity conditioning for allogeneic hematopoietic stem cell transplantation in patients with acute myeloid leukemia: disease status by marrow blasts is the strongest prognostic factor. Bone Marrow Transplant 2003; 31:1089–95. Kroger N, Bornhauser M, Ehninger G, et al. Allogeneic stem cell transplantation after a fludarabine/busulfan-based reduced-intensity conditioning in patients with myelodysplastic syndrome or secondary acute myeloid leukemia. Ann Hematol 2003; 82:336–42. Tauro S, Craddock C, Peggs K, et al. Allogeneic stem-cell transplantation using a reduced-intensity conditioning regimen has the capacity to produce durable remissions and long-term disease-free survival in patients with high-risk acute myeloid leukemia and myelodysplasia. J Clin Oncol 2005; 23:9387–93. Subira M, Sureda A, Ancin I, et al. Allogeneic stem cell transplantation with reduced-intensity conditioning is potentially feasible as an outpatient procedure. Bone Marrow Transplant 2003; 32:869–72. Lawler M. Prospective chimerism analysis, the time is now but can we respond? Leukemia 2001; 15:1992–4.
References 1115
◆59
60
61
62
63
❊❊64
❊
65
❊◆66
❊67
◆68
69
70
71
◆72
73
Bader P, Niethammer D, Willasch A, Kreyenberg H, Klingebiel T. How and when should we monitor chimerism after allogeneic stem cell transplantation? Bone Marrow Transplant 2005; 35:107–19. Peggs KS, Thomson K, Hart DP, et al. Dose-escalated donor lymphocyte infusions following reduced intensity transplantation: toxicity, chimerism, and disease responses. Blood 2004; 103:1548–56. McCann SR, Gately K, Conneally E, Lawler M. Molecular response to imatinib mesylate following relapse after allogeneic SCT for CML. Blood 2003; 101:1200–01. Fuge R, Bird JM, Fraser A, et al. The clinical features, risk factors and outcome of thrombotic thrombocytopenic purpura occurring after bone marrow transplantation. Br J Haematol 2001; 113:58–64. Leung AY, Yuen KY, Kwong YL. Polyoma BK virus and haemorrhagic cystitis in haematopoietic stem cell transplantation: a changing paradigm. Bone Marrow Transplant 2005; 36:929–37. Spielberger R, Stiff P, Bensinger W, et al. Palifermin for oral mucositis after intensive therapy for hematologic cancers. N Engl J Med 2004; 351:2590–8. Blazar BR, Weisdorf DJ, DeFor T, et al. A phase I/II randomized, placebo-control trial of Palifermin to prevent graft-versus-host disease (GVHD) after allogeneic hematopoietic stem cell transplantation (HSCT). Blood 2006. Einhorn LH, Rapoport B, Koeller J, et al. Antiemetic therapy for multiple-day chemotherapy and high-dose chemotherapy with stem cell transplant: review and consensus statement. Support Care Cancer 2005; 13:112–16. Cox GJ, Matsui SM, Lo RS, et al. Etiology and outcome of diarrhea after marrow transplantation: a prospective study. Gastroenterology 1994; 107:1398–1407. Strasser SI, McDonald GB. Hepatobiliary complications of hematopoietic stem cell transplantation. In: Schiff ER, Sorrell MF, Maddrey WC (eds) Schiff’s Diseases of the Liver. Philadelphia: Lippincott-Raven Publishers, 1999, 1636–63. McDonald GB, Hinds MS, Fisher LD, et al. Veno-occlusive disease of the liver and multiorgan failure after bone marrow transplantation: a cohort study of 355 patients. Ann Intern Med 1993; 118:255–67. Carreras E, Bertz H, Arcese W, et al. Incidence and outcome of hepatic veno-occlusive disease after blood or marrow transplantation: a prospective cohort study. European Group for Blood and Marrow Transplantation Chronic Leukemia Working Party. Blood 1998; 92:3599–604. Lee JL, Gooley T, Bensinger W, Schiffman K, McDonald GB. Veno-occlusive disease of the liver after busulfan, melphalan, and thiotepa conditioning therapy: incidence, risk factors, and outcome. Biol Blood Marrow Transplant 1999; 5:306–15. DeLeve LD, Shulman HM, McDonald GB. Toxic injury to hepatic sinusoids: sinusoidal obstruction syndrome (venoocclusive disease). Semin Liver Dis 2002; 22:27–42. Richardson PG, Murakami C, Jin Z, et al. Multi-institutional use of defibrotide in 88 patients after stem cell transplantation with severe veno-occlusive disease and
◆74
◆75
◆76
◆77
78
❊79
80
❊81
◆82
◆83
84
85
86
87
88
multisystem organ failure: response without significant toxicity in a high-risk population and factors predictive of outcome. Blood 2002; 100:4337–43. Crawford SW. Critical care and respiratory failure. In: Forman SJ, Blume KG, Thomas ED (eds) Hematopoietic Cell Transplantation. Oxford: Blackwell Science, 1999, 712–22. Shanholtz C. Respiratory complications of blood and marrow transplantation. Clin Pulm Med 1999; 6:254–62. Krowka MJ, Rosenow EC III, Hoagland HC. Pulmonary complications of bone marrow transplantation. Chest 1985; 87:237–46. Horak DA. Pulmonary complications after hematopoietic cell transplantation. In: Blume KG, Appelbaum FR, Forman SJ, Thomas ED (eds) Thomas’ Hematopoietic Cell Transplantation. Malden: Malden, MA; Oxford: Blackwell Publishing, 2004, 873–82. Bach PB, Schrag D, Nierman DM, et al. Identification of poor prognostic features among patients requiring mechanical ventilation after hematopoietic stem cell transplantation. Blood 2001; 98:3234–40. Groeger JS, White P Jr, Nierman DM, et al. Outcome for cancer patients requiring mechanical ventilation. J Clin Oncol 1999; 17:991–7. Junghanss C, Marr KA, Carter RA, et al. Incidence and outcome of bacterial and fungal infections following nonmyeloablative compared with myeloablative allogeneic hematopoietic stem cell transplantation: a matched control study. Biol Blood Marrow Transplant 2002; 8:512–20. Hughes WT, Armstrong D, Bodey GP, et al. 2002 guidelines for the use of antimicrobial agents in neutropenic patients with cancer. Clin Infect Dis 2002; 34:730–51. Gratwohl A, Brand R, Frassoni F, et al. Cause of death after allogeneic haematopoietic stem cell transplantation (HSCT) in early leukaemias: an EBMT analysis of lethal infectious complications and changes over calendar time. Bone Marrow Transplant 2005; 36:757–69. Winston DJ, Ho WG, Champlin RE. Cytomegalovirus infections after allogeneic bone marrow transplantation. Rev Infect Dis 1990; 12 Suppl 7:S776–92. Meyers JD, Flournoy N, Thomas ED. Infection with herpes simplex virus and cell-mediated immunity after marrow transplant. J Infect Dis 1980; 142:338–46. Zutter MM, Martin PJ, Sale GE, et al. Epstein-Barr virus lymphoproliferation after bone marrow transplantation. Blood 1988; 72:520–9. Bruno B, Gooley T, Hackman RC, Davis C, Corey L, Boeckh M. Adenovirus infection in hematopoietic stem cell transplantation: effect of ganciclovir and impact on survival. Biol Blood Marrow Transplant 2003; 9:341–52. Liakopoulou E, Mutton K, Carrington D, et al. Rotavirus as a significant cause of prolonged diarrhoeal illness and morbidity following allogeneic bone marrow transplantation. Bone Marrow Transplant 2005; 36:691–4. Whimbey E, Champlin RE, Couch RB, et al. Community respiratory virus infections among hospitalized adult bone marrow transplant recipients. Clin Infect Dis 1996; 22:778–82.
1116 Introduction to haematopoietic stem cell transplantation
❊89
◆90
91
92
●93
❊94
❊95
◆96
◆97
98
❊99
100
❊101
❊102
103
Einsele H, Ehninger G, Hebart H, et al. Polymerase chain reaction monitoring reduces the incidence of cytomegalovirus disease and the duration and side effects of antiviral therapy after bone marrow transplantation. Blood 1995; 86:2815–20. Griffiths PD, Emery VC. Cytomegalovirus. In: Richman DD, Whitley RJ, Hayden FG, editors. Clinical Virology. Washington, D.C: ASM Press, 2002, 433–61. Junghanss C, Boeckh M, Carter RA, et al. Incidence and outcome of cytomegalovirus infections following nonmyeloablative compared with myeloablative allogeneic stem cell transplantation, a matched control study. Blood 2002; 99:1978–85. Holmberg LA, Boeckh M, Hooper H, et al. Increased incidence of cytomegalovirus disease after autologous CD34-selected peripheral blood stem cell transplantation. Blood 1999; 94:4029–35. Riddell SR, Watanabe KS, Goodrich JM, Li CR, Agha ME, Greenberg PD. Restoration of viral immunity in immunodeficient humans by the adoptive transfer of T cell clones. Science 1992; 257:238–41. Walter EA, Greenberg PD, Gilbert MJ, et al. Reconstitution of cellular immunity against cytomegalovirus in recipients of allogeneic bone marrow by transfer of T-cell clones from the donor. N Engl J Med 1995; 333:1038–44. Einsele H, Roosnek E, Rufer N, et al. Infusion of cytomegalovirus (CMV)-specific T cells for the treatment of CMV infection not responding to antiviral chemotherapy. Blood 2002; 99:3916–22. Lin SJ, Schranz J, Teutsch SM. Aspergillosis case-fatality rate: systematic review of the literature. Clin Infect Dis 2001; 32:358–66. Marr KA, Carter RA, Crippa F, Wald A, Corey L. Epidemiology and outcome of mould infections in hematopoietic stem cell transplant recipients. Clin Infect Dis 2002; 34:909–17. Offner F, Cordonnier C, Ljungman P, et al. Impact of previous aspergillosis on the outcome of bone marrow transplantation. Clin Infect Dis 1998; 26:1098–103. Ascioglu S, Rex JH, de Pauw B, et al. Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and hematopoietic stem cell transplants: an international consensus. Clin Infect Dis 2002; 34:7–14. Williamson EC, Oliver DA, Johnson EM, Foot AB, Marks DI, Warnock DW. Aspergillus antigen testing in bone marrow transplant recipients. J Clin Pathol 2000; 53:362–6. Einsele H, Hebart H, Roller G, et al. Detection and identification of fungal pathogens in blood by using molecular probes. J Clin Microbiol 1997; 35:1353–60. Hebart H, Loffler J, Reitze H, et al. Prospective screening by a panfungal polymerase chain reaction assay in patients at risk for fungal infections: implications for the management of febrile neutropenia. Br J Haematol 2000; 111:635–40. Williamson EC, Leeming JP, Palmer HM, et al. Diagnosis of invasive aspergillosis in bone marrow transplant recipients
104
❊❊105
◆106
◆107
108
109
◆110
111
112
◆113
❊114
115
116
❊117
❊118
by polymerase chain reaction. Br J Haematol 2000; 108:132–9. Parkman R. Clonal analysis of murine graft-vs-host disease. I. Phenotypic and functional analysis of T lymphocyte clones. J Immunol 1986; 136:3543–48. Walsh TJ, Pappas P, Winston DJ, et al. Voriconazole compared with liposomal amphotericin B for empirical antifungal therapy in patients with neutropenia and persistent fever. N Engl J Med 2002; 346: 225–34. van Burik JA. Role of new antifungal agents in prophylaxis of mycoses in high risk patients. Curr Opin Infect Dis 2005; 18:479–83. Strasfeld L, Weinstock DM. Antifungal prophylaxis among allogeneic hematopoietic stem cell transplant recipients: current issues and new agents. Expert Rev Anti Infect Ther 2006; 4:457–68. Kerr JP, Liakopolou E, Brown J, et al. The use of stimulated granulocyte transfusions to prevent recurrence of past severe infections after allogeneic stem cell transplantation. Br J Haematol 2003; 123:114–8. Hood AF, Vogelsang GB, Black LP, Farmer ER, Santos GW. Acute graft-vs-host disease. Development following autologous and syngeneic bone marrow transplantation. Arch Dermatol 1987; 123:745–50. Reddy P, Ferrara JL. Immunobiology of acute graft-versus-host disease. Blood Rev 2003; 17: 187–94. Klingemann HG, Storb R, Fefer A, et al. Bone marrow transplantation in patients aged 45 years and older. Blood 1986; 67:770–6. Glucksberg H, Storb R, Fefer A, et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HLA-matched sibling donors. Transplantation 1974; 18:295–304. Vogelsang GB, Hess AD, Santos GW. Acute graft-versus-host disease: clinical characteristics in the cyclosporine era. Medicine (Baltimore) 1988; 67:163–74. Rowlings PA, Przepiorka D, Klein JP, et al. IBMTR Severity Index for grading acute graft-versus-host disease: retrospective comparison with Glucksberg grade. Br J Haematol 1997; 97:855–64. Al Ghamdi H, Leisenring W, Bensinger WI, et al. A proposed objective way to assess results of randomized prospective clinical trials with acute graft-versus-host disease as an outcome of interest. Br J Haematol 2001; 113:461–9. Storb R, Deeg HJ, Pepe M, et al. Methotrexate and cyclosporine versus cyclosporine alone for prophylaxis of graft-versus-host disease in patients given HLA-identical marrow grafts for leukemia: long-term follow-up of a controlled trial. Blood 1989; 73:1729–34. Goker H, Haznedaroglu IC, Chao NJ. Acute graft-vs-host disease: pathobiology and management. Exp Hematol 2001; 29:259–77. Deeg HJ, Henslee-Downey PJ. Management of acute graft-versus-host disease. Bone Marrow Transplant 1990; 6:1–8.
References 1117
119 Van Lint MT, Uderzo C, Locasciulli A, et al. Early treatment of acute graft-versus-host disease with high- or low-dose 6-methylprednisolone: a multicenter randomized trial from the Italian Group for Bone Marrow Transplantation. Blood 1998; 92:2288–93. 120 Tivol E, Komorowski R, Drobyski WR. Emergent autoimmunity in graft-versus-host disease. Blood 2005; 105:4885–91. 121 Lee SJ, Klar N, Weeks JC, Antin JH. Predicting costs of stem-cell transplantation. J Clin Oncol 2000; 18:64–71. ●122 Bonini C, Ferrari G, Verzeletti S, et al. HSV-TK gene transfer into donor lymphocytes for control of allogeneic graftversus-leukemia. Science 1997; 276:1719–24. 123 Berger C, Blau CA, Clackson T, Riddell SR, Heimfeld S. CD28 costimulation and immunoaffinity-based selection efficiently generate primary gene-modified T cells for adoptive immunotherapy. Blood 2003; 101:476–484. ◆124 Thirman MJ, Larson RA. Therapy-related myeloid leukemia. Hematol Oncol Clin North Am 1996; 10:293–320. 125 Travis LB, Weeks J, Curtis RE, et al. Leukemia following low-dose total body irradiation and chemotherapy for nonHodgkin’s lymphoma. J Clin Oncol 1996; 14:565–571. 126 Kumar L. Secondary leukaemia after autologous bone marrow transplantation. Lancet 1995; 345:810. 127 Stone RM. Myelodysplastic syndrome after autologous transplantation for lymphoma: the price of progress. Blood 1994; 83:3437–40. 128 Witherspoon RP, Deeg HJ, Storer B, Anasetti C, Storb R, Appelbaum FR. Hematopoietic stem-cell transplantation for treatment-related leukemia or myelodysplasia. J Clin Oncol 2001; 19:2134–41. 129 Govindarajan R, Jagannath S, Flick JT, et al. Preceding standard therapy is the likely cause of MDS after autotransplants for multiple myeloma. Br J Haematol 1996; 95:349–53.
130 Bhatia S, Ramsay NK, Steinbuch M, et al. Malignant neoplasms following bone marrow transplantation. Blood 1996; 87:3633–39. 131 Kolb HJ, Duell T, Socie G, et al. New malignancies in patients surviving more than 5 years after marrow transplantation. Blood 1995; 86:460a. 132 Deeg HJ, Socie G. Malignancies after hematopoietic stem cell transplantation: many questions, some answers. Blood 1998; 91:1833–44. ◆133 Blume KG, Amylon MD. The evaluation and counselling of candidates for hematopoietic cell transplantation. In: Blume KG, Appelbaum FR, Forman SJ, Thomas ED (eds) Thomas’ Hematopoietic Cell Transplantation. Malden: Malden, MA; Oxford: Blackwell Publishing, 2004, 449–62. 134 Lee SJ, Fairclough D, Antin JH, Weeks JC. Discrepancies between patient and physician estimates for the success of stem cell transplantation. JAMA 2001; 285:1034–38. 135 Molassiotis A. Further evaluation of a scale to screen for risk of emotional difficulties in bone marrow transplant recipients. J Adv Nurs 1999; 29:922–7. 136 Wagner ND, Quinones VW. Allogeneic peripheral blood stem cell transplantation: clinical overview and nursing implications. Oncol Nurs Forum 1998; 25:1049–55. ◆137 Urbano Ispizua A. Requirements for a clinical BMT unit. In: Apperley JF, Carreras E, Gluckman E, Gratwohl A, Masszi T (eds) The EBMT Handbook: Haemopoietic Stem Cell Transplantation. Genoa: European School of Haematology, Forum Service Editore, 2004, 66–77. ◆138 Pasquini M. Report on state of the art in blood and marrow transplantation – Part I – CIBMTR Summary Slides, 2005. CIBMTR Newsletter 2006; 12:5–7. ❊139 Guidelines for preventing opportunistic infections among hematopoietic stem cell transplant recipients. MMWR Recomm Rep 2000; 49:1–7.
46 Paediatric oncology STEPHEN LOWIS, EDDY ESTLIN AND KEITH SIBSON
Introduction Epidemiology Specific paediatric malignant diseases CNS tumours in infants and young children Supportive care Conclusion
1118 1119 1121 1142 1148 1149
INTRODUCTION Children with malignant disease are numerically few, representing perhaps one per cent of all patients. Childhood cancer is, however, the second most common cause of death in childhood, and many children cured of their disease are left with significant problems that have effects throughout their lives. A diagnosis of malignancy affects parents, siblings and the extended family more directly than for an adult patient, and for this reason, care for the family unit has been central to paediatric oncology practice for many years. The nature of most paediatric tumours differs from ‘adult-type’ tumours, being principally of mesenchymal origin (leukaemias, sarcomas, ‘blastomas’) rather than carcinomas. They are typically chemosensitive, and a greater emphasis on chemotherapy has developed because of this. The adverse effects of radiotherapy on developing tissues, which may be profound, have further encouraged the development of strategies that minimize dose and field of radiation, and for many, it has been possible to avoid radiotherapy entirely. Many chemotherapy regimens for children are highly intensive and lead to many acute complications. Supportive care for these patients represents a major component of overall management, such that inpatient oncology beds represent a disproportionately large part of acute paediatric beds in most large hospitals. The long potential life expectancy of a child cured of cancer leaves ample time for long-term sequelae to develop. The loss of expected years of healthy life in a patient who is not cured, or cured with
Late effects of treatment Specific sequelae of therapy Palliative care Psychological aspects of childhood cancer References
1150 1151 1154 1154 1156
disability may also be great. Viewed in these terms, the numerical imbalance between adult and paediatric oncology is to some extent redressed. The rarity of paediatric tumours is such that meaningful information regarding management can rarely be obtained in a single centre. For this reason, multi-centre collaborative approaches have been in place almost from its inception as a specialty, and it has long been accepted that all patients should be treated according to investigational studies wherever possible. This has been advantageous to the population as a whole, with a progressive improvement in survival of patients with many types of malignancy, and to individuals, for whom a monitored and coherent treatment strategy offers the best chance of cure. There is evidence too that young adult patients with paediatric-type tumours benefit from a paediatric approach. The borders between adult and paediatric patients are increasingly blurred, and a new specialty of adolescent or young adult oncology has grown. This chapter will review the more common malignancies seen in children, and will highlight some of the key problems still seen. Increasingly, as treatment strategies become more effective, the emphasis is placed upon the sequelae of therapy, the limitation of long-term damage and the preservation of a high quality of life for affected children and their families. Stratification, based upon diagnosis, site and stage is being augmented by biological risk factors such as Myc-N amplification status, proliferation index, chromosomal abnormalities, growth factor receptor expression and in recent years, gene expression profiling.
Epidemiology 1119
The benefits of these have already been seen for some tumours, and it is hoped will bring similar benefits for many others. Finally, there are increasingly, tumourspecific gene alterations identified for paediatric tumours. Clone-specific immunoglobulin gene re-arrangements allow estimation of so-called minimal residual disease for patients with acute lymphoblastic leukaemia, and this allows dose intensification for certain high-risk groups in the current MRC UKALL protocol. Similar strategies are being developed for solid tumours, using now well-recognized translocations to identify residual malignant cells. The potential benefits of dose reduction for good-risk patients or dose intensification for high-risk patients are clear. The development of new treatments for paediatric patients has historically lagged behind that for adult patients. Development of new chemotherapy drugs in children, for example, has always been an area of great sensitivity, given the vulnerable nature of children, and the fact that consent is given on behalf of the child rather than by the child themselves. This situation has improved substantially in recent years, as a result of legislation in the USA that provides a substantial incentive for paediatric drug development, and also because of the further organization of collaborative groups to include specific ‘drug development’ working groups. As a result, novel drug development in children can often follow closely behind that in adults.
EPIDEMIOLOGY The total incidence of childhood cancer (cancer in children less than 15 years of age) is about 1 per cent that of the adult population. The total age-standardized annual incidence of childhood cancer in the UK has been measured at 118.3 per million children less than 15 years of age, with a risk of developing a malignancy in childhood of 1 in 581.1 In the USA, there has been no substantial change in incidence for the major paediatric cancers since the mid-1980s, when modest increases, which probably reflected diagnostic improvements or reporting changes, were reported for CNS tumours, leukaemia and infant neuroblastoma.2 In the paediatric setting, the most frequently encountered diagnostic tumour groups are acute leukaemia, central nervous system tumours, lymphomas and soft tissue sarcomas (Table 46.1). The incidence of overall and individual cancers can vary internationally. For example, the rate of childhood cancer in Ibadan, Nigeria is four times higher than that reported for the Indian population of Fiji.3 For acute lymphoblastic leukaemia, CNS tumours and neuroblastoma, higher rates are found in western Europe and the USA than Africa and Asia. Racial differences in the incidence of individual cancers within a single country are recognized, with acute leukaemia and Ewing’s found more commonly in the white compared to the black population of the USA. Overall, childhood cancer is more common in boys than girls.3
Table 46.1 Age-standardized incidence rates of common childhood cancers. (Adapted from Stiller et al. 19951) Diagnostic group
Leukaemia Acute lymphoblastic Acute non-lymphocytic Lymphomas Hodgkin’s disease Non-Hodgkin’s lymphoma Central nervous system Astrocytoma Primitive neuroectodermal tumour Ependymoma Symphathetic nervous system Neuroblastoma Renal Wilms’ tumour Bone Osteogenic sarcoma Ewing’s tumour Soft tissue sarcoma Rhabdomyosarcoma Fibrosarcoma Other tumours Retinoblastoma Germ cell tumours Epithelial
Age-standardized incidence rates (per million/year)
32.3 5.9 4.6 6.2 10.0 6.0 3.1 8.1 7.6 2.5 2.3 5.2 1.0 3.7 3.6 3.0
Unlike cancer in adults, where the overwhelming majority of cancers are carcinomas that originate in epithelial surfaces, malignancy in children rarely takes the form of a carcinoma. In general, other than in leukaemia, most of the common forms of childhood cancer mimic developing or embryonal tissue development. For example, rhabdomyosarcoma and Wilms’ tumour resemble developing myogenic mesenchyme and renal tissue, respectively. Moreover, certain childhood tumours, such as neuroblastoma and Wilms’ tumour are more common in the first 5 years of life,1 suggesting that many cases of childhood cancer represent gestation-related defects in tissue growth and differentiation. Whereas CNS tumours and acute lymphoblastic leukaemia also have a higher incidence in early childhood, the peak incidence of Ewing’s tumour, Hodgkin’s disease and osteogenic sarcoma is found in early adolescence.4 Despite the histological features and age of onset of many childhood cancers, less than 5 per cent of cases are associated with a known genetic or cancer-predisposition syndrome.4 However, certain conditions are associated with an increased predisposition to cancer for children. For example, increased head circumference at birth is related to the development of CNS tumours in childhood5 and the development of
1120 Paediatric oncology
intra-abdominal embryonal tumours is well described for the over-growth syndromes of Perlman, Beckwith–Wiedemann and hemihypertrophy.6 Moreover, morphological abnormalities, such as cleido-cranial dysostosis and Wilms’ tumour are increasingly recognized for children with a wide variety of cancers.7 Furthermore, children with the constitutional chromosomal abnormality of Down syndrome have a 20-fold increased risk of developing acute leukaemia during the first 10 years of life. Sex chromosome abnormalities also confer a risk for developing certain malignancies. Any phenotypic female with part or all of a Y chromosome, such as testicular feminization syndrome and girls with mosaic 46 XO/XY Turner’s syndrome are at increased risk of gonadoblastoma. Similarly, males with Klinefelter syndrome (47 XXY) are at risk of developing dysgerminomas. The study of the genetic abnormalities found in childhood malignancies and the identification of certain cancerpredisposition genes is providing invaluable information for understanding of the pathogenesis of childhood cancer.8 For example, the inappropriate activation of normal growth-promoting genes, or cellular proto-oncogenes, is increasingly recognized as playing a role in the pathogenesis of childhood cancers. An example of this is seen with the t(8,14) translocation found with Burkitt’s lymphoma, where the c-myc gene is brought under the influence of immunoglobulin heavy chain enhancers at the break point region on chromosome 14, which results in the inappropriate overexpression of this transcription factor. Alternatively, the functional inactivation of tumour suppressor genes can cause a cancer predisposition phenotype with autosomal recessive characteristics, and the example of the retinoblastoma tumour suppressor gene (RB-1) on chromosome 13q14 has become a paradigm for the analysis of the inherited cancer predisposition syndromes. For example, genetic predisposition to cancer is recognized in the Wilms-aniridia-genitourinary abnormalities-retardation (WAGR) syndrome and the Beckwith–Wiedeman syndrome, where loss of the putative tumour suppressor genes WT-1 (chromosome 11p13) and WT-2 (chromosome 11p15), respectively, are associated with the development of Wilms’ tumour. Approximately 2 per cent of Wilms’ tumour patients have a family history of this condition, and candidate familial Wilms’ tumour genes (FWT) are described for FWT1 at 17q12-q21 and FWT2 at 19q13.4.9 Similarly, the Li–Fraumeni syndrome, where there are germ-line mutations in the p53 gene, is characterized by the familial clustering of multiple malignancies, including pediatric sarcomas, breast cancer, leukaemias, CNS tumours and adrenocortical carcinoma.8 Other conditions that are known to predispose to childhood cancer include multiple endocrine neoplasia (MEN) 2, certain phakomatoses and disorders that are associated with defects in DNA replication or repair. Children with MEN 2A and the related MEN 2B are at risk of developing medullary thyroid carcinoma, and the molecular genetic studies suggest that these children may inherit a mutation of 10q11.2, which activates the RET oncogene.4 Children
with neurofibromatosis type 1 (NF-1) have a risk of developing low-grade glioma, especially of the optic nerve/optic pathway, malignant peripheral nerve sheath tumours, rhabdomyosarcoma and acute myeloid leukaemia. Overall, the incidence of NF-1 for children with cancer has been found to be sixfold to eightfold higher than the general population.10 Tuberous sclerosis, which is classically characterized by seizures, mental retardation and facial angiofibroma, is a phakomatous condition associated with the development of retinal hamartomas, giant cell astrocytomas, other CNS tumours and rhabdomyosarcoma.11 Unlike tuberous sclerosis and NF-1, Von Hippel–Lindau disease (VHL) is not associated with any specific dermatological or developmental abnormalities. Children with VHL are at increased risk of developing cerebellar haemangioblastoma, retinal haemangioma, renal cell carcinoma, and phaeochromocytoma. Finally, children with DNA repair defects such as ataxia-telangiectasia, Fanconi’s anaemia and hereditary immunodeficiency diseases are known to be at increased risk for the development of leukaemia and lymphoma.4 In recent years, the list of tumour suppressor genes that are inactivated frequently by epigenetic events such as promoter methylation rather than classic mutation/deletion events has been growing. In relation to childhood cancers, this phenomenon has been described in relation to the development of medulloblastoma,12, and also the clinical risk group status for neuroblastoma, with methylation of cell cycle control and pro-apoptotic genes relating to tumour progression.13 Exogenous factors such as exposure to ultraviolet radiation and ionizing radiation are also associated with the development in malignancy in children. Most children with skin cancer are white and are genetically predisposed because of xeroderma pigmentosum, dysplastic nevoid syndrome or albinism.4 However, exposure to ultraviolet light in childhood is related to the development of malignant melanoma in later life.14 Children are especially sensitive to ionizing radiation-induced leukaemia and thyroid cancer. At present, only weak evidence exists for an association between individual or paternal exposure to electromagnetic fields and various chemical carcinogens and cancer in childhood.15 However, genetically determined responses to infection may be important in the aetiology of childhood acute lymphoblastic leukaemia.16 A more defined association exists between hepatitis B infection/carriage and the development of hepatocellular carcinoma in children, and recent studies have defined the interaction for viral antigen and the inactivation of tumour suppressor genes such as Rb and p53 alongside the activation of signal transduction pathways.17 In summary, epidemiological studies have played an important role in the clinical characterization of individual childhood cancers. Although the vast majority of childhood cancer occurs in children who do not have a predisposing factor, and the importance of environmental factors are largely uncertain, the identification of cancer predisposition syndromes has allowed the evolution of the molecular genetic and epigenetic characterization of
Specific paediatric malignant diseases 1121
diseases such as Wilms’ tumour, medulloblastoma and neuroblastoma. Such information is providing an invaluable insight into the pathogenesis of childhood cancer.
SPECIFIC PAEDIATRIC MALIGNANT DISEASES Childhood leukaemia Leukaemia accounts for approximately one third of all cases of malignancy in childhood. Of these, around 80 per cent are classified as acute lymphoblastic leukaemia (ALL) and 15 per cent as acute myeloid leukaemia (AML). Most of the remainder comprise chronic myeloid leukaemia. In the vast majority of cases a complete remission can be achieved, but a substantial number will at some stage relapse and around 25 per cent in total are currently not cured. Much has been learnt in recent years regarding specific prognostic features, and treatment protocols now reflect this by incorporating different strategies for different risk groups. For many patients this will mean an increase in intensity of treatment in an attempt to achieve cure. However, for some the outlook is now so good that a reduction in treatment (and, hopefully, toxicity) is the goal.
Acute lymphoblastic leukaemia ALL has an annual incidence of 3.5 cases per 100 000 in Europe and the USA. This equates to an individual risk of 1 in 2000 of developing ALL before the age of 15 years. There is a marked peak in incidence (up to 10 per 100 000) between the ages of 2 and 6 years, which is more noticeable within affluent societies. This prompted the hypothesis that the development of ALL may in part be the result of an abnormal immune response to an infectious agent, within genetically susceptible children who had previously been underexposed to common infections. This was supported by a large epidemiological case–control study, involving over 1700 children with leukaemia and carried out over a 15-year period.18 This showed that lower levels of day care activity (as a proxy measure for infections) within the first year of life were associated with a small but significantly
increased risk of developing ALL. In contrast, previously reported associations with increased exposure to other environmental agents – such as naturally occurring radon, electromagnetic fields, and ionizing radiation in fathers prior to conception of the child with leukaemia – have not been substantiated. Children with certain congenital chromosomal abnormalities, or abnormalities of DNA repair or immune regulation, have a notably higher risk of developing ALL – for example Down syndrome, Fanconi anaemia and Wiskott– Aldrich. In addition, the risk to a monozygous twin of a child with ALL is very much higher, particularly in the case of infant ALL, where the risk approaches 100 per cent.19 However, in the vast majority of cases of ALL there is no underlying disease or family history, and no aetiological factor can be identified. CLASSIFICATION
The French–American–British (FAB) classification from the 1970s described three categories of cell morphology (L1, L2 and L3). The L3 subtype retains some clinical significance as it identifies mature B-cell leukaemia, which requires treatment on a more intensive protocol. However, this classification system has now been superseded by the use of immunophenotyping by flow cytometry, which can more accurately classify over 95 per cent of all cases of ALL. In addition, cytogenetic analysis of the malignant clone (using karyotyping, FISH and RT-PCR) can be very useful in categorizing borderline cases, and is increasingly used to define risk groups. Furthermore, gene expression profiling has been shown to correlate closely with different types of leukaemia and may become a routine part of the diagnostic work up in the future.20 TREATMENT
Precursor B-cell (85 per cent) and T-cell (10 per cent) comprise the vast majority of childhood ALL. They are treated on common protocols with variations based on risk groups defined by age, sex, white cell count at diagnosis, presence or absence of CNS disease, cytogenetic abnormalities and early response to treatment (Table 46.2). Although international
Table 46.2 Risk stratification in current UKALL 2003 triala Standard risk
Intermediate risk
High risk
Age at diagnosis Highest white cell count Cytogenetic abnormalities
1–10 years 50 109/L None of those in HR group
10 years 50 109/L None of those in HR group
Response to induction therapy
Rapid early response
Rapid early response
Any Any BCR-ABL Hypodiploidy (44) MLL rearrangement AML1 amplification Slow early response
a
In addition, T-cell lineage and CNS disease confer a poor prognosis but these are linked to other variables within the table. Infants and patients with mature B-cell disease are treated on separate protocols.
1122 Paediatric oncology
differences exist, treatment in general consists of a 3–4-drug induction phase (vincristine, dexamethasone, asparaginase
an anthracycline), followed by CNS-directed therapy, one or more post-remission intensification blocks (usually 4–5 drugs), and then oral maintenance therapy with 6mercaptopurine and methotrexate for up to 2–3 years. The introduction of prophylactic cranial radiotherapy in the 1960s led to a dramatic reduction in the CNS relapse rate (previously over 60 per cent) and the regular use of intrathecal methotrexate, with or without systemic high-dose methotrexate and cytarabine, have further reduced the incidence to below 5 per cent. Cranial radiotherapy has recently been shown to be no more effective than intrathecal and high-dose systemic chemotherapy for CNS prophylaxis.21 Because of this, as well as its significant neurotoxicity in young children, it is now reserved for the 5 per cent of cases with CNS disease at diagnosis and those with CNS relapse. The value of post-remission intensification blocks was first recognized by the Berlin–Frankfurt–Munster (BFM) group in the 1970s. Since then many studies have examined the optimum number and composition of these blocks. Whilst it remains clear that high-risk patients benefit from one or more of these, it is apparent that a substantial number of patients are currently being overtreated. For example, in the MRC UKALL VIII trial, 50 per cent of children were cured without any delayed intensifications at all.22 The recent aim has therefore been to identify those patients who genuinely require post-remission intensification, and to stratify therapy accordingly.
to note that this latter group comprises 40–50 per cent of patients, and it is likely that there is a close correlation with the patients on UKALL VIII who were cured without any post-remission intensifications. However, this also is yet to be proven and is one of the questions posed by the current MRC UKALL 2003 trial. RELAPSE
Approximately 20 per cent of patients with ALL relapse, the majority with disease restricted to the bone marrow. The prognosis for these patients is hugely variable and is dependent on the length of first remission, the site of relapse and the disease lineage. For example, in the MRC R1 study, the 5-year EFS for patients who suffered a bone marrow relapse on therapy was 7 per cent, compared with 77 per cent in those relapsing outside the bone marrow greater than 21⁄2 years from diagnosis.24 Therefore risk groups have also been constructed within patients who relapse, in an attempt to optimize salvage therapy. Treatment begins in all patients with reinduction chemotherapy, which tends to be more intensive than at diagnosis (and therefore more toxic). This is followed by consolidation, then maintenance chemotherapy in the case of low-risk patients, and allogeneic stem cell transplantation for high-risk patients. Again there are ongoing studies (e.g. the MRC R3 trial in the UK) testing whether analysis of MRD at the end of reinduction can helpfully contribute to risk group stratification. Cranial radiotherapy is administered for CNS disease following consolidation if not previously given.
MINIMAL RESIDUAL DISEASE
On current treatment protocols around 98 per cent of patients achieve a complete morphological remission during induction. It has been known for many years that the time taken to achieve this remission is a strong prognostic factor to combine with the diagnostic features described earlier. However, because most patients do respond quickly to treatment and have no other adverse risk factors, the majority of relapses (in absolute numbers) occur in the good risk group. A more accurate predictor of relapse was therefore required. Over the last decade techniques have been developed to measure submicroscopic levels of the leukaemic clones, namely flow cytometry of their specific constellation of CD antigens and PCR detection of their known gene rearrangements. These techniques can monitor the malignant clone down to a level of 1 in 104–105 cells and have now proved powerful in predicting outcome.23a–c For example, children with a high level of this minimal residual disease (MRD) at the end of induction have a significantly higher risk of relapse, and this is irrespective of the traditional risk groups. It is hoped these patients may benefit from further intensification of therapy at an early stage, but this is not yet known. Similarly, almost all those with low level MRD (1 in 104) do not relapse. It is interesting
ROLE OF STEM CELL TRANSPLANTATION
Allogeneic stem cell transplantation is indicated in a small group of patients in first complete remission who have a very poor prognosis with chemotherapy alone. This group comprises children with certain cytogenetic abnormalities (e.g. BCR-ABL, AML1 amplification), a group of high-risk infants (see later), and the tiny number of patients who fail to remit during standard induction therapy. In addition, patients who relapse with high-risk features may benefit from receiving an allogeneic transplant in CR2. There is no evidence for the benefit of autologous stem cell transplantation and this is no longer routinely performed. NEW AGENTS
The dramatic increase in overall survival for children with ALL over the past 30 years has been due to established drugs being used in more effective ways on standardized protocols with improved supportive care. However, several new drugs are now becoming available for use in paediatric ALL (e.g. clofarabine, dasatinib). These are currently being studied in phase I and II trials in patients with refractory or multiply relapsed disease. It remains to be seen whether
Specific paediatric malignant diseases 1123
any will subsequently make the transition to up-front or relapse phase III trials. INFANT ALL
Infants commonly possess the 11q23 MLL gene rearrangement within their malignant clone. This partly explains their poor prognosis, which is worsened if they are less than 6 months of age or have a presenting white cell count 300. If all three features are present the EFS is a dismal 16 per cent. Treatment again consists of induction, intensive blocks and a prolonged maintenance phase, but there is a stronger reliance on cytarabine to which the leukaemic cells are more sensitive.
CLASSIFICATION
As with ALL the classification for a very long time was based on the FAB criteria of the 1970s. This system divided AML up into eight subtypes (named M0–M7) on the basis of the morphology of the blasts and their immunohistochemical staining patterns. These can still have clinical relevance, particularly in the case of acute promyelocytic leukaemia (APL, FAB subtype M3) where a rapid diagnosis may be crucial in guiding management. However, as more was learnt about the clinical outcome with relation to specific genetic abnormalities it became clear that the old system was inadequate. The World Health Organization (WHO) therefore devised a new classification in 2001, in which AML is divided into four major categories:
ALL IN ADOLESCENCE
There is increasing evidence that patients aged 15–19 years have a better outcome if treated on paediatric protocols.25 The next adult ALL trial in the UK will therefore recommend this as routine practice. MATURE B-CELL ALL
Only 2 per cent of childhood ALL is due to mature B-cell disease. These patients tend to be older (median age 10 years) and are more commonly male (ratio 3:1). The cells display L3 morphology and usually possess the t(8;14) translocation. They should be treated in the same way as B-cell NHL, with 5–6 months of intensive block chemotherapy, following which around 80 per cent are cured. In those that relapse, the anti-CD20 monoclonal antibody rituximab may be helpful, but currently the outcome is very poor. SUMMARY
In general the outlook for children with ALL is relatively good. Nearly all achieve complete remission and the vast majority are cured. The advent of techniques to monitor MRD has led to a more accurate definition of risk groups, and it is hoped that these will subsequently enable further improvement in cure rates with a concurrent reduction in the late effects of treatment.
Acute myeloid leukaemia Acute myeloid leukaemia is far less common than ALL in childhood, causing just 70 new cases in the UK each year. It is commonest in children under the age of 2 years and in adolescence, during which the incidence gradually rises. It can either arise de novo, or on the background of myelodysplastic syndrome (MDS). In a significant proportion of patients it occurs as a result of previous chemotherapy (particularly topoisomerase II inhibitors and alkylating agents) or in relation to a congenital syndrome, most commonly Down syndrome (see later), neurofibromatosis type I, or chromosomal fragility disorders.
1. 2. 3. 4.
AML with recurrent genetic abnormalities AML with multilineage dysplasia Therapy-related AML AML, not otherwise categorized (modified version of the old FAB system).
Along with this new definition, the WHO lowered the limit of percentage blasts in the marrow from 30–20 per cent for the diagnosis of AML to be made. TREATMENT
As with nearly all cases of paediatric malignancy, children with AML are treated within highly regulated national or international trials. Although variations exist between different trials, they share the common features of remission induction followed by consolidation. In general, induction consists of 1–2 courses of intensive chemotherapy, with a heavy reliance on the use of anthracyclines and ara-C. The MRC and CCG trials have consistently shown the benefit of providing high-intensity induction therapy, although the optimum drug combination is an ongoing source of debate.26,27 Current studies are also looking at whether the addition of gemtuzumab ozogamicin (mylotarg, a calicheamicin-conjugated anti-CD33 monoclonal antibody) can improve remission and survival rates. Consolidation is either in the form of further short courses of chemotherapy (up to a maximum of five courses in total) or, for some high-risk patients, stem cell transplantation. Some groups advocate the use of prolonged oral maintenance therapy in a similar way to that used in ALL. However, with the possible exception of APL, there is little evidence to support this, and it is not practised in non-APL cases in the UK. CNS prophylaxis is adequately achieved with intrathecal chemotherapy and cranial irradiation is reserved for the very few cases with CNS disease at presentation. APL is characterized by myeloid arrest and an abnormal clonal proliferation at the promyelocyte stage. It is unusual in being frequently associated with a severe coagulopathy, which responds to treatment with all-trans retinoic acid (ATRA). This causes differentiation and subsequent
1124 Paediatric oncology
apoptosis of the leukaemic clone, and should be given continuously with chemotherapy until complete remission is obtained. It is not known at present what form the rest of the chemotherapy should take. The current MRC trial (AML 15) is planning to compare the standard UK approach (4–5 courses of chemotherapy) with that of the Spanish group, which employs a relatively low-intensity schedule with a long maintenance phase. The latter has recently yielded an impressive 5-year EFS of 82 per cent.28
In addition, allogeneic stem cell transplantation is widely employed in patients achieving a second CR, where it can result in a minority being cured. It has also been attempted in the setting of refractory disease, where there is hope that a KIR mismatch between the haploidentical donor and the recipient might produce an exaggerated graft vs. leukaemia effect. However, this is highly experimental and at present the prognosis for these patients is dismal. DOWN SYNDROME AND AML
RELAPSE AND REFRACTORY DISEASE
With the regimens mentioned earlier, complete remission can be expected in around 90 per cent of cases of AML, with roughly 5 per cent dying from infection or haemorrhage during induction and the remainder having resistant disease. A further 30 per cent will go on to relapse. To a degree these can be predicted by the following risk groups, identified by AML 10 and 12: ●
●
●
GOOD RISK: Patients with favourable cytogenetic abnormalities – t(8;21) or inv16/t(16;16) (Relapse rate 25 per cent, OS 76 per cent at 5 years) POOR RISK: Patients with adverse cytogenetic abnormalities – -5, -7, del(5q), abn(3q) or complex – or 15 per cent blasts in the BM at the end of course 1 (Relapse rate 73 per cent, OS 21 per cent) STANDARD: Patients in neither the good risk nor poor risk groups (Relapse rate 52 per cent, OS 48 per cent)
In general, patients with refractory or relapsed disease are treated with a combination of fludarabine and ara-C, with or without an anthracycline, in an attempt to induce a second remission. If this is unsuccessful, alternative intensive chemotherapy combinations or novel therapies (such as mylotarg) may be tried. In the case of APL, arsenic trioxide has been reported to be very successful.29 If remission is achieved, most patients then proceed to allogeneic stem cell transplantation.
The risk of AML in children with Down syndrome (DS) is increased 10–20 fold, around 80 per cent of which is megakaryoblastic (i.e. M7 subtype). This disease is particularly sensitive to chemotherapy (especially cytarabine) and yields an EFS of 80 per cent.31 This contrasts with M7 found in patients without DS (10 per cent of patients) in whom the prognosis is poor (EFS 25 per cent). It is currently unclear as to why M7 leukaemic cells in only DS patients are so sensitive to therapy, but clearly this could be a fruitful area of research. Because of their chemosensitivity, as well as their tendency to experience worse toxicity, these patients are now treated on separate protocols. In addition there is a unique condition called transient abnormal myelopoiesis (TAM) that occurs in at least 10 per cent of neonates with DS. This condition is characterized by an abnormal clonal proliferation of megakaryoblasts that resembles AML but spontaneously resolves. Some 30–40 per cent of such patients will go on to develop megakaryoblastic AML within the next 3 years. Of note, mutations in the GATA-1 gene on the X chromosome (encoding for a transcription factor essential for normal erythroid and megakaryoblastic differentiation) are found exclusively in TAM and DS-AML M7.32 Such mutations have subsequently been discovered in blood samples from Guthrie cards of patients with M7 and their identical twins, making it highly likely that they are an example of the socalled ‘first hit’ in the process of leukaemogenesis. SUMMARY
ROLE OF STEM CELL TRANSPLANTATION
Autologous stem cell transplantation has not been shown to improve survival in any group of patients with AML. Although some studies have shown reduced relapse rates for those treated with such in consolidation, these are offset by an increase in treatment-related mortality. The value of allogeneic transplantation in first remission has yet to be clarified. In the CCG-2891 study, patients were allocated to transplantation following induction therapy if they had a matched family donor. In this group there was a small but significant improvement in OS and DFS at 8 years in comparison with the chemotherapy-only and autologous stem cell transplantation patients.30 In contrast the AML 10 trial did not show a significant difference in outcome. This is going to be further examined in the poor risk group of patients within AML 15.
As in the case of ALL, there has been a marked improvement in the outcome of patients with AML over the last 30 years, such that over 65 per cent are now long-term survivors. Cytogenetic analysis and early morphological response to therapy can identify three distinct risk groups. Monitoring of MRD, while less well developed than in ALL, may in future provide additional guidance for riskdirected therapy. However, a significant rate of treatment related mortality suggests that further intensification of therapy may not be possible. Therefore novel targeted therapies are likely to be required to provide the way forward. Finally, AML in DS has very distinct characteristics, which not only provide insights into the pathogenesis of AML, but also may prove helpful in our understanding of how to improve the efficacy of chemotherapeutic agents in all patients.
Specific paediatric malignant diseases 1125
Rhabdomyosarcoma Rhabdomyosarcoma (RMS) is the most common soft tissue sarcoma of childhood, accounting for 4–5 per cent of paediatric malignancies overall. Most children will present before the age of 10 years, although there are two peaks in incidence: a major one at pre-school age between 2 and 5 years, and a smaller one during adolescence. Although its aetiology remains unknown, there are wellrecognized associations with certain cancer predisposition syndromes (neurofibromatosis, Gorlin’s syndrome) and it is one of the tumour types encountered in the Li–Fraumeni syndrome. RMS has occasionally been associated with other anomalies such as congenital lung cysts. RMS is a tumour of embryonic mesenchymal origin. Common primary sites include the head and neck area (40 per cent), genitourinary tract (20 per cent) and extremities (20 per cent). RMS is an aggressive tumour, locally invasive into surrounding organs and along tissue planes. Distant dissemination through the lymphatic and haematogenous systems (to lung, bone and bone marrow) is recognized in up to 20 per cent of patients at presentation, particularly in the alveolar variant. The major concern for newly diagnosed patients is one of local control, since initial recurrence is principally local in the majority of cases. Historically RMS carried a poor prognosis, but multinational collaborative studies have improved the survival for localized disease significantly over the last 30 years. Sadly, the prognosis for disseminated disease has remained poor over the same interval.33,34
brings together two transcription factors, PAX3 and FKHR (seen in 70 per cent of tumours), and t(1;13)(p36;q14), involving PAX7 (approximately 20 per cent of alveolar RMS). Even where classical cytogenetic translocation is not seen, it may be possible to identify abnormal fusion RNA. There is no single genetic marker for embryonal RMS, although LOH for 11p15 is sometimes identified. SYMPTOMS AND SIGNS
As site of origin is so varied, so too are presenting symptoms. The primary site often determines both timing and mode of presentation. Orbital tumours, for example, will typically present early, with proptosis and diplopia. Nasopharyngeal tumours may cause nasal/airway obstruction, or a polypoid extrusion with discharge, as can also be found with lesions of the vagina. A painless mass may be the only finding, as with many paratesticular lesions. In contrast, deep-seated pelvic masses may grow to considerable size before presentation, often with obstructive symptoms. Most patients will present with a painless mass, but in certain sites, particular symptoms or signs may be characteristic. Parameningeal tumours have a significant risk of direct extension through bony tissue into the central nervous tissue, and may present with cranial nerve palsies. Prostatic RMS is typically aggressive, infiltrating adjacent bladder and urethra, and may present as acute urinary obstruction, which is most unusual in young males. Occasionally, primary tumours may arise in the biliary tree, causing obstructive jaundice. These sites present particular difficulties with management, but do not preclude cure.
HISTOPATHOLOGY
The separation of histological subtypes for RMS carries prognostic significance, and is used to stratify therapy in all current studies. Embryonal RMS accounts for approximately two thirds of all tumours, typically arising in head and neck or genitourinary sites. The botryoid variant of embryonal RMS is found where tumours arise under the mucosal surface of a viscus such as the vagina, bladder or nares. The spindle cell variant is most often seen at the paratesticular site, and like the botryoid form, is associated with a more favourable outcome. Alveolar RMS (ARMS) comprises approximately 25 per cent of all cases and has a less favourable prognosis. The characteristic appearance histologically of ‘alveolar’ spaces within the tumour, and the presence of even a single such area within a tumour biopsy is considered sufficient to confer a poor prognosis. Histological diagnosis can be particularly difficult in some cases, and there is also a ‘solid variant’ of ARMS. Diagnosis has been greatly helped by the identification of characteristic translocations present in ARMS, identifiable either by conventional cytogenetic methods and more recently (with greater sensitivity) by identification of fusion mRNA by RT-PCR. The two characteristic chromosomal translocations seen are t(2;13)(q35;q14), which
STAGING INVESTIGATIONS
Diagnostic work-up should include accurate assessment of the primary lesion and lymph nodes by CT or MRI scan, and examination of potential metastatic sites by CT chest, bone marrow aspirate and trephine and bone scan. Brain CT or MRI is indicated for limb primary tumours and alveolar histology. Two major staging systems by the North American (IRS) collaborative group, and the European (SIOP) group. IRS risk groups are based on initial surgical resectability, whereas the SIOP staging system is a TNM-based system. Reports from these groups cannot therefore be directly compared, although considerable progress has been made in producing a universally applicable system. PROGNOSTIC VARIABLES
Tumour site and stage at diagnosis is of great importance.33,34 With current multi-agent chemotherapy, surgery and radiotherapy, outcome for orbital RMS may be in excess of 80 per cent at 5 years. Despite progressively more intensive regimens, the prognosis for a patient with stage 4 disease remains poor, although those patients with metastases confined to the lungs may fare better.
1126 Paediatric oncology
Age at presentation affects prognosis. Poor prognostic variables (alveolar histology, limb primaries, metastatic disease) are much more frequent in the older child and certainly associated with a dismal outlook if all are present. In contrast, metastatic disease in younger children (10 years) with embryonal histology may be curable in up to 50 per cent of cases. Conventional histological classification (ERMS, ARMS) is highly significant, and therapy is stratified according to this. Where there is doubt, cytogenetic or molecular genetic classification may help, and within the two main variants of ARMS, there seems to be some prognostic difference. The translocation t(1;13) tends to be found more often in younger children and may be associated with a better prognosis than t(2;13).35 Hyperdiploid tumours have a favourable outcome and are often associated with embryonal histology. Diploid/ tetraploid tumours are associated with alveolar histology, and fare worse.36 The most important prognostic factor for RMS, as for all tumours, is the efficacy of therapy. Rapid improvements were obtained through the 1970s, 80s and 90s through large, multi-centre working groups in Europe and the USA, and these groups continue to develop rigorous, wellrun studies which it is hoped will lead to further improvements in survival with reduction in long-term morbidity. TREATMENT
Surgery Surgery for paediatric sarcoma is less aggressive than for adults. It is unusual for primary resection to be attempted, and secondary resection after chemotherapy, with or without radiotherapy is typical. Diagnosis should be by incisional biopsy in preference to needle or ‘Trucut’ biopsy, since the architecture of the specimen has great prognostic significance. Collection of fresh tissue for molecular biological analysis may be of critical diagnostic value. Primary excision may be attempted where this is not mutilating, and a complete clearance can be confirmed with marginal biopsy. This is rarely the case, except in the occasional paratesticular or peripheral limb tumours. Secondary surgery is of great importance, however. The results of the IRS III study showed that three quarters of all patients with a partial response could achieve CR, and that even where there was no apparent tumour response to chemotherapy, a high proportion could still be rendered free from tumour. Radiotherapy The importance of radiotherapy has been understood for many years. High-dose radiotherapy – 60–65 Gy – will achieve local control in 90 per cent of orbital tumours, but the risk of late effects following high local doses in a young person are often unacceptably high.37 Omitting radiation was associated with a high relapse frequency in most risk groups,38 although the IRS-I study reported no difference in
survival in completely resected (group I) patients treated with or without radiotherapy. It is clear that alveolar RMS requires radiotherapy and intensive chemotherapy, as do parameningeal tumours, for which a significant survival benefit was seen in IRS III with whole-brain radiotherapy.39,40 The timing and precision of radiotherapy has improved with successive collaborative studies. There is some evidence that delay in radiotherapy for parameningeal tumours may compromise local control, and central review of radiotherapy fields and doses in early studies showed significant variation. Hyperfractionation of radiotherapy dosing, which may offer an improved late effects profile with similar or improved local activity, has not to date been shown to give better local control and is not part of current strategies. Radiotherapy fields have been significantly reduced from 5 cm margins in IRS-1 to the current 2 cm margin, without evidence of any reduction in local control. The value of newer modalities of radiation therapy, such as proton beam radiation is not yet defined. Brachytherapy, which has a role for some tumours in adults is undertaken extremely rarely, given the high level of cooperation required. Chemotherapy Chemotherapy will significantly improve the prognosis for the majority of patients with RMS. Drugs with demonstrated activity in phase II studies include cyclophosphamide, vincristine, actinomycin D and doxorubicin, and more recently, ifosfamide, etoposide, cisplatin, carboplatin, epirubicin and vinorelbine have been shown to give responses. Topotecan was reported to have a high response rate in a phase II window study of previously untreated patients. Window studies cannot be compared directly with standard phase II studies, but a response rate of 46 per cent is highly encouraging. Combination chemotherapy has been based upon the combination of vincristine, actinomycin and cyclophosphamide for many years. The introduction of ifosfamide was associated with an improvement in response rates in the CWS-86 and SIOP 84 studies, and the combination of ifosfamide, vincristine and actinomycin forms the backbone of the recently closed SIOP study MMT95, and of the current, EpSSG RMS 2005. This latter examines whether the addition of anthracycline to IVA, and of a maintenance phase of vinorelbine and cyclophosphamide will improve the prognosis for patients with high-risk disease. METASTATIC DISEASE AND RELAPSE STRATEGIES
The outlook for children with metastatic disease remains poor, with current 5-year survival rates of around 20 per cent. Although clinical complete remission is achievable in up to two thirds, early relapse is a frequent occurrence. High-dose therapy with autologous stem cell rescue following conventional treatment as a method of eliminating the assumed minimal residual disease has certainly been employed. Although it would appear to prolong the time interval to progression in comparison to conventional therapy, there has been no effect on overall survival.41
Specific paediatric malignant diseases 1127
LONG-TERM EFFECTS
Survival from RMS is in excess of 60 per cent, and for certain sites may exceed 90 per cent. The overall burden of treatment is of greater significance for these young children, who may undergo intensive and cosmetically damaging therapy. The long-term effects of radiotherapy in particular may be profound for a child whose face will grow substantially after therapy. To some degree, these issues, reflect the main philosophical differences between North American and European strategies, the latter adopting a more intensive approach to local treatment in good responders with the aim of gaining quality of life in survivors. The cost of this approach is the fact that local relapse rates in the SIOP studies are higher than those documented elsewhere;34 a significant number of these may be salvaged with further therapy, and the majority of survivors overall have a reduced burden of therapy.
Liver tumours Primary malignant hepatic tumours represent 1.2 per cent of all cancers in children. There are two main types of malignant tumours, namely hepatoblastoma and hepatocellular carcinoma, plus the rarer mesenchymal tumours such as rhabdomyosarcoma and undifferentiated sarcoma.42 Benign hepatic tumours include haemangioendothelioma, mesenchymal hamartoma and the rare hepatic adenoma and focal nodular hyperplasia. HEPATOBLASTOMA
Hepatoblastoma is the most common liver tumour of childhood, with an incidence worldwide of 0.5–1.5 cases per million children.42 Hepatoblastoma is essentially a disease of young children, with a median age of presentation of 16 months. Cytogenetic and loss of heterozygosity analyses have uncovered frequent deletions of 1p and 1q, which may suggest the location of putative tumour suppressor genes at these sites.43 Although most cases of hepatoblastoma appear to be sporadic, there are known associations with Beckwith–Wiedeman syndrome, familial adenomatous polyposis, Gardner syndrome and extreme low birth weight.42 In addition, an increased risk of hepatoblastoma has been reported for infants with very low birth weights.44 Hepatoblastoma may present as an symptomatic abdominal mass, although weight loss, anorexia, vomiting and abdominal pain are features of more advanced disease. Many patients are anaemic and thrombocytosis is common. An elevated level of serum alpha fetoprotein (αFP) is found in 90 per cent of cases. For the Liver Tumour Study Group of the International Society of Paediatric Oncology (SIOPEL), where therapeutic strategy is based on primary chemotherapy, the pretreatment extent of disease (PRETEXT) system describes the site and size of the tumour, along with invasion of vessels
and distant spread as defined by radiological evaluation. The system identifies four categories, which reflect the sections of the liver that are free of tumour, plus any extension of the disease beyond the liver into the IVC/hepatic veins (V), portal vein (P), extra-hepatic abdomen (E) and distant metastases (M). For hepatoblastoma, only complete resection of the tumour offers a chance of cure. However, surgery alone cures very few patients; 50 per cent of cases present with unresectable primary tumours or distant metastases, and a 30 per cent relapse rate is found for children treated with surgery alone.45 Therefore, several study groups prefer to defer surgery until after 2 or 3 months of chemotherapy as this may improve tumour resectability. For example, for the SIOPEL I study, 85 per cent of patients responded to preoperative chemotherapy with doxorubicin and cisplatin, and a complete tumour resection was achieved for 69 per cent of patients,46 and the vent-free survival and overall survivals of 66 per cent and 75 per cent, respectively represent a gold standard for international comparison.47 Indeed, similar success rates have been reported in North American studies employing cisplatin in combination with vincristine and fluorouracil.48 The extent of disease at diagnosis has been shown to relate to survival for children with hepatoblastoma. For the SIOPEL I study, children with PRETEXT I disease had a 100 per cent 3-year event-free survival, which compared to an event-free survival of 83, 59, and 44 per cent for patients with PRETEXT II–IV disease.49 Chemotherapy with cisplatin and doxorubicin, in combination with surgery, offers a 25 per cent chance of cure for children with pulmonary metastases at presentation.50 Orthotopic liver transplantation is currently undergoing evaluation as a treatment modality for children with unresectable hepatoblastoma, and an overall 5-year survival rate of 62 per cent has been reported for children without extrahepatic metastases.51 In particular, survival in the region of 80 per cent compares with that of 30 per cent when an orthotopic liver transplant is performed with primary as opposed to rescue intent.52 HEPATOCELLULAR CARCINOMA
In most countries, hepatocellular carcinoma (HCC) is less common than hepatoblastoma. However, considerable geographic variation in incidence exists, with rates varying from 0.2 per million children in England and Wales to 2.1 per million children in Hong Kong. HCC is commoner in males and is associated with cirrhosis and other preexisting parenchymal liver disorders, and more rarely for children with Gardner syndrome or familial adenomatous polyposis.42 In comparison to hepatoblastoma, HCC is a tumour of older children, with a peak incidence between 10 and 14 years. Loss of heterozygosity on the distal part of chromosome 1p, which has been mapped to 1p35–1p36, may indicate that certain HCC and hepatoblastoma tumours share a molecular pathway in their pathogenesis.43
1128 Paediatric oncology
HCC usually presents as an abdominal mass, with jaundice and abdominal pain and serum αFP is elevated in 60–90 per cent of cases. As with hepatoblastoma, treatment is primarily surgical. However, HCC does not appear to be as chemoresponsive as hepatoblastoma, with poorer resection rates achieved post chemotherapy.46 Therefore, the outlook for HCC remains poor, with only 20 per cent of children surviving 5 years.53,54
neuroblastoma to mature ganglioneuroma, continues to fascinate scientists and clinicians alike. On the one hand it demonstrates not only both spontaneous and induced maturation but also the ability to undergo spontaneous regression. On the other hand, however, it still has one of the poorest outcomes overall in the case of metastatic disease, accounting for 15 per cent of all deaths in children from cancer.
Neuroblastoma
CLINICAL FEATURES
Neuroblastoma, accounting for 6–8 per cent of all paediatric malignancy, is predominantly a disease of early childhood. It is the most common malignant disease of infancy, has a median age of presentation of 2 years, and is rarely encountered over the age of 10 years. There is a slight increased incidence in boys (M:F; 1.2:1) and up to 70 per cent of children present with metastatic disease. The aetiology of the condition remains unknown. It has been associated with other neural crest lesions such as neurofibromatosis and Hirschsprung disease, and familial cases are recognized although infrequent. A pertinent observation is the high incidence of neuroblastoma in situ, microscopic nodules of neuroblasts within the adrenals found exclusively in fetuses and infants. This may be a normal part of embryonic adrenal development that regresses in health but persists in disease, i.e. that a dysregulation of cellular differentiation plays a part in the aetiology. There have undoubtedly been many advances in the understanding of various aspects of this disease over recent years, with the recognition of prognostic variables and moves towards stratification of treatment. This tumour group as a whole, however, ranging from the undifferentiated
Symptoms at presentation are extremely variable. The tumour arises from neural crest cells, which migrate to form the adrenal medulla and sympathetic chain, thus the primary can arise from cranium to pelvis. An adrenal (50 per cent) or abdominal (25 per cent) primary is more frequent in older children, whereas infants have a high incidence of thoracic (33 per cent) tumours. A common presentation is with obstructive symptoms such as respiratory embarrassment, and local invasion of neural foramina with spinal cord compression at any level is a well-recognized complication. As the disease is disseminated at presentation in over two thirds of the cases, more generalized symptoms such as anorexia, bone pain, fever and anaemia from bone marrow infiltration are common, and periorbital ecchymoses and proptosis are frequent signs of retrobulbar metastases. In addition to haematogenous dissemination of tumour to bone and bone marrow, spread to lung and brain parenchyma is now increasingly recognized, and regional or distant lymphadenopathy may be present. In the particular case of stage 4S neuroblastoma confined to infants (Table 46.3), disease commonly extends to the liver (massive hepatomegaly) and skin (subcutaneous nodules).
Table 46.3 INSS staging Stage 1
Stage 2A Stage 2B Stage 3
Stage 4 Stage 4s
Localized tumour with complete gross excision, with or without microscopic residual disease; representative ipsilateral lymph nodes negative for tumour microscopically (nodes attached to and removed with the primary tumour may be positive) Localized tumour with incomplete gross excision; representative ipsilateral non-adherent lymph nodes negative for tumour microscopically Localized tumour with or without gross excision, with ipsilateral non-adherent lymph nodes positive for tumour. Enlarged contralateral lymph nodes must be negative microscopically Unresectable unilateral tumour, infiltrating across the midline,a with or without regional lymph node involvement; or localized unilateral tumour with contralateral regional lymph node involvement; or midline tumour with bilateral extension by infiltration or lymph node involvement Any primary tumour with dissemination to distant lymph nodes, bone, bone marrow, liver and/or other organs (except as defined in stage 4s) Localized primary tumour (as defined for stages 1, 2A or 2B), with dissemination limited to liver, skin, and/or bone marrowb (limited to infants 1 year of age)
Multifocal primary tumours (e.g. bilateral adrenal primary tumours) should be staged according to the greatest extent of disease, as defined above, and be followed by a subscript ‘M’ (e.g. stage 3M). a The midline is defined as the vertebral column. Tumours originating on one side and crossing the midline must infiltrate to or beyond the opposite side of the vertebral column. b Marrow involvement in stage 4s should be minimal, i.e. less than 10 per cent nucleated cells on bone marrow biopsy or quantitative assessment of nucleated cells on marrow aspirate. More extensive bone marrow involvement should be considered stage 4. The MIBG scan (if carried out) should be negative in the marrow for stage 4s.
Specific paediatric malignant diseases 1129
STAGING
TREATMENT
There have been multiple staging systems utilized over the last few decades. In an attempt to rationalize these and permit accurate comparison of data, an International Neuroblastoma Staging System (INSS) has been adopted)55 (Table 46.3).
The three main treatment modalities of surgery, chemotherapy and radiotherapy all have their place in the management of this complex disease. Chemotherapy is certainly widely used for this chemosensitive tumour, and the chemotherapy agents vincristine, cyclophosphamide, cisplatin, doxorubicin, etoposide, and carboplatin are employed in various treatment regimens worldwide63,64 and for high-dose consolidation regimens based on busulphan/melphalan65 or carboplatin/etoposide/melphalan66 followed by autologous stem cell or bone marrow rescue are usually employed for high-risk disease. Surgery remains an important component of the treatment of this solid tumour, and is the treatment of choice for low-risk and localized tumours. As will be discussed later, conventional radiotherapy remains a useful palliative option, but is also important for localregional control of high-risk tumours. From observations clinically and biologically, Brodeur proposed classification of this disease into three main subtypes with excellent, intermediate and very poor outlook.67 These have been increasingly adopted, somewhat modified, and essentially form the basis of current broad treatment strategies.
INVESTIGATIONS
As a result of the complex nature of this disease, diagnostic work-up and monitoring of treatment progress involves an elaborate schedule of investigations. Urinary catecholamines are raised in up to 90 per cent of tumours, providing a relatively simple diagnostic aid. However, confirmation of the disease requires additional pathological evidence. In the light of an increasing panel of prognostic markers, biopsy material is necessary not only for light microscopy confirmation, but also recommended for tumour biology (MYCN amplification status, chromosomal gains and deletions and DNA ploidy). Since disseminated disease is a frequent finding at diagnosis, an extensive metastatic work-up is recommended. This should incorporate bone marrow examination (aspirates/trephines) in addition to thorough radiological assessment, which may include a combination of plain X-rays, ultrasound, CT/MRI scan, 131I-labelled metaiodobenzylguanidine (MIBG) scan and technetium bone scan if the MIBG scan is unavailable or negative. PROGNOSTIC FACTORS
Disease stage, patient age and certain biological variables are the most important clinical prognostic factors in this disease. The tumour-related findings of MYCN amplification is well recognized as a high-risk adverse factor in terms of survival for both low-stage56 and disseminated disease,57 and the chromosomal findings of gain of chromosome 17q58 and loss of 1p36 and 11q23 are also associated with an adverse outcome59 Whereas survival is excellent for localized disease that does not demonstrate adverse tumours with adverse biological characteristics do not fare as well. In terms of DNA ploidy, children with hyperdiploid disease can confer an improved survival in the absence of MYCN amplification.60 In relation to age, it has long been recognized that infants with stage 4 neuroblastoma that does not have an adverse biological signature have higher cure rates than older children, and recent studies have further defined the contribution of age to contribution to be continuous in nature, and suggests that a cut off point of 460 days could be applied in terms of prognosis.61,62 Although various serum markers (ferritin, LDH, neuronspecific enolase) have been investigated for prognostic value, none are specific and probably reflect tumour burden rather than any strong relationship. Of far more importance, it would seem, are the biological variables, although an unfavourable histological appearance is also an independent factor for adverse outcome.57
Low risk A low-risk group (localized tumours and infants with stage 4S, all with favourable biology) has an excellent prognosis (90–95 per cent progression-free survival), and the aim of treatment should be the avoidance of toxicity from treatment and minimal intervention.56 Surgery alone is recommended for localized tumours within this group. Although some children will have residual disease post surgery, there appears to be no survival advantage documented with the routine use of adjuvant therapy,68 and this should be reserved for relapses. Because of the high rate of spontaneous regression, the majority of infants with stage 4S disease can be managed with supportive care alone. Limited chemotherapy should be reserved for those patients with life-threatening symptoms, the aim obviously being cure but avoiding long-term toxicity if at all possible. For children with localized and low-risk tumours that are inoperable, low-dose primary chemotherapy confers an excellent progression-free survival and overall survival rates of 90 per cent and 100 per cent, respectively.69 Intermediate risk This group includes stage 3 disease (all ages) and stage 4 disease (infants and young children up to 18 months of age), but only if biological factors are favourable. These patients are treated with conventional chemotherapy (number of courses variable) together with surgery. However, ultimately they still have a good chance of cure from their disease, with progression-free survival in the area of 80–100 per cent.60,70 High risk High-risk patients (consisting mainly of children over 1 year of age with stage 4 disease, but also lower age and stage
1130 Paediatric oncology
with unfavourable biology), evidently form the bulk of patients. Historically they have a very poor outlook with progression-free survival rates of 25–45 per cent.71,72 However, there has probably been some improvement in survival over recent years, with benefit derived from strategies such as early gross total resection of primary tumour,73 loco-regional radiotherapy74,75 in the prevention of local recurrence. The inclusion of the differentiation-inducing agents 13-cis retinoic acid also improves the event-free survival for high-risk neuroblastoma to 45–50 per cent.66 Other current alternative approaches to treatment of this aggressive disease include the use of therapeutic 131 I-MIBG, which has been explored both as a palliative measure76 and also as an adjunct to high-dose chemotherapy in recent studies,77 and novel combinations of agents such as topotecan-cyclophosphamide have also shown encouraging activity for patients with high-risk neuroblastoma.78 Future studies may incorporate the negative prognostic findings that relate to a poor response to induction chemotherapy79 and persisting minimal residual disease,80 and facilitate the rational introduction of new therapies for this tumour that carries a poor prognosis. SCREENING
Mass screening of urinary catecholamines in infants was first carried out in Japan in the 1970s, and subsequently in Europe and North America, under the assumption that delayed clinical detection of low-stage disease ultimately evolves into high-risk disease and therefore earlier diagnosis could reduce mortality. This does not in reality appear to be the case – these screening programmes in infants have not, in fact, resulted in a reduction in either the incidence or mortality of older children with neuroblastoma. Indeed, there has probably been increased morbidity from higher detection of infants with favourable disease, which, if left undisturbed, might subsequently have regressed spontaneously. Ongoing programmes are currently investigating whether delaying the screening to 13 months confers any advantage, but otherwise present data do not support the value of neuroblastoma screening.81
Ewing’s tumour Ewing’s tumours comprise 10–15 per cent of primary bone tumours in childhood and adolescence, affecting 0.6 people per million population, or 1.7 per million children. There is a peak incidence at 10–15 years of age, affecting boys more than girls. It is rare below 5 years and after 30 years, and in those of Afro-Caribbean and Asian origin. Ewing’s tumour will typically present as a bony lump, often with pain. There may be associated neurological deficit, and a fever, due either to tumour load or to superinfection. The majority of tumours arise (in order of frequency) in the distal extremities, proximal limb, pelvis, chest wall and axial skeleton. Extra-osseous Ewing’s tumour
arises in the trunk, extremities, head and neck, and retroperitoneum. Extra-osseous Ewing’s tumours are likely to be larger and less amenable to definitive local surgery than bony Ewing’s. Approximately 25 per cent of Ewing’s tumours will show metastases at the time of diagnosis. The incidence of systemic symptoms – fever, weight loss – is greater for Ewing’s tumour than for osteosarcoma, and in part accounts for the difficulty in separation from osteomyelitis for some patients. A plain X-ray of the affected region will usually show bone destruction. An associated soft tissue mass may be evident, but this requires MR imaging for accurate assessment. Calcification is not prominent, but its presence does not exclude a diagnosis of Ewing’s sarcoma. HISTOPATHOLOGY
Pathologically, the Ewing’s tumour family fall within the group of malignancies referred to as the small round blue cell tumours of childhood, although microscopic, immunohistochemical and molecular diagnostic techniques allow this group to be identified with accuracy. Ewing’s tumour arises within the medullary cavity of bone, but erodes the cortex and at presentation may have a highly variable soft tissue component. In the extreme, Ewing’s tumour may have a barely identifiable or no recognizable bony component – the so-called extra-osseous Ewing’s tumour. The major differential diagnosis is from other primary bone tumours, osteogenic sarcoma, malignant fibrous histiocytoma, and osteomyelitis. Where the soft tissue component predominates, other small round blue cell tumours must be distinguished. These include rhabdomyosarcoma and other non-rhabdomyomatous sarcomas, neuroblastoma and lymphoma. Ewing’s tumour may show elements of neural differentiation, and in this, overlaps with primitive neuro-ectodermal tumours. A third tumour, the Askin tumour, presents as a chest wall tumour in adolescence, and similarly shares many features. Characteristic pathological features include the presence of intracellular glycogen (PAS positivity), neurosecretory granules and the expression of NSE, S100 and CD99. All three tumours are characterized by translocations involving chromosome 22q12 and in the majority of cases, chromosome 11q24. The breakpoint region of this translocation has been cloned and the transcript has been sequenced.82 The novel transcript includes the DNA-binding domain of the (human homologue) Fli-1 gene and the EWS gene, bringing the Fli-1 gene under the control of the EWS promotor and producing a transforming capacity not present in the wild-type Fli-1 gene product. In about 20 per cent of cases, the EWS gene is translocated to the ERG region of chromosome 21. This is a similar DNA binding protein to Fli-1. Many different fusion products are seen even within the EWS-Fli-1 rearrangements. The most common (type 1, 72 per cent) links Exon 6 of Fli-1 with Exon 7 of EWS, but
Specific paediatric malignant diseases 1131
at least eight other transcripts are known. Presence of the type 1 transcript seems to be an independent prognostic variable for localized tumours, with significantly better relapse-free survival.83 Approximately 30 per cent of tumours show a secondary change, trisomy 8, the significance of which is unclear. Translocation of the EWS region to other chromosomes leads to other tumour types, such as desmoplastic small round cell tumour of adolescence, associated with an EWS-WT1 translocation. Previous clinical practice for PNETs was to treat according to either a Ewing’s protocol, or a soft tissue sarcoma protocol. Whilst aspects of these therapies are similar, the importance of local therapy – radical surgery or radiotherapy – is emphasized more for Ewing’s-type strategies, and it is currently recommended that a Ewing’stype strategy be adopted for such tumours. Information regarding the comparative benefits of such approaches is not available at present. Accurate diagnosis is essential for appropriate therapy to be instigated, and the importance of a combined approach by oncologist, surgeon, radiologist and pathologist cannot be overstated. Typical features and the appropriate assessment of a suspected Ewing’s tumour is summarized in Table 46.4. MANAGEMENT
In his original reports of diffuse endothelioma of bone, Ewing described the radiosensitivity of the tumour,84
Table 46.4 Clinical features of Ewing’s sarcoma Presentation Pain and swelling of affected bone: mainly pelvis, femur, tibia and fibula Radiology Plain X-ray shows permeative destruction of bone. May show ‘onion skinning’ Soft tissue swelling may be apparent, best seen by MRI. CT chest, Tc bone scan show metastases in 25% Pathological features ‘Small round blue cell tumour’. Closely packed, rounded nuclei. Intracellular glycogen (PAS pos), variable neuronal differentiation. Immunohistochemistry CD99 positive Cytogenetics 85% t(11,22)(q24;q12) 10% t(7,22)(q22;q12)
Ews-Fli-1 Ews-erg
Within Ews-Fli-1, breakpoint varies: may have prognostic significance Prognostic groups 5 years efs Localized disease 65% Isolated pulmonary metastases 30% Bone or bone marrow metastases 20%
although with radiotherapy alone, the large majority of patients relapsed with disseminated disease within 2–5 years. Numerous agents have been shown to have activity in Ewing’s tumour including cyclophosphamide, doxorubicin, ifosfamide, vincristine, actinomycin D, melphalan, BCNU and 5-fluorouracil. Etoposide and ifosfamide administered in combination have been reported to have high activity.85,86 Combination chemotherapy has been used for many years, and overall survival has progressively increased. It is clear, however, that cure cannot be achieved by chemotherapy alone, and either radiotherapy or surgery is required to eradicate local disease.87 The issues of chemotherapy, radiotherapy and surgery have been addressed in the major national and international trials, and it is appropriate to discuss these here. US INTERGROUP EWING’S SARCOMA STUDY (IESS I AND II) TRIALS
IESS I randomized patients with localized disease to receive three drug (vincristine, actinomycin D, cyclophosphamide: VAC) or four-drug chemotherapy (VAC doxorubicin, VACA), or three drugs with pulmonary irradiation. A significant survival and relapse-free benefit was seen with four-drug therapy, with EFS at 5 years being 60 per cent, compared with 44 per cent with VAC radiation, and 24 per cent after VAC alone. IESS II study88 examined the effect of intensity of treatment on survival, comparing VACA given in a high dose intermittent schedule with a moderate dose continuous schedule. Intensity of chemotherapy was important, and doxorubicin was believed to be the major factor in this. EFS at 5 years was 68 per cent in the high intensity, and 48 per cent in the moderate intensity arms. CESS TRIALS
The first Co-operative Ewing’s Sarcoma Study, CESS 81, ran from 1981 to 1985, and enrolled 93 patients.89 Patients were treated with four-drug chemotherapy (VACA) in four, 9-week courses. Surgical resection of the primary tumour was performed, if possible, and radiotherapy administered at a dose of 36 Gy for residual disease postoperatively. Where radiotherapy was the only local therapy, a dose of 46–60 Gy was administered, with a 5 cm safety margin. Whole-compartment radiotherapy was also given, at a dose of 36 Gy. EFS at 10 years was 53 per cent, with greater survival in those patients who received both surgery and radiotherapy (69 per cent) than with radiotherapy (44 per cent) or surgery (48 per cent) alone. CESS 81 also demonstrated the possibility of defining risk groups other than metastasis. Patients with small (100 mL) tumours had EFS of 80 per cent, compared with 32 per cent for those with tumours measuring 100 mL. In 54 patients, an assessment of histological response to surgery was made, and a good response (90 per cent tumour cell kill) associated with a significantly
1132 Paediatric oncology
better outcome (79 per cent vs. 31 per cent for poor responders).89,90 The rate of local relapse in CESS 81 was high, particularly in patients where DXT alone was used (22 per cent), next where both were used (14 per cent), and lowest with surgery alone (3 per cent). This led to central radiotherapy planning such that radiation portals were subsequently planned according to X-rays defining disease at presentation, and failure rates subsequently fell.91 Following CESS 81, CESS 86 enrolled 177 patients with localized tumour between 1986 and 1991. Based on the results of CESS 81, treatment intensity was adapted to tumour volume and site, with high-risk patients (100mL volume at diagnosis) and all central tumours receiving four-drug chemotherapy using ifosfamide rather than cyclophosphamide (VAIA). In addition, the preferred local therapy of larger tumours was surgery, where possible, and local therapy was brought forward to week 9. A larger radiation dose was given postoperatively (45 Gy rather than 40 Gy) and for definitive therapy (60 Gy). In addition, patients received either conventional or hyperfractionated radiotherapy. With this approach, the prognostic significance of tumour volume was changed, such that patients with a tumour volume of 100 mL did not differ in outcome from those with a smaller volume. Instead, a new stratification at 200 mL was produced.92 Patients with tumour volume 200 mL had an EFS at 8 years of 42 per cent, which was significantly worse than those with a volume of 100–200 mL (70 per cent) or 100 mL (63 per cent). Note that the greatest survival was seen in patients previously regarded as being at high risk, who received intensified therapy with ifosfamide. Age, gender and tumour site were no longer prognostic. In contrast to CESS 81, the histological response to chemotherapy was no longer a significant prognostic factor (EFS: 64 per cent for good and 50 per cent for poor responders, respectively). UK-ET TRIALS
The first UK Ewing’s tumour (ET) study recruited 144 patients, of whom 15 per cent had metastases at diagnosis. All patients received local irradiation and four-drug therapy (VACA) for 1 year.93 EFS at 10 years for patients with local disease at presentation was found to be 41 per cent, but only 31 per cent for those with axial tumours. The second UK trial enrolled 201 patients with localized tumour between 1987 and 1993.94 ET-2 was identical to ET-1, except that ifosfamide replaced cyclophosphamide (VACA vs. VAIA). Relapse occurred in 61 of 191 evaluable patients (5-year DFS of 67 per cent). No significant difference was seen between patients receiving surgery alone as local therapy (DFS 62 per cent, n 114), radiotherapy alone (67 per cent, n 56) or both (78 per cent, n 20), although dual therapy is again associated with the highest survival. In the 24 patients with incomplete surgical resection, 3 relapses occurred in the 17 patients who received radiotherapy, compared with 4 relapses in the 7 patients
who did not. Radiotherapy is clearly of benefit in the setting of incomplete surgical resection. EICESS 92
EICESS 92 enrolled patients until January 2000, and a total of 631 patients were registered.95 Patients were stratified according to the information available in 1992 into standard risk (volume 100 mL) or high risk (volume 100 mL
metastatic disease). Standard-risk patients received initial therapy with four drugs (VAIA), and were randomized to receive four-drug therapy with either cyclophosphamide or ifosfamide. HR patients received either four-drug (VAIA) or five-drug (VAIA and etoposide). Total therapy was for 14 cycles over 44 weeks. Patients were assessed after 2 cycles, and surgery performed after 4 cycles. Guidelines indicated surgery to be the preferred local therapy where response was slow, and for large tumours. Preoperative radiotherapy (45 Gy) was given to poor responders. 369 patients were randomized. The 3-year EFS has been reported to be 66 per cent for patients with localized tumours, 43 per cent for patients with primary pulmonary/ pleural metastases, and 29 per cent for patients with other metastases. Either large tumour volume or pelvic tumour site appear to be adverse prognostic factors. Paulussen has reported preliminary data for 171 patients with metastatic disease at diagnosis registered up to 1995.96 A total of 36 received myeloablative megatherapy with stem cell rescue following conventional treatment. Bilateral whole-lung irradiation was administered in 57 with pulmonary involvement. Event-free survival at 4 years from diagnosis for all 171 patients was 27 per cent. EFS for isolated lung metastases was 34 per cent, for bone/bone marrow (BM) metastases, 28 per cent, and for combined lung plus bone/BM metastases, 14 per cent (P 0.005). Wholelung irradiation improved outcome in case of isolated pulmonary involvement (40 vs. 19 per cent at 4 years, P 0.05). In patients with combined pulmonary/skeletal metastases, intensification by megatherapy and/or whole-lung irradiation improved EFS from 0 to 27 per cent (P 0.0001). Whether intensification of therapy will lead to improved overall survival is as yet unclear. Smith et al.97,98 have reported analyses of dose intensity in published trials of Ewing’s sarcoma and osteosarcoma therapy, and identified doxorubicin dose intensity as the most important parameter. Marina et al. reported 53 patients treated at St Jude’s hospital with a dose intense regimen escalating ifosfamide dose intensity to 2.5 times and cyclophosphamide dose intensity to 1.5 times previous protocols.99 Overall survival for patients with localized and metastatic disease was good (90 6 per cent and 35 13 per cent), although because of small numbers of patients, no statistical difference was demonstrated. The role of high-dose chemotherapy with stem cell support is under investigation in the current European study of Ewing’s tumour, EuroEWINGS 99. Patients with intermediate risk disease, defined by initial tumour volume and
Specific paediatric malignant diseases 1133
the presence of pulmonary metastases are randomized to receive a maintenance phase of chemotherapy (VAI or highdose therapy with busulphan and melphalan. Studies of HD chemotherapy radiotherapy in patients with highrisk or relapsed disease have given promising results.100–102
Osteogenic sarcoma Osteogenic sarcoma (OGS) is the most common primary tumour of bone, with an incidence reported to be between 6.2 and 7.2 cases per 10 000 population between the age of 10 and 24 years. This is substantially less in people of black Afro-Caribbean descent. The peak incidence is seen between ages 15 and 20 years, and there is a male predominance of approximately 3:2. Overall, there are 100–130 new osteogenic sarcomas diagnosed in the UK annually. The majority of cases of OGS are sporadic, but the tumour is recognized to be associated with a number of underlying disorders. The incidence of OGS is high in patients treated with radiotherapy for retinoblastoma. Retinoblastoma is seen in patients with mutations involving the Rb gene (chromosome 13q14). Whilst karyotypic abnormalities involving chromosome 13q14 are seen in sporadic OGS, this is not typically the case. More recently, however, evidence for a high incidence of loss of heterozygosity (LOH) of the Rb gene has been reported.103 OGS is also a component of the Li–Fraumeni syndrome, attributed to abnormal P53 protein, encoded by the region 17q13. Karyotypic abnormalities involving this region are also seen with sporadic OGS, but overall only a small proportion of OGS are likely to be attributable to a known underlying predisposition.104 OGS is recognized as a second primary malignancy following prior treatment with radiotherapy.105 In a large cohort study (4400 survivors of a first solid tumour), an overall risk of 1 per cent was identified, a relative risk of 100 compared to the normal population.106 The risk of second malignancy was found to rise linearly with the dose of radiation received, such that an increased relative risk of 1.8 per Gray was seen. The total exposure to ‘electrophilic’ chemotherapy – those agents that interact directly with DNA forming covalent bonds – was also found to correlate directly to risk. CLINICAL PRESENTATION
The typical presentation of OGS is with pain and swelling of the affected limb or bone. OGS typically affects the most rapidly growing areas of the skeleton, the metaphyseal regions of the femur, tibia and proximal humerus (knees and elbows), although it is not known why this is the case. Possibly because of this pattern of disease, the delay between onset of symptoms and diagnosis tends to be less than for Ewing’s tumour (6–10 weeks compared to 5–6 months for Ewing’s). A smaller proportion of patients present with metastases – 10–20 per cent compared with 25 per cent for
Ewing’s – and patients have fewer systemic symptoms such as cough or dyspnoea. RADIOLOGICAL ASSESSMENT
Plain radiographs may show cortical destruction, elevation of the periosteum and new bone formation. Pathological fracture may be present, and may be associated with a greater incidence of local recurrence in patients treated by limb salvage procedures. Oedema and a variable amount of soft tissue may be visible. MRI is the preferred investigation to assess primary disease, since this will allow definition of soft tissues, the neurovascular bundle and epiphysis. The extent of marrow involvement can be identified, and skip metastases identified. For this reason, the whole of the affected bone should be imaged, and areas where the diagnosis is uncertain should be biopsied. Plain chest X-ray may demonstrate metastases, and CT, particularly spiral CT scanning is likely to identify smaller deposits, causing an increase in identified metastatic patients. Migration between stages will affect reported outcomes for patients in both the metastatic and non-metastatic groups of patients. 99 Tc bone scanning may also show bony deposits. Skeletal metastases carries a very poor prognosis, and any area of abnormality seen on bone scan should have a plain radiograph to confirm this. DIAGNOSIS
The diagnosis of OGS is histological, supported by appropriate clinical and radiological evidence. A biopsy is necessary, and this should be done by the surgeon who will perform the subsequent definitive procedure. The biopsy track will need to be resected at definitive surgery, and must avoid the apex of the tumour. Failure to do so risks causing tumour ulceration and fungation. A Trucut needle is usually sufficient for the biopsy. Diagnosis can often be made rapidly from cytology of the biopsy smear, but otherwise from conventional histological staining for alkaline phosphatase. Osteogenic sarcoma is characterized by the presence of bone or osteoid tissue by the tumour cells. A distinction between central (medullary) and surface (peripheral) tumours is made, although the majority of tumours fall into the central osteosarcoma subgroup. Surface tumours, such as periosteal or parosteal sarcoma, although high grade, have a much smaller chance of metastasizing. The most common pathological subtype is conventional central osteosarcoma, which is characterized by areas of necrosis, atypical mitoses, and malignant cartilage.107 Subtypes of high-grade osteoblastic, chondroblastic and fibroblastic OGS are identified, based upon the relative amounts of each cell type, but do not seem to carry significant prognostic value with the exception of chondroblastic osteosarcoma in one series.108 Intraosseous
1134 Paediatric oncology
well-differentiated osteosarcoma and parosteal osteosarcoma are associated with a favourable prognosis and can be treated successfully with radical primary excision alone, but are uncommon. Occasionally, little or no osteoid may be seen, and differentiation from Ewing’s tumour may be problematic. In this, as in the majority of paediatric tumours, the simultaneous processing of material for immunohistochemical and cytogenetic analysis is likely to allow an accurate diagnosis. Biopsy should not be performed by a surgeon who does not have appropriate pathological, immunohistochemical and cytogenetic support.
STAGING
The Enneking system remains in use, although its value prognostically is limited109 (Table 46.5). The majority of patients fall into group IIb.
BIOLOGICAL MARKERS
As for many primary solid tumours, biological factors have been sought that may offer prognostic value at diagnosis. Demonstration of LOH of the Rb gene in newly diagnosed patients wth osteosarcoma has been reported to predict early treatment failure.103 In a series of 34 patients for whom evaluation was possible, event-free survival at 5 years was 100 per cent for patients without LOH, 43 per cent for all patients with Rb LOH, and 65 per cent for non-metastatic patients with Rb LOH. Similarly, LOH p53 has been reported to correlate with lack of chemoresponsiveness in vitro110 and in vivo.111 The dominant oncogene c-fos is implicated from transgenic mice which constitutively overexpress the gene, and almost always develop osteosarcoma.112
TREATMENT
Surgery Osteogenic sarcoma treated by surgical excision alone leads to an overall survival of at best, 20 per cent.113 Pulmonary metastases develop in the large majority of
patients so treated, indicating that control of metastatic disease is the major factor in determining survival. It is not surprising that intra-arterial therapies have produced no improvement in outcome.114,115 Local control is essential, however, and failure to remove the primary tumour completely is associated with a high risk of subsequent local and metastatic relapse even with adjuvant chemotherapy.108 Local relapse is often associated with simultaneous metastatic recurrence.116 Limb salvage is now the expected approach and low rates of local recurrence, and favourable local and metastatic relapse rates are expected,117 even where there is a pathological fracture, provided adequate surgical margin can be attained.118 It is beyond the space of this chapter to discuss the orthopaedic approaches taken for limb conservation surgery, but the patient and their family need to be aware of the benefits and drawbacks of a given procedure. For some amputation may offer the best and most reliable chance of return to normal life. One particular approach worthy of consideration where an above-knee amputation might otherwise be necessary is the van Ness rotation plasty, for which excellent functional outcome may be obtained.119 A metallic endoprosthesis is most commonly used, and technological improvements in these have allowed the failure rate to be reduced. Currently, ‘growing’ endoprosthetic replacements are available, which will reduce later morbidity in patients who grow significantly after their definitive procedure. For some patients, where a tumour is away from a growth point, the possibility exists to use a vascularized endoprosthetic graft, usually from the contralateral fibula. Such grafts have considerable advantages over metallic implants, in their ability to repair after injury, and are at no significant risk of infection. Chemotherapy Single chemotherapy agents with activity against osteosarcoma include: doxorubicin,120 methotrexate,121 cisplatin,122–124 ifosfamide,125–129 carboplatin,130,131 etoposide in combination with ifosfamide.85,132 A beneficial effect of melphalan in high-dose therapy has also been reported.133 The value of combination chemotherapy in addition to surgery has been demonstrated in the two major studies
Table 46.5 Enneking staging system
I II III
Stage
Grade
Site
Metastases
A B A B A B
Low Low High High Low High
Intracompartmental (T1) Extracompartmental (T2) Intracompartmental (T1) Extracompartmental (T2) Any Any
None (M0) None (M0) None (M0) None (M0) Any (M1) Any (M1)
Specific paediatric malignant diseases 1135
performed in the 1970s. In the MIOS study, using the T-10 regimen, event-free survival at 2 years was 66 per cent with chemotherapy and 17 per cent without chemotherapy.134 Overall survival, after patients who had relapsed had been treated with further surgery and chemotherapy, was 71 per cent for the initial chemotherapy group, and 51 per cent for the group with no initial chemotherapy.135 Similar benefit of chemotherapy was reported by Eilber.136 The first European Osteosarcoma Intergroup study (EORTC 80831) compared the efficacy of a two-drug regimen (cisplatin and doxorubicin) with the same drugs, with the addition of methotrexate at a dose of 8 g/m2.137 Toxicities for both regimens did not differ, and overall survival was not significantly different (at 53 months: 64 per cent for two drug, 50 per cent for three drug), but diseasefree survival for the two-drug arm was significantly greater (57 per cent vs. 41 per cent). The second European Osteosarcoma Intergroup study (EORTC 80861) compared the efficacy of this two-drug regimen (cisplatin and doxorubicin) with a multi-agent regimen similar to that used by Rosen in the T10 protocol (preoperative vincristine, high-dose methotrexate, and doxorubicin; postoperative bleomycin, cyclophosphamide, dactinomycin, vincristine, methotrexate, doxorubicin, and cisplatin).138 All patients receiving the multi-agent chemotherapy arm received cisplatin, rather than only those identified to have a poor histological response. A total of 407 patients were entered in this randomized study, and no benefit from the multi-drug arm over two-drug arm was seen. At 5 years, overall survival was 55 per cent and progression-free survival was 44 per cent in both groups. The two-drug regimen was shorter in duration and better tolerated. An analysis of relative dose intensity in two randomized trials run by the EOI failed to show a relationship between this and progression-free or overall survival, but there was a trend towards worse outcome for those patients who received incomplete chemotherapy, or had the lowest overall intensity.139 It was not possible to distinguish the relative effect of absolute dose or dose intensity from this study, but the importance of doxorubicin and methotrexate dose intensity has been reported elsewhere.98,140,141 There is little evidence to support an effect of cisplatin dose intensity. Following a pilot study that demonstrated acceptable toxicity,142 EOI 931 involved a randomization to receive either 3-weekly or 2-weekly cisplatin and doxorubicin, with G-CSF support for the rapid schedule. Despite the negative results of the first EOI study, there is good evidence that cumulative dose and dose intensity of methotrexate treatment affects response, and by implication, outcome. Much of the work in this area comes from the group of Delepine. The importance of dose intensity of methotrexate is reviewed in reference.141 First, this group have demonstrated a correlation between a pharmacokinetic parameter, methotrexate concentration, at the end of infusion, and tumour response as measured by the degree of tumour necrosis.108,143 A concentration of methotrexate
of 700 μM after a 6-hour infusion,108 or 1000 μM at the end of a 4-hour infusion143,144 has been shown to be predictive of response, allowing stratification according to a surrogate endpoint early on in treatment. Furthermore, dose adaptation based upon pharmacokinetic parameters has been performed, and a survival benefit seen. In a study of 44 patients receiving conventional dosing, and 27 with individual dose adaptation, Delepine reported significantly greater response rates and disease-free survival (76 per cent with conventional dosing compared with 92 per cent with pharmacokinetically guided dosing.145 For patients treated with a methotrexate dose of 8, 10 or 12 g/m2, the proportion of patients attaining a methotrexate concentration in excess of 700 μM has been reported to be 44 per cent, 59 per cent and 85 per cent,146 and one explanation for the variable reports of efficacy of methotrexate stems from this variability. STRATIFICATION OF PATIENTS
The concept of stratification of patients based upon histological response of the resected tumour was introduced early in the history of chemotherapy, although it has yet to be proven whether modification of therapy affects overall prognosis.147–149 In the majority of series, the presence of more than 90 per cent necrosis in the resection specimen identifies a good prognostic group, for whom a high (75 per cent) overall survival may be expected. In the EOI first study, OS for patients with 90 per cent necrosis was 85 per cent, compared with 40 per cent for those with 90 per cent. In the second EOI study comparing a two-drug (cisplatin – doxorubicin) regime with a multi-agent regime similar to the T10 protocol, 30 per cent of patients had 90 per cent necrosis with the two-drug regime, and 27 per cent with multiagent therapy (ns).138 The event-free survival of this study as a whole was disappointing, 55 per cent at 8 years, and no significant difference was seen between the two arms. These findings have been found repeatedly, but unfortunately, most authors report only a minority of patients achieve a good response. Other groups have reported higher proportions of patients showing good histological response.147,150
CURRENT RECOMMENDED THERAPY
The current strategy, adopted by the European Osteosarcoma Intergroup (EOI), Cooperative Osteosarkomstudiengruppe, Germany (COSS), Combined Oncology Group, United States (COG) and Scandinavian Sarcoma Group (SSG) groups is a multi-national collaborative study, which will assess the effectiveness of multi-agent chemotherapy and optimal surgery in all patients with osteosarcoma. All patients receive a three-drug induction with methotrexate, cisplatin and doxorubicin (MAP), after which they undergo definitive surgery. Good responders, defined by a residual viability of 10 per cent, are randomized to receive either the same chemotherapy, or MAP followed by maintenance
1136 Paediatric oncology
interferon-alpha. Poor responding patients, for whom residual viability is 10 per cent, will receive either MAP or an intensified regimen with additional ifosfamide and etoposide. WHAT DOES PERCENTAGE NECROSIS MEAN?
A poor response to initial chemotherapy may be explained by the tumour itself, or by the type of therapy used. The tumour may be intrinsically more resistant to chemotherapy, because of underlying biological differences, or the chemotherapy used is inadequate: the concentration of drug, total duration of exposure or some other parameter may be insufficient. The identification of adverse biological factors, such as loss of heterozygosity for the Rb gene may allow an appropriate aggressive approach to be taken early in the course of treatment, and may in time point to novel therapeutic strategies. More immediately, the identification that dose intensity, and further, concentration of drug correlates with response emphasizes the need to use chemotherapy optimally, and perhaps to individualize therapy to each patient. Strategies to include extra agents such as methotrexate have often led to reduction in dose intensity of all agents, with a consequent failure to improve outcome. POOR PROGNOSIS DISEASE
Although the majority of tumours arise in peripheral long bones, at least 20 per cent of patients present with flatbone, axial or metastatic disease at diagnosis. These patients have a mixed prognosis. Chemotherapy alone is unlikely to be curative, and for some, surgical resection is not possible. The prognosis for patients with primary metastatic disease is dependent upon site. In the study of Harris et al., patients with primary bone metastasis had a particularly poor outcome.151 Most patients will have pulmonary metastasis, and for this group, surgical resection seems to be of benefit.152–154 Bacci et al. reported a 5-year OS of only 14 per cent with an aggressive regimen using methotrexate (8 g/m2), cisplatin, doxorubicin, ifosfamide and etoposide.155 All patients who did not achieve a complete surgical or chemotherapeutic remission died. In contrast, patients in the series of Harris et al. who presented with less than eight pulmonary nodules had a relatively high chance of cure. 12 of 18 were alive at 5 years, and patients with unilateral disease had a 5-year event-free survival of 75 per cent. These data are encouraging, and indicate that an aggressive, multimodality approach may overcome previously adverse prognostic factors.
Renal tumours Renal tumours constitute 6–8 per cent of childhood cancer in the USA.156 Wilms’ tumour (WT), the most common renal tumour of childhood is a paradigm for the multimodal treatment of solid tumours in children.
WILMS’ TUMOUR
The incidence of WT is 8.1 cases per million Caucasian children less than 15 years of age.157 WT usually presents before 5 years of age, and is associated with congenital abnormalities such as anhiridia, hemihypertrophy and hypospadias, and with the Denys–Drash, Beckwith– Weidemann and WAGR syndromes.157 At least three genes are associated with Wilms’ tumour, and since the incidence of familial WT is less than 1 per cent, the genetics of WT do not always follow the simple two-hit model of tumour suppressor genes.8 Patients with the WAGR (hemihypertrophy, anhiridia, genitourinary malformation and mental retardation) syndrome have a constitutional deletion of the WT suppressor gene, WT1, which is located at 11p13.8 Although constitutional mutations of WT1 are described for children with the rare Denys–Drash syndrome, specific mutations of WT1 have been found in less than 10 per cent of sporadic WT.158 WT2 has been found to map to 11p15.5, which is also the location of the Beckwith–Weidemann gene abnormality, and familial WT genes are also located at 17q12 and 7p13.157 Nephrogenic rests, which are small, usually microscopic, clusters of blastemal cells, tubules and stromal cells found at the periphery of the renal lobe, are thought to be precursor lesions for WT.156 In terms of prognosis, loss of heterozygosity for chromosomes 1p and 16q confers an adverse outcome for children who present with otherwise favourable histological features.159 Most children present with a history of an abdominal mass or swelling, and ultrasound screening for children with conditions that predispose to the development of Wilms’ tumour can result in the detection of early stage disease.160 Abdominal pain, gross haematuria and pyrexia are frequently observed, and hypertension is encountered in 25 per cent of cases.156 Investigations performed at diagnosis include abdominal ultrasound and CT, with the aim of determining the extent of spread of tumour into adjacent organs such as the liver, the involvement of the inferior vena cava with tumour, and radiological abnormalities of the opposite kidney. A chest X-ray is also indicated to exclude pulmonary metastases. However, staging of WT also depends on the involvement of regional lymph nodes and direct examination of the contralateral kidney by the surgeon. The presence of tumour cells in retroperitoneal lymph nodes is an important prognostic factor for children with WT, and lymph node sampling forms part of the staging of WT according to the National Wilms’ Tumour Study Group (Table 46.6). Classically, WT is made up of varying proportions of three cell types, namely blastemal, stromal and epithelial. WT histology is designated favourable or anaplastic, depending on the presence of gigantic polypoid nuclei within the tumour sample in the latter case. The finding of diffuse or focal tumour anaplasia, which occurs in approximately 5 per cent of cases of WT generally results in more aggressive treatment.156
Specific paediatric malignant diseases 1137
Table 46.6 The staging of Wilms’ tumour by the National Wilms’ Tumour Study Group Stage I
Stage II
Stage III
Stage IV Stage V
Tumour limited to the kidney and completely excised. The renal capsule has an intact outer surface. The tumour is not ruptured or biopsied prior to removal (fine-needle aspiration allowed). Renal sinus vessels not involved. No evidence of tumour at or beyond the margins of resection. Tumour extends beyond the kidney, but was completely excised. There may be regional extension of tumour (i.e. penetration of the renal capsule or extensive invasion of the renal sinus). The blood vessels outside the renal parenchyma, including those of the renal sinus, may contain tumour. The tumour is biopsied (except for fine-needle aspiration), or there is spillage of tumour before or during surgery that is contained to the flank, and does not involve the peritoneal surface. There must be no evidence of tumour at or beyond the margins of resection. Residual non-haematogenous tumour is present, and confined to the abdomen. (1) Lymph nodes within the abdomen or pelvis are found to be involved with tumour (renal hilar, para-aortic or beyond); (2) the tumour has penetrated the peritoneal surface; (3) tumour implants are found in the peritoneal surface; (4) there is residual or microscopic tumour postoperatively; (5) the tumour is not completely resectable because of local infiltration into vital structures; (6) tumour spill is not confined to the flank either before or during surgery. Haematogenous metastases (lung, liver, bone, brain etc.) or lymph node metastases outside the abdomino-pelvic region are present. Bilateral renal involvement is present at diagnosis. An attempt should be made to stage each side according to the above criteria on the basis of the extent of disease prior to biopsy or treatment.
The therapy of WT comprises a combination of surgery, chemotherapy and radiotherapy that is dependent on the stage and histology of the tumour. Whereas in the USA, immediate nephrectomy is performed, followed by treatment with chemotherapy radiotherapy, more recent trials conducted by SIOP have employed pre-nephrectomy chemotherapy.157 The use of pre-nephrectomy chemotherapy may confer an advantage both in terms of a lower tumour rupture rate at operation, therefore possibly reduce the risk of local relapse,161 and the identification of good prognostic subgroups based on tumour response.162 However, immediate nephrectomy is advocated by the National Wilms’ Tumour Study Group (NWTSG) to avoid possible modification of tumour histology and staging, and the administration of chemotherapy to children with nonWilms’ malignancies or benign lesions.157 Approximately 1 per cent of children with unilateral WT develop disease in the contralateral kidney, and as for children with bilateral WT at presentation, consideration is given to renal sparing surgery. The chemotherapy for European and NWTSG studies is largely based on combination therapy with vincristine, actinomycin-D and doxorubicin, with radiotherapy generally being employed for stage III or stage IV disease. However, the excellent survival of approximately 80 per cent for children with advanced disease has led national collaborative groups devising studies to try to identify minimal necessary therapy for this disease. For example, for the National Wilms’ Tumour Study-3 (NWTS-3), the 87 per cent relapse-free survival for stage II patients receiving 15 months of therapy with vincristine and actinomycin-D was not improved by the addition of an anthracycline and/or addition of 20 Gy radiotherapy.163 Similarly, shortening post-nephrectomy chemotherapy from 18–4 weeks for children with stage I, including
intermediate-risk or anaplastic disease, had no effect in terms of prognosis.164,165 For children relapsing after chemotherapy with vincristine and actinomycin alone, salvage therapy with radiotherapy and cyclophosphamide and doxorubicin-containing chemotherapy results in a good outcome for 70 per cent of cases.166 In addition, for children with relapsed WT, highdose chemotherapy with carboplatin, etoposide and melphalan has been reported to confer a prolonged disease-free survival in approximately 50 per cent of cases.167 OTHER RENAL TUMOURS
Clear cell sarcoma of the kidney is a primary renal tumour which has a significantly higher relapse and death rate than favourable histology WT.156 As with WT, clear cell sarcoma metastasizes most frequently to the lungs, and has a tendency to metastasize to bone and brain. The tumour is more common in boys, and presents at a median age of 1.5 years.168 The overall survival for clear cell sarcoma is 69 per cent, and survival has been found to relate to stage, age at diagnosis and treatment with doxorubicin, with stage I patients having a 98 per cent event-free survival rate on NWTSG studies.168 The rhabdoid tumour of the kidney is a distinctive and highly malignant tumour that metastasizes to the lungs and CNS, is more common in males and presents at a median age of 13 months.156 Age at diagnosis is an important determinant for outcome, with infants having a very poor prognosis.169 Clear cell sarcoma and the rhabdoid tumour of the kidney are distinct entities; they are often treated as ‘unfavourable’ histological subtypes on WT protocols and receive triple chemotherapy and radiotherapy.156 For this disease survival rates are improving, and the relapse-free survival is improved by prolonged schedules of administration for chemotherapy.170
1138 Paediatric oncology
Malignant germ cell tumours Germ cell tumours account for approximately 3 per cent of all childhood malignancies in the UK, with an annual incidence of 2.4 per million children per year.171 Malignant germ cell tumours (MGCTs) usually arise in the midline structures, i.e. the sacrococcygeal region, retroperitoneum, mediastinum and midbrain. The morphological subtype of a MGCT reflects the differentiation pathway to which a cell becomes committed (Fig. 46.1). An increased incidence of germ cell tumours is found for children with dysgenic gonads, Klinefelter’s syndrome and defects in the urogenital tract such as cryptorchidism and sacral agenesis. The relative incidence of GCT according to age, sex and pathological subtype is shown in Table 46.7. The cytogenetics of germ cell tumours differ with respect to age, sex and tumour location. For testicular or mediastinal tumours in adolescent boys, isochromosome 12p is a common finding, and deletion of 1p/gain of 1q and chromosome 3 are the most common abnormalities among
Primitive germ cell
Unipotential
Gonadoblastoma
Multipotential
Germinoma Dysgerminoma Seminoma
Teratoma
Embryonal carcinoma
Embryonal
Extraembryonal
Mature Immature
Yolk sac Choriocarcinoma
Figure 46.1 Differentiation pathway for malignant germ cell tumours.
the MGCTs from both sexes.172 Staging of MGCT can be based on conventional TMN stage I–IV criteria, or on the basis of risk groupings.171 Patients with surgically completely resected tumours, and in whom the tumour markers αFP and β-hCG rapidly normalize can be observed without further therapy, and children with metastatic disease require chemotherapy. MGCT may present as an asymptomatic mass, or with symptoms due to compression/obstruction, i.e. respiratory distress due to a mediastinal tumour, and visual disturbances, diabetes insipidus, and hypopituitarism in the case of intracranial germ cell tumours.173 Elevated serum levels of, or positive immunohistochemical staining of, germ cell tumours for αFP indicates the presence of a malignant component, i.e. yolk sac tumour or embryonal carcinoma. Elevations of β-hCG in patients with germ cell tumours occur with choriocarcinoma and germinomas.173 Whereas surgical resection is the treatment of choice for benign germ cell tumours, radical resection of malignant lesions is generally limited to gonadal sites,171 but the extent of resection is important for disease at sites such as the mediastinum.174 With the exception of MGCT of the CNS, radiotherapy has little role in the management of MGCT, except in the treatment of residual tumour following second-line chemotherapy and surgery.171 Chemotherapy with bleomycin, etoposide and cisplatin (BEP) has become the gold standard regimen for the therapy of MGCT.171 In addition, a dose–response relationship has been found for cisplatin and outcome for high-risk malignant germ cell tumours, albeit at the cost of increased ototoxicity.175 However, chemotherapy based on bleomycin, carboplatin and etoposide (JEB) is undergoing evaluation, and preliminary results suggest that JEB has reduced long-term nephro- and ototoxicity than BEP, without compromising efficacy.171,176 For children receiving platinum-based chemotherapy regimens, their prognosis is usually excellent for sites of disease as diverse as the gonads177 and retroperitoneum.178 Whereas the majority of intracranial germinomas are cured by radiotherapy alone, secreting intracranial tumours such as yolk sac tumour, embryonal carcinoma and choriocarcinoma have a poorer prognosis with radiotherapy
Table 46.7 Relative incidence of childhood germ cell tumours Site Sacrococcyx Vagina Ovary Testis Retroperitoneum Mediastinum Head and neck Cranium
Age
Relative incidence (%)
Pathology
Neonate Infant Adolescence Infant and adolescent Infant Adolescent Infant and neonate Infant
35 2 25 20 5 5 3 5
Teratoma (malignant 10–30%) Yolk sac Teratoma (malignant 30%) Teratoma (malignant 80%) Teratoma (rarely malignant) Teratoma (malignant 20–40%) Teratoma (rarely malignant) Germinoma, embryonal carcinoma, mature teratoma
Specific paediatric malignant diseases 1139
alone. A combination of chemotherapy with cisplatin, ifosfamide and etoposide (PEI), followed by surgery and craniospinal radiotherapy has shown very promising results for these tumours, with an event-free survival of 75 per cent.179 In particular, cumulative cisplatin dose and craniospinal radiotherapy have a positive effect on outcome, but extent of surgery appears to be less important.180 For children with relapsed disease, a regimen such as ifosfamide, vincristine, and doxorubicin may be curative when combined with surgery,171 and high-dose therapy with carboplatin and etoposide may be curative for patients with relapsed or resistant disease.181 Also, the complete resection of locally relapsed disease represents a cornerstone of management for these patients.182 In summary, although the current therapy for paediatric MGCT confers a high cure rate, further studies are needed to define, in relation to internationally agreed risk classifications, the minimal effective therapy for this group of childhood tumours.
Hodgkin’s disease Hodgkin’s disease (HD) accounts for approximately 5 per cent of childhood malignancies with an incidence of 4.6 cases per year per 1 000 000 population aged less than 15 years. HD is very unusual in the under-5-year age group and has a marked male predominance in younger children, being 10 times more common in boys under 7 years. The bimodal distribution of cases according to socio-economic group (a high incidence in low socio-economic groups with a large sibling number, and in high socio-economic groups with a small sibling number) suggests an infective aetiology. HD is more common in those with prior exposure to Epstein–Barr virus (EBV), but this is not adequate to explain all cases in the western world in particular. Current understanding is that chronic antigenic stimulation (EBV or other virus-related) leads to aberrant gene rearrangement and therefore abnormal gene expression. PRESENTATION
The majority of children will present with painless cervical lymphadenopathy. Mediastinal involvement is less common in younger children, but is seen in 70 per cent over the age of 12 years. The diagnosis of cervical lymphadenopathy in a child is a cause of much discussion: other causes of lymphadenopathy include reactive adenopathy (a large majority), infective, (including cat-scratch disease and occasionally atypical tuberculosis), EBV infection, acute leukaemia, non-Hodgkin’s lymphoma, and rarely, phenytoin-induced pseudo-lymphoma. Reactive lymphadenopathy in young children can be marked, and cause considerable anxiety, prompting biopsy. A diagnosis of HD requires a knowledge of the lymph node architecture, whilst other malignant causes of lymphadenopathy often fall into the category of small round blue
cell tumours, which will require excision biopsy. Although well established in adult practice, fine-needle aspirate is unlikely to give a diagnosis, is likely to lead to significant delay in diagnosis and can therefore never be justified in paediatric oncology. PATHOLOGY
Histologically, paediatric HD is classified according to the Rye classification, now incorporated into the WHO classification. The main categories are: lymphocyte rich, nodular sclerosing, mixed cellularity and lymphocyte depleted. Mixed cellularity type HD is more common in children than adults, and lymphocyte-depleted HD is uncommon. The unusual variant, nodular lymphocyte predominant HD seems to have a better prognosis than other forms, and in recent years an attempt to reduce the intensity of therapy for this subgroup has been made. The classic lesions, Reed–Sternberg and its mononuclear variant, the Hodgkin cell are always positive for CD30, and often for CD15, but negative for CD45. Cytogenetics are often abnormal, but a characteristic abnormality is not seen. Most will have some rearrangement of the immunoglobulin genes, or occasionally of the T-cell receptor. Differentiation from large cell anaplastic lymphoma (ALCL) can be difficult, but ALCL more commonly has systemic features such as fever and cutaneous involvement. Staging is according to the Ann Arbor classification183 (Table 46.8). Staging investigations should include chest X-ray and abdominal ultrasound, CT or MRI of the primary site, chest and abdomen. Bone marrow aspirates and trephines are indicated only where a patient is thought likely to have stage III or IV disease. The role of FDG-PET scanning in children is still developing, not least because of limited availability of this modality for children in the UK, but it will have a role in assessing residual disease in particular. TREATMENT
Treatment of paediatric HD may involve surgery, radiotherapy and chemotherapy. Surgical excision alone may have a place in low-stage NLPHD, where complete excision of the affected lymph node can be achieved without mutilation. Radiotherapy is highly effective in achieving control, and is a major component of most strategies. Radiotherapy alone is inferior to combined chemotherapy and radiotherapy for patients with advanced disease, but may be equivalent for localized stage I and II disease.184 There have been many attempts to reduce the dose and radiation fields for paediatric patients. Used as a single modality, a dose of 40–44 Gy with extensive fields, even for stage I disease is optimal. The addition of combination chemotherapy with MOPP (mechlorethamine, VCR, procarbazine, prednisone), effective in patients with highstage disease with extended field radiotherapy, has been
1140 Paediatric oncology
Table 46.8 Ann Arbor staging classification for Hodgkin disease Stage I II
III
IV
Description Involvement of a single lymph node region (I) or of a single extralymphatic organ or site. (IE) Involvement of two or more lymph node regions on the same side of the diaphragm (II) or localized involvement of an extralymphatic organ or site and one or more lymph node regions on the same side of the diaphragm. (IIE) Involvement of lymph node regions on both sides of the diaphragm (III), which may be accompanied by involvement of the spleen (IIIS) or by localized involvement of an extralymphatic organ or site (IIIE), or both. (IIISE) Diffuse or disseminated involvement of one or more extralymphatic organs or tissues with or without associated lymph node involvement
The absence or presence of fever, night sweats, or unexplained loss of 10 per cent or more of body weight in the 6 months preceding admission are to be denoted in all cases by the suffix letters A or B, respectively.
shown to be effective for patients with low stage,185–187 even with involved field radiotherapy only. The 5-year OS for both series was reported to be 93 per cent, which was equivalent to extended field radiation, and markedly better than OS with radiotherapy alone. Similar results were reported by Donaldson.188 Reduction in the overall dose of RT has also proven to be possible, often with a tailored approach, with reduction limited to good responders.189–192 The adverse side effects of chemotherapy – particularly infertility in boys – have prompted a search for different, effective agents. The hybrid regime of MOPP/ABVD is as effective as MOPP alone, with fewer long-term sequelae, although even with this regime abnormalities of lung function are likely. The combination of chemotherapy with busulphan and lung field irradiation is particularly likely to lead to falls in FEV1, vital capacity and gas exchange. Attempts to replace alkylating agents have been mixed. High rates of control with a combination of vinblastine, bleomycin, etoposide and prednisolone and radiation to 20 Gy were seen in good responders, and hence both anthracyclines and alkylating agents may be avoidable in selected groups.191 In contrast, the VEEP regime (vincristine, etoposide, epirubicin, prednisone) without adjuvant radiotherapy was reported to have unacceptably poor local control.193,194 In the UK, the combination ChlVPP (chlorambucil, vinblastine, procarbazine, prednisolone) was well tolerated, but current recommendations are to use the combinations used in the DAL-HD-90 study, OPPA (vincristine, procarbazine, prednisolone, doxorubicin [Adriamycin]) and OEPA (vincristine, etoposide, prednisolone, doxorubicin).195 The
importance of radiotherapy is recognized, but it is hoped that use of PET scanning in patients with residual disease will allow its more selective use in patients with residual disease. PROGNOSTIC FACTORS
Stage IV disease, B-type symptoms and the presence of a mediastinal mass or other bulky disease are prognostic in children, and most protocols will stratify treatment accordingly.196 Response to chemotherapy is also prognostic and stratification of therapy based on response, and more recently on PET scan activity is indicated, with radiotherapy dose and field adjusted accordingly.192 Overall 5-year survival for stages I–IIIA disease is of the order of 90 per cent. Stage IV disease has a significantly poorer outcome. Survival figures from different groups using combination treatment vary from 60–79 per cent disease-free survival. REFRACTORY AND RELAPSED DISEASE
Salvage regimens for relapsed HD offer a chance of cure. These may not necessarily involve high-dose therapy, although this has been the approach taken for many patients, particularly poor prognostic groups such as progressive disease on treatment and early second relapse.197 Recently, allogeneic transplantation has been reported to offer some benefit to patients at very high risk.198
Non-Hodgkin’s lymphoma Non-Hodgkin lymphoma is a malignant tumour arising from proliferation of lymphoid cells at various stages of activation and differentiation. In childhood, most present as extranodal disease, with rapid growth and non-contiguous spread. The large majority of lymphomas in childhood are high grade, and may be of B or T lineage. NHL has an annual incidence of 7 cases per million children. There is a male predominance and a peak incidence between the ages of 7 and 10 years. Burkitt’s lymphoma is endemic in Africa and has an association with EBV. NHL in the West may show evidence of EBV in approximately 20 per cent of cases. Certain immunodeficiency syndromes such as ataxia telangiectasia, Wiskott–Aldrich and acquired immunodeficiency secondary to HIV infection or post-transplant immunosuppression are associated with NHL. PATHOLOGY
The Revised European American classification of Lymphoid neoplasms (REAL) classification was accepted by the Société Françdaise d’Oncologie édiatrique (SFOP), the United Kingdom Children’s Cancer Study Group (UKCCSG) and
Specific paediatric malignant diseases 1141
the United States Children’s Cancer Group (USCCG).199 This has now been incorporated into the WHO classification.200,201 Classification in childhood is somewhat more straightforward than in adults, given that almost all such tumours fall into one of four categories. Morphology and immunophenotyping is necessary, since morphology alone is prone to marked variation.202 Approximately 50 per cent will be Burkitt or Burkitt-like tumours. Ten per cent will be diffuse large B-cell, and the identification of these is important because of the worse overall outcome for these patients. Thirty per cent of tumours will be of precursor T or B lymphoblastic origin, and are treated according to protocols that mirror those for ALL. Ten per cent will be ALCLs, for which a separate therapeutic strategy is also identified.
Table 46.9 St Jude staging system Stage I
II
III
CLINICAL PRESENTATION
Most patients with NHL will present with an abdominal mass or intussusception. Intussusception occurring after 3 years of age is likely to be due to lymphoma, and appropriate imaging should be performed before a surgical intervention. Some patients may present with ascites or as an acute appendicitis. Mediastinal compression and obstruction is often seen at presentation, and such tumours are typically of T-cell origin. A differential diagnosis includes HD, neuroblastoma, mediastinal germ cell tumour and mediastinal B-cell NHL. Adenopathy affecting Waldeyer’s ring and cervical chains may be present. Occasionally disease may present because of involvement of bones, kidney or epidural space causing cord compression. ALCL may present with painful lymph node involvement, skin involvement, and often systemic symptoms such as fever. ALCL may have episodes of spontaneous regression, with adenopathy waxing and waning, which is otherwise unusual for malignancy. Assessment of high-grade lymphomas must include consideration of the patient’s airway (with mediastinal disease) and renal function, given the high likelihood of tumour lysis syndrome once treatment begins. Staging of paediatric NHL is according to the St Jude’s staging system (Table 46.9). The majority of patients will present with stage 3 disease on the basis of a large mediastinal or unresectable abdominal primary. Primary CNS or marrow involvement is relatively uncommon. The principal diagnostic groups (Burkitt, Large cell B, precursor t or B lymphoblastic and ALCL) are discussed later.
Burkitt’s lymphoma and diffuse large B-cell lymphoma Burkitt’s lymphoma is characterized by a translocation involving the long arm of chromosome 8, causing transcriptional deregulation of c-myc.203 Almost all will show a
IV
Description Single tumor (extranodal) or single anatomic area (nodal), excluding mediastinum or abdomen Single tumor (extranodal) with regional node involvement, or Primary gastrointestinal tumor
associated mesenteric node involvement, with gross total resection, or on same side of diaphragm: 2 or more nodal areas, or 2 single (extranodal) tumors regional node involvement Any primary intrathoracic tumor (mediastinal, pleural, thymic), or any extensive abdominal tumor (unresectable), or any primary paraspinous or epidural tumor, regardless of other sites, or on both sides of the diaphragm: 2 or more nodal areas, or 2 single (extranodal) tumors regional node involvement Any of the above with initial CNS or marrow (25 per cent) involvement
translocation affecting t(8,14)(q24;q32) (heavy chain Ig locus), t(2,8)(q1;q24) (kappa light chain) or t(8,22)(q24;q11) (lambda light chain). Malignant cells arise from the germinal centre of the lymph node, and are phenotypically and immunohistochemically of B-cell origin, with expression of CD10, 19, 20, 22 and 79a. Tumours have an extremely high proliferative rate. Endemic Burkitt’s lymphoma is associated with EBV and frequently involves the face. Non-endemic Burkitt’s is not associated with EBV and tends to present with abdominal disease. Particular attention should be paid to a B-cell lymphoma presenting in the chest, as this may be diffuse large B-cell disease, for which the prognosis is more guarded than for Burkitt. Morphologically, both tumours show effacement of the lymph node and a ‘starry sky’ appearance, although nuclei are typically larger in the former. Diffuse large B-cell lymphoma typically shows translocations which lead to deregulation of the gene bcl-2 or bcl-6, both intimately involved in regulation of apoptosis.
TREATMENT STRATEGIES FOR B-NHL
Burkitt’s and diffuse large cell B-cell lymphomas are treated according to CHOP-based (cyclophosphamide, doxorubicin, vincristine and prednisolone) regimens. Since the 1980s the best reported survival has been reported from France, and the LMB studies have therefore provided the backbone of subsequent therapy throughout Europe, and more recently, North America.
1142 Paediatric oncology
The LMB protocols began in 1981 and consisted of cytoreduction (cyclophosphamide, vincristine and prednisolone), intensive induction (cyclophosphamide, vincristine, prednisolone, doxorubicin and high-dose methotrexate) and consolidation (high-dose cytarabine infusions).204 All patients receive CNS-directed therapy with intrathecal chemotherapy and high-dose methotrexate. Patients with high-risk disease (primary CNS or bone marrow involvement) would also receive additional maintenance chemotherapy. These studies have been highly effective, with event-free survival in excess of 90 per cent for patients without CNS involvement, and over 80 per cent with CNS involvement. Given such impressive results, it has been possible to reduce the intensity of treatment for the majority of patients, with the use of less anthracyclines and less cyclophosphamide.205 At the other end of the spectrum there is evidence that dose intensification improves outcome in stage IV B-cell disease.206 Patients who show nonresponse to COP, and those who did not achieve a CR after induction therapy have a poor prognosis (efs 22 per cent for poor responders). The importance of initial response has prompted interest in the pharmacokinetics of induction chemotherapy agents, in particular, cyclophosphamide. The role of megatherapy in B-cell disease remains unclear. Relapsed stage I and II disease is salvageable with additional cycles of conventional chemotherapy. Megatherapy using BEAM (BCNU, etoposide, cytosine arabinoside and melphalan) is reserved for relapse of more advanced disease or disease poorly responsive to conventional doses. Relapsed stage IV disease is virtually incurable except for isolated CNS relapse. The role of rituximab (monoclonal anti-CD20 antibody) in paediatric patients with NHL is limited, given the high efficacy of current regimens, although its use in adult high-risk patients may lead to inclusion in studies in coming years.
Precursor T- and B-cell lymphoblastic lymphoma Precursor T- and B-lymphoblastic lymphoma is best regarded as a localized leukaemia. Cells are medium-sized, FAB L1 or 2 lymphoblasts and stain according to B- or T-lineages (B lineage positive for CD19, CD79a, negative for surface immunoglobulin, cytoplasmic μ , and T lineage positive for CD7, CD2 and CD3c). T-cell disease and precursor lymphoblastic B-cell disease is treated using leukaemia protocols. The most effective of these have been piloted by the BFM group, and form the basis of the current European approach. Patients receive a cytoreductive pre-phase with prednisolone and intrathecal methotrexate. They then receive induction chemotherapy similar to that seen in current ALL treatment protocols, followed by ‘protocol M’, comprising 6-mercaptopurine and high-dose methotrexate. Patients with stage III or IV
disease undergo further reinduction, followed by a maintenance phase, randomized to 2 or 3 years. This approach, closely modeled on NHL-BFM 90 is expected to lead to a high rate of overall cure. BFM-90 reported an event-free survival of 82 5 per cent at 5 years.207 Poor initial response was taken as an adverse prognostic factor, but was uncommon, and with intensified therapy did not lead to worse outcome. Failures for patients with T-cell disease all occurred in the first 2 years, whereas those for B-cell disease occurred up to 4 years after diagnosis. There is a randomization in the study comparing prednisolone and dexamethasone, as has been studied already in ALL, and between treatment duration of 2 years or a reduction to 18 months from diagnosis.
Anaplastic large cell lymphoma ALCL accounts for around 10 per cent of childhood NHL. Patients may present with mediastinal or abdominal disease, but there is a greater likelihood of systemic symptoms and cutaneous involvement. Histologically, cells may resemble carcinoma or other undifferentiated tumours, Hodgkin’s and other forms of NHL. Immunophenotyping is necessary to make the diagnosis for most. Cells are characteristically positive for CD30 and epithelial membrane antigen (EMA).208 Cells have T-helper phenotype (CD3 and CD4 positive). The translocation t(2;5) is specific to ALCL, and is found in at least 80 per cent of patients.209 The translocation leads to expression of a novel gene product, NPMALK, and ALK-1 positivity appears to be associated with a better clinical outcome for systemic disease. Both localized ALCL and primary CNS involvement are uncommon. B-symptoms, however, are common and staging according to conventional systems is not helpful. ALCL is stratified instead according to histology and site of disease in to standard-risk and high-risk (with skin or visceral organ involvement) groups. The current UK strategy adopts an approach based upon previous BFM protocols, for which efs of 79 per cent was reported at 3 years.210 Patients receive multi-agent chemotherapy with dexamethasone, cyclophosphamide, vincristine, doxorubicin, ifosfamide, cytarabine and etoposide, with a dual randomization to look at the effect of vinblastine, and low (1 g/m2) or high (3 g/m2) dose methotrexate.
CNS TUMOURS IN INFANTS AND YOUNG CHILDREN Approximately 12–15 per cent of all childhood CNS tumours occur in children younger than 2 years of age, and two thirds of these are supratentorial in location.211 The predominant histological diagnoses are medulloblastoma-PNET (30–40 per cent), ependymoma (8–28 per cent) and low-grade gliomas (21–36 per cent) and malignant glioma (15 per cent). Because infants and very young children with CNS
CNS tumours in infants and young children 1143
tumours are at risk of catastrophic long-term intellectual sequelae if radiotherapy is administered at such an early age, chemotherapy schedules have been investigated with the aim of delaying, or even avoiding, radiotherapy. Duffner et al.212 described effective postoperative chemotherapy with cyclophosphamide/vincristine and cisplatin/etoposide that delayed the need for radiotherapy in a significant number of cases. The overall PFS was 33 per cent at 2 years, and children with medulloblastoma, ependymoma and malignant glioma fared better than children with supratentorial PNET. Moreover, a PFS of approximately 45 per cent has also been described for children with anaplastic astrocytoma who received treatment with the 8-in-1 regimen prior to planned radiotherapy.213 This indicates that chemotherapy may confer considerable benefit for this group of patients even without radiotherapy. In the case of infants with medulloblastoma and PNET, improvement in survival with intraventricular methotrexatecontaining multi-agent chemotherapy, which also avoids radiotherapy, has been described and offers promise for this difficult patient group.214 As a group CNS tumours represent the second most frequent malignancy in children under the age of 15 years, with an approximate annual incidence in the USA of 2.8 cases per 100 000 children per year. Within the first 10 years of life a predominance of embryonal CNS neoplasms is found, with a relative absence of gliomas. After early adolescence there is an increase in the incidence of typically adult CNS tumours. The WHO classification for paediatric CNS tumours (Table 46.10) recognizes both traditional morphological entities and degree of anaplasia, as well as providing for the designation of location within the CNS.215 CLINICAL PRESENTATION
The clinical presentation of children with CNS tumours varies considerably with age, development and the site of
origin of the tumour. Infratentorial (brainstem and cerebellar) tumours may present with disturbances of truncal steadiness, upper extremity co-ordination and gait and cranial nerve function. Children with supratentorial tumours may present with features of raised intracranial pressure, irritability, seizures, regression of developmental milestones and upper motor neurone signs such as hemiparesis. Tumours of the optic chiasm may result in visual field defects such as a bitemporal hemianopia, nystagmus and head tilt. In addition, hypothalamic tumours may give rise to the diencephalic syndrome (failure to thrive, euphoria and hyperactivity), and endocrine disorders such as diabetes insipidus, hypogonadism and precocious puberty.216 Magnetic resonance imaging (MRI) scanning is now the investigation of choice for tumours of the central nervous system in childhood.
Glial tumours Glial tumours are classified as Grade I (pilocytic astroctoma), Grade II (fibrillary astrocytoma), Grade III (anaplastic astrocytoma) and Grade 4 (glioblastoma multiforme). The designation of a glial tumour as low grade (I and II) and high grade (II and IV) carries an important impact in terms of both treatment and prognosis.216 LOW-GRADE GLIOMAS
Low-grade gliomas comprise the most common form of CNS tumour in children,216 and factors such as site and operability define the balance of the treatment modalities of surgery, chemotherapy and radiotherapy and also prognosis. In their review of a decade of experience at St Jude Children’s Research Hospital, Gajjar et al.217 found lowgrade gliomas to arise from the cerebral hemispheres (20 per cent), cerebellar hemispheres (35 per cent), hypothalamus (12 per cent), thalamus (12 per cent), brainstem
Table 46.10 Classification of childhood CNS tumours Glial tumours
Astrocytic astrocytoma Anaplastic astrocytoma Subependymal giant cell tumour Giagnatocellular glioma
Neuronal tumours
Primitive neuroectodermal tumours (PNETs)
Pineal cell tumours
Gangliocytoma Ganglioglioma Anaplastic ganglioglioma
PNET not otherwise specified PNET with differentiation (astrocytic, ependymal, neuronal)
Pineocytoma Pineoblastoma (PNET)
Medulloepithelioma Oligodendoglioma Anaplastic oligodendroglioma Ependymoma Anaplastic ependymoma Myxopapillary ependymoma
1144 Paediatric oncology
(12 per cent), spinal cord (4 per cent) and optic nerve/chiasm (3 per cent). Although the median age at diagnosis was 7 years, 32 per cent of children were younger than 5 years at diagnosis. In addition, although the overall 4-year survival rate was 90 per cent at 4 years, children below the age of 5 years had a poorer progression-free survival (PFS) than those diagnosed after the age of 5 years, and this difference was most pronounced in children with hypothalamic or thalamic tumours. Surgery is the mainstay of therapy for low-grade glioma of the cerebral hemispheres218 or cerebellum,217 with radiotherapy or chemotherapy reserved for recurrent or progressive inoperable disease. Although there have been no randomized trials to compare radiotherapy and observation alone for residual tumours,216 local field radiotherapy, at a dose of 54 Gy administered over 30 fractions, is generally used for incompletely resected and progressive lowgrade glioma. Although effective for the long-term control of low-grade glioma, radiotherapy to areas such as the hypothalamus/optic chiasm is known to result in vasculoapthy (especially for young children and those with neurofibromatosis-type 1), endocrinopathy and cognitive impairment.219 More recently, conformal radiotherapy techniques have been employed with the hope of reducing the late effects burden for children with low-grade glioma who require this treatment modality.220 For tumours of the optic chiasm/hypothalamus, curative surgery is not usually possible, and surgery is usually indicated to debulk symptomatic tumours, relieve obstruction at the foramen of Monro and to obtain a diagnostic biopsy. Of children with optic chiasm/hypothalamus, 50 per cent present before the age of 5 years, and although approximately 33 per cent of these children have NF1 and a more indolent course, the majority of cases progress within 6 years of diagnosis.221 Chemotherapy with vincristine and carboplatin been shown to be effective in controlling disease progression at this site, thereby postponing radiotherapy until after the age of 5 years in the majority of cases.221 However, evidence is emerging that children below the age of 1 year at diagnosis and with non-pilocytic histologies are at an increased risk of treatment failure with vincristine and carboplatin.222 Although therapy with vincristine and carboplatin remains the mainstay of first-line therapy for children with low-grade glioma, experience with combinations such as cisplatin/etposide,224 thioguanine/CCNU/procarbazine/vincristine224 and weekly vinblastine are defining treatments with potential second-line use or as an alternative to carboplatin in the face of the development of hypersensitivity to this drug.225 HIGH-GRADE GLIOMAS
High-grade gliomas comprise 7–11 per cent of childhood CNS tumours.211 The vast majority are supratentorial, with 63 per cent occurring in the superficial cerebral hemisphere, 28 per cent in the midline cerebrum, and 8 per cent in the posterior fossa.226 Unlike adult high-grade gliomas, very
few of these tumours in children demonstrate p53 mutation, LOH for chromosome 10 and amplification of epidermal growth factor receptor.216 However epidermal growth factor receptor is expressed by high-grade glioma in children, both for non-brainstem227 and brainstem tumour types228 and may play a role in the pathogenesis of this cancer in children. The prognosis for high-grade glioma has been shown to relate to the extent of surgical resection.226 Thus, for children with anaplastic astrocytoma, 5-year PFS rates of 44 per cent and 22 per cent were observed for children achieving a radical resection and less extensive resection, respectively. Similarly, for children with glioblastoma multiforme (GBM), 5-year PFS rates of 26 per cent and 5 per cent were observed for children achieving a radical resection and less extensive resection, respectively. The role of chemotherapy in the management of high-grade glioma in children is uncertain at present. However, treatment regimens that incorporate multi-agent chemotherapy generally show a small (approximately 10–15 per cent) improvement in survival over that which results from surgery and radiotherapy alone and in the context of a good surgical resection initially.229,230 Similarly, the benefit of high-dose chemotherapy with thiotepa and etoposide for the treatment of children with high-grade glioma at presentation or relapse remains uncertain, but may play a role where a complete remission is achieved prior to highdose therapy.216 Moreover, a recently published study from the Milan group has reported encouraging survival rates of 45 per cent for children who received pre-irradiation chemotherapy with cisplatin, etoposide, cyclophosphamide, vincristine, high-dose methotrexate and myeloablative therapy with thiotepa. Vincristine and CCNU were employed as post-irradiation therapy also.231 However, the majority of contemporary studies demonstrate that, particularly in the setting of relapsed disease, high-grade glioma has a poor prognosis in the face of therapy with alkylating agent/etoposide combinations,232 temozolomide233 and oral topotecan.234 Tumours arising in the midbrain, pons and medulla oblongata account for 10–20 per cent of all childhood CNS tumours, and usually present in children between the age of 5–9 years.211 High-grade gliomas of the brainstem are chemoresistant and radioresistant, and although clinical improvement is observed in the majority of children with 54 Gy radiotherapy, this is not improved by a hyperfractionated dose of 70 Gy,235 and the tumours continue to have an extremely poor prognosis.236 Therefore, improvements in survival for high-grade glioma are likely to at least partly result from the development of drugs that interact with molecular targets such as vascular endothelial growth factor.
Ependymomas Ependymoma accounts for 6–12 per cent of CNS tumours in childhood, with 50 per cent of cases occurring before the
CNS tumours in infants and young children 1145
age of 5 years.237 Ependymomas develop from the neuroepithelial lining of the ventricles and central canal of the spinal cord and are classified as subependymoma and myxopapillary ependymoma (grade I tumour), low-grade ependymoma (grade II) and anaplastic ependymoma (grade III). However, the relationship between tumour anaplasia and prognosis is uncertain.237 The clinical presentation of ependymoma is usually due to the signs and symptoms of raised ICP, although posterior fossa ependymomas commonly adhere to and invade the brainstem causing palsies of cranial nerves VI–X. Surgical resection is the mainstay of therapy for ependymoma, and the importance of the extent of surgical resection has been demonstrated in multi-centre trials of therapy for this disease.238,239 For example, children with no evidence of disease on imaging postoperatively have a 5-year PFS of 60–70 per cent, which compares with 5–10 per cent for children with residual macroscopic disease.240 However, for children with supratentorial tumours, a complete resection is possible for approximately 50 per cent of cases, which compares with only 30 per cent for children with infratentorial tumours.237 Thus, second-look surgery is increasingly advocated in order to achieve a radiological complete remission.239 Although no randomized study has been conducted to measure the benefit of radiotherapy in the treatment of ependymoma, local radiotherapy remains the standard postoperative treatment for this tumour.234 The role of chemotherapy in the treatment of ependymomas remains unknown, and there are no reports that indicate a survival advantage for chemotherapy when compared with historical controls, or patients receiving radiotherapy alone.237 However, encouraging pilot results have been obtained with the use of carboplatin, vincristine, ifosfamide and etoposide in addition to radiotherapy.241 As with highgrade glioma, there is no clear role for high-dose chemotherapy in the treatment of ependymoma.237 In the setting of relapsed disease, chemotherapy with oral etoposide has been shown to confer some benefit,242 and re-operation may also prolonged periods of disease control.243
Other glial tumours Two recently described glial tumours include the dysembryoplastic neuroepithelial tumour (DNET) and ganglioglioma. These tumours have limited growth potential and tend to occur in infants and young children, and the tumours may cause intractable epilepsy if they arise in the frontal or temporal lobes.216 The prognosis for these tumours is good, and surgery can result in improvement in seizure control for both ganglioglioma244 and DNET.245
Primitive neuroectodermal tumours Primitive neuroectodermal tumour (PNET) is the most common malignant primary brain tumour of childhood,
comprising approximately 30–40 per cent of such tumours, and affecting 0.5 per 100 000 children each year. The peak incidence is at the age of 7 years, although cases are reported into late adult life. There is a slight male preponderance. The most common site is in the posterior fossa, and such tumours are referred to as medulloblastoma or PNET-MB. Biologically these are not different from supratentorial PNETs or pineal parenchymal tumours (pineoblastoma), although the prognosis for medulloblastoma is generally better than at other sites. PRESENTATION
Clinical features relate to the site of the tumour. Most PNETs arise in the posterior fossa, involving the cerebellar vermis and causing obstruction to the aqueduct or IVth ventricle. Clinical features will typically include truncal ataxia and signs of raised intracranial pressure. Spread is by local invasion, to the cerebellar peduncle, floor of the fourth ventricle, and cervical spine. Occasionally, rostral extension is seen. Involvement of the brainstem may give rise to cranial nerve palsies. Brainstem involvement has been reported to carry an adverse outcome, relating to the difficulty in attaining a complete surgical excision. Spread may also be through the CSF, and leptomeningeal deposits are seen in approximately 30 per cent of patients. Rarely, tumour will be present in bone marrow at diagnosis. Supratentorial PNET may cause raised intracranial pressure, or present because of epeilepsy or after acute haemorrhage. ASSESSMENT
The primary tumour should be imaged by MRI, and appears as a solid, often homogeneously enhancing mass. All patients should have imaging of the spinal cord and whole brain by gadolinium-enhanced MRI: the entire spine must be imaged in at least two planes. Preoperative assessment is preferred to postoperative, but an interval of no more than 72 hours after operation is generally agreed to be acceptable in the assessment of metastatic disease. Lumbar puncture has been shown to have a higher diagnostic value than ventricular CSF sampling, and lumbar CSF should be obtained before treatment commences. Initial tumour assessment is of importance in stratification of patients into standard- or high-risk groups, for which outcome is significantly different. Dissemination is the most powerful adverse prognostic marker, but in addition, younger age at diagnosis, incomplete resection and a supratentorial tumour indicate a likely poor outcome.246,247 Brainstem involvement has been reported to carry an adverse outcome, relating to the difficulty in attaining a complete surgical excision.248,249 In general, patients with average risk are children older than 3 years of age with posterior fossa tumors, totally or ‘near-totally’ resected and no dissemination. Patients with poor risk are children younger than 3 years of age or those
1146 Paediatric oncology
with metastatic disease and/or subtotal resection and/or non-posterior fossa location. As for all tumours, these prognostic variables must be evaluated in the context of the treatment received. It is clear that ineffective chemotherapy which delays radiotherapy will adversely affect prognosis.250
PATHOLOGY
PNET is a highly malignant embryonal tumour of the CNS, with the capacity for differentiation along diverse pathways. Cells appear as undifferentiated round cells, but may mark with antibodies to synaptophysin, vimentin, GFAP, N-CAMs, retinal S-antigen and the low affinity (p75-)NGF-receptor. The cell of origin is debated, but may be subependymal. Desmoplastic PNET is said to have a more favourable prognosis than classical PNET, possibly because it is more lobulated and therefore more readily removed, although this is not in itself used to stratify therapy.
BIOLOGICAL FACTORS AND PROGNOSIS
Approximately 50 per cent of patients demonstrate isochromosome 17q, with a consequent loss of 17p. LOH 17p has been localized to the region 17p13.3, and a candidate gene affected is the hypermethylated in cancer (HIC)-1 gene, but this remains unproven. A predisposition to medulloblastoma is a feature of Gorlin’s syndrome, in which there is mutation of the PTCH gene. Mutation of PTCH is seen in approximately 8 per cent of cases of sporadic PNET-MB, and loss of heterozygosity for the PTCH locus (9q22.3-q310) is reported in some PNET-MB. The importance of the class I receptor tyrosine kinase, ErbB2 in both tumorigenesis and prognosis has been investigated extensively. Dysregulation of both ErbB2 and ErbB4 is seen in medulloblastoma, and elevated levels of ErbB2 receptor expression were associated with reduced patient survival in patients treated by radiotherapy with or without chemotherapy.251 ErbB2 expression was associated with a 5-year overall survival of 54 per cent, compared to 100 per cent without.252 The possibility of stratification of therapy, based upon biological characteristics is close, and the possibility of reducing the overall morbidity of this tumour an achievable goal. THERAPY
Standard therapy for PNET-MB involves initial neurological stabilization of the patient, which may require external ventricular drainage, and surgical resection. Complete surgical resection is a positive feature, but aggressive resection is not recommended where this would lead to increased morbidity. Postoperative craniospinal radiotherapy is almost always recommended. Attempts to avoid radiotherapy in ‘at risk’ groups such as infants have generally been associated
with poor outcomes. In the UK, treatment of infants with chemotherapy alone was associated with a survival of less than 10 per cent, whereas long-term disease-free survival of between 55 and 70 per cent is reported with craniospinal radiotherapy and local tumour boost. Encouraging results were however reported by Rutkowski et al., in 43 children treated without radiotherapy, and the possibility remains that radiotherapy may safely be avoided with appropriate chemotherapy regimes in the future.245 In poor risk disease, chemotherapy improves survival,246,247,253 and in recent years there has been increasing acceptance of chemotherapy for patients with metastatic disease. It is accepted that chemotherapy has a role for patients with localized, completely resected disease, although the value of this depends upon the use of effective chemotherapy agents. The SIOP first study (CCNU, vincristine) showed no benefit for localized disease, although a benefit for chemotherapy was seen for patients with partial or subtotal surgery (P 0.007), brainstem involvement (P 0.001), and stage T3 and T4 disease (P 0.002).248 In the second SIOP study, no benefit from pre-radiotherapy chemotherapy (with vincristine, procarbazine and methotrexate given in a 6-week module) was seen for any group. In addition, a particularly poor outcome was seen for those children receiving chemotherapy and reduced-dose radiotherapy.250 In the SIOP third study, patients were randomized to receive immediate radiotherapy alone or ‘sandwich chemotherapy’ consisting of a 12-week regimen of four pulses of chemotherapy, followed by CSRT and a boost to the tumour bed. Despite poor recruitment to the randomization, a significant difference in PFS was seen for patients treated by chemotherapy and RT (EFS of 73 per cent at 5 years) compared with RT alone (60 per cent, P 0.04). The 5-year OS for the two arms were 76 per cent for patients treated with chemotherapy and RT, compared with 67 per cent for treatment with RT alone (P 0.17). For patients who had undergone a total resection, eventfree survival was significantly better with chemotherapy RT than RT alone (P 0.035).254 It is unlikely that any subsequent study will be performed for which chemotherapy is not a standard component, and this study represents an important milestone in the development of therapy. The best series so far was reported by the CCG, who reported a 5-year disease free survival rate of 90 per cent for ‘good risk’ patients treated with standard radiotherapy and a prolonged (48-week) course of chemotherapy with cisplatin, CCNU and vincristine.255 Only patients with no evidence of residual disease, or ambiguity regarding staging were included in this group, but clearly, such results indicate that progress can be made. The adverse long-term consequences of radiotherapy, particularly in young children are potentially profound, and attempts have been made to reduce overall radiation doses used.246,250,256,257 Whilst some studies have found non-significant differences when patients received reduced-dose radiotherapy,
CNS tumours in infants and young children 1147
more generally there is a trend towards increasing rates of relapse. A combined CCSG/POG study of craniospinal radiotherapy to a dose of 23.4 vs. 36 Gy (without chemotherapy) was halted prematurely because a statistically significant difference in early relapse of good prognosis patients. Reduced-dose radiotherapy with adjuvant chemotherapy may avoid an increased rate of relapse. This was not the case in the second SIOP study,250 but encouraging results have been reported by others.257,258
Infant PNET The treatment of young children with PNET is more problematic than for older children, given the adverse consequences of craniospinal radiotherapy at this age. Attempts to eliminate radiotherapy have for the most part been unsuccessful, with relapse-free survival of 10 per cent at best, whilst survival of infants with metastatic PNET treated with craniospinal radiotherapy approaches 40 per cent. There is, therefore, an ethical dilemma in the choice of therapy, which the physician cannot make alone. The chemosensitivity of PNET has lead to the development of high dose and high dose–intensity regimens, but as yet these include small patient numbers. The UKCCSG currently recommends patients be treated according to a rapid schedule regimen utilizing vincristine, carboplatin and cyclophosphamide and peripheral whole-blood reinfusion, and this has proven relatively well tolerated and effective in the authors’ experience, but the long-term outcome is unknown. Studies of high-dose therapy with busulphan and thiotepa are hopeful, albeit with severe toxicity.259
towards neuroectodermal tumour, sarcoma or carcinoma.269 In addition, 60 per cent of tumours show an abnormality of chromosome 22, and more recently, mutations in the gene hSNF5/INI1 have been identified in this and other tumours.270 Familial cases have been reported; Figure 46.2 was of an infant with ATRhT who had siblings with choroid plexus carcinoma and a renal rhabdoid tumour. Tumours arise in all parts of the CNS, and spread contiguously or by dissemination: approximately one third are metastatic at diagnosis. These are highly malignant tumours, and can spread through dense fibrous structures such as tentorium. EPIDEMIOLOGY
ATRhT is rare, and there are fewer than 200 cases reported worldwide. Diagnosis is made increasingly, however, and is assisted by the availability of tumour-specific cytogenetic translocations. Most tumours present in early childhood, median age 20 months, and show a preponderance of males (M F 1.6:1). The majority arise within the posterior fossa or cerebral hemisphere; rarely, they may arise in the spinal cord or be multifocal. PRESENTATION
As for all tumours, the presentation of ATRhT depends on where the tumour arises and the age of the child. These tumours tend to grow rapidly, and may be extremely invasive. Symptoms and signs of raised intracranial pressure
Relapsed PNET Recurrent PNET is rarely curable, and median survival is generally less than 1 year.234,260 Recurrent PNET-MB may show further chemoresponsiveness, and a role for etoposide in particular has been identified, both intravenously and orally.238,261 A phase I study of irinotecan showed some activity, and there may be some benefit from selective use of streotactic radiosurgery.262–264 There does seem to be a role for high-dose therapy in such patients, provided tumour remission can be achieved, and increasing reports of long-term survivors are appearing. Regimes using high-dose cyclophosphamide, busulphan, melphalan, thiotepa and carboplatin have been used.265–268 ATYPICAL TERATOID RHABDOID TUMOURS
Atypical teratoid rhabdoid tumours (ATRhTs) are rare embryonal tumours showing some similarities to renal rhabdoid tumour of infancy. This is complicated by the fact that there is an association of renal rhabdoid tumours with other CNS embryonal tumours, such as PNET, and by the presence within ATRhTs of areas showing differentiation
Figure 46.2 Sagittal MRI image with gadolinium enhancement of a patient with a primary atypical teratoid rhabdoid tumour. The tumour appears to have arisen in the posterior fossa and eroded through the tentorium.
1148 Paediatric oncology
are expected. Radiological features are not characteristic, and such tumours appear as large, hyperdense lesions that enhance strongly, reflecting a highly vascular structure. There may be haemorrhage or calcification, and occasionally, signs of leptomeningeal spread. MOLECULAR GENETICS
A high prevalence of chromosomal aberrations involving chromosome 22 is seen in ATRhT, and the gene, INI1/hSNF5 has been linked to this site. INI1/hSNF5 is a component of the ATP-dependent chromatin remodeling complex, hSWI/SNF. INI1/hSNF5 appears to cause transcriptional repression of cyclin D1 gene, through histone deacetylase. TREATMENT AND PROGNOSIS
In the series reported by Rorke, median time to progression was 4.5 months, and median OS was 6 months.271 In recent years, long-term survivors have been reported, and overall it seems that perhaps 10 per cent of patients may survive. Most successful strategies have adopted combination chemotherapy, often with craniospinal radiotherapy. The degree of surgical resection seems to have a significant bearing on outcome, and radiotherapy is usually used to achieve local control. Concerns regarding the long-term problems for infants exposed to radiation at a young age are relevant only once patients can be cured, but this is a matter currently under discussion in Europe and North America.
SUPPORTIVE CARE Acute complications of therapy for malignant disease are common, reflecting the narrow therapeutic index of most anti-cancer drugs, and the variable patterns of organ involvement of paediatric malignancies. The importance of complications lies in the immediate danger to life, as with acute infection, and with the delays and reduction of subsequent therapy often caused: intensity of treatment is a determinant of therapeutic success for many tumours. Hence there is a possibility to improve treatment outcomes for patients by optimizing therapy and reducing complications. The most frequently encountered complications are nausea and vomiting, bone marrow failure with its consequences of infection, and poor nutrition. The management of nausea, of nutritional problems and of infection is of particular importance to paediatric practice, given the high dose intensity of many regimens. Specific complications such as convulsions and other neurological problems are not specific to paediatric oncology, but the management of neurological change is perhaps most worthwhile to be discussed in the present chapter. Bone marrow failure has been dealt with elsewhere in this book, but it is important to recognize that haematological problems are central to
many emergencies in oncology, at presentation, during conventional and high-dose chemotherapy.
Nausea and vomiting Nausea and vomiting are of major significance in chemotherapy and radiotherapy, being both common and at times severe. The early and effective control of symptoms is of major importance both in determining the patient’s ability to tolerate a given regimen, and their long-term nutritional state. Compliance with therapy, especially that of adolescent patients, may be affected by the nauseating effect of some therapeutic regimens. Early control is also important because the association of nausea and vomiting with administration of chemotherapy may lead to significant problems with anticipatory symptoms. Control of nausea and vomiting is more readily maintained by prevention than by increasing therapy once it has become established. Control of nausea and vomiting may be achieved with a single drug, but more commonly, two or more agents may be necessary, depending on the emesis-inducing potential of the chemotherapy agents, and on the individual patient. Anti-emetic agents may be grouped as given later. Their use according to an emetic control ‘ladder’, as reported by Foot and Hayes is recommended.272 The high efficacy, lack of adverse side effects and ease of administration for the 5HT3 anatagonists has led to these being the favoured agents. Antidopaminergic and antihistaminic agents, corticosteroids, benzodiazepines and cannabinoids have a place in management, as for adult oncology patients. Although current experience with the newer neurokinin inhibitors is limited, favourable reports have been published in patients with otherwise difficult nausea.
Nutritional support BACKGROUND
Poor nutrition, or rather the inability to maintain nutritional status in children with cancer is common, and relates to the malignant disease, acute and long-term effects of therapy, poor oral intake, and intercurrent complications such as infection. It is a major cause of morbidity and mortality, leading to progressive wasting, weakness and debilitation. Dietary intake may present a significant psychological problem for a child and their carers. For many children, food refusal may represent the only way available to them for control over their body. This may be made worse by learned patterns of behaviour because of previous experience of nausea or vomiting, and it may prove difficult to break a cycle of gradual starvation. The incidence of protein energy malnutrition (PEM) is common, and is more common and more severe in patients with extensive, progressive or unresponsive disease273 and in those receiving intensive chemotherapy,
Conclusion 1149
abdominal or pelvic irradiation, abdominal surgery or with psychiatric depression. NUTRITIONAL ASSESSMENT OF THE CHILD WITH CANCER
Nutritional assessment should be made at the time of diagnosis, and frequently throughout the course of therapy in order to assess any immediate or subsequent need for intervention. It must be remembered that for a child, lack of weight gain constitutes a deterioration in nutritional status. Assessment should include a detailed dietary history, calculation of energy and protein intake and of the nutritional requirements, anthropometric measurement and consideration of biochemical indices. Each patient should have their height and weight, and in young children, head circumference, recorded and plotted on centile charts along with any previously recorded measurements, and this should continue at regular intervals throughout treatment. As a general rule, a weight-for-height which is less than 80 per cent of the 50th centile value for the child’s age is a reasonable indicator that nutritional intervention is required. Similarly, any recent loss of 5 per cent of body weight warrants investigation and possible intervention. Anthropometry is valuable, particularly where a child has a large tumour which may contribute to their overall body weight. Measurement of mid-arm circumference (MAC) and skin fold thickness in triceps, biceps, subscapular and supra-iliac regions allows accurate estimation of body fat deposition. Any such measurement must always be done by a trained, and if possible the same operator on each occasion. Centile charts for values at each site are available. BIOCHEMICAL PARAMETERS USED FOR NUTRITIONAL ASSESSMENT
All biochemical parameters will to some extent illustrate metabolic processes, but some are particularly useful in defining or monitoring patients requiring nutritional support. Albumin, pre-albumin, transferrin, retinol binding protein may indicate levels of protein nutrition. Blood glucose concentration is particularly important where a child is significantly underweight. Nutritional intervention has many potential benefits: the well nourished child is better able to withstand the effects of chemo- and radiotherapy,274 has a better immune response275 and is better tempered and more able to play and participate in activities of his or her family. Most importantly, the loss of more than 10–15 per cent of body weight is itself associated with significant mortality, this being due to a combination of increased susceptibility to infection, metabolic disturbance and specific deficiencies. Hyper-alimentation has been reported to allow administration of chemo- and radiotherapy on time and with fewer complications than with inadequate nutrition.276 Initial nutritional support is with high calorie normal foods, to which glucose polymers in solid or liquid form may be used. Complete nutritional supplements may also
be added, although acceptance of these is variable. Patients with significant and continuing weight loss, those receiving highly intensive therapy, and very young children, may require naso-enteric feeding. Flexible fine-bore and silk tubes are generally well tolerated, but in patients with specific pharyngeal problems causing difficulties with oral administration of feeds (including severe mucositis, oesophagitis, radiotherapy to the oropharynx) gastrostomy or jejunostomy feeding may prove valuable. Commercially produced sterile feeds are most appropriate. These are in general nutritionally complete, but the formulation must be appropriate to the specific children. Energy requirements for children receiving chemotherapy or suffering side effects from this are greater than for normal children, and oral intake may be greatly reduced. Initial target oral intake should be 50–70 per cent of estimated average requirement for energy (EAR, 1991 Department of Health figures) for age. The majority of malnourished children will gain weight on this amount, but intake should be increased to 100 per cent of EAR, or more if necessary. PARENTERAL NUTRITION
Parenteral nutrition may be indicated where weight loss continues despite enteral nutrition, or where gastrointestinal abnormalities prevent this route from being used. Most patients with these requirements will already have appropriate central venous access, although multiple concomitant medication may severely impair the ability to administer the required volumes.
CONCLUSION Protein energy malnutrition in children receiving chemotherapy is common, may be severe, and may continue for prolonged periods of time. Nutritional intervention will contribute to an improved feeling of well-being for the patient and their family. The possibility that improved nutrition may adversely affect tumour responsiveness must be balanced against the improved ability of a well child to tolerate intensive chemo- and radiotherapy, and appropriate intervention should begin at diagnosis and continue throughout the course of treatment. The long-term effects of a period of protein energy malnutrition, and of other nutritional deficiency must be considered with other sequelae of treatment.
Management of acute neurological complications Alterations in the level of consciousness of a patient, confusion, disorientation and convulsions must always be regarded as serious events. Changes in neurological states are more common in patients with brain tumours, but may
1150 Paediatric oncology
occur in any patient undergoing treatment for malignant disease. The process involved in managing such events is similar to that for any other patient, but important additional considerations must be made, reflecting a different spectrum of potential causes of such problems. Assessment of neurological function may be difficult, particularly in young children given wide variability in normal development. A change in function must be recognized and responded to promptly, however, and all staff caring for children must be appropriately trained in this. Initial assessment of the child requires quantification of the level of change of consciousness. The Glasgow Coma scale is appropriate for all children, and is familiar to most medical and nursing staff. An adaptation of this, suitable for use with children below the age of 5 years has been reported to be reproducible between observers277 and is given in Table 46.11. A low score in a child of any age indicates a high risk of further deterioration, and urgent need for investigation and treatment. Response of a child to pain only indicates a GCS of around 8, and immediate resuscitation should follow.
LATE EFFECTS OF TREATMENT The majority of children diagnosed with cancer are cured, and current estimates are that almost 80 per cent of patients will reach 5 years from diagnosis. This positive statistic is balanced by a high likelihood that survivors will have at least one chronic medical condition thoughout their life, and that even without recognized illness, many patients
will have significant physical, emotional or intellectual difficulties. Sequelae relate to the patient’s malignancy, to their treatment, to the patient and their family. The recognition of these late effects depends upon our ability to identify specific problems, which requires structured follow up for extended periods of time, and our having the appropriate tools to investigate these. The recognition of cardiotoxicity or second primary malignancy after 10 or 15 years from diagnosis is important in defining treatment strategies for patients in subsequent years, and this has been a strength of paediatric oncology since it first arose. Detailed, wellrun research studies of endocrine late effects have helped to identify the many deficits that are produced in developing children. In recent years, the impact of treatment, particularly for children with brain tumours, has been recognized to be far greater than had been thought. Neurodevelopmental and educational problems for this group are likely to cause great difficulties for the child as they move into the adult world. In contrast to the acute side effects of therapy, late sequelae for children are often more profound than for adult patients. Long-term survivors of childhood cancer will be in the most active and productive phase of their lives, and the contrast between aspiration and achievement for some of these young adults may be great. Tumour-related problems may arise at presentation, as a complication of direct organ damage, or due to secondary effects such as tumour lysis syndrome, sepsis or haematological problems at the time of diagnosis. Brain and spinal tumours may cause specific neurological deficit, and even
Table 46.11 Modified Glasgow Coma Scale. Modified James Scale277 5 years
5 years
Eye opening 4 3 2 1
Spontaneous To voice To pain None
Verbal 5 4 3 2 1
Orientated Confused Inappropriate words Incomprehensible sounds No response to pain
Alert, babbles, coos, words or sentences normal Less than usual ability, irritable cry Cries to pain Moans to pain No response to pain
Motor 6 5 4 3 2 1
Obeys commands Localizes to supra-ocular pain (9 months) Withdraws from nailbed pressure Flexion to supraocular pain Extension to supraocular pain No response to supraocular pain
Normal spontaneous movements
Specific sequelae of therapy 1151
with neurological recovery, subsequent orthopaedic deformity such as scoliosis or kyphosis may develop. Optic pathway gliomas may cause no problem other than visual failure, which may affect the individual in a profound manner. Surgery may itself cause disability if an aggressive, mutilating approach is taken. This is rarely indicated, and the prevalent view for almost all patients is to achieve a diagnosis, to remove a tumour without causing disability, and for definitive surgery to follow tumour control by chemotherapy and radiotherapy. Even for tumours such as osteosarcoma, where complete surgical resection is an absolute requirement of therapy, there is a strong focus on reduction of long-term morbidity with limb-salvage procedures. For primary CNS tumours, improved visualization and image-guidance methods have allowed better, more radical but less destructive surgery to be achieved. Radiotherapy is well known to cause widespread longterm effects, affecting any organ that receives a radiation dose. Reduction of limb or body growth, reduction of mobility at irradiated joints and early onset of degenerative conditions such as osteoarthritis and osteoporosis may be seen. Hypoplasia of affected tissues may render them more susceptible to damage. Osteopenia and osteonecrosis may cause fractures many years after treatment. Cranial radiotherapy is associated with a greater risk of stroke and second primary tumours such as meningioma or glioma. Neuro-endocrine effects of radiation to the pituitary region include loss of one or many pituitary hormones, particularly of growth hormone and thyroid-stimulating hormone. Radiation to the heart alone or in association with anthracycline chemotherapy greatly increases the risk of cardiac impairment, and of ischaemic heart disease in later life. Chemotherapy may cause numerous sequelae, although it is hoped that many of these can at least be minimized by careful monitoring of patients during active treatment. Cardiac toxicity can be avoided for most patients provided anthracycline chemotherapy is given to previously documented safe levels (300 mg/m2), and as an infusion over several hours rather than as a bolus, but this risk cannot be eliminated completely for these drugs. Cisplatin causes significant renal impairment in many patients, as both a reduction in GFR and also a loss of tubular function. Ifosfamide has similar nephrotoxic activity, predominantly with regard to tubular function. Sensorineural hearing loss from cisplatin may not recover, and is a function of cumulative dose. The effects of specific adverse side effects on the individual depend also on the developmental stage of the child. Hearing loss at an early age may cause significant disruption to language development and consequent developmental progress, even in the absence of other sequelae. The long-term consequences of therapy for primary CNS tumours have been underestimated until recently, but it is hoped that assessment of disability and intervention at a young age will allow many of these children to achieve their full potential in the future.
SPECIFIC SEQUELAE OF THERAPY Second malignancies Second malignant neoplasms (SMNs) are the second most common cause of death in survivors of primary cancer in childhood.278 A lifetime risk of between 10 and 20 times that of age-matched controls is seen, and an overall probability of between 2–12 per cent at 25 years. With more prolonged follow up, this proportion is likely to rise. Certain groups, such as survivors of Hodgkin’s disease have an actuarial risk which may exceed 20 per cent. The risk is also high in patients with cancer predisposition syndromes such as neurofibromatosis type I, Li–Fraumeni syndrome or Gorlin syndrome. In the latter, for example, which is associated with a germ line mutation of the PTCH gene, patients are at extremely high risk of basal cell carcinoma in areas of previous irradiation. Irradiation is associated with many SMNs.278,279 The majority of sarcomas, carcinomas, and brain tumours arising as SMNs (85 per cent) will arise within the radiation field. In patients with familial retinoblastoma, with mutation of the RB gene, there is a greater risk of second sarcomas, both in the radiation field and outside. Chemotherapy-related SMNs include myelodysplasia (tAML) or leukaemia, which appear to be dose dependent.280 Two main groups of tAML can be identified, associated with alkylating agents or with inhibitors of topoisomerase II. tAML following alkylator therapy will typically present several years (peak 4–7 years) after treatment, is most frequently associated with deletions of chromosomes 5 and/or 7.281 tAML associated with topoisomerase-II inhibitors typically occurs much sooner (less than 2 years from treatment) and is associated with rearrangements involving chromosome 11q23. GROWTH
Abnormalities of growth hormone (GH) secretion may be seen after doses as low as 10 Gy. Irradiation to the hypothalamus or pituitary at a dose above 30 Gy is likely to cause permanent loss of GH secretion. This may become apparent soon after treatment, but more commonly, GH deficiency will evolve slowly, depending on age, dose and fractionation of treatment.282–284 Monitoring of growth at 6-monthly intervals, until the end of puberty, and early liaison with a paediatric endocrinologist is essential for all patients at risk. Early replacement of deficient growth hormone will minimize the loss of height for an affected individual. Growth of the spine and long bones will also be affected by radiotherapy, and monitoring for secondary effects – scoliosis and limb length difference in particular – must continue until growth is complete. In addition, chemotherapy-induced growth failure is seen, in particular after long-term corticosteroid therapy. Early overgrowth
1152 Paediatric oncology
may be followed by early fusion of endplates and an overall reduction in final height. These effects may be compounded in patients, particularly those with CNS malignancy, where puberty may be entered early, leading to premature growth arrest. Finally, normal growth requires ongoing adequate nutrition. This may be impossible during intensive chemotherapy, and it is common for children to fail to grow in this phase. After completion of therapy, ongoing effects of chemotherapy, such as renal damage affecting tubular function, GFR and other functions such as vitamin D metabolism, persistent poor nutrition or other organ impairment may lead to a continuing failure to achieve normal growth. Monitoring is again essential for the period of normal growth, which may therefore be for over 15 years.
Fertility and sex hormone function Radiotherapy and chemotherapy both influence gonadal function. Damage to germ cells is common with treatment and may lead to loss of both endocrine function and fertility in females. Leydig cell failure is rare in males and endocrine function is often preserved, but spermatogenesis is more sensitive to chemotherapy and radiotherapy, such that infertility is a significant risk for many children. Even if females continue to menstruate regularly, they remain at risk of early menopause. Not only should they be advised accordingly if contemplating a family, but they will also be predisposed to developing osteoporosis and coronary heart disease and should be carefully monitored for both. Prior to commencing radiotherapy which might affect fertility, consideration of both the potential to avoid damage to or preserve germ cells should be undertaken. Whenever possible consideration of shielding or repositioning of ovaries should be contemplated before treatment. Uterine dysfunction following radiotherapy may also jeopardize future pregnancies – a factor that should be borne in mind if in vitro fertilization programmes are contemplated. Finally, the health and well being of their actual offspring is an understandable area of concern for survivors of childhood malignancy. As yet, there is little published mature data in this field, although it would appear that there is no current evidence of appreciable excess risk of either malignancy or congenital malformations.285 Cranial irradiation to the pituitary may cause damage to GnRH-secreting cells. Doses in excess of 50 Gy may causes loss of GnRH secretion, with consequent pubertal failure, whereas smaller doses (25–50 Gy) may paradoxically cause precocious puberty.282,283 In part this may arise because of associated obesity, an increase in insulin and the adipocyte hormone, leptin, but a direct central effect may also be responsible.286
Organ toxicity CARDIAC
Cardiotoxicity associated with anthracycline therapy has been well recognized for over 20 years.287 Cumulative dose, rate of infusion, administration schedule, mediastinal irradiation and young age at the time of exposure are all predictive of cardiac risk. Acute toxicities, arrhythmias or acute impairment of contractility are rare. Chronic toxicity leading to cardiomyopathy is more common and clinically the most important. Cardiac sequelae may vary from a subclinical impairment of contractility, evident on echocardiogram, to a dilated cardiomyopathy. The occurrence of sudden death in otherwise healthy young patients during childbirth or heavy exercise, particularly weightlifting emphasizes the need for long-term monitoring and caution even many years after exposure to anthracycline. RENAL
Surgical resection of one kidney will be tolerated without clinical consequence for the majority of patients, and it is doubtful whether long-term follow up of such patients is necessary. One or both kidneys may be involved by a nonrenal primary tumour, particularly neuroblastoma, and there is a recognized risk to renal vascular supply when such tumours are resected. Coincident nephrotoxic chemotherapy and radiotherapy will potentially be of greater importance for most patients. Radiotherapy may cause renal dysfunction, but is less likely if the dose is restricted to less than 20 Gy. More commonly, chemotherapy – particularly cisplatin and ifosfamide – may lead to long-term problems. Cisplatin causes glomerular and tubular dysfunction, whereas ifosfamide most commonly affects tubular function. The risk for both of these agents is related to administration and total dose.288 Methotrexate and high-dose cyclophosphamide may also cause significant renal dysfunction. PULMONARY
Although limited in its use in the paediatric field, radiotherapy commonly results in pulmonary toxicity, affecting both the lung directly and subsequent growth of the chest wall. Acute pneumonitis and long-term clinically significant restriction is uncommon provided the radiation dose is less than 30 Gy, but radiation to the chest is always used cautiously in paediatric patients. The risk of later malignant change in breast or thyroid, and of the marked increase in cardiotoxicity seen with radiation and anthracyclines are significant concerns. Bleomycin, a component of germ cell tumour chemotherapy regimes may cause acute pneumonitis, and potentially long-term fibrosis. BCNU is well known to cause
Specific sequelae of therapy 1153
fibrosis, and is rarely used in childhood cancer, but CCNU is an important component of therapy for CNS primitive neuro-ectodermal tumours and for some astrocytomas. High-dose chemotherapy is associated with a risk of pulmonary damage. The aetiology is often unclear, but exposure to respiratory viruses – particularly adenovirus – use of high-dose busulphan and of radiotherapy to the lungs are associated with development of pulmonary inflammatory disease and bronchiolitis obliterans with obstructive pneumonia within 1 year of treatment. Although reversible if mild, this condition may progress and lead to death.
Neurocognitive effects EDUCATIONAL
Disruption of education as a consequence of treatment and ill health may have secondary effects on attendance at school, and subsequent performance, but in the long term for most children, this risk seems to be small. In contrast, children who have recovered from primary brain tumours are at significant risk, and it is appropriate for these children to have specific support at an early stage. Although most will complete secondary education, there is a far smaller proportion who move on to university or higher education courses.289,290 Many survivors, particularly those who were young at diagnosis and those who had brain tumours or who received CNS directed therapy (including intrathecal methotrexate) will need special educational support. Social isolation may develop because of time off school, physical limitations or the experience of a lifethreatening event, and this may lead to secondary difficulties in later life. The effects of a tumour are variable and depend upon tumour location, the age and developmental stage of the affected child and the therapy needed to achieve cure. Hypothalamic tumours in particular are likely to cause significant intellectual problems, presumably as a result of the critical importance of this region in aspects of memory, and close proximity of visual pathways in particular. The presence of hydrocephalus and the need for CSF diversion were associated with late cognitive impairment in some series but not others.291 Patients who have a complicated recovery period postoperatively are also at risk of more severe late cognitive and psychological sequelae. Young age at diagnosis, neurofibromatosis type 1, hydrocephalus at presentation, surgical complications, increasing interval since treatment and importantly, the use of cumulative dose and volume of cranial irradiation were all described as ‘certain’ risk factors for cognitive impairment.292 Socio-economic status, supratentorial location, chemotherapy and complications of treatment, including growth hormone deficit, hypothyroidism and epilepsy were classified as ‘probable’ risk factors.
Radiotherapy is often necessary for these children, and the long-term effects of radiotherapy in intellectual development are now well documented. The effects are a function of the site and extent of the radiation field and appear to be more significant in very young children, below the age of 7 years, and most marked in those under 3 years.293–297 Very significant reduction in IQ may be seen after craniospinal radiotherapy, not found in patients undergoing surgery alone, such that a child may have great difficulty to remain integrated in society in adult life. Radiation dose and the total volume of brain irradiated are important, and there is therefore a dilemma for many children, their parents and their physicians. There is no ‘safe’ age for radiotherapy in childhood, and a balance between the benefit (in terms of likely cure) and harm (in terms of neuropsychological impairment, neuro-endocrine and other sequelae) needs to be made for each child. Attempts have been made to reduce overall radiation doses used.246,250,257,258
QUALITY OF LIFE
Measurements of morbidity following survival from childhood cancer are most often objective, documentary records based upon physical findings. The survivor’s subjective view is at least as important, but has been relatively underreported due to difficulties of definition, and limited availability of appropriate tools. This has begun to change, and in recent years, substantial progress made. Indeed, major clinical trials sponsors such as the Medical Research Council and National Cancer Institute, require new trials to consider quality of life (QOL) assessments as part of their central strategy.298,299 QOL is determined by a person’s disabilities, their abilities, in terms of achieving those things that are important to them, their expectations and achievements. It is influenced by circumstances, relationships and the person’s physical and mental health. Many of the sequelae discussed earlier will affect QOL, but the overall impact on the patient may be extremely variable. QOL is appropriately expressed in terms of fields relating to aspects of normal life. In the field of physical functioning, the majority of survivors have been reported to have one or more physical late effects,300,301 but the majority of survivors report good physical health. The exceptions are survivors of CNS, bone and other sarcomas.302–305 Fatigue has been reported to be of variable significance in survivors of cancer. In contrast to studies from adult survivors, there seemed to be no excess in fatigue in the study of Langeveld, or for patients with previous leukaemia reported by Meeske et al.306,307 A greater prevalence of psychological problems compared to population norms and normal or sibling controls has been reported by some authors,308,309 although no difference, or indeed a lower prevalence has been reported.310 Gray et al. found that survivors reported significantly more
1154 Paediatric oncology
positive affect, more perceived personal control, and greater satisfaction with life than controls.311 Post-traumatic stress disorder (PTSD) may help to explain the psychosocial consequences of having cancer as a child.312, 313 A prevalence of between 4.5 and 25 per cent of survivors has been reported. Interestingly, post-traumatic stress symptoms seem to be more common in parents of survivors, than in the survivors themselves.313
PALLIATIVE CARE The Joint Working Party of the Association for Children with Life-Threatening or Terminal Conditions and their Families (ACT) and the Royal College of Paediatrics and Child Health 1997,314 defined palliative care for children with life-limiting conditions as: ‘an active and total approach to care, embracing physical, emotional, social and spiritual elements (1997). It focuses on enhancement of quality of life for the child and support for the family and includes the management of distressing symptoms, provision of respite and care through death and bereavement.’ It is clear that palliative care needs to address the whole of a child’s life, including the extended family, all of whom will affect and be affected by the dying of a child. The transition from optimism, looking for cure, to an acceptance that this is not possible will be gradual, and may proceed at different rates for different members of a family. This may lead to difficulties, if one person’s expectations are markedly at odds with those of the child or a close relative, or indeed the members of the palliative care team. End-of-life decisions are more challenging for parents than for those based around other areas of treatment,315 and this may lead to exploration of wide-ranging, sometimes inappropriate therapies. It has been reported that parents of children with cancer have difficulty in distinguishing between research and treatment and consequently can feel that they are not in control of the choices available.316 It is a role of the physician to move with the child and their family, and to support them at each stage. Symptom control for palliative care in a child is similar to that for any child or indeed, adult. Analgesia is of great importance, and many children will progress quickly from non-steroidal analgesics, paracetamol, and local measures to needing opiates. These may not always be effective, and careful use of agents that address local disease, and appropriate consideration of the pathological processes that are going on is essential in children as for all patients. Steroids may be rapidly effective in patients with raised intracranial pressure, with bone pain secondary to marrow infiltration, or with neuropathic pain. Neuropathic pain, in particular, may present difficulties in management. It arises in most patients as a result of infiltration or compression of the
spinal cord or nerve roots, causing shooting or burning pain and sometimes, altered sensation (dysaesthesia) or increased sensitivity to normal stimuli (allodynia).317 Co-analgesics, such as tricyclic antidepressants or anticonvulsant drugs may be helpful.318–320 Transcutaneous electrical nerve stimulation (TENS), and anxiolytic agents may also allow better control for resistant pain. The importance of continuing to provide the child with appropriate play should not be underestimated as a means of reducing the experience of pain.321 The majority of children will die at home in the UK. For many children this is by far the preferred option for the child and their parents alike, but for adolescents in particular, physical difficulties and the return to dependency upon parents may mean that a hospital or hospice environment is preferred.
PSYCHOLOGICAL ASPECTS OF CHILDHOOD CANCER The psychological impact of the diagnosis of cancer in a child is enormous. There will be feelings of fear, guilt, anxiety, anger, denial, hopelessness and depression. It is the responsibility of the multidisciplinary team to support the child and their family. This should include doctors, nurses, psychologists, the clergy and social workers. The parents should be together, to support one another, when the doctor explains the diagnosis. Discussions with the family should be in the presence of an experienced nurse or social worker, who may later help them assimilate the information. Information should be repeated and reinforced where possible with written material. The initial response of parents will be grief, despair and helplessness. They may feel disempowered as parents and an experienced medical team needs to recognize their feelings, reassure them that they are appropriate, normal responses and help them see the importance of their role in the child’s care.
Talking to the child Talking to the child with cancer is difficult for parents and they look to the extended team for guidance. The approach will depend on the child’s chronological age, their developmental age and their past experiences. Children may be unable to express their feelings in words, so need to explore other means of communication such as art therapy, music therapy and play therapy. Prior to 1970 most of those dealing with childhood cancer believed that unless a child was older than 10 years, they were incapable of understanding death and therefore did not experience anxiety about it. It is clear now that even young pre-school age children have a concept of the seriousness of their illness. Those children who had not had the opportunity to express their feelings may demonstrate a psychological distance from those around them, leaving them isolated. Play therapists and
Psychological aspects of childhood cancer 1155
child psychologists are important members of the wider multidisciplinary team for every oncology department dealing with children. The child with cancer should be well informed at a level appropriate to their age and understanding. This is a legal requirement of care. The Children’s Act of 1989 in the UK and the United Nations Convention on the Rights of the Child (also 1989) exhort those involved in the care of children to inform the child of their situation, to solicit their opinion and when appropriate to regard their opinion as determinative. Clearly this may seem a huge responsibility for the child. They will look for advocates such as their parents, nursing staff or a social worker to help support them with making decisions regarding their care. By not soliciting the child’s opinion, their autonomy is violated and the natural pathway to emerging independence in adult life is interrupted.
Educating the child with cancer Continuing to attend school whilst receiving chemotherapy or radiotherapy allows the child to keep up with and maintain identity within their peer group. It presents an image of hope to teachers, classmates, parents and the child and enhances social and coping skills to help the child better deal with their illness. Classmates will also need counseling and ongoing support. Teachers need to be supported in supporting the child, and for most, specialist oncology nurses provide this. The child needs realistic goals to prevent further loss of self-esteem through failure, and often attendance can only be part time because of fatigue, illness or attendances at hospital.
they should be kept well informed. There are likely to be repeated episodes where the affected child and one or both parents will be away from home. The sibling left behind will deal with this better if they are aware of why it is happening. There are a number of support groups for siblings. In general, the support available to the affected child is also extended to the siblings.
Ambulatory paediatrics The multidisciplinary paediatric oncology team members have been the pioneers of ambulatory paediatrics. Minimizing time spent in hospital has been for the psychological benefit of the child and their family rather than the economic benefit to the health service. In fact, the resources needed to provide such a service are enormous. Specialist domiciliary care nurses do far more than just travel to the child’s home or school to take blood samples or administer treatment. Integral to the success of ambulatory care is extensive education of all those involved in caring for the child. This includes parents, teachers and general practitioners. In addition, the child and their family need to know that they have rapid and easy access to medical support.
KEY LEARNING POINTS ●
●
Parent support groups There are a number of parent support groups in existence. Early in 1999 a National Alliance of Childhood Cancer Parents Organizations (NACCPO) was formed to model the International Confederation of Childhood Cancer Parent Organizations (ICCCPO). Such organizations link parent groups in the same way professionals are linked through national and international organizations, such as the United Kingdom Children’s Cancer Study Group (UKCCSG) and Société Internationale d’Oncologie Pédiatrique (SIOP). In addition, NACCPO seeks to be the voice of parents and families of children affected by cancer at a national level through interactions with government, health authorities and social services, to communicate the issues and represent their needs.
Supporting siblings The siblings of a child with cancer will be significantly affected. Again, according to their age and understanding,
●
●
●
Cancer in childhood is uncommon, but is the second most common cause of death. This group of patients utilizes substantial resources in secondary and tertiary hospitals. Predisposition syndromes account for a minority of cases, but it is important to consider these for the child and other family members. Acute lymphoblastic leukaemia has a high likelihood of cure. Stratification of therapy according to minimal residual disease means that therapy can be reduced for good responders, and intensified for high risk patients. Paediatric tumours are diverse, and appropriate diagnosis requires a combination of morphology, immunohistochemistry and molecular genetics with an experienced paediatric pathologist. Identification of tumourspecific translocations is valuable diagnostically for many tumours, and may allow stratification of therapy according to MRD in the future. Stratification of therapy according to conventional risk factors, and increasingly, biology is central to the approach for most paediatric tumours. For example, amplification of the Myc-N oncogene remains the most important biological risk factor for patients with neuroblastoma. Tumour necrosis after
1156 Paediatric oncology
●
●
●
●
●
●
chemotherapy is of great importance in Ewing family tumours. PET positivity is used to stratify treatment of Hodgkin’s disease. For most tumours, progressive intensification and dose intensification of therapies has given improvement in overall survival. Osteosarcoma and high grade astrocytoma have shown only small improvements. The majority of paediatric non-Hodgkin’s lymphomas are high grade, Burkitt or Burkittlike, for which a high likelihood of cure is expected with current French-American-British strategies. Delays in treatment of even a few days may be associated with increased risk of relapse and should be avoided at all costs. Most primary brain tumours in childhood are glial. CNS metastasis is relatively uncommon, but may change as a result of improvements in survival of patients with high risk tumours. Low grade astrocytoma affecting the posterior fossa is expected to be cured. Visual pathway LGG is unlikely to be life threatening, but causes severe morbidity and treatment requires extensive multi-disciplinary involvement. Improving survival in ependymoma therapy has arisen from better surgical resection and radiotherapeutic technique. In infants, chemotherapy to delay radiotherapy, aiming to reduce intellectual impairment has been adopted by most european centres, but radiotherapy appears to be still important for most. Developing understanding of biological risk factors for CNS PNET will allow risk stratification, such that long term morbidity of treatment may be reduced for good prognosis is patients. Reduction of whole brain radiation dose to 24 Gy has not been associated with unacceptable increase in relapse, and further reductions may follow.
5
6
7
8 9 10
11 12
13
14
15 16
17
18
REFERENCES 1 Stiller CA, Allen MB, Eatock EM. Childhood cancer in Britain: The national registry of childhood tumours and incidence rates 1978–1987. Eur J Cancer 1995; 31A:2028–2034. 2 Linet MS, Ries LA, Smith MA, Tarone RE, Devesa SS. Cancer surveillance series: recent trends in childhood cancer incidence and mortality in the United States. J Natl Cancer Inst 1999; 91:1051–1058. 3 Robison LL. General principles of the epidemiology of childhood cancer. In Pizzo, P.A., Poplack, DG (ed.), Principles and practice of pediatric oncology. Lippincottt-Raven Publishers, 1997; Philadelphia, 1–10. 4 Plon SE, Peterson LE. Childhood cancer, hereditary and the environment. In Pizzo PA, Poplack, DG (ed.), Principles and
19 20
21
22
practice of pediatric oncology. Lippincott-Raven Publishers, 1997; Philadelphia, 11–36. Samuelsen SO, Bakketeig LS, Tretli S, Johannesen TB, Magnus P. Head circumference at birth and risk of brain cancer in childhood: a population-based study. Lancet Oncol 2006; 7:39–42. Lapunzina P. Risk of tumorigenesis in overgrowth syndromes: a comprehensive review. Am J Med Genet C Semin Med Gene 2005; 137:53–71. Merks JH, Caron HN, Hennekam RC. High incidence of malformation syndromes in a series of 1,073 children with cancer. Am J Med Genet A 2005; 134:132–43. Pritchard-Jones K. Genetics of childhood cancer. Br Med Bull 2006; 52:704–723. Ruteshouser EC, Huff V. Familial Wilms tumor. Am J Med Genet C Semin Med Genet 2004; 129:29–34. Matsui I, Tanimura M, Kobayashi N, Sawada T, Nagahara N, Akatsuka J. Neurofibromatosis type 1 and childhood cancer. Cancer 1993; 72:2746. Narod S, Stiller C, Lenior G. An estimate of the heritable fraction of childhood cancer. Br J Cancer 1991; 63:993. Lindsey JC, Lusher ME, Anderton JA, et al. Identification of tumour-specific epigenetic events in medulloblastoma development by hypermethylation profiling. Carcinogenesis 2004; 25:661–8. Alaminos M, Davalos V, Cheung NK, Gerald WL, Esteller M. Clustering of gene hypermethylation associated with clinical risk groups in neuroblastoma. J Natl Cancer Inst 2004; 96:1192–3. Oliveria SA, Saraiya M, Geller AC, Heneghan MK, Jorgensen C. Sun exposure and risk of melanoma. Arch Dis Child 2006; 91:131–8. Stiller CA. Epidemiology and genetics of childhood cancer. Oncogene 2004; 23:6429–44. McNally RJ, Eden TO. An infectious aetiology for childhood acute leukaemia: a review of the evidence. Br J Haematol 2004; 127:243–63. Feitelson MA, Pan J, Lian Z. Early molecular and genetic determinants of primary liver malignancy. Surg Clin North Am 2004; 84:339–54. Gilham C, Peto J, Simpson J, et al. Day care in infancy and risk of childhood acute lymphoblastic leukaemia: findings from UK case-control study. Br Med J 2005; 330(7503):1279–80. Greaves MF, Maia AT, Wiemels JL, Ford AM. Leukaemia in twins: lessons in natural history. Blood 2003; 102(7):2321–33. Armstrong SA, Staunton JE, Silverman LB, et al. MLL translocations specify a distinct gene expression profile that distinguishes a unique leukaemia. Nat Gene 2002; 30(1):41–7 Clarke M, Gaynon P, Hann I, et al. CNS-directed therapy for childhood acute lymphoblastic leukaemia: Childhood ALL Collaborative Group overview of 43 randomized trials. J Clin Oncol 2003; 21(9):1798–809. Eden OB, Lilleyman JS, Richards S, et al. Results of Medical Research Council Childhood Leukaemia Trial UKALL VIII (report to the Medical Research Council on behalf of the
References 1157
23a
23b
23c
24
25
26
27
28
29
30
31 32
33 34
35
Working Party on Leukaemia in Childhood). Br J Haematol 1991; 78(2):187–96. van Dongen JJ, Seriu T, Panzer-Grumyer ER, et al. Prognostic value of minimal residual disease in acute lyphoblastic leukaemia in childhood. Lancet 1998; 352:1731–1738. Cave H, van der Werff ten Bosch J, Suciu S, et al. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia. European Organization for Research and Treatment of Cancer – Childhood Leukemia Cooperative Group. New Engl J Med 1998; 339:591–598. Goulden NJ, Knechtli CJC, Garland RJ, et al. Minimal residual disease analysis for the predication of relapse in children with standard-risk acute lymphoblastic leukaemia. British Journal of Haematology 1998; 100:235–244. Lawson SE, Harrison G, Richards S, et al. The United Kingdom experience in treating relapsed childhood acute lymphoblastic leukaemia: a report on the MRC UKALL R1 study. Br J Haematol 2000; 108(3): 531–43. Ramanujachar R, Richards S, Hann I, Webb D. Adolescents with acute lymphoblastic leukaemia: emerging from the shadow of paediatric and adult treatment protocols. Pediatr Blood Cancer 2006; Feb 8 [Epub ahead of print]. Woods WG, Kobrinsky N, Buckley JD, et al. Timed-sequential induction therapy improves post-remission outcome in acute myeloid leukaemia: a report from the Children’s Cancer Group. Blood 1996; 87(12):4979–89. Gibson BE, Wheatley K, Hann IM, et al. Treatment strategy and long-term results in paediatric patients treated in consecutive UK AML trials. Leukaemia 2005; 19(12):2130–8. Ortega JJ, Madero L, Martin G, et al. Treatment with alltrans retinoic acid and anthracycline monochemotherapy for children with acute promyelocytic leukaemia: a multicenter study by the PETHEMA Group. J Clin Oncol 2005; 23(30):7632–40. Sanz MA, Fenaux P, Lo Coco F. European APL Group of Experts. Arsenic Trioxide in the treatment of acute promyelocytic leukaemia. A review of the current evidence. Haematologica 2005; 90(9):1231–5. Smith FO, Alonzo TA, Gerbing RB, et al. Children’s Cancer Group. Long-term results of children with acute myeloid leukaemia: a report of consecutive Phase III trials by the Children’s Cancer Group: CCG 251, CCG 213 and CCG 2891. Leukaemia 2005; 19(12):2054–62. Webb DK. Optimizing therapy for myeloid disorders of Down syndrome. Br J Haematol 2005; 131(1):3–7. Gurbuxani S, Vyas P, Crispino JD. Recent insights into the mechanisms of myeloid leukaemogenesis in Down syndrome. Blood 2004; 103(2):399–406. Crist W, Gehan EA, Ragab AH, et al. The third Intergroup Rhabdomyosarcoma Study. J Clin Oncol 1995; 13:610–630. Flamant F, Rodary C, Rey A, et al. Treatment of non-metastatic rhabdomyosarcomas in childhood and adolescence. Results of the second study of the International Society of Paediatric Oncology: MMT84. Eur J Cancer 1998; 34:1050–1062. Kelly KM, Womer RB, Sorensen PH, et al. Common and variant gene fusions predict distinct clinical phenotypes in rhabdomyosarcoma. J Clin Oncol 1997; 15:1831–1836.
36 DeZen L, Sommaggio A, d’Amore E, et al. Clinical relevance of DNA ploidy and proliferative activity in childhood rhabdomyosarcoma: a retrospective analysis of patients enrolled onto the Italian Cooperative Rhabdomyosarcoma Study RMS88. J Clin Oncol 1997; 15:1198–1205. 37 Paulino AC, Simon JH, Zhen W, Wen BC. Long-term effects in children treated with radiotherapy for head and neck rhabdomyosarcoma. Int J Radiat Oncol Biol Phys 2000; 48(5):1489–95. 38 Schuck A., Mattke AC, Schmidt B, et al. Group II rhabdomyosarcoma and rhabdomyosarcomalike tumors: is radiotherapy necessary? J Clin Oncol 2004; 22(1):143–9. 39 Maurer HM, Gehan EA, Beltangady M, et al. The Intergroup Rhabdomyosarcoma Study-II. Cancer 1993; 71(5): 1904–22. 40 Wolden SL, Anderson JR, Crist WM, et al. Indications for radiotherapy and chemotherapy after complete resection in rhabdomyosarcoma: A report from the Intergroup Rhabdomyosarcoma Studies I to III. J Clin Oncol 1999; 17(11):3468–75. 41 Carli M, Colombatti R, Oberlin O, et al. High-dose melphalan with autologous stem-cell rescue in metastatic rhabdomyosarcoma. J Clin Oncol 1999; 17:2796–2803. 42 Perilongo G, Shafford EA. Liver Tumours. European Journal of Cancer 1999; 35:953–959. 43 Kraus JA, Albrecht S, Wiestler OD, von Schweinitz D, Pietsch T. Loss of heterozygosity on chromosome 1 in hepatoblastoma. Int J Cancer 1996; 67:467–471. 44 Reynolds P, Urayama KY, Von Behren J, Feusner J. Birth characteristics and hepatoblastoma risk in young children. Cancer 2004; 100:1070–6. 45 Feusner JH, Krailo MD, Haas JE, Campbell JR, Lloyd DA, Ablin TR. Treatment of pulmonary metastases of initial stage I hepatoblastoma in childhood. Report from the Children’s Cancer Group. Cancer 1993; 71:859–864. 46 Plaschkes J, Perilongo G, Shafford E. Pre-operative chemotherapy – cisplatin (PLA) and doxorubicin (DO) PLADO for the treatment of hepatoblastoma and hepatocellular carcinoma – results after 2 years’ follow up. Med Pediatr Oncol 1996; 256. 47 Pritchard J, Brown J, Shafford E, et al. Cisplatin, doxorubicin, and delayed surgery for childhood hepatoblastoma: a successful approach – results of the first prospective study of the International Society of Pediatric Oncology. J Clin Oncol 2000; 18:3819–28. 48 Ortega JA, Douglass EC, Feusner JH, et al. Randomized comparison of cisplatin/vincristine/fluorouracil and cisplatin/continuous infusion doxorubicin for treatment of pediatric hepatoblastoma: A report from the Children’s Cancer Group and the Pediatric Oncology Group. J Clin Oncol 2000; 18:2665–75. 49 Plaschkes J, Perilongo G, Shafford E. Childhood hepatoblastoma: an investigation into variables of prognostic relevance using data from the SIOP liver tumour study (SIOPEL 1). SIOP XXVII meeting abstract. Med Pediatr Oncol 1995; 25:256. 50 Perilongo G, Brown J, Shafford E, et al. Hepatoblastoma presenting with lung metastases: treatment results of the
1158 Paediatric oncology
51
52
53
54
55
56
57
58
59
60
61
62
63
64
first cooperative, prospective study of the International Society of Paediatric Oncology on childhood liver tumors. Cancer 2000; 89:1845–53. Al-Qabandi W, Jenkinson HC, Buckels JA, et al. Orthotopic liver transplantation for unresectable hepatoblastoma: a single centres experience. J Paediatr Surg 1999; 34:1261–1264. Otte JB, Pritchard J, Aronson DC, et al. International Society of Pediatric Oncology – SIOP. Liver transplantation for hepatoblastoma: results from the International Society of Pediatric Oncology (SIOP) study SIOPEL-1 and review of the world experience. Pediatr Blood Cancer 2004; 42:74–83. Haas JE, Muczynski KA, Krailo M, et al. Histopathology and prognosis in childhood hepatoblastoma and hepatocellular carcinoma. Cancer 1989; 64:1082–95. Czauderna P, Mackinlay G, Perilongo G, et al. Hepatocellular carcinoma in children: results of the first prospective study of the International Society of Pediatric Oncology group. J Clin Oncol 2002; 20:2798–804. Brodeur GM, Seeger RC, Barrett A, et al. International criteria for diagnosis, staging, and response to treatment in patients with neuroblastoma. J Clin Oncol 1988; 6:1874–81. Simon T, Spitz R, Faldum A, Hero B, Berthold F. New definition of low-risk neuroblastoma using stage, age, and 1p and MYCN status. J Pediatr Hematol Oncol 2004; 26:791–6. George RE, Variend S, Cullinane C, et al. United Kingdom Children Cancer Study Group. Relationship between histopathological features, MYCN amplification, and prognosis: a UKCCSG study. United Kingdom Children Cancer Study Group. Med Pediatr Oncol 2001; 36:169–76. Bown N, Cotterill S, Lastowska M, et al. Gain of chromosome arm 17q and adverse outcome in patients with neuroblastoma. N Engl J Med 1999; 340:1954–1961. Attiyeh EF, London WB, Mosse YP, et al. Children’s Oncology Group (2005). Chromosome 1p and 11q deletions and outcome in neuroblastoma. N Engl J Med 2005; 353:2243–53. George RE, London WB, Cohn SL, et al. Hyperdiploidy plus nonamplified MYCN confers a favorable prognosis in children 12 to 18 months old with disseminated neuroblastoma: a Pediatric Oncology Group study. J Clin Oncol 2005; 23:6466–73. London WB, Castleberry RP, Matthay KK, et al. Evidence for an age cutoff greater than 365 days for neuroblastoma risk group stratification in the Children’s Oncology Group. J Clin Oncol 2005; 23:6459–65. Schmidt ML, Lal A, Seeger RC, et al. Favorable prognosis for patients 12 to 18 months of age with stage 4 nonamplified MYCN neuroblastoma: a Children’s Cancer Group Study. J Clin Oncol 2005; 23:6474–80. Cheung NK, Kushner BH, LaQuaglia M, et al. N7: a novel multi-modality therapy of high risk neuroblastoma (NB) in children diagnosed over 1 year of age. Med Pediatr Oncol 2001; 36:227–30. Tweddle DA, Pinkerton CR, Lewis IJ, et al. OPEC/OJEC for stage 4 neuroblastoma in children over 1 year of age. Med Pediatr Oncol 2001; 36:239–42.
65 Valteau-Couanet D, Benhamou E, Vassal G, et al. Consolidation with a busulfan-containing regimen followed by stem cell transplantation in infants with poor prognosis stage 4 neuroblastoma. Bone Marrow Transplant 2000; 25:937–42. 66 Matthay KK, Villablanca JG, Seeger, RC, et al. Treatment of high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous bone marrow transplantation, and 13-cis-retinoic acid. N Engl J Med 1999; 341:1165–1173. 67 Brodeur GM. Molecular basis for heterogeneity in human neuroblastomas. Eur J Cancer 1995; 505–510. 68 Evans AE, Silber, JH, Shpilsky A, et al. Successful management of low stage neuroblastoma without adjuvant therapies: a comparison of two decades, 1972 through 1981 and 1982 through 1992, in a single institution. J Clin Oncol 1996; 14:2504–2510. 69 Rubie H, Coze C, Plantaz D, et al. Neuroblastoma Study Group, Societe Francaise d’Oncologie Pediatrique. Localised and unresectable neuroblastoma in infants: excellent outcome with low-dose primary chemotherapy. Br J Cancer 2003; 89:1605–9. 70 Matthay KK, Perez C, Seeger RC, et al. Successful treatment of stage III neuroblastoma based on prospective biologic staging: a Children’s Cancer Group study. J Clin Oncol 1998; 16:1256–64. 71 Berthold F, Boos J, Burdach S, et al. Myeloablative megatherapy with autologous stem-cell rescue versus oral maintenance chemotherapy as consolidation treatment in patients with high-risk neuroblastoma: a randomised controlled trial. Lancet Oncol 2005; 6:649–58. 72 Escobar MA, Grosfeld JL, Powell RL, Long-term outcomes in patients with stage IV neuroblastoma. J Pediatr Surg 2006; 41:377–81. 73 La Quaglia MP, Kushner BH, Su W, et al. The impact of gross total resection on local control and survival in high-risk neuroblastoma. J Pediatr Surg 2004; 39:412–7. 74 Haas-Kogan DA, Swift PS, Selch M, et al. Impact of radiotherapy for high-risk neuroblastoma: a Children’s Cancer Group study. Int J Radiat Oncol Biol Phys 2003; 56:28–39. 75 Laprie A, Michon J, Hartmann O, et al. High-dose chemotherapy followed by locoregional irradiation improves the outcome of patients with international neuroblastoma staging system Stage II and III neuroblastoma with MYCN amplification. Cancer 2004; 101:1081–9. 76 Kang TI, Brophy P, Hickeson M, Heyman S, Evans AE, Charron M, Maris JM. Targeted radiotherapy with submyeloablative doses of 131I-MIBG is effective for disease palliation in highly refractory neuroblastoma. Pediatr Hematol Oncol 2003; 25:769–73. 77 Matthay KK, Tan JC, Villablanca JG, et al. Phase I dose escalation of iodine-131-metaiodobenzylguanidine with myeloablative chemotherapy and autologous stem-cell transplantation in refractory neuroblastoma: a new approach to Neuroblastoma Therapy Consortium Study. J Clin Oncol 2006; 24:500–6. 78 Kretschmar CS, Kletzel M, Murray K, et al. Response to paclitaxel, topotecan, and topotecan-cyclophosphamide in
References 1159
79
80
81
82
83
84 85
86
87
88
89
90
91
92
93
children with untreated disseminated neuroblastoma treated in an upfront phase II investigational window: a pediatric oncology group study. J Clin Oncol 2004; 22:4119–26. Matthay KK, Edeline V, Lumbroso J, et al. Correlation of early metastatic response by 123I-metaiodobenzylguanidine scintigraphy with overall response and event-free survival in stage IV neuroblastoma. J Clin Oncol 2003; 21:2486–91. Reynolds CP. Detection and treatment of minimal residual disease in high-risk neuroblastoma. Pediatr Transplant Suppl 2004; 5:56–66. Philip T. Early detection of neuroblastoma in infants: Research? Yes. Routine screening? No. Report of the 1998 consensus conference on neuroblastoma screening. Med Pediatr Oncol 1999a; 33:355–359. Delattre O, Zucman J, Melot T, et al. The Ewing family of tumors – a subgroup of small-round-cell tumors defined by specific chimeric transcripts [see comments]. N Engl J Med 1994; 331(5):294–9. de Alava E, Kawai A, Healey JH, et al. EWS-FLI1 fusion transcript structure is an independent determinant of prognosis in Ewing’s sarcoma [published erratum appears in J Clin Oncol 1998 Aug;16(8):2895] [see comments]. J Clin Oncol 1998; 16(4):1248–55. Ewing J. Diffuse endothelioma of bone. Proc NY Pathol Soc 1921; 21:17–24. Miser JS, Kinsella TJ, Triche TJ, et al. Ifosfamide with mesna uroprotection and etoposide: an effective regimen in the treatment of recurrent sarcomas and other tumors of children and young adults. J Clin Oncol 1987; 5(8):1191–8. Meyer WH, Kun L, Marina N, et al. (1992). Ifosfamide plus etoposide in newly diagnosed Ewing’s sarcoma of bone. J Clin Oncol 1992; 10(11):1737–42. Thomas PR, Perez CA, Neff JR, et al. The management of Ewing’s sarcoma: role of radiotherapy in local tumor control. Cancer Treat Rep 1984; 68(5):703–10. Burgert EO Jr, Nesbit ME, Garnsey LA, et al. Multimodal therapy for the management of nonpelvic, localized Ewing’s sarcoma of bone: intergroup study IESS-II [see comments]. J Clin Oncol 1990; 8(9):1514–24. Jurgens H, Gobel V, Michaelis J, et al. [The Cooperative Ewing Sarcoma Study CESS 81 of the German Pediatric Oncology Society—analysis after 4 years]. Klin Padiatr 1985; 197(3):225–32. Jurgens H, Exner U, Gadner H, et al. Multidisciplinary treatment of primary Ewing’s sarcoma of bone. A 6-year experience of a European Cooperative Trial. Cancer 1988; 61(1):23–32. Sauer R, Jurgens H, Burgers JM, et al. Prognostic factors in the treatment of Ewing’s sarcoma. The Ewing’s Sarcoma Study Group of the German Society of Paediatric Oncology CESS 81. Radiother Oncol 1987; 10(2):101–10. Ahrens S, Hoffmann C, Jabar S, et al. Evaluation of prognostic factors in a tumor volume-adapted treatment strategy for localized Ewing sarcoma of bone: the CESS 86 experience. Cooperative Ewing Sarcoma Study. Med Pediatr Oncol 1999; 32(3):186–95. Craft A, Cotterill S, Imeson J. Improvement in survival for Ewing’s sarcoma by substitution of ifosfamide for
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
cyclophosphamide. Am J Pediatr Hematol Oncol 1993; 15(Suppl A):531–35. Shankar AG, Pinkerton CR, Atra A, et al. Local therapy and other factors influencing site of relapse in patients with localised Ewing’s sarcoma. United Kingdom Children’s Cancer Study Group (UKCCSG). Eur J Cancer 1999; 35(12):1698–704. Paulussen M, Ahrens S, Braun-Munzinger G, et al. [EICESS 92 (European Intergroup Cooperative Ewing’s Sarcoma Study) – preliminary results]. Klin Padiatr 1999; 211(4):276–83. Paulussen M, Ahrens S, Burdach S, et al. Primary metastatic (stage IV) Ewing tumor: survival analysis of 171 patients from the EICESS studies. European Intergroup Cooperative Ewing Sarcoma Studies. Ann Oncol 1998; 9(3):275–81. Smith MA. The impact of doxorubicin dose intensity on survival of patients with Ewing’s sarcoma [letter; comment]. J Clin Oncol 1991; 9(5):889–91. Smith MA, Ungerleider RS, Harowitz ME, Simon R. Influence of doxorubicin dose intensity on response and outcome for patients with osteogenic sarcoma and Ewing’s sarcoma [see comments]. J Natl Cancer Inst 1991; 83(20):1460–70. Marina NM, Pappo AS, Parham DM, et al. Chemotherapy dose-intensification for pediatric patients with Ewing’s family of tumors and desmoplastic small round-cell tumors: a feasibility study at St. Jude Children’s Research Hospital. J Clin Oncol 1999; 17(1):180–90. Burdach S, Jurgens H, Peters C, et al. Myeloablative radiochemotherapy and hematopoietic stem-cell rescue in poor-prognosis Ewing’s sarcoma. J Clin Oncol 1993; 11(8):1482–8. Stewart DA, Gyonyor E, Peterson AH, et al. High-dose melphalan +/- total body irradiation and autologous hematopoietic stem cell rescue for adult patients with Ewing’s sarcoma or peripheral neuroectodermal tumor. Bone Marrow Transplant 1996; 18(2):315–8. Atra A, Whelan JS, Calvagna V, et al. High-dose busulphan/ melphalan with autologous stem cell rescue in Ewing’s sarcoma. Bone Marrow Transplant 1997; 20(10):843–6. Feugeas O, Guriec N, Babin-Boilletot A, et al. Loss of heterozygosity of the RB gene is a poor prognostic factor in patients with osteosarcoma [published erratum appears in J Clin Oncol 1996 Aug;14(8):2411]. J Clin Oncol 1996; 14(2):467–72 Carnevale A, Lieberman E, Cárdenas R, et al. Li-Fraumeni syndrome in pediatric patients with soft tissue sarcoma or osteosarcoma. Arch Med Res 1997; 28(3):383–6. Newton WA, Meadows AT, Shimada H, et al. Bone sarcomas as Second Malignant Neoplasms following childhood cancer. Cancer 1991; 67:193–201. Le Vu B, de Vathaire F, Shamsaldin A, et al. Radiation dose, chemotherapy and risk of osteosarcoma after solid tumours during childhood. Int J Cancer 1998; 77(3):370–7. Schajowicz F, Sissons H, Sobin LH, et al. The World Health Organization’s histologic classification of bone tumors: a commentary on the second edition. Cancer 1995; 75(5):1208–1214. Bacci G, Ferrari S, Delepine N, et al. Predictive factors of histologic response to primary chemotherapy in osteosarcoma
1160 Paediatric oncology
109 110
111
112
113
114
115
116
117
118
119
120 121
122
123
124
125
of the extremity: study of 272 patients preoperatively treated with high-dose methotrexate, doxorubicin, and cisplatin [see comments]. J Clin Oncol 1998; 16(2):658–63. Enneking WF. A system of staging musculoskeletal neoplasms. Clin Orthop 1986; (204):9–24. Asada N, Tsuchiya H, Tomita K, et al. De novo deletions of p53 gene and wild-type p53 correlate with acquired cisplatin-resistance in human osteosarcoma OST cell line. Anticancer Res 1999; 19(6B):5131–7. Goto A, Kanda H, Ishikawa K, et al. Association of loss of heterozygosity at the p53 locus with chemoresistance in osteosarcomas. Jpn J Cancer Res 1998; 89(5):539–47. Wang ZQ, Liang J, Shellander K, et al. C-fos-induced osteosarcoma formation in transgenic mice: cooperativity with c-jun and the role of endogenous c-fos. Cancer Res 1995; 55(24):6244–51. Sweetnam R. Amputation in osteosarcoma. Disarticulation of the hip or high thigh amputation for lower femoral growths? J Bone Joint Surg [Br] 1973; 55(1):189–92. Winkler K, Bielack S, Delling G, et al. Effect of intraarterial versus intravenous cisplatin in addition to systemic doxorubicin, high-dose methotrexate, and ifosfamide on histologic tumor response in osteosarcoma (study COSS-86). Cancer 1990; 66(8):1703–10. Bielack SS, Bieling P, Erttmann R, et al. Intraarterial chemotherapy for osteosarcoma: does the result really justify the effort? Cancer Treat Res 1993; 62:85–92. Bacci G, Donati D, Manfrini M, et al. [Local recurrence after surgical or surgical-chemotherapeutic treatment of osteosarcoma of the limbs. Incidence, risk factors and prognosis]. Minerva Chir 1998; 53(7–8):619–29. Szendroi M, Papai Z, Koos R, et al. Limb-saving surgery, survival, and prognostic factors for osteosarcoma: the Hungarian experience. J Surg Oncol 2000; 73(2):87–94. Abudu A, Sferopoulos NK, Tillman RM, et al. The surgical treatment and outcome of pathological fractures in localised osteosarcoma. J Bone Joint Surg Br 1996; 78(5):694–8. Kawai A, Hamada M, Suqihara S, et al. Rotationplasty for patients with osteosarcoma around the knee joint. Acta Med Okayama 1995; 49(4):221–6. Cores E, Holland J, Wang JJ, et al. Doxorubicin in disseminated osteosarcoma. JAMA 1972; 221:1132–8. Edmonson J, Green S, Ivins JC, et al. A controlled pilot study of high-dose methotrexate as postsurgical adjuvant treatment for primary osteosarcoma. J Clin Oncol 1984; 2(3):152–156. Shirkhoda A, Jaffe N, Wallace S, et al. Computed tomography of osteosarcoma after intraarterial chemotherapy. Am J Roentgenol 1985; 144(1):95–9. Epelman S, Estrada J, Jaffe N, et al. Pediatric osteosarcoma. Successful retreatment of relapsed primary tumor and soft tissue recurrence with intraarterial cisdiamminedichloroplatin-II. Cancer 1990; 66(4):801–5. Jaffe N. Pediatric osteosarcoma: treatment of the primary tumor with intraarterial cis-diamminedichloroplatinum-II (CDP) – advantages, disadvantages, and controversial issues. Cancer Treat Res 1993; 62:75–84. Marti C, Kroner T, Remagen W, et al. High dose ifosfamide in advanced osteosarcoma. Cancer Treat Rep 1985; 69:115.
126 Bowman LC, Meyer WH, et al. Activity of ifosfamide in metastatic and unresectable osteosarcoma. Proc ASCO 1987; 6:214. 127 Pratt CB, Horowitz ME, Meyer WH, et al. Phase II trial of ifosfamide in children with malignant solid tumors. Cancer Treat Rep 1987; 71(2):131–5. 128 Pratt CB, Douglass EC, Etcubanas EC, et al. Ifosfamide in pediatric malignant solid tumors. Cancer Chemother Pharmacol 1989; 24(Suppl 1):S24–7. 129 Petrilli AS, Kechichian R, Broniscer A, et al. Activity of intraarterial carboplatin as a single agent in the treatment of newly diagnosed extremity osteosarcoma. Med Pediatr Oncol 1999; 33(2):71–5. 130 Foster BJ, Clagett-Carr K, Leyland-Jones B, et al. Results of NCI-sponsored phase I trials with carboplatin. Cancer Treat Rev 1985; 12 Suppl A:43–9. 131 Bacha DM, Caparros-Sison B, Allen JA, et al. Phase I study of carboplatin (CBDCA) in children with cancer. Cancer Treat Rep 1986; 70(7):865–9. 132 Gentet JC, Brunat-Mentigny M, Demaille MC, et al. Ifosfamide and etoposide in childhood osteosarcoma. A phase II study of the French Society of Paediatric Oncology. Eur J Cancer 1997; 33(2):232–7. 133 Ohira M. [Autologous bone marrow transplantation in pediatric cancer]. Gan To Kagaku Ryoho 1990; 17(12):2299–306. 134 Link MP, Goorin AM, Miser AW, et al. The effect of adjuvant chemotherapy on relapse-free survival in patients with osteosarcoma of the extremity. N Engl J Med 1986; 314(25):1600–6. 135 Link MP. The multi-institutional osteosarcoma study: an update. Cancer Treat Res 1993; 62:261–7. 136 Eilber F, Giuliano A, Eckardt J, et al. Adjuvant chemotherapy for osteosarcoma: a randomized prospective trial. J Clin Oncol 1987; 5(1):21–6. 137 Bramwell VH, Burgers M, Sneath R, et al. A comparison of two short intensive adjuvant chemotherapy regimens in operable osteosarcoma of limbs in children and young adults: the first study of the European Osteosarcoma Intergroup. J Clin Oncol 1992; 10(10):1579–91. 138 Souhami RL, Craft AW, Van der Eijken JW, et al. Randomised trial of two regimens of chemotherapy in operable osteosarcoma: a study of the European Osteosarcoma Intergroup [see comments]. Lancet 1997; 350(9082):911–7. 139 Lewis IJ, Weeden S, Machin D, et al. Received dose and doseintensity of chemotherapy and outcome in nonmetastatic extremity osteosarcoma. European Osteosarcoma Intergroup. J Clin Oncol 2000; 18(24):4028–37. 140 Bacci G, Picci P, Ferrari S, et al. Influence of adriamycin dose in the outcome of patients with osteosarcoma treated with multidrug neoadjuvant chemotherapy: results of two sequential studies. J Chemother 1993; 5(4):237–46. 141 Delepine N, Delepine G, Bacci G, et al. Influence of methotrexate dose intensity on outcome of patients with high grade osteogenic osteosarcoma. Analysis of the literature [see comments]. Cancer 1996; 78(10):2127–35. 142 Ornadel D, Souhami RL, Whelan J, et al. Doxorubicin and cisplatin with granulocyte colony-stimulating factor as adjuvant chemotherapy for osteosarcoma: phase II trial of
References 1161
143
144
145
146 147
148
149
150
151
152
153
154
155
156
the European Osteosarcoma Intergroup. J Clin Oncol 1994; 12(9):1842–8. Graf N, Winkler K, Betlemovic M, Fuchs N, Bode U. Methotrexate pharmacokinetics and prognosis in osteosarcoma. J Clin Oncol 1994; 12(7):1443–51. Delepine N, Delepine G, Jasmin C, et al. Importance of age and methotrexate dosage: prognosis in children and young adults with high-grade osteosarcomas. Biomed Pharmacother 1988; 42(4):257–62. Delepine N, Delepine G, Cornille H, Brion F, Arnaud P, Desbois JC. Dose escalation with pharmacokinetics monitoring in methotrexate chemotherapy of osteosarcoma. Anticancer Res 1995; 15(2):489–94. Bacci, G. (Letter). J Clin Oncol 1998; 16(6):2290–1. Bielack S, Kempf-Bielack B, Schwenzer D, et al. [Neoadjuvant therapy for localized osteosarcoma of extremities. Results from the Cooperative osteosarcoma study group COSS of 925 patients]. Klin Padiatr 1999; 211(4):260–70. Meyers PA, Gorlick R, Heller G, et al. Intensification of preoperative chemotherapy for osteogenic sarcoma: results of the Memorial Sloan-Kettering (T12) protocol. J Clin Oncol 1998; 16(7):2452–8. Bacci G, Briccoli A, Ferrari S, et al. Neoadjuvant chemotherapy for osteosarcoma of the extremities with synchronous lung metastases: treatment with cisplatin, adriamycin and high dose of methotrexate and ifosfamide. Oncol Rep 2000; 7(2):339–46. Philip T, Iliescu C, Demaille MC, et al. High-dose methotrexate and HELP [Holoxan (ifosfamide), eldesine (vindesine), platinum] – doxorubicin in non-metastatic osteosarcoma of the extremity: a French multicentre pilot study. Federation Nationale des Centres de Lutte contre le Cancer and Societe Francaise d’Oncologie Pediatrique. Ann Oncol 1999; 10(9):1065–71. Harris MB, Gieser P, Goorin AM, et al. Treatment of metastatic osteosarcoma at diagnosis: a Pediatric Oncology Group Study. J Clin Oncol 1998; 16(11): 3641–8. Carter SR, Grimer RJ, Sneath RS, Matthews HR. Results of thoracotomy in osteogenic sarcoma with pulmonary metastases. Thorax 1991; 46(10):727–31. van Rijk-Zwikker GL, Nooy MA, Taminiau A, Kappetein AP, Huysmans HA. Pulmonary metastasectomy in patients with osteosarcoma. Eur J Cardiothorac Surg 1991; 5(8):406–9. Tabone MD, Kalifa C, Rodary C, Raquin M, Valteau-Couanet D, Lemerle J. Osteosarcoma recurrences in pediatric patients previously treated with intensive chemotherapy. J Clin Oncol 1994; 12(12):2614–20. Bacci G, Briccoli A, Mercuri M, et al. Osteosarcoma of the extremities with synchronous lung metastases: long-term results in 44 patients treated with neoadjuvant chemotherapy. J Chemother 1998; 10(1):69–76. Green DM, Coppes MJ, Breslow NF, et al. Wilms tumour. In Pizzo PA, Poplack DG (ed.), Principles and practice of pediatric oncology. Lippincott-Raven Publishers, Philadelphia, 1997a; 733–759.
157 Green DM. Wilms’ tumour. Eur J Cancer 1997b; 33:409–418. 158 Li FP, Breslow NE, Morgan JM, et al. Germline WT1 mutations in Wilms’ tumor patients: preliminary results. Mes Pediatr Oncol 1996; 27:404–7. 159 Grundy PE, Breslow NE, Li S, et al. Loss of heterozygosity for chromosomes 1p and 16q is an adverse prognostic factor in favorable-histology Wilms’ tumor: a report from the National Wilms’ Tumor Study Group. J Clin Oncol 2005; 23:7312–21. 160 Choyke PL, Siegel MJ, Craft AW, Green DM, DeBaun MR. Screening for Wilms tumor in children with BeckwithWiedemann syndrome or idiopathic hemihypertrophy. Med Pediatr Oncol 1999; 32:196–200. 161 Shamberger RC, Guthrie KA, Ritchey ML, et al. Surgeryrelated factors and local recurrence of Wilms’ tumour in National Wilms’ Tumour Study 4. Ann of Surg 1999; 229:292–297. 162 Boccon-Gibod L, Rey A, Sandstedt B. Complete necrosis induced by preoperative chemotherapy in Wilms’ tumour as an indicator of low risk: report of the international society of paediatric oncology (SIOP) nephroblastoma trial and study 9. Med Pediatr Oncol 2000; 34:183–90. 163 D’Angio GJ, Breslow N, Beckwith JB, et al. The treatment of Wilms’ tumour: results of the Third National Wilms’ Tumour Study. Cancer 1989; 64:349. 164 de Kraker J, Graf N, van Tinteren H, Pein F, Sandstedt B, Godzinski J, Tournade MF. Reduction of postoperative chemotherapy in children with stage I intermediate-risk and anaplastic Wilms’ tumour (SIOP 93-01 trial): a randomised controlled trial. Lancet 2004; 364:1229–35. 165 Reinhard H, Semler O, Burger D, et al. Results of the SIOP 93-01/GPOH trial and study for the treatment of patients with unilateral nonmetastatic Wilms’ Tumor. Klin Padiatr 2004; 216:132–40. 166 Green DM, Cotton CA, Malogolowkin M, et al. Treatment of Wilms’ tumor relapsing after initial treatment with vincristine and actinomycin D: A report from the National Wilms’ Tumor Study Group. Pediatr Blood Cancer 2006; Mar 17; [Epub ahead of print]. 167 Pein F, Michon J, Valteau-Couanet D, et al. High-dose melphalan, etoposide and carboplatin followed by autologous stem-cell rescue in paediatric high-risk recurrent Wilms’ tumour: a French Society of Pediatric Oncology study. J Clin Oncol 1998; 16:3295–3301. 168 Argani P, Perlman EJ, Breslow NE, Browning NG, Green DM, D’Angio GJ, Beckwith JB. Clear cell sarcoma of the kidney: a review of 351 cases from the National Wilms’ Tumour Study Group Pathology Center. Am J Surg Pathol 2000; 24:4–18. 169 Tomlinson GE, Breslow NE, Dome J, et al. Rhabdoid tumor of the kidney in the National Wilms’ Tumor Study: age at diagnosis as a prognostic factor. J Clin Oncol 2005; 23:7641–5. 170 Seibel NL, Li S, Breslow NE, et al. Effect of duration of treatment on treatment outcome for patients with clear-cell sarcoma of the kidney: a report from the National Wilms’ Tumor Study Group. J Clin Oncol 2004; 22:468–73. 171 Pinkerton CR. Malignant Germ Cell Tumours in childhood. Eur J Cancer 1997; 33:895–902. 172 Bussey KJ, Lawce HJ, Olson SB, et al. Chromosome abnormalities of eighty-one pediatric germ cell
1162 Paediatric oncology
173
174
175
176
177
178
179
180
181
182
183
184
185
tumours: sex-, age-, site- and histopathology-related differences – A Children’s Cancer Group Study. Genes Chromosomes Cancer 1999; 25:134–146. Castleberry RP, Cushing B, Perlman E, Hawkins EP. Germ cell tumours. In Pizzo PA, Poplack DG (ed.), Principles and practice of pediatric oncology. Lippincott-Raven Publishers, Philadelphia, 1997; 921–945. Billmire D, Vinocur C, Rescorla F, et al. Malignant mediastinal germ cell tumors: an intergroup study. J Pediatr Surg 2001; 36:18–24. Cushing B, Giller R, Cullen JW, et al. Randomized comparison of combination chemotherapy with etoposide, bleomycin, and either high-dose or standard-dose cisplatin in children and adolescents with high-risk malignant germ cell tumors: a pediatric intergroup study – Pediatric Oncology Group 9049 and Children’s Cancer Group 8882. J Clin Oncol 2004; 22:2691–700. Mann JR, Raafat F, Robinson K, et al. The United Kingdom Children’s Cancer Study Group’s second germ cell tumor study: carboplatin, etoposide, and bleomycin are effective treatment for children with malignant extracranial germ cell tumors, with acceptable toxicity. J Clin Oncol 2000; 18:3809–18. Rogers PC, Olson TA, Cullen JW, et al. Treatment of children and adolescents with stage II testicular and stages I and II ovarian malignant germ cell tumors: A Pediatric Intergroup Study – Pediatric Oncology Group 9048 and Children’s Cancer Group 8891. J Clin Oncol 2004; 22:3563–9. Billmire D, Vinocur C, Rescorla F, et al. Malignant retroperitoneal and abdominal germ cell tumors: an intergroup study. J Pediatr Surg 2003; 38:315–8. Calaminus G, Andreussi L, Garre ML, et al. Secreting germ cell tumours of the central nervous system (CNS). First results of the cooperative German/Italian pilot study (CNS sGCT). Klin Padiatr 1997; 209:222–227. Calaminus G, Bamberg M, Jurgens H, et al. Impact of surgery, chemotherapy and irradiation on long term outcome of intracranial malignant non-germinomatous germ cell tumors: results of the German Cooperative Trial MAKEI 89. Klin Padiatr 2004; 216:141–9. Nichols CR, Anderson J, Lazarus HM, et al. High dose carboplatin and etoposide with autologous bone marrow transplantation in refractory germ cell cancer: an Eastern Cooperative Oncology Group protocol. J Clin Oncol 1992; 10:558. Schneider DT, Wessalowski R, Calaminus G, et al. Treatment of recurrent malignant sacrococcygeal germ cell tumors: analysis of 22 patients registered in the German protocols MAKEI 83/86, 89, and 96. J Clin Oncol 2001; 19:1951–60. Carbone PP, Kaplan HS, Musshoff K, et al. Report of the Committee on Hodgkin’s Disease Staging Classification. Cancer Res 1971; 31(11):1860–1. Longo DL, Glatstein E, Duffey PL, et al. Radiation therapy versus combination chemotherapy in the treatment of earlystage Hodgkin’s disease: seven-year results of a prospective randomized trial. J Clin Oncol 1991; 9(6):906–17. Gehan EA, Sullivan MP, Fuller LM, et al. The intergroup Hodgkin’s disease in children. A study of stages I and II. Cancer 1990; 65(6):1429–37.
186 Oberlin O, Leverger G, Pacquement H, et al. Low-dose radiation therapy and reduced chemotherapy in childhood Hodgkin’s disease: the experience of the French Society of Pediatric Oncology. J Clin Oncol 1992; 10(10):1602–8. 187 Longo DL, Glatstein E, Duffey PL, et al. Alternating MOPP and ABVD chemotherapy plus mantle-field radiation therapy in patients with massive mediastinal Hodgkin’s disease. J Clin Oncol 1997; 15(11):3338–46. 188 Donaldson SS, Link MP. Combined modality treatment with low-dose radiation and MOPP chemotherapy for children with Hodgkin’s disease. J Clin Oncol 1987; 5(5):742–9. 189 Bonadonna G, Valagussa P, Santoro A, Viviani S, Bonfante V, Banfi A. Hodgkin’s disease: the Milan Cancer Institute experience with MOPP and ABVD. Recent Results Cancer Res 1989; 117:169–74. 190 Schellong G, Bramswig JH, Hornig-Franz I. Treatment of children with Hodgkin’s disease – results of the German Pediatric Oncology Group. Ann Oncol 1992; 3 Suppl 4:73–6. 191 Landman-Parker J, Pacquement H, Leblanc T. Localized childhood Hodgkin’s disease: response-adapted chemotherapy with etoposide, bleomycin, vinblastine, and prednisone before low-dose radiation therapy – results of the French Society of Pediatric Oncology Study MDH90. J Clin Oncol 2000; 18(7):1500–7. 192 Ruhl U, Albrecht M, Dieckmann D, et al. Response-adapted radiotherapy in the treatment of pediatric Hodgkin’s disease: an interim report at 5 years of the German GPOHHD 95 trial. Int J Radiat Oncol Biol Phys 2001; 51(5):1209–18. 193 Shankar AG, Ashley S, Atra A, Kingston JE, Mott M, Pinkerton CR. A limited role for VEEP (vincristine, etoposide, epirubicin, prednisolone) chemotherapy in childhood Hodgkin’s disease. Eur J Cancer 1998; 34(13):2058–63. 194 Ekert H, Toogood I, Downie P, Smith PJ, Macfarlane S, White L. High incidence of treatment failure with vincristine, etoposide, epirubicin, and prednisolone chemotherapy with successful salvage in childhood Hodgkin’s disease. Med Pediatr Oncol 1999; 32(4):255–8. 195 Schellong G, Potter R, Bramswig J, et al. High cure rates and reduced long-term toxicity in pediatric Hodgkin’s disease: the German-Austrian multicenter trial DAL-HD-90. The German-Austrian Pediatric Hodgkin’s Disease Study Group. J Clin Oncol 1999; 17(12):3736–44. 196 Dieckmann K, Potter R, Hofmann J, Heinzl H, Wagner W, Schellong G. Does bulky disease at diagnosis influence outcome in childhood Hodgkin’s disease and require higher radiation doses? Results from the German-Austrian Pediatric Multicenter Trial DAL-HD-90. Int J Radiat Oncol Biol Phys 2003; 56(3):644–52. 197 Schellong G, Dorffel W, Claviez A, et al. Salvage therapy of progressive and recurrent Hodgkin’s disease: results from a multicenter study of the pediatric DAL/GPOH-HD study group. J Clin Oncol 2005; 23(25):6181–9. 198 Akpek G, Ambinder RF, Piantadosi S, et al. Long-term results of blood and marrow transplantation for Hodgkin’s lymphoma. J Clin Oncol 2001; 19(23):4314–21. 199 Harris NL, Jaffe ES, Stein H, et al. A revised EuropeanAmerican classification of lymphoid neoplasms: a proposal
References 1163
200
201
202
203 204
205
206
207
208
209
210
211
212
from the International Lymphoma Study Group. Blood 1994; 84(5):1361–92. Harris NL, Jaffe ES, Diebold J, Flandrin G, Muller-Hermelink HK, Vardiman J. Lymphoma classification – from controversy to consensus: the REAL and WHO Classification of lymphoid neoplasms. Ann Oncol 2000; 11 Suppl 1:3–10. Harris NL, Jaffe ES, Diebold J, et al. The World Health Organization classification of neoplasms of the hematopoietic and lymphoid tissues: report of the Clinical Advisory Committee meeting – Airlie House, Virginia, 1997. Hematol J 2000; 1(1):53–66. Armitage JO, Weisenburger DD. New approach to classifying non-Hodgkin’s lymphomas: clinical features of the major histologic subtypes. Non-Hodgkin’s Lymphoma Classification Project. J Clin Oncol 1998; 16(8):2780–95. Croce CM, Nowell PC. Molecular basis of human B cell neoplasia. Blood 1985; 65(1):1–7. Vannier JP, Patte C, Philip T, et al. Treatment of extensive B-cell lymphoma in children: studies of the French Pediatric Oncology Society. Bull Cancer 1988; 75(1):61–8. Patte C, Auperin A, Michon J, et al. The Societe Francaise d’Oncologie Pediatrique LMB89 protocol: highly effective multiagent chemotherapy tailored to the tumor burden and initial response in 561 unselected children with B-cell lymphomas and L3 leukemia. Blood 2001; 97(11):3370–9. Patte C, Philip T, Rodary C, et al. Improved survival rate in children with stage III and IV B cell non-Hodgkin’s lymphoma and leukemia using multi-agent chemotherapy: results of a study of 114 children from the French Pediatric Oncology Society. J Clin Oncol 1986; 4(8):1219–26. Reiter A, Schrappe M, Ludwig WD, et al. Intensive ALL-type therapy without local radiotherapy provides a 90% eventfree survival for children with T-cell lymphoblastic lymphoma: a BFM group report. Blood 2000; 95(2):416–21. Delsol G, Al Saati T, Gatter KC, et al. Coexpression of epithelial membrane antigen (EMA), Ki-1, and interleukin-2 receptor by anaplastic large cell lymphomas. Diagnostic value in so-called malignant histiocytosis. Am J Pathol 1988; 130(1):59–70. Lamant L, Meggetto F, al Saati T, et al. High incidence of the t(2;5)(p23;q35) translocation in anaplastic large cell lymphoma and its lack of detection in Hodgkin’s disease. Comparison of cytogenetic analysis, reverse transcriptasepolymerase chain reaction, and P-80 immunostaining. Blood 1996; 87(1):284–91. Reiter A, Schrappe M, Tiemann M, et al. Successful treatment strategy for Ki-1 anaplastic large-cell lymphoma of childhood: a prospective analysis of 62 patients enrolled in three consecutive Berlin-Frankfurt-Munster group studies. J Clin Oncol 1994; 12(5):899–908. Heideman RL, Packer RJ, Albright LA, Freeman CR, Rorke LB. Tumours of the central nervous system. In Pizzo PA, Poplack DG (ed.), Principles and practice of pediatric oncology. Lippincott-Raven Publishers, Philadelphia, 1997; 633–697. Duffner PK, Horowitz ME, Krischer JP, et al. Postoperative chemotherapy and delayed radiation in children less than three years of age with malignant brain tumours. New Engl J Med 1993; 328:1725–1731.
213 Geyer JR, Finlay JL, Boyett JM, et al. Survival of infants with malignant astrocytomas. A report from the Children’s Cancer Group. Cancer 1995; 75:1045–50. 214 Rutkowski S, Bode U, Deinlein F, et al. Treatment of early childhood medulloblastoma by postoperative chemotherapy alone. N Engl J Med 2005; 352:978–86. 215 Rorke LB, Gilles FH, Davis RL, Becker LE. Revision of the World Health Organisation classification of brain tumours for childhood brain tumours. Cancer 1985; 56:1869. 216 Cokgor I, Friedman AH, Friedman HS. Paediatric Update: Gliomas. Eur J Cancer 1998; 12:1910–1918. 217 Gajjar A, Sanford RA, Heideman R, et al. Low-grade astrocytoma: A decade of experience at St Jude Children’s Research Hospital. J Clin Oncol 1997;15:2792–2799. 218 Pollack IF, Claassen D, al-Shboul Q, Janofsky JE, Deutsch M. Low-grade gliomas of the cerebral hemispheres in children: an analysis of 71 cases. J Neurosurg 1995; 82:536–547. 219 Kortmann RD, Timmermann B, Taylor RE, et al. Current and future strategies in radiotherapy of childhood low-grade glioma of the brain. Part II: Treatment-related late toxicity. Strahlenther Onkol 2003; 179:585–97. 220 Marcus KJ, Goumnerova L, Billett AL, et al. Stereotactic radiotherapy for localized low-grade gliomas in children: final results of a prospective trial. Int J Radiat Oncol Biol Phys 2005; 61:374–9. 221 Janss AJ, Grundy R, Cnaan A, et al. Optic pathway and hypothalamic/chiasmatic gliomas in children younger than age 5 years with a 6-year follow up. Cancer 1995; 75:1051–1059. 222 Gnekow AK, Kortmann RD, Pietsch T, Emser A. Low grade chiasmatic-hypothalamic glioma-carboplatin and vincristin chemotherapy effectively defers radiotherapy within a comprehensive treatment strategy – report from the multicenter treatment study for children and adolescents with a low grade glioma – HIT-LGG 1996 – of the Society of Pediatric Oncology and Hematology (GPOH). Klin Padiatr 2004; 216:331–42. 223 Massimino M, Spreafico F, Cefalo G, et al. High response rate to cisplatin/etoposide regimen in childhood low-grade glioma. J Clin Oncol 2002; 20:4209–16. 224 Lancaster DL, Hoddes JA, Michalski A. Tolerance of nitrosurea-based multiagent chemotherapy regime for lowgrade pediatric gliomas. J Neurooncol 2003; 63:289–94. 225 Lafay-Cousin L, Holm S, Qaddoumi I, et al. Weekly vinblastine in pediatric low-grade glioma patients with carboplatin allergic reaction. Cancer 2005; 103:2636–42. 226 Wishoff JH, Boyett JM, Berger MS, et al. Current neurosurgical management and the impact of the extent of resection in the treatment of malignant gliomas of childhood: a report of the Children’s Cancer Group Trial No. CCG-1945. J Neurosurg 1998; 89:52–59. 227 Bredel M, Pollack IF, Hamilton RL, James CD. Epidermal growth factor receptor expression and gene amplification in high-grade non-brainstem gliomas of childhood. Clin Cancer Res 1999; 5:1786–92. 228 Gilbertson RJ, Hill DA, Hernan R, et al. ERBB1 is amplified and overexpressed in high-grade diffusely infiltrative pediatric brain stem glioma. Clin Cancer Res 2003; 9:3620–4.
1164 Paediatric oncology
229 Finlay JL, Boyett JM, Yates AJ, et al. Randomised Phase III trial in childhood high-grade astrocytoma comparing vincristine, lomustine and prednisone with the eightdrugs-in-1 day regimen. J Clin Oncol 1995; 13:112–123. 230 Wolff JE, Gnekow AK, Kortmann RD, Pietsch T, Urban C, Graf N, Kuhl J. Preradiation chemotherapy for pediatric patients with high-grade glioma. Cancer 2002; 94:264–71. 231 Massamino M, Gandola L, Luksch R, et al. Sequential chemotherapy, high-dose thiotepa, circulating progenitor cell rescue, and radiotherapy for childhood high-grade glioma. Neuro Oncol 2005; 7:41–8. 232 MacDonald TJ, Arenson EB, Ater J, et al. Phase II study of highdose chemotherapy before radiation in children with newly diagnosed high-grade astrocytoma: final analysis of Children’s Cancer Group Study 9933. Cancer 2005; 104:2862–71. 233 Broniscer A, Chintagumpala M, Fouladi M, el al. Temozolomide after radiotherapy for newly diagnosed highgrade glioma and unfavorable low-grade glioma in children. J Neurooncol 2006; 76:313–319. 234 Wagner S, Erdlenbruch B, Langler A, et al. Oral topotecan in children with recurrent or progressive high-grade glioma: a Phase I/II study by the German Society for Pediatric Oncology and Hematology. Cancer 2004; 100: 1750–7. 235 Mandell LR, Kadota R, Freeman C, et al. There is no role for hyperfractionated radiotherapy in the management of children with newly diagnosed diffuse intrinsic brainstem tumors: results of a Pediatric Oncology Group phase III trial comparing conventional vs. hyperfractionated radiotherapy. Int J Radiat Oncol Biol Phys 1999; 43:959–64. 236 Hargrave D, Bartels U, Bouffet E. Diffuse brainstem glioma in children: critical review of clinical trials. Lancet Oncol 2006; 7:241–8. 237 Bouffet E, Perilongo G, Canete A, Massimino M. Intracranial ependymomas in children: a critical review of prognostic factors and a plea for cooperation. Medical and Pediatric Oncology 1998; 30:319–331. 238 Robertson PL, Zeltzer PM, Boyett JM, et al. Survival and prognostic factors following radiation therapy and chemotherapy for ependymomas in children: a report of the Children’s Cancer Group. J Neurosurg 1998; 88:695–703. 239 Massimino M, Gandola L, Giangaspero F, et al. Hyperfractionated radiotherapy and chemotherapy for childhood ependymoma: final results of the first prospective AIEOP (Associazione Italiana di Ematologia-Oncologia Pediatrica) study. Int J Radiat Oncol Biol Phys 2004; 58:1336–45. 240 Perilongo G, Massimino M, Sotti G, et al. Analyses of prognostic factors in a retrospective review of 92 children with ependymoma: Italian Paediatric Neuro-Oncology Group. Med Pediatr Oncol 1997; 29:79–85. 241 Needle MN, Goldwein JW, Grass J, et al. Adjuvant chemotherapy for the treatment of intracranial ependymoma of childhood. Cancer 1997; 80:341–347. 242 Sandri A, Massimino M, Mastrodicasa L, et al. Treatment with oral etoposide for childhood recurrent ependymomas. J Pediatr Hematol Oncol 2005; 27:486–90.
243 Vinchon M, Leblond P, Noudel R, Dhellemmes P. Intracranial ependymomas in childhood: recurrence, reoperation, and outcome. Childs Nerv Syst 2005; 21:221–6. 244 Johnson JH Jr, Hariharan S, Berman J, Sutton LN, Rorke LB, Molloy P, Phillips PC. Clinical outcome of pediatric gangliogliomas: ninety-nine cases over 20 years. Pediatr Neurosurg 1997; 27:203–7. 245 Chan CH, Bittar RG, Davis GA, Kalnins RM, Fabinyi GC. Longterm seizure outcome following surgery for dysembryoplastic neuroepithelial tumor. J Neurosurg 2006; 104:62–9. 246 Packer RJ, Sutton LN, D’Angio G, Evans AE, Schut L. Management of children with primitive neuroectodermal tumors of the posterior fossa/medulloblastoma. Pediatr Neurosci 1985; 12(4–5):272–82. 247 Evans AE, Jenkin RD, Sposto R, et al. The treatment of medulloblastoma. Results of a prospective randomized trial of radiation therapy with and without CCNU, vincristine, and prednisone. J Neurosurg 1990; 72(4):572–82. 248 Tait DM, Thornton-Jones H, Bloom HJ, Lemerle J, MorrisJones P. Adjuvant chemotherapy for medulloblastoma: the first multi-centre control trial of the International Society of Paediatric Oncology (SIOP I). Eur J Cancer 1990; 26(4):464–9. 249 Carrie C, Lasset C, Blay JY, et al. Medulloblastoma in adults: survival and prognostic factors. Radiother Oncol 1993; 29(3):301–7. 250 Bailey CC, Gnekow A, Wellek S, et al. Prospective randomised trial of chemotherapy given before radiotherapy in childhood medulloblastoma. International Society of Paediatric Oncology (SIOP) and the (German) Society of Paediatric Oncology (GPO): SIOP II. Med Pediatr Oncol 1995; 25(3):166–78. 251 Gilbertson RJ, Pearson AD, Perry RH, Jaros E, Kelly PJ. Prognostic significance of the c-erbB-2 oncogene product in childhood medulloblastoma. Br J Cancer 1995; 71(3): 473–7. 252 Gajjar A, Hernan R, Kocak M, et al. Clinical, histopathologic, and molecular markers of prognosis: toward a new disease risk stratification system for medulloblastoma. J Clin Oncol 2004; 22(6):984–93. 253 Packer RJ. Chemotherapy for medulloblastoma/primitive neuroectodermal tumors of the posterior fossa. Ann Neurol 1990; 28(6):823–8. 254 Taylor RE, Bailey CC, Robinson K, et al. Results of a randomized study of preradiation chemotherapy versus radiotherapy alone for nonmetastatic medulloblastoma: The International Society of Paediatric Oncology/United Kingdom Children’s Cancer Study Group PNET-3 Study. J Clin Oncol 2003; 21(8):581–91. 255 Packer RJ, Sutton LN, Elterman R, et al. Outcome for children with medulloblastoma treated with radiation and cisplatin, CCNU, and vincristine chemotherapy. J Neurosurg 1994; 81(5):690–8. 256 Deutsch M, Thomas P, Boyett J, et al. Low stage medullobastoma: a Children’s Cancer Study Group (CCSG) and Pediatric Oncology Group (POG) randomized study of standard vs. reduced neuraxis radiation. Proc Am Soc Clin Oncol 1991; 10:A363.
References 1165
257 Goldwein JW, Radcliffe J, Johnson J, et al. Updated results of a pilot study of low dose craniospinal irradiation plus chemotherapy for children under five with cerebellar primitive neuroectodermal tumors (medulloblastoma). Int J Radiat Oncol Biol Phys 1996; 34(4):899–904. 258 Packer RJ, Goldwein J, Nicholson HS, et al. Treatment of children with medulloblastomas with reduced-dose craniospinal radiation therapy and adjuvant chemotherapy: A Children’s cancer group study [In Process Citation]. J Clin Oncol 1999; 17(7):2127. 259 Dupuis-Girod S, Hartmann O, Benhamou E, et al. Will high dose chemotherapy followed by autologous bone marrow transplantation supplant cranio-spinal irradiation in young children treated for medulloblastoma? J Neurooncol 1996; 27(1):87–98. 260 Torres CF, Rebsamen S, Silber JH, et al. Surveillance scanning of children with medulloblastoma [see comments]. N Engl J Med 1994; 330(13):892–5. 261 Boor R, Huber A, Gutjahr P. Etoposide treatment in recurrent medulloblastoma. Neuropediatrics 1994; 25(1):39–41. 262 Vassal G, Doz F, Frappaz D, et al. A phase I study of irinotecan as a 3-week schedule in children with refractory or recurrent solid tumors. J Clin Oncol 2003; 21(20):3844–52. 263 Abe M, Tokumaru S, Tabuchi K, Kida Y, Takagi M, Imamura J. Stereotactic radiation therapy with chemotherapy in the management of recurrent medulloblastomas. Pediatr Neurosurg 2006; 42(2):81–8. 264 Zwerdling T, Krailo M, Monteleone P, et al. Phase II investigation of docetaxel in pediatric patients with recurrent solid tumors: a report from the Children’s Oncology Group. Cancer 2006; 106(8):1821–8. 265 Graham MJ, Herndon JE 2nd, Casey JR, et al. High-dose chemotherapy with autologous stem-cell rescue in patients with recurrent and high-risk pediatric brain tumors. J Clin Oncol 1997; 15(5):1814–23.998. 266 Dunkel IJ, Boyett JM, Yates A, et al. High-dose carboplatin, thiotepa, and etoposide with autologous stem- cell rescue for patients with recurrent medulloblastoma. Children’s Cancer Group. J Clin Oncol 1998; 16(1):222–8. 267 Zia MI, Forsyth P, Chaudhry A, Russell J, Stewart DA. Possible benefits of high-dose chemotherapy and autologous stem cell transplantation for adults with recurrent medulloblastoma. Bone Marrow Transplant 2002; 30(9):565–9. 268 Perez-Martinez A, Lassaletta A, Gonzalez-Vicent M, Sevilla J, Diaz MA, Madero L. High-dose chemotherapy with autologous stem cell rescue for children with high risk and recurrent medulloblastoma and supratentorial primitive neuroectodermal tumors. J Neurooncol 2005; 71(1):33–8. 269 Rorke LB, Packer R, Biegel J. Central nervous system atypical teratoid/rhabdoid tumors of infancy and childhood. J Neurooncol 1995; 24(1):21–8. 270 Sevenet N, Sheridan E, Amram D, et al. Constitutional mutations of the hSNF5/INI1 gene predispose to a variety of cancers. Am J Hum Genet 1999; 65(5):1342–8. 271 Rorke LB, Packer RJ, Biegel JA. Central nervous system atypical teratoid/rhabdoid tumors of infancy and childhood: definition of an entity. J Neurosurg 1996; 85(1):56–65.
272 Foot AB, Hayes C. Audit of guidelines for effective control of chemotherapy and radiotherapy induced emesis. Arch Dis Child 1994; 71(5):475–80. 273 Rickard KA, Baehner RL, Provisor AJ, Weetman RM, Grosfeld JL. The effects of hyperalimentation on immune function and tumor growth. Prog Clin Biol Res 1981; 67:339–347. 274 Rickard KA, Coates TD, Grosfeld JL, Weetman RM, Baehner RL. The value of nutrition support in children with cancer. Cancer 1986; 58(8 Suppl):1904–1910. 275 Neumann CG, Jelliffe DB, Zerfas AJ, Jelliffe EF. Nutritional assessment of the child with cancer. Cancer Res 1982; 42(2 Suppl):699s–712s. 276 Rickard KA, Loghmani ES, Grosfeld JL, et al. Short- and longterm effectiveness of enteral and parenteral nutrition in reversing or preventing protein-energy malnutrition in advanced neuroblastoma. A prospective randomized study. Cancer 1985; 56(12):2881–2897. 277 Tatman A, Warren A, Williams A, Powell JE, Whitehouse W. Development of a modified paediatric coma scale in intensive care clinical practice. Arch Dis Child 1997; 77(6):519–521. 278 Neglia JP, Friedman DL, Yasui Y, et al. Second malignant neoplasms in five-year survivors of childhood cancer: childhood cancer survivor study. J Natl Cancer Inst 2001; 93(8):618–29. 279 Rosso P, Terracini B, Fears TR, et al. Second malignant tumors after elective end of therapy for a first cancer in childhood: a multicenter study in Italy. Int J Cancer 1994; 59(4):451–6. 280 Hawkins MM, Wilson LM, Stovall MA, Marsden HB, Potok MH, Kingston JE, Chessells JM. Epipodophyllotoxins, alkylating agents, and radiation and risk of secondary leukaemia after childhood cancer. Br Med J 1992; 304(6832):951–8. 281 Blayney DW, Longo DL, Young RC, et al. Decreasing risk of leukemia with prolonged follow-up after chemotherapy and radiotherapy for Hodgkin’s disease. N Engl J Med 316(12):710–4. 282 Ogilvy-Stuart AL, Shalet SM. Effect of chemotherapy on growth. Acta Paediatr Suppl 1995; 411:52–6. 283 Ogilvy-Stuart AL, Shalet SM. Growth and puberty after growth hormone treatment after irradiation for brain tumours. Arch Dis Child 1995; 73(2):141–6. 284 Schmiegelow M, Lassen S, Poulsen HS, Feldt-Rasmussen U, Schmiegelow K, Hertz H, Muller J. Cranial radiotherapy of childhood brain tumours: growth hormone deficiency and its relation to the biological effective dose of irradiation in a large population based study. Clin Endocrinol (Oxf) 2000; 53(2):191–7. 285 Hawkins MM. Pregnancy outcome and offspring after childhood cancer. Br Med J 1994; 309(6961):1034. 286 Lustig RH, Rose SR, Burghen GA, et al. Hypothalamic obesity caused by cranial insult in children: altered glucose and insulin dynamics and reversal by a somatostatin agonist. J Pediatr 1999; 135(2 Pt 1):162–8. 287 Shan K, Lincoff AM, Young JB. Anthracycline-induced cardiotoxicity. Ann Intern Med 1996; 125(1):47–58. 288 Skinner R, Sharkey IM, Pearson AD, Craft AW. Ifosfamide, mesna, and nephrotoxicity in children. J Clin Oncol 1993; 11(1):173–90.
1166 Paediatric oncology
289 Charlton A, Larcombe IJ, Meller ST, et al. Absence from school related to cancer and other chronic conditions. Arch Dis Child 1991; 66(10):1217–22. 290 Hays DM, Landsverk J, Sallan SE, et al. Educational, occupational, and insurance status of childhood cancer survivors in their fourth and fifth decades of life. J Clin Oncol 1992; 10(9):1397–406. 291 Glauser TA, Packer RJ. Cognitive deficits in long-term survivors of childhood brain tumors. Childs Nerv Syst 1991; 7(1):2–12. 292 Grill J, Kieffer V, Kalifa C. Measuring the neuro-cognitive side-effects of irradiation in children with brain tumors. Pediatr Blood Cancer 2004; 42(5):452–6. 293 Ellenberg L, McComb JG, Siegel SE, Stowe S. Factors affecting intellectual outcome in pediatric brain tumor patients. Neurosurgery 1987; 21(5):638–44. 294 Radcliffe J, Packer RK, Atkins TE, Bunin GR, Schut L, Goldwein JW, Sutton LN. Three- and four-year cognitive outcome in children with noncortical brain tumors treated with wholebrain radiotherapy. Ann Neurol 1992; 32(4):551–4. 295 Anderson V, Smibert E, Ekert H, Godber T. Intellectual, educational, and behavioural sequelae after cranial irradiation and chemotherapy. Arch Dis Child 1994; 70(6):476–83. 296 Silber JH, Radcliffe J, Peckham V, et al. Whole-brain irradiation and decline in intelligence: the influence of dose and age on IQ score [see comments]. J Clin Oncol 1992; 10(9):1390–6. 297 Mulhern RK, Kepner JL, Thomas PR, Armstrong FD, Friedman HS, Kun LE. Neuropsychologic functioning of survivors of childhood medulloblastoma randomized to receive conventional or reduced-dose craniospinal irradiation: a Pediatric Oncology Group study. J Clin Oncol 1998; 16(5):1723–8. 298 Nayfield SG, Ganz PA, Moinpour CM, Cella DF, Hailey BJ. Report from a National Cancer Institute (USA) workshop on quality of life assessment in cancer clinical trials. Qual Life Res 1992; 1(3):203–10. 299 Machin D. Assessment of quality of life in clinical trials of the British Medical Research Council. J Natl Cancer Inst Monogr 1996; (20):97–102. 300 Lackner H, Benesch M, Schagerl S, Kerbl R, Schwinger W, Urban C. Prospective evaluation of late effects after childhood cancer therapy with a follow-up over 9 years. Eur J Pediatr 2000; 159(10):750–8. 301 Oeffinger KC, Eshelman DA, Tomlinson GE, Buchanan GR, Foster BM. Grading of late effects in young adult survivors of childhood cancer followed in an ambulatory adult setting. Cancer 2000; 88(7):1687–95. 302 Mostow EN, Byrne J, Connelly RR, Mulvihill JJ. Quality of life in long-term survivors of CNS tumors of childhood and adolescence. J Clin Oncol 1991; 9(4):592–9. 303 Apajasalo M, Sintonen H, Siimes MA, et al. Health-related quality of life of adults surviving malignancies in childhood. Eur J Cancer 1996; 32A(8):1354–8. 304 Moe PJ, Holen A, Glomstein A, et al. Long-term survival and quality of life in patients treated with a national all protocol 15–20 years earlier: IDM/HDM and late effects? Pediatr Hematol Oncol 1997; 14(6):513–24.
305 Novakovic B, Fears TR, Horowitz ME, Tucker MA, Wexler LH. Late effects of therapy in survivors of Ewing’s sarcoma family tumors. J Pediatr Hematol Oncol 1997; 19(3):220–5. 306 Langeveld NE, Grootenhuis MA, Voute PA, de Haan RJ, van den Bos C. No excess fatigue in young adult survivors of childhood cancer. Eur J Cancer 2003; 39(2):204–14. 307 Meeske K, Katz ER, Palmer SN, Burwinkle T, Varni JW. Parent proxy-reported health-related quality of life and fatigue in pediatric patients diagnosed with brain tumors and acute lymphoblastic leukemia. Cancer 2004; 101(9):2116–25. 308 Zebrack BJ, Zeltzer LK, Whitton J, Mertens AC,Odom L, Berkow R, obison LL. Psychological outcomes in long-term survivors of childhood leukemia, Hodgkin’s disease, and nonHodgkin’s lymphoma: a report from the Childhood Cancer Survivor Study. Pediatrics 2002; 110(1 Pt 1):42–52. 309 Glover DA, Byrne J, Mills JL, Robison LL, Nicholson HS, Meadows A, Zeltzer LK. Impact of CNS treatment on mood in adult survivors of childhood leukemia: a report from the Children’s Cancer Group. J Clin Oncol 2003; 21(23):4395–401. 310 Eiser C, Hill JJ, Blacklay A. Surviving cancer; what does it mean for you? An evaluation of a clinic based intervention for survivors of childhood cancer. Psychooncology 2000; 9(3):214–20. 311 Gray RE, Doan DB, Shermer P, FitzGerald AV, Berry MP, Jenkin D, Doherty MA. Psychologic adaptation of survivors of childhood cancer. Cancer 1992; 70(11):2713–21. 312 Hobbie WL, Stuber M, Meeske K, et al. Symptoms of posttraumatic stress in young adult survivors of childhood cancer. J Clin Oncol 2000; 18(24):4060–6. 313 Kazak AE, Alderfer MA, Rourke T, Simms S, Streisand R, Grossman JR. Posttraumatic stress disorder (PTSD) and posttraumatic stress symptoms (PTSS) in families of adolescent childhood cancer survivors. J Pediatr Psychol 2004; 29(3):211–9. 314 Association for children with life-threatening or terminal conditions and their families (ACT) and Royal College of Paediatrics and Child Health (RCPCH). A guide to the development of Children’s Palliative Care Services. 1997 315 Hinds P, Oakes L, Furman W, et al. Decision making by parents and healthcare professionals when considering continued care for pediatric patients with cancer. Oncol Nurs Forum 1997; 24(9):1523–1528. 316 Deatrick JA, Angst DB, Moore C. Parents’ Views of their Children’s Participation in Phase 1 Oncology Clinical Trials. J Pediatr Oncol Nurs 2002; 19(4):114–121. 317 Kenner DJ. Pain forum. Part 2. Neuropathic pain. Aust Fam Physician 1994; 23(7):1279–83. 318 Rosenberg JM, Harrell C, Ristic H, Werner RA, de Rosayro AM. The effect of gabapentin on neuropathic pain. Clin J Pain 1997; 13(3):251–5. 319 McGraw T, Stacey BR. Gabapentin for treatment of neuropathic pain in a 12-year-old girl. Clin J Pain 1998; 14(4):354–6. 320 Watson CP. The treatment of neuropathic pain: antidepressants and opioids. Clin J Pain 2001; 16(2 Suppl):S49–55. 321 McQuay H., Carroll D, Jadad AR, Wiffen P, Moore A. Anticonvulsant drugs for management of pain: a systematic review. Br Med J 1995; 311(7012):1047–52.
47 AIDS-related malignancy MARK BOWER AND TOM NEWSOM-DAVIS
Introduction Kaposi’s sarcoma Systemic NHL Primary CNS NHL Castleman’s disease Hodgkin’s disease Anal cancer
1167 1168 1173 1176 1178 1178 1179
INTRODUCTION The human immunodeficiency virus (HIV) was recognized as a human pathogen just over 20 years ago but in that time it has infected millions of people, resulting in a global pandemic. The consequence of HIV infection is a relentless destruction of the immune system culminating in the diagnosis of acquired immune deficiency syndrome (AIDS). The World Health Organization estimated that at the end of 2005, 21.8 million people had died of HIV/AIDS and that 40.3 million people were living with HIV. In many established market economy countries, a significant reduction in the death rate for people with AIDS has been seen following the introduction of potent combination antiretroviral drugs. Thus in these countries there are increasing numbers of people living longer with their HIV. Both primary congenital immunosuppression and iatrogenic secondary immunosuppression have long been recognized to predispose patients to particular malignancies and it was therefore not surprising that these malignancies occurred more frequently in people with HIV. Many of the malignancies associated with immunosuppression are thought to have a viral aetiology with herpes viruses (Kaposi’s sarcoma, non-Hodgkin’s lymphoma) and papilloma viruses (cervical cancer, anal cancer) implicated. It is of interest that some of the tumours (such as Burkitt’s lymphoma and Hodgkin’s disease) that occur more frequently in HIV are not associated with other
Cervical cancer Leiomyosarcoma in children Squamous cell carcinomas of conjunctiva Other skin tumours Testicular germ cell tumours Other tumours References
1179 1181 1181 1181 1181 1181 1182
forms of immunosuppression and that the incidence of some common malignancies appears to be reduced in HIV-positive populations (for example breast and prostate cancers). The development of effective antiretroviral therapies commenced in the mid-1980s with the introduction of the nucleoside analogue reverse transcriptase inhibitors such as zidovudine. Since then there has been a rapid expansion of the therapeutic armamentarium and there are now three further classes of drugs currently licensed for the treatment of HIV: non-nucleoside analogue reverse transcriptase inhibitors, protease inhibitors, and fusion inhibitors. The introduction of the first two classes in the late 1990s led to the use of combination highly active antiretroviral treatment (HAART). This advance was associated with profound and sustained suppression of HIV viral replication, a dramatic reduction in opportunistic infections, AIDSdefining illnesses and mortality amongst HIV-infected persons. However, only 1 million of the estimated 40 million people infected with HIV worldwide are receiving HAART as the majority of affected people live in developing countries. Hence the WHO initiated the ‘3 by 5’ global target of treating 3 million people in low and middle income countries with antiretrovirals by the end of 2005. In addition to reducing the incidence of opportunistic infections, HAART has been associated with a reduction in the incidence of Kaposi’s sarcoma and non-Hodgkin’s lymphoma, especially primary cerebral lymphomas.
1168 AIDS-related malignancy
KAPOSI’S SARCOMA Epidemiology The era of AIDS was heralded in 1981 by the emergence of two previously uncommon diseases: Pneumocystis carinii pneumonia (subsequently renamed Pneumocystis jirovecii) and an aggressive form of Kaposi’s sarcoma (KS).1 Prior to this three different clinical expressions of KS had been recognised: (1) classical KS, an indolent variant predominantly affecting elderly men of Mediterranean and Jewish descent; (2) endemic KS, a more severe form affects children and young adults in sub-Saharan Africa; and (3) iatrogenic KS, which comprises up to 5 per cent of malignancies in immunosuppressed allogeneic transplant recipients. A fourth clinical variant – epidemic (AIDS related) KS – has since been added to the list. KS has also been reported in HIV-negative homosexual men, in whom it follows the indolent course found in classical KS. Although the incidence of KS in a meta-analysis of HIV cohort studies has declined markedly, it remains the commonest AIDS-associated cancer.2*** A substantial cohort study has confirmed that HAART protects against the development of KS and this may account for the falling incidence.3*
HHV8 discovery Early in the HIV pandemic it was noted that the epidemiology of AIDS-KS pointed to an infectious agent transmitted independently of HIV. An analysis of 13 000 persons with AIDS reported to the Centre for Disease Control, Atlanta, GA, USA up to 1989 revealed that the incidence of KS in homosexual and bisexual men infected with HIV was ten times greater than in other seropositive transmission groups.4* In 1994 Chang and Moore isolated unique DNA sequences from KS biopsies using representational difference analysis, a PCR-based technique that enriches DNA fragments (‘representations’) present in the tumour but absent from normal tissues. The original sequences that were found were homologous to, but distinct from, capsid and tegument protein genes of the gammaherpes viruses herpesvirus saimiri and Epstein–Barr virus (EBV).5 Complete genomic sequencing of this novel herpes virus,
initially named Kaposi’s sarcoma herpes virus (KSHV) and subsequently renamed human herpes virus-8 (HHV8) confirms that HHV8 is most closely related to herpes saimiri virus, which induces lymphoid malignancies in New World primates, and EBV, a human oncogenic herpes virus.6 The HHV8 genome consists of a 140 kb unique coding region flanked by terminal repeat sequences. The unique region includes at least 81 open reading frames (ORFs) that potentially encode proteins. In addition to structural proteins and viral enzymes, HHV8 encodes an unusually large number of pirated eukaryotic cellular proteins that may help it utilize host cellular processes and avoid antiviral responses. These pirated proteins include homologues for a G-protein coupled receptor, cyclin D2, bcl-2, and interferon regulatory factor (Fig. 47.1). HHV8 also encodes a viral homologue of interleukin (IL)-6, which is thought to act in a paracine manner and stimulate an acute phase response. In addition, HHV8 has been detected in two rare HIV-associated lymphoproliferative disorders: multicentric Castleman’s disease and primary effusion lymphoma.7,8
Molecular virology of HHV8 Molecular epidemiological studies and serologic assays have demonstrated that HHV8 is not ubiquitous. The virus appears more prevalent in those populations at increased risk for developing KS, including homosexual men, Africans and certain populations in Mediterranean Europe, while the prevalence of HHV8 appears quite low in the general population of the USA and UK. The mode of transmission of HHV8 remains unclear. Initial studies suggested that the virus was present in semen, however numerous investigators have failed to confirm this finding and the only frequent source of HHV8 from patients with KS is peripheral blood mononuclear cells and saliva. Seroepidemiological studies from areas of high prevalence have demonstrated some evidence of both horizontal and vertical transmission: HHV8 exists at high concentrations in the saliva and serpositivity rates in gay men are associated with the number of sexual partners, whilst data from Uganda suggests that vertical transmission does occur. The presence of HHV8 DNA in the peripheral blood
MIP-2 DHFR vIL-6
vCyclin D vFLIP
KSbcl-2 vIRF
TK
vOX2 GPCR
TR 0
TR 10
20
30
40
50
60
70 80 Kilobases
90
100
110
120
130
140
Figure 47.1 HHV8 genome and location of pirated cellular homologues. TR, Terminal repeat; TK, thymidine kinase; vIL-6, viral interleukin-6; vIRF, viral interferon regulatory factor; DHFR, dihydrofolate reductase; vFLIP, viral FLICE inhibitory protein; MIP-2, macrophage inhibitory protein-2; GPRC, G-protein-coupled receptor; KSbcl2, Kaposi’s sarcoma B-cell lymphoma 2.
Kaposi’s sarcoma 1169
of HIV-positive patients without KS has been shown to predict for the later development of clinical KS.9 Thus a group of patients at high risk of KS may be identified and could form the focus of preventative treatment strategies. The evidence for a causal role for HHV8 in the pathogenesis of KS is compelling. HHV8 infection precedes the clinical development of KS and the epidemiology and risk factors for HHV8 infection and KS overlap. HHV8 has been demonstrated in KS spindle cells and in the endothelial cells in KS lesions. Finally, HHV8 infects and transforms primary human endothelial cells, thought to be the precursor cells of malignant KS spindle cells.
Is KS a malignancy? All forms of KS have the same characteristic histology comprising spindle-shaped stromal cells, abnormal proliferating endothelial cells and extravasated erythrocytes. The major component of KS lesions is composed of endothelial cells, fibroblasts and inflammatory cells that form slit-like vascular channels to resemble neo-angiogenesis. The cell of origin for KS remains controversial. Angioblasts, mesenchymal cells, vascular and lymphatic endothelial cells and smooth muscle cells have all been proposed as potential spindle cell precursors, but more recently these cells have been shown to express lymphatic endothelial markers.10 It is uncertain whether KS is a polyclonal proliferation or a true malignancy. Spindle cells contain a normal chromosomal complement and lack nuclear atypia that might favour a non-malignant process. There is however, increasing evidence supporting the definition of KS as a true malignancy. KSY-1 a cell line derived from the pleural effusion of a man with AIDS-KS has metastatic properties in animal models. In addition tumour clonality has been shown in some human tumour biopsies by X chromosome inactivation patterns in women patients with AIDS-KS.11
Other pathogenetic factors CYTOKINES
The expression of growth factors and their receptors by KS cells has been demonstrated by immunocytochemical staining and in culture. Amongst the many growth factors that are expressed by KS cells are basic fibroblast growth factor (bFGF), IL-1β, IL-6, IL-8, oncostatin-M, and vascular endothelial growth factor (VEGF). These cytokines are mitogenic for KS cells in culture, suggesting that they may act as autocrine growth factors. In contrast, transforming growth factor-beta (TGF-β) which is also expressed by KS cells, acts as an autocrine growth inhibitor. Moreover, bFGF acts not only as an autocrine growth factor for KS cells but also stimulates endothelial cell migration and proliferation. Antisense oligonucleotides to bFGF mRNA block KS cell growth and lesion formation in nude mice
xenografts. Furthermore, subcutaneous injections of bFGF induce KS-like skin lesions in nude mice.12 Thus host cytokines appear to play an important role in the development of KS lesions and form a second potential target for novel therapeutic strategies. HIV VIRAL FACTORS
The role of HIV itself in the pathogenesis of KS has been a focus of research since early in the AIDS epidemic. The tat gene product of HIV is a potent transactivator which up-regulates viral gene expression by transcriptional and post-transcriptional enhancement and is necessary for viral replication. HIV tat is released from HIV-infected T-lymphocytes. In cell culture, tat stimulates KS cell growth and promotes the migration and proliferation of cytokineactivated endothelial cells. When injected subcutaneously into nude mice tat causes KS-like lesions and there is a synergy between tat and bFGF in this effect.12 Moreover, about 15 per cent male transgenic mice overexpressing the tat gene developed skin tumours resembling KS at age 12–18 months.13 These lesions were multifocal and contained spindle-shaped cells in the dermis and slit-like spaces with extravasated blood cells. The effect of tat in KS is thought to be mediated by integrin receptors. These are receptors for extracellular matrix (ECM) proteins that induce cell adhesion and invasion, facilitating angiogenesis. The tat protein contains an RGD domain, which is the receptor binding sequence for integrins and tat is able to bind to the VEGF receptor Flk-1 and induce a mitogenic signal. It is believed that these actions of tat may account for the rather aggressive course of AIDS-KS compared with the more indolent behaviour of KS in the HIV-seronegative population.
Clinical features and differential diagnosis AIDS-related KS has a wide variety of clinical presentations. The earliest cutaneous lesions are frequently asymptomatic, innocuous looking, macular pigmented lesions that vary in colour from faint pink to vivid purple. Larger plaques occur usually on the trunk as oblong lesions following the line of skin creases. Lesions may develop to form large plaques and nodules, which can be associated with painful oedema. Lymphatic infiltration is a common feature in the limbs and causes lymphoedema and ulceration. Oral lesions are a frequent accompaniment that may lead to ulceration, dysphagia and secondary infection. Gastrointestinal lesions are usually asymptomatic but may bleed. Pulmonary KS is a life-threatening complication that usually presents with dyspnoea, with or without fever and may cause haemoptysis. Chest X-ray typically reveals a diffuse reticulo-nodular infiltrate and pleural effusion. Visceral KS has been described in all organs at postmortem examinations. In making the diagnosis of visceral KS the complications of biopsy (particularly haemorrhage) must
1170 AIDS-related malignancy
Table 47.1 The modified AIDS Clinical Trials Group staging of KS14 Good risk (all of the following)
Poor risk (any of the following)
Tumour (T)
Confined to skin, lymph nodes or minimal oral disease
Tumour-associated oedema or ulceration Extensive oral KS Gastrointestinal KS KS in other non-nodal viscera
Immune status (I)
CD4 count 150/mm3
CD4 150/mm3
be carefully weighed up against the benefits of confirming the diagnosis histologically. The main differential diagnosis is bacillary angiomatosis or epithelioid angiomatosis, which is caused by a fastidious Gram-negative Rickettsia-like organism Bartonella henselae (previously known as Rochalimaea henselae). This infection can be effectively treated with erythromycin. These diagnoses can only be reliably distinguished histopathologically and so a biopsy is essential to confirm the diagnosis of KS.
Staging The staging system described for AIDS-KS does not follow the standard TNM approach but instead includes an assessment of immune function as determined by the CD4 cell count. In the modified AIDS Clinical Trails Group staging classification, poor risk tumour (T1) includes ulcerated KS, KS-associated oedema, nodular oral KS or KS involvement of any visceral organ. Poor immune status (I1) is defined as a CD4 T-lymphocyte count of 150/mm3 14* (see Table 47.1).
Treatment options The prognosis of patients with KS depends on the stage of disease, the level of immunosuppression, and the response to anti-HIV therapy. The management of AIDS patients with KS must balance the benefits of tumour regression with the potential effects of treatment upon the patient’s immunological and haematological status. For patients with symptomatic disease or life-threatening visceral disease prompt effective therapy is usually merited, whilst for patients with asymptomatic indolent lesions HAART alone may result in complete remission. HAART
The epidemiological evidence for a reduction in the incidence of AIDS-related KS as a first AIDS diagnosis was confused by the alterations to the definition of AIDS by the US Centres for Disease Control, most recently in 1993. However, early in the AIDS epidemic it was recognized through anecdotal reports that regression of KS occasionally occurred on anti-retroviral monotherapy with zidovudine.
Moreover, response rates in KS improved when zidovudine was added to interferon alpha. This led to speculation that immune reconstitution following the initiation of HAART therapy may lead to regression of KS and a number of case reports and small studies documenting responses of KS to HAART were published. The introduction of HAART in 1996 has indeed been followed by a fall in incidence of KS both as a first AIDS diagnosis and as a subsequent manifestation in HIVseropositive cohorts from established market economies. Larger cohort studies have demonstrated that HAART protects against the development of KS in people with HIV,3* and that in patients with established KS, the introduction of HAART therapy is associated with a prolongation of the time to treatment failure of KS.15* The latter effect complements the reported reduction in the incidence of KS developing in patients on HAART. Thus HAART therapy has a major influence both on the epidemiology and clinical progression of KS without apparently having a direct effect upon the causative herpesvirus, HHV8. The postulated mechanism of this effect is the immune reconstitution of cytotoxic T-lymphocyte responses to HHV8 and suppression of HIV replication.16 There is also evidence that antiretroviral agents may possess a direct anti-tumour effect.17 Protease inhibitors have been suggested to have specific anti-angiogenic and thus anti-tumour effects. However multivariate logistic regression analyses of 1204 cases of AIDS-KS seen between 1996 and 2002 found the incidence of KS in patients receiving non-PI-based HAART to be lower than that of patients receiving PI-containing HAART.3* This demonstrates that non-PI regimens are at least as effective as PI-based HAART in terms of protection against KS. A proportion of HIV-seropositive patients who subsequently commence HAART exhibit a deterioration in their clinical status despite control of their virological and immunological parameters. This response, known as immune reconstitution inflammatory syndrome (IRIS), occurs secondary to an immune response against previously diagnosed pathogens. One study of patients with HIV-associated KS starting HAART found that 6 per cent of patients developed progressive KS, which was identified as IRIS-KS, risk factors being higher CD4 counts and KSassociated oedema. The prognosis in the IRIS-KS group was no worse.18*
Kaposi’s sarcoma 1171
INTRALESIONAL CHEMOTHERAPY
As the regression of KS with HAART may take 6–12 months, localized therapies are advocated for patients with cosmetically significant limited cutaneous disease. Intralesional injection of a dilute solution of vinblastine (0.2 mg/mL) using volumes of up to 0.5 mL per lesion is an effective easy and well-tolerated treatment for lesions 1 cm in diameter. This treatment may cause pain at the site of injection and this may be reduced by prior use of local anaesthetic. There is frequently an initial flare reaction and lesions then regress, flatten and become fainter over the ensuing fortnight. However, post-treatment hyperpigmentation leaves brown spots at the sites and may be cosmetically unsatisfactory. Intralesional vinblastine has no significant systemic effects and injections may be repeated two or three times. This approach is also valuable for small intra-oral lesions and gingival lesions. RADIOTHERAPY
Larger cutaneous or oral lesions may be treated with radiotherapy and local control is generally achieved. For cutaneous lesions either a single fraction of 8 Gy or 16 Gy in four fractions is routinely used. Although the response rate and duration of local control may be better with fractionated regimens compared with single fraction treatment, toxicity and patient convenience are worse. Cosmetic improvement is usually achieved although there may be a halo appearance on account of the margin around treated lesions. Severe mucositis and acute oedema reactions may follow radiation treatment of the oral cavity and feet and for this reason treatment is given in four fractions of 4 Gy each at weekly intervals. Recurrent tumour within radiation fields is common and therefore radiotherapy treatment is usually reserved for symptomatic and cosmetically disturbing lesions. IMMUNOTHERAPY
Prior to the introduction of HAART, immunotherapy for good risk disease was been advocated in patients with well preserved immune function (generally CD4 counts 200/mm3). Interferons inhibit HIV replication and angiogenesis and interferon alpha (IFN-α) was the first agent licensed for use in AIDS-KS. Low-dose IFN-α (3–5 MU) produces response rates in KS of 10 per cent whilst higher doses (20 MU) yield response rates of 30 per cent. The response rates were highest in patients with higher CD4 levels, no B symptoms and no opportunistic infections. The response may take several weeks to achieve and there is considerable toxicity associated with this treatment including myalgia, arthralgia, fevers, chills, anorexia, weight loss, nausea, diarrhoea, anaemia, neutropaenia and elevated liver enzymes. Pegylated forms of IFN may be associated with a more favourable side-effect profile. In combination with nucleoside analogues, IFN acts synergistically on KS cells in vitro. Initial studies have shown higher response rates for
the combination of AZT and IFN-α, and lower IFN-α doses may be used. The combination causes considerable myelosuppression and G-CSF may be needed to compensate for this. The considerable toxicity of immunotherapy and its lack of efficacy in patients with advanced immunosuppression severely limit the benefits of this strategy. CHEMOTHERAPY
Chemotherapy is advocated for advanced cutaneous and visceral KS but is not merited for early disease in view of the brief response durations observed in randomized studies from the pre-HAART era.19** Early single-agent studies confirmed the activity of a number of cytotoxic agents, including anthracyclines, vinca alkaloids, bleomycin and etoposide. However although observed response rates were 20–60 per cent, responses were seldom durable and treatment was associated with significant toxicity. Consequently, liposomal anthracyclines have more recently become the established gold standard on the basis of guidelines for evaluating responses20* (Table 47.2). LIPOSOMAL ANTHRACYCLINES
Liposome encapsulation of anthracyclines constitutes a considerable advance in the chemotherapy of KS. The advantages of liposomal formulation include increased tumour uptake and hence favourable pharmacokinetics. Both liposome encapsulated daunorubicin (DaunoXome 40 mg/m2 every 2 weeks) and the pegylated liposomal doxorubicin (Doxil, Caelyx 20 mg/m2 every 3 weeks) have been shown to have good antitumour activity. The toxicity profile is better than for other anthracyclines, with no reported cardiotoxicity even with high cumulative dosages and rarely significant alopecia, however there remains considerable myelosuppression, and occasional emesis. In addition, infusionrelated hypotension and hand/foot syndrome are novel side effects seen with these liposomal formulations. Randomized comparisons of liposomal doxorubicin compared to conventional combination chemotherapy as first-line therapy for KS in patients not on HAART, found response rates were higher in the Doxil arms but were often not sustained.21,22** In the post-HAART era, response rates for liposomal anthracycline average around 70 per cent and are usually more prolonged.23* A phase III randomized comparison of DaunoXome and ABV in the pre-HAART era demonstrated equivalent response rates (25 per cent), time to treatment failure and survival duration.24** Only one study has investigated DaunoXome in patients receiving HAART.25* Overall response rate was 38 per cent, although one third of patients studied received DaunoXome as part of combination chemotherapy. Both studies found significant myelosuppression. Table 47.3 summarizes the results of phase III trials of liposomal anthracyclines for KS. These results may not be directly comparable but it is unlikely that a head-to-head comparison of Doxil and DaunoXome will be completed.
1172 AIDS-related malignancy
Table 47.2 Response criteria for HIV-associated Kaposi’s sarcoma20 Complete response (CR) The complete resolution of all KS with no new lesions, lasting for at least 4 weeks. A biopsy is required to confirm the absence of residual KS in flat lesions containing pigmentation. Endoscopies must be repeated to confirm the complete resolution of previously detected visceral disease Clinical complete response (CCR) Patients who have no detectable residual KS lesions for at least 4 weeks but whose response was not confirmed by biopsy and/or repeat endoscopy Partial response (PR) One or more of the following in the absence of (i) new cutaneous lesions, (ii) new visceral/oral lesions, (iii) increasing KS-associated oedema, (iv) a 25% or more increase in the product of the bi-dimensional diameters of any index lesion: 1. A 50% or greater decrease in the number of measurable lesions on the skin and/or in the mouth or viscera. 2. A 50% or greater decrease in the size of the lesions as defined by one of the following three criteria (a) a 50% or more decrease in the sums of the products of the largest bi-dimensional diameters of the index lesions; (b) a complete flattening of at least 50% of the lesions; (c) where 75% or more of the nodular lesions become indurated plaques. Stable disease (SD) Any response that does not meet the above criteria Progressive disease (PD) Any of the following: 1. 2. 3. 4.
A 25% or more increase in the product of the bi-dimensional diameters of any index lesion The appearance of new lesions Where 25% or more of previously flat lesions become raised The appearance of new or increased KS-associated oedema
Table 47.3 The results of phase III trials of liposomal anthracyclines for KS26 Agent
Daunoxome Doxil/Caelyx Doxil/Caelyx Doxil/Caelyx
Dose
Schedule
Assessable patients
Response rate
Median response duration
Ref
40 mg/m2 20 mg/m2 20 mg/m2 20 mg/m2
Every 2 weeks Every 2 weeks Every 2 weeks Every 3 weeks
116 133 62 121
25% 46% 79% 58%
3.8 m 3.0 m 8.1 m 5.0 m
24**
Based on the response rates, median response durations and the toxicity profile, liposomal anthracyclines are considered first-line chemotherapy for advance KS. PACLITAXEL
Interest in the use of taxanes as therapy for KS was fuelled by an improved tolerance to the side effects of chemotherapy following the introduction of HAART. Furthermore, evidence suggests an important role for angiogenesis in the pathogenesis of KS, whilst the anti-angiogenic effects of taxanes have been described in detail.27 Paclitaxel (100 mg/m2, every 2 weeks) has a valuable role in the management of refractory KS. The toxicities of taxanes are well recognized (neuropathy, myelosuppression, nausea/vomiting, fatigue, and alopecia)
21** 26** 22**
although appear to be no worse in patients with HIV than in other groups treated with equivalent dosages. A number of phase II studies of single-agent paclitaxel in the treatment of refractory KS have shown response rates of between 53 per cent and 71 per cent, and median response durations of 7.4–10.4 months.28–31* These studies included patients previously treated with anthracyclines or liposomal anthracyclines and have led to the rapid acceptance of paclitaxel as the treatment of choice for anthracycline refractory KS.
Novel therapies for KS The importance of angiogenesis and host cytokine activity in the pathogenesis of KS lesions and the recently identified
Systemic NHL 1173
role of HHV8 has led to several novel approaches to the management of KS.
SYSTEMIC NHL Epidemiology
TOPICAL RETINOIDS
In vitro retinoids inhibit the proliferation of KS-derived spindle cell lines and all-trans retinoic acid has been shown to induce apoptotic cell death in these lines. Moreover, retinoids down-regulate the expression of IL-6 receptors, which are thought to play a role in the autocrine stimulation of KS cells. Systemic therapy with both all-trans retinoic acid and 13-cis retinoic acid has been disappointing. More promising results have been reported for 9-cis retinoic acid, which binds to both classes of retinoid receptors: retinoic acid receptors (RARs) and retinoid X receptors (RXRs). In phase II and II trials, topical and oral 9-cis retinoic acid were associated with clinical response rates of 35–40 per cent, although application site reactions (e.g. erythema, bruising, flaking) were common.32–34* Indeed, the US Food and Drug Administration recently approved 0.1 per cent 9-cis-retinoic acid gel as a topical treatment for AIDS-related KS. ANTI-ANGIOGENIC THERAPIES
KS is an easily accessible and evaluable tumour that is composed predominantly of proliferating endothelial cells and is a valuable model of tumour angiogenesis. For this reason a number of anti-angiogenic agents have been studied in Kaposi’s sarcoma. Thalidomide inhibits bFGF-induced vascular proliferation, production of tumour necrosis factor alpha (TNF-α), intercellular adhesion and vascular maturation. Two small studies of thalidomide in AIDSrelated KS have documented regression of lesions.35,36* Numerous other anti-angiogenic agents have been investigated in clinical trials in KS. ANTI-HHV8 APPROACHES
The discovery of HHV8 opens up a potential therapeutic target for the management and prevention of HHV8associated malignancies. There are anecdotal reports of regression of KS with the anti-herpetic agent foscarnet and studies that demonstrate a reduced incidence of KS among patients previously treated with anti-herpes drugs.37* These data suggest that inhibition of HHV8 could have a role in the treatment and prophylaxis of KS. In vitro the most effective antiviral agent tested has been cidofovir, which in addition inhibits vIL-6 production by KS cells in culture. Furthermore, cidofovir inhibits the development of KS xenografts in nude mice. However early clinical studies with cidofovir have been disappointing.38 Although it is unlikely that antiviral agents will be of significant benefit in established tumours, their role in cancer prophylaxis amongst HHV8-seropositive patients may prove valuable.
The Centre for Disease Control (CDC) included highgrade B-cell non-Hodgkin’s lymphoma (NHL) as an AIDS-defining illness in 1985 following the description of NHL in 90 men from a population at risk for AIDS.39 Registry linkage studies in the pre-HAART era found that NHL in HIV-positive patients was 60–200 times commoner than in the general population.40,41* NHL was the AIDS-defining diagnosis in 16 per cent of patients in Europe in 1998. The annual rate of developing NHL after a diagnosis of AIDS is 2.4 per cent per year, remaining constant over a 5-year period. An estimated 5–10 per cent of patients will therefore develop HIV-NHL at some time during their illness. HIV-associated NHL accounts for 12–16 per cent of all deaths attributable to AIDS. It appears to be slightly commoner in haemophiliacs and less common in IV drug users than in other HIV transmission groups. The age incidence is bimodal with peaks in the 10–19-year and 50–59-year age groups, reflecting peaks in Burkitt’s lymphoma and diffuse large cell/immunoblastic lymphomas, respectively. The introduction of HAART has been associated with a dramatic reduction in the incidence of KS primary cerebral lymphoma (PCL) while the effects on NHL are less dramatic. A meta-analysis of 20 cohort studies between 1992–96 and 1997–99 confirmed an overall reduction in PCL and systemic immunoblastic lymphoma, but not in Burkitt’s lymphoma.2 Moreover a cohort study has confirmed that HAART protects against the development of systemic NHL.42*
Biology HIV-associated NHLs are B-cell aggressive lymphomas of high or intermediate grade. Approximately one third are classified as small non-cleaved cell (SNCCL) lymphomas (Burkitt or Burkitt-like lymphomas). The remaining two thirds are diffuse large cell (DLC) lymphomas. These may be immunoblastic lymphomas or large non-cleaved cell lymphomas. EPSTEIN–BARR VIRUS
Primary infection of epithelial cells by EBV is associated with the infection of some resting B lymphocytes via the CD3 receptor. Most infected B lymphocytes express EBV latently with a type 3 latency expression, producing up to 6 EBV nuclear antigens (EBNAs) as well as latent membrane proteins 1, 2A and 2B. These lymphocytes are destroyed by cytotoxic T lymphocytes. Some lymphocytes appear to switch to a latency 1 pattern expressing only EBNA-1 a nuclear antigen and these cells persist and are believed to be the origin of EBV-positive reactivation and NHL. Indeed,
1174 AIDS-related malignancy
EBV genomic terminal analysis has shown clonal EBV infection in HIV-associated NHL, implying that EBV infection precedes clonal expansion. HIV-associated diffuse large cell and immunoblastic lymphomas are frequently associated with EBV. The EBV genome in these lymphomas expresses latency type 3 antigens including EBNA-2 and LMP-1 and 2, which have transforming activity in vitro. Aciclovir has mild anti-EBV activity in vivo and one case–control study found that high-dose acyclovir for 1 year was associated with a decreased incidence of NHL. However, the EBV genome can be detected in only 60 per cent of these HIV-associated large cell lymphomas compared to almost all cases of posttransplantation NHL. This implicates other factors in the aetiology of these malignancies associated with HIV infection including polyclonal B-cell expansion and impaired T-cell immunosurveillance. In contrast, EBV is present in around 30 per cent HIV-associated Burkitt’s lymphoma and in these tumours EBV is found in the latency type 1 profile expressing only EBNA-1. c-Myc
Burkitt’s lymphoma (BL) is not associated with other forms of immunosuppression. The HIV-associated BL resembles sporadic BL in that only 30 per cent are EBV positive and c-myc/immunoglobulin gene translocations are found in most cases. The translocation breakpoints on chromosome 8 occur most frequently in exon1 or intron1 of the c-myc gene in AIDS-related NHL rather than 5 upstream of the gene. This pattern of translocation breakpoints on c-myc do not alter the peptide sequence since exon1 is not translated, and resemble the breakpoints observed in sporadic BL rather than endemic BL. The reciprocal breakpoint most often found lies in the Sμ portion of the immunoglobulin heavy chain gene on chromosome 14. This pattern, which is also found in sporadic BL, suggests that the mechanism of translocation is defective recombination during isotype class switching of the constant region of the heavy chain of immunoglobulin, which occurs late in B-lymphocyte ontogeny. HHV8
In addition to its role in KS, HHV8 has been demonstrated in both HIV-associated primary effusion lymphoma and Castleman’s disease.7,8 p53
The tumour suppressor gene p53 has a central role in cell cycle control and hence regulates cell replication. About 40 per cent of HIV-associated NHLs have been found to have mutations of p53 gene. These mutations are found most commonly associated with the small non-cleaved cell or Burkitt-like variants rather than the diffuse large cell histologies. In contrast mutations of the retinoblastoma gene, which is also a cell cycle regulator, have not been found in HIV-associated NHL.
bcl6
Chromosome 3q27 translocations in diffuse large cell lymphomas led to the identification of a novel oncogene bcl-6. This gene is rearranged in 30–40 per cent diffuse large cell lymphomas and rearrangements have been found in 20 per cent HIV-associated NHL, including mutations of the 5regulatory sequences. In contrast rearrangements of Bcl1 and Bcl-2 have not been demonstrated in AIDS-associated NHL.
Prognostic factors The staging of HIV-associated lymphomas follows the Ann Arbor system and is identical to that employed for the staging of non-HIV-related NHL. The majority of patients present with advanced stage, B symptoms and/or extra nodal disease and these factors are therefore less discriminatory regarding prognosis. The most frequent extranodal sites of lymphoma are the gastrointestinal tract, liver, bone marrow and cerebrospinal fluid. In the pre-HAART era, the most influential prognostic factors in patients with HIV-associated NHL related to the severity of immunosuppression rather than lymphoma-related factors. The International Prognostic Index (IPI) for lymphoma was introduced in 1993 to segregate aggressive lymphomas in terms of survival and this scoring system has been evaluated in HIV-associated lymphomas. Elevated serum LDH, age over 40 years and CD4 lymphocyte count 100/μL were confirmed as adverse prognostic variables. However, recent analysis has suggested that the prognostic variables in HIV-NHL closely resemble those in the IPI for non-HIV associated non-Hodgkin’s lymphoma and combining the IPI with the CD4 cell count has produced a valuable prognostic model43* (see Table 47.4). Indeed in the era of HAART, the prognosis of systemic HIV-NHL is approaching that for high-grade NHL in the general population. Table 47.4 Prognostic weightings for construction of the model in the HAART era43 Variable
Weighting
CD4 count 100/mm3 CD4 100/mm3 IPI score High IPI score High/intermediate IPI score Low/intermediate IPI score Low
1.34 0 2.9 1.84 1.0 0
Add up score: Total score 1 1–1.83 1.83–2.90 2.90
2-year overall survival 82% 47% 20% 15%
Systemic NHL 1175
Clinical presentation The majority of patients with HIV-associated lymphomas present with advanced stage and B symptoms however a biopsy is essential for the diagnosis of systemic lymphomas as they can be mimicked by many AIDS-related illnesses. Extranodal disease (especially hepatic), bone marrow involvement and leptomeningeal disease (particularly with Burkitt’s lymphoma) are all common features. Patients with systemic NHL should therefore be staged with a CT scan of chest/abdomen/pelvis, a bone marrow biopsy and a lumbar puncture. Since the introduction of HAART there has been no change in the stage at presentation, presence of B symptoms, bone marrow infiltration or performance status. However, patients who developed NHL in the HAART era were less likely to have had a prior AIDS diagnosis, were older, and had a high CD4 count at the time of diagnosis.44*
Management During the 1980s, conventional chemotherapy schedules were used at full dosages for patients with better prognostic factors. However, marked toxicity and an increased incidence of opportunistic infections led to modifications of the standard lymphoma regimens. The subsequent development of haematopoietic growth factors allowed more myelotoxic schedules to be studied. The appreciable death rate from opportunistic infections generally offset any decline in NHL-related deaths and most centres persisted with either dose-reduced chemotherapy schedules or prognostic stratification that reserved full-dose therapy for patients with the best prognostic factors only. Recently, a number of groups have described an improvement in the overall survival compared to historical controls since the introduction of HAART. The complete remission rate and overall survival with cyclophosphamide, doxorubicin, vincristine, prednisolone (CHOP) chemotherapy has improved with the addition of HAART to the chemotherapy45,46* and the goal of therapy is clearly complete remission, not palliation.47 However, in other series there has been no change in the lymphoma response rates and the improvements in survival duration may be related to reduced deaths from opportunistic infections amongst patients who achieve durable tumour remissions.48–50* Nonetheless these encouraging findings have led to a more aggressive approach to the management of AIDS-associated NHL. Infusional chemotherapy for high-grade lymphoma was pioneered at the Albert Einstein Cancer Centre in New York using the combination of cyclophosphamide, doxorubicin and etoposide (CDE) administered as a 96-hour continuous infusion together with granulocyte colony stimulating factor (G-CSF).51* Early reports of a selected group of 25 patients with AIDS-related lymphomas who were treated with CDE and didanosine produced an impressive median survival of
18.4 months and was widely heralded as a breakthrough.52 The same schedule was then combined with saquinavir, with similar results although there was more mucositis with the protease inhibitor. A large multicentre phase II trial of infusional CDE has been conducted by the Eastern Cooperative Oncology Group and the results have been far less impressive as is so often the case following encouraging initial single-centre studies.53* Moreover, there are both toxicity and pharmacokinetic drawbacks to the concomitant administration of chemotherapy and HAART. For example, the potentiation of myelotoxicity with infusional CDE chemotherapy combined with protease inhibitors may be a consequence of microsomal enzyme inhibition; reducing the metabolism of cytotoxics in this regimen PI-based HAART appears to significantly potentiate the myelotoxicity of CDE chemotherapy.54* At the National Cancer Institute, etoposide, prednisolone, vincristine, cyclophosphamide and doxorubicin (EPOCH), a dose-adjusted schedule has been developed which omits all HAART for the duration of the chemotherapy. Initial reports have been encouraging with a complete response rate of 79 per cent. However there was a dramatic fall in CD4 cell count during chemotherapy and even with restarting the HAART at the end of the chemotherapy this took 12 months to recover to baseline levels.55* This phase II study has been expanded to 39 selected patients now56* and EPOCH is currently under investigation in a multicentre study. As there are no comparative studies it is difficult to recommend an optimal gold standard therapy and there are advocates of conventional CHOP as well as supporters of infusional therapies. The high rate of leptomeningeal disease at presentation, which may be asymptomatic led to the widespread use of staging lumbar punctures and prophylactic intrathecal chemotherapy for patients considered to be at high risk of relapse in the cerebrospinal fluid. The prophylactic administration of intrathecal chemotherapy to patients with these risk factors but without meningeal disease at presentation prevented meningeal relapse in 81 per cent.57* Chemotherapy results in a decline in CD4 cell counts in both the immunocompetent and immunocompromised and prophylaxis to prevent opportunistic infections particularly in this patient group requires careful attention. It is well established in the management of HIV infection that prophylaxis against Pneumocystis jirovecii pneumonia should commence when the CD4 cell count falls below 200/mm3 and against Mycobacterium avium complex when it falls below 50/mm3.58,59*** The prolonged T-cell depletion recorded following EPOCH was previously demonstrated for patients receiving chemotherapy in the pre-HAART era.60 The concomitant use of chemotherapy and HAART has been widely practiced and when used together the CD4 cell count declines by 50 per cent during chemotherapy but recovers rapidly within 1 month of completing chemotherapy. The CD8 and natural killer (CD16 and CD56) cell counts follow a similar profile whilst the B-cell (CD19) count recovers more slowly
1176 AIDS-related malignancy
Table 47.5 Summary of recent studies reporting the outcome of treatment for HIV-associated systemic lymphomas63–68 Chemotherapy regimen
HAART
No. of patients
Ref.
Complete response rate
Median survival
Overall survival
CHOP CHOP-R CHOP/ACVBP
100% 100% 100%
50 99 35
62* 62* 63*
47% 58% 54%
110 weeks 139 weeks 22 months
CDE-R CHOP (liposomal dox) CDE (ECOG trial: HAART patients only) CDE CHOP CHOP EPOCH
76% 100%
74 24
64* 65*
70% 75%
23 months 16 months
NS NS 1 year 54% 2 year 49% 2 years 64% 1 year 58%
100%
55
66**
44%
14 months
100% 100% 100% 0%
46 24 25 39
54* 67* 68* 56*
50% 50% 45% 74%
26 months NYR NS NYR
1 year 57% 2 years 44% 2 years 61% 2 years 55% 60% at median follow up of 4.5 years)
NS, not stated; NYR, not yet reached.
but is restored to pre-chemotherapy levels by 3 months. There was no change in the HIV mRNA viral load during the chemotherapy.61* In view of the decline in CD4 cell count by 50 per cent, PCP prophylaxis should commence at CD4 cell counts of 400/mm3 and MAC at CD4 counts of 100/mm3. The improved survival described since the introduction of HAART and the preservation of immune function suggests that the combination of chemotherapy with HAART is an important step forward in the management of AIDSrelated lymphomas (see Table 47.5). However, there are both toxicity and pharmacokinetic drawbacks to the concomitant administration of chemotherapy and HAART. For example, the potentiation of myelotoxicity with CDE combined with protease inhibitors may be a consequence of microsomal enzyme inhibition reducing the metabolism of cytotoxics in this regimen.54*
New developments The improvements in the treatment of HIV infection have led to a more aggressive management strategy for AIDS-related lymphomas and this has resulted in better outcomes. Further refinements mirror those seen in immunocompetent patients with high-grade lymphoma including the addition of anti-CD20 antibodies as first-line therapy and the use of high-dose chemotherapy with autologous stem cell transplantation at first relapse. Rituximab in addition to chemotherapy has yielded increased response rates (70 per cent complete response with a 59 per cent 2-year survival).46* However, a randomized phase II study of CHOP rituximab has found no difference in response rates or durations, and a higher incidence of neutropenia appears to be associated with rituximab use.62** Patients
have also undergone successful autologous stem cell transplantation for AIDS-related lymphomas despite predictions that adequate harvesting would prove difficult on account of myelodysplasia.69–72
Primary effusion lymphoma Primary effusion lymphoma (PEL) or body cavity based lymphoma (BCBL) is a rare variant of HIV-associated lymphoma that is characterized by effusions in serosal cavities (pleura, pericardium, peritoneum) in the absence of solid nodal masses. PEL express an indeterminate immunophenotype with clonal immunoglobulin gene rearrangements. All PELs are associated with HHV8 infection and tumour cells carry a high HHV8 viral copy number per cell.7 In addition many PELs are co-infected with EBV. The clinical management of PEL does not differ from the HIVassociated NHL. Patients present with a median CD4 count of 90/μL whilst the median survival in one study is just 5 months.
PRIMARY CNS NHL Epidemiology Primary central nervous system lymphoma (PCNSL) is defined as NHL that is confined to the cranio-spinal axis without systemic involvement. This diagnosis is rare in immunocompetent patients but occurs more frequently in patients with both congenital and acquired immunodeficiency. Since 1985 high-grade B-cell NHL including PCNSL has been an AIDS-defining diagnosis. AIDSrelated PCNSL occurs equally frequently across all ages and
Primary CNS NHL 1177
transmission risk groups. A meta-analysis of cohort studies has shown a significant decline in the incidence of PCL following the introduction of HAART (relative risk 0.42)2*** and is attributed to the protective effects of HAART.73
Mass on cerebral CT scan
Toxoplasma serology CSF serology and PCR for EBV
Pathogenesis The presence of EBV is a universal feature of HIV-associated primary cerebral NHL, which are monoclonal immunoblastic lymphomas, but is not found in other primary cerebral lymphomas.74,75 EVB may be detected by immunocytochemical staining of biopsy tissue or by polymerase chain reaction (PCR) amplification of cerebrospinal fluid (CSF) using EBV-specific oligonucleotide primers.
Toxoplasma seropositive
Toxoplasma seronegative EBV seronegative
EBV seropositive
Trail of anti-toxoplasma therapy
Biopsy
Differential diagnosis Toxoplasmosis and lymphoma are the commonest causes of cerebral mass lesions in HIV-seropositive patients and the differential diagnosis often proves difficult. Both diagnoses occur in patients with advanced immunodeficiency (CD4 count 50/mm3) and present with headaches and focal neurological deficits. Clinical features that favour a diagnosis of PCNSL include a more gradual onset over 2–8 weeks and the absence of a fever. CT and MRI scanning usually reveal solitary or multiple ring enhancing lesions with prominent mass effect and oedema. Again these features occur in both diagnoses although PCNSL lesions are usually periventricular whilst toxoplasmosis more often affects the basal ganglia. Thus even the combination of clinical findings and standard radiological investigations rarely provide a definitive diagnosis. Moreover, toxoplasma serology (IgG) is falsely negative in 10–15 per cent of patients with cerebral toxoplasmosis. More than 85 per cent of patients with cerebral toxoplasmosis will respond clinically and radiologically to 2 weeks of anti-toxoplasma therapy and this has become the cornerstone of the diagnostic algorithm for cerebral masses in severely immunodeficient patients. In these patients it has been standard practice to commence empirical anti-toxoplasmosis treatment, and resort to a brain biopsy if there is no clinical or radiological improvement. This strategy avoids the routine use of brain biopsy in these patients who frequently have a very poor performance status and prognosis (see Fig. 47.2). The discovery that all HIV-associated PCNSLs, which histologically are diffuse large cell NHLs, are associated with EBV infection has led to the development of a PCR method that can detect EBV-DNA in the CSF. The detection of EBV-DNA in the CSF by PCR in patients with PCNSL has become established as a diagnostic test with a high sensitivity (83–100 per cent) and specificity (90 per cent).75–77 Radionuclide imaging by 201-thallium (201Th) singlephoton emission computed tomography (201Th-SPECT)
Figure 47.2 Diagnostic algorithm for primary cerebral non-Hodgkin’s lymphoma.
or 18F-fluorodeoxyglucose-positron emission tomography (FDG-PET) is able to differentiate between PCNSL and cerebral toxoplasmosis. PCNSLs are thallium avid and demonstrate increased uptake on PET scanning, however although both techniques have high specificity for PCNSL neither are highly sensitive and thus cannot be used as a single test but in combination with PCR are emerging as a diagnostic alternative to brain biopsy. The application of PCR and 201Th-SPECT in the diagnosis of contrast enhancing brain lesions in 27 patients was shown to result in a positive and a negative predictive value of 100 and 88 per cent, respectively, which supports their combined value as an alternative to brain biopsy.78 Further studies are now required to compare effectiveness of PCR with 201ThSPECT or FDG-PET.
Treatment The standard treatment modality for PCNSL in HIV patients is whole-brain irradiation. This treatment is associated with progressive impairment of cognitive function and neurological status secondary to chronic encephalopathy and the median survival time is just 2.5 months. The use of chemotherapy for PCNSL is limited by the poor penetration of cytotoxics into brain parenchyma on account of the blood–brain barrier and the toxicity, especially myelosuppression of these agents in patients with advanced immunosuppression and poor performance status. Combination chemotherapy regimens may prolong the median survival of PCNSL in immunocompetent patients but at the cost of severe myelotoxicity. Singleagent chemotherapy with IV high-dose methotrexate and folinic acid rescue was recently studied in AIDS patients with PCNSL in the context of a prospective uncontrolled study that included 15 patients. The results showed a complete response in 47 per cent of patients, a median survival
1178 AIDS-related malignancy
of 19 months; a low relapse rate of approximately 14 per cent and no evidence of neurological impairment nor treatment limiting myelotoxicity.79* A controlled trial of IV methotrexate vs. whole-brain irradiation is needed to confirm these encouraging results. Now that anti-retroviral therapies are improving survival it is necessary to reassess currently available diagnostic and treatment modalities aiming to cure HIV-associated brain lymphomas.
CASTLEMAN’S DISEASE Castleman’s disease is a rare lymphoproliferative disorder originally described in 1954 and characterized by angiofollicular lymphoid hyperplasia.80 It is divided into localized disease and multicentric Castleman’s disease (MCD), which is characterized by lymphadenopathy and multiorgan involvement. Histologically, two variants are recognized, a hyaline vascular variant and a less common plasma cell variant. The former is more common in localized disease and the latter more common in MCD. Castleman’s disease has germinal centre hyalinization or atrophy surrounded by concentric layers of lymphocytes with prominent vascular hyperplasia, hyalinization of small vessels and interfollicular sheets of plasma cells and immunoblasts. There is an association between HIV infection, Kaposi’s sarcoma and plasma cell MCD, and HHV8 has been found in nearly all MCD samples.8 The virus encodes a homologue of interleukin (IL)-6, a pro-inflammatory cytokine that is thought to mediate the constitutional symptoms including fever, weight loss and night sweats. Clinical findings include lymphadenopathy, hepatosplenomegaly, ascites, oedema, effusions, and rashes. Investigations frequently reveal microcytic anaemia, hypoalbuminaemia and polyclonal hypergammaglobulinaemia. Ultimately the diagnosis is made histologically. MCD is associated with a 15-fold increased incidence of NHL. The optimum treatment for Castleman’s disease remains uncertain and treatment is often based on case reports in the literature as there are no randomized trials. Surgery is often curative in localized Castleman’s disease, but has a limited role in MCD. Early reports of MCD in seropositive patients in the pre-HAART era suggested a median survival of less than 6 months. However, quicker diagnosis and treatment with splenectomy followed by singleagent chemotherapy may prolong survival. Variable benefit has been achieved with single-agent chemotherapy (most commonly vinblastine or etoposide), combination anthracycline-based chemotherapy, interferon-α and thalidomide.81* The IL-6 receptor blocking monoclonal antibody atlizumab, and the anti-CD20 monoclonal antibody rituximab have also been used, the latter being associated with a fall in IL-6 levels with resolution of clinical symptoms.82* Given the viral pathogenesis of MCD, the anti-herpes virus agent ganciclovir has been successful in reducing HHV8 viral load with a corresponding resolution in symptoms.
HODGKIN’S DISEASE Epidemiology The incidence of Hodgkin’s disease (HD) appears to be about ten times higher in individuals with HIV, although HD is not an AIDS-defining diagnosis. No significantly increased incidence of HD was detected in the earliest large study from San Francisco in 1984 of single men aged 20–49 and the New York cancer registry data of the same era. However, subgroup analysis of the latter data revealed an increased incidence of HD amongst HIV patients who were IV drug users (IVDUs). In contrast, in 1988 the Italian Co-operative Group for AIDS-related tumours described a cohort of 35 HIV patients with HD, almost all of whom were IVDUs.83 More recent studies have included larger numbers of patients and suggest an increased incidence of HD but the number of reported cases remains small and therefore the confidence intervals wide. Perhaps the most comprehensive data to date comes from the large linkage studies across the USA and Puerto Rica, which gave a relative risk of 8.6.84 This figure has been repeated in other large linkage studies and is significantly lower than that for NHL in patients with HIV. As yet there have been no reports of a change in the incidence of HD since the introduction of HAART. As with other malignancies associated with HIV, a virus (in this case EBV) has been implicated in the pathogenesis of HD and there appears to be a higher incidence of EBV detected in HIV-HD tissues than in HD samples from HIVseronegative patients. In addition, the natural history and treatment outcomes in HIV-HD may differ from those in the general population, although this seems less clear in the era of HAART. This may in part reflect the less favourable histology, advanced stage at presentation as well as the high incidence of opportunistic infection in HIV-HD.
Clinical presentation HIV-seropositive patients with HD generally present with more advanced lymphoma, with a higher incidence of stage III/IV disease, B-symptoms, and extra-nodal involvement. The bone marrow, skin, liver and central nervous system are the commonest extra-nodal sites. HD tends to present at an earlier stage of immunosuppression than HIV-associated NHL, with a higher median CD4 cell count at diagnosis, with between 50 and 90 per cent of patients being previously diagnosed with HIV, and with 11 per cent of patients having a prior AIDS-defining diagnosis.
Treatment The optimal chemotherapy schedule for HIV-HD has not been determined. However, MOPP, ABVD and MOPP/ ABVD hybrid regimens have been used in a number of series. No regimen has been shown to produce better results
Cervical cancer 1179
than any other, with 45–60 per cent complete remission rates and median survivals ranging from 8 to 18 months described in the pre-HAART era.85–87* These figures are significantly worse than the results published for the HD in the general population. A major cause of this poor survival is infectious complications; CD4 function, although often normal at presentation, becomes depressed during the progression of disease resulting in opportunistic infections (OIs) both during and after chemotherapy. Fortunately, recent reports suggest that the prognosis has improved following the introduction of HAART.88*
The therapeutic options for anal HSIL include surgical excision and laser ablation. These procedures are performed under general anaesthesia and although complications are uncommon, postoperative pain may persist for several days or weeks. Anal HSIL appears to be a field effect and hence recurrence is frequent, suggesting the need for medical therapies. The prolonged survival of HIV-infected people in the era of HAART and the lack of regression of anal HSIL may account for the observed trend towards an increasing incidence of anal cancer.
Invasive anal cancer ANAL CANCER Epidemiology The incidence of anal carcinoma amongst HIV-positive patients is 120 times higher that in the age- and gendermatched general population89* but it has been shown that homosexual men were at increased risk of this malignancy before the onset of the AIDS epidemic. Indeed the incidence of anal cancer amongst gay men in the pre-AIDS era was estimated to be 35/100 000, which resembles the incidence of cervical cancer before the introduction of routine Pap smear screening. Nonetheless, anal cancer is twice as common in HIV-positive gay men as it is in HIV-negative gay men,90 although unlike invasive cervical cancer, invasive anal cancer is not an AIDS-defining diagnosis. US AIDS cancer registry matching calculated that the relative risk of invasive anal cancer is 37 in HIV-positive men and 6.8 in HIV-positive women. The incidence of invasive anal cancer has not fallen in the era of HAART and there is no correlation between relative risk of anal cancer and CD4 cell count.89,91*
Pathogenesis Anal cancer shares many features with cervical cancer, including a strong association with human papilloma virus (HPV) infection and similar histology. High-grade squamous intraepithelial lesion (HSIL) or anal intra-epithelial neoplasia (AIN) of the anus is believed to progress to invasive anal cancer in a fashion analogous to the progression from cervical high-grade squamous intraepithelial lesion (HSIL) or cervical intraepithelial neoplasia (CIN) to invasive cervical cancer.
Pre-invasive anal lesions Cohort studies of men with anal HSIL have demonstrated that in the post-HAART era these lesions do not regress with HAART despite the established benefit of HAART on other viral infections and associated diseases in HIV-infected patients, nor has the prevalence of AIN declined.92,93 The optimal management of anal HSIL remains unclear.
The standard approach to anal cancer in immunocompetent patients was surgical until the report in 1974 that combined modality therapy (CMT) of chemotherapy (5FU and mitomycin C) with radiation treatment could result in microscopic and histological tumour ablation with sphincter preservation. Subsequently it was demonstrated that the resulting survival rates were at least as good as those achieved with surgery alone, most centres reporting that 85 per cent of tumours can be controlled locally with 5-year survival rates in the range of 65–85 per cent. It has been observed that HIV-infected people tolerate both chemotherapy and radiotherapy poorly. For this reason, coupled with the limited prognosis of patients with HIV prior to the introduction of more active antiretroviral regimens, there was some reluctance to treat anal carcinoma in the setting of HIV with standard-dose CMT. However, more recently standard CMT has been adopted for the management of HIV-associated anal cancer. Studies have found durable complete remission rates of between 71 per cent and 94 per cent and toxicity, although greater than in the HIV-negative population, is acceptable.94–97* Reductions in intensity of radiation or chemotherapy are seldom merited and indeed would be expected to compromise local control rates. A 5-year disease-free survival rate of 66 per cent has been reported for these patients.89*
Future The future of anal cancer in HIV-seropositive people may lie with effective screening of the at-risk population and early intervention. Anoscopic cytology has been found to be an effective method of screening for AIN and if therapeutic interventions could be shown to reduce progression and mortality this would be an attractive strategy.
CERVICAL CANCER Epidemiology Invasive cervical cancer was included as an AIDS-defining diagnosis in 1993, although at that time the incidence of
1180 AIDS-related malignancy
cervical cancer was not increased significantly in HIVseropositive women. Nonetheless there was good epidemiological evidence that the precursor lesions, cervical intra-epithelial neoplasia (CIN) or squamous intraepithelial lesion (SIL) occurred more frequently in women with HIV. The prolonged incubation between CIN and invasive cervical cancer of over 10 years may have accounted for the low incidence of invasive cervical cancer in HIV-seropositive women whose life expectancy in many parts of the world will be shorter than this. More recent studies in the era of HAART have determined a 4–6 fold increased relative risk of invasive cervical cancer in HIV-seropositive women.91,98* HPV has a central role in the pathogenesis of both CIN and invasive cervical cancer. In the USA, 30 per cent of female college students have cervical HPV infection and the modes of transmission of HIV and HPV are similar. Two oncoproteins present in HPV are believed to be responsible for the oncogenic properties of this virus, E6 and E7. E6 transforms cells by binding to the host cellular regulatory protein p53, forming a complex of p53, E6 and E6 associated protein. This complex is rapidly degraded in cellular proteosomes, resulting in depletion of p53, which leads to loss of the G1 checkpoint of cell cycle control. E7 binds the retinoblastoma protein Rb, releasing E2F transcription factor and this overcomes the G2/M checkpoint, resulting in the activation of mitosis. The Women’s Interagency HIV Study (WIHS) has studied cases of HIV-positive women and matched HIV-negative controls. An increased rate of HPV infection often with multiple HPV genotypes and a higher rate of cytological and histological precursor lesion has been found in these cases. Within this study the risk of SIL was greatest amongst women with CD4 cell counts 200/μL, where HPV was detected in cervical lavage specimens by PCR amplification, when multiple HPV genotypes were present and where HPV genotypes 16, 18, 31, 33 and 35 were detected99 (see Table 47.6). HIV is associated not only with a higher prevalence of HPV in the cervix, a high frequency of multiple HPV genotypes and persistence of HPV in the cervix, but also a higher prevalence of CIN/SIL, a higher progression from Table 47.6 Findings of WIHS study of cervical screening for HPV and CIN in HIV seropositive women and matched controls99
Number HVP detected in cervical lavage by PCR Multiple HPV genotypes present in lavage Abnormal PAP smear Cervical SIL on biopsy
HIV-positive cases
HIV-negative controls
2015
577
58%
26%
42% 49% 30%
16% 17% 7%
low-grade SIL to high-grade SIL and a greater likelihood of relapse of CIN II/III after therapy.
Effects of HAART on pre-invasive cervical cancer The effect of HAART on the natural history of CIN has been addressed in 49 women with advanced HIV. At 5 months after starting HAART the prevalence of CIN fell from 66 per cent to 49 per cent, regression of HGSIL to LGSIL occurred in 23 per cent and from LGSIL to normal in 43 per cent. These changes occurred without a significant change in the level of HPV DNA in cervical tissue.100 In a recent WIHS cohort from five US cities, the effect of HAART on CIN was assessed by 6-monthly smear testing. After adjustment for CD4 cell count and Papanicolaou smear status, women on HAART were 40 per cent (95 per cent confidence interval, 4–81 per cent) more likely to demonstrate regression and less likely (odds ratio, 0.68; 95 per cent confidence interval, 0.52–0.88) to demonstrate progression.101 However the benefits of HAART have not been reproduced in all studies and HAART appears to have limited ability to clear HPV infection and induce regression of CIN in HIV-positive women. These findings suggest that frequent cervical smears should be offered to all HIV-positive women, although the frequency remains controversial.102 Furthermore, CIN 2/3 should be aggressively treated, followed up by frequent colposcopic surveillance, and should be considered an indication for starting HAART therapy.
Invasive cervical cancer management In most centres HIV-seropositive women with invasive cervical cancer are treated using the same protocols as immunocompetent women. One retrospective series compared 16 HIV-positive women with 68 seronegative women treated at the same institution and time period. The results demonstrated that women with HIV had more advanced cervical cancer at presentation, relapsed more frequently and had a worse median survival (see Table 47.7).103 There is as yet little data to suggest that the introduction of HAART has improved the survival of invasive cervical cancer.104,105 Table 47.7 Comparison of HIV-negative and HIV-positive women with invasive cervical cancer from one centre103
Number Stage 3/4 Relapsed after definitive therapy Median survival
HIV positive
HIV negative
16 70% 100% 9 months
68 28% 49% 28 months
Other tumours 1181
Future The most attractive strategies aimed at reducing the mortality of cervical cancer have focused on vaccination against HPV both as prophylaxis and potential treatment. Virus-like particles (VLPs) are composed of HPV capsid proteins that autoassemble in the absence of viral DNA and these may be manipulated to incorporate additional proteins. These modified VLPs are highly antigenic and have been shown to be efficient inducers of neutralizing antibodies to HPV. In a recent high profile study of a bivalent HPV-16/18 VLP vaccine, an according-to-protocol analysis after 2 years found a 96 per cent protection against incidental infection, and a 100 per cent protection against persistent infection.106** In this intention-to-treat analysis, vaccine efficacy was 93 per cent against cytological abnormalities associated with HPV-16/18. The vaccine was safe, well-tolerated and highly immunogenic.
LEIOMYOSARCOMA IN CHILDREN Smooth muscle tumours, including leiomyomas and leiomyosarcomas are very uncommon tumours in childhood but are found more frequently in children with HIV107 where they are the second commonest tumour and are an AIDS-defining diagnosis. Leiomyosarcomas are also rarely reported in young adults and may be associated with EBV.108
SQUAMOUS CELL CARCINOMAS OF CONJUNCTIVA Squamous cell carcinoma of the conjunctiva typically presents with ocular surface epithelial dysplasia most frequently on the nasal aspect of the eye. Metastases are very rare and the prognosis with local excision is good. These tumours occur more frequently in subSaharan Africa, where the incidence is up to 12/106 per year and have been thought to relate to UV light exposure. A number of case–control studies have been conducted in Africa demonstrating an increased incidence of these tumours in HIV-positive people with a relative risk of around 10.109* In a cattle model for squamous cell carcinoma of the conjunctiva an association has been found with bovine papilloma virus, suggesting that HPV might have a role in the pathogenesis of this malignancy in humans.
of immunosuppression.110,111 Similarly dysplastic naevi and melanoma have been reported with HIV and may occur more frequently than expected. The lesions are frequently thicker and metastatsize early compared to non-HIV-infected persons, particularly when the CD4 count is low.
TESTICULAR GERM CELL TUMOURS Testicular tumours including lymphoma, non-seminoma germ cell tumour (NSGCT) and seminoma occur more frequently in HIV-positive men than in the general population. Wilson et al reported an incidence of 0.2 per cent of testicular tumours in 3015 HIV-positive men seen over a 2-year period, which was 57 times greater than in the general population.112 The introduction of HAART has not had a significant effect on incidence. In the pre-HAART era there was a reduced overall survival for HIV-positive patients with testicular cancer compared to those in the general population. However since the introduction of HAART, HIV-positive patients have similar response rates and tumour-free survival to those in the general population with both localized and metastatic testicular cancer.113*
OTHER TUMOURS Rates of other cancers appear to be marginally increased in HIV-positive patients, including myeloma, melanoma and non-small cell lung cancer, especially adenocarcinoma of the lung.114,115 Intriguinely, there are also reports of a reduced risk of breast cancer and prostate cancer in HIV disease.116 In general since the introduction of HAART, the management and prognosis of these tumours resembles that in the general population.117
KEY LEARNING POINTS ●
●
●
OTHER SKIN TUMOURS The risk for development of skin tumours other than KS appears to be greater in HIV-positive persons. The second commonest skin tumour is basal cell carcinoma and these frequently occur on the trunk, are superficial and multicentric. There is no correlation between basal cell cancers and degree
●
Kaposi sarcoma and high-grade NHL are AIDSdefining cancers. The incidence of both is reduced by HAART and the prognosis has improved with this therapy. Human papillomavirus-associated malignancies including cervical and anal cancer also occur at increased incidence in HIV-positive people but HAART does not appear to reduce this risk or improve the survival. The management of malignancy in this population is complicated by the increased risk of opportunistic infections during cancer therapy and potential interactions between anticancer and anti-HIV therapies. In experienced centres, the prognosis of these malignancies now approaches that seen in the general population.
1182 AIDS-related malignancy
REFERENCES ●1
●2
3
4
●5
6
7
8
9
10
11
12
13
❊14
15
Friedman-Kien A, Laubenstein L, Marmor M, et al. Kaposi’s sarcoma and pneumocystis pneumonia among homosexual men – New York and California. Morbid Mortal Week Rep 1981; 30:250–4. International Collaboration on HIV and Cancer: Highly active antiretroviral therapy and incidence of cancer in human immunodeficiency virus-infected adults. J Natl Cancer Inst 2000; 92(22):1823–30. Portsmouth S, Stebbing J, Gill J, Mandalia S, Bower M, Nelson M, Gazzard B. A comparison of regimens based on non-nucleoside reverse transcriptase inhibitors or protease inhibitors in preventing Kaposi’s sarcoma. AIDS 2003; 17(11):17–22. Beral V, Peterman TA, Berkelman RL, Jaffe HW. Kaposi’s sarcoma among persons with AIDS: a sexually transmitted infection? Lancet 1990; 335:123–8. Chang Y, Cesarman E, Pessin MS, Lee F, Culpepper J, Knowles DM, Moore PS. Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi’s sarcoma. Science 1994; 266(5192):1865–9. Russo JJ, Bohenzky RA, Chien MC, et al. Nucleotide sequence of the Kaposi sarcoma-associated herpesvirus (HHV8). Proc Natl Acad Sci U S A 1996; 93(25):14862–7. Cesarman E, Chang Y, Moore PS, Said JW, Knowles DM. Kaposi’s sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas. N Engl J Med 1995; 332(18):1186–91. Soulier J, Grollet L, Oskenhendler E, et al. Kaposi’s sarcomaassociated herpesvirus-like DNA sequences in multicentric Castleman’s disease. Blood 1995; 86:1276–80. Whitby D, Howard MR, Tenant-Flowers M, et al. Detection of Kaposi’s sarcoma-associated herpesvirus (KSHV) in peripheral blood of HIV-infected individuals predicts progression to Kaposi’s sarcoma. Lancet 1995; 364:799–802. Carroll PA, Brazeau E, Lagunoff M. Kaposi’s sarcomaassociated herpesvirus infection of blood endothelial cells induces lymphatic differentiation. Virology 2004; 328(1):7–18. Rabkin CS, Bedi G, Musaba E, Sunkutu R, Mwansa N, Sidransky D, Biggar RJ. AIDS-related Kaposi’s sarcoma is a clonal neoplasm. Clin Cancer Res 1995; 1:257–60. Ensoli B, Gendelman R, Markham P, et al. Synergy between basic fibroblast growth factor and HIV-1 Tat protein in induction of Kaposi’s sarcoma. Nature 1994; 371:674–80. Vogel J, Hinrichs SH, Reynolds RK, Luciw PA, Jay G. The HIV tat gene induces dermal lesions resembling Kaposi’s sarcoma in transgenic mice. Nature 1988; 335(6191):606–11. Krown SE, Testa MA, Huang J. AIDS-related Kaposi’s sarcoma: prospective validation of the AIDS Clinical Trials Group staging classification. AIDS Clinical Trials Group Oncology Committee. J Clin Oncol 1997; 15(9):3085–92. Bower M, Fox P, Fife K, Gill J, Nelson M, Gazzard B. Highly active anti-retroviral therapy (HAART) prolongs time to treatment failure in Kaposi’s sarcoma. AIDS 1999; 13(15):2105–11.
16 Osman M, Kubo T, Gill J, et al. Identification of human herpesvirus 8-specific cytotoxic T-cell responses. J Virol 1999; 73(7):6136–40. 17 Monini P, Sgadari C, Toschi E, Barillari G, Ensoli B. Antitumour effects of antiretroviral therapy. Nat Rev Cancer 2004; 4(11):861–75. 18 Bower M, Nelson M, Young AM, et al. Immune reconstitution inflammatory syndrome associated with Kaposi’s sarcoma. J Clin Oncol 2005; 23(22):5224–8. 19 Uthayakumar S, Bower M, Money-Kyrle J, et al. Randomized cross-over comparison of liposomal daunorubicin versus observation for early Kaposi’s sarcoma. AIDS 1996; 10(5):515–9. ❊20 Krown S, Metroka C, Wernz JC. Kaposi’s sarcoma in the acquired immune deficiency syndrome: a proposal for uniform evaluation, response, and staging criteria. AIDS Clinical Trials Group Oncology Committee. J Clin Oncol 1989; 7:1201–7. 21 Northfelt DW, Dezube BJ, Thommes JA, et al. Pegylatedliposomal doxorubicin versus doxorubicin, bleomycin, and vincristine in the treatment of AIDS-related Kaposi’s sarcoma: results of a randomized phase III clinical trial. J Clin Oncol 1998; 16(7):2445–51. 22 Stewart S, Jablonowski H, Goebel FD, et al. Randomized comparative trial of pegylated liposomal doxorubicin versus bleomycin and vincristine in the treatment of AIDS-related Kaposi’s sarcoma. International Pegylated Liposomal Doxorubicin Study Group. J Clin Oncol 1998; 16(2) :683–91. 23 Lichterfeld M, Qurishi N, Hoffmann C, et al. Treatment of HIV-1-associated Kaposi’s sarcoma with pegylated liposomal doxorubicin and HAART simultaneously induces effective tumor remission and CD4+ T cell recovery. Infection 2005; 33(3):140–7. 24 Gill PS, Wernz J, Scadden DT, et al. Randomized phase III trial of liposomal daunorubicin versus doxorubicin, bleomycin, and vincristine in AIDS-related Kaposi’s sarcoma. J Clin Oncol 1996; 14(8):2353–64. 25 Rosenthal E, Poizot-Martin I, Saint-Marc T, Spano JP, Cacoub P. Phase IV study of liposomal daunorubicin (DaunoXome) in AIDS-related Kaposi sarcoma. Am J Clin Oncol 2002; 25(1):57–9. 26 Mitsuyasu R, Von Roenn J, Krown S, et al. Comparison study of liposomal doxorubicin alone or with bleomycin and vincristine for treatment of advanced AIDS-associated Kaposi’s sarcoma (AIDS-KS): AIDS Clinical Trial Group (ACTG) protocol 286. Proc Am Soc Clin Oncol 1997; 16:55a. 27 Sgadari C, Barillari G, Toschi E, et al. HIV protease inhibitors are potent anti-angiogenic molecules and promote regression of Kaposi sarcoma. Nat Med 2002; 8(3):225–32. 28 Saville MW, Lietzau J, Pluda JM, et al. Treatment of HIVassociated Kaposi’s sarcoma with paclitaxel. Lancet 1995; 346(8966):26–8. 29 Gill PS, Tulpule A, Espina BM, et al. Paclitaxel is safe and effective in the treatment of advanced AIDS-related Kaposi’s sarcoma. J Clin Oncol 1999; 17(6):1876–83. 30 Stebbing J, Wildfire A, Portsmouth S, et al. Paclitaxel for anthracycline-resistant AIDS-related Kaposi’s
References 1183
31
32
33
34
35
36
37
38
●39
40 41
42
43
44
45
sarcoma: clinical and angiogenic correlations. Ann Oncol 2003; 14(11):1660–6. Tulpule A, Groopman J, Saville MW, et al. Multicenter trial of low-dose paclitaxel in patients with advanced AIDSrelated Kaposi sarcoma. Cancer 2002; 95(1):147–54. Walmsley S, Northfelt DW, Melosky B, Conant M, Friedman-Kien AE, Wagner B. Treatment of AIDS-related cutaneous Kaposi’s sarcoma with topical alitretinoin (9-cisretinoic acid) gel. Panretin Gel North American Study Group. J Acquir Immune Defic Syndr 1999; 22(3):235–46. Aboulafia DM, Norris D, Henry D, et al. 9-cis-retinoic acid capsules in the treatment of AIDS-related Kaposi sarcoma: results of a phase 2 multicenter clinical trial. Arch Dermatol 2003; 139(2):178–86. Miles SA, Dezube BJ, Lee JY, et al. Antitumor activity of oral 9-cis-retinoic acid in HIV-associated Kaposi’s sarcoma. AIDS 2002; 16(3):421–9. Fife K, Howard MR, Gracie F, Phillips RH, Bower M. Activity of thalidomide in AIDS-related Kaposi’s sarcoma and correlation with HHV8 titre. Int J STD AIDS 1998; 9(12):751–5. Welles L, Little R, Wyvill K, et al. Preliminary results of a phase II study of oral thalidomide in patients with HIV infection and Kaposi’s sarcoma (KS). J Immune Defic Syndr Hum Retrovir 1997; 14:A21. Mocroft A, Youle M, Gazzard BG, Morcinek J, Halai R, Phillips AN. Antiherpesvirus treatment and risk of Kaposi’s sarcoma in HIV infection. Royal Free/Chelsea and Westminster Hospitals Collaborative Group. AIDS 1996; 10:1101–5. Little RF, Merced-Galindez F, Staskus K, et al. A pilot study of cidofovir in patients with Kaposi sarcoma. J Infect Dis 2003; 187(1):149–53. Epub 2002 Dec 13. Ziegler JL, Beckstead JA, Volberding PA, et al. Non-Hodgkin’s lymphoma in 90 homosexual men. Relation to generalized lymphadenopathy and the acquired immunodeficiency syndrome. N Engl J Med 1984; 311(9):565–70. Beral V, Peterman T, Berkelman R, Jaffe H. AIDS-associated non-Hodgkin lymphoma. Lancet 1991; 337(8745):805–9. Biggar RJ, Rosenberg PS, Cote T. Kaposi’s sarcoma and non-Hodgkin’s lymphoma following the diagnosis of AIDS. Multistate AIDS/Cancer Match Study Group. Int J Cancer 1996; 68(6):754–8. Stebbing J, Gazzard B, Mandalia S, et al. Antiretroviral treatment regimens and immune parameters in the prevention of systemic AIDS-related non-Hodgkin’s lymphoma. J Clin Oncol 2004; 22(11):2177–83. Bower M, Gazzard B, Mandalia S, et al. A prognostic index for systemic AIDS-related non-Hodgkin lymphoma treated in the era of highly active antiretroviral therapy. Ann Intern Med 2005; 143(4):265–73. Matthews GV, Bower M, Mandalia S, Powles T, Nelson MR, Gazzard BG. Changes in acquired immunodeficiency syndrome-related lymphoma since the introduction of highly active antiretroviral therapy. Blood 2000; 96(8):2730–4. Navarro JT, Ribera JM, Oriol A, et al. Influence of highly active anti-retroviral therapy on response to treatment and survival in patients with acquired immunodeficiency syndrome-related non-Hodgkin’s lymphoma treated with
46
47
48
49
50
51
52
53
54
55
56
57
❊58
❊59
cyclophosphamide, hydroxydoxorubicin, vincristine and prednisone. Br J Haematol 2001; 112:909–15. Spina M, Jaeger U, Sparano JA, et al. Rituximab plus infusional cyclophosphamide, doxorubicin, and etoposide in HIV-associated non-Hodgkin lymphoma: pooled results from 3 phase 2 trials. Blood 2005; 105(5):1891–7. Spina M, Carbone A, Vaccher E, et al. Outcome in patients with non-Hodgkin lymphoma and with or without human immunodeficiency virus infection. Clin Infect Dis 2004; 38(1):142–4. Antinori A, Cingolani A, Alba L, et al. Better response to chemotherapy and prolonged survival in AIDS-related lymphomas responding to highly active antiretroviral therapy. AIDS 2001; 15(12):1483–91. Besson C, Goubar A, Gabarre J, et al. Changes in AIDSrelated lymphoma since the era of highly active antiretroviral therapy. Blood 2001; 98(8):2339–44. Bower M, Stern S, Fife K, Nelson M, Gazzard BG. Weekly alternating combination chemotherapy for good prognosis AIDS-related lymphoma. Eur J Cancer 2000; 36(3):363–7. Sparano JA, Lee S, Henry DH, Ambinder RF, von Roenn J, Tirelli U. Infusional cyclophosphamide, doxorubicin and etoposide in HIV associated non-Hodgkin’s lymphoma: A review of the Einstein, Aviano and ECOG experience in 182 patients. J AIDS 2000; 23:A11. Sparano JA, Wiernik PH, Hu X, et al. Pilot trial of infusional cyclophosphamide, doxorubicin and etoposide plus didanosine and filgrastim in patients with HIV associated non-Hodgkin’s lymphoma. J Clin Oncol 1996; 14:3026–35. Sparano JA, Anand K, Desai J, Mitnick RJ, Kalkut GE, Hanau LH. Effect of highly active antiretroviral therapy on the incidence of HIV-associated malignancies at an urban medical center. J Acquir Immune Defic Syndr 1999; 21 Suppl 1:18–22. Bower M, McCall-Peat N, Ryan N, et al. Protease inhibitors potentiate chemotherapy-induced neutropenia. Blood 2004; 104(9):2943–6. Little R, Pearson D, Gutierrez M, Steinberg S, Yarchoan R, Wilson W. Dose-adjusted chemotherapy with suspension of antiretroviral therapy for HIV-associated non-Hodgkin’s lymphoma. J AIDS 2000; 23:A11. Little RF, Pittaluga S, Grant N, et al. Highly effective treatment of acquired immunodeficiency syndrome-related lymphoma with dose-adjusted EPOCH: impact of antiretroviral therapy suspension and tumor biology. Blood 2003; 101(12):4653–9. Levine AM, Wernz JC, Kaplan L, et al. Low dose chemotherapy with central nervous system prophylaxis and zidovudine maintenance in AIDS-related lymphoma. J Am Med Assoc 1991; 266:84–8. Kaplan JE, Masur H, Holmes KK. Guidelines for preventing opportunistic infections among HIV-infected persons–2002. Recommendations of the U.S. Public Health Service and the Infectious Diseases Society of America. MMWR Recomm Rep 2002; 51(RR-8):1–52. Yeni PG, Hammer SM, Hirsch MS, et al. Treatment for adult HIV infection: 2004 recommendations of the International AIDS Society-USA Panel. JAMA 2004; 292(2):251–65.
1184 AIDS-related malignancy
60 Zanussi S, Simonelli C, D’Andrea M, et al. The effects of antineoplastic chemotherapy on HIV disease. AIDS Res Hum Retroviruses 1996; 12(18):1703–7. 61 Powles T, Imami N, Nelson M, Gazzard BG, Bower M. Effects of combination chemotherapy and highly active antiretroviral therapy on immune parameters in HIV-1 associated lymphoma. AIDS 2002; 16(4):531–6. 62 Kaplan LD, Lee JY, Ambinder RF, et al. Rituximab does not improve clinical outcome in a randomized phase III trial of CHOP with or without rituximab in patients with HIVassociated non-Hodgkin’s lymphoma: AIDS-malignancies consortium trial 010. Blood 2005; 24:1538–43. 63 Lascaux AS, Hemery F, Goujard C, et al. Beneficial effect of highly active antiretroviral therapy on the prognosis of AIDS-related systemic non-Hodgkin lymphomas. AIDS Res Hum Retroviruses 2005; 21(3):214–20. 64 Spina M, Jaeger U, Sparano JA, et al. Rituximab plus infusional cyclophosphamide, doxorubicin, and etoposide in HIV-associated non-Hodgkin lymphoma: pooled results from 3 phase 2 trials. Blood 2005; 105(5):1891–7. Epub 2004 Nov 18. 65 Levine AM, Tulpule A, Espina B, et al. Liposomeencapsulated doxorubicin in combination with standard agents (cyclophosphamide, vincristine, prednisone) in patients with newly diagnosed AIDS-related non-Hodgkin’s lymphoma: results of therapy and correlates of response. J Clin Oncol 2004; 22(13):2662–70. 66 Sparano JA, Lee S, Chen MG, et al. Phase II trial of infusional cyclophosphamide, doxorubicin, and etoposide in patients with HIV-associated non-Hodgkin’s lymphoma: an Eastern Cooperative Oncology Group Trial (E1494). J Clin Oncol 2004; 22(8):1491–500. 67 Vaccher E, Spina M, di Gennaro G, et al. Concomitant cyclophosphamide, doxorubicin, vincristine, and prednisone chemotherapy plus highly active antiretroviral therapy in patients with human immunodeficiency virus-related, non-Hodgkin lymphoma. Cancer 2001; 91(1) :155–63. 68 Ratner L, Lee J, Tang S, et al. Chemotherapy for human immunodeficiency virus-associated non-Hodgkin’s lymphoma in combination with highly active antiretroviral therapy. J Clin Oncol 2001; 19(8):2171–8. 69 Molina A, Krishnan AY, Nademanee A, et al. High dose therapy and autologous stem cell transplantation for human immunodeficiency virus-associated non-Hodgkin lymphoma in the era of highly active antiretroviral therapy. Cancer 2000; 89(3):680–9. 70 Gabarre J, Azar N, Autran B, Katlama C, Leblond V. Highdose therapy and autologous haematopoietic stem-cell transplantation for HIV-1-associated lymphoma. Lancet 2000; 355(9209):1071–2. 71 Kentos A, Vekemans M, Van Vooren JP, Lambermont M, Liesnard C, Feremans W, Farber CM. High-dose chemotherapy and autologous CD34-positive blood stem cell transplantation for multiple myeloma in an HIV carrier. Bone Marrow Transplant 2002; 29(3):273–5. 72 Re A, Cattaneo C, Michieli M, et al. High-dose therapy and autologous peripheral-blood stem-cell transplantation as
73
74
75
76
77
78
❊79
◆80
81 82
83
84
85 86
87
salvage treatment for HIV-associated lymphoma in patients receiving highly active antiretroviral therapy. J Clin Oncol 2003; 21(23):4423–7. Bower M, Mazhar D, Nelson M, Mandalia S, Gazzard BG, Stebbing J. The influence of HAART on HIV-associated primary cerebral lymphoma. J Natl Cancer Inst 2006; 98:1088–91. MacMahon EM, Glass JD, Hayward SD, et al. Epstein–Barr virus in AIDS-related primary central nervous system lymphoma. Lancet 1991; 338(8773):969–73. Cinque P, Brytting M, Vago L, et al. Epstein–Barr virus DNA in cerebrospinal fluid from patients with AIDS-related primary lymphoma of the central nervous system. Lancet 1993; 342:398–401. Arribas J, Clifford D, Fichtenbaum C, Roberts R, Powderly W, Storch G. Detection of Epstein–Barr virus DNA in cerebrospinal fluid for diagnosis of AIDS-related central nervous system lymphoma. J Clin Microbiol 1995; 33(6):1580–3. De Luca A, Antinori A, Cingolani A, et al. Evaluation of cerebrospinal fluid EBV-DNA and IL-10 as markers for in vivo diagnosis of AIDS-related primary central nervous system lymphoma. Br J Haematol 1995; 90(4):844–9. Castagna A, Cinque P, d’Amico A, Messa C, Fazsio F, Lazzarin A. Evaluation of contrast-enhancing brain lesions in AIDS patients by means of Epstein–Barr virus detection in cerebrospinal fluid and 201thallium single photon emission tomography. AIDS 1997; 11:1522–23. Jacomet C, Girard P, Lebrette M, Farese V, Monfort L, Rozenbaum W. Intravenous methotrexate for primary central nervous system non-Hodgkin’s lymphoma in AIDS. AIDS 1997; 11:1725–30. Castleman B, Towne VW. Case records of the Massachusetts General Hospital: Case No. 40231. N Engl J Med 1954; 250(23):1001–5. Waterston A, Bower M. Fifty years of multicentric Castleman’s disease. Acta Oncol 2004; 43(8):698–704. Newsom-Davis T, Bower M, Wildfire A, Thirlwell C, Nelson M, Gazzard B, Stebbing J. Resolution of AIDS-related Castleman’s disease with anti-CD20 monoclonal antibodies is associated with declining IL-6 and TNF-alpha levels. Leuk Lymphoma 2004; 45(9):1939–41. Tirelli U, Vaccher E, Rezza G, et al. Hodgkin disease and infection with the human immunodeficiency virus (HIV) in Italy. Ann Intern Med 1988; 108(2):309–10. Goedert JJ, Cote TR, Virgo P, et al. Spectrum of AIDSassociated malignant disorders. Lancet 1998; 351(9119):1833–9. Rubio R. Hodgkin’s disease associated with HIV: A clinical study of 46 cases. Cancer 1994; 73:2400–7. Tirelli U, Vaccher E, Rezza G, et al. Hodgkin’s disease and infection with the human immunodeficiency virus in Italy. Ann Intern Med 1988; 108:309–10. Levy R, Colonna P, Tourani J-M, et al. Human immunodeficiency virus associated Hodgkin’s disease: Report of 45 cases from the French Registry of HIV-associated tumours. Leuk Lymphoma 1995; 16:451–6.
References 1185
88 Hoffmann C, Chow KU, Wolf E, et al. Strong impact of highly active antiretroviral therapy on survival in patients with human immunodeficiency virus-associated Hodgkin’s disease. Br J Haematol 2004; 125(4):455–62. 89 Bower M, Powles T, Newsom-Davis T, et al. HIV-associated anal cancer: Has highly active antiretroviral therapy reduced the incidence or improved the outcome? J Acquir Immune Defic Syndr 2004; 37(5):1563–5. ●90 Palefsky JM, Holly EA, Ralston ML, Jay N, Berry JM, Darragh TM. High incidence of anal high-grade squamous intraepithelial lesions among HIV-positive and HIV-negative homosexual and bisexual men. AIDS 1998; 12(5):495–503. 91 Frisch M, Biggar RJ, Goedert JJ. Human papillomavirusassociated cancers in patients with human immunodeficiency virus infection and acquired immunodeficiency syndrome. J Natl Cancer Inst 2000; 92(18):1500–10. 92 Palefsky JM, Holly EA, Efirdc JT, Da Costa M, Jay N, Berry JM, Darragh TM. Anal intraepithelial neoplasia in the highly active antiretroviral therapy era among HIV-positive men who have sex with men. AIDS 2005; 19(13):1407–14. 93 Fox P, Stebbing J, Portsmouth S, et al. Lack of response of anal intra-epithelial neoplasia to highly active antiretroviral therapy. AIDS 2003; 17(2):279–80. 94 Chadha M, Rosenblatt EA, Malamud S, Pisch JAB. Squamous- cell carcinoma of the anus in HIV-positive patients. Dis Colon Rectum 1994; 37:861–5. 95 Holland JM. Tolerance of patients with human immunodeficiency virus and anal carcinoma to treatment with combined chemotherapy and radiation treatment. Radiology 1994; 193:251–4. 96 Hoffman R, Welton ML, Klencke B, Weinberg VRK. The significance of pretreatment CD4 count on the outcome and treatment tolerance of HIV-positive patients with anal cancer. Int J Radiat Oncol Biol Phys 1999; 44:127–31. 97 Cleator S, Fife K, Nelson M, Gazzard B, Phillips R, Bower M. Treatment of HIV-associated invasive anal cancer with combined chemoradiation. Eur J Cancer 2000; 36(6):754–8. 98 Clifford GM, Polesel J, Rickenbach M, et al. Cancer risk in the Swiss HIV Cohort Study: associations with immunodeficiency, smoking, and highly active antiretroviral therapy. J Natl Cancer Inst 2005; 97(6):425–32. 99 Palefsky JM, Minkoff H, Kalish LA, et al. Cervicovaginal human papillomavirus infection in human immunodeficiency virus-1 (HIV)-positive and high-risk HIV-negative women. J Natl Cancer Inst 1999; 91(3):226–36. 100 Heard I, Schmitz V, Costagliola D, Orth G, Kazatchkine MD. Early regression of cervical lesions in HIV-seropositive women receiving highly active antiretroviral therapy. AIDS 1998; 12(12):1459–64. 101 Minkoff H, Ahdieh L, Massad LS, et al. The effect of highly active antiretroviral therapy on cervical cytologic changes associated with oncogenic HPV among HIV-infected women. AIDS 2001; 15(16):2157–64.
102 Harris TG, Burk RD, Palefsky JM, et al. Incidence of cervical squamous intraepithelial lesions associated with HIV serostatus, CD4 cell counts, and human papillomavirus test results. JAMA 2005; 293(12):1471–6. 103 Maiman M, Fruchter R, Guy L, Cuthill S, Levine P, Serur E. Human immunodeficiency virus infection and invasive cervical carcinoma. Cancer 1993; 71:402–6. 104 Maiman M, Fruchter RG, Clark M, Arrastia CD, Matthews R, Gates EJ. Cervical cancer as an AIDS-defining illness. Obstet Gynecol 1997; 89(1):76–80. 105 Robinson WR, Freeman D. Improved outcome of cervical neoplasia in HIV-infected women in the era of highly active antiretroviral therapy. AIDS Patient Care STDS 2002; 16(2):61–5. ●106 Harper DM, Franco EL, Wheeler C, et al. Efficacy of a bivalent L1 virus-like particle vaccine in prevention of infection with human papillomavirus types 16 and 18 in young women: a randomised controlled trial. Lancet 2004; 364(9447):1757–65. 107 Granovsky M, Mueller B, Nicholson H, Rosenberg P, Rabkin C. Cancer in human immunodeficiency virus-infected children: a case series from the children’s cancer group and the National Cancer Institute. J Clin Oncol 1998; 16:1729–35. 108 McClain KL, Leach CT, Jenson HB, et al. Association of Epstein–Barr virus with leiomyosarcomas in children with AIDS. N Engl J Med 1995; 332(1):12–8. 109 Goedert J, Cote T. Conjunctival malignant disease with AIDS in USA. Lancet 1995; 346:257–8. 110 Wang C, Brodland D, Su W. Skin cancers associated with acquired immunodeficiency syndrome. Mayo Clin Proc 1995; 70:766–72. 111 Remick S. Non-AIDS defining cancers. Hematol Oncol Clin North Am 1996; 10:1203–13. 112 Wilson WT, Frenkel E, Vuitch F, Sagalowsky AI. Testicular tumors in men with human immunodeficiency virus. J Urol 1992; 147(4):1038–40. 113 Powles T, Bower M, Daugaard G, et al. Multicenter study of human immunodeficiency virus-related germ cell tumors. J Clin Oncol 2003; 21(10):1922–7. 114 Gunthel CJ, Northfelt DW. Cancers not associated with immunodeficiency in HIV infected persons. Oncology (Huntingt) 1994; 8(7):59–64. 115 Bower M, Powles T, Nelson M, Shah P, Cox S, Mandelia S, Gazzard B. HIV-related lung cancer in the era of highly active antiretroviral therapy. AIDS 2003; 17(3):371–5. 116 Herida M, Mary-Krause M, Kaphan R, et al. Incidence of non-AIDS-defining cancers before and during the highly active antiretroviral therapy era in a cohort of human immunodeficiency virus-infected patients. J Clin Oncol 2003; 21(18):3447–53. 117 Powles T, Thirwell C, Newsom-Davis T, et al. Does HIV adversely influence the outcome in advanced non-smallcell lung cancer in the era of HAART? Br J Cancer 2003; 89(3):457–9.
This page intentionally left blank
PART
3
MANAGEMENT
48 Medical care Clare E Green, Richard H J Begent 49 Interstitial brachytherapy J Roger Owen, David bottomley 50 Principles of external beam radiotherapy planning techniques Stephen L Morris, H Jane Dobbs 51 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy Vincent Khoo 52 Concomitant chemo-radiotherapy principles and management Syed A Hussain, Anjali Zarkar, Nicholas D James 53 Palliative care Anne Naysmith, Karol Sikora 54 Communications with the cancer patient Justin Stebbing, Maurice L Slevin 55 Clinical cancer genetics Lisa J Walker, Ros A Eeles 56 Large-scale randomized evidence: trials and overviews Richard Gray, Rory Collins, Richard Peto, Keith Wheatley 57 The economics of cancer care Nick Bosanquet 58 Medical audit Amit K Bahl, Gareth J G Rees 59 The organisation of cancer services: a UK perspective Ian Kunkler 60 New drug development George Blackledge 61 Late effects of cancer therapy Susan E Davidson 62 Surgical oncology David Merrilees, David E Neal 63 Cancer prevention strategies William P Steward, Andreas Gescher
1189 1210 1233 1254 1280 1290 1305 1318 1338 1352 1365 1377 1407 1415 1423 1433
This page intentionally left blank
48 Medical care CLARE E. GREEN AND RICHARD H.J. BEGENT
Plan of treatment Respiratory system Cardiovascular system Urogenital system Alimentary system Central nervous system
1189 1190 1194 1196 1197 1199
Management of a patient with cancer requires an holistic approach, including the characterization of the tumour, its eradication or control, and attention to the full range of personal, family, psychological and general medical issues. Study of patients with cancer shows us that there is a whole range of medical problems which are more or less peculiar to these diseases. Particular patterns of the development of the disease and of disruption of normal functions caused by malignant infiltration and the paraneoplastic phenomena are sufficiently complex to be a specialist subject in their own right. The oncologist is in a crucial position to develop expertise in the recognition and management of these syndromes with a central role in the multidisciplinary team. This chapter will review the more important syndromes and problems of management. Paraneoplastic syndromes which are an integral part of the presentations of particular malignancies are covered in the separate chapters relating to their tumour types. Others, relevant to several tumour types, are dealt with here. Improved understanding of the nature of paraneoplastic phenomena must improve our comprehension of the underlying nature of cancer. Recognition of their often subtle features in the face of symptoms and signs caused more directly by the tumour may be demanding but, if diagnosed, many can be treated with benefit to the patient. Other syndromes produced directly by tumour involvement in various systems present major problems in management and deserve special
Endocrine system Skin Haematological manifestations of malignancy Infection Conclusions References
1201 1203 1204 1205 1206 1206
consideration of the best means of treating them in the presence of the underlying disease.
PLAN OF TREATMENT A patient with cancer may expect cure, prolonged survival without cure, or only palliation. The choice of management of his general medical problems will often depend on these broader issues. Everything that follows needs to be considered in this light. The knowledge and experience of the oncologist are tested particularly by the patient whose tumour can be expected to respond, with some prolongation of survival, but in whom cure is extremely unlikely. As systemic therapy improves in effectiveness and reduced toxicity, the case for treating such patients becomes stronger. In these cases, the patient and his family need to be well informed by their oncologist to help them to decide whether to have treatment aimed at prolonging survival. However, the onus is on the oncologist to advise whether the duration and quality of life that may be expected is sufficient to justify the anticipated toxicity of therapy and inconvenience associated with hospital visits. Whatever decision is taken, time must be given to ensure that medical and nursing staff, as well as the patient and family, understand the reasons for the course of action chosen and know that the policy may be changed if circumstances alter.
1190 Medical care
RESPIRATORY SYSTEM Bronchial obstruction Management depends on the site, the pathology and the aims of the treatment. Early endoscopy or imaging and biopsy of the obstructing lesion are nearly always essential. More modern and advanced CT imaging techniques such as ‘virtual bronchoscopy’ can characterize bronchial lesions in a non-invasive manner.1 Occasionally the obstruction will be found to be caused by a non-malignant lesion. Tumours arising in the bronchial tree tend to present at an early stage and may cause obstruction, which may present as an emergency. The treatment of choice is bronchoscopy, when the obstruction may be relieved with neodymium yttrium– aluminium–garnet (Nd-YAG) laser therapy2 (Fig. 48.1), or possibly with electrocautery, cryotherapy or placement of airway stents3–6,7* (Fig. 48.2). Radiotherapy is often effective in relieving obstruction of a bronchus caused by a variety of tumour types, and may be combined with bronchoscopy to be delivered intra-bronchially as brachytherapy.4 Bronchoscopic procedures are invaluable both as emergency treatment and in palliation so that definitive treatment can be planned and given after the immediate distress has been relieved. Surgical resection of a primary pulmonary tumour may give a 10-year survival in 10–20 per cent of patients, and should always be considered if disease appears to be localized. Chemosensitive tumours, such as small cell carcinoma of the bronchus, lymphomas and testicular germ cell tumours, are best treated with cytotoxic drugs, which will often give a rapid response and also deal with disease at distant sites, as is commonly present with these tumour types (Fig. 48.3).
(a)
Parenchymal lung disease Diffuse lung infiltration in a patient with cancer may be caused by tumour, infection, drugs, radiation, various other causes or a mixture of pathologies. Patients are frequently so ill that full investigation is not possible. Where biopsy material cannot be obtained, serum tumour markers (human chorionic gonadotrophin – hCG, and alpha-fetoprotein – AFP) can be valuable in making a diagnosis of potentially curable malignancies such as choriocarcinoma or a germ cell tumour. Trends in serum carcinoembryonic antigen (CEA) and other markers of common epithelial malignancies can also indicate whether there is tumour progression. A systematic search for infection is important, but it is often necessary to start treatment for infection before a microbiological diagnosis is made. Multiple tumour deposits can predispose to areas of pulmonary infection distal to obstruction. Therefore, patients in severe respiratory failure who have a known widespread lung tumour are best given a broad-spectrum antibiotic as well as antitumour therapy, even though it may be impossible to
(b)
Figure 48.1 Obstructing squamous cell carcinoma of trachea (a) and patent lumen after tumour resection with Nd-YAG (b).
prove infection. Opportunistic infection, particularly with Pneumocystis carinii, and tuberculosis should be considered. When cytotoxic chemotherapy is begun for a drugsensitive tumour such as choriocarcinoma or a germ cell tumour, the respiratory failure may deteriorate for the first few days (Fig. 48.4). This is thought to be caused by oedema and inflammation around necrotic tumour cells. In patients
Respiratory system 1191
Clinical Case One
(a)
(b)
(d)
(c)
(e)
(f)
Figure 48.2 a–c Pre-treatment. (a) Bilateral hilar adenopathy with atelectasis of right middle lobe (arrows). (b) CT shows near occlusion of left main stem bronchus (arrow). (c) Bronchoscopy shows near occlusions of right and left main stem bronchi (arrows). d–f Post-treatment. (d and e) CT and plain films show bilateral self-expanding Wallstent in the right (dark arrows) and left (light arrows) main stem bronchi. (f) Bronchography shows widely patent right main stem bronchus with stent in place.
(a)
with dyspnoea at rest, chemotherapy should be started less intensively than usual. Arterial PO2 should be monitored from the start of treatment to detect such deterioration as early as possible. Oxygen is often helpful and, although ventilation is occasionally successful in supporting a patient until the lungs improve, in our experience most patients who are ventilated die. It can often be difficult to discriminate between tumour progression, infection and drug-induced lung fibrosis (Fig. 48.5). However, the clinician may be alerted by a careful history and knowledge of the types of allergic lesion seen with bleomycin, docetaxel, methotrexate, mitomycin C and procarbazine, and the diffuse interstitial pneumonitis with fibrosis which may result from bleomycin, cyclophosphamide, chlorambucil, busulfan, BCNU (1, 3-bis (2-chloroethyl)-1-nitrosurea), and methotrexate.8,9 Combination chemotherapy often makes it difficult to be certain of the causative drug. Lung biopsy may discriminate between drug-induced, infective and neoplastic causes of diffuse interstitial pneumonitis. It is unclear if the concomitant use of oxygen, itself toxic to the lung, may exacerbate bleomycin toxicity. There is anecdotal evidence that high-dose corticosteroids may prevent or reverse the pulmonary toxicity,10 but there have been no formal trials. Radiation-induced lung disease (Fig. 48.6) is characterized by dry cough, low-grade fever and dyspnoea occurring
(b)
Figure 48.3 (a) Chest X-ray, showing collapse of the right upper lobe in a patient with a malignant germ cell tumour of the testis. Biopsy of a tumour obstructing the left upper lobe bronchus contained metastatic germ cell tumour, and cytotoxic chemotherapy was given. (b) After chemotherapy as the only treatment, the tumour resolved completely, leading to re-expansion of the lobe.
1192 Medical care
(a)
(c)
2–3 months after radiation. A diffuse pulmonary opacity closely corresponding to the radiation field is typical. This may be followed at 9–12 months by fibrosis with loss of lung volume. The picture may easily be complicated by
(b)
Figure 48.4 (a) Chest X-ray, showing diffuse pulmonary shadowing caused by metastases of gestational choriocarcinoma. Dyspnoea and arterial PO2 deteriorated 2 days after starting a shortened course of chemotherapy. (b) Thirteen days after starting chemotherapy there was no improvement in the radiograph nor in the patient’s dyspnoea. (c) The metastases did eventually resolve. The patient is tumour-free 3 years later.
the presence of recurrent tumour or drug-induced fibrosis. Although its efficacy has not been proved, it is customary to give a course of corticosteroids for the acute reaction.
Respiratory system 1193
Figure 48.5 Computed tomographic scan of the lungs in a patient receiving chemotherapy, including bleomycin, for a malignant germ cell tumour of the testis. Multiple opacities are present in the parenchyma of the lungs. The differential diagnosis was between progressing tumour, bleomycin-induced lung fibrosis and infection. The blurred edges of the lesions are atypical for tumour metastases, and after a rise in antibody titres to cytomegalovirus was shown, the changes were attributed to infection with this virus. The changes resolved and did not recur when chemotherapy was resumed.
Figure 48.6 Chest X-ray of a patient who had a pneumonectomy for carcinoma of the left lung. A recurrence at the right hilum was treated with radiation and an area of radiation fibrosis corresponding to the field is shown. The straight edges of the opacity are typical.
Pleural effusion Rational management of a pleural effusion requires an understanding of the cause. At one level, it is important to exclude non-malignant conditions causing effusions in patients with cancer. At another level, management will depend on whether the malignant effusion is a transudate caused by venous or lymphatic obstruction in the lung or mediastinum, an exudate from malignant infiltration of the pleura or a chylous effusion from rupture of the thoracic duct by tumour. Examination of the pleural fluid by culture and cytology is important. The presence of malignant cells in the pleural fluid is generally a clear indication of direct pleural involvement. However, the pleura may be infiltrated without cells being detectable. Here, evidence of an exudate is in favour of pleural infiltration. The level of tumour markers in the pleural fluid compared to serum may also be helpful.11 Although different criteria for an exudate have been used, those of Light et al. 197212 have been widely cited. They are: a ratio of pleural fluid to serum protein 0.5, a ratio of pleural fluid to serum lactate dehydrogenase (LDH) 0.6, and pleural fluid LDH 200mg/dL. The presence of opaque white pleural fluid strongly suggests a chylothorax. However, a similar appearance can occur with desquamating pleural
and tumour cells. It is prudent, therefore, to confirm the chylous nature of the fluid by staining for fat with Sudan III. When a pleural effusion is a manifestation of a disseminated malignancy for which there is an effective systemic therapy and it is not causing dyspnoea, systemic treatment alone is often the best management. Monitoring of the effusion may contribute to assessment of effectiveness of treatment (Fig. 48.7). When the therapy includes methotrexate, toxicity may be increased by the presence of the effusion. Methotrexate diffuses into the pleural fluid when serum concentrations are high but is then cleared more slowly than the drug in the circulation. The result is a depot effect, methotrexate being released into the circulation over several days. If possible, methotrexate therapy should be avoided, particularly in high doses, and even prolonged folinic acid rescue may not be protective. In the absence of effective systemic therapy, palliation can often be given by local means. Aspiration of the effusion close to dryness at an early stage usually gives immediate relief of symptoms while the results of investigations are obtained, the rate of re-accumulation observed and definite treatment planned. When there is also solid tumour in the thorax, aspiration under ultrasound control can assist in optimal location of fluid for removal. In a
1194 Medical care
minority, the rate of re-accumulation is very slow and occasional aspirations give adequate palliation. Generally, however, measures are required to prevent or slow down fluid formation. The standard therapy is drainage of the pleural cavity to dryness with an intercostal drain followed by instillation of either bleomycin (60 mg) or tetracycline (1000 mg), although historically other cytotoxic drugs and sclerosants have been used. The response rate varies from 30 to 70 per cent13,14**,15** with some studies showing a clear superiority of bleomycin over tetracycline and others showing no significant differences. Bleomycin tends to cause more fevers, whereas tetracycline is associated with more chest pain requiring morphine as analgesia. There is evidence to suggest that insufflated talc under thoracoscopic guidance may give the best results.14** The addition of intra-pleural fibrinolytic agents (streptokinase/urokinase) to intercostal tube drainage to try to break down fibrin bands causing fluid loculation may increase the overall amount of fluid drained and reduce time to re-accumulation16*** but numbers in the trials are too small to make this a standard recommendation. Patients who fail standard chemical pleurodesis may be referred to the cardiothoracic service for either talc pleurodesis or pleurectomy. Transudates caused by pulmonary or mediastinal tumour invasion of veins or lymphatics are best treated by radiotherapy or systemic chemotherapy. Chylothorax will sometimes resolve after mediastinal radiotherapy or successful chemotherapy.
Hypertrophic pulmonary osteoarthropathy This paraneoplastic syndrome characterized by clubbing, periostitis of long bones and sometimes a polyarthropathy,
Figure 48.7 Chest X-ray showing a large right pleural effusion.
occurs not uncommonly with non-small cell carcinoma of the lung and various other primary and secondary neoplasms in the thorax. Polyarthropathy may be the presenting feature without clubbing and can give a positive bone scan. Hypertrophic pulmonary osteoarthropathy (HPOA) can be associated with severe pain, and treatment of the underlying malignancy will often control the pain, but where this fails or is not possible, radiotherapy can be given to the affected bony site.17
CARDIOVASCULAR SYSTEM Superior vena caval obstruction (SVCO) The syndrome of non-pulsatile distension of cervical and thoracic veins with facial, cervical and upper thoracic oedema and dyspnoea is well known. Sometimes there is conjunctival oedema and oral mucosal engorgement in addition.18 Central nervous system features such as headache, papilloedema and altered consciousness are attributed to raised intracranial pressure resulting from venous distension. A histological diagnosis is important since benign goitres, aortic aneurysms, thrombotic syndromes and idiopathic sclerosing mediastinitis may cause superior vena caval obstruction (SVCO) in about 3 per cent of cases.19 The first case described by William Hunter in 1757 was caused by a syphilitic aortic aneurysm, but infective causes were much more common before antibiotics were available. Bronchial carcinoma is the most common causative malignancy (80 per cent) with small cell carcinoma more frequent than squamous cell carcinoma, followed by lymphoma (17 per cent), with non-Hodgkin’s lymphoma as the most frequent type.18 Treatment should be started urgently and it may be justifiable to do this without a histological diagnosis if the patient is very unwell. Under these circumstances superior vena caval stenting is the treatment of choice as it gives prompt relief and preserves tumour tissue for biopsy before definitive treatment.20*,21* If this is not available or possible then radiotherapy should be given and a biopsy obtained as soon as possible. Radiotherapy produces symptomatic relief in up to 85 per cent of patients22* and is the treatment of choice for squamous cell carcinoma of the bronchus. However, small cell carcinoma and non-Hodgkin’s lymphoma make up half of all the malignant causes of SVCO, and for these tumour types chemotherapy should be strongly considered as the primary therapy because of the chemosensitivity of the tumour types and because these are essentially systemic diseases. Comparison between the effectiveness of radiotherapy, chemotherapy and stenting has been investigated.23*** In non-small cell lung cancer, 60 per cent of patients had relief of SVCO following chemotherapy and/or radiotherapy; 19 per cent of those treated had a recurrence of SVCO. Insertion of an SVC stent relieved SVCO in 95 per cent; 11 per cent of those treated had further SVCO. So,
Cardiovascular system 1195
chemotherapy and radiotherapy are effective in relieving SVCO in a proportion of patients whilst stent insertion may provide relief in a higher proportion and more rapidly. If stenting is not performed up front and the tumour fails to respond to treatment, or if treatment cannot be given, stenting may give good palliation at a later date.24 When SVCO fails to respond to treatment, venous thrombosis rather than persistent tumour should also be considered.
Pericardial tumour The frequent involvement of the pericardium by metastatic tumour, familiar in the post-mortem room, is now a common incidental finding with CT or ultrasound examinations. It is unusual for symptoms to occur, but cardiac tamponade produced either by effusion or constrictive pericardial tumour may have an insidious onset and easily be missed in a patient with disseminated malignancy. Emergency treatment of tamponade by aspiration of pericardial fluid is followed by chemotherapy or radiotherapy if the tumour is likely to respond. Intrapericardial mustine, thiotepa, 5-fluorouracil, tetracycline and radioactive colloidal gold or phosphorus have been used historically. In chronic recurring effusions, fenestration of the pericardium into the pleural cavity is worth considering.25*
(a)
This is a relatively simple procedure with little morbidity and a high probability of preventing re-accumulation of the effusion. An example is shown in Figure 48.8.
Cardiac toxicity This may occur with anthracycline therapy, biological agents such as trastuzumab (Herceptin), cardiac radiotherapy, cyclophosphamide and iron overload from repeated transfusion. The most important and common cardiotoxicity follows anthracycline chemotherapy.26 Acute toxicity after doxorubicin most commonly takes the form of arrhythmias or conduction defects. An acute myocarditis–pericarditis syndrome has also been described. Chronic toxicity dependent upon the cumulative dose was clinically evident in 1–10 per cent of patients who received a cumulative dose of 550 mg/m2.27 Cardiomyopathy followed by congestive cardiac failure is the principal manifestation, and once it has developed the prospects for improvement are poor. Therefore prevention of cardiac toxicity is vital. Cardiac dysfunction can be measured at an earlier stage by echocardiography or radionuclide imaging, where the first sign is a decrease in left ventricular ejection fraction, particularly with exercise stress testing. Cardiotoxicity may be reduced by using liposomal doxorubicin, or doxorubicin analogues
(b)
Figure 48.8 (a) Chest X-ray, showing a large pericardial effusion which caused cardiac tamponade in a patient with pericardial metastases of hypernephroma. Fenestration was performed and 7 months later (b) the effusion had not re-accumulated and there was no recurrence of the tamponade.
1196 Medical care
such as epirubicin and mitoxantrone, although these have varying degrees of cardiotoxicity.28 Mitoxantrone is one of the least cardiac toxic analogues but its anti-tumour effect may not always be comparable. There has been interest in the cytoprotectant dexrazoxane which, when co-administered with doxorubicin, appears to reduce the incidence of cardiac toxicity, but dexrazoxane may potentiate doxorubicin-induced myelosuppression and it is not clear if it interferes with the activity of chemotherapy or will protect against late cardiac toxicity.29 Trastuzumab causes a decrease in left ventricular ejection fraction in a minority of patients, approximately 2–3 per cent.30**,31* Incidence is increased if given in conjunction with paclitaxel or anthracyclines. It differs from anthracycline-induced cardiotoxicity in that it is not cumulative or dose-dependent and often improves after withdrawal of treatment. Re-treatment with trastuzumab is often possible. Taxol can cause bradycardia, heart block or other arrhythmias during or shortly after infusion.28 5-Fluorouracil may cause precordial pain, S-T and T wave changes on ECG, atrial or ventricular arrhythmias or sudden death. The incidence of significant effects may reach 5 per cent in patients with previous heart disease.
UROGENITAL SYSTEM Obstructive uropathy Renal failure caused by ureteric obstruction in gynaecological, urinary and gastrointestinal malignancy and lymphoma has, in the past, been welcomed as a relatively gentle way for the patient to die. Modern chemotherapy, and radiotherapy particularly for carcinoma of the cervix and urinary bladder, lymphoma and germ cell tumours has changed this so that many patients may be offered several months of good-quality life or even cure. Good renal function is essential for effective chemotherapy and optimal management of the ureteric obstruction becomes crucial for success. Ultrasound is the best method for demonstrating the hydronephrosis. When obstruction is not complete, an isotope renogram is useful to show the extent of the renal obstruction and the degree of impairment to each kidney. After correcting hyperkalaemia, ureteric stenting or percutaneous nephrostomy is the most satisfactory emergency measure to relieve obstruction.32,33,34* Chemotherapy or radiotherapy may then be given in safety provided that the glomerular filtration rate recovers. If there is a good tumour response but persisting obstruction due to fibrosis, a permanent urinary diversion may be considered at a later stage. This is sometimes valuable for long-standing urinary fistulas as well. When the obstruction is incomplete and the response rate to drugs or radiation is high, obstruction may be rapidly relieved by these means. Drugs metabolized by the kidneys must be used with great care or, particularly in the case of methotrexate, avoided altogether. Etoposide and low-dose
cisplatin have proved useful for treatment of germ cell tumours in this context.35*** The full range and dose of drugs may then be given as soon as renal function has improved.
Renal parenchymal disease Renal impairment is infrequently caused by direct tumour involvement, but lymphoma and chronic lymphocytic leukaemia can occasionally have this effect (Fig. 48.9). The renal failure of myeloma is complex; direct damage by immunoglobulin light chains, amyloid and hypercalcaemia may contribute. Amyloid has been reported in hypernephroma and Hodgkin’s disease. Nephrotic syndrome is a rare complaint of a wide range of malignancies. Hodgkin’s disease is the most common and is usually associated with minimal change or lipoid nephropathy.36 Membranous glomerulonephritis is the more common type of lesion in carcinomas and non-Hodgkin’s lymphoma. Deposition of immune complexes containing tumour-associated antigens has been shown in some instances of carcinomas. Resolution of nephrotic syndrome has been reported when the causative tumour is successfully treated. Disseminated intravascular coagulation in malignancy is associated with renal impairment in 25 per cent of cases.37* Like the nephrotic syndrome, it may resolve when a causative malignancy is successfully treated. Urate nephropathy results from release of purines by necrosing tumour cells and consequent precipitation of urate in the renal tubules. It occurs in myeloproliferative disorders but most strikingly after treatment of tumours, such as leukaemias and lymphomas, which
30
25
20
15
Urea (mmol/L)
10
Lymphocytes ( 109/L)
5
Chlorambucil, vincristine and prednisolone 0 1
3
5
7
9
11
14
16
Weeks
Figure 48.9 Renal failure caused by biopsy-confirmed infiltration of the kidney by chronic lymphocytic leukaemia. Treatment with chlorambucil, vincristine and prednisolone led to a fall in the lymphocyte count, with a corresponding fall of plasma urea.
Alimentary system 1197
respond rapidly to chemotherapy. It is prevented by good hydration during chemotherapy, administration of allopurinol and ensuring an alkaline urine.38 Tumour lysis syndrome occurs when rapid tumour destruction leads to hyperkalaemia, hyperphosphataemia with hypocalcaemia, lactic acidosis and urate nephropathy, and may be fatal. It occurs after the start of treatment of tumours such as Burkitt’s lymphoma and lymphoblastic leukaemia in which response may be dramatic. With care to ensure good hydration, alkaline urine and allopurinol prophylaxis, it should be a manageable problem.39 However, where there is pre-existing renal impairment, haemodialysis may be necessary. Hepato–renal syndrome is a progressive oliguric renal failure due to renal hypoperfusion secondary to liver failure. In the oncological setting it may be caused by a co-existing medical condition, commonly advanced cirrhosis, or as a complication of treatment, such as chemo-embolization of liver lesions. The prognosis is grave, as the only effective treatment is liver transplantation, nearly always contraindicated for oncology patients. Treatments such as transjugular intra-hepatic porto–systemic shunts and renal vasodilators may temporarily improve renal function and should be considered.40 The nephrotoxicity of radiation, cytotoxic drugs such as cisplatin, methotrexate and nitrosoureas, and antibiotics such as the aminoglycosides, is dealt with elsewhere.
ALIMENTARY SYSTEM
Those who can be helped by simple dietary advice, mouth care or a liquid diet will also have been identified. A clear idea of the aims of treatment is essential before deciding on nutritional support, and, where possible, nutrition should always be supplied in the least invasive and most physiological way. If therapy is palliative, total parenteral nutrition or forced enteral nutrition is usually inappropriate and the patient and his family can be reassured that it is not necessary to eat if to do so is distressing. However, measures to relieve the discomfort of oral ulceration or infection and relief of oesophageal obstruction may be of great benefit in palliation. If the patient is undergoing active treatment, then nutritional support appears logical, but in fact has been difficult to prove in clinical trials.41 Nevertheless, if patients are temporarily unable to swallow, it is justifiable to insert a nasogastric tube for feeding. Patients who cannot tolerate a nasogastric tube, or in whom one cannot be passed, can be satisfactorily fed through a percutaneous gastrostomy or gastrojejunostomy tube.42 Total parenteral nutrition can also be justified in selected patients with intestinal obstruction from ovarian or other carcinomas in which there is a reasonable chance of a remission of some months if the patient can be supported for the first few weeks. The decision to start such support should be made early if it is going to be undertaken, and it is not appropriate for all. Nevertheless, for some patients the chance of a few extra months of life out of hospital will be of great worth.
Causes of catabolic states
Nutrition Weight loss is so clearly a stigma of cancer that the whole topic of nutrition easily arouses confusion and strong feelings in the patient and carers. The weight loss may be due to gastrointestinal tract involvement with subsequent difficulties in eating, or it may be due to the anorexia–cachexia syndrome, or both. The anorexia–cachexia syndrome is a major clinical problem and is not the same as simple starvation, as discussed further below. Therefore it is not surprising that ensuring an adequate diet alone rarely reverses the syndrome, although successful treatment of the tumour commonly does. The patient’s nutritional status must be assessed before treatment to decide whether there are sufficient reserves to prevent the patient becoming severely, or even fatally, malnourished during surgery, radiotherapy or chemotherapy. The criteria for making such an assessment are not well defined. Weight is influenced by hydration, and plasma albumin by hepatic function and renal loss. Other parameters, such as creatinine/height ratio, are similarly unreliable. This is not an excuse for not considering malnutrition when they are abnormal. A period of observation by nurses and dieticians in hospital will identify patients ingesting less than 1000–1500 calories daily. Such patients will need nutritional support if intensive therapy is to be given over a prolonged period.
The anorexia–cachexia syndrome has a complex pathophysiology which is only now being appreciated.41 There is suppression of hunger in the face of weight loss, and one of the key features is equal mobilization of fat and muscle, whereas in simple starvation there is preferential mobilization of fat. Cytokines are now thought to play a major role in the syndrome, but extrapolating from animal models to patients has not proved straightforward. Tumour necrosis factor alpha (cachectin, TNF-α) induces cachexia in animal tumour models in which anti-TNF antibodies reduce cachexia, but there is no clear association between serum TNF levels and cachexia in humans. Interleukin-6 correlates best with the development of cachexia in patients, but by itself IL-6 does not induce cachexia in animal models. Insulin resistance and decreased insulin:glucagon ratio are commonly documented, but the cause of these changes is not clear, and attempts to correct them by insulin therapy have not been successful. The most common agents in clinical use for anorexia–cachexia syndrome are corticosteroids and progestogens. Clinical studies suggest that they are of equal efficacy but have different toxicity profiles, with progestogens having more thromboembolic and oedema problems, and corticosteroids having classical corticosteroid toxicities.43,44** Doses commonly used in studies have been dexamethasone (4 mg once daily), megestrol acetate (800 mg once daily) and
1198 Medical care
medroxyprogesterone acetate (500 mg twice daily). There is clearly a need for new agents, and these will come from our increasing understanding of this distressing condition.
Malabsorption Tumour infiltration (particularly by lymphoma), surgery and radiotherapy may contribute to malabsorption. Some patients with a variety of malignancies appear to have a paraneoplastic type of malabsorption characterized by a flat mucosa or, less often, subtotal villous atrophy. The syndrome usually resolves with successful treatment of the tumour. Patients undergoing pancreaticoduodenectomy for carcinoma of the pancreas can become malnourished due to pancreatic insufficiency. Fat-soluble vitamin deficiency is common, especially vitamin A,45 and easily treated with supplements. Pancreatic enzyme supplements (Creon) are also essential.
Diets with putative anti-tumour effect The many claims for therapeutic effect on tumours of particular diets appear to lack satisfactory scientific support, although with the renewed interest in alternative medicine we may expect more properly conducted clinical trials.46,47 Where the diet is innocuous and the patient has faith in it, there seems little to gain by discouraging its use. However, there are a number of “alternative medicine” diets which clearly do not maintain an adequate nutritional intake and in these situations it may be appropriate to encourage resumption of a more nutritional diet, particularly if the patient is undergoing active treatment.
Obstruction of the gastrointestinal tract Management is generally by well-established surgical principles. It is salutary occasionally to see patients with intestinal obstruction who are being given anti-emetics for their vomiting in the mistaken belief that their symptoms are caused by cytotoxic drugs or abdominal radiotherapy. Also it is easy to overlook common non-malignant causes of intestinal obstruction, such as strangulated hernia, in patients with cancer. Obstruction by malignant disease is, of course, best treated by definitive therapy of the tumour (Fig. 48.10). When this is not possible, surgical bypass of biliary, duodenal, small intestinal and large bowel obstruction can often be achieved at the initial operation. However, where surgery is not possible, good palliation may still be achieved either by using the Nd-YAG laser through an endoscope or by the placement of stents, either endoscopically or radiologically (Fig. 48.11).48 Laser endoscopy has the advantage that it may be repeated as often as required, although there is a risk of bowel perforation.49 Stenting with self-expandable metal stents (SEMS) is particularly suitable for shortterm palliation due to limited life expectancy or expected
Figure 48.10 Plain abdominal X-ray showing small bowel obstruction caused by a mass (M) in the right upper quadrant.
response to active treatment because in the longer term the tumour tends to overgrow and obstruct the stent.50 If the obstruction is not bypassable by surgery or stenting, and response to chemotherapy unlikely, then conservative management to palliate symptoms is appropriate. This situation is particularly common with advanced ovarian and gastrointestinal malignancies and is highly distressing to the patient. The initial management involves nasogastric tube suction and intravenous fluids. A trial of dexamethasone (16 mg daily intravenously for up to a week) may help to relieve the obstruction, presumably by reducing inflammation and possibly a direct anti-tumour effect, but should be discontinued if the obstruction persists.51*** Octreotide is a long-acting analogue of somatostatin, and, amongst its other effects, it reduces intestinal secretions and promotes absorption. There is good evidence that it may improve nausea and vomiting associated with intestinal obstruction, probably by these mechanisms,52 but it is less clear if octreotide actually helps to relieve the obstruction itself. Patients in whom these pharmacological approaches fail are often forced to retain their nasogastric tubes, although percutaneous endoscopic gastrostomy has been used safely for decompression in unresolving intestinal obstruction and allows the removal of the nasogastric tube.53
Central nervous system 1199
Ascites
Figure 48.11 This patient had a pancreatic cancer causing both biliary and gastric outlet obstruction. Stent A is a mesh metal stent in the biliary tree. Stent B is a mesh metal stent through the gastric outlet and extending into the duodenum; a guide-wire is still in place in the duodenum.
Jaundice The various causes of jaundice in patients with cancer are well covered in standard medical texts. Awareness of hepatic dysfunction is important because of enhanced toxicity of drugs metabolized in the liver. Obstructive jaundice in patients with cancer is worthy of special mention because it is common with primary and metastatic tumours of the liver and bile ducts. Pruritus can be distressing and obstruction probably accelerates the progress of renal failure. Relief of obstruction is often worthwhile, therefore, even if little anti-tumour therapy is possible. Biliary diversion when the diagnostic laparotomy shows an unresectable tumour causing obstruction will usually relieve symptoms and give a time of improved quality of life. When the diagnosis is known, laparotomy is rarely justified to relieve obstructive jaundice. However, biliary stents can be placed either endoscopically or percutaneously, and can be made of either plastic or metal (Fig. 48.11).48 Endoscopic insertion is preferred, where possible, as there is a higher success rate with fewer complications.54** Plastic stents are cheaper but may require changing every 3–6 months due to blockage. Metal stents are more costly but last longer because of their wider bore and lower tendency to migrate.
Non-malignant causes for ascites should always be considered in patients with cancer. The presence of malignant cells on cytology should always be sought. A protein content over 2.5 g/100 mL or a raised CEA value11 also supports a diagnosis of malignancy, although the former may, of course, be a feature of inflammatory ascites, tuberculous ascites being particularly difficult to discriminate from malignancy on clinical grounds. A diagnosis of chylous ascites, with a typical milky appearance of the fluid, should be confirmed by staining fluid with Sudan III to show the presence of lipid. Repeat aspiration leads to protein and fluid loss. Treatment of the tumour with systemic therapy is the most satisfactory solution and is often applicable, for instance in breast and ovarian cancer. Intra-peritoneal cytotoxic chemotherapy and radionuclide therapy have a long history and there is an attraction as these deliver a high dose of therapy locally. However, this route of administration is not in common usage outside clinical trials because of difficulties with the administration and distribution of the agent within the peritoneal cavity. If systemic therapy is not possible, the ascites may be tapped and spironolactone started.55 If there is re-accumulation, subsequent management depends upon the rate of re-accumulation and the prognosis of the patient. If there is a slow rate of re-accumulation, ascitic taps as required may provide good palliation. If there is rapid re-accumulation with a relatively long prognosis, a LeVeen shunt should be considered. The LeVeen shunt drains fluid from the peritoneal cavity into the superior vena cava via a subcutaneously tunnelled catheter. A pressure-sensitive one-way valve permits flow when the pressure intra-peritoneally exceeds that in the superior vena cava by 2–3 cm water. It is simply inserted and appears satisfactory for chylous ascites and for malignant ascites that relapses chronically. When the ascites are blood-stained or particularly rich in malignant cells or protein then clotting is apt to occur, making the shunt unusable. Disseminated intravascular coagulation has been reported with LeVeen shunts but this appears to be unusual in the absence of pre-existing hepatic dysfunction. For carefully selected patients this technique can give excellent palliation.56
CENTRAL NERVOUS SYSTEM Spinal cord compression Epidural metastases, usually with adjacent bony involvement, are the most common cause of spinal cord compression, which may be the presenting feature of malignancy. Radiation myelitis should be remembered in the differential diagnosis. Most patients have previously known bone metastases and there is an association between cerebral metastasis and spinal cord compression in small cell lung carcinoma.57* In a representative recent series the most common presenting
1200 Medical care
symptoms were motor weakness (96 per cent), pain (94 per cent), sensory disturbance (79 per cent) and sphincter disturbance (61 per cent). Ninety-one per cent of patients had had symptoms for at least 1 week.58* Early diagnosis is critical because patients who are ambulatory at the time when treatment is started have a higher chance of remaining ambulant. Patients with known bony metastases should be warned to report pain, weakness or sensory disturbance urgently. Magnetic resonance imaging (MRI) is the imaging modality of choice, and the initial treatment is with steroids, typically dexamethasone 16 mg once daily, which are thought to reduce oedema at the site of compression. However, the benefits of steroids at these doses have not been proven in randomized trials. The only positive data are for much higher doses of dexamethasone (96 mg once daily) but the benefit was countered by significant side effects.59 It is critical to involve both the radiotherapist and neurosurgeon at an early stage, as one or both specialities may be required. Generally, if there is spinal instability, compression of the cord by bone, previous radiotherapy or the histology is unknown, then surgery is indicated. Tumours arising posteriorly may be suitable for laminectomy, but 85 per cent of epidural tumours arise anteriorly in the vertebral body, and the operation of choice is vertebral body resection and stabilization, which gives better results than the more traditional laminectomy which was supposed to decompress the spinal cord but often led to further destabilization.60,61 The response to radiotherapy is well established. Early clinical trials showed that radiotherapy alone gave comparable results to laminectomy plus radiotherapy,62 although later studies suggest that vertebral body resection may give better results in selected patients.60 A recent paper63 compared treatment of cord spinal compression with decompressive surgery followed by radiotherapy with radiotherapy alone. More patients in the surgery group were able to walk after treatment and retained this ability for longer than those in the radiotherapy alone arm. Also the need for corticosteroids and opioid analgesics was significantly reduced in the surgical group, so it seems that in selected patients decompressive surgery and radiotherapy is better than radiotherapy alone. The optimal dose and schedule for radiotherapy remains unclear, although cord tolerance is the limiting factor. The indications for radiotherapy include known radiosensitive tumour with no spinal instability, and following surgery. Chemotherapy has a place in the management of spinal cord compression of drug-sensitive tumours. Friedman et al.64 have reviewed reports of 51 such patients. Improvement in spinal cord function occurred after chemotherapy with or without radiotherapy in nearly all patients with lymphomas, Hodgkin’s disease, myeloma, neuroblastoma and testicular germ cell tumours. Choriocarcinoma is also successfully treated by chemotherapy. Moderate success was reported with carcinoma of the breast and prostate. The advantage of chemotherapy is that it is treating the disease systemically, and indeed a disadvantage of radiotherapy is
that by ablating part of the bone marrow it can make subsequent chemotherapy more difficult to give by making the patient more sensitive to myelosuppression.
Raised intracranial pressure When a cerebral tumour meant almost certain death this topic was of little relevance, except for the palliation which could be achieved with dexamethasone. However, advances in neurosurgery, chemotherapy and radiotherapy have changed this, and in selected patients more aggressive treatment is indicated. Death is usually caused by raised intracranial pressure which results in herniation of the cerebral hemispheres and presents as headache and vomiting. Palsies of the third and sixth cranial nerves are particularly suggestive of imminent cerebral herniation. Dexamethasone (16–32 mg daily) should be commenced immediately and an MRI or CT scan performed. The subsequent optimal management depends upon the mode of presentation and the prognosis for the tumour type.65 If the presentation is de novo, surgery is indicated or, if not possible, a biopsy should be attempted because the histology is crucial to subsequent treatment. In certain primary brain tumours, such as medulloblastoma, surgery may be curative. If the presentation is of cerebral metastases from a known primary tumour, then the appropriate treatment depends upon the tumour type and condition of the patient. In potentially curable tumours such as choriocarcinoma, germ cell tumours and lymphomas, one aim of surgery is to protect against raised intracranial pressure by ventriculo–peritoneal shunting, craniotomy or resection of the metastasis to allow further treatment with chemotherapy and radiotherapy. Chemotherapy may actually exacerbate raised intracranial pressure by causing necrosis and bleeding into cerebral metastases, and there should be a low threshold for neurosurgical intervention. For example, choriocarcinoma with cerebral metastases has a mortality rate of 25 per cent in the first 2 weeks, with most deaths due to raised intracranial pressure and herniation, and survivors have a high chance of cure.66* In patients with less drug-sensitive tumour types, the appropriate treatment depends upon the condition of the patient. If there is progressive systemic disease, or where surgery is not possible, such as in the presence of multiple metastases, then whole-brain radiotherapy (WBRT) is indicated. If the cerebral metastases are the only manifestation of disease or the systemic disease is stable, then consideration should be given to surgery followed by WBRT, and in selected patients this gives superior results in terms of quality of life and survival. There is some evidence that stereotactic radiosurgery may be equivalent to surgery plus WBRT and superior to WBRT alone, and in centres with access to stereotactic radiosurgery this should be considered.67,68 There is also accumulating evidence that chemotherapy combined with radiotherapy may be superior to radiotherapy alone, and clearly more clinical trials are needed to evaluate this.69
Endocrine system 1201
Table 48.1 Paraneoplastic syndromes of the nervous system (from Harrison’s Principles of Internal Medicine) Autoantibody Anti-Hu Anti-Ma Anti-Ta Anti-Yo Anti-Ri Anti-CAR Anti-VGCC
Clinical presentation
Commonly associated cancer(s)
Encephalomyelitis, sensory neuronopathy Cerebellar dysfunction, brainstem dysfunction Limbic encephalitis, brainstem dysfunction Cerebellar degeneration Cerebellar ataxia, opsoclonus Photoreceptor degeneration Lambert–Eaton myasthenic syndrome
Small cell lung cancer (SCLC) and neuroblastoma Breast, lung, colon Testicular Gynaecological, breast Breast, gynaecological, SCLC SCLC and others SCLC
Paraneoplastic syndromes of the nervous system These are listed in Table 48.1 and a full discussion is beyond the scope of this text. The reader is referred to the review by Bunn and Ridgway.70 Advances in tumour immunology have recently shown that, in most cases, disordered humoral and cellular immunity as well as the expression of neural antigens leading to autoimmune attack have been demonstrated as causes of neurological injury.71,72 Paraneoplastic syndromes often precede other manifestations of tumours, and therefore one practical application is that diagnostic anti-neuronal antibody tests may define malignancy, and even the type of tumour, early on as the cause of the neurological disorder. Treatment is of the underlying malignancy and/or immunosuppression.
ENDOCRINE SYSTEM Eutopic hormone production Tumours of many endocrine glands are characterized by secretion of their eutopic hormones. The associated syndromes often produce the presenting features of such tumours. Their investigations and therapy are dealt with elsewhere under the separate tumour types. Very high levels of human chorionic gonadotrophin (hCG) are worthy of special mention for the hyperthyroidism with which they are associated. This is believed to be caused by a cross-reaction between hCG and thyroid-stimulating hormone.73
Molecular mechanisms responsible for the syndrome are poorly understood. Mutational events not only may initiate neoplastic transformation but may also lead to the activation (re-expression) of genes responsible for hormone production. Additionally, epigenetic events such as methylation may also be responsible for the development of these syndromes.75 The ectopic ACTH syndrome is characterized by oedema, muscle weakness, hypertension, hypokalaemia, alkalosis and hyperglycaemia. Moon facies, striae and buffalo hump may be seen in patients with rapidly progressing tumours. The syndrome may be the presenting feature of a malignancy. The measurement of 24-hour urinary free cortisol is the most sensitive and specific test to confirm Cushing’s syndrome. Serum ACTH levels usually exceed 200 pg/mL. A high-dose dexamethasone suppression test will distinguish between pituitary and ectopic ACTH secretion. Treatment of the tumour is the most effective means of control. Where this is not possible, medical means of suppression of the adrenal cortex should be tried first. The steroid biosynthesis inhibitors aminoglutethimide, metyrapone and ketoconazole have been used effectively,76 or the adrenal cortex may be ablated with mitotane (o,p’-DDD). If these fail, then surgery should be considered. Corticosteroid replacement is required if the adrenal cortex is to be successfully suppressed or ablated. Ectopic secretion of hCG is worthy of mention here, not because it commonly causes troublesome clinical features, but because measurements of serum hCG concentration can be used as a tumour marker.77 However, serum hCG levels above 100 000 IU/L may be associated with thyrotoxicosis. Ectopic hormone secretion in relation to hypercalcaemia and hyponatraemia is dealt with below.
Ectopic hormone production: ectopic ACTH syndrome
Hypercalcaemia
Some 5 per cent of patients with small cell lung cancer have clinical features of Cushing’s syndrome.74 Carcinoid, medullary thyroid carcinoma, neuroblastoma and phaeochromocytomas are also associated with the syndrome, which is also seen occasionally with a variety of other tumour types. Production of ‘big ACTH’ by the tumour produces the effect through adrenocortical stimulation. Other derivatives of this hormone may function as melanocyte-stimulating hormone or opiate-like hormones.
Hypercalcaemia occurs in about 10 per cent of patients with cancer at some stage in their disease, and is more common with breast cancer, myeloma and squamous cell carcinomas. Although bony metastases are present in the majority of cases of hypercalcaemia, this is not invariably the case, and there is no correlation between hypercalcaemia and the extent of skeletal involvement. Hypercalcaemia is caused by humoral agents released by the tumour, such as parathyroid hormone-related peptide that binds to the
1202 Medical care
parathyroid hormone receptor but is not detected by standard parathyroid hormone assays. In assessing hypercalcaemia, correction of the serum value according to serum protein levels is important because the effects of calcium are dependent on the ionized fraction and can be calculated from the formula: Corrected calcium (mmol/L) measured calcium (mmol/L) {[40 albumin (g/L)] 0.02} Malaise, nausea, vomiting, polydipsia, polyuria, hyporeflexia, weight loss and constipation are common presenting features. Hypovolaemia from the loss of sodium and water associated with excess calcium excretion is followed by renal failure. Excretion of calcium is decreased at this stage and the progression of hypercalcaemia accelerates. Confusion, psychosis, seizures and death may ensue. Therapy depends on the severity of hypercalcaemia. If mild, without dehydration and detected at the outset of a treatable malignancy, therapy of the tumour with attention to good hydration will suffice. With dehydration, renal impairment or markedly raised serum calcium levels, a more intensive approach is required. Correction of hypovolaemia is the first step, with 3–4 L/day of 0.9 per cent sodium chloride with potassium supplementation as required. Furosemide should only be used to prevent or treat heart failure. Fluid therapy alone will improve patients, but will not control the hypercalcaemia in the long term. Effective anti-tumour therapy is required, but in many patients this will not be available or will not act sufficiently quickly. The bisphosphonates inhibit osteoclast-mediated bone resorption and are highly effective in the treatment of hypercalcaemia.78 Two are currently commonly used in the treatment of malignant hypercalcaemia – disodium pamidronate and sodium clodronate – and both should be used only when the dehydration and renal impairment have been corrected. Disodium pamidronate is given as a single infusion of 30–90 mg, dependent upon serum calcium concentration, at a rate of 1 mg/min. Sodium clodronate is given as a single infusion of 1.5 g over 4 hours. The serum calcium concentration falls and reaches a nadir between 3 and 5 days. Side effects are few, although fever and bone pain have been reported with pamidronate, and mild gastrointestinal upsets with clodronate. Hypercalcaemia tends to recur within a few weeks unless the underlying tumour has been treated, and if this is not possible, then infusions of pamidronate or clodronate may be given every 3–4 weeks, or oral clodronate 1.6–3.2 g/day can be started when the acute episode has been treated. Bisphosphonates are also valuable in relieving pain from bony metastases and may delay the development of bony metastases in breast, lung and prostatic carcinoma and multiple myeloma.78,79 Complications such as fractures and cord compression can be reduced by the regular use of bisphosphonates.80 Zoledronic acid is a newer, more potent bisphosphonate with the added advantage of a shorter infusion time of 15 minutes. Oral agents such as ibandronate
are now available and appear to be as efficacious but perhaps more cost/time efficient than intravenous bisphosphonates.81 Further trials are investigating the adjuvant use of bisphosphonates in patients with primary breast cancer to prevent the development of bone metastases and cancer treatment-induced bone loss (AZURE trial). Other drugs have been used to control hypercalcaemia, such as mithramycin and calcitonin,82 but these have greater side effects and poor efficacy and are rarely required now. Corticosteroids are only partially effective and work better with haematological malignancies.83,84 Gallium nitrate has few side effects (except renal toxicity in 10 per cent) but is effective in the treatment of hypercalcaemia of malignancy. The data are limited and further trials are necessary.85
Hyponatraemia Hyponatraemia is a common problem in patients with cancer. Its causes are diverse and a systematic approach to investigation is crucial. Figure 48.12 gives a scheme of points to consider for this purpose. The syndrome of inappropriate anti-diuretic hormone (SIADH) secretion deserves special mention because the hormone responsible may be secreted by tumours, particularly small cell carcinoma of the bronchus and head and neck cancers.86 Caution must be exercised in making an assumption of this cause because there are a large number of non-malignant causes, e.g. infection and drugs, including cytotoxics such as vincristine and cyclophosphamide.87 The selective retention of water under the influence of ADH results in hyponatraemia and low plasma osmolarity with inappropriate concentration and high osmolarity of urine. The retained water is principally intracellular so that oedema or raised venous pressure is unusual. The intracellular Is there volume depletion?
Yes
No
Is there inappropriate concentration of urine? Yes No
Consider: diuretics vomiting diarrhoea haemorrhage ileus ascites adrenal insufficiency sodium-losing renal disease
Consider: SIADH, adrenal and thyroid deficiency
Consider: excess of sodium-free intravenous fluids (e.g. 5% dextrose) Water redistribution e.g. hyperglycaemia, mannitol therapy Pseudohyponatraemia e.g. hyperlipidaemia (TPN) hyperproteinaemia (myeloma)
Figure 48.12 The causes of hyponatraemia. TPN, total parenteral nutrition.
Skin 1203
hypo-osmolarity presents the chief danger, causing seizures and death if extreme. Removal of the cause by therapy of the tumour or other causative factor is central to management. However, when the serum sodium is very low, other direct measures to control this are important. Water restriction to 500 mL daily will usually bring improvement. The tetracycline derivative, demeclocycline (600–1200 mg daily) is also effective and has been found to be superior to lithium.88 If a patient is comatose or having convulsions, more rapid means of increasing the plasma sodium are imperative. The problem here is that rapid correction of hyponatraemia appears to be capable of causing permanent brain damage.89 The best course appears to be only to use a minimum of isotonic or hypertonic (308–462 mmol/L) saline, with care not to cause a rapid rise of plasma sodium (more than 12 mmol/L a day). Furosemide is usually required if hypertonic saline is given. It is very unusual, however, for intravenous saline to be justifiable and the other measures given are usually adequate.
Hypoglycaemia Although classically associated with insulinomas, hypoglycaemia is occasionally seen with almost any type of malignancy, particularly when there is a large tumour mass. Weakness, fatigue, confusion and occasionally focal neurological signs, most marked after fasting, are presenting features. Over-secretion of insulin is classically associated with insulinomas and very occasionally other non-islet cell tumours,90 but otherwise hypoglycaemia is caused by incompletely processed insulin-like growth factor-II (IGFII), which binds to and activates the insulin receptor.91 In addition, it also binds to IGF-I receptors in the pituitary, down-regulating growth hormone secretion, which further
exacerbates the hypoglycaemia by reducing gluconeogenesis in the liver. Treatment in emergency is by infusion of 50 per cent dextrose. Therapy should then be directed at tumour control. When this is not rapidly possible, frequent meals will often prevent symptoms and it is occasionally necessary for a patient to be wakened to eat in the early hours to avoid hypoglycaemia at the normal waking time. Glucocorticoids and growth hormone may also be effective if these measures are not sufficient.
Hypomagnesaemia This came to light as a frequent problem with the use of cisplatin, which causes impairment of renal absorptive capacity for magnesium.92 The renal loss can be compensated for with an adequate diet, but in anorexic patients hypomagnesaemia could cause muscular twitching, tetany and convulsions. The condition was often compounded by hypocalcaemia. Macaulay et al.93 showed that the problem could effectively be eliminated by giving magnesium sulphate (3 g in 3 litres of 154 mmol saline) as part of the intravenous infusion accompanying cisplatin therapy.
SKIN Many skin lesions have been proposed as paraneoplastic phenomena. For some, the association has not been proved, while others are so strongly associated that their recognition should be followed by a systematic search for malignancy. The subject has been well reviewed.94,95 Tables 48.2 and 48.3 summarize these different categories.
Table 48.2 Cutaneous conditions frequently associated with malignancy Condition Acanthosis nigricans Weber–Christian disease Gardner’s syndrome Tylosis Necrolytic migratory erythema Erythema gyratum repens Paget’s disease of nipple Bowen’s disease Hypertrichosis lanuginosa (malignant down) Dermatomyositis Fanconi’s anaemia Chediak–Higashi syndrome
Description
Associated malignancy
Pigmented hyperkeratosis in skin folds Crops of tender, erythematous subcutaneous areas of fat necrosis Epidermal and sebaceous cysts, dermoid tumours, lipomas, fibromas Hyperkeratosis of palms and soles after age of 10 Blistering and erosive erythema on limbs, mucositis Rapidly changing erythema Keratotic erythematous patch on or near nipple Non-elevated, scaly squamous carcinoma Rapidly growing long hair on ears, forehead and other parts Purplish erythema of face and hands Patchy hyperpigmentation Recurrent pyoderma, pigmentary dilution of skin and hair colour
Gastric and other GI carcinomas Pancreatic carcinoma Colonic carcinoma and polyposis (hereditary) Oesophageal cancer (hereditary) Glucagonoma Breast, lung Breast Internal malignancy Internal malignancy Various malignancies Leukaemia (hereditary) Lymphoma (hereditary)
1204 Medical care
Table 48.3 Cutaneous conditions occasionally associated with malignancy Condition Generalized melanosis Pigmentation of Addison’s disease Striae of Cushing’s syndrome Porphyria cutanea tarda Dermatitis herpetiformis Flushing Herpes zoster
Multiple seborrhoeic keratoses Exfoliative dermatitis Hirsutism Pruritus Peutz–Jeghers syndrome (pigmentation of lips, face, oral mucosa and digits) (hereditary) Neurofibromatosis
Associated malignancy Melanoma or chronic liver disease Infiltration of adrenals Ectopic ACTH, adrenal tumours Liver tumours Lymphomas and others Carcinoid tumours Lymphomas, immunosuppressive therapy Lymphomas, gastrointestinal tumours Lymphomas Adrenal tumours Lymphoma and others Gastrointestinal
Phaeochromocytoma
HAEMATOLOGICAL MANIFESTATIONS OF MALIGNANCY Anaemia Tumours commonly cause anaemia by direct blood loss from the tumour or bone marrow infiltration. Tumours can also be associated with an ‘anaemia of chronic disorders’, characterized by normal iron stores. Blood transfusion remains the mainstay of treatment for anaemia. However, clinical trials have shown that recombinant human erythropoietin (EPO) can reduce the degree of anaemia and the transfusion requirements in cancer patients, both on treatment with chemotherapy or not being treated, with a concomitant improvement in quality of life assessments.96,97 The use of EPO has been limited because it is only effective in a proportion of cancer patients (47 per cent of those with solid tumours), and for economic considerations. Red cell aplasia is reported with thymomas and a variety of other malignancies.98,99 Megaloblastic anaemias are associated with cytotoxic chemotherapy in the absence of vitamin B12 deficiency. When folate deficiency is shown in patients with cancer, it is customary not to treat it for fear that folate will permit more rapid tumour growth. Autoimmune haemolytic anaemias are associated
particularly with B-cell lymphomas but occur with many other malignancies and occasionally with cytotoxic drugs. When caused by tumour, they are usually refractory to treatment with corticosteroids but tend to resolve if the tumour is successfully treated.
Venous thrombosis The high incidence of venous thrombosis in cancer was recognized by Trousseau in the nineteenth century.100 Venous thromboembolism is the second most common cause of mortality in cancer patients, with high levels of coagulation factors and cellular components contributing to the phenomenon of hypercoagulability. Diagnosis is made using duplex ultrasound. D-dimers are the products of fibrin hydrolysis with elevated levels in fibrinolytic processes. However, levels of D-dimer can be artificially raised in patients with a malignancy such that a positive result does not verify the diagnosis of a deep vein thrombosis or pulmonary embolism. Its high predictive negative value is proved for the exclusion of thrombosis or pulmonary embolism, however, which is where it has value in the assessment of a cancer patient with possible thromboembolism.101 Initial treatment of venous thromboses to prevent thrombus growth and recurrence is low molecular weight heparin injections daily. Prophylactic anti-coagulation has generally not been given because of concern about haemorrhage, but a study of low-dose warfarin (starting dose 1 mg daily), in which international normalized ratio (INR) was maintained at 1.3–1.9 after the first 6 weeks, showed a relative risk reduction of 85 per cent by comparison with placebo-treated controls.102** The use of low molecular weight heparin for long-term secondary prevention/prophylaxis is thought to be as effective, with fewer bleeding complications and no monitoring requirements and it is now used routinely as an alternative to warfarin.103 Indefinite anti-coagulation is recommended for cancer patients.104 Several studies indicate that low molecular weight heparin may also improve the prognosis of cancer patients overall, but this needs to be confirmed in other studies. The cause of thromboembolism may not be evident but disseminated intravascular coagulation (DIC) is often implicated.
Pulmonary embolus Pulmonary embolism is an extremely common and highly lethal condition in patients with underlying malignancy. Prompt diagnosis and treatment can dramatically reduce mortality. Pulmonary thromboembolism is a complication of underlying venous thrombosis which can occur in any vessel including the calf veins, iliac and pelvic veins, subclavian system or from a thrombus forming around indwelling central venous catheters. Studies suggest that nearly every patient with a thrombus in the upper leg will have a
Infection 1205
pulmonary embolus if it is sought with a sensitive enough test.105 Current imaging techniques allow us to demonstrate PEs in 60–80 per cent of these patients. The classical triad of symptoms of a pulmonary embolism consists of haemoptysis, dyspnoea and pleuritic chest pain, but many patients are completely asymptomatic and others have atypical presentations (new onset atrial fibrillation, hypotension, etc.). Approximately 10 per cent of patients in whom a PE is diagnosed die in the first 60 minutes. Of the remainder, one third are correctly diagnosed and treated and one twelth still die from PE, but in the remaining, undiagnosed two thirds, one third will die from their PE. An ECG usually shows a sinus tachycardia, but in cases of chronic pulmonary emboli or massive acute pulmonary embolism there may be evidence of right heart strain. Arterial PO2 may be low. Chest X-ray is almost invariably normal, although rarely the Westermark sign is seen – dilatation of pulmonary vessels proximal to an embolism with collapse of distal vessels, sometimes with a sharp cut-off. . . A nuclear scintigraphic ventilation–perfusion (V/Q) scan is usually the primary investigation of choice, giving the probability of an embolus being present. A nondiagnostic or ‘intermediate prognosis’ scan report may require further imaging with a pulmonary angiogram, which if negative allows us to rule out a pulmonary embolus with 90 per cent certainty. High-resolution spiral computed tomographic angiography (CTA) has largely replaced angiography; it is as sensitive and specific but less invasive.. Patients with mediastinal disease will not . be assessable by V/Q scanning and should proceed straight to CTA. In a patient who has had a large PE, demonstrable by hypotension, severe hypoxia and . . right-sided heart strain, a confirmatory test such as a V/Q scan should not delay immediate administration of a fibrinolytic agent unless there is an absolute contraindication such as active internal bleeding, recent cerebrovascular accident or intracranial bleed, recent spinal surgery or trauma, uncontrolled hypertension or an aneurysm. Fibrinolytic agents include two forms of tissue plasminogen activator – alteplase/reteplase, which are faster acting and more e ffective than the more traditional agents – streptokinase/ urokinase. In patients with right heart strain, prompt fibrinolytic infusion can reduce mortality by 50 per cent. Heparin should then be given. Heparin will not help to disperse clot that has already formed, but by activating antithrombin III will prevent ongoing embolization of new thrombi. Effective anti-coagulation with heparin has been shown to decrease overall mortality from 30 per cent to less than 10 per cent for survivors of a PE. Low molecular weight heparin is now well established as being more effective and safer than unfractionated heparin. Patients should be given oxygen as required and encouraged to wear compression stockings. Warfarin can be commenced on days 1–3 of effective heparinization, and anti-coagulation should continue for at least 6 months but probably be lifelong for patients with underlying malignancy.
INFECTION Patients with cancer are commonly immunocompromised, either directly due to the disease or indirectly from treatment. Infections are common in this population and the most common manifestation is fever, although fever can be absent in the profoundly immunosuppressed or as a result of drugs such as steroids. Fever in such patients can also have non-infectious causes and, unfortunately, there is no pattern of fever which is pathognomonic of infections. The most important decisions regarding fever in the immunocompromised patient are the urgency of evaluation and empirical antibiotic therapy. For example, patients who are neutropenic from chemotherapy or bone marrow infiltration by tumour require urgent evaluation. The majority of patients (60 per cent) will not have an identifiable focus of infection, and while there are no definitive guidelines, the likely organisms can be identified from the type, degree and duration of immunosuppression as well as other factors, such as the presence of indwelling intravenous access catheters. For most chemotherapy regimens the period of neutropenia lasts less than 1 week, and bacteria are the most common organisms, with coagulasenegative staphylococci as the leading cause of acute bacterial infections in this population. The pattern of bacterial infections has changed significantly in the last 20–30 years in that the prevalence of Gram-negative organisms has decreased, whereas the isolation of Gram-positive organisms has increased and more recently the threat of MRSA (meticillin resistant Staphylococcus aureus) is foremost.106 Pseudomonas aeruginosa is still a large problem, although overall it is becoming less common. The increasing use of indwelling intravenous access catheters may be partly responsible for the dominance of coagulase-negative staphylococci. Patients with bone marrow transplants in the immediate post-transplant period have the same risk profile, but in the late post-transplant period (more than 100 days) become more susceptible to encapsulated bacteria such as Streptococcus pneumoniae. Treatment regimens must be dependent upon an institution’s own guidelines, but, in general, broadspectrum antibiotic therapy with intravenous combination antibiotics or monotherapy with third-generation cephalosporins, such as ceftazidime, is started while awaiting the results of the septic screen. Teicoplanin is routinely added if patients have an indwelling central venous catheter and, if MRSA is isolated, the combination of vancomycin and imipenem may be effective.107 There is also evidence that in selected patients treatment with oral antibiotics on an outpatient basis may be safe.108 The use of prophylactic antibiotics during predicted neutropenic periods is controversial but considered in certain higher risk patient groups.109*** The use of recombinant human granulocyte colony stimulating factor (G-CSF) has become routine in oncology where there is proven benefit in prophylaxis to prevent neutropenic sepsis110*** During an episode of neutropenic
1206 Medical care
sepsis there is evidence that G-CSF may shorten the duration of the neutropenic episode, but there is no evidence that there is any clinical benefit.111 However, in a severely ill neutropenic patient, administration of G-CSF remains common clinical practice, although there is a lack of clinical trial data to support this, largely due to difficulties in conducting trials in these situations. Description of the presentation and treatment of the many different types and manifestations of infections in patients with cancer is beyond the scope of this text. The reader is referred to specific reviews.112,113
CONCLUSIONS The medical manifestations of cancer have a fascinating diversity and include many syndromes which are more or less exclusive to patients with cancer. Recognition of the condition calls for considerable clinical acumen in patients who may have a range of other symptoms directly attributable to the malignancy. Our understanding of these conditions is increasing at the cellular and molecular level and will doubtless lead to better therapies that will improve the symptoms and save the lives of patients through appropriate diagnosis and treatment.
KEY LEARNING POINTS ●
●
●
●
●
Cancer is a multi-system disease and as such the oncologist should remain open-minded and think laterally about diagnosis and treatment in all situations. Medical aspects of care contribute significantly to quality of life. In the metastatic setting, this should be the priority. Simple steps such as good palliation of ascites and pleural effusions as well as the correction of metabolic disturbances and poor nutrition have a huge impact on quality of life. Medical aspects of care contribute significantly to the prospects of giving effective anti-tumour therapy so that the correction of metabolic abnormalities and optimization of organ function can affect outcome. Life-threatening conditions such pulmonary emboli need to be investigated and treated swiftly and aggressively. Newer techniques such as stenting and laser treatment for bronchial obstruction, stents for the superior vena cava and sophisticated surgery for spinal cord compression have reduced morbidity and improved outcome for many patients.
REFERENCES Seminal primary article Key review paper ❉ Management guideline 1 Finkelstein SE, Summers RM, Nguyen DM, Schrump DS. Virtual bronchoscopy for evaluation of airway disease. Thorac Surg Clin 2004; 14:79–86. Review. 2 Turner JF, Wang KP. Endobronchial laser therapy. Clin Chest Med 1999; 20:107–22. 3 Becker HD, Wagner B, Lierman D, et al. Stenting of the central airways. In: Liermann DD (ed.) Stents – State of the art and future developments. Boston: Boston Scientific Corporation, 1995, 249–55. ◆4 Prakash UB. Advances in bronchoscopic procedures. Chest 1996; 116:1403–8. 5 Saito Y, Imamura H. Airway stenting. Surg Today 2005; 35:265–70. ◆6 Simoff MJ. Endobronchial management of advanced lung cancer. Cancer Control 2001; 8:337–43. 7 Asimakopoulous G, Beeson J, Evans J, Maiwand M. Cryosurgery for malignant endobronchial tumours: analysis of outcome. Chest 2005; 127:2007–14. 8 Weiss RB, Muggia FM. Cytotoxic drug-induced pulmonary disease: update 1980. Am J Med 1980; 68:259–66. 9 Batist G, Andrews JL. Pulmonary toxicity of antineoplastic drugs. JAMA 1981; 246:1449–53. 10 Maher J, Daly PA. Severe bleomycin lung toxicity: reversal with high dose corticosteroids. Thorax 1993; 48:92–4. 11 Begent RHJ, Rustin GJS. Tumour markers: from carcinoembryonic antigen to products of hybridoma technology. Cancer Surv 1989; 8:107–21. 12 Light RW, MacGregor MI, Luchsinger PC, Ball WC. Pleural effusions: the diagnostic separation of transudates and exudates. Ann Int Med 1972; 77:507–13. ◆13 Moores DW. Malignant pleural effusion. Semin Oncol 1991; 18:59–61. ◆14 Hartman DL, Gaither JM, Kesler KA, et al. Comparison of insufflated talc under thorascopic guidance with standard tetracycline and bleomycin pleurodesis for control of malignant of pleural effusions. J Thorac Cardiovasc Surg 1993; 105:743–7. ●15 Martinez-Moragon E, Aparicio J, Rogado MC, et al. Pleurodesis in malignant pleural effusions: a randomised study of tetracycline versus bleomycin. Eur Resp J 1997; 10:2380–3. 16 Cameron R, Davies HR. Intra-pleural fibrinolytic therapy versus conservative management in the treatment of parapneumonic effusions and empyema. Cochrane Database Syst Rev 2004; (2):CD002312. 17 Yeo W, Leung SF, Chan AT, Chiu KW. Radiotherapy for extreme hypertrophic pulmonary osteoarthropathy associated with malignancy. Clin Oncol 1996; 8:195–7. ◆18 Perez CA, Presant CA, Amburg AL. Management of superior vena cava syndrome. Semin Oncol 1978; 5:123–34. 19 Lokich JJ, Goodman RL. Superior vena cava syndrome. JAMA 1975; 231:58–71. ● ◆
References 1207
20 Greillier L, Barlesi F, Doddoli C, et al. Vascular stenting for palliation of superior vena caval obstruction in non-small cell lung cancer patients: a future ‘standard’ procedure? Respiration 2004; 71:178–83. 21 Wilson E, Lyn E, Lynn A, Khan S. Radiological stenting provides effective palliation in malignant central venous obstruction. Clin Oncol 2002; 14:228–32. ◆22 Davenport D, Ferree C, Blake, D. Radiation therapy in the treatment of superior vena cava obstruction. Cancer 1978; 42:2600–3. ◆23 Rowell NP, Gleeson FV. Steroids, radiotherapy, chemotherapy and stents for superior vena caval obstruction in carcinoma of the bronchus: a systematic review. Clin Oncol (R Coll Radiol) 2002; Oct 14:338–51. 24 Zollikofer CL, Anonucci F, Stuckman G, et al. The use of stents in venous vessels. In: Liermann DD (ed) Stents – state of the art and future developments. Boston: Boston Scientific Corporation, 1995, 73–8. ❉25 Hankins JR, Satterfield JR, Aisner J, et al. Pericardial window for malignant pericardial effusion. Ann Thorac Surg 1980; 30:465. ◆26 Singal PK, Iliskovic N. Current concepts: doxorubicininduced cardiomyopathy. N Engl J Med 1998; 339: 900–5. 27 Von Hoff D, Layard D. Risk factors for doxorubicin induced congestive cardiac failure. Ann Int Med 1979; 91: 710–17. 28 Ewer MS, Benjamin RS. Cardiotoxicity of chemotherapeutic drugs. In: Perry MC (ed) The Chemotherapy Source Book, 2nd edn. Baltimore: Williams and Wilkins, 1996, 649–64. ❉29 Hensley ML, Schucter LM, Lindley C, et al. American Society of Clinical Oncology clinical practice guidelines for the use of chemotherapy and radiotherapy protectants. J Clin Oncol 1999; 17:3333–5. 30 Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al. Trastuzumab after standard adjuvant chemotherapy in HER-2 positive breast cancer. N Engl J Med 2005; 353:1659–72. 31 Suter TM, Cook-Bruns N, Barton C. Cardiotoxicity associated with trastuzumab (Herceptin) therapy in the treatment of metastatic breast cancer. Breast 2004; 13:171–2. ◆32 Sing B, Kim H, Wax SJ. Stent versus nephrostomy: is there a choice? J Urol 1979; 121:268–70. 33 Fallon B, Olney L, Culp DA. Nephrostomy in cancer patients: to do or not to do. Br J Urol 1980; 52:237–42. 34 Ho PC, Talner LB, Parsons CL, Schmidt JD. Percutaneous nephrostomy: experience in 107 kidneys. Urology 1980; 16:532–5. 35 Newlands ES, Begent RHJ, Rustin GJS, et al. Further advances in the management of malignant teratomas of the testis and other sites. Lancet 1983; i:948–51. ◆36 Zimmerman SW. Glomerulopathies associated with malignant disease. In: Garnick MB, Rieselbach RE (eds) Cancer and the kidney. Philadelphia: Lea and Febiger, 1982. ●37 Siegal T, Seligsohn U, Aghai E, Modan M. Clinical and laboratory aspects of disseminated intravascular
38
◆39
◆40
41 42
43
44
45
46 47
48 49 ◆50
◆51
52
53
54
coagulation (DIC), a study of 118 cases. Thromb Haemost 1978; 39:122. Crittenden DR, Akerman GL. Hyperuricaemia acute renal failure in disseminated carcinoma. Arch Intern Med 1977; 137:97–9. Warrell RP. Metabolic emergencies. In: DeVita VT, Hellman S, Rosenberg SA (eds) Cancer: Principles and Practice of Oncology, 4th edn. Philadelphia: Lippincott, 1993, 2128–37. Gentilni P, La Villa G, Casini-Raggi V, et al. Hepatorenal syndrome and its treatment today. Eur J Gastroent Hepatol 1999; 11:1061–5. Body JJ. The syndrome of anorexia–cachexia. Curr Opin Oncol 1999; 11: 255–60. Larson DE, Burton DD, Schroder KW, DiMagno EP. Percutaneous endoscopic gastrostomy. Gastroenterology 1987; 93:48–52. Ottery FD, Walsh D, Strawford A. Pharmacologic management of anorexia–cachexia. Semin Oncol 1998; 25:35–44. Loprinzi CL, Kugler JW, Sloan JA, et al. Randomised comparison of megestrol acetate versus dexamethasone versus fluoxymesterone for the treatment of cancer anorexia/cachexia. J Clin Oncol 1999; 17:3299–306. Armstrong T, Walters E, Varshney S, Johnson CD. Deficiencies of micronutrients, altered bowel function and quality of life during late follow-up after pancreaticoduodenectomy for malignancy. Pancreatology 2002; 2:528–34. Fugh-Berman A. Alternative Medicine: What Works. Baltimore: Lippincott Williams and Wilkins, 1997. Ahmed FE. Effect of diet, life style, and other environmental/chemoprotective factors on colorectal cancer development, and assessment of the risks. J Environ Sci Health C Environ Carcinog Ecotoxicol Rev 2004; 22:91–147. Zollikofer CL, Jost R, Schoch E, Decurtins M. Gastrointestinal stenting. Eur Radiol 2000; 10:329–41. Bown SG. New techniques in laser therapy. BMJ 1998; 316:754–7. Baron TH. Colonic stenting: technique, technology, and outcomes for malignant and benign disease. Gastrointest Endosc Clin N Am 2005; 15:757–71. Feuer DJ, Broadley KE. Systematic review and metaanalysis of corticosteroids for the resolution of malignant bowel obstruction in advanced gynaecological and gastrointestinal cancers. Ann Oncol 1999; 10:1035–41. Khoo D, Hall E, Motson R, et al. Palliation of malignant intestinal obstruction using octreotide. Eur J Cancer 1994; 30:28–30. Campagnutta E, Cannizzaro R, Gallo A, et al. Palliative treatment of upper intestinal obstruction by gynecological malignancy: the usefulness of percutaneous endoscopic gastrostomy. Gynecol Oncol 1996; 62:103–5. Speer AG, Cotton PB, Russell RCG, et al. Randomised trial of endoscopic versus percutaneous stent insertion in malignant obstructive jaundice. Lancet 1987; ii:57–62.
1208 Medical care
55 Greenway B, Johnson PJ, Williams R. Control of malignant ascites with spironolactone. Br J Surg 1982; 69:441–2. 56 Strauss AK, Roseman DL, Shapiro TM. Peritoneovenous shunting in the management of malignant ascites. Arch Surg 1977; 114:489–91. ●57 Goldman JM, Ash CM, Souhami RL, et al. Spinal cord compression in small cell lung cancer: a retrospective study of 610 patients. Br J Cancer 1989; 59:591–3. 58 Hill ME, Richards MA, Gregory WM et al. Spinal cord compression in breast cancer: a review of 70 cases. Br J Cancer 1993; 68:969–73. ❉59 Sorenson S, Helweg-Larsen S, Mouridsen H, Hansen HH. Effect of high-dose dexamethasone in carcinomatous metastatic spinal cord compression treated with radiotherapy: a randomized trial. Eur J Cancer 1994; 30A:22–7. ◆60 Siegal T, Siegal T. Treatment of malignancy epidural cord and cauda equina compression. Prog Exp Tumor Res 1985; 29:225–34. ◆61 Sundaresan N, Galicich JH, Lane JM, et al. Treatment of neoplastic epidural spinal cord compression by vertebral body resection and stabilisation. J Neurosurg 1985; 63:676–84. 62 Gilbert RW, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumour: diagnosis and treatment. Ann Neurol 1987; 3:40–51. ●63 Patchell RA, Tibbs PA, Regine WF, et al. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer. Lancet 2005; 366:643–8. ❉64 Friedman HM, Sheetz S, Levine HL, et al. Combination chemotherapy and radiotherapy: the medical management of epidural spinal cord compression. Arch Intern Med 1986; 146:509–12. ❉65 Vecht CJ. Clinical management of brain metastasis. J Neurol 1998; 245:127–31. ●66 Athanassiou A, Begent RHJ, Newlands ES, et al. Central nervous system metastases of choriocarcinoma: 23 years experience at Charing Cross Hospital. Cancer 1983; 52:1728–35. 67 Young RF. Radiosurgery for the treatment of brain metastases. Semin Surg Oncol 1998; 14:70–8. ◆68 Kondziolka D, Martin JJ, Flickinger JC, et al. Long-term survivors after gamma knife radiosurgery for brain metastases. Cancer 2005; 104:2784–91. ◆69 Postmus PE, Smit EF. Chemotherapy for brain metastases of lung cancer: a review. Ann Oncol 1999; 10:753–9. ◆70 Bunn PA, Ridgeway EC. Paraneoplastic syndromes. In: DeVita VT, Hellman S, Rosenberg, SA (eds) Cancer: Principles and Practice of Oncology, 4th edn. Philadelphia: Lippincott, 1993, 2026–71. 71 Voltz R, Gultekin H, Rosenfeld MR, et al. A serologic marker of paraneoplastic limbic and brain stem encephalitis in patients with testicular cancer. N Engl J Med 1999; 340:1788–95. ◆72 Lieberman FS, Schold SC. Distant effects of cancer on the nervous system. Oncology (Williston Park) 2002; 16:1539–48.
73 Nisula BC, Taliadouros GS. Thyroid function in gestational trophoblastic neoplasia: evidence that the thyrotrophic activity of chorionic gonadotrophin mediates the thyrotoxicosis of choriocarcinoma. Am J Obstet Gynecol 1980; 138:77. 74 Lokich JJ. The frequency and clinical biology of the ectopic ACTH syndromes of small cell carcinoma. Cancer 1980; 50:2111–14. ◆75 DeLellis RA, Xia L. Paraneoplastic endocrine syndromes: a review. Endocr Pathol 2003; 14:303–17. 76 Engelhardt D, Weber MM. Therapy of Cushing’s syndrome with steroid biosynthesis inhibitors. J Steroid Biochem Mol Biol 1994; 49:261–7. 77 Crawford SM, Ledermann JA, Turkie W, et al. Is production of ectopic human chorionic gonadotrophin (hCG) and alpha fetoprotein (AFP) by tumours a marker of chemosensitivity? Eur J Cancer Clin Oncol 1986; 22:1483–7. ❉78 Body JJ, Bartl R, Burckhardt P, et al. Current use of bisphosphonates in oncology: International Bone and Cancer Study Group. J Clin Oncol 1998; 16:3890–9. 79 Coleman RE. How can we improve the treatment of bone metastases further? Curr Opin Oncol 1998; 10 (suppl. 1): S7–13. ❉80 Michaelson MD, Smith MR. Bisphosphonates for treatment and prevention of bone metastases. J Clin Oncol 2005; 23:8219–24. 81 De Cock E, Hutton J, Canney P, et al. Cost-effectiveness of oral ibandronate versus IV zoledronic acid or IV pamidronate for bone metastases in patients receiving oral hormonal therapy for breast cancer in the United Kingdom. Clin Ther 2005; 27:1295–310. 82 Ralston SH, Alzait AA, Gardner MD, Boyle IT. Treatment of cancer associated hypercalcaemia with combined diphosphonate and calcitonin. BMJ 1986; 292:1549–50. 83 Thallasionos N, Joplin GF. Phosphate treatment of hypercalcaemia due to carcinoma. BMJ 1968; 4:14–19. ◆84 Mundy GR, Wilkinson R, Heath DA. Comparative study of available medical therapy for hypercalcaemia of malignancy. Am J Med 1983; 74:421–32. 85 Pecherstorfer M, Brenner K, Zojer N. Current management strategies for hypercalcaemia. Treat Endocrinol 2003; 2:273–92. 86 Sorensen JB, Andersen MK, Hansen HH. Syndrome of inappropriate secretion of antidiuretic hormone in malignant disease. J Int Med 1995; 238:97–110. 87 DeFronzo RA, Thier SO. A pathophysiologic approach to hyponatraemia. Arch Intern Med 1980; 147:897–902. 88 Forrest JN, Cox M, Hong C, et al. Superiority of demeclocycline over lithium in treatment of chronic syndrome of inappropriate secretion of antidiuretic hormone. N Engl J Med 1978; 298:173–7. 89 Swales JD. Dangers in treating hyponatraemia. BMJ 1987; 294:261–2. 90 Seckl MJ, Mulholland PJ, Bishop AE, et al. Hypoglycaemia due to an insulin-secreting small cell carcinoma of the cervix. N Engl J Med 1999; 341:733–7.
References 1209
91 Le Roith D. Tumour-induced hypoglycaemia. N Engl J Med 1999; 341:757–9. 92 Schlinsky RL, Anderson T. Hypomagnesaemia and renal magnesium wasting in patients receiving cisplatin. Ann Intern Med 1979; 90:929–31. 93 Macaulay VM, Begent RHJ, Philip ME, Newlands ES. Prophylaxis against hypomagnesaemia induced by cisplatinum combination chemotherapy. Cancer Chemother Pharmacol 1982; 9:179–81. 94 Helm F, Helm J. Cutaneous markers of internal malignancies. In Helm F (ed) Cancer Dermatology. Philadelphia: Lea and Febiger, 1979, 247–83. 95 Callen JP. Cutaneous complications of cancer. In: Calabresi P, Schein P, Rosenberg SA (eds) Medical Oncology. Basic Principles and Clinical Management. New York: Macmillan, 1985, 223–34. 96 Thatcher N. Management of chemotherapy-induced anaemia. Semin Oncol 1998; 25 (3 suppl. 7):23–6. 97 Groopman JE, Itri LM. Chemotherapy-induced anemia in adults: incidence and treatment. J Natl Cancer Inst 1991; 91:1616–34. 98 Jacobs EM, Hutter RVP, Pool JL, Ley AB. Benign thymoma and selective erythroid aplasia of the bone marrow. Cancer 1959; 12:47–57. 99 Guthrie TH, Thornton RM. Pure red cell aplasia obscured by a diagnosis of carcinoma. South Med J 1983; 76: 632–4. 100 Rickles FR, Edwards RL. Activation of blood coagulation in cancer: Trousseau’s syndrome revisited. Blood 1983; 63:14–31. 101 Ziegler T, Murzik M, Schau A, et al. Interpretation of increased D-dimer values. Hamostaseologie 2004; 24:144–6. ●102 Levine M, Hirsh J, Gent M, et al. Double blind randomised trial of very low dose warfarin for prevention of
103 ◆104
❉105
106
107
❉108
◆109
❉110
111
112 113
thromboembolism in stage IV breast cancer. Lancet 1994; 343:886–9. Luxembourg B, Bauersachs R. Malignancy and thrombosis: a double-sided clinical relationship. Vasa 2005; 34:225–34. Sanchez O, Wermert D, Allain YM, et al. Treatment of venous thromboembolic disease. Presse Med 2005; 34:1427–34. Feied C, Handler JA. Pulmonary embolism. eMedicine 2004. Kanabmaru A, Tatsumi Y. Microbiological data for patients with febrile neutropenia. Clin Infect Dis 2004; 39 Suppl 1: S7–S10. Totsuka K, Shiseki M, Kikuchi K, Matsui Y. Combined effects of vancomycin and imipenem against methicillin resistant Staphylococcus aureus (MRSA) in vitro and in vivo. J Antimicrob Chemother 1999; 44:455–60. Finberg RW, Talcott JA. Fever and neutropenia – How to use a new treatment strategy. N Engl J Med 1999; 341:362–3. Van de Wetering MD, de Witte MA, Kremer LC, et al. Efficacy of oral antibiotics in neutropenic afebrile oncology patients: a systematic review of randomised controlled trials. Eur J Cancer 2005; 41:1372–82. American Society of Clinical Oncology. Update of recommendations for the use of hematopoietic colonystimulating factors: evidence-based, clinical practice guidelines. J Clin Oncol 1996; 14:1957–60. Hartmann LC, Tschetter LK, Habermann TM, et al. Granulocyte colony-stimulating factor in severe chemotherapy-induced afebrile neutropenia. N Engl J Med 1997; 336:1776–80. Rosenberg AS, Brown AE. Infection in the cancer patient. Disease a Month 1993; 39:505–69. Pizzo PA. Current concepts: fever in immunocompromised patients. N Engl J Med 1999; 341: 893–900.
49 Interstitial brachytherapy J. ROGER OWEN AND DAVID BOTTOMLEY
Historical introduction General considerations Radiation safety Dosimetry Pre-planning an implant Preparation of implant patients and management of complications
1210 1210 1211 1211 1216 1218
HISTORICAL INTRODUCTION The story of brachytherapy begins in 1898 when Marie and Pierre Curie, working in their laboratory in Paris, succeeded in isolating radium from pitchblende. Within a few years Robert Abbe, an American surgeon, had used radium, with an afterloading technique, to treat cancer. The first English-language textbook of radium therapy, by Wickham and Degrais, was published in 1910.1 Between the end of the First World War and the early 1930s, the major developmental work was carried out in Paris. Basic implantation techniques and clinical indications were defined. Clinical studies in patients with cervical cancer led to an optimization of the radiation dose and dose rate. A little later, a major development took place in England where painstakingly careful measurement of exposure rates in air, at certain distances from radiation sources, by Meredith in the early 1930s led to the development of the ‘Manchester system’.2 This didactic set of rules rid brachytherapy of empiricism and permitted implants to be performed according to a predetermined plan. The importance of this system, which became universally accepted, cannot be overemphasized. For the first time it set limits within which an implant would be safe, but outside of which the risks to the patient were too high. This principle holds true today with other more modern systems, such as the ‘Paris system’. The advent of megavoltage machinery resulted in a decline in brachytherapy during the 1950s and 1960s. However, the development of artificial radioactive isotopes brought new
Dose and dose-rate considerations Dosimetry in vivo Clinical indications Prostate brachytherapy Conclusion Key references References
1221 1222 1225 1227 1230 1230 1231
possibilities. Afterloading was developed using smaller, flexible sources. In Europe it was again the Parisians who took a leading role, particularly at the Institut Gustav Roussy, where Bernard Pierquin, following his father and godfather before him, Andrée Dutreix and Daniel Chassagne, opened new possibilities for brachytherapy, using iridium-192 wire with afterloading techniques.3 These methods, and the accompanying Paris system of dosimetry, form the basis for this chapter.
GENERAL CONSIDERATIONS There are two essential characteristics of interstitial brachytherapy, the first of which is probably the more important: 1. Physical: by placing the radioactive sources directly into the tumour-bearing tissue, very high and localized doses are achieved. The rapid fall-off of dose with distance ensures that surrounding normal tissues are relatively spared and that the irradiated volume is precisely tailored to the tumour-bearing tissues. 2. Biological: low, continuous dose-rate treatment and short overall treatment time. Radical doses can be given over a few days only. Brachytherapy techniques tend to be time-consuming and therefore relatively costly when compared with a short course of external irradiation. Their use in palliation is
Dosimetry 1211
therefore seldom justified. The primary indication is the radical treatment of relatively small, well-circumscribed tumours in accessible sites. Radical doses may be given by the implant alone, or the implant may be used to boost the dose locally after external irradiation. Evidence is accumulating, although not from randomized clinical trials, that the chances of cure are highest when the total dose is given by the implant. The mean dose within an implanted volume may be in the region of twice the dose prescribed to the reference isodose. Thus, a radical implant delivering 60 Gy to the treatment isodose could result in a mean tumour dose of 120 Gy or more. In comparison, an implant boost of 25 Gy after 50 Gy external irradiation to the same tumour leads to an estimated total tumour dose of only 100 Gy (with no allowance for the different dose rates). In cases where a tumour is fairly large and there is a significant risk of subclinical lymph-node metastases, external irradiation may be given first, in sufficient doses to give a high chance of sterilizing these small-volume tumour deposits. In general, the volume of tissue implanted to give the ‘boost’ in these circumstances is substantially the same volume as would have been implanted had no external irradiation been given. However, the implant dose is lower, thus reducing the risk of normal tissue necrosis, but at the same time risking a reduced chance of local cure. Another way of maximizing the benefits of brachytherapy is to perform a perioperative implant. Thus, the bulk of the tumour is removed surgically and the excision margins can be implanted under direct vision. Although high doses can be given by implantation techniques, there is inevitably a rapid fall-off and cases where the tumour margin is poorly defined or very irregular may not be suitable because of the risk of underdosing the tumour edge. Very large-volume implants are poorly tolerated and also there will be significantly higher doses in the central part of the treatment volume compared to the periphery. Some normal tissues have a poor tolerance of implantation (e.g. bone and skin at pressure points or of the lower leg). The risk of normal tissue necrosis at such sites may be high. Inevitably some tumour sites are poorly accessible and geometrically satisfactory implants in these regions are difficult to achieve.
Loading and unloading the implant 1. Implants should only be loaded or unloaded by experienced trained staff – it is not a job for an inexperienced or unsupervised junior doctor. 2. Loading and unloading must take place under strictly controlled conditions. The following are essential: (a) a relaxed, pain-free patient; (b) the operator should not be working alone as he/she would be unable to leave the patient to call for help if necessary; (c) some local anaesthetic and catgut on a roundbodied needle should be immediately available in case of haemorrhage that does not respond to pressure (rare); (d) a good light and the correct equipment; (e) as far as possible these procedures should be timed to occur during normal working hours; (f) radiation monitoring of the patient and room after removal of sources to ensure that no source, or part of a source, is left in the patient or protected room.
DOSIMETRY The dose received at a point in the vicinity of a brachytherapy source diminishes rapidly with distance. This is because of the inverse square law, which dominates the dosimetry of interstitial therapy. A dosimetry system is similar in principle to the ‘Highway Code’, which does not tell you how to drive a car, but does set boundaries beyond which it is unsafe to stray. Similarly, dosimetry systems set the parameters within which it is safe to work, where the risks of causing necrosis due to very high doses, or of tumour recurrence as a result of underdosing, are minimized. All dosimetry systems consist of two parts:1 the rules of implantation; and2 the method of calculating the time the sources must remain in place to deliver the prescribed dose to the intended volume of tissue. Although other dosimetry systems exist, the Paris system is best adapted to interstitial brachytherapy using iridium wire (Table 49.1).4 This system was born of the twin needs to extend the indications of brachytherapy by using newly available miniaturized sources, but at the same time Table 49.1 Physical properties of iridium-192
RADIATION SAFETY Each radiotherapist engaged in brachytherapy must develop, with his physicists and radiographers, a system of safe practice to minimize the chances of preventable errors.
Production Half-life of 192Ir Decay
Iridium-192 wire
Monitoring the patient A system for checking the position and number of sources at regular intervals during the implant must be established and a record sheet maintained by nursing staff.
Protection
Ir (nγ) : 192Ir 74 days 16 gamma-ray energies, 4 with an abundance of greater than 10% Weighted mean energy 0.370 MeV Iridium/platinum alloy in platinum sheath Overall diameter 0.3 mm for wire; 0.5 or 0.6 mm for ‘hairpins’ A range of linear activities is available The wire is springy but easily bent Half-value thickness 3 mm lead 191
1212 Interstitial brachytherapy
basing the dosimetry calculations on the dose at certain fixed points within the implanted volume. The system can be used for straight or curved wires, loops or hairpins. It is flexible and allows the separation between sources to be varied, within limits, to suit the volume being treated. Above all, the Paris system allows the brachytherapist to rapidly pre-plan his implant without needing to resort to complicated calculations or tables.
Principles of the Paris system 1. The radioactive sources must be parallel and arranged so that their centres lie in the same plane, called the ‘central plane’. This plane is the perpendicular bisector of straight radioactive source wires, or passes through the centres of curved wires or the legs of hairpins and loops (Fig. 49.1). 2. The linear activity of each source wire must be uniform along its length and identical for all the sources. However, seed ribbons are permitted within the system provided the distance between adjacent seeds is equal or less than 1.5 times the active length of the seed and, similarly, a mobile stepping source is permitted as long as the steps and dwell time in each position are identical for each of the vectors. 3. The radioactive sources must be equidistant. When more than one plane of sources is required, these are positioned so that their pattern as they cross the central plane is either equilateral triangles or squares (Fig. 49.2).
Dose specification
1. The reference isodose line (to which the treatment prescription is applied) is a fixed percentage (85 per cent) of the basal dose rate, and its position coincides with the borders of the target volume. 2. The diameter of the high-dose sleeve around each source wire is not allowed to become too large. If it were to do so, there would be an unacceptable risk of necrosis. By these means, the tumour receives the intended dose but, at the same time, the risk of necrosis is controlled.
(a)
d
(a)
(b)
0.87 d
separation d
(c)
In the Paris system dosimetry calculations are always made in the central plane. Around each linear source wire lies a high-dose zone or sleeve; midway between adjacent sources there will lie a minimum dose-rate point (basal dose rate) and surrounding the implant as a whole there are continuous isodose lines or envelopes (Fig. 49.3).
d
Figure 49.2 The principle of equidistance. (a) Single plane of wires; (b) wires arranged ‘in triangles’; (c) wires arranged ‘in squares’. Adapted from Pierquin and Marinello (1992).
(b)
Central plane
Height difference 1 cm
1 cm (c)
The system enables the radiotherapist to perform an implant that has the following characteristics:
(d)
Figure 49.1 Definition of the central plane with examples. (a) Single plane of wires, e.g. skin implant; (b) curved plane of wires, e.g. anal canal implant; (c) hairpins in tongue; (d) curved wires, e.g. soft palate implant. Adapted from Pierquin and Marinello (1992).
Dosimetry 1213
85 98
Central plane
102 170
Central 1
plane
3
100% BD
4 Central plane
170 102
2
AP
LAT
98 85%
Figure 49.3 Dose distribution in the central plane for three linear coplanar sources. BD, basal dose. Adapted from Pierquin and Marinello (1992).
Definitions ●
●
●
●
●
The central plane is the plane at right angles to the plane of the wires and passes through their mid points. In practice, the mid points of each wire seldom lie exactly in one plane and a mean central plane is used (Fig. 49.4). Basal dose rate within the implant: between the sources are points where the dose rate is minimal. These are elementary minimal dose rates. The basal dose rate of the implant is the arithmetic mean of these individual dose rates, which should not vary from the mean by 10 per cent. Examples of basal dose rates are shown in Figures 49.5 and 49.6. Reference isodose: the prescribed dose is applied to the reference isodose line, which is defined as 85 per cent of the mean basal dose rate, thus inextricably linking the prescribed dose to the minimal doses in the interior of the implant. Tumour volume is that volume known to be occupied by tumour and it may be estimated by direct measurement or radiography. Target volume is the known tumour volume with a surrounding margin of tissue, which may contain tumour cells at the microscopic level.
Figure 49.4 implants. ●
●
●
●
●
Central plane – examples from radiographs of
Treated volume is that volume encompassed by the reference isodose. This can be defined by its thickness, length and width. An appreciation of these parameters and their relation to the positions of the sources is the central concept of the Paris system. Figures 49.7–49.11 show how the thickness treated, length treated and width treated are defined. Thickness treated (T) is the average thickness of tissue encompassed by the reference isodose (Figs 49.7, 49.9–49.11). Length treated (L) is the average length of tissue covered by the reference isodose, midway between wires for single-plane and midway between the planes for multiplane implants (Figs 49.9 and 49.10). High-dose volume is the volume around each source wire that is encompassed by the isodose that is twice the reference isodose, i.e. 170 per cent of the basal dose rate (Figs 49.7–49.10). Lateral safety margin is the distance, measured in the central plane of the implant and midway between the planes of wires for multi-plane implants, from the outmost source wire to the position of the reference isodose (Figs 49.7, 49.9 and 49.10).
1214 Interstitial brachytherapy
Central plane BD
BD1 BD2
BDn
(b)
(a)
90
BD3
BD2
BD4
BD1
Figure 49.5 Calculation of basal dose rate for straight sources in flat or curved arrangement. Adapted from Pierquin and Marinello (1992).
(c)
BD2
d d
BD1
BD
BD3
BD1 BD2 BD3 3
(a)
BD1
d
d
(b)
BD
BD1 BD2 2
BD2
Figure 49.6 Calculation of basal dose rate for implants in (a) triangles or (b) squares. BD, basal dose. Adapted from Pierquin and Marinello (1992).
How to find the central plane As the Paris system allows the sources to be of different lengths, this gives rise to some difficulties in finding the central plane. In practice, a mean central plane is normally used. Figure 49.4 gives some examples.
For hairpins, the central plane is that plane passing through the mid points or the straight ‘legs’ of each hairpin (Fig. 49.6). Similarly, loops have to be divided into a curved bridge piece and straight legs, and the central plane passes through the mid points of the legs. A loop must
Dosimetry 1215
High-dose sleeve
Treated length BD1
BD2
BD
Length treated
BD3
c b RI Reference isodose
High-dose sleeve
a
Lateral safety margin
RI
Plane a BD1
BD2
BD3
Thickness treated
Thickness treated BD
Plane b
Figure 49.7 Definition of the volume treated and the high-dose sleeve for implants of several co-planar lines. BD, basal dose. Adapted from Pierquin and Marinello (1992).
Plane c
Safety margin
Figure 49.9 Definition of dimensions of volume treated and high-dose sleeve for implantation of triangles. BD, basal dose; RI, reference isodose. Adapted from Pierquin and Marinello (1992).
their heights (Fig. 49.1). The plane used to reconstruct the implant and for dosimetry must not be more than 10 degrees beyond the perpendicular, or the reconstruction will make it appear that the hairpins are significantly further apart than they are in reality and the patient will be overdosed as a result.
D ref 85% BD B A
B
Basal dose rate points (BD)
High dose rate sleeve
Safety margin A
Central plane
Figure 49.8 Definition of treated volume and high-dose sleeve for straight sources arranged around a curve. BD, basal dose. Adapted from Pierquin and Marinello (1992).
correspond to certain criteria before it can be used for volume implants (Fig. 49.12). Although the central plane is ideally perpendicular to the wires, situations arise when this is not the case, e.g. when hairpins are placed in the tongue and there is a difference in
Relationship between source positions and reference isodose In the Paris system the prescribed dose is applied to the reference isodose, which has a fixed relationship to the average basal dose rate within the radioactive set-up. In their turn, the basal dose rates depend upon the distances between each source and the basal dose rate points, the linear activity of the sources and the radioisotope used. Thus, given iridium-192 wire of uniform activity, there are easily definable relationships between the wire positions and the treated volume, as described by the reference isodose. These are summarized in Table 49.2.
Relationship between source positions and high-dose volume Figure 49.13 shows that, as the distance between adjacent source wires increases, the diameter of each sleeve of high
1216 Interstitial brachytherapy
High-dose sleeves d/2
b 1.5 d
BD1
Length treated
BD2
Spacing (d)
Half the bridge section
1.0 cm
0.8 cm
1.5 cm
1.2 cm
2.0 cm
1.6 cm
2.5 cm
2.0 cm
d Half the active length half the bridge section length of one leg (b)
Figure 49.12 Conditions that need to be satisfied for a loop to be used for a ‘volume implant’. Adapted from Pierquin and Marinello (1992).
a b (Reference isodose)
Safety margin
Plane b
Thickness treated
BD2
BD1
Plane a
Figure 49.10 Definition of dimensions of treated volume and high-dose sleeves in cases of implantation in squares. BD, basal dose. Adapted from Pierquin and Marinello (1992).
dose around the sources also becomes progressively greater; the relationship is non-linear. Clinical experience over 30 years has led to the understanding that, when this highdose sleeve is more than 8 mm in diameter, there is a significantly higher risk of necrosis. The graph shows that this critical point is reached at a separation of 22 mm for wires of 10 cm length but at about 20 mm for 5 cm wires. For wires of less than 5 cm length it is reached when their separation is 15–16 mm.
Practical constraints on the wire separation Just as the excess risk of necrosis occurring when wires are placed far apart limits their maximum spacing, so there are limitations that govern how close the sources may be placed. These are partly practical and partly dosimetric. First, it is, in practice, very difficult to achieve a good geometry when implanting many wires very close together. Secondly, if the wires are too near, then minor deviations from parallelism will produce major variations in dose. Table 49.3 shows the practical safe ranges for wire separation.
Thickness treated
Length treated
PRE-PLANNING AN IMPLANT Faced with a patient whose tumour is suitable for implantation, the brachytherapist must pre-plan how the desired dose distribution will be achieved and which implant technique will be used.
Width treated
Figure 49.11 Definition of dimensions of treated volume in the case of hairpins or loops. The ‘thickness’ is the dimension parallel to the bridge of the loop. Adapted from Pierquin and Marinello (1992).
Steps in achieving a correct dose distribution 1. Careful clinical evaluation of the tumour volume. 2. Evaluation of the desired tumour margin around the tumour that is to be treated to the prescribed dose, i.e. the target volume. Measurement of the thickness, width and length of this target volume.
Pre-planning an implant 1217
Table 49.2 Relationships between wire positions and treated volume (adapted from Pierquin and Marinello, 1992) Arrangement 2 lines ‘n’ lines in one plane ‘n’ lines in squares ‘n’ lines in triangles
Treated length/ active length
Thickness treated/ separation
Lateral safety margin separation
Safety margin/ separation
0.7 0.7 0.7 0.7
0.5 0.6 1.55–1.60 1.3
0.37 0.33 – –
– – 0.27 0.20
Diameter 200% isodose (mm)
(a) 10 8 6
5 cm wire 10 cm wire
4 2 0
8
10
12
14 16 18 20 Wire separation (mm)
22
24
26
(b) Volume treated (cm3)
5
can be irradiated by a single plane implant is 20 0.612 mm. Target volumes thicker than 12 mm require two or more planes of wires. 4. If a multi-plane implant is to be used, a decision must be made whether to implant the sources ‘in triangles’ or ‘in squares’. This is determined by the shape of the target volume and also by the tumour site. 5. With the aid of the ratio of thickness treated to source separation (Table 49.2) corresponding to the source arrangement chosen, determination of the minimal separation to adhere to and the number of sources needed. 6. Calculation of the length of each wire by multiplying the target volume length by 1.4 and multiplying this length by the number of wires to obtain the total amount of wire needed.
4
Thus, by relatively simple calculations the brachytherapist now knows the position and length of each wire which, when implanted, will produce a reference isodose envelope that exactly covers the intended target volume.
3 2 1 0
0
5
10 15 20 Wire separation (mm)
25
30
Figure 49.13 (a) Diameter of 200 per cent isodose around central wires (four-wire, single-plane implant); (b) volume enclosed by 170 per cent for four wires 5 cm long at different separations.
Table 49.3 Minimum and maximum source spacing Source spacing Source length (cm) Short (1–4) Long (101)
Minimum (cm)
Maximum (cm)
0.8 1.5
1.5 2.2
3. Determination of the number of planes to be used. Using the information in Tables 49.2 and 49.3, it is clear that, with a maximum safe separation of source wires of 20 mm, the maximum thickness of tissue that
Choice of implant technique The tumour site and size will determine which technique is used. MINIATURE PLASTIC-TUBE TECHNIQUE
The iridium wire is loaded inside plastic tubes of external diameter 0.8 mm. This technique is admirably suited to superficial tumours, particularly where the surface is irregular and there are advantages to allowing the sources to conform to the shape of the tissues, e.g. basal cell carcinoma of the nose. HYPODERMIC NEEDLE TECHNIQUE
Simple hypodermic needles or spinal needles of diameter 0.8 mm are used as the vector. Bare iridium wire may be slid inside the needles, which then need to be blocked at both ends. Small tumours in soft parts of the body are particularly suitable, e.g. lip, ear and penile tumours. By using rigid needles the geometry of the implant is maintained.
1218 Interstitial brachytherapy
Ir wire
0.8 mm
0.55 mm
0.3 mm
12 16 mm mm
Figure 49.14 Diagram of outer and inner plastic tubes. Adapted from Pierquin et al. (1987).
GUIDE-NEEDLE TECHNIQUE
Needles of internal diameter 1.2 mm are used, left in the tissue and usually connected to templates to maintain the geometry. This is suitable for large tumour volumes in soft tissue, e.g. breast, anal canal. PLASTIC-TUBE TECHNIQUE
This is the most suitable technique for larger tumours or intra-operative brachytherapy. The vector plastic tube has an external diameter of 1.6 mm; through this is afterloaded the smaller plastic tube of 0.8 mm external diameter, which is preloaded with iridium wire (Fig. 49.14). PLASTIC-TUBE LOOPS
The same system can be formed into loops. This has the effect of crossing one end of the implant and is particularly useful in positions where there is limited access or where it is impossible to exit the wire internally (with its necessary extra length to allow for the ‘uncrossed’ end), e.g. base of tongue, floor of mouth. GUIDE GUTTER TECHNIQUE
These metal guides are commonly used for implanting smaller tumours of the tongue or tonsil. They carry wire pins, either single or double ‘hairpins’. It is normal to remove the metal guide leaving bare iridium wire in the tissues. The hairpins have a fixed separation between their legs of 12 mm. This means that they can only be used to treat target volumes of 18 mm thickness or less. Iridium wire for hairpins and single pins is manufactured in diameters of 0.5 or 0.6 mm so is stiffer than the usual wire, which is 0.3 mm external diameter.
Ordering the wire The following steps need to be done before ordering the iridium wire: 1. Construction on graph paper of the central plane of the proposed implant.
2. Measurement of the distance between each wire and each basal dose rate point. 3. Determination of the dose rate contribution from each wire to each dose rate point. This is done by using tables or graphs of the absorbed dose rate as a function of distance, assuming wire of reference linear activity (Table 49.4). 4. Calculation of the average basal dose rate. 5. Calculation of reference dose rate by multiplying by 0.85. 6. Allowance for filtration of source vectors, e.g. metal needles if left in situ may reduce the dose rate by approximately 2 per cent. 7. Calculation of activity of wire at mid-implant by the fraction: required activity
required reference dose rate ⎛ reference dose rate for ⎟⎞ ⎜⎜ ⎟⎟ ⎜⎝ unit wire activity ⎠
8. Adjustment of mid-implant wire activity to activity on the day that the iridium is calibrated and despatched by the manufacturers. This is necessary because iridium-192 decays with half-life of 74 days. Mid-implant activity should be increased by approximately 1 per cent/day for every day between the date the manufacturer specifies the activity and the mid-implant date. 9. Allowance for the fact that 10–20 per cent of the wire may be wasted during the process of cutting and loading into the inner plastic tubes. 10. Order for total required length of iridium wire of activity that corresponds most closely to that required.
PREPARATION OF IMPLANT PATIENTS AND MANAGEMENT OF COMPLICATIONS General preparation General preparation is both psychological and physical. Implant techniques are not commonly understood by the general public or even by non-specialist colleagues, so a full
Preparation of implant patients and management of complications 1219
Table 49.4 Absorbed dose rate as a function of the distance to 192Ir wires of different lengths and 0.3 mm in diameter (from Pierquin et al. [1987] with permission) Distance to the source (cm) Source length (cm) 1 2 3 4 5 6 7 8 10 12 14
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
13.1 14.8 15.3 15.5 15.6 15.7 15.8 15.8 15.8 15.9 16.0
7.63 9.30 9.86 10.1 10.3 10.3 10.4 10.4 10.5 10.5 10.6
5.00 6.55 7.13 7.40 7.55 7.65 7.71 7.76 7.82 7.86 7.88
3.52 4.90 5.48 5.76 5.92 6.02 6.09 6.14 6.21 6.25 6.28
2.60 3.82 4.37 4.66 4.83 4.94 5.01 5.06 5.13 5.17 5.20
1.99 3.06 3.59 3.87 4.04 4.16 4.23 4.29 4.36 4.41 4.44
1.57 2.50 3.00 3.28 3.45 3.57 3.65 3.70 3.78 3.83 3.86
1.27 2.09 2.55 2.82 2.99 3.11 3.19 3.25 3.33 3.38 3.41
1.05 1.76 2.19 2.46 2.63 2.74 2.82 2.88 2.96 3.01 3.05
0.875 1.51 1.91 2.16 2.33 2.44 2.52 2.58 2.67 2.72 2.75
0.743 1.30 1.67 1.91 2.08 2.19 2.27 2.33 2.42 2.47 2.50
Distance to the source (cm) Source length (cm) 1 2 3 4 5 6 7 8 10 12 14
1
2
3
4
5
6
7
8
10
12
15
1.05 1.76 2.19 2.46 2.63 2.74 2.82 2.88 2.96 3.01 3.05
0.278 0.523 0.725 0.882 1.003 1.096 1.170 1.228 1.312 1.368 1.408
0.125 0.242 0.348 0.441 0.520 0.587 0.642 0.689 0.762 0.814 0.853
0.070 0.138 0.202 0.260 0.313 0.360 0.401 0.437 0.496 0.542 0.577
0.045 0.088 0.130 0.170 0.206 0.240 0.271 0.299 0.347 0.385 0.415
0.031 0.061 0.090 0.118 0.145 0.170 0.194 0.215 0.253 0.285 0.311
0.022 0.044 0.066 0.087 0.107 0.126 0.144 0.161 0.191 0.218 0.240
0.017 0.033 0.050 0.066 0.081 0.096 0.110 0.124 0.148 0.170 0.189
0.010 0.020 0.030 0.040 0.049 0.059 0.068 0.076 0.093 0.108 0.122
0.007 0.013 0.020 0.026 0.033 0.039 0.045 0.051 0.063 0.074 0.084
0.004 0.008 0.012 0.016 0.020 0.023 0.027 0.031 0.038 0.045 0.052
Absorbed dose rate in cGy/h in water at the mid-plane of iridium-192 wires 0.3 mm in diameter with a reference linear kerma rate in air equal to 1 mGy/h/m2/cm, taking into account oblique filtration of the g radiations by the platinum sheath and the attenuation and diffusion of the g in the water.
explanation of the process must be given to each patient. Alternative methods of treatment, where such exist, should be explained. Patients are often frightened about being isolated in a hospital room, sometimes without windows, where they cannot have visits from family or friends. Some are not prepared to give up smoking even for the short period of hospitalization. Dietary factors are important. Implants in the oral cavity prevent normal eating and the patient should be allowed to get accustomed to a liquid or semi-solid diet before admission. If the anal canal is the site of brachytherapy, then a low-residue diet needs to be started 5–6 days before the procedure is carried out. In general, skin care is uncomplicated and no special precautions are necessary. Any patient with valvular disease of the heart requires suitable antibiotic cover to prevent the risk of subacute bacterial endocarditis. Patients undergoing implant for tumours of the rectum or anal canal need full bowel preparation and oral constipants, e.g. codeine phosphate 60 mg three times daily, as they will be unable to defecate for 3–4 days.
Preoperative Preoperative investigations should include chest X-ray, electrocardiogram, full blood count and clotting screen.
Premedication Atropine (or scopolamine) is very helpful in preventing excessive viscous salivation, which can hinder an oral cavity implantation.
Operation At the time of the operation skin cleaning with aqueous chlorhexidine is sufficient. Parenteral analgesia should be commenced towards the end of the anaesthetic and prescribed routinely for the first 12 hours or so postoperatively.
1220 Interstitial brachytherapy
Postoperative In the postoperative period adequate analgesia must be maintained. Reactive oedema may occur and is treatable by dexamethasone or non-steroidal anti-inflammatory drugs. Oral hygiene of a high standard needs to be maintained; patients with implants in the oral cavity may be unable to brush their teeth for 4–5 weeks overall and will need to use a mild antiseptic mouthwash several times daily. The hospital dietician is a very important member of the team. Careful attention to the individual patient’s dietary needs is essential for their general comfort and well-being. Normal hospital food is often totally unsatisfactory for these patients.
Complications Complications can be categorized as acute and delayed. ACUTE
Haemorrhage Haemorrhage due to puncturing a vessel during the implant procedure causes concern but is seldom dangerous. The needle should be withdrawn and pressure applied. Venous bleeds stop within 1–2 min but pressure on an artery should be maintained for a full 5 min. Oedema Reactive oedema takes some hours to develop and may last for the majority of the implant duration. Sources may become moved apart as the tissues swell up, significantly altering the dosimetry. If this is not checked and allowed for, the tumour may be underdosed. It is often advantageous to repeat check X-rays 24 hours after implantation. Infection Infection is remarkable by its absence during the acute phase. However, an acute radiation reaction may develop surface infection. Systemic antibiotics are seldom, if ever, required. DELAYED
Fibrosis One of the characteristics of an implant treatment is the relative lack of post-radiation fibrosis compared to external-beam radiotherapy. It is minimized by careful dosimetry, selection of the appropriate dose and dose rate, and by accurately conforming the high-dose volume to the tumourbearing tissue. Delayed haemorrhage Delayed haemorrhage due to necrosis of a vessel wall is a rare complication. When it occurs the usual cause is recurrent
tumour in association with the poor vascularity secondary to radiation-induced small-vessel damage. Immediate surgical management is required. Infection Infection is only a significant problem if it accompanies a mucosal skin or bone necrosis. Under these circumstances adequate doses of systemic antibiotics are essential, both to control pain and also to set the conditions under which the necrosis may heal. Necrosis A degree of normal tissue necrosis commonly occurs. If this is in the centre of an area of breast tissue, e.g., it is usually asymptomatic and undetectable. A mucosal necrosis in the oral cavity may be acutely painful, requiring some strong analgesics as well as measures to promote healing. All mucosal necroses of 1 cm or less have a good chance of healing with conservative management. The important measures are to abolish surface infection, prevent the patient from smoking and prevent the patient from drinking wines or spirits. Necroses of more than 1 cm seldom heal and should be referred for surgical management. Bone necrosis of the mandible is a potential complication of implants to the floor of the mouth or alveolus. The risk is directly proportional to the number of wires placed within 2–3 mm of the periosteum, so careful attention to the exact placing of sources is critically important. Bone necrosis is a serious complication, which causes the patient several months of severe pain. For this reason, although small tongue tumours are optimally treated by an implant, equally small and curable floor of mouth tumours may be better dealt with by primary surgical excision. This is particularly true with the modern techniques of microvascular anastamosis that have so improved surgery to this area. The other factor that may predispose to bone necrosis, is poor dental hygiene. Dental care and prevention of necrosis Prior to brachytherapy patients must be examined by an experienced dentist or stomatologist to determine the risk of developing bone exposure or full-blown osteoradionecrosis. The areas of highest risk are points of underlying bone having sharp bony exostoses or a sharp ridge covered by a thin layer of mucosa, such as the mylohyoid ridge. Pre-treatment evaluation includes clinical and radiological assessment. A dental treatment programme should include considerations of the patient’s socio-economic situation, general health status, degree of motivation and oral hygiene. Patients who have all or most of their natural teeth present in reasonably good condition should not normally be considered for extractions. They should be motivated to raise their level of oral hygiene to prevent any further loss of teeth. In patients with teeth in poor condition or with many missing teeth several factors must be evaluated. A dentition in poor periodontal condition presents a greater risk than one with many carious lesions but with an acceptable periodontal
Dose and dose-rate considerations 1221
state. Lower molars in the radiation field present a high-risk area. Any infection located in the bifurcation area is particularly difficult to eradicate. The factors to be considered in decision making are: 1. the general condition of the mouth and teeth as well as the specific condition of each individual tooth, with particular attention to teeth that will be in the radiation field; 2. the patient’s level of oral hygiene and the degree of compliance to oral hygiene measures that is to be expected; 3. mandibular teeth should be scrutinized more closely than maxillary teeth; 4. factors relating to the tumour, such as its rate of growth, the prognosis, the size of the radiation field, and the dosage and the mode of application of the radiation. If the tumour is rapidly growing, delay for dental treatment may not be in the patient’s best interest. The technical factors in the treatment determine the severity of the changes in the oral environment. Preventative protocol If the teeth are to be maintained, a preventative protocol is started. The patient must be educated and trained to maintain excellent oral hygiene. Daily application of fluoride gel to the teeth is mandatory in order to reduce the occurrence of ‘radiation caries’. The method of preference is a daily, 5-minute, self-application by the patient using a custom-made fluoride carrier. The fluoride treatment is started during the radiation therapy and is ideally continued for the rest of the patient’s life. When mucositis is at its worst during treatment the use of carriers and gel may be temporarily replaced by a fluoride mouth-rinsing solution. Special considerations in interstitial therapy Special custom-made protection devices may be made that may keep the dentition out of the radiation field, e.g. in the case of the lower lip implantation or a tongue lesion. These consist of individually made acrylic protectors embedded with lead foil. A 2 mm foil reduces the transmitted dose by 50 per cent, while the extra spacing will further reduce the dose considerably. Critical aspects in this management procedure are the vulnerable area of the mylohyoid ridge, which should be entirely covered, as well as making sure of the patient’s compliance in maintaining the device in place during the treatment time. It is advisable to prepare a second acrylic protector having an identical shape but without lead shielding.
DOSE AND DOSE-RATE CONSIDERATIONS Dose and dose rate Empirical experiments by Regaud and his team at the Foundation Curie during the early 1920s led to standardization in the filtration of radium sources and, compromising
between tumoricidal effects and normal tissue tolerance, an ‘optimal’ dose was established. This was described in terms of ‘millicuries equivalent destroyed’. With the establishment of the Manchester system this dose became described as 7000 roëntgens (approximately 60 Gy) in 7 days at the reference isodose. Excellent results were achieved using this prescription, but inevitably clinicians sought to modify and refine their experience, mainly in an effort to reduce the long overall treatment times which were difficult for patients to tolerate. In 1952 Paterson5 recommended a dose reduction for treatment times of less than 7 days (e.g. 46 Gy for a treatment time of 3 days) and a dose increase for treatment times of greater than 7 days (e.g. 62 Gy for 9 days). His model was used extensively for more than 20 years. Pierquin et al.,6 on the other hand, found no apparent effects on local control and necrosis rates when a dose of 70 Gy was given to oropharyngeal tumours with overall treatment times that varied between 3 and 8 days, (0.9–0.3 Gy/hour approximately). He advised against dose adjustment that might lead to underdosing of tumours. During the 1970s other authors studying a variety of tumour sites confirmed that it was difficult to demonstrate a significant effect of dose rate on tumour control or necrosis rates in the clinical setting.7,8 However, in recent years several reports have appeared that have demonstrated just such effects. Larger numbers of cases treated in the same hospital by closely controlled methods and the modern statistical tools of univariate and multivariate analyses may be responsible for bringing a greater degree of precision.9–11 For example, the study of Mazeron et al.11 on T1 and T2 squamous carcinomas of the anterior two thirds of the tongue and floor of the mouth was able to recommend a dose prescription that, in their hands, gave the optimum results as: ● ● ●
dose: 65 Gy to reference isodose (Paris system) dose rate: 0.4–0.5 Gy/h at reference isodose source spacing: 12–15 mm.
Quality measures To be able to compare one implant with another, internationally agreed rules for describing dose and volume need to be developed. Such a set of recommendations has been published by the International Commission on Radiation Units (ICRU). However, the presence of significant ‘hot spots’ or lack of equality in source spacing – to mention just two parameters – are quality measures that are difficult to standardize. Dose–volume histograms (DVHs) represent an attempt to describe implants in such a way as to enable further analysis and comparison. A dose–volume histogram which suppresses the inverse square law effects has been developed by Anderson12 and is called the ‘natural’ dose–volume histogram.
1222 Interstitial brachytherapy
3.0
76.3 cGy/hr 62.5 cGy/hr
15
100 cGy/hr
14 13
75.6 cGy/hr
100
12
2.0
cm3
cm3
11 10 9 8 1.0
7 6 5 4 3
27
30
34
40
48
63
100159
Dose rate (cGy/h)
2
27
30
34
40
48
63
100 159
Dose rate (cGy/h)
Figure 49.15 Two clinical examples of natural dose–volume histograms. (a) Two ‘hairpins’ in tongue; (b) three loops of wire in anal canal/rectum.
Figure 49.15 shows two such dose–volume histograms, which should demonstrate a clear peak. Greater dose uniformity is implied by higher narrower peaks. The prescribed dose rate is lower than the DVH maximum since, in this region, the volume treated by particular dose rates changes very rapidly.
Radiobiological aspects of brachytherapy Irradiation from a source or sources implanted within a tumour has a distinct geometrical advantage as it inevitably spares the surrounding normal tissues, which receive a lower dose. The non-uniformity of radiation dose and dose rate around an implant has important radiobiological consequences. Close to the sources the dose rate is high and the degree of cell killing will be close to that indicated by the acute radiation survival curve. As you move away, however, two changes will occur. First, cells will be less sensitive at the lower dose rates and, secondly, within the given period of implantation the accumulated dose will be less. Close to sources the level of cell killing is so high that cells of any radiosensitivity will be killed. Further out, the effects will be so slight that even the most radiosensitive cells will all survive. Between these two extremes there is a critical
zone where, over a very short distance, the probability of local cure will pass from nearly 100 per cent to virtually zero. Steel et al.13 described this change from high to low local cure probability as being like a cliff-face. The distance from the source to the cliff edge is determined by the radiosensitivity of the cells at low dose rate, being further for radiosensitive cells and nearer for those more radioresistant. Systems of implant rules such as the Manchester or Paris systems must therefore function by enabling the brachytherapist to place his sources in such a way that these ‘cliffs’ lie beyond the edges of the tumour. Apart from geometrical optimization of dose, the other clear advantage for brachytherapy in radiobiological terms is that the overall treatment time is very short – a radical dose is given over just a few days. Tumour cell repopulation is therefore minimized. This will confer a therapeutic advantage for the treatment of the more rapidly repopulating tumour types, such as squamous tumours of the oropharynx.
DOSIMETRY IN VIVO The dose distribution of a completed implant can be determined from knowing the source positions, geometry and source strengths.
Dosimetry in vivo 1223
Localization of source positions Source positions may be determined by several different methods: 1. Direct measurement from the implanted wires or vector material. 2. Tomography (including CT scanning) through the central plane of the implant. 3. Orthogonal radiographs. 4. Isocentric radiographs, i.e. two radiographs separated by an angle not equal to 90 degrees. Direct measurement is suitable only for wires in a superficial location, offering easy access for accurate measurement. The wires must also be straight and parallel so that the measurements in one plane are representative of the wire separation along their whole length. Tomography is most suitable for wires, which are known to be straight and parallel, since information is obtained in the central plane only. Care must be taken to ensure that the tomographic slice is taken perpendicular to the wires, since if it is taken at an angle, the wire separations appear to be larger than their true separations. Angles of up to 10 degrees from the perpendicular will not introduce a serious error. Pairs of radiographs, whether orthogonal or not, can be used to provide full three-dimensional coordinate information of the wires along their full length. Orthogonal and isocentric radiographs must be used if the wires are not parallel or are curved. Whenever pairs of films are used for dosimetry purposes, a few simple rules must be followed to achieve accurate results: 1. The patient must not move or be moved between radiographs. 2. A magnification marker must be placed close to the implants so that the radiograph magnification at the position of the implant can be determined on each film. If the radiographs are taken on a radiotherapy simulator, the implant can be set at the isocentre of the machine and the field-defining wires are used to determine the magnification. 3. If possible, the patient should be radiographed in a position which is likely to be representative of his or her position throughout the treatment time. 4. When using an afterloading technique it is often helpful to load the vectors with dummy sources of different design. This will assist with the identification of individual wires on the two views. For example, one dummy source may consist of 10 mm of lead wire alternating with 10 mm of plastic tube, and another may be 20 mm of lead wire alternating with 10 mm of plastic tube. If there is difficulty in identification of the individual wires on the two radiographs it may be helpful to adopt the following technique on a radiotherapy simulator. The patient is screened while rotating the simulator from one required view to the other. Each person present is allocated
one wire on the initial image and observes it carefully as the simulator is rotated. By observing only one wire as the simulator rotates the observers are able to identify the position of ‘their’ wire on the final image.
Manual reconstruction of the central plane of the implant from orthogonal radiographs A three-dimensional object can be represented by the coordinates x, y and z measured along orthogonal axes. One of the orthogonal pair of radiographs can be considered to provide the x and y coordinates, and the second provides the z and y coordinates. Work will be facilitated if the films are placed side by side with the common y axes parallel. Before starting the reconstruction, each wire must be clearly identifiable from the others. Then, on one view, a line is drawn through the approximate centre of the implant perpendicular to the average wire direction. This line represents the central plane. The points at which the central plane intersects the wires can then be marked on view 2. This may most easily be done by calculating the y coordinate of the intersection point on view 1 and then marking a point with the calculated y coordinate on the same wire on view 2 (Fig. 49.16). This is repeated for all the wires. It is generally most convenient to use one end of the wire marked as the coordinate origin when carrying out this procedure. If the films have different magnifications, this must be taken into account when calculating the y coordinates. A reconstruction of the central plane may then be carried out by measuring the distances to the wires on view 1 along the marked plane from an arbitrary origin, and marking these distances along one axis of a piece of graph paper. It is usually convenient to use one of the wire positions as the origin. On view 2 the marked intersections should then be projected down onto the z axis. (This will be a line parallel to the patient couch). The distances between the projected marks should then be transferred to the second axis of the graph paper using the same origin as for view 1. Using these points, the positions of each wire may then be marked. A plot is thus obtained of the marked central plane, which is approximately at right angles to the average wire direction. The described method may be inaccurate if the wires on at least one view are not approximately parallel to the common axis designated y.
Calculation of dose rate The dose distribution around a particular source is a function of the source geometry, i.e. its diameter construction material and length. Data of dose rate against perpendicular distance from the wire for iridium-192 wires of different lengths are available in tabular form4 and graphical form14 for unit source strength. To calculate the dose rate at a particular point in the central plane, the distance from each wire to the dose point is measured on the cross-section of the implant. The data
1224 Interstitial brachytherapy
Y
Y
c
b
b a
c
X
a c
b
a View 1
Figure 49.16
View 2
X
Z
Z Cross-section through central plane
Manual reconstruction of the central plane of an implant from orthogonal radiographs.
appropriate for each length of wire are then selected. From the table or graph for the required wire length the measured distances are used to read off the dose rate that a wire of unit source strength will give at that distance. This process is repeated for all the wires in the implant and the contributions from each wire are summed to find the total dose rate at the point required for unit source strength wire. This calculation is best presented in the form of a table. The dose rate is finally modified to allow for the source strength actually used. Iridium-192 has a half-life of 74 days and decays by just under 1 per cent/day. To obtain a dose rate that is representative of the implant throughout its entire insertion time, it is necessary to use the mid-implant activity: Dose rate dose rate for unit source strength mid implant source strength To calculate the treatment time in the Paris system the dose rate at each of the basal dose rate points is calculated as described above. The average basal dose rate is calculated and the reference dose rate which is 85 per cent of the average basal dose rate is found. The treatment time is then calculated by dividing the prescribed dose by the true reference dose rate. When the calculation has been completed it must be independently checked by a second qualified person. Particular attention must be paid to the accuracy of the reconstruction since small changes in wire position result in large differences in dose rate to the patient.
SUMMARY Basal dose basal dose rate for unit source strength rate (BDR ) mid implant source strength Average basal dose rate
(BDR1 BDR 2 L BDRn ) n
Reference dose rate 0.85 average basal dose rate Treatment time
prescribed dose reference dose rate
EXAMPLE Four wires are implanted, each separated by 12 mm. The most lateral wire is 50 mm long and the others are 70 mm in length. The wire used has a source strength per mm of 587 nGy/h per mm at 1 m on 15 April. The implant will deliver 60 Gy to the reference isodose and the wires will be inserted on 19 April. How long must it be left in place? The dose rates for unit source strength are given in Table 49.5. Actual source strength 15 April 587 nGy/h/mm at 1 m Unit source strength 1000 nGy/h per mm at 1 m Average basal dose rate (1.165 1.225 1.173)/3 1.188 Gy/h Expected mid-implant date 22 April 1 week decay factor 0.937 Basal dose rate corrected for activity at mid-implant 1.189 0.937 587/1000 0.65 Gy/h Reference dose rate 0.85 0.65 0.56 Gy/h Time for 60 Gy 60/0.56 107 h 4 days 11 hours
Source strength specification The recommended quantity for source strength specifications is air kerma rate at 1 m/mm; the units are mGy/h per mm at 1 m. There are many other ways in which source strength has been specified in the past, e.g. milligrams radium equivalent, millicuries, exposure rate at 1 m/mm. To convert from one quantity to another requires the use of a conversion factor, and care must be taken in their application as not all the published values of these factors are in agreement. Ideally one would specify source strengths only in air kerma rate to prevent any possible confusion but, unfortunately, some computer programs do not have the facility to accept source strengths in this way, so it is inevitable that a variety of units will continue to be used for some time.
Clinical indications 1225
Table 49.5 Dose rates for unit source strength Wire number and length 1 (70 mm) 2 (70 mm) 3 (70 mm) 4 (50 mm) Total
Distance to DB1 (mm)
Dose rate to DB1 (Gy/h)
Distance to DB2 (mm)
Dose rate to DB2 (Gy/h)
Distance to DB3 (mm)
Dose Rate to DB3 (Gy/h)
6 6 18 30
0.497 0.497 0.120 0.051 1.165
18 6 6 18
0.120 0.497 0.497 0.111 1.225
30 18 6 6
0.063 0.12 0.497 0.493 1.173
Use of computers Calculations of the dose distribution are frequently carried out by computer. The computed dose distributions are usually obtained by regarding the iridium wires as a very large number of point sources. As with any computer dosimetry system, extensive checking should be carried out by the user at the commissioning of the system before it is used for the calculation of any patient treatment times. The simplest way of carrying out these checks may be to compare the computer results with the results of the hand calculations made by methods similar to those described earlier. Even when computers are used for patient dosimetry, an entirely independent check must still be made of the reconstruction of the implant and the other data entered by the operator, to ensure that no error has been made.
Checking of the wire prior to use It is the user’s responsibility to ensure that the wire supplied is of the correct activity and satisfactory uniformity. To achieve this it is necessary not only to measure the total source strength of the delivered source but also to check that the expected length of wire has been received and to make some other check of the source strength per unit length. It is recommended that the following checks are carried out: 1. Compare the order, delivery note and measurement certificate to check that they agree. 2. Measure the source strength in a radionuclide calibrator with a calibration traceable to the national standard. Allowing for radioactive decay. This measurement should agree with the manufacturer’s measurement prior to despatch. 3. Make a check to ensure that not only is the total activity of the source correct; but that the source strength per unit length is as expected. This can be done by: (a) measuring the strength of individual sources cut from the wire to see that the total strength of the individual wires is in agreement with that expected from the source strength per unit length; or (b) weighing the wires after the total source strength has been measured. The weight of the wire is accurately proportional to its length and so this measurement will confirm that the expected length of wire has been
delivered. This may not detect whether a small length of the active core is missing, since the radioactive core of the 0.3 mm diameter wire contributes only 10 per cent of the weight of the wire; or (c) autoradiograph the entire length of the newly delivered source. This is likely to be possible for hairpins and loops but is difficult to achieve for long, coiled wires.
CLINICAL INDICATIONS The principal reasons for choosing interstitial brachytherapy in preference to external-beam treatment relate to dose delivery and dose distribution. Suitable sites clearly need to be easily accessible and the tumour-bearing normal tissue needs to be one that tolerates radiation well. Even in the era of today’s sophisticated computer planning and modern multi-leaved linear accelerators, a well executed implant will achieve the most conformal and intensity modulated radiation dose treatment that is possible. The following is a brief and by no means exhaustive list of tumour sites that are particularly well suited to treatment by interstitial brachytherapy. For more detailed description of techniques and results the reader should consult a textbook on brachytherapy.4,15
Cancer of the tongue Squamous carcinoma usually occurs on the lateral borders and ventral surfaces. Tumours of 3 cm maximum diameter or less are suitable for implanting, with a very high chance of local cure. The ‘thickness’ dimension of the target volume is its lateral dimension. Target volumes of 18 mm ‘thickness’ or less can be implanted using hairpins: thicker lesions need the wider separation that can be obtained with loops of iridium wire in plastic tubes. When this technique is used, the plastic tubes enter and exit in the submental area, with the apex of the loop arching over the dorsum of the tongue. Spacing between the legs of the loop of more than 15 mm is not recommended, which means in effect that target volumes ‘thicker’ than 24 mm are not ideally suited for radical treatment by interstitial implantation. They may, however, be boosted by brachytherapy after some external irradiation.
1226 Interstitial brachytherapy
There is a general view, however, that although this combination of external radiotherapy with a brachytherapy boost probably does reduce the risk of necrosis, it also reduces the chance of local control. Pernot et al.16 have shown that the local control rate of T2N0 cases treated by implant alone (70 cases) was 89 per cent at 5 years, whereas 72 cases of T2N0 tumours treated by external radiotherapy with iridium-192 boost was only 50 per cent. Published local control rates for T1N0 tumours are in the range of 95–100 per cent at 5 years for patients treated by initial iridium implant and surgery for salvage.17,18 These excellent figures, coupled with the superb functional result seen after brachytherapy, suggest strongly that the optimum management for smaller squamous carcinomas of the tongue is interstitial implant to dose of 65 Gy followed by surgery to salvage the few recurrences that occur. Radiotherapists not practised in performing good implants should refer such cases on to a colleague who subspecializes in brachytherapy. Treatment by external radiotherapy alone is associated with poorer results. In areas where implantation techniques are not available, patients may be better managed by primary surgical excision and reconstruction, followed by external irradiation in the postoperative period if appropriate, even though the functional result is less satisfactory. All patients with T2N0 tumours should be considered for prophylactic treatment to the upper neck nodes. If the T2 group is subdivided into those 3 cm diameter and those 3 cm, a significant difference in local control and locoregional control emerges in the majority of published series. Although these differences may be partly due to tumour biological factors, many brachytherapists believe that the higher doses achievable by radical dose implant treatment of the ‘smaller’ tumours is largely responsible.
Cancer of the floor of the mouth Similar indications and techniques apply for floor of the mouth tumours as for tongue lesions. The floor of the mouth is a horseshoe shape but narrower posteriorly than anteriorly. This leads to some difficulties in placing the hairpins or loops of iridium-192 parallel. The one significant problem is the proximity of the mandible. When a tumour is fixed to and/or involving bone, surgery is recommended. In cases where more than three iridium wires lie close (3 mm or nearer) to the mandible there is a significant risk of causing an area of osteoradionecrosis, and surgical treatment may well be the better option. Small tumours, however, are very curable, with local control rates of 92 per cent when the lesion is small enough to be treated by a radical dose implant. For larger tumours, where a combination of external-beam radiation and implant was used, the control rates fell to 55 per cent.19
Cancer of the lip The lip tolerates low dose-rate irradiation extremely well and the functional and cosmetic results of brachytherapy
are usually excellent. The implant is done transversely and it may be necessary to leave one of the radioactive wires ‘in air’ in order to pull the treatment isodose upwards to cover an exophytic tumour. This is achieved by using rigid afterloading needles held in place at each end by small perspex templates to maintain the correct spacing. The teeth and gingiva may be protected by putting some distance between them and the source wires. Simply inserting a roll of cotton wool in the anterior gingival sulcus achieves the desired effect. Results of treatment show local control rates of 95 per cent.20 Locoregional control rates are similar because lymph-node involvement is rare in this tumour.
Skin cancers Skin cancers in some sites are well adapted to treatment by brachytherapy techniques. Tumours that site over the bridge of the nose, like a saddle on a horse’s back, and those on the free edge of the pinna are two good examples as they are difficult to treat by external irradiation. Furthermore there appears to be no significant risk of cartilagenous necrosis with low dose-rate radiation. The miniature plastic tube technique or hypodermic needle technique is used. Basal cell carcinomas respond well to 60 Gy to the reference isodose, but 65 Gy is prefered for squamous cell lesions.
Cancers of the anal canal and lower rectum The use of brachytherapy to boost the dose to the anal canal or lower rectum after chemoradiation is associated with high rates of local control (and thus low rates of abdominoperineal resection for salvage). After a dose of 45 Gy external irradiation (25 treatments over 5 weeks) the patient is allowed a significantly long rest period of 6 weeks before being given the implant. Afterloading rigid needles are implanted along the axis of the anal canal at least 3–4 mm deep to the mucosal surface and are held in place by a perspex template, which is sutured to the perineum. Not more than 50 per cent of the circumference of the anal canal should be implanted, to reduce the risk of post-radiation stricture formation. It should be remembered that the ends of these implants are ‘uncrossed’ and the active wire must be implanted far enough up into the rectal wall to cover the target volume. A dose of 20–25 Gy is prescribed.
Cancer of the penis Smaller tumours of the glans penis can be treated by a volume implant with local control rates of 90 per cent.21 After adequate prior circumcision, the implant is performed using rigid afterloading needles held in place by two perspex templates placed dorsally and ventrally. Care must be taken not to insert any needle through the urethra (patient catheterized). The skin and mucosal reaction that occurs after 65 Gy is sometimes prolonged. Patients must be aware of this beforehand and should be well motivated to tolerate it. Conservation of the penis is achieved in more than 80 per
Prostate brachytherapy 1227
cent of cases. The majority of sexually active patients retain their potency.
Cancer of the breast The breast tolerates low dose-rate radiation well and up to one third of its volume may be implanted safely. Indications vary. Commonly, a boost dose to the site of an excised tumour is given after external irradiation to the whole breast. Alternatively, brachytherapy may be used in patients who will not accept any surgical excision or as salvage therapy for local recurrence. Its main advantage is the ability to deliver high doses without causing unsightly skin changes. This gives the combination of good cosmetic appearance and high rate of tumour control. It must be said, however, that external radiation with an electron field boost gives good results for most patients who have undergone a complete surgical excision with accurate histological control of the margins. Implant techniques should, however, be considered for those cases where the radiotherapist feels that a higher dose is required, for example, in cases where there has been no tumour excision or incomplete tumour excision, in high-grade tumours and in young patients with a high risk of local recurrence who will not accept a mastectomy. Rigid afterloading needles are used, together with predrilled perspex templates to maintain the separation of the wires. These volume implants can be large, with two or sometimes three planes and commonly 7–9 radioactive lines. After 45–50 Gy external irradiation a boost dose of 25–30 Gy to the reference isodose gives local control rates of 97 per cent for T1 tumours at 15 years.22,23 Clinical trials of partial breast irradiation by perioperative or intra-operative brachytherapy techniques are currently in progress and the results are awaited. If successful, these techniques may allow some patients to avoid the need for a course of post-operative whole breast external beam radiotherapy.
PROSTATE BRACHYTHERAPY Introduction Prostate brachytherapy is the insertion of radioactive sources directly into the prostate. It was first performed in 1932 using radium sources.24 Prostate brachytherapy was rarely used until the development of transrectal ultrasound, which gave rise to increasingly performed prostate brachytherapy in the late 1980s.25 Transrectal ultrasound facilitated the accurate placement of seeds, which reduces the chances of underdosage to some parts of the prostate and excessive dose to other areas. Prostate brachytherapy can be subdivided into a permanent radioactive source implant, e.g. using iodine-125 or palladium-103 seeds, and temporary radioactive source implantation using hollow catheters or needles in which radioactive sources, usually iridium, are placed in the prostate for a short period of time.
Figure 49.17 Pre-treatment measurement of prostate volume in Blasko and Ragde’s trans-perineal permanent radioactive seed implantation of the prostate.
The potential advantage of brachytherapy compared to external radiotherapy of the prostate is that the dose can be prescribed with much closer margins than external radiotherapy, thus reducing the volume of rectum and bladder that receives the prescribed dose. This permits dose intensification, which increases the likelihood of eradication of prostate cancer.26 Furthermore the risk of geographical miss because of internal prostate movement during external beam radiotherapy is eliminated using brachytherapy. Another advantage in the case of permanent radioactive seed implants is that this can be given in one hospital visit compared to multiple visits that are required with external beam radiotherapy.
Permanent radioactive source implant Blasko and Ragde in Seattle in the late 1980s popularized transperineal permanent radioactive seed implantation of the prostate. Their technique required a pre-treatment measurement of the prostate volume (Fig. 49.17) so that the optimum positioning of seeds could be ascertained prior to the actual insertion of the radioactive seeds (Figs 49.18 and 49.19). Stock and Stone in New York and Batterman in Rotterdam in Holland developed a technique where needles are placed into the prostate; the positions of the needle are then verified in real time prior to the seeds being loaded. Both techniques have potential advantages and disadvantagesbutwhenperformedwellbothresultinaccurateradioactive seed placement so that an appropriate dosage of radiotherapy is received by the prostate gland (Figs 49.20 and49.21).
Permanent radioactive seed prostate implantation planning The aim in planning is to position the seeds in or near to the prostate so that the periphery of the prostate receives a
1228 Interstitial brachytherapy
Figure 49.21 Figure 49.18 Prostate mid gland showing isodose map. The whole prostate receives a minimum dose of 145 grays.
Post Implant: CT/MRI
Figure 49.22 Figure 49.19 Taking the Z-axis measurement to determine how deeply each needle should be inserted.
Technique for 125seed implantation.
Post seed loading CT/MRI.
Powerful computer software is required to calculate the dosimetry resulting from the numerous radioactive sources that are to be placed. The number of sources that are used depends on the activity of the source and the size of the prostate gland (conventionally the whole prostate gland is treated and not just the tumour-bearing part of the prostate). Where higher activity seeds are used fewer seeds are required to achieve the desired dosage, which is less expensive. The disadvantage of this however, is that since fewer radioactive seeds are used slight inaccuracies in positioning of individual seeds will have a more significant impact on the dosimetry as compared to the situation where a high number of lower activity seeds are used. In practice most centres compromise and use medium activity seeds, e.g. 0.36 mCi.
Temporary radioactive source implantation Figure 49.20 Insertion of needle to correct position in gland for delivery of radioactive seeds.
minimum dosage of radiation (with iodine-125 145 Gy and with palladium-103 120 Gy). The preferred seed placement arrangement is a modified peripheral loading technique where the majority of seeds are placed at or near the periphery of the prostate (Fig. 49.22). This avoids an excessive dose to the prostatic urethra.
This is almost always given at high dose rate (HDR) and is usually given as a fractionated course and usually in combination with external beam radiotherapy. There are a number of disadvantages compared to a permanent seed implantation. For example once the hollow needles or catheters are inserted into the prostate they may need to remain in for many hours so that a second fraction of HDR brachytherapy can be given. Secondly, patients may have to return for a second or third procedure depending on the fractionation
Prostate brachytherapy 1229
schedule. Furthermore it is possible for the needle or catheter position to slightly change between fractions, although usually this can be easily corrected manually. A significant advantage over permanent seed implantation is that the brachytherapy plan can be optimized by adjusting the dwell time and position of the radioactive sources within the catheters or needles to optimize the radiotherapy plan; therefore even if the needle or catheter positions are not absolutely correct, optimization may allow complete coverage of the prostate by the prescribed dose and avoidance of excess dosage to normal tissue. A further advantage over permanent seed implantation is that catheters and needles hold their position in extraprostatic tissue better than radioactive seeds, and thus should achieve an adequate radiotherapy dose for microscopic extracapsular spread of cancer more reliably. Theoretically high-dose brachytherapy may have radiobiological advantages following the recent suggestion from several sources that prostate cancers have a low alpha–beta ratio and therefore may be more sensitive to larger dose per fraction radiotherapy as occurs with HDR brachytherapy.27 Several centres have reported excellent results in terms of biochemical control and low rates of major toxicity using a variety of schedules of external beam radiotherapy and HDR brachytherapy. The rest of the discussion will concentrate on the more widely practised permanent radioactive seed implantation.
Patient selection The American Brachytherapy Society (ABS)28 and the European Organisation for the Radiotherapeutic Treatment of Cancer (EORTC)29 have recommended that prostate brachytherapy using permanent radioactive seed placement is appropriate for patients with prostate-specific antigen (PSA) less than 10 and a Gleason score of 6 or less. There is no consensus regarding patients with higher risk prostate cancer. The results from several major centres suggest that the patients with at least one adverse factor, e.g. PSA between 10–20 with a Gleason sum score of 6, or Gleason score of 7 with a PSA less than 10 have a good outcome using permanent iodine-125 seed implantation. Patients with two adverse features, i.e. PSA greater than 10 and Gleason grade 7 or above have tended to have a relatively high risk of biochemical relapse using permanent radioactive seed implantation. In general it is not recommended as a sole treatment for this group of patients. Generally accepted contraindications to seed implantation are transurethral resection of prostate (TURP) within the last 6 months, significant prostatic urinary outflow obstruction which would benefit from TURP and a previous AP rectal resection. Relative contraindications include moderate obstructive urinary symptoms, an increased propensity to bleeding and diabetes because of its increased association with infection.28,29
Centres vary in terms of their cut off in terms of prostate size with almost all centres excluding patients with a prostate size greater than 60 ml and most having a cut off of 50 ml. The reasons that this restriction exists are: 1. because the pubic arch is more likely to prevent the access of needles into the anterior aspect of a prostate that is enlarged; 2. the treatment of larger prostates is associated with both an increased risk of urinary retention requiring urinary catheterization and of moderate to severe urinary symptoms in the short to medium term. Most large prostates can be reduced in size by approximately one third after 3 months of treatment with a lutenizing hormone releasing hormone (LHRH) agonist. The increased risk of urinary retention and moderate to severe urinary symptoms remains however.
Side effects of prostate brachytherapy URINARY SIDE EFFECTS
The commonest side effect of prostate brachytherapy is short- to medium-term obstructive and irritative urinary symptoms, e.g. slow urine flow, urgency, hesitancy and frequency of urination, which are experienced by the majority of patients for weeks to months. Symptoms are expected to return to pre-treatment levels by 12 months in most patients.30 Approximately 10 per cent of patients require some form of urinary catheterization for a period of a few weeks because of acute urinary retention. This usually takes place within the first 2 weeks of the procedure and is highly unlikely to occur more than 4 weeks after permanent seed implantation. Dysuria sometimes occurs and if this persists it raises the possibility of an underlying urinary tract infection. Most patients receive an alpha blocker, e.g. tamsulosin or alfuzosin. These medications can be continued for as long as they produce symptomatic benefit which, in some cases, can be many months. Analgesics, e.g. paracetamol and anti-inflammatory drugs are helpful in some patients. In addition most centres routinely prescribe prophylactic antibiotics at the time of the procedure. Significant long-term urinary complications are uncommon and include a risk of urethral stricture of approximately 4 per cent which can be treated by simple dilatation. Stress incontinence is uncommon unless TURP is required following prostate brachytherapy, where the risk of incontinence can be up to 50 per cent. The risk of incontinence can be reduced by avoidance of transurethral surgery in the first year after brachytherapy. If transurethral surgery remains necessary for chronic urinary retention then the extent of resection should be kept to a minimum, e.g. channel-type TURP.30
1230 Interstitial brachytherapy
BOWEL SIDE EFFECTS
Prostate brachytherapy is generally associated with a low level of problematic radiation proctitis (approximately 4 per cent). At proctoscopy proctitis can often be documented but if symptomatic should be treated conservatively, e.g. with the use of stool bulking agents and occasionally topical steroid or anti-inflammatory medication. Minimal rectal bleeding requires no specific treatment but if bleeding were significant then topical application of formalin can be considered.31 Biopsy of the area of suspected proctitis should be avoided as this can commonly lead to recto-prostatic fistula.
Biochemical relapse following all forms of radiotherapy has in general been defined by the ASTRO consensus definition of three consecutive PSA rises over a 6-month period. In the case of prostate brachytherapy this cannot be confirmed until a minimum of 2 years after treatment. The biochemical relapse-free survival at 5 years ranges from 75 per cent to 90 per cent for intermediate- to goodrisk patients in most reports, and these results persist beyond 10 years.35 No randomized comparative studies exist comparing prostate brachytherapy with the other standard treatments for good to moderate risk prostate cancer but in non-randomized reports prostate brachytherapy appears to be at least as effective.36
SEXUAL COMPLICATIONS
The ability to get erections after prostate brachytherapy is reduced. This can occur temporarily in the first year but loss of erectile strength continues annually following prostate brachytherapy. The risk of erectile dysfunction appears to be related to the quality of erections prior to treatment and the age of the patient. For a patient of 60–65 with good erections before treatment the risk of impaired erections 5 years later would range from 25–50 per cent.32,33 Fortunately the response rate to Sildenafil and related drugs is in the order of 60–80 per cent.32
CONCLUSION Prostate brachytherapy using permanent radioactive seed implantation is increasingly being used in North America and Western Europe for treatment of patients with goodto moderate-risk prostate cancer. The reasons behind this are due to its relative convenience for patient and clinician, its high cure rate and relatively low risk of long-term complications. Prostate brachytherapy may have a future role in dose escalation in conjunction with external beam radiotherapy for higher risk prostate cancers.
OTHER COMPLICATIONS
In the immediate postoperative period patients may feel tired. They commonly experience localized discomfort in the perineum at the site of needle insertion. In the medium to long term, ejaculation becomes dry and subsequently patients become infertile, although this usually occurs several months after brachytherapy. There is a theoretical risk of second malignancy which, to date, has not been quantified. PROSTATE CANCER OUTCOME FOLLOWING PROSTATE BRACHYTHERAPY
Since patients with relatively good risk prostate cancer are selected to have prostate brachytherapy as a sole treatment, the likelihood of death from prostate cancer in the first 10 years is very low. Therefore, the commonly applied measure of effective treatment with brachytherapy is biochemical failure (PSA failure). Biochemical failure following prostate brachytherapy cannot generally be confirmed until at least 2 years after treatment because of the occurrence of benign PSA rises that usually arise in the second year following prostate brachytherapy. This is an increase in PSA, up to as high as 10 ng/ml in exceptional circumstances, which with further follow up reduces and remains low permanently. These benign PSA rises occur in approximately 40 per cent of men and do not predict a worse prognosis.34 The benign PSA rises are managed simply by reassurance and observation of PSA.
KEY LEARNING POINTS ●
●
●
●
●
● ●
Brachytherapy is an essential technique for the clinical oncologist. Physical optimization of dose leads to high cure rates in suitable tumours. The clinician must understand the dosimetry system so that he can adapt each implant to the individual tumour being treated. Computers save time but cannot be used without the brachytherapist also having a good understanding of the dosimetry system being used. Good clinical indications for interstitial brachytherapy include tumours of the tongue, floor of the mouth, lip, breast, anal canal and prostate. Pre-planning the implant is essential. Above all, meticulous attention to detail is needed for best results.
KEY REFERENCES Pierquin B, Marinello G. Manuel practique de Curietherapy. Paris: Hermann, Editeurs des Sciences at des Arts, 1992.
References 1231
Pierquin B, Marinello G. A Practical Manual of Brachytherapy. Madison, WI: Medical Physics Publishing, 1997. Pierquin B, Wilson J-F, Chassagne D. Modern Brachytherapy. New York: Mason Publishing Inc., 1987. Steel GG. Basic Clinical Radiobiology. London: Edward Arnold, 1993.
REFERENCES 1 Wickham L, Degrais P. Radium Therapy. New York: Funk and Wagnalls, 1910. 2 Paterson R, Parker HM. A dosage system for gammaray therapy. Br J Radiol 1934; 7:592. 3 Pierquin B, Chassagne D, Perez R. Precis de Curietherapie. Paris: Masson and Co., 1964. 4 Pierquin B, Wilson JF, Chassagne D. Modern Brachytherapy. New York: Masson Publishing Inc., 1987. 5 Paterson R. Studies in optimum dosage. Br J Radiol 1952; 25:505–16. 6 Pierquin B, Chassagne D, Baillet F, Paine C. Clinical observations on the time factor in interstitial radiotherapy using 192-iridium. Clin Radiol 1973; 24:506–9. 7 Barklay HY, Fletcher G. Volume and time factors in interstitial gamma-ray therapy. Am J Roentgenol 1976; 126:163–70. 8 Larra F, Dixon B, Couette JE, et al. Facteur temps en curietherapie. J Radiol Electrol 1977; 58:329–33. 9 Fontanesi MD, Hetzler D, Ross J. Effects of dose rate on local control and complications in the re-irradiation of head and neck tumours with interstitial 192-iridium. Int J Radiat Oncol Biol Phys 1989; 17:365–9. 10 Mazeron JJ, Simon JM, Crook J, et al. Influence of dose rate on local control of breast carcinoma treated by external beam irradiation plus 192-iridium implant. Int J Radiat Oncol Biol Phys 1991a; 21:1173–7. 11 Mazeron JJ, Simon JM, Le Péchoux C, et al. Effect of dose rate on local control of T1–T2 squamous cell carcinomas of mobile tongue and floor of mouth with interstitial 192-iridium. Radiother Oncol 1991b; 21:39–47. 12 Anderson LL. A ‘natural’ dose–volume histogram for brachytherapy. Med Phys 1986; 13:898–903. 13 Steel GG, Kelland LR, Peacock JH. The radiobiological basis of low dose-rate radiotherapy. In: Mould RF (ed.) Brachytherapy 2, Proceedings of the 5th International Selectron Users’ Meeting 1988. Leersum, The Netherlands: Nucietron International BV 1989; 15–25. 14 Welsh A, Dixon-Brown A, Stedeford B. Calculation of dose distributions for iridium 192 implants. Acta Radiol Oncol 1983; 22:331. 15 Pierquin B, Marinello G. Manuel pratique de Curietherapy. Paris: Hermann, Editeurs des Sciences et des Arts, 1992. 16 Pernot M, Malissard L, Hoffstetter S, et al. 455 cases of carcinoma of the tongue: results of different modes of treatment. Abstracts of ESTRO (Den Haag), 1988, 468. 17 Mazeron JJ, Crook J, Beuk V, Walop W, Pierquin B. Iridium 192 implantation for T1 T2 epidermoid carcinomas of the mobile tongue: the Creteil experience. Int J Radiat Oncol Biol Phys 1989; 17(Suppl. 1):225.
18 Pernot M, Mallisard L, Aletti P, et al. (1990) Brachytherapy in the management of 147 T2N0 oral tongue carcinoma treated with irradiation alone. Int J Radiat Oncol Biol Phys 19(Suppl. 1): 139–40. 19 Owen JR, Maylin C, Le Bourgeois JP, et al. 192 Iridium implantation of tumours of the anterior two-thirds of tongue and floor of mouth. A retrospective analysis of treatment results and sites and causes of failures. J Eur Radiother 1981; 2:93–102. 20 Mazeron JJ, Richaud P. Compte-rendu de la 18e reunion du Groupe Européen de Curie-thérapie. Session consacrá aux cancers de la lévre. Padoue (Italie). J Eur Radiother 1984; 5:50–6. 21 Daly NJ, Douchez J, Coombes PF. Treating carcinoma of the penis by iridium 192 wire implant. Int J Radiat Oncol Biol Phys 1982; 8:1239–43. 22 Leung S, Otmezguine Y, Calitchi E, et al. Locoregional recurrences following radical external beam irradiation and interstitial implantation for operable breast cancer. A twenty-three year experience. Radiother Oncol 1986; 5:1–10. 23 Pierquin B, Huart M, Raynal M, et al. Conservative treatment for breast cancer: long term results (15 years). Radiother Oncol 1991; 20:16–23. 24 Barringer BS. Radium in the treatment of carcinoma of the bladder and prostate. JAMA, 1917; 68:1227–30. 25 Holm HH, Juul N, Penderson JF, Hansen H, Stroyer I. Transperineal 125 iodine seed implantation in prostate cancer guided by transrectal ultrasonography. J Urol 1983; 130(2):283–6. 26 Zietman AL, DeSilvio M, Slater JD, Rossi CJ, Miller DW, Adams JA, Shipley WU. Comparison of conventional-dose vs. highdose conformal radiation therapy inclinically localised adenocarcinoma of the prostate: a randomised controlled trial. JAMA 2005; 294:1233–9. 27 Fowler J, Chappell R, Ritter M. Is α/β for prostate tumors really low? Int J Radiat Oncol Biol Phys 2001; 50:1021–31. 28 Nag S, Beyer D, Friedland J, Grimm P, Nath R. American brachytherapy society (ABS) recommendations for transperineal permanent brachytherapy of prostate cancer. Int J Radiat Oncol Biol Phys 1999; 44:789–99. 29 Ash D, Flynn A, Battermann J, de-Reijke T, Lavagnini P, Blank L. ESTRO/EAU/EORTC Recommendations on Permanent Implantation for Localised Prostate Cancer. Radiother Oncol 2000; 57:315–21. 30 Gelblum DY, Potters L, Ashley R, Waldbaum R, Wang XH, Leibel S. Urinary morbidity following ultrasound-guided transperineal prostate seed implantation. Int J Radiat Oncol Biol Phys 1999; 45: 59–67. 31 Counter SF, Froese DP, Hart MJ. Prospective evaluation of formalin therapy for radiation proctitis. Am J Surg 1999; 177:396–8. 32 Merrick GS, Butler WM, Galbreath RW, Stipetich RL, Abel LJ, Lief JH. Erectile function after permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2002a; 52: 893–902.
1232 Interstitial brachytherapy
33 Potters L, Torre T, Fearn PA, Leibel SA, Kattan MW. Potency after permanent prostate brachytherapy for localized prostate cancer. Int J Radiat Oncol Biol Phys 2001; 50:1235–42. 34 Merrick GS, Butler WM, Wallner KE, Galbreath RW, Anderson RL. Prostate-specific antigen spikes after permanent prostate brachytherapy. Int J Radiat Oncol Biol Phys 2002b; 54:450–6. 35 Potters L, Morgenstern C, Calugaru E, Fearn P, Jassal A, Presser J, Mullen E. Twelve year outcome following
permanent prostate brachytherapy in patients with clinically localised prostate cancer. J Urol 2005; 173: 1562–6. 36 Kupelian PA, Potters L, Khuntia D, et al. Radical prostatectomy, external beam radiotherapy 72 Gy, permanent seed implantation, or combined seeds/external beam radiotherapy for stage T1–T2 prostate cancer. Int J Radiat Oncol Biol Phys 2004; 58:25–33.
50 Principles of external beam radiotherapy planning techniques STEPHEN L. MORRIS AND H. JANE DOBBS
Introduction Clinical evaluation – ‘who and when to treat’ Radiotherapy planning process Patient positioning and immobilization Tumour localization
1233 1233 1234 1235 1238
INTRODUCTION Radiotherapy has been used in the management of cancer since the turn of the twentieth century, and continues to have a very important role in the twenty-first century. X-rays were first discovered in 1895. The first reported cure of a malignant tumour using radiotherapy was reported in 1899.1 A patient with a histologically confirmed basal cell carcinoma of the nose was treated with 150 radiations over 9 months. She was cured and alive and well 30 years later. The International Commission on Radiation Units and Measurements (ICRU) was formed in 1925 recognizing that it was essential to define and measure the delivery of radiotherapy. The discovery of the new radionuclide cobalt-60 led to the development of the Cobalt Teletherapy unit, the first of which was installed in Ontario in 1951. Technological advances in radar led to the development of high-energy microwave sources such as the Klystron and Magnetron, and to the invention of the modern linear accelerator, the first of which was built at the Hammersmith Hospital in London, June 1952.2 The further major advances adding to modern radiotherapy planning techniques include the invention of computed tomography (CT) by Hounsfield in 1973 and the first use of CT and computers for intensitymodulated radiation therapy (IMRT) in 1990. Alongside the technical advances, results of clinical trials provided an evidence base for the establishment of radiotherapy as a key
Definitions of tumour and target volume Computerized dose planning Verification of treatment Key references References
1243 1245 1249 1250 1250
treatment for cancer. Developments in radiobiology increased the understanding of radiotherapy and formalized dosimetric methods. Currently 1 in 3 people will be diagnosed with cancer at some time during their lifetime, and 1 in 4 will die of cancer. About 40% of cancer patients receive radiotherapy and it remains one of the most successful ways of curing and controlling cancer.3 Radiotherapy has a curative role in two thirds of patients, either alone or in conjunction with surgery or drug treatment. In addition radiotherapy has an important role in palliation of symptoms in patients with recurrent or metastatic disease. This chapter will describe the importance of clinical evaluation, patient positioning and immobilization, the acquisition of patient and tumour data using different imaging modalities, definitions of tumour and target volume, use of computerized dose planning and verification of treatment in order to obtain the high standards of radiotherapy treatment planning.
CLINICAL EVALUATION – ‘WHO AND WHEN TO TREAT’ The decision to use radiotherapy as a treatment modality is that of the clinical oncologist after a thorough clinical assessment of the whole patient including an accurate histological diagnosis and definition of the site, size and
1234 Principles of external beam radiotherapy planning techniques
extent of tumour. The role of radiotherapy for a particular tumour type is determined partly by the radiosensitivity of the tumour relative to adjacent normal tissues and partly by the probability that the tumour is truly confined to allow treatment with localized radiotherapy. Radiotherapy has a curative role in many cancers as the sole treatment, or in combination with chemotherapy or hormone therapy, or as an adjuvant to surgery. Palliative radiotherapy is an important treatment in many cancers, for example to relieve pain from bone metastases, control bleeding or treat spinal cord compression. Perhaps the most important decision is whether treatment is radical or palliative in intent. Radical treatment attempts to kill or remove all malignant cells present with the aim of cure or local control. Some morbidity may be acceptable if cure is to be achieved. Palliative treatment aims to relieve symptoms or temporarily restrain the growth of the tumour. Any side effects should therefore be of minor impact on the patient’s quality of life. The choice between radical or palliative treatment may be straightforward, but in many cases is difficult. For example, if the general condition of the patient is poor, even if the disease is localized, they may not tolerate the stresses of radical treatment. When making this decision the following factors are important: ●
●
Tumour factors: site, size, spread (local or metastases), operability, radiosensitivity, histology. Patient factors: age, performance status, physical and mental condition, co-morbidity, functional and cosmetic outcome of treatment, reliability, patient preference.
All newly diagnosed patients should be assessed within a multidisciplinary setting where the roles of surgery, radiotherapy (RT), chemotherapy, hormone therapy, new agents, clinical trials and palliative care can be considered at the start of the patient’s journey. All patients should have their tumour classified according to a recognized staging system such as tumour, node, metastasis (TNM), the International Federation of Gynecology and Obstetrics (FIGO) or the American Joint Committee on Cancer Staging (AJCCS). Histology should be reviewed and discussed. Following this discussion the patient needs to be informed of the diagnosis and through clear communication and information giving, the best course of treatment offered for that individual patient. The decision to offer treatment with RT followed by consent of the patient is the first in a series of a large number of steps in the process of planning, prescribing and delivering RT.
dose prescription is made by the clinical oncologist. The process then of planning radiotherapy from preparation to delivery requires close liaison between all members of the team, including physicists, radiographers, planning technicians, diagnostic radiologists, mould room technicians and clinical oncologists to ensure the highest standards of care. The transfer of data during each step of the procedure is critical, and human errors of transcription or of data interpretation or data input at any stage can lead to the creation of systematic errors.4 A quality assurance programme to ensure accurate treatment outcome should be set up as part of the radiotherapy procedure and is necessary in order to achieve optimum treatment in terms of maximizing tumour control probability and minimizing normal tissue complications. Figure 50.1 illustrates the different steps that have to be taken successively during the radiotherapy planning process, and there should be a continuous feedback between all the different steps so that new information can be integrated into the final treatment policy. It is essential that at each step there is recording of information and full documentation. When the final decision to use radiotherapy is made, the treatment prescription is given, including a statement of the aim of therapy, the definition of target volume to be treated and the specification of dose, fractionation, overall time and other treatment parameters, as described in the International Commission on Radiation Units and Measurements (ICRU) Report 50,5 ICRU Report 626 and ICRU report 71.7
Pre planning
Clinical Evaluation/staging/MDM
Decision to treat/consent
Planning
Acquisition of data using simulator/CT/MRI with immobilization in treatment position
Delineation of volumes GTV/CTV/PTV
Computation of dose distribution Select optimal plan
Delivery
Dose prescription
Set-up at therapy unit
RADIOTHERAPY PLANNING PROCESS Once the treatment decision is made, a preliminary plan of action for the choice of target volume, position, immobilization, technique, machine, beam energy and
Verification Online portal imaging/in vivo dosimetry
Figure 50.1 process.
Steps of the radiotherapy treatment planning
Patient positioning and immobilization 1235
PATIENT POSITIONING AND IMMOBILIZATION It is paramount at the outset of the planning process that a comfortable and reproducible patient position is developed. This position must be the same at the time of acquiring images for localization and planning as that used on a daily basis during treatment delivery. Any alteration in position may change internal or external anatomy and risk over- or under-dosage. The extent of day-to-day variation in position of the patient differs from one part of the anatomy to another. This is illustrated by a wealth of studies in the literature recording the accuracy of reproducibility of patient positioning during a course of treatment for a variety of different tumour sites. Care given to the immobilization of the patient and development of innovative fixation devices is essential. It has also been clearly shown that displacement of tumour and normal organs occurs with variation in the position of the patient.8 Examples of organ movement are displacement of the larynx on swallowing, cardiac pulsation, movement of the lungs and the kidneys on respiration, variation in position of the prostate, seminal vesicles and bladder with change in bladder and rectal filling. The difference between patient movement and organ motion is an important consideration when planning radiotherapy. These variations are dealt with separately by using a set-up margin (SM) and an internal margin (IM) respectively around the target volumes for radiotherapy planning as defined by the ICRU. The degree of immobilization required varies according to the technique and volume being irradiated. High-dose, small-volume techniques might require millimetre precision, e.g. stereotactic radiotherapy, whereas some largevolume techniques such as total body irradiation may allow a larger tolerance. There are a number of techniques used to aid positioning and immobilization, and many currently being developed. Patient positioning aids such as three-point laser set-up, pillows, headrests, foam cushions and customdesigned couch attachments are used. It is possible for the patient position to change during treatment (intra-fraction movement). A review of patient set-up immediately post treatment and monitoring techniques such as real-time electronic portal imaging give an indication of the stability of the patient. Complex immobilization techniques used include Perspex casts, shells, and stereotactic head frames. These restrict patient positioning and give a higher degree of precision than positioning aids. Where the patient has kyphosis, scoliosis, frozen shoulder or limitation of joint movement extra limb pads or immobilization devices may be necessary. Any pain or discomfort to the patient must be considered. Metallic prosthesis and abdominal stoma can cause problems and should be excluded from the radiation volume. Pacemakers pose specific problems during radiotherapy and a full pacemaker check and cardiology advice should be sought prior to treatment.9 Other conditions may have to be addressed. For example,
radical radiotherapy for head and neck cancer in a patient with severe Parkinson’s was made possible using the aid of a deep brain stimulator.10 Close liaison is necessary with mould room technicians preparing fixation devices, so that details of the position of the patient and tumour site to be treated are all defined; the most relevant device can then be made and, wherever possible, markings made on the fixation device rather than the patient’s skin. Full use must be made of immobilization devices, lasers and placing permanent skin markers, such as tattoos, at optimum sites. All of these details must be carefully documented to ensure accurate transfer of data from localization through all the stages of the planning process and subsequent treatment. A photograph of the patient at the time of localization in the simulator or CT scanner aids in recording the set-up parameters.
Head and neck The immobilization technique used to position the head and neck is crucial to the accuracy of radiotherapy delivery. The simplest technique is the thermoplastic shell (Orfit) that requires immersion in warm water prior to stretching over the patient to provide a good fit and attachment to the couch. The most frequently used technique is a Perspex shell, custom made over a plaster of paris impression of the patient and fixed reproducibly to a head board on the couch, see Figure 50.2. A study comparing the two techniques found the average reproducibility using Orfit was 5–10 mm, and using a Perspex shell was 3–4 mm.11* Reproducibility depended upon the fit, hair length, the number of fixing points and the expertise and experience of the technician. However, another study in 43 patients found no significant difference between set-up accuracy using Orfit or a Perspex shell with an average reproducibility of 2 mm.12* The ultimate immobilization device is a stereotactic frame. A study that
Figure 50.2 impression.
Vacuum-moulded Perspex shell with plaster-cast
1236 Principles of external beam radiotherapy planning techniques
compared immobilization using a Perspex shell system and a stereotactic frame found the mean displacement of the lateral fields was 1.8 mm compared to 1 mm respectively.13* The Perspex shell with an uncertainty of 1–2 mm is the optimum fixation device, as it is relatively comfortable for the patient, has the advantage of being transparent and provides a vehicle for field centre and portal delineation marks to aid set-up of treatment. Sections can be removed from the shell to allow for skin sparing thereby reducing the skin reaction. If the neck is treated it is important to immobilize the shoulders and extend the Perspex shell with a good lower neck fixation point to reduce spinal cord movement. A study in 20 patients using a Perspex shell immobilization system showed that 94% of translational displacements were 3 mm and 99% were 5 mm, sufficiently accurate for use with IMRT treatments.14*
Thorax Immobilization of the thorax is important for treatment of lung cancer, breast cancer and tumours of the mediastinum. When treating patients with lung cancer, opinions differ as to the type of immobilization necessary to ensure maximum set-up accuracy and minimum thoracic movement during treatment. The system should eliminate lateral rotation and fix the arms in a comfortable, stable and reproducible position. In most cases the patient will be supine with the arms extended above the head with an arm board or cradle system.15 Variations in normal respiration occur during treatment for lung and breast cancer causing movement of the underlying breast, lung and heart. There have been attempts to control this with deep inspiration breath hold techniques (DIBH)16 or the automatic breathing control technique (ABC),17 and studies are ongoing. The accuracy of set-up for breast irradiation has been addressed in many publications.18,19 A study that looked at 376 portal films performed on 14 patients undergoing tangential breast radiotherapy with tangential fields showed that the standard deviation in a cranio-caudal direction between portal films and simulator films decreased from 5.8 mm to 3.7 mm with a new technique using fixation of one arm to an arm pole.20* This study provides objective data to show that fixation of the patient during breast radiotherapy is essential to improve accuracy of field reproducibility. A variety of immobilization techniques are used, depending on whether localization is performed on a simulator or CT scanner. If localization for breast irradiation is to be performed on a CT scanner, then the patient needs to be supine with one or both arms raised and secured above the head in order to pass through the aperture of the scanner. Gagliardi et al.21* presented an analysis of their treatment techniques of patients with node-negative breast cancer using three-dimensional CT treatment planning. The patient was immobilized supine with one arm above the
head using a vacuum-moulded polystyrene bag. They analysed the shift on a grid drawn on the skin of the patient’s breast according to movement in the position of the ipsilateral arm. When the arm was abducted to 130°, as for entry through a CT scanner for planning, the central mammillary plane was shifted upwards by 2 cm and the displacement was up to 4 cm in the axilla. If the arm was abducted to 90°, the displacement in the cranio-caudal direction was small over the breast, but in the region of the shoulder joint there was a difference of 2 cm between the grids. This illustrates that, if the arm position is varied between localization in a CT scanner and treatment on a therapy unit, there will be an error in set-up. It also shows that the use of skin tattoos as landmarks for setting up treatment fields is highly dependent on the arm position and that fixation of the arm(s) is essential. Location of the breast for radiotherapy can also be performed on the simulator using an isocentric technique and manual or automatic contouring at multiple levels. A breast board with a wedge can then be used to bring the sternum horizontal, so obviating the need for collimator angle on the tangential fields and reducing the complexity of dosimetric calculations. Figure 50.3 shows a patient on a purpose-built wedge with two arm poles for symmetry, each of which has a full range of movement in all directions to maximize the comfort of individual patients. A foot board ensures support for the patient in the cranio-caudal plane and the angle of the wedge is fully adjustable to allow for individual patient variation. Other immobilization techniques such as a customized hemibody foam cradle, plastic fixation shell, or alpha cradle rest have not shown advantages over the standard angled breast board, arm support with floor, knee and bottom supports.15 The exception is in women with large pendulous breasts. Techniques treating the patient prone22*,23* or using a micro shell that minimizes skin dose24* have been shown to be of value.
Figure 50.3 Immobilization device for breast irradiation with variable wedge angle and adjustable arm poles suitable for localization using a CT scanner.
Patient positioning and immobilization 1237
Abdomen and pelvis Immobilization of patients for abdominal and pelvic radiotherapy is unsatisfactory and no ideal technique has yet been devised. The patient is positioned prone or supine depending on the position of the tumour and the beam arrangement used to treat the tumour and avoid normal tissues. The supine position is standard for treating prostate, bladder and gynaecological tumours. The prone position is standard for treating rectal tumours. When treating pelvic tumours the prone position has the advantage of displacing the small bowel, but the supine position is more comfortable for the patient. Olofsen-van Acht et al.25* treated patients with gynaecological tumours in the prone position with the aid of a bellyboard and showed significant reduction in the amount of irradiated small bowel; daily set-up variations were small. Griffiths et al.26* reported a study of pelvic radiotherapy showing that errors in movement were greatest in the cranio-caudal direction and that the use of lasers improved lateral shift errors, particularly in prone patients. All patients should have lateral tattoos, which should be aligned with lasers as a minimum requirement to prevent lateral rotation (Fig. 50.4). The study also showed that it is important to examine the type of mattress used on the treatment couch as, if these vary between simulator and treatment unit, they can introduce error, as can the use of a ‘tennis racquet’ and other couch windows necessary for under-couch treatments. The use of carbon-fibre couch inserts on both simulator and treatment units has largely eliminated this type of variation. When treating prostate cancer it has been shown in a randomized controlled trial that there is significantly less prostate motion in the supine position.27** The prone position required a larger planning target volume (PTV) with resulting increased doses to critical organs such as small bowel, rectal wall and bladder wall. The supine position was associated with improvement in patient comfort and therapists’ convenience.
Figure 50.4 Patient positioning using laser lights to align midline and lateral skin tattoos.
For treatment of prostate cancer, some centres used to treat patients with a full bladder in order to distend the bladder mucosa away from the prostate target volume and to displace small bowel from the irradiated area. However, Ten Haken et al.28* showed that by filling the bladder or rectum using catheters, to simulate treatment conditions of urine or stool contents, the prostate gland moved 0–2 cm (mean 0.5 cm). This displacement of the prostate gland was shown to occur out of the high-dose radiation zone, leading to inadequate dosage of the tumour. Other studies29–31 have confirmed and quantified the movement of the prostate gland and seminal vesicles during radiotherapy. Zelefsky et al.32* evaluated 50 patients with serial CT scans and showed that prostatic and seminal vesicle displacement during radiotherapy was more pronounced amongst patients whose initial planning scans had large rectal and bladder volumes. Patients in their study were in the prone position with an empty bladder. They showed that prostatic displacement was more common in the AP and cranio-caudal directions and less so laterally. It would appear from these studies that asking patients to open their bowels and empty their bladders before both planning CT scan and treatment each day reduces movement of the prostate and seminal vesicles. Pinkawa et al.33* showed in a study of 34 patients treated supine that prostate position stability was the same with a full bladder compared with an empty bladder, but with an empty bladder there was an increase in the amount of bladder in the high-dose region and higher doses to bowel loops. Exact filling of the bladder is probably unnecessary but a comfortably full bladder will maintain consistency of prostate motion, reduce genitourinary complications and ensure that small bowel is displaced from the treatment volume.15 The impact of patient immobilization on set-up accuracy for prostate irradiation is controversial.34 Soffen et al.35 are among those convinced that rigid immobilization is necessary. However, equally good results have been reported by Catton et al.36 and Dearnaley et al.,37 who advocate using supports for the knees and lower limbs only. Song et al.38* compared the set-up accuracy of four immobilization devices versus none, and concluded that laser-aligned multiple tattoos with meticulous patient positioning by trained radiographers was as good as the use of formal immobilization devices. Several studies have evaluated changes in bladder volume and position during radiotherapy for bladder cancer.39–42 The most significant directions for bladder movement were in the anterior-posterior (AP) and cranio-caudal directions and appeared to be random in time and direction. Turner et al.41* found that bladder movement occurred in up to 60 per cent of patients as a displacement greater than 1.5 cm between the bladder wall and 95 per cent isodose, and they recommend a margin of 2 cm, at least for tumour-bearing regions of the bladder. Most centres require patients to empty the bladder by voluntary micturition just before CT scanning and each day before radiotherapy for treatment of bladder cancer. It is important to avoid large volumes of
1238 Principles of external beam radiotherapy planning techniques
oral contrast media during scanning as these lead to distension of the bladder, causing inappropriately enlarged volumes. Muren et al.43 investigated organ motion, set-up and treatment margins in radical RT for bladder cancer in 20 patients with weekly CT scans and electronic portal images (EPIs). They found substantial patient set-up variation and large internal motion of the bladder leading them to recommend that an additional 2–6 mm be added to conventional margins. A review of set-up error verification using portal imaging describes a ‘state of the art’ set-up accuracy of 2 mm for head and neck treatments (one standard deviation of the random and systematic set-up error), 2.5 mm for prostate, 3.0 mm for pelvis and 3.5 mm for lung treatments.44
TUMOUR LOCALIZATION When planning radiotherapy, localization of the target volume within the patient is done in relation to external reference points under exactly the same conditions as subsequent treatments. Acquisition of tumour data involves the use of optimum imaging modalities for the particular tumour site. Traditionally, clinical data for radiotherapy treatment planning have been collected in the cross-sectional plane. The conventional method of tumour localization involves the use of AP and lateral radiographs, often with the use of contrast media to define the target volume, usually obtained with the use of a simulator. A transverse outline of the patient at the centre of the target volume is then taken and the tumour and anatomical data transferred to the cross-sectional map. However, this method of localization fails to visualize the tumour itself in most cases, as orthogonal radiographs merely show bony landmarks with contrast outlining hollow viscera. Tumours such as those of the bladder, oesophagus, prostate, rectum and vagina may be outlined directly, but in most cases the contrast media outlines anatomical structures only. The advent of X-ray computed tomography (CT) scanners in the late 1970s provided a method of obtaining the required tumour information accurately, rapidly and in a transverse axial plane ideal for radiotherapy planning. In almost all cases, the primary tumour and its extensions can be visualized, together with the precise positions of sensitive organs such as kidneys, lungs and spinal cord and the patient’s external body contour. In the past decade CT scanning has become truly three-dimensional, not only for graphic visualization but also for three-dimensional dosimetric calculations. This enables both visualization and localization of the tumour, target volume and normal organs as a threedimensional display with subsequent three-dimensional planning. This has led to more complex treatment field arrangements with detailed field shaping and the ability to perform conformal therapy (see Chapter 51). The increased use of magnetic resonance imaging (MRI) in the staging of malignant disease has shown it to be particularly useful for tumours of the central nervous
system (CNS), head and neck and musculoskeletal tumours, and prostate and cervical cancer, where it provides new information about the site and extent of the gross tumour.45 Additional information may be obtained from ultrasound examination, positron emission tomography (PET) or single positron emission computed tomography (SPECT) scanning. Image registration methods are being developed to combine these data from different modalities using a variety of algorithms. However, it should be remembered that all available information must be used when performing localization of the target volume, including clinical examination (under general anaesthesia in some instances) and surgical and histopathological details which, for some tumour sites, may provide the most important data. In any particular radiotherapy department, the choice of equipment used for localization will depend on the workload and range of patients and tumour sites to be treated, the techniques employed in planning and irradiation and the resources available.46 Recommendations on the optimum imaging strategies for localization, staging and planning have been published in a report by the Royal College of Radiologists.47 For example, it recommends that lung tumours should be localized with CT (except for superior sulcus tumours where MRI is preferred), staged with CT (and PET where available) and planned using CT with IV contrast and PET co-registration if available. It also discusses special issues such as the use of PET to resolve uncertainties in defining areas of consolidation, collapse and fluid from tumour. Current methods of tumour and target localization include use of the simulator, CT scanning, simulator CT and CT simulator, which are discussed below.
Simulator The simulator is an isocentrically mounted diagnostic X-ray machine which can reproduce treatment parameters and has the facility for screening by the use of an image intensifier.48 It has a couch, which is flat and capable of all the movements of the therapy unit, and a gantry, which can be rotated through 360°. The patient is set up on the simulator in the treatment position and any tumour masses are marked with wire and contrast medium placed in the bladder, rectum, vagina or oesophagus. A skin tattoo is placed over the nearest immobile bony landmark to the tumour, such as the pubic symphysis in the pelvis or the xiphisternum in the chest, to act as a reference point. Lateral tattoos should also be placed on the skin at fixed distances up from the couch on each side and laser lights are needed in the simulator to align the patient and prevent rotation. If field sizes are defined using the simulator guide-wires on the skin surface, they will appear larger than the target volume at depth because of the inward bowing of the isodose curves. Sufficient margin around the planning target volume must be allowed in order to give the required dose at depth.
Tumour localization 1239
Similarly, when an isocentric treatment plan is verified on the simulator, the field sizes on the skin will appear larger than the target volume in order to ensure adequate dose at depth, and this should be remembered and no adjustment made.
CT planning CT scans taken for radiotherapy treatment planning require different considerations from those taken for diagnostic use, where the prime object is to detect the presence of malignant disease. The radiation oncologist must determine the exact position of the tumour, its extent in all directions, the site of adjacent structures and the relationship between the organs, and the tumour and external landmarks which are to be used for setting up the patient on the therapy unit. The acquisition of tumour data involves the use of optimum imaging modalities defined in tumour site protocols and developed in collaboration with diagnostic radiologists.45,47,49 Interpretation of data by an expert in oncological diagnostic imaging is essential for ensuring accurate tumour delineation. The patient position and set-up and immobilization must be identical for CT scanning as for the treatment. The reproducibility of the patient’s position on the CT scanner can best be ensured by the presence of a therapy radiographer assisting the diagnostic team in the positioning of the patient on the CT couch. Exactly the same immobilization devices, e.g. Perspex shell, vacuum bag, arm pole, must be used on the CT scanner and a record made of these devices and any polystyrene head or knee pads or pillows used. Sagittal and lateral lasers should be available in the CT scanning room to align the patient and permanent skin markers are placed on the skin over the nearest immobile bony landmark to the tumour site. Lateral tattoos are also made to prevent lateral rotation of the patient. All of these support devices should be radiolucent, so as not to produce artefacts. Skin markers should be covered with either radio-opaque catheters or barium paste, in order that their site is recorded on both topogram and CT scans (Fig. 50.5). There must be a flat insert on the CT scanner couch and good-quality control of the image and spatial resolution. It must be remembered that even though the aperture size of the CT scanner may be up to 85 cm, the reconstructed diameter, which describes the area over which an image can be obtained or a scan reconstructed, may be smaller than this and limit the information that can be gained. The topogram (scanogram) produced by CT scanners is a digital radiograph obtained by moving the patient on the couch through the stationary fan beam of the machine. This is used to localize the anatomical position of the resulting CT slices (Fig. 50.6), but it must be remembered that the image is distorted laterally due to fan beam geometry and that the geometrical representation is not the same as that of the treatment beam since it is not divergent in all directions.
Figure 50.5 Photographic record of patient in treatment position on CT scanner. Tattoo over pubic symphysis marked with barium paste, midline radiopaque catheter and lateral tattoos are all illustrated.
Figure 50.6 Topogram (scanogram) of the pelvis with CT sections recorded. Barium paste landmarks skin tattoo over pubic symphysis.
Contrast media are used orally to outline small bowel and oesophagus, but can affect dose calculations if these are based on CT density data. For treatment of bladder cancer, patients should empty the bladder before CT scanning and treatment. For prostate tumours, it has been shown that excess gas or faeces in the rectum causes variation in position of the prostate, and a repeat CT scan should be performed on another day if this occurs. Use of contrast in a rectal catheter should be avoided, in view of the fact that it distorts the anatomy for planning and will not be used during the treatment. It is important to mark structures such as the introitus,
1240 Principles of external beam radiotherapy planning techniques
anal margin, vagina and any tumour masses or operative scar with barium or radio-opaque catheters, so that they are visualized both on the scan and topograms. With modern multi-slice CT scanners, images are taken very rapidly in a fraction of a respiratory cycle. First, a topogram is obtained over the part of the body of interest and then CT scans are taken at 4–10 mm intervals using a slice thickness of 4–8 mm. It is important that close liaison is made with the diagnostic radiologist to make sure that the CT scans obtained have the maximum amount of diagnostic information, as well as full body contour necessary for treatment planning. Misinterpretation of CT data can be a source of error in the delineation of the target volume due, for example, to erroneous evaluation of vascular structures, unopacified bowel and the partial volume edge effect.50 It is usual now for diagnostic scans to already have been taken as part of the initial staging procedure and subsequent additional CT scans for treatment planning are then performed (Fig. 50.7). This is because of the necessity
(a)
of having the patient under conditions for therapy with accurate external markers on the skin and to obtain data, not only of the tumour itself, but also in relation to external landmarks.51 The CT scanner and treatment-planning computer should be networked to allow for direct transfer of CT images. Outlining of body contour, tumour (gross tumour volume, GTV), and target volume (clinical target volume, CTV, and planning target volume, PTV) (Fig. 50.8) and normal organs takes place at the planning computer on each individual CT slice with an interactive software programme. For three-dimensional treatment planning, multiple slices must be taken using contiguous thin (2–4 mm) slices in order to obtain the maximum amount of information, as this is necessary to permit accurate three-dimensional dose calculations and for the formation of good-quality digitally reconstructed radiographs (DRRs). An automatic facility allows the body contour to be drawn rapidly, but all other structures have to be outlined by the clinician. This is a time-consuming procedure as well as one which is prone to inter-clinician variability.52,53 For two-dimensional planning, a dose distribution is calculated on the CT image corresponding to the centre of the proposed target volume (Fig. 50.9). Off-axis beam data can be used to determine the dose distribution at other levels within the target volume, in order to record the homogeneity of the dose throughout the entire volume. A beam’s eye view facility may help in the choice of beam configuration. Three-dimensional graphic display and dose volume histograms give additional information for the comparison of three-dimensional treatment plans. Immobilization techniques become critical and quality assurance studies are essential to verify patient and organ movement when using three-dimensional planning and conformal therapy. It has been shown that CT scanning information alters the radiation fields used in approximately 30 per cent of patients receiving radical radiotherapy treatment.54,55 These studies compared conventional planning with CT scanning as used for localization of the target volume, and
(b) Figure 50.7 CT scans of carcinoma of the bladder: bladder full, diagnostic images, (b) after micturition, CT planning images taken under treatment conditions, showing full body contour.
Figure 50.8 CT image of the pelvis for prostate tumour irradiation showing clinical target volume (CTV), planning target volume (PTV) and rectum (R), and midline and lateral skin tattoos.
Tumour localization 1241
showed that in around 30 per cent of cases a change in the margins of the target volumes was necessary in order to ensure coverage of the tumour and avoid geographical miss. Two studies have addressed the difficult question of whether this improvement in accuracy of targeting radiotherapy translates into an improvement in survival, although there is as yet no randomized study in the literature looking at improvement in local tumour control. Rothwell et al.56* made a retrospective comparison of the survival rate of patients whose bladder tumours were included within a 90 per cent treatment isodose with those who, by CT criteria, had tumour lying outside the 90 per cent isodose, i.e. a geographical miss. Although CT scans were obtained before radiotherapy treatment was given, the comparison was made following treatment, which had been delivered using the conventional plan. A significant difference was found in survival rate in favour of those whose tumours lay within the 90 per cent isodose. Overall the study predicted a 4–5 per cent increase in survival of the whole group of patients with bladder cancer at 3 years, if all tumours were included in all patients. However, the histopathological staging of the two groups of patients was found to be different. The incidence of positive lymphadenopathy was 73 per cent in the group treated to less than the 90 per cent isodose, compared with 21 per cent in the group with adequate tumour coverage, suggesting that the survival difference might be due to this prognostic factor rather than to the volume irradiated. Tsujii et al.57* reported a study of patients with post-nasal sinus tumours in which those treated with the use of Perspex shell before 1979 were compared with those treated after 1979 with the addition of CT planning – approximately 80 patients in each group. A significant difference in survival rate was reported in favour of the patients with CT planning, particularly for supra-structure tumours. The study also showed a reduction in field sizes and complication rates with the use of CT planning. There is no doubt that
CT information improves localization of the target volume for many tumour sites, even though the translation of this accuracy into improved survival rates may be modest.58
Simulator CT A CT modality attached to the gantry of a simulator can be used to produce images with a relatively limited resolution but at the same time as the simulation process.59 Simulator CT scanners therefore have a larger aperture size than CT scanners, but still have a limited reconstruction diameter. The scan time is much slower than the CT scanner and the spatial resolution reduced.60 However, they combine the facilities available on the simulator with a limited CT facility which can provide body contour and outline of structures such as the lungs, heart and bone contour, and can be used for simple inhomogeneity corrections. Images do not produce detailed tumour information or accurate CT numbers. However, the cost is low compared with that of the CT scanner and is particularly appropriate for techniques such as tangential breast irradiation, where it provides CT slices at the central, superior and inferior axes of the target volume. A recent study has shown that 80 per cent of plans for tangential breast irradiation using three level outlines are as good as those using an entire three-dimensional data set in terms of dose inhomogeneity assessment.61 It is possible to use multiple level outlines not only to improve breast dose homogeneity but also to localize the position of the lungs for tissue inhomogeneity corrections. Two studies have shown that the central lung distance (CLD) on a simulator film was the most useful predictor of ipsilateral lung volume included in tangential fields of breast radiotherapy, and correlated well with the lung volume measured on CT scanning.62,63 The maximum heart distance (MHD) on a simulator film of tangential breast fields correlates well with normal tissue control probability (NTCP) values for excess cardiac mortality from breast radiotherapy.64 Hence CLD and MHD, which are relatively easy to measure on the simulator film and portal image, can be used as good predictors of normal tissue toxicity. The spatial resolution of a simulator CT is not satisfactory in the pelvis or abdomen at present either for normal organ or tumour localization, and it is not possible to use this equipment to produce fully three-dimensional computerized plans.
CT simulator
Figure 50.9 Dose distribution using three fields produced by integrated CT planning system on the central slice for twodimensional planning.
CT scanners that provide state-of-the-art imaging can be combined with software to produce virtual simulation from a beam’s eye perspective with images which are equivalent to conventional simulator images. These can be used to replace the simulation process, but have to be related to an internal isocentre within the patient, anchoring data to external landmarks. A multi-image display system and
1242 Principles of external beam radiotherapy planning techniques
three-dimensional treatment-planning computer are linked with a portal imaging system to verify the dosimetric plan. The ability to derive CT scans and to provide target volume definition with margin generation, dose calculation and simulation all on one workstation is a major advance. The CT simulator provides maximum tumour information as well as full three-dimensional planning capabilities. The major use of the CT simulator is for three-dimensional computerized treatment planning and conformal therapy in the development of innovative treatment techniques. A small study of ten patients undergoing palliative RT for lung cancer showed a significant improvement in tumour volume coverage using virtual CT simulation compared to conventional simulation.65* As more patient groups undergo virtual CT simulation it will be necessary for each department to tailor its planning methods to the requirements of those patient groups.
MRI The main advantage of MRI is that it gives additional diagnostic information, particularly in tumours of the central nervous system (Fig. 50.10) and head and neck region, musculoskeletal tumours, cervix and prostate tumours. These data are available in multiple planes, including those not directly imaged by CT, and also give superior soft-tissue contrast. However, there is no bone signal, making comparison with conventional X-rays difficult. MRI does have certain disadvantages when used for RT planning, e.g. it does not provide electron density information required for dosimetry calculations and there is distortion inherent in the images. Integration of CT and MRI data into a single workstation using image registration is now possible and is recommended as the optimum imaging strategy when planning RT for brain, some head and neck, rectal, cervix, endometrium, bladder, prostate, soft tissue sarcomas and primary bone tumours.45,47 Experience of MRI alongside CT for prostate planning has shown that it allows better definition of all the structures relevant to prostate planning, i.e. prostate apex, seminal vesicles and rectum.66,67 Other advantages are that smaller volumes are often drawn using MRI and the inter-observer variation is smaller.15
Ultrasound The main use of ultrasound for treatment planning is in the localization of superficial tumour masses. For instance, in patients who present with stage 3 carcinoma of the breast there may be an excellent response to chemotherapy and this may be followed by subsequent radical radiotherapy. Clinical examination may then fail to detect an abnormality but ultrasound can be used to localize any small residual tumour masses within the breast and mark
Figure 50.10 Magnetic resonance image of tumour of the fourth ventricle.
the skin for targeting the booster dose of irradiation. Ultrasound has also been useful in the choice of electron energy for adjuvant radiotherapy to the breast tumour bed.68*
Positron emission tomography Positron emission tomography (PET) is a functional imaging technique based on the detection of altered tissue metabolism. 18F-fluoro-2-deoxyglucose (FDG) is the most commonly used tracer. Malignant cells take it up readily and the local concentration can be measured. Positron emission tomography has been available in some centres for over a decade and much experience has been accumulated. Its role is becoming established in brain, lung and colorectal cancer, lymphoma, melanoma and unknown primary tumours. There is growing interest in oesophageal, breast, pancreas, head and neck and ovarian tumours.69 Radiotherapy planning relies on anatomical landmarks, and PET has the potential to augment these images and aid in tumour localization. Positron emission tomography has been used to assist radiotherapy planning of lung cancer and shown to define gross tumour more clearly by differentiating tumour from atelectasis.70 It has also been used to determine whether sites of potential nodal involvement, not suspicious on CT,
Definitions of tumour and target volume 1243
should be treated.71 Studies on target delineation have shown a reduction in inter-clinician variability.72 A method of registration and fusion of CT and PET through automatic lung surface contouring by a threshold technique and chamber matching of extracted lung surfaces has been described.73 The new wave of PET/CT scanners now being introduced to departments is likely to play a large role in the future. Current lung cancer radiotherapy dose escalation trials are using PET to define the treatment target.74 However, there are no data at present on the impact of PET on survival, toxicity and local control. Positron emission tomography is currently being investigated in a NCRI Lymphoma Group study of early Hodgkin’s lymphoma. It is seeking to determine whether PET can predict early response to chemotherapy and define a group of patients who may not need radiotherapy. A study at St Thomas’ Hospital has shown that such patients may be those who have a negative PET scan after just two cycles of chemotherapy.75* Positron emission tomography has an established role in identifying active areas in brain tumours, and may help in the delineation of GTV and CTV. A dose escalation study in 40 patients using PET to delineate the target for boosting to 79.4 Gy has shown it is feasible. However, no improvement in overall survival or progression-free survival was seen.76 Compared to CT and MRI alone, fusion with PET allows more accurate delineation of the GTV.77
Image registration At present, CT remains the mainstay for acquisition of anatomical and tumour data for radiotherapy treatment planning, partly because correlation of CT numbers with electron densities makes it ideal. However as discussed above, for many tumour sites MRI, PET, ultrasound or SPECT may provide optimum tumour information, rather than CT. It has therefore become imperative to find methods of combining optimum tumour imaging data with CT data for radiotherapy treatment planning. This correlation is difficult because of differences in patient positioning, geometric distortions of magnetic resonance images, varying anatomical boundaries and differences in spatial resolution. Image registration methods can be used to find a match between two image sets from different modalities. Extrinsic landmarks can be matched using reference points, which may be located using fiducial markers or systems such as stereotactic frames, Perspex helmets, dental moulds or ear plugs for the brain. Markers can be glued to the skin, but skin movement limits the accuracy of this technique. Co-registration of CT and MRI is now being used successfully in many centres for planning radiotherapy for CNS and prostate tumours. Combining the information from CT with that of functional imaging such as PET is being investigated for many tumours as discussed above.
DEFINITIONS OF TUMOUR AND TARGET VOLUME The next step in the planning process is to define the extent of gross tumour, as delineated by clinical examination and optimum imaging, and then to decide on the target volume, which is designed to receive a radical dose of radiotherapy aimed at eradicating the tumour. The process of determining these volumes for treatment of malignant disease consists of several distinct steps, which have been well described in the ICRU Reports 50, 62 and 71. The first volume is that of the gross tumour volume (GTV), which is the gross, palpable, visible or clinically demonstrable extent of malignant tumour and corresponds to the site where the tumour cell concentration is at its maximum. Staging systems such as the TNM and FIGO classifications describe the site, size and extent of GTV; however, delineation of the GTV is very dependent on imaging. The UICC TNM staging classification states that imaging should be used to define the T stage. A recent review by the Royal College of Radiologists has recommended the optimum imaging strategies for radiotherapy planning.47 Expert advice from a diagnostic radiologist is essential to gain maximum information from any imaging modality, such as CT and MRI, used to locate the site and extent of the tumour.49 A margin is then added around the GTV in order to include direct local subclinical microscopic spread. This margin is the clinical target volume (CTV) that contains the GTV(s) and/or subclinical malignant disease at a certain probability level. This volume needs to be treated adequately. This margin usually has a decreasing malignant cell density towards the periphery, where it should reach zero. If the tumour has been removed prior to radiotherapy (e.g. operable breast cancer), then no GTV can be defined, and the volume of subclinical disease constitutes a CTV. If treatment is planned to a primary tumour and lymph node areas too, then there may be two clinical target volumes in a patient. These additional volumes with presumed subclinical spread may be designated CTV–N (if necessary CTV–N1, CTV–N2). Adding the letter T, N or M to identify volumes may better clarify their clinical significance. The size of the clinical margin chosen for subclinical disease depends primarily on tumour characteristics such as histological type, grade and features such as lymphatic, vascular and perineural infiltration. The size of this margin appears to vary widely, depending on the clinician’s experience and interpretation of the histological features of the tumour and its projected natural history. Data for defining this margin are sparse, and it is the most subjective step in the planning process. Knowledge from surgical and postmortem specimens, patterns of tumour recurrence, as well as from clinical experience are used to quantify the margin. It has been shown that there can be marked differences in the size and site of the clinical target volume for treatment of the same nasopharyngeal tumour when prescribed by two different clinicians.52 A study of patients planned
1244 Principles of external beam radiotherapy planning techniques
for radiotherapy for non-small-cell lung cancer showed a wide inter-clinician variation in the choice of tumour and target volumes, even with the use of CT scanning.50 The main causes of inter-clinician variation were found to be radiological interpretation, margin described for microscopic disease, and margin allowed for geometric variables. Studies of inter-clinician variation in delineation of the target volume for CNS tumours78 and prostate tumours79 show that for some tumours it has to be accepted that the GTV cannot be drawn as a well-demarcated volume with precise borders. There may remain a zone of uncertainty around the GTV due to inherent limitations of the imaging technique, inexperience or interpretation of imaging data and this should be expressed in a qualitative or quantitative way. The next step is to add a margin around the CTV to account for variation in size and position of tissues relative to the treatment beams due to organ movement, patient movement and variation in daily set-up, i.e. physiological or technical factors. For instance, swallowing causes movement of a laryngeal tumour, respiration leads to movement of bronchial tumours and the breast during irradiation. Variations in bladder or rectal filling can cause movement of bladder and prostate tumours. Verification studies have shown organ motion varying from 1 mm for the brain, 3 mm for the breast, 8 mm for the prostate, and 15 mm for bladder wall, as two standard deviations from the mean. ICRU Report 62 and 71 defines an internal margin (IM) to allow for the physiological variations in the shape, position and size of the organ, therefore defining an internal target volume (ITV) when added to the CTV. Image-guided radiotherapy techniques may reduce the internal margin needed in the future. Intra-fractional and inter-fractional variations occur in patient position and in alignment of the beams with external marks. To account for these changes the ICRU Report 62 and 71 defines a set-up margin (SM) to compensate for uncertainties in patient position and alignment of the therapeutic beams during treatment planning and throughout all treatment sessions. This set-up margin can be defined for each technique, using verification studies or a quality assurance programme. For example, set-up variations of 10 mm for the pelvis, 13 mm for the breast, and 2–4 mm for the head and neck have been reported as two standard deviations from the mean. Combining the internal margin for physiological changes and the set-up margin for technical variations with the CTV leads to the planning target volume (PTV) (Fig. 50.11). The PTV is a geometrical concept used for planning treatment. It is used to select the appropriate beam sizes and arrangements, and should ensure that the prescribed dose is actually delivered to the CTV when variations are taken into account. The penumbra of the radiation beam is not included in the PTV, and the beam aperture has to be increased in order to compensate for this. The beam sizes are defined by the 50% isodose.50 Organs at risk need to be defined when planning radiotherapy. Organs at risk are normal tissues whose radiation sensitivity may significantly influence treatment planning and/or prescribed dose, e.g. spinal cord.80 As is the case
with the PTV, any movements of the organs at risk during treatment, as well as uncertainties in the set-up during the whole treatment course must be considered. This leads to the planning organ at risk volume (PRV) using the same principles of internal and set-up margins as used for the PTV. Note that the PTV and PRV may overlap. Due to the limitations of techniques, in some specific clinical situations the volume receiving the prescribed dose may not match accurately with the PTV. It is in general of a simpler shape and larger (sometimes much larger). This leads to the concept of treated volume (TV), defined by the ICRU as the tissue volume which receives at least the absorbed dose selected as the minimum dose to the PTV. For example, if the prescribed dose is 60 Gy, and the minimum dose considered adequate is 5% below the central dose, the TV is enclosed by the 57 Gy isodose surface. In ‘conformal therapy’ the tissue volume receiving the TV dose level should match the PTV. The conformity index is the ratio of the TV to the PTV and can be used to optimize planning. The TV should not be confused with the irradiated volume (IV), which is the tissue volume that receives a dose that is considered significant in relation to normal tissue tolerance. Of note, when the TV is made smaller by the use of many beams (e.g. IMRT) the IV gets larger. During the planning process when all of these uncertainties, volumes and margins are defined it is recognized that simple (linear) addition of all factors of geometric uncertainty will lead to an excessive PTV incompatible with the tolerance of the surrounding normal tissues. The risk of missing part of the cancer cell population and disease control must be balanced against the risk of complications. The selection of a composite margin and delineation
Gross tumour volume Clinical target volume (CTV) Planning target volume (PTV) CTVcombined IMSM
Figure 50.11 Definitions of volumes according to ICRU Reports 50 and 62. IM, internal margin; SM, set-up margin. (Reproduced with permission from International Commission on Radiation Units and Measurements 1993, ICRU Report 50; and 1999, ICRU Report 62. Prescribing, Recording and Reporting Photon Beam Therapy. Bethesda, Maryland: ICRU.)
Computerized dose planning 1245
of the border of the PTV involves a compromise that relies upon the experience and the judgement of the clinical oncology team. If all the relevant uncertainties and their standard deviation are known then estimates of their combined effect can be made using simple algorithms.30 In practice, there is a tendency to use standard margins added to the CTV to define the PTV, and much work is in progress.81 The margins may be asymmetrical. For instance, for a lung tumour the margin may be 11 mm in the transverse plane and 15 mm in the cranio-caudal direction, due to the effect of respiration.82 McKenzie81 examines the issue of breathing-induced motion and suggests that the margin to allow for respiration should be added linearly to the quadrature sum of the other contributing errors. It is recommended that each institution should evaluate local variations and uncertainties (which are dependent on immobilization, set-up, technique, etc.) and define a reasonable level of probability for the different components. In 1976, ICRU Report 24 considered that 5 per cent accuracy was required in the delivery of absorbed dose. Brahme et al.83 and Mijnheer et al.84 suggested that the requirement for accuracy should be 3.5 per cent, one standard deviation, in the dose value at the specification point in the target volume. This requirement is based on the sigmoid shape and steepness of dose–response curves for both tumour control and normal tissue complications. These curves show that a small variation in dose level can have a considerable influence on the probability of tumour control and also on the likelihood of normal tissue complications. However, within the target volume, there is a variation in tumour cell density. There also exists heterogeneity of cell type and radiosensitivity as well as distribution within the cell cycle, so that clinical dose–effect data curves have a reduced steepness. A clinical study reported from the Institut Gustave Roussy suggested that an increase in dose of 7–10 per cent resulted in clinically detectable reactions.8,85 These studies show that accuracy is important, and a multidisciplinary team effort with a careful quality assurance programme is essential to ensure optimum quality throughout the whole radiotherapy process and so achieve a successful treatment outcome.
● ●
● ●
Dose specification point Prescribed target dose, maximum and minimum doses acceptable, maximum permissible doses to vital organs outlined on the plan Need for bolus (e.g. to maximize skin dose) Need for compensators, beam shaping.
Treatment machines Megavoltage treatment machines include cobalt-60 machines producing 1.25 MV gamma irradiation and linear accelerators operating between 4 and 25 MV. These vary in their build-up depth (Dmax) and percentage depth dose characteristics, as well as other features such as availability of half-beam blocking, or multi-leaf collimators, ease of lead shielding, rotational facility, effect of penumbra on beam definition and couch attachments. Cobalt units produce beams of less penetration than linear accelerators, have less skin sparing and a greater penumbra, longer treatment times but require less maintenance. Electron therapy may be useful for single-field treatments of superficial tumours, particularly where sparing of underlying cartilage or bone is important (e.g. nose, ear). High-energy electron beams (15–25 MeV) are used in the treatment of head and neck tumours, such as cervical lymph nodes, thyroid carcinomas and parotid tumours, to avoid dose to the underlying spinal cord. Electron beams have a sharp fall off in dose beyond the 90 per cent isodose and so the energy of the electron is chosen to encompass the target volume by the 90–95 per cent isodose. The effective treatment depth in centimetres is approximately one third of the beam energy in MeV, and total range about half, dependent on field size (Fig. 50.12). Electron beams provide a small amount of skin sparing and if this is not required, then bolus material must be used to increase dose to the skin surface. The volume included by the 90 per cent isodose is less than the field size on the surface, and hence wider margins must be allowed when delineating the target volume. However, at depth, the 40 per cent isodoses and below tend to bow outwards so increasing dose laterally. If the target volume is near a critical structure such as the eye, shielding against lateral scatter must be applied or an alternative modality of treatment considered. Doses
COMPUTERIZED DOSE PLANNING Data are transferred from the CT scanner or simulator via a direct line, magnetic tape or digitizer to the computerized planning system. Quality control measures are essential for both the method of transfer and the planning system, in order to guard against systematic errors being introduced at this stage. Discussion then follows between clinician and physicist or planning technician on the following: ● ●
Choice of treatment machine Possible configuration of beams for treatment planning
Surface
8 MeV 100 90 20 5
50 10
Figure 50.12 Isodose curves for an 8 MeV electron beam 10 cm 10 cm.
1246 Principles of external beam radiotherapy planning techniques
beyond cavities may be higher than expected even after density corrections, e.g. when using electron therapy to treat the chest wall after mastectomy where there is lung beneath the ribs. Superficial radiotherapy is used to treat up to 90% of superficial tumours such as basal cell carcinomas with low energy X-rays of between 50 and 160 kV. Owing to the dominance of photoelectric absorption in tissues at these energies there is high bone absorption. The advantages, however, are the simplicity of the superficial treatment machine in design and operation, the wide range of field defining applicators and the ease of collimation and field shaping to individual requirements with only 1–2 mm of lead.
hypopharynx, opposing lateral fields with the use of wedges often give a satisfactory distribution for the target volume and may be ideal for radical treatment. Doses are prescribed at the midplane point on the central axis of the beam. It must be remembered that there is narrowing of the isodoses in the centre of the volume and hence a larger field size on the skin must be chosen to adequately treat a central volume. The maximum dose occurs peripherally due to the build-up effects, and care must be taken to ensure that this dose is calculated as it may lie over a vital structure, e.g. spinal cord. ICRU Report 50 requires the
Posterior
8MV Prone
Configuration of beams SINGLE FIELD
114
A single megavoltage field may be used for palliative treatment, e.g. bone metastases, for ease of set-up and convenience to the patient. The fall off of dose across the target volume produces an inhomogeneous effect, but this is adequate for achieving pain relief. However, the site of the lesion, e.g. vertebral metastasis, must be considered when specifying the dose, as well as the dose-limiting structure, e.g. spinal cord. Table 50.1 shows the variations in the depth of the spinal cord along the vertebral column and differences in percentage depth dose from different megavoltage machines. If the dose is prescribed as an ‘applied dose’, i.e. 100 per cent at the Dmax, then the dose to the anterior vertebral body may be inadequate. For megavoltage irradiation the ICRU dose specification point, which lies at the centre of the target volume, is therefore chosen and the maximum and minimum doses carefully recorded (Fig. 50.13).
110 100 90
Left
Right
80
10
Anterior
Figure 50.13 Isodose distribution from an 8 MV single direct posterior beam. Centre of target volume (100 per cent) is ICRU dose specification point, maximum dose 114 per cent, minimum dose 90 per cent.
OPPOSING FIELDS
The isodose distribution from two parallel opposing fields (Fig. 50.14) is not homogeneous, does not conform well to the target volume and irradiates much normal tissue to the same dose or greater than the tumour. Hence this technique is more commonly used for palliative treatments. However, for tumours such as the larynx and
Anterior
8MV
95
95 102 100
Right
Table 50.1 Variation in depth of spinal cord along the vertebral column and differences in percentage depth doses from treatment machines
Structure Cervical canal Thoracic canal Lumbar canal Lumbar vertebral body
Depth (cm)
Percentage depth dose Cobalt-60 6 MV
5 6 7 10
75–80 70–75 65–70 52–57
83–88 81 79 64
50 10
10 50 80
102
Left
80
Posterior
Figure 50.14 Isodose distribution from 8 MV opposing anterior and posterior beams. Centre of target volume (100 per cent), maximum dose 102 per cent, minimum dose 95 per cent.
Computerized dose planning 1247
midplane dose and maximum and minimum doses as specified doses for opposing fields. The applied dose to each field is determined from depth dose charts, which are available for each field size for each treatment machine and are normalized to 100 per cent at the depth of maximum build-up for a 10 10 cm field at the standard SSD. When irregular or rectangular fields are used, the contribution from scattered irradiation is different from that from square fields, and tables must therefore be consulted to obtain equivalent square fields. Sample calculation for parallel opposing fields treated on a 6 MV linear accelerator, using an isocentric technique: Field size 12 17 cm
used to alter the dose distribution to compensate for missing tissue due to body contour (Fig. 50.16) and to avoid high-dose areas when beams are combined. Wedges attenuate the beam and hence, for the same incident dose, a greater number of monitor units are set compared with when an open field is used. In order to concentrate dose at the target volume and reduce dose to critical organs, ‘weighting’ may be used so that different amounts of radiation are given by each of the beams. Compensators may be designed individually and placed in the beam to correct dose inhomogeneity due to obliquity of body contour and varying depth of the target volume in the patient (e.g. oesophagus). Compensators are often made of aluminium alloy as this makes them easy to handle.
Equivalent square 14 cm2 Interplanar distance (separation of patient) 18 cm
Anterior
8 MV
Depth of midpoint 9 cm Midpoint corresponds to centre of target volume (ICRU point) and is therefore placed at the isocentre. SSD 100 9 91 cm
50
Right
20
80
95
103
20
Left
100
Output factor for a 14 cm2 field at 9 cm 0.86. For 2 Gy per fraction at ICRU dose specification point monitor units per field 100/0.86 116
Posterior
MULTIPLE FIELDS
Combinations of two, three or more fields are commonly used for most radical treatments (Fig. 50.15). Wedges are
100 90
100 10
80 70 60
Figure 50.15 Isodose distribution from 8 MV linear accelerator, anterior, posterior and two lateral beams. Centre of target volume ICRU dose specification point 100 per cent, maximum target volume 103 per cent, minimum target volume 95 per cent.
130 120 110 100 90 80 70 60
50
50 40
40
30 30
5 10
5 10 20
20
(a)
(b)
Figure 50.16 (a) 12 12 cm 6 MV open field; (b) same field with 60° wedge.
1248 Principles of external beam radiotherapy planning techniques
Beam shaping Conventional shielding is achieved by Cerrobend blocks, which may be standard shapes or individually designed for each patient. For the mantle technique, blocks are made to allow for beam divergence to improve the accuracy of shielding at the interface between beam and block.86 Halfbeam blocking can be achieved with independent collimator jaws, eliminating divergence so that a straight beam edge can be used. Multi-leaf collimators provide detailed and dynamic beam shaping and, if available, can be used instead of Cerrobend blocks with reduction in treatment delivery times. The beam’s eye view facility may improve selection of field configurations and provides detailed information about the projection of the beam in relation to normal organs. However, the use of field shaping to match the target volume is dependent on the reliability of CT interpretation and the accuracy of definition of margins around the tumour. Great care must be used to ensure that elaborate shielding does not inadvertently obscure dose to the tumour.
Heterogeneity corrections Lung tissue attenuates the radiation beam less than other soft tissues and this factor alters both the shape of the dose distribution and the value of the isodoses. For irradiation of tumours of the breast, chest wall, lung, oesophagus, etc. in the thorax it is recommended that the lung tissue is localized on to the outline with the target volume. Either a correction factor of 0.25–0.3 can be applied or, using CT planning, a pixel by pixel correction can be made, taking into account the presence of pleural fluid, atelectasis and other abnormal as well as normal lung densities. It is not usually necessary to make similar corrections in the pelvis, as differences may be variable, e.g. gas in the bowel and rectum. If iodinated contrast medium has been used for CT planning scans (e.g. to outline small bowel or bladder), this may affect heterogeneity corrections if they are applied automatically, thereby introducing an error.
Choice of dose specification and reporting The International Commission for Radiation Units and Measurements (ICRU) published Report 50 in 1993 to encourage a common international language for the reporting and recording of radiation dose.5 The recommendations of this report have been maintained in Reports 62 and 71. The recommendation is that doses are reported to the ICRU reference point. This point is selected according to the following general criteria: 1. The dose at the point should be clinically relevant and representative of the dose throughout the planning target volume (PTV).
2. The point should be easy to define in a clear and unambiguous way. 3. The point should be selected where dose can be accurately determined (physical accuracy). 4. The point should be in a region where there is no steep dose gradient. These criteria are fulfilled if the ICRU reference point is located at the centre (or in a central part) of the PTV and (when possible) on (or near) the central axis of the beam(s). The dose to the ICRU reference point as a basic requirement should be reported with the maximum and minimum dose to the PTV. This indicates the dose profile to the PTV and describes the heterogeneity of the irradiation throughout the CTV. It is recommended that the dose variation in the PTV should be within 7% and 5% of the prescribed CTV dose. The minimum target dose is an important parameter because it correlates with the probability of tumour control. However, there is heterogeneity of tumour cell population throughout the target volume, commonly with the highest cell density at the centre of the GTV and the lowest cell density at the periphery of the clinical target volume. Although it is difficult to predict the biological effect of a given dose because of these variations, it is important to have an internationally accepted method of dose specification in order to assess effectiveness of treatment and to allow comparison between radiotherapy centres. Traditionally, many centres used the minimum isodose around the target volume as a dose specification point. However, this has several disadvantages. The minimum isodose frequently lies in a region of steep dose gradient at the periphery of the volume and it also lies at the edge of the beam, where calculation of dose is less accurate than on the central axis of the beam. The minimum isodose lies on the periphery of the target volume where there is great inter-clinician variability as to its site and size and where the cell density is believed to be lowest. The minimum isodose may also be 10–15 per cent lower than the maximum isodose and, as such, is not a good description of the entire target volume. When changes are made in a radiotherapy department from specifying dose at the minimum isodose to the new ICRU point, then for each treatment technique, dose plans should be normalized both to the old and new specification point and a clinical decision made as to whether there needs to be a change in the prescribed dose. For example, if a dose of 50 Gy has hitherto been given to the minimum or 90 per cent isodose, this will have given a dose of 55 Gy to the 100 per cent isodose. If the dose specification point is changed to the intersection of the beams, and this lies on or near the 100 per cent isodose, then a decision has to be made as to whether the dose prescription is changed to 55 Gy to the new dose specification point, which will give the same clinical result. The ICRU reports recommend three levels of dose evaluation for reporting. Level 1 constitutes the standard below
Verification of treatment 1249
100
% Volume
80 60 40 20 0 0
20
40
60
80
100
Isodose
Figure 50.17 Example of a dose–volume histogram for a treatment of the bronchus. The upper curve is the PTV and the lower curve is the DVH for the lung. (Reprinted from Williams JR, Thwaites DI (ed) Radiotherapy Physics: in Practice, 2nd edn. Oxford: Oxford University Press, 2000.)
which safe and accurate radiotherapy cannot be performed and the dose to the ICRU reference point with an estimate of maximum and minimum doses to the PTV are reported. At level 2, information such as absorbed doses to the GTV, CTV, OR, PTV, PRV, and complete dose distributions within a full quality assurance programme are reported. This allows the exchange of more complete and relevant information between different centres. Level 3 includes the development of new techniques for which reporting criteria are not yet established. At all levels any relevant additional information should be given, e.g. dose–volume histograms (DVHs) and the absorbed doses to the organs at risk. A cumulative dose–volume histogram is a useful graphical way of displaying dose distributions (see Fig. 50.17) It can show how much of the PTV is above/below the prescribed dose and also display the DVH for each organ at risk or PRV. This is particularly useful in assessing which of several plans is the best. The ICRU has also made recommendations on reporting series of patients. When reporting treatments in scientific journals the prescribed CTV and PTV and corresponding doses should be illustrated in an isodose distribution chart, giving the total absorbed doses in Gy.
VERIFICATION OF TREATMENT On the first day of treatment all the treatment planning parameters are checked and the patient is positioned on the treatment couch according to the instructions on the treatment prescription sheet, which includes details of immobilization, fixation devices, position of the arms, full or empty bladder, and a photograph as a helpful visual record. Laser beams are used to align the patient according to permanent skin tattoos in both the midline and lateral
planes in order to ensure reproducibility of the treatment set-up. Machine parameters are checked and treatment verified, both for geometry and for dose delivered. Geometric verification can be achieved with a radiograph taken on the treatment unit (portal film) or by real-time visualization of the patient using an on-line electronic portal imaging device. In 2002, the Royal College of Radiologists, Institute of Physics and College of Radiographers produced a joint document The Development and Implementation of Conformal RT in the UK. This includes guidelines on sitespecific verification protocols. An audit of current UK practice shows wide variation at individual sites and between departments.87 It is essential that local guidelines are drawn up to identify the acceptable limits of variation, correlated with the electronic margin around the clinical target volume that has been chosen to ensure that the planning target volume allows for day-to-day variability. In vivo dosimetry can be performed using either lithium fluoride thermoluminescent dosimetry or silicon diodes, and this provides a final check of the overall accuracy of the dose delivered. A quality assurance programme for therapy equipment and simulators includes check of geometry, dosimetry, beam data, wedges and other accessories. National and regional dosimetry intercomparison studies between centres ensure the validity of output measurements. Quality control protocols for computerized planning systems should be adhered to, in order to provide accurate dose planning calculations. Radiographers are instructed in the importance of daily checks on patient positioning using laser alignment, measurement of patient separation, fitting the immobilization devices and attention to detailed instructions for individual patients. The aim of an ongoing quality assurance programme is to eliminate any errors during the multiple steps in the planning and delivery of radiation treatment and is an essential part of the work of a radiation oncology department. Despite improvements in the quality of electronic portal imaging, their main limitation continues to be the twodimensional nature and quality of the images they provide. Several studies confirm that good bony alignment does not automatically translate into good soft-tissue alignment.88 Newer imaging modalities, such as X-ray volumetric imaging (XVI), have the ability to provide detailed threedimensional kV cone beam computed tomography images of a patient on the treatment couch, enabling soft-tissue registration to overcome problems produced by organ motion and deformation. McBain et al. reported on single XVIs acquired from 30 patients undergoing RT at ten different anatomical sites.89 Soft-tissue definition of organ boundaries allowed direct assessment of 3D target volume coverage at all sites. Individual image quality depended on both imaging parameters and patient characteristics. Radiation dose ranged from 0.003 Gy in the head to 0.03 Gy in the pelvis. This technology sets a new standard in treatment verification and will facilitate novel adaptive radiotherapy techniques.
1250 Principles of external beam radiotherapy planning techniques
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
The choice of equipment used for localization will depend on the workload and range of patients and tumour sites to be treated, the techniques employed in planning and irradiation and the resources available. It has been shown that CT scanning information alters the radiation fields used in approximately 30 per cent of patients receiving radical radiotherapy treatment. The process of determining tumour and target volumes for treatment of malignant disease consists of several distinct steps, which have been well described in ICRU Report 50 (5), ICRU Report 62 (6) and ICRU Report 71 (7). Integration of CT, MRI and SPECT/PET data using image registration may revolutionize the acquisition of clinical and tumour data. Dose–response curves show that a small variation in dose level can have a considerable influence on the probability of tumour control and also on the likelihood of normal tissue complications. The ICRU Report 24 considers that 5 per cent accuracy is required in the delivery of absorbed dose. Brahme et al.83 and Mijnheer et al.84 suggested that the requirement for accuracy should be 3.5 per cent, one standard deviation, in the dose value at the specification point in the target volume. The ICRU dose specification point is chosen at the centre of the target volume because it is easy to determine, representative of the dose distribution and does not lie in a peripheral steep dose gradient. It is essential that the dose at the dose specification point is accompanied by the maximum and minimum target doses in order to describe the homogeneity of the irradiation throughout the target volume. For three-dimensional treatment plans, it is recommended that the dose at the ICRU point, maximum and minimum doses and, in addition, the mean dose and dose–volume histogram are all recorded. The aim of an ongoing quality assurance programme is to eliminate any errors during the multiple steps in the planning and delivery of radiation treatment and is an essential part of a radiation oncology department.
KEY REFERENCES Dobbs J, Barrett A, Ash D. Basic principles of treatment planning. In: Dobbs J, Barrett, A, Ash D (eds) Practical Radiotherapy Planning, 3rd edn. London: Arnold, 1999. Hurkmans CW, Remeijer P, Lebesque JV, Mijnheer BJ. Set-up verification using portal imaging: review of current clinical practice. Radiother Oncol 2001; 58: 105–20. Husband JES, Reznek RH (eds) Imaging in Oncology, vols I and II. Oxford: Isis Medical Media, 1998. Landberg T, Chavaudra J, Dobbs HJ, et al. International Commission on Radiation Units and Measurements, ICRU Report 62, Supplement to ICRU Report 50. Prescribing, Recording, and Reporting Photon Beam Therapy. Bethesda, Maryland: ICRU, 1999. Emami B, Lyman J, Brown A, Coia L, Goitein M, Munzenrider JE, Shank B, Solin LJ, Wesson M. Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 1991; 21(1):109–22.
REFERENCES 1 Williams FH. The Roentgen Rays in Medicine and Surgery. New York: McMillan Co, 1901. 2 Ginzton EL, Craig SN. History of microwave electron linear accelerators for radiotherapy. Int J Radiat Oncol Biol Phys 1984; 11:205–216. 3 Cancer Research UK (www.cancerresearchuk.org). 4 Bel A, Bartelink H, Vijlbrief RE, Lebesque J. Transfer errors of planning CT to simulator: a possible source of set-up inaccuracies? Radiother Oncol 1994; 31: 176–80. 5 * Landberg T, Chavaudra J, Dobbs HJ, et al. International Commission on Radiation Units and Measurements ICRU Report 50. Prescribing, Recording, And Reporting Photon Beam Therapy. Bethesda, Maryland: ICRU, 1993. 6 Landberg T, Chavaudra J, Dobbs HJ, et al. International * Commission on Radiation Units and Measurements, ICRU Report 62, Supplement to ICRU Report 50. Prescribing, Recording, And Reporting Photon Beam Therapy. Bethesda, Maryland: ICRU, 1999. 7 Landberg T, Chavaudra J, Dobbs HJ, et al. International * Commission on Radiation Units and Measurements, ICRU Report 71, Supplement to ICRU Report 62. Prescribing, Recording, And Reporting Electron Beam Therapy. J ICRU 2004; 4(1):1–100. 8 Dutreix A. When and how can we improve precision in radiotherapy? Radiother Oncol 1984; 2:275–92. ◆9 Last A. Radiotherapy in patients with cardiac pacemakers. Br J Radiol 1998; 71(841):4–10. 10 Mazdai G, Stewart DP, Hounsell AR. Radical radiation therapy in a patient with head and neck cancer and severe Parkinson’s disease. Clin Oncol (R Coll Radiol) 2006; 18:82–4. 11 Bentel GC, Marks LB, Hendren K, Brizel DM. Comparison of two head and neck immobilization systems. Int J Radiat Oncol Biol Phys 1997; 38(4):867–73.
References 1251
12 Weltens C, Kesteloot K, Vandevelde G, Van den Bogaert W. Comparison of plastic and ORFIT masks for patient head fixation during radiotherapy. Int J Radiat Oncol Biol Phys 1995; 33(2):499–507. 13 Graham JD, Warrington AP, Gill SS, Brada M. A non invasive, relocatable stereotactic frame for fractionated radiotherapy and multiple imaging. Radiother Oncol 1991; 21:60–2. 14 Humphreys M, Guerrero Urbano MT, Mubata C, Miles E, Harrington KJ, Bidmead M, Nutting CM. Assessment of a customised immobilisation system for head and neck IMRT using electronic portal imaging. Radiother Oncol 2005; 77(1):39–44. ◆15 McKenzie AL, Dobbs HJ, Landberg T, et al. Geometric Uncertainties in Radiotherapy Planning. 36 Portland Place, London W1B 1AT, UK: The British Institute of Radiology, 2003. ●16 Hanley J, Debois MM, Mah D, et al. Deep inspiration breath hold technique for lung tumours: the potential value of target immobilization and reduced lung density in dose escalation. Int J Radiat Oncol Biol Phys 1999; 45:603–11. ●17 Wong JW, Sharpe MB, Jaffray DA, et al. The use of active breathing control (ABC) to reduce margin for breathing motion. Int J Radiat Oncol Biol Phys 1999; 44:911–9. 18 Van Tienhoven G, Lanson JH, Crabeels D, et al. Accuracy in tangential breast treatment set-up: a portal imaging study. Radiother Oncol 1991; 22:317–22. 19 Holmberg O, Huizenga H, Idzes MHM, et al. In vivo determination of the accuracy of field matching in breast cancer irradiation using an electronic portal imaging device. Radiother Oncol 1994; 33:157–66. 20 Mitine C, Dutreix A, Van der Schueren E. Tangential breast irradiation: influence of technique of set up on transfer errors and reproducibility. Radiother Oncol 1991; 22:308–10. 21 Gagliardi G, Lax I, Rutqvist LE. Radiation therapy of stage 1 breast cancer: analysis of treatment technique accuracy using three dimensional treatment planning tools. Radiother Oncol 1992; 24:94–101. 22 Grann A, McCormick B, Chabner ES, et al. Prone breast radiotherapy in early stage breast cancer: a preliminary analysis. Int J Radiat Oncol Biol Phys 2000; 47:319–25. 23 Merchant TE, McCormick B. Prone position breast irradiation. Int J Radiat Oncol Biol Phys 1994; 30:197–203. 24 Latimer JG, Beckham W, West M, Holloway L, Delaney G. Support of large breasts during tangential irradiation using a micro-shell and minimizing the skin dose – a pilot study. Med Dosim 2005; 30(1):31–5. 25 Olofsen-van Acht M, van den Berg H, Quint S, et al. Reduction of irradiated small bowel volume and accurate patient positioning by use of a bellyboard device in pelvic radiotherapy of gynaecological cancer patients. Radiother Oncol 2001; 59(1):87–93. 26 Griffiths SE, Khoury GG, Eddy A. Quality control of radiotherapy during pelvic irradiation. Radiother Oncol 1991; 20:203–6.
●27
28
29
*30
31
32
33
34
35
36
●37
38
39
40
Bayley AJ, Catton CN, Haycocks T, et al. A randomized trial of supine vs. prone positioning in patients undergoing escalated dose conformal radiotherapy for prostate cancer. Radiother Oncol 2004; 70(1):37–44. Ten Haken RK, Former JD, Heimberger DK, et al. Treatment planning issues related to prostate movement in response to differential filling of the rectum and bladder. Int J Radiat Oncol Biol Phys 1991; 20:1317–24. Balter JM, Sandler HM, Lam K, et al. Measurement of prostate movement over the course of routine radiotherapy using implanted markers. Int J Radiat Oncol Biol Phys 1995; 31:113–8. van Herk M, Remeijer P, Rasch C, Lebesque JV. The probability of correct target dosage: dose-population histograms for deriving treatment margins in radiotherapy. Int J Radiat Oncol Biol Phys 2000; 47(4):1121–35. Beard CJ, Kijewski P, Bussiere M, et al. Analysis of prostate and seminal vesicle motion: implications for treatment planning. Int J Radiat Oncol Biol Phys 1996; 34:451–8. Zelefsky MJ, Crean D, Mageras GS, et al. Quantification and predictors of prostate position variability in 50 patients evaluated with multiple CT scans during conformal radiotherapy. Radiother Oncol 1999; 50:225–34. Pinkawa M, Asadpour B, Gagel B, Piroth MD, Holy R, Eble MJ. Prostate position variability and dose-volume histograms in radiotherapy for prostate cancer with full and empty bladder. Int J Radiat Oncol Biol Phys 2005; 63:S306–7. Fiorino C, Reni M, Bolognesi A, et al. Set-up error in supine positioned patients immobilised with two different modalities during conformal radiotherapy of prostate cancer. Radiother Oncol 1998; 49:133–41. Soffen D, Hanks G, Huang C, Chu J. Conformal static field therapy for low volume low grade prostate cancer with rigid immobilisation. Int J Radiat Oncol Biol Phys 1991; 20:141–6. Catton C, Lebar L, Warde P, et al. Improvement in total positioning error for lateral prostatic fields using a soft immobilisation device. Radiother Oncol 1997; 44(3):265–70. Dearnaley D, Shoo V, Norman A, et al. Comparison of radiation side-effects of conformal and conventional radiotherapy in prostate cancer: a randomised trial. Lancet 1999; 353:267–71. Song P, Washington M, Vaida F. A comparison of four patient immobilisation devices and the treatment of prostate cancer patients with three-dimensional conformal radiotherapy. Int J Radiat Oncol Biol Phys 1996; 34:213–19. Dobbs HJ. From GTV (gross tumour volume) to PTV (planning tumour volume). In: Minet P (ed) Three Dimensional Treatment Planning. Proceedings of 5th Workshop organized by Commission Informatique, European Association of Radiology, Liège, 1993. Sur RK, Clinkard J, Jones WG, et al. Changes in target volume during radiotherapy treatment of invasive bladder carcinoma. Clin Oncol 1993; 5:30–3.
1252 Principles of external beam radiotherapy planning techniques
41 Turner SL, Swindell R, Bowl N, et al. Bladder movement during radiation therapy for bladder cancer: implications for treatment planning. Int J Radiat Oncol Biol Phys 1997; 39(2):355–60. 42 Harris SJ, Buchanan RB. An audit and evaluation of bladder movements during radical radiotherapy. Clin Oncol 1998; 10:262–4. 43 Muren LP, Smaaland R, Dahl O. Organ motion, set-up variation and treatment margins in radical radiotherapy of urinary bladder cancer. Radiother Oncol 2003; 69(3):291–304. 44 Hurkmans CW, Remeijer P, Lebesque JV, Mijnheer BJ. Set-up verification using portal imaging: review of current clinical practice. Radiother Oncol 2001; 58:105–20. *45 Husband JES, Johnson RJ, Reznek RH. A Guide to the Practical Use of MRI in Oncology. London: Board of the Faculty of Clinical Radiology, The Royal College of Radiologists, 1999. 46 Dobbs HJ, Parker RP. The respective roles of the simulator and computed tomography in radiotherapy planning: a review. Clin Radiol 1984; 35:433–9. *47 Royal College of Radiologists. Imaging for Oncologists. London: RCR, 2004. 48 Williams JR, Thwaites DI (eds). Radiotherapy Physics: in Practice, 2nd edn. Oxford: Oxford University Press, 2000. 49 Husband JES, Reznek RH (eds) Imaging in Oncology, vols I and II. Oxford: Isis Medical Media, 1998. 50 Denham JW, Hamilton CS, Joseph DJ, et al. The use of simulator and CT information in the planning of radiotherapy for non-small cell lung cancer: Australasian patterns of practice study. Lung Cancer 1992; 8:275–84. 51 Ash DV, Andrews B, Stubbs B. A method for integrating a CT scanner into radiotherapy planning and treatment. Clin Radiol 1983; 34:99–101. 52 Urie MM, Goitein M, Doppke K, et al. The role of uncertainty analysis in treatment planning. Int J Radiat Oncol Biol Phys 1991; 21:91–107. 53 Perez CA, Purdy JA, Harris W, et al. Three-dimensional treatment planning and conformal radiation therapy: preliminary evaluation. Radiother Oncol 1995; 36:32–43. 54 Goitein M. Applications of computed tomography in radiotherapy treatment planning. NewYork: Plenum Press, 1982, 195–287. 55 Dobbs HJ, Parker RP, Hodson NJ, et al. The use of CT in radiotherapy treatment planning. Radiother Oncol 1983; 1:133–41. 56 Rothwell RI, Ash DV, Thorogood J. An analysis of the contribution of computed tomography to the treatment outcome in bladder cancer. Clin Radiol 1985; 36:369–72. 57 Tsujii H, Kamada T, Matsuoka Y, et al. The value of treatment planning using CT and an immobilising shell in radiotherapy for paranasal sinus carcinomas. Int J Radiat Oncol Biol Phys 1989; 16:243–9. 58 Goitein M. The utility of computed tomography in radiation therapy: an estimate of outcome. Int J Radiat Oncol Biol Phys 1979; 5:1799–807.
59 Verellen D, Vinh-Hung V, Bijdekerke P, et al. Characteristics and clinical application of a treatment simulator with CT-option. Radiother Oncol 1999; 50:355–66. 60 Dobbs HJ, Webb S. Clinical applications of X-ray computed tomography in radiotherapy planning. In: Webb S (ed) The Physics of Medical Imaging. Bristol: Adam Hilger, 1988. 61 Vincent D, Beckham W, Delaney G. An assessment of the number of CT slices necessary to plan breast radiotherapy. Radiother Oncol 1999; 52:179–83. 62 Bornstein BA, Cheng CW, Rhodes LM, et al. Can simulation measurements be used to predict the irradiated lung volume in the tangential fields in patients treated for breast cancer? Int J Radiat Oncol Biol Phys 1990; 18:181–7. ●63 Neal AJ, Yarnold JR. Estimating the volume of lung irradiated during tangential breast irradiation using the central lung distance. Br J Radiol 1995; 68:1004–8. 64 Hurkmans CW, Borger JH, Bos LJ, et al. Cardiac and lung complication probabilities after breast cancer irradiation. Radiother Oncol 2000; 55:145–51. 65 Driver DM, Drzymala M, Dobbs HJ, Faulkner S, Harris S. Virtual simulation in palliative lung radiotherapy. Clin Oncol (R Coll Radiol) 2004; 16(7):461–6. 66 Khoo VS, Padhami AR, Tanner SF, Finnigan DJ, Leach MO, Dearnaley DP. Comparison of MRI with CT for the radiotherapy planning of prostate cancer: a feasibility study. Br J Radiol 1999; 72:590–7. 67 Milosevic M, Voruganti S, Blend R, et al. MRI for localisation of the prostate apex: a comparison to CT and urethrography. Radiother Oncol 1998; 47(3):227–84. 68 Helyer SJ, Moskovic E, Ashley S, et al. A study testing the routine use of ultrasound measurements when selecting the electron energy for breast boost radiotherapy. Clin Oncol 1999; 11(3):164–8. 69 Rohren EM, Turkington TG, Coleman RE. Clinical applications of PET in oncology. Radiology 2004; 231(2):305–32. 70 Nestle, U, Walter, K, Schmidt, S, et al. 18F-deoxyglucose positron emission tomography (FDG-PET) for the planning of radiotherapy in lung cancer: high impact in patients with atelectasis. Int J Radiat Oncol Biol Phys 1999; 44:593–7. 71 Rosenzweig KE, Dladla N, Schindelheim R, et al. Threedimensional conformal radiation therapy (3D-CRT) for early-stage non-small-cell lung cancer. Clin Lung Cancer 2001; 3:141–4. 72 Balogh JB, Caldwell CB, Ung YC, et al. Interobserver variation in contouring gross tumour volume in carcinoma of the lung. The impact of 18FDG – hybrid PET fusion. Radiother Oncol 2000; 56(suppl1):59. 73 Cai J, Chu JCH, Recine D, et al. CT and PET lung image registration and fusion in radiotherapy treatment planning using the chamfer-matching method. Int J Radiat Oncol Biol Phys 1999; 43:883–91. 74 Morris, DE, Rosenman, JG, Halle, JS, et al. Induction chemotherapy followed by dose-escalated thoracic 3-dimensional conformal radiation therapy to 90 Gy with concurrent chemotherapy in stage IIIA/B non-small cell
References 1253
75
76
77
78
79
◆80
81
lung cancer: a phase I radiation dose escalation trial. Int J Radiat Oncol Biol Phys 2003; 57:S138–9. Hutchings M, Mikhaeel NG, Fields PA, Nunan T, Timothy AR. Prognostic value of interim FDG-PET after two or three cycles of chemotherapy in Hodgkin lymphoma. Ann Oncol 2005; 16(7):1160–8. Douglas JG, Stelzer KJ, Mankoff DA, et al. [F-18]fluorodeoxyglucose positron emission tomography for targeting radiation dose escalation for patients with glioblastoma multiforme: Clinical outcomes and patterns of failure. Int J Radiat Oncol Biol Phys 2006; 64(3):886–91. Epub 2005 Oct 19. Grosu AL, Piert M, Weber WA, et al. Positron emission tomography for radiation treatment planning. Strahlenther Onkol 2005; 181(8):483–99. Leunens G, Menten J, Weltens C, Verstraete J, Van der Scheuren, E. Quality assessment of medical decision making in radiation oncology: variability in target volume delineation for brain tumours. Radiother Oncol 1993; 29:169–75. Seddon B, Bidmead M, Wilson J, Khoo V, Dearnaley D. Target volume definition in conformal radiotherapy for prostate cancer: quality assurance in the MRC RT-01 trial. Radiother Oncol 2000; 56:73–83. Emami B, Lyman J, Brown A, et al. Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 1991; 21(1):109–22. McKenzie AL. How should breathing motion be combined with other errors when drawing margins around clinical target volumes? Br J Radiol 2000; 73:973–7.
82 Ekberg L, Holmberg O, Wittgren L, et al. What margins should be added to the clinical target volume in radiotherapy treatment planning for lung cancer? Radiother Oncol 1998; 48:71–7. 83 Brahme A, Chavaudra J, Landberg T, et al. Accuracy requirements and quality assurance of external beam therapy with photons and electrons. Acta Oncol 1988; 27(suppl. 1). 84 Mijnheer BJ, Battermann JJ, Wambersie, A. What degree of accuracy is required and can be achieved in photon and neutron therapy? Radiother Oncol 1987; 8:237–52. 85 Chassagne D, Dutreix J, Dutreix A. Report on a systematic overdosage of patients in 1970 and 1971. Internal report. Villejuif: Institut Gustave Roussy, 1976. 86 Dobbs J, Barrett A, Ash D. Basic principles of treatment planning. In: Dobbs J, Barrett A, Ash D (eds) Practical Radiotherapy Planning, 3rd edn. London: Arnold, 1999. 87 Stratford J, Ball K, Henry AM, Cullen JN, Swindell R, Price P, Jain P. Radiotherapy treatment verification in the UK: an audit of practice in 2004. Clin Oncol (R Coll Radiol) 2006; 18(1):15–22. 88 Vigneault E, Pouliot J, Laverdiere J, Roy J, Dorion M. Electronic portal imaging device detection of radioopaque markers for the evaluation of prostate position during megavoltage irradiation: a clinical study. Int J Radiat Oncol Biol Phys 1997; 37:205–12. ●89 McBain CA, Henry AM, Sykes J, et al. X-ray volumetric imaging in image-guided radiotherapy: The new standard in on-treatment imaging. Int J Radiat Oncol Biol Phys 2006; 64(2):625–34.
51 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy VINCENT KHOO
Introduction Conformal radiotherapy Intensity-modulated radiotherapy
1254 1254 1260
INTRODUCTION The aim of radiotherapy is to deliver the desired tumoricidal dose to the intended target as accurately and reliably as possible and avoid unnecessary dose to surrounding normal tissues. This principle was recognized when irradiation was utilized therapeutically over 100 years ago and has been the guiding force for development and refinement of modern radiotherapeutic practice and strategy. Radiotherapy has always relied heavily on the fields of medical physics, biology, computer science, and engineering for its development. Prior to the 1960s, radiotherapy was limited by factors that included suboptimal definition of tumour/target volumes, inadequate treatment fields shaping, poor depth–dose profiles, time consuming manual planning and dose calculations as well as limited tools for evaluation and verification. This was due to limitations in technology and their translation into practice. Since the 1960s, there have been several major sequential technological advances in physics, imaging, computing and medical hardware that have led to the current modern radiotherapy standards. Firstly, low energy photons from cobalt-60 and betatron units were replaced by higher energy photons produced by linear accelerators (LINACs). These modern LINACs provide substantially improved depth–dose profiles with better skin sparing effects, suitable dose delivery rates, are computer controlled, more reliable and form the backbone of modern radiotherapy departments. Secondly, the development of cross-sectional imaging with computed tomography (CT) has enabled the introduction of 3D planning. Thirdly, these
Image-guided radiotherapy References
1268 1276
developments have relied on and paralleled developments in physics and computer technology for increased microprocessor capacity, multileaf collimators for automated beam shaping, 3D image processing, enhanced anatomical graphics, more efficient data storage and increased computational power to run complex sophisticated algorithms. These advances have heralded 3D radiotherapy as the standard method for modern centres. More recent developments include 3D imaging on the LINAC itself or during radiotherapy delivery. This new technology is currently being investigated for image-guided radiotherapy strategies. This chapter will describe the concept, rationale, and methodology for conformal radiotherapy (CFRT), intensitymodulated radiotherapy (IMRT) and image-guided radiotherapy (IGRT). Some examples of their application in clinical practice and future developments will be outlined. The physical basis of these techniques is reliant on the principles of radiotherapy treatment planning outlined in Chapter 50 (Planning techniques) and builds upon many of its standard components.
CONFORMAL RADIOTHERAPY Concept and rationale It can be an immense and difficult technical task to deliver a tumoricidal dose to a tumour as they are often located close to important radiosensitive organs, such as nerves, bowels, kidneys, lungs, and heart tissue. The radiotolerance of these organs-at-risk (OAR) needs to be respected to avoid damaging them irrevocably and to permit adequate doses to the
Conformal radiotherapy 1255
I
II
III
Percentage (%)
100 Local Tumour Control Curve
Radiotherapyrelated Normal Tissue Complication Curve
50
0
A
B
C
Increasing dose
Figure 51.1 The therapeutic ratio in radiotherapy showing sigmoid dose-response curves for local tumour control (solid curve I) and radiation-related normal tissue complications (dotted curves II and III). Different clinical outcomes may be produced by different clinical prescriptions depending on where the two curves are situated. Consider the situation for curves I and II only. For example, the dose at A would produce no complications but only achieve minimum local control. The dose at C would produce higher local control but with considerable complications. The dose at B would be considered optimal because it is would produce the highest local control rate with the least complications. If the radiation-related normal tissue complications curve can be displaced further to the right (as demonstrated by dotted curve III) this would produce a more favourable therapeutic ratio as the dose prescription at C can now offer a higher probability of local control with equivalent or lower rate of complications compared with the dose prescription at B.
tumour. It needs to be targeted appropriately to achieve this aim. The concept of CFRT was first coined in the 1960s.1 As this descriptive term suggests, the prescribed high-dose region is made to ‘conform’ (or shaped) to the intended tumour or target volume. The goal of CFRT is to improve the radiotherapy therapeutic ratio. The therapeutic ratio in radiotherapy relates dose to local control and expected treatment complications (Fig. 51.1). The curves of local control and treatment-related complications are sigmoid in shape.2 The gradient of the curves will vary for different tumours and complications in different OAR. The therapeutic ratio can be improved by displacing the local tumour control and treatment-related complication curves further apart from each other as illustrated by the three different scenarios outlined. This is an important strategy as the dose that can be delivered to any target volume is limited by normal tissue tolerance. Conformal radiotherapy attempts to improve the therapeutic ratio in two ways. Firstly, CFRT tailors each radiotherapy field to the shape of the tumour so that the high-dose distribution is also shaped appropriately and consequently the quantity of adjacent normal tissue/OAR can be reduced. In turn, this can reduce side effects or decrease the normal tissue complication probability (NTCP). Secondly, by more accurate ‘target’ shaping and
simultaneous normal tissue sparing, CFRT can potentially permit safer radiation dose escalation for improved local control or tumour control probability (TCP) with clinically acceptable morbidity rates as described in Figure 51.1. It is postulated that if the dose can be escalated by 20 per cent, this may provide improved TCP and therefore increase cancer specific survival for some cancer types.3–6* In order to achieve this in radiotherapy, there must be a dose-response effect for TCP and the constraints on dose delivery or volume effect for NTCP must be overcome. The clinical benefit for CFRT is evidenced from a randomized trial that compared conventional unblocked fields and CFRT fields in men with localized prostate cancer.7** In this trial, CFRT treatments were significantly associated with less radiation-related proctitis and bleeding compared with the conventional treatments. Toxicity for RTOG Grade 1 or higher was 37 per cent compared with 56 per cent for the CFRT and conventional techniques respectively (p 0.004) and RTOG Grade 2 toxicity or higher was 5 per cent compared with 15 per cent respectively (p 0.01). Conformal radiotherapy reduced the volume of irradiated rectum without compromising local tumour control (median follow-up 3.6 years; local control from CFRT 78 per cent with 95 per cent confidence interval [CI]: 66–86 and conventional 83 per cent with 95 per cent CI: 69–90). This Level I evidence provided the basis for the National Institute of Clinical Excellence (NICE) UK to recommend CFRT as the minimum radiotherapeutic technique for prostate cancer.8 There is no reason not to believe that the benefits of CFRT can also be translated to other clinical sites.
Basic principles in CFRT The radiotherapy treatment planning process is outlined in Figure 50.1 (see Chapter 50, Planning techniques) and is similar for CFRT. For CFRT, there are some issues worth emphasizing. In CFRT, each treatment field aperture will be shaped to the profile of the target volume in the treatment portal or the beam’s-eye-view (BEV). This BEV is that of a virtual observer as seen along the projection of the treatment field from the source of radiation within the LINAC. This visual perspective allows a projection of the target in relation to the treatment field edges. Multiple treatment fields will be individually tailored for its designated beam orientation in this manner to create a better high-dose region ‘fit’ to the target compared with conventional methods where treatment fields are either rectangular or square shapes (Fig. 51.2). Conformal radiotherapy provides for better dose conformity and high-dose sparing to OAR than conventional techniques. In certain cases, where the target volume is concave in shape, the dose-limiting OAR is too close to the target volume or if the target is moving, then CFRT field configurations may not provide the necessary therapeutic advantage and more complex beam techniques such as IMRT or strategies such as IGRT may be needed.
1256 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
Figure 51.2 This is a beam’s-eye view (BEV) of a single treatment portal used in prostate radiotherapy. The target volume of the prostate gland (green) is located between the organs at risk, which are the bladder (blue) anteriorly and the rectum (red) posteriorly. Using conformal radiotherapy (CFRT) techniques, the treatment field (blue line) is shaped around the profile of the prostate gland in the BEV compared with the conventional treatment field (black dotted line), which is an unshaped rectangular field. It is clear that the unshaped rectangular field will irradiate substantially more of the bladder and rectum compared with the conformal field. (See Plate Section.)
Treatment planning and delivery in CFRT The planning principles outlined in Chapter 50 (Planning techniques) also applies to CFRT. In CFRT some of the planning steps are quite involved because of the need for field shaping that necessitates 3D imaging, 3D planning, use of 3D evaluation tools and optimization algorithm(s) where appropriate, application of multileaf collimators and quality assurance (QA). 3D IMAGING IN CFRT
The foundation for CFRT is through spatial and 3D assessment of the tumour location and shape. This concept is simple and is close to being understated. Evaluation of the target volume with its surrounding anatomy in 3D aids not only in the treatment field shaping process but also the treatment plan optimization in terms of number of fields, beam orientations, treatment set-up and radiotherapy verification. Inherent in the definition of these volumes for planning are the guidelines from the International Commission for Radiation Units (ICRU),9,10 which are discussed in depth in Chapter 50 (Planning techniques). They are applicable for CFRT, IMRT and IGRT.
Figure 51.3 A model view of a 3-field coplanar conformal radiotherapy (CFRT) plan within the abdomen. This 3D reconstruction of the patient shows the orientation of the beams and their coverage of the clinical target volume (solid blue colour) located in the middle of the picture and covered by the planning target volume (red wire colour). The surrounding normal structures have been outlined: the bowel (translucent brown colour), spinal cord (solid light blue colour), right kidney (solid yellow colour) and liver (translucent green colour). The conformal field shapes are shown as they should appear projected against the bony anatomy of the patient and are used to verify the positioning of the beams daily during patient set-up with electronic portal imaging. (See Plate Section.)
The use of CT has enable CFRT. Early non-randomized studies that compared the contribution of CT with conventional imaging studies (X-rays, xerograms, lymphangiograms, arteriograms, air contrast studies, isotopic studies and ultrasound) reported that 52 per cent of cases had their conventional radiotherapy plans changed as a result of using CT, due to inadequate coverage of the target volume in 42 per cent of plans, and 31 per cent of cases had some portion of the target volume outside of the 50 per cent isodose region with conventional planning.11* Using an empirical mathematical model, it was suggested that improved target volume definition using CT could result in an increase in local TCP by an average of 6 per cent, with the chance of 5-year survival increasing by an average of 3.5 per cent simply by avoiding missing the target.12 Another advantage of CT-based planning is that dosimetry can now take into account the different 3D profiles through the body and its contours (Fig. 51.3); internal tissue inhomogeneities such as bone, air and soft tissues via direct correlation of the Hounsfield units into electron density measurements; permit more complex dose calculation algorithms and provide 3D displays of dose distributions (see Value of planning evaluation tools). Correct treatment volume definition forms the foundation in the treatment planning process. It is important to emphasize that target volume definition may not be wholly adequate using CT alone. Other imaging modalities such as magnetic resonance imaging (MRI) and positron emission tomography (PET) can provide improved means of
Conformal radiotherapy 1257
assessing and defining tumour extent as well as providing biological and functional information unavailable from morphological imaging. These imaging aspects are discussed in Chapter 6 (Tumour imaging in oncology). The ability to combine the best aspects from each imaging modality (i.e., multimodality imaging) can augment as well as provide complementary and distinct information to further improve the clinician’s ability in accurately and reliably defining the appropriate tumour or subclinical tumour volumes for radiotherapy. An important consideration in the use of these new imaging modalities for radiotherapy is that clinicians need to have the necessary training to utilize them appropriately.13 It is of value to liaise closely with diagnostic colleagues who have cancer subsite(s) expertise even if the oncologist has relevant expertise. This approach has been endorsed by the Royal College of Radiologists, UK.14 In the ideal situation, there should be an oncological team approach for the definition of radiotherapy target volumes, similar in concept to the multidisciplinary team arrangement for subsite cancer management. USE OF OPTIMIZATION ALGORITHMS AND CLASS SOLUTIONS
Once the planning target volume (PTV) has been created, a treatment plan must be devised. Conventionally, a ‘forward’ planning process is used, whereby the planner/clinician would initiate a probable plan based on experience and intuition. Variations or iterations of this initial plan and its parameters (such as number of fields, beam orientations and weightings, wedge angles and treatment energies) are then computed to determine the final plan. This is done manually. Due to time constraints, once an acceptable distribution has been obtained, this plan is used instead of having the luxury to search for the best plan for the particular case. In some instances, it is possible that a superior plan might exist to that actually used to treat the patient. The complexity of CFRT has posed new issues with more variables in treatment planning but offers opportunities for individual optimization. Determination of the treatment plan still relies heavily on past experience and patterns of practice within each centre. A variety of optimization algorithms have been proposed to aid the 3D planning process. For radiotherapy planning, the term optimization is usually applied to the method whereby the most appropriate treatment parameters have been chosen to achieve the desired or nominated 3D dose distribution. It is assumed that by obtaining the desired 3D dose distribution, this will reflect the optimal therapeutic ratio. There are too many treatment parameters in CFRT optimization for the forward planning or ‘trial and error’ method to be effective. To solve the difficulties associated with forward planning, the concept of ‘inverse’ planning was developed.15 Inverse planning is the method whereby the desired dose prescription is specified in 3D and an optimization algorithm is constructed to evaluate and determine the
best treatment parameters that will achieve the predefined goal. This type of optimization is undertaken by a computer that is able to perform hundreds or thousands of iterations of potential plans in much less time than a human planner will take to perform a few plan iterations. There are many mathematical methods proposed to satisfy dose constraints determined by the user in order to provide a solution to the optimization criteria. Detailed descriptions of these various methods are outside the scope of this chapter and general information can be found in this reference.16 This inverse planning methodology is also provides the basis for IMRT planning. There has been considerable work in the development of optimization algorithms for CFRT. Algorithms for optimization of beam weights17–19 and wedge angles20,21 for prostate CFRT have been evaluated and reported to produce improved planning outcomes compared with manual or forward planning methods. Algorithms to optimize beam orientations have also been investigated for the brain as the full 3D treatment space can be considered for non-coplanar beam orientations.21–25 These algorithms have been reported to produce a clinically acceptable solution – as good as an experienced planner would have devised but in a fraction of the time.25* This time saving has implications for busy radiotherapy centres and would be a useful option in difficult planning cases. Another alternative is to consider using planning class solutions designed for known clinical scenarios. There have been many class solutions published, with some derived from extensive institute experience whilst others have been evaluated using optimization algorithms. One example is for central sellar and suprasellar tumours of the brain. This class solution of four non-coplanar maximally spaced CFRT fields was found to produce a good clinically acceptable result compared with more than four fields.26* Class solutions are also available for prostate CFRT. Prostate CFRT plans vary substantially between centres and countries. Three or four coplanar fields with varying beam orientations are commonly used in Europe, whilst 4–6 fields are more common in the USA. Due to this large planning variation, investigators have examined two different PTVs (prostate gland alone and prostate with seminal vesicles) at two different doses (64 Gy and 74 Gy) for a class solution plan that provides equivalent target coverage and best sparing of the rectum, the dose-limiting organ in prostate CFRT.27–30* Series of different beam orientations for 3-, 4-, and 6-field prostate CFRT plans were systematically assessed (Fig. 51.4), and it was found that either 3-field (Fig. 51.4A & B) or 4-field (Fig. 51.4C & D) class solution plans provided acceptable results with good rectal sparing, and both were better than the 6-field class solution plan (Fig. 51.4E & F).31* The 3-field class solution plan had the added advantage that it could be used either for prostate alone or prostate plus seminal vesicle volumes without further modification for different treatment volumes, and this arrangement was simpler than either the 4-field or 6-field class solution plans, thereby translating
1258 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
Prostate Only
Prostate Seminal Vesicles
3F (0,90,270)
3F (0,90,270)
A
B
4F(35,90,270,325) C
4F(20,90,270,340) D
55
55
70
70
(a)
6F(65/lat/30)
6F(50/lat/25) E
F 50 25
50
65 25
30
65 30
Figure 51.4 Class solution plans for conformal coplanar prostate radiotherapy for two different treatment volumes: prostate only and prostate plus seminal vesicles. The class solution 3-field plans for both prostate only and prostate plus seminal vesicles are the same and shown by A and B, respectively. The class solution 4-field plans for both prostate only and prostate plus seminal vesicles are shown by C and D, respectively. The class solution 6-field plans for both prostate only and prostate plus seminal vesicles are shown by E and F, respectively. Note that the anterior beam orientations for the 4-field and 6-field plans are different when considering the two different prostate volumes. The angles of the anterior fields to the mid-coronal plane are larger so as to avoid the femoral heads when the larger treatment volumes are being considered.
into easier treatment set-up, delivery and field verification. Non-coplanar prostate CFRT plans can produce further rectal sparing compared with coplanar plans, but this requires optimization by beam orientation algorithms that can be computationally heavy.32 Although class solutions are designed to produce the best outcome for a group of patients, it can also be a valuable starting point where only minor alterations may be needed to individualize the plan appropriately to the clinical case. This has the advantage of saving the planner considerable time, an important consideration in any busy radiotherapy department. VALUE OF PLANNING EVALUATION TOOLS
Evaluation is an essential component of the planning process. The use of 3D data permits the assessment of dose distributions to be visualized in several ways. Dose distributions may be reviewed interactively in 2D and superimposed over CT images in different planes, or as 3D volumetric
(b)
(c)
Figure 51.5 A conformal radiotherapy (CFRT) plan for a post-operative bed recurrence in renal cancer. In this case, the dose distribution from a 3-field plan is shown in three different central planes: trans-axial (a), sagittal (b), coronal (c), with a dose–volume histogram.
or colour wash overlaid on the PTV or OAR (Fig. 51.5). The PTV dose distribution is assessed in terms of the conformity and homogeneity of its coverage, the presence of hot/cold dose spots and sparing of OAR. Dose statistics will provide mean, minimum and maximum assessments for each structure in the treatment region (Fig. 51.6). A histogram plot of dose/volume of tissue irradiated known as dose–volume histogram (DVH) can be created (Fig. 51.7). This DVH plot provides a simple estimation of volume coverage by dose but it does not provide spatial localization of the isodoses
Conformal radiotherapy 1259
Figure 51.6 A table of dose statistics arising from the treatment plan shown in Figure 51.5.
Figure 51.7 A dose volume histogram (DVH) is shown summarizing the coverage of the planning target volume and extent of irradiation received by the surrounding organs at risk (OAR). This DVH is derived from the treatment plan and table of dose statistics shown in Figures 51.5 and 51.6, respectively.
specifically for low or high dose spots. Therefore dose statistics, 3D isodose distributions and DVHs must be used together for complete plan evaluation. The weakest aspect of plan evaluation is that current methods rely on simple dose distribution as a surrogate of clinical outcomes. Much work is ongoing to develop reliable measures for using radiobiological data or quantifiable clinical outcomes to assess treatment plans. The use of models for TCP and NTCP could be useful in evaluation and in guiding optimization algorithms. The use of TCP and NTCP models for planning and evaluation remains controversial. It must be emphasized that current TCP and NTCP models are still limited by the lack of reliable clinical data that guide their development. However, the use of 3D dosimetry data over the next decade with thorough and careful collection of clinical outcomes can be used to
develop improved models and better biological planning algorithms. QUALITY ASSURANCE
All steps in the treatment planning chain of events are subject to inaccuracies and errors. Appropriate QA is needed to ensure that treatment outcomes for patients are not compromized, and it must cover every component in the planning process from inputting and transfer of clinical data to dosimetry, and from software to hardware testing of radiotherapy equipment. The QA programme must provide for clear standardization, documentation and archiving for each undertaken step in the process.33 These principles are reviewed in Chapter 50 (planning techniques). For CFRT, the general QA procedures are summarized in Table 51.1.
1260 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
Table 51.1 Quality assurance (QA) in the treatment planning process Process
QA procedures
Positioning/immobilization Simulation Clinical data acquisition Data transfer Definition of planning target volumes Aperture design Prescription Plan computation Plan evaluation Computation of monitor units Production of blocks, multi-leaf collimators (MLCs) Plan implementation Machine QA Patient QA
Port films, laser alignment, immobilization devices Simulator QA to include image and mechanical integrity Computed tomography (CT) and magnetic resonance imaging (MRI) QA to include image and mechanical integrity QA of the entire transfer process Peer review process by chart review with appropriate personnel and/or weekly clinical audit meetings Independent check of delivery and peer review Written, signed and dated. Peer review process by chart review with appropriate personnel and/or weekly clinical audit meetings Peer review process by chart review with appropriate personnel or weekly clinical audit meetings. QA of treatment planning systems to include dose calculation algorithms and production of plans. Peer review process by chart review with appropriate personnel and/or weekly clinical audit meetings Treatment planning system QA and independent checks QA for block cutting and compensator systems, MLCs, port film reviews. Peer review process by chart review with appropriate personnel and/or weekly clinical audit meetings QA of treatment machine and machine data from commissioning Peer review process by patient and chart review with appropriate personnel and weekly clinical audit meetings to include dose to dose-limiting or critical organs at risk (OAR), review for any plan modifications (i.e., dose prescription, new or modified fields, etc.), electronic portal imaging, patient status, and follow-up procedures.
Modified from Kutcher et al. American Association of Physics in Medicine (AAPM) Comprehensive QA for radiation oncology: report of AAPM Radiation Therapy Committee Task Group 40. Med Phys 1994; 21: 581–618.33
Radiation
Source
Target
Figure 51.8 A schematic representation of a multi-leaf collimator (MLC) device showing that the leaves can be individually altered and positioned to conform to the shape of the tumour or the target.
INTENSITY-MODULATED RADIOTHERAPY Concept and rationale The incorporation of inverse planning techniques with CFRT and routine use of multi-leaf collimators (MLCs) for automated field shaping (Fig. 51.8) have led to a new treatment paradigm termed IMRT.34,35 The aim of IMRT is to
further improve the physical basis of radiation delivery by allowing the intensity profile of the delivered radiation beam that is usually uniform to be variably altered or modulated across any portion of its treatment field (Fig. 51.9). Modulation of the beam’s intensity is not new, as this has been achieved in the past through simple methods such as beam wedges where modulation of the beam’s intensity is constant and gradual. However IMRT methodology provides substantially more variability of the beam’s intensity over many different portions or segments of the treatment field. This permits increased flexibility in distributing the beam’s intensity to fit concave or irregularly shaped targets. This flexibility in dose delivery can also be manipulated to deliver different dose per fractions to different target regions simultaneously with the effect of delivering a higher biological dose as a simultaneous integrated boost.
Basic principles of IMRT The IMRT rationale can be explained geometrically. The overlap of the prescription/high-dose contour from CFRT beams will produce a dose distribution that is essentially convex in shape (Fig. 51.10). If the target volume is concave in shape, or if there is an irregular 3D shape with several concavities within it, then all tissue/OAR nestled within these concavities will be enclosed by this convex dose envelope and be irradiated to the prescribed dose (Fig. 51.11a).
Intensity-modulated radiotherapy 1261
100
0 (b)
(a)
Figure 51.9 The beam intensities from (a) a conformal radiotherapy (CFRT) and (b) an image-modulated radiotherapy (IMRT) field are compared in a 3D column graph, where the height of the columns represent the relative beam intensities. The CFRT field has a constant beam intensity throughout its shaped field compared with the IMRT field in which the beam intensity is variable across its field. Each different column height and different shading represents a different dose intensity.
Conformal beam
OARs
OARs
Conformal beam
T OARs
Dose
(a)
Depth
(b)
Figure 51.10 (a) A schematic illustration for the basic principle of multiple fields and conformal radiotherapy. The shaded area (T) represents the tumour or target volume and the shaded circular areas represent different organs at risk (OAR) located within the head, such as a pair of eyes anteriorly and spinal cord posteriorly. Each treatment field is individually and conformally shaped to the profile of the target within the beam’s-eye view of the treatment portal. In 3D, the resultant high-dose coverage resulting from the overlap of the many conformal fields is a convex dose cloud envelope, which is located in the centre of Figure 51.10b.
This is an important concept. The use of IMRT avoids this conundrum by reducing the dose within this concavity through intensity modulation. The effect of IMRT is to produce a series of many smaller beamlets for each treatment beam or multiple segments for each treatment field. Each of these beamlets or segments can now be individually controlled in terms of its dose intensity. Eventually, these smaller beamlets can then be combined to create a single treatment beam that will deliver the desired dose distribution for that
field. Several of these intensity modulated fields can be combined. In this manner, the resultant high-dose distribution can be modulated to give better cover for concave or irregularly shaped 3D regions compared with CFRT. Much steeper high- to low-dose gradients at the junction between the target and adjacent OAR can be produced by IMRT to further reduce radiation to OAR. More importantly, this methodology can also spare the concavity regions that house critical OAR from unnecessary irradiation and promote the concept
1262 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
Radiation
Using constant beam intensity, the dose distribution will result in the OAR being within the high dose region.
Target Conformal shaped blocks
OAR
(a) Radiation
Using intensity-modulated, the resultant dose distribution can permit relative dose sparing of the OAR.
Target Intensity modulation
OAR
(b)
Figure 51.11 Schematic illustration for the principle of (a) conformal radiotherapy (CFRT) compared with (b) intensity-modulated radiotherapy (IMRT). For CFRT, the beam intensities are constant and field shaping is undertaken in the beams-eye view (BEV). For IMRT, both the field and beam intensities can be altered or shaped to ‘sculpt’ the dose distribution better to the 3D shape of the target.
of ‘OAR avoidance’ (Fig. 51.11b). Consequently, IMRT adds a new degree of freedom to treatment planning for safer dose escalation and reduction in radiation complications that can further improve the therapeutic ratio in radiotherapy.
Treatment planning for IMRT The steps in the planning process in IMRT will be similar to CFRT as IMRT is simply a refinement of the CFRT planning process. Steps common to CFRT planning become even more crucial in IMRT, particularly for patient set-up, immobilization and definition of target volumes and OAR, as IMRT produces substantially steeper dose gradients such that a geographical miss will invalidate the potential benefits of IMRT. Issues of target motion are essential considerations for any IMRT program. There are several differences in the IMRT planning process. These aspects include pre-planning considerations, inverse planning methodology, optimization, evaluation of IMRT plans and treatment delivery. PRE-PLANNING CONSIDERATIONS
Considerations such as patient set-up, immobilization devices and definition of planning volumes are extensively covered in Chapter 50 (Planning techniques). In IMRT, there may be a need to define more volumes-of-interest that would not normally be considered in CFRT. This is to
‘inform’ or guide the optimization algorithm as to where it is allowed or not allowed to deposit dose during its optimization. Any tissue region that is not qualified will be considered by the computer to be ‘fair game’ in its attempt to achieve the prescription dose aims and constraints. In this scenario, the algorithm may unwittingly deposit more dose in unspecified tissue regions than is clinically appropriate or desired by the clinician in order to achieve its aims. For example, in brain or head/neck IMRT, it may be appropriate to identify not only the eyes, optic chiasma and spinal cord but also hearing structures, all salivary glands, bones or segments of bones and portions of brain tissue. It may also be necessary to also create sub-structure volumes that will be used to drive the optimization algorithm to its goal or to create steeper dose gradients for conformal avoidance of dose-limiting OAR adjacent to the PTV. One example is if a dose-limiting OAR lies partially within the PTV. In this situation, there is both a need to treat the PTV as well as to reduce as much dose to the OAR within the PTV as possible. This overlap region can be defined separately to limit dose to this region and to increase the dose gradients between these regions.
PLANNING AND INVERSE ALGORITHMS FOR IMRT
Treatment planning follows the methodology described for CFRT. The use of algorithms to devise the entire plan
Intensity-modulated radiotherapy 1263
Table 51.2 Normal tissue/organ volumes and tolerances to therapeutic irradiation
Organ Kidney Brain Brain stem Spinal cord Lung Heart Oesophagus Stomach Small intestine Large intestine Rectum Liver
TD5/5 (Gy) for organ volume
TD50/5 (Gy) for organ volume
1/3
2/3
3/3
1/3
2/3
3/3
Clinical complication
50 60 60 50 45 60 60 60 50 55 50
30 50 53 50 30 45 58 55 35
23 45 50 47 17.5 40 55 50 40 45 60 30
75 70 65 70 72 70 60 65 55
40 65 70 40 55 70 67 45
28 60 65 24.5 50 68 65 55 55 80 40
Clinical nephritis Necrosis/infarction Necrosis/infarction Myelitis/necrosis Pneumonitis Pericarditis Stricture/ulceration/perforation Ulceration/perforation Obstruction/ulceration/perforation/fistula Obstruction/ulceration/perforation/fistula Severe proctitis/necrosis/fistula Liver failure
TD5/5 is the tolerance dose for the probability of a 5 per cent complication rate within 5 years; TD50/5 is the tolerance dose for the probability of a 50 per cent complication rate within 5 years. These TD rates are rough estimations and relate to the volume of organ treated. Note that the tolerance ranges are described for a dose-fractionation scheme using daily fraction sizes of 2 Gy given 5 times a week. It also assumes uniform irradiation of all or part of an organ, normal baseline organ function, no influence from chemotherapy or surgery and has been derived from data related to adults. Note also that both TD5/5 and TD50/5 are defined for severe end-points or those complications that are classified as RTOG Grade 3. Modified from Emami et al. Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 1991 21:109–122 (Ref. 36).
including optimization for number of fields, beam orientations and the dose distribution is still under investigation. Currently most IMRT plans require some pre-selection to be made by the radiotherapy planner for the number of fields and beam orientations. These choices are usually selected from institute protocols based on experience or using previously defined class solutions as a starting point. This pre-selection is often less important for IMRT than for CFRT as the inherent flexibility in dose modulation can often compensate if a less than optimal choice was made by the planner. However, having an adequate number of fields and appropriate beam orientations permits best functioning of the inverse optimization algorithms, particularly for difficult cases such as concave shapes or where there is minimal separation between critical OAR and the PTV. In IMRT, beam intensities for each field in their beam orientations will be calculated by the inverse optimization algorithm. The planner will specify the desired dose goals and range for the target volume(s) and different volumes-of-interest. The optimization algorithm divides each beam into many smaller beamlets and iteratively manipulates the intensity of each beamlet from all fields until the resultant 3D dose distribution best resembles the desired prescription or matches the specified dose objectives. The inverse planning concept (see Use of optimization algorithms and class solutions) is crucial for IMRT. In inverse planning, cost functions and importance factors are usually employed to drive the optimization algorithm to its desired solution. Cost function is a mathematical definition of the ‘goodness’ of the optimized plan as it progresses through many iterations to the desired solution. Cost functions may be based on objective dose criteria or biological criteria. Objective dose-based criteria usually take into
account parameters such as PTV coverage and OAR tolerances. For many tissues, a single dose value may not accurately describe its tolerance, and a range of values may be needed. Often a profile of dose–volume constraints may be needed as described by a series of dose-point thresholds or DVH. Some dose–volume tolerances for normal tissues have been published36 and act as guidelines (Table 51.2). It is important to realise that many of these stated values are conservative estimations based on limited data and extrapolation from clinical experience. These thresholds are still being defined and refined today. Biologically based cost functions use models of calculated radiobiological response rates as a measure of the goodness of the plan but this type of modelling is still being investigated for its full validity in accurately predicting radiation outcomes. Appropriate parameter specification is crucial to the success of optimization. Unrealistic specification of physical goals and constraints will cause the optimization algorithm to produce a markedly inferior plan as it will attempt to use extreme values for the beamlets or use excessive numbers of beamlets to meet a prescription that it is impossible to achieve. Experience of inverse planning methodologies specific to each planning system is needed to determine appropriate and realistic dose constraints that correlate with the eventual dose distributions. Templates of dose constraints for common tumour treatment sites can be developed and shared, but it is beneficial for all initial users of any optimization or inverse planning system to be guided through this learning process. This is useful because, although solutions to easy cases can be readily handled by these templates, there often is enough variation in difficult cases to limit these templates to being merely good starting points that will need further refinement.
1264 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
PLAN EVALUATION AND OPTIMIZATION FOR IMRT
The tools utilized in the evaluation of IMRT plans are similar to those for CFRT (see Use of optimization algorithms and class solutions). These tools can be used to determine if coverage of the target has been achieved whilst satisfying the dose constraints for the OAR. In addition, biological indices such as NTCP and TCP thresholds may also be used, taking into consideration the previously highlighted caveats. In IMRT plans, assessment of the intensity profiles is another useful method to determine the contribution of an individual beam to the dose distribution. This can be undertaken using an observer’s view for each field where the intensity profiles can be visualized in 3D (Fig. 51.12a) or a dose intensity colour/greyscale wash in 2D (Fig. 51.12b). Although the evaluation of IMRT plans is similar to CFRT, the method of improving IMRT plans may not be as straightforward for a number of reasons. It may not be obvious to the novice planner if a bad IMRT plan is the result of poorly constructed dose specifications and constraints or due to an inappropriate and unrealistic prescription that the inverse planning algorithm simply cannot achieve. Subsequent optimization for the IMRT plan will have to involve adjustment of the planning constraints or tissue penalties and may not be as intuitive as for forward planning situations. This will have to depend on knowledge of the methodology of the inverse planning algorithm in question. Occasionally, artificial volumes may have to be defined (see Pre-planning considerations) in order to drive the inverse planning algorithm appropriately. For example, if an undesirable hot spot cannot be reduced by increasing the penalty weighting for overdose of that region without adversely affecting the overall plan, then this region may be contoured specifically and assigned different dose constraints to allow the algorithm achieve this. Similarly, homogeneity for the PTV or inhomogeneity for the PTV may be created by sub-contouring sections of the PTV and prescribing different dose thresholds for the different regions in order to achieve the desired aims. This sub-volume concept can also be applied to OAR where there may be overlap of the OAR within the PTV and the inverse planning algorithm will have difficulty satisfying the need to deliver full dose to the PTV whilst maintaining the dose constraints for the OAR in question. Sub-volume planning method is based on the fact that DVH-based thresholds are not able to take account of the spatial distribution for cold and hot spots or regions. This sub-volume planning strategy will aid the inverse planning algorithm to achieve a better plan rather than searching for an impossible solution.
Methods of delivering intensity-modulated beams There are several methods whereby intensity-modulated beams can be delivered. One previous method was with a modulating LINAC known as a ‘tracking unit’, which controls the beam fluence directly by rotating its gantry
(a)
(b)
Figure 51.12 A single intensity-modulated radiotherapy (IMRT) field can be represented as a 3D dose-intensity map (a) or a 2D grey-scale wash map (b).
head over a treatment couch that had full range of movement.37 The dose delivered at any one point is proportional to the duration of time at that point and inversely proportional to the tracking speed of that beam. Therefore, the intensity delivered could be modified or varied according to the dwell time of the beam, the rotational speed of the gantry and the position of the patient on the couch, but it was very time consuming to deliver a treatment. This machine, which pre-dates the application of MLCs, has now been superseded by more sophisticated methodology. Another method is using collimators to modulate the beam as the beam is delivered by the treatment machine.38 One traditional method is the partial transmission block or physical compensators. The block thickness is custommade and shaped in 3D to produce variable transmission and thus variable beam intensity within the treatment field. However, the creation of these partial transmission blocks is labour intensive and requires sophisticated
Intensity-modulated radiotherapy 1265
milling machines. Although the thickness of these modulators can be varied continuously to create very high resolution, in practice, this resolution is comparable with the use of smaller size MLCs. These partial transmission blocks or 3D compensators need to be physically inserted for each field, can also harden the beam and cause radiation scatter for increased skin doses. They are now largely superseded by MLCs. It is important to understand that, in the delivery of IMRT treatments using MLCs, only a proportion of the treatment field is exposed at any given time. Once the dose prescription goal has been achieved, a ‘leaf sequencer’ algorithm within the IMRT planning system calculates the delivery of the beam intensity patterns depending on the method of beam delivery. This algorithm generates a set of instructions that specify the order and sequence of MLC leaf motion during beam delivery for the intensity patterns in question. This process takes into consideration other physical parameters such as the number of segments, the number of monitor units, restrictions on the MLCs being used (i.e., leaf motion, overlap of leaves and leaf speed) and the time for delivery.39–41
mu
11 10
11 10
9
9
8 6
4
4
8
7
6
7
5
5
3
3
2
2
1
1
(a)
position 11 10
mu
11 10
9
9
8
8
7 6
7
5
6
5
4
4 3
3
2
2
1
1 position
(b)
MULTI-LEAF COLLIMATOR METHODS
There are several MLC techniques used to create intensitymodulated beams and these can be grouped as: 1. the static ‘step and shoot’ method; 2. dynamic collimation – leaf sweeping with variable leaf gaps; and 3. the arcing or tomotherapy method with dynamic collimation. In the ‘step and shoot’ method,42 each treated field is subdivided into a set of smaller ‘static’ fields, each with uniform beam intensity. Each uniform subfield is sequentially delivered in a ‘step and shoot’ sequence with the LINAC turned off in-between each subfield delivery. The dose from each subfield is summed to provide the final modulated beam profile. The concept of the ‘step and shoot’ methodology is illustrated in Figure 51.13. The planner selects the number of intensity steps and hence the number of subfields or segments. Increasing the number of intensity steps will increase the number of segments and the delivered dose distribution will better approximate the intended dose computed. However, segments used have to be pragmatically determined such that dose delivery can be achieved within a sensible time duration. It is obvious that separating out the dose intensities using a finite number of subfields only approximates the continuous 2D intensity distribution calculated by the IMRT planning system such that there will be a difference between the calculated and delivered doses. Some IMRT planning systems have overcome this known difference by introducing a discretized optimum intensity calculator that inputs the user-selected number of intensity levels needed before optimization to reduce this dose difference. The dynamic MLC method for fixed fields can be achieved in two ways. The dynamic ‘step and shoot’ method43 is similar to the process described above except
11 mu
1 2
10 9
3
8
5 7
7 5
8 9
6
10
4 3 2 1 (c)
11 4 6 position
Figure 51.13 A schematic illustration for the principle of the step and shoot intensity-modulated radiotherapy (IMRT) method. The desired intensity profile is shown in the background as the shaded region (a). This intensity profile can then split into a series of static fields with each field delivering a specified dose increment or monitor units (mu). For field segment, the multi-leaf collimator (MLC) leaf positions are labelled for their left and right positions respectively. Each of these field segments is delivered in turn. When all the field segments have been delivered, this should amount to the original intensity profile. Another method of delivering step and shoot is called ‘close-in’ (b). Here, each peak is created and treated in turn to amount to the original intensity profile. A more efficient method of delivery is to have the MLC leaves move in one direction (c). Here each leaf starts from the left-hand position and is delivered in order of position in a process called ‘leaf sweep’. For all of these techniques, the radiation beam is suspended between each field segment whilst the leaves are set to their next position and ready for delivery of the next field segment, hence the description ‘step and shoot’.
1266 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
that the LINAC is kept running throughout the procedure and the movement along the different sequential static field positions is kept constant. In the dynamic ‘leaf sweeping’ or ‘sliding window’ method,39,44 a pair of MLC leaves are swept unidirectionally across each treatment field at different leaf speeds creating a leading and trailing leaf. The sliding window methodology is illustrated in Figure 51.14. The size of the gap or aperture created by these two leaves together with the dwell time of the aperture allows a variable intensity to be created for each point of the treatment field. The computer will deliver the calculated intensity pattern by controlling the position and speed of the MLC leaves as a function of the monitor units needed to be delivered during the beam-on time for every point in the beam. The constraints on this method relate to the mechanical limitations of the MLC device being used due to MLC leaf design, maximum leaf speed and clinical dose rates for the LINAC. Table 51.3 compares and contrast some of the differences between these two methods. There are two main approaches for dynamic arcing techniques using the tomotherapy method or intensitymodulated arc therapy approach. In the tomotherapy methodology (‘tomos’ meaning slice in Greek), treatment intensity is modulated by a rotating fan or slit beam delivered in a slice-by-slice method. This concept was developed initially by the NOMOS Corporation45 using a purpose-built MLC device called the MIMiC (Multi-leaf Intensity-Modulating Collimator). The MIMiC device consisted of two banks of 20 MLC leaves with each leaf being 1–2 cm in length and 1 cm wide. The MIMiC uses an electro-pneumatic system to open or close the individual leaves whilst the gantry is rotating around the patient. This allows for continuous delivery of fixed intensity fields for every 5–10 degrees of the gantry arc. The therapy bed with the immobilized patient is then indexed along the z-axis after each slit beam is delivered to cover the whole length of the treatment field. This system has the advantage that the MIMiC device can be easily mounted to a conventional LINAC, avoiding the expense of a dedicated tomotherapy machine (Fig. 51.15). The other method for tomotherapy is the Tomotherapy LINAC developed by the University of Wisconsin group.46 This machine has a temporally modulated MLC device mounted within the gantry of a spiral CT machine and incorporates a modified 6-MV LINAC in place of the conventional kilovoltage (kV) X-ray tube (Fig. 51.16). This tomotherapy machine will permit a 6-MV photon fan beam to rotate through a full circle with the patient being treated by a spiral technique whereby the treatment couch is moved continuously though the aperture of the machine during gantry rotation for 3D intensity-modulation treatments, radiotherapy planning, simulation and treatment verification. A cross between the methods described above is the ‘intensity-modulated arc therapy’ or IMAT. The IMAT method combines features of the rotating gantry principle with the ‘step and shoot’ and ‘sliding window’ methods.43
mu
position
(a)
mu Shielded by left leaf
Shielded by right leaf (b)
position
dark mu
light
(c)
position
Figure 51.14 A schematic illustration for the principle of ‘sliding windows’ or dynamic multi-leaf collimator (MLC). In this method, the beam intensity is modulated by moving the MLC leaves continuously during radiation delivery. The intensity profile is again split into different sections as seen in (a) with a positive gradient (dark grey) and negative gradient (light grey). These segments are rejoined in (b) to enable a continuous gradient of leaf movement with the left leaf moving first and followed at the appropriate time interval and space by the right leaf. This is similar in concept to the step and shoot leaf sweep shown in Figure 51.13 but, in this method, the radiation beam is not interrupted. The leaf pathways are seen in (c) and the eventual result is a dose-intensity distribution that resembles the intensity profile in (a).
In IMAT, modulation of the beam is created by superimposing different field shapes instead of using leaf speed, so it is similar in concept to the ‘step and shoot’ technique. However, with the gantry rotating through its arc, the
Intensity-modulated radiotherapy 1267
Table 51.3 Comparison of intensity-modulated radiotherapy methods Method
Advantages
Disadvantages
Step and shoot
Simpler concept Simpler multi-leaf collimator (MLC) control system as there is no need to control individual leaf speeds. Simple delivery of subfields Easier to resume if there is any interruption in treatment delivery Simpler to verify the intensity pattern for each subfield Fewer monitor units compared with dynamic methods
May require longer treatment times for complex intensity patterns Complex intensity patterns can only be approximated due to inherent method of step and shoot (see text) Longer verification time if many segments are used
Sliding window
The intensity patterns including complex profiles can be smoothly delivered Only one fields is required per direction
The delivery is more complex and requires precise control of the individual leaf speeds and dose-rate modulation Small errors of leaf calibration may result in substantial errors in delivered dose More difficult to cope with intensity patterns that have regions of no dose intensity as it is difficult to close leaves in dynamic mode More extensive quality assurance program is needed
Figure 51.15 The MIMiC device. (Courtesy of NOMOS Radiation Oncology Division of North American Scientific)
MLC leaves are in constant motion while the beam is on so it is similar to the ‘sliding window’ technique. This method delivers all subfields of equal intensity sequentially with the subfield shape being changed continuously for each point along the arc. Once the specified intensity level has been completed, a new arc is begun for the next intensity level and so on. The arcs are approximated as multiple shaped fields spaced every 5–10 degrees around the patient and may be superimposed coplanar or non-coplanar. There is no indexing of the therapy couch. Some limitations with this method include the lack of robust automated planning tools that can produce efficient IMAT treatment plans, but algorithms have been developed to optimize leaf positions and aperture weighting in IMAT planning.47 Use of these algorithms to permit rotational IMRT can be achieved using a conventional linear accelerator and a conventional MLC, thus avoiding the costs of a tomotherapy machine.
Figure 51.16 The TomoTherapy machine. (Courtesy of TomoTherapy Inc, Madison, WI, USA)
Quality assurance for IMRT The complexity of IMRT beam delivery entails a rigorous QA program that includes planning checks, dosimetric assessments, mechanical tests, and appropriate record-verification systems. All intensity-modulated fields must undergo periodic dosimetric and geometrical verification. Assessments of the
1268 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
intensity fields must form part of the routine physics QA program. Intensity patterns can provide a quick assessment of that the MLC settings are correct. Dose checks via ion chamber, diodes, thermo-luminescent dosimeters (TLD) or film measurements are needed to assess the dose output, the produced IMRT plan and its stability for different locations in the process. Phantom measurements are useful to check that each treatment beam of the plan is correct. In vivo measurements on the patients are another valuable tool to confirm the reliability of the delivery process. All current dosimetry verification methods are labour intensive and have limitations. For systems that utilize MLC for beam modulation, mechanical tests for geometric accuracy of the MLC leaves are essential. Calibration and verification is needed not only before treatment but also needs to be monitored during delivery itself. Although this applies to both static and dynamic methods of intensity modulation, it is more important in the sliding window technique. Software systems are available to continuously monitor the individual leaf positions during delivery, and deviations beyond the nominated threshold will invoke a beam-off command by the computer. All sections of the planning process to production of the MLC leaf segment or motion files and their data transference to the LINAC needs to be recorded and verified. This must include checks on patient identification and appropriate labelling of all related data files.
Clinical experience with IMRT Intensity-modulated radiotherapy is a time consuming and expensive process, and it is important to focus efforts on tumour sites where the greatest potential benefit may be derived. In general, cases requiring coverage of irregular and concave-shaped 3D targets, dose escalation where nearby OAR limit dose and cases to reduce relevant OAR toxicity will benefit most from IMRT. The clinical data arising from applications of IMRT are too extensive to review in this chapter and is addressed in the respective subsite cancer chapters. Some examples for the clinical utility of IMRT will be outlined to illustrate its potential benefit. Initial applications of IMRT were utilized in central nervous system (CNS), prostate, head and neck, thyroid and breast cancers. Success in these subsites has led to the application of IMRT to many other subsites such as sarcomas, lung, gynaecological, abdominal and bladder cancers. In planning studies for complex shaped tumours, IMRT planning was able to increase target coverage by up to 36 per cent with conformality improved by 10 per cent compared with CFRT.48 These investigators also noted that IMRT can increase dose escalation by a mean dose of 4–6 Gy and target coverage by 19 per cent for the same degree of conformality. Investigators have noted that IMRT offered the opportunity to adequately treat complex base of skull meningiomas safely whilst better sparing nearby OAR, compared with CFRT techniques.49*
In prostate cancer, the dose has been safely escalated to 81 Gy with clinically acceptable rates of rectal toxicity.50*,51 These investigators have reported that IMRT planning was able to improve dose coverage of the target compared with CFRT, with reduced irradiated volumes of the rectum and bladder in the high-dose region (p 0.01). It was also reported that the 2-year actuarial risk of grade 2 bleeding was 2 per cent using IMRT, compared with 10 per cent for CFRT techniques (p 0.001). In addition, strategies have been developed to boost multiple intra-dominant prostatic lesions to 90 Gy without exceeding normal tissue tolerances.52,53 In head and neck cancer, IMRT has been employed to reduce the development of xerostomia. Investigators from several centres have reported good and acceptable recovery of xerostomia following treatment using parotid sparing IMRT techniques.54–57* These studies have demonstrated that the partial dose sparing of the salivary glands is able to improve treatment related xerostomia to mild or zero levels by 2 years post irradiation.55*,56 This reduction in rates of xerostomia can improve other head and neck functions such as swallowing and speech and lead to improvements in broad aspects of quality of life for patients.57* This is particularly important when chemoradiotherapy regimens are used in head and neck cancers. Although IMRT has been intensively investigated in several large treatment centres, it is still a relatively new treatment method for many centres throughout the world. Despite the growing expertise in IMRT, there will still be a steep learning curve for all members of the radiotherapy department and attention needs to be paid to the issues highlighted in the subsections above. There are many considerations to be taken into account when initiating an IMRT program. These considerations should include a decision on the initial clinical site to commence IMRT, methodology of inverse planning, method of IMRT delivery, and commissioning of the IMRT system. Once IMRT expertise has been gained in one site, attention can be expanded to other clinical sites. Each clinical site will have its own peculiar issues and challenges in IMRT implementations, from patient immobilization to treatment planning, delivery and verification.
IMAGE-GUIDED RADIOTHERAPY Concept and rationale The central aim for radiotherapy is to be able to deliver its prescribed dose to its nominated target accurately and reliably at every fraction on a day-to-day basis. It is clear that temporal variations for spatial localization and geometry exist within the human body during radiotherapy, and these issues represent the main cause of treatment uncertainties for radiotherapy particularly IMRT. Strategies to overcome or minimize these patient and organ temporal spatial variations have been collectively termed image-guided radiotherapy
Image-guided radiotherapy 1269
(IGRT). A variety of terminologies have also been applied to this concept, including 4D radiotherapy (where the fourth dimension being evaluated is time), adaptive radiotherapy or predictive radiotherapy. The essence of all these different terms is to account for the temporal–spatial treatment variations that exist in radiotherapy and to develop suitable strategies to overcome them. Conceptually, the term IGRT can have a broader meaning and may relate not only to strategies of reliable beam targeting but also to improved means of identifying the extent of the tumour and OAR. This section on IGRT will deal with the former issue. Developments in both offline and online imaging to improve each facet within the radiotherapy process are encompassed by IGRT. Temporal– spatial uncertainties in localization of the target and OAR as well as in patient set-up can lead to geographical misses. The resultant outcome will be a decrease in local control rates, leading to higher treatment failures and increased treatment related complications from unnecessary and excessive irradiation of OAR. Most radical radiotherapy plans rely on a single imaging ‘snap-shot’ of the treatment region that has the target ‘frozen’ in its relationship to its surrounding anatomy. Imaging is usually performed using CT scanning taken 1–2 weeks before actual treatment delivery to organize planning. This single scan is unable to account for the temporal–spatial variations in the treatment region due to set-up uncertainties and, more importantly, the effects of internal physiological activity such as breathing motion and organ filling and emptying (e.g., bladder and rectum) that can displace the target from its initial imaged position in the treatment planning scan. In order to compensate for this uncertainty, treatment margins are used to ensure that the target is adequately irradiated. The size of this safety margin depends on the treatment region. Irradiation of a malignant lesion in the thorax may require a larger treatment margin (to avoid a geographical miss due to respiratory motion) than irradiation of a lesion in the brain where there is little internal motion. A treatment margin that is too large will result in excessive irradiation of adjacent normal organs and limit the curative dose that can be delivered. Excessive normal-tissue dose may also limit the use of combined modality radiation–chemotherapy regimens. The opportunity to correctly apportion treatment margins by reducing margins by a few to several millimetres can effectively reduce the unnecessary irradiated volume by as much as one third. Currently, the benefits of CFRT, especially sophisticated techniques such as stereotactic and IMRT, may be limited by our ability to reliably and accurately deliver these treatment plans on a daily basis, and IGRT strategies are needed to account for the temporal–spatial variations and treatment uncertainties in radiotherapy treatment delivery not only ‘between’ (inter-fraction) daily radiotherapy treatments but also ‘during’ (intra-fraction) radiotherapy delivery. The aim of IGRT is to minimize or reduce the extent of systematic errors and limit the impact of
random errors in treatment delivery. By reducing both these errors, the PTV may be significantly reduced. Estimations of the potential benefit for the individual patient may be calculated using NTCP or TCP. These can then be used to select or deselect the patient for more aggressive radiotherapy regimes, i.e., dose escalation.
Basic principles The temporal–spatial variations that occur during a course of radiotherapy can be divided into a systematic and a random component. The systematic component comprises the discrepancies between that defined on the pre-treatment scan and that during treatment. The random component comprises the variation and discrepancies between each treatment and will always be present in varying degrees. Evaluation of these two components can only be determined by repeat imaging during the course of radiotherapy and ideally whilst the patient is on the LINAC treatment bed during radiation delivery. The systematic component of error may be estimated from repeat imaging and this component can be substantially minimized or eliminated compared with the random component. The implications of these temporal–spatial variations are both site and patient specific. Knowledge of the magnitude, distribution, pattern of organ motion and physiological activity within each site and the impact on radiotherapy delivery is important. This is relevant not only for organs that fill or empty such as lungs, stomach, bladder and rectum but also organs that may be displaced or deformed by physiological activity, such as the liver and prostate. Imageguided strategies need to be site specific but certain general principles can be followed. First, the target, the radiotherapy technique and the site of irradiation need to be examined to estimate the potential magnitude of temporal–spatial variation for both systematic and random components of treatment uncertainty. This evaluation also needs to take into account the proximity of the critical OAR to the target and its radiation tolerance. Depending on the clinical scenario, this needs to be examined and measured for each individual patient. Assessment of variations can be undertaken in a number of ways using repeated cross-sectional imaging, continuous monitoring with surrogate markers, fluoroscopy or cine imaging. The target and/or the critical OAR or surrogates of target/OAR motion should be imaged. Second, image evaluations can be performed before radiotherapy commences or before each fraction, during and/or after each daily radiotherapy fraction. These assessments should be undertaken at the start of radiotherapy. These parameters of temporal–spatial variations can then be assessed in terms of patient set-up, plan design and treatment delivery. It is also possible to estimate the potential impact on TCP or NTCP calculations by comparing standard radiotherapy methods versus IGRT programs. This pattern(s) of variation and its components can be
1270 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
measured, and based upon these measurements an imageguided strategy can be designed to minimize or adjust/ adapt to the problem at hand.
Current imaging systems in radiotherapy Portal imaging is the current method of treatment verification in the majority of radiotherapy centres (see Chapter 50, Planning techniques). The development of electronic portal imaging devices (EPIDs) has greatly increased the efficiency and quality of this verification process by permitting off-line treatment set-up verifications and error corrections that can compensate for some of the known systematic and random treatment uncertainties. However, this method can only provide verification of the patient’s positioning based on anatomical landmarks, and it is inadequate if internal anatomical information is needed. Another major limitation of portal imaging is the low image contrast, which provides inferior quality compared with the simulator image taken using kV sources or high-resolution digital reconstructed radiographs. The need for better image quality and, more importantly, spatial verification of internal target positioning has spurred developments of 3D on-line imaging systems. Ideally, imaging should provide cross-sectional/3D information. There are several on-line strategies to address the issues of temporal spatial variation of the target in radiotherapy. Ultrasound-guided systems stereotactically linked to the LINAC may permit on-line daily localization for some tumour sites (e.g., prostate radiotherapy). One such system introduced in the late 1980s is the BAT (B-mode Acquisition and Targeting) device engineered by NOMOS Corporation (Fig. 51.17). In prostate radiotherapy, it has been reported that anterior–posterior displacements 5 mm from the original treatment position could be reduced by a factor of two using BAT, but there was significant interoperator variability in the contour alignment process.58 In a study of over 5300 BAT procedures in 147 consecutive patients receiving prostate IMRT, this process has been reported to add an extra 5 minutes to the allocated 20 minute schedule, with the quality of images being acceptable in 95 per cent of cases with only 3 per cent major alignment errors.59 It is clear that QA procedures are needed to ensure adequate operator technique and expertise to limit potential errors that may arise with this system. More recent hardware developments include kV conebeam imagers mounted on the LINAC gantry permitting kV radiography that can provide improved superior bony identification compared with MV portal imaging. These advances have been supported by developments in amorphous silicon (a-Si) flat panel detector technology for large area imaging. The availability of in-room kV sources has also provided the opportunity for on-line fluoroscopy for real-time assessment of target and organ motion. Importantly, kV cone-beam technology can also provide cross-sectional imaging for on-line 3D verification.
On-line strategies can be grouped into those that provide landmark identification, such as surrogates of breathing patterns; radio-opaque fiducial marker localization, such as with ‘stereoscopic’ kV fluoroscopy; and those that provide cross-sectional information, such as the combination of a CT scanner in a therapy room or integrated in the LINAC to provide MV-CT images or cone-beam CT on a LINAC. Several systems have been developed to enable the localization of internally implanted markers or fiducials using kV imaging systems that are housed on the gantry of the LINAC or within the therapy room. The quality of the images obtained is excellent and permits the identification of fiducials down to a size of 3 mm in length. Some of these systems are described below (see Tracking methods). If implanted radio-opaque markers are not used to provide surrogate information on target/internal organ motion, then cross-sectional information is needed. One system places a conventional CT scanner and the treatment couch on rails, with both devices in a known geometrical relationship to the LINAC’s isocentre for set-up reproducibility. Inter-fractional variation of the target can be determined prior to treatment and corrected on-line. One potential limitation is that movement of the patient and target motion may occur in the time between image acquisition and treatment delivery. Another development is the integration of a CT scanner within the LINAC produced by Tomotherapy Inc (Wisconsin, USA). This system avoids the issue above and provides MV CT imaging, permits on-line treatment verification as well as tomotherapy treatments. Single or dual kV sources mounted on the LINAC gantry with a digital a-Si flat panel can be rotated 360 degrees to acquire volumetric cross-sectional information. A volume
Figure 51.17 The BAT system for stereotactic ultrasound guided localisation. (Courtesy of NOMOS Radiation Oncology Division of North American Scientific)
Image-guided radiotherapy 1271
equivalent to the length of the panel field of view (20–25 cm) in use can be doubled by off-setting the panel in opposite directions for the first 180 degrees and subsequent 180 degrees to complete the 360 degrees acquisition. The resulting image quality using this method is lower as the scanned structures are only imaged once during the 360-degree rotation compared with the standard acquisition where the scanned structures are scanned twice. However, volumetric acquisition means that this data can be reconstructed in 3D and any plane to assess the target or the adjacent OAR within the irradiated region. Jaffray et al.60,61 reported the initial development of a single kV system with an a-Si EPID device sited perpendicular to the treatment head of the LINAC gantry with excellent image resolution and definition of soft tissue structures. These systems are now commercially available from a variety of manufacturers such as the Synergy (Elekta Oncology Systems Crawley, UK) and On-board Imager (Varian Medical Systems Inc, Palo Alto, California). Examples of these single kV systems are shown in Figures 51.18 and 51.19. Another variation in the placement of the kV system is being investigated by Siemens Medical Solutions (Erlangen, Germany). The Siemens Artiste system incorporates its kV imager along the same axis as the gantry head instead of perpendicular to the irradiation axis (Figure 51.20). This is a very interesting concept currently under development and provides many other radiotherapy opportunities including entry and exit dosimetry. The advantage of on-line cone-beam kV systems is the multi-functionality that they can provide. First, kV radiographs can provide improved portal imaging quality, and this may be particularly useful for small treatment or boost fields. Second, on-line fluoroscopy permits the opportunity to track and gate treatments to implanted
radio-opaque markers. This may allow tighter treatment margins for reduced dose to OAR and permit safer dose escalation. Third, cross-sectional volumetric imaging can define the spatial positioning not only of the target but also any OAR to allow positional variation to be corrected online and allow adaptive or predictive IGRT strategies.
Figure 51.18 A kilovoltage X-ray imager mounted perpendicular to the gantry head of a medical linear accelerator. Example shown is of the Elekta Synergy XVI system (Elekta Oncology Systems, Crawley, UK) installed in the Christie Hospital.
Figure 51.20 A concept illustration of the Siemens Artiste system with kilovoltage and megavoltage cone-beam imaging with the opportunity for transit dosimetry (Courtesy of Siemens Oncology Heidelberg, Germany).
Image-guided radiotherapy strategies Strategies for image guidance in radiotherapy is a relatively new and very active field of research and development that is continuously being investigated, challenged and modified
Figure 51.19 The On-board Imager, a cone beam system mounted on a linear accelerator (LINAC) (Courtesy of Varian Medical Systems Inc, Palo Alto, California).
1272 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
at the current time. The IGRT strategies include adaptive, predictive, gated and tracking methods. ADAPTIVE METHODS
The adaptive IGRT strategy relies on repeat imaging to determine the magnitude of the systematic components responsible for temporal–spatial treatment uncertainties for the individual patient. This process generally involves several feedback loops (that may be off-line or on-line) to reduce the treatment uncertainty inherent in the initial treatment plan by ‘adapting’ the treatment plan to the temporal–spatial variations for the individual patient. Multiple repeat cross-sectional imaging can estimate the mean or systematic component and standard deviation or random component of treatment uncertainties. Once sampling of the anticipated temporal–spatial discrepancies is quantified, off-line modifications to the treatment plan can be made and implemented. This has been examined for a number of sites such as prostate and bladder radiotherapy. The adaptive radiotherapy (ART) strategy was initially developed at William Beaumont Hospital by Yan et al.62,63 In prostate ART, 4–5 repeat imaging scans performed early in the course of radiotherapy, usually within the first 1–2 weeks, has been reported to provide the optimal sampling schedule.63 Using ART, the mean systematic error was reduced from 4 mm (range 2–7 mm) to 0.5 mm (range 0.2–1.4 mm). The investigators suggested that ART permitted appropriate treatment plan re-optimization for prostate radiotherapy, with 95 per cent of patient-specific set-up margins being predictable to within 1 mm. A subsequent study confirmed the suitability of ART for conventional 4-field prostate radiotherapy but suggested that IMRT treatments required 2 weeks of daily imaging to provide the same level of optimization.64 Whilst large systematic errors may be reduced by the initial adaptive modification, it is still important to continue measurements throughout the treatment following modification so as to provide QA for the adaptive strategy. Bladder radiotherapy is another site for adaptive strategy (Fig. 51.21). Using multiple daily CT scans, investigators at the NKI have developed a model to describe the 3D RTP
Day 1
Day 2
Day 3
Average
Figure 51.21 An schematic 2D example of the adaptive radiotherapy strategy for bladder radiotherapy showing the bladder sizes taken from the radiotherapy treatment planning scan (RTP) and consecutive days at the start of radiotherapy (days 1–3) with the calculated average bladder volume produced from the repeated scans. The average bladder volume will then be used for the rest of the radiotherapy treatment.
change in bladder shape.65 This model was tested with better approximation for the inferior portions of the bladder than the cranial and posterior bladder portions, which were more affected by bowel activity.66 In a feasibility study of partial bladder irradiation, the adaptive process was reported to able to reduce significantly treatment volumes by 40 per cent compared with conventional techniques, with coverage of the GTV in 95 per cent of the time.67 Implantations of fiducial markers within the bladder wall have been described by investigators at the Royal Marsden Hospital to aid both localisation of boost volumes and verification of bladder position.67a These strategies demonstrate the potential of ART for targeted therapies including reduced volume and boost treatments. PREDICTIVE METHODS
In this IGRT strategy, it is assumed that temporal–spatial variations for a particular site and organ in individual patients may be modelled in a predictive manner as it deforms with time from physiological activity. Using this ‘a priori’ knowledge of the pattern of change for a particular organ, predictions may be undertaken to estimate the potential shape and location of the target at the point of irradiation. This strategy, termed POLO or Predictive Organ Localization, initially developed at the University of Manchester and subsequently at the RMH, is most suitable for organs that progressively change during irradiation i.e., bladder filling. There is substantial variation in bladder size/shape between daily bladder radiotherapy fractions, with a change in bladder dimensions of up to 2 cm.68 In addition, filling of the bladder will also cause change in the bladder size and shape during delivery of radiotherapy.69 Traditionally, patients are requested to either empty or fill their bladders prior to irradiation to minimize this variation. However, for the individual patient, this technique will still result in inconsistent bladder volumes as patients cannot reliably empty or fill their bladder to the exact same volume on their planning scan. The POLO strategy aims to define the temporal–spatial deformity of the bladder from an empty bladder state as it fills over a period of time similar to that needed to deliver a fraction of radiotherapy. Using on-line LINAC 3D imaging, the initial 3D bladder size/shape can be confirmed prior to the irradiation.70 With this information, predictive (POLO) methods can be utilized to predict for the appropriate bladder field at the time of irradiation based on the patient’s own pattern of deformity. Alternatively a combined adaptive–predictive strategy (A-POLO) developed at RMH can be applied using a pre-selection of likely estimated plans derived from the patient’s patterns. These strategies remains under investigation. BREATH-HOLD GATED METHODS
One major issue for thoracic and abdominal including pelvic radiotherapy is respiratory movement. This limitation is well acknowledged and many methods have been developed to monitor, compensate, arrest and gate respiration.
Image-guided radiotherapy 1273
One gated method is patient based breath hold. In this method, the patient and the therapeutic radiographer are active participants to the process. First, the patient is trained via means of a visualized self-spirometry graph to develop a reproducible pattern of breathing and voluntary breath hold. This is illustrated in Figure 51.22. At breath hold, and when the threshold level has been reached, radiation is turned on until the threshold level is breached. Activation of the beam may be taken at different stages of the respiratory cycle. Investigators at Memorial Sloan Kettering Cancer Center have instigated breath hold at deep inspiration/maximum inhalation.71–73 Using this method, compliant patients have been able to maintain reproducible maximum inhalation breath hold for periods of 10–20 s. Inter- and intra-breath hold reproducibility was reported at 1.3 5.3 mm and 1.8 1.0 mm respectively. Another approach is from the William Beaumont Hospital whereby the level of breath hold is monitored and controlled by computer pre-programmed to the individual patient.74 This active breath control (ABC) system monitors the patient’s breathing at the start of the session, and then, when the airflow is matched to the present level, the airflow is temporarily blocked to immobilize the breathing motion. In the ABC system, the level and length of breath hold is expected to be more reproducible per session compared with voluntary systems. Inter- and intra-breath hold reproducibility was reported at 2.0 2.2 mm and 2.6 2.0 mm, respectively, in an early report.74 This system is now commercially available as the Active Breathing Coordinator (Elekta Oncology Systems). An obvious aspect for breath hold is that planning CT scans must be taken in breath hold. This would require 3–6 separate breath-hold scans as well as a normal breathing scan and a normal end-inhalation scan to determine the magnitude of the tumour displacement at the different amplitudes of the breathing cycle. The number of breath holds needed to deliver a radiotherapy fraction depends on fraction size, number of fields, LINAC dose rate and radiation technique. Using 2 Gy per fraction with more than three fields and a LINAC dose rate around 500–600 monitor units per minute, one breath hold may be adequate to allow for delivery of a Inspiration Breath hold
Expiration Radiation Beam 'on'
Time
Figure 51.22 An example of breathing spirometry used to gate the radiation beam to voluntary breath hold.
static CFRT field. Slightly longer treatment times are needed to deliver IMRT fields, due to the complexity. The advantage of breath-hold techniques is that radiation is better targeted, and treatment margins may be reduced with less normal lung volume within the treatment region. There are reports that deep inspiration substantially increases lung expansion,71 and this may be useful in expanding more normal lung out of the treatment region whilst the target volume remains unchanged. This may further reduce radiation lung morbidity whilst permitting the opportunity for dose escalation with or without concomitant chemotherapy. One disadvantage is that not all patients are able to learn this, comply with the procedure or hold their breath for a suitable period of time. This may be an issue for those who are unwell, elderly or with other respiratory illnesses such as emphysema. Specially trained personnel are needed to train patients and additional patient sessions are required. The actual treatment sessions are expected to be longer than a standard radiotherapy session. Equipment set-up for this treatment means that patients using this method should be grouped together rather than being slotted at different times through the daily schedule. Ideally, there should be a dedicated set-up room. RESPIRATORY GATED METHODS
Another method to deal with respiratory motion is to introduce a gating system whereby the radiation beam is correlated with the respiratory cycle. In gating, the respiratory cycle is tracked by methods such as surface chargecoupled device (CCD) camera, infra-red marker, or body trace system and assumes that these surrogate markers of breathing provide a direct correlation between the external chest wall anatomy and the motion of the internal organs such as the tumour and OAR. It is also expected that this relationship determined at simulation is reproducible throughout the radiotherapy course. The first gating system reported was developed at the Tsukaba Proton Therapy Facility. This system monitored respiratory motion using a combination of an airbag and a strain gauge taped onto the patient’s abdomen. The electric signal produced by pressure effects on the system from breathing motion is feedback to either a LINAC75 or proton accelerator76 whereby the gating system triggers the beam ‘on’ or ‘off’ according to whether the signal is located within the pre-set gate ‘on’ window. These investigators reported that the average diaphragm excursion was between 30 mm and 70 mm in deep inspiration and between 10 mm and 25 mm in quiet respiration, and gating of the beam was positioned at expiration phase of quiet respiration. This phase was selected as it was reported to represent the most reproducible position. Subsequently other gating systems have been developed. One system uses a video camera to detect the spatial position of an infra-red light source positioned on the patient’s upper torso. The respiratory cycle is indirectly tracked by the vertical motion of the infra-red light source and is used to gauge
1274 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
when the signal is within the gate window to trigger the beam. Different versions of the same concept of respiratory monitoring have been reported.77–79 Another method of respiratory monitoring is via plethysmography, which is a common method of monitoring breathing in critical care wards. This method uses a strap wrapped around the patient (Respiband®) and the signal is generated from the expansion and shrinkage of the area covered by the strap.79 This method avoids the need for a video camera and can be easily transported from one room to another using a laptop. One potential disadvantage of gated systems is that patients are relied upon to maintain a regular pattern of respiration with acceptable amplitude between inspiration and expiration. Some patients may become too relaxed and their breathing amplitude may decrease. This can be prevented by constant audio feedback and/or instructions to keep the patient alert. Audio instructions for breathing coaching can also help to develop suitable regular breathing patterns.77,80 TRACKING METHODS
Tracking of the tumour is another IGRT method. Surrogates of tumour position are used to identify its position and the treatment beam is tracked to the spatial position of the marker. These radiographic markers or fiducials are generally metal in composition. Gold or titanium is the metal usually used. In most cases, these fiducial(s) will need to be implanted, so it is an invasive procedure requiring either a local or general anaesthetic depending on where the implant is needed. This is potentially the main drawback. The principle is to track the fiducial(s) in real time on the LINAC with the beam guided to the position of the fiducial either as the fiducial moves or is within a ‘zone’ for treatment.81–85 The latter concept is a variation of the gated system described above. In the track and gate methodology, a spatial volumetric zone is generated around the known spatial co-ordinates of the marker within the patient from the planning scan. The beam is turned on when the fiducial enters the zone of treatment. The parameters of this volumetric zone are dictated by the anatomical region, the nature and sensitivity of the surrounding OAR, the proximity of the OAR to the target and the dose-fractionation schedule. Investigators at the University of Hokkaido were the first team to design and implement a room-based kV imaging and tracking system.81 This system employed four pairs of diagnostic X-ray television systems integrated within the LINAC treatment room. Each fluoroscopic imager is permanently fixed in its location with its axis adjusted such that the central axis of the individual systems would cross at the beam’s isocentre (Fig. 51.23). Although only two pairs of fluoroscopic imagers are needed to locate the spatial position of the fiducial, the other two pairs of fluoroscopic imagers provided extra imaging flexibility if the LINAC gantry head blocked any other imagers by its position. This system utilized the volumetric zonal treatment concept described previously. The pattern of motion for each fiducial is calculated by continuous monitoring of its activity,
Figure 51.23 The University of Hokkaido Medical School fluoroscopic real-time tumour tracking system (Reproduced from Shirato et al. Int J Radiation Oncology Biol Phys, 48(2):435–442, 2000).
and this information is then used to characterize the target or organ motion. Using a moving object recognition software system, this system has been reported to have an accuracy of
1 mm in determining the position of the implanted marker every 0.033 seconds during radiotherapy.82 These investigators reported that they were able to reduce fiducial motion from a range of 9.6–38.4 mm down to 2.5–5.3 mm.82 Using fluoroscopy, they were able to also correct for systematic and random errors in treatment delivery. This tracking schema has been examined for prostate,84,86 bladder,84 liver,87,88 pancreatic,89 stomach,90 oesophagus90 and lung radiotherapy,85,88 demonstrating that this is technically feasible. Preliminary studies in hypofractionated prostate radiotherapy to a total dose of 70 Gy in 2.5 Gy per fraction suggested that real time tracking was able to minimized the incidence of acute morbidity.91 Assessment of implanted fiducial markers to the bronchial tree was reported to be useful for the setup of peripheral lung radiotherapy fields with an accuracy of
2 mm but it was also noted the relationship between the markers and tumour can alter substantially after 2 weeks thereby indicating that a combined adaptive and tracking method is required.92 Other investigators are working on the ability to track the fiducial with treatment to the target in real time. This has been explored by Tohoku University Hospital in Japan.93 In this system, 2 kV X-ray tubes are mounted perpendicular to the LINAC gantry head axis with corresponding a-Si EPID devices with the axis of all the devices intersecting with the beam’s isocentre. This team have tracked fiducials using real-time fluoroscopy and translated this motion into coverage with dynamic MLC sequences. One potential advantage of this method is that the beam duty cycle needs to be 100 per cent, but it is possible that MLC shaping may not be
Image-guided radiotherapy 1275
Figure 51.24 The Cyberknife system. (Courtesy of Accuray Inc, Sunnyvale, CA, USA).
efficient enough to accommodate this in high dose rate mode. Using this system for prostate IMRT, it was reported the intra-fractional mean displacement of the fiducial markers could be reduced from 0.45 mm to 0.1 mm in the right–left direction, from 1.08 mm to 0.2 mm in the cranial–caudal direction and from 1.45 mm to 0.2 mm in the anterior–posterior direction.94 This study demonstrates the potential for substantially reducing intra-fractional uncertainties, which can increase the benefits of IMRT dose escalation programs in prostate cancer. Another example of an in-room kV monitoring system is the Cyberknife (Accuray Inc, Sunnyvale, CA, USA). This system incorporates a 6 MV beam from an accelerator mounted on a computer-controlled robotic arm (Fig. 51.24). It is equipped with two pairs of fluoroscopic imagers fixed to the ceiling and floor, respectively, with their imaging axis matched at the beam’s axis. These enabled localization and monitoring of any implanted fiducial markers. The investigators report that the CyberKnife allows delivery with a precision of 0.5 mm and a tracking error of 1 mm to the position of the radioopaque implanted markers.95 This system can be used to track breathing cycles and can be designed to incorporate individual breathing patterns after several cycles for prediction of target location, such that the computercontrolled robotic arm can anticipate the next location almost continuously. The benefit of having the beam treating continuously is that the beam duty cycle can be at its maximum, i.e., 100 per cent. Some limitations for this system are the projected time for treatment of around 40 minutes for prostate radiotherapy, and the collimators are limited by size with the maximum circle of 5 cm.
deliveries, or to compensate for temporal–spatial variations of internal organs, must have rigorous QA to ensure that any modifications are appropriate. This is needed to confirm that the principles on which the methods are based correctly reflects the daily scenario during treatment and is reproducible. Simulation of the individual cases is important particularly for breath-hold and gating methodologies as these sessions now have to incorporate patient training, immobilization, assessment of IGRT systems to planning scans, fluoroscopic movies and gated images. For breath-hold systems, the accuracy of the methodologies employed relies on the reliability of patient positioning and the reproducibility of the residual lung volume during guided breathing at treatment. Patient positioning can be substantially improved using fluoroscopic imaging for treatment set-up and most systems have now incorporated some form of on-line imaging in their system. The key to IGRT programs is experienced and diligent radiation therapists well accustomed to the IGRT system in question. Team members will need to be adequately trained in order to gain the expertise for this undertaking. Continued monitoring of the IGRT program including repeated checks for all stages of the process is essential including the need to perform acceptance testing and commissioning of the equipment. This is essential to maintain the precision of the system and to permit corrections of any errors that may have arisen during the process. Policies and standard operating procedures specific to IGRT will need to be written.
KEY LEARNING POINTS ●
●
●
●
Quality assurance in IGRT Any system designed to modify the position of the patient and their physical actions during or between radiotherapy
●
An important step in radiotherapy treatment planning is to utilize the best imaging modality for the cancer subsite being treated or to incorporate multi-modality imaging to enable more accurate delineation of target volumes and critical normal structures. Conformal radiotherapy (CFRT) is a technique whereby the treatment fields are shaped to the reconstructed profile of the 3D target in the ‘beam’s-eye view’. Randomized data supports the use of CFRT compared with unshaped radiotherapy fields to reduce unnecessary irradiation of organs at risk (OAR) and decrease radiation-related toxicity. The intensity-modulated radiotherapy (IMRT) paradigm, regardless of the method of inverse planning optimization or method of delivery, is based on the ability to deliver a non-uniform beam fluence that can provide improved coverage of irregularly shaped targets, permit more homogenous target-dose distribution and conformal avoidance of OAR. Both CFRT and IMRT techniques can be used to improve the therapeutic ratio in radiotherapy by
1276 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
●
●
●
●
●
permitting dose escalation with lower or acceptable rates of normal tissue morbidity. Internal organ motion and/or organ deformation is an important issue for CFRT and IMRT as it can limit the potential benefits of these techniques. Image-guided radiotherapy (IGRT) or 4D strategies are being designed in many radiotherapy sites to overcome the limitations placed by internal organ motion such as respiration or physiological emptying and filling of internal abdominal organs. Strategies under investigation include adaptive, predictive, gated and tracking methods. Major technology developments have permitted the installation of on-line imaging devices on the linear accelerator or the development of radiotherapy systems for improved real-time imaging using on-line fluoroscopy and cone-beam imaging to assist IGRT and treatment verification. Treatment verification is a crucial part of the radiotherapy process and must form an integral component of the radiotherapy quality assurance program. The development of systems and protocols to incorporate temporal–spatial information to optimize radiotherapy treatment planning and treatment delivery is likely to replace many current radiotherapy schemes with the promise of improved outcomes for patients. These new technology and systems will need to be carefully studied to document its advantages and disadvantages before incorporation into routine practice.
REFERENCES 1 Takahashi S. Conformation radiotherapy: rotation techniques as applied to radiography and radiotherapy of cancer. Acta Radiol 1965;Suppl:242. 2 Hall EJ. Radiobiology for the Radiologist. 4th ed. Philadelphia: Lippincott-Raven, 1994. 3 Williams MV, Denekamp J, Fowler JF. Dose response relationships for human tumours: implications for clinical trials of dose modifying agents. Int J Radiat Oncol Biol Phys 1984; 10:1703–7. 4 Suit HD, Westgate SJ. Impact of improved local control on survival. Int J Radiat Oncol Biol Phys 1986; 12(4):453–8. 5 Thames HD, Schultheiss TE, Hendry JH, Tucker SL, Dubray BM, Brock WA. Can modest escalations of dose be detected as increased tumor control? Int J Radiat Oncol Biol Phys 1992; 22:241–246. 6 Armstrong J, Raben A, Zelefsky M, et al. Promising survival with three-dimensional conformal radiation therapy for nonsmall cell lung cancer. Radiother Oncol 1997; 44(1):17–22.
7 Dearnaley DP, Khoo VS, Norman AR, et al. Comparison of radiation side effects of conformal and conventional radiotherapy in prostate cancer: a randomised trial. Lancet 1999; 353(9149):267–72. 8 NICE. (National Institute of Clinical Excellence). Improving outcomes in urological cancers. The Manual. In. London: NICE; September 2002, 67. 9 ICRU-50. International Commission on Radiation Units and Measurements. ICRU Report 50: Prescribing, recording, and reporting photon beam therapy. Bethesda, MD: International Commission on Radiation Units and Measurement, 1993, 3–16. 10 ICRU-62. International Commission on Radiation Units and Measurements. ICRU Report 62: Prescribing, recording, and reporting photon beam therapy. Bethesda, MD: International Commission on Radiation Units and Measurement, 1999, 3–20. 11 Goitein M, Wittenberg J, Mendiondo M, et al. The value of CT scanning in radiation therapy treatment planning: A prospective study. Int J Radiat Oncol Biol Phys 1979; 5:1787–1798. 12 Goitein M. The utility of computed tomography in radiation therapy; an estimate of outcome. Int J Radiat Oncol Biol Phys 1979; 5:1799–1807. 13 Sundar S, Symonds RP. Diagnostic radiology for radiotherapist: the case for structured training in crosssectional imaging (CT and MRI). Clin Oncol (R Coll Radiol) 2002; 14(5):413–4. 14 Board of the Faculty of Clinical Oncology. Imaging for oncologists: collaboration between clinical radiologists and clinical oncologists in diagnosis, staging and radiotherapy planning. London: Royal College of Radiologists, 2004. 15 Brahme A. Inverse radiation therapy planning: principles and possibilities. In: The use of computers in radiation therapy. Proceedings of the 11th Conference, Manchester, 1994. 16 Webb S. Treatment Plan Optimisation. In: The physics of three-dimensional radiation therapy: conformal radiotherapy, radiosurgery and treatment planning. Bristol: IOP Publishing, 1993, 65–124. 17 Oldham M, Neal AJ, Webb S. A comparison of conventional forward planning with inverse planning for 3D conformal radiotherapy of the prostate. Radiother Oncol 1995; 35:248–262. 18 Oldham M, Webb S. The optimization and inherent limitations of 3D conformal radiotherapy treatment plans of the prostate. Br J Radiol 1995; 68(812):882–93. 19 Xing L, Hamilton RJ, Pelizzari C, Chen GTY. A threedimensional algorithm for optimizing beam weights and wedge filters. Med Phys 1998; 25:1858–1865. 20 Oldham M, Neal AJ, Webb S. The optimisation of wedge filters in radiotherapy of the prostate. Radiother Oncol 1995; 37:209–220. 21 Li JG, Boyer AL, Xing L. Clinical implementation of wedge filter optimization in three-dimensional radiotherapy treatment planning. Radiother Oncol 1999; 53: 257–264. 22 Chen GTY, Spelbring DR, Pelizzari C, et al. The use of beam eye view volumetrics in the selection of noncoplanar
References 1277
23
24
25
26
27
28
29
30
31
32
33
34 35 36
37 38
39
40
radiation portals. Int J Radiat Oncol Biol Phys 1992; 23:153–163. Gokhale P, Hussein EMA, Kulkarni N. Determination of beam orientation in radiotherapy planning. Med Phys 1994; 21:393–400. Rowbottom CG, Webb S, Oldham M. Beam-orientation customization using an artificial neural network. Phys Med Biol 1999; 44:2251–2262. Rowbottom CG, Khoo VS, Webb S. Simultaneous optimization of beam orientations and beam weights in conformal radiotherapy. Med Phys 2001; 28(8):1696–702. Perks JR, Jalali R, Cosgrove VP, et al. Optimization of stereotactically guided conformal treatment planning of sellar and parasellar tumors based on normal brain dose volume histograms. Int J Radiat Oncol Biol Phys 1999; 45:507–513. Bedford JL, Khoo VS, Oldham M, Dearnaley DP, Webb S. A comparison of coplanar four-field techniques for conformal radiotherapy of the prostate. Radiother Oncol 1999; 51(3):225–5. Khoo VS, Bedford JL, Webb S, Dearnaley DP. An evaluation of three-field coplanar plans for conformal radiotherapy of prostate cancer. Radiother Oncol 2000; 55(1):31–40. Bedford JL, Khoo VS, Webb S, Dearnaley DP. Optimization of coplanar six-field techniques for conformal radiotherapy of the prostate. Int J Radiat Oncol Biol Phys 2000; 46(1):231–. Khoo VS, Bedford JL, Webb S, Dearnaley DP. Evaluation of the optimal co-planar field arrangement for use in the boost phase of dose escalated conformal radiotherapy for localized prostate cancer. Br J Radiol 2001; 74(878):177–82. Khoo VS, Bedford JL, Webb S, Dearnaley DP. Class solutions for conformal external beam prostate radiotherapy. Int J Radiat Oncol Biol Phys 2003; 55(4):1109–20. Bedford JL, Henrys AJ, Dearnaley DP, Khoo VS. Treatment planning evaluation of non-coplanar techniques for conformal radiotherapy of the prostate. Radiother Oncol 2005; 75(3):287–292. Kutcher GJ, Coia L, Gillin M, et al. Comprehensive QA for radiation oncology: report of AAPM Radiation Therapy Committee Task Group 40. Med Phys 1994; 21(4):581–618. Brahme A. The importance of non-uniform dose delivery in treatment planning. Phys Med Biol 1994; 39A(1):498. Mackie TR, Holmes TW, Deasy JO, Reckwert PJ. New trends in treatment planning. Phys Med Biol 1994; 39A(1):480. Emami B, Lyman J, Brown A, et al. Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 1991; 21(1):109–22. Brucer M. An automatic controlled pattern cesium-137 teletherapy machine. Am J Roentgenol 1956; 75:49–55. Kallman P, Lind B, Eklof A, Brahme A. Shaping of arbitrary dose distributions by dynamic multileaf collimation. Phys Med Biol 1988; 33:1291–1300. Spirou SV, Chui CS. Generation of arbitrary intensity profiles by dynamic jaws or multileaf collimators. Med Phys 1994; 21(7):1031–41. Que W. Comparison of algorithms for multileaf collimator field segmentation. Med Phys 1999; 26(11):2390–6.
41 Budgell GJ. Temporal resolution requirements for intensity modulated radiation therapy delivered by multileaf collimators. Phys Med Biol 1999; 44(7):1581–96. 42 Bortfeld TR, Kahler DL, Waldron TJ, Boyer AL. X-ray field compensation with multileaf collimators. Int J Radiat Oncol Biol Phys 1994; 28(3):723–30. 43 Yu CX. Intensity-modulated arc therapy with dynamic multileaf collimation: an alternative to tomotherapy. Phys Med Biol 1995; 40(9):1435–49. 44 Convery DJ, Rosenbloom ME. The generation of intensitymodulated fields for conformal radiotherapy by dynamic multileaf collimation. Phys Med Biol 1992; 37:1359–1374. 45 Carol MP. PEACOCK: a system for planning and rotational delivery of intensity-modulated fields. Int J Imaging Syst Technol 1995; 6:56–61. 46 Mackie TR, Holmes T, Swerdloff S, et al. Tomotherapy: a new concept for the delivery of dynamic conformal radiotherapy. Med Phys 1993; 20(6):1709–19. 47 Earl MA, Shepard DM, Naqvi S, Li XA, Yu CX. Inverse planning for intensity-modulated arc therapy using direct aperture optimization. Phys Med Biol 2003; 48(8):1075–89. 48 Pirzkall A, Carol M, Lohr F, Hoss A, Wannenmacher M, Debus J. Comparison of intensity-modulated radiotherapy with conventional conformal radiotherapy for complex-shaped tumors. Int J Radiat Oncol Biol Phys 2000; 48(5):1371–80. 49 Pirzkall A, Debus J, Haering P, et al. Intensity modulated radiotherapy (IMRT) for recurrent, residual, or untreated skull-base meningiomas: preliminary clinical experience. Int J Radiat Oncol Biol Phys 2003; 55(2):362–72. 50 Zelefsky MJ, Fuks Z, Happersett L, et al. Clinical experience with intensity modulated radiation therapy (IMRT) in prostate cancer. Radiother Oncol 2000; 55(3):241–9. 51 Leibel SA, Fuks Z, Zelefsky MJ, et al. Technological advances in external-beam radiation therapy for the treatment of localized prostate cancer. Semin Oncol 2003; 30(5):596–615. 52 Pickett B, Vigneault E, Kurhanewicz J, Verhey L, Roach M. Static field intensity modulation to treat a dominant intra-prostatic lesion to 90 Gy compared to seven field 3-dimensional radiotherapy. Int J Radiat Oncol Biol Phys 1999; 44(4):921–9. 53 Xia P, Pickett B, Vigneault E, Verhey LJ, Roach M, 3rd. Forward or inversely planned segmental multileaf collimator IMRT and sequential tomotherapy to treat multiple dominant intraprostatic lesions of prostate cancer to 90 Gy. Int J Radiat Oncol Biol Phys 2001; 51(1):244–54. 54 Chao KS, Deasy JO, Markman J, et al. A prospective study of salivary function sparing in patients with head-and-neck cancers receiving intensity-modulated or three-dimensional radiation therapy: initial results. Int J Radiat Oncol Biol Phys 2001; 49(4):907–16. 55 Lee N, Xia P, Quivey JM, et al. Intensity-modulated radiotherapy in the treatment of nasopharyngeal carcinoma: an update of the UCSF experience. Int J Radiat Oncol Biol Phys 2002; 53(1):12–22. 56 Eisbruch A, Ship JA, Dawson LA, et al. Salivary gland sparing and improved target irradiation by conformal and intensity modulated irradiation of head and neck cancer. World J Surg 2003; 27(7):832–7.
1278 Conformal radiotherapy, intensity-modulated radiotherapy and image-guided radiotherapy
57 Lin A, Kim HM, Terrell JE, Dawson LA, Ship JA, Eisbruch A. Quality of life after parotid-sparing IMRT for head-and-neck cancer: a prospective longitudinal study. Int J Radiat Oncol Biol Phys 2003; 57(1):61–70. 58 Langen KM, Pouliot J, Anezinos C, et al. Evaluation of ultrasound-based prostate localization for image-guided radiotherapy. Int J Radiat Oncol Biol Phys 2003; 57(3):635–44. 59 Chandra A, Dong L, Huang E, et al. Experience of ultrasoundbased daily prostate localization. Int J Radiat Oncol Biol Phys 2003; 56(2):436–47. 60 Jaffray DA, Drake DG, Moreau M, Martinez AA, Wong JW. A radiographic and tomographic imaging system integrated into a medical linear accelerator for localization of bone and soft-tissue targets. Int J Radiat Oncol Biol Phys 1999; 45(3): 773–89. 61 Jaffray DA, Siewerdsen JH, Wong JW, Martinez AA. Flat-panel cone-beam computed tomography for image-guided radiation therapy. Int J Radiat Oncol Biol Phys 2002; 53(5): 1337–49. 62 Yan D, Wong JW, Gustafson G, Martinez A. A new model for “accept or reject” strategies in off-line and on-line megavoltage treatment evaluation. Int J Radiat Oncol Biol Phys 1995; 31(4):943–52. 63 Yan D, Ziaja E, Jaffray D, et al. The use of adaptive radiation therapy to reduce setup error: a prospective clinical study. Int J Radiat Oncol Biol Phys 1998; 41(3):715–20. 64 Yan D, Lockman D, Brabbins D, Tyburski L, Martinez A. An offline strategy for constructing a patient-specific planning target volume in adaptive treatment process for prostate cancer. Int J Radiat Oncol Biol Phys 2000; 48(1):289–302. 65 Lotz HT, Remeijer P, van Herk M, et al. A model to predict bladder shapes from changes in bladder and rectal filling. Med Phys 2004; 31(6):1415–23. 66 Lotz HT, van Herk M, Betgen A, Pos F, Lebesque JV, Remeijer P. Reproducibility of the bladder shape and bladder shape changes during filling. Med Phys 2005; 32(8):2590–7. 67 Pos FJ, Hulshof M, Lebesque J, et al. Adaptive radiotherapy for invasive bladder cancer: a feasibility study. Int J Radiat Oncol Biol Phys 2006; 64(3):862–8. 67a Mangar S, Thompson A, Miles E, et al. A feasibility study of using gold seeds as fiducial markers for bladder localization during radical radiotherapy. Br J Radiol 2007; 80(952): 279–83. 68 Turner SL, Swindell R, Bowl N, et al. Bladder movement during radiation therapy for bladder cancer: implications for treatment planning. Int J Radiat Oncol Biol Phys 1997; 39(2):355–60. 69 McBain CA, Sykes JS, Buckley DL, et al. Optimising bladder radiotherapy: MR assessment of time-dependent organ motion. Clin Oncol (R Coll Radiol) 2003; 15(S2):13. 70 Henry AM, Stratford J, McCarthy C, et al. X-ray volume imaging in bladder radiotherapy verification. Int J Radiat Oncol Biol Phys 2006; 64(4):1174–8. 71 Hanley J, Debois MM, Mah D, et al. Deep inspiration breathhold technique for lung tumors: the potential value of target immobilization and reduced lung density in dose escalation. Int J Radiat Oncol Biol Phys 1999; 45(3):603–11.
72 Rosenzweig KE, Hanley J, Mah D, et al. The deep inspiration breath-hold technique in the treatment of inoperable nonsmall-cell lung cancer. Int J Radiat Oncol Biol Phys 2000; 48(1):81–7. 73 Mah D, Hanley J, Rosenzweig KE, et al. Technical aspects of the deep inspiration breath-hold technique in the treatment of thoracic cancer. Int J Radiat Oncol Biol Phys 2000; 48(4):1175–85. 74 Wong JW, Sharpe MB, Jaffray DA, et al. The use of active breathing control (ABC) to reduce margin for breathing motion. Int J Radiat Oncol Biol Phys 1999; 44(4):911–9. 75 Ohara K, Okumura T, Akisada M, et al. Irradiation synchronized with respiration gate. Int J Radiat Oncol Biol Phys 1989; 17(4):853–7. 76 Inada T, Tsuji H, Hayakawa Y, Maruhashi A, Tsujii H. [Proton irradiation synchronized with respiratory cycle]. Nippon Igaku Hoshasen Gakkai Zasshi 1992; 52(8):1161–7. 77 Kubo HD, Shapiro EG, Seppi EJ. Potential and role of a prototype amorphous silicon array electronic portal imaging device in breathing synchronized radiotherapy. Med Phys 1999; 26(11):2410–4. 78 Minohara S, Kanai T, Endo M, Noda K, Kanazawa M. Respiratory gated irradiation system for heavy-ion radiotherapy. Int J Radiat Oncol Biol Phys 2000; 47(4):1097–103. 79 Kubo HD, Len PM, Minohara S, Mostafavi H. Breathingsynchronized radiotherapy program at the University of California Davis Cancer Center. Med Phys 2000; 27(2):346–53. 80 Kubo HD, Wang L. Introduction of audio gating to further reduce organ motion in breathing synchronized radiotherapy. Med Phys 2002; 29(3):345–50. 81 Shirato H, Shimizu S, Shimizu T, Nishioka T, Miyasaka K. Real-time tumour-tracking radiotherapy. Lancet 1999; 353(9161):1331–2. 82 Shirato H, Shimizu S, Kitamura K, et al. Four-dimensional treatment planning and fluoroscopic real-time tumor tracking radiotherapy for moving tumor. Int J Radiat Oncol Biol Phys 2000; 48(2):435–42. 83 Shirato H, Shimizu S, Kunieda T, et al. Physical aspects of a real-time tumor-tracking system for gated radiotherapy. Int J Radiat Oncol Biol Phys 2000; 48(4):1187–95. 84 Shimizu S, Shirato H, Kitamura K, et al. Use of an implanted marker and real-time tracking of the marker for the positioning of prostate and bladder cancers. Int J Radiat Oncol Biol Phys 2000; 48(5):1591–7. 85 Harada T, Shirato H, Ogura S, et al. Real-time tumor-tracking radiation therapy for lung carcinoma by the aid of insertion of a gold marker using bronchofiberscopy. Cancer 2002; 95(8):1720–7. 86 Kitamura K, Shirato H, Seppenwoolde Y, et al. Threedimensional intrafractional movement of prostate measured during real-time tumor-tracking radiotherapy in supine and prone treatment positions. Int J Radiat Oncol Biol Phys 2002; 53(5):1117–23. 87 Kitamura K, Shirato H, Seppenwoolde Y, et al. Tumor location, cirrhosis, and surgical history contribute to
References 1279
88
89
90
91
tumor movement in the liver, as measured during stereotactic irradiation using a real-time tumor-tracking radiotherapy system. Int J Radiat Oncol Biol Phys 2003; 56(1):221–8. Shirato H, Seppenwoolde Y, Kitamura K, Onimura R, Shimizu S. Intrafractional tumor motion: lung and liver. Semin Radiat Oncol 2004; 14(1):10–8. Ahn YC, Shimizu S, Shirato H, et al. Application of real-time tumor-tracking and gated radiotherapy system for unresectable pancreatic cancer. Yonsei Med J 2004; 45(4):584–90. Hashimoto T, Shirato H, Kato M, et al. Real-time monitoring of a digestive tract marker to reduce adverse effects of moving organs at risk (OAR) in radiotherapy for thoracic and abdominal tumors. Int J Radiat Oncol Biol Phys 2005; 61(5):1559–64. Kitamura K, Shirato H, Shinohara N, et al. Reduction in acute morbidity using hypofractionated intensitymodulated radiation therapy assisted with a fluoroscopic real-time tumor-tracking system for prostate cancer:
92
93
94
95
preliminary results of a phase I/II study. Cancer 2003; 9(4):268–76. Imura M, Yamazaki K, Shirato H, et al. Insertion and fixation of fiducial markers for setup and tracking of lung tumors in radiotherapy. Int J Radiat Oncol Biol Phys 2005; 63(5):1442–7. Takai Y, Mitsuya M, Nemoto K, et al. Development of real time tumor tracking system with cMLC with dual x-ray fluoroscopy and amorphous silicon flat panel on the gantry of linear accelerator. Int J Radiat Oncol Biol Phys 2002; 54S:193. Britton KR, Takai Y, Mitsuya M, Nemoto K, Ogawa Y, Yamada S. Evaluation of inter- and intrafraction organ motion during intensity modulated radiation therapy (IMRT) for localized prostate cancer measured by a newly developed on-board image-guided system. Radiat Med 2005; 23(1):14–24. King CR, Lehmann J, Adler JR, Hai J. CyberKnife radiotherapy for localized prostate cancer: rationale and technical feasibility. Technol Cancer Res Treat 2003; 2(1):25–30.
52 Concomitant chemo-radiotherapy principles and management SYED A. HUSSAIN, ANJALI ZARKAR AND NICHOLAS D. JAMES
Introduction Models of interaction between chemotherapy and radiotherapy Mechanisms of radiosensitization by chemotherapy Cisplatin
1280 1281 1282 1283
INTRODUCTION The goal of combining systemic therapy (mainly chemotherapy, but also hormone therapy as well as, increasingly, the newer targeted molecular therapies) and radiotherapy is to increase local tumour control, relapse-free survival and overall survival. Clinical trials have shown that systemic therapy given concurrently with radiotherapy can result in improved local control in a range of different disease sites, and in some cases this has also translated into improved survival.1 It is probably under-explored as a therapeutic strategy since much of the thrust of radiation research has been on issues such as oxygenation, fractionation, conformation of dose to target volume and dose escalation. Whilst these developments in radiation oncology are important, the chemotherapy literature has arguably demonstrated much bigger therapeutic gains from successful combination therapy, ranging from the seminal results with Hodgkin’s disease to recent gains in the adjuvant therapy of HER-2 positive breast cancer. In the radiation context, this is best exemplified by the progressive dose escalation studies in early prostate cancer, which to date have only shown improvement in biochemical outcomes with no overall survival benefit. In contrast, a range of studies using combinations of radiation and hormone therapy have, in locally advanced prostate cancer, consistently shown significant survival gains (Table 52.1). There is now increasing evidence that combining chemotherapy with radiotherapy can be achieved safely and with substantial therapeutic gains, summarized in the second part of this chapter.
Targeted molecular therapies Clinical evidence from studies with chemo-radiotherapy Conclusion References
1283 1283 1286 1287
Table 52.1 Interaction of hormone therapy with radiotherapy (RT) Study Bolla 5 yr EORTC 228631 Pilepich 10 yr RTOG 85-312 EPC 7.4 yr3
RT ⴙ hormones (%)
RT only (%)
hazard ratio (HR)
87
55
0.42
62
44
N/A
59
40
0.56
1
Ref. 2; 2Ref. 3; 3Iverson, personal communication.
This first part of this chapter will cover the theoretical basis of combining these two treatment modalities, and the second part will examine the specific clinical settings in which it may be employed on the basis of current evidence. This will clearly need to be examined in the context of the various disease-specific chapters that cover the management of these tumours in more detail. Chemotherapy on its own is rarely curative in solid tumours, often because of the development of drug resistance. Conversely, surgery or radiotherapy may fail as a sole modality of treatment because of micrometastatic disease already present at the time of treatment or because malignant cells are present beyond the surgical or target volume margin. In addition, radiotherapy may fail due to inherent radioresistance in a tumour initially confined within the treated volume. Radiation therapy aimed at controlling
Models of interaction between chemotherapy and radiotherapy 1281
a tumour as a sole modality of treatment or as an adjuvant treatment is limited by tolerance of the irradiated normal tissues. Compiled data over many years suggests that tissues are able to tolerate certain doses with acceptable toxicities.4 Once dose is increased beyond these nominal ‘tolerance’ doses, the probability of serious radiation induced toxicity increases rapidly. The therapeutic ratio can be improved by improved targeting of radiation or by the combination of radiotherapy with a second means of cell killing with complementary toxicity. Therapeutic gain depends on the differential effects between tumour and normal tissue. Some of the tumour characteristics such as genetic instability, rapid proliferation, cell distribution within the cell cycle, hypoxia and low pH can be manipulated to optimize the probability of tumour control. A second mechanism for therapeutic gain is spatial co-operativity whereby micrometastatic spread is targeted by the systemic therapy whilst bulk local disease is addressed by radiation. We shall now examine these mechanisms in more detail.
MODELS OF INTERACTION BETWEEN CHEMOTHERAPY AND RADIOTHERAPY The principles underlying the combination of chemotherapy with radiotherapy can be summarized under four different headings first proposed by Steele.5 Before running through this classification, it is worth outlining the possible levels of interaction between treatments, which is best summarized in Table 52.2.6
Independent cell kill In this model, there is no direct interaction between the chemotherapy and radiotherapy. Radiotherapy produces a tumour response limited by normal tissue toxicity as summarized above. Similarly, chemotherapy produces a tumour response and is limited by distinct tissue toxicities such as bone marrow suppression. The combination of the two modalities will thus produce a greater response than if only one modality were used but may nonetheless be sub-additive5 and is the concept that largely underlies combination chemotherapy in terms of independent toxicity but simple addition of anti-tumour effects. This model, however, breaks down to a degree when combining chemotherapy and radiotherapy, as there may be an interaction between the two elements resulting in increased toxicity within the radiation field and consequent reduction in the maximum achievable radiation effect. If this is avoided by the non-synchronous delivery of the two modalities (e.g., neoadjuvant chemotherapy followed by radical radiotherapy) the gains from combination may be lost by accelerated tumour cell repopulation during the recovery phase between treatments. Whilst accelerated repopulation has been most extensively studied in the context of radiotherapy,7,8 it is likely to apply just as readily to chemotherapy. For example, contrast the results of neoadjuvant cisplatin with radiotherapy9,10 where there were no important outcome differences with those obtained with synchronous administration of the same doses of cisplatin with radiotherapy11 where a large improvement in local control was observed. Concomitant chemo-radiotherapy (CRT) may thus avoid the problem of accelerated repopulation with enhanced tumour control probability.
Spatial co-operation Protection of normal tissues The term spatial co-operation can be used when chemotherapy and radiotherapy are used to treat a disease at spatially different anatomical locations; this is based on the concept that chemotherapy may help eradicate micro-metastatic disease, while radiotherapy may improve local control of the irradiated site. The classical example is the treatment of acute lymphoblastic leukaemia. Chemotherapy is used as the main line of treatment and radiotherapy is used to treat the central nervous system, which is considered as a sanctuary site. There is no direct interaction between the two modalities of treatment.
Usually combining two modalities of treatments produces more toxicity but certain drugs in experiments have shown to increase the resistance of normal tissues to radiation thus offering radio-protective effect. This can be exploited to deliver a higher dose of radiotherapy to achieve the increased tumour response. In experiments performed on intestinal micro-colonies, cytarabine given 12 hours before irradiation offered protection to the intestinal stem cells from radiation.12 Its effectiveness is dependent on the timing of drug administration and delivery of radiotherapy. This effect has not been exploited to any significant extent in the clinic.
Table 52.2 Models if interaction between two treatments Enhancement Synergism Additive Sub-additive Interference Antagonism
Example
Enhancement of tumour response
214 213 2 1 2.5 2 1 1.5 2 1 0.5
The enhancement of tumour response may be viewed at a number of levels and may partly come under the heading of spatial co-operation. Possible mechanisms are complex and are dealt with in detail in the next section. In the context of CRT, it may be best thought of as occurring when the addition of one agent (in this case chemotherapy) increases the
1282 Concomitant chemo-radiotherapy principles and management
effectiveness of the other agent (radiotherapy) producing an effect that amounts to more than simple addition of effects.6 Possible mechanisms of enhancement are considered in the next section.
MECHANISMS OF RADIOSENSITIZATION BY CHEMOTHERAPY Chemotherapy can act as a radiosensitizing agent by a variety of mechanisms: 1. direct enhancement of the initial radiation damage by incorporating drugs into DNA, 2. inhibiting cellular repair mechanisms, 3. accumulating cells in a radiosensitive phase or eliminating radioresistant phase cells, 4. eliminating hypoxic cells or 5. inhibiting the accelerated repopulation of tumour cells. The lack of overlapping toxicities for these treatments will be an important consideration here as it allows their combination. Acute toxicities have been generally well reported, but long-term toxicities are less well understood as many trials have not really been set up to adequately examine the late effects of combination therapy but instead have focussed on ‘harder’ endpoints such as local control and survival.
Hypoxic cell sensitizers Solid tumours contain poorly oxygenated cells in proportions ranging from 0 to up to 50 per cent,13 and these are disproportionately resistant to therapeutic radiation. Several methods for overcoming this problem have been tested clinically, including the administration of hyperbaric oxygen, spacing of multiple doses of conventional radiation and use of hypoxic cell sensitizers, which may mimic the sensitizing effect of oxygen and are active only against hypoxic cells and not well-oxygenated normal tissues.14 Misonidazole is a nitroimidazole compound and has been found to sensitize hypoxic cells in vitro as well as animal tumours.15 Although promising results were seen in early clinical studies, subsequent clinical testing was disappointing, failing to show any overall advantage.15 This appears to be due to dose-limiting toxicity of peripheral neuropathy leading to misonidazole dose reduction to sub-therapeutic levels. Trials continue with newer agents with positive results reported in head and neck cancer with nimorazole with acceptable toxicity.16 The other group of drugs that selectively kill hypoxic cells are known as bioreductive drugs. Mitomycin has been used as a chemotherapeutic agent along with radiotherapy.17 Clinical results with this agent are discussed in detail below. Bioreductive drugs are reduced preferentially to cytotoxic compounds in hypoxic regions, which occur more commonly in rapidly growing tumours. Tirapazamine has been extensively tested in vitro as well as in vivo18 and has shown promising results in clinical trials used concurrently with radiation.19,20
Reduction in hypoxic fraction Chemotherapy causes a reduction in the number of viable cells leading to a potential decrease in the number of hypoxic cells by a mechanism analogous to that seen with tumour reoxygenation after a radiotherapy fraction. Improved oxygenation may thus lead to an improved response to ionizing radiation when combined with chemotherapy.21 It may of course at the same time reduce repair in normal tissue causing increased normal tissue toxicity and consequently the final net clinical effect will depend on the balance of toxicity to benefit.
Cell synchronization Cells in certain phases of the cell cycle (S, G0 and 1 phase) are relatively resistant to ionizing radiation.22 Chemotherapeutic agents that show selectivity for these phases of the cell cycle will potentially synchronize rapidly dividing cells. Using ionizing radiation to then target these cells in their radio-sensitive phase could therefore achieve better results. Taxanes are potent microtubule stabilizing agents that block or prolong G2/M phase of the cell cycle. This is a radiosensitive phase and is associated with radiosensitization.23
Inhibition of repair of sublethal damage by radiation Radiation causes sublethal and potentially lethal damage. Fractionated treatment allows this damage to be repaired. Combining chemotherapy and certain pharmacological substances that prevent repair of this damage could thus enhance tumour response.24 This is likely to have a beneficial effect only if it has differential effect on the normal tissue repair. There are various drugs that may have effects in this class, which are summarized below. The ideal radiosensitizer would have a synergistic or additive effect on the tumour but a sub-additive effect on normal tissues. HALOGENATED PURINES OR PYRIMIDINES
5-Bromo-deoxyuridine and 5-iodo-deoxypyrimidine are halogenated pyrimidines. They work by replacing thymidine at the time of DNA synthesis. This makes cells more susceptible to X-rays or ultraviolet rays.25 Although these substances have shown some promising results in clinical trials using intra-arterial infusion to treat high-grade glioblastomas and large resectable sarcomas,25 they have not found widespread acceptance due to the difficulty of administration and toxicity. Much more widely used is 5-fluorouracil (5-FU), which is now extensively employed in the clinic as a radiosensitizer and is discussed in detail in the clinical section below.
Clinical evidence from studies with chemo-radiotherapy 1283
ACCELERATED RADIOTHERAPY WITH CARBOGEN AND NICOTINAMIDE (ARCON)
Accelerated radiotherapy with carbogen and nicotinamide (ARCON) is a new approach designed to improve therapeutic gain based on knowledge of dynamics of tumour repopulation and tumour oxygenation. It combines accelerated fractionated radiotherapy to reduce clonal repopulation, the hyperoxic gas carbogen (95 per cent oxygen and 5 per cent carbon dioxide) to reduce diffusion-limited hypoxia and nicotinamide to reduce perfusion-limited hypoxia. This approach uses the combination of different agents to maximize the benefit. Pre-clinical work has shown that ARCON may reduce the radiation dose required for a given effect by up to 50 per cent. 26,27 The enhancement ratios for normal tissues in pre-clinical studies were much lower than the tumours suggesting that this approach may be beneficial27,28,29. Phase II trials in head and neck cancer and bladder carcinoma have demonstrated the feasibility and tolerability of the approach. 29,30 A Phase III, randomized, multicentre trial (BCON) is underway currently in the United Kingdom to evaluate these findings in bladder cancer. Gemcitabine Gemcitabine is a deoxycytidine analogue that is known to have cytotoxic activity against various cancers, such as pancreas, lung, bladder, breast and head and neck. The exact mechanism of radiosensitization is still unclear and various mechanisms have been suggested. One of the suggested mechanism is early ‘S’ phase block by gemcitabine, making tumour cells to synchronously progress into ‘S’ phase after treatment leading to radiosensitization by subsequent radiation.31,32 The other suggested mechanism is induction of apoptosis.33 In addition to its known cytotoxic activity, it has been considered to have potent radiosensitizing effect. This has been confined to in vitro studies on human tumour cells from pancreatic, head and neck cancer and squamous cell carcinoma cell lines.33–35 The radiosensitizing effect of gemcitabine has also been confined to in vivo studies23 and there are various ongoing clinical trials.36–38
CISPLATIN Cisplatin is widely employed in CRT schedules and is discussed in more detail in the section on clinical studies with CRT below.
TARGETED MOLECULAR THERAPIES In this fast developing era of newer targeted therapies, it is logical to examine their combination with exiting modalities such as radiation. There are a number of ongoing clinical trials but more are required. Possible targets include the epidermal growth factor receptor (EGFR), the vascular endothelial growth factor receptor (VEGFR) and p53. Over expression of EGFR is common in many solid tumours, such as colorectal and lung carcinomas as well as
cancers of the head and neck. It correlates with increased metastasis, decreased survival and a poor prognosis. The over-expression of EGFR may protect malignant tumour cells from the cytotoxic effects of chemotherapy and radiotherapy, making these treatments less effective. A range of agents targeting EGFR either via monoclonal antibody or small molecule targeted inhibitors is now available. Targeting of EGFR in head and neck cancer is discussed in more detail below. Angiogenesis plays an important role in tumour growth. Vascular endothelial growth factor and its receptors on endothelial cells play a pivotal role in development of neo-vascularization leading to further tumour growth. Targeting VEGFR therefore makes logical sense as a combination strategy, and early laboratory data has shown that inhibition of VEGFR signalling causes reversal of tumour resistance to radiotherapy.39 p53 is a well known tumour suppressor gene. This tumour suppressor gene has an important role to play in apoptosis and cell cycle regulation. Apoptosis is the pathway of cell regulation in radiation-induced cell damage. It is well known that many solid tumours have an altered expression of p53. A number of gene therapy studies have used p53 as an effector gene, and trials in combination with radiation are therefore feasible.
CLINICAL EVIDENCE FROM STUDIES WITH CHEMO-RADIOTHERAPY Chemo-radiotherapy has been used in different disease settings with mixed success. For anal cancer concomitant CRT now represents optimum primary treatment, obviating the need for surgery in most cases. Similarly promising data exist for cervical cancer and head and neck cancer. Results of phase II trials in bladder cancer are also supporting the design of larger phase III studies, which are now ongoing. These four models provide a paradigm for CRT, where studies have provided evidence of activity of combined modality treatment at the expense of little additional acute or long-term toxicity. Nevertheless, the search for more active and less toxic agents continues.
Anal cancer Nigro and co-workers first described treatment with concomitant radiotherapy and chemotherapy using 5-FU and mitomycin in 1974 in the pre-operative setting with the aim of improving operability and reducing the risk of subsequent recurrence. They observed that the surgical specimens were pathologically free of disease. The response obtained was so convincing that the authors proposed synchronous CRT as a first-line treatment modality for anal cancer.40–42 Surgery is now reserved for salvage and cases for which organ preservation is unsuitable. In 1996, the United Kingdom Co-ordinating Committee on Cancer Research (UKCCCR) published the result of the
1284 Concomitant chemo-radiotherapy principles and management
first randomized trial that compared radiotherapy alone with radiotherapy plus 5-FU and mitomycin in anal carcinoma. A total of 585 patients entered the study; 290 patients received external irradiation only and 295 received combinedmodality treatment. The authors observed a 46 per cent reduction of local failure rate (59 per cent in the radiotherapy arm vs. 36 per cent in CRT arm), after a median follow up of 42 months. Early toxicity was more frequent in the CRT arm but late toxicity was similar in both groups. The risk of death from anal cancer was also significantly reduced in the combined treatment group (17). However, overall survival was not significantly improved in the combined therapy arm. In the trial by the European Organization for Research and Treatment of Cancer (EORTC), 110 patients were randomly assigned radiotherapy alone (45 Gy in 25 fractions, followed 6 weeks later by a boost of 15–20 Gy), or the same radiotherapy schedule with synchronous chemotherapy (mitomycin, 15 mg/m2 on day 1; fluorouracil, 750 mg/m2 on days 1–5 and days 29–33). The combined modality group had better local control (68 per cent vs. 55 per cent at 3 years, p 0.02) and colostomy-free interval (72 per cent vs. 47 per cent at 3 years, p 0.002) than the radiotherapy group.43 Another phase III trial was undertaken by the Radiation Therapy Oncology Group and Eastern Co-operative Oncology Group (RTOG/ECOG), which recruited 310 patients, and showed better local control (72 per cent vs. 64 per cent) and disease-free survival (73 per cent vs. 51 per cent) with mitomycin (10 mg/m2 on days 1 and 29) and fluorouracil (1 g/m2 on days 1–4 and days 29–32), than with fluorouracil alone in combination with radiotherapy (45–50.4 Gy in 25–28 fractions)44. At 4 years, colostomy rates were lower (9 per cent vs. 22 per cent; p 0.002), colostomy-free survival higher (71 per cent vs. 59 per cent; p 0.014), and disease-free survival higher (73 per cent vs. 51 per cent; p 0.0003) in the MMC arm. A significant difference in overall survival was not observed at 4 years. Though in these studies a significant improvement in local control was achieved, overall survival was not significantly better. A significant number of patients still die of metastatic disease. This may be due to inadequate chemotherapy doses to achieve systemic effects. The UKCCCR ACT-II trial addresses this question; it is a randomized trial of a 2 2 design that is comparing mitomycin and 5FU with cisplatin and 5-FU in combination with radiotherapy. The second part of the randomization compares two further cycles of adjuvant cisplatin and 5-FU chemotherapy with no further chemotherapy after completion of CRT.45 In addition, a French trial (Fondation Française de Cancerologie Digestive [FFCD] Trial 9804) randomizes patients between a low-dose or high-dose radiotherapy boost with cisplatin-based CRT, and between neoadjuvant chemotherapy versus no neoadjuvant chemotherapy.46
Bladder cancer Currently, the only randomized evidence for organ preservation in bladder cancer comes from a trial by the National
Cancer Institute of Canada (11). This study randomized 99 patients to receive cisplatin 100 mg/m2 with radiotherapy or radiotherapy alone to a dose of 40 Gy followed by elective cystectomy or further radiotherapy. The CRT arm showed statistically non-significant improvements in complete response rate compared with the radiotherapy arm (24/51 vs. 14/48; difference 16 per cent, p 0.11) and in overall survival (47 per cent vs. 33 per cent at 3 years, log rank p 0.34). Significant differences were seen in the pattern of relapse. The risk of pelvic failure was significantly reduced in the chemo radiotherapy arm (15/51 vs. 36/48, p 0.026) with a corresponding improvement in pelvic progression free survival (67 per cent vs. 47 per cent at 2 years, log rank p 0.038) and bladder preservation (70 per cent vs. 36 per cent, log rank p 0.16). In a published surgical series of 1054 patients with bladder cancer treated between 1971 and 1997 with radical cystectomy, Stein et al. (47) reported excellent overall survival rates that, according to the authors, should be considered as the surgical standard to which other treatment modalities should be compared. They reported a 5- and 10year overall survival in the range of 60 per cent and 43 per cent, respectively, for the whole group of patients, which compared favourably to the results from Rodel et al. 48 who described a large experience with organ-sparing treatment of invasive bladder cancer, documenting long-term outcome of 415 patients treated over a 20-year time period with a 51 per cent overall survival rate at 5 years and 31 per cent at 10 years. However, this cystectomy series included 213 patients (20 per cent) with non-invasive (T0, Ta, Tis) bladder cancer and excluded 112 patients with inoperable tumour, non-radical surgery, or intra-operative detection of distant metastases. Conversely, Rodel’s series did not include superficial noninvasive tumours and did not exclude patients who were, for different reasons, unsuitable for surgery. If comparison is restricted to the operable patients with invasive disease, the 5-year overall survival rates of this radical cystectomy series and Rodel’s bladder-sparing approach become equal: 74 per cent and 75 per cent, respectively, for TI-tumours, and 47 per cent and 45 per cent, respectively, for muscle-invasive disease (48). Although this is a nonrandomized comparison, this analysis indicates that selective bladder preservation by multi-modality treatment may result in long-term cure and survival rates comparable to the best cystectomy series, allowing this treatment to be considered a reasonable treatment option for patients who are deemed medically unfit for cystectomy and for those seeking an alternative to radical cystectomy. Zietman et al.49 and Shipley et al.50,51 from Massachusetts General Hospital and Sauer (52) from Erlangen have reported promising results with organ-preservation strategies over the last decade that offered optimism for an option of bladder preservation in a significant proportion of patients who currently undergo cystectomy. Zietman reported that when radiation and transurethral surgery are combined with cisplatin-based chemotherapy local control rates increase such that the majority of patients preserve a tumour-free functional bladder. Up to 85 per cent of
Clinical evidence from studies with chemo-radiotherapy 1285
patients selected for bladder sparing therapy on the basis of their initial response to CRT kept their bladders. This figure could increase further when other powerful prognostic factors such as the presence of hydronephrosis and the presence of carcinoma in situ are also taken into account in initial patient selection.49 Sauer52 presented a 14-year multivariate analysis of prognostic factors influencing survival and bladder preservation after transurethral resection of a bladder tumour (TURBT) and radiotherapy chemotherapy. Cause-specific survival (CSS) was 59 per cent and 43 per cent after 5 and 10 years; 79 per cent of patients survived with preserved bladder. Five-year CSS after radiotherapy, CRTcisplatin, and CRT-carboplatin was 40 per cent, 64 per cent, and 54 per cent, respectively and 10-year CSS 31 per cent, 48 per cent, and 27 per cent, respectively (p 0.04–0.045, univariate). For relapsed patients after cystectomy, the 5- and 10-year CSS were 40 per cent and 33 per cent.52 Compared with historic controls, the addition of cisplatin or carboplatin in this series led to significantly more complete remissions and better survival. Survival rates reported here were similar to those achieved by primary cystectomy and possibly even better for selected subjects, such as patients with T3b and T4 tumours. The authors concluded that cystectomy should be restricted to only those patients who fail after CRT. Shipley51 reported long-term outcome of patients with invasive bladder cancer who were treated on combined modality organ-preservation strategy. A total of 190 patients were treated using concurrent cisplatin-containing CRT after transurethral resection of the bladder tumour. The 5- and 10-year actuarial overall survival rates were 54 per cent and 36 per cent, respectively (Stage T2, 62 per cent and 41 per cent; Stage T3–T4a, 47 per cent and 31 per cent, respectively). The 5- and 10-year disease-specific survival rates were 63 per cent and 59 per cent (Stage T2, 74 per cent and 66 per cent; Stage T3–T4a, 53 per cent and 52 per cent), respectively. The 5- and 10-year disease-specific survival rates for patients with an intact bladder were 46 per cent and 45 per cent (Stage T2, 57 per cent and 50 per cent; Stage T3-T4a, 35 per cent and 34 per cent), respectively.51 The pelvic failure rate was 8.4 per cent. No patient required cystectomy because of bladder morbidity. The 10-year overall survival and disease-specific survival rates were comparable with the results reported for contemporary radical cystectomy for patients of similar clinical and pathologic stage. These results strengthen the evidence for bladder preservation strategies. A variety of phase II studies have reported the efficacy and toxicity of this approach in bladder cancer generally with results better than that seen with radiotherapy alone, and forms a basis for further work in organ preservation strategies.53,54 It should be emphasized that with organ preservation strategies there is a necessity for long-term follow-up, as a proportion of patients will develop recurrence or second malignancy in the bladder, and patients need to be counselled about the risk of salvage cystectomy and the importance of regular follow-up cystoscopies.51 Clinical criteria helpful in determining patients for bladder preservation include such variables as early tumour
stage, presence or absence of hydronephrosis and a visibly and microscopically complete TURBT. In the United Kingdom recruitment is ongoing to a national phase III trial on CRT, BC2001 (www.bc2001.bham.ac.uk), in which patients are randomized to concurrent CRT versus radiotherapy alone. This trial uses mitomycin C and 5-FU, based on encouraging phase II data in terms of efficacy and short- and long-term tolerability.55,56 The design is pragmatic in that centres can opt for one of the two fractionation schemes: 55 Gy in 20 fractions or the longer 64 Gy in 32 fractions. The trial incorporates quality-of-life and bladder capacity measures and will be able to assess whether the more intensive treatment schedules have a long-term impact on bladder function or symptoms. An alternative approach to improve local control is to use other radiosensitizers targeting tumour hypoxia. A current randomized trial, B-CON, is utilizing carbogen and nicotinamide (ARCON) in conjunction with the same fractionation schedules as those used for BC2001. Together these two trials have now recruited over 600 patients in 4 years indicating that large-scale bladder radiotherapy trials can recruit effectively. The use of newer cytotoxic drugs such as gemcitabine and taxanes synchronous with radiation remains experimental at present but may be incorporated into refined treatment protocols in the future. Improved techniques in delivery of radiation therapy will allow further dose escalation with little additional toxicity, facilitating the addition of further chemotherapeutic drugs and an increase in the dose intensity of concurrent chemotherapy.
Cervical cancer Four large randomized studies on the use of synchronous CRT for cervical cancer 57–60 showed reductions in odds of death or recurrence of around 50 per cent with the addition of synchronous chemotherapy to radiotherapy. This prompted the National Cancer Institute to issue a rare clinical announcement that strong consideration should be given to the incorporation of concurrent cisplatin therapy in women who require radiation therapy for treatment of cervical cancer. A review of randomized clinical trials in locally advanced cervical cancer comparing concomitant CRT with radiotherapy in the experimental arm was recently reported. This included 24 trials (21 published, three unpublished) and 4921 patients, although due to patient exclusion and differential reporting 61–75 per cent were available for the analyses. The review strongly suggested CRT improves overall survival and progression-free survival, whether or not platinum was used, with absolute benefits of 10 per cent and 13 per cent, respectively. There was, however, statistical heterogeneity for these outcomes. There was some evidence that the effect was greater in trials including a high proportion of stage I and II patients. Chemoradiation also showed significant benefit in reducing local recurrence and a
1286 Concomitant chemo-radiotherapy principles and management
suggestion of a benefit for distant recurrence. Acute haematological and gastrointestinal toxicity was significantly greater in the concomitant CRT group. Late effects of treatment were not well reported and so the impact of CRT on these effects could not be determined adequately. Treatment-related deaths were rare. Concomitant CRT appears to improve overall survival and progression-free survival in locally advanced cervical cancer. It also appears to reduce local and distant recurrence, suggesting concomitant chemotherapy may afford radiosensitization and systemic cytotoxic effects. Some acute toxicity is increased, but the long-term side effects are still not clear.61
Head and neck cancers Meta-analyses have reported a consistent survival advantage in favour of concomitant CRT compared with radiotherapy alone in locally advanced squamous cell head and neck cancer (SCHNC). Clinical practice guidelines from several sources endorse the use of platinum-based CRT as standard treatment for SCHNC.62 Two of three early literature-based meta-analyses detected a small overall survival benefit favouring concurrent treatment.63–65 The most recent and largest meta-analysis, using individual patient data,66 has been updated to include 87 randomized clinical trials (RCTs) of CRT versus radiotherapy alone in over 16 000 patients studied between 1965 and 2000.67 Fifty of these trials tested radical radiotherapy with or without concurrent chemotherapy. There was less heterogeneity on subset analysis across studies conducted between 1994 and 2000 compared with those between 1965 and 1993. The updated meta-analysis confirmed the findings of the original analysis, with a 4 per cent overall survival benefit with chemotherapy but a relative 19 per cent improvement in survival (hazard ratio, 0.81; p 0.0001) for concomitant therapy translating into an 8 per cent benefit in 5-year absolute survival rate.67 The update confirmed that the greatest benefit was conferred by platinum-based regimens. Unfortunately, despite encouraging survival results with primary CRT, almost half of patients experience loco-regional relapse. Consequently, several studies have been conducted to evaluate the role of intensified radiotherapy in the context of concurrent CRT as a means of improving loco-regional control with no significant benefit in local control or overall survival.68 Combination of neoadjuvant therapy to surgery is being considered to improve local control. There is a need for further research into improving quality of life and optimizing nutritional support, and in understanding tumour biology in order to stratify patients according to the risk/benefit ratio of various regimens to move towards tailored therapies.
to synergism of drugs targeting the EGFR pathway with radiation therapy. The phenotypic cellular changes associated with EGFR inhibition are impressively consistent between different model systems, with almost all studies showing that EGFR inhibitors affect proliferation, angiogenesis and cell survival. Further research to clarify whether EGFR inhibitors influence response to radiation directly or whether it is the pooled effect of the treatment modalities leading to synergism is required. However, cell-cycle arrest, endothelial cell sensitivity and apoptotic potential are all important factors in radiation response of epithelial tumours that may be modified by the EGFR inhibition. Furthermore, effects of EGFR inhibitors on DNA repair suggest that modulation of DNA damage response to cytotoxic injury might result in radio- or chemosensitization.69 Preliminary data from a phase III trial of cetuximab, a human–murine chimeric monoclonal antibody that inhibits epidermal growth factor receptor signalling in combination with conventional radiotherapy, reported an improvement in both loco-regional control and overall survival with minimal additive toxicity compared with radiotherapy alone.70 Because the control arm in this trial did not include chemotherapy, it is unclear how cetuximab plus radiotherapy compares with CRT. Moreover, given the non-overlapping toxicities of cetuximab and cisplatin, the effect of combining both agents with radiotherapy would be of interest.62
CONCLUSION A multidisciplinary approach with as many eligible patients as possible to enrol into clinical trials to identify further potential radiosensitizers is the key to further progress. Search for better radiosensitizers in this disease setting continues to further improve non-surgical management. Progress in molecular characterization in different disease sites will continue, and definition of new-targeted therapies will provide new opportunities to tailor treatment to the specific needs of the individual. This will further improve the efficacy of CRT and improve organpreserving strategies.
KEY LEARNING POINTS ●
●
EPIDERMAL GROWTH FACTOR RECEPTOR AND HEAD AND NECK CANCER Pre-clinical studies have provided important observations with regard to potential mechanisms of interaction leading
●
Combined systemic therapy with radiotherapy has been shown to improve outcomes in a range of disease sites. Further trials are necessary to expand the knowledge in this fast moving field. New targeted molecular therapies hold great promise for combination with radiotherapy in view of their low toxicity
References 1287
REFERENCES 1 Nishimura Y. Int J Clin Oncol 2004 Dec;9(6):414–20. Bolla M, Gonzalez O, Warde P, et al. Improved survival in patients with locally advanced prostate cancer treated with radiotherapy and goserelin [see comments]. N Engl J Med 1997; 337:295–300. ◆3 Pilepich MY, Caplan R, Byhardt RW, et al. Phase III trial of androgen suppression using goserelin in unfavorableprognosis carcinoma of the prostate treated with definitive radiotherapy: report of Radiation Therapy Oncology Group Protocol 85–31. J Clin Oncol 1997; 15:1013–1021. 4 Emami B, Lyman J, Brown A, et al. Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol BioI Phys 1991; 21(1):109–22. 5 Steel GG. Terminology in the description of drug-radiation interactions. Int J Radiat Oncol Biol Phys 1979; 5(8):1145–50. 6 Phillips TL. Chemical modification of radiation effects. Cancer 1977; 39(2 Suppl):987–98. ◆7 Withers HR, Taylor JM, Maciejewski B. The hazard of accelerated tumor clonogen repopulation during radiotherapy. Acta Oncol 1988; 27(2):131–46. 8 Petersen C, Zips O, Krause M, et al. Repopulation of FaDu human squamous cell carcinoma during fractionated radiotherapy correlates with reoxygenation. Int J Radiat Oncol Bioi Phys 2001; 51(2):483–93. 9 Wallace DMA, Raghavan D, Kelly KA, et al. Neo-adjuvant (pre-emptive) cisplatin therapy in invasive transitional cell carcinoma of the bladder. Br J Uro1 1991; 131:477–482. 10 Cooke PW, Dunn JA, Latief T, Bathers S, James NO, Wallace OM. Long-term risk of salvage cystectomy after radiotherapy for muscle-invasive bladder cancer. Eur Urol 2000; 38: 279–286 11 Coppin CM, Gospodarowicz MK, James K, et al. Improved local control of invasive bladder cancer by concurrent cisplatin and preoperative or definitive radiation. The National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 1996; 14:2901–2907. 12 Phelps TA, Blackett NM. Protection of intestinal damage by pretreatment with cytarabine (cytosine arabinoside). Int J Radiat Oncol Biol Phys 1979; 5(9):1617–20. 13 Moulder JE, Rockwell S. Hypoxic fractions of solid tumors: experimental techniques, methods of analysis, and a survey of existing data. Int J Radiat Oncol BioI Phys 1984; 10(5):695–712. 14 Adams GE, Dische S, Fowler JF, Thomlinson RH. Hypoxic cell sensitisers in radiotherapy. Lancet. 1976; I (7952):186–8. 15 Overgaard J. Clinical evaluation of nitroimidazoles as modifiers of hypoxia in solid tumors. Oncol Res 1994; 6(10–11):509–18. ◆16 Overgaard J, Hansen HS, Overgaard M, et al. A randomized double-blind phase III study of nimorazole as a hypoxic radiosensitizer of primary radiotherapy in supraglottic larynx and pharynx carcinoma. Results of the Danish Head
◆17
◆2
18
19
20
21
22
23
24
25
26
27
28
29
30
and Neck Cancer Study (DAHANCA) Protocol 5–85. Radiother Oncol 1998; 46:135–146 Epidermoid anal cancer: results from the UKCCCR randomised trial of radiotherapy alone versus radiotherapy, 5-fluorouracil, and mitomycin. UKCCCR Anal Cancer Trial Working Party. UK Co-ordinating Committee on Cancer Research. Lancet. 1996; 348(9034):1049–54. Brown JM, Lemmon MJ. Potentiation by the hypoxic cytotoxin SR 4233 of cell killing produced by fractionated irradiation of mouse tumors. Cancer Res 1990; 50(24):7745–9. Treat J, Johnson E, Langer C, et al. Tirapazamine with cisplatin in patients with advanced non-small-cell lung cancer: a phase II study. J Clin Oncol 1998(11):3524–7 Rischin D, Peters L, Fisher R, et al. Tirapazamine, cisplatin, and radiation versus fluorouracil, cisplatin, and radiation in patients with locally advanced head and neck cancer: a randomized phase II trial of the Trans-Tasman Radiation Oncology Group (TRaG 98.02). J Clin Oncol 2005; 23(1):79–87. Palcic B, Skarsgard LD. Reduced oxygen enhancement ratio at low doses of ionizing radiation. Radiat Res 1984; 100(2):328–39. Sinclair WK, Morton RA. X-ray sensitivity during the cell generation cycle of cultured Chinese hamster cells. Radiat Res 1966; 29(3):450–74. Milas L, Milas MM, Mason KA. Combination of taxanes with radiation: preclinical studies. Semin Radiat Oncol 1999; 9(2 Suppl 1):12–26. Kelland LR, Steel GG. Inhibition of recovery from damage induced by ionizing radiation in mammalian cells. Radiother Oncol 1988; 13(4):285–99. Kinsella TJ, Mitchell JB, Russo A, Morstyn G, Glatstein E. The use of halogenated thymidine analogs as clinical radiosensitizers: rationale, current status, and future prospects: non-hypoxic cell sensitizers. Int J Radiat Oncol BioI Phys 1984; 10(8):1399–406. Simon JM, LartigauE, Guichard M. Nicotinamide and carbogen: major effect on the radiosensitivity of EMT6 and HRT18 tumours. Radiother Oncol 1993; 28(3):203–7 Kjellen E, Joiner MC, Collier JM, Johns H, Rojas A. A therapeutic benefit from combining normobaric carbogen or oxygen with nicotinamide in fractionated X-ray treatments. Radiother Oncol 1991; 22(2):81–91. Horsman MR, Siemann OW, Chaplin OJ, Overgaard J. Nicotinamide as a radiosensitizer in tumours and normal tissues: the importance of drug dose and timing. Radiother Oncol 1997; 45(2):167–74. Hoskin PI, Rojas AM, Phillips H, Saunders MI. Acute and late morbidity in the treatment of advanced bladder carcinoma with accelerated radiotherapy, carbogen, and nicotinamide. Cancer 2005; 103(11):2287–97. Kaanders JR, Pop LA, Marres HA, et al. ARCON: experience in 215 patients with advanced head-and-neck cancer. Int J Radiat Oncol BioI Phys 2002; 52(3):769–78.
1288 Concomitant chemo-radiotherapy principles and management
31 Mose S, Class R, Weber HW, Rahn A, Brady LW, Bottcher HO. Radiation enhancement by gemcitabine-mediated cell cycle modulations. Am J Clin Oncol 2003; 26(1):60–9. 32 Ostruszka LJ, Shewach OS. The role of cell cycle progression in radiosensitization by 2 ,2-difluoro-2- deoxycytidine. Cancer Res 2000; 60(21):6080–8. 33 Swisher SG, Roth JA, Nemunaitis J, et al. Adenovirusmediated p53 gene transfer in advanced non-small-cell lung cancer. J Natl Cancer Inst 1999; 91(9):763–71. 34 Rosier JF, Beauduin M, Bruniaux M, et al. The effect of 2 - 2 difluorodeoxycytidine (dFdC, gemcitabine) on radiation-induced cell lethality in two human head and neck squamous carcinoma cell lines differing in intrinsic radiosensitivity. Int J Radiat BioI 1999; 75(2):245–51 35 Mose S, Karapetian M, Juling-Pohlit L, et al. Radiation enhancement of gemcitabine in two human squamous cell carcinoma cell lines. Anticancer Res 2000; 20(1A):401–5. 36 Price A, Dixon B, Erridge SC, Mohammed N. GriN: A trial and tribulation in respiratory radiotherapy research. Clin Oncol 2005; 17(5):328–31. 37 Ikeda M, Okada S, Tokuuye K, Veno H, Okusaka T. A phase I trial of weekly gemcitabine and concurrent radiotherapy in patients with locally advanced pancreatic cancer. Br J Cancer 2002; 86(10):1551–4. 38 Trodella L, Granone P, Valente S, et al. Phase I trial of weekly gemcitabine and concurrent radiotherapy in patients with inoperable non-small-cell lung cancer. J Clin Oncol 2002; 20(3):804–10. 39 Geng L, Donnelly E, McMahon G, et al. Inhibition of vascular endothelial growth factor receptor signaling leads to reversal of tumor resistance to radiotherapy. Cancer Res 2001; 61(6):2413–9. 40 Nigro ND, Vaitkevicius VK, Considine B Jr. Combined therapy for cancer of the anal canal, a preliminary report. Dis Colon Rectum 1974; 17:351–6. 41 Dean GT, McAleer JJA, Spence RAJ. Malignant anal tumours. Br J Surg 1994; 81:500–8 42 Willet CG. Organ presentation in anal and rectal cancers – current opinion. Oncology 1996; 8:329–33 43 Bartelink H, Roelofsen F, Eschwege F, et al. Concomitant radiotherapy and chemotherapy is superior to radiotherapy alone in the treatment of locally advanced anal cancer: results of a phase III randomized trial of the European Organization for Research and Treatment of Cancer Radiotherapy and Gastrointestinal Cooperative Groups. J Clin Oncol 1997; 15:2040–2049. 44 Flam M, John M and Pajak TF. Role of mitomycin in combination with fluorouracil and radiotherapy, and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: results of a phase III randomized intergroup study. J Clin Oncol 1996; 14:2527–2539. 45 James R, Meadows H, for the NCRI Anal Cancer Trials Management Group. The second UK phase III anal cancer trial of chemoradiation and maintenance therapy (ACT II): Preliminary results on toxicity and outcome. Proc Am Sac Clin Oncol 2003; 22:287 (abstr).
46 Clark MA, Hartley A, Geh LL. Cancer of the anal canal. Lancet Oncol 2004; 5(3):149–157. 47 Stein JP, Lieskovsky G, Cote R, et al. Radical cystectomy in the treatment of invasive bladder cancer: long-term results in 1,054 Patients. J Clin Oncol 2001: 666–675. 48 Rodel C, Grabenbauer GG, Kuhn R, et al. Combined-modality treatment and selective organ preservation in invasive bladder cancer: Long-term results. J Clin Oncol 2002; 20:3061–3071. 49 Zietman AL, Shipley WU, Kaufman DS. The combination of cisplatin based chemotherapy and radiation in the treatment of muscle-invading transitional cell cancer of the bladder. Int J Radiat Oncol Bioi Phys 1993; 27:161–170. 50 Shipley WU, Rose MA, Perronen, et al. Full-dose irradiation for patients with invasive bladder carcinoma: Clinical and histological factors prognostic of improved survival. J Urol 1985; 134:679–683. 51 Shipley WU, Kaufman DS, Zehr E, et al. Selective bladder preservation by combined modality protocol treatment: long-term outcomes of 190 patients with invasive bladder cancer. Urology 2002; 60(1):62–7 52 Sauer R, Birkenhake S, Kuhn R, Wittekind C, Schrott KM, Martus P. Efficacy of radiochemotherapy with platin derivatives compared to radiotherapy alone in organsparing treatment of bladder cancer. Int J Radiat Oncol Bioi Phys 1998; 40:121–7 53 Hussain SA, James ND. Organ preservation strategies in bladder cancer. Expert Rev Anticancer Ther 2002; 2(6):641–51. 54 James ND, Hussain SA. Management of muscle invasive bladder cancer – can we improve the results of bladder preserving therapy? Semin Radiat Oncol 2005; 15(1):19–27. 55 Hussain SA, Moffitt DD, Glaholm JG, Peake D, Wallace DM, James ND. A phase I-II study of synchronous chemoradiotherapy for poor prognosis locally advanced bladder cancer. Ann Oncol 2001; 12:929–35 56 Hussain SA, Stocken D, Peake D, Glaholrn J, Zarkar A, Wallace DMA, James N. Long term results of a phase II study of synchronous chemoradiotherapy in advanced muscle invasive bladder cancer. Br J Cancer 2004; 90(11):2106–11 ◆57 Keys HM, Bundy BN, Stehman FB, et al. Cisplatin, radiation, and adjuvant hysterectomy compared with radiation and adjuvant hysterectomy for bulky stage IB cervical carcinoma [see comments]. N Engl J Med 1999; 340(15):1154–61. ◆58 Rose PG, Bundy BN, Watkins EB, et al Concurrent cisplatinbased radiotherapy and chemotherapy for locally advanced cervical cancer [see comments]. N Engl J Med 340(15): 1999; 1144–53. ◆59 Morris M, Eifel PL, Lu J, et al. Pelvic radiation with concurrent chemotherapy compared with pelvic and para-aortic radiation for high-risk cervical cancer [see comments]. N Engl J Med 1999; 340(15):1137–43. ◆60 Thomas G, Dembo A, Ackerman I, Franssen E, Balogh J, Fyles A, Levin W. A randomised trial of standard versus partially hyperfractionated radiation with or without concurrent 5fluorouracil in locally advanced cervical cancer. Gynecol Oncol 1998; 69(2):137–45.
References 1289
61 Green J, Kirwan J, Tierney J, et al. Concomitant chemotherapy and radiation therapy for cancer of the uterine cervix. Cochrane Database Syst Rev 2005 Jul 20(3):CDO02225. 62 Hao D, Ritter MA, Oliver T, et al. Platinum-based concurrent chemoradiotherapy for tumors of the head and neck and the esophagus. Semin Radiat Oncol 2006; 16(1):10–19. 63 Stell PM, Rawson NSB. Adjuvant chemotherapy in head and neck cancer Br J Cancer 1990; 61:779–787. 64 Munro AJ. An overview of randomized controlled trials of adjuvant chemotherapy in head and neck cancer. Br J Cancer 1995; 71:83–91. 65 EI-Sayed S, Nelson N Adjuvant and adjunctive chemotherapy in the management of squamous cell carcinoma of the head and neck region: A meta-analysis of prospective and randomized trials. J Clin Oncol 1996; 14:838–847. ◆66 Pignon JP, Bourihis J and Domenge C. Chemotherapy added to locoregional treatment for head and neck squamous-cell
67
68
69
70
carcinoma Three meta-analyses of updated individual data. Lancet 2000; 355:949–955. Bourhis J, Amand C, Pignon JP. Update of MACH-NC (metaanalysis of chemotherapy in head & neck cancer) database focused on concomitant chemoradiotherapy. Proc Am Soc Clin Oncol 2004; 23:489s (abstr) Minsky BD, Pajak TF and Ginsberg RJ, et al. !NT 0123 (Radiation Therapy Oncology Group 94–05) phase III trial of combined-modality therapy for esophageal cancer Highdose versus standard-dose radiation therapy, J Clin Oncol 2002; 20:1167–1174. Sartor CI. Mechanisms of disease: Radiosensitization by epidermal growth factor receptor inhibitors. Nat Clin Pract Oncol 2004; 1(2):80–7. Bonner JA, Harari PM, Giralt J. Cetuximab prolongs survival in patients with locoregionally advanced SCCHN A randomized phase III trial. Proc Am Soc Clin Oncol 2004; 22:489s (abstr).
53 Palliative care ANNE NAYSMITH AND KAROL SIKORA
Introduction Planning management Physical symptoms Pain Respiratory symptoms Gastrointestinal problems Cachexia–anorexia syndrome Lymphoedema Sweats Skin problems
1290 1290 1291 1291 1294 1295 1298 1299 1299 1299
INTRODUCTION Palliative care is the active total care of patients and their families by a multi-professional team when the patient’s disease is no longer responsive to curative treatment.1 Many patients with cancer can benefit from the comprehensive palliative care approach, which aims to relieve symptoms and promote physical and psychosocial wellbeing.2 even if cure is the goal of their anti-tumour treatment. Cancer patients at all stages of the illness should have access to palliative care, and referral to specialist palliative care services if appropriate. All clinicians working with cancer patients need the basic skills to deal with the issues described in this chapter.
Emergency situations Rehabilitation Psychological care Spiritual support Social care The dying phase Support for staff Bereavement Key references References
1299 1300 1300 1301 1301 1301 1302 1302 1303 1303
Similar models of care, with structures developed in line with the local health-care system, exist in many other countries. Patients may be referred to a specialist palliative care service at any stage of their illness, if there is a clinical need for specialist advice or care. Access to specialist services should be solely on the basis of clinical need. There should be equity of access for all patients, irrespective of their ethnic or cultural background or the presence of other problems, such as severe psychiatric illness. Referral should not be on the basis of prognosis alone, particularly as it is rarely possible to estimate this with any accuracy.3 In general, specialist palliative care services for children are separate from adult services. Many of the needs of children with a terminal disease are specific to the paediatric setting, and require appropriately trained staff.
Specialist palliative care PLANNING MANAGEMENT Specialist palliative care services in the UK are multiprofessional and involve clinicians with appropriate specialist training and experience. There are four core service structures: hospital support teams, community support teams (which also cover residential and nursing homes), specialist day care and specialist in-patient care. The majority of in-patient beds are provided by voluntary hospices.
When disease is advanced and progressive, there should be a clear overall management plan for the final phases of the illness. Although flexibility is essential, and the plan may need to be developed over a period of time, it should be agreed between the patient and family and the multi-professional team involved in the care. The goals of management
Pain 1291
should be negotiated with the patient, and be focused on quality of life. The management plan, and any later amendments, including what information has been given to the patient and the family, must be shared with all the professionals involved in the patient’s care. This is particularly important when planning the discharge from hospital of terminally ill patients.
Preferred place of death Not all patients can acknowledge that death is inevitable. But many do, at some stage in their illness. The majority would prefer to die at home, if sufficient support is available. Some prefer to die in a hospice, rather than an acute hospital. If the patient has made a choice (or revised an earlier choice), it is important that this is communicated to everyone involved in the patient’s care, so that inappropriate and unnecessary hospital admissions can be avoided.
Communication with patient and family Good communication is an essential part of the care of all cancer patients. For many patients, there is a succession of disappointments and setbacks along their cancer pathway. Whilst it is important that patients are given truthful information, it is also essential that they remain aware that they are important and cared about, and that they will not be abandoned, even after all attempts at anti-cancer treatment have ended. Most people need to feel that they retain some element of control over their illness, and that they are valued partners in treatment, even in very late stages of the illness.
PHYSICAL SYMPTOMS The majority of patients with advanced and progressive cancer have multiple physical symptoms. Pain is often the most feared, although only two thirds of cancer patients experience significant pain. Breathlessness, anorexia and weakness are also common. Although good palliative care involves the whole person, their psychological, spiritual and social needs as well as their physical ones, good symptom control is essential if the patient is to have a reasonable quality of life.
Barriers to pain control The principles of analgesia are well known to most clinicians. Nevertheless, a significant proportion of patients with advanced cancer continue to have poorly controlled pain. Barriers to good pain control are created by the beliefs and behaviours both of clinicians, and of patients and their carers. Clinicians often underestimate the severity of cancer pain. Although effective rating scales have been described .4 they are seldom used.5 Doctors and patients share unrealistic fears about treatment with opioids, particularly morphine, and may be reluctant to start treatment, or to increase it to an effective dose. Morphine is associated with the terminal phase, so that starting morphine seems equivalent to a death sentence.
World Health Organization analgesic ladder Although criticized for its lack of a basis in randomized controlled trials, the World Health Organization (WHO) ‘analgesic ladder’ has stood the test of time and clinical experience, and is the most straightforward way to consider control of chronic pain. The principles are: ● ●
●
●
●
●
●
●
Analgesics are given by mouth wherever possible. Analgesia for chronic pain should be given regularly, at intervals corresponding to the drug’s duration of action. Extra medication should be available for breakthrough pain. Side effects should be actively managed, so that the analgesic can be titrated upwards until the pain is controlled. Simple analgesics are tried initially (the ‘first step’ of the ladder). If inadequate, opioids such as dihydrocodeine are given in low doses (the ‘second step’ of the ladder). Finally, opioids such as morphine are given regularly and titrated upwards to achieve pain control (the ‘third step’ of the ladder). Adjuvant drugs such as non-steroidal anti-inflammatory drugs (NSAIDs), or drugs for neuropathic pain, may be needed at any point for pain that is not fully opioid responsive.
PAIN Good pain control can only be achieved if there is a thorough, preferably multidisciplinary, assessment of the patient’s pain. The assessment should include the patient’s own description of the severity of the pain. It should also document the type, site, radiation and timing of each pain, any precipitating or relieving factors and the response to previous analgesia, both regular and rescue medication. The assessment should be repeated regularly and should form the basis of changes in therapy.
Simple analgesics, NSAIDs and coxibs Mild pain may be adequately treated with paracetamol 1 g qds. Drugs with anti-inflammatory activity, the traditional NSAIDs and the newer COX-2 inhibitors (coxibs), are useful alone for mild pain, and as adjuvant drugs for pain arising from bones, joints or an inflammatory process such as hepatic metastases. There are many NSAIDs available. Most patients find more potent drugs, such as diclofenac,
1292 Palliative care
more helpful than ibuprofen. For oral use, a drug that can be used once or twice daily should be selected, e.g. piroxicam, diclofenac s.r. (slow release). Ketorolac trometamol is a potent NSAID that can be used parenterally, including by continuous subcutaneous infusion. Some NSAIDs are also available in suppository form. They can be helpful to relieve muscular stiffness and the discomfort of immobility in the last hours or days of life.
Low-dose opioids If moderate pain is not controlled by a simple analgesic alone, an opioid is given in low dose, often together with an anti-inflammatory adjuvant drug. Although combinations of paracetamol and an opioid, such as codeine or dihydrocodeine, are often used, this is seldom a logical approach, particularly if the patient is also taking an NSAID. The paracetamol content becomes dose limiting, and the combination has to be given four times a day because of the short plasma half-life of paracetamol. It is usually preferable to use an opioid alone, together with an NSAID if appropriate. Modified-release dihydrocodeine can be helpful. Tramadol, an atypical opioid that is both a μ-opioid receptor agonist and also inhibits noradrenaline (norepinephrine) and serotonin (5-hydroxytryptamine [5HT]) reuptake, can be given orally in doses of 200–400 mg/day. Normal-release and modified-release formulations are available, allowing its use for baseline and rescue analgesia.
Higher-dose opioids Morphine is the gold standard drug in the management of severe cancer pain. It should be given regularly by mouth, titrated upwards as necessary to achieve adequate pain relief. Common side effects, such as nausea and constipation, must be treated effectively to allow effective analgesic doses to be given. The dose requirement should be established using normal-release morphine before changing to a modified-release preparation. Unless patients have previously taken another opioid analgesic, 5–10 mg 4-hourly is a normal starting dose. Oral morphine undergoes considerable first-pass metabolism, so that its systemic availability is 20–30 per cent.6 The main active metababolite is morphine-6-glucuronide (M6G), which is renally excreted. Both morphine and M6G bind to the μ-opioid receptor. M6G accumulates in renal failure and is the main contributor to morphine toxicity in this situation. Morphine should be used extremely cautiously in patients with renal failure. There is clinically no ceiling to the effective dose of morphine. As the dose increases, either pain relief is achieved or side effects become dose limiting. Neither respiratory depression nor dependence is significant in the management of cancer pain. Sedation, confusion and ultimately delirium are the major dose-limiting side effects.
Breakthrough pain It is rare for patients to remain pain free over 24 hours as a result of their baseline analgesia only. The majority of patients with severe pain will experience some breakthrough pain from time to time, and all patients who have chronic pain need to be prescribed appropriate rescue analgesia. Breakthrough pain can be assigned to one of three categories. End-of-dose pain occurs when the next dose of baseline analgesia is almost due. It is an indicator that the dose of baseline analgesia is too low and should be increased. Incident pain is provoked in a predictable way by certain stimuli, which may be physical, such as movement or weight bearing, or psychological, such as anxiety. The pain can be reduced by modifying the behaviour that provokes the pain. Increasing the dose of baseline analgesia may produce only an increase in side effects, such as sedation, when the incident pain is not present. Bone metastases are a frequent cause of incident pain. Adjuvant drugs, radiation therapy, nerve blocks or spinal analgesia may be helpful. True breakthrough pain arises spontaneously and unpredictably. It is typically of rapid onset, may be severe, and often lasts less than 30 minutes. The reported prevalence in hospice patients varies from 40 per cent to 86 per cent.7 The brevity of many episodes of breakthrough pain means that rescue medication needs to have a rapid onset and short duration of action to minimize side effects, and needs to be readily available to the patient, with no delay in administration. Rescue medication should be the normal-release equivalent of the baseline analgesia, and usually the same drug is used. The next regular dose of analgesia should not be delayed. If the baseline analgesic dose is changed, the dose of the rescue medication also needs to be adjusted so that it remains equivalent.
Opioid switching Some patients experience severe adverse effects with morphine. Although in some cases this may be due to dehydration, renal impairment and the accumulation of morphine metabolites, this does not seem an adequate explanation in all cases. Sometimes, reducing the opioid dose relieves the adverse effects without loss of analgesia. Changing to a different strong opioid may reduce the side effects, so that the patient can take an effective analgesic dose. This is known as opioid switching. The explanation is not clear. There seems to be incomplete cross-tolerance between opioid agonists. Also, there are differences between the drugs in their affinity for different opioid receptors. A number of strong opioid drugs are available. Hydromorphone and oxycodone are used orally in ways similar to morphine. Fentanyl is available as a transdermal preparation for baseline analgesia and a buccal lozenge for rescue medication. This may be useful in patients intolerant of oral medication or poorly compliant. Methadone has been claimed to be more useful than other opioids for neuropathic
Pain 1293
pain and other difficult pain problems. However, this is not at present supported by data from clinical trials. The long elimination half-life, leading to accumulation, makes methadone hydrochloride a more hazardous drug to use in routine clinical practice, particularly in the elderly. When switching from one opioid to another, caution should be used in selecting the equivalent dose, because of the likelihood of incomplete cross-tolerance.
Adjuvant therapy It is often not possible to control cancer pain using opioids alone. Incident pain, bone pain, neuropathic pain, gut colic and other non-nociceptive pain syndromes may contain a component of pain that is poorly responsive to opioids by oral or parenteral administration. In this situation, adjuvant drugs are commonly used both to improve pain control and to minimize the side effects of the high doses of opioids that might otherwise be required. There is little randomized controlled trial data for the use of adjuvant drugs in cancer patients. Drug selection is based mainly on evidence from non-malignant pain patients and clinical experience. BONE PAIN
Bone pain is the major cause of incident pain, one of the factors that predict a poorer response to standard opioid analgesia. Anti-inflammatory drugs Despite a lack of evidence from controlled trials in cancer patients, anti-inflammatory drugs are widely believed to be a useful adjuvant to opioid therapy for bone pain. The NSAIDs are usually thought to act peripherally, but there is evidence that they also produce effects centrally. In small studies, they have been shown to be effective for cancer pain, with efficacy comparable to 5–10 mg of intramuscular morphine in single-dose studies but fewer side effects. The published trials are too small to confirm whether or not NSAIDs are particularly useful for bone pain. The side effects of NSAIDs are broadly similar. Gastric erosions or peptic ulceration can lead to perforation or life-threatening haemorrhage. Risk factors for gastrointestinal problems include age 60 years, past history of peptic ulceration and significant ill-health. Most cancer patients are at increased risk of gastrointestinal problems and should be given prophylaxis with either misoprostol 200 mg twice daily or a proton pump inhibitor. The simultaneous administration of NSAIDs and corticosteroids is particularly risky. Neither aspirin nor anticoagulant therapy should be given concomitantly. It is not yet clear that coxibs offer significantly safer treatment for patients with major gastrointestinal risk factors. Both NSAIDs and coxibs have an equal risk of causing renal impairment in chronic use. Elderly patients are at particular risk. In general, anti-inflammatory drugs should
be avoided in patients with pre-existing renal impairment. Altough there are several useful reviews of current drug utilization in a palliative care setting.8 there are very few controlled trials because of the subjective nature of pain and the complexity of assessing its severity.9 Palliative radiotherapy Radiotherapy is the single most effective treatment for a painful bone metastasis.10 The form of irradiation, the radiation dose and the primary tumour type seem not to influence significantly the likelihood of response. Improvement in localized bone pain has been reported after radiotherapy in 70–80 per cent of patients, although pain relief is rarely complete. Pain may recur after a few months, and re-treatment may be indicated. Scattered bone pain can be treated with wide-field irradiation or systemic treatment with a radioisotope. Both are effective. The latter causes fewer side effects. Neither is commonly used. Bisphosphonates for bone pain In patients with bone metastases, treatment with bisphosphonates reduces both pain and pathological fractures, and leads to improved functional status.11 The evidence is best for multiple myeloma and breast cancer. It is less good for other cancers, and there is little evidence to suggest effectiveness in cancer of the prostate. For rapid management of metastatic bone pain, 50–60 per cent of patients will get significant reduction in pain within 14 days after the first dose of a bisphosphonate. A quarter of the remainder may respond after a second course. Patients who do not respond after two courses are unlikely to do so. Both pamidronate and clodronate are effective. Intravenous treatment is better tolerated and more effective than oral. A dose response has been demonstrated for sodium pamidronate. An intravenous infusion of sodium pamidronate 90–120 mg or sodium clodronate 600–1500 mg is therefore recommended. A new oral biphosphonate, zoledronic acid, may supersede these because it is more convenient to administer. NEUROPATHIC PAIN
A variety of neuropathic pain syndromes occurs in patients with advanced cancer. Opioid therapy can be effective. Trials in patients with non-malignant neuropathic pain have shown benefit from tramadol hydrochloride, oxycodone hydrochloride and fentanyl. Methadone, because it is a non-competitive NMDA (N-methyl D-aspartate) receptor antagonist, has also been recommended, but the supporting clinical evidence is lacking. Opioids should be titrated to maximum effectiveness before considering adjuvant medication. The majority of drugs used as adjuvants for neuropathic pain are unlicensed, but many have a supporting evidence base in non-malignant neuropathic pain. Tricyclic antidepressants are the most effective group. Amitriptyline hydrochloride is commonly used, starting with a small dose and titrating up to 75 mg at night. Response may be
1294 Palliative care
seen within 5 days of starting treatment. Comparative trials suggest selective serotonin reuptake inhibitors (SSRIs) are less effective.12 Anticonvulsants have been used for neuropathic pain on the basis of experience with trigeminal neuralgia. There is insufficient evidence to determine whether antidepressants or anticonvulsants are more effective, or whether the response rate would be higher if both were combined. Gabapentin, a recently introduced anticonvulsant, has been licensed for the treatment of neuropathic pain, but trials suggest it is no more effective than amitriptyline hydrochloride. Dizziness is the major side effect, and the drug is better tolerated if upward titration is done slowly to a dose of 900–1800 mg/day, or sometimes more. Systemically administered local anaesthetic-type drugs have been evaluated, including systemic lignocaine and oral mexiletine. Although they are effective in non-malignant pain, the few studies reported so far in cancer-related neuropathic pain have failed to show benefit. Drugs antagonistic at the NMDA receptor might theoretically be expected to be helpful, and ketamine has been shown to be effective in a variety of non-malignant pain states. Ketamine may be used parenterally or orally. Early studies suggest NMDA receptor antagonists may be more useful in com-bination with morphine than used alone. The drugs currently used for neuropathic pain are at best partly effective, and cancer-related neuropathic pain remains a difficult management problem. A number of new agents are being studied, and the situation may improve in the near future. INTERVENTIONAL PAIN TREATMENTS
Improvements in drug treatment have reduced the number of patients for whom invasive therapies are indicated. Nevertheless, for the small proportion of cancer patients with intractable pain syndromes, these therapies should be considered. Local anaesthetic and steroid injections usually have only a transient effect. Neurolytic coeliac plexus block has been reported to have a high success rate, but is rarely performed; serious side effects may occur. The most useful technique is epidural or spinal infusion, either of opioids alone or of a mixture of opioid and other drugs, frequently bupivacaine. Implantable pump systems have been used very successfully, but side effects and complications have been reported in 15–25 per cent of patients.13 These techniques are probably indicated in fewer than 2 per cent of patients with advanced cancer. They will often be offered near the end of life, when the patient is unfit for an implantable pump, and the infusion will be maintained via a tunnelled catheter.
Tenesmus Tenesmus is defined as ‘ineffectual, painful straining’ and may arise from rectum or bladder. Some patients have
spasms of excruciating pain, and may constantly feel that they need to defaecate or urinate. It is associated with a tumour mass, usually from a rectal primary, in the pelvis. Local radiotherapy or endoscopic or laser resection may provide symptomatic relief. Corticosteroids may relieve the pressure caused by the tumour mass. There have been reports of responses to antidepressants and anticonvulsants, as in neuropathic pain, and to calcium channel blocking agents, such as diltiazem and nifedipine. Some patients need intraspinal treatment, either with a local anaesthetic infusion or a neurolytic block. There is little good evidence to guide further treatment, although a small study of lumbar sympathectomy in twelve patients reported pain relief in ten.14
RESPIRATORY SYMPTOMS Patients with advanced cancer may have a variety of respiratory symptoms, including breathlessness, wheeze, cough, haemoptysis and stridor. The rapid onset of new respiratory symptoms may indicate a complication such as bronchial obstruction, superior vena caval obstruction, infection or pulmonary embolism.
Dyspnoea Breathlessness is common in patients with advanced cancer. The prevalence varies with the underlying primary site and with prognosis. Dyspnoea becomes more common as death approaches, and the overall reported incidence rises to 70 per cent in patients in the last 6 weeks of life. Risk factors include primary or secondary lung cancer, pleural metastases and pre-existing respiratory or cardiac disease.15 Dyspnoea is usually multi-factorial16 and, in advanced cancer, respiratory muscle weakness may be a major contributing factor. Anxiety is rarely the sole cause, but may worsen the subjective sensation of dyspnoea and lead to panic attacks when the patient is afraid of choking to death. Subjective improvement may follow treatment of any reversible factors, good nursing care and physiotherapy. Although patients often use oxygen, if they are not hypoxaemic there is little evidence that it helps. The mainstay of management is drug treatment. Low-dose morphine (2–4 mg 4-hourly or as required) is the most effective agent for relieving the sensation of breathlessness. The dose can be titrated upwards as for pain, but lower doses and smaller increments are likely to be needed. Placebo-controlled trials have not shown either slow-release morphine or nebulized morphine to be beneficial. Benzodiazepines do not directly improve dyspnoea, but many patients benefit from relief of anxiety. Lorazepam 0.5–1 mg sublingually can help acute episodes of breathlessness. Regular diazepam (5–10 mg daily) may be needed if chronic anxiety is a symptom.
Gastrointestinal problems 1295
Corticosteroids may be useful in lymphangitis carcinomatosis and in superior vena caval obstruction. If sputum retention is a problem, nebulized saline may help to produce an increased volume of more fluid sputum, or nebulized bronchodilators if there is bronchoconstriction.
temporary relief. Breathing a mixture of helium and oxygen (in a ratio of 4:1) reduces the work of breathing and may be symptomatically helpful.
GASTROINTESTINAL PROBLEMS Cough
Mouth problems
Cough in patients with advanced cancer may be productive or non-productive. Productive cough is usually a problem only if the patient is too weak, or the sputum too thick and viscid, for the cough to be effective. Persistent coughing can disturb sleep (of the family as well as the patient), and can cause chest pain, vomiting, rib fractures and exhaustion. If the productive cough is due to underlying infection, it may be appropriate to give antibiotics even in very ill patients in order to control the very distressing symptoms, but each situation has to be managed individually and in discussion with the patient and family. Nebulized saline can be used to thin sputum, in order to help patients expectorate effectively. However, this should not be done if the patient is completely unable to produce an effective cough, as it can produce very large volumes of sputum, which add to the patient’s distress. Cough suppressants are appropriate for chronic, unproductive cough. Although codeine phosphate is traditionally used, it is not an effective cough suppressant.17 Strong opioids such as morphine do seem to be useful. Methadone hydrochloride at night can be helpful because it has a long half-life. Nebulized local anaesthetics, e.g. lignocaine hydrochloride (up to 5 mL of 2 per cent solution every 6 hours) or bupivacaine hydrochloride (up to 5 mL of 0.25 per cent solution every 8 hours), can relieve intractable unproductive cough. They may cause bronchospasm, and a nebulized bronchodilator should be available. They also reduce the sensitivity of the gag reflex, and patients should not eat or drink for an hour afterwards.
A dry, dirty or painful mouth is common in advanced cancer, and can reduce the patient’s oral intake. If the mouth is not infected, local measures including brushing the teeth and tongue, using a mouthwash, chewing chunks of fresh or tinned pineapple (which contains the proteolytic enzyme ananase) or dissolving effervescent vitamin C (0.5 g) on the tongue can all help. Ulceration may be due to mucositis from recent antitumour treatment, aphthous ulceration or infection. Regular use of a benzydamine mouthwash or benzocaine lozenges may make the mouth less painful, but benzocaine must be avoided before eating. Severe pain may require opioid analgesia. Bacterial infection can be improved with regular mouthwashes of chlorhexidine gluconate or tetracycline. Herpes simplex may be difficult to diagnose, and requires specific antiviral treatment. Aphthous ulceration may respond to a short course of thalidomide. ORAL CANDIDIASIS
This is common in patients with advanced cancer. It may present as the typical white plaques, erythema or angular cheilitis. Patients find nystatin suspension difficult to comply with; it is only effective if used four times daily for more than 5 days. Ketoconazole 200 mg once daily for 5 days is easier for patients and more cost-effective, but interacts with a number of commonly prescribed drugs. Fluconazole 150 mg orally as a single dose.18 is equally effective though more expensive. Dentures must be cleaned and sterilized to prevent re-infection.
Haemoptysis DRY MOUTH
Minor haemoptysis is a common symptom of primary, and sometimes secondary, tumours of the lung. If radiotherapy of the tumour is not indicated, an oral haemostatic drug such as tranexamic acid or etamsylate is often helpful.
Stridor Stridor results from obstruction of the larynx or major airways. It is frightening both for the patient and for his family. The patient’s position should be adjusted to minimize the obstruction. If relief of the obstruction with radiotherapy or stenting is not possible, high-dose corticosteroid therapy, such as dexamethasone 16 mg daily, can give
Dry mouth (xerostomia) is a common and distressing symptom. It has been reported in 30 per cent of patients receiving palliative care and 77 per cent of patients admitted to a hospice. It is often assumed to be a sign of dehydration, but the correlation between xerostomia and the patient’s state of hydration is very poor. It is much more often due to drug treatment or to the effects of previous surgery or radiotherapy. Drugs commonly implicated include analgesics (e.g., morphine), antidepressants (e.g., amitriptyline) and hypnotics (e.g., temazepam). Most patients find frequent sips of water or semi-frozen drinks as effective as any other treatment. Artificial saliva sprays are widely prescribed. Mucin-based artificial salivas are more effective than carboxymethylcellulose-based ones,
1296 Palliative care
but the effect lasts only minutes. Offered artificial saliva or a low-tack, sugar-free chewing gum, patients found the gum acceptable and more effective than artificial saliva.19 In a comparative study, pilocarpine 5 mg tds was more effective than artificial saliva.20 but produced a higher incidence of side effects. The side effects of pilocarpine are dose related, and patients with drug-induced xerostomia may respond to a lower dose. After a dose of pilocarpine, saliva flow peaks at 1 hour and remains increased for up to 4 hours. DROOLING
Unlike xerostomia, drooling is fairly uncommon but can occur either following head and neck tumours or as a result of primary or secondary brain tumours. A normal amount of saliva is produced, but inability to manage the saliva interferes with eating, and can lead to irritation of the lips and chin, angular cheilitis or bacterial infection. The saliva may also be aspirated, leading to choking. Treatment aims to reduce the amount of saliva. Inhibition of salivation can be produced with very low doses of anticholinergic drugs. Hyoscine hydrobromide is given either sublingually or transdermally (1 mg/72 hours). Glycopyrronium does not cause central side effects. It can be administered orally or via percutaneous gastrostomy tubes and is absorbed sufficiently to reduce saliva production for up to 6 hours. Low dose amitriptyline hydrochloride (10–25 mg at night) is often effective.
generates the stimulus to the act of vomiting. Input to the vomiting centre comes from: ●
● ●
●
It is important to identify the likely underlying mechanism(s) and the route through which they are provoking vomiting. An anti-emetic with the appropriate site of action can then be prescribed. Any treatable underlying cause, such as raised intracerebral pressure or hypercalcaemia, should be treated effectively if possible. Nausea has to be assessed separately. A patient who remains nauseated, despite control of vomiting, will still be extremely symptomatic and distressed. Common causes of nausea and vomiting in advanced cancer include: ●
● ● ●
Dysphagia
●
Dysphagia may arise as a result of intrinsic or extrinsic tumour compression. If stenting or other intervention is not possible in very ill patients, corticosteroids may temporarily reduce the volume of the tumour and relieve the dysphagia slightly. Some patients experience severe pain on swallowing, not always associated with dysphagia. The pain may be due to oesophageal spasm, infection or benign ulceration in the oesophagus. Endoscopy is useful for accurate diagnosis. Infection is most commonly candidiasis, and responds to an imidazole or triazole antifungal. Ulceration may be due to reflux or corticosteroid therapy, best treated with a proton pump inhibitor. Nifedipine modified release (10–20 mg twice daily) has been used as a smooth muscle relaxant to relieve oesophageal spasm. Hypotension is the dose-limiting side effect.
●
Nausea and vomiting Nausea and vomiting are common in advanced cancer, affecting 40–70 per cent of patients. Good initial assessment and frequent reassessment are essential. Vomiting is mediated via the vomiting centre in the brain, which receives and integrates input from various sources and
the chemoreceptor trigger zone, stimulated by drugs or chemicals in plasma such as opioids, urea or calcium the vagus and vestibular apparatus mechanoreceptors and chemoreceptors in gut, liver and other organs, mainly via 5-HT3 and dopamine receptors higher centres, which are involved both in conditioned vomiting and in stress-induced delay in gastric emptying.
disease or obstruction within the gastrointestinal tract, including delayed gastric emptying due to opioids, hepatomegaly or the anorexia-cachexia syndrome drugs, e.g., opioids, NSAIDs, digoxin, metronidazole metabolic causes, e.g., hypercalcaemia, renal failure anti-tumour therapy, e.g., chemotherapy, radiotherapy to the abdomen raised intracerebral pressure, or brain metastases affecting the vestibular system pharyngeal irritation due to chronic cough or sputum.
There is a wide range of anti-emetic drugs. They can be grouped according to their main site or mechanism of action. Recent National Institute for Clinical Excellence (NICE) guidance is helpful21 and offers guidelines for anti-emetic selection. Good first-line drugs: ●
●
●
are prokinetic (gastric stasis, opioid-induced ileus) – metoclopramide hydrochloride 10 mg tds/qds orally or 30–100 mg/24 hours by subcutaneous infusion act on the chemoreceptor trigger zone (opioids, hypercalcaemia, renal failure) – haloperidol 1.5–5 mg nocte orally or 5–10 mg/24 hours by subcutaneous infusion act on the vomiting centre (mechanical bowel obstruction, raised intracranial pressure, pharyngeal irritation, movement-induced nausea) – cyclizine 50 mg tds orally or 150 mg/24 hours by subcutaneous infusion.
Since prokinetic drugs act through cholinergic receptors, they should not be prescribed at the same time as anticholinergic drugs such as cyclizine or hyoscine. The route of administration is important. Nauseated patients may have very poor absorption via the oral route, and may
Gastrointestinal problems 1297
require rectal or parenteral administration until the nausea and vomiting are controlled. The patient should be regularly reassessed. If the underlying mechanism seems to have been wrongly identified, the anti-emetic should be changed. If the patient has more than one underlying cause for vomiting (30 per cent of patients) a second anti-emetic, such as haloperidol added to metoclopramide hydrochloride, may be needed. If first-line anti-emetics are not effective, adding dexamethasone may be helpful. Alternatively, therapy can be changed to levomepromazine (methotrimeprazine), a very broad-spectrum anti-emetic. It has a long half-life and can usually be given as a single oral dose of 6.25–25 mg at night or as a subcutaneous infusion. In intractable nausea or vomiting, 5-HT3 antagonists, e.g., granisetron, may be tried, but in view of their cost should be rapidly discontinued if not effective. Young patients with chemotherapyinduced vomiting often find cannabinoids helpful; large studies in advanced cancer have not been carried out.
Constipation Decreased oral intake, increasing immobility and drug therapy, especially with opioids and anticholinergic drugs, are the major contributory factors to constipation in cancer patients. But confused or immobile patients also have difficulty reaching a toilet or commode, and may be embarrassed to ask for help. In severe constipation, the clinical findings may mimic intestinal obstruction. Plain X-ray of the abdomen is diagnostic if there is any doubt about the diagnosis of constipation. Laxatives can be classified as either predominantly softening or predominantly peristalsis stimulating. Most patients require both a softener, such as lactulose or docusate sodium, and a stimulant, such as senna or bisacodyl. Co-danthramer and co-danthrusate suspensions and capsules are convenient combinations of a softener and the stimulant, dantron, and are licensed for the terminally ill. Hydrophilic bulk-forming agents should be avoided. Patients can rarely drink the volume of fluid required with these agents, and they may precipitate obstruction. If possible, the patient should be encouraged to increase their fibre and fluid intake; some sipfeeds are enriched with fibre. True diarrhoea is relatively rare in patients with advanced cancer, and the commonest cause is over-use of laxatives, although infective and antibiotic-induced diarrhoea also occur. Once overflow diarrhoea and infection have been ruled out, if an anti-diarrhoeal agent is required loperamide is safe and effective.
Intestinal obstruction Patients presenting with signs of intestinal obstruction must be carefully assessed to distinguish malignant obstruction from drug-induced ileus, faecal impaction or obstruction
secondary to benign pathology such as adhesions. Few patients with advanced cancer and malignant obstruction have a localized site of blockage and are fit enough to be considered for surgery. Mortality of palliative surgery is high (12–30 per cent), average survival is short and there is a high incidence of postoperative complications such as fistula formation. There is limited evidence that malignant obstruction, particularly if it is high in the gastrointestinal tract, may be more likely to resolve after corticosteroids. Dexamethasone has been given parenterally in doses of 6–16 mg daily. There is no evidence on which to select a suitable dose.22 Most patients are best managed medically. Oral drugs are poorly absorbed, and treatment should be given parenterally, preferably by continuous subcutaneous infusion. The major symptoms are continuous abdominal pain, colic and vomiting. Patients usually require diamorphine for analgesia, and either haloperidol or cyclizine, or both, for control of nausea. Hyoscine butylbromide (60 mg/24 hours), titrated upwards as required, is used to control colic and reduce the volume of intestinal secretions. If the patient continues to have large volume vomits, octreotide (300–600 mg/24 hours) can be substituted for the hyoscine. It is effective in reducing the volume of fluid in the obstructed bowel and also reduces peristalsis. Dry mouth and thirst may be major problems. Patients whose nausea is controlled can drink at will, and regular mouth care is helpful. A few patients, not in the terminal stage of their disease but with intractable obstruction, may need parenteral hydration for symptom control. Patients rarely find a nasogastric tube helpful, unless their large volume vomiting cannot be controlled by any other means; if drainage is needed for more than a few days, the possibility of a venting gastrostomy should be considered. If drugs and fluids are both given by subcutaneous infusion, it is often possible to keep the patient at home despite the obstruction.
Ascites Ascites is commonest in tumours of the breast, ovary or gastrointestinal tract. Progressive abdominal distension produces a tight discomfort. In addition, there is early satiety, and splinting of the diaphragm can cause respiratory embarrassment. Although the ascites can be drained percutaneously, it usually re-accumulates relatively quickly; repeated drainage leads to a progressive fall in the serum albumin. Diuretic therapy with spironolactone (100–400 mg/day), usually combined with a small dose of furosemide, has been used to slow down the re-accumulation of fluid. It is unlikely to be effective if the ascites is due solely to peritoneal tumour deposits. Peritoneo-venous shunts can be inserted to drain the ascites continuously back into a central vein. This prevents the fall in serum albumin seen with external drainage. Malignant effusions usually contain tumour cells and have a high protein content, so that blockage of the shunt is common. Shunt insertion is associated with significant
1298 Palliative care
morbidity and should only be considered in patients who are still quite fit.
and an increased tendency to thromboembolism are the main side effects. The onset of benefit is slower than after corticosteroids.
CACHEXIA–ANOREXIA SYNDROME Prokinetic agents This poorly understood symptom complex is common in patients with advanced cancer. Eighty per cent of patients have developed it by the time of death, although the disease burden is not correlated with the degree of cachexia. Cachexia is used to describe the wasting of body tissues, both fat and lean mass. Whereas in normal starvation fat is lost preferentially and there is relative preservation of muscle tissue, in cancer cachexia there is equal loss of both tissue types and the overall wasting is out of proportion to the degree of anorexia. Parenteral nutrition or supplemental oral feeding does not reverse the weight loss. Cancer cachexia is due to major metabolic abnormalities, caused mainly by cytokines and other products released by the tumour. In most patients, these tumour products are the main cause of the anorexia. Cachectic patients have poorer survival, and higher complication rates after anti-tumour treatment. Cachexia aggravates the fatigue and chronic weakness seen in advanced cancer. It often causes severe psychological distress both to patients and their families. Because people often assume the patient is starving to death, families may put great pressure on the patient to eat, leading to anxiety and conflict. Weight loss occurs earlier and is more severe in patients with dysphagia, pain on swallowing, abnormalities of taste or chronic nausea. Management therefore involves controlling nausea and other swallowing problems, as well as providing the patient with small quantities of appetizing food and removing any pressure, from staff or families, to eat more than can be managed. Drug treatment has limited effectiveness.
Corticosteroids Corticosteroids improve appetite in many cancer patients. They also have a beneficial effect on well-being. Improvement is rapid, so that a short trial of 5–7 days is sufficient to establish if the drug is useful. The appetite stimulation is not translated into weight gain in most patients, and is often short-lived (3–6 weeks). There are no trial data to indicate the most useful dose. Dexamethasone is usually used in doses of 4–8 mg daily.
Progestational agents Both megestrol acetate and medroxyprogesterone acetate have been shown to stimulate appetite, food intake, energy level and weight gain in controlled trials. The weight gain is mainly fat. The optimal dose of megestrol acetate is reported to be 800 mg/day, although effects are seen with doses as low as 160 mg. Medroxyprogesterone acetate should be given in a dose of 500 mg twice daily. Oedema
Chronic nausea and early satiety are common components of the anorexia-cachexia syndrome. The regular administration of a prokinetic anti-emetic, metoclopramide or domperidone, is helpful in reducing both symptoms.
Other drugs Although the anorexia-cachexia syndrome is believed to be produced by the release of cytokines, trials of antagonists to various cytokines have so far been disappointing. However, positive results have been reported from trials of thalidomide in patients with human immunodeficiency virus (HIV) disease, and studies of cannabinoids are also in progress.
Chronic anaemia Chronic anaemia in advanced cancer often coexists with weakness and fatigue, but is not always the cause. Chronic blood loss from lung or gastrointestinal tract may be reduced by regular tranexamic acid 1g three times daily. Etamsylate 500 mg four times daily is safe when bleeding is from the kidney or bladder. The benefit of transfusion should be regularly reassessed, and transfusion only continued if the patient is symptomatically improved. In other countries, patients have been maintained on erythropoietin, but the high cost limits use of this drug in the UK.
Fatigue Fatigue is associated with more advanced cancer, radiotherapy and chemotherapy. No specific treatment is effective. Corticosteroids are often prescribed, and some patients seem to find them helpful, but there is no supporting trial data. Patients should be advised about adjusting their lifestyle and pacing their activity. Any associated depression should be treated. Magnesium deficiency is probably commoner in advanced cancer than is generally realized.23 It may present as generalized weakness, possibly with hypokalaemia and a variety of other symptoms and signs. It may follow platinum chemotherapy. The serum magnesium is maintained within the normal range until there is severe whole-body depletion, so that a normal serum magnesium does not rule out the diagnosis. Intravenous repletion followed by oral maintenance treatment reverses the symptoms.
Emergency situations 1299
LYMPHOEDEMA Lymphoedema may follow anti-cancer treatment or disease recurrence. It is worsened by infection. The skin in a lymphoedematous limb should be cared for meticulously and kept moisturized to reduce the risk of skin breaks and cellulitis. Any infection that develops should be treated promptly. Light superficial massage and compression bandaging usually improve the swelling and reduce any leakage, although the benefits are only apparent after several weeks of treatment. A lymphoedematous limb may feel heavy and stiff, but lymphoedema itself is not painful. Nor does it cause nerve damage. Pain or neurological signs indicate that there is another underlying pathology, often progressive tumour.
SWEATS The prevalence of abnormal sweating in advanced cancer has been reported to be 16–28 per cent. Some patients have sweating of hormonal origin because of a menopause (natural or treatment-induced) in women, or anti-androgen therapy in men with prostate cancer. Low-dose megestrol acetate (40 mg/day) has been shown to be effective and safe in both groups; benefit is seen after 2–3 weeks. More commonly, sweating is assumed to represent a paraneoplastic phenomenon. Simple measures such as cool surroundings and fresh air or a fan may make the sweating tolerable. Unless there is associated fever, anti-pyretics are not indicated. The main symptomatic agents used are not supported by a large volume of evidence.24 and thioridazine, the drug with the highest reported rate of success, has been withdrawn from general clinical use. Hyoscine has been used successfully in opioid-induced sweating, and the use of cimetidine 400–800 mg daily has also been reported. In one case report, thalidomide 200 mg at night was used successfully; the side effects of teratogenicity (though rarely a problem in palliative care patients) and peripheral neuropathy have to be remembered.
antihistamines rarely relieve itch in uraemia, cholestasis or lymphoma, although their sedating effects may reduce associated anxiety and promote sleep. Placebo-controlled trials in cholestasis, and case reports in uraemia and opioid-induced pruritus, suggest that 5-HT3 antagonists such as granisetron in normal anti-emetic doses can be effective. Small studies suggest that paroxetene, opioid antagonists and thalidomide may also be worth trying; there is a risk of opioid withdrawal effects with opioid antagonists, especially naloxone.
Fungating wounds Fungating wounds cause psychological distress. Odour is the most distressing symptom; the wounds can produce a profuse exudate, which requires repeated dressings, may bleed and may be painful and itchy. The pain may be poorly responsive to opioids. Bacterial infection should be treated with a topical preparation such as metronidazole gel to reduce the odour. If the surrounding skin is inflamed and itchy, corticosteroid cream may be helpful. Bleeding can be controlled by a topical application of sucralfate (the suspension is placed on a non-adherent dressing) or of tranexamic acid (using the parenteral preparation). Very absorbent dressings, such as alginate, reduce both the bulk and the frequency of dressings. Pain during dressings may be reduced by using diluted bupivacaine rather than saline to wash the wound, or teaching the patient to breathe a mixture of oxygen and nitrous oxide (Entonox). A top layer of a carbonimpregnated dressing, to absorb any remaining odour, may then make the situation more tolerable and manageable both for the patient and for the family.
EMERGENCY SITUATIONS Not all acute problems are reversible in patients with advanced cancer. The adverse effects of treatment may outweigh a benefit that may be short-lived. If the patient is too ill to make his wishes clear, the situation should be discussed with those close to him so that agreement can be reached on the best course of action.
SKIN PROBLEMS Pruritus
Severe infections
About 5 per cent of palliative care patients have troublesome itch. The commonest causes are cholestasis and renal failure. Itch is also common in lymphomas and haematological malignancies. If possible, the underlying cause should be relieved, e.g., by stenting of the common bile duct. Otherwise, treatment is symptomatic. The itch is usually exacerbated by heat and by dry skin; cool temperatures, light cotton clothing and sheets, and emulsifying ointment or aqueous cream instead of soap should be advised. There is little evidence to support the topical treatments commonly used. Of the systemic treatments available,
Infection may follow treatment, e.g., after chemotherapy, or arise as a consequence of the underlying illness. It is often difficult in the emergency situation to be clear whether the patient or his family would want aggressive treatment of the infection. Sometimes, when chemotherapy has been entirely palliative, the oncologist’s judgement may be that it is better not to treat aggressively, but this can be psychologically very difficult to accept when active antitumour treatment is still being given. If possible, it is helpful to discuss with the patient in advance what his wishes would be in this situation. Some patients write Advance
1300 Palliative care
Directives, and their existence should be communicated to all the clinicians caring for the patient.
Hypercalcaemia Treatment should be considered if the corrected serum calcium is 3 mmol/L, the patient is symptomatic or the serum calcium is rising very rapidly. Patients may present with confusion and vomiting, and are also likely to have thirst, polyuria, dehydration and constipation. Intravenous bisphosphonates are the treatment of choice. After initial intravenous rehydration, a biphosphonate such as intravenous disodium pamidronate 90 mg or sodium clodronate 1500 mg is infused over 2–4 hours, followed if necessary by further intravenous fluids. The serum calcium begins to fall 3–5 days after treatment, and 80 per cent of patients will have a normal serum calcium within a week. Benefit is often transient. Oral clodronate may delay recurrence of hypercalcaemia. Further intravenous bisphosphonates can be given regularly or as required, depending on the speed with which the serum calcium rises again. Alternative treatments are rarely required. Calcitonin acts very quickly, but tolerance usually develops within days or a very few weeks. Its only role is probably in the patient with life-threatening hypercalcaemia, who cannot wait 3 days for bisphosphonates to be effective. Gallium nitrate has been shown to be as effective as bisphosphonates, but is less well tolerated and rarely used. Corticosteroids are probably only effective when they have an anti-tumour effect, e.g., in lymphomas.
Superior vena caval obstruction Most cases of superior vena caval obstruction (SVCO) are due to tumour within the mediastinum, usually from a primary cancer of the bronchus. If the obstruction cannot be relieved, patients need symptomatic relief of breathlessness. High-dose corticosteroids are often helpful. Lowdose oral morphine can reduce the dyspnoea. Patients may also need a benzodiazepine for anxiety.
resuscitation measures can be considered, but may not always be appropriate, particularly if further bleeding is likely and cannot be prevented. If a major haemorrhage can be anticipated, it may be helpful to warn the patient’s family, especially if he is at home. However, it can be difficult to do this without causing panic. In a hospital or hospice situation, patients at risk of major haemorrhage should have on their drug charts a single large dose of sedative, which can be injected if there is a major haemorrhage, to reduce the patient’s distress and fear.
REHABILITATION Patients with advanced cancer are often perceived to be progressing along a trajectory of deterioration, which ends inevitably in death. This is an over-simplification. Despite progressive cancer, many patients can be helped to maximize their functional ability and independence by individualized programmes of rehabilitation, often involving physiotherapy, occupational therapy and psychological support. Rehabilitation is particularly important after events such as a pathological fracture or spinal cord compression, but often appropriate after radiotherapy or chemotherapy if there have been major side effects. Good links between oncologists and palliative care or other rehabilitation facilities help to ensure patients achieve the best possible level of independence.
PSYCHOLOGICAL CARE Cancer patients and their families need good psychological care throughout their illness. As the disease progresses, patients may have to deal with repeated sharing of information, much of it bad news, and may have to deal with a great deal of uncertainty, particularly regarding prognosis. Psychiatric disorders, particularly adjustment reactions, are common, and up to 80 per cent of this morbidity is unrecognized and untreated.25 Clinical psychiatric disease, particularly depression, is also common, but its prevalence is difficult to determine from published studies, as it varies markedly according to the methods and criteria used for assessment.
Massive haemorrhage Depression Those most at risk of massive haemorrhage include patients with carcinomas of the bronchus and haemoptysis, patients with a tumour that is eroding a major artery (usually in the neck or groin) and patients with a primary tumour of the upper gastrointestinal tract. It is rarely possible to stop a torrential haemorrhage, although some patients, as the massive blood loss results in hypotension, will stop bleeding spontaneously. If bleeding is from the lung, the patient should be positioned on his side with the diseased lung downwards, to reduce the chance that he will drown in his own blood. If the bleeding can be stopped,
Depression is often unrecognized in an out-patient population. Among patients admitted to specialist palliative care units, estimates of prevalence range from 25 to 37 per cent. As somatic symptoms may be due to the physical illness, the diagnosis has to be established solely on psychological symptoms, including suicidal ideation, which should always be asked about. Tricyclic antidepressants and SSRIs can both be used selectively in patients with advanced cancer, provided they are likely to live for the 2–4 weeks needed to see benefit.
The dying phase 1301
Tricyclic drugs tend to be used where patients are having difficulty sleeping. A long-acting SSRI such as fluoxetene may be preferable to a short-acting one, to avoid a withdrawal syndrome in the dying phase when the patient can no longer swallow. Response rates are difficult to assess, but have been reported to be lower than in patients without physical illness, and psychiatric help may be needed. If the patient’s prognosis is very short, a psychostimulant such as methylphenidate may be considered. Severe anxiety is often associated with depression. But if it is diagnosed independently, or does not respond to a sedating antidepressant, benzodiazepines are safe and effective.
Organic confusional states An organic confusional state (delirium) may develop at any time, but becomes more prevalent as patients approach death. Common causes include brain tumours, infection, metabolic disturbances (including renal failure) and drugs, particularly opioids and psychotropic drugs. If possible, an underlying cause should be treated. No active management may be needed if the patient is not distressed by the confusion. But many patients become agitated, and require sedation. Delirium may be accompanied by hallucinations and/or delusions. A neuroleptic such as haloperidol or, if more sedation is needed, levomepromazine, can be given orally or by subcutaneous infusion. If only sedation is required, a subcutaneous infusion of midazolam can be titrated to give the patient comfort, and can be rapidly withdrawn if the underlying cause resolves.
SPIRITUAL SUPPORT Being brought face to face with impending death often makes people want to explore existential and spiritual issues. These need not necessarily involve any formal religion, or even a faith in any God. People may wonder why they have been singled out for such an unfair punishment, or question the value of their life. They may feel guilty for deeds in the past, particularly if quarrels or problems have not been resolved. They may even feel their cancer is a punishment for living an evil life. Those who have a religious faith may find it a source of comfort, but may conversely question the existence of a God who allows these things to happen. They may have tremendous fear and doubt about what is on the other side of death. Sometimes, well-meaning friends or fellow church members express the belief that the cancer will be cured if the sufferer has enough faith and prays enough. When the cancer is not cured, the patients are often left feeling guilty that they have not had sufficient faith, an unnecessary additional burden. We live in a society in which people come from many cultural backgrounds and follow a wide variety of religions.
As part of caring for the whole person, and their family, it is essential to respect religious beliefs and practices, even if that means adapting the normal routine of a hospital ward. In some religions, illness and death are seen as determined by fate, rather than the patient’s deeds. This can be helpful in absolving the patient from any sense of personal guilt. Spiritual issues are often left for a chaplain or trained counsellor to explore. Clinicians often feel out of their depth in such conversations. This is not necessarily helpful for the patient, who may prefer to discuss things with someone he knows well rather than with an unfamiliar professional, particularly if he does not have an identifiable religious faith. It is important for everyone to remain open to such conversations.
SOCIAL CARE People rarely exist in isolation. They often have close family, and usually a social network. These carers also need support, and should be involved in the patient’s care if that is what they and the patient want. Families cope better with death if they have been kept well informed as the patient’s condition deteriorates, and can contribute to the discussions around management. This is particularly important if the patient wants to spend his last period of life at home, as he will be dependent on family for much of the care, no matter how good a package of community care is organized.
THE DYING PHASE Death from cancer is rarely sudden, unless it is precipitated by an event such as a pulmonary embolus. In the majority of patients there is a dying phase, which may last for hours or days. It is characterized by increasing sleepiness (and often finally loss of consciousness), confusion, weakness, difficulty in coughing and inability to swallow. It is essential that control of symptoms such as pain and vomiting is maintained, even when the patient is unable to swallow his usual medication. But other medication, particularly that for long-term conditions such as hypertension, should be discontinued. Cancer sufferers come from a variety of cultural and religious backgrounds. While it is helpful to be familiar with the basic customs of the major religions around death, there is considerable individual variation in how, or whether, these are observed. Staff should always ask, as sensitively as possible, either the patient or a relative what they would like done during the dying period and also immediately after death. If a relative wants to take part in laying out the body, or the family needs to take over this function, this should be encouraged.
Assessment Careful, preferably multi-professional assessment is essential, and should be regularly repeated. The situation may
1302 Palliative care
change very rapidly. The focus should be on the patient’s problems. However, by this stage many patients are too weak to take part in discussions. Relatives often speak for them. There is evidence that both professionals and relatives may have views that differ from those of the patient.26 Relatives in particular may transfer their own distress to the patient, and describe his pain as worse than it actually is.
Maintaining symptom control by non-oral routes Dying patients commonly lose the ability to swallow medication. All non-essential medication should be stopped. This will include replacement hormones, and often corticosteroids too. Some drugs can be given by the transdermal or rectal route, or by subcutaneous injection. Repeated injections should be avoided in cachectic patients. Many patients need a parenteral analgesic, such as diamorphine, together with additional drugs, to maintain symptom control over the dying phase. The simplest way to administer these is as a continuous subcutaneous infusion, using a small, batterydriven infusion pump. Pharmacists can advise on compatibilities when mixing drugs in one syringe.
Terminal agitation Dying patients may experience overwhelming distress, or fear of impending death. This may be associated with their severe weakness, and a sense of being helpless and out of control. Sometimes it is related to unresolved conflict or to guilt over real or imagined past sins, or reflects their sadness at leaving behind a loved and dependent family. As confusion is common at this stage of life, agitated patients may not be able to express their anguish verbally, or be comforted or reassured. The subcutaneous route can be used for sedation, either with a benzodiazepine such as midazolam (10 mg/24 hours titrated upwards as required) or, if an anti-emetic effect is also needed, with the phenothiazine, levomepromazine (12.5–25 mg/24 hours titrated upwards as required). Either can be mixed with diamorphine for infusion.
Death rattle Noisy breathing, with a rattling sound in the throat or chest, may develop in the last hours or days. It mainly reflects the patient’s inability to cough effectively. Fluid overload, particularly in hypoalbuminaemic patients, or bronchopneumonia may worsen it. Either glycopyrronium bromide (0.6–1.2 mg/24 hours) or hyoscine given subcutaneously, by injection or infusion, can reduce the secretions. Hyoscine is given as the butylbromide in a dose of 20–40 mg/24 hours, or as the hydrobromide in a dose of 1.2–2.4 mg/24 hours.
Fluid needs Dying patients are rarely able or willing to drink. Staff and relatives often worry that the patient’s death will be accelerated, or made more uncomfortable, by dehydration. The expected symptoms of dehydration are a dry mouth and thirst. Research does not show any correlation between a dry mouth and biochemical indicators of dehydration. Instead, a dry mouth is associated with treatment with particular drugs, such as opioids or tricyclic antidepressants. Water requirements are decreased at the end of life, and there is a risk that parenteral rehydration may cause increased pulmonary oedema, particularly in patients with hypoalbuminaemia. It may be very difficult to convince relatives that the patient is not dying as a result of lack of food and water. A great deal of explanation and reassurance is often needed.
SUPPORT FOR STAFF Caring for cancer patients who are terminally ill can be very stressful for professional staff. They may see their role as curative, and feel guilty about failing to cure particular patients, especially if the primary tumour was one with a high cure rate. Caring for a dying patient may also cause staff to consider their own future death, and this may lead to considerable anxiety. Staff are often particularly distressed if the dying patient is young, or has a very young family, or if in some other way they identify with the particular patient. It is essential not to deny the reality of these feelings. They can sometimes lead professionals to avoid patients who are dying, or to deny that the patient’s prognosis is short. Inappropriate management decisions may then be made. If staff are to continue to work well with all cancer patients, and not merely those in the early stages of the disease, they need support in dealing with their feelings of grief, anger, guilt and bereavement when patients are dying, and with their feelings of personal vulnerability.
BEREAVEMENT Because the care of the terminally ill patient includes the care of his family, involvement should not end with the death of the patient. It is important to ensure that those who have been bereaved are grieving normally, and that help is available for those in whom there is a high risk of an abnormal or exceptionally severe grief reaction, such as parents who have lost a child, and those with a previous history of alcohol misuse or depressive illness. In addition to feelings of numbness, disbelief and loss, many bereaved people feel strong anger. This may be displaced on to staff who were involved in the patient’s care. This is particularly difficult to deal with if there is some basis for it, such as a delay in diagnosis. Grieving may also involve inappropriate feelings of guilt, whether for imagined
References 1303
failings in caring for the dead person or for estrangements which were not healed before death. It is helpful if it is made easier for the family to have an interview with a member of staff who knew the patient and the circumstances of the death. If the anger can be dealt with in an understanding way, rather than by confrontation, and honest explanations given for anything that is worrying the family, these difficult reactions will usually come to a natural end. Families should be given information about community resources that offer support in bereavement, including the primary care team and voluntary organizations.27 Specialist help may be needed for children who have lost a parent, who need counselling appropriate to their age group.
KEY LEARNING POINTS ●
●
●
●
●
●
The majority of cancer patients will need palliative care, which should run alongside anticancer therapy. Palliative care is multi-professional and involves caring for the family as well as the patient. Good communication, both with the patient and among the clinical team, is essential. The management plan should reflect the goals of the patient and family, and should be shared with all the professionals involved. Good palliative care demands meticulous symptom control and requires a thorough knowledge of the pharmacology of analgesics, anti-emetics and drugs for the terminal phase. The patient’s autonomy must be respected. This involves the right to information and to make choices, including the choice of where to die.
KEY REFERENCES Doyle D, Hanks GWC, Chernon N, Calnman K (eds) Oxford textbook of palliative medicine, 3rd edn. Oxford: Oxford University Press, 2005. Fallon M, O’Neill, B. ABC of Palliative Care. London: Blackwell/ BMJ Books, 2003. Faull C, Carter Y, Daniels L (eds) Handbook of Palliative Care. Oxford: Blackwells Publishing, 2005. National Institute for Clinical Excellence. Guidance on supportive and palliative care for adults with cancer, 2004. www.nice.org.uk The goldstandards framework programme, 2006. www.goldstandardsframework.nhs.uk Twycross R, Wilcocks A (eds) Symptom management in advanced cancer. Oxford: Radcliffe Press, 2001.
REFERENCES 1 World Health Organization. Cancer pain relief and palliative care. Report of a WHO Expert Committee. WHO Tech Rep Series 1990; 804:1–75. 2 National Council for Hospice and Specialist Palliative Care Services Specialist palliative care: a statement of definitions, Occasional Paper 8. London: National Council for Hospice and Specialist Palliative Care Services, 1995. 3 Viganò A, Dorgan M, Buckingham J, et al. Survival prediction in terminal cancer patients: a systematic review of the medical literature. Palliat Med 2000; 14:363–74. 4 Bruera E, Pereira J. Recent developments in palliative cancer care. Acta Oncol 1998; 37:749–57. 5 Pargeon KL, Hailey BJ. Barriers to effective pain management: a review of the literature. J Pain Sympt Manage 1999; 18: 358–68. 6 Hanks G, Hawkins C. Agreeing a gold standard in the management of cancer pain: the role of opioids. In: Hillier R, Finlay I, Welsh J, Miles A. (eds), The effective management of cancer pain. London: Aesculapius Medical Press, 2000, 57–77. 7 Zeppetella G, O’Doherty CA, Collins S. Prevalence and characteristics of breakthrough pain in cancer patients admitted to a hospice. J Pain Sympt Manage 2000; 20: 87–92. 8 Llanes L, Fassbender K, Baracos V, Watanabe. Drug utilisation review on a tertiary palliative care unit. J Symptom Manag 2006; 31:457–464. 9 Bell R, Wisloff T, Eccleston C and Kalso E. Controlled clinical trials in cancer pain. Br J Cancer 206; 94:1559–67. 10 Hoskin PJ. Radiotherapy for bone pain. Pain 1995; 63:137–9. 11 Michaelson M, Smith M. Biphosphonates for treatment and prevention of bone metastases. J Clin Oncol 2005; 23:8219–24. 12 Quigley C. New approaches to the management of neuropathic pain. CME Bull Palliat Med 2000; 2:35–40. 13 Williams J. Critical appraisal of invasive therapies used to treat chronic pain and cancer pain. Eur J Palliat Care 2000; 7:121–5. 14 Rich A, Ellershaw J. Tenesmus/rectal pain – how is it best managed? CME Bull Palliat Med 2000; 2:41–4. 15 Neerkin J, Riley J. Ethical aspects of palliative care in lung cancer and end stage lung disease. Chronic Resp Dis 2006; 3:93–101. 16 Rawlinson F. Dyspnoea and cough. Eur J Palliat Care 2000; 7:161–4. 17 Herbert ME. Myth: codeine is an effective cough suppressant for upper respiratory tract infections. West J Med 2000; 173: 283. 18 Sulica L. Laryngeal thrush. Ann Otol Rhinol Laryngol 2005; 114:369–75 19 Ellershaw JE, Sutcliffe JM, Saunders CM. Dehydration and the dying patient. J Pain Symp Manage 1995; 10:192–7. 20 Davies AN, Daniels C, Pugh R, Sharma K. A comparison of artificial saliva and pilocarpine in the management of xerostomia in patients with advanced cancer. Palliat Med 1998; 12:105–11.
1304 Palliative care
21 National Institute for Clinical Excellence Guidance on Cancer services: Improving Supportive and Palliative Care for Adults with Cancer. www.nice.org.uk 22 Feuer DJ, Broadley KE. Corticosteroids for the resolution of malignant bowel obstruction in advanced gynaecological and gastrointestinal cancer (Cochrane review). Cochrane Database Syst Rev 2000; 2:CD001219. 23 Crosby V, Wilcock A, Lawson N, Corcoran R. The importance of low magnesium in palliative care. Palliat Med 2000; 14:544.
24 Hami F, Trotman I. The treatment of sweating. Eur J Palliat Care 1999; 6:184–7. 25 Abraham A, Kutner J, Beaty B. Suffering at the end of life in the setting of low physical symptom distress. J Palliat Med 2006; 9:658–65 26 Terry W, Olson L, Wilss L, Boulton Lewis G. Experience of dying: concerns of dying patients and carers. Intern Med 2006; 36:338–46 27 King D Quill T. Working with families in palliative care: one size does not fit all. J Palliat Med 2006; 9:704–15
54 Communication with the cancer patient JUSTIN STEBBING AND MAURICE L. SLEVIN
Introduction Communication skills Issues relating to cancer medicine
1305 1305 1310
INTRODUCTION Communication, the activity of conveying information, is often defined as ‘to impart’ or ‘make known’, but its Latin derivation is helpful in emphasizing the ‘sharing’ of information. ‘Communis’ means ‘in common’ and although the main aim remains to elicit and impart information,1 the way that this is accomplished can have a profound effect on the relationship between the doctor and the patient, and on the patient’s approach to his or her disease and treatment. Effective communication is increasingly recognized as a core clinical skill; cancer patients and their families demand a high level of involvement in discussions concerning treatment and outcome. This chapter is arranged in two parts: first, a review of the various aspects relating to effective communication with cancer patients and, secondly, communication issues that relate particularly to the practice of cancer medicine.
Conclusion Key references References
1315 1315 1315
Box 54.1 General Medical Council The General Medical Council requires that doctors make the care of their patient their primary concern, and treat every patient politely and considerately, with respect given to patient’s dignity and privacy. The patients should be listened to, and respect be given to their views with information delivered in a way that they can understand. The patients should be fully involved in decisions about their care. Doctors have a responsibility to keep their professional knowledge and skills up to date, to recognize the limitations of their professional competence, and to be honest and trustworthy. Doctors should respect patient’s confidentiality, and make sure that their personal beliefs should not prejudice their patients’ care. They should act quickly to protect patients from risks if they have good reason to believe that they themselves, or a colleague, may not be fit to practice. Because of their particular responsibility, doctors should avoid abusing their position, and work with their colleagues to best serve a patient’s interest.
COMMUNICATION SKILLS How important is communication? Many aspects of communication are obscured by myths and moralizing, but communication is one of the most important influences on the quality of medical practice. The General Medical Council has stated that communication skills are fundamental to patient care.2 Sir William Osler reputedly penned ‘listen to the patient, he is telling you the diagnosis’. The history is still, justifiably, regarded as the most important part of the consultation.3
Good communication improves the accuracy of diagnosis, enables better management decisions and reduces unnecessary investigations and inappropriate treatment. For the doctor it is professionally rewarding and personally satisfying. For the patient it reduces anxiety4 and uncertainty. There is evidence that good communication improves compliance5 and that improving doctor–patient communication may be the most effective way of reducing the incidence of litigation.6 Results from large multi-centre studies provide
1306 Communication with the cancer patient
conclusive evidence that the vast majority of individuals living with cancer want a great deal of specific information concerning their illness and treatment.7 Positive adjustment to the diagnosis of cancer has been cited as an independent prognostic factor,8 and it is even suggested from a randomized trial, as yet unconfirmed, that counselling for cancer patients may improve survival.9
The current situation In the past decade the swing from protective paternalism towards unsolicited candour10 has continued, though with more calls for a practice marked by congruent sharing of information.11 Galen described a ‘confidence and hope that do more good than Physic’. Good medical practice should embrace both that high ideal and the technical quality of modern medicine. Yet dissatisfaction is high. Ley, in a review of hospital-based surveys, found a median 38 per cent (11–65 per cent) of patients dissatisfied with their consultations with doctors.12 Patients’ most common complaints continue to revolve around communication. The two most common criticisms of hospital practice are: ‘doctors don’t listen’,13 and ‘not being told what’s wrong’.14 Patients can relate a history without interruption in less than 2 minutes if allowed to talk13 and yet are, in the main, interrupted within 18 seconds.15 With the rather sobering suggestion that policemen may break bad news better than doctors16 and that up to one quarter of junior doctors fail to elicit the main problem in taking a history,17 criticisms conspire to overwhelm the medical profession. There has been a boom in the teaching of communication skills as it has been incorporated increasingly into the undergraduate medical curriculum, with the inclusion of roleplay and video feedback to facilitate learning. In 1983, one third of British medical schools offered no communication skills training. In 1992, this figure had fallen to only 3 of the 28 medical schools.18 For many doctors practising today, however, there was little undergraduate training in communication skills. For physicians in practice, the Cancer Research Campaign (CRC) supports Professor Fallowfield’s communication skills courses,19 and a number of other courses are available.20 Although there is presently a relatively small literature on the subject, this is increasing. Research in communication is severely hampered by the lack of a systematic and accepted model of psychological functioning, and the need for accurate controls to reduce bias.
What limits communication? Physicians’ ineffective communication skills have detrimental consequences for patients and their relatives, such as insufficient detection of psychological disturbances, dissatisfaction with care, poor compliance and increased risks of litigation for malpractice. These poor communication
skills also contribute to everyday stress, lack of job satisfaction and burnout among physicians. Literature shows that communication skills training programs may significantly improve matters, contributing to improvements in patients’ satisfaction with care and physicians’ professional satisfaction.21 In this section, the most important factors that limit effective communication will be considered. Areas that doctors and patients wish to leave uncharted will always be present.
Box 54.2 Improved communication skills Improved communication skills: ● Improve the accuracy of diagnosis and quality of management ● Improve the eliciting and imparting of information ● Improve patients’ understanding, retention of information and compliance ● Reduce anxiety, uncertainty and litigation ● Improve doctor and patient satisfaction ● Can be learned.
FROM THE DOCTOR’S PERSPECTIVE
Communication is stressful Communication, particularly face-to-face interaction, is fundamental to the art of medicine, and yet arguably the biggest source of stress for both doctors22 and patients.23 The scars of previous bad experiences, the fear of future wounds or the attrition of the work of the caring profession can create a significant aversion to communicating at any more than a superficial level and inhibit self-awareness. The stigma of death Although it is easy to graciously accept the credit for cures and remissions, the sting of the blame for failure is bitter. In an age marked by high expectations of modern technology, outcomes short of cure are very reasonably perceived as failure, and death still remains a taboo subject in our society. Few are comfortable with their own mortality.24 Doctors, as well as patients, are uncertain, or even fearful, of feelings about such emotive issues, and this naturally limits the freedom with which they can be discussed. The innate limits to good advice Treatment almost always involves advice, either explanation or exhortation. Brief advice with the aim of changing behaviour has been shown to be effective, although with rather limited lasting success. A meta-analysis of 39 controlled smoking cessation trials showed that success was most likely when individualized advice was repeated in different forms by several sources on multiple occasions, and yet success rates were found to most frequently be in the range of
Communication skills 1307
5–10 per cent .25 It is becoming apparent that in circumstances when such ‘lifestyle advice’ is given there is considerable ambivalence. Taking too dictatorial a stance is risky in two ways. First, patients may tend to side with what they see as the other side of the coin, the alternative argument in their conflict of interests. Second, motivation varies, with a wide spectrum of readiness to change. Success may come by using a negotiation method in which the patient, rather than the doctor, articulates the benefits and costs involved, even if all that is achieved is movement toward the goal, in terms of pre-motivation.26 In one study, physicians underestimated the patient’s satisfaction with the overall communication process, the personal contact, the perceived sufficiency of the specific treatment information and their ability to decide on entry into certain studies.27 Words In recent times considerable emphasis has been put on nonverbal communication. A media experiment serves to show the power of the words themselves. During the ‘Mega-lab’ week (25 March 1994), in which science was advertized, Sir Robin Day gave accounts of why Gone with the wind and Some like it hot were his favourite films, one account true and one account fictitious. These were broadcast on TV and radio and published in a daily newspaper. Those with the ‘benefit’ of the non-verbal clues, because of seeing the interview on television, guessed correctly less frequently (TV, 52 per cent; newspaper, 64 per cent; and radio, 73 per cent), perhaps because such non-verbal behaviour can add to what is actually said in an unhelpful way (R. Wiseman, personal communication). Words themselves can, however, be a snare. For many the word ‘cancer’ equates with death, and it is often important to explain that there are over 200 common varieties of cancer, which all behave differently and have different treatments. Words do not mean the same to all people. ‘Sorry’ is a word dangerously contaminated with ambiguity. It can suggest sympathy or guilt. A lot of cancer treatment related issues are about the choice of different approaches each with their own advantages and risks. Medical jargon can leave patients unable to understand or retain information on which important decisions are based. Time A lack of time is often blamed for limiting good communication, and rightly so. However, Richard Asher helpfully commented that ‘… to give the patient the impression that you could spare him an hour and yet make him satisfied with five minutes is an invaluable gift, and of much more use than spending half an hour with him during every minute of which he is made to feel he is encroaching on your time’.28 Privacy The lack of privacy may so militate against the safe disclosure by patients of difficult issues and important information
that arranging another time and another place may be the only constructive way to proceed. For example, ‘business’ or ‘teaching’ ward rounds are not the place to dabble in emotionally charged issues. FROM THE PATIENT’S PERSPECTIVE
Emotional chaos Insensitive communication skill teaching can be de-skilling, leaving competent doctors vulnerable and embarrassed. This can happen particularly during role-play sessions, because of critical feedback. This, however, pales into insignificance beside the degree to which reversing roles and truly becoming a patient leaves members of the medical profession bereft of their insight and equanimity.29 At times of severe stress, taking in a mountain of facts and making important decisions can become impossible, and sensitivity to this can prevent it becoming a barrier to effective communication.
Box 54.3 Information needs of cancer patients ● ● ● ● ● ● ●
Specific name of the illness Is it cancer? Week by week progress Chances of cure Possible treatments Possible side effects How the treatment works.
Denial and collusion Most barriers to open communication are genuine attempts at damage limitation. Denial and collusion in cancer patients are often reasonable defence mechanisms but come with a high emotional cost when the situation eventually deteriorates. How much a doctor should interfere with this process is a very difficult area and requires empathy, experience and good judgement. Dense denial of psychiatric proportions is rare and, to an extent, partial denial is universal – the hope for optimistic goals. Collusion is the attempt by the patient, or more commonly the family or friends, to protect others by denying reality. This can be challenged, to enable the transition from fighting cancer separately to fighting it together. Depression Depression in cancer patients should not be dismissed as an understandable reaction.30 In patients with cancer, where the diagnosis is complicated by organic symptoms and the effects of treatment, a clear and persistent change in mood, hopelessness and an inability to enjoy anything are in fact symptoms that distinguish depression from fatigue. Active treatment often proves to be effective. In situations where there is not only anhedonia but also impaired concentration, feelings of worthlessness or inertia, the use of psychotherapy
1308 Communication with the cancer patient
Box 54.4 Reactions 1. 2. 3. 4. 5.
Anger Denial Depression Anxiety and fear Resolution.
should be deferred pending possible improvement with prescription of antidepressants.31 Age Older patients may want less information and less involvement in decision making.32 Particular account of this should be taken, both in offering to go into more detail with younger patients and in offering ‘escape clauses’ to older patients. Such escape clauses could be in terms of euphemisms or a positive response to the request to ‘Do what you think is right, doctor’. Though this older group may offer fewer ‘signposts’ to the doctor hoping to steer a successful course through difficult issues, they tend to be more easily satisfied.33
Improving communication skills and training There is a wide consensus regarding the need to improve communication skills of health-care professionals dealing with cancer patients.34 Communication is aided by an appropriate knowledge base, skills and attitude. Being a doctor is not instant accreditation in effective communication. Editorials and consensus statements are continual reminders that experience is insufficient as the sole qualification of a communicator.35 We must practice what we preach.36 Many health- and social-care professionals do not feel adequately trained in communicating and in hand-ling interpersonal issues that arise in the care of patients with cancer.37 There is evidence that how a diagnosis is delivered influences the impact of bad news. Cunningham reported five times as many parents satisfied with the way news that their child had Down syndrome was broken, when staff had been trained to be unhurried, honest, balanced and empathic.38 Ensuring that such studies are well controlled is very difficult. The ability to establish rapport with patients and to find out what it is that they are really asking can be taught effectively in a relatively short period of time.39 Maguire has shown, with medical students, that teaching by feedback through audio- and videotape is significantly superior to teaching by the apprenticeship method, although the students need an experienced tutor. A subgroup of 36 were followed up 4–6 years later. All had improved, but the group who had received feedback training maintained their superiority in a number of key skills, notably precision and the exploring of verbal and non-verbal cues.40 Although skills can be learned, that does not ensure that they are used. Fear of patients’ reactions or the perception that
elicited concerns would not be addressed creates a reluctance to apply learned skills.39 As with cardio-pulmonary resuscitation (CPR), having done it, and even doing it, does not guarantee high standards and, without attention, communication skills deteriorate. Courses in communication skills are available, such as those run by the Medical Interview Teaching Association and the Cancer Research UK (CRUK).19 There is no regulating authority, and most are recommended by word of mouth. Adequate training of health-care staff in the practice of communication has clear benefits to health-care professionals and patients. In a systematic review of training methods that covered 47 studies assessing communication training, including 13 interventions, the best results occurred using training programmes that were carried out over a longer time period. Learner-centred programmes using several methods combining a didactic component focusing on theoretical knowledge with practical rehearsal and constructive feedback from peers and skilled facilitators proved to be very effective. Small groups encouraged more intensive participation. Training in communication for medical or nursing students and senior health-care professionals is advisable.37 When randomized studies were examined, results confirmed the usefulness of learner-centred, skills-focused and practice-oriented communication skills training programs organized in small groups of a maximum six participants and lasting at least 20 hours.40 There is only one study assessing the efficacy of posttraining consolidation workshops (six 3-hour consolidation workshops conducted after a 2.5 day basic-training program). In this randomized study of patient interviews, results revealed, in those who took part in the consolidation workshop, a significant increase in acknowledgements and empathic statements, in educated guesses and in negotiations. Importantly, using the Cancer Research UK (CRUK) 14-item questionnaire, patients interacting with physicians who had benefited from consolidation workshops reported higher scores concerning their physicians’ understanding of their disease.41
Good communication: essential elements NON-VERBAL COMMUNICATION
Introductions should be courteous and help the patient see how this consultation fits in to their management. Privacy, eye contact, posture, tone of voice, pauses and nods facilitate communication. Sensitivity to the patient’s needs and expectations, careful negotiation of time and topic and a responsive style prevent avoidable mistakes. Note-taking needs to be unobtrusive and interruptions prevented or kept to a minimum. As a nation, the British probably have the worst reputation for disquiet about touch,42 although even this is now being taught in medical schools.
Communication skills 1309
EMPATHY
Empathy is essentially making a connection with someone, experiencing their emotions as an extension of your own and communicating an understanding of their position and feelings. It can be as simple as listening carefully to how the patient responds to your questions, or as profound as the unspoken understanding of friends. A breadth of experience is an invaluable mentor.43 VERBAL COMMUNICATION
An invitation to talk and when to move on can be negotiated with the patient. Open questions that invite an explanation enable patients to explain what is important to them. When it comes to what the patient thinks and feels, the patient is the expert. However, wide-open questions can be confusing or invite rambling responses. It is important to have clear objectives and to anchor open questions by being specific or directing the patient to the particular area of interest. Encouraging the expression of feelings early on in the consultation gives the message that it is all right to talk about them. Leading questions, value judgements and premature advice or reassurance tend to limit what patients feel able to say. Responding to verbal or non-verbal cues that there are other unvoiced issues that patients wish to discuss is important if the patient is to be given the opportunity to disclose all of his or her concerns. The use of excess in adjectives – ‘devastated, furious, can you bear it?’ – communicates a genuine attempt to understand, rather than risking positive reinforcement of the distress. There are three particularly helpful ways to proceed when difficult or distressing things are being discussed: reflection, clarification and summarizing. Reflection The repetition of the last few words, or word, that the patient said can give the patient the opportunity to say more if they wish. It is especially helpful in emotionally charged interviews, but risks confronting the patient with strong emotions. Sufficient control of the interview needs to be maintained to enable the patient to be moved on through to other matters if they appear to be wallowing in misery. Clarifying Checking that you understand correctly what has been said aids precision and avoids errors. Gut feelings that you have about the link between a particular cause and effect or the underlying emotion, should be explored: ‘so you felt … because of the …?’. Summarizing Summarizing what has been said fundamentally improves understanding while giving the opportunity to order some of the chaos that cancer causes. There may be a lot to be gained in asking the patient to summarize.
Box 54.5 Basic communication skills 1. Have clear objectives 2. Introduce yourself and establish some rapport 3. Establish how much they know, want to know and what they were expecting 4. Use open-ended, directive questions 5. Assume nothing and be flexible 6. Go at an appropriate pace 7. Feedback: what they say and possibly what they can’t say 8. Ask questions. Do they understand you? Do you understand them? 9. Summarize, and ask if there are any other concerns.
ASSESSMENT
It remains the case that in the assessment of any symptom, the essentials of character, severity, frequency, duration and impact, as well as exacerbating and relieving factors, have to be established. However, in communicating, the goal is not simply a precise transcription. The aim, once the patient feels genuinely understood, is to establish that you are aware of all the problems and to prioritize them. Encouraging the patient to prioritize the problems and generate solutions, and talking through these strategies and goals can be very valuable. COPING WITH REACTIONS
Patients exhibit the full spectrum of reaction, from anger and fear to depression and anxiety or denial. It is necessary to acknowledge distressing emotions and to try to understand their impact. If you don’t understand why a patient has reacted in a particular way, then ask. Expressing strong emotions verbally is very therapeutic, and distancing tactics are more likely to exacerbate the problem. CONTINUITY OF CARE
Patients frequently complain that they never see the same doctor. Improving this may be difficult, but rewarding. If consistent personal attention is not possible, then communication with colleagues about patients and accurate recording in the notes of assessments, plans and what patients have been told serves to promote good continuity of care. SURVIVING COMMUNICATION AND BURN-OUT
Successfully engaging in the task of accompanying people through one of the most threatening of life events carries with it an inevitable cost in terms of time and emotions. Believing in what you do, positive feedback, confronting or even avoiding your own reactions and planning rewarding elements to daily practice can help prevent ‘burnout’. Burn-out is typically identified as emotional exhaustion,
1310 Communication with the cancer patient
depersonalization and a sense of low personal accomplishment. Ramirez et al.44 investigated burn-out and psychiatric morbidity among gastroenterologists, surgeons, radiologists and oncologists in the UK, using a questionnaire-based survey. Psychiatric morbidity was estimated using the General Health Questionnaire and the Maslach Burnout Inventory. Questionnaires were returned by 882 of 1133 consultants. The estimated prevalence of psychiatric morbidity was 27 per cent. Burn-out appeared to be associated with feeling overloaded and poorly managed and dealing with patients’ suffering. Better training in communication and management skills was identified as potentially important in protecting against burn-out. A number of organizations, including the British Medical Association (BMA; Tel.: 020 7387 4499) offer anonymity and support to health-care professionals who find themselves in a position of intolerable strain.
ISSUES RELATING TO CANCER MEDICINE How much talk is too much? There is a lot of talk about cancer, with high-profile media coverage of celebrity cases. For most people, this anxious preoccupation remains at arm’s length, but for individuals who develop cancer, for whom it becomes all too real, how much talk is too much?.45 It is an oversimplification to say that there are ‘tellers’ and ‘non-tellers’. The majority of doctors will, to an extent, tailor what is said despite favourite allegories and familiar personalized patter. For those who ‘tell’ there are three particular dangers. First, that they dwell on the morbid, not offering practical and positive advice; second, that they tell patients what they already know but do not want to hear, or hear again; and third, that they fail to appreciate that, for some, considerable time is needed to adjust to bad news. For ‘non-tellers’, who may very compassionately maintain that ‘the truth but not the whole truth’ is important, there are also risks. Patients may be left with half truths, irrational fears and negative previous experience to colour what information they do have. The Hippocratic practice of telling the family the stark truth while presenting a rosy interpretation to the patient invites collusion. Although this can remain a tenet of good practice at times of intolerable strain, this policy should be continually reviewed, and the option for patients to explore more of the ‘whole truth’ should given at a later date. Patients very reasonably over-estimate their prognosis and yet can be bitter at being denied medium-term goals by unexpected deteriorations in their health. Being forewarned with respect to treatment side-effects is being forearmed, and to an extent this is also true of the effects of the illness. There is no gold-standard answer to the question of how much is too much. It varies between doctors, between patients and often, for any one patient, varies with time. Patients are amazingly long suffering, they should be trusted
to guide doctors in difficult interpersonal areas and we should apologize to them when mistakes are made. In one study,46 women with breast cancer regarded their doctors as attachment figures who would care for them. They sought communication that would not compromise that view, and that enhanced confidence in their care.
Breaking bad news There are many advocates of a right way to break bad news. However, common to each school of thought, there are a number of helpful essential elements.47–50 Breaking bad news is painful. Successfully approximating expectations and reality almost always takes more than one consultation. Although the presence of friends and relatives can inhibit the disclosure of difficult things that patients may wish to discuss, the person may retain little information and may benefit from subsequently going over what was said again, with the spouse, friend or relative who was present.51 Recall and adjustment may be helped by a tape if the interview was recorded52 or by written information and illustrations. Patients often fear the worst about symptoms or investigations, and asking for their thoughts on the matter can help to establish how much the patient knows and wants to know. It also creates the opportunity to confirm bad news, rather than delivering bad news, unprepared for possible reactions. Rarely is someone’s reaction to bad news purely dependant on the news itself. Everyone has ‘personal baggage’ that colours how they respond. By giving the patient the opportunity to explain their particular situation in advance, useful information may be made available that enables the interview to be structured to pre-empt problems. The person needs to be prepared, if not, he or she needs to be warned that there is serious news. Giving the details about the diagnosis, stage and prognosis in a sequence of acceptably small packets offers the opportunity to tailor the information to the individual’s needs. What the patient says, asks and how he or she reacts should determine how much is said at any one time. The danger with giving information in a ‘ladder’ like this is that patients may be reluctant to ask questions. One study, although not in the specific context of breaking bad news, found that 42 per cent of patients did not ask the questions about their diagnosis that they had planned to ask.12 The patient needs to be informed of the facts and assessed with respect to consequent emotions and problems. Reinforce the elements that they perceive correctly and gently educate in areas of ignorance or error. Continually check for understanding. Have the confidence to ask them to help you understand how they feel. If they feel that you do understand their distress, then reassurance will be effective. Only if they want to move on can you succeed in making progress. The options have to be clearly explained: what can be, and what might not be, achieved.
Issues relating to cancer medicine 1311
Box 54.6 Breaking bad news Ideally, confirm bad news 1. Be prepared: what do they know and want to know? 2. Warn that you have serious news 3. Be simple and clear. Tailor the information to the patient 4. Has the message been understood? If not check how much more information the patient wishes to know 5. Pause to let it sink in, then respond to their reactions and to difficult questions 6. Pick up the threads so that there is a plan for how to continue discussions at the next meeting.
Bad news inevitably brings a considerable number of problems. Establishing how the patient prioritizes them ensures that they are addressed in an appropriate order. It may be more productive to tackle most of them on another occasion. However, the first opportunity may be the best opportunity to hear all of the issues.53 It can be very helpful to ensure that it is not only the doctor who is available to pick up the pieces. Nurses and nurse specialists provide invaluable support. Patients need to be orientated with respect to what happens next, both
with reference to the medical agenda and to their agenda. Establishing the point of next contact and the ground rules of how discussions might proceed, giving a telephone number through which you can be contacted and fostering realistic hopefulness is investing towards success.
Hope SOURCES OF HOPE
It is often not the case that the diagnosis of cancer can be softened in the same breath with talk of cure. Honest information, not unduly negative or falsely reassuring, offers the safest path for patients to establish their own hopes. There is an obvious aversion to saying that there is no hope. Such statements should be clearly focused ‘… there is no chance of X, Y or Z but, yes, a very real chance of …’. Very much as the profession reports response rates rather than failure rates, patients also wish to minimize the impact of ‘failure’: the half-full rather than half-empty glass. Patients consistently overestimate prognosis and are keen to opt for radical treatment with minimal chance of benefit (Fig. 54.1). The chance that new treatments might become available, however unlikely this may be, is an
Accept Intensive Chemotherapy for a 1% Chance of Cure 60 50 40 % 30 20 10
(a)
Cancer Nurses
GPs
Radiotherapists
Medical Oncologists
Controls
Patients
0
Accept Intensive Chemotherapy for a 3 month Prolongation of Life 50 40 30 % 20 10
Cancer Nurses
GPs
Medical Oncologists
Radiotherapists
(b)
Controls
Patients
0
Figure 54.1 (a) Chances of cure between 1 per cent and 100 per cent. (Slevin M, Stubbs L, Plaut HJ, et al. Attitudes to chemotherapy: comparing views of patients with cancer with those of doctors, nurses and general public. Br Med J 1990; 300: 1458–60.). (b) Chances of prolongation of survival of 3 months to 5 years. (Slevin M, Stubbs L, Plaut HJ, et al. Attitudes to chemotherapy: comparing views of patients with cancer with those of doctors, nurses and general public. Br Med J 1990; 300: 1458–60.)
1312 Communication with the cancer patient
extremely common source of hope. Keeping the ‘options open’ is a very constructive way of coping for patients, and a common practice for doctors. It offers uncommitted hope and supports patients who often live ‘a day at a time’. Perhaps all that separates hope from denial in some situations is that the latter is destructive or obstructive. Many realize that hopes for a cure are not realistic and yet they remain optimistic and appear to find meaning and value in life. In the palliative setting, patients consistently do realize short-term goals. Things that provide enjoyment, or at the least distraction, continuing support and the fact that they will not be abandoned by their doctors contribute to a hopeful attitude. It maybe helpful to conclude with the description by the psychiatrist, Victor Frankl, of his experience of the concentration camps of the Second World War.55 He observed that people who felt that life had real purpose and meaning coped with the atrocities and the almost certain threat of death, while others, for whom life had lost all meaning, quickly succumbed to malnutrition and infection. The latter could be helped to regain the meaning in their lives, with a consequent improvement in its quality. ENGENDERING HOPE
The way that we communicate can engender hope or minimize the risk of hopelessness. Information and order Information fills voids that can otherwise get filled with despair. Positive and practical information banishes fear of the unknown. Simply categorizing issues can chop up overwhelming distress into manageable worries. A new perspective Patients are often faced with what initially appear to be insurmountable problems. By bringing some objectivity, difficult issues can be divorced from the imagination, which so easily fuels despair. This also gives the patient the opportunity of a little distance from reality and the possibility of establishing a new perspective and positive readjustment. Patients’ fears often centre around symptoms that can be well controlled, such as pain. Reassurance can dispel unreasonable fears. Very reasonable fears about death are often not addressed and giving the opportunity to voice these may be constructive.
To get in and out of emotionally laden issues and move people on through such distress is undoubtedly a skill to be cherished. Leaving the patient knowing that their concerns have been understood enables the sharing of painful issues to be a therapeutic manoeuvre. Spirituality Frequently a distinction is drawn between religion (an organized system of beliefs with authority figures, rules, rituals and traditions) and spirituality (a personal belief system related to the transcendent, to a search for meaning in one’s life, and to that which gives one hope, joy, peace, contentment and energy). The diagnosis of cancer can precipitate a search for meaning in the patient’s life. Questions such as ‘Why me?’, ‘Why this?’ and ‘Why now?’ are common. Such existential questions can also reflect specifically spiritual concerns, such as ‘What is the quality of my relationship to myself, others and God?’ In addition to patients’ spiritual beliefs, many have religious practices from which they derive comfort (e.g., prayer) or that enforce constraints (e.g., blood transfusions prohibited). Given the potential depth and breadth of spiritual and religious issues with which patients may be grappling, it may be very helpful to have an understanding of patients’ ideas of their worth, their philosophy and their faith. Simply inquiring of a patient, ‘What does this illness mean for you?’ or ‘What aspects of your religion or spirituality would be helpful for me to know?’ can access important information for a therapeutic relationship. In assessing patients’ strategies to cope with their illness, one qualitative study based on in-depth interviews identified three overarching attitudes to their management: faith, hope and charity.56 Faith in their doctor’s medical expertise precluded the need for patients to seek further information themselves. Hope was essential for patients to carry on with life as normal, and could be maintained through silence and avoiding information, especially too detailed or ‘unsafe’ information. Charity to fellow patients, especially those seen as more needy than themselves, was expressed in the recognition that scarce resources – including information and explanations – had to be shared and meant that limited information was accepted as inevitable. Interestingly, psychologists ‘hated’ this article.
Talking about prognosis
Confidence Patients should be encouraged to take some control through involvement in decisions about major or minor matters, if this promotes confidence. There is a tremendous encouragement in not being alone. Patients need the support of friends and relatives and medical, psychological, social and spiritual support. They may find valuable support in an organized support group. At the very least, patients can be encouraged that many have faced the same fears and yet coped in the absence of other sources of confidence. Sadly, this is a well-trodden path.
How many cancer patients wish to know about prognosis? In a national survey, 85 per cent of Americans wanted a ‘realistic estimate’ of how long they had to live if their type of cancer ‘usually leads to death in less than a year’.57 It is not known whether this figure falls when detached reasoning is challenged by imminent threat, or by crossing the Atlantic. The prognosis of all cancer patients is uncertain. Statistics apply to populations, not individuals, and cannot predict survival for an individual patient. Fixed life expectancies can discolour remaining time, and it has been suggested that
Issues relating to cancer medicine 1313
uncertainty as to when life ends is a prerequisite for life to have meaning and value.58 It is common practice to say in response to questions about prognosis that it is genuinely not known, while checking that the patient’s hopes are not completely unrealistic. Informing a patient is a continuous process. Although the first time may be the best time to tackle issues that patients wish to discuss, there are often other opportunities for patients to ask difficult questions. Some doctors respond to such questions by asking patients if they wish to know of signs of deterioration. Others give rough estimates, while explaining that over-estimates risk procrastination on important issues and bitterness at unexpected deterioration, and that under-estimates risk leaving the patient with a death sentence hanging over them once they exceed the deadline. The Californian Supreme Court has debated traditional medical paternalism versus information sharing and patientcentred decision making with respect to information about prognosis. The family of a patient who had received chemotherapy after the incidental discovery of carcinoma of the pancreas at a laparotomy for a non-functioning kidney successfully sued after his death because the patient was not specifically told that he had a less than 5 per cent chance of surviving 5 years.59 The court challenged the aphorism that if patients don’t ask they don’t want to know. The court ruling was that patients have to specifically state that they do not want to be told of the prognosis to be denied information about actuarial survival statistics. The ruling affirmed information sharing and patient-centred decision-making in the context of a fiduciary (based on trust) physician–patient relationship, and clearly stated that the weighing of the impact of distressing news against a patient’s individual hopes and fears was a non-medical judgement reserved for the patient alone. This ruling was in line with other American legislation, for example, with respect to withholding CPR. Such legislation has been criticized for ignoring medical issues in favour of legal considerations.60 Ethics is, to an extent, the codification of the will of the people. We live in an age increasingly dominated by autonomy. However, patient-centred ethics has to be balanced with benevolence, non-malevolence and justice. Constructing health care to mirror patients and their needs doesn’t always work. Dr Solomon Papper tells an amusing story of an unkempt drug addict who refused to be seen by a particular medical student who looked rather like him, complaining that he was a ‘slob’. He wanted a starchy doctor in a tie.61
Box 54.7 Therapeutic dialogue 1. Listen, and ask open, but directive, questions. 2. Question and summarize until you have the whole picture. 3. Acknowledge and address issues. 4. Reinforce realistic hopefulness.
Talking with children Being told that your child has cancer is one of the hardest things for parents to face. Children generally suspect that there is something seriously wrong despite being shielded from the diagnosis. They have florid imaginations and often feel guilty that the illness is a punishment. Explanations, by necessity, are age dependent. The illness can be explained in terms of cells that are ‘good guys’ and ‘bad guys’. Cartoons or talking to other patients can help. Keeping in touch with schooling and school friends is important. Parents should be encouraged to tell their children that it is alright to feel sad and cry, to ask questions and to talk with the child about the child’s thoughts and feelings. Letting them have some control over things that don’t interfere with their health can be helpful, but setting limits is important. ‘Bending the rules’ tends to provoke the anxiety that things are worse than they seem.62
Informed consent NHS Management Executive guidelines have been a reminder of the legal requirement to obtain informed consent from patients undergoing treatment.63 One paper purporting to investigate whether the British patients or their doctors are afraid of informed consent, compared a simple and a detailed information sheet about complications of elective inguinal hernia repair. The latter was eight times as long, giving the statistical risk of discomfort, failure and permanent damage as well as stating unlikely complications, including death. No increase in anxiety was documented in the group who received the detailed information sheet. However, nearly a quarter (8/33) of these patients felt that they had been given ‘too much’ information. The subgroup of patients who were more anxious prior to the study appeared to be significantly less anxious when given the simpler information (P50.05).64 Fully informed consent can be ‘needlessly cruel’65 and intolerable indecision in anxious patients can be immobilizing, even when the patient is the editor of the New England Journal of Medicine.29 After being diagnosed as having cancer, Dr Franz Ingelfinger was inundated with well intentioned but contradictory advice. He benefited enormously from a wise physician, whom he trusted, who took responsibility for the medical decisions. The extent to which patients wish to participate in the decision-making process is highly variable, as found in a much quoted study, where at least one third of patients preferred to leave decision-making to the doctor alone.32 The injudicious involvement of all patients in medical decisions is likely to heighten anxiety in an already fraught situation. In clinical practice, it is wise to be guided by the patient’s verbal and non-verbal cues.
1314 Communication with the cancer patient
Accrual to randomized trials and phase I and II studies Randomized trials, though rightly hailed as the only solution to many medical problems, generate significant problems themselves, not only in workload. The trust that patients give doctors is grounded on the expectation that the doctors motive is to make them better. When secondary motives, such as ‘learning something’, become priorities this needs to be clearly justified and the significant benefits of external audit, attention and advancing the science of medicine explained. Tackling uncertainty in such a detached way needs to be tempered with information, assurances and reassurance. In one study of patients with inoperable lung cancer, more depression was found in the observation group, with many of those who were not formally depressed appearing ‘puzzled and unhappy’. Those in the treatment groups expressed appreciation of radiotherapy and chemotherapy even if obtaining little symptomatic benefit and suffering unpleasant side effects.66 Good research is difficult. Balancing the need for randomized trials so that ‘the most’ can benefit, forwarding science, making wise individualized judgements and containing uncertainty while being the patient’s advocate is obviously extremely difficult. Collins et al.67 draw attention to the fact that the ISIS-2 trial of aspirin and streptokinase in myocardial infarction was delayed by poor recruitment in the United States because of ‘humanly inappropriate’ written informed consent procedures, compared with the UK, where consent was obtained in the manner considered to be in the best interests of individual patients. The authors highlight the thousands of deaths that may have resulted world-wide from the unnecessary delay in completing the study. The expeditious neglect of the ethical imperative to inform patients is still hotly debated.68 Many patients in whom cure is considered highly unlikely and the chance of palliation and prolongation of life is unknown are eligible for entry to phase I and II studies of new drugs. Toxic treatments known to have some anti-cancer activity, yet which may have a negative impact on quality of life with an unknown chance of remission, are weighed against manoeuvres aimed at symptom relief. Treatment may be extremely inappropriate for some, yet reluctance to treat these people may in some cases lead to disappointment, resentment and perhaps despair.69 In one study, patients with non-resectable non-small-cell lung cancer or colorectal cancer informed about experimental chemotherapy completed a questionnaire on satisfaction with the communication process, general attitude towards experimental treatments, the substance of information and personal contact with the physician following their first consultation in a medical oncology unit. Physicians completed a questionnaire on their perception of the patients’ satisfaction. Among 68 physician–cancer patient pairs, 29 patients were informed on chemotherapy in randomized trials and 39 in non-randomized studies. The general attitude towards experimental treatment was positive
or very positive in 71 per cent of patients. Information on the treatment was perceived as completely adequate in 93 per cent of patients informed on randomized and in 67 per cent informed on non-randomized trials. Considerable differences were observed between patients informed about experimental chemotherapy in randomized and non-randomized trials, both with respect to their perception of how adequate the information on the specific treatments were, and whether it was sufficient for decisions on study entry. Importantly, these data supported the fact that cancer patients have a desire for and ability to understand rather detailed and comprehensive treatment information.27
Assessing quality of life The constituents of quality of life (QOL) are personal and will always remain immutably subjective. In palliative oncology, QOL is rightly paramount when considering the worth of treatment. It may be helpful to make patients aware that not only does more effective treatment improve QOL, but that side effects may be less important than control of disease and that QOL often improves despite the absence of an objective response, possibly due to minimal shrinkage of the tumour.70 There are devotees of questionnaires to support history taking;3 they may not increase objectivity but improve completeness and precision. They are now very well-validated measures of QOL.68,71 Asking patients to fill in such questionnaires outside the context of clinical trials may improve the detection of unspoken emotional distress, functional limitation or social deprivation. These may be things that should be addressed or treated.
Communication issues and screening for cancer Screening potentially offers a fairly rapid and important impact on cancer mortality. Although screening provokes considerable anxiety, it offers the hope of cure with less radical treatment to some who would have died. The majority eventually find screening reassuring. Effective communication may improve accrual to screening programmes. The potential impact of screening on cures, the procedure and likelihood of positive and false-positive results, lag and lead time and the overtreatment of borderline abnormalities need to be clearly explained.
Sexuality Sexual dysfunction is common among cancer patients because of anxiety, ill health and organic causes. It is difficult to treat because of the large psychogenic element. However, sexual counselling has been shown to be effective in reducing long-term morbidity.72 The message has been aptly
References 1315
put: talk about it.73 Patients do not volunteer sexual problems, and specific enquiry should be made of sexual function beyond broader issues such as ‘Who are the most important people in your life?’ This is often met with relief rather than embarrassment, particularly in those with problems.74
should not be afraid to say ‘I don’t know’,77 but never leave it there.
KEY LEARNING POINTS ●
Racial issues
● ●
Health-care disparities have been identified in the treatment of ethnic minorities. Black patients report lower trust in physicians than white patients, but this difference has been poorly studied. In a recent questionnaire study, there was a perception that physician communication was less supportive, less partnering and also less informative accounting for this lower trust.75 Further research on tailored messages for cancer communication, taking into account individualized minority issues, are required here.
Multidisciplinary teams, the general practitioner and support organizations Ensuring discussion about patients’ cases between different specialists can contribute greatly to good standards of care. Working alongside colleagues in a structured way offers a safety net against error and enables continuing education as well as support. Timely, clear and, perhaps, structured correspondence with the general practitioner can avoid distancing the family doctor during a crucial illness. The general practitioner needs clear information about treatment, side effects, prognosis and what the patient has been told. There are many useful organizations that offer information, advice and emotional support for patients. Patients feel less anxious through the anonymity of the telephone and can benefit from talking to an independent cancer nurse (Cancer BACUP, Tel.: 0208 800 1234).
CONCLUSION It is increasingly accepted that communication plays a significant role in many aspects of the care experience, and that poor communication can have a significantly negative influence on the patient’s psychosocial experience, symptom management, treatment decisions and quality of life. Poor communication may also have an economic impact worthy of attention, and existential and material costs associated with poor communication in cancer care may well be considerable.76 Good communication potentially offers the most rewarding aspect of total patient care, but at not inconsiderable cost. Patients have a very important and valid contribution to their ongoing care. The way in which they are involved and the way in which information is elicited and imparted can maximize the quality of their treatment. With sensitivity and a readiness to learn we
●
● ● ●
●
Have clear objectives, assume nothing and be flexible. Introduce yourself and establish some rapport. Listen and ask open, but directive, questions. Question and summarize until you have the whole picture. Acknowledge and address issues. Summarize and screen for other issues. Be simple and clear, tailor the information to the patient. Reinforce realistic hopefulness.
KEY REFERENCES Buckman R. I don’t know what to say. Vancouver, WA: Vintage books, 1992. Davis H, Fallowfield L. Counselling and communication in health care. Chichester: John Wiley and Sons, 1991. Faulkner A. Effective interaction with patients. Edinburgh: Churchill Livingstone, 1992. Simpson M, Buckman R, Stewart M, et al. Doctor–patient communication: the Toronto consensus statement. Br Med J 1991; 303: 1385–7. Slevin ML. Talking about cancer: how much is too much? Br J Hosp Med 1987; 38: 58–9.
REFERENCES 1 Spence J. Function of the hospital out-patient department. Lancet 1953; 261:275. 2 General Medical Council. Recommendations on general clinical training. London: GMC, 1987. 3 Short D. History taking. Br J Hosp Med 1993; 50:337–9. 4 Fallowfield L J, Baum M, Maguire GP. Effects of breast conservation on psychological morbidity associated with diagnosis and treatment of early breast cancer. Br Med J 1986; 293:1331–4. 5 Davis H, Fallowfield L. Counselling and communication in health care. Chichester: John Wiley and Sons, 1991. 6 Shapiro R, Simpson DE, Lawrence SL, et al. A survey of sued and nonsued physicians and suing patients. Arch Intern Med 1989; 149:2190–6. 7 Jenkins V, Fallowfield L, Saul J. Information needs of patients with cancer: results from a large study in UK cancer centres. Br J Cancer 2001; 84:48–51. 8 Greer S, Watson M. Mental adjustment to cancer; its measurement and prognostic importance. Cancer Surv 1987; 6:439–53.
1316 Communication with the cancer patient
9 Spiegel D, Bloom JR, Kraemer HC, et al. Effects of psychological treatment on survival of patients with metastatic breast cancer. Lancet 1989; ii:888–91. 10 Novack D, Plumber R, Smith RL, et al. Changes in physicians’ attitudes towards telling the cancer patient. JAMA 1979; 241:897. 11 Schain WS. Physician–patient communication about breast cancer. A challenge for the 1990s. Surg. Clin. North Am. 1990; 70(4):917–36. 12 Ley P. Communicating with patients. London: Croom Helm, 1988. 13 Blau J. Time to let the patient speak. Br Med J 1989; 298:39. 14 Fletcher C. Listening and talking to patients. Br Med J 1980; 281:994. 15 Beckman H, Frankel R. The effect of physician behaviour on the collection of data. Ann Intern Med 1984; 101:692–6. 16 Finlay I, Dallimore D. Your child is dead. Br Med J 1991; 302:1524–5. 17 Maguire P, Fairbairn S, Fletcher C. Consultation skills of young doctors: II – most young doctors are bad at giving information. Br Med J 1986; 292:1576–8. 18 McManus I, Vincent CA, Thom S, et al. Teaching communication skills to clinical students. Br Med J 1993; 306:1322–7. 19 Fallowfield L, Lipkin M, Hall A. Teaching senior oncologists communication skills: results from phase I of a comprehensive longitudinal program in the United Kingdom. J Clin Oncol 1998; 16:1961–8. 20 Baile WF, Lenzi R, Kudelka AP, et al. Improving physician–patient communication in cancer care: outcome of a workshop for oncologists. J Cancer Educ 1997; 12:166–73. 21 Bragard I, Razavi D, Marchal S, et al. Teaching communication and stress management skills to junior physicians dealing with cancer patients: a Belgian Interuniversity Curriculum. Support Care Cancer 2006; 18:1–8. 22 Firth-Cozens J. Emotional distress in junior house officers. Br Med J 1987; 295:533–6. 23 McLaughlan C. Handling distressed relatives and breaking bad news. Br Med J 1990; 301:1145–9. 24 Feifel H. Toward death: a psychological perspective. In Schneidmen, E.S. (ed.), Death: current perspectives. Palo Alto, California: Mayfield Publishing, 1976. 25 Kottke T, Battista RN, De Grise G, et al. Attributes of successful smoking cessation interventions in medical practice. A meta analysis of 39 controlled trials. JAMA 1988; 259:2882–9. 26 Rollnick S, Kinnersley P, Stott N. Methods of helping patients with behaviour change. Br Med J 1993; 307:188–90. 27 Sorensen JB, Rossel P, Holm S. Patient-physician communication concerning participation in cancer chemotherapy trials. Br J Cancer 2004; 90:328–32. 28 Asher R. Talking sense. Tunbridge Wells: Pitman, 1972. 29 Ingelfinger F. Arrogance. N Engl J Med 1980; 303:1507–11. 30 Maguire P, Hopwood P, Tarlier N, et al. Treatment of depression in cancer patients. Acta Psychiatr Scand Suppl 1985; 320(81):81–4.
31 Goldberg R, Cullen L. Use of psychotropics in cancer patients. Psychosomatics 1986; 27:687–700. 32 Cassileth B, Zupkis RV, Sutton-Smith K, et al. Information and participation preferences among cancer patients. Ann Intern Med 1980; 92:832–6. 33 Blanchard C, Ruckeschel MD, Fletcher BA. The impact of oncologists’ behaviour on patient satisfaction with morning rounds. Cancer 1986; 58:387–93. 34 Delvaux N, Razavi D, Marchal S, et al. Effects of a 105 hours psychological training program on attitudes, communication skills and occupational stress in oncology: a randomised study. Br J Cancer 2004; 90:106–14. 35 Simpson M, Buckman R, Stewart M, et al. Doctor–patient communication: the Toronto consensus statement. Br Med J 1991; 303:1385–7. 36 Roter DL, Larson S, Fischer GS, Arnold RM, Tulsky JA. Experts practice what they preach: a descriptive study of best and normative practices in end-of-life discussions. Arch Intern Med 2000; 160(22):3477–85. 37 Gysels M, Richardson A, Higginson IJ. Communication training for health professionals who care for patients with cancer: a systematic review of training methods. Support Care Cancer 2005; 13:356–66. 38 Cunningham C, Morgan P, McGucken R. Down’s syndrome: is dissatisfaction with disclosure of diagnosis inevitable? Dev Med Child Neurol 1984; 26:33–9. 39 Maguire P. Can communication skills be taught? Br J Hosp Med 1990; 43:215–16. 40 Maguire P, Fairbairn S, Fletcher C. Consultation skills of young doctors: I – benefits of feedback training in interviewing as students persist. Br Med J 1986; 292:1573–6. 40 Merckaert I, Libert Y, Razavi D. Communication skills training in cancer care: where are we and were are we going? Curr Opin Oncol 2005; 17:319–30. 41 Razavi D, Merckaert I, Marchal S, et al. How to optimize physicians’ communication skills in cancer care: results of a randomized study assessing the usefulness of posttraining consolidation workshops. J Clin Oncol 2003; 21:3141–9. 42 Heylings P. The no touch epidemic – an English disease. Br Med J 1973; 2:111. 43 Albom M. Tuesdays with Morrie. New York: Doubleday, 1997. 44 Ramirez AJ, Graham J, Richards MA, et al. Mental health of hospital consultants: the effects of stress and satisfaction at work. Lancet 1996; 347:724–8. 45 Slevin ML. Talking about cancer: how much is too much? Br J Hosp Med 1987; 38:56, 58–9. 46 Wright EB, Holcombe C, Salmon P. Doctors’ communication of trust, care, respect in breast cancer: qualitative study. Br Med J 2004; 328:864. 47 Maguire P, Faulkner A. Communicate with cancer patients: 1. Handling bad news and difficult questions. Br Med J 1988; 297:907–9. 48 Faulkner A. Effective interaction with patients. Edinburgh: Churchill Livingstone, 1992. 49 Fallowfield L. Giving sad and bad news. Lancet 1993; 341(8843):476–8.
References 1317
50 Baile WF, Buckman R, Lenzi R, et al. SPIKES – A six-step protocol for delivering bad news: application to the patient with cancer. Oncologist 2000; 5(4):302–11. 51 Fallowfield L, Baum M, Maguire GP, et al. Addressing the psychological needs of the conservatively treated cancer patient. J. R. Soc. Med. 1987; 80:696–700. 52 Hogbin B, Fallowfield L. Getting it taped: the ‘bad news’ consultation with cancer patients. Br J Hosp Med 1989; 41:330–33. 53 Haven CM, Maguire P. Disclosure of concerns by hospice patients and their identification by nurses. Palliat Med 1997; 11:283–90. 54 Slevin M, Stubbs L, Plaut HJ, et al. Attitudes to chemotherapy: comparing views of patients with cancer with those of doctors, nurses and general public. Br Med J 1990; 300:1458–60. 55 Frankl VE. Man’s searching for meaning. Washington DC: Washington Square Press, 1946. 56 Leydon D, Boutlon M, Moynihan C, et al. Cancer patients’ information needs and information seeking behaviour: in depth interview study. Br Med J 2000; 320:909–13. 57 President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. Making health care decisions: the ethical and legal implications of informed consent in the patient-practitioner relationship, vol. 2, Appendices. Washington DC: Government Printing Office, 1982, 245–6. 58 Harris J. The value of life – an introduction to medical ethics. London: Routledge and Kegan Paul, 1985, 87–110. 59 Annas GJ. Informed consent, cancer and truth in prognosis. N Engl J Med 1994; 330(3):223–5. 60 Snider G. The do not resuscitate order – ethical and legal imperative or medical decision? Am Rev Respir Dis 1991; 143:665–74. 61 Papper S. Doing right. Everyday medical ethics. Boston: Little, Brown and Co, 1983, 125. 62 US Department of Health and Human Services. Talking with your child about cancer. Bethesda, Maryland: National Institutes of Health, 1994.
63 Delamothe A. Consenting patients. Br Med J 1990; 301:510. 64 Kerrigan D, Thevasagayam RS, Woods TO, et al. Who’s afraid of informed consent? Br Med J 1993; 306:298–300. 65 Tobias J, Souhami R. Fully informed consent can be needlessly cruel. Br Med J 1993; 307:1199–201. 66 Hughes J. Depressive illness and lung cancer. II. Follow-up of inoperable patients. Eur J Surg Oncol 1985; 11:21. 67 Collins R, Doll R, Peto R. Ethics in clinical trials. In: Williams CJ (ed.) Introducing new treatments of cancer: practical, ethical and legal problems. Chichester: Wiley, 1992, 49–65. 68 Kiebert GM, Kaasa S. Quality of life in clinical cancer trials: experience and perspective of the European Organization for Research and Treatment of Cancer. J Natl Cancer Inst Monographs 1996; 20:91–5. 69 Cody M, Slevin M. Treatment decisions in advanced ovarian cancer. Br J Cancer 1989; 60:155–6. 70 Slevin M, Plaut HJ, Lynch D. Who should measure quality of life, the doctor or the patient? Br J Cancer 1988; 57:109–12. 71 Maguire P, Selby P. Assessing quality of life in cancer patients. Br J Cancer 1989; 60:437–40. 72 Capone M, Good RS, Wentie KS, et al. Psychosocial rehabilitation of gynaecologic oncology patients. Arch Phys Med Rehabil 1980; 61:128–32. 73 Crowther M, Corney R, Shepherd J. Psychosexual implications of gynaecological cancer: talk about it. Br Med J 1994; 308:869–70. 74 Tomlinson J. ABC of sexual health: taking a sexual history. Br Med J 1998; 317:1573–6. 75 Gordon HS, Street RL Jr, Sharf BF, et al. Racial differences in trust and lung cancer patients’ perceptions of physician communication. J Clin Oncol 2006; 24:904–9. 76 Thorne SF, Bultz BD, Baile WF, et al. Is there a cost to poor communication in cancer care?: a critical review of the literature. Psychooncology 2005; 14:875–84. 77 Buckman R. I don’t know what to say. London: Papermac, 1988.
55 Clinical cancer genetics LISA J. WALKER AND ROS A. EELES
Introduction Cancer pre-disposition genes Research approaches for the identification of cancer pre-disposition genes Cancer risks associated with cancer pre-disposition genes
1318 1318 1322 1323
Management of a known or suspected cancer pre-disposition gene mutation carrier 1326 Summary 1334 Key references 1334 References 1334
INTRODUCTION
CANCER PRE-DISPOSITION GENES
Cancer is a common disease, affecting up to a third of the population at some time in their lives. All cancer can be termed ‘genetic’ as cancer is caused by genetic mutations (alterations in the DNA code), which result in abnormal cellular growth and/or proliferation. The majority of these mutations are sporadic (only occur in the cancer cell) and only a small proportion of these cases (approximately 5–10 per cent)1 are due to the inheritance of a germ-line mutation in a high-penetrance cancer pre-disposition gene. This mutation is then present in every somatic cell and, on average, in half the gametes (a gamete only contains half of the total genes) and therefore has a chance of being passed onto the offspring of an affected individual. These mutated cancer pre-disposition genes have a well-defined pattern of inheritance. Approximately a further 20 per cent of cancer cases can be described as familial, i.e., there is a clustering of cancer cases within the family, but they do not show a welldefined pattern of inheritance. These family clusters may be due to the chance clustering of common cancers, the inheritance of genes that are associated with only a slightly increased cancer risk, the sharing of common environmental influences, or they may be of multifactorial origin, possibly as a result of the inheritance of genes that render an individual more susceptible to environmental influences. This chapter will focus on cancer pre-disposition genes, cancer risks associated with these genes and the management of suspected cancer pre-disposition gene mutation carriers.
Cancer pre-disposition genes are mutated genes, the normal function of which is to regulate cell growth or the detection and/or repair of DNA damage. These are germline mutations and are present in all nucleated cells, including on average half of the germ cells. The risk that a cancer pre-disposition gene gives rise to the development of cancer is designated the penetrance, and the fact that many of these genes do not universally result in cancer development is termed incomplete penetrance. Familial clustering of the same type of cancer may be due to more than one type of cancer pre-disposition gene. This is termed genetic heterogeneity. For example, familial breast cancer, in which there are clusters of four cases of breast cancer at 60 years of age in the same lineage, may be due to mutations in either the BRCA1 or BRCA2 gene (breast cancer genes 1 and 2). However, calculations from the Breast Cancer Linkage Consortium,2* which collates international data from breast cancer families, have suggested that in 63 per cent of such families cases are due to mutations in either BRCA1 or BRCA2. Cases in the remaining 37 per cent of families are likely to be due to gene(s) that remain to be discovered. Cancer pre-disposition genes can be associated with syndromes that predominantly consist of clustering of cancers at either one or multiple associated sites (Table 55.1). For example, BRCA1/2 genes pre-dispose to breast, ovarian, prostate and other cancers, and hereditary non-polyposis colon cancer syndrome (HNPCC) mismatch repair genes
Cancer pre-disposition genes 1319
Table 55.1 Syndromes associated with increased risk of malignancy where the major feature associated with the syndrome is the development of cancer Malignancies
Risk (%)1
Mode2
Chromosomal location3
Gene name
Reference
Melanoma Familial polyposis coli
Melanoma Large bowel Upper GI tract Desmoid tumour
53 ⬃100 5 20
D D
9p 5q
CDKN2 (P16) APC
3 4
Familial breast/ovarian cancer syndrome 1
Breast (female)
80
D
17q
BRCA1
2*,5
Ovary Colon Prostate
60 6 6
Breast (female)
80
D
13q
BRCA2
2*,6
Ovary Prostate Breast (male) Pancreas Other cancers, e.g., cutaneous and ocular melanoma, gall bladder, bile duct, fallopian tube, stomach
40 14 5 4 ? 1
Colon
70–80
D
2p 3p 2p 2q 7q
hMSH2 hMLH1 hMLH6 PMS1 PMS2
7 7 8 9 9
Endometrium Ovary Stomach Urothelial Melanoma, head and neck, brain, small bowel
43 10 11–19% 5
2p
hMSH2
10
hMSH2 hMLH1 APC PMS2
11
Syndrome name
Familial breast/ovarian cancer syndrome 2
Hereditary non-polyposis colorectal cancer (HNPCC)
Muir–Torre syndrome
As HNPCC with skin lesions Keratoacanthoma/ sebaceous cysts
D
Turcot’s syndrome
CNS malignancies Very early onset colon cancer (20 years) with café-au-lait patches
R
MYH-associated polyposis
Colorectal
R
1p
MYH
12
Hereditary prostate cancer
Prostate
85
D/R/X-linked
?
13
D
Multiple loci on linkage 13q
BRCA2
14
Li-Fraumeni syndrome
Sarcoma
24 childhood cancer Overall cancer risk: 74 in men, 95 in women
D
17p
TP53
15
Early-onset breast cancer Brain tumour Leukaemia Adrenocortical tumour Other cancers
1320 Clinical cancer genetics
Table 55.1 (Continued) Malignancies
Risk (%)1
Mode2
Chromosomal location3
Gene name
Multiple endocrine neoplasia type 1
Parathyroid, endocrine pancreas, pituitary
70–90
D
11q
MEN1
16
Multiple endocrine neoplasia type 2
Medullary carcinoma of thyroid Phaeochromocytoma (type 2A)
70
D
10q
RET
17
Retinoblastoma Osteosarcoma Other cancers
90 6 8
D
13q
RB1
18
Syndrome name
Retinoblastoma
Reference
50
1
The risk is either the ‘lifetime risk’, as quoted in the reference articles, or the ‘risk to age 70 years’ in those studies which have performed detailed agespecific calculations. Where possible, risks by set ages or a ‘per site’ risk is given; in the absence of such figures a ‘syndrome’ penetrance estimate (risk of cancer development) is provided. These risks are approximate and may vary between different populations with different mutation profiles. 2 Mode of inheritance is classified as autosomal dominant (D) or autosomal recessive (R). 3 Chromosomal arms: q, long arm; p, short arm.
pre-dispose to gastrointestinal, gynaecological, urinary and other cancers. Other genetic syndromes are associated with an increased risk of cancer in addition to other, nonmalignant, features of a syndrome such as neurofibromatosis, multiple endocrine neoplasia (MEN) 1 and 2 and ataxia telangiectasia syndromes (Table 55.2). The ability to recognize clustering of cancers at different sites as being part of a syndrome is an important part of recognising the possible presence of a cancer pre-disposition gene in a family.
Inheritance Inheritance of germ-line mutations in cancer pre-disposition genes may be either dominant or recessive at the genetic level or X-linked. We all carry two copies (alleles) of every gene, one copy from each parent, and as only one allele can be passed down to the next generation, there is a 50:50 chance as to which allele is inherited (Figure 55.1). In dominant inheritance the presence of a single mutated allele is usually sufficient to cause the associated disease and approximately 50 per cent of all offspring develop the disease. In recessive inheritance the presence of a single mutated allele is not usually sufficient for disease expression and two mutated alleles are required. Usually both parents have to carry the mutated allele for the creation of an offspring affected by disease, but they themselves are unaffected, as their ‘normal’ allele effectively acts as a backup for the mutated one. Two ‘carrier’ parents therefore have a 25 per cent chance of having an affected child. Examples of noncancer conditions inherited in this way include cystic fibrosis and β-thalassaemia. The majority of cancer predisposition genes are recessively inherited at the cellular level, but dominantly inherited in families. In X-linked inheritance the mutated gene is carried on the X-chromosome. Females have two X-chromosomes,
and can therefore be carriers of the condition but not usually affected by it. As males only have one copy of the X chromosome, if they inherit a mutated gene on the X from their mother, they will inherit the condition. A carrier female would therefore have a 50 per cent chance of passing the condition on to each of their sons, and a 50 per cent chance that their daughters would, themselves, be carriers. The best-known example of this mode of inheritance outside cancer is Duchenne muscular dystrophy, but X-linked familial prostate cancer has also been observed in a small number of families, although the exact gene has not yet been identified.44
Mechanisms of action Cancer pre-disposition genes can be oncogenes, tumour suppressor genes or mismatch repair genes. Oncogenes are mutated normal genes (proto-oncogenes) in which mutation tends to cause a ‘gain in function’ effect resulting in increased growth or proliferation of the affected cells. Most oncogenes tend to act in a dominant manner and those causing cancer include the RET oncogene in the multiple endocrine neoplasia 2A (MEN2A) syndrome or MET oncogene in familial papillary renal cancer. Tumour suppressor genes are normal genes in which mutation tends to cause a ‘loss of function’ effect in the control mechanisms of growth and/or cellular proliferation pathways. Examples include RB1, involved in retinoblastoma. Most cancer pre-disposition genes are tumour suppressor genes and are recessively inherited at the cellular level. However, they tend to manifest dominant inheritance (the chance of a mutation being inherited by the offspring is 50 per cent). A sporadic mutation of the remaining normal allele occurs in a somatic cell during the lifetime of the germ-line mutation carrier to lead to cancer development. This two-stage process in the development of cancers
Table 55.2 Some of the ‘rare’ genetic syndromes associated with an increased risk of malignancy Syndrome
Neoplasia or malignancy
Risk (%)1
Neurofibromatosis type 1
Plexiform neurofibroma, optic glioma, neurofibrosarcoma
4–5
Neurofibromatosis type 2
Acoustic neuroma (vestibular schwannoma) ● bilateral ● unilateral Meningioma Spinal tumours Astrocytomas Ependymomas
85 6 45 26 4 3
von Hippel–Lindau
Cerebellar haemangioblastoma Retinal angioma Renal cell carcinoma Phaeochromocytoma Renal, liver and pancreatic cysts
Ataxia telangiectasia Fanconi anaemia
Mode2
Chromosomal location3
Gene name
Reference
D
17q
NF1
19*
D
22q
NF2
20
44–72 25–60 25–60 15 16–50
D
3p
VHL
21
Lymphoid malignancy Breast cancer
10–15 ?
R
11q
ATM
22
Lymphoid malignancy Myelodysplastic syndrome
9 7
R
16q ? 9q
FANCA FANCB FANCC
23
Wilms tumour
20 (FANCD1 only)
13q
25
3p 6p 11p 9p 17q 2p16 14q21
FANCD1/ BRCA2 FANCD2 FANCE FANCF FANCG/XRCC9 FANCJ/BRIP1 FANCL FANCM
24
26 27 28 29 30 31 32
Bloom syndrome
Many sites Immunodeficiency
40
R
15q
BLM
33
Xeroderma pigmentosum
Skin cancers (basal cell carcinoma, squamous cell carcinoma, melanoma)
⬃100
R
PTEN hamartoma tumour syndrome/ Cowden syndrome
Breast cancer (female) Thyroid cancer Bowel cancer Multiple hamartomas of skin, tongue and bowel
30–501 10 ?3 ⬃100
D
9q22 2q21 3p25 19q13 11p 16p 13q 6p 10q
XPA XPB XPC XPD XPE XPF XPG XPV PTEN
34 35 36 37 38 39 40 41 42
Basal cell naevus/Gorlin’s syndrome
Basal cell carcinoma Ovarian fibroma Medulloblastoma Falx calcification, bifid ribs, macrocephaly
90 24 5 85
D
9q
PTCH
43*
1
Lifetime risk of neoplasia or cancer. Mode of inheritance is classified as autosomal dominant (D) or autosomal recessive (R). 3 Chromosomal arms: p, short arm; q, long arm. 4 Risk by age 50 years. PTEN, phosphatase and tensin homologue. 2
1322 Clinical cancer genetics
Father, has two alleles of each gene
Mother, has two alleles of each gene
Gametes, contain single allele
Offspring, carry two alleles
(where one stage is germline and the other is somatic) is known as Knudson’s Two-Hit Hypothesis.45 Mismatch-repair genes (HNPCC syndrome) maintain the integrity of the genome and mutations in them permit acquired genetic damage to accumulate resulting in the creation of a cancer cell. Hereditary cancer pre-disposition genes have also been divided into ‘gatekeeper genes’ and ‘caretaker genes’. ‘Gatekeeper’ genes are those that regulate progression through the cell cycle. Disturbance of their function leads to an imbalance of cell division over cell death. This cellular proliferation is followed by the accumulation of multiple somatic genetic events leading to the tumour. Examples of ‘gatekeeper’ genes include RB1 and TP53. ‘Caretaker’ genes maintain the integrity of the genome. Mutation occurring in these genes simply gives rise to genetic instability, causing mutation in other genes, including ‘gatekeeper’ genes. The DNA mismatch repair genes causing HNPCC are examples of ‘caretaker’ genes.
RESEARCH APPROACHES FOR THE IDENTIFICATION OF CANCER PREDISPOSITION GENES When a cancer pre-disposition gene is thought to be the cause of familial clustering of cancer cases, there are several approaches to locate the gene. Once located and characterized, genetic testing can then be offered in the clinical setting.
Cytogenetic alterations Gross chromosomal changes can be seen on cytogenetic analysis. Rarely, a study of a constitutional chromosomal alteration seen on cytogenetic analysis in an individual who has an unusually early onset of cancer and other
Figure 55.1 Principles of genetic inheritance (See Plate Section.).
unusual phenotypic features can indicate the location of a cancer pre-disposition gene. The chromosomal study of a man with mental retardation and polyposis led to the finding of a loss of part of chromosome 5, subsequently found to be the location of the polyposis gene, APC.4
Linkage analysis The concept of genetic linkage was first recognized by Gregor Mendel, who noted that certain characteristics of his experimental plants tended to be co-inherited. The explanation for this became clear once it was recognized that chromosomes contain the genetic material and two traits are linked only if the corresponding genes for them reside close together on the same chromosome. The search for cancer pre-disposition genes using linkage relies on collections of families with numerous cancer cases of the same cancer type. Co-inheritance of specific genetic markers with the disease is said to show evidence of linkage if the co-inheritance is greater than would be expected by chance. This is expressed as a ‘LOD score’ (logarithm to base 10 of the odds). A LOD score is similar to a p-value in clinical trials, and a LOD score of 3 is statistically significant and equivalent to odds of linkage of 1000 to 1 (log101000 3).
Phenotypic features A physical characteristic associated with a cancer predisposition syndrome may give a clue as to the location of the cancer pre-disposition gene. An example of this is the co-existence of aniridia and genitourinary abnormalities with Wilms tumour in the WAGR (Wilms – Aniridia – Genitourinary – Retardation) syndrome. This is caused by a contiguous gene deletion on chromosome 11.46
Cancer risks associated with cancer pre-disposition genes 1323
Association studies A number of disease susceptibility loci have been identified through direct testing of candidate genes, looking for associations between particular alleles and disease, by comparing allele frequencies in affected individuals and controls. A candidate gene can be identified by a number of methods including knowledge of the natural history of specific cancers; for example, androgen receptor gene polymorphisms (variants in the genetic code) are associated with prostate cancer risk in some studies.47
CANCER RISKS ASSOCIATED WITH CANCER PRE-DISPOSITION GENES Cancer risks depend on the presence of mutations in a cancer pre-disposition gene and its penetrance. Penetrance may be affected by external factors, such as lifestyle, and may depend on the ethnic origin of an individual due to population-specific mutation risks. Different ethnic populations may have a different gene penetrance, which is illustrated by the breast cancer penetrance estimates for BRCA1/2 mutation carriers. Using data from the Breast Cancer Linkage Consortium based on breast and ovarian cancer families identified from a world-wide population of highrisk families with breast cancer, the risk of breast cancer is estimated to be 85 per cent by 80 years.2* Estimates based on the Ashkenazim are 60 per cent by 70 years48* and for the Icelandic founder mutation carriers to be 37 per cent by 80 years,49* in contrast to a general population risk of breast cancer of around 11 per cent by 80 years in the UK (www.statistics.gov.uk). The ethnic population differences may be due to a founder mutation dependent risk, the effect of other modifying genes in a population or the added effect of environmental influences, which may be shared within specific populations. It is therefore extremely important to ascertain the genetic origin of the patient before formal genetic counselling is initiated. Tables 55.1 and 55.2 summarize the current penetrance/risk estimates associated with known cancer pre-disposition genes. The estimate of penetrance can be confounded by the presence of phenocopies when research into the identification of a cancer pre-disposition gene is undertaken. Phenocopies are people who have developed the disease of interest but are found not to carry the disease pre-disposition gene, therefore the disease occurred by chance alone or may have been due to environmental influences. Phenocopies are a particular problem in syndromes associated with common cancers such as breast or prostate cancer.
Risk assessment This is arguably the most difficult part of cancer genetic counselling: first, to arrive at a risk estimate and second, to convey this information in the most appropriate manner
to the individual so that they can understand and retain this information and are not made inappropriately anxious about their risks. The first risk estimation is the chance that a familial cluster is due to genetic pre-disposition. This is called the prior probability of a genetic pre-disposition gene being present in a family. This estimation can be based upon published data or clinical experiences when published data are lacking, which unfortunately is often the case with rare genetic conditions. In the case of breast cancer families, we are in the fortunate position of having risk estimates available from the Cancer and Steroid Hormone (‘CASH’ or ‘Claus’) study.50* This was a study of 4700 women with breast cancer who had their family history taken. A statistical model (Claus model, named after the author of the study) was developed that estimated the chance that a cancer pre-disposition gene was present in a family. This model can be used to generate the curves in Figure 55.2, which can be used easily in a genetics clinic to estimate risk. For example, if an individual has two first degree relatives with breast cancer at 45 years (Fig. 55.3), there is a 60 per cent chance that there is a breast cancer pre-disposition gene present. More recently, other models have been developed to estimate the likelihood that a high-risk breast cancer pre-disposing gene is present in the family. These models vary in complexity of analysis and ease of use. Examples include the Frank model, the BOADICEA model and the Manchester scoring system. Research is currently underway to assess which is the most accurate model for use in the cancer genetics clinic. The Frank model is a logistic regression model developed to estimate BRCA1 and BRCA2 probability based on family history. It was originally based on 238 high-risk women. The BOADICEA model was developed using complex segregation analysis of breast and ovarian cancer in a combined dataset of more than 1500 families. It allows for the simultaneous effects of BRCA1 and BRCA2 as well as the effects of low penetrance genes. It also considers the effect of genetic modifiers that cluster in families and alter the breast cancer risks in BRCA1 and BRCA2 carriers. The recently developed Manchester scoring system was developed using a combination of BRCA1 and BRCA2 screening results from 422 families and data regarding mutation positive and negative kindreds. It was devised to discriminate at the 10 per cent likelihood level, and takes the form of a numerical scoring system that is easy to implement in clinical practice.51–54 The second risk estimation is the chance the individual has inherited a particular gene based upon their position in the family tree, if they are affected by cancer and their current age. This is termed the posterior probability. The risk that the individual in Figure 55.3 may have inherited this gene is therefore half of 60 per cent (the chance there is a cancer pre-disposition gene in the family), i.e., 30 per cent. The final calculation is the chance that cancer will develop. Penetrance estimates are essential to calculate this.
1324 Clinical cancer genetics
Proportion of cases due to a breast cancer predisposition gene
100 Single first degree relative affected Two first degree relatives affected
90 80 70 60 50 40 30 20 10 0 0
10
20
30
40 50 60 Age at diagnosis (years)
Breast cancer x 45
70
80
90
Figure 55.2 Graph showing the probability that breast cancer is due to a predisposition gene by age at diagnosis of breast cancer.50* Graph courtesy of Prof D.T. Bishop.
to consider, and there may be intervening unaffected individuals between affected individuals as demonstrated in Figure 55.5.
Risk perception
Breast cancer x 45
Figure 55.3 A sample family tree demonstrating a family history of breast cancer in two individuals. (Clear symbol, unaffected by cancer; shaded symbol, affected by cancer; stroke-through, deaths Δ, age at diagnosis; consult and indicated by arrow.)
Box 55.1 Standard notation for a family tree Male – square Female – circle ● Deceased – diagonal line through symbol ● Proband – arrow indicates consultand who is giving the family history. There is variation in notation of the shading of the symbols between clinics and, as the shading is not standardized, a legend should be attached to a family tree if referring to a family history in a medical report. ● ●
Using breast cancer as an example, penetrance curves for BRCA1 have been derived from the Breast Cancer Linkage Consortium data set2* (Fig. 55.4). These calculations can be complex, particularly if there are multiple generations
Expressing these risks in a form that is meaningful for the consultand is difficult. The uptake of preventive strategies may depend upon an individual’s perception of risk; for example, Croyle et al.55 have shown that individuals who perceive themselves to be at increased risk of heart disease were more likely to express their intentions to modify their lifestyle than those at perceived population risk. The understanding and retention of this information may depend upon the format in which it is presented and the individual’s attitudes to risk. The expression of this risk can be delivered in a number of formats (Table 55.3). The optimal format for conveying risk information is unknown. Currently, risk estimates tend to be given as a percentage risk or a ‘1 in X’ value and followed up with a written summary, incorporating this risk estimate, to the individual attending the genetics consultation. Unfortunately, there are data which suggest that women prefer not to have, or remember, numerical information. For example, 98 per cent of women attending a cancer family clinic because of a family history of breast cancer could not remember their percentage annual risk, even when this was given both verbally in the clinic, and by follow-up letter. They were somewhat better at remembering their own lifetime risk, but 35 per cent still gave an incorrect figure. More importantly though, they were able to report the qualitative category of their risk (low, medium, high) with reasonable accuracy, but this did not relate to their perception that they were more or less likely to get cancer.56 This suggests that clients have a poor understanding of the risk information being given. Green and Brown57
Cancer risks associated with cancer pre-disposition genes 1325
90 80
Cumulative Risk (%)
70 60 50 40 30 20 10 0 25
30
35
40
45
50 55 Age (years)
60
65
70
75
Figure 55.4 Estimates of breast cancer risk by age in carriers of germline BRCA1/2 mutations [from Breast Cancer Linkage Consortium,].2*
? Ovarian cancer x 65
Breast cancer x 50
Breast cancer x 60
Breast cancer x 45
have suggested that the qualitative aspect of risk is more important than the quantitative aspect. However, this finding contrasts with that of Josten et al.58 who report from a cancer family clinic in Wisconsin, US that ‘clients say that a number gives them boundaries rather than having an ambiguous sense of being high risk’. The main problem
Figure 55.5 Complex family tree involving many individuals affected by cancer, but also intervening unaffected individuals.
with the quantitative approach is that one person’s high risk is another’s moderate risk. The individual’s background information, sociodemographic factors (e.g., educational level) and psychological profile could conceivably alter the optimal method of risk presentation since these can act as barriers to adequate
1326 Clinical cancer genetics
Table 55.3 Methods of presentation of cancer risk estimates Method of presentation Numerical
Expression of risk Risk per year Risk by certain age 1 in x value or percentage format Relative risk corrected for age
General categorization Situation analogy
High/moderate/low risk A situation carrying an equivalent risk without any numerical information e.g. the chances of picking an ace if one card is chosen blind from a card pack
The risk figure measures
Risk of developing cancer Risk of not developing cancer Risk of death from cancer (this is rarely given in clinics as it is perceived as too distressing)
information content.59 In cancer families, it is possible that a larger cancer burden (the number, age at diagnosis, and closeness of relationship of the cancer cases) may distort the perceived risk above the true level.60 Many people in cancer families think erroneously that their risk of developing cancer is 100 per cent, and the only uncertainty is the point in time when the disease will occur. Lerman et al.59 have reported that members of cancer families distort their risk, even when their family history consists of only one affected relative. The points above have been summarized by Vlek61 who claims that there are five factors underlying perception of risk 1. the potential degree of harm or lethality associated
with the risk 2. the controllability through safety/rescue measures (i.e., prevention/early detection) 3. the number of people exposed (this would equate to the cancer burden in the family) 4. the familiarity with the effects of the risk 5. the degree to which exposure to the risk is voluntary. There are reports that suggest that those at highest risk have a lower rate of adoption of health preventive measures due to avoidance behaviours instigated by high levels of anxiety.62 If cancer family clinics are to provide a useful service it is important to ensure that those counselled understand the risk information and advice they are given. Lack of understanding of their risk could impact on their ability to use this information when making decisions about the future management of their health, and may also affect their mental health if cancer-related worries are increased through misunderstanding of information given in the clinic.
MANAGEMENT OF A KNOWN OR SUSPECTED CANCER PRE-DISPOSITION GENE MUTATION CARRIER Identification of an at-risk family A family at genetic risk of cancer must first be identified. There are many potential sources of identification, for example, through consultation with a general practitioner (GP) or a hospital clinic while under treatment for an associated disease, through conversation with an associated professional such a practice or clinic nurse, radiographer or doctor or through an individual’s own perception of a potential genetic problem in their family precipitating contact with a health-care professional. In the hospital oncology setting, if an individual has presented with cancer at a young age, it is now becoming more common for that individual’s family history to be requested by cancer geneticists working as part of the multidisciplinary team co-ordinating the patient’s care. If there is significant family history in addition to the presenting cancer, a referral is made. Sometimes, a referral is made simply because an individual presents with an unusual type of cancer at a young age, for example, medullary thyroid cancer in childhood. In other settings, unless an individual directly expresses concern about their perceived risk, then the only way an at-risk family will be identified is by systematic questioning of all patients about a family history of cancer while eliciting a general medical history. As a quick guideline, taking a history of all first degree relatives only (parents, siblings and children) and then asking if there are any other cancers in the family will detect 95 per cent of familial syndromes. Due to the limited time available during most consultations, it would not be appropriate to obtain a detailed family history from the patient. From this quick family history it should, however, be possible to make an assessment of whether the family history warrants further investigation. Sometimes, the family history may look suspicious but not obviously fall into a specific cancer pre-disposition syndrome. In these cases, it is important to seek further advice from the local Clinical Genetics Department. These departments are mainly located at tertiary referral centres, and they mostly have departmental websites containing detailed referral guidelines. (Box 55.2 shows an example of the current Royal Marsden Hospital NHS Foundation Trust genetics clinic referral guidelines.) Telephone advice is also available from the local Clinical Genetics Department. Referral guidelines are also important as it is not currently possible to see all individuals with a family history of cancer in a cancer genetics clinic due to the limited availability of resources for genetic and cancer services. The recent guidelines published by the National Institute for Clinical Excellence (NICE) (Guideline CG014;63*** provide a pathway of referral for patients with a family history of breast and ovarian cancer, from the primary to the tertiary setting. These guidelines also set out management plans for
Management of a known or suspected cancer pre-disposition gene mutation carrier 1327
Box 55.2 Suggested Referral Guidelines to the Royal Marsden NHS Foundation Trust Familial Cancer Clinic Breast cancer families Moderate risk ● One first degree relative with breast cancer diagnosed under 40 years ● One first degree relative and one second degree relative (or two first degree) relatives diagnosed with breast cancer at any age. ● Two second degree relatives with breast cancer on the father’s side ● Three second degree relatives with breast cancer on the mother’s side ● Any male relative with breast cancer ● One first or second degree relative with breast cancer and one first or second degree relative with ovarian cancer. High risk Two first (or one first and one second) degree relatives diagnosed with breast cancer under 50 years ● Three first degree and second degree relatives diagnosed with breast cancer under 60 years ● Four first, second and third degree relatives diagnosed with breast cancer at any age ● One relative with ovarian cancer and another with breast cancer diagnosed under 50 years ● Two relatives with ovarian cancer and one with breast cancer diagnosed at any age ● One relative with ovarian cancer and two with breast cancer diagnosed under 60 years ● One relative with bilateral breast cancer under the age of 50 years ● One relative with bilateral breast cancer and another relative with breast cancer under the age of 60 years ● A male relative diagnosed with breast cancer at any age and a relative with breast cancer under 50 years or two relatives with breast cancer under 60 years ● A case of both breast cancer and ovarian cancer in the same individual ● Ashkenazi Jewish women affected with breast cancer under 40 years or their close relatives. ●
Li-Fraumeni families/Li-Fraumeni-like families Sarcoma at 45 years with any cancer in a first degree relative at less than 45 years plus another close relative with any cancer at less than 45 years or sarcoma at any age or ● Childhood cancer in an individual with two close relatives with: – Sarcoma, brain tumour, adrenocortical cancer, prostate cancer, melanoma, germ cell tumour, leukaemia at less than 60 years. ●
●
Sarcoma at any age with two of the following in first or second degree relatives: – Breast cancer or stomach cancer at less than 50 years of age – Brain tumour, adrenocortical cancer, prostate cancer, melanoma, germ cell tumour, leukaemia at less than 60 years of age – Sarcoma at any age.
Prostate cancer families Any family with two cases of prostate cancer where one is at less than 65 years of age when diagnosed ● Families with three or more cases of prostate cancer at any age. ●
Colon cancer families Both of the following sets of guidelines/criteria need to be considered. Referral is appropriate if either set of criteria is fulfilled. The Amsterdam criteria are much more specific for HNPCC (hereditary non-polyposis colorectal cancer), however, the Bethesda guidelines are more sensitive. Amsterdam criteria The family is said to meet Amsterdam criteria if the following apply: ● Three or more relatives with HNPCC-associated cancer: colorectal, endometrial, small bowel, ureter or renal pelvis plus all of the following: – One affected patient should be a first degree relative of the other two – Two or more successive generations should be affected – Cancer in one or more affected relatives should be diagnosed before the age of 50 years – Another cancer syndrome called FAP (familial adenomatous polyposis) should be excluded in any case of colorectal cancer – Tumours should be verified by pathological examination. Bethesda guidelines The family fulfils the Bethesda guidelines if it contains: ● An individual diagnosed with colorectal or endometrial cancer before the age of 45 years ● An individual seen with two cancers related to the cancer syndrome HNPCC, such as endometrial, ovarian, gastric, hepatobiliary, small bowel cancer or transitional cell carcinoma of the renal pelvis or ureter, or more than one primary colorectal cancer at the time of diagnosis or another primary diagnosed within 6 months of the first ● An individual with a diagnosis of colorectal cancer and a first degree relative with an HNPCC related cancer (as above), with one of the cancers being diagnosed before the age of 45 years or a colorectal adenoma diagnosed before the age of 40 years ● An individual with colorectal adenomas diagnosed by the age of 40 years.
1328 Clinical cancer genetics
Ovarian cancer families ● Families with two or more cases of ovarian cancer at any age.
but computer software packages are now also under trial to aid risk assessment in primary care (www.cmgp.org.uk).
Genetics clinics Other families ● Any other families with an unusual pattern of cancer in the family or rare syndromes, e.g., testicular or melanoma cancer in two or more relatives at any age or families with rare cancers e.g. MEN2 (multiple endocrine neoplasia). Relatives: First degree – mother, sister, daughter, etc. Second degree – grandmother, aunt, half siblings, etc. Third degree – great grandmother, first cousin, etc.
individuals at increased risk of cancer, but indicate when the risk is not sufficiently high as to warrant referral to a specialist centre. These can be managed in primary or secondary care. As breast cancer is such a common disease, associated with a high level of public anxiety, particularly regarding familial clustering of the disease, it is important to avoid overwhelming the clinical genetics clinics because referrals to the clinics have been precipitated by anxious women with only a moderately increased risk of breast cancer. The NICE guidelines aim to delineate the management and referral pathway according to the Kenilworth model (www.macmillan.org.uk) whereby individuals whose risk does not exceed that of the general population are managed by their GPs, individuals at moderate risk are managed in secondary care, and individuals at high risk are managed in tertiary care. In order to find out more information about a family, to establish their level of risk of a cancer pre-disposition syndrome, it is common to ask the consultand to complete a full family history out to third degree relatives. Often help from other family members has to be requested by the consultand in order to complete this, so this is best undertaken by them in their own time. Using the family history, a doctor may be able to reassure an individual, suggest an increased cancer screening schedule or initiate referral to a genetics clinic for the minority of patients who may benefit from the specialist services of such a genetics clinic. Taking breast cancer as an example, a number of women may qualify for earlier mammographic screening according to the recent NICE guidelines, but do not require specialist genetic services due to the limited family history. A woman with a single first degree relative with breast cancer diagnosed under the age of 40 years or two relatives diagnosed with breast cancer under the age of 60 (but not 50) years would be at increased risk of breast cancer (around 3 population risk) but not at high enough risk to consider genetic testing. The main difficulty for the non-geneticist is making an assessment of the level of genetic risk. Guidelines are helpful,
Following the identification of an at-risk individual, referral to a specialist clinic should be made. Formal assessment of the individual or family risk can be undertaken, and screening and management strategies can be discussed, along with the possibility of genetic testing. AIMS OF GENETICS CLINICS
Cancer genetics counselling aims to provide an explanation of how cancer develops (most commonly as a result of somatic mutation), the principles of genetic inheritance, an estimation of the chance that a familial cluster is due to genetic pre-disposition, information about the likely specific pre-disposition gene present, an estimation of cancer risk, options for managing the risk and the opportunity for genetic testing. STRUCTURE OF THE GENETICS CLINIC
As in many other areas of oncology, the multidisciplinary approach is being increasingly used in cancer genetic counselling clinics. Most of these clinics are located within, or in close association with, the regional clinical genetics service, however, it is desirable that cancer genetic counsellors have training in both genetics and oncology. In the UK, genetic counsellors/clinical nurse specialists in cancer genetics counselling conduct the majority of the routine counselling, working closely with medical personnel who provide medical back up, diagnostic skills where necessary and formal risk assessment. MEDICAL HISTORY AND EXAMINATION
It must be established from the history and examination whether the consultand is an affected or an at-risk member of the family. The consultand should also be questioned on any symptoms indicative of cancer or congenital abnormalities. The presence of congenital abnormalities or developmental delay in other members of the family may also provide an important clue to the diagnosis. It is extremely important to ascertain the ethnic origin of the patient as mutations in some cancer genes are more common in certain populations, particularly when the individual is part of a ‘closed’ population, where founder mutations may be present. The taking of a full family history is central to the practice of the management of familial cancer. The individuals with cancer should be noted in a family tree (Fig. 55.6). The standard notation is given in Box 55.1. Verification of the cancer cases in the family is then considered, as recall of family history may not be accurate. Studies have shown that recall of family history is superior for first degree relatives compared with more distant
Management of a known or suspected cancer pre-disposition gene mutation carrier 1329
Breast cancer x 50
'Abdominal' cancer x 55
Breast cancer x 45
relatives and recall of breast cancer family history is approximately 90 per cent accurate.64 However, accuracy falls for cancer at more indeterminate sites such as ovarian or endometrial cancer. These types of cancer are often misreported, for example as ‘abdominal’ ‘stomach’ or ‘bowel’.64 Verification of diagnosis would therefore be important in individuals such as that with ‘abdominal cancer’ in Figure 55.6, since if this were shown to be ovarian cancer, this family would have a high probability of a mutation in BRCA1/2. Not all cancer genetics clinics verify all breast cancer cases because of the high accuracy of recall and the fact that recall is more likely to be an over-recall, since mastectomy may have been performed for benign disease. This would result in an over-screening of only 5–10 per cent of patients, and it is often not cost effective to verify all breast cancer cases. An exception would be if an individual wishes to undertake more extreme measures such as prophylactic surgery since cases of Munchausen syndrome (where an individual wishes to undergo medical procedures for fictitious reasons) or Munchausen syndrome by proxy (where individuals pass on to other family members a fictitious family history of cancer in order to provoke them to take preventive measures) have been reported, although these are rare (1 per cent).65 Douglas et al.64 showed that verification of all family histories in their cancer genetics clinic resulted in an 11 per cent change in recommendations for management and most of the changes were related to verification of cancers at abdominal sites. Initial clinical examination involves looking for any dysmorphic features and congenital anomalies. The skin should be carefully examined, as many cancer syndromes are associated with dermatological features, such as pigmentary abnormalities, e.g., freckles on the lip in Peutz–Jeghers
Figure 55.6 Sample family tree obtained from family history questionnaire.
syndrome, café au lait patches in neurofibromatosis type I or Turcot’s syndrome, basal cell naevi in Gorlin’s syndrome. Skin tumours, such as the epidermoid cysts seen in FAP (familial adenomatous polyposis), keratoacanthomas seen in Muir–Torre syndrome or trichilemmomas of Cowden syndrome can be indicators that the individual is very likely to be a gene carrier before conformation by formal DNA genetic testing. Throughout the consultation, it is important to be sensitive to any psychopathology that may be occurring. Frequently, there will have been bereavement due to the premature death of close relatives, particularly a parent or child. Unresolved bereavement may make it difficult for people to accept their own risks and make decisions about their own management. Some individuals are particularly worried when they are approaching the age at which their relatives were diagnosed. Others assume that they are more likely to have inherited the cancer pre-disposition gene because they resemble their affected relative, either physically or in temperament. Patients are sometimes unable to cope with their worries, and referral for formal psychological counselling may be needed. Of particular concern are those individuals who have prophylactic surgery because of excess anxiety but who, while being temporarily relieved, could return at a later date with further cancer phobic symptoms. A psychological assessment and counselling should be part of the referral process before prophylactic mastectomy.
Clinical management The subsequent management of an individual and their family will depend upon the final risk estimates regarding
1330 Clinical cancer genetics
the inheritance of a cancer pre-disposition gene and the potential cancer risks associated with this. In general, management strategies fall into four categories, cancer screening, lifestyle changes, preventive strategies and genetic testing. CANCER SCREENING
Cancer screening strategies can be advised for many individuals at increased risk of developing cancer. Table 55.4 outlines guidance protocols for screening in individuals with a high probability of mutations in cancer pre-disposition genes. Not all of the screening schedules described have been proven to reduce mortality from the relevant cancer, but these schedules represent a pragmatic approach to the management of individuals at risk. There is, however, some evidence that screening individuals with HNPCC by colonoscopy reduces mortality due to colorectal cancer, as any suspicious lesions observed on colonoscopy may be removed at an early stage.66* LIFESTYLE CHANGES
Lifestyle changes may take many forms such as the avoidance of known cancer causing factors such as sunlight in Gorlin syndrome and X-ray exposure in the Li-Fraumeni syndrome. Other lifestyle changes are less well established in the prevention of cancer but have been suggested based upon current understanding of tumour biology and a small number of epidemiological studies. Although lifestyle changes may have a greater role in the prevention of sporadic cancers, they may still be of benefit in cancers associated with a genetic pre-disposition. For example, a diet rich in plant phytoestrogens may protect against the development of breast cancer67* or the avoidance of excessive exposure to ovarian hormones may reduce the risk of breast cancer in genetically pre-disposed individuals.68* A high-fibre diet may reduce the risk of colorectal cancer.69*** A European-wide study is currently in progress (CAPP2) to investigate the role of aspirin and/or soluble starch in the prevention of colonic cancer in individuals from HNPCC families. PREVENTION STRATEGIES
Prevention strategies can take many forms including prophylactic surgery and chemoprevention. The evidence in support of the efficacy of these measures is variable, mainly due to rarity of the genetic mutation making clinical trials difficult to perform. Established measures include total colectomy in the familial adenomatous polyposis syndrome70* and total thyroidectomy in the MEN2 syndrome,71* although genotype–phenotype correlations in these two conditions are now able to inform decision-making as to which patients require these drastic measures.72 More contentious roles for prophylactic surgery include mastectomy or oophorectomy in patients with known or suspected BRCA1/2 mutations. Limited retrospective data suggest that
the risk of breast cancer is reduced by 90 per cent following prophylactic mastectomy although there is still a residual risk due to the inability to remove all breast epithelial cells at mastectomy.73* Prophylactic oophorectomy has been shown to reduce ovarian cancer risk74* in addition to a reduction in breast cancer risk.75* Recent papers76*,77* suggest that prophylactic bilateral oophorectomy reduces the incidence of both ovarian and breast cancers in women who are BRCA1/2 mutation carriers. A risk of peritoneal carcinomatosis remains due to the shared embryonic origin of both peritoneum and ovarian epithelium.78 The role of chemoprevention is much less certain, but includes a reduction of ovarian cancer risk in users of the combined oral contraceptive pill.79* Data on the role of tamoxifen in the prevention of breast cancer in high-risk women are conflicting. A large American study (NSABP1) has suggested a 45 per cent reduction in breast cancer risk in women at increased risk who took tamoxifen chemoprevention.80* However, the effect was not replicated by two European studies.81**,82** Tamoxifen also has potentially life-threatening side effects (pulmonary emboli, endometrial cancer, etc.), which has resulted in its use being patchy, even in the USA. More recent evidence, (that for women who are BRCA1/2 gene mutation carriers, tamoxifen significantly reduces the risk of breast cancer in both pre-menopausal and post-menopausal women;83**) has led to the establishment of the STAR trial in the USA, involving tamoxifen and raloxifene, to determine which drug is most effective in preventing breast cancer. In the field of colorectal cancer, there has been demonstration of a reduction in incidence of colonic cancer in long-term users of non-steroidal anti-inflammatory drugs,84* together with evidence regarding the inverse relationship between starch intake and colorectal cancer.85* These two lines of evidence have led to the current study (CAPP2) of aspirin and starch prevention of colonic cancer in HNPCC families. GENETIC TESTING
Genetic testing is possible for most cancer pre-disposition genes (Table 55.5) and is performed on DNA from venous blood. Genetic testing may either be diagnostic (the detection of a mutation in an individual affected by cancer) or predictive (the detection of a mutation in a clinically unaffected individual). Mutations in cancer pre-disposition genes often occur throughout the gene and the vast majority of mutations so far have only been observed in limited numbers of families, except in specific ethnic groups with known founder mutations such as the Icelandic86 and Ashkenazi87 populations with BRCA1/2 mutations. Hence, unless an individual is a member of such a group, the specific mutation for that family must first be identified. An affected family member is tested first because they are the family member most likely to have the cancer pre-disposing mutation. Once a mutation is suspected, it is important to check that the ‘mutation’ is likely to be cancer causing and
Management of a known or suspected cancer pre-disposition gene mutation carrier 1331
Table 55.4 Screening in cancer genetic syndromes Disease von Hippel–Lindau (affected)
von Hippel–Lindau (at risk relative)
Familial adenomatous polyposis (affected)
Screening/management
Frequency
Age at start of screening/age range for screening
Physical examination Urine testing Direct ophthalmoscopy/ fluorescein angiography 24-hr urinary VMA/plasma catecholamines Abdominal (renal) ultrasound MRI brain CT kidneys
Annual Annual Annual
From diagnosis From diagnosis From diagnosis
Annual
From diagnosis
Annual 3 yearly If renal cysts present
From diagnosis From diagnosis
Annual Annual Annual Annual
5 years upwards 5 years upwards 5 years upwards 10–60 years 11 years upwards
CT kidneys
Annual 3 yearly 5 yearly If renal cysts present
20–65 years 15–40 years 40–60 years 20–65 years
Colonoscopy
Annual
11 years upwards
Annual
Average 16–18 years (depending on extent of polyposis) Following colectomy
Physical examination Urine testing Direct ophthalmoscopy/ fluorescein angiography 24-hr urinary VMA/plasma catecholamines Abdominal (renal) ultrasound MRI brain
Offer total colectomy with ileo-rectal anastomosis Rectal stump screening (if conserved in surgery) Upper gastrointestinal endoscopy Familial adenomatous polyposis (at risk)
Gorlin’s syndrome (affected) (at-risk children usually have abnormal skull or spine X-rays by 5 years)
Multiple endocrine neoplasia type 2 (affected)
Frequency depends on extent of polyposis found
20 years upwards
Colonoscopy Offer total colectomy with ileo-rectal anastomosis
Annual
11 years upwards If polyposis extensive
Dermatological examination
Annual 3 monthly
Infancy upwards Puberty onwards
Orthopantomogram (for jaw cysts) ?MRI brain (not CT due to radiosensitivity)
Annual
From diagnosis
Annual
In infancy only
Annual
By age 2 years if indicated by genotype 8–70 years
Annual Annual Annual Annual
8–70 years 8–70 years 8–70 years 8–70 years
Offer genetic testing if possible
Offer genetic testing if possible Prophylactic thyroidectomy Plasma calcium, phosphate, parathormone Pentagastrin test Thyroid ultrasound Abdominal ultrasound and CT 24-hr urinary VMA/plasma catecholamines
1332 Clinical cancer genetics
Table 55.4 (Continued) Disease
Age at start of screening/age range for screening
Screening/management
Frequency
Symptom enquiry (dyspepsia, diarrhoea, renal colic, fits, amenorrhoea, galactorrhoea) and physical examination Serum calcium, parathormone Fasting gastrin Fasting serum glucose, insulin, proinsulin Pancreatic polypeptide, VIP, glucagon Serum prolactin, IGF-1 CT thorax and abdomen MRI brain
Annual
5 years upwards
Annual Annual Annual
8 years upwards 20 years upwards 5 years upwards
Annual
20 years upwards
Annual 2–3 yearly 3 yearly
5 years upwards 20 years upwards 5 years upwards
Wilms’ tumour (at-risk individuals)
Renal ultrasound
3–4 monthly
Birth until 7–8 years (depending on syndromic diagnosis)
Li-Fraumeni syndrome
Breast examination Breast MRI Physical examination
Annual Annual Annual
18–60 years 20–60 years Lifelong
Retinoblastoma (siblings and offspring of affected)
Offer genetic testing if possible Monthly
Birth to 3 months
3 monthly 4 monthly
3 months to 2 years 2–3 years
6 monthly
3–5 years
Annual
5–11 years
Multiple endocrine neoplasia type 1
Retinal examination (without anaesthetic) Retinal examination under anaesthetic Retinal examination under anaesthetic Retinal examination (without anaesthetic) Retinal examination (without anaesthetic) Examination for sarcoma
Annual
Early teens to lifelong
Neurofibromatosis type 1
Physical examination Blood pressure measurement Visual field assessment
Annual Annual Annual
Lifelong Lifelong Birth to 6 years
Neurofibromatosis type 2
Physical examination Ophthalmological assessment (for congenital cataracts) Audiometry Brainstem auditory evoked potentials MRI brain and spine MRI brain and spine
Annual Once
Early childhood
Annual Annual Annual 2 yearly
10–40 years 10–40 years Until age 30 years Until age 40 years, then 3 yearly
Colonoscopy
2 yearly
25 years upwards
Ovarian screening (CA 125 and TV USS) Endometrial screening Consider urine cytology/ mammography (depending on cancers in family)
Annual at least
35 years upwards
Annual Annual
35 years upwards 35 years upwards
Hereditary non-polyposis colorectal cancer
Management of a known or suspected cancer pre-disposition gene mutation carrier 1333
Table 55.4 (Continued) Screening/management
Frequency
Age at start of screening/age range for screening
Familial melanoma
Skin examination
Annual
Teens upwards
Familial breast/ovarian cancer
Mammography
Annual
Mammography Ovarian screening (CA 125 and TV USS) Consider PSA and prostate biopsy
18 monthly Annual at least
35–50 years (or 5 years younger than youngest case if 35 years) 50–70 years 35 years upwards
Annual
40 years upwards
Disease
CT, computed tomography; IGF-1, insulin-like growth factor; MRI, magnetic resonance imaging; PSA, prostate serum antigen; TV USS, transvaginal ultrasound; VMA, Vanillyl mandelic acid.
Table 55.5 Location of and status of testing for cancer predisposition genes in 2006
Disease Familial breast/ovarian cancer Familial breast/ovarian cancer von Hippel–Lindau Familial adenomatous polyposis MYH-associated polyposis Gorlin’s syndrome PTEN hamartoma tumour syndrome (Cowden syndrome) Multiple endocrine neoplasia type 2 Multiple endocrine neoplasia type 1 Wilms’ tumour Retinoblastoma Li-Fraumeni syndrome Neurofibromatosis I Neurofibromatosis II Hereditary non polyposis colorectal cancer (HNPCC)
Location
Mutation analysis available
17q21 13q12 3p25 5q21 1p34 9q22 10q23
BRCA1 BRCA2 VHL APC MYH PTCH PTEN
10q11 11q13 11p13 13q14 17p13 17q11 22q12 2p22
RET MEN1 WT1 RB1 TP53 NF1 NF2 hMLH1
3p21 2p16 2q31 7p22
hMSH2 MSH6 PMS1 PMS2
not a rare normal variant of the gene (polymorphism). Occasionally, the significance of a mutation is uncertain, and its pathogenicity may be then further investigated by testing other affected and unaffected family members to establish whether or not the mutation segregates with disease. The presence of phenocopies may complicate this issue further. If a pathogenic mutation is identified,
however, predictive testing may be offered to unaffected family members for the identified mutation. Misleading results may occur if an unaffected individual has a genetic test in order to identify a mutation without first identifying it in an affected relative. A negative result (i.e., no mutation is identified in the cancer predisposition gene tested) may not be a true negative for several reasons: ●
●
●
The family history is due to a gene other than that being tested. The alteration may be regulatory which means that it controls how the gene is expressed but the gene itself (and therefore the test which looks at the gene code) is normal. The genetic test sensitivity is not 100 per cent for the genetic coding mutations and may therefore have missed mutations.
When the specific mutation has been identified in an affected individual, if it is not found in an unaffected relative, this is then a truly negative result. Unfortunately, due to the high penetrance of certain pre-disposition genes, for example, BRCA1/2, it is not uncommon to be presented with a family with multiple cases of cancer but all the affected individuals have died from their cancer. In this difficult situation it is not usually possible to offer mutation testing to unaffected cases, and their future clinical management will depend upon their probability of having inherited a mutation based on their family history. An exception is the rare circumstance of a closed ethnic group (e.g., the Ashkenazim) in which the suspected mutations are known to occur with a high frequency in that population. Commercial companies exist that will offer genetic testing to unaffected individuals where there is no living affected person in the family, but for the reasons outlined above, this is fraught with difficulty.
1334 Clinical cancer genetics
Genetic testing should only take place following full genetic counselling to outline the implications of genetic testing. A recognized counselling schedule allows at least a month of reflection between two counselling sessions prior to taking blood for mutation analysis. The personal and wider social implications of positive and negative results are issues discussed during these sessions. A positive result could have psychological implications as well as widespread repercussions involving the rest of the family. A negative test result may have psychological consequences due to the recognized ‘survivor guilt syndrome’, which has been documented in the setting of Huntington’s disease.88 The social implications of the ability to purchase life and medical insurance or mortgages and a possible effect on employment opportunities may be just as important as, if not more than, the personal and familial implications. At present, effects on employment are theoretical. Following a recent statement from the Association of British Insurers in March 2005, an individual may apply for a total of up to £500 000 of life insurance, £300 000 of critical illness and £30 000 annual benefit of income protection insurance without having to disclose to the insurer the results of any predictive genetic test previously taken.89 This moratorium is currently scheduled to remain in place until 2011, at least. For genes pre-disposing to adult onset cancers, testing of young children is not advised as the age of cancer onset permits the individual to make their own decision to have genetic testing once they have reached adulthood, following full genetic counselling. Children are offered genetic testing when it may alter management, for example, in MEN2A syndrome when thyroidectomy is offered before the age of 5 years for some gene carriers as it is totally protective against medullary thyroid cancer; or in familial adenomatous polyposis where regular colonoscopies or colectomy may be avoided.
SUMMARY Cancer is a common disease but only a small proportion of cases can be attributed to the inheritance of specific cancer pre-disposition genes. However, in absolute terms, this represents a significant number of families or individuals due to the high population frequency of cancer. Only a few cancer pre-disposition genes have been identified although the existence of many more is suspected. Currently, we can offer limited advice regarding risks, genetic testing and general management of suspected gene carriers, but cancer genetics is one of the most rapidly expanding areas of oncological knowledge. Any advice given regarding cancer pre-disposition genes may be liable to change with advances in research and knowledge, and specialists in the field of cancer genetics should facilitate the dissemination of any changes in advice to other colleagues likely to be in contact with individuals at risk.
KEY LEARNING POINTS ●
●
● ●
●
The majority of cancer cases are sporadic and not related to the inheritance of a cancer predisposition gene. Most cancer families who may benefit from referral to a specialist cancer genetics unit will be detected by taking a limited family history. This should include all first degree relatives (parents/siblings/children) and any other cancer cases in other relatives. Cancer site and age of onset are important. Early cancer screening may be available for specific cancer pre-disposition syndromes. Genetic testing for the presence of a familial cancer pre-disposition gene can usually only be performed by direct DNA analysis using venous blood from a living relative affected by cancer.
KEY REFERENCES Eeles RA, Ponder BAJ, Easton DF, Eng C (eds). Genetic Predisposition to Cancer. London: Arnold, 2004. Morrison PJ, Hodgson SV, Haites NE (eds). Familial Breast Cancer: Genetics, Screening and Management. Cambridge University Press, 2002.
REFERENCES 1 Easton DF, Peto J. The contribution of inherited predisposition to cancer incidence. Cancer Surveys 1990; 9:395–416. ●2 Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium. American Journal of Human Genetics 1998; 62:676–689. 3 Cannon-Albright LA, Meyer LJ, Goldgar DE, et al. Penetrance and expressivity of the chromosome 9p melanoma susceptibility locus (MLM). Cancer Research 1994; 54:6041–6044. ●4 Bodmer WF, Bailey CJ, Bodmer J, et al. Localization of the gene for familial adenomatous polyposis on chromosome 5. Nature 1987; 328:614–616. ●5 Miki Y, Swensen J, Shattuck-Eidens D, et al. Isolation of BRCA1, the 17q-linked breast and ovarian cancer susceptibility gene. Science 1994; 266:66–71. ●6 Wooster R, Bignell G, Lancaster J, et al. Identification of the breast cancer susceptibility gene, BRCA2. Nature 1995; 378:789–792. ◆7 Lynch HT and Lynch JF 25 years of HNPCC. Anticancer Research 1994; 14:1617–1624.
References 1335
8 Papadopoulos N, Nicolaides NC, Liu B, et al. Mutations of GTBP in genetically unstable cells. Science 1995; 268:1915–1917. ●9 Nicolaides NC, Papadopoulos N, Liu B, et al. Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 1994; 371:75–80. 10 Hall NR, Murday VA, Chapman P, et al. Genetic linkage in Muir-Torre syndrome to the same chromosomal region as cancer family syndrome. European Journal of Cancer 1994; 30A:180–182. 11 Itoh H, Hirata K, Ohsato K. Turcot’s syndrome and familial adenomatous polyposis associated with brain tumour: review of related literature. International Journal of Colorectal Disease 1993; 8:87–94. ●12 Sieber OM, Lipton L, Crabtree M, et al. Multiple colorectal adenomas, classic adenomatous polyposis, and germ-line mutations in MYH. New England Journal of Medicine 2003; 348:791–799. ◆13 Edwards SM, Eeles RA. Unravelling the genetics of prostate cancer. American Journal of Medical Genetics C 2004; 129:65–73. 14 Edwards SM, Kote-Jarai Z, Meitz J, et al. Two percent of men with early-onset prostate cancer harbor germline mutations in the BRCA2 gene. American Journal of Human Genetics 2003; 72:10–12. ●15 Malkin D, Li FP, Strong LC, et al. Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas and other neoplasms. Science 1990; 250:1233–1238. ●16 Thakker RV, Bouloux P, Wooding C, et al. Association of parathyroid tumours in multiple endocrine neoplasia type 1 with loss of alleles on chromosome 11. New England Journal of Medicine 1989; 321:218–224. ●17 Ponder BAJ, Smith D. The MEN2 syndromes and the role of the RET proto-oncogene. Advances in Cancer Research 1996; 70:179–222. 18 Draper GJ, Sanders BM, Brownbill PA, Hawkins MM. Patterns of risk of hereditary retinoblastoma and applications to genetic counselling. British Journal of Cancer 1992; 66:211–219. ◆19 Huson SM, Compston DA, Harper PS. A genetic study of von Recklinghausen neurofibromatosis in south east Wales. Journal of Medical Genetics 1989; 26:712–721. ◆20 Evans DGR. Neurofibromatosis type 2: genetic and clinical features. Ear Nose Throat Journal 1999; 78:97–100. ◆21 Maher ER, Webster AR, Moore AT. Clinical features and molecular genetics of von Hippel-Lindau disease. Ophthalmic Genetics 1995; 16:79–84. ◆22 Johnson JA. Ataxia telangiectasia and other alphafetoprotein-associated disorders. In: Lynch, HT, Hirayama T (Eds.): Genetic Epidemiology of Cancer. Boca Raton: CRC Press 1989,145–147. 23 Lo Ten Foe JR, Rooimans MA, Bosnoyan-Collins L, et al. Expression Cloning of a cDNA for the major Fanconi anaemia gene FAA. Nature Genetics 1996; 14:320–323. 24 Strathdee CA, Gavish H., Shannon WR, et al. Cloning of cDNA’s for Fanconi’s anaemia by functional complementation. Nature 1992; 358:763–767.
25 Howlett NG, Taniguchi T, Olson S, et al. Biallelic inactivation of BRCA2 in Fanconi anaemia. Science 2002; 297:606–609. 26 Whitney M, Thayer M, Reifsteck C, et al. Microcell mediated chromosome transfer maps the Fanconi anaemia group D gene to chromosome 3p. Nature Genetics 1995; 11:341–343. 27 Waisfisz Q, Saar K, Morgan NV, et al. The Fanconi anaemia group E gene, FANCE, maps to chromosome 6p. American Journal of Human Genetics 1999; 64:1400–1405. 28 De Winter JP, Rooimans MA, van der Weel L, et al. The Fanconi anaemia gene FANCF encodes a novel protein with homology to ROM. Nature Genetics 2000; 24:15–16. 29 De Winter JP, Waisfisz Q, Rooimans MA, et al. The Fanconi anaemia group G gene FANCG is identical with XRCC9. Nature Genetics 1998; 20:281–283. 30 Levran O, Attwooll C, Henry RT, et al. The BRCA1-interacting helicase BRIP1 is deficient in Fanconi anaemia. Nature Genetics 2005; 37:931–933. 31 Meetei AR, de Winter JP, Medhurst AL, et al. A novel ubiquitin ligase is deficient in Fanconi anaemia. Nature Genetics 2003; 35:165–170. 32 Meetei AR, Medhurst AL, Ling C, et al. A human ortholog of archeal DNA repair protein Hef is deficient in Fanconi anaemia complementation group M. Nature Genetics 2005; 37:958–963. ●33 Ellis NA, German J. Molecular genetics of Bloom’s syndrome. Human Molecular Genetics 1996; 5:1457–1463. 34 Tanaka K, Satokata I, Ogita Z, et al. Molecular cloning of a mouse DNA repair gene that complements the defect of group-A xeroderma pigmentosum. Proceedings of the National Academy of Sciences 1989; 86:5512–5516. 35 Weeda G, van Ham RCA, Vermeulen W, et al. A presumed DNA helicase encoded by ERCC-3 is involved in the human repair disorders xeroderma pigmentosum and Cockayne’s syndrome. Cell 1990; 62:777–791. 36 Li L, Bales ES, Peterson CA, et al. Characterization of molecular defects in xeroderma pigmentosum group C. Nature Genetics. 1993; 5:413–417. 37 Sung P, Bailly V, Weber C, et al. Human xeroderma pigmentosum group D gene encodes a DNA helicase. Nature 1993; 365:852–855. 38 Nichols A F, Ong P, Linn S. Mutations specific to the xeroderma pigmentosum group E Ddb- phenotype. Journal of Biological Chemistry 1996; 271:24317–24320. 39 Sijbers AM, de Laat WL, Ariza RR, et al. Xeroderma pigmentosum group F caused by a defect in a structure-specific DNA repair endonuclease. Cell 1996; 86:811–822. 40 Shiomi T, Harada Y, Saito T, et al. An ERCC5 gene with homology to yeast RAD2 is involved in group G xeroderma pigmentosum. Mutation Research 1994; 314:167–175. 41 Masutani C, Kusumoto R, Yamada A, et al. The XPV (xeroderma pigmentosum variant) gene encodes human DNA polymerase eta. Nature 1999; 399:700–704. ◆42 Eng C. Genetics of Cowden’s syndrome: through the looking glass of oncology. International Journal of Cancer 1998; 12:701–710.
1336 Clinical cancer genetics
◆43
44
●45
46
47
48
49
●50
51
52
53
54
55
56
57 58
59
Kimonis VE, Goldstein AM, Pastakia B, et al. Clinical manifestations in 105 persons with nevoid basal cell carcinoma syndrome. American Journal of Medical Genetics 1997; 69:299–308. Xu J, Meyers D, Freije D, et al. Evidence for a prostate cancer susceptibility locus on the X-chromosome. Nature Genetics 1998; 20:175–179. Knudson AG. Mutation and Cancer: statistical study of retinoblastoma. Proceedings of the National Academy of Sciences 1971; 68:820–823. Riccardi VM, Sujansky E, Smith AC, et al. Chromosomal imbalance in the Aniridia-Wilms’ tumour association: 11p interstitial deletion. Pediatrics 1978; 61:604–610. Platz EA, Giovannucci E, Dahl DM, et al. The androgen receptor gene GGN microsatellite and prostate cancer risk. Cancer Epidemiology, Biomarkers and Prevention 1998; 7:379–384. Struewing JP, Hartge P, Wacholder S, et al. The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. New England Journal of Medicine 1997; 336:1401–1408. Thorlacius S, Struewing JP, Hartge P, et al. Population based study of risk of breast cancer in carriers of BRCA2 mutations. Lancet 1998; 352:1337–1339. Claus EB, Risch NJ, Thompson WD. Genetic analysis of breast cancer in the cancer and steroid hormone study. American Journal of Human Genetics 1991; 48:232–242. Frank TS, Manley SA, Olopade OI, et al. Sequence analysis of BRCA1 and BRCA2: correlation of mutations with family history and ovarian cancer risk. Journal of Clinical Oncology 1998; 16:2417–25. Antoniou AC, Pharoah PDP, McMullen G, et al. A comprehensive model for familial breast cancer incorporating BRCA1, BRCA2 and other genes. British Journal of Cancer 2002; 86:76–83. Evans DGR, Eccles DM, Rahman N, et al. A new scoring system for the chances of identifying a BRCA1/2 mutation, outperforms existing models including BRCAPRO. Journal of Medical Genetics 2004; 41:474–480. Tyrer J, Duffy SW, Cuzick J. A breast cancer prediction model incorporating familial and personal risk factors. Stat Med 200; 23:1111–1130. Croyle RT, Sun YC, Louie DH. Psychological minimization of cholesterol test results: moderators of appraisal in college students and community residents. Health Psychology 1993; 12:503–507. Lloyd S, Watson, M, Waites B, et al. Familial breast cancer: a controlled study of risk perception, psychological morbidity and health beliefs in women attending for genetic counselling. British Journal of Cancer 1996; 74:482–487. Green H, Brown RA. Counting lives. Journal of Occupational Accidents 1978; 2:55. Josten DM, Evans AM, Love RR. The cancer prevention clinic: a service program for cancer-prone families. Journal of Psychosocial Oncology 1985; 3:5–20. Lerman C, Daly M, Masny A, Balshem A. Attitudes about genetic testing for breast-ovarian cancer
60
61
62
❋
63 64
65 ❋
67
68
69
70
●71
❋
72
●73
74
75
susceptibility. Journal of Clinical Oncology 1994; 12:843–850. Ardern-Jones A. Living with a cancer legacy: the experience of hereditary cancer in the family. Institute of Cancer Research, London University. MSc Dissertation. 1998. Vlek C. Risk assessment, risk perception and decision making about courses of action involving genetic risk: an overview of concepts and methods. Birth Defects Original Article Series 1987; 23:171–207. Kash KM, Holland JC, Halper MS, Miller DG. Psychological distress and surveillance behaviours of women with a family history of breast cancer. Journal of the National Cancer Institute 1991; 84:24–30. National Institute for Clinical Excellence (UK) Guidelines on Familial Breast Cancer (2004), www.nice.org.uk. Douglas FS, O’Dair LC, Robinson M, et al. The accuracy of diagnoses as reported in families with cancer: a retrospective study. Journal of Medical Genetics 1999; 36:309–312. Evans DG, Kerr B, Cade D, et al. Fictitious breast cancer family history. Lancet 1996; 348:1034. Dove-Edwin I, Sasieni P, Adams J, Thomas HJ. Prevention of cancer by colonoscopic surveillance in individuals with a family history of colorectal cancer: 16 year, prospective, follow-up study. British Medical Journal 2005; 331:1047–1050. Ingram D, Sanders K, Kolybaba M, Lopez D. Case-control study of phyto-oestrogens and breast cancer. Lancet 1997; 350:990–994. Ursin G, Henderson BE, Haile RW, et al. Does oral contraceptive use increase the risk of breast cancer in women with BRCA1/BRCA2 mutations more than in other women? Cancer Research 1997; 57:3678–3681. Howe GR, Benito E, Castelleto R, et al. Dietary intake of fiber and decreased risk of cancers of the colon and rectum: evidence from the combined analysis of 13 case-control studies. Journal of the National Cancer Institute 1992; 84:1887–1896. Nyam DC, Brillant PT, Dozois RR, et al. Ileal pouch-anal canal anastomosis for familial adenomatous polyposis: early and late results. Annals of Surgery 1997; 226:514–519. Lallier M, St-Vil D, Giroux M, et al. Prophylactic thyroidectomy for medullary thyroid cancer in gene carriers of MEN2 syndrome. Journal of Pediatric Surgery 1998; 33:846–848. Brandi M, Gagel RF, Angeli A, et al. Guidelines for diagnosis and therapy of MEN type 1 and type 2. Journal of Clinical Endocrinology and Metabolism 2001; 86:5658–5671. Hartmann LC. Efficacy of bilateral prophylactic mastectomy in women with a family history of breast cancer. New England Journal of Medicine 1999; 340:77–84. Struewing JP, Watson P, Easton DF, et al. Prophylactic oophorectomy in inherited breast/ovarian cancer families. Journal of the National Cancer Institute Monographs 1995; 17:33–35. Rebbeck TR, Levin AM, Eisen A, et al. Breast cancer risk after bilateral prophylactic oophorectomy in BRCA1 mutation
References 1337
76
77
78
79
80
81
82
carriers. Journal National Cancer Institute 1999; 91:1475–1479. Kauff ND, Satagopan JM, Robson ME, et al. Risk-reducing salpingo-oophorectomy in women with a BRCA1 or BRCA2 mutation. New England Journal of Medicine 2002; 346:1609–1615. Rebbeck TR, Lynch HT, Neuhausen SL, et al. Prophylactic oophorectomy in carriers of BRCA1 or BRCA2 mutations. New England Journal of Medicine 2002; 346:1616–1622. Kemp GM, Hsiu JG, Andrews MC Papillary peritoneal carcinomatosis after prophylactic oophorectomy. Gynecol Oncol 1992; 47:395–397. Narod SA, Risch H, Moslchi R, et al. Oral contraceptives and the risk of hereditary ovarian cancer. Hereditary Ovarian Cancer Clinical Study Group. New England Journal of Medicine 1998; 339:424–428. Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for prevention of breast cancer: Report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. Journal of the National Cancer Institute 1998; 90:1371–1388. Powles T, Eeles R, Ashley S, et al. Interim analysis of the incidence of breast cancer in the Royal Marsden Hospital tamoxifen randomised chemoprevention trial. Lancet 1998; 352:98–101. Veronesi U, Maisonneuve P, Costa A, et al. Prevention of breast cancer with tamoxifen: preliminary findings from the
83
84
85
86
87
88
89
Italian randomised trial among hysterectomised women. Lancet 1998; 352:93–97. King MC, Wieand S, Hale, et al. Tamoxifen and breast cancer incidence among women with inherited mutations in BRCA1 and BRCA2: National Surgical Adjuvant Breast and Bowel Project (NSABP-P1) Breast Cancer prevention Trial. Journal of the American Medical Association 2001; 286:2251–2256. Giovannucci E, Rimm EB, Stampfer MJ, et al. Aspirin use and the risk for colorectal cancer in male health professionals. Annals of Internal Medicine 1994; 121:241–246. Cassidy A, Bingham SA, Cummings JH. Starch intake and colorectal cancer risk: an international comparison. British Journal of Cancer 1994; 69:937–942. Thorlacius S, Sigurdsson S, Bjarnadottir H, et al. Study of a single BRCA2 mutation with high carrier frequency in a small population. American Journal of Human Genetics 1997; 60:1079–1084. Tonin P, Weber BL, Offit K, et al. Frequency of recurrent BRCA1 and BRCA2 mutations in Ashkenazi Jewish breast cancer families. Nature Medicine 1996; 2:1179–1183. Demyttenaere K, Evers-Kiehooms G, Decruyenaere M. Pitfalls in counselling for predictive testing in Huntington’s disease. Birth Defects: Original Article Series 1992; 28:105–111. Association of British Insurers. Concordat and Moratorium on Genetics and Insurance. Declaration by Association of British Insurers, March 2005, www.abi.org.uk.
56 Large-scale randomized evidence: trials and overviews† RICHARD GRAY, RORY COLLINS, RICHARD PETO AND KEITH WHEATLEY
Introduction and summary Avoiding moderate biases Avoiding moderate random errors Minimizing both bias and random error: systematic overviews (meta-analyses) of randomized trials Some examples of important results in the treatment of vascular and of neoplastic disease that could have been
1338 1340 1342 1345
INTRODUCTION AND SUMMARY This chapter is intended principally for clinicians who need to use the results of clinical grails in their routine practice, and who want to know why some types of evidence are much more reliable than others. It is concerned with treatments that might improve survival (or some other really major aspect of long-term disease outcome), and its chief point is that as long as doctors start with a healthy scepticism about the many apparently striking claims that appear in the medical literature, large-scale trials do make sense. The main enemy of common sense is overoptimism: there are a few striking exceptions of treatments for serious disease that really do turn out to work extremely well, but in general most of the claims of vast improvements from new therapies turn out to be evanescent. Hence, clinical trials need to be able to detect or to refute really reliably the more moderate differences in long-term outcome that it is medically realistic to expect. Once this common-sense idea is explicitly recognized, the rest follows naturally, and it becomes fairly obvious what types of evidence can and cannot be trusted. Although this chapter may also be of some interest or encouragement to doctors who are considering participating in (or even planning) †
reliably established only by large-scale randomized evidence Results from large, anonymous trials are relevant to real clinical practice Key references References
1346 1350 1351 1351
some large trial, its main intended audience is the practising clinician, for even the most definite results from largescale randomized evidence cannot save lives unless practising clinicians accept and apply them. The chapter does not go into a lot of statistical details: instead, it tries to communicate the spirit that underlies the increasing emphasis over the past decade or so on large-scale randomized evidence. Unrealistic hopes about the chances of discovering big treatment effects can be a serious obstacle, not only to appropriate patient care but also to good clinical research, for such hopes may misleadingly suggest to some research workers or funding agencies that small, or even nonrandomized, studies may suffice. In contrast, realistically moderate expectations of what treatment might achieve (or, if one treatment is to be compared with another, realistically moderate expectations of how large any difference between those treatments is likely to be) should, in contrast, tend to foster the design of studies that aim to discriminate reliably between, first, differences in outcome that are realistically moderate but still worthwhile, and, second, differences in outcome that are too small to be of any material importance.1 Studies with this particular aim must guarantee strict control of bias (which, in general, requires proper randomization
Adapted from Collins R, Peto R, Gray R, Parish S. Large-scale randomised evidence: trials and overviews. In: Weatherall D, Ledingham JGG, Warrell DA (eds) Oxford Textbook of Medicine, Vol. 1. Oxford: Oxford University Press, 1996.
Introduction and summary 1339
and appropriate statistical analysis, with no unduly ‘datadependent’ emphasis on specific parts of the overall evidence), and must guarantee strict control of the play of chance (which, in general, requires large numbers rather than a lot of detail). The conclusion is obvious: moderate biases and moderate random errors must both be avoided if moderate benefits are to be assessed or refuted reliably. This leads to the need for large numbers of properly randomized patients, which in turn leads both to large simple randomized trials (or ‘mega-trials’) and to large systematic overviews (or ‘meta-analyses’) of related randomized trials.2 Non-randomized evidence, unduly small randomized trials or unduly small overviews of trials are all much inferior as sources of evidence about current patient management or as foundations for future research strategies, for they cannot discriminate reliably between moderate (but worthwhile) differences and negligible differences in outcome, and the mistaken clinical conclusions that they engender could well result in the world-wide under-treatment, over-treatment or other mismanagement of millions of future patients. In contrast, hundreds of thousands of premature deaths a year could be avoided by seeking appropriately large-scale randomized evidence about various widely practicable treatments for the common conditions leading to death, and by disseminating such evidence appropriately. Likewise, appropriately large-scale randomized evidence could substantially improve the management of many important, but non-fatal, medical problems. The value of large-scale randomized evidence is illustrated in this chapter by the trials of fibrinolytic therapy for acute myocardial infarction, of anti-platelet therapy for a wide range of vascular conditions, and of hormonal therapy for early breast cancer. In these examples, proof of benefit that could not have been achieved either by smallscale randomized evidence or by non-randomized evidence has led to widespread changes in practice that are now preventing tens of thousands of premature deaths each year.
Moderate (but worthwhile) effects on major outcomes are generally more plausible than large effects Some treatments do have large, and hence obvious, effects on survival – for example, it is clear without randomized trials that insulin treatment of diabetic coma or surgery for breast and colorectal cancer saves lives. But, perhaps in part due to these striking successes, in the past few decades the hopes of large treatment effects on mortality and major morbidity in other serious diseases have been unrealistically high. Of course, treatments do quite commonly have large effects on various less fundamental measures: for example, many tumours can be shrunk temporarily by radiotherapy or chemotherapy; giving 5-fluorouracil by continuous infusion rather than bolus, or in combination
with folinic acid, can double the number of responding patients; in acute myocardial infarction, tissue plasminogen activator dissolves coronary thrombi more rapidly than streptokinase; drugs readily reduce blood pressure, blood lipids and blood glucose. But, although all of these effects are large, any effects on mortality are much more modest – indeed, there is still dispute as to whether any net improvement in survival results from these interventions. In general, if substantial uncertainty remains about the efficacy of a practicable treatment, its effects on major end points are probably either negligibly small, or only moderate, rather than large. Indirect support for this rather pessimistic conclusion comes from many sources, including: 1. the previous few decades of disappointingly slow progress in the curative treatment of cancer and the common chronic diseases of middle age; 2. the heterogeneity of each single disease, as evidenced by the unpredictability of survival duration even when apparently similar patients are compared with each other; 3. the variety of different mechanisms in certain diseases that can lead to death, only one of which may be appreciably influenced by any one particular therapy; and 4. the modest effects often suggested by systematic overviews (or ‘meta-analyses’; see later) of various therapies. Having accepted that, with many currently available interventions, only moderate reductions in mortality are plausible, how worthwhile might such effects be, if they could be detected reliably? To some clinicians, reducing the risk of death from breast cancer from 50 per 100 women down to 45 per 100 women treated may not seem particularly worthwhile – and, indeed, if such a reduction was only transient, or involved an extremely expensive or toxic treatment, this might well be an appropriate view. World-wide, however, 800 000 women a year are diagnosed with breast cancer, and if just half of these women were to be given a simple, non-toxic, widely practicable treatment that reduced the risk of death from 50 per cent down to 45 per cent (i.e. a 10 per cent proportional reduction), then this would avoid 10 000 or 20 000 deaths. For example, more than two million women world-wide are currently taking tamoxifen as adjuvant therapy of breast cancer, and this must be avoiding tens of thousands of deaths each year (see below). Such absolute gains are substantial – and might, indeed, considerably exceed the numbers of lives that could be saved by a much more effective treatment for a less common disease.
Reliable detection or refutation of moderate differences requires avoidance both of moderate biases and moderate random errors If realistically moderate differences in outcome are to be reliably detected or reliably refuted, then the errors in comparative assessments of the effects of treatment need to be
1340 Large-scale randomized evidence: trials and overviews
Table 56.1 Requirements for reliable assessment of moderate effects: negligible biases and small random errors NEGLIGIBLE BIASES (i.e. GUARANTEED AVOIDANCE OF MODERATE BIASES) Proper randomization (non-randomized methods might suffer moderate biases) Analysis by allocated treatment (including all randomized patients: ‘intention-to-treat’ analysis) Chief emphasis on overall results (no unduly data-dependent emphasis on particular subgroups) Systematic overview of all relevant randomized trials (no unduly data-dependent emphasis on particular studies) SMALL RANDOM ERRORS (i.e. GUARANTEED AVOIDANCE OF MODERATE RANDOM ERRORS) Large numbers in any new trials (to be really large, trials should be ‘streamlined’) Systematic overviews of all relevant randomized trials (which yields the largest possible total numbers)
much smaller than the difference between a moderate but worthwhile effect on the one hand, and, on the other, an effect that is too small to be of any material importance. This, in turn, implies that moderate biases cannot be tolerated, and that moderate random errors cannot be tolerated. The only way to guarantee very small random errors is to study really large numbers, and this can be achieved in two main ways: by making individual studies large, and combining information from as many relevant studies as possible in systematic overviews (Table 56.1). However, it is not much use having very small random errors if there may well be moderate biases, so even the large sizes of some non-randomized analyses of computerized hospital records cannot guarantee medically reliable comparisons between the effects of different treatments.
AVOIDING MODERATE BIASES Proper randomization avoids systematic differences between the types of patients in different treatment groups The fundamental reason for randomization is to make possible the avoidance of moderate biases by ensuring that each type of patient is allocated in similar proportions to the different treatment strategies that are to be compared, so that only random differences should affect the final comparisons of outcome. Non-randomized methods, by contrast, cannot in general guarantee that the types of patient given the study treatment do not differ systematically in any important ways from the types of patient given other treatment(s) with which the study treatment is to be compared. For example, moderate biases might arise if the study treatment was novel and doctors were afraid to use it for the most seriously ill patients – or, conversely, if
they were more ready to try it out on those who were desperately ill. There may also be other ways in which the severity of the condition differentially affects the likelihood of being assigned to different treatments by the doctor’s choice (or by any other non-random procedure). It might at first sight appear that by collecting enough information about various prognostic features it would be possible to make some mathematical adjustments that would correct for any such differences between the types of patients who, in a non-randomized study, actually get given the different treatments that are to be compared. The hope is that such methods (which are sometimes called ‘outcomes analyses’) might achieve comparability between those entering the different treatment groups, but in general they cannot be guaranteed to do so. For some important prognostic factors may be unrecorded, while some other prognostic features may be difficult to assess exactly, and hence difficult to adjust for properly. There are two reasons for this difficulty. First, it is often not realized that, even if there are no systematic differences between one treatment group and another in the accuracy with which prognostic factors are recorded, purely random errors in assessing prognostic factors can introduce systematic biases into the statistically adjusted comparison between treatments in a non-randomized study. Second, in a nonrandomized comparison the care with which prognostic factors are recorded may differ between one treatment group and another. Doctors studying a novel treatment may investigate their patients particularly carefully, and, perhaps surprisingly, this extra accuracy can introduce a moderate bias. For example, an unusually careful search of the axilla among women with early breast cancer will sometimes result in the discovery of tiny deposits of cancer cells that would normally have been overlooked, and hence some women who would have been classified as ‘stage I’ will be reclassified as ‘stage II’. The prognosis of these ‘down-staged’ women is worse than that of those who remain as stage I, but better than that of those already classified as stage II by less intensive investigation. Paradoxically, therefore, such down-staging not only improves the average prognosis of ‘stage I’ breast cancer but also improves the average prognosis of ‘stage II’ breast cancer, biasing any non-randomized comparison with other women with stage I or stage II disease for whom the staging was less careful.
The machinery of a properly randomized trial: no foreknowledge of treatment allocation, no bias in patient management, unbiased outcome assessment and no post-randomization exclusions NO FOREKNOWLEDGE OF WHAT THE NEXT TREATMENT WILL BE
In a properly randomized trial, the decision to enter a patient is made irreversibly, in ignorance of which of the
Avoiding moderate biases 1341
trial treatments that patient will be allocated. The treatment allocation is made after trial entry has been decided upon. (The purpose of this sequence is to ensure that foreknowledge of what the next treatment is going to be cannot affect the decision to enter the patient: if it did, then those allocated one treatment might differ systematically from those allocated another.) Ideally, any major prognostic features should also be irreversibly recorded before the treatment is revealed, especially if these are to be used in any analyses of treatment. For, if the recorded value of some prognostic factor might be affected by knowledge of the trial treatment allocation, then treatment comparisons within subgroups that are defined by that factor might be moderately biased. In particular, treatment comparisons just among ‘responders’ or just among ‘non-responders’ can be extremely misleading, unless the response is assessed before treatment allocation. NO BIAS IN PATIENT MANAGEMENT OR IN OUTCOME ASSESSMENT
An additional difficulty, both in randomized and in nonrandomized comparisons of various treatments, is that there might be systematic differences in the use of other treatments (including general supportive care), or in the assessment of major outcomes. A non-randomized comparison, especially if it merely involves retrospective review of medical records, may well suffer uncorrectably from moderate biases due to such systematic differences in ancillary care or assessment. In the context of a randomized comparison, however, it is generally possible to devise ways to keep any such biases small. For example, placebo tablets may be given to control-allocated patients and certain subjective assessments may be ‘blinded’ (although this is less important in studies assessing mortality). ‘INTENTION-TO-TREAT’ ANALYSES WITH NO POST-RANDOMIZATION EXCLUSIONS
Even in a properly randomized trial, unnecessary biases could be introduced by inappropriate statistical analysis. One of the most important sources of bias in the analysis is undue concentration on just one part of the evidence (i.e. on ‘data-derived subgroup analyses’, see below) instead of on the totality of the evidence. Another bias, which is easily avoided, is caused by post-randomization exclusion of patients, especially if the type (and prognosis) of those excluded from one treatment group differs from that of those excluded from another. The fundamental statistical analysis of a trial should, therefore, generally compare all those originally allocated one treatment (even though some of them may not have actually received it) with all those allocated the other treatment (i.e. it should be an ‘intention-to-treat’ analysis). Additional analyses can also be reported: for example, in describing the frequency of some very specific side effect it may be preferable to describe its incidence only among those who actually
received the treatment. (This is because strictly randomized comparisons may not be needed to assess extreme relative risks.) However, in assessing the overall outcome, such ‘on-treatment’ analyses can be misleading, and ‘intentionto-treat’ analyses are generally a more trustworthy guide as to whether there is any real difference between the trial treatments in their effects on long-term outcome (for further discussion, see reference 3 and reference 4).
Problems produced by unduly data-dependent emphasis on particular results The treatment that is appropriate for one patient may be inappropriate for another. Ideally, therefore, what is wanted is not only an answer to the question ‘Is this treatment helpful on average for a wide range of patients?’, but also an answer to the question ‘For which recognizable categories of patient is this treatment helpful?’. This ideal is, however, difficult to attain directly because the direct use of clinical trial results in particular subgroups of patients is surprisingly unreliable. Even if the real sizes of the effects of treatment in specific subgroups are importantly different, standard subgroup analyses are so statistically insensitive that they may well fail to demonstrate these differences. Conversely, even if there is a highly significant ‘interaction’ (i.e. an apparent difference between the sizes of the therapeutic effects in different subgroups) and the results seem to suggest that treatment works in some subgroups but not in others (thereby giving the appearance of a ‘qualitative interaction’), this may still not be good evidence for subgroup-specific treatment preferences. Questions about such ‘interactions’ between patient characteristics and the effects of treatment are easy to ask, but surprisingly difficult to answer reliably. Apparent interactions can often be produced just by the play of chance and, in particular subgroups, can mimic or obscure some of the moderate treatment effects that might realistically be expected. To demonstrate this, a subgroup analysis was performed based on the astrological birth signs of patients randomized in the very large ISIS-2 trial of the treatment of acute myocardial infarction.5 Overall in this trial, the 1-month survival advantage produced by aspirin was particularly clearly demonstrated (804 vascular deaths among 8587 patients allocated aspirin vs. 1016 among 8600 allocated controls; 23 per cent reduction; p 0.000001). But, when these aspirin analyses were subdivided by the patients’ astrological ‘birth signs’, to illustrate the unreliability of subgroup analyses, aspirin appeared to be totally ineffective for those born under Libra or Gemini (Table 56.2). It would obviously be unwise to conclude from such a result that patients born under the sign of Libra or Gemini should not be given this particular treatment. Yet, similar conclusions based on ‘exploratory’ dataderived subgroup analyses that, from a purely statistical viewpoint, are no more reliable than these, are often reported and believed, with inappropriate effects on practice.
1342 Large-scale randomized evidence: trials and overviews
Table 56.2 False negative mortality effect in a subgroup defined only by the astrological ‘birth sign’: the ISIS-2 trial of aspirin among over 17 000 acute myocardial infarction patients Astrological ‘birth sign’ Libra or Gemini All other signs Any birth signa
No. of 1-month deaths (aspirin vs. placebo)
Statistical significance
150 vs. 147 654 vs. 869 804 (9.4%) vs. 1016 (11.8%)
NS 2p 0.000001 2p 0.000001
a
Appropriate overall analysis for assessing the true effect in all subgroups.
There are three main remedies for this unavoidable conflict between the reliable subgroup-specific conclusions that doctors want and the unreliable findings that direct subgroup analyses can usually offer. However, the extent to which these remedies are helpful in particular instances is one on which informed judgements differ. First, where there are good a priori reasons for anticipating that the effect of treatment might be different in different circumstances, then a limited number of subgroup analyses may be pre-specified in the study protocol, along with a prediction of the direction of such proposed interactions. For example, it was expected that the benefits of tamoxifen would be greater for women with hormonesensitive (oestrogen-receptor positive) breast tumours; and that the benefits of fibrinolytic therapy for acute myocardial infarction would be greater the earlier patients were treated (and so some studies pre-specified analyses subdivided by time from onset of symptoms to treatment, see later). These pre-specified subgroup-specific analyses are then to be taken much more seriously than other subgroup analyses. The second approach is to emphasize chiefly the overall results of a trial (or, better still, of all such trials) for particular outcomes as a guide – or at least a context for speculation – as to the qualitative results in various specific subgroups of patients, and to give less weight to the actual results in each separate subgroup. This is clearly the right way to interpret the findings in Table 56.2, but it is also likely in many other circumstances to provide the best assessment of whether one treatment is better than the other in particular subgroups. Of course, the extrapolation needs to be done in a medically sensible way. For example, if one treatment has substantial side effects then it may be inappropriate for low-risk patients. (In this case, the side effects in a particular subgroup and the proportional benefit in that subgroup should be estimated separately, but the estimation for both might be more reliable if based on an appropriate extrapolation from the overall results rather than on the results in that one subgroup alone.) The third approach is to be influenced, in discussing the likely effects on mortality in specific subgroups, not only by the mortality analyses in these subgroups but also by the analyses of recurrence-free interval or some other major ‘surrogate’ outcome. For, if the overall results are similar but much more highly significant for recurrence-free
interval than for mortality, subgroup analyses with respect to the former may be more stable and may provide a better guide as to whether there are any big differences between subgroups in the effects of treatment, (again particularly if such subgroup analyses were specified before results were available).
AVOIDING MODERATE RANDOM ERRORS The need for large-scale randomization To distinguish reliably between the two alternatives – either that there is no worthwhile difference in survival, or that treatment confers a moderate, but worthwhile, benefit (e.g. 10 or 20 per cent fewer deaths) – not only must systematic errors be guaranteed to be small (see above) compared with such a moderate risk reduction, but so too must any of the purely random errors that are produced just by chance. Random errors can be reliably avoided only by studying large enough numbers of patients. It is not, however, sufficiently widely appreciated just how large clinical trials really need to be, in order to detect moderate differences reliably. This can be illustrated by a hypothetical trial that is actually quite inadequate – even though by previous standards it is moderately large – in which a 20 per cent reduction in mortality (from 50 per cent down to 40 per cent) is supposed to be detected among 400 cancer patients (200 treated and 200 controls). In this case, one might predict finding about 100 deaths (50 per cent) in the control group and 80 (40 per cent) in the treated group. Even if exactly this difference were to be observed, however, it would not be conventionally significant (p value 0.1; indicating that even if there is no real difference between the effects of the trial treatments, it would still be relatively easy for a result at least as extreme as this to arise by chance alone). Although the play of chance might well increase the difference enough to make it conventionally significant (e.g. to 110 deaths versus 70 deaths; p 0.001), it might equally well dilute, obliterate (e.g. to 90 deaths vs. 90 deaths) or even reverse it. The situation in real life is often even worse, as the average trial size may be only several dozen patients, rather than the several thousand that would ideally be needed.
Avoiding moderate random errors 1343
Mega-trials: how to randomize large numbers One of the chief techniques for obtaining appropriately large-scale randomized evidence is to make trials extremely simple, and then to invite hundreds of hospitals to collaborate. The first of these large simple trials (or ‘mega-trials’) were the ISIS and GISSI studies5,6 in heart attack treatment, and a few mega-trials have now been undertaken in cancer, such as the QUASAR (‘Quick And Simple And Reliable’) study of colorectal cancer chemotherapy.7 But, in terms of medically significant findings, what has been achieved so far is only a fraction of what could quite readily be achieved by the wholehearted pursuit of such research strategies. Any obstacle to simplicity is an obstacle to large size, so it is worth making enormous efforts at the design stage to simplify and streamline the process of entering, treating and assessing patients. Many trials would be of much greater scientific value if they collected ten times less data, both at entry and during follow-up, on ten times more patients. It is particularly necessary to simplify the entry of patients, for if this is not done then rapid recruitment may be difficult. The current fashions for unduly complicated eligibility criteria, overly detailed ‘informed’ consent, over-burdensome regulatory procedures, extensive auditing of data, and excessive quality-oflife assessments and measurements of the economic costs of treatment are often inappropriate.1,8
Simplification of economic and ‘quality-of-life’ assessments Often, the effectiveness of various treatments needs to be balanced against the costs, but that does not necessarily imply that costs should be assessed in the same studies in which effectiveness is to be assessed, especially if attempts to assess costs seriously damage attempts to assess the effects on mortality and major morbidity sufficiently reliably. Moreover, what really matters is the cost of a treatment in routine practice, not its cost when given in the particular circumstances of a randomized trial. For this reason, it is better to measure differences in resource usage between treatments rather than differences in costs. Likewise, of course, any important ways in which treatments affect the quality of life need to be understood, but again that does not necessarily imply that ‘quality-of-life’ indices should be assessed in the same trials that assess the main effects of treatment. In particular, although several thousand patients may be required for reliable assessment of the effects of treatment on mortality and major morbidity, only a few hundred are likely to be needed for sufficiently reliable assessment of the effects of treatment on quality-of-life measures (or on costs of treatment). Because of these different sample size requirements, if such assessments are to be incorporated within a large mortality study, then they can be included as small sub-studies. Also, as for clinical outcome data, the work involved in
collecting economic and quality-of-life data should be kept to a minimum so recruitment is not compromised. However, even this may be difficult in practice, and there are many instances where what should be a large, simple trial of clinical efficacy should not be jeopardized by the measurement of such factors. Moreover, the effects of a treatment on quality of life in a trial when both the doctors and the patients are uncertain about any clinical benefits of the treatment may differ substantially from its effects on quality of life after the treatment has been shown to improve survival. Hence, it may be better to assess these other outcome measures only after having determined whether the treatment has any worthwhile effects on mortality and major morbidity, and if (as is often the case) it does not, then any costs and adverse effects on quality of life may be largely irrelevant.
Simplification of entry procedures for trials: the ‘uncertainty principle’ For ethical reasons, patients cannot have their treatment chosen at random if either they or their doctor are already reasonably certain what treatment to prefer. Hence, randomization can be offered only if both doctor and patient feel substantially uncertain as to which of the trial treatments is best. The question then arises: of those about whose treatment there is such uncertainty, which categories should be offered randomization? The obvious answer is all of them, welcoming the heterogeneity that this will produce. (For example, either the treatment of choice will turn out to be the same for men and women, in which case the trial might as well include both, or it will be different, in which case it is particularly important to study both sexes.) This approach of randomizing a wide range of patients in whom there is substantial uncertainty as to which treatment option is best was used in the MRC’s European Carotid Surgery Trial (ECST), which compared a policy of immediate carotid endarterectomy versus a policy of ‘watchful waiting’ in patients with partial carotid artery stenosis and a recent minor stroke in the part of the brain supplied by that carotid artery.9 If a patient was prepared at least to consider surgery then the neurologist and surgeon responsible for that individual patient’s care considered in their own way whatever medical, personal or other factors seemed to them to be relevant (Fig. 56.1), including, of course, the patient’s own preferences and values: 1. If they were then reasonably certain, for any reasons, that they did wish to recommend immediate surgery for that particular individual, then the patient was ineligible and was not part of the ECST. 2. Conversely, if they were reasonably certain, for any reason, that they did not wish to recommend immediate surgery, then that patient was likewise ineligible.
1344 Large-scale randomized evidence: trials and overviews
Transient ischaemic attack, retinal infarction or non-disabling ischaemic stroke in one or both carotid territories in past 6 months
Patient prepared to consider surgery
Carotid angiogram performed
Doctors 'REASONABLY CERTAIN' surgery IS indicated: INELIGIBLE
Doctors 'REASONABLY CERTAIN' surgery NOT indicated: INELIGIBLE
'SUBSTANTIALLY UNCERTAIN' whether carotid surgery is indicated: Telephone CTSU to RANDOMIZE 60% IMMEDIATE SURGERY vs 40% NOT
TRIAL CENTRE: categorize % stenosis on relevant side of ENTRY ANGIOGRAM (i.e. of the 'symptomatic' carotid artery) Mild (30%) stenosis reported 1991
Moderate (30–69%) stenosis reported until 1996
Severe (70–99%) stenosis reported 1991
Figure 56.1 Example of the ‘uncertainty principle’ for trial entry: the chief eligibility criterion for the European Carotid Surgery Trial (ECST)9 was that the doctors and patient should be substantially uncertain whether to risk immediate or deferred surgery. (Partly because this criterion was appropriately flexible, ECST became the largest ever trial of vascular surgery.)
3. If, but only if, they were substantially uncertain what to recommend, then that individual patient was automatically eligible for randomization between immediate versus no immediate surgery (with all patients receiving whatever their doctors judged to be the best available medical care – which generally included advice to stop smoking, treatment of any hypertension, and the use of aspirin as an anti-thrombotic drug). There were substantial differences between individual doctors in the types of patients they were uncertain about (in terms of the severity of carotid stenosis, as well as in various other characteristics). This guaranteed that no category – mild, moderate or severe stenosis – would be wholly excluded, and hence that the trial would yield at least some direct evidence in each. As a result of the wide and simple entry criteria adopted by ECST, 3000 patients were randomized and the study was therefore able to provide some clear answers about who needed carotid endarterectomy. For patients with only mild (0–29 per cent) carotid artery stenosis on their pre-randomization angiogram, there was little risk of ipsilateral ischaemic stroke, even in the absence of surgery, so the benefits of surgery over the next few years were small and outweighed by its early risks. Conversely, for patients with severe
(70–99 per cent) stenosis, the risks of surgery were significantly outweighed by its later benefits over the next few years. For both of these categories the trial stopped early, but for the intermediate category of patients with moderate (30–69 per cent) stenosis, the balance of surgical risk and eventual benefit remained uncertain, so recruitment into the study continued, with entry still governed by the ‘uncertainty principle’ as before.10 In large trials, homogeneity of patients is generally a defect, while heterogeneity is generally a strength. Consider, for example, the trials of fibrinolytic therapy for acute myocardial infarction: if the coronary artery has been occluded for long enough, then the heart muscle that it supplies will have been irreversibly destroyed: how late after the heart attack starts is fibrinolytic treatment still worth risking – 3 hours? 6 hours? 12 hours? 24 hours? Before the large trials of fibrinolysis, it was forcefully, but mistakenly, argued that such treatments could not possibly be of any worthwhile benefit if given more than a few hours after the onset of symptoms. Consequently, some trials had restrictive entry criteria that allowed inclusion of only those patients who presented 0–6 hours after pain onset, so those trials contributed almost nothing to the key question of how late such treatment can still be useful. In contrast, trials with wider, more heterogeneous entry criteria that included some patients with longer delays between pain onset and randomization assessed this question prospectively, and were able to show that fibrinolytic therapy can have definite protective effects when given not only 0–6 but also 7–12 hours after pain onset11 (see below). The longer that fibrinolytic treatment for such patients was delayed, however, the less benefit it seemed to produce. The benefit was greatest (about 30 lives saved per 1000) among those randomized 0–6 hours after the onset of pain (Fig. 56.2), but the mortality reduction was still substantial and significant (about 20 per 1000, 2p 0.003) when such patients were randomized 7–12 hours after pain onset. Indeed, if they were randomized 13–18 hours after pain onset there still appeared to be some net reduction in mortality (about 10 per 1000, but not statistically definite). The regression line in Figure 56.2 reinforces, in a more reliable way, these separate subgroup analyses. The ‘uncertainty principle’ meets simultaneously the requirements of ethicality, heterogeneity, simplicity and maximal trial size. It states that the fundamental eligibility criterion is that both patient and doctor should be substantially uncertain about the appropriateness for this particular patient of each of the trial treatments. With such uncertainty as the fundamental principle of eligibility, informed consent can also be simplified, for the degree of ‘informed consent’ that is humanly appropriate in a randomized comparison of different treatments that is governed by the ‘uncertainty principle’ should probably not differ greatly from that which is humanly appropriate in routine practice outside of trials, where treatment is being chosen haphazardly – or, to put it another way, ‘double standards’ between trial and non-trial situations are not appropriate.
Absolute benefit per 1000 patients with ST elevation or BBB allocated fibrinolytic therapy ( 1SD)
Minimizing both bias and random error: systematic overviews (meta-analyses) of randomized trials 1345
40
3000 14 000
30
Loss of benefit per hour of delay to randomization: 1.6 SD 0.6 per 1000 patients
12 000 9000
20
10 7000 0 0
6 12 18 Hours from symptom onset to randomization
24
Figure 56.2 Benefit versus delay (0–1, 2–3, 4–6, 7–12, or 13–24 hours) in the nine largest randomized trials of fibrinolytic therapy versus control in patients with acute myocardial infarction. One-month mortality results for 45 000 patients with ST elevation or bundle branch block (BBB) when randomized, showing the definite net benefit even for the 9000 randomized 7–12 hours after the onset of pain.11
The haphazard nature of many non-randomized treatment choices is reflected in the wide variations in practice between and within countries (and, even when practice is similar, it may be similarly wrong: for example, before the ISIS-2 results became available [see later], almost all doctors around the world were not using fibrinolytic therapy for acute myocardial infarction). Provided that trials are governed by the ‘uncertainty principle’, there is an approximate parallelism between good science and good ethics. Indeed, in such circumstances, excessively detailed consent procedures (which can be distressing and inhumane, and so would not be considered appropriate in routine practice) would not be either scientifically or ethically appropriate (for further discussion, see reference 8). This ‘uncertainty principle’ is just one of many ways to simplify trials and thereby help them to avoid becoming enmeshed in a mass of wholly unnecessary traditional complexity. If randomized trials can be vastly simplified, as has already been achieved in a few major diseases, and thereby made vastly larger, then they will play an appropriately central role in the development of rational criteria for the planning of health care throughout the world.
MINIMIZING BOTH BIAS AND RANDOM ERROR: SYSTEMATIC OVERVIEWS (META-ANALYSES) OF RANDOMIZED TRIALS When several trials have all addressed much the same therapeutic question, then the traditional procedure of choosing only a few of them for emphasis and fame may be a source of serious bias, since chance fluctuations for or
against treatment may affect which trials become famous. To avoid this, it is appropriate to base inference chiefly on a systematic overview (or ‘meta-analysis’) of all the results from all the trials that have addressed a particular type of question (or on an unbiased subset of such trials), and not on some potentially biased subset of the trials.2,12 Such overviews will also minimize random errors in the assessment of treatment, since, in general, far more patients are involved in an overview than in any individual trial that contributes to it. The separate trials may well be heterogeneous in their entry criteria, their treatment schedules, their follow-up procedures, their methods of treating relapse, etc. At one extreme, each trial might, in view of this heterogeneity, be considered in virtual isolation from all others, while at the opposite extreme all might be considered together. Both of these extreme views have some merit, and the pursuit of each by different people may prove more illuminating than too definite an insistence on any one particular approach. The heterogeneity of the different trials, however, merely argues for careful interpretation of any overviews of different trial results, rather than arguing against any such overviews. For, whatever the difficulties of interpretation of overviews may be, without systematic overviews moderate biases and random errors that may obscure any moderate treatment effects (or, conversely, may imply effects where none exist) cannot reliably be avoided.
Which overviews are trustworthy? Over the past decade or two, a large (and rapidly increasing) number of meta-analyses of randomized trial results have been reported, not all of which are trustworthy. The two fundamental questions are how carefully the metaanalysis has been done, and how large it is. The simplest approach is merely to have collected and tabulated the published data from whatever randomized trial reports can be found easily in the literature, and sometimes this may suffice. At the opposite extreme, extensive efforts may have been made by those organizing the overview to locate every potentially relevant randomized trial, to collaborate closely with the trialists to seek individual data on each patient ever randomized into those trials, and then (after extensive checks and corrections of such data) to produce, in collaboration with those trialists, agreed analyses and publications. The results of some of the largest such collaborations will be described later: the Antiplatelet Trialists’ (APT) Collaborative Group,13 the Fibrinolytic Trialists’ (FTT) Collaborative Group,11 the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG)14 and the Colorectal Cancer Collaborative Group.15 Collaboration of the original trialists in the overview process, with collection of individual patient data, can help to avoid or minimize the biases that could be produced by missing trials (e.g. due to the greater likelihood of extremely good, or extremely bad, results being particularly widely known about and
1346 Large-scale randomized evidence: trials and overviews
SOME EXAMPLES OF IMPORTANT RESULTS IN THE TREATMENT OF VASCULAR AND OF NEOPLASTIC DISEASE THAT COULD HAVE BEEN RELIABLY ESTABLISHED ONLY BY LARGE-SCALE RANDOMIZED EVIDENCE Definite result from a single very large trial: benefit from medium-dose aspirin for patients with suspected acute myocardial infarction (and benefits among other groups of patients indicated by overviews of trials) In the ISIS-2 trial, half of 17 000 patients with suspected acute myocardial infarction were allocated aspirin tablets (162 mg/day for 1 month, which virtually completely inhibits cyclooxygenase-dependent platelet inhibition), and half were allocated placebo (i.e. dummy) tablets.5 Before 1988, when the ISIS-2 results were published, aspirin was not routinely used in acute myocardial infarction, and no other major trial had (or has subsequently) assessed aspirin in suspected acute myocardial infarction. But, the effects of 1 month of aspirin were so definite in ISIS-2 (804/8587
Placebo tablets: 1016 vascular deaths (11.8%) 1000
Cumulative number of vascular deaths
published), by inappropriate post-randomization withdrawals or by failure to allocate treatment properly at random. If randomization was done properly in the first place, then post-randomization withdrawals can often be followed up and restored to the study for an appropriate ‘intention-to-treat’ analysis. Knowledge of the exact methods of treatment allocation (backed up by checks on whether the main prognostic factors recorded are non-randomly distributed between the treatment groups in a particular trial) may help to identify trials that were not, in fact, properly randomized, and that should therefore be excluded from an overview of randomized trials. Overviews based on individual patient data may also provide more information about treatment effects than the more usual overviews of grouped data, for they allow more detailed analyses – indeed, if they are really large then they may actually yield statistically reliable subgroup analyses of the effects of treatment in particular types of patient. Conversely, however, even a perfectly conducted overview may not be large enough to be reliable. An overview that brings together complete data from all the trials that have ever been done of a certain treatment but still (because the trials were all small) includes a total of only 100 deaths will have random errors that are no smaller than those for a single trial with 100 deaths among such patients. Small-scale evidence, whether from an overview or from one trial, is often unreliable, and will often be found in retrospect to have yielded wrong answers. What is needed is large-scale randomized evidence; it does not matter much whether that evidence comes from a properly conducted overview or a properly conducted trial. The practical medical value of such evidence will now be illustrated by a few recent examples.
800
600 Aspirin: 804 vascular deaths (9.4%) 400
200
0 0
7 14 21 28 Days from randomization
35
Figure 56.3 Effect of administration of aspirin for 1 month on 35-day mortality in the ISIS-2 trial among over 17 000 acute myocardial infarction patients. (Absolute survival advantage: 24 [SD 5] lives saved per 1000 allocated aspirin; 2p 0.00001.)
vascular deaths among those allocated aspirin vs. 1016/8600 among those not) that even the lower 99 per cent confidence limit would have represented a worthwhile benefit from so simple and inexpensive a treatment (Fig. 56.3). As a result, world-wide treatment patterns changed sharply when the ISIS-2 results emerged, and aspirin is now routinely used in many different countries for the majority of emergency hospital admissions with suspected acute myocardial infarction. In the UK, for example, two British Heart Foundation surveys found cardiologists reporting that routine aspirin use in acute coronary care had increased from under 10 per cent in 1987 to over 90 per cent in 1989.16 World-wide, the annual number of patients with suspected myocardial infarction who would nowadays be given such treatment must be well over a million a year, suggesting that in this clinical context alone aspirin is already preventing tens of thousands of premature deaths each year. But, if the ISIS-2 trial had been 10 times smaller (i.e. 1700 instead of 17 000 patients), then exactly the same proportional reduction in mortality as in Figure 56.3 would not have been conventionally significant, and therefore would have been much less likely to influence medical practice – indeed, the result might by chance have appeared exactly flat, greatly damaging future research on aspirin in this context. Likewise, if the ISIS-2 trial had been non-randomized, then it might well have got the wrong answer (since in a non-randomized study doctors might tend to give active treatment to patients who were particularly ill, or who were in various other ways
Some examples of important results in the treatment of vascular and of neoplastic disease 1347
Table 56.3 Summary of overall results in trials of aspirin (or other antiplatelet drugs)a for the prevention of vascular events: the Antiplatelet Trialists’ Collaboration, involving a total of about 100 000 randomized patients in over 100 trials
Type of patient studied
Average scheduled treatment duration (and approximate numbers of patients randomized)
Proportions who suffered a non-fatal stroke, non-fatal heart attack or vascular death during the trials Anti-platelet (%)
Control (%)
Events avoided in these trials
10 13 18
14 17 22
40 per 1000 (2p 0.00001) 40 per 1000 (2p 0.00001) 40 per 1000 (2p 0.00001) 20 per 1000 (2p 0.00001)
High risk: Suspected acute heart attack Previous history of heart attack Previous history of stroke or transient ischaemic attack Other vascular diseaseb
1 year
(20 000)
7
9
Low risk: Primary prevention in low-risk people
5 years (30 000)
4.4
4.8
1 month (20 000) 2 years (20 000) 3 years (10 000)
4 per 1000 (2p 0.05)
a
The most widely tested regimen was medium-dose aspirin, involving an average daily dose of 75–325 mg, and no other anti-platelet regimen appeared to be significantly more or less effective than this at preventing such vascular events. b For example, angina, peripheral vascular disease, arterial surgery or angioplasty, etc.
somewhat different from those not given active treatment). And, even if a non-randomized study did happen to get an unbiasedly correct answer, it would be impossible to be sure that it had actually done so, and hence again a nonrandomized study might have had much less influence on medical practice than did ISIS-2. In the ISIS-2 trial, aspirin significantly reduced the 1-month mortality, but it also significantly reduced the number of non-fatal strokes and of non-fatal re-infarctions that were recorded in hospital. Combining all these three outcomes into ‘vascular events’ (stroke, death or re-infarction), 13 per cent of those allocated aspirin, and 17 per cent of those who were not, were known to have suffered a vascular event in the month after randomization (Table 56.3: an absolute difference of 40 events per 1000 treated – or, perhaps more relevantly, of 40 000 per million).13 The randomized trials of aspirin, or of other anti-platelet regimens, in other types of high-risk patients (e.g. a few years of aspirin for those who have survived a myocardial infarction or stroke) have not been as large as ISIS-2, and so, taken separately, most have yielded false negative results. But, when the results from many such trials are combined, statistically definite reductions in ‘vascular events’ are seen (Table 56.3). Since such treatments do not appear to increase non-vascular mortality, all-cause mortality is also significantly reduced. In principle these findings could, if appropriately widely exploited, prevent about 100 000 premature vascular deaths a year in developed countries alone, and there are probably at least as many vascular deaths in less-developed as in developed countries. So, with realistically achievable levels of use of ‘medium-dose’ aspirin (75–325 mg/day) for the secondary prevention of vascular disease, it might well be possible in practice to ensure that aspirin is used in enough high-risk patients to prevent, or substantially delay,
at least 100 000 vascular deaths a year world-wide, and such use of aspirin would, in addition, prevent a comparable number of non-fatal strokes or heart attacks. (Mediumdose aspirin was the least expensive and most widely tested anti-platelet regimen: it is of proven efficacy and, on review of all the anti-platelet trials, no other anti-platelet regimen has been shown to be of greater efficacy in preventing vascular events: see notes to Table 56.3.) This large-scale randomized evidence about medium-dose aspirin is now changing world-wide clinical practice in ways that will, at low cost, prevent much death and disability in high-risk patients. But, small trials, small overviews or non-randomized studies (however large) could not possibly have provided appropriately reliable evidence about such moderate risk reductions.
Definite result from a very large overview of trials: benefit from ‘adjuvant’ hormonal therapy with tamoxifen for patients with ‘early’ breast cancer (and further benefits with ovarian ablation in younger women) By definition, in ‘early’ breast cancer all detectable deposits of disease are limited to the breast and the loco-regional lymph nodes, and can be removed surgically. But, experience shows that undetectably small deposits may remain elsewhere that eventually, perhaps after a delay of several years, cause clinical recurrence at a distant site, which is then usually followed by death from the disease. These micrometastatic deposits may have been stimulated by the body’s own hormones during the years before recurrence became detectable. So, among women who have had the detectable deposits of breast cancer removed by surgery (or by surgery with radiotherapy), there have been many
1348 Large-scale randomized evidence: trials and overviews
trials of ‘adjuvant’ treatments that either reduce the production of endogenous oestrogens (e.g. various forms of ovarian ablation), or that block the access of those oestrogens to the tumour cells (e.g. tamoxifen, which blocks the oestrogen receptor protein in some breast cancer cells). Taken separately, most of these adjuvant trials have been too small to provide reliable evidence about long-term survival.14 But, if the results of all of them are combined, then some very definite differences in 10-year survival do emerge (Fig. 56.4). Among women with stage II disease who were less than 50 years old (and, therefore, generally pre- or peri-menopausal), ovarian ablation appears to produce about a 10 per cent absolute difference in 10-year survival (e.g. 50 per cent versus 40 per cent). This finding is based on the analysis of only a few hundred deaths, so it is still not as reliable as might ideally be wished, and, because substantial uncertainty remains, much larger trials are now in progress. Among older stage II women, ovarian ablation is unlikely to be of much relevance (since most of the endogenous oestrogen at older ages comes from sources other than the ovaries) but, in aggregate, the randomized trials among such women have shown very definitely that five years of tamoxifen likewise produces about a 10 per cent absolute difference in 10-year survival. A smaller, but still highly significant, reduction in mortality by tamoxifen is also seen among the 10 000 randomized women with stage I disease. Taken separately, however, 37 of the 42 tamoxifen trials were too small to have yielded statistically reliable evidence on their own (2p 0.01) and the five other trials were significant only because, by chance, they had results that were too good to be true. Tamoxifen 5 yrs vs. Not ALL DEATHS 100
TAMOXIFEN for 5 yrs 80 65% CONTROL
60
2P0.00001 57%
%
8000 randomised 40 24% proportional reduction in annual risk of death (se 4), yielding 8% absolute reduction in mortality at 15 yrs
ER-positive women
20
0
0
5
10
15
years
Figure 56.4 Effects of hormonal adjuvant treatments for early breast cancer on 5-year survival in a world-wide overview of randomized trials.14
These tamoxifen overview results have already changed clinical practice substantially, and have re-directed research towards large randomized trials (e.g. ATLAS and aTTom) of the effects of different durations of tamoxifen: should tamoxifen in asymptomatic women continue for 5 years, or for 10 or more years. Large randomized studies of the primary prevention of breast cancer among highrisk women by tamoxifen have also been undertaken, encouraged by the results from the tamoxifen trials overview. In 30 000 women with established cancer (stage I or stage II) in one breast, there was a highly significant reduction of one third in the likelihood of development of contralateral breast cancer, but a small absolute increase in endometrial cancer. Again, this degree of trustworthy detail would not have been attained without large-scale randomized evidence.
Factorial (2 ⴛ 2) trial designs: separate assessment of more than one treatment in the same trial In ISIS-2, not only were patients randomly allocated to receive aspirin or placebo tablets (as described above: Fig. 56.3), but they were also separately allocated to receive intravenous streptokinase or a placebo infusion.5 In this ‘factorial’ design (which allows the separate assessment of more than one treatment without any material loss in the statistical reliability of each comparison), one quarter of patients were allocated aspirin alone, one quarter were allocated streptokinase alone, one quarter were allocated both streptokinase and aspirin, and one quarter were allocated neither (i.e. placebo tablets and placebo infusion). Streptokinase, like aspirin, produced a highly significant reduction in mortality (and the combination of streptokinase and aspirin was highly significantly better than either aspirin or streptokinase alone: Fig. 56.5). It might appear, from Figure 56.5, that there was no need for any more randomized evidence about fibrinolytic therapy, but this ignores the potential hazards of such treatment and the heterogeneity of patients. Taken separately, even ISIS-2, the largest of these trials, was not big enough for statistically reliable subgroup analyses, but when the nine largest trials were all taken together they included a total of about 60 000 patients, half of whom were randomly allocated fibrinolytic. Those entering a coronary care unit with a diagnosis of suspected or definite acute myocardial infarction range from patients who are already in cardiogenic shock, with low blood pressure and a fast pulse (half of whom will die rapidly), to those who have merely got a history of chest pain and no very definite changes on their ECG (of whom ‘only’ a small percentage will die before discharge). Fibrinolytic therapy often causes a frightening blood pressure drop: should it be used in patients who are already dangerously hypotensive? It occasionally causes serious strokes: should it be used in patients who are elderly or hypertensive, and therefore already have
Some examples of important results in the treatment of vascular and of neoplastic disease 1349
an above-average risk of stroke (or who have only slight changes on their ECG, and therefore have only a low risk of cardiac death)? These questions needed to be answered reliably before appropriate and generally accepted indications for, and against, such an immediately hazardous but potentially effective therapy could be devised. To address them, all fibrinolytic therapy trialists have collaborated in a systematic overview of the randomized evidence.11 On review of the 60 000 patients randomized between fibrinolytic therapy and control in trials of more than 1000 patients, some of the therapeutic questions were relatively easy to answer satisfactorily. For example, it appeared that most of those whose ECG was still fairly normal (or showed some other pattern that indicated only a low risk of death) might as well be left untreated, leaving open the option of starting fibrinolytic treatment urgently if their ECG changed suddenly for the worse in the following few hours. Conversely, among those who already had ‘high-risk’ ECG changes when they were randomized, the absolute benefit of immediate fibrinolytic therapy was, if anything, slightly greater than is indicated by Figure 56.5, and age, sex, blood pressure, heart rate, diabetes and previous history of myocardial infarction could not identify reliably any group that would not, on average, have their chances of survival appreciably increased by treatment.
Routine hospital care alone 13% dead (568/4300) Aspirin only
Percentage dead
10
Streptokinase only
Routine care combination of both streptokinase and aspirin 8% dead (343/4292)
5
0 0
1 2 3 4 Weeks from starting treatment
5
Figure 56.5 Effects of 1-hour streptokinase infusion (and of 1 month of aspirin) on 35-day mortality in ISIS-2 among 17 187 acute myocardial infarction patients who would not normally have received streptokinase or aspirin, divided at random into four similar groups to receive aspirin only, streptokinase only, both, or neither. (Any doctor who believed that a particular patient should be given either treatment gave it, and did not include that patient in ISIS-2.)
Such detailed inferences are difficult enough with largescale, properly randomized evidence, and would be impossible without it; because of their unknowable biases (see above), non-randomized database analyses are simply not a viable alternative to large-scale randomized evidence. Nor would randomization of ‘only’ several thousand patients have been sufficient. Indeed, in several important respects what is still needed is more, rather than less, randomized evidence about the effects of fibrinolytic therapy in various particular types of patient. For example, it is still not clear whether patients who present 12–18, or even 18–24, hours after pain onset should be treated: more randomized evidence is still needed (Fig. 56.2). Nevertheless, substantial progress has been made by the past decade of mega-trials of fibrinolytic agents. World-wide, in the mid 1990s, about half a million patients per year were given fibrinolytic therapy, avoiding about 10 000 early deaths each year.
Small trials refuted by a mega-trial: lack of significant benefit from magnesium infusion in suspected acute myocardial infarction It had been suggested that, in patients with suspected acute myocardial infarction, an infusion of a magnesium salt might reduce early mortality. Several small trials, involving between them a total of only about 1500 patients, had addressed this question by 1990, and their aggregated results indicated a statistically significant, but implausibly large, benefit (42/754 deaths among those allocated magnesium versus 86/740 among the controls; 2p 0.001).17 Some argued that such results constituted proof beyond reasonable doubt that magnesium was of sufficient value to justify widespread usage without seeking further randomized evidence, but others remained sceptical, arguing that the apparent results were far too good to be true. Two trials, one (LIMIT-2) (18) involving 2000 patients and one (ISIS-4) (17) involving 58 000, were therefore set up to test more reliably the possible effects of magnesium. The former yielded a moderately promising result (Table 56.4) indicating avoidance of about one quarter of the early deaths, but because of its small size this result was statistically compatible with a true benefit that ranged from about zero to about a halving of early mortality. The much larger ISIS-4 trial, however, yielded a completely unpromising result, so the overall evidence, based on about 60 000 randomized patients, is now non-significantly adverse.17 In view of the striking disparity between the apparent effects of magnesium before and after ISIS-4 had provided large-scale randomized evidence, it is of interest to recall some of the expert views that were expressed while ISIS-4 was in progress. Some felt so strongly that magnesium was already of proven benefit (and hence that further randomization was unethical) that the data monitoring committee
1350 Large-scale randomized evidence: trials and overviews
Table 56.4 Magnesium in acute myocardial infarction: contrast between the results of the smaller and the larger randomized trials One-month mortality
Nine small trials LIMIT-2 trial ISIS-4 trial All trials
Number of patients randomized
Allocated magnesium
Allocated control
1500 2300 58 000 62 000
42/754 (5.6%) 90/1159 (7.8%) 2216/29 011 (7.6%) 2348/30 924 (7.59%)
86/740 (11.6%) 118/1157 (10.2%) 2103/29 039 (7.2%) 2307/30 936 (7.46%)
There is highly significant heterogeneity (p 0.001) between the group of small trials, the ‘hypothesis-generating’ results of which led to the testing of magnesium in ISIS-4, and the pair of larger trials (ISIS-4 and LIMIT-2), the results of which tested that hypothesis.
of ISIS-4 was lobbied to try to have the study stopped early and all future patients given magnesium. In contrast, the ISIS-4 steering committee was sufficiently sceptical to want large-scale randomized evidence. They believed that the available evidence was consistent with a negligible benefit, or even a small net hazard – although they all thought it more likely that at least some net benefit would be seen. Even after the LIMIT-2 result was available, they continued to hold these opinions, and thought that if there was any real benefit then this was likely to be less than LIMIT-2 had suggested (and hence very much less than the other small trials had suggested). Those who had trusted the implausibly extreme results from the previous small trials may well have been disappointed by the results of the ISIS-4 mega-trial, which now provide strong evidence that the routine use of magnesium has little or no effect on mortality in acute myocardial infarction. But, in a world where moderate benefits are much more plausible than large benefits, it will commonly happen that striking results in small-scale trials, in smallscale overviews or in small subgroups prove evanescent. The medical assumption that both a moderate mortality difference or a zero mortality difference may be plausible, but that an extreme mortality difference is much less so, has surprisingly strong consequences for the interpretation of randomized evidence. In particular, it implies that even quite highly significant (e.g. 2p 0.001) mortality differences that are based on only relatively small numbers of deaths may provide untrustworthy evidence of the existence of any real difference.17
RESULTS FROM LARGE, ANONYMOUS TRIALS ARE RELEVANT TO REAL CLINICAL PRACTICE A clinician is used to dealing with individual patients, and may feel that the results of large trials somehow deny the individuality of each patient. This is almost the opposite of the truth, for one of the main reasons why trials have to be large is just because patients are so different from one another. Two apparently similar patients may run entirely different clinical courses, one remaining stable and the other progressing rapidly to severe disability or early death.
Consequently, it is only when really large groups of patients are compared that the proportion of truly good and bad prognosis patients in each can be relied on to be reasonably similar. One commonly hears statements such as: ‘If a treatment effect isn’t obvious in a couple of hundred patients then it isn’t worth knowing about.’ As the previous examples demonstrate, such statements may reveal not clinical wisdom but statistical naivety. It is also said that what is really wanted is not a blanket recommendation for everybody, but rather some means of identifying those few individuals who really stand to benefit from therapy. If any criteria (e.g. short-term response to a non-placebo-controlled course of some disease-modifying agent) can be proposed that are likely to discriminate between people who will and who will not benefit, then these can, of course, be recorded prospectively at entry and the eventual trial results subdivided with respect to them. There is, however, a danger in too detailed an analysis of the apparent response of small subgroups chosen for separate emphasis because of the apparently remarkable effects of treatment in those subgroups. Even if an agent brought no benefit, it would have to be acutely poisonous for it not to appear beneficial in one or two such subgroups! Conversely, if an intervention really avoids an approximately similar proportion of the risk in each category of patient, it will, by chance alone, appear not to do so in some category or other. The surprising extent to which this happens is evident from the example in Table 56.2. A large, anonymous trial will at least still help answer the practical question of whether on average a policy of widespread treatment (except where clearly contraindicated) is preferable to a general policy of no immediate use of the treatment (except where clearly indicated). Moreover, without a few really large trials it is difficult to see how else many such questions could be resolved over the next few years. For example, digitalis has already been in use for over two centuries, and there is still no reliable consensus as to its net long-term effects on mortality. Trials are at least a practical way of making some solid progress, and it would be unfortunate if desire for the perfect (i.e. knowledge of exactly who will benefit from treatment) were to become the enemy of the possible (i.e. knowledge of the direction and approximate size of the effects of the treatment of many large categories of patient).
References 1351
KEY LEARNING POINTS 6 ●
●
●
●
Proper randomization avoids systematic differences between different types of patient in different treatment groups. Randomized control trials have resulted in changes in cancer care which have prevented tens of thousands of premature deaths. Non-randomized evidence is far inferior as a source of evidence. Moderate benefits of major outcomes are more plausible than large effects.
7
8
9
KEY REFERENCES Chalmers I. The Cochrane Collaboration: preparing, maintaining and disseminating systematic reviews of the effects of health care. Ann NY Acad Sci 1994; 703:156–63. Collins R, Doll R, Peto R. Ethics of clinical trials. In: Williams CJ (ed.) Introducing New Treatments for Cancer: Practical, Ethical and Legal Problems. Chichester: John Wiley & Sons, 1992, 49. Peto R, Pike MC, Armitage P, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Part I: Introduction and design. Br J Cancer 1976; 34:585–612. Peto R, Pike MC, Armitage P, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Part II: Analysis and examples. Br J Cancer 1977; 35:1–39. Yusuf S, Collins R, Peto R. Why do we need some large, simple randomized trials? Stat Med 1984; 3:409–20.
10
11
12
13
14
REFERENCES 1 Yusuf S, Collins R, Peto R. Why do we need some large, simple randomized trials? Stat Med 1984; 3:409–20. 2 Collins R, Gray R, Godwin J, Peto R. Avoidance of large biases and large random errors in the assessment of moderate treatment effects: the need for systematic overviews. Stat Med 1987; 6:245–50. 3 Peto R, Pike MC, Armitage P, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Part I: Introduction and design. Br J Cancer 1976; 34:585–612. 4 Peto R, Pike MC, Armitage P, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Part II: Analysis and examples. Br J Cancer 1977; 35:1–39. 5 ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187
15
16
17
18
cases of suspected acute myocardial infarction: ISIS-2. Lancet 1988; ii:349–60. Gruppo Italiano per lo Studio della Streptochinasi nell’infarto miocardico (GISSI). Effectiveness of intravenous thrombolytic treatment in acute myocardial infarction. Lancet 1986; i:397–402. QUASAR Collaborative Group. Comparison of fluorouracil with additional levamisole, higher-dose folinic acid, or both, as adjuvant chemotherapy for colorectal cancer: a randomised trial. Lancet 2000; 355:1588–96. Collins R, Doll R, Peto R. Ethics of clinical trials. In: Williams CJ (ed.) Introducing New Treatments for Cancer: Practical, Ethical and Legal Problems. Chichester: John Wiley & Sons, 1992, 49. European Carotid Surgery Trialists’ Collaborative Group. MRC European Carotid Surgery Trial: interim results for symptomatic patients with severe (70–99%) or with mild (0–29%) carotid stenosis. Lancet 1991; 337:1235–43. European Carotid Surgery Trialists’ Collaborative Group. Endarterectomy for moderate symptomatic carotid sterosis: interim results from the MRC European Carotid Surgery Trial. Lancet 1996; 347:1591–3. Fibrinolytic Therapy Trialists’ Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet 1994; 343:311–22. Chalmers I. The Cochrane Collaboration: preparing, maintaining and disseminating systematic reviews of the effects of health care. Ann NY Acad Sci 1994; 703:156–63. Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy. I: Prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. BMJ 1994; 308:81–106. Early Breast Cancer Trialists’ Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomized trials. Lancet; 351:1451–1467. Colorectal Cancer Collaborative Group. Adjuvant radiotherapy for rectal cancer: a systematic overview of 8507 patients from 22 randomized trials. Lancet 2001; 358:1291–304. Collins R, Julian D. British Heart Foundation Surveys (1987 and 1989) of United Kingdom: policies for acute myocardial infarction. Br Heart J 1991; 66:250–5. ISIS-4 (Fourth International Study of Infarct Survival) Collaborative Group. ISIS-4. A randomised factorial trial assessing early oral captopril, oral mononitrate, and intravenous magnesium sulphate in 58,050 patients with suspected acute myocardial infarction. Lancet 1995; 345:669–85. Woods KL, Fletcher S, Roffe C, Haider Y. Intravenous magnesium sulphate in suspected acute myocardial infarction: results of the second Leicester Intravenous Magnesium Intervention Trial (LIMIT-2). Lancet 1992; 339:1553–8.
57 The economics of cancer care NICK BOSANQUET
Introduction Supply of cancer services Case study of the economics of cancer: the USA Case study: improving services in New Europe
1352 1353 1353 1356
INTRODUCTION What are the main economic issues that will shape cancer care over the next few years, exercising an influence on actual patient care that is often occult but powerful? The requirement for services grows partly out of demand as requested by patients themselves and partly from need – the definition by professionals of the quality of service and level of access to services which is professionally acceptable. Cancer services will be affected by compelling new pressures both from rising demand and from changed definitions of need. Additional demand will rise mainly from a factor which is usually termed expectations, but expectations are shaped in different ways. They could rise because of improved or faster access to information through the media or internet. They could reflect changed perceptions of the right or fair level of service so that there was less willingness to take no for an answer and less deference, or expectations might change because of altered perceptions of risk, return and side effects. In general, cancer services have a more rapid growth of expectations with all these factors strongly at work. Internet and media information about cancer treatment is widely available through the very comprehensive and comprehensible web site of the National Cancer Institute. There are some hits a month from outside the USA on this web site. Perceptions of the right to service have shifted in part because there are many more younger survivors who are likely to have stronger demand than when cancer was seen as a mainly short-term and terminal illness, and
Case study: improving the patient experience The new cost pressures: new therapies The new agenda References
1358 1360 1362 1364
perceptions of risk have changed with a tendency to minimize the chance of side effects. On average, expectations have risen strongly; however, there may still be great variation between specific patients with some patients being wary of treatments that may cause them pain or discomfort without very much survival gain. Thus for some patients active palliative care which leads to privacy, dignity and control in the late phase of life may be more important than treatment which might lead to a few months more survival. For clinicians it is not only the average rise in expectations that is important but their greater variability. The demand for services is also affected by their price or cost to the patient and carers. Services in the UK are free at the point of delivery: in the USA however they are going to be affected by level of co-payment, which could be as high as 20 per cent. In the UK and in other countries the indirect costs have risen as treatments become longer and more complex. They may involve a range of costs from loss of earnings to increased car park charges and loss of earnings by carers. Definitions of ‘need’ are also likely to be rising strongly with new standards of clinical governance and change in technology. The use of care programmes and protocols involves multi-professional scrutiny of care and there is also more external audit and peer review. Definitions of need affect diagnostics as well as treatment. The intensity of treatment has also risen sharply in the UK: one estimate is that breast cancer patients are being treated with six times more activity and cost than 10 years ago.
Case study of the economics of cancer: the USA 1353
SUPPLY OF CANCER SERVICES A new staged model of cancer services is emerging. This covers: ● ● ●
● ●
prevention; screening; treatment with different options in surgery, chemotherapy and radiotherapy; risk/recurrence management; follow-up and palliative care.
There are more options for spending in each of these areas and a challenge of ensuring adequate integration between the stages. For example, rapid increases in screening may have side effects in delaying diagnosis for patients who have been referred in other channels, as diagnostic workload on false positive leads to longer queues. Heavy expenditure on treatment may mean fewer resources for programmes that might improve the quality of life of survivors. The cancer services are being affected by a wave of change in technology that is increasing treatment options and in most cases cost. Among the key areas for change are: 1. Surgery. Much cancer surgery is now being done on a minimally invasive basis and in a day or short stay setting. This may create a potential for reductions in unit costs in the longer term but for these to be achieved there have to be investments in new skills, new teamwork and different kinds of facilities. 2. Radiotherapy. Brachytherapy offers a potential for treatment that improves targeting on affected areas and reduces side effects such as incontinence or impotence in treatment of prostate cancer; however, it requires more planning time by physicists and investment in new more powerful linear accelerators. 3. Chemotherapy. There is a range of new and more powerful therapies and an increasing gap in costs between standard treatments using generics and those which use the new biotech therapies. There is more change likely in therapies available in the next five years than in the previous two decades.
4. Care pathways. The increasing range and effectiveness of treatment is creating new standards for integration and quality of care.
CASE STUDY OF THE ECONOMICS OF CANCER: THE USA Services in the USA have been the pathfinders for world cancer services in terms of improving outcomes changing models of care – and rising costs. The US experience can show how the economic dimension is likely to impact. Data from the 1980s show how the burden of cancer was very large (Table 57.1). Since 1985 we can estimate that the pattern of costs has changed as follows: ●
●
●
●
Reduction in lung cancer has changed mortality costs but they still remain high. Even so, cancer was still the most important cause of life years lost with a total of 8.4 million life years lost compared with 7.8 million for heart disease and 7.0 million for all other causes. Mortality costs will be affected mainly by the changing incidence of lung cancer, which accounted for 2.3 million of the life years lost in 2001 – far more than for any other kind of cancer. Morbidity costs have risen with increased prevalence/ survival. Changes in insurance/Medicare have increased the share of out-of-pocket costs in this total. Treatment costs have been stable as a proportion of total health-care spending.
After the War on Cancer, initiated by President Nixon in 1971, spending on cancer treatment initially rose as a proportion of the total: this increase was not sustained (Table 57.2). The data show that the cancer treatment has been treating more patients with a constant share of spending. Some of the pressures towards spending on technology and new drugs have been offset by reductions in hospital stay and admissions, made possible in part by new antiemetic drugs. Total spending on cancer care has risen faster than direct treatment costs with increased spending on prevention and screening. In addition to the published
Table 57.1 Economic costs of all illnesses and cancer by type and cost, United States, 1985 Type Total Direct Indirect Morbidity Mortality
All illnesses amount
Per cent (%)
Cancer amount
Per cent (%)
Per cent of all
679 712 371 400 308 312 80 850 227 462
100 54.6 45.4 11.9 33.5
72 494 18 104 54 390 7170 47 220
100 25.0 75.0 9.9 65.1
10.7 4.9 17.6 8.9 20.8
Source: Brown et al. 1996.1
1354 The economics of cancer care
Table 57.2 National cancer treatment expenditures in billions of dollars, 1963–2004
Year 1963 1972 1980 1985 1990 1995 2004
Cancer treatment spending
Total health-care spending
Per cent of total (%)
1.3 3.9 13.1 18.1 27.5 41.2 72.1
29.4 78.0 217.0 376.4 614.7 879.3 15 340.7
4.4 5.0 6.0 4.8 4.5 4.7 4.7
Source: NCI.2
estimates, which are due for up-dating from the NCI, it is possible to make estimates of the total spending at each stage of the care process. There are also estimates for spending on specific types of cancer, which show differences in cost between cancer types (Table 57.3). There have also been estimates of cancer spending on patients under 65, which could be compared with Medicare-based data on spending for older patients. These data show wide differences depending on the levels of insurance coverage for patients under 65. They estimate that the cost of a government programme to cover all patients under 65 would range from $30 billion to 41 billion. For countries outside the USA there are some clear pointers to the future from the US pattern of costs. Cancers with either a high initial cost (colorectal) or longer-term recurrent cost are likely to rise as proportions of the total. Increasing subspecialization is likely to raise total costs. The US data show that for many less common cancers spending fractions are higher than incidence. Cancer treatment has been relatively low cost in the past as a declining proportion of total expenditure was being allocated over a rising total number of patients, but this position is likely to change in the future. More active development of new therapies for an increasing patient base will increase spending on cancer care much more rapidly. Cancer screening costs have risen from a low level 20 years ago to $10 billion or more, which is 20–25 per cent of spending on active treatment. US practice has combined increasing concentration of services for rarer surgical procedures with the development of ambulatory oncology centres for local access for the more common cancers and procedures. The Surveillance Epidemiology and End Results (SEER) database has allowed accurate identification of success rates in surgery. For removal of the pancreas the mortality rate in low-volume hospitals was double that of highvolume hospitals. Lung resection was also more likely to be successful in high-volume hospitals. ‘The magnitude of the benefit was large, equating to more than 100 lives a year’.3
There is less research evidence on the performance of ambulatory care centres, but these centres have been able to achieve very rapid moves from symptoms to diagnosis to treatment (two weeks or less.) These centres have also had incentives to reduce hospital admissions, which have contributed to the US shift away from in-patient treatment. One centre attributed the change to increased use of anti-emetic drugs. ‘This has been a revolution – we have been able to treat twice as many patients ten years ago with virtually no hospital admissions.’ In 2002, 7 per cent of adults in the USA had been diagnosed with cancer, which accounted for at least 30 per cent of premature life years lost. Against this background the actual use of health services is quite limited. Patients with cancer made 16 million visits to office-based physicians in 2002 (1.8 per cent of all visits.) They had 1.2 million hospital discharges or 3.5 per cent of discharges. The number of discharges had fallen by 8.5 per cent since 1994. Many of the admissions which did take place were by elderly patients so that the average length of stay was higher than the average – at 7.1 days. The US evidence may also show how difficult it will be to achieve further improvements when survival has reached towards 90 per cent for many early-stage cancers (from this perspective cancer was already becoming a survivable illness). In addition to the successes with early-stage cancer there are also some more specific successes to record: ●
●
● ●
A fall in the incidence of lung cancer among men. Lung cancer incidence rates among men declined by 1.8 per cent a year from 1991 through 2000 and death rates declined by 1.7 per cent a year. By 2000 there were fewer patients diagnosed with lung cancer and fewer deaths compared with 1980. However, the incidence of lung cancer rose among women. Improvements in cure and survival for children with cancer. Improvements in cure for testicular cancer. Reductions in incidence of stomach cancer.
For lung cancer the gains are due mainly to smoking cessation. For children’s cancer and testicular cancer the gains would seem to be due to improved treatment. For stomach cancer the causes are contested but due at least in part to changes in diet. There is new evidence on the gains to screening. Screening programmes were evaluated in terms of the gains in numbers of cases detected, but it is now clear that possible later gains in survival through detection of disease at an earlier stage should be taken into account. Earlier estimates based just on gains in sensitivity are now inadequate. The US evidence allows tracking of the interactions between screening and treatment in ways which were not possible with early studies such as the Swedish two counties study which were carried out before there was evidence on the longer-term effects of treatment.
Case study of the economics of cancer: the USA 1355
Table 57.3 Estimates of national expenditures for medical treatment for the 13 most common cancers, 2004 Type of cancer Breast Colorectal Lung Prostate Lymphoma Bladder Cervix Head and neck Leukaemia Ovary Melanoma Pancreas Oesophagus All other Total
Per cent of all new cancers (%)
Expenditure in 2004 (US dollars)
Per cent of all expenditure (%)
Average medicare payment in first year following diagnosis
15.9 10.7 12.7 16.8 4.6 4.4 0.8 2.8 2.4 1.9 4.0 2.3 1.0 14.0 100.0
9.6 8.4 9.6 8.0 6.3 2.9 1.7 3.2 2.6 2.2 1.5 1.5 0.8 13.4 72.1
13.3 11.7 13.3 11.1 6.3 4.0 2.4 4.4 3.7 3.1 2.0 2.1 1.1 18.5 100.0
11 200 24 200 24 700 11 000 21 500 12 300 20 100 18 000 18 100 36 800 4800 26 600 30 500 20 400
Source: NCI.2
Table 57.4 Relative 5-year survival rates (%) for breast cancer Country Canada (Manitoba 1985–89) Canada (Ontario 1985–89) France (1985–89) Italy (1985–89) Japan (1992) Norway (1990–94) Sweden (1989) UK (England 1993–95) USA (1989–95)
40–49 yrs
50–59 yrs
60–64 yrs
65–69 yrs
70–79 yrs
80 yrs
All
78.5 79.4 82.6 82.2 90.5 80.5 81.0 79.5 82.6
76.5 75.7 79.6 75.8 85.9 79.2 79.0 81.7 82.5
76.9 75.9 88.0 77.6 86.3 75.2 88.0 77.5 84.7
82.1 80.9 81.2 78.6 N/a 79.8 N/a N/a N/a
77.7 77.5 83.2 82.2 81.4 74.1 85.0 69.6 82.7
79.4 68.4 78.4 75.7 76.4 74.6 73.0 53.0 N/a
78.4 76.5 82.0 79.0 84.9 77.9 82.2 74.1 83.8
Source: OECD 2003.4
Health Maintenance Organization (HMO)-based data now allow evidence on the longer-term effects of screening. A retrospective study analysed persons detected with colorectal cancer from 1993 to 1999 in Group Health Cooperative – a large HMO in Washington State. Total health-care costs were compared for screen-detected versus symptom-detected individuals. In the three months before diagnosis total costs were $7436 for persons with screen detected cancer compared to $10,042. 53 per cent of screen-detected cancers were at an early stage compared to 30 per cent of symptom-detected cases. Overall cost for the screendetected group was $22,639 in the 12 months after diagnosis compared to $29,471. There are still questions, however, as to whether the US results are any better than those in other developed health systems. What is the record so far in terms of economic
evidence? Has the US bought better outcomes for patients with the distinctive scale and intensity of treatment? For coronary heart disease treatment, activism allied to lifestyle change seemed by the end of the 1990s to have created a clear pattern by which outcomes and survival were much better in the US. Age-adjusted death rates for heart disease and stroke have fallen by 59 and 69 per cent since 1950, but even in breast cancer treatment, which has received more funding than other types of cancer, there was little distinctive about results in the US compared with other advanced countries. Some direct evidence is now available from a special study carried out by the Organisation for Economic Cooperation and Development (OECD).4 The OECD study of international outcomes in breast cancer is detailed and thorough: it shows a similarity of results for younger patients across nations with the UK as an outlier for its poor results with older patients (Table 57.4). The US survival rates are almost the same as for France and Sweden. Other evidence within the US confirms
1356 The economics of cancer care
that gains in improving outcomes in breast cancer have been slow: ●
●
A study of patients with breast cancer treated at the MD Anderson Center in Texas showed that there were some survival gains for patient with metastatic or later-stage cancer, but the rate of improvement seemed to be similar to or even slightly below that found in a large-scale study of patients with metastatic cancer at 20 cancer centres in France from 1980 to 1999. ‘Using statistical adjustment techniques, we estimate that there has been a reduction in risk of approximately 1 per cent for each increasing year although the change was not statistically significant.’ Overall, the improvement in breast cancer survival has averaged 3 per cent per year. Earlier detection with treatment of more disease has been the key variable.5 A cost effectiveness study of technological change in medicine concluded that for patients with heart attacks change in treatment costs of $10 000 had contributed to a year-on-year increase in life expectancy valued at $70 000. For breast cancer a change in treatment costs of $20 000 had only brought about a 4-month increase in life expectancy valued at $20 000 so that technological change was neither beneficial nor harmful. In other areas such as the treatment of heart attack, low-birthweight infants and cataracts, the net benefits for additional treatment had ranged from $70 000 to 240 000.6
In summary, there are also some significant areas in the impact of cancer services. 1. There has been little progress in improving survival for patients with lung cancer. 2. There are still significant differences in outcomes (in the US as in the UK) between social and ethnic groups. For all cancers in 2001, mortality was 25 per cent higher for black Americans, mainly reflecting poorer survival rather than higher incidence. The National Cancer Institute is now concentrating much more development effort on these issues. 3. The long-term quality of life of cancer survivors is much poorer than was hoped for7 with 50 per cent of long-term survivors recorded as having very poor quality of life. An official report in 2004 estimated there were 9.8 million cancer survivors in the US in 2001.8 The report showed that: – Sixty-four per cent of adults whose cancer is diagnosed today can expect to be living in 5 years time. Omitting lung cancer, the percentage would rise to 80–90 per cent. – Breast cancer survivors make up the largest group of cancer survivors (22 per cent) followed by prostate cancer survivors (17 per cent) and colorectal cancer survivors (11 per cent). – The majority (61 per cent) of cancer survivors are aged over 65 and cancer survivors constitute 1 in 6 of the age group. Thus there are very significant new challenges in raising quality of life for longer-term survivors.
CASE STUDY: IMPROVING SERVICES IN NEW EUROPE We would sum up key findings as follows: 1. The current disease burden in cancer for the new member states (NMS) is not widely recognized. Cancer is responsible for 30–40 per cent of lives lost prematurely under 65 and as mortality from coronary heart disease can be expected to reduce with smoking prevention and change in diet, cancer is going to become relatively more important as a cause of premature mortality. 2. With current trends in old (lung and colorectal) and new (breast and prostate) cancers the disease burden is set to increase. There will be a high contribution from early mortality but there will be an increased contribution from disability and longer-term morbidity. 3. Evidence is available on outcomes from the EUROCARE 3 survey. The data cover patients diagnosed in 1990–94 who survived for 5 years. Much of the treatment would have taken place before recent increases in spending on health. In all, the results are a remarkable tribute to the abilities of clinicians in the NMS in achieving such survival levels with very limited resources. In effect, the survival rates were roughly the same as those being achieved in the EU in the late 1970s but with much more limited resources. 4. The level of expenditure is low but spending has risen, driven by equipment purchase and therapy expansion. Services are fragmented and there is little sense of coherent strategy or investment. On current outlook there is a very real chance that expenditure will increase but with little gain in improved outcomes and real value. The main strength in the past has been in the availability of dedicated clinicians who were working for very low salaries. However, this strength will decline with increased migration in the EU and the movement of younger doctors into other occupations and countries. The NMS cannot expect that approaches that have served them well in the past will automatically carry on in the future. First, it was clear from our visits to leading cancer institutes in the Czech Republic, Hungary and Poland, that in all three countries at the highest level there was an appreciation of these challenges and a degree of investment in new diagnostic techniques, such as genetic profiling associated with products such as Herceptin. One would anticipate with some confidence that, in these leading tier centres, the necessary funding and development of new skills and multi-functional team approaches will take place as more new products of this type become available. In second- and third-tier hospital oncology centres, already embroiled in complex and uncertain re-organizational schemes requiring concentration and a general
Case study: improving services in new Europe 1357
upgrading of existing chemotherapy services, responding quickly to the potential of new wave of high cost drug therapies will be a formidable challenge. A key element in surmounting this challenge is likely to be a national plan that places even greater emphasis on networking between leading and secondary centres. It was also clear that good progress has been made over the past decade in overcoming decades of neglect. In many of our interviews there was complaint about the fragmentation of services. There are divisions between specialties, with urologists and gynaecologists involved in treating cancers as well as oncologists. There is also fragmentation between oncology centres with little coordination of role or workload. These comments reflect the situation as it was in 2004. Among the specific comments were: Poor follow-up after initial treatment. (Patient group, Czech Republic) There is a big problem of fragmentation between centres – a weakness in follow-up and adjuvant therapy. (Leading specialist, Prague) We often meet very advanced cancer. (Specialist in regional centre in older industrial area, Czech Republic) In the last 10 years there has been good progress – patients are being recognized earlier: but diagnosis is frequently left too late. (Specialist, Silesia) Life expectancy here is worse than North Africa. (Specialist, Hungary) Only 25 per cent of breast cancer patients get the most modern follow-up treatment. (Specialist, Hungary) The system faces very divergent pressures from producers....Money does not follow the patient especially in oncology. It is divided among many hospitals. (Health fund manager, Czech Republic) Of 200 hospitals in the Czech Republic some 80 treat cancer. (Health fund manager, Czech Republic) The situation of a patient in the last stages of cancer is quite terrible. There is no system of home care and we only have 2 hospices and 120 terminal care beds rather than the 200 needed. (Health fund manager, Czech Republic) On a more positive note: We act as a virtual centre for South Moravia. We are the only comprehensive cancer centre in the Czech Republic. (Specialist, Brno) In 1997 there was a new law on patient’s rights. This changed a culture of belief that it was better to lie. The new move requires cooperation between physician and patient. (Patient association, Hungary)
The Soros Foundation has been active in funding palliative care and a specialist nursing service. (Specialist, Hungary) Until the 1990s palliative care, as an organized specialty with dedicated staff teams, was limited to a very few countries (mainly the UK, the USA and Australia). Now there are signs of wider interest. There is greater interest both in hospices and in home care teams. For Hungary, there has been important new funding from the Soros Foundation. With improved survival, the role of palliative care becomes more important and also cost effective in saving on emergency admissions and treatment in acute hospitals. ‘The aim can be a highly positive one – that of securing quality of life and freedom from pain in the last phase’.9 Palliative care can also give support to carers through respite care and help with disability. The NMS now have an opportunity to learn from the experience of others for new local initiatives in palliative care. In the short to medium term, the growing incidence of cancer cases and only modest improvements in outcomes for some major categories such as lung cancer clearly suggest that the prevalence of terminally ill patients is also likely to rise. A common social pattern was noted by many of the interviewees, both administrative and professional, on the care for this type of patient. Traditional locally resident extended families are in decline even in rural areas. As a result, the traditional approach to terminal care at home, under the supervision of a general practitioner, is also in decline. Modern nuclear families, particularly those with children, are increasingly reluctant to take on this role. Thus, the proportion of patients in need of palliative care, dying in hospital, is correspondingly increasing. In terms of hospital bed availability, currently this is not an issue as there is substantial surplus capacity but, if current reform plans to streamline and pursue ‘cost efficient’ strategies to reduce hospital bed capacity and hospital stay periods come to fruition, in parallel a new, more pro-active, strategy for palliative care for cancer patients will be needed. At the present time, initiatives aimed at creating specialist facilities with dedicated staff, trained in techniques such as psychological support and pain relief, could best be described as embryonic in all countries. It should be noted in fairness that, across the EU as a whole, there is a considerable variation in this regard, ranging from highly developed systems similar to the USA, to very few limited care facilities, run by charitable and religious organizations. It appears self-evident that, within health-care reform plans reviewing or further developing a holistic national cancer plan, attention needs to be given to provision of palliative care facilities and staff. This is likely to require shifting the balance away from patients remaining in hospital to care at home with a much stronger specialist support system, or in dedicated ‘hospice’-like facilities. The NMS face challenges in reducing the disease burden from cancer. It is already much higher than in the old EU
1358 The economics of cancer care
and is set to rise further without effective action. Levels of expenditure on cancer services, mainly for treatment, are currently low at 3–5 per cent of total health expenditures. ●
●
●
●
Survival outcomes achieved in the 1990s were about two thirds of those in Western Europe. This was a remarkable achievement given the low level of resources available for treating patients diagnosed in 1990–94. Without modernization, the core strength – of commitment from dedicated clinicians – is likely to be eroded as some younger professionals migrate. There will be problems in meeting new and challenging standards for clinical governance. There is a new model of cancer care, which leads towards balanced and co-ordinated investment in prevention, screening, and treatment and follow-up care. The model also supplies many opportunities for international partnership, where greater participation in clinical trials has already been a good start. Use of this model is essential to achieving reductions in cancer incidence and mortality. We are impressed by the potential for services in the NMS to catch up, but the opportunities must be taken soon.
CASE STUDY: IMPROVING THE PATIENT EXPERIENCE The cancer treatment services have been developed for short episodes of treatment and have been distinguished by intensive professional direction: now they have to adapt to a new situation where they will need to offer more information on options and continuing support.10 New approaches are needed and a new commitment in order to improve the quality of life of survivors. In the past, cancer patients have complained more than others of lack of information during treatment. Survivors have been found to have a quality of life that is significantly poorer than that of their peers.7 Data for the USA8 show that: ●
●
●
●
The number of cancer survivors rose from 3 million to 10 million from 1971 to 2001. Sixty-four per cent of adults whose cancer is diagnosed today can expect to be living in 5 years time. Breast cancer survivors make up the largest group of cancer survivors (22 per cent) followed by prostate cancer survivors (17 per cent) and colorectal cancer survivors (11 per cent). The majority (61 per cent) of cancer survivors are aged 65 or older.
There have been particular concerns about support for adolescent cancer patients.11 In the USA, longer-term followups have shown that survivors of childhood cancer are at risk of a range of complications later in life, sometimes up to 20 years after treatment.12
This increased number of survivors, which will also follow across Europe in the next few years, will result in significant changes for patients who will experience living with cancer as a chronic disease for many years or even decades, not just days or months. Patients will be concerned about the impact of treatment on quality of life in the short term and on their ability to function in the longer term. They will fear loss of income and, for younger patients, problems of employment. They will have to deal with the social and economic effects of cancer treatment as well as the physical side effects of the disease and its treatment. At present there is survey evidence that cancer patient services will have to develop in a number of different dimensions. These will include: 1. communication with patients and carers on treatment choices; 2. increasing the role of primary care in providing support for cancer patients as well as in early identification and diagnosis; 3. developing choice in location and timing of treatment; 4. improving cancer rehabilitation and longer-term support; 5. improving access to complementary therapies; 6. improving palliative care and end-of-life support. Patients often express frustration at the minimal information they receive. The President’s Cancer Panel recently convened a meeting with European cancer survivors.13 Speakers, particularly those who were longer-term survivors, noted that when they were diagnosed, they were given little or no information about their disease. Further there were no readily available information resources, and no internet, now often the most relied upon information source by both newly diagnosed and longer-term survivors.13 Some of the testimony from patients showed just how far there is to go: I was always . . . looking for his right ward, for the room in which to undergo the CAT scan . . . for scintigraphy, for x-rays, bronchoscopy, et cetera. And all of this without synergy, and often, I myself had to explain to the doctor on duty what was wrong with me . . . Everything is in my hands, my fight against cancer, my fight against the treatment by the medical professionals, and lastly, my fight against the system of how patients are followed up. (Antonio Toscano, 47, lung cancer survivor, diagnosed 2002, Italy)13 It is so easy to do it differently: an open attitude to questions, interest in your personal situation, a telephone number to call in case of panic. These are really experiences that help. (Fenna Postma-Schit, 60, thyroid cancer survivor, diagnosed 1978, the Netherlands)13
Case study: improving the patient experience 1359
There are quite simple changes that could make a difference for most cancer patients in helping them through the patient journey. There are also some specific treatments where it may be vitally important to give patients much more information. Five-year follow-up of patients with prostate cancer has shown that those who are treated by surgery are much more likely to have serious problems with incontinence and impotence than those treated with radiotherapy. In the UK the number of transurethral resection of the prostate (TURP) operations has in fact halved since there was more awareness of these effects. It is now part of good professional practice – even minimum clinical governance – to ensure that patients are informed about the risks and benefits of treatment. A patient in the UK diagnosed with colorectal cancer in 2002 set out his views on how communication could be improved: Primarily I would have liked the initial phase to have been more rapid! Waiting six months for a diagnosis was not really acceptable and it seems amazing that the ulcer was not seen with the rigid proctoscope. Another doctor I spoke to suggested that I was lucky that it didn’t take a year. I was only seen at the hospital when I was because I kept pushing. Once everything was underway, I have no complaints about my treatment, though the thought that I would not have had oxaliplatin other than as part of the EXPERT trial is very frightening! I would have liked a formal mechanism to obtain copies of all tests, scans, etc. at every stage. Some of these I obtained, but this was only on the basis of asking and hoping that someone did something about it. The colorectal specialist nurse and research nurse were very good at organizing this early on but it hasn’t been maintained. (Personal communication, Dr. Andrew Martin, 2004). Among the most powerful accounts of poor communication and denial of access to treatment was the President’s Cancer Panel Report Voices of a Broken System.14 Even well-informed patients in the USA report great problems about access to care. When you are fighting for your life which I was, it is virtually more than you can do to also fight the system every inch of the way . . . I also want to point out that I am a middle class person. I have supportive family. I have an incredible network of friends. I’m also about to serve in my eleventh year in the legislature, so I have skills, and knowledge, and at least perceived power and perceived access to the press that many people don’t have, yet despite all of that I had an incredible struggle to get all I needed for myself. (Karen Kitzmiller, stage IV breast cancer patient and state legislator, Vermont [deceased])14 There is also some survey evidence showing more positive developments. One survey of long-term survivors of
lung cancer showed that 50 per cent said that cancer had helped them to view their lives more positively and 71 per cent described themselves as hopeful about the future.15 Even so, 22 per cent of respondents had distressed mood and scored low on depression scales. A study of long-term survivors (average of 6.3 years earlier) also showed that quality of life was good: ‘This was a large group of women, and the good news is that we could find very little wrong with the emotional or physical health of most of them’.16 For patients at the treatment stage there is now strong evidence from many countries that patients want more information. Many studies have suggested that patients with cancer believe they are not given enough information. Cassileh and colleagues found that most cancer patients in the USA wanted detailed information about their cancer whether the news was good or bad. This finding has been confirmed in surveys of patients with cancer in the UK, North America and Australia. A large study in the UK, involving 2231 patients found that 87 per cent of participants preferred to have as much information as possible, good or bad.17 There are a number of simple ways in which communication could be improved. Mills and Sullivan18 listed six functions of information for patients: to gain control, to reduce anxiety, to improve compliance, to create realistic expectations, to promote self care and participation, and to generate feelings of safety and security. The key step here is to start and build on e-mail communication so that each patient can have a care programme. A recent survey of breast surgeons in the UK showed that they were making virtually no use of the internet at present – in fact there was far less use than in many more mundane areas of life.19 E-mail communication can give a list of useful web sites for patient support groups and information on treatment. It could give a contact point for patients with queries in the event of emergency. We would see such a service as being well within the reach of most larger centres. In systems and countries with primary care gate-keeping, the role of primary care in cancer services is bound to expand from initial referral through to palliative care. The accuracy of referral affects the speed of diagnosis and staging. Family doctors will often be asked for advice on therapies. They will have to manage longer-term support programmes and they will have the main role in shared care. In the UK the original Calman Hine report stressed that the role of the GP was central but little seems to have happened to follow this up effectively. A Report by the National Audit Office in the UK rightly emphasized the priority – it remains to work out how to achieve it in practice.20 Cancer care is often regarded as a province for specialists only, yet with increased numbers of patients and a requirement for longer-term support the role of the primary care team is set to increase. Initial referral is a key area where
1360 The economics of cancer care
there has been a set of guidelines since 2000 for the referral of suspected cancer patients, providing detailed guidance on how to identify patients requiring urgent referral. Evaluation is needed as to whether the guidelines actually achieved the results expected from earlier detection but there is a positive view of the results so far for breast cancer. The aspiration is clear enough – it remains to carry out more projects on how to develop it in the most effective way. We would see the opportunity to develop shared-care models and for cancer centres to involve GPs with special interests. For some types of cancer, such as prostate cancer, some of the initial diagnostic assessment is being carried out in primary care. Clearer protocols and clinical governance standards will make it much easier to involve primary care teams in the cancer journey. Until now the role of primary care in cancer services has been mainly concerned with palliative care,21 but there could be a more extensive role in developing care pathways and in monitoring risk. More generally, patient preferences and capabilities are going to become much more important as cancer becomes a longer-term illness with many more elderly patients. The design and the location of services so as to ensure maximum patient involvement will become major issues. With more oral chemotherapy it will be possible to develop more local access at community hospitals and primary care centres. For older patients needing radiotherapy it may be possible to re-design programmes or to organize transport to more distant centres. Patients will want to know more about risks and benefits of treatment. Even though antiemetics have improved the patient experience there may still be important side effects from treatment. Some of the symptoms may clear up quite quickly but they will be quite painful. Until now, the key issues affecting organization and location have been those concerned with clinical specialization and concentration of equipment. These considerations have usually created an irresistible pressure towards concentration of services at a few locations. Of course such issues remain critical but it is surely time to ensure that evidence can be reviewed on customer/patient perspectives. When patients have to make many attendances, convenience of location may start to affect the take-up of therapies quite apart from considerations of patient convenience. Where there is scope for meeting clinical requirements in different ways it is surely time to search for degrees of freedom in how services are designed. In fact there may be unexpected gains to having more access and treatment at different points. The US system has had multiple access points with treatment much more widely available in physicians’ offices. Apart from some specialized surgery there has not been the same pull to specialization as in the UK where most treatment is being concentrated on as few as 50 hubs. The greater use of web and information technology systems may in fact mean that it would be much more possible to develop networks and to manage treatment programmes on different sites. It will also make it far easier to set consistent standards for quality.
Cardiac services include a strong element of rehabilitation. The presumption is that most patients can be helped back to the lifestyle of their choice. For the UK rehabilitation is one key standard in the National Service Framework for Coronary Heart Disease.22 As cancer becomes a chronic illness it is surely time to add in rehabilitation as part of the core programme together with advice on future options, financial arrangements, etc. For many patients cancer is an adverse economic event, even a disaster, as well as a disease. Within the USA patients may face heavy costs in funding drug therapies; in Europe economic effects may include loss of employment and inability to get insurance policies or credit. One study in the UK examined the social problems of cancer patients under eight headings.23 The categories were: 1. 2. 3. 4. 5. 6. 7. 8.
problems of managing in the home; health and welfare services; finances; employment; legal matters; relationships; sexuality and body image; recreation.
Most commonly experienced were difficulties in relationships, appearance and travel or getting around. Women patients experienced somewhat more problems than men and palliative care patients experienced more than others. It is likely that more such problems will be experienced with longer-term illness and declining general health. Urgent investment in better support is required if cancer is not to be associated with longer-term disability and loss of quality of life.
THE NEW COST PRESSURES: NEW THERAPIES By 2025, cancer treatment will be shaped by a new generation of drugs. What this new generation will look like is not apparent today and will depend on the relative success of agents currently in development. Over the next 3–5 years, we will understand more fully what benefits these compounds such as the kinase inhibitors are likely to provide. It is estimated that in 2004 there were about 500 oncology drugs being tested in clinical trials. Of these, around 300 were against specific molecular targets. But this number is set to rise dramatically. Two thousand compounds will be available to enter clinical trials by 2006, and 5000 by 2010. Many of these drug candidates will be directed at the same molecular targets and industry is racing to screen those most likely to make it through the development process. Tremendous pressures are coming from the loss of patent protection from the majority of high-cost chemotherapy drugs by 2008. Unless new premium-priced innovative drugs are available, cancer drug
The new cost pressures: new therapies 1361
provision will come from global generic manufacturers currently gearing up for this change. The challenge of funding future innovation is well illustrated by the projection of the likely global trend in expenditure on chemotherapy, which could well increase threefold by 2010. So what will these drug candidates look like? In 2004, small molecules were the main focus of research. Most of these were designed to target specific gene products that control the biological processes associated with cancer such as signal transduction, angiogenesis, cell cycle control, apoptosis, inflammation, invasion and differentiation. Treatment strategies involving monoclonal antibodies, cancer vaccines and gene therapy are also being explored. Although we do not know exactly what these targeted agents will look like, there is growing confidence that they will work. More uncertain is their overall efficacy at prolonging survival. Many could just be expensive palliatives. In the future, advances will be driven more by biological understanding of the disease process. Already we are seeing the emergence of drugs targeted at a molecular level: Herceptin, directed at the Her2 protein; Glivec, which targets the Bcr-Abl tyrosine kinase; and Iressa and Tarceva, directed at epidermal growth factor receptor (EGFR) tyrosine kinase. These therapies will be used across a range of cancers. What will be important in 2025 is whether a person’s cancer has particular biological or genetic characteristics. Traditional categories will continue to be broken down and genetic profiling will enable treatment to be targeted at the right patients. Patients will understand that treatment options are dependent on their genetic profile. The risks and benefits of treatment will be much more predictable than today. Therapies will emerge through our knowledge of the human genome and the use of sophisticated bio-informatics. Targeted imaging agents will be used to deliver therapy at screening or diagnosis. Monitoring of cancer patients will also change as technology allows the disease process to be tracked much more closely. Treatment strategies will reflect this and drug resistance will become much more predictable. Biomarkers will allow those treating people with cancer to measure if a drug is working on its target. If it is not, an alternative treatment strategy will be sought. Tumour regression will become less important as clinicians look for molecular patterns of disease and its response. By 2010, there will be more of a focus on therapies designed to prevent cancer. A tangible risk indicator and risk-reducing therapy, along the lines of cholesterol and statins, would allow people to monitor their risk and intervene. Delivering treatment early in the disease process will also be possible because subtle changes in cellular activity will be detectable. This will lead to less aggressive treatment. The role of industry in the development of new therapies will continue to change. Smaller, more specialized companies linked to universities will increasingly deliver drug candidates to large pharmaceutical companies for marketing.
In 2025, people will be used to living with risk and will have much more knowledge about their propensity for disease. Programmes will enable people to determine their own predisposition to cancer. This in turn will encourage health-changing behaviour and will lead people to seek out information about the treatment options available to them. Patients will also be more involved in decision-making as medicine becomes more personalized. Indeed, doctors may find themselves directed by well-informed patients. This, and an environment in which patients are able to demonstrate choice, will help drive innovation towards those who will benefit. However, inequity based on education, wealth and access will continue. At the present time there are large numbers of products for the major cancers based upon a range of different therapeutic strategies: monoclonal antibodies, anti-angiogenesis agents, kinase inhibitors, apoptosis inducers, gene therapy and vaccines. On the one hand, past experience suggests that many of these may well fail to deliver; on the other, the sheer scale and range of innovative activity would suggest that surely some will succeed. It is beyond the scope of this chapter to elaborate further on the technological aspects of these advances. However, what is important is to try to foresee, however tentatively, how these changes may affect future clinical practice and resource patterns in terms of funding, human resources and infrastructure. There will be a multiplicity of treatments that are clearly linked to the genetic profile of the patient, the specific characteristics of the tumour and its stage of evolution. This will herald the coming of ‘personalized therapy’ for the individual patient, or at least many more narrowly defined patient segments. To take full advantage of these new opportunities will require a major upgrading and investment in diagnostic techniques, both in terms of laboratory analysis and clinical imaging, and in the widespread adoption of more efficient management processes, which link diagnosis to the therapeutic team responsible for planning, implementing and following up on individual patient treatment regimens. The prospect of this potential for technological change raises even more difficult questions with regard to the economics of funding cancer in the future, which we introduced at the outset. More new technologies for cancer diagnosis and associated new drug treatment strongly suggest that, inevitably, cancer will require a bigger share of total health-care expenditure. Also, within cancer funding, shifts in planning priorities and allocation of funds will be needed to efficiently optimize their introduction for the benefit of patients. Even though new compounds may offer only a small health-care gain, they are avidly taken up because of the huge unmet medical need in this therapeutic area. The estimated 2004 sales of recently released molecular targeted therapies are staggering and will drive up cancer budgets in all EU countries.
1362 The economics of cancer care
The increase in the range and cost of drugs will raise new issues of choice. There will be new programmes to measure cost effectiveness through the use of benefit measures such as the quality adjusted life year (QALY). The National Institute for Clinical Excellence (NICE) has carried out a number of evaluations of new cancer drugs. These have mainly led to positive results for use of the new drugs. NICE has now completed 59 difference guidance documents on cancer treatment, including over 20 technology appraisals on the use of new drugs. The 100th appraisal was on the use of capecitabine and oxaliplatin for the adjuvant treatment of patients with stage III (Dukes’ C) colon cancer. The report concluded that standard treatments already used – 5-FU/FA – could raise overall 5-year survival from 51 to 64 per cent compared to surgery alone. The issue is whether the new more costly drugs are justified in terms of the lower rate of relapse and better survival. NICE accepted estimates that the cost per life year gained from using oxaliplatin was £24 952. The estimated number of eligible patients was likely to be 3868 and the likely cost impact of all the proposals taken together was £10.332 million or £14 000 for each 100 000 population.24
●
●
●
THE NEW AGENDA We have sought to set out the challenges to the teams who make cancer services across the world. In each nation there are few key decision makers and agents of change. They now have to face up to a new agenda. At first sight there might seem to be some rather major, indeed insuperable, difficulties in the way of continued progress in the twin aims of improving outcomes and quality of life. In developed countries we are already achieving 90 per cent survival rates for most early-stage cancers. There has already been a considerable expansion in screening which has produced very clear results only in cancer of the cervix. Access to chemotherapy has improved. The World Cancer Report left an abiding impression of the sheer scale of the problem with mortality of 2 million a year in developed countries and 6 million world-wide. On estimates for the USA, 46 per cent of men and 36 per cent of women will be diagnosed with cancer in the course of their lives and better survival will lead to a doubling in prevalence. We can identify two possible scenarios. In one we use limited resources in an effective way to achieve key aims within the model of staged cancer services. The international evidence is that a number of key actions can make a real difference to mortality in developed countries. Research and international service development over the last two decades have shown that the following actions are likely to be effective: ●
Prevention through smoking cessation. If we aim for a 20 per cent reduction in total mortality in developed countries and 40 per cent in premature mortality then
reductions in smoking rates are vital. On the evidence, there is no other way that such targets are going to be reached. Targeted screening. Selective expansion of screening programmes on a population basis can assist earlier detection both directly and indirectly through raising awareness more widely of the importance of early diagnosis. Developing one-stop ambulatory cancer centres. The international evidence from the USA points to effective results if there can be a maximum of two weeks from diagnosis to treatment. Most cancer patients in the UK and in Europe are taking months to move through the patient pathway. Speed in the diagnosis/treatment process certainly contributes to favourable outcomes in the USA and it also reduces the cost of treatment. Current services outside the USA are based on high levels of hospital admissions and lengths of stay for cancer patients. With less invasive surgery and the availability of anti-emetics the use of such high levels of admission in early-stage cancer is quite wasteful. There is a world-wide challenge of freeing up resources from hospital treatment to fund more intensive treatment. Activism works. More intensive chemotherapy and radiotherapy, sometimes in combination, improves survival. Again, the use of anti-emetics has reduced side effects for patients. Activism is also important in monitoring and treating recurrence.
The patient experience can and must be improved through improved communication, increasing access to complementary therapies and better palliative care. There is an alternative scenario in which we simply muddle on through expanding the current disconnected modules of service. There is a failure to achieve key targets in prevention so that mortality from lung cancer continues to be high. Screening increases mainly through repeated opportunistic screening of people in low-risk groups while full population coverage is not achieved. The treatment process is slow and fragmented and there is a multiplication of high-cost capital equipment in a few big cities. There are differences in access to new high-cost therapies and underuse of effective generics. Although survival improves, many survivors have a high level of disability and poor quality of life. Last, the aim of privacy, dignity and control in the last phase of life is not achieved. There is a shortage of palliative care and poor access to morphine. We have it in our power to bring about the first scenario. In developed countries there will have to be a very significant re-engineering of services to achieve the aims. In middleincome countries there can be a catching up so that they can raise 70 per cent survival rates to 90 per cent. They can also make great gains in prevention. In this chapter we set out feasible, fundable changes that could move forward. Successful change does not require large amounts of new spending or high-cost capital expenditure. We need a one-off
The new agenda 1363
investment in the new staged model of care followed by steady development over a period of years. This has to start from an appreciation of feasible options, based upon an emerging model of international best practice: ● ● ● ● ● ●
national cancer strategies; new integrated cancer programmes; screening; early detection and treatment; follow-up and secondary prevention; palliative care.
Accepting likely constraints of funding/structure, etc., how can there be movement towards the model so as to achieve improved outcomes, reduce disease burdens and progress even with likely fiscal strains? The need to develop and refine such holistic models as a basis for consensus building and investment planning is now gaining recognition with key contributions from the National Cancer Institute in the USA, Scandinavia and more recently the UK. This type of model replaces older ones, in which cancer treatment is seen as matter of an individual clinician waiting for patients to bring in symptoms. It involves a cancer strategy that seeks to maximize prevention, early detection, effective treatment and follow-on care involving management of risk and/or palliative care. This new approach rests on an evidence base which has been greatly strengthened. Two separate developments will determine the patient’s experience of future cancer care. Increasing expectations of patients as consumers will lead health and social care services to become much more responsive to the individual, in the way that other service industries had started to become in the latter years of the twentieth century, even in Eastern Europe. Coupled with this, targeted approaches to diagnosis and treatment will individualize care. People will have higher personal expectations, be less deferential to professionals and more willing to seek alternative care providers if dissatisfied. As a result, patients will be more involved in their care. They will take more responsibility for decisions rather than accepting a paternalistic ‘doctor knows best’ approach. This will partly be fuelled by the internet and competitive provider systems. In 20 years the overwhelming majority of people in their 70s and 80s will be familiar with using the internet to access information through the massive computing power that they will carry personally. With patients having access to so much health information, they will need someone to interpret the huge volumes available, helping them to assess the risks and benefits as well as determining what is relevant to them. These patient brokers will be compassionate but independent advocates who will act as patients’ champions, guiding them through the system. They will be helped by intelligent algorithms to ensure patients understand screening and the implications of early diagnosis. They will spell out what genetic susceptibility means and guide patients through the treatment options.
Patients and health professionals will have confidence in computer-aided decision-making because they will see that the programmes work, evidenced by: ●
●
●
●
● ●
●
● ●
●
new kinds of partnership between patients and health professionals; a stronger evidence base on care and treatment options, which will be accessible through web sites to local professionals; a shift from emphasis on items of hardware towards software systems; new informatics for measuring quality and clinical governance; greater emphasis on networking and teamwork; a commitment to learning and adaptation through national and international links; a shift away from in-patient care towards ambulatory care; targeting of patients at high risk for genetic reasons; targeting of disease prevention and treatment in socially deprived areas; a strategic role for health funds with use of managed care systems.
The social insurance funds and national ministries in many countries face a massive challenge in formulating long-term strategies for allocating limited funds to a range of services. Many face uncertain futures, in terms of income streams that are heavily dependent upon national economic performance. There are many other competing priorities besides cancer and, above all, the need to push through funding and organizational reforms on a grand scale. Somehow, within this context, the concept of a national cancer investment plan, supported by the relevant professional bodies, needs to be established with a clearer set of investment and funding priorities that are compatible with the wider reform programme. We would suggest that the following components should be high priorities in this plan: ●
●
● ●
●
● ●
●
●
stronger management capability in key local centres, especially in areas of systems design and utilization; expansion of screening programmes and management to ensure population coverage; funding survey research on the patient experience; designing new informatics for quality measurement and for networking among professionals; training of doctors and health professionals in new skills; design of protocols and care pathways; investing in new management and upgrading of diagnostic/treatment systems so as to ensure most effective use; developing international links and increasing research/clinical trial participation; organizing workshops and conferences to increase international cooperation in improving cancer services.
1364 The economics of cancer care
KEY LEARNING POINTS ●
●
●
●
●
Health professionals in cancer are going to be faced with new economic pressures from patient expectations and from increasing costs. As cancer becomes ever more a survivable longer-term illness there will also be new requirements for treatment planning over years rather than months and new choices about costly therapies to reduce recurrence risk. So far, cancer treatment can be counted as an economic and clinical success story within developed countries, but now there is a new set of challenges which could mean that future success could be compromised. There will be new pressures from increased workload and the greater intensity of treatment for many patients. There will be pressure to contain costs in systems where there are very limited amounts of funding for new therapies with many options for extending services in screening and prevention. There will be little increase in the hours available from experienced staff. Key issues are likely to be those of productivity and of targeting of therapies. The productivity depends on the system issue of the appropriate investments across the stages from prevention to after. There will be a challenge in finding the most effective range of investments. The new model offers the promise of better outcomes for patients from effective investments in prevention, screening and short-stay/ambulatory care. There is also the potential for improved quality of life for survivors from better care and even in the last phase of life for ensuring dignity, privacy and control for patients. Productivity also depends on the organization of services in specific units. There are opportunities here to improve patient flows and patient services so as to make most effective use of the time of experienced staff and to improve the experience of care. The balance between hub and spoke is likely to change with new technology, making it possible to have virtual hubs with more actual treatment locally. Technology may reverse the strong tendencies towards centralization in fewer units. For drug therapies the key challenge will be that of targeting so as to ensure that full use is made of information about genetic factors and that high-cost therapies are used where they are most likely to improve outcomes.
2 3
4 5
6 7
8 9 10 11 12 13
14 15
16 17
18
19
20 21
22 23
REFERENCES 24 1 Brown ML, Hodgson TA, Rice DP. Economic impact of cancer in the United States. In: Schottenfeld D, Fraumeni JF (eds)
Cancer Epidemiology and Prevention. New York: Oxford University Press, 1996. NCI 2006. http://www.cancer.gov/ Begg CB, Cramer LD, Hoskins WJ, Brennan MF. Impact of hospital volume on operative mortality for major cancer surgery. JAMA 1998; 280:1747–51. OECD. Health Care Systems: Lessons from the Reform Experience. Paris: OECD, 2003. Giordano SH, Buzdar AU, Smith TL, Kau SW, Yang Y, Hortobagyi GN. Is breast cancer survival improving? Cancer 2004; 100:44–52. Cutler D, McClellan M. Is technological change worth it? Health Aff 2001; 20:11–29. Yabroff KR, Lawrence WF, Clauser S, Davis WW, Brown ML. Burden of illness in cancer survivors. Findings from a population-based national sample. J Natl Cancer Inst 2004; 96:1322–30. Centers for Disease Control. Cancer Survivorship – United States, 1971–2001. Atlanta: CDC, 2004. Bosanquet N, Salisbury C. Providing a Palliative Care Service: Towards an Evidence Base. Oxford: Oxford University Press, 1999. Bosanquet N, Sikora K. The Economics of Cancer Care. Cambridge: Cambridge University Press, 2006. Hollis R, Morgan S. The adolescent with cancer – at the edge of no-man’s land. Lancet Oncol 2001; 2(1):423–48. Boughton B. Childhood cancer treatment causes complications later in life. Lancet Oncol 2002; 3(1):390. President’s Cancer Panel. Living Beyond Cancer: A European Dialogue. Washington: US Department of Health and Human Services, 2004. President’s Cancer Panel. Voices of a Broken System. National Cancer Institute 2001–2002. Sarna L, Padilla G, Holmes C, Tashkin D, Brecht ML, Evangelista L. Quality of life of long-term survivors of nonsmall-cell lung cancer. J Clin Oncol 2002; 20:2920–29. Senior K. Good quality of life found in long-term breast cancer survivors. Lancet Oncol 2002; 3(2):66. Jefford M, Tattersall MHN. Informing and involving cancer patients in their own care. Lancet Oncol 2002; 3(10): 629–37. Mills ME, Sullivan K. The importance of information giving for patients newly diagnosed with cancer: a review of the literature. J Clin Nurs 1999; 8:631–42. Pleat JM, Bailey J, Dunkin C. Postoperative web advice for UK patients with breast cancer. Lancet Oncol 2003; 4(9):527–8. National Audit Office. Tackling Cancer in England; Saving More Lives. London: NAO, 1964. Leese B, Din I, Darr A, Walker R, Heywood P, Allgar V. ‘Early days yet’. The Primary Care Lead Physician (PCCL) Initiative. University of Leeds, 2004. Department of Health. Coronary Heart Disease. National Service Framework. London: DOH, 2001. Wright EP, Kiely MA, Lynch P, Cull A, Selby PJ. Social problems in oncology. Br J Cancer 2002; 87:1099–1104. National Institute for Clinical Excellence. Capecitabine and Oxaliplatin in the Adjuvant Treatment of Stage III (Dukes’ C) Colon Cancer. Technology Appraisal 100. London: NICE, 2006.
58 Medical audit AMIT K. BAHL AND GARETH J.G. REES
Introduction Methodology
1365 1366
INTRODUCTION Sporadic audit has always been part of medical practice, observations of the results of interventions leading to improvements in care for future patients. In recent years there has been a growing recognition of the importance of devoting adequate time to audit, and of doing it in an organized, systematic and cost-effective way. Regular participation in audit is now a contractual requirement for doctors working in the National Health Service (NHS). The climate for the assurance of standards has moved on considerably in the past few years. Under clinical governance, consultants are required to demonstrate audit of their performance and results.1 Clinical audit is central to the NHS approach on clinical effectiveness, and the informed participation of health authorities, GP fundholders and NHS trusts is essential to its success and further development. Medical audit of radiotherapy dates back to 1898, when the Roentgen Society appointed a committee to collect data on the effects of X-rays. A report of the results of treatment was a condition of the loan of radioactive material to hospitals in the late 1920s by the National Radium Commission, and this led to the establishment of national cancer registration. Radiotherapists can therefore be regarded as being among the leaders of medical audit in the UK.2 Audit of a business involves preparing a profit and loss account that enables assessment of its success or failure, and identification of which elements of the business are fruitful and which are not. Audit in medicine has essentially the same objectives. It is an examination of activity
Examples of published audits References
1370 1375
that attempts to measure success or failure, and to identify remediable shortcomings. Improving care is dependent to a large extent on discovering what is being achieved at present. Most providers of cancer care still have little or no objective quantitative information on what they are achieving for most of their patients in terms of quantity or quality of life. This degree of ignorance about achievement would be surprising in any other area of human endeavour utilizing comparable resources. Medical audit has been defined as: 1. ‘The systematic, critical analysis of the quality of medical care, including procedures used for diagnosis and treatment, the use of resources and the resulting outcome and quality of life for the patients.’ 2. ‘The sharing by a group of peers of information gained from personal experience and/or medical records in order to assess the care provided to their patients, to improve their own learning and to contribute to medical knowledge.’ Medical audit is different from clinical audit. The former is concerned with the quality of care provided and supervised by doctors. The latter involves assessment of the service provided by the healthcare system as a whole and is a multidisciplinary activity. The purpose of medical audit is to improve the quality of care by striving to ensure that individual doctors practise medicine to the highest standard that can reasonably be achieved with the resources available, and that they use those resources most effectively. Audit should be regarded principally as an educational
1366 Medical audit
process, mainly through promoting discussion between colleagues about clinical practice. The quality of medical work can, to a large extent, only be reviewed by a doctor’s peers. Medical audit is an activity that is done by doctors and led by doctors. Attempts are often made to draw a clear dividing line between medical audit and research. The quest for discovering new and better ways of caring for patients is research. Audit is concerned with analyzing how effectively the knowledge gained from research is applied, with assessing and changing routine medical practice and improving standards. However, much of what has been described hitherto as clinical research has involved retrospective or prospective studies of groups of patients, their management and how they fared, with the aim of acquiring new knowledge which might help the management of future patients. This is essentially the first stage of medical audit. Clinical research and medical audit merge into one another. Indeed, the entry of patients into some types of prospective clinical trials, particularly those that involve already wellestablished treatments, should be viewed as an especially effective form of audit. Sitting somewhere in between the two disciplines of audit and research (although admittedly closer to research), health technology assessment (HTA) evaluates whether a piece of medical technology which has worked in a laboratory setting is actually effective in general use. Audit is central to any programme to enhance the quality of care. An effective programme of audit will help to provide reassurance to patients, doctors and managers that the best possible care is being offered with the resources available. While the primary purpose of medical audit is clinical rather than managerial, and its focus is the process and results of medical care rather than resources, the results of audit may have implications for strategic decision-making by administrators and politicians on the allocation of resources and on the organization of services.
METHODOLOGY General Many techniques can be used for audit. These include such different activities as informal discussion of individual cases; large surveys of the management of groups of patients; or the use of particular treatments and healthcare systems, comparative studies of the services provided by different doctors, departments, hospitals or even national healthcare systems. Audit can be concerned with examining the resources provided and the way in which they are used for patients, in the belief that these – ‘structure’ and ‘process’ – will have a bearing on the result, or ‘outcome’ of care. However, for cancer patients what matters most is the outcome – are they being given the maximum chance of both quantity and quality of life? Outcome is the ideal indicator of the
quality of patient care. Some measurements of outcome, e.g. survival, can be fairly easy to measure, while others, such as symptom relief or quality of life, are more difficult to quantify. Audit of structure involves examination of the quantity and quality of resources, including both staff and equipment. Structure is not in itself an indicator of the quality of care and should only be used in conjunction with assessments of process and outcome. It can be helpful to identify significant variations in the provision of resources to examine whether they do affect results. Surveys of provision of oncologists and radiotherapy equipment have identified substantial intra-national and international variations. In some instances, it is self-evident or already established, as a result of previous clinical research or medical audit, that certain structures and processes do influence outcome. However, the relationship between structure and process and outcomes is often more a matter of belief than science. Effective audit will establish areas where outcome, process or structure is unsatisfactory, identify causal relationships between process/structure and outcome, and set in train alterations in management aimed at improving the care for future patients. However, audit is only truly successful if it goes on to establish that the improvements hoped for have been achieved. In audit jargon this is known as ‘closing the loop’. There are few aspects of care that might not benefit from critical review. Audit can involve analysis at many different levels of organization, from examining the care given to a single patient to comparisons of outcomes for patients treated in different continents. In most instances, medical audit is primarily of local value, but sometimes the methods and findings are relevant to a wider audience and suitable for presentation at meetings or for publication. However, the extent to which published audits have impact elsewhere is probably overestimated. Someone else’s audit results are usually less powerful than a person’s own as an incentive to alter clinical practice. It is important that all doctors working within a department, including those in training, are fully involved in medical audit. It is also vital that paramedical staff, e.g. nurses, radiographers and physicists, are fully involved in clinical audit dealing with topics of interdisciplinary interest. Medical audit must be conducted within a clearly defined organizational framework if it is to be fully effective. It is best led by senior doctors. It will be more successful if there is maximum and enthusiastic participation. This is substantially dependent on the provision of adequate amounts of time and support. The latter includes personnel and appropriate computer hardware and software to facilitate the gathering and processing of statistical information. Enthusiasm also depends on the participants having freedom to decide those areas of medical care worth examining. Confidentiality is extremely important for participation as well as for medico-legal reasons. Peer review findings concerning individual cases and doctors should usually be absolutely confidential, but it is often desirable for the
Methodology 1367
more general or aggregated results of medical audit to be made more widely available. There must also be mechanisms for remedial action where mistakes are made repeatedly by individuals or groups. The knowledge that errors will be discussed may, in itself, be a stimulus to improve quality of care, but it is difficult to maintain high levels of attendance if audit meetings are not stimulating and enjoyable. The atmosphere at meetings should not be threatening or adversarial. Criticism should be constructive and it should be acknowledged that everyone makes mistakes and that many variations in medical practice are acceptable. It is also important that audit should not be allowed to discourage doctors from undertaking difficult clinical work just because there is an element of risk involved. The data for medical audit should ideally be readily accessible, timely, accurate, clinically relevant and as complete as possible. The quality of information produced by analysis of data is directly related to the quality of the data provided. The quality may be enhanced by involving clinicians in selecting the data to be collected and by collecting it as close to the activity that generates it as possible. It also helps to have regular involvement of those involved in data recording and collection in the audit process, and by keeping data collection within a realistic range. It is easy for medical audit to be overambitious. It is usually more sensible to audit particular aspects of care thoroughly and sequentially, rather than to try to examine several different areas simultaneously. Good audit is a continuous cycle. This involves observing practice, setting desirable standards, comparing practice with standards, implementing change and observing the new practice. Ascertaining what has been or is being achieved is really only the beginning of audit. Audit is not just about measuring where a doctor is – it is about measuring how far away the doctor is from where they want to be. Doctors could spend so long auditing themselves or other people that there would no longer be time to do the work that they are really meant to be doing. It makes sense to concentrate on those aspects that need looking at and to examine them in detail. If the time and resources made available for audit are to be used most effectively, the purpose of the audit should be made clear to all involved and topics for analysis should be chosen carefully, reflecting genuine interest and relevance. It is easy to waste large amounts of time and money on audit through a lack of clarity of purpose, unfocused data collection and inadequate motivation. Lasting change is usually very difficult to achieve and is greatly dependent on motivation. The topics most suitable for audit will be those concerned with activities that are common, risky or expensive, or with areas where there are wide variations in management. Audit should provide an important means for the review and development of departmental treatment policies. Ideally a rolling programme should be established, enabling the sequential review of the management of particular conditions. It is important that no major area of care is excluded from audit. Ideally audit should be
concerned not only with process and outcome, but with cost as well. Inequitably high expenditure on some patients will deprive a larger number of other patients of their entitlement, thereby reducing the average quality of care. The management of individual patients or groups of patients should ideally be judged against predetermined guidelines, which may have been established locally or nationally. The production of clinical management guidelines following consensus development is now commonplace across most areas of medical practice, but such guidelines are often not fully effective in promoting change without locally developed motivation and incentives. Medical audit may reveal deficiencies in care that involve a few patients at random, deficiencies that are confined to those patients treated under particular circumstances, e.g. by a particular doctor or unit or at particular times, or deficiencies involving the care of all patients. The cause of deficiencies must be discovered before they can be remedied. Most instances of deficient care will be caused by a lack of one or more of the following: 1. feedback (doctors do not know how their performance compares with what can be achieved) 2. knowledge or skill 3. motivation (e.g. discouraging working environment, personal health or emotional problems) 4. appropriate procedures (e.g. insufficient clarity about agreed clinical management policies or protocols, inadequate communication or documentation) 5. sufficient resources (equipment, drugs, staffing). It is important to keep written records of audit meetings. Records should include lists of those attending, some broad information on the topics discussed and the resulting conclusions and recommendations. These must be strictly confidential and the identity of patients, clinicians and other hospital staff must not be entered or be capable of being traced. A strict code of confidentiality must be established and maintained for all staff involved in audit, medical and non-medical.
Audit of audit Not all audit studies are as effective as they could be. Completed audit studies should themselves be audited. An assessment of any audit study should involve asking why it was done, how it was done, and what it found. The question asked should be simple and the methodology straightforward. All audit studies should have a clearly stated purpose. Ideally, the topic chosen should concern routine medical practice, be well defined and amenable to standard setting. This means the establishment of a consensus on what constitutes acceptable care and on the proportion of cases (which may be all of them) in which the care given should be at least as good as this.
1368 Medical audit
Such clear statements about the desired standard of care facilitate objective and quantitative comparison of the care actually given with that standard. This, where possible, is preferable to subjective assessment of the quality of care, dependent as it is on the vagaries of individual clinical judgement. Meaningful conclusions from audit studies are as dependent on sample size as research studies. There is little point in trying to form conclusions about clinical management policies and quality on the basis of comparisons of the outcome for different groups of patients if the observed differences could easily have come about by chance. In drawing conclusions it is important to remember possible causes of bias in the study, such as case mix, non-response from sending questionnaires, the failure of some patients to attend for follow-up, and the unavailability of certain types of case notes (e.g. those of dead patients). An ideal audit study will place emphasis on the implementation of change. It should not only identify room for improvement in the quality of care but should lead to recommendations, preferably specific, on how to remedy deficiencies, and then demonstrate that the recommendations have successfully been put into practice, thus completing the ‘audit cycle’.
Computers Systematic audit often requires the generation, storage and manipulation of large volumes of clinical information. Computerized clinical information systems are usually essential to support this type of activity. Collection and storage of data on patients and their management may also be required for resource management or research. Many different software packages have been developed to support audit. Most have been designed to run on personal computers, but others have been developed for multi-user operation as part of, or integrating with, hospital or regional information systems. Such multi-user systems can offer a more cost-effective approach to audit, particularly as medical audit and hospital administration have some common information requirements. However, it is essential that linkages between systems be such that access to audit information is strictly controlled. Several features may be important in choosing software. These include: the flexibility to tailor the programme to individual requirements; a word processing facility; the ability to select cases against preset criteria indicating unexpected or adverse events and to identify cases falling outside a defined range of values; and the potential for review of a particular topic (e.g. by diagnosis, complication or procedure). The system should be capable of simple data analysis or designed to be used with a statistical package. The mode of presentation of information is also important and a graphics facility is usually desirable. It is essential to take advice from appropriate experts before choosing a system but, where possible, it makes sense to use a system that has already been
proven successful. Benghiat et al have described a PCbased computer network that caters for the clinical information needs of a cancer centre, crossing speciality boundaries and involving all members of the multidisciplinary team.3 Data are captured at all stages of patients’ progress, from diagnosis through to treatment and followup. The system produces information on workload and process, as well as on clinical outcomes. Data are entered prospectively at the point of care by healthcare professionals, ensuring a high degree of clinical confidence. A dedicated computer information system for use in British oncological departments was developed through the Audit Office of the Royal College of Radiologists.4 Known as the Clinical Oncology Information Network (COIN), this system was designed to computerize clinical records, facilitate patient management through monitoring progress through the various processes of care, and summarize treatments and store data for audit. The use of a common system with agreed uniform standards for record keeping would facilitate the comparison of data between different centres.
Oncological audit Much of the management of cancer patients, particularly non-surgical treatment, is centralized in regional cancer centres. Oncological audit should not be confined merely to what goes on in the cancer centre. It should embrace the wider geographical perspective and also the multidisciplinary nature of the total management of cancer patients. Collaboration with general practitioners and with clinical and non-clinical hospital colleagues can be very fruitful. Useful audit can also be conducted through collaboration between centres, looking particularly at variations in process and outcome, and possible justifications and reasons for them. National and supranational societies and Royal Colleges can play a useful co-ordinating role. Wide variations still exist in the UK in the management of common cancers in adults. Development of, and adherence to, nationally agreed treatment protocols is a key measure in reducing variations in treatment and in outcomes for patients with cancer.5 The Royal College of Radiologists’ COIN project was created in 1994 on behalf of the Faculty of Clinical Oncology and the Joint Council for Clinical Oncology, to coordinate the production of evidence-based clinical practice guidelines in oncology and to identify data sets to audit compliance with the guidelines. The guidelines for prostate, lung, breast and breast cancer and generic radiotherapy and chemotherapy have been published in Clinical Oncology. These are available on the website of the Royal College of Radiologists, UK. INDIVIDUAL CASE REVIEW
This is one of the simplest types of audit. Cases are selected for discussion at regular meetings. They may be chosen at
Methodology 1369
Table 58.1 Examples of criteria for selecting cases for discussion Delay of more than a specified duration between referral and clinic appointment Delay of more than a specified duration between clinic consultation and start of treatment Delay of more than a specified duration between consultation and letter to GP Histology report not filed in notes Lack of specification in notes of TNM stage for patients receiving radical treatment Target symptom(s) not specified for patients receiving palliative treatment Lack of documentation of information given to patient and/or relatives No up-to-date serum haemoglobin estimation for patients embarking on radical radiotherapy Treatment not completed as planned Lack of documentation of termination or alteration of systemic treatment following evidence of tumour progression Death within a short time of completing radiotherapy or cytotoxic chemotherapy Duration of palliative radiotherapy more than 10% of remaining survival Death within 6 months of radical treatment Lack of documentation of contact with GP following death in hospital
random but some selection is required to avoid repeated similar discussion about the management of patients with common conditions. It may for example be useful to select patients with complications, or those who have died unexpectedly. Discussion on such cases is different from traditional ‘staff rounds’ in that the cases are selected and the discussion led by someone not involved in the patient’s management. A structured approach to case review is often helpful. Cases may be chosen for discussion on the basis of a list of preset well-defined criteria. Examples of such criteria are listed in Table 58.1. Large numbers of case notes may be screened by a non-medically qualified audit assistant, thereby making most effective use of doctors’ time for audit. The management of many of the cases selected for discussion on the basis of such criteria will not be found to have been deficient – often there will be acceptable reasons for what did or did not happen. However, such a process can highlight individual and systematic deficiencies in the service, and it can also illuminate areas where available resources are not being used effectively. Necropsies can be a fruitful component of individual case review. Many studies have shown that about one in 10 cases coming to necropsy have pathological lesions that would have materially altered clinical management if they had been identified before death. However, there
has been a decline in hospital necropsies in the UK in recent years. Communication of results of necropsies to hospital doctors, general practitioners and relatives is often inadequate. A joint working party of the Royal Colleges of Pathologists, Physicians and Surgeons recommended in 1991 that the relevant clinicians should receive a summary of the significant findings within a couple of days of the necropsy and a complete report within 3 weeks, but these standards are frequently not attained.6 Poor communication to clinicians may have contributed to the general decline in necropsy rates. In a study of lay perceptions of the results of necropsy, 88 per cent of relatives were reassured about the adequacy of medical care and benefited from explanations of their deceased relative’s illness. It is often not practical for doctors to attend necropsies, but there is scope for increasing considerably the use of photographic and video material at clinicopathological conferences. It is important that necropsies are themselves properly performed and audited. The working party recommended that histological examination should be part of every examination. It also recommended that, as well as necropsies performed for specific reasons such as verifying the cause of death, a necropsy rate of at least 10 per cent of other general hospital deaths should be the target. These two categories might together amount to a total of about a third of all hospital deaths. Discrepancies between antemortem and post-mortem diagnoses should be monitored and made available to consultants on an individual basis.
ANALYSIS OF GROUPS OF PATIENTS
Quantitative measurement of the process or outcome of care for defined groups of patients is a particularly important type of audit. The findings are often very illuminating in themselves, but sometimes they may also usefully be compared with results obtained elsewhere, and with consensus opinions on what are the optimal achievable standards of care. Variations in process tend to be greater than variations in outcome. The identification of substantial variations in the use of resources, e.g. radiotherapy fractionation, can lead to more cost-effective management. However, where there is variation in resource usage it is important to relate this to outcome before concluding that the cheapest policy is the most cost effective. The simplest and most important example of auditing outcome in oncology is the calculation of survival figures for patients with given tumours, stage by stage. The measurement of survival is particularly appropriate for cancer patients treated with curative intent and this long-established tradition accounts for oncological practice being in the vanguard of audit in twentieth-century medicine. Several cancer centres around the world now publish survival figures routinely. It is important that survival figures are evaluated critically. It may not be reasonable to expect figures quoted in
1370 Medical audit
Table 58.2 Examples of subjects for systematic audit in oncology Survival, stage by stage, for patients treated radically Intervals between presentation, diagnosis and treatment Waiting time in hospital departments Usefulness of investigations Adequacy of histopathological reports Frequency of on-treatment review of patients receiving radiotherapy Adequacy of excision margins Incidence and severity of complications of treatment, including iatrogenic death Prevention, management and outcome of complications Success in relieving target symptom(s) with palliative treatment Comparison of frequency of particular treatments, duration of treatment and hospital stay and policies of follow up between different doctors, institutions and geographical areas Conformity of investigations, drug usage and radiotherapy with agreed protocols Costs and cost effectiveness of care Patient satisfaction with total service provided, and with particular aspects such as the provision of information and involvement in decisions GP satisfaction with service provided
clinical research publications to be achieved in routine management elsewhere. There may have been deliberate or unwitting case selection, different staging procedures and publication bias. In comparing the results between different institutions, it is important to be satisfied that, as far as is possible, what is compared is like with like. Such studies must be particularly rigorous in their methodology and subject to monitoring by external assessors. Examples of other types of systematic analysis are shown in Table 58.2.
EXAMPLES OF PUBLISHED AUDITS Structure and process RESOURCE ALLOCATION
The Board of Faculty of Clinical Oncology of the Royal College of Radiologists conducted a survey of medical manpower and workload in clinical oncology in the UK.7 There were 240 consultants in post, no more than a decade before, despite an increase of 20 per cent in the number of new patients referred during the decade and increased complexity of clinical management. On average each consultant saw 560 new patients per year, 2–3 times as many as their counterparts in the USA and Europe. On average each consultant had 2000 patients under his or her care at any one time, of which approximately half would be receiving active medical care. Subsequently the Board of Faculty of Clinical Oncology of the Royal College of
Radiologists audited the changes in workload, megavoltage treatment facilities and radiographer staffing levels in all radiotherapy departments in the UK over a 5-year interval (1992–1997).8 It revealed a picture of growing workload outstripping the resources available to meet this demand. Treatment machine capacity had only increased by 3.6% despite an 18% increase in workload over the previous 5 years. This report provided a precise baseline assessment of the situation, suggested national targets to remedy the inadequacy, and detailed methods for individual departments to calculate their future resource requirements. An audit of clinical resources available for the care of head and neck cancer patients in England conducted by the Cancer Services Collaborative-Improvement Partnership involved a postal survey of cancer networks to define the services available for the diagnosis, treatment and support of patients with head and neck cancer in England. It was reported that multidisciplinary team working was evident in all centres. However, the teams were not always fully staffed, especially in areas of national staff shortages such as pathology, radiology, dietetics and speech therapy. Rapid referral pathways were present for initial assessment of patients but clinically significant delays were identified in the provision of complex investigations, in access to intensive care beds and for oncological care, especially radiotherapy. The authors concluded that major changes in service configuration are likely to be required if the current national targets of time to diagnosis and treatment are to be met.9
ORGANIZATION OF SERVICES
The Clinical Oncology Patients’ Liaison Group of The Royal College of Radiologists audited the ‘patient-friendliness’ of radiotherapy departments in the UK. The results demonstrated considerable variation in departmental practice and they were used as the basis for writing a booklet entitled ‘Making Your Radiotherapy Service More Patient-Friendly’, which was circulated to all departments. This was intended to give practical suggestions on improving services, recommending only what had been shown to be feasible. A year later the re-audit of the same services showed that in the great majority of departments the booklet had influenced staff and led to changes in practice and improved care. The service areas most influenced were the provision of facilities for waiting patients, the provision of information about treatment, the maintenance of privacy and dignity, and surveying the views of patients.10
DIAGNOSIS
Schmidt et al audited the reporting of 26 711 diagnostic breast-imaging studies performed over a period of 5 years.11 During the first 18 months several general radiologists reported on mammograms, whereas during the rest of the evaluated period only one radiologist was responsible for all mammography reporting. In the first 2 years the
Examples of published audits 1371
proportions of small cancers detected (Tis or T1a, b) were 27.2% and 25.7%, respectively. In the third, fourth and fifth years, these increased to 38.8%, 34.5% and 38.8%, respectively. The authors concluded that the detection of curable early stage breast carcinomas requires the dedication and commitment of a small group of radiologists who are willing to spend most of their time on this single subject. This will increase considerably the number of early stage cancers found and reduce the number of false-positive diagnoses. Many groups have studied delay in the diagnosis of cancer. An audit of the development of the nurse biopsy role conducted in Nottingham demonstrated that creating nurse surgery posts could enhance continuity of care and facilitate inter-professional collaboration, thus helping to ensure that patients receive both appropriate and timely care. Before the introduction of the nurse biopsy role there was an 8-week wait after an initial consultation for a biopsy by the doctor, and then a further 8-week wait for the removal of the lesion. Patients with suspected skin cancers now have a biopsy on their first visit to the clinic, potentially reducing both morbidity and possibly even mortality. It took 3 years to develop the nurse biopsy role in Nottingham. The nurse-led biopsy service is now being adopted nationwide, as recommended by the All Parliamentary Group on Skin (2003).12 PATHOLOGY REPORTING
Dey et al conducted a regional survey on the completeness of reporting on prognostic factors for breast cancer.13 The study population was 885 patients with invasive breast cancer diagnosed in Lancashire and Greater Manchester. Histological type, tumour size, presence or absence of tumour in vascular channels, and adequacy of excision were recorded for 95%, 91%, 49% and 86% of cases, respectively. The authors concluded that non-screening and low throughput laboratories were significantly less likely to record certain histopathological features. Keating et al audited the quality and completeness of histopathology reports of rectal cancer resections submitted to the National Cancer Registry in 2000 in New Zealand.14 Scores from teaching hospitals, public nonteaching hospitals and private laboratories did not differ significantly. Multiple staging systems were used in 40% of reports and no stage was allocated in 31% of reports. Circumferential margin involvement was recorded in 63% of reports. The authors concluded that no significant differences existed in the quality of pathology reporting of rectal cancer between different laboratory types, either public or private. There was however an unacceptable lack of uniformity in reporting tumour stage. Circumferential margin involvement was frequently omitted in spite of its documented value as an indicator of quality of rectal cancer surgery, as an important predictor of local recurrence, and its more recently established value as a marker for distant metastasis and survival.
RADIOTHERAPY FRACTIONATION
A survey of British radiotherapeutic practice undertaken by Priestman et al on behalf of the Royal College of Radiologists discovered a wide variety of dose/fractionation regimens.15 In only one situation – the palliative treatment of carcinoma of the bronchus – did more than 25 per cent of oncologists use the same treatment schedule. Training and established local policies, rather than the results of clinical trials, emerged as the major influences on practice. While the degree of therapeutic variation in radiobiological terms was far less marked, the survey did identify a need for critical appraisal of treatment schedules to ensure the most cost-effective use of resources. Almost 15 years later a 1-week national survey of radiotherapy fractionation practice in 2003 conducted by the audit sub-committee of the Faculty of Clinical Oncology of the Royal College of Radiologists concluded that UK radiotherapy practice had become more uniform and had moved closer to practice in North America and Europe.16 For radical radiotherapy, 54% of prescriptions were for a fraction size of 1.8–2.0 Gy but the distribution was bi-modal and 20% of patients were prescribed fraction sizes of 2.7–3.0 Gy. Evidence-based practice now supports hypofractionated palliative treatment favouring single fractions for bone metastases and one or two fractions for many patients with advanced lung cancer. Two fractions are advised for some patients with brain metastases. If these guidelines had been applied uniformly the number of treatments prescribed for palliation would have been reduced by 36%. This would have represented a 6% reduction in the overall radiotherapy workload. Not all patients are suitable for such hypo-fractionated treatments, but this is an area in which resource use can be improved. In the postoperative management of breast cancer, a change in practice to use 15 fractions uniformly would reduce overall radiotherapy workload by 4%. By contrast, a change to 25 fractions would increase overall workload by 7%. WAITING TIMES
The results of a national audit of waiting times for radiotherapy were published by the Board of Faculty of Clinical Oncology of The Royal College of Radiologists in 1998.17 In 1993 the Joint Council for Clinical Oncology (JCCO) had defined targets for waiting times for radiotherapy. Four categories of patient were specified: urgent cases, those requiring radical (curative) radiotherapy, those requiring palliative treatment, and women requiring adjuvant radiotherapy after surgery for breast cancer. Overall, 28% of patients were waiting longer than the maximum times advised by the JCCO. The report documented considerable regional variation in waiting times and there was evidence to suggest that inequalities in the provision of resources were a major factor. Junor et al audited the travel and waiting times for outpatients receiving radiotherapy at a single centre in
1372 Medical audit
Glasgow.18 They found unacceptable treatment waiting times for many patients and concluded that the service might be improved by the provision of hostel or hotel accommodation and a computerized appointment system which would take into account the treatment machine, mode of transport and geographical area in which the patient lived. Jones and Dudgeon investigated the time between a person presenting to a general practitioner (GP) with a symptom of cancer and that person starting treatment.19 They analyzed retrospectively the records of 1465 cancer patients registered with 245 GPs, assuring confidentiality by using a coding system. They discovered considerable variation in processing, particularly in the time taken in different (unnamed) hospitals from hospital appointment to treatment. The authors believed that their study highlighted the need for GPs to review their diagnostic procedures on a regular basis and for hospital staff to review their own work.
Process and outcome CARCINOMA OF THE CERVIX
A retrospective analysis of 140 patients with stage Ib cervical carcinoma treated by either surgery or radiotherapy demonstrated the two modalities to have equivalent efficacy, but there was increased late morbidity in irradiated patients.20 The audit resulted in increased multidisciplinary collaboration, agreed protocols for staging, treatment and follow-up and prospective recording of the results of investigations, full treatment details and assessment at follow-up, including standardized documentation of bowel, bladder and sexual function. The authors considered that the improved documentation would help to ensure that future audits would be more reliable. An audit evaluating late toxicities in patients with carcinoma of the cervix treated with radical radiotherapy in India reported on 871 women with stage I-IVA disease treated with a combination of external-beam radiotherapy (EBRT) and intra-cavitary radiotherapy (ICRT) and 198 treated with EBRT alone.21 The median follow-up was 34 months. The median dose to point A was 81 Gy. The 5-year actuarial incidence of overall (all grades) and severe (grade 3/4) late toxicities in the rectum, bladder, small intestine and subcutaneous tissue were 12.3% and 1.1%, 11.2% and 1.2%, 9.2% and 0.2%, and 23.1% and 1.2%, respectively. Vaginal adhesions were seen in 29.6% of patients and stenosis in 33.9%. On multivariate analysis, factors adversely affecting the overall incidence of proctitis were an anterior-posterior (AP) separation more than 18 cm and the presence of co-morbidities. The latter was the only factor affecting the incidence of severe proctitis. An AP separation of more than 18 cm was also associated with an increased incidence of cystitis. Late small bowel toxicity was increased in some groups, for example in women
younger than 50 years and in women with co-morbidities, but no factor emerged that was significantly related to the incidence of severe toxicity. Subcutaneous fibrosis was significantly higher in patients with an AP separation over 18 cm, those treated by cobalt machines and those who received EBRT only. Severe subcutaneous fibrosis was associated with the use of EBRT alone. The overall incidence of vaginal toxicity was higher in women whose overall treatment time was shorter and in women who received ICRT. Vaginal stenosis was higher in elderly women and in women who received ICRT using a low dose rate. The authors concluded that even with telecobalt machines, impressive results with acceptable late toxicity can be achieved in the treatment of cancer of the cervix using an ideal combination of EBRT with ICRT. TESTICULAR CANCER
The care of 429 patients with malignant teratoma treated in five different units in the west of Scotland was audited by Harding et al.22 The proportion of men receiving nationally agreed protocol treatment was higher in the authors’ unit than elsewhere. Their unit treated the largest number of patients. The survival of patients treated at this unit, adjusted for other important prognostic variables, was also better than in the other units and the benefit seemed to be additional to any advantage resulting from protocol treatment. These patients had been treated over a 14-year period beginning 18 years earlier. The findings suggested strongly that centralization of treatment for malignant teratoma improves outcome. The authors concluded that the cumulative expertise in pathology, radiology, biochemistry and surgery, in addition to that of the oncology staff, contributed to the improved survival. HEAD AND NECK CANCER
An audit of handicap after radical neck dissection in 46 patients showed that half had to give up work because of problems with the shoulder due to sacrifice of the accessory nerve.23 Appreciable pain was experienced by one third of patients. The authors believed that accessory nerve function should be preserved whenever possible. Hong et al audited the treatment and outcome of 545 patients with head and neck cancer during the 8-year period from 1980 to 1987.2 Their local control and survival rates compared favourably with results reported from other institutions. The authors drew attention to the resources necessary to perform such an audit. Medical staff, data managers, cancer registry staff and statisticians together spent almost 1 hour for each patient. The increasing prospective computerization of patient data, treatment details and outcome would make similar audits much less labour intensive in future. Robertson et al audited the effect of the length and position of unplanned gaps in radiotherapy treatment schedules on 5-year local control of laryngeal cancer and the disease-free
Examples of published audits 1373
period.24 They reported that unplanned gaps in treatment were associated with poorer local control rates and an increased hazard of a local recurrence through extending the treatment time. A gap of 1 day was potentially damaging but the greatest effect was with treatment extensions of 3 or more days. The treatment extension as a result of the gap was more important than the position of the gap in the schedule. The authors concluded that gaps in the treatment schedule have a detrimental effect on the disease-free period and that any gap in the treatment is potentially damaging. BREAST CANCER
An audit of the incidence of arm lymphoedema after prophylactic level I/II axillary dissection without division of the pectoralis minor muscle demonstrated that lymphoedema is a common complication. Whole limb volume is often considered the main outcome measure for detecting lymphoedema and determining success of treatment, yet swelling may be restricted to regions of the limb and site-specific circumference measurements were therefore preferred. The authors recommended that pre- and postoperative circumference measurements are likely to be the most sensitive way of determining the presence of lymphoedema following surgery for breast cancer.25 The impetus for optimizing outpatient provision of breast-care services has come both from the patient and from management, with the aim of reducing anxiety and making the best and most timely use of scarce resources. The one-stop diagnostic clinic for the investigation of symptomatic breast lesions is a relatively recent concept with well-known service benefits. Berry et al audited patient acceptance of one-stop diagnosis for symptomatic breast disease and concluded that there was a high level of patient satisfaction.26 SOFT-TISSUE SARCOMA
Clasby et al audited the management of soft-tissue sarcoma by reviewing the records of 377 patients with primary softtissue sarcoma treated in the South-East Thames region between 1986 and 1992.27 53.6% were treated by general surgeons, irrespective of tumour location. Overall only 21.3% were investigated optimally, with wide variation among specialities. Only 60% were treated adequately (wide excision or surgery with radiotherapy). The uptake of adjunctive therapy and follow-up were variable and the authors concluded that investigation and management of many patients with soft-tissue sarcoma was both variable and suboptimal, having implications for patient outcome, resource usage and costs. They argued that patients with sarcoma are best managed in specialist centres. PANCREATIC CANCER
Pancreatic cancer carries a very poor prognosis, but in selected patients resection can improve survival. Several
recent reports have demonstrated that centralization of treatment can result in improved outcomes. The benefits of specialization in the management of pancreatic cancer in Scotland were reported with regard to both 30-day mortality and longer term survival. The 30-day mortality rate following resection was 8%, and was not influenced by hospital or consultant volume. The 30-day mortality rate following palliative surgical operations was 20%, and consultants with higher case loads or with a specialist pancreatic practice had significantly fewer postoperative deaths. For patients undergoing potentially curative or palliative surgery, the adjusted hazard of death was higher in patients with advanced years, increased co-morbidity and metastatic disease, and was lower for those managed by a specialist or by a clinician with an interest in pancreatic disease. The risk of death 3 years after diagnosis of pancreatic cancer was higher among patients undergoing surgery by nonspecialists.28
PALLIATIVE CARE
A clinical audit of the treatment of cancer-related pain, ordered by Stockholm County Council and the Karolinska Institute, was performed at two Stockholm hospitals.29 Of 153 consecutive cancer patients, 93 (61%) reported pain, varying in intensity from 2.4 to 6.6 on a 10-point visual analogue scale. The pain was cancer-related in 20 patients, treatment-related in 28 patients, and caused by nonmalignant disease in 40 patients (e.g. post-herpetic neuralgia, urethritis, decubital ulcer or constipation). Nine patients had pain with a previously undetected neuropathic component, and 18 patients reported both significant pain intensity and dissatisfaction with their treatment. The auditors found these patients to have persistent pain problems despite the availability of both time and opportunity to resolve them. The care offered in both hospitals was found to be characterized by the same deficiencies: lack of pain analysis or diagnosis, failure to detect neuropathic pain components, and underdosing of opioid analgesics irrespective of pain intensity. The auditors concluded that there was a need for pain education, particularly for doctors, as fewer doctors than nurses had attended pain courses. Malignant spinal cord compression (MSCC) is a particularly challenging area of cancer care where early diagnosis and expert multi-professional care and rehabilitation are paramount in optimizing quality of life. An audit of 174 patients with MSCC referred to the West of Scotland Cancer Centre examined current practice for symptom assessment, multi-professional care and rehabilitation of patients with MSCC admitted to the centre. Areas of concern identified by the authors included poor assessment of pain, the poor ambulatory status of patients on admission and the lack of clear plans for mobilization and rehabilitation for the majority of patients. Recommendations from the audit included the development of regional guidelines for referral, treatment and rehabilitation, the development
1374 Medical audit
of a pathway of care for use in all care settings across the region, and improvements in patient information, staff education, audit and research. These are now being taken forward through the West of Scotland Cancer Network with dedicated funding from Macmillan Cancer Relief.30
COMMUNICATION
An audit of 394 cancer patients to evaluate multidisciplinary team communication, information needs, decisionmaking preferences and information experiences was undertaken by the Cancer Research UK Psychosocial Oncology Group.31 A total of 87% of patients wanted all possible information, both good and bad news. Assuming that all clinicians had equal skill, 89% of patients expressed no preference for the sex of their doctor; 39% wanted to share responsibility for decision making. Preference was significantly influenced by age, with older patients more likely to prefer their doctor to make decisions on their behalf. While most or all patients reported receiving information regarding their initial tests (100%), diagnosis (100%), surgery (100%) and prognosis (87%), fewer recalled discussions concerning clinical trials (43%), family history (28%) or psychosocial issues, notably sexual well-being (37%). The authors also recommended that multidisciplinary cancer teams should ensure that whenever appropriate patients are provided with information about clinical trials, familial risk and psychosocial issues.
EFFECTIVENESS OF FOLLOW-UP
Rathmell et al examined the effectiveness of follow-up examinations and investigations for 29 patients who had relapsed following treatment for metastatic germ-cell tumours of the testis.32 The analysis showed that routine estimation of serum tumour markers was the single most important follow-up procedure, including patients who had been previously marker negative. It was the first indicator of relapse in 55 per cent of patients. Regular clinical examinations and chest radiography in asymptomatic patients were of little value. Chest radiography gave the first evidence of relapse in only two patients, but computed tomographic (CT) scanning of the chest and abdomen was the first abnormal investigation in seven patients. The authors also undertook a cost analysis. The costs per relapse detected by tumour marker assay and by CT scanning were £1800 and £12 880, respectively. With such a large discrepancy in cost effectiveness it was concluded that the CT scan interval could be extended with a large cost saving, and to the potential detriment of only the few patients with marker-negative relapse. An audit of 38 patients with breast cancer recurrence after treatment for primary breast cancer was conducted at the University of Sydney. 31 recurrences were symptomatic, and 32 presented at unscheduled appointments. 17 of these appointments were arranged following self-referral
and ten were requested by GPs. Of the seven patients with asymptomatic recurrences, two were discovered on specialist examination and five on routine mammography. The authors concluded that long-term routine clinical follow-up after treatment for breast cancer is inefficient in detecting recurrence.33 GEOGRAPHICAL AND SOCIAL INFLUENCES
Marked geographical variations in disease-specific mortality rates were observed among the 98 area health authorities in Wales and England.34 Variations persisted after standardization for social factors. There was a statistically significant fourfold difference in mortality from carcinoma of the cervix. Large socio-demographic differences in cancer survival have also been discovered.35 Statistically significant regional differences in survival were found for bladder, breast and colon cancer, survival being better in the south and east of England and worse in the north of England and in Wales. For cancers of poor prognosis, survival differences between socio-economic groups were small, but in some good-prognosis cancers the differences were wide, e.g. more than 2 years between owner-occupiers and council tenants for cancer of the corpus uteri. For the majority of cancers, people in better socio-economic circumstances had a greater probability of surviving their cancer than those in poorer circumstances, when account was taken of sex and age. Marked international variations in cancer survival in Europe have been demonstrated in an analysis based on data from 33 cancer registries in 17 countries, covering the period 1978–89.36 For all the common cancers – lung, breast, colorectal and prostate – the British survival figures are well below the European average. It seems likely that this reflects an inferior quality of care and the need for increased resources.37
KEY LEARNING POINTS ●
●
● ●
●
●
●
Good audit addresses important clinical activities (e.g. those that are common, high risk or high cost). It has a clearly stated purpose, is simple in design and uses resources efficiently. It assesses quality of care using objective criteria. It aims to bring about an improvement in outcome through a change in structure and/or process. It should be free of significant bias and be statistically sound. It requires regular and enthusiastic participation by all those involved. It stimulates the development and review of protocols for clinical management.
References 1375
REFERENCES ●1
2
●3
●4
5
6
●7
8
9
10
11
12 13
14
15
●16
17
Department of Health. The New NHS: Modern, Dependable. London: Department of Health, 1998. Hong A, Saunders MI, Dische S, et al. An audit of head and neck cancer treatment in a regional centre for radiotherapy and oncology. Clin Oncol 1990; 2:130–7. Benghiat A, Saunders V, Steele WV. Computerizing the cancer centre. Clin Oncol 1999; 11:33–9. Rosalki JR, Karp SJ. The COIN workstation: state of the future art. Clin Oncol 1999; 11:15–27. Kunkler IH. Variations in the management of cancer in the NHS: a legitimate cause for concern? J Eval Clin Pract 1997; 3:173–7. Royal College of Pathologists, the Royal College of Physicians of London and the Royal College of Surgeons of England. The Autopsy and Audit. The Report of the Joint Working Party of the Royal College of Pathologists, the Royal College of Physicians of London and the Royal College of Surgeons of England. London: The Royal College of Pathologists, 1991. Board of the Faculty of Clinical Oncology. Medical manpower and workload in clinical oncology in the United Kingdom. London: The Royal College of Radiologists, 1991. Board of Faculty of Clinical Oncology. Equipment, Workload and Staffing for Radiotherapy in the UK 1992–1997. London: The Royal College of Radiologists, 1998. Bradley PJ, Zutshi B, Nutting CM. An audit of clinical resources available for the care of head and neck cancer patients in England. J Laryngol Otol 2005; 119:620–6. Rees G. Lay influence on service provision: impact of a Royal College Patient Liaison Group on radiotherapy departments. J Radiother Pract 2003; 3:5–90. Schmidt F, Hartwagner KA, Spork EB, Groell R. Medical Audit after 26 711 breast imaging studies: improved rate of detection of small breast carcinomas (classified as Tis or T1a,b). Cancer 1998; 83:2516–20. Godsell GA. The development of the nurse biopsy role. Br J Nurs 2005; 14:690–2. Dey P, Woodman CB, Gibbs A, Coyne J. Completeness of reporting on prognostic factors for breast cancer: a regional survey. J Clin Pathol 1997; 50:829–31. Keating J, Lolohea S, Kenwright D. Pathology reporting of rectal cancer: a national audit. NZ Med J 2003; 116:514. Priestman TJ, Bullimore JA, Godden TP, Deutsch GP. The Royal College of Radiologists’ fractionation survey. Clin Oncol 1989; 1:39–46. Williams MV, James ND, Summers ET, Barrett A, Ash DV. National survey of radiotherapy fractionation practice in 2003. Clin Oncol 2006; 18:3–14. Board of Faculty of Clinical Oncology. A National Audit of Waiting Times for Radiotherapy. London: The Royal College of Radiologists, 1998.
18 Junor EJ, Macbeth FR, Barrett A. An audit of travel and waiting times for outpatient radiotherapy. Clin Oncol 1992; 4:174–6. 19 Jones RV, Dudgeon TA. Time between presentation and treatment of six common cancers: a study in Devon. Br J Gen Pract 1992; 42:419–22. 20 Gaze MN, Kelly CG, Dunlop PR, et al. Stage IB cervical carcinoma: a clinical audit. Br J Radiol 1992; 65: 1018–24. 21 Saibishkumar EP, Patel FD, Sharma SC. Evaluation of late toxicity of patients with carcinoma of the cervix treated with radical radiotherapy: an audit from India. Clin Oncol 2006; 18:30–7. 22 Harding MJ, Paul J, Gillis CR, Kaye SB. Management of malignant teratoma: does referral to a specialist unit matter? Lancet 1993; 341:999–1002. 23 Shone GR, Yardley MP. An audit into the incidence of handicap after unilateral radical neck dissection. J Laryngol Otol 1991; 105:760–2. 24 Robertson AG, Robertson C, Perone C, et al. Effect of gap length and position on results of treatment of cancer of the larynx in Scotland by radiotherapy: a linear quadratic analysis. Radiother Oncol 1998; 48:65–73. 25 Querci della Rovere G, Ahmad I, Singh P, Ashley S, Daniels IR, Mortimer P. An audit of the incidence of arm lymphoedema after prophylactic level I/II axillary dissection without division of the pectoralis minor muscle. Ann R Coll Surg Engl 2003; 85:158–61. 26 Berry MG, Chan SY, Engledow A, et al. An audit of patient acceptance of one-stop diagnosis for symptomatic breast disease. Eur J Surg Oncol 1998; 24:492–5. 27 Clasby R, Tilling K, Smith MA, Fletcher CD. Variable management of soft tissue sarcoma: regional audit with implications for specialist care. Br J Surg 1997; 84:1692–6. 28 Parks RW, Bettschart V, Frame S, Stockton DL, Brewster DH, Garden OJ. Benefits of specialisation in the management of pancreatic cancer: results of a Scottish population-based study. Br J Cancer 2004; 91:459–65. 29 Arner S, Killander E, Westerberg H. Poor leadership behind poor pain relief. Medical audit of cancer-related pain treatment. Lakartidningen 1999; 96:33–6. 30 McLinton A, Hutchinson C. Malignant spinal cord compression: a retrospective audit of clinical practice at a UK regional cancer centre. Br J Cancer 2006; 94:486–91. 31 Cox A, Jenkins V, Catt S, Langridge C, Fallowfield L. Information needs and experiences: An audit of UK cancer patients. Eur J Oncol Nurs 2006; 10:263–72. 32 Rathmell AJ, Brand IR, Carey BM, Jones WG. Early detection of relapse after treatment for metastatic germ cell tumour of the testis: an exercise in medical audit. Clin Oncol 1993; 5:34–8. 33 Hiramanek N. Breast cancer recurrence: follow up after treatment for primary breast cancer. Postgrad Med J 2004; 80:172–6.
1376 Medical audit
34 Charlton JR, Hartley RM, Silver R, Holland WW. Geographical variation in mortality from conditions amenable to medical intervention in England and Wales. Lancet 1983; 26:691–6. 35 Kogevinas M. Longitudinal Study: Sociodemographic Differences in Cancer Survival. Office of Population Censuses and Surveys. London: HMSO, 1990.
36 Coebergh J, Sant M, Berrino F, Verdecchia A. Survival of adult cancer patients in Europe diagnosed from 1978–1989: the Eurocare II study. Eur J Cancer 1998; 34:2137–278. 37 Sikora, K. Cancer survival in Britain is poorer than that of her comparable European neighbours. BMJ 1999; 319:461–2.
59 The organisation of cancer services: a UK perspective IAN KUNKLER
Introduction Historical perspective The Calman–Hine report National cancer plans
1377 1377 1379 1380
INTRODUCTION UK cancer services have undergone unprecedented structural reform over the last decade with marked improvements in the quality of care provided to patients. In this chapter the historical background to these changes and the principal government reforms of cancer services in the UK are described. An assessment is made of progress on implementation, further challenges and of developments in cancer care that may shape the future organization of cancer services. While the importance of cancer prevention is acknowledged, the focus of this chapter is on services to patients with an established diagnosis of cancer.
Why is organization important? The management of patients with cancer involves multiple steps in diagnosis and treatment. This ‘cancer journey’ (Fig. 59.1) stretches all the way from primary care through district general hospitals to cancer centres and palliative care units. The pathway is a complex one. Up to thirty different people may be involved in delivering the patient’s care.1 For patients, this complex pathway must seem confusing and difficult to navigate. As the Commission for Health Improvement/National Audit Office (CHI/NAO) report points out: ‘patient centred care is about more than simply the patient’s experience of an individual consultation. It is
The effectiveness of reorganization of UK cancer services Future directions in organization of cancer services References
1385 1403 1404
about the gearing of the whole cancer service to the needs of the patient with all the professionals working together to that end.’
Patterns of cancer services in the UK The pattern of organization of cancer services across the UK remains generally broadly similar. However, Scotland and Wales have devolved powers over health care and some differences in approach to delivery of cancer services have emerged. In England there is a stronger commitment to developing a plurality of providers of NHS care from both state and independent sectors. In Scotland and Wales the emphasis has been on integrated state-funded cancer care. In addition there are different national administrative frameworks for the delivery of cancer services. Initially the ‘hub and spoke’ model of cancer centres and cancer units was adopted in England2 and managed cancer clinical networks in Scotland.3 The ‘hub and spoke’ model in England has been replaced by cancer networks, a concept ranking equally the contribution of all cancer professionals within a geographical area.
HISTORICAL PERSPECTIVE There were a number of reports produced by the Royal Colleges4,5 that provided guidance on the provision of cancer
1378 The organization of UK cancer serivces: a UK perspective
Screening Accident & emergency
Consultant referral
Patient with symptoms
GP referral
Subsequent treatments
1st hospital assessment
Diagnosis
Decision to treat
services which preceded the landmark Calman– Hine recommendations2 for reform of cancer services in 1995. However, they had little discernible impact on government policy. At that time the focus was on specific services such as radiotherapy and breast screening.6
An emerging crisis Evidence began to accumulate in the 1990s of a growing crisis in the provision of cancer services. There were a number of contributory factors: evidence of deficiencies in services, professional concern about standards, a media focus on the treatment of cancer in the UK being akin to a lottery, and unfavourable comparisons of UK cancer survival with continental European countries.
The cancer lottery Breast cancer was a cameo sketch of shortcomings generic to the organization of cancer services for a wide range of malignancies. The high quality of organization of breast screening contrasted with the poorly organized services for symptomatic women.6 Intra- and inter-regional differences in outcomes for breast cancer became apparent. A study from Yorkshire by Sainsbury et al.7 showed considerable differences in survival depending on the individual surgeon. Approximately 26 per cent of the difference was explained by the use of chemotherapy and endocrine therapy. Poorer outcomes were associated with surgeons treating fewer than 30 patients per year. In addition, differences in management of breast cancer were identified between teaching and non-teaching hospitals.8 A retrospective study by Gillis and Hole in 19969 of 3786 breast cancer patients under the age of 75 from the West of Scotland cancer registry who were treated between 1980 and 1988 compared the survival of patients treated by specialist and non-specialist breast surgeons. The 5- and 10-year survival rates of patients treated by specialist breast surgeons were, respectively, 9 per cent and 8 per cent higher.
1st treatment
Figure 59.1 The cancer journey. (Reproduced with permission from the Department of Health.)30
A number of publications showed better outcomes in patients managed in a specialist centre.10–12 While most of the published literature endorsed the notion that specialist cancer care was associated with better clinical outcomes, for example for breast,9 testicular13 and ovarian cancer,14 this was not true for all cancer sites. Surgeons in district general hospitals with a specialist interest in coloproctology could achieve similar outcomes to comparable colleagues in teaching hospitals.15 A study from Finland showed that some of the geographical differences in survival were attributable to the type of central hospital facilities, especially the availability of radiotherapy and the presence of a medical school.16 The probability of survival was 10 to 15 per cent higher in specialist teaching hospitals. In Northern Ireland, it was concluded that there was no evidence that management by cancer specialists was associated with poorer outcomes. On this basis the Northern Ireland Cancer Working Group17 favoured specialized cancer care for both common and rarer cancers. In addition there was growing professional unease with the state of cancer services.18,19 A study by the Association of Cancer Physicians showed that 40 per cent of UK cancer patients never saw an oncologist.18 There was pressure from patient groups for change and evidence of poorer survival for common malignancies in the UK compared with continental Western European countries.20
Cancer specialization, multidisciplinary teams and delivery of care Up to the early 1990s, cancer surgery was largely undertaken by general surgeons. Some subspecialization did exist in urology, gynaecology and thoracic surgery21 and within clinical oncology, medical oncology and haematology. There are many elements to specialization. These include the composition of a multidisciplinary team, training and skills, workload by unit or individual and the provision of care within a teaching hospital or a district general hospital. 21 A multidisciplinary team has been defined as: ‘Primarily it is that group of professionals which meets together at a given time and place and together makes
The Calman–Hine report 1379
decisions regarding treatment of individual patients, which can therefore be multidisciplinary.’22 The need for multidisciplinary care was highlighted by the British Breast Group23 as a key element in ensuring that each patient had the benefit of input to decisions on management from breast surgeons, oncologists, radiologists, pathologists and breast care nurses. At that time there was an extensive literature on the effectiveness of cancer services. However, much of it was inconclusive.2,21 Concerns were also raised about the unevenness of dissemination of research findings.24
Inadequate radiotherapy capacity The legacy of inadequate capacity planning for radiotherapy has been long waiting times for radiotherapy in the UK. An audit of waiting times25 showed that 28 per cent (556/1988) of patients were waiting for more than the maximum waiting times for radical and palliative radiotherapy (28 and 14 days, respectively) advised by the Joint Council for Clinical Oncology. Only 26 per cent of patients met the criteria for good practice for radical radiotherapy and 20 per cent for palliative radiotherapy. Considerable regional variation in waiting times was identified.
Comparisons with European cancer survival European epidemiologists are due substantial credit as catalysts for reform of UK cancer services. The relatively poor survival figures for common malignancies such as breast, bowel and lung cancer were shown by the first EUROCARE study.26 These were based on 30 cancer registries reporting on 800 000 patients diagnosed between 1978 and 1985. For 18 of the 25 cancer sites, survival of UK patients was poorer than the majority of other European countries. While these studies did not take account of casemix, they implied that the quality of cancer care in the UK was inferior to that of many continental European countries.
THE CALMAN–HINE REPORT In response to these various pressures for reform outlined above, an Expert Advisory Group (EAG) was established in 1994 to provide the direction of cancer services for England and Wales. Its recommendations were given in the Calman– Hine Report,2 which was a landmark in the organization of cancer services since it set out a template for major reform. The main recommendations are summarized in Box 59.1. The report highlighted the published benefits of specialist treatment for uncommon cancers and, although to a lesser degree, for common cancers. It also recognized the importance of ensuring that all patients had access to the
Box 59.1 Calman–Hine recommendations on UK cancer services General ● Network of expertise between primary care, cancer units in district hospitals and cancer centres ● Three levels of care: 1. Primary care – with referral guidelines and follow up to be agreed with cancer units and centres 2. Designated cancer units in many district general hospitals with sufficient clinical teams and expertise to manage common cancers 3. Designated cancer centres with expertise in the management of all cancers including common and uncommon malignancies. These centres will have specialized diagnostic and therapeutic facilities including radiotherapy ● Cancer site specific delivery of surgical services within cancer unit ● Appointment of lead clinician for cancer services in each cancer unit ● Multidisciplinary management with input from medical oncologist or clinical oncologist who also holds an appointment in a cancer centre ● A minimum of five non-surgical oncology sessions per week ● Establishment of Joint Council for Clinical Oncology between the Royal College of Radiologists and the Royal College of Physicians ● Establishment of cancer site specific nursing service ● Delivery of chemotherapy and biological therapy where appropriate facilities and sufficiently experienced multidisciplinary teams are present ● Sharing of common treatment protocols between cancer centres and cancer units. Cancer centres Cancer site specialization in cancer centres ● Minimum population base of 2/3 of a million for cancer centres ● Review of hospitals providing radiotherapy services but not conforming to the definition of a cancer centre ● Specialist surgical services including reconstructive surgery ● Intensive chemotherapy including bone marrow transplantation ● Important role in training, continuous professional development, clinical audit and entry of patients into clinical trials. ●
Children and adolescents with cancer Maintenance of network of centres providing paediatric oncology.
●
Palliative care Seamless integration with all cancer treatment service ● Establishment of community and hospital-based palliative care services. ●
1380 The organization of UK cancer serivces: a UK perspective
best standard of care. While it was acknowledged that the timescale for implementation of change would be several years, more immediate benefits could accrue from better organization and communications between patients, purchasers, providers and professionals. The EAG recommended a network of specialized expertise extending from primary care to district general hospitals and cancer centres. Particular emphasis was placed on the role of primary care and its closer interface with cancer centres and cancer units. The Calman–Hine Report recommended a ‘hub and spoke’ arrangement between cancer centre and cancer units. Cancer units were to be based in district general hospitals. Each cancer unit would appoint a lead clinician for cancer services. He/she would be responsible for coordinating all cancer services and the implementation of cancer site specific protocols. It was envisaged that cancer units would provide basic local services. Cancer centres would provide both basic services and more specialist services including radiotherapy, complex surgery, specialized imaging and complex chemotherapy. The Calman–Hine report made a number of recommendations on professional practice. It required more surgical subspecialization, and provision of specialist nurses. The report espoused a set of principles for cancer services ensuring that they were of good quality and accessible irrespective of distance. Systems needed to be in place for the early recognition of cancer. Patients required good information about treatment options and support during the process of care.
Response to the Calman–Hine Report In response to the Calman–Hine recommendations, reports from Scotland,27 Wales28 and Northern Ireland17 were published, endorsing its recommendations and providing guidance on its application. SCOTLAND
The Scottish report endorsed the general principles of Calman–Hine and the need to formalize the relationships between cancer centres and feeder district general hospitals, whether or not the latter were designated as cancer units.27 Scotland subsequently elected not to adopt the ‘hub and spoke’ model of cancer centres and cancer units but adopted managed clinical networks as the administrative framework of cancer services in cancer centres and district general hospitals. The report also acknowledged the need for investment, particularly in specialized nursing staff and non-surgical oncologists, and for a comprehensive monitoring and quality assurance system. WALES
The Cameron report28 on cancer services in Wales identified that from a patient’s perspective communications
between patients and cancer professionals were poor. Patients and general practitioners often lacked information about where specialist services were available and who provided the service. NORTHERN IRELAND
The Campbell Report17 on cancer services in Northern Ireland recommended that there should be three levels of care for patients with cancer: ● ●
●
Primary care, which was seen as the focus of care Cancer Units, where patients with the more common cancers would receive their care A Cancer Centre, which would provide a regional service for patients with less common cancers or those who needed more complex treatment and care.
NATIONAL CANCER PLANS It was the publication of the EUROCARE 2 study in 199929 that persuaded the UK government that a national cancer plan was needed to address the shortcomings of UK cancer services. The same pattern emerged of poorer 5-year survival for common cancers compared with many continental countries, similar to that in the first EUROCARE study.20 Following a cancer summit convened by the Prime Minister in May 1999, a national cancer director was appointed for England. A single target was set of reducing the mortality from cancer in individuals under the age of 75 by 20 per cent by 2010 from a baseline in 1997. A 10-year national cancer plan was established in England 30 and equivalent plans in Scotland,31 Wales39 and Northern Ireland.17 The National Cancer Plan in England was intended to provide a comprehensive strategy to improve prevention, screening, diagnosis, treatment and care of patients with cancer. It was accompanied by a plan of investment for staffing, equipment, drugs, other treatments and information systems. The national cancer plan in Scotland31 covered similar areas to the English national plan but, in general, was less specific about the timescales for changes to be achieved. The aims of the national cancer plan in England are to: ● ●
●
●
Save more lives Ensure people with cancer receive the right professional support and care as well as the best treatment Tackle inequalities in health that result in unskilled workers being twice as likely to die from cancer as professionals Build for the future by investing in the workforce and in research and development to ensure that the NHS does not fall behind in cancer care again.
In conjunction with national cancer plans, the Royal Colleges have provided guidance on issues such as the appropriate provision and replacement of radiotherapy,32
National cancer plans
skills mix33 and dose-fractionation regimes for different cancers.34
Reorganization of cancer services in England The organizational tree for cancer services is summarized in Fig. 59.2. The National Cancer Director reports to ministers and the Permanent Secretary. A Cancer Taskforce gives advice to the National Cancer Director, identifies needs for developments in policy and monitors progress of the implementation of the national cancer plan. Below the National Director is a triad of Cancer Policy Team, Cancer Action Team and the Cancer Collaborative Improvement Partnership, a branch of the Modernisation Agency for the NHS. The 34 cancer networks involving NHS trusts, primary care trusts (PCTs), the voluntary sector and cancer site specific clinical groups report to the Cancer Action Team. The cancer networks are responsible for implementing the NHS Cancer Plan. Cancer networks were established to integrate patient care, improve clinical outcomes, provide cost effective equitable service and improve the patient’s experience of cancer care. These are matrix organizations responsible for developing and planning all aspects of cancer services (Fig. 59.3). They have inputs from tumour-specific site specialists, acute and primary care trusts, patients and user groups and the voluntary sector. Each network has a management team led by a network board. The main responsibilities of the network boards include: ● ● ●
Strategic planning Clinical governance development Support strategies for workforce planning and planning of facilities.
Managed clinical networks in Scotland In Scotland managed clinical networks (MCNs) are the administrative framework for cancer services.3,35 Their development is in line with the priority given in Scotland to integration of health and related services.36 Partnership in the NHS was also emphasized in the White Paper Partnership for Care.37 Managed clinical networks were defined as: ‘Linked groups of health professionals and organisations from primary, secondary and tertiary care working in a coordinated manner unconstrained by existing health board boundaries to ensure the equitable provision of high quality clinically effective services throughout Scotland‘38 The concept of MCNs is based on the sharing of patients, expertise and resources rather than unidirectional centripetal flow. The design was intended to underpin subspecialty medical cover where the population was too small to sustain this in isolation. The Acute Service Review in Scotland recommended: ‘connection and partnership rather than isolation and self-sufficiency, on distribution of resources rather than centralisation, maximising benefits for large numbers of patients.’3 Managed clinical networks have been developed in the three geographical regions of Scotland. Each network is composed of cancer site specific professionals. The inclusive approach to membership is intended to foster collaboration across traditional geographic boundaries. The roll out of these networks started in 1999. They provide a
Ministers & Permanent Secretary
Cancer Policy Team (Department of Health) Develops/monitors/reviews policy; advises ministers etc
National Cancer Director (NCD)
Cancer Taskforce Advises NCD & ministers; monitors progress; identifies policy development needs
Cancer Action Team (NHS) Supports implementation of Plan and development of networks; leads on quality assurance of cancer services
Cancer Services Collaborative Improvement Partnership (Modernisation Agency) National programme of service improvement
34 Cancer Networks (NHS trusts, primary care trusts, voluntary sector, clinical groups etc)
Figure 59.2 Organizational tree of National Cancer Plan in England. (Reproduced with permission from the Department of Health.)67
1382 The organization of UK cancer serivces: a UK perspective
Cancer Network Functions Prevention
Screening
Community Services
Waiting Times
Treatment & Care
Research
Inputs from SHAs Acute Trusts PCTs Tumour Groups Voluntary Sector Patients & Carer Groups Local Authorities
Figure 59.3 Cancer networks as matrix organizations. (Reproduced with permission from the Department of Health.)67 PCTs, Primary Care Trusts; SHAs, Strategic Health Authorities
forum for regional service planning, clinical audit and redesign of cancer services. The principles of managed clinical networks by which MCNs should be governed35 are shown in Box 59.2. Key mechanisms for improving outcomes include multidisciplinary working, the development and implementation of evidence-based protocols and redesigning patient pathways. The managed clinical networks aim to break down the barriers between primary and secondary care to create seamless delivery of care. A particular emphasis is put on
Box 59.2 Principles of managed clinical networks in Scotland35 ●
●
●
●
●
● ● ● ●
Clarity about management of the network and designation of person in overall charge Defined structure defining points at which services are delivered and connections between them Clear statement of specific clinical and service improvements anticipated Use of documented evidence base, e.g., Scottish Intercollegiate Guidelines Network (SIGN) with commitment to expand the evidence base through appropriate research and development Multi-professional membership including representation from patients’ organizations Education and training provision Audit of performance Staff rotation to promote development Consideration for providing better value for money.
the managed element of the network to ensure lines of responsibility and accountability are clearly defined. These networks are not intended to be static structures but dynamic and capable of responding and adapting to changes in health-care needs. They may create virtual communities of cancer professionals within a region with a shared commitment to developing common approaches to managing specific malignancies.38
Reorganization of cancer service in Wales Following the Calman–Hine report,2 the Cameron Report was published in 199628 reviewing cancer services in Wales. As its baseline, the report acknowledged a polarization between the need for specialization of skills and equipment on one hand and community-based generic services on the other. It recognized too the perception of patients and their families that only cancer specialists gave appropriate advice. There was a lack of computerized workforce data and little inter-trust audit of cancer outcomes. With few exceptions, hospitals could not provide comprehensive information on ‘who does what to whom and where?’ The existing contractual framework, which focussed on overall quantum of cost, precluded calculation of the real cost of patient-specific protocols. The existing contracting process in which different aspects of care were split into different contracts mitigated against seamless cancer care. While acknowledging that improvements to equipment were needed, it was felt that changes in methods of treatment and cancer site specific protocols could be implemented more quickly. It concluded that key targets would be the establishment of multidisciplinary team working and service
National cancer plans 1383
development based on implementation of standards for cancer services. Its main recommendations are summarized in Box 59.3.
Box 59.3 Recommendation of the Cameron Report, 1996 on cancer services in Wales28 ●
●
●
●
●
●
An all-Wales service for cancer patients to be integrated at Cancer Centre level with services provides at three locations, N Wales, SE Wales, SW Wales Common guidelines, protocols and standards to be adopted between the three Cancer Centres in Wales Specialist cancer care to be provided by multidisciplinary teams for a specific tumour site to maximize patient access to comprehensive specialist cancer care Tumour site specific professional groups to be retained at an all-Wales level who will have a key role, in collaboration with the Cancer Services Coordinating Group in monitoring outcome, updating guidance and providing impartial advice to Health Authorities, Trusts and Local Health Groups Information systems with the ability to collect common data to be used across all Trusts in Wales and feed the Welsh Cancer Surveillance and Intelligence Unit General practitioners and patients to be provided with information on where specialist cancer services are available.
The Cameron report did not attempt to set national standards or appropriate indicators of the quality of cancer care. This task has been undertaken by the Cancer Services Coordinating Group (CSCG). An NHS plan for Wales39 including cancer was published in 2001. A commentary on its aims by the National Assembly for Wales focuses on building on the work of the CSCG to achieve the following targets for cancer patients by December 2001:
Reorganization of cancer services in Northern Ireland The Campbell Report (1996) in response to the Calman– Hine report recommended major changes in how cancer services were organized and delivered in the province. A regional advisory committee on cancer was established in 1997 to take forward the recommendations of the Campbell Report. The Campbell Commissioning Project was set up to oversee the changes in partnership between Trusts, Boards and other stakeholders. The Northern Ireland Cancer Forum was set up in 1999 as a meeting point for all voluntary and statutory bodies dealing with cancer in Northern Ireland. Its remit was to: ● ● ● ●
Working groups were established for workforce planning, guidelines for chemotherapy, the development of a Clinical Oncology Information System, palliative care services and communications with the public on cancer related topics. In addition, a group of lead cancer clinicians were established to implement changes between the centre and cancer units, particularly in relation to quality assurance, clinical governance and clinical audit. An important and unique development has been the Memorandum of Understanding signed in October 1999 establishing the Cancer Consortium – a tripartite agreement between the National Cancer Institute of the National Institute of Health in the USA with the Departments of Health in the North and South of Ireland. The purpose of this collaboration is to create an All Ireland infrastructure for cancer research and clinical cancer investigations. Initial objectives included co-operation between the cancer registries, development of a clinical trials network between the three participants, high speed links via telemedicine and a scholar exchange programme. Significant improvements in the staffing and specialist multidisciplinary management of cancer and of palliative care have occurred. A Northern Ireland Cancer Network (NICaN), a managed clinical network, was established in 2004 to secure progressive improvements in the quality of care for patients. The guiding values of NICaN are: ●
●
●
●
●
Consultant appointment within 10 days of receipt by the hospital of an urgent referral by their general practitioner Receive a diagnosis and appointment for treatment as set out in published CSCG minimum standards for cancer care Treatment and care will be discussed by the multidisciplinary team specializing in the disease Increasing access to specialist nurses with extended skills enabling more of their care to be carried out at home.
Empower the participant organizations Facilitate a team approach to cancer care Enable resources to be shared Promote better communication between organizations.
●
● ● ● ● ●
Creating a person centred service Securing involvement of service users and those with an interest in cancer services Fostering cultural change and development Joining ‘everything’ up to provide seamless care Promoting population health/wellbeing Providing clinically effective treatment, care and support Developing clinical leadership and meaningful teamwork.
The early work NICaN was focussed on obtaining the support of key individuals across the region who had the power and influence to promote the development of the
1384 The organization of UK cancer serivces: a UK perspective
Network.40 Northern Ireland Cancer Network tumour and crosscutting theme groups are responsible for the development of regional protocols, audit, patient pathways, service redesign, quality assurance and identification of funding priorities. The NICaN board has commissioned an external evaluation of the processes and outcomes of the cancer network. A number of key questions (Box 59.4) raised by over 250 stakeholders and by the NICaN board to be considered in developing the cancer network plan for 2005 to 2008, are highly relevant to all UK cancer networks.
Clinical standards All the home countries have developed frameworks for defining and monitoring clinical standards. Evidence-based standards have been established in England for breast,41 colorectal,42 lung,43 gynaecological44 and upper gastrointestinal45 cancers. The Clinical Standards Board for Scotland (now part of NHS Quality Improvement Scotland) has established minimum standards for lung,46 ovarian,47 colorectal48 and breast49 cancers, breast screening50 and specialist palliative care.51 Monitoring of compliance with standards is undertaken through a regular cycle of site visits of cancer professionals and patient representatives. In Wales, the Cancer Services Coordinating Group was established in 1997 to coordinate the implementation of the Cameron Report ‘Cancer Services in Wales’. Minimum standards were developed including breast,52 colorectal,53 gastrooesophageal,54 gynaecological,55 haematological,56 head and neck, lung,57 skin58 and urological59 cancers and palliative care60. In England, the Manual of Cancer Services and Standards22 has been developed to provide a framework by which local cancer networks can assess the quality of cancer services they provide. The manual spans ten topics along the patient pathway. For each topic, there are specified objectives and standards defining the structures and/or processes to be implemented to ensure the objectives can be achieved. These standards are very comprehensive and should therefore minimize significant differences in implementation across England if correctly followed.
Clinical trials The National Cancer Plan in England30 recognized the importance of clinical research to improvement of outcomes for patients. It is generally true that patients often have better outcomes in clinical trials. In large part this is explained by the strict following of treatment protocols required in clinical trials. However, a critical review of this issue suggests that some of the comparisons of trial and nontrial patients have not taken account of potential confounding factors.61 A National Cancer Research Institute (NCRI) was established to coordinate clinical cancer trials. The NCRI groups review new trial proposals, which then are submitted for peer reviewed funding to national bodies such
Box 59.4 Key questions for stakeholders for the national development plan for cancer services 2005–8 (NiCAN report, 200440) 1. How will the network ensure an effective flow of communications between and within all organizations, sectors of care and disciplines? 2. How will the network lever the required resources to provide an equitable regional cancer service for all? 3. How will the network, as a complex arrangement, make a real difference to the quality of service, patient experience and patient outcomes? 4. How will truly integrated and effective multidisciplinary/multi-professional team-working be further developed across the network (despite sectors and geographical spread)? 5. How will the network facilitate evidence-based regional standards, guidelines and protocols, including their implementation and monitoring? 6. What mechanisms, arrangements and processes are required to ensure the delivery of the network development plan? 7. What is the impact of service user involvement on service development and professional practice? 8. How will cancer prevention/health promotion be integrated into all aspects of the network? 9. How will the network prioritize its developments? 10. How will the network ensure that the development of standards/protocols will lead to improvement in services? 11. How will the network attain relevant and appropriate data in line with regionally agreed guidelines and standards (quality indicators) and monitor quality of service provision? 12. How will the network ensure that the work is integrated/aligned with existing regional processes, including departmental thinking? 13. How can the network facilitate the full integration of primary and secondary care? 14. How can the network board ensure: ● That all those who contribute to the strategic thinking and further development of the network are identified and fully engaged? ● That the cancer network is recognized and aligned with regional processes and planning? 15. How can the network harness the potential of available media to ensure timely and effective communications? 16. How can the network enable meaningful service user involvement? 17. How will the network support the development of more effective MDT working?
The effectiveness of reorganization of UK cancer services 1385
as Cancer Research UK and the Medical Research Council. Each NCRI group has lay representation to ensure that clinical studies take account of patients’ interests, particularly in relation to patient information sheets and consent forms. The National Cancer Research Network (NCRN) was established to provide infrastructure for running cancer trials. Funding has been provided for research nurses to assist with the administrative support and data collection involved in clinical trials. In 2001, the NCRI and its partners agreed a target of increasing accrual into cancer trials from less than 4 per cent to at least 7.5 per cent of new patients with cancer each year. This target has been met. Each research network has to provide a balanced portfolio of clinical trials within the budget available to it. The NCRN badged studies have priority in access to local NCRN support.
THE EFFECTIVENESS OF REORGANIZATION OF UK CANCER SERVICES Evaluation of the effectiveness of the UK cancer service, whose organizational networks and services are developing at different rates, is a complex task. It is critically dependent on reliable, timely information on clinical activity, investigations, clinical outcomes and patient experience of care received. Despite the huge strides in improving cancer services in the UK, the acquisition of this information has been handicapped by the lack of national information systems. Progress is being made in organizing prospective audit and in providing a national information technology (IT) system for the NHS. However, the roll-out of the NHS national programme for IT in England – for example, providing electronic medical records to authorized users in primary, secondary, tertiary and community care – has been slower than planned.62 Equivalent clinically relevant data need to be available to cancer professionals and managers at local, regional and national levels on all aspects of cancer service delivery. In this way a much more accurate and contemporary picture of cancer care could be captured. Trends in lengthening waiting times for treatment could prompt earlier review of the efficiency of services or the need for additional resources. The integration of electronic feedback from patients linked to data on their passage along cancer site specific care pathways will be essential if the NHS commitment to providing patient focussed care is to be meaningfully assessed.
Cancer mortality A reduction in cancer mortality, a principal endpoint against which the reorganization of cancer services is to be judged, has been achieved in England with provisional data from 2001 to 2003 showing that the death rate in people under the age of 75 had fallen by 12.2 per cent from the 1995 to 1997 baseline.63 One of the major targets of reducing waiting times for all cancers from first urgent GP referral to treatment to two months by 2005 has not been met in England or Scotland. In England 78 per cent of all urgently referred patients were treated within
62 days and 73 per cent in Scotland. Additional service expansion will be needed to reach these targets. An equally important endpoint, to which cancer planners in the UK have appropriately devoted much effort, is improving the patient’s experience of contact with cancer services. The analysis of the patient’s cancer journey (see Fig. 59.1) from presentation to their general practitioner through diagnosis, treatment and supportive care to death or long-term remission has been the starting point for redesigning better cancer services. The more streamlined and seamless diagnostic and treatment services become, the better the patient’s experience is likely to be. The application of many of the key elements of the reorganization of UK cancer services (such as networks) in the reorganization and research frameworks of a number of non-malignant diseases is testament to their quality and impact. The NAO audit64 comments that the NHS Cancer Plan in England compares favourably with the cancer plans in other countries.65
Quantitative and qualitative aspects of organization of cancer services Cancer services embrace both quantitative and qualitative elements. The quantitative elements are straightforward (expansion of oncology posts, diagnostic or treatment equipment) and easy to measure against stated targets in cancer plans. The qualitative aspects, particularly the patient’s experience of contact with cancer services, are more subjective but perhaps the more important arbiter of the bene-fits of change. Patient surveys and focus groups have been used to capture the patients’ experiences of cancer care. There are qualitative aspects too of the functioning and dynamics of multidisciplinary teams and cancer networks. Despite the centrality of multidisciplinary teams to UK cancer policy, relatively little information is available about the clinical effectiveness,66 culture and team dynamics of multidisciplinary teams.
Progress in implementing national cancer plans An important common feature of UK cancer policy is a commitment to regularly review progress against targets or milestones of organizational reform including feedback from patients on whether they have received high-quality care. These reviews provide snapshots of the service over particular periods. The NAO in England for example has undertaken reviews64,67,68 of progress against the target of the National Cancer Plan including surveys of patients’ experience of the service.69 In addition, the CHI undertook a joint assessment of cancer services in England and Wales.1 In Scotland, a 6-monthly review of progress against goals of its own cancer plan is published, as are annual reports. 70–72 Considerable progress has been made in implementing national cancer plans in each of the home countries. Progress has been exemplified from England and Scotland (see Table 59.1).
1386 The organization of UK cancer serivces: a UK perspective
Table 59.1 Progress against targets of national cancer plans in England (2000) and Scotland (2001). ENGLAND TARGET
TARGET YEAR
TARGET MET
PROGRESS
OVERALL
20% reduction in cancer mortality in under 75s from 1997.
2010
NA
NA
IMPROVING PREVENTION (I) SMOKING
Comprehensive smoking cessation services in Primary Care Trusts.
2001
Smoking amongst manual groups reduced from 33% in 1998 to 31% in 2003. (NB – the method of calculation has changed for smoking prevalence in manual groups)1.
800,000 smokers successfully quitting at 4 week stage between 2003–2004 & 2005–2006.
2005–6
NK
204,151 quitters reported in 2003–2004.
Roll out of ‘5-A-Day’ programme to increase fruit and vegetable consumption.
2002
National School Fruit Programme – free piece of fruit for 4–6 year olds each school day
2001
Health authorities to have physical activity prevention schemes
2001
Department of Health to issue guidance on supervised programme of physical activity
2001
(II) DIET
ALCOHOL PHYSICAL ACTIVITY
IMPROVING SCREENING BREAST
Department of Health to NS encourage more walking and cycling, particularly in deprived areas
NA
Extension of Breast Screening to age 70.
2003
All 85 units expected to meet target by April 2005.
All women will have two views of the breast taken at every screening.
2003
81 out of 85 breast screening units met targets. All will achieve target by August 2005.
The effectiveness of reorganization of UK cancer services 1387
SCOTLAND TARGET
OVERALL
TARGET YEAR 20% reduction in cancer mortality in under 75s from 1999.
TARGET MET
TARGET YEAR ACHIEVED
2010
NA
1995–2005
2010
NA
Tobacco Control Action Plan & extending and improving smoking cessation services.
2004
NK
NK
Partnership Action on Tobacco Health (PATH)
NS
NA
2002
Nicotine replacement therapy on prescription.
NS
NA
2001
Food and Health coordinator.
NS
NA
2001
Expert panel on nutritional standards.
NS
NA
2001
Breakfast club challenge fund – free fruit to all Primary 1 and 2 pupils.
NS
NA
2003
ALCOHOL
Plan for Action on Alcohol.
2001
2002
PHYSICAL ACTIVITY
National Physical Activity Coordinator.
NS
NA
2002
National Physical Activity Strategy.
NS
NA
2003
Increase proportions of men and women active for 20 minutes per day to 50% and 40% respectively
2005
2004
2004
IMPROVING Reduce smoking rate PREVENTION (I) from 35% to 33%. SMOKING Reduce smoking rate to average of 31%.
(II) DIET
IMPROVING SCREENINGBREAST
Introduction of extension of 2003/04 breast screening up to age 70. Roll out over one 3 year round of screening.
NA
PROGRESS
2002: 150.6 deaths per 100,000 (10.4% improvement over 1995 baseline).
1388 The organization of UK cancer serivces: a UK perspective
Table 59.1 (Continued) ENGLAND TARGET
- CERVICAL
- BOWEL
- PROSTATE
IMPROVING CANCER SERVICES IN COMMUNITY
CUTTING WAITING FOR DIAGNOSIS & TREATMENT
TARGET YEAR
TARGET MET
PROGRESS
All PCTs to review cervical screening coverage in deprived and minority ethnic groups.
2001
Department of Health does not monitor target. Good activity in some areas.
Health authorities to achieve 80% cervical screening coverage by 2002.
2002
Target dropped – considered unfair expectation for all PCTS, because of smaller size.
Bowel screening introduced in 2003 if pilot study supports this.
NA
NA
Insufficient evidence by 2003 to take a decision.
Training initiatives in Endoscopy.
2001
3 national and 7 regional training centres began training programme in 2004.
Prostate Cancer Risk Management programme.
2001
Good progress in all areas.
Prostate Cancer (early diagnosis, treatment, care and research) programme.
NS
NA
Initiated 2003.
Lead cancer clinician in each Primary Care Trust.
2001
By April 2004, 279 out of 303 PCTs had lead cancer clinicians in post. But questions over continued funding.
Training programmes in palliative care.
NS
Training programmes for palliative care delivered in all 34 cancer networks. Funding increased to £2m p.a. for three years. 10,000 nurses participated to date.
2 week wait for outpatient By end 2000 appointment after urgent GP referral.
Now 99.2% of people with suspected cancer seen by a specialist within two weeks. 93.5% of people with suspected cancer seen by a specialist within target by 2001.
Maximum 1 month from urgent GP referral to treatment for leukaemia, testicular and children’s cancers.
By 2001
Between 95% and 100% now achieved (91.5%–100% by 2002).
Maximum 1 month wait from diagnosis to treatment for all cancers by 2005.
By end 2005
89.9% of patients diagnosed with cancer treated within 31 days by June 2004.
Maximum 1 month from diagnosis to treatment for breast cancer by 2001, and maximum 2 months from urgent GP referral to treatment by 2002.
2002
97% of both targets for patients achieved. 94.2% by 2002 for 1 month target, and 96.5% by 2003 for 2 month target.
The effectiveness of reorganization of UK cancer services 1389
SCOTLAND TARGET
TARGET YEAR
TARGET MET
TARGET YEAR ACHIEVED
PROGRESS
Introduction of liquid based cytology.
2003
National IT based call-recall system for cervical cancer screening.
2003
- BOWEL
Consider introduction of bowel screening if trials show significant impact
NS
NA
NA
- PROSTATE
Await results of prostate cancer screening trials
NS
NA
NA
Maximum wait from urgent referral for all cancers will be 2 months.
2005
75% for 01/07/65–30/09/05
Maximum 1 month from diagnosis of breast cancer to treatment
2001
Approximately 80% by 2004.
- CERVICAL
Acceptance testing of callrecall functionality in June 2003
2001 Scottish Executive approves PSA testing for symptomatic men
IMPROVING CANCER SERVICES IN COMMUNITY
CUTTING WAITING FOR DIAGNOSIS & TREATMENT
1390 The organization of UK cancer serivces: a UK perspective
Table 59.1 (Continued) ENGLAND TARGET
IMPROVING TREATMENT
IMPROVING CARE
TARGET YEAR
TARGET MET
PROGRESS
Maximum 2 month wait from urgent GP referral to treatment for all cancers by 2005.
By end 2005
78% of all urgently referred patients with cancer treated within 62 days by June 2004.
Roll out to Cancer Networks of Cancer Services Collaborative programme of service improvement by 2003–04.
2003–2004
Service improvement leads in every cancer network and service improvement underway.
2003
NICE appraisals of cancer drugs to end the postcode lottery.
NS
NA
16 cancer drugs appraised and 11 appraisals in progress. National Cancer Director in 2004 reported increasing drug use, but unacceptable variations between networks remain. New measures being introduced to address this.
Care of cancer patients to be reviewed by a specialist multi-disciplinary team.
NS
NA
More than 95% of trusts report providing care within teams for lung, breast, upper GI and bowel cancers.
National Cancer Standards published. Peer review visits begin.
2000 2001
Standards published in 2000; revised and extended in 2004. All cancer networks reviewed in 2001. Second review began in November 2004 to assess progress.
Plans to strengthen cancer registries.
By 2000
Action plan for cancer registries published in 2001. Cancer registration improved and links to cancer networks established.
National Cancer Datasets for all cancers.
By 2002
Largely complete. Datasets for some rarer cancers will not be completed until 2005.
Health authorities to agree palliative care investment with networks in 2001.
2001
Cancer network investment strategies for palliative care in place for 3 year period 2003–2006.
National Institute of Clinical Excellence (NICE) to commission Improving Outcomes guidance on all cancers by 2003 for Primary Care Trusts to implement.
4 new reports and 2 updates since 2000. Programme to be completed during 2005–06. During 2004 cancer networks developed action plans for implementing guidance setting firm milestones for delivery.
The effectiveness of reorganization of UK cancer services 1391
SCOTLAND TARGET
IMPROVING TREATMENT
TARGET YEAR
Cancer Service Improvement Programme (CSIP). Service redesign – 3 year programme from 2002.
NS
TARGET MET
NA
TARGET YEAR ACHIEVED
NA
PROGRESS
Regional facilitators in place since June 2003 – process mapping initiated in lung, colorectal and gynaecological cancer.
Scottish Medicines Forum established (2001) to review all new drugs and indicators
IMPROVING CARE
Scottish referral guidelines for suspected cancer.
2002
Implementation of SIGN guidelines and monitoring of services against standards of Clinical Standards Board for Scotland (now NHS Quality Improvement Scotland – NHS QIS).
NS
NA
Baseline NHS QIS assessment for breast, lung, colorectal and ovarian cancer 2001–2.
Completed by many but not all boards
All NHS boards to undertake 2002 comprehensive needs assessment for palliative care.
2002
1392 The organization of UK cancer serivces: a UK perspective
Table 59.1 (Continued) ENGLAND TARGET
INVESTING IN FACILITIES
TARGET YEAR
TARGET MET
PROGRESS
£50 million for hospices and specialist palliative care services by 2004.
2004
£50 million was allocated in 2003–4, with £38.5 million being spent to date.
Supportive care strategy to be developed, including standards for supportive and palliative care in 2001.
2001
Draft strategy published as Cancer Standards in 2002. Due to be completed in 2005.
NICE to develop guidance on supportive care in 2001.
2001
Guidance published March 2004.
Advanced communication skills training programmes.
NS
NA
Advanced communication skills training programmes developed. Pilot successful; roll-out started.
Cancer networks to take account of views of patients/carers in planning services by 2001.
2001
Partnership groups established in 30 out of 34 networks by 2004.
Cancer information advisory group will identify gaps and develop guidance on patient information in 2001.
2001
NA
Remit changed to focus on dissemination and delivery. Accreditation processes for information providers being developed, and cancer information leaflets are now available centrally for local use.
£400 million invested in new linear accelerators, MRI scanners and other cancer equipment.
By end 2004
Additional 50 MRI scanners, 200 CT scanners and 45 linear accelerators.
2000–3
Over £400 million invested in new facilities so far. By the end of 2004 the NHS had received 68 MRI scanners, 177 CT scanners, 83 linear accelerators and over 700 items of breast screening equipment since April 2000.
Additional £28 million invested to support modernisation of 23 pathology services.
NS
£28 million invested to support upgrades and reconfigurations in 29 pathology sites.
National cancer facilities strategy based on an audit by cancer networks by 2001.
2001
Audits undertaken. Single facilities strategy document not produced. New facilities strategy in development incorporating, for example, radiotherapy stocktake and PET-CT framework.
The effectiveness of reorganization of UK cancer services 1393
SCOTLAND TARGET
INVESTING IN FACILITIES
TARGET YEAR
TARGET MET
TARGET YEAR ACHIEVED
PROGRESS
Establishment of managed clinical networks for palliative care.
NS
NA
NA
Board for Scotland (NHS QIS) investment of £5 million in palliative care services 2000–2003.
Development of core standards for palliative care by clinical standards.
NS
NA
2002
Published 2002. Peer review process in process 2003. Report 2004.
NS
NS
Communication skills training for senior clinicians.
NS
NA
NA
Facilitated training course 2003
Provision of high quality information for patients.
NS
NA
Additional investment in imaging equipment.
NS
NS
Health Technology Board for Scotland review of PET technology.
NS
NA
£5 million investment for PET services from 2004–5.
Modernisation programme (£13 m) to replace ageing linear accelerators in Scotland.
NS
NA
£33.07 million on additional radiotherapy equipment since 1997. Scotland should have 5 linears/million by 2005–6.
1394 The organization of UK cancer serivces: a UK perspective
Table 59.1 (Continued) ENGLAND TARGET
INVESTING IN STAFF
INVESTING IN FUTURE
RESEARCH & GENETICS
INFORMATION TECHNOLOGY
TARGET YEAR
TARGET MET
PROGRESS
Nearly 1000 extra cancer specialists.
2006
Met ahead of schedule. 975 extra consultants in post by June 2004.
Increase specialist trainees.
2008
NA
No specific target in plan but 36% increase in training places between 1999 and 2003.
Training scheme to increase SHOs in histopathology. Extra 18 specialist registrars to be trained over 3 year pilot programme.
2001–2004
£1.3 million invested to fund 3 training centres. An intensive training and assessment centre for histopathologists from overseas was set up in 2003–04. In 2004–05 3 new training centres have been set up and 6 additional centres will come on stream in 2005–06.
Training places for radiographers to be increased; attrition rates reduced.
NS
NA
Training places have been more than doubled. Average attrition rates reduced significantly between 2001 and 2002 in diagnostics and radiotherapy.
Pilot sites for new radiography skill mix.
2000
NA
TBA
National Cancer Research Institute, including National Cancer Research Network, to be established.
2003
NCRI established 2001, and NCRN fully established in October 2002.
National translational cancer network (NTRAC)
2001
Established 2001.
Development of cancer genetic services (in partnership with Macmillan).
2001
Pilot work under way
The effectiveness of reorganization of UK cancer services 1395
SCOTLAND TARGET
INVESTING IN STAFF
INVESTING IN FUTURE RESEARCH & GENETICS
INFORMATION TECHNOLOGY
TARGET YEAR
TARGET MET
TARGET YEAR ACHIEVED
PROGRESS
Establishment of Human Resources subgroup of Scottish Cancer Group.
NS
NA
Additional 24 radiology and 16 pathology training posts 2001–3. Publication of Scottish Health Workforce Plan 2004
Commissioning by Scottish Cancer Group of modes of capacity planning for radiology, pathology and other services.
NS
NA
Development of capacity planning model 2002; Scottish Health Workforce Plan: the baseline, 2004 24 radiology and 16 pathology training posts created from 2001–2003
Role development and skills enhancement in diagnostic and therapeutic radiography.
NS
NA
Joint workshop of Scottish Executive, Royal College of Radiologists and Society of Radiographers on capacity and demand issues, 2004
Consultation on targets for accrual to clinical trials.
NS
NA
2002 Investment of £1 million in Scottish Cancer Research Network
Scottish Cancer Clinical Trials network to be established.
2002
ECCI system for protocol-based electronic referrals.
NS
NS
2002
Publication (2002) of Guidance on the Evaluation of IT systems for the Management of Cancer in Scotland. Establishment of IM & T project board.
1396 The organization of UK cancer serivces: a UK perspective
Table 59.1 (Continued) ENGLAND TARGET
IMPLEMENTING THE NHS PLAN
TARGET YEAR
TARGET MET
PROGRESS
Additional £570 million for cancer services by 2003–04.
2003–4
Preliminary results of current tracking exercise show target met.
Networks to develop service delivery plans, underpinned by workforce, education and training and facilities strategies.
2001
Strategic plans produced but of variable quality. By late 2003 a third of networks had a workforce and/or education & training strategy. Improving outcomes guidance and other initiatives are seeking to address this.
NOTES 1 General Household Survey data is now weighted. This roughly increases prevalence by one percentage point. KEY NK not known NA not applicable NS not stated
Addressing long waiting times for diagnosis and treatment is one of the most challenging tasks of all the national cancer plans. Delays can occur in presentation of patients, referral from primary care, initial assessment to establish a histological diagnosis, subsequent staging and delivery of treatment. Dissection of the ‘patient pathway’ to identify and address causes of delay has been central to reducing waiting times. Box 59.5 shows progress against stated targets for prevention, screening, cancer services in primary care, waiting lists for diagnosis and treatment and improving care in England and Scotland. All these reports attest to the substantial progress in the reorganization of UK cancer services. Some of the remaining major challenges in England are integration between the cancer networks and strategic health authorities tasked with commissioning cancer services and, in all the home countries, maintaining the necessary budget to ensure that targets set within each cancer plan are met.
Commission for Health Improvement and Audit Commission Report on cancer services in England and Wales The CHI and the Audit Commission (AC) were asked in 1999 by the UK government and National Assembly for Wales to assess the implementation of the Calman–Hine report published in 1995.1 While the Calman–Hine report2 set out the principles for reorganization of UK cancer
services, it did not provide a central plan for implementing its proposals. Each region in England and Wales developed its own strategy, reflecting the devolved structure of NHS management. The regions also carried responsibility for monitoring progress in implementing change. There was no centralized oversight nor additional resources for auditing progress against the objectives of Calman–Hine. It is not surprising that regional variations developed in how the proposals were implemented. The lesson for the future is that central cancer planners need to be more prescriptive about the process of the implementation to ensure consistency across the UK. The remit of the CHI/AC report1 covered the range of services that patients received from their first point of contact with the NHS in England and Wales. Their focus was what happened in practice, the reasons for it and how services might be improved and resources optimally used. Its findings on diagnosis, treatment and palliative care are shown in Box 59.5. From the patient’s perspective there was still poor communication and a lack of systematic planning of care between cancer professionals. There were geographical variations in the way cancer services for different tumours were provided. The report called for a fundamental change in the behaviour of staff, and priority to be given to addressing gaps in the planning of the management of individual patients. The CHI/NAO report highlighted differences in how active regions were in assigning cancer centre and cancer unit status. Some were very active, while others devolved responsibility to trusts at a local level.
The effectiveness of reorganization of UK cancer services 1397
SCOTLAND TARGET
TARGET YEAR
IMPLEMENTING Cancer managed THE NHS PLAN clinical networks will be fully established. Establish 3 Regional Cancer Advisory Groups
TARGET MET
TARGET YEAR ACHIEVED
PROGRESS
2002
Well developed for most cancers.
NS
NA
3 RCAGs established in south east, west and north of Scotland.
Box 59.5 Findings of commission for Health Improvement/National Audit Office review of cancer services in England and Wales (diagnosis, treatment and palliative care1) Diagnosis and treatment ● For the most part, these are given sensitively but a few patients have poor experiences ● Specialist nurses are valuable but in short supply ● In most hospitals, patients are seen by specialist within 2 weeks of referral ● Long delays in obtaining tests, especially CT and MRI ● Limited information from consultants to inform general practitioner about diagnosis and prognosis. Treatment and care ● Most patients are operated on relatively quickly ● Waiting for radiotherapy, even for urgent cases, is a problem in some areas ● Many patients are seen by non-specialists ● Most medical and clinical oncology professionals see patients with more than one type of cancer, and some are seen by total generalists ● Radiotherapy machines are unevenly distributed and many are old
National Audit Office report on patients’ experience of cancer services in England In 1999 to 2000, the Department of Health commissioned a baseline survey of the experience of cancer patients involving all trusts in England for six cancers and involving 65 000 patients.69 The report found high levels of satisfaction among patients in relation to dignity, respect and privacy
●
●
●
●
There is significant variation in machine use arising from differences in prescribing, hours of operation and availability of staff There is a lack of agreed policies for management of many cancers; even if they exist, practice is often not audited Hospital discharge is often not properly planned with appropriate briefing of general practitioners and district nurses Psychologists and counsellors are in short supply
Palliative and terminal care There is considerable variation in availability of palliative care consultants and nurses ● It is often unclear who has the principal responsibility for the patient ● There is limited availability of services out of hours ● Provision of hospices across the UK is patchy. ●
but identified areas for improvement. A follow up survey in 2004 concentrated on four cancers (breast, bowel, lung and prostate cancer); 7800 patients were invited to participate, and the response rate was 55 per cent. This was supplemented by visits to hospitals and hospices, focus groups and interview with GPs, nurses, patients and carers. Overall progress had been made. However, for a minority of patients experiences of communicating information, symptom relief
1398 The organization of UK cancer serivces: a UK perspective
Table 59.2 Key items from surveys of cancer patients’ experience of treatment and care: comparison between 2000 and 200468 with permission. 2000 Patients did not see a worsening in their condition while waiting to see a specialist Patients told what was wrong with them with sufficient sensitivity and care (n/s) Doctors or nurses discussed the purpose of treatment with patients, and patients completely understood the explanation Patients found doctors’ explanation of condition, treatment or tests very easy to understand Patient always had trust and confidence in nurses Patients with strong religious beliefs felt beliefs were taken into consideration by hospital staff Printed information given to patient at discharge covered all the issues Patient told about support or self-help group Patient had enough privacy at examination at their last out-patient visit Patient had a lot of confidence and trust in the doctor at their last out-patient visit
2004
74 94 82
80 94 86
62 79 * * 61 99 68
68 81 91 96 60 97 84
Items are drawn from representative questions for each ‘theme’ within the 2004 NAO Cancer patient survey. For the items marked (n/s), the year on year change is not statistically significant. Questions marked * not asked in 2000.
and lack of options at the end of life were suboptimal. The comparison of patients’ experiences in 2000 and 2004 are shown in Table 59.2. In the second report by the NAO on cancer services in England,67 it was noted that substantial progress had been made with a substantial number of targets met. The downward trend in mortality was ahead of the forecast of the Department of Health target of a 20 per cent reduction in mortality from cancer in individuals under the age of 75 by 2010. There was a mixed picture in the functioning of the cancer networks with some making substantial progress but others less effective in terms of staffing, planning of cancer services or the necessary support from other elements of the health care system. Cancer networks preceded the development of PCTs or strategic health authorities and working arrangements within these still need to develop further. Virtually all the cancer networks report progress in meeting objectives of the NHS plan. The following areas of progress were reported: ● ● ● ● ● ● ● ●
●
Boosting the downward trend in smoking Extending the breast screening programme Accelerating access to cancer diagnosis and treatment Establishing specialist cancer teams Reducing variation in access to cancer drugs Increasing specialist palliative care services Appointing more cancer specialists Modernizing and expanding cancer diagnostic and treatment facilities Increasing the pace of research.
with fundamental issues such as clarity of purpose, consistency of senior membership and objectives.
Network management teams As a minimum each network should include a lead clinician, lead nurse, lead manager and lead for service development. In practice this was often extended to a lead for pharmacy, audit, palliative care, public health and patient advocacy. The NAO survey67 showed that all networks had a lead clinician in place but the creation of additional lead posts (pharmacy, IT, palliative care) varied. Networks reported that the main reason for vacancies not being filled was inadequate finance. There was wide variation in the perception among networks about the adequacy of funding. In some cases, PCTs would not pay for posts. An assessment of network management teams’ assessment of the effectiveness of relationships between member organizations showed relatively low levels for Strategic Health Authorities (Table 59.3).
Network service delivery plans All networks were required to provide a network service plan by the end of October 2001. Some were well developed and comprehensive. Others did not show any strategic awareness, were characterized by omissions and varied in structure and content.
Workforce strategy Cancer networks Discussion between the NAO and the Cancer Action Team indicated that a number of network boards were still dealing
Despite the priority given to workforce planning, only 12 of the 34 networks had developed an education and training strategy by 2003.
The effectiveness of reorganization of UK cancer services 1399
Table 59.3 Network management teams’ assessment of the effectiveness of relationships with constituent organizations [Source: NAO survey,67 with permission]. Constituent organization Acute trusts PCTs SHAs Voluntary sector Local authorities
Very good
Good
Adequate
Poor
Very poor
11 (32%) 8 (24%) 11 (32%) 7 (21%) 0
20 (50%) 15 (44%) 15 (44%) 19 (56%) 0
3 (9%) 8 (24%) 12 (35%) 7 (21%) 10 (34%)
0 3 (9%) 2 (6%) 1 (3%) 10 (34%)
0 0 0 0 9 (31%)
PCTs, Primary Care Trusts; SHAs, Strategic Health Authorities.
Commissioning cancer services Most networks were adopting or planning to adopt a network-wide process of commissioning cancer services. The Cancer Action Team observed that some PCTs were producing their own commissioning plans independently of other PCTs. The risk of the latter strategy is that the priorities identified by the network may not be addressed. There was also variation in how network management teams provided input to the process of delivery of local services. Some networks were actively engaged with PCTs to help them develop local services, but others were not. Major difficulties experienced in commissioning included variable completeness, quality and consistency of local data. The quality of PCT planning documents varied widely. In addition, the approach to monitoring progress in local planning also was inconsistent. There was often confusion within the network as to who was responsible for monitoring progress against targets. The NAO found inconsistency on the frequency of monitoring progress of the plan. This is an important weakness in the implementation process where there needs to be absolute clarity about who is responsible for tracking progress against targets. In four of the 34 networks progress was monitored monthly, in around half quarterly and in seven less frequently than quarterly. In five no monitoring was carried out at all. However, all the PCTs visited by the NAO were monitoring progress on achieving key targets on a regular basis.
Dividends from the establishment of networks Examples of dividends of networks cited by the NAO67 are the introduction of new drugs, funding of specialist palliative care, action plans for the development of cancer services in accordance with guidance of the National Institute of Clinical Excellence (NICE) and facilitating the development of multidisciplinary teams. However, the effectiveness of each network varied. Obstacles to progress included insufficient resources for a network to function effectively, suboptimal staffing of network management teams, difficulties in establishing cross-boundary working and inadequate representation from acute and PCTs. The
Department of Health in conjunction with strategic health authorities needed to underpin the functioning of networks by ensuring that the roles of responsibilities of its member organizations were well defined and adhered to. In the third of its reports68 patients were broadly positive about their experience with GPs and how they were informed about cancer. However, patients with prostate cancer and those who lived in London were less positive about their experience. Levels of satisfaction with GP care were lower for patients with bowel cancer (61 per cent) compared with those with other cancers (67 per cent).
Modernizing cancer services Systematic approaches to improving the ‘patient journey’ have been adopted. In England, the Cancer Services Collaborative ‘Improvement Partnership’ in association with the Department of Health: Cancer Waits Project has identified from a number of demonstration sites factors that help to achieve the national target of a max-imum 62day wait from referral to treatment for all patients. As the report recognizes,73 whole system change in the NHS is complex. Cancer care is only one part of the NHS and intimately linked to it. However, the 26 demonstration sites show that redesigning patient pathways is feasible, and sustainable reductions in waiting times can be achieved. This is illustrated for colorectal cancer where referral to first definitive treatment was achieved in 54 days (Fig. 59.4). Commitment of clinicians and managers to achieving improvement was essential. Executive clinical and Chief Executive/Board leadership was needed to unblock constraints and ensure proposals were actioned. Redesigning a single patient pathway or overcoming an obstacle such as access to imaging may lead to improvement in that specific area. However, there are often wider constraints across the pathway, such as theatre capacity, that need to be addressed. Patient pathway management systems have to be integrated with hospital-wide systems to be effective. Scaling up from demonstration projects to a whole service requires a combination of sustained focus, organizational effort and leadership. A whole system approach needed for sustainable change is shown in Fig. 59.5.
Ref–1st def 54 days 8/12 1st seen Flexi Sig
2ww ref
Ref–1st seen 10 days
29/11 26/11/2004
6/12
19/1 surgery 1st def
22/12 diag to pt dec treat 15/12 19/12 CT MDM
13/12
20/12
Dec trt–1st def 28 days
27/12
3/1
10/1
14 days
17/1
24/1
31/1
7/2
21/2 27/02/2005
31 days
14 day target
14/2
31 day target
62 days
Figure 59.4 Application of best practice across the patient pathway. Colorectal pathway from referral to first definitive treatment in 54 days, Queen Mary’s Sidcup NHS Trust. (Reproduced with permission from the Department of Health.)74
e System Approach Whol ORGANISATIONAL FACTORS
Effective Communication
Executive Leadership and Focus (Clinical & Managerial)
Service Improvement
Clinical & Managerial Engagement and Partnership
Performance Improvement
Organisational Culture
DELIVERING EFFECTIVE PATHWAYS
Effective evidence-based pathway design
Prospective patient management and navigation
Robust data information and administrative systems
Figure 59.5 Whole system approach to improving cancer services. (Reproduced with permission from the Department of Health.)74
The effectiveness of reorganization of UK cancer services 1401
Box 59.6 Guidelines on reducing waiting at referral, diagnostic, treatment and follow stages74
Box 59.7 Key approaches to reducing waiting times for radiotherapy75 1. Are there referral guidelines and clinical protocols for treatment? 2. Are there system links to the MDT meeting (e.g., booking appointments to see the oncologist from the MDT meeting to reduce administrative delay)? 3. How far in advance are patients booked? Scheduling months in advance increases non-attendance due to death, patients forgetting, holidays. Only book as far as longest patient pathway and ‘pend’ patients until that time. 4. How many times does the patient visit? Can the steps/visits be reduced? 5. Is the doctor/skill in the right part of the process? Scan/simulate the patient prior to the scheduled doctor slot, schedule time for them to define, check and prescribe treatment 6. What fractionation regimen is used? A shorter evidence-based palliative (5 instead of 10 fractions) or radical radiotherapy regimen might be used? 7. Is the workforce being used appropriately? Appropriate administrative, booking and helper support can enable therapy radiographers to increase capacity. 8. Have patients been involved? There is no point in planning to work 12-hour days if most patients come by hospital transport or prefer not to come after 4.30 pm.
The following should be in place: ● ● ● ● ● ●
Demand management systems Referral protocols for all tumour sites Single referral pathway, single queue Pooling of referrals Defined patient pathways Robust booking and scheduling systems.
At the diagnostic phase, the following should be in place: ●
● ● ● ●
● ●
Triaging patients – ‘straight to test’ before first out-patient visit Capacity matched to demand for diagnostic tests Combined test and visits (one or two stops) Agreed protocols for diagnosis and staging Extended roles for nurses and radiographers (endoscopy, TRUS biopsy and imaging) Results communication system Proactive pathway management – using trackers and navigators.
Effective multidisciplinary working requires that: ● ●
● ●
All relevant information is available Decisions are recorded and disseminated to all relevant parties Waiting times are monitored proactively Further steps in pathway are planned, booked and coordinated.
Effective strategies for reducing consultant-led follow-up, so releasing capacity for other essential tasks include: ●
● ●
●
Gaining clinical and managerial support to redesign service New/extended roles, e.g., nurse-led follow-up Discharging patients from routine follow-up but with arrangements for patients to contact service if needed Implementing agreed protocols.
Primary care There was a recognition in the National Cancer Plan for England30 that cancer services in the community were patchy. Patients with symptoms related to cancer were often not identified early enough or patients referred sufficiently quickly for diagnosis and treatment. A primary care cancer lead (PCCL) was to be appointed in each PCT to coordinate improvement of cancer services in the community. There are five principal aspects of the role of the PCCL: ●
Considerable experience has been built up in the Cancer Services Collaborative Improvement Partnership focussed on referral, diagnosis, treatment planning and follow up.74 To reduce waiting at referral, diagnostic, treatment and follow up stages,75 practical guidelines have been produced (Box 59.6). A number of useful ‘how to do it’ toolkits have been devised. For reducing waiting times for radiotherapy some of the key questions and solutions75 are shown in Box 59.7. A similar approach to service redesign has been adopted by the Cancer Service Improvement Programme in Scotland.76
●
●
●
●
To provide strategic leadership within the primary care group/trust (PCG/T) in line with the NHS plan to develop services for cancer patients To contribute to network development and, where appropriate, to represent the PCG/T and primary care in the cancer network To contribute to the development of network service delivery plans in order to streamline patient care To raise standards of cancer care with the PCG/T in collaboration with the clinical governance lead To ensure the services are responsive to the needs of people affected by cancer.
A review in 200477 indicated that it was premature to judge whether PCCLs had succeeded in improving the experience
1402 The organization of UK cancer serivces: a UK perspective
of patients experience since they were but a small part of a number of cancer initiatives. Just over half of PCCLs had focused on a specific aspect of cancer service, particularly palliative care. Some PCCLs were unable to identify any improvements, while others felt there had been improvement in palliative care services, better access and speed of referrals, including 2-week wait changes.77 Remaining challenges were the engagement of general practitioners in cancer services in the community and inadequate sessional time for PCCLs. Organizational barriers, lack of capacity, competing priorities and poor infrastructure were identified as problems by some PCT managers.
progress against its stated objectives. Four disease groups were established in 2001 (breast, colorectal, lung and gynaecological cancer). Subsequently four additional groups were added (skin, haematology, head and neck, and urology) in 2002 and upper gastrointestinal in 2003. Each disease site group has a lead clinician, a lead GP, a lead nurse and a patient lead. The SCAN is supported by an administrative team, an audit manager, a telemedicine project manager and a Cancer Information Network project manager and team. Breast, colorectal and urology groups were selected for evaluation based on:
Progress in implementation of national cancer plan in Scotland
●
●
●
In Scotland significant progress has been made in reducing waiting times for diagnosis and treatment by a combination of new investment (£25 million recurring ring-fenced until at least 2005 to 2006) and redesign of services. By 2004,72 approximately 80 per cent of women were treated within 1 month of diagnosis. However, the maximum wait from urgent referral to treatment for all cancers within 62 days had only been met in 73 per cent of cases. A programme of the Centre for Change and Innovation underway aimed at improving experiences and outcomes for patients with suspected or confirmed cancer.72 By undertaking process mapping of the steps along the patient journey, substantial changes have reduced waiting times, for example, for radiotherapy in the South East Scotland Cancer Network. A programme of development of national networks has been implemented focussing on quality improvement, redesign of services, protocols, audit and patient involvement and information.
Evaluation of Managed Clinical Networks in Scotland The first formal evaluation78 of managed cancer clinical networks in Scotland was undertaken in the South East Cancer Network (SCAN). This network was launched in 2000 and was independently evaluated in 2004. At the time of evaluation there were no specific benchmarks of a successful managed clinical network (MCN). For an organization as complex as an MCN it is difficult to assign particular weight to factors that have or are likely to influence outcomes. However, it seems probable that development of and adherence to common protocols of evidence-based best practice are likely cumulatively to make a significant contribution to improving clinical outcomes. Since the MCNs are new structures, there is little information on what constitutes a successful MCN. The evaluation of the SCAN MCN sought views from members of the network on how they perceived it to be working as well as assessing
●
Length of establishment and number of activities (such as date of establishment, availability of audit data and production of protocols Disease factors (complexity of treatment and of presentation Staff issues ( e.g., difficulties with recruitment) External factors (e.g., alternative data sets for validation).
Assessment was based on semi-structured interviews, observation of tumour group meetings and documentary evidence (correspondence, minutes of minutes and audit reports). Interviewees in general had a common understanding of the concept of MCNs. Multidisciplinary working, cancer site specialization and protocols were seen as important goals for effecting change in cancer care. There was similar consensus on achieving equity of access to care across the region. Most interviewees felt that the membership of each tumour group was appropriate. Significant benefits were seen as contact with other members of the network, a culture of regional working and MCNs being less hierarchical with members being seen as equal partners. A large minority of interviewees felt that patient involvement had influenced strategy or services, while many others felt that it had as yet had little impact. A key concern expressed by SCAN members was its ability to allocate funds according to regional priorities. For many, this was central to the survival of the network and engagement of its members. Key individuals, particularly the clinical leads, were seen as important in getting members to pull together and making a success of the network. Overall, progress within the SCAN tumour groups was mixed. However, the authors of the evaluation point out that progress of different groups does not necessarily indicate success or failure of particular tumour groups. Some cancer site networks may take longer to establish than others because of their inherent complexity. The successes of SCAN included improved relationships, connections and working across the region, group work plans, administrative support, formal organizational structure and associated networks (such as nursing). Barriers were identified as lack of time available for meeting and participation in working groups in already busy schedules; and lack of control over resources to effect change and sustainability. The Regional Cancer Advisory Group was perceived as the real decision maker in the region.
Future directions in organization of cancer services 1403
In terms of implementing key processes and core principles, progress was mixed. Significant progress on cancer site specialization had been made in all three tumour groups. The collection of regional audit data was well established. Data quality monitored by the Scottish Cancer Intelligence Unit was high at 95 per cent. Observation of the tumour groups suggested that results of the audit had resulted in changes to the configuration of services. Redesign of services had occurred. However, no information was available for evaluation on patient experience. All tumour groups had some patient representation. There were mixed views among interviewees about the impact of patient involvement on services or strategy. No negative views were expressed on patient involvement, but it was not a priority for most respondents. A quality assurance programme has been developed by NHS Quality Improvement Scotland for all cancer MCNs. The network states the level of performance required in relation to explicit standards, based on the core principles governing MCNs. The MCN has to justify the targets set and the criteria that indicate whether the target has been met. Each tumour group has education and training on its work plan but neither SCAN nor its tumour groups had yet addressed the issue. SCAN was directly involved in the distribution and monitoring of 50 per cent of the new funding for cancer in the region, with the remaining 50 per cent being assigned by the NHS areas involved. There was no data available to indicate whether or not the network has been able to generate better value for money.
would be concentrating oncological expertise, a critical mass of individuals for research and development, economies of scale in mass processing of blood and tissue sampling and imaging and treatment modalities. At the other extreme is a much more geographically distributed cancer service with as many services available locally as possible. This might include radiotherapy for the simpler radical and palliative treatment techniques. This would require a network of radiotherapy facilities under the aegis of the cancer centre for the purposes of quality assurance. All the radiotherapy planning could be carried out in the cancer centre and treatment delivered locally in centres with one or two linear accelerators. The benefits for patients would be the avoidance of lengthy journeys for daily radiotherapy over many weeks or hotel-style accommodation far from friends and family. Critical to these planning decisions will be clarity that the same high standards of radiotherapy are available in cancer units and cancer centres. An aging population will present challenges of delivering cancer service to older patients where issues of mobility and co-morbidities will create pressure to provide cancer services locally.81 Satellite radiotherapy centres may be appropriate to minimize the need to travel to metropolitan radiotherapy centres. The following categories of patients might be considered for treatment in a linked unit if the patient is fit and unlikely to need additional oncological review:82 ● ● ● ●
FUTURE DIRECTIONS IN ORGANIZATION OF CANCER SERVICES The landscape of cancer care will be fashioned by the impact of preventive strategies, techniques of early diagnosis, novel therapies and the resources available to support them.
Predicting demand for radiotherapy services Various attempts have been made to predict the national requirements for radiotherapy capacity based on the evidence base of indications for radiotherapy. Delaney and colleagues79 calculated that 53 per cent of patients with cancer would require external beam radiotherapy. Similar modelling in Scotland suggests that radiotherapy will be required for approximately 14 000 to 15 000 of the estimated 31 500 new cancer patients diagnosed annually by 2015.80
A centralized or devolved pattern of cancer services? Highly centralized cancer regional or supraregional supercentres are one possible future scenario. The benefits
● ● ● ●
●
Metastatic disease Breast cancer (post-mastectomy) Breast cancer (post-conservation surgery) Prostate cancer (radical external beam local therapy) Bladder cancer (radical external beam therapy) Lymphoma (palliative treatment) Gynaecological cancer (pelvic external beam irradiation) Colorectal cancer (preoperative and postoperative pelvic therapy) Oesophageal cancer (palliative external beam therapy)
It is predicted81 that the costs of caring for a patient with cancer (£20 000 in 2003 for direct medical care of which 70 per cent is spent in the last 6 months of life without covering the costs of other medical conditions and social care) will rise markedly. Balanced against the increased costs of treating individual patients, the costs of in-patient care should fall as more patients are treated on an out-patient basis. Fewer oncological beds may be needed. Hotel-style accommodation may need to be expanded for patients needing to travel for complex treatments. Whether the state will continue to fund the whole infrastructure of cancer care in the future remains uncertain. In 2000, the UK government increased investment in the NHS by 7 per cent per year for a period of 7 years, unparalleled in any other health system. If the trend in England for a plurality of providers spreads to the rest of the UK, the challenge will be to coordinate the patient journey through networks of public and private providers with
1404 The organization of UK cancer serivces: a UK perspective
shared information systems, clinical accountability and systems of quality assurance to agreed standards. New patterns of collaborative working between state and independent health sectors will be needed if cancer care is to be seamless. It is too early to assess the net gain or loss to the effectiveness of the organization of cancer services from the diverging patterns of funding and changing NHS administrative structures. The fall in cancer mortality in the UK by 2010 will be the test of the robustness of the evolving organization of cancer services.
KEY LEARNING POINTS ●
●
●
●
●
●
●
●
●
●
National cancer plans have provided a road map for improving services. Improvements in cancer services have led to an improvement in cancer mortality. Patients’ experience of cancer care is improving but communication and waiting for diagnosis and treatment are still significant challenges. Consistent, systematic planning and delivery of cancer services is key to improving cancer outcomes. Multidisciplinary management according to evidence-based protocols reduces regional and intra-regional variations in practice. Regional networks of cancer professionals provide a successful template for service development and clinical research. Setting and monitoring of clinical standards are integral to improving care. Establishment of the National Cancer Research Network has contributed to significant increase in recruitment to clinical trials. Better information systems are needed to monitor cancer services. High-quality cancer services require sustained investment in estate, personnel and equipment.
REFERENCES ●1
Commission for Health Improvement and National Audit Office. NHS Cancer Care in England and Wales. National Service Framework Assessments No 1. London: Her Majesty’s Stationery Office (HMSO), December 2001. ●2 Expert Advisory Group on Cancer. A Policy Framework for Commissioning Cancer Services: a Report to the Chief Medical Officers of England and Wales (The Calman-Hine Report). London: Department of Health, 1995. 3 Carter D. Acute Services Review Report. Edinburgh: The Scottish Office, 1998. 4 Bagshawe KD on behalf of the Standing Medical Advisory Committee. Acute services for cancer: report of a Working Group. London: Standing Medical Advisory Committee, 1984.
5 Joint Council for Clinical Oncology, Cancer Care and Treatment Services. Advice for Purchasers and Providers. Report of an Expert Group. London: Royal College of Radiologists, September 1991. ◆6 Haward RA. The Calman-Hine Report: a personal retrospective on the UK’s first comprehensive policy on cancer services. Lancet Oncology 2006; 7:336–46. ●7 Sainsbury R, Haward R, Rider L et al. Influence of clinician workload and patterns of treatment on survival from breast cancer. The Lancet 345:1256–70. 8 Basnett I, Gill M, Tobias JS. Variations in breast cancer management between a teaching and non-teaching district. Eur J Cancer 1986; 28A:1945–50. ●9 Gillis CR, Hole DJ. Survival outcome of care by specialist surgeons in breast cancer: a study of 3786 patients in the West of Scotland. Br Med J 1996; 312:145–8. ●10 Stiller CA, Draper GJ. Treatment centre size, entry to trials and survival in acute lymphoblastic leukaemia. Arch Dis Child 1989; 64:657–61. 11 Matthews HR, Powell DJ, McKonkey CC. Effect of surgical experience on the results of resection for oesophageal cancer. Br J Surg 1986; 73:621–3. 12 Stiller CA. Centralisation of treatment and survival rates for cancer. Arch Dis Child 1988; 63:23–30. 13 Harding MJ, Paul J, Gillis CR, Kaye SB. Management of malignant teratoma: does referral to a specialist unit matter? Lancet 1993; 341:999–1003. ●14 Junor EJ, Hole DJ, Still RM, et al. Management of ovarian cancer: referral to a multidisciplinary team matters. Br J Cancer 1994; 70:363–70. 15 Kingston RD, Walsh S, Jeacock J. Colorectal surgeons in district general hospitals produce similar survival outcomes to their teaching hospital colleagues: review of 5 year survivals in Manchester J Roy Coll Surg Edin 1992; 37:235–37. 16 Hakama M, Karjalainen S, Hakuliinen T. Outcome based equity in the treatment of colon cancer patients in Finland. Int J Technol Assess Health Care 1989; 5:619–30. ●17 Department of Health and Social Services. Investing for the future. Report of cancer working group. Belfast: Department of Health and Social Services, 1996. 18 Review of the pattern of cancer services in England and Wales. London: Association of Cancer Physicians, 1994. 19 Sikora K. Cancer survival in Britain. BMJ 1999; 319:461–2. ●20 Coebergh J, Sant M, Berrino F, Verdecchia A. Survival of adult cancer patients in Europe diagnosed from 1978-1989: the EUROCARE II study. Eur J Cancer 1998; 34:2137–278. ◆21 Selby P, Gillis C, Haward R. Benefits from specialised cancer care. Lancet 1996; 348:313–18. ●22 Manual of Cancer Services Standards. London: NHS Executive, December 2000. ●23 Richards MA, Baum M, Dowsett, et al. Provision of breast services in the UK: the advantages of specialist breast units. Report of a working party of the British Breast Group. London: British Breast Group, 1994. 24 Haines A, Jones R. Implementing findings of research. BMJ 1994; 308:1488–92.
References 1405
25 Cancer Guidance Sub-Group of the Clinical Outcomes Group. Improving outcomes in breast cancer: guidance for radiotherapy. London: Royal College of Radiologists, 1993. ●26 Berrino F, Sant M, Verdecchia A, et al. Survival of cancer patients in Europe –The Eurocare Study. IARC Scientific Publication No.132. Lyon: IARC, 1995. ●27 Scottish Cancer Coordinating and Advisory Committee. Commissioning Cancer Services Subcommittee (1995), Interim Report to the Chief Medical Officer. Edinburgh: Scottish Office, Department of Health, 1995. ●28 Cancer Services Expert Group. Cancer Services in Wales (Cameron Report). Cardiff: NHS Wales, 1996. ●29 Coebergh J, Sant M, Berrino F, Verdecchia A. Survival of adult cancer patients in Europe diagnosed from 1978-1989: the EUROCARE II study. Eur J Cancer 1998; 34: 2137–278. ●30 Department of Health. The national cancer plan. London: Department of Health, 2000. ●31 Cancer in Scotland: Action for Change. Edinburgh: Scottish Executive, July 2001. 32 Board of Faculty of Clinical Oncology. Equipment, workload and staffing for radiotherapy in Scotland 1992–7. London: Royal College of Radiologists, 2000. 33 Board of the Faculty of Clinical Oncology. The Royal College of Radiologists, The Society and College of Radiographers, The Royal College of Nursing, The Institute of Physics and Engineering in Medicine. Breaking the mould: roles, responsibilities and skills mix in department of Clinical Oncology. London: Royal College of Radiologists, 2002. 34 Board of Faculty of Clinical Oncology. Radiotherapy dose-fractionation London: Royal College of Radiologists, April 2006. ●35 Scottish Office. Department of Health. Cancer. Introduction of managed clinical networks within the NHS in Scotland. NHS MEL (1999) 10. Edinburgh: The Scottish Office, 1999. 36 Scottish Office. Designed to Care. Edinburgh: The Stationery Office, 1997. 37 Scottish Executive. Partnership for Care: Scotland’s Health White Paper Edinburgh: The Stationery Office, 2003. 38 Kunkler IH. Managed Clinical Networks: a new paradigm for clinical medicine. J Roy Coll Physicians London 2000; 34:230–3. 39 National Assembly for Wales. Improving Health in Wales: A plan for the NHS and its partners. Cardiff: National Assembly for Wales, 2001. 40 NICaN, Northern Ireland Cancer Network. Annual Report Belfast: NICaN, 2004–05. ❋41 Guidance on Commissioning Cancer Services. Improving outcomes in Breast Cancer. London: NHS Executive, July 1996. ❋42 Guidance on Commissioning Cancer Services. Improving outcomes in Colorectal Cancer. London: NHS Executive, November 1997. ❋43 Guidance on Commissioning Cancer Services. Improving outcomes in Lung Cancer. London: NHS Executive, June 1998.
❋44
❋45
❋46
❋47
❋48
❋49
❋50
❋51
❋52
❋53
❋54
❋55
❋56
❋57
❋58
❋59
❋60
61
62 63
●64
65
●66
Guidance on Commissioning Cancer Services. Improving outcomes in Gynaecological Cancer. London: NHS Executive, July 1999. Guidance on Commissioning Cancer Services. Improving outcomes in Upper Gastro-intestinal Cancers. London: NHS Executive, January 2001. Clinical Standards for Lung Cancer. Clinical Standards Board for Scotland, January 2001. Clinical Standards for Ovarian Cancer. Edinburgh: Clinical Standards Board for Scotland, January 2001. Clinical Standards for Colorectal Cancer. Edinburgh: Clinical Standards Board for Scotland, January 2001. Clinical Standards for Breast Cancer. Edinburgh: Clinical Standards Board for Scotland, January 2001. Clinical Standards for Breast Screening. Edinburgh: Clinical Standards Board for Scotland, December 2002. Clinical Standards for Palliative Care. Edinburgh: Clinical Standards Board for Scotland, June 2002. Breast Cancer Services: All Wales Minimum Standards. Cardiff: Cancer Services Co-ordinating Group, 2000. Colorectal Cancer Services: All Wales Minimum Standards. Cardiff: Cancer Services Co-ordinating Group, 2000. Gastro-oesophageal Cancer Services: All Wales Minimum Standards. Cardiff: Cancer Services Co-ordinating Group, 2000. Gynaecological Cancer Services: All Wales Minimum Standards. Cardiff: Cancer Services Co-ordinating Group, 2000. Haematological Cancer Services: All Wales Minimum Standards. Cardiff: Cancer Services Co-ordinating Group, 2000. Lung Cancer Services: All Wales Minimum Standards. Cardiff: Cancer Services Co-ordinating Group, 2000. Skin Cancer Services: All Wales Minimum Standards. Cardiff: Cancer Services Co-ordinating Group, 2000. Urological Cancer Services: All Wales Minimum Standards. Cardiff: Cancer Services Co-ordinating Group, 2000. Specialist Palliative Care as applied to Cancer Services: All Wales Minimum Standards. Cardiff: Cancer Services Co-ordinating Group, 2000. Peppercorn JM, Weeks JC, Cook EF, Joffe S. Comparison of outcomes in patients treated within and outside clinical trials: conceptual framework and structural review. Lancet 2004; 363:263–70. Cross M. Will ‘Connecting for Health’ deliver its promises? BMJ 2006; 332:599–601. Cardy P, Sikora K. Cancer 2025: Conclusion. In: Sikora K. (ed). Cancer in 2025: The future of cancer care. Expert Rev. Anticancer Ther 4 (3) Suppl, 2004, S75–77. Department of Health. Tackling cancer in England: Saving More Lives. London: National Audit Office, March 2004. Bennett JA, Brewer C, Beyko S, Bennett SAL. Cancer Control Strategies in Eleven OECD Countries. Ottawa: University of Ottawa, 1999. Haward R, Amir Z, Borrill C, et al. Breast cancer teams: the impact of constitution, new cancer workload, and methods of operation on their effectiveness. Br J Cancer 2003; 89:15–22.
1406 The organization of UK cancer serivces: a UK perspective
●67
●68
●69
70 71 72 ●73
●74
●75
Department of Health. The NHS Cancer Plan: a progress report. London: National Audit Office, March 2005. Department of Health. Tackling cancer: Improving the patient journey. London: National Audit Office, February 2005. Department of Health. National surveys of NHS Patients. Cancer. National Overview 1999/2000. London: National Audit Office. Cancer in Scotland. Action for Change. Annual Report. Edinburgh: Scottish Executive, 2002. Cancer in Scotland. Action for Change. Annual Report. Edinburgh: Scottish Executive, 2003. Cancer in Scotland. Action for Change. Annual Report. Edinburgh: Scottish Executive, 2004. Department of Health. The challenge of implementing sustainable improvement in cancer services Cancer Services Collaborative Improvement Partnership. London: National Audit Office, May 2006. Department of Health. Applying high impact changes to cancer care Cancer Services Collaborative Improvement Partnership. London: National Audit Office, February 2005. Department of Health. The How To Guide: achieving cancer waiting times Cancer Services Collaborative Improvement Partnership. London: National Audit Office, February 2005.
●76
77
●78
●79
●80
◆81
82
Cancer Service Improvement Programme. Edinburgh: Scottish Executive, November 2004. Leese B, Din I, Darr A, et al. ‘Early Days Yet’ the Primary Care Cancer Lead Clinician (PCCL) Initiative. Final Report. Leeds: Centre for Research in Primary Care, University of Leeds, June 2004. Livingston M, Woods K. Evaluation of South East Scotland Cancer Network. Feedback paper 3. Glasgow: University of Glasgow, 2004. Delaney G, Jacob S, Featherstone C, Barton M. The role of radiotherapy in cancer management: estimating optimal utilisation from a review of evidence-based clinical guidelines. Cancer 2005; 104:1129–37. Cancer in Scotland. Radiotherapy activity planning for Scotland 2011–2015, Edinburgh: Scottish Executive, 2005. Bosanquet N, Sikora K. The economics of cancer care in the UK. Lancet Oncol 2004; 5:568–74. Board of the Faculty of Clinical Oncology. The Royal College of Radiologists. Guidance on the development and management of devolved radiotherapy services. London: Royal College of Radiologists, 2004.
60 New drug development GEORGE BLACKLEDGE
Introduction What does a clinician expect from drug development? The clinical investigator’s view of drug discovery and development
1407 1407 1407
INTRODUCTION Since surgery and radiation therapy can only treat and cure localized malignancy, there remains a need for systemic treatments for cancer. Existing systemic treatments include small molecules, such as cytotoxic chemotherapy, hormonal treatments and drugs targeting other pathways, large molecules including antibodies, cytokines, proteins and other potential approaches, which could include vaccines, gene therapy, etc. All these approaches can be regarded as drug therapy, and the process of making them available for cancer patients and their carers is termed drug development. Since the aim of any scientist involved in drug discovery and development should ultimately be to change clinical practice, drug development is a critical part of improving the treatment of cancer.
The importance of a multidisciplinary approach Specific issues for different disciplines Summary References
1408 1408 1413 1413
experiments. The next assumption is that the safety profile of the drug has been studied in a sufficient number of subjects that potential side effects can be predicted and managed appropriately. If the drug is to be used in combination then any possible interactions with other agents should have been studied and reported. A principal assumption is that the activity and efficacy of the drug has been evaluated in clinical trials, which can be extrapolated to the patient about to be treated.1 Although many of the assumptions above will be taken for granted most of the time, the work and research that allows this must have been carried out, reviewed by appropriate authorities, and decisions taken about whether the drug should be made available for use in the particular setting of the clinician and patient. That work and research is drug development, and involves many different disciplines that must work together to deliver a medicine and the information about that medicine in a timely manner, so that a clinician can safely treat the patient.2
WHAT DOES A CLINICIAN EXPECT FROM DRUG DEVELOPMENT? When a clinician decides with a patient that a particular drug should be used to treat the patient’s condition, he or she makes a number of assumptions. The first is that the tablet or injection contains what it says on the bottle – i.e., that the drug has been manufactured in a consistent and reliable way, and that the shelf life and storage conditions have been tested and confirmed. The second is that the way that the drug is handled in the body has been studied, and that the recommendations for dosing, schedule and route of administration have basis in proven and published scientific
THE CLINICAL INVESTIGATOR’S VIEW OF DRUG DISCOVERY AND DEVELOPMENT From the clinical investigator’s point of view, drug discovery and development appear to be a linear series of events (Fig. 60.1). Pre-clinical scientists identify a potential target gene whose product is involved in the malignant process, either as a direct stimulus (e.g. testosterone in prostate cancer), or over-expressed (e.g. Her-2-neu in breast cancer), or mutated (e.g. epidermal growth factor receptor [EGFR] in non-small-cell lung cancer). This target is then used to
1408 New drug development
Gene
Protein Screen Lead
Target Selection
Candidate
Phase I Phase II Phase III
Clinical Drug Candidate Selection
File
Launch
Proof of Concept
Figure 60.1 The sequential process of drug discovery and development from a clinical point of view Pre-clinical research E
Synthesis and purification
Clinical studies
NDA Review
Phase 1
E
Phase 2 Phase 3
Accelerated development/review Animal testing Short-term
E
workers and suppliers so that things get done at the right time, so does the process of drug development. Most groups involved in drug development will have a team structure with representatives from each of the disciplines represented on the team. One person will be chosen to be the leader of the team, and there will also be a project manager. That person should be familiar with planning multidisciplinary groups’ activities and coordinating the groups to ensure delivery of their commitments.4 For a small project this could be a part time role, but when a new drug is in the late stages of development there are multiple activities ongoing, the project manager and probably the project leader will have a full time role fulfilling the needs of the extended team and coordinating the project to ensure delivery. In the early stages of drug development, this level of infrastructure for a project may seem excessive and feel over-controlling, but assuming success of the project, investment in proper planning and leadership will prove invaluable.
Treatment IND Parallel track
SPECIFIC ISSUES FOR DIFFERENT DISCIPLINES
Long-term Institutional review boards Industry time FDA time IND Submitted Sponsor/FDA meetings encouraged Advisory committees
Bulk drug and manufacturing NDA Submitted Early Access: E Subpart E
Review decision Sponsor answers any questions from review
Figure 60.2 Drug development is series of complex, parallel and sequential, inter-related activities, involving internal teamwork and external authorities
screen potential agents active against it, resulting in the discovery of a molecule that could be a potential drug. Clinical studies will then be carried out resulting in a set of results that convince regulatory agencies and potential prescribers that a useful new agent has been developed. In reality the process is far more complicated. Many activities are carried out in parallel to enable the necessary clinical and scientific data to be generated (Fig. 60.2). Drug development is therefore multidisciplinary.3
THE IMPORTANCE OF A MULTIDISCIPLINARY APPROACH Well-conducted new drug development will involve coordinated, planned work from a number of different disciplines. Typically these include basic discovery research, bulk drug manufacturing, formulation and testing of finished product, toxicology, drug metabolism and pharmacokinetics, clinical development, regulatory affairs, intellectual property, manufacturing and supply, and project management. Project management is a key skill within these groups, since the activities of any one discipline will be dependent on the delivery of work from others. In just the same way as the building of a house requires the coordination of different
One critical issue in drug development is to ensure that there are quantities of drug available during the different development phases adequate to support the activities of other disciplines. In early phases this will be measured in kilograms; there needs to be enough to support the animal toxicology studies, to allow the formulation group to research and develop an optimal formulation and to perform the first clinical studies. As the clinical studies increase in size, then further campaigns of manufacturing will be required, producing tens or hundreds of kilograms. Finally, there must be manufacturing facilities in place to provide for consistent supplies of the drug in the marketing environment. Often two manufacturing facilities are built to ensure continuity of supply. The final route of manufacture needs to be defined before the phase III studies are carried out. Regulatory authorities insist that the pivotal studies on which approval will be based have a route of manufacture and formulation of finished drug identical to that which will be marketed. As a result of this, decisions about the chemical processes and the starting materials involved in making the drug need to be made early in the development, so that a reliable and tested route is established. This can be a major issue with biopharmaceuticals. The cell lines will need to be laid down right at the beginning of development, and there could be a need for major investment in complex manufacturing equipment before the drug has proven its activity.
Formulation and assessment of new agents Having quantities of the compound available is of little value until it has been formulated in a preparation acceptable
Specific issues for different disciplines 1409
for human administration. Formulation can present many challenges, since some molecules prove to be stable only in extreme conditions of pH, humidity, etc. The half-life of a drug can be altered by formulation so that slow release preparations can give large benefits if sustained exposure to the drug is required. In oncology, parenteral formulations are widely accepted, and many cytotoxic agents are only available in parenteral form. This has the advantage that there will be good compliance, but administration can be complex and timeconsuming. The stability and sterility of agents that are reconstituted for administration are also potential areas for concern, and such agents must be administered by specially trained staff. The team that develops new formulations will normally also develop assays to assess stability, shelf life and bioequivalence. These are all factors that need to be known about the drug at a single dose level but also at a batch level during manufacture.
Toxicology and exposure in animals Ensuring that animal toxicology studies are carried out to good laboratory practice (GLP) and can support the desired human doses is an important part of pre-clinical work, and is a pre-requisite for obtaining an investigational new drug (IND) or clinical trial application approval. Regulations exist defining the exposure and duration of toxicology studies prior to human administration in all territories.3 Usually, for short term phase I studies in humans, 1-month toxicology studies in a rodent species and a larger mammal (usually dog) will be required. Toxicology studies investigating specific observations made pre-clinically will also be required. If cardiac abnormalities have been seen (e.g., QTc prolongation), then specific electrocardiographic monitoring will be required both in the animal studies and subsequently in early human studies to estimate the likelihood of this being a significant safety issue.5 Other organ-specific or biochemistry-specific (e.g., cytochrome p450) studies will also be needed.6 It is important that toxicology studies are carried out in the right species. This is particularly relevant for biopharmaceuticals such as antibodies. Since mice or dogs will develop anti-human antibodies to biopharmaceuticals that are either human or humanized, studies in primate species will be required to determine toxicities unrelated to immune response. It remains a fact that animal and in-vitro toxicology studies can never completely predict toxicity in man, and, therefore, the initial early human studies need to be carefully designed and monitored for unexpected side effects.3
Generating clinical data
effects are established in maximum tolerated dose (MTD) phase I studies. Activity in a specific tumour type is sought in phase II studies, and phase III studies confirm the comparative efficacy of a new agent. This is a good starting point, but the situation is evolving.
Guidelines for anti-cancer drug development Both the FDA (Food and Drug Administration) and EMEA (European Medicines Agency) provide guidelines for clinical drug development, and also specific guidelines for anticancer drugs.7 Until the end of the twentieth century, these were focused on cytotoxic agents – i.e., agents that would be used at a dose close to MTD, and where the most common toxicities were usually myelosuppression and gastrointestinal (GI) toxicity, and were often genotoxic. More recently, it has become clear that many newer agents being developed are chronically administered, given at doses well below MTD and have a different pattern of toxicity. Many of these agents are not genotoxic, and can therefore potentially be evaluated in early studies in healthy volunteers. This has resulted in a re-evaluation of oncology guidelines, particularly for the early phases of development.8 In common with non-oncology drugs, there is greater emphasis on healthy volunteers for establishing pharmacokinetic parameters, and also the use of biomarkers to establish pharmacodynamic endpoints of activity.
Potential changes in early anti-cancer drug development With better understanding of biological targets, and molecules being designed specifically and selectively to inhibit those targets, early clinical studies are changing. If a target is only expressed in a specific cancer (e.g., BCR-ABL mutation in chronic myeloid leukaemia), then early studies might only include patients suffering from that particular disease. As well as assessing pharmacokinetics and side effects, pharmacodynamic measurements can also be made.9 Biomarker inclusion in early studies has been described as translational science.
Translational science ADVANTAGES
The increase in scientific knowledge about the biological mechanisms involved in cancer could now enable the developer to change the scale and pace of development. Some areas where the integration of pre-clinical observations and new markers of human disease can be seen below: ●
It is taught that there is a standard way of carrying out oncology clinical development. Dose and common side
The disease or diseases to be studied can be pre-selected on the basis of the presence, over-expression or mutation of a gene and the gene product.
1410 New drug development
●
●
●
●
●
Selection of patients whose tumours have the gene defect could ‘enrich’ activity; i.e., patient with no chance of responding to the new treatment are not entered into clinical studies. This has been demonstrated most elegantly with the use of trastuzumab in breast cancer, where the monoclonal antibody could have been dropped from development if a selected population had not been studied whose breast cancers strongly over-expressed Her-2-neu.10 Serious side effects might be avoided if patients were pre-screened for metabolic defects.11 The speed of development could be increased since safety and activity can be studied in early trials, in addition to the conventional pharmacokinetics The chance of regulatory success might be increased if surrogate markers, predictive of relevant clinical outcome, are assessed early. Other indications, e.g., other tumour types that have a specific mutation, could be identified and thus studied earlier than would be the case if identification depended on broad phase II studies.
It can, therefore, be seen that the use of pre-clinical data, coupled with the incorporation of biomarkers for patient selection, tumour selection and activity could provide an acceleration of development and a greater chance of positive pivotal trials for regulatory approval. CHALLENGES REMAIN FOR TRANSLATIONAL SCIENCE
Whilst all the advantages described above apply to translational science, there are also some challenges integrating translational science into a development project, particularly in terms of integrating pre-clinical and clinical observations. There are some surrogate endpoints that do not necessarily predict for biological activity. An example of this is establishing a dose based on achieving a plasma concentration of a drug that in vitro achieved an IC90. There are many factors that could affect this correlation, including different concentrations of drug in the tumour compared with plasma, protein binding, and the kinetics and distribution of the drug in humans.9 A second example is the use of surrogate target tissue, such as skin or buccal membrane biopsies, to predict effects of a drug in tumour tissue. Whilst they may be able to show an effect at the mechanism level, this does not necessarily predict anti-cancer effects, because of other factors with the tumour.12 As well as scientific reasons why surrogate biomarkers of response may disappoint, there are a number of practical issues that make the use of biomarkers problematic. Relevant paired biopsies, whether from tumour or normal tissue, are difficult to obtain. Even in a study where paired biopsies are mandated in the protocol, investigators do not usually succeed in achieving a 50 per cent success rate. It is also critical that the biopsies are treated in the same way; many of the assays carried out are sensitive and depend on the correct handling of the biopsy, rapid freezing or fixing of
the tissue and consistent transportation to the laboratory where the assays are carried out. Surrogate marker assessments make the early protocols more complicated. The statistics become more involved because of the multiplicity of endpoints, resulting in studies being underpowered for some of the endpoints. Complex protocols also are more difficult to carry out and more difficult to recruit to. As a result, time penalties are added to the project. There are further potential time penalties. When a surrogate endpoint has been validated, then it can be used in early studies with confidence. An example of this is oestrogen suppression and aromatase inhibition. The principal, and probably only action of aromatase inhibitors is oestrogen suppression, and therefore if this is achieved there can be reasonable confidence that that dose is appropriate clinically.13 If, however, a new biomarker is being developed for a new target, it will almost certainly not be validated before the first studies begin. This creates the dilemma of waiting to validate the biomarker or proceeding with clinical studies not using, or not depending on the biomarker. Since time to successful launch is usually the most important factor in developing a competitive new agent, development will probably proceed without the biomarker information.9 The final concern is the attraction of science. Pre-clinical and clinical investigators can ‘take their eye off the ball’ because of a wish fully to understand the science behind a drug development. This can introduce unnecessary delays in the project, leading to a non-competitive product. The attraction of biomarker-directed development is great, but the pitfalls involved in biomarkers, either for surrogate efficacy or patient and tumour selection are also great. The role of translational science in a development should always be assessed very carefully and the risks and benefits clearly identified.
What is needed to enter development for regulatory approval and commercial launch of a new anti-cancer agent? Conventional phase I and II studies will have established pieces of key information about a new agent. These include a dose, or dose range, a schedule and route of administration, safety, usually in several hundred patients, and some evidence of efficacy in a patient population and tumour type or types that is felt likely will predict a successful outcome in pivotal trials. Translational science and clinical pharmacology may also have provided more detailed information about which patients can be safely given the drug, and which tumours at a genetic level are most likely to benefit.
Pivotal studies (phase III) for regulatory submission and approval In spite of all these data, pivotal studies remain a major risk. Data generated in the past decade show that two
Specific issues for different disciplines 1411
Chart 1: Valuation
Cost of Sales & Marketing
Phase III Phase II Phase I Cost Begin Clinical
Outcome success 50%
7
Cost
Outcome success 60%
37
Cost
Outcome success 70%
125
PV (Revenues) given three demand scenarios
632
($150)
hi 30%
820
($150)
lo 40%
failure 30%
($25)
med 30%
$1,400 $800 $400
failure 40%
($10) failure 50%
Cancer success 30%
Cancer success 20%
Exit t=0
1
3
8
T = 8.5 years
Figure 60.3 Flow chart showing industry average spend ($millions), patient numbers (in circles), success and value for a typical new drug development (cumulative chance of success 20%), and the average success rates for a cancer development (cumulative chance of success 6%)
thirds of all phase III studies in oncology failed.14 Not only that, it can be seen (Fig. 60.3) that the cost of failure increases as large trials are carried out and the risk of failure also increases. A better understanding of the science behind the development may influence this, but there is no evidence of this at the present time. In terms of an ultimate endpoint to demonstrate efficacy, increased survival in the patient population being studied remains the gold standard. This remains the situation in the EU, but the FDA has been re-evaluating its guidelines in view of the fact that many patients now receive a number of different treatments. Therefore, in some defined situations time to progression or time to relapse may become acceptable endpoints for regulatory approval, because subsequent treatment will dilute or confound any survival benefit. The acceptability of different efficacy endpoints has been demonstrated in a review of reasons for approval of new agents by the FDA.15 This shows that although a survival benefit represents the highest level of evidence of benefit, other endpoints are acceptable in different situations, particularly when the natural history of the cancer being evaluated is well understood. There is also the concept of non-inferiority, which is a method of defining what colloquially would be termed equivalence with an existing treatment.16 Approval of new drugs has been on the basis of non-inferiority of survival where there were demonstrable safety benefits.
Traditionally, in oncology, side effects have been observed and graded. The UICC and CTC systems were established to do this. With cytotoxic chemotherapy, side effects are extremely common and grading made sense. Side effects were going to occur; what was important was their severity. This system is not particularly good at detecting infrequent and unexpected side effects. In order to do this, adverse events must be collected. Adverse events are any events that have happened to the patient, irrespective of cause. Certainly in the stages of clinical drug development up to the time of first marketing approval, potential side effects should be collected in this way. It is often the case that when adverse events are collected in this unbiased manner, patterns emerge and unexpected side effects are identified. In early clinical studies, only common side effects will be identified (frequency greater than 1 in 10). Side effects that occur at a 2–9 per cent frequency will require many hundreds of patients to have been studied. (The average number of patients in an oncology new drug application [NDA] is around 1000–3000). For unexpected side effects occurring with a 1 per cent incidence many thousands or tens of thousands of patients will need to be studied. An example of this is the association between tamoxifen and endometrial cancer. It was not until tens of thousands of patients had been treated over about 10 years that the association was established.17
Reporting side effects and pharmacovigilance
Regulatory issues
It is usually only necessary to prove the efficacy of a new drug once, usually at the time of first market approval. Assessing and measuring the safety and side effects of a new drug is a continuous process. This is partly because uncommon side effects will not be detected until many thousands of patients have been treated, and also that in order to detect unexpected side effects information must be collected in a specific way.
A critical factor to be considered in drug development is the role that regulatory authorities play in all countries. Following the thalidomide tragedy and other drug scandals, the pharmaceutical industry has become one of the most intensively regulated industries. There are guidelines for regulation in all major countries, and these differ in detail, although all address issues of efficacy, safety and quality of the information provided for a particular medicine. Because
1412 New drug development
these regulations frequently change, the best sources for these guidelines and advice are the websites of these authorities. Examples include in the USA www.fda.gov/cder/index.html, in the European Union www.emea.eu.int/ and in the United Kingdom www.mhra.gov.uk/. In the USA, and now in Europe, regulation extends to academic research, and, for example, the recent EU Clinical Trial Directive18 does not differentiate between industry and academic sponsored trials, demanding the same levels of conduct and patient protection. This has implications for some of the early drug development that has been carried out in an academic setting, and which now requires increased resource and a different level of commitment with respect to safety reporting, patient protection and legal responsibilities. Regulatory authorities take slightly different approaches from country to country. The best defined approach is probably that of the FDA in the USA, which has clearly defined, transparent processes laid down by law. The International Committee for Harmonisation (ICH) has been in existence for many years, set up by the USA, EU and Japan, whose purpose is to define common standards for the development of new medicines, allowing the use of data generated in one country to be acceptable elsewhere (www.ich.org/cache/compo/276-254-1.html). No potential medicine can be administered to humans until regulatory approval has been granted in that country. Generally regulators require all the available data about a given molecule to be available to them. This is then summarized in an Investigator Brochure (IB), which is made available to all clinicians taking part in clinical investigations of that molecule. In the USA this is termed opening an IND (Investigational New Drug application). Other countries have different nomenclatures, but the principle is the same. As further information becomes available from whatever source, this is then filed into the IND, allowing a living data bank about that particular molecule. When a sponsor believes that sufficient information has been generated about a potential new medicine, they combine all the information and present an NDA to regulatory authorities. This application must be robust enough to convince the assessors appointed by each regulatory authority that the applicant can make the drug reliably and to a defined specification on an ongoing basis; that there is sufficient data to justify the dose and schedule of administration; and that an acceptable balance of benefit and risk (i.e., efficacy and safety) has been established in the indication for which the drug will be prescribed. After consideration by the assessors, who may use external experts as part of that assessment, a decision will be made whether or not the drug for a specific indication can be marketed in that territory. A decision on the approvability of a submission will normally be primarily based on the outcome of the pivotal (phase III) studies. These will normally be survival studies. Both FDA and EMEA have guidelines which state that in some situations, where there is an unmet medical need and no other available treatment, conditional approval may be
given on the basis of an intermediate endpoint, potentially in phase II studies. Further definitive studies must subsequently be carried out to confirm the intermediate endpoint.
Post-marketing approval Regulatory approval is only the first step to successful marketing of a new drug. There needs to be a price set, which is a formal process in some countries, and health economic studies need to have been carried out to convince reimbursement authorities that the cost of the drug is justified by the benefits it gives. Increasingly, this is of importance in countries concerned about the rising costs of health care In the UK, for example, the National Institute for Health and Clinical Excellence (NICE) (www.nice.org.uk) issues guidelines for the treatment of specific diseases and for the use of individual medicines. As well as justifying the price of the drug, there will be a need to convince physicians to prescribe the new agent. This can require data that is different from the clinical trials that were carried out for regulatory approval. Clinical practice varies from country to country and in regions within a country. Further clinical trials relevant to clinical practice in a particular territory may therefore be necessary. Other development work is like to be necessary. As experience grows with a new agent it is likely to be tested earlier in a disease process. A good example of this is in breast cancer, where most new agents over the past 40 years have an initial market approval in advanced, metastatic disease. Once sufficient safety data has been generated, successful agents have then been evaluated as adjuvant therapy, requiring very large studies. Now, following the data generated with tamoxifen, new agents are also being evaluated as prevention, or risk reduction treatment.19 With clinical experience, there may be a need for evaluation of a different dose or a different formulation. These would represent supplementary NDAs, and would require the necessary data to support them. It can be seen that the process of new drug development does not end with the first marketing approval, and is an ongoing process for as long as the sponsor wishes to support the drug on the market (normally until patent expiry). Even after that the original sponsor of the agent will have pharmacovigilance responsibility.
Building regulatory assets Filing information into an IND can be regarded as building regulatory assets. An asset can be defined as generating information that will contribute to a successful NDA. If a regulatory authority’s rules specify that certain toxicology studies are required, then the delivery of those toxicology studies done to GLP and with drug that is manufactured to the final marketing specification can be regarded as regulatory assets. Many assets that will be required can be carried
References 1413
out in parallel with other activities, e.g., clinical pharmacology studies of impaired renal or hepatic function in patients. Nevertheless, there should be an early view of what the final NDA will contain, and a plan of how to provide the information in a timely way. Equally, there may be information generated that is not a regulatory asset. Academic research is not always of value in a regulatory submission, even thought it may be scientifically valid and interesting. A sponsor is probably going to want to control non-relevant research, since any results generated will have to be included in a NDA, and if they are counter to the overall information in the NDA this will require explanation.
Controlling development The cost of new drug development is considerable. The risks associated with development are also high. It has been estimated that a development costs a minimum of £300 million, and that only 1 in 10 of molecules that enter development will get to market. Much of the cost is incurred at the later stages of development, when large clinical trials are required, and manufacturing capacity is being built. Over the past 5 years, two thirds of phase III oncology trials have failed in the sense that they did not provide a positive enough answer for regulatory approval and hence marketing of the drug, by which time around £250 million will have been spent (Fig. 60.3). No organization, irrespective of its resources, can afford regular failures at that level. Most organizations, therefore, put in place a system of control to ensure that projects likely to fail are stopped as soon as possible. This means that resources are then available for other projects more likely to succeed. The forms of control or governance vary with the organization in question. Most will have specific points during the development when the project is reviewed. These normally occur when new data is available. Examples of this are when the first toxicology data are available. This is an opportunity to determine whether the first human exposure can be in healthy volunteers or patients. There would usually be a review when the results of the phase I studies are available, especially if there is evidence of pharmacodynamic effect, or other evidence of activity. A critical review point for an organization is when a decision is needed whether or not to begin phase III studies. At this point, the cost of further development increases approximately 10-fold, and, if the organization is a commercial business, the whole of the company needs to prepare its staff for a launch and also prepare the market. This is a highly significant go–no go point, which should be made at the highest level of any organization. A final decision will need to be made when the results of the phase III studies are known, whether to file for regulatory approval, and then, based on the label approved by regulatory authorities, whether to launch the drug. Positive or negative decisions at each decision point will not be based on scientific data alone. The external
environment, competition, relative timing of results opposite competitors and patent expiry, financial considerations and manufacturing costs and the shape of an organization’s portfolio will all influence whether a positive or negative decision is taken at each point. Small organizations, either academic units or start-up companies, deliberately plan only to take the development to a certain point, with the intention of licensing the potential drug to an organization that has the resources to complete the development. The more work that has been done on a molecule, the greater its value. Some small organizations, however, deliberately carry out the absolute minimum needed to achieve what is usually a clinical result. This can result in a loss on time, since whoever licenses the molecule will need to go back and carry out further toxicology studies, scale up of manufacturing and further formulation work before further clinical work can be begun.
SUMMARY New drug development is a complex multidisciplinary process. It is also highly regulated with the primary purpose of protecting patients and providing convincing evidence of efficacy and safety. Increasingly health-care providers are looking for incremental value with new agents, and this has created a further hurdle to be surmounted. Nevertheless, there remains a need for new anti-cancer agents and oncology new drug development will continue to be an area for clinical research and scientific progress.
KEY LEARNING POINTS ●
●
●
●
New drug development is a complex multidisciplinary process. There is intensive regulation, but extensive guidelines to follow. Translational science has the opportunity to improve early development but there are also challenges associated with it. In oncology, two thirds of phase III studies fail making drug development very high risk.
REFERENCES 1 Schilsky RL. Principles of Antineoplastic Drug Development and Pharmacology. New York: Marcel Dekker, 1996. 2 Welling PG. The Drug Development Process. New York: Marcel Dekker, 1996. 3 Teicher BA, Andrews PA. Anticancer Drug Development Guide. New York: Humana Press, 2004. 4 Jack R. Meredith, Samuel J. Mantel. Project Management: A Managerial Approach 5th ed., Rockville, Maryland: Wiley, 2002.
1414 New drug development
5 Algra A, Tijssen JG, Roelendt JR, et al. QTc prolongation measured by standard 12-lead electrocardiography is an independent risk factor for sudden death due to cardiac arrest. Circulation 1991; 83:1888–94. 6 Kvisto KT, Koemer HK, Eichelbaum M. The role of human cytochrome P450 enzymes in the metabolism of anticancer agents: implications for drug interactions. Br J Clin Pharmacol 1995; 40:523–30. 7 Center for Drug Evaluation and Research. Guidance for Industry U.S. Department of Health and Human Services Food and Drug Administration, 1997. 8 CPMP/EWP/205/95 Guideline on the Evaluation of Anticancer Medicinal Products in Man (Adopted by CHMP December 2005) European Medicines Agency, London. 9 Figg WD, McLeod HL. Handbook of Anticancer Pharmaco kinetics and Pharmacodynamics. New York: Humana Press, 2004. 10 Pegram MD, Koneeny G, Slamon DJ. The molecular and cellular biology of the HER2/neu gene amplification/ overexpression and the clinical development of herceptin (trastuzumab) therapy for breast cancer. Cancer Treatment Research 2000; 103:57–75. 11 Johnson MR, Yan J, Shao L, et al. Semi-automated radioassay for determination of dihydropyrimidine dehydrogenase (DPD) activity. Screening cancer patients for DPD deficiency, a
12
13
14 15
16
17
18 19
condition associated with 5-fluorouracil toxicity. J Chromatogr Biol Biomed Sci Appl 1997; 696:183–91. Baselga J. Skin as a surrogate tissue for pharmacodynamic end point: is it deep enough? Clin Cancer Res 2003; 9:2389–90. Lonning PE, Geisler J, Bhatnager A. Development of aromatase inhibitors and their pharmacologic profile. Am J Clin Oncol 2003; 26 Suppl 1:S3–8 Market Resarch, CMR International, Surrey, 2002. Johnson JR, Williams, G, Pazdur R. End Points and United States Food and Drug Administration Approval of Oncology Drugs. J Clin Oncol 2003; 21:1404–11. Rothmann M, Ning L, Gang C, et al. Design and analysis of non-inferiority mortality trials in oncology. Stat Med 2002; 22:239–64. Van Leeuwen FE, Benraadt J, Coebergh JW, et al. Risk of endometrial cancer after tamoxifen treatment of breast cancer. Lancet 1994; 343:448–52. Directive 2001/20/EC of the European Parliament and of the Council. Official J Eur Commun 2001; 121:34–44. Winer EP, Hudis C, Burstein HJ, et al. American society of Clinical Oncology Technology Assessment on the use of aromatase inhibitors as adjuvant therapy for women with hormone receptor-positive breast cancer: status report 2002. J Clin Oncol 2002; 20:3317–27.
61 Late effects of cancer therapy SUSAN E. DAVIDSON
Late effects of treatment: general considerations Measuring and recording the late effects of treatment Management: general principles
1415 1415 1418
LATE EFFECTS OF TREATMENT: GENERAL CONSIDERATIONS The improvements in cancer curability, long-term survival and organ preservation after radiation treatment, as well as improvements in combined modality therapy have resulted in many patients living with the late adverse effects of treatment.1 The number of people diagnosed with cancer each year in the US is over 1 million and is expected to double by 2050, reflecting an aging population and increase in population size.2 Long-term follow-up of these individuals reveals progressive fibrosis and functional impairment of organs along with associated co-morbidities such as reduced tissue compliance, accelerated senescence of organs, the decline of endocrine function, cardiac dysfunction, joint stiffness, myelopathy, cognitive impairment and others.3 Evidence suggests that long-term survival after treatment for cancer in childhood is associated with increased risk of impaired quality of life with adverse consequences for many aspects of everyday life including personal care, education, employment and family and social relationships.4 Late adverse effects after cancer treatment during childhood cause substantial morbidity. Even with modern techniques of radiotherapy, normal tissues will be irradiated within the planning target volume, and the normal tissue volume that receives the prescribed dose may exceed the gross tumour volume. It is not possible to cure a tumour by radiotherapy without the risk of normal tissue complications. When assessing tumour outcomes, normal tissue reactions have to be assessed and documented when evaluating treatment outcomes.
Clinical outcomes research References
1419 1420
By convention, late effects of treatment are defined as those reactions seen more than 90 days after the start of radiotherapy,5 though this is not absolute as the time for early radiotherapy reactions to heal may require longer, particularly if severe with combined chemoradiotherapy or after accelerated radiotherapy.6 In the new paradigm of normal tissue injury, a central theme is the interaction between various cell populations mediated by cytokines and growth factors,7 in contrast to the target mechanisms, which focused on the loss of a specific cell type.8 The mechanisms and cellular interactions of normal tissue injury are complex and have been studied over the last decade and more recently may be considered as potentially amenable to therapeutic interventions. Recent progress in molecular pathology and normal tissue radiobiology has improved the mechanistic understanding of late normal tissue effects and shifted the focus from initial-damage induction to damage recognition and tissue remodeling.9 Preliminary evidence suggests that pharmacological or other interventions may be possible to reverse the manifestation of the injury and restore function to tissues.10
MEASURING AND RECORDING THE LATE EFFECTS OF TREATMENT Evolution of scoring systems Several attempts have been made to create a comprehensive system for the grading and reporting of normal tissue complications of treatment, though there is no universally accepted system. The World Health Organization system
1416 Late effects of cancer therapy
published in 1979 covered chemotherapy effects with 28 criteria.11 The National Cancer Institute’s Common Toxicity Criteria (NCI CTC) attempted to score chemotherapy effects with 18 criteria, but this was markedly expanded in 1998 to 260 criteria to score chemotherapy acute effects. This did not score late effects of treatment and the NCI relied on the Radiation Therapy Oncology Group European Organisation for Research and Treatment of Cancer (RTOG/EORTC) Late Morbidity System created in 1984 that was appended to the CTC.12 The CTC version 2 was widely adopted by co-operative and industry groups but this highlighted the need to improve the reporting of treatment effects, particularly surgical, late and paediatric treatment effects. In 1995 working groups of the RTOG and EORTC developed the Late Effects of Normal Tissue scales LENT SOMA (Subjective, Objective, Measurement and Analytic) in an attempt to provide a comprehensive system for assessment and recording of radiotherapy related morbidity.13,14 This system introduced the patients’ views on ranking or grading their effects from treatment (subjective scoring), which is not included in any other scoring system. By addressing both subjective and objective morbidity LENT SOMA provides a more comprehensive measure of late treatment effects and enables comparison of doctors’ and patients’ views of ‘tolerable damage’.13 Quality-of-life measures have been used as a surrogate endpoint for treatment toxicity in a few studies, and this is not justifiable. Quality-of-life measurements are indeed important but they are not a measure of treatment toxicity; instead they are a measure of the way patients are living with the consequences of treatment.15 Comparison of the LENT SOMA to the RTOG/EORTC and CTC systems has shown inconsistent concordance and correlations, indicating the need for a single system. In 2002 an effort was made to revise the LENT SOMA late criteria for inclusion into the CTC at the Late Effect Workshop (LENT IV).16 Late effects and acute effects were merged into a single system the Common Terminology Criteria Adverse Effects v3.0 (now to be known as CTCAE) and is available at the NCI Cancer Therapy Evaluation Program web site http://ctep.info.nih.gov/CTC3/ctc.htm. It is organized into 28 organ systems (e.g., gastrointestinal, genitourinary) that may be used for injury regardless of modality. This version of the CTC contains many new items that are applicable to surgery, though only the unintended severe and lifethreatening (grades 3–4) consequences of surgery should be the focus of surgical reporting. Adverse events are no longer pre-designated as ‘acute’ or ‘late’ and so the ‘90 day rule’ has been dropped.17 However, from a biological and clinical view point, it is useful to have a division of normal tissue effects into early and late effects with early effects during or immediately after therapy and late effects, which may appear after months to years.18 The scoring system should be reproducible (inter- and intra-observer variability should be low compared with variation between patients) and sensitive (should be able to detect changes in treatment toxicity from increasing the
dose for example). A scoring system should undergo study of its feasibility, reliability, validity, responsiveness and specificity before being introduced in the clinic, but this has largely been neglected for the systems developed so far. Each aspect of validity has components that should be addressed, and some of these have been developed by social scientists and psychologists.18
How and when should data be collected? The CTCAE does not address patient screening or methods of data collection. There is no standard approach for screening and collecting patient information from the clinic. It may range from abstracting the clinical record after the consultation (‘passive’ data collection) to a detailed assessment using pre-defined templates to structure the interview (‘active’ data collection).5,17 The questionnaire approach appears a feasible, reliable and consistent method to collect morbidity data with good compliance. It is a method that has been used in quality-of-life research and is well received by patients with good reliability and completeness of data collection.20, 21 It is also a practical way to obtain detailed normal tissue outcome data in a busy clinic.19 Electronic methods to facilitate adverse event data collection may promote more uniform data collection and reporting practices.17 The relationship between observer- and patient-based toxicity reporting has received little attention in oncology. Jensen and colleagues described a study that explored the relationship between observer-based toxicity scoring and patient-assessed symptom severity after treatment for head and neck cancer.22 The Danish head and neck (DAHANCA) study group scoring technique was effective in assessing objective treatment induced toxicity in head and neck patients but insensitive and non-specific with regard to patient-assessed subjective endpoints (EORTC qualityof-life questionnaire).22 The DAHANCA study observed objective symptoms such as oedema, atrophy and fibrosis correlated poorly with quality-of-life endpoints. This finding is likely to apply to other observer-based scoring systems. The Groupe d’Oncologie Radiothérapie Tête et Cou (GORTEC) reported long-term survival and late effects from a pivotal randomized trial of chemoradiotherapy in advanced head and neck cancer.23 The trial data supports the superiority of concurrent chemoradiotherapy over radiotherapy alone in head and neck cancer, but combined modality approaches are associated with higher toxicity. This group reported a 30 per cent high-grade late effects24 at 5 years in the radiation-alone arm compared with 56 per cent in the chemoradiotherapy arm using the RTOG/ EORTC5 grading system for late effects and the NCI CTC for selected late effects not covered by the RTOG/EORTC system. The GORTEC group reported a more detailed evaluation of late effects in forty-four 5-year survivors in this trial using a combination of RTOG/EORTC, the NCI CTC system and the LENT SOMA system.25 The rate of late
Measuring and recording the late effects of treatment 1417
effects were 47 per cent in the control arm and 82 per cent in the chemoradiotherapy arm when the data from the three systems was combined. This was significantly different from the rate observed when using the RTOG scale alone. The initial paper in 1999 of the GORTEC study reported 9 percent and 14 per cent grade 3–4 late toxicity in the radiotherapy and chemoradiotherapy arms, respectively.26 This study illustrates that the choice of a grading instrument and reporting methodology can dramatically alter the observed rates of normal tissue effects. The importance of establishing reporting standards is illustrated by the large range of reported values from 9 to 47 per cent in the radiotherapy arm and from 14 to 82 per cent in the chemoradiotherapy arm in these three reports. We lack sufficient methods to effectively communicate the full extent of the potential and expected adverse effects of cancer treatment, and we need to develop more meaningful and reliable reporting methods.27 Late effects of treatment (radiation and chemotherapy) can occur after many years and, therefore, patients need to be followed for long periods of time. Five-year follow-up is widely used to report outcomes of cancer therapy, but this is not sufficient for follow-up for late occurring sequelae such as coronary artery damage and second tumours induced by treatment as well as, for example, bladder and bowel damage within the pelvis. Patients treated at the MD Anderson hospital for stage IB cervical cancer had a serious morbidity incidence of 9.3 per cent at 5 years and this rose by 0.34 per cent per year to give an actuarial risk of 14.4 per cent of major complications at 20 years from a hospital records review of 1784 patients.28 It is also well recognized that there may be a progression of severity of treatment effects with time.29 When reporting the treatment complications it is important to choose an appropriate time point after treatment to obtain reliable incidence figures. For example, prostate brachytherapy patients need to be followed for at least 3 years, as a rectal fistula can develop after several years as a late complication of this procedure.30 The study of radiation injury and long-term consequences of cancer treatment is costly. It can be relatively inexpensive to study overall survival as an endpoint in a clinical trial but it is expensive to study toxicity, particularly late-toxicity.17 An important aim of follow-up after treatment for cancer is to detect various events associated with disease recurrence, metastatic spread or severe treatment-related complications as early as possible. Each tumour type may show a specific pattern and timing of these events related to different prognostic factors. A working group of the European Society for Therapeutic Radiology and Oncology was funded by the EU for a project on Recording, providing Education, and Ameliorating the Consequences of Treatment (REACT). It carried out a study to propose a way of defining an optimal timing schedule for follow-up after treatment based on the analysis of failure patterns determined from follow-up data from prospective clinical trials.31 The study suggested timings of follow-up visits but did not address who should do
this or what investigations should be done. Another aim of long-term follow-up is to facilitate timely diagnosis and appropriate management of late adverse effects, thereby reducing the frequency of severe complications.4 We need to be able to measure toxicity of treatment better than we do at present, not just within research, but also in clinical practice during follow-up visits.32,33
Examples of need for prolonged follow-up of cancer survivors CARDIAC MORBIDITY
The leading cause of non-cancer mortality amongst survivors of Hodgkin’s disease is cardiovascular disease.34 A study of 474 childhood cancer survivors treated at a single institution in the US found that after mortality from the original cancer, cardiac disease and second malignancy were the next most common causes of death.35 Mortality from cardiovascular disease is also increased in survivors of left-sided breast cancer who received adjuvant radiotherapy that included the internal mammary nodes in the treatment field.36 While earlier studies reported on cardiac mortality,37 later ones reported on significant cardiac morbidity in cancer survivors with various incidences between 0 and 57 per cent.38 Many series reported results from single institutions and with older treatment techniques. Pihkala39 reported cardiac toxicity in children and adolescents treated with anthracyclines with or without the addition of mediastinal radiotherapy: 32 per cent of those receiving anthracyclines alone had abnormal left ventricular function and 50 per cent of those receiving combined modalities had abnormal function. A detailed study of 115 children treated for acute lymphoblastic leukaemia40 revealed that cardiac abnormalities were persistent and progressive following doxorubicin therapy. The deficits were worst after highest cumulative doses of doxorubicin, but appeared even after low doses. Late cardiomyopathy generally presents 5 or more years after therapy with anthracyclines and is characterized by a dose-dependent, symptomatic or asymptomatic, progressive decrease in left ventricular function often resulting in congestive cardiac failure.41 Risk factors for anthracyclineinduced cardiac toxicity included cumulative dose, hypertension, pre-existing cardiac disease, advancing age and prior mediastinal irradiation.41,42 Patients who have radiotherapy to the mediastinum have an increased risk of earlyonset coronary artery disease.43,44 Mortality from radiationinduced coronary disease usually occurs at least a decade after radiation treatment.45 Prevention is the best way to treat radiation-induced cardiotoxicity, and modern radiotherapy may help to reduce the frequency of cardiac disease along with establishing best practice to reduce toxicity, such as employing techniques that will minimize irradiation of the coronary arteries. Dexrazoxane is an agent that appears to decrease the cardiotoxicity of doxorubicin in patients with breast
1418 Late effects of cancer therapy
cancer and is being used in a trial in children treated for Hodgkin’s Disease in the US, though there is no long term data with this agent and the equivalence of tumour outcomes needs to be established. Evidence is needed to find out if patients with asymptomatic reduced left ventricular function benefit from treatment to reduce the cardiac afterload with beta-blockers and angiotensin converting enzyme (ACE) inhibitors.46,47 Radiotherapy can lead to a broad range of cardiac abnormalities if it is included in the treatment field. Survivors of cancer treated with chest radiotherapy should be followed up and screened periodically for heart disease. SECOND MALIGNANCIES
Information concerning radiation-induced malignancies comes from the Japanese A-bomb survivors and from medically exposed individuals, including second cancers in patients treated with radiotherapy. The A-bomb survivors show an excess incidence of carcinomas in the gastrointestinal tract, thyroid, breast and bladder, which increase linearly with dose. Radiotherapy patients show an excess incidence of carcinomas, often at sites remote from the radiation fields, and in addition there is an excess of sarcomas in the high-dose areas of the treatment volume.48 With the increase in intensity-modulated radiation therapy (IMRT), which involves more radiation fields, there is an increase in the volume of normal tissue that is exposed to lower doses of radiation. This factor may increase the risk of second cancers.49 There is increased risk of second malignancies in survivors of Hodgkin’s lymphoma treated in childhood. Adult malignant tumours occurred at a relatively young age after radiotherapy, with tumours of the breast, colon, lung and CNS, for example.50 In adults treated successfully for Hodgkin’s lymphoma, lung cancer is the most frequently occurring second tumour. High-risk patients can be identified based on age at treatment and smoking status, with the expected incidence of lung cancer in these patients between 50 and 150 per 1000 patients. Chemotherapy, especially with alkylating agents, increases this risk. In a Cochrane review of data from 37 trials in Hodgkin’s disease with 9312 patients, second malignancies were seen with chemotherapy alone, but less so than with combined modality treatment; acute leukaemia was seen more with chemotherapy, and solid tumours following radiotherapy and combined modality treatments.51 In a study of 6671 Swedish patients with soft tissue sarcoma there were 650 second malignancies, and soft tissue sarcomas were seen as second tumours in patients after primary cancers of the bone, ovary, nervous system, cervix, thyroid gland, skin, endometrium, breast and upper gastrointestinal tract, and after Hodgkin’s disease, non-Hodgkin’s lymphoma and leukaemia,52 though genetic pre-disposition may partly explain the excesses. Of the 3.5 per cent of the US population that has survived cancer, second malignancies in this group now account for 16 per cent of all US cancer incidence.53 Few data
exist currently regarding the molecular mechanisms for second primary cancers and other late outcomes after cancer treatment. CO-MORBIDITY
Co-morbidities are defined as diseases, illnesses or conditions affecting the patient but unrelated to their index cancer. Co-morbidity has direct impact on the care of patients, selection of initial treatment, and evaluation of treatment effectiveness with the severity of co-morbid conditions being significantly related to survival.54 Various instruments exist for classifying co-morbidity, but the Adult Co-morbidity Evaluation-27 (ACE-27), modified from the Kaplan–Feinstein Index55 developed for diabetic patients, is currently the only co-morbidity instrument designed specifically for cancer patients.56,57 The issue of co-morbidity in cancer becomes particularly significant in older patients and, if current trends continue, 70 per cent of all neoplasms will develop in persons in the age group of 65 years and over by 2020.58 There is some evidence in the literature for various co-existing conditions (co-morbidities) to influence normal tissue effects of treatment (increase the risk of late effects) though the reports are plagued by methodological weaknesses.59 A review carried out by Hölscher et al. on the evidence for the effects of connective tissue disorders (CTDs) on the expression of radiation side effects came to the conclusion that CTDs are associated with an increased risk of late radiation effects but a large coordinated study is required. There is concern that the long-term adverse effects of treatment may interact unfavourably with the normal ageing process.
MANAGEMENT: GENERAL PRINCIPLES An essential part of management of late effects is prevention. Techniques such as IMRT may minimize normal tissue in the high-dose volume that is irradiated, which may be the critical dose-limiting organ, for example, the small bowel in the pelvis.60 Patient positioning using a belly board and the prone position may avoid the inclusion of some of the small bowel within the treatment volume. Management strategies for symptoms following treatment require careful diagnosis of the problem and patients have multiple causation contributing to their symptoms.61 Following pelvic irradiation, the symptom of diarrhoea can have many causes, with bile salt malabsorption and small bowel bacterial overgrowth. It is important, therefore, to determine causation if the symptoms are to be managed successfully. Developing patient awareness of the potential for these problems and self-management strategies for unresolved late effects have been a focus of patient support groups and nursing interventions. In the management of chronic conditions, guided self-management has been shown to reduce morbidity and disability.62,63 Barriers to
Clinical outcomes research 1419
effective self-management have been shown to be a lack of awareness by patients and poor physician communication,64 which has a resonance to reports on the management of late effects.65 The lack of evidence is apparent when exploring the management of late effects, with the focus of most radiotherapy research on new technology and delivery rather than side-effects management.66 With no consensus on recording and analysing normal tissue effects in patients and with a general tendency to under-report toxicity, there is a reduction in the possibility of synthesizing information from different studies and clinical sources.32 A Cochrane review of non-surgical interventions for late radiation cystitis revealed no randomized controlled trials and concluded that it was not possible to draw any firm conclusion for management.67 Similarly a Cochrane review of non-surgical management of late radiation proctitis revealed small studies and lack of placebo-controlled studies making it difficult to establish whether particular treatments are effective.68 A review of hyperbaric oxygen therapy revealed six small trials in which the data was examined. There was no evidence of benefit to clinical outcomes with established radiation injury to neural tissue, but there appeared to be some benefit to radiation injuries of the head and neck, anus and rectum.69 Further research is required to establish optimum treatment. Paediatric oncologists (UK Children’s Cancer Study Group) have published the therapy-based long-term follow-up statement, to give clinical recommendations for long-term follow-up based on published work, clinical experience and expert opinion.49 Wallace and colleagues have suggested three possible levels of long-term follow-up care for survivors: postal or telephone follow-up (level 1); nurse-led or primary care follow-up (level 2); and attendance at a medically supervised long-term follow-up clinic (level 3).4,70 The level of follow-up would be determined by the treatment that survivors had received and the associated risk of long-term problems from that treatment. Guidance from the UK National Institute of Health and Clinical Excellence recommends that long-term follow-up should be given by a multidisciplinary team that includes appropriate doctors and a specialist nurse for improving outcomes in children and young people with cancer.71
Specialist clinics to which patients can be referred
follow-up of people who have survived cancer treatment as children in the UK and USA has been done by paediatric oncologists.72,73 Oncologists are well placed to coordinate follow-up and have the knowledge of the treatment carried out, for example, the irradiated volume, and understand the radiobiology and pathological processes that underlie the late effects and their evolution. One study involved a prospective audit of over two thousand patients and involved 194 physicians to examine the effectiveness of follow-up from both the doctors’ and patients’ perspectives.74 Forty per cent of patients had symptoms or medical problems relating to their disease. There was wide variation in follow-up practice among European centres with a low incidence of positive findings on investigations.
CLINICAL OUTCOMES RESEARCH There is a glaring need for clinical, epidemiological and outcomes studies in well-defined cohorts of cancer survivors to define the overall and organ-specific prevalence of late effects of radiation, as well as the medical, qualityof-life related, social and financial consequences of the radiation late effects.75 Studies like these are required to set priorities for prophylactic and interventional clinical trials as well as identifying areas for focussing basic and preclinical research. Clinical trials should be designed with outcome measures, and consistent and accurate reporting of treatment effects is required.
KEY LEARNING POINTS ●
●
●
●
●
WHO SHOULD DO FOLLOW-UP? EFFICACY OF FOLLOW-UP
The issue of who does long-term follow-up needs to be addressed from the patients’ perspectives and needs. A 5-year follow-up is widely used for reporting the outcome of cancer therapy, but it must be stressed that this period may not be sufficient to give reliable estimates on very late-occurring consequences of treatment. To date, most long-term
●
Late effects of treatment (radiotherapy and chemotherapy) can occur after many years, and patients should be followed up. Late side effects and symptoms may progress with time. The choice of a grading instrument and reporting methodology can dramatically alter the observed rates of normal tissue effects. Management strategies for symptoms following treatment require careful diagnosis of the problem – patients may have multiple causation contributing to their symptoms. There is a glaring need for clinical, epidemiological and outcomes studies in welldefined cohorts of cancer survivors to define prevalence of late treatment effects, as well as the medical, social and financial consequences of late radiation effects. Preliminary evidence suggests that it may be possible to reverse the manifestation of late injury by pharmacological or other intervention.
1420 Late effects of cancer therapy
REFERENCES 1 Hewitt M, Rowland JH, Yancik R. Cancer survivors in the United States: age, health and disability. J Gerontol 2003; 58:82–91. 2 Yabroff K, Lawrence WF, Clauser S, Davis WW, Brown ML. Burden of Illness in Cancer Survivors: Findings from a population-based national sample. J Natl Cancer Inst 2004; 96:1322–30. 3 Aziz N, Rowland JH. Trends and advances in cancer survivorship research: Challenge and opportunity. Semin Radiat Oncol 2003; 13:248–66. ◆4 Skinner R, Wallace WH, Levitt GA. Long-term follow-up of people who have survived cancer during childhood. Lancet Oncol 2006; 7:489–98. 5 Cox JD, Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int J Radiat Oncol Biol Phys 1995; 31: 1341–6. ◆6 Dorr W, Hendry JH. Consequential late effects in normal tissues. Radiother Oncol 2001; 61:223–31. 7 Martin M, Lefaix J, Delanian S. TGF-beta1 and radiation fibrosis: a master switch and a specific therapeutic target? Int J Radiat Oncol Biol Phys 2000; 47:277–90. 8 Rodemann HP, Bamberg M. Cellular basis of radiationinduced fibrosis. Radiother Oncol 1995; 35:83–90. ◆9 Bentzen SM. Preventing or reducing late side effects of radiation therapy: radiobiology meets molecular pathology. Nat Rev Cancer 2006; 6:702–13. 10 O’Sullivan B, Levin W. Late radiation-related fibrosis: pathogenesis, manifestations, and current management. Semin Radiat Oncol 2003; 13:274–89. 11 World Health Organization World Health Organization Handbook for Reporting Results of Cancer Treatment WHO Offset Publication no. 48. Geneva, Switzerland; 1979. 12 http://ctep.cancer.gov/reporting/ctc.html NCICAa. CTC v2.0 and CTCAE v3.0; 2005. 13 Pavy JJ, Denekamp J, Letschert J, et al. EORTC Late Effects Working Group. Late effects toxicity scoring: the SOMA scale. Radiother Oncol 1995; 35:11–15. 14 Rubin P, Constine LS, Fajardo LF, et al. RTOG Late Effects Working Group. Overview. Late Effects of Normal Tissues (LENT) scoring system. Int J Radiat Oncol Biol Phys 1995; 31:1041–2. 15 Baumann M, Bentzen, SM. Clinical manifestations of normal-tissue damage. In: Steel GG, editor. Basic Clinical Radiobiology. London: Arnold; 2002. pp. 105–119. 16 Trotti A, Colevas AD, Setser A, et al. CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol 2003; 13:176–81. 17 Chen Y, Trotti A, Coleman CN, et al. Adverse event reporting and developments in radiation biology after normal tissue injury: International Atomic Energy Agency consultation. Int J Radiat Oncol Biol Phys 2006; 64:1442–51.
◆18
19
20
21
22
23
24
25
26
27
28
29
30
Bentzen SM, Dorr W, Anscher MS, et al. Normal tissue effects: reporting and analysis. Semin Radiat Oncol 2003; 13:189–202. Routledge JA, Burns MP, Swindell R, et al. Evaluation of the LENT-SOMA scales for the prospective assessment of treatment morbidity in cervical carcinoma. Int J Radiat Oncol Biol Phys 2003; 56:502–10. Velikova G, Wright EP, Smith AB, et al. Automated collection of quality-of-life data: a comparison of paper and computer touch-screen questionnaires. J Clin Oncol 1999; 17:998–1007. Drummond HE, Ghosh S, Ferguson A, et al. Electronic quality of life questionnaires: a comparison of pen-based electronic questionnaires with conventional paper in a gastrointestinal study. Qual Life Res 1995; 4:21–6. Jensen K, Bonde Jensen A, Grau C. The relationship between observer-based toxicity scoring and patient assessed symptom severity after treatment for head and neck cancer. A correlative cross sectional study of the DAHANCA toxicity scoring system and the EORTC quality of life questionnaires. Radiother Oncol 2006; 78:298–305. Denis F, Garaud P, Bardet E, et al. Final results of the 94-01 French Head and Neck Oncology and Radiotherapy Group randomized trial comparing radiotherapy alone with concomitant radiochemotherapy in advanced-stage oropharynx carcinoma. J Clin Oncol 2004; 22:69–76. Adelstein DJ, Li Y, Adams GL, et al. An intergroup phase III comparison of standard radiation therapy and two schedules of concurrent chemoradiotherapy in patients with unresectable squamous cell head and neck cancer. J Clin Oncol 2003; 21:92–8. Denis F, Garaud P, Bardet E, et al. Late toxicity results of the GORTEC 94-01 randomized trial comparing radiotherapy with concomitant radiochemotherapy for advanced-stage oropharynx carcinoma: comparison of LENT/SOMA, RTOG/EORTC, and NCI-CTC scoring systems. Int J Radiat Oncol Biol Phys 2003; 55:93–8. Calais G, Alfonsi M, Bardet E, et al. Randomized trial of radiation therapy versus concomitant chemotherapy and radiation therapy for advanced-stage oropharynx carcinoma. J Natl Cancer Inst 1999; 91:2081–6. Trotti A, Bentzen SM. The need for adverse effects reporting standards in oncology clinical trials. J Clin Oncol 2004; 22:19–22. Eifel P, Levenback, C, Taylor-Wharton, J, Oswald, MJ. Time course and incidence of late complications in patients treated with radiation therapy for FIGO stage 1B carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1995; 32:1289–1300. Pedersen D, Bentzen SM, Overgaard J. Early and late radiotherapeutic morbidity in 442 consecutive patients with locally advanced carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 1994; 29:941–52. Shah SA CR, Benoit E, Breen EL, Bleday R. Rectal complications after prostate brachytherapy. Dis Colon Rectum 2004; 47:1487–92.
References 1421
31 Ataman OU, Barrett A, Filleron T, et al. Optimization of follow-up timing from study of patterns of first failure after primary treatment. An example from patients with NSCLC: a study of the REACT working group of ESTRO. Radiother Oncol 2006; 78:95–100. 32 Bentzen SM. Towards evidence based radiation oncology: improving the design, analysis, and reporting of clinical outcome studies in radiotherapy. Radiother Oncol 1998; 46:5–18. 33 Dische S. Revealing morbidity. Radiother Oncol 1999; 53:173–5. 34 Hancock SL, Hoppe RT. Long-term complications of treatment and causes of mortality after Hodgkin’s disease. Semin Radiat Oncol 1996; 6:225–42. 35 Green DM, Hyland A, Chung CS, et al. Cancer and cardiac mortality among 15-year survivors of cancer diagnosed during childhood or adolescence. J Clin Oncol 1999; 17:3207–15. 36 Adams MJ, Lipshultz SE, Schwartz C, et al. Radiationassociated cardiovascular disease: manifestations and management. Semin Radiat Oncol 2003; 13:346–56. 37 Boivin JF, Hutchison GB, Lubin JH, et al. Coronary artery disease mortality in patients treated for Hodgkin’s disease. Cancer 1992; 69:1241–7. ◆38 Kremer LC, van der Pal HJ, Offringa M, et al. Frequency and risk factors of subclinical cardiotoxicity after anthracycline therapy in children: a systematic review. Ann Oncol 2002; 13:819–29. 39 Pihkala J, Saarinen UM, Lundstrom U, et al. Myocardial function in children and adolescents after therapy with anthracyclines and chest irradiation. Eur J Cancer 1996; 32A:97–103. 40 Lipshultz SE, Lipsitz SR, Sallan SE, et al. Chronic progressive cardiac dysfunction years after doxorubicin therapy for childhood acute lymphoblastic leukemia. J Clin Oncol 2005; 23:2629–36. ◆41 Floyd JD, Nguyen DT, Lobins RL, et al. Cardiotoxicity of cancer therapy. J Clin Oncol 2005; 23:7685–96. 42 Theodoulou M, Seidman AD. Cardiac effects of adjuvant therapy for early breast cancer. Semin Oncol 2003; 30:730–9. 43 Green DM, Gingell RL, Pearce J, et al. The effect of mediastinal irradiation on cardiac function of patients treated during childhood and adolescence for Hodgkin’s disease. J Clin Oncol 1987; 5:239–45. ◆44 van der Pal HJ, van Dalen EC, Kremer LC, et al. Risk of morbidity and mortality from cardiovascular disease following radiotherapy for childhood cancer: a systematic review. Cancer Treat Rev 2005; 31:173–85. 45 Taylor CW, McGale P, Darby SC. Cardiac risks of breastcancer radiotherapy: a contemporary view. Clin Oncol (R Coll Radiol) 2006; 18:236–46. 46 Silber JH, Cnaan A, Clark BJ, et al. Enalapril to prevent cardiac function decline in long-term survivors of pediatric cancer exposed to anthracyclines. J Clin Oncol 2004; 22:820–8.
47 Lipshultz SE, Lipsitz SR, Sallan SE, et al. Long-term enalapril therapy for left ventricular dysfunction in doxorubicintreated survivors of childhood cancer. J Clin Oncol 2002; 20:4517–22. 48 Hall EJ, Wuu CS. Radiation-induced second cancers: the impact of 3D-CRT and IMRT. Int J Radiat Oncol Biol Phys 2003; 56:83–8. 49 Skinner R WW, Levitt GA et al. Therapy based long-term follow-up: practice statement UK Children’s Cancer Study Group. http://www.ukccsg.org.uk/public/followup/ PracticeStatement/index.html (accessed April 24 2006); 2006. 50 Bhatia S, Yasui Y, Robison LL, et al. High risk of subsequent neoplasms continues with extended follow-up of childhood Hodgkin’s disease: report from the Late Effects Study Group. J Clin Oncol 2003; 21:4386–94. ◆51 Franklin JG, Paus MD, Pluetschow A, et al. Chemotherapy, radiotherapy and combined modality for Hodgkin’s disease, with emphasis on second cancer risk. Cochrane Database Syst Rev 2005:CD003187. 52 Ji J, Hemminki K. Second primary malignancies among patients with soft tissue tumors in Sweden. Int J Cancer 2006; 119:909–14. 53 Travis LB, Rabkin CS, Brown LM, et al. Cancer survivorship— genetic susceptibility and second primary cancers: research strategies and recommendations. J Natl Cancer Inst 2006; 98:15–25. 54 Piccirillo JF. Importance of comorbidity in head and neck cancer. Laryngoscope 2000; 110:593–602. 55 Kaplan MH, Feinstein AR. The importance of classifying initial co-morbidity in evaluatin the outcome of diabetes mellitus. J Chronic Dis 1974; 27:387–404. 56 Borggreven PA, Kuik DJ, Quak JJ, et al. Comorbid condition as a prognostic factor for complications in major surgery of the oral cavity and oropharynx with microvascular soft tissue reconstruction. Head Neck 2003; 25:808–15. 57 Paleri V, Narayan R, Wight RG. Descriptive study of the type and severity of decompensation caused by comorbidity in a population of patients with laryngeal squamous cancer. J Laryngol Otol 2004; 118:517–21. 58 Balducci L, Extermann M. Cancer and aging. An evolving panorama. Hematol Oncol Clin North Am 2000; 14:1–16. ◆59 Holscher T, Bentzen SM, Baumann M. Influence of connective tissue diseases on the expression of radiation side effects: a systematic review. Radiother Oncol 2006; 78:123–30. 60 Keller-Reichenbecher MA, Bortfeld T, Levegrun S, et al. Intensity modulation with the “step and shoot” technique using a commercial MLC: a planning study. Multileaf collimator. Int J Radiat Oncol Biol Phys 1999; 45: 1315–24. 61 Andreyev J. Gastrointestinal complications of pelvic radiotherapy: are they of any importance? Gut 2005; 54:1051–4. 62 Kennedy A, Rogers A. Improving self-management skills: a whole systems approach. Br J Nurs 2001; 10:734–7. 63 Health Do. Self-care support: a compendium of practical examples across the whole system of health and social care. London: Department of Health; 2005.
1422 Late effects of cancer therapy
64 Jerant AF, von Friederichs-Fitzwater MM, Moore M. Patients’ perceived barriers to active self-management of chronic conditions. Patient Educ Couns 2005; 57:300–7. 65 Maher EJ. Late radiation damage – whose point of view? Radiother Oncol 2000; 57:S1–S2. 66 Faithfull S. Developments in radiotherapy treatment for cancer: implications for practice Eur J Cancer Care 2005; 14:91–100. ◆67 Denton AS, Clarke NW, Maher EJ. Non-surgical interventions for late radiation cystitis in patients who have received radical radiotherapy to the pelvis. Cochrane Database Syst Rev 2002:CD001773. ◆68 Denton A, Forbes A, Andreyev J, et al. Non surgical interventions for late radiation proctitis in patients who have received radical radiotherapy to the pelvis. Cochrane Database Syst Rev 2002:CD003455. ◆69 Bennett MH, Feldmeier J, Hampson N, et al. Hyperbaric oxygen therapy for late radiation tissue injury. Cochrane Database Syst Rev 2005:CD005005. 70 Wallace WH, Blacklay A, Eiser C, et al. Developing strategies for long term follow up of survivors of childhood cancer. BMJ 2001; 323:271–4.
71 NICE. National Institute for Health and Clinical Excellence. Improving outcomes in children and young people with cancer. Guidance on cancer services. http://www.nice.org.uk/page.aspx?0=268841 2006:(accessed 24th April 2006). 72 Taylor A, Hawkins M, Griffiths A, et al. Long-term follow-up of survivors of childhood cancer in the UK. Pediatr Blood Cancer 2004; 42:161–8. 73 Oeffinger KC, Eshelman DA, Tomlinson GE, et al. Programs for adult survivors of childhood cancer. J Clin Oncol 1998; 16:2864–7. 74 Ataman OU, Barrett A, Davidson S, et al. Audit of effectiveness of routine follow-up clinics after radiotherapy for cancer: a report of the REACT working group of ESTRO. Radiother Oncol 2004; 73:237–49. ◆75 Hauer-Jensen M, Wang J, Denham JW. Bowel injury: current and evolving management strategies. Semin Radiat Oncol 2003; 13:357–71.
62 Surgical oncology DAVID MERRILEES AND DAVID NEAL
Introduction Diagnosis of malignancy Staging of malignancy Surgery and surgical oncology Preservation of structure and function
1423 1423 1424 1425 1427
INTRODUCTION Surgical oncology encompasses the surgical treatment of malignancy and has a major role in providing curative treatment for solid malignancy. Surgical oncology has evolved from the situation where generic surgeons performed radical surgery in isolation, to what should be now be a multidisciplinary team with an emphasis on combining modalities such as radiotherapy and chemotherapy. Expert surgical oncologists aim to achieve outcomes focused on cancer cure and preservation of organ structure and function. In this chapter we will use urological malignancies to illustrate the principles of modern surgical oncology. We also discuss the emerging roles of minimally invasive surgery, robotics and the future, as we see it, of surgical oncology. Surgery in isolation remains appropriate for some malignancies, albeit an ever-decreasing number. Radical surgery for renal cell cancer is one example, although renal preservation strategies or minimally invasive approaches are increasingly used. There are also small molecule tyrosine kinase inhibitors in development that will significantly change the way in which metastatic renal cell cancer is treated. Many malignancies are best treated with a combination of therapies. Surgical oncology should be integrated with other treatments provided by medical and radiation oncologists who also need to have a good understanding of surgical oncology. Neoadjuvant chemotherapy followed by surgery is one approach to combined systemic and local treatment that has been shown to confer survival benefits for some solid cancers including urothelial cell cancer of the bladder. Adjuvant therapy also has defined roles in other malignacies
Provider volume and outcome Minimally invasive surgery Follow-up protocols Conclusions References
1429 1429 1431 1431 1431
including colonic adenocarcinoma. These combination treatments need to be used appropriately because some elderly patients may not be able to tolerate current toxic regimens following major surgery such as cystectomy. Urology includes hormone sensitive cancers. Adjuvant hormonal deprivation is useful in the treatment of high-risk locally advanced prostatic cancer prior to radical radiotherapy and has been shown to improve disease-free survival and freedom from biochemical recurrence in this setting.1** Testicular cancer provides perhaps the best example of a malignancy where outcomes have improved as a result of the collaboration of medical and radiation oncologists with surgeons. Although much of this improvement is down to the results of platinum-based chemotherapy, the multidisciplinary approach to this malignancy provides a good example of modern surgical oncology where radical surgery is completely integrated into the patient’s management. Changing practice in surgical oncology in recent times has resulted from clinical research, randomized controlled trials and the advent of new technology. Improvements in perioperative care have also been highly significant. This, associated with increased surgical sub-specialization has led to ‘high volume’ centres from which clinical outcomes have improved.2 Although randomized clinical trials provide the highest level of evidence around which the practice of medicine is altered, good quality information is also obtained through large collaborative groups such as the Wilms’ tumour registry.
DIAGNOSIS OF MALIGNANCY In the UK, cancer remains the second most common cause of death behind cardiovascular disease (Office for National
1424 Surgical oncology
Statistics, General Register Office, Edinburgh). There have been improvements in the outcomes for many different malignancies but almost universally cure for metastatic disease remains uncommon. We have yet to develop effective, curative systemic therapy to deal with many common malignancies. However, as discussed elsewhere in this book, the paradigm is changing but with the acceptance that for many patients cancer will remain a chronic disease. Early diagnosis of malignancy remains critical. Early diagnosis implies good access to high-quality health care and caries with it significant cost implications. As effective therapies, both surgical and medical, have been developed to treat localized cancer, surgical oncologists have naturally shifted to examine the role of screening asymptomatic populations for the detection of early stage, curable malignancy. The role of screening for prostate cancer has provided a fertile platform for opinions and debate. There are markedly different practices between the UK and North America. Two large-scale randomized trials, the PLCO and European Randomised Screening for Prostate Cancer trial, have randomized over 200 000 men into control and intervention arms.3** These trials will help to answer whether screening for prostate cancer is beneficial for the population. Screening for prostate cancer illustrates many of the difficulties encountered in introducing a test such as that for prostate serum antigen (PSA), which has imperfect sensitivity and specificity, and using it to diagnose a malignancy where little was understood about the natural history of the disease or treatment outcomes. There is evidence that treatment of localized prostate cancer with surgery improves disease-specific survival and overall survival compared with monitoring or watchful waiting.4** Such results will increase the numbers of men having a PSA test and will increase the numbers of early-stage prostate cancers. The question of whether these men with screen-detected disease should be treated by monitoring, surgery or radiotherapy remains open. If outcomes for the treatment of malignancy continue to improve and the morbidity of surgery declines, through advances such as laparoscopy and robotics, screening may take on new implications. Genetic profiling of populations may result in the ability to predict an individual’s lifetime risk of developing, for example, breast, thyroid or prostatic cancer and, therefore, offer minimally invasive preventative surgery, or perhaps less invasive chemoprevention strategies. This will require a dramatic change in assessment of surgical risk on behalf of both the surgeon and patient. Such possibilities are fraught with ethical dilemma but may eventually become part of surgical reality. For instance, for many years women with BRAC gene mutations have had to consider the role of prophylactic bilateral mastectomy.
Assessment of malignant potential and staging Histological diagnosis of tumour subtype is important in helping to shape surgical treatment. For instance, the
histological subtype of Wilms’ tumour, for example, is one of the critical factors in predicting outcome and therefore the requirement for more or less aggressive neo-adjuvant/ adjuvant therapy. The histological subtype of bladder cancer including whether there are small cell features or sarcomatoid differentiation predicts a poor treatment response to both surgical therapy and chemotherapy.5* It is likely that in the future there will be a greater role for in vivo testing of malignant cell lines against treatment regimens including chemotherapy and more importantly the development of cancer vaccines. The histopathologist is vital in breast cancer histology to provide information on receptor positivity/negativity in tumours – perhaps the most relevant current example is that of Her-2 receptor-positive breast cancer and response to subsequent treatment with trastuzumab (Herceptin). Treatment with trastuzumab has been shown to significantly delay clinical progression in Her-2 receptor-positive breast cancer (up to 50 per cent reduction).6**,7**,8** In a pivotal phase III trial published in 2001, it was also shown to improve mortality with a median follow-up of 29 months.6 This improvement in survival has been reinforced by recently published data.7,8 Tyrosine kinase inhibitors (TKIs) have been developed that are dual inhibitors of c-erbB1 and c-erbB2 (lapatinib) and may help to reduce problems of drug resistance encountered using single agent inhibitors (trastuzumab and gefitinib). The development of TKIs that are cytostatic rather than cytotoxic will change the way cancer is treated in the future. Metastatic renal cell carcinoma is likely to be treated with debulking nephrectomy and subsequent small molecule therapy that will result in a prolonged period of clinical non-progression – a chronic disease situation. Parallels can be seen between cancer treatments and the treatment of other chronic potentially fatal diseases such as human immunodeficiency virus (HIV) infection, for which there is now effective antiretroviral therapy and a long latent stage of infection.
STAGING OF MALIGNANCY Staging of malignancy following diagnosis often follows the TNM classification. Clinical staging is largely based on radiological imaging. Imaging modalities continue to improve in their definition with high-resolution computed tomography (CT) scanning and three dimensional reformatting. Perhaps the greatest drawback of our reliance on CT and magnetic resonance imaging (MRI) in the staging of malignancy is that these modalities do not provide data on tissue function/ activity. This is particularly relevant for lymph node involvement in for example testicular, bladder and penile cancer. Positron emission tomography (PET) scanning and lymphotrophic nano-particle MRI are imaging modalities that have attempted to accurately stage urological and other malignancies preoperatively. These functional studies have not perhaps fully lived up to promised expectation and
Surgery and surgical oncology 1425
currently have limited role in assessing in particular nodal status. The major interest in PET scanning revolved around a potential use in post-chemotherapy residual mass in testicular cancer to help differentiate between residual tumour/teratoma and fibrosis. Thus far PET scanning has not altered treatment plans and therefore has only a minor role in assessment of residual mass.9 Lymphotropic nano-particle MRI using ferumoxtran-10 enhancement is probably more effective than standard MRI in assessing lymph node status in bladder cancer10 but, with increasing evidence that lymphadenectomy improves survival in bladder cancer, detection of micrometastatic disease in normal sized lymph nodes may be of little clinical use. The role of functional staging tools in surgical oncology is therefore yet to be defined, at least in urological surgery. There will undoubtedly be a greater role for this approach in the future.
Tumour markers Tumour markers in current use consist of onco-foetal antigens or serum glycoproteins. Measurement of serum tumour markers provides valuable information for diagnosis, treatment response, follow-up and prognostic purposes. However, the introduction of new bio-markers will impact significantly on our use of markers for prognosticating and diagnosis. The use of tumour markers has been particularly important for testicular cancer where beta-human chorionic gonadotrophin (β-hCG), alpha-fetoprotein (AFP) and lactate dehydrogenase (LDH) reflect tumour subtype and the response to treatment. The absolute level of tumour markers reflect the tumour burden (LDH is used in seminoma), prognosis (TNMS staging where S represents tumour marker level) and the response to treatment (falling levels following orchidectomy or chemotherapy for nodal/visceral metastases). The rate at which the tumour markers fall reflects both the serum half-life of the particular marker, and also treatment response. The search for novel serum markers that reflect prognosis in different malignancies is the focus of intense research. Potential markers include circulating cytokines, cancer antigens/proteins and angiogenesis factors. The great hope for serum bio-markers is that they will allow us to predict which patients require treatment with which modality. As an example, a bio-marker may be discovered that stratifies the risk of TIG3 TCC bladder and therefore directs clinicians and patients toward aggressive therapy with early cystectomy or bladder preservation with intravesical BCG.
SURGERY AND SURGICAL ONCOLOGY The traditional ‘Halsted’ approach to surgery for malignancy involved radical surgery to the affected organ coupled
with regional lymphadenectomy. This approach remains valid for tumours such as renal cell cancer and penile cancer. Robson staged renal cell cancer and studied the impact on prognosis. Although his staging has been modified, the standard surgery for renal cell cancer has changed little – it still involves radical removal of the diseased kidney and adrenal gland outside Gerota’s fascia with early control of the renal vasculature. De-bulking radical surgery for renal cell cancer has been shown to improve life expectancy when combined with immunotherapy, versus immunotherapy alone. Debulking surgery was abandoned for many malignancies after recognizing that the patient morbidity outweighed any potential small gain in survival – it is possible, however, that in surgical oncology de-bulking surgery will have a greater role again in the future as macromolecules such as interferon and inhibitors for tyrosine kinase receptors and the HIF (hypoxiainducible pathways) are developed. The greatest recent advance in the management of renal cell cancer is the use of laparoscopic or minimally invasive surgery. The surgical dissection of laparoscopic nephrectomy aims to completely recapitulate its open counterpart and has now expanded to laparoscopic partial nephrectomy. The results of such surgery for T1 tumours are identical to open surgery and offer the benefits of reduced wound morbidity, decreased pain and, therefore, earlier return to activity. It is now accepted that laparoscopic nephrectomy for smaller tumours is the ‘gold standard’ rather than open nephrectomy. The place of open partial nephrectomy remains the gold standard for smaller tumours in patients who select a renal preserving approach or those with impaired renal function. However, expert laparoscopic surgeons are moving to carrying out laparoscopic partial nephrectomy.11 Advanced renal cell cancer with inferior vena cava thrombus extending above the liver exemplifies a planned surgical approach with cooperation between surgical disciplines with urologists operating with cardiac surgeons under cardiopulmonary bypass and hypothermic circulatory arrest. Figure 62.1 illustrates a renal tumour with the distal extent of tumour thrombus shown in a coronal reformatted CT scan. The role of lymphadenectomy in renal cell cancer remains controversial. Lymph node status provides prognostic information that accurately allows pathological staging but probably confers no survival benefit. However, the role of lymphadenectomy in surgical oncology has become more important because, as systemic therapies improve, it allows identification of patients who may benefit from chemotherapy. It provides accurate staging and also, for some cancers of the penis and bladder, it provides survival benefit. Extended lymphadenectomy in bladder cancer has recently been shown to provide a survival benefit and relates directly to the number of lymph nodes retrieved at the time of surgery.12–14 Lymphadenectomy plays an important role in many other malignancies including breast, colon and melanoma. In urology, lymphadenectomy is most relevant for penile cancer where early ilio-inguinal nodal dissection for clinical
1426 Surgical oncology
mapping is accurate18 but as yet has not been shown to result in a survival benefit.
Neo-adjuvant therapy and surgery
(a)
(b)
Figure 62.1 A & B A renal tumour with the distal extent of tumour thrombus shown in a coronal reformatted CT scan.
N0 disease provides a clear survival benefit as opposed to delayed node dissection.15 However, extended lymph node dissections carry morbidity in the setting of penile cancer with lymphoedema, wound collections/infections and flap necrosis. Morbidity is also common in complete axillary dissection for breast cancer. Sentinel node sampling is used in a variety of malignancies to prevent complications from unnecessary extended lymph node dissections. The concept of the sentinel lymph node was introduced by Cabanas for the treatment of penile cancer in 1977. It is very attractive as a concept for eliminating unnecessary extended nodal dissection. The technique and further refinements were reintroduced in 1992 by Morton for the treatment of melanoma. In penile cancer there is evidence that the use of sentinel lymph node biopsy may improve survival for T2 and T3 penile cancer.16* There is, however, a false negative rate of up to 15 per cent.17 It is therefore difficult to reconcile routine early complete ilio-inguinal lymph node dissection with attendant complications versus sentinel lymph node mapping. In melanoma, sentinel lymph node
Another important treatment strategy in surgical oncology is that of neo-adjuvant therapy followed by surgery. This approach combines systemic and local therapy and is therefore a very attractive way to improve survival through treatment of micro-metastatic disease not detected on routine staging. Neo-adjuvant therapy can take the form of either chemotherapy or radiotherapy. Bladder cancer is a moderately chemosensitive tumour with 5-year survival rates of only 65 per cent for surgically treated disease. With such poor survival data chemotherapy has been used either in the neo-adjuvant or adjuvant setting. Randomized trial data has been difficult to interpret because of small numbers of patients, but recent meta-analyses have provided clear evidence that neo-adjuvant chemotherapy has an overall absolute survival advantage of 5 to 6.5 per cent when compared with surgery alone.19***,20*** In addition to the survival benefit, neo-adjuvant chemotherapy downstages tumours and may allow resection of a previously un-resectable T4 tumour. Some patients are rendered pT0 on final histology indicating complete response to chemotherapy. Equally important is the fact the people who respond poorly to neo-adjuvant chemotherapy with disease progression or with residual positive nodes at histology do poorly – the responders to neo-adjuvant chemotherapy tend to do well after subsequent surgery, as opposed to non-responders. If patients do not respond it is best to stop neo-adjuvant chemotherapy early and proceed to surgery, as delay in time to cystectomy has also been associated with poorer long term survival, at least in high-grade superficial disease (pT1 G3).21 The role of adjuvant chemotherapy in bladder cancer is not as well defined and awaits further data from soon-tobe completed randomized trials. Adjuvant therapy may be more appealing since some patients in good prognostic groups may avoid unnecessary chemotherapy and side effects. However, some patients are not able to accept significant doses of chemotherapy in the early weeks after major pelvic surgery. In vivo testing of tumour against chemotherapeutic regimens is likely to help decide who should receive neo-adjuvant therapy. There are, however, problems with assuming that metastatic lesions will have the same chemosensitivity as the primary tumour – such has recently been highlighted by the fact the some Her-2 negative breast primary tumours may in fact have Her-2 positive metastatic lesions.22 This recent finding has major implications for testing of both primary and metastatic lesions for receptor expression and sensitivity to various chemotherapy and immunotherapeutic regimens.
Preservation of structure and function 1427
Multimodal team-based approach to surgery Testicular cancer provides an excellent example of modern multimodality surgical oncology. In this disease, excellent outcomes are only achieved through effective cooperation amongst medical and radiation oncologists, and surgeons. Much credit for improvement in survival can be taken by medical oncologists with the advent of platinum-based chemotherapy. This seminal work was reported by Einhorn and Donohue23 who also provided significant leadership in the collaboration between surgeons and oncologists. This cooperation has led to refinements in treatment and continued small improvement in therapy and side-effect profiles.
(a)
PRESERVATION OF STRUCTURE AND FUNCTION Surgeons have become more aware of the potential impact of radical surgery on the quality of life of their patients. In urological oncology emphasis in the past decade has been on improving functional outcomes from radical prostatectomy and on providing an effective bladder substitute. Patient morbidity has also been a driving factor for the development of alternative treatment strategies including interstitial brachytherapy and high-intensity focused ultrasound (HIFU) for prostate cancer treatment. Preservation of organ structure and function is, therefore, an important consideration when discussing treatment alternatives with patients considering surgery. Organ preservation is one of the main aims when considering treatment options for small localized renal cell carcinoma (RCC). Partial nephrectomy for RCC less than 4 cm is the treatment option of choice for patients with a solitary kidney, renal impairment, hereditary RCC syndromes or diseases that may threaten future renal function – the oncological outcome from partial nephrectomy for tumours less than 4 cm is equivalent to radical nephrectomy. The role of partial nephrectomy in the presence of a normal contralateral kidney is a source of ongoing debate. Figure 62.2 demonstrates arterial reconstruction for renal vessel anatomy prior to surgery in kidneys with double renal arteries bilaterally. Preservation of organ function and structure is also well demonstrated when discussing the principles of management of testicular cancer. The incidence of metachronous testicular cancer is 1.9 per cent in the largest reported US series from SEER.24 With a detectable pre-malignant change in the testis, namely tubular intra-epithelial neoplasia, patients at risk for a second tumour can be identified and ‘prophylactically’ treated so that the testis may be preserved maintaining, in the most part, at least hormone function. Alternatively, the patient may be closely followed with clinical examination and ultrasound allowing for a longer period in which to maintain spermatogenesis and the chance to conceive naturally. In this circumstance, it is also possible to perform a partial orchidectomy for a small metachronous
(b)
Figure 62.2 a & b Arterial reconstruction for renal vessel anatomy prior to surgery in kidneys with double renal arteries bilaterally (See Plate Section.).
tumour (less than 2 cm) thus potentially preserving endocrine and reproductive function of the testis. Patients and surgeons will continue to improve treatment strategies in the future to allow preservation of structure and function while maintaining cancer cure.
Anatomical knowledge Sound anatomical knowledge is an important part of surgical oncology. The way in which medical students acquire anatomical knowledge has changed dramatically. Dissection room teaching has been reduced by many medical schools and in its place prosections and electronic images are used. We hope that this does not impact on the anatomical knowledge gained by surgeons during their training. A sound anatomical knowledge is crucial for the application of surgical principles when treating malignancy. This is particularly important in a surgical dissection of the retroperitoneum. Anatomical mapping of the lymphatic drainage of the left and right testicle and the course of the
1428 Surgical oncology
(a)
(b)
Figure 62.3 a & b The retroperitoneal anatomy and delicate network of sympathetic nerves responsible for antegrade ejaculatory function. From Innovations In Urologic Surgery 2nd edition; Hohenfellner R, Novick A, Fichtner J. Informa Healthcare, 1997. Reproduced by permission of Taylor & Francis Books UK.
sympathetic plexus of nerves in the retroperitoneum has been important in allowing template dissection of the retroperitoneum for post-chemotherapy residual masses in testicular cancer. Figure 62.3 illustrates the retroperitoneal anatomy and delicate network of sympathetic nerves responsible for antegrade ejaculatory function. A unilateral template dissection allows preservation of the contralateral ejaculatory nerves and, therefore, antegrade ejaculation/seminal emission. Radical surgery is vital when performing retroperineal lymph node dissection (RPLND) – incomplete resection of the retroperitoneum is associated with risk of recurrence in the retroperitoneum and decreased survival. However, even in the postchemotherapy setting it is possible in carefully selected patients to carry out unilateral template dissection and preserve the appropriate nerves. There are proponents of the use of laparoscopy for RPLND, but it is fair to say that whilst ‘lumpectomy’ is feasible the recapitulation of full complete RPLND by means of laparoscopy has yet to be reported.
Nomograms and other methods of predicting outcome With an ever expanding array of different potential treatments for malignancy, nomograms have helped to focus clinicians and patients on outcomes and, therefore, to make decisions on surgical risk and treatment preference. Nomograms play an important role in estimating the risk of
extra-prostatic and nodal disease in prostate cancer and have also been developed to aid in decision making after biochemical recurrence. Table 62.1 shows the updated Partin nomogram for needle-detected prostate cancer with PSA in the range 4–10 ng/ml – the most commonly detected prostate cancer treated with curative intent. Utilizing the two variables above and the Gleason grade gives each patient their likelihood (percentage probability) of having organ-confined, locally advanced or metastatic disease. These tools are a useful adjunct to patient counselling but are often difficult to apply to the individual patient because the figures are population-based and merely another form of risk/benefit ratio for the patient to consider. There are also other concerns with the use of nomograms – the data provided is from a cohort of patients who were diagnosed and treated in the past, often a decade beforehand, and therefore to use this data is to look back at the way disease was diagnosed and treated by previous surgical oncologists, rather than to look forward to the future. The Partin tables have therefore been updated recently (2001) to provide a better reflection of the significant stage migration that prostate cancer has undergone in the past decade.
The role of the multidisciplinary team A major change in surgical oncology has been the development of multidisciplinary teams (MDTs) that have been formed in an attempt to develop protocol driven and an
Minimally invasive surgery 1429
Table 62.1 Clinical stage T1c (nonpalpable, PSA elevated) Pathologic stage
2–4
Gleason score 5–6 347
437
8–10
Organ confined Extraprostatic extension
90 (78–98) 10 (2–22)
80 (78–83) 19 (16–21)
63 (58–68) 32 (27–36)
52 (43–60) 42 (35–50)
46 (36–56) 45 (36–54)
4.1–6.0
Seminal vesicle () Lymph node () Organ confined Extraprostatic extension
87 (73–97) 13 (3–27)
1 (0–1) 0 (0–1) 75 (72–77) 23 (21–25)
3 (2–5) 2 (1–3) 54 (49–59) 36 (32–40)
3 (1–6) 3 (1–5) 43 (35–51) 47 (40–54)
5 (3–9) 3 (1–6) 37 (28–46) 48 (39–57)
6.1–10.0
Seminal vesicle () Lymph node ()
2 (2–3) 0 (0–1)
8 (6–11) 2 (1–3)
8 (4–12) 2 (1–4)
13 (8–19) 3 (1–5)
PSA range (ng/mL)
Adapted from: http://urology.jhu.edu/prostate/partintables.php; PSA, prostate-specific antigen.
evidence-based approach to surgical oncology. Within the National Health Service in the UK, it is now a requirement that every new patient with cancer is presented and discussed at an MDT that comprises surgeons, oncologists, radiologists, pathologists and specialist cancer nurses.
PROVIDER VOLUME AND OUTCOME There is increasing evidence that patient outcome is linked to provider volume in urological oncology. Better patient outcomes with respect to morbidity and mortality are associated with both hospital volume and surgeon volume. This is particularly the case for major pelvic cancer surgery including cystectomy and radical prostatectomy.25*** Mortality from RPLND is also decreased in high-volume hospitals.26*** Interestingly, the impact of high volume translating into better outcomes from radical nephrectomy has been more difficult to prove.25 The value of the MDT environment and regionally based cancer networks is therefore vital to create an environment with high-volume cancer surgery. The UK is making significant progress in this regard. Although there is some interest in trying to define the reasons for better outcomes in high-volume centres, translating such success to small-volume hospitals will not be possible. On this basis, the old-fashioned generalist is fading from surgical oncology. Super-specialization is now a reality for most surgical disciplines with urological oncologists who perform pelvic surgery, others who perform nephrectomy laparoscopically and others who are RPLND experts. This subdivision of labour results in experts treating within their field of interest and performing a high number of repetitive cases, creating an environment to improve patient outcome. The number of major cases required to maintain surgical skills per year can be debated, but small numbers do not allow a surgeon to develop his/her technique in an operation with functional outcomes that are directly related to surgical anatomy – preservation of the neuro-vascular bundle and precise apical dissection of the prostate during radical prostatectomy.
The difficulties of developing regional cancer networks and disenfranchising previously independent urological and surgical units is not to be underestimated. There are talented surgeons who lose major open surgical oncology as part of their independent urological practice. This has been recognized as a future problem for workforce numbers and skill requirements for urologists in the UK who will now be trained as office urologists or surgical urologists in a twintrack system. It is difficult to predict future requirement for the number of surgeons who perform major cancer surgery as, with improvements in systemic therapy with macromolecules and chemotherapy, radical surgery in isolation is not a significant part of a surgeon’s practice. We may therefore require fewer rather than more ‘surgeons’ in the future who specialize in oncology. The alternative viewpoint is that with minimally invasive surgery and genetic mapping, prophylactic surgery will mean that the surgical workforce will need to expand. Screening for malignancies will become more cost effective in the future and will also impact dramatically on workforce numbers, particularly in the primary care setting where one imagines a sub-speciality of general practitioners with expertise in screening will emerge.
MINIMALLY INVASIVE SURGERY In urology Clayman performed the first laparoscopic nephrectomy in 1991. There has been an exponential increase in the uptake of minimally invasive techniques across all surgical disciplines. The explosion in minimally invasive surgery has been driven partly by patients, who are interested in new technology and less wound morbidity, and also by surgeons. The magnitude of the learning curve has been a great obstacle to surgeons taking up minimally invasive surgery. The application of laparoscopy to urological treatment was partly hampered by the fact that in urological disease there is no common pathology that is easily treated with laparoscopic surgery – compared with appendicitis and cholecystitis in general surgery. There are now many different ways in which surgeons can shorten the learning curve when beginning a laparoscopic
1430 Surgical oncology
practice. The first step in this process involves using a trainer box to learn basic laparoscopic skills, including intracorporeal knot tying. The second step usually requires the use of live animal models. The use of animal models is controversial and provokes debate and, occasionally, violence toward the individuals and institutions involved. The use of animals in the UK for the purpose of surgical training has long been outlawed, meaning that surgeons in the UK who would prefer not to learn a new operation by first performing it on a patient have to travel overseas to gain this experience. Mentoring with an established laparoscopic surgeon is the final step in becoming a solo operator. The mentor has traditionally been present on site in the operating theatre with the surgeon training in laparoscopic surgery. A novel approach to mentorship is to link via satellite as was the case for the introduction of laparoscopic donor nephrectomy in Christchurch, New Zealand. In this case, the mentor was seated at a console in Brisbane, Australia, while the surgeon operated in New Zealand. The surgical planning involved to introduce new technology and surgical techniques is time consuming but rewarding if nurses, anaesthetists and surgeons have all taken part in the preparation; it leads to a strong team structure where all have a sense of achievement. It highlights the modern attitude to surgical oncology where surgeons, once qualified, no longer introduce a new procedure to their skill set without first attaining and mastering the appropriate skills. Minimally invasive surgery may now be performed laparoscopically with either two-dimensional images or via a robotic master–slave three-dimensional system (Fig. 62.4). The robotic system used is the da Vinci master–slave and has the advantage of six degrees of freedom for hand movement as opposed to two degrees of freedom for traditional laparoscopy. In addition to this, the robotic systems use a two-camera lens, from which the view is translated to a threedimensional viewing screen. This has allowed rapid uptake of minimally invasive procedures by surgeons previously
(a)
unskilled at laparoscopy and has eliminated the difficulties of laparoscopic suturing. The console and instruments are ergonomic and more comfortable for the surgeon to use than laparoscopic instruments. The use of robotic arms also eliminates physiological tremor allowing very precise surgery, for example, at the prostatic apex. The major concern with the introduction of minimally invasive surgery has been whether the results match the results of open cancer surgery – are margins as wide and lymph node dissections as complete? In urological circles, the oncological outcomes for open versus laparoscopic nephrectomy and well selected radical prostatectomy (lowvolume impalpable disease) are equal.27*,28* Functional outcomes are similar and wound morbidity is much less. Cost is a particularly important factor for robotic surgery where the initial outlay is approximately £500 000 with significant disposable costs and annual service charges. This is weighed against the actual savings as a result of fewer nights in hospital. Where savings are also relevant is in the patient’s earlier return to activity including employment. Robotic surgery in particular has potentially fascinating applications in the future. The master can sit at the controls in America and the slave can be positioned at the patient in say Australia and the operation proceed exactly as if the surgeon was present in the operating room. The operating surgeon/technician may not have to leave his/her home to perform such surgery. This is dependent on appropriate satellite links and cable speed but will become a reality. We believe that patients will again be the one of the main driving forces behind the application of such technology – if they can have the surgeon of their choice from a global market and not have to leave their country of origin to gain the benefit of his/her expertise then it is likely that this is an option they will choose. Potential difficulties may arise with timing of surgery across time zones; it may require the surgeon to operate or the patient to be anaesthetized at times that are outside standard
(b)
Figure 62.4 The master unit with surgeon seated at the console (a) and the slave unit with surgical assistants positioned at the patient’s side (b).
References 1431
working hours. Fatigue and performance of complex tasks are closely related – medical errors increase at times of fatigue – and hence this will need to be carefully considered by the medical profession if patients do pursue such a possibility. Minimally invasive therapeutic techniques including radiofrequency ablation and cryotherapy are also exciting surgical options. Both of these modalities have been applied to the treatment of small renal cell cancers – lesions incidentally discovered in the elderly population. The natural history of such lesions is that they are slow growing and as such have been ideal lesions to treat with minimally invasive therapies. Although data is immature it is apparent that such technology is safe, and when used to treat small lesions, appears to be effective. Minimally invasive options such as these will continue to be applied to appropriately selected tumours, with little morbidity or functional impairment.
disease in combination with small molecule therapy to maintain low-volume disease burden will again allow metastatic cancer in the future to become a truly chronic disease.
CONCLUSIONS Surgical oncology is a dramatically changing field where minimalism without compromised cancer outcomes is becoming a reality. Radical surgery currently would appear to have a smaller and smaller role in the practice or oncology. Single modality surgical therapy for cancer has almost vanished as an option for cure, as novel molecular targeted therapies emerge.
KEY LEARNING POINTS FOLLOW-UP PROTOCOLS ●
After primary treatment of malignancy, patients are routinely surveyed with physical examination, tumour markers and cross-sectional imaging. In many cancers a rigorous protocol has not been shown to improve overall survival. Selected patients do however benefit from a strict protocoldriven follow-up. In the setting of renal cell carcinoma, resection of metachronous lung metastases is associated with improved survival particularly is the patient is rendered R0 following treatment.29,30 Survival is also associated with fewer than seven metastatic lung lesions and small size of lesion (4 cm). Likewise, a solitary hepatic metastasis in colorectal cancer is often treatable surgically with little morbidity. Although follow-up protocols have not as yet been proven to significantly affect outcome, early detection of metastatic
REFERENCES ●1
2
3
●4
5
Bolla M, Collette L, Blank L, et al. Long term results with immediate androgen suppression and external irradiation in patients with locally advanced prostate cancer (an EORTC trial): a phase III randomised trial. Lancet 2002; 360(9327):103–6. Birkmeyer JD, Stukel TA, Siewers AE, et al. Surgeon volume and operative mortality in the United States. N Engl J Med 2003; 349:2117–27. de Koning HJ, Auvinen A, Berenguar Sanchez A, et al. Large-scale randomized prostate cancer screening trials: program performances in the European Randomized Screening for Prostate Cancer trial and the Prostate, Lung, Colorectal and Ovary Cancer trial. Int J Cancer 2002; 97:237–44. Bill-Axelson A, Holmberg L. Radical prostatectomy versus watchful waiting in early prostate cancer. N Engl J Med 2005; 352:1977–84. Chen L, Pan CX, Yang XJ, et al. Small cell carcinoma of the urinary bladder: a clinicopathologic analysis of 64 patients. Cancer 2004; 101:957–62.
●
●
●
●
●
●6
●7
●8
9
10
◆11
Surgery in isolation to cure malignancy is now uncommon. Combination therapy with surgery and chemotherapy/radiotherapy provides better outcomes in surgical oncology. Minimally invasive therapy has an ever expanding role in surgical oncology. Small molecule inhibitors are likely to transform many cancers into chronic diseases in the future. Surgical sub-specialization, cancer networks and multidisciplinary teams provide a platform of improved patient care. High-volume surgeons and high-volume hospitals have better outcomes.
Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344:783–92. Romond EH, Perez EA, Bryant J, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 2005; 353:1673–84. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 2005; 353:1659–72. Karapetis CS, Strickland AH, Yip D, et al. Use of fluorodeoxyglucose PET scans in patients with advanced germ cell tumour following chemotherapy: single-centre experience with long-term follow up. Intern Med J 2003; 22:427–35. Deserno WM, Harisinghani MG, Jaupitz M, et al. Urinary bladder cancer: preoperative nodal staging with ferumoxtran-10-enhanced MR imaging. Radiology 2004; 233:449–56. Link RE, Bhayani S, Allaf M, et al. Exploring the learning curve, pathological outcomes and perioperative morbidity of laparoscopic partial nephrectomy performed for renal mass. J Urol 2005; 173:1690–4.
1432 Surgical oncology
12 Herr HW. Extent of surgery and pathology evaluation has an impact on bladder cancer outcomes after radical cystectomy. Urology 2003; 61:105–8. 13 Stein JP, Cai J, Groshen S, Skinner DG. Risk factors for patients with pelvic lymph node metastases following radical cystectomy with en bloc pelvic lymphadenectomy: concept of lymph node density. J Urol 2003; 170:35–41. 14 Sanderson KM, Stein JP, Skinner DG. The evolving role of pelvic lymphadenectomy in the treatment of bladder cancer. Urol Oncol2004; 22:205–11. 15 Kroon BK, Horenblas S, Lont AP, et al. Patients with penile carcinoma benefit from immediate resection of clinically occult lymph node metastases. J Urol 2005; 173:816–9. 16 Lont AP, Horenblas S, Tanis PJ, et al. Management of clinically node negative penile carcinoma: improved survival after the introduction of dynamic sentinel node biopsy. J Urol 2003; 170:783–6. 17 Kroon BK, Horenblas S, Meinhardt W, et al. Dynamic sentinel node biopsy in penile carcinoma; evaluation of 10 years experience. Eur Urol 2005; 47:601–6. 18 Morton DL, Cachran AJ, Thompson JF, et al. Sentinel node biopsy for early-stage melanoma: accuracy and morbidity in MSLT-1, an international multicentre trial. Ann Surg 2005; 242:302–11. ●19 Vale C. Advanced Bladder Cancer Meta-analysis Collaboration. Neoadjuvant chemotherapy in invasive bladder cancer: a systematic review and meta-analysis. Lancet 2003; 361:1927–34. 20 Winquist E. Kirchner TS, Segal R, et al. Neoadjuvant chemotherapy for transitional cell carcinoma of the bladder: a systematic review and meta-analysis. J Urol 2004; 171:561–9.
21 Herr HW, Sogani PC. Does early cystectomy improve the survival of patients with high risk superficial bladder tumours? J Urol 2001; 166:1296–9. 22 Zidan J, Dashkovsky I, Stayerman C, et al. Comparison of HER-2 overexpression in primary breast cancer and metastatic sites and its effect on biological targeting therapy of metastatic disease. Br J Cancer 2005; 93:552–6. ●23 Donohue JP, Einhorn LH, Perez JM. Improved management of nonseminomatous testis tumours. Cancer 1978; 42:2903–8. 24 Fossa SD, Chen J, Schonfeld SJ, et al. Risk of contralateral testicular cancer: a population-based study of 29,515 US men. J Natl Cancer Inst 2005; 97:1056–66. ◆25 Nuttall M, van der Meulen J, Phillips N, et al. A systematic review and critique of the literature relating hospital or surgeon volume to health outcomes for 3 Urological cancer procedures. J Urol 2004 Dec; 172:2145–52. 26 Joudi FN, Konety BR. The impact of provider volume on outcomes from urological cancer therapy. J Urol 2005; 174:432–8. 27 Permpongkosol S, Chan DY, Link RE, et al. Laparoscopic radical nephrectomy: long-term outcomes. J Endourol 2005; 19:628–33. 28. Rassweiler J, Schulze M, Teber D, et al. Laparoscopic radical prostatectomy: functional and oncological outcomes. Curr Opin Urol 2004; 14:75–82. 29 Piltz S, Meimarakis G, Wichmann MW, et al. Long-term results after pulmonary resection of renal cell carcinoma metastases. Ann Thorac Surg 2002; 73:1082–7. 30 Hofmann HS, Neef H, Krohe K, et al. Prognostic factors and survival after pulmonary resection of metastatic renal cell carcinoma. Eur Urol 2005; 48:77–81.
63 Cancer prevention strategies WILLIAM P. STEWARD AND ANDREAS GESCHER
The influence of diet on cancer incidence Chemoprevention Role of surgery
1433 1436 1438
It has been estimated that approximately 60 per cent of all cancers are related to environmental exposures or lifestyle choices.1 Components of the diet and cigarette smoking are, by far, the most important of these. Increasingly, information is becoming available on genetic causes of several tumours with hereditary cancer syndromes being identified. The marked increase in the use of chemotherapy and radiotherapy and improved cure rates of several malignancies have resulted in a large number of individuals who are at a greater risk of developing second primary tumours. Despite the increasing knowledge of the causes of cancer, and the ability to identify those who are at higher risk of developing malignancies, there is limited information or ongoing research to develop cancer prevention strategies for these individuals. This chapter will focus on the potential for cancer prevention through modification of dietary intake, use of chemoprevention strategies and the application of surgery in high-risk individuals. The evidence base for these approaches is often uncertain and will be reviewed. In contrast, the role of cigarette smoking in the aetiology of a variety of cancers is well established2,3* and is discussed in earlier chapters. It is clear that cessation of cigarette smoking would have a major impact on cancer incidence. Approaches to modify smoking behaviour will not be discussed in this chapter but have been reviewed extensively in the literature.4
THE INFLUENCE OF DIET ON CANCER INCIDENCE Important information on the potential impact of diet on the aetiology of cancer has come from studies of population
Conclusions References
1439 1440
migration.5* The incidence of many cancers varies strikingly between and within continents.6 For the majority of cancers that have been studied, the incidence among migrants changes rapidly7* and approaches that in the host country, sometimes within a single generation. The age at the time of migration has an important influence on the degree to which first-generation migrants have an altered risk. The earlier the age at migration, the greater is the likelihood of adopting the cancer incidence of the indigenous population of the country to which they have migrated. Such observations clearly indicate that environmental and lifestyle factors play an important part in causing cancer, potentially by modifying the likelihood of DNA mutation, and they argue against a genetically inherited risk. Many putative environmental and lifestyle factors have been suggested as important, predominantly from retrospective studies of varying exposure, but unequivocal proof of their impact remains lacking for the majority.
Dietary fibre Dietary fibre comprises non-starch plant-derived polysaccharides that are resistant to hydrolysis by human digestive enzymes.8 Many studies have focused on its potential to modify the risk of colorectal cancer. The presence of fibre in the bowel leads to biological effects that could have an impact on carcinogenesis. It leads to increased stool bulk, which reduces transit time and decreases the concentration of potential carcinogens within the stool.9 Both of these factors together reduce exposure of the colonic epithelium to carcinogens. Fibre also stimulates anaerobic bacterial fermentation, leading to the production of short-chain
1434 Cancer prevention strategies
fatty acids, acetate, propionate and butyrate.10 This leads to a reduction in bowel pH and conversion of primary to secondary bile acids. These changes have been shown to reduce cell proliferation and induce apoptosis in cell lines. THE ROLE OF DIETARY FIBRE IN COLORECTAL CANCER INCIDENCE
It has long been recognized that the incidence of colorectal cancer has a huge variation around the world. Burkitt11* investigated the low incidence in Africa and hypothesized that this was related to a high fibre intake. Numerous retrospective case-control studies followed and showed a correlation between increased fibre intake and reduced risks of colorectal cancer. A meta-analysis of 16 of these studies demonstrated a significant inverse association between dietary fibre and colorectal cancer risk.12*** There are, however, major concerns with such retrospective studies, which predominantly relate to the accuracy of assessments of fibre intake. The major methodological issue involves potential recall bias when individuals may over- or underestimate components of the diet. There are also concerns with confounding variables – for example, those with high levels of exercise may take a high fibre content in their diet. Prospective studies should reduce recall bias and several have been undertaken to date.13**,14**,15**,16**,17** The majority of these have failed to show an association between fibre intake and the risk of colorectal cancer although concerns have been raised as to their validity. Two major problems have been the failure, in many, to correct for potential confounding factors (such as exercise, weight, other dietary factors) in the analysis and the use of single populations in whom dietary habits are more or less homogenous. The latter is a serious concern in that the extent of measurement error could easily obscure all but very large underlying associations with diet. The European Prospective Investigation of Cancer and Nutrition (EPIC) Study, reported in 2003,16** examined the association between dietary fibre intake and incidence of colorectal cancer in 519 978 individuals recruited from 10 European countries. There was a much wider variation in dietary fibre intake in this population than had been reported in previous studies. There was a significant correlation between increasing dietary fibre content and reduced risk of colorectal cancer (most strikingly for carcinomas of the descending colon). There was a 40 per cent reduction of risk when average fibre intakes increased from the lowest to the highest quintile but no obvious difference in outcome depending upon the source of dietary fibre. Of the other prospective studies, only the US Nurses’ Health Study17** had sufficient follow-up to observe a large number (904) of cases of colorectal cancer. This study also included potential confounding factors in its analysis. No significant correlation between fibre intake and colorectal cancer risk was detected but, unlike the EPIC study, the population was homogenous with a much lower spread between the lowest and highest dietary fibre intakes and, overall, a much lower average daily fibre
content. It is thus possible that the variation in fibre intake was too small to significantly alter cancer incidence. In summary, retrospective case-control studies would support a role for increasing dietary fibre to protect against colorectal cancer risk. All but one of the prospective studies have failed to confirm this but, by far the largest of these, EPIC, has shown a highly significant correlation between increasing dietary fibre and reduced colorectal cancer risk. The design of this study appears to correct for many potential deficiencies of the negative studies and in particular contained a broad spread of populations with highly differing fibre intakes. Taken with the results of retrospective data and the observed biological effects of fibre in preclinical systems it seems sensible to follow consensus recommendations18*** that intake of fibre in populations should be increased to reduce the risk of colorectal cancer. The findings of the EPIC study would suggest that a doubling of total fibre intake from food could reduce the risk of colorectal cancer by 40 per cent.
FIBRE INTAKE AND OTHER CANCERS
Oestrogens are secreted by the biliary tract, and fibre has been shown to reduce intestinal absorption of these hormones. Breast cancer risk has been correlated with dietary fibre intake in several retrospective case-control studies,19* and an inverse correlation between breast cancer risk and dietary fibre intake has been observed. All prospective studies have failed to confirm this association. In particular, a large Canadian National Breast Screening Study20** with a 16.2 year follow up period, which included 2536 cases of breast cancer, revealed no association between dietary fibre intake and risk. The incidence of gastric cancer was investigated in the Netherlands cohort study,21* and no correlation was detected between the level of dietary fibre intake and risk of gastric cancer.
Dietary fat There is a large amount of preclinical data in animal models to suggest that fat intake has an effect both on tumour incidence and growth.22 These observations led to large retrospective studies that compared varying fat intake with cancer incidence in different communities. Significant correlations were described for many tumour types, particularly those which are hormone-sensitive (breast, prostate and endometrium),23* but also for carcinoma of the colon. Animal fat appeared to be much more important than vegetable fat in these studies. As with all retrospective analyses, several concerns have been raised about the validity of these data, particularly with regard to potential confounding factors and the methods for estimating fat intake. For these reasons, prospective studies with standardized approaches to estimating fat intake are necessary to properly assess the linkage between fat intake and cancer risk. Prospective
The influence of diet on cancer incidence 1435
studies are, however, extremely difficult and expensive to undertake. Nevertheless, several have been instituted and are beginning to provide important information.24** FAT INTAKE AND BREAST CANCER RISK
Diet is thought to play a major role in the aetiology of breast cancer. This is suggested by striking observations of international variations of incidence and changes in risk with migration such that migrants adopt the risk within the host country in one to two generations.5 Dietary practices and cancer risk were compared between several countries by Armstrong and Dole in 1975.6* This study showed a strong correlation between fat consumption and national breast cancer mortality. These results are, however, not supported by many other surveys. In some Chinese communities where fat provides approximately 25 per cent of energy intake,25 breast cancer mortality is only approximately one fifth that in US women who have a similar fat intake. Likewise, fat consumption as a percentage of total energy intake has fallen over the past 30 years in Western nations whereas the incidence of breast cancer has steadily risen. A meta-analysis reported in 1990 combined results from over 10 000 individuals in 12 retrospective studies.26*** The risk of breast cancer was elevated by 35 per cent when total daily fat intake increased by 100 g. Cohort studies have not shown an association between dietary fat intake (as a percentage of total energy) and breast cancer risk.27*,28*,29* A pooled analysis of 337 000 women, reported in 1996,24*** showed no association between breast cancer incidence and daily fat intake between 20 per cent and 45 per cent of total energy. The US Nurses’ Health Study30** included 2956 women with breast cancer that developed over a 14year follow-up period. There was no evidence of a link between rising fat intake and breast cancer risk – indeed there was a weak inverse correlation.31** There is also data from studies that have investigated dietary intake of monounsaturated fats, which show an inverse correlation between intake and breast cancer risk. This appears to be particularly important where the source of mono-unsaturated fat is olive oil. The hypothesis that olive oil has a protective effect is backed up by animal studies.32 FAT AND COLORECTAL CANCER RISK
There is a strong correlation between national intakes of meat and animal fat and the incidence of colorectal cancer.6* Experimental data have shown that bile acids can be converted to carcinogens, and it is known that dietary fat increases bile acid secretion. Several retrospective studies have demonstrated an association between the intake of fat or red meat and the incidence of colorectal cancer. The studies are usually confounded by the fact that high energy intake from fat or meat may lead to obesity and a reduction in exercise, both of which have also been correlated with cancer risk. There is also the problem of recall bias inherent in any retrospective assessment of dietary intake.
Prospective studies have failed to show a consistent correlation between dietary fat intake and colorectal cancer risk. The Nurses’ Health Study33** included almost 90 000 women, among whom 150 cases of colorectal cancer developed. There was an almost two-fold increase in the risk of colorectal cancer among those with a high versus low fat intake. However, further analysis revealed that the level of red meat intake also correlated with colorectal cancer risk and multivariate analysis retained only red meat intake as a risk factor. In a large Dutch study34** that included over 120 000 men and women, the intake of processed meat produced a significant increase in colon cancer risk whereas total fat had no effect. The majority of prospective studies have failed to show a correlation between fat intake and the risk of colorectal cancer,35** although a majority have shown a correlation with red meat consumption.36**,37** It is unclear whether it is the fatty acid or a non-fat component of meat that is the primary factor related to cancer incidence. FAT AND PROSTATE CANCER RISK
The risk of prostate cancer internationally has been linked to dietary fat intake.6*,38* This is only true for animal fat, and similar associations have been seen in several retrospective case-control studies. Five prospective cohort studies have been reported, the largest being a study from the Netherlands39** that included 58 279 individuals who developed 642 cases of prostate cancer. No correlation was seen between prostate cancer risk and the levels of intake of fat. In contrast, the Health Professionals Follow Up Study,36** which included 52 000 men, demonstrated a significant correlation between the level of intake of fat and red meat and prostate cancer risk, an 80 per cent increase being seen in the group with high versus low intake. Two other studies have shown a correlation between saturated fat intake (but not total fat) and prostate cancer risk.40**,41** There is some evidence from these studies that fat intake may be correlated with a more aggressive biological subtype of prostate cancer.
Other dietary factors A variety of constituents of the diet have been implicated as having a role in cancer prevention and prominent among these have been fruits and vegetables.42 Many retrospective analyses have shown inverse relationships between daily fruit and vegetable consumption and cancer risk in a variety of disease sites. As with all such retrospective analyses, interpretation is difficult because of the inaccuracies of the recall of dietary intake by participants. Interpretation is not only affected by such recall bias but by numerous confounding factors including the observation that those with high fruit and vegetable intake will often choose a lifestyle containing factors that may reduce cancer risk, such as being less likely to smoke, taking regular physical exercise
1436 Cancer prevention strategies
and having a lower alcohol consumption. It is thus almost impossible to isolate the impact of fruit and vegetable intake on cancer incidence and outcome. Prospective studies in breast,43** colorectal44** and stomach cancers45** have not shown a clear relationship between fruit and vegetable consumption and cancer incidence or mortality. Although the majority of studies that have examined lung cancer have shown an inverse relationship between intake of fruit and vegetables and lung cancer risk,46** interpretation is confounded by the observation that the likelihood of smoking is also related to fruit and vegetable intake. It has long been postulated that carotenoids, which are found in fruit and vegetables, may be the constituent in these that has a cancer preventive effect.47,48 These agents have beneficial preclinical activity, which may be associated with their ability to promote cell differentiation and exert antioxidation. Intake of β-carotene has been shown in several retrospective studies to be associated inversely with the risk of cancer. This has been especially strong for carcinoma of the bronchus.49* This association led to prospective randomized studies in which β-carotene was administered and compared with control groups. Two of these large randomized prospective studies50**,51** showed an increase in the risk of lung cancer and mortality in the β-carotene supplemented group compared with controls, and the increase in risk appeared to be predominant among smokers. Folic acid is also found in fruits and vegetables and appears to have important roles in the regulation of DNA synthesis and repair; deficiency in animals increases the risk of carcinogenesis.52 Retrospective studies have shown an inverse relationship between folate intake and the risk of colorectal cancer.49* Data from prospective randomized studies has suggested that folic acid supplementation in the diet may indeed reduce the risk of colorectal cancer, in particular for those with a high alcohol intake.53** Studies that have examined selenium and vitamin E supplementation in skin and lung cancer prevention demonstrated reductions in the risk of prostate malignancy in men who were included in these trials.54**,55** These were secondary analyses and are therefore open to criticism. Nevertheless, associations were strong enough to lead to one of the largest cancer prevention intervention trials ever initiated. This trial, the Selenium Vitamin E Chemoprevention Trial (SELECT), was established by the NCI (US) and has almost completed its planned recruitment of 32 400 men.
CHEMOPREVENTION The term ‘chemoprevention’ was derived to define the use of natural or synthetic chemicals that reverse, suppress or prevent the process of carcinogenesis.56,57 More recently it has been appreciated that such approaches may not prevent the development of cancer but cause significant delay.58 Delay of carcinogenesis could have an important beneficial impact on society. The concept of chemoprevention should therefore also incorporate delay as an endpoint.
The process of carcinogenesis has classically been separated into three phases: initiation, promotion and progression. Each is linked with the accumulation of multiple sequential mutations, predominantly leading to DNA damage, which ultimately causes invasive neoplasia.59 The whole process of carcinogenesis may take many years, and there is, therefore, a prolonged period of time during which an effective chemopreventive agent could be administered with the hope that it may interfere with the carcinogenic cascade. This notion is in contrast to treatment of established disease when therapy has to be given rapidly with only a short window of opportunity to have an effect on tumour growth and prolong survival. Chemopreventive agents may block initiation by modifying the access of carcinogens to DNA binding sites, usually through modifying their metabolism. Agents that modify promotion or progression may have an effect through a variety of targets and, because these processes are chronic, are of more potential clinical utility. The field of chemoprevention focuses on two major approaches: the use of drugs with specific oncogenic targets; and phytochemicals, which are often diet-derived. Diet-derived agents, which tend to engage a variety of potential mechanisms of activity, have the potential advantage of a lower risk of toxicity on longterm administration. In contrast, the use of molecularly targeted drugs is based on more precise information on mechanisms and appropriate dosing.
Targets for chemoprevention TARGETING CYCLO-OXYGENASE 2
Mechanisms for the prevention of promotion and progression involve inhibition of events controlled by complex cascades of cellular signal-transduction molecules. Kinases are pivotal in cell signalling and form attractive targets for cancer chemoprevention. Many putative chemopreventive agents have been shown to interact directly with specific enzymes or exert an effect upstream of a kinase. Reactive oxygen species, cell cycle regulatory proteins and nitric oxide are all involved in the regulation of cell proliferation and constitute other potential targets for chemopreventive agents. Cyclo-oxygenase 2 (COX-2) has received considerable attention as a target for chemopreventive agents as it is over-expressed in a number of malignancies.60 Overexpression of COX-2 leads to a number of effects associated with carcinogenicity or pre-neoplasia including the up-regulation of angiogenesis and inhibition of apoptosis. Inhibitors of COX-2 have been shown to stimulate apoptosis in pre-malignant cells and to suppress the growth of capillary blood vessels.61 Non-steroidal anti-inflammatory drugs (NSAIDs) have been investigated for their potential chemopreventive efficacy in several studies over the past 20 years. This is based on their abilities to inhibit COX-2 and to suppress the growth of chemically induced intestinal malignancies in
Chemoprevention 1437
rodents. The autosomal dominant inherited condition, familial adenomatous polyposis (FAP), has been used as a model for colorectal carcinogenesis. In several studies NSAIDs have been investigated in patients with FAP to observe whether a biological effect occurred. Sulindac was found to inhibit the growth of adenomatous polyps and also to induce regression in existing adenomas.62*,63* Intriguingly, in a subsequent study, sulindac administered to young genotypically positive but phenotypically negative FAP patients failed to affect adenoma development, suggesting a narrow time window for successful intervention.64 The selective inhibitor of COX-2 celecoxib was included in a small short-term randomized phase II study and caused, similar to sulindac, significant regression of polyps compared with placebo.65** This result led to the registration of celecoxib as a treatment for FAP. More recent data on the cardiotoxicity of chronic exposure to COX-2 inhibitors has caused considerable concern and the termination of a large number of ongoing trials for the use of COX-2 inhibitors in the prevention of colorectal and other cancers.66 Several epidemiological studies have been reported comparing NSAID intake with the risk of colorectal cancer or adenoma development. Almost all of these studies have demonstrated between 30 per cent and 50 per cent reduction in risk of adenomas or cancer among regular users of NSAIDs.67* Although different studies have reported varying dose levels of NSAIDs, it appears that the duration of exposure may be more important than daily dose. Ten prospective randomized clinical trials have been undertaken with COX inhibitors in patients with FAP or those with previous colorectal cancer or adenomatous polyps. Three studies68**,69**,70** incorporating aspirin showed significant chemopreventive effects on adenoma and colorectal cancer recurrence and, interestingly, in a study reported by Baron et al.69** the best effect was observed in the group treated with the lowest dose of aspirin (81 mg). The majority of these studies were, however, small (less than 100 individuals in all but two). Two of the studies incorporated sulindac and have found no evidence that it suppresses adenoma recurrence.71**,72** Only one randomized trial utilizing aspirin contained sufficient individuals to measure mortality rates from colorectal cancer. This was the Physicians’ Health Study,73 which incorporated 22 071 individuals randomly assigned to either aspirin (325 mg) or placebo every other day for five years. There was no difference between the two groups in the incidence or mortality of colorectal cancer. Unfortunately, the trial was discontinued after only five years of aspirin exposure, which is likely to have been too short to have an impact on mortality. Interpretation is also confounded by the fact that mortality from colorectal cancer was not a primary endpoint for this study. There are therefore compelling preclinical data to suggest that COX-2 may be an important target in terms of reducing adenomatous polyps and delaying carcinogenesis.74 Observational results from epidemiological studies would also suggest that COX inhibitors, in particular
aspirin, may reduce the incidence of colorectal cancer in humans, although such retrospective analyses are always open to criticism. Nevertheless, the consistency of an association between NSAID exposure and reduction in the incidence of colorectal cancer is striking. The prospective studies that have been undertaken to date have suggested that COX-2 inhibition can reduce adenomas in patients with FAP, but so far effects have not been demonstrated on either mortality or conversion to malignancy. Likewise, the majority of studies that have examined adenoma recurrence in non-FAP individuals have shown a reduction in recurrence rates, but all of these studies have been too small or have had too short follow-up to allow demonstration of an alteration in incidence of malignancy or mortality. Until prospective randomized trials confirm these key endpoints, particularly given the cost and potential toxicity75 of long-term NSAID exposure, routine provision of NSAIDs to the general population cannot be recommended at the present time for the chemoprevention of colorectal cancer. ROLE OF NON-STEROIDAL ANTI-INFLAMMATORY DRUGS IN MALIGNANCIES OTHER THAN COLORECTAL CANCER
There is a large body of preclinical data suggesting that a variety of tumours including those arising in lung, prostate, stomach, oesophagus and bladder may be inhibited by NSAIDs. Numerous epidemiological studies have correlated NSAID intake with incidence of a variety of malignancies. The most compelling data relates to the incidence of oesophageal and stomach cancer with all but one study in each disease showing an inverse relationship between NSAID intake and incidence or death rates76*,77* A meta-analysis has been performed for oesophageal cancer78*** showing a relative risk of 0.57 in terms of the incidence of oesophageal malignancy and regular NSAID use. In another meta-analysis,79*** the relative risk of gastric cancer was 0.78 for regular NSAID use. There is a suggestion for a small reduction in risk of incidence and mortality in breast80*,81* and ovarian82* cancers, although large prospective studies have not confirmed these findings. Case-control and cohort studies have shown conflicting data for lung, urinary tract and prostate cancers. Five studies examined the relationship between NSAID intake and incidence or mortality from all malignancies combined.76*,77*,83*,84*,85* Only one of these suggested a reduced risk from NSAID use, and in two studies there was actually an increased risk of cancer among individuals receiving one or more prescriptions for NSAIDs. HORMONES AS TARGETS FOR CHEMOPREVENTION
By far the most advanced area of chemoprevention involves the use of hormone manipulation for the prevention of breast cancer. Tamoxifen studies have shown that this agent can reduce breast cancer incidence in healthy women who have been determined to be at increased risk,
1438 Cancer prevention strategies
those with ductal carcinoma in situ, and patients who have previously been treated for breast cancer.86** A recent meta-analysis87*** that included 28 406 healthy subjects in the four major primary prevention trials of tamoxifen showed a 38 per cent reduction of breast cancer incidence. This reduction was at the cost of a 2.4-fold increase in endometrial cancer and a 1.9-fold increase in venous thromboembolic events. Ongoing approaches are utilizing lower doses of tamoxifen (down to 5 mg per day in a current Italian study) in an attempt to reduce side effects. There is also evidence that the combination of hormone replacement therapy with tamoxifen88** leads to a reduction in venous thrombosis and endometrial cancer without having a negative impact on breast cancer prevention. Indeed, breast cancer risk appears to be further reduced by this combination. The related agent, raloxifene, was incorporated in the Multiple Outcome Raloxifene Endpoint (MORE) study89** in 7705 women with osteoporosis. There was a 62 per cent reduction in invasive breast cancer in post-menopausal women and, encouragingly, there was no increased incidence of endometrial cancer. Unfortunately, venous thromboembolic events were increased. There is an ongoing randomized study comparing raloxifene with tamoxifen as primary prevention in post-menopausal highrisk women and this study will include 19 000 patients. A further large international randomized trial (IBIS II) was opened in 2004 and will investigate the efficacy of the third generation aromatase inhibitor, anastrozole. Success has also been seen with hormone manipulation in the prevention of prostate cancer. The 5-α-reductase inhibitor, finasteride, was given for 7 years in the Prostate Cancer Prevention Trial (PCPT) sponsored by the NCI (US),90** and 19 882 healthy volunteers aged over 55 years were entered. The primary endpoint was the prevalence of biopsy-detected prostate cancer and a highly significant reduction of 24.8 per cent was seen in the finasteride treatment arm. Of some concern was the fact that in the finasteride treated group, there was a significantly higher proportion of high-grade tumours and sexual side effects were also more problematic. A recent analysis has estimated that widespread adoption of finasteride in US men over the age of 55 years would save 316 760 person-years.91 It is clearly important to consider finasteride usage for men who are anxious about the development of prostate cancer after due counselling about the potential risks of the development of high-grade lesions and the side effects which can occur.
Diet-derived agents for chemoprevention There is extensive epidemiological data to suggest that dietary factors may play a role in both the causes and prevention of cancers. Several studies have identified dietary constituents that may be chemopreventive. Such observations must be interpreted with great caution as no single constituent of the diet is taken in isolation, and effects may
easily be confounded by associated dietary factors or lifestyle choices. Nevertheless, such epidemiological observations have led to the isolation of several dietary constituents and their examination in preclinical systems for potential mechanisms which may be associated with chemoprevention. An example of such an agent is curcumin, a major constituent of the spice turmeric, which has been associated with the potential reduction in risk of colorectal cancer. It has been shown to inhibit the activities of protein kinase C, epidermal growth factor receptor kinase, mitogen-activated protein kinases, and c-JUN terminal kinase.92 It also modifies cell cycle regulatory proteins, reduces angiogenesis and promotes apoptosis. Curcumin inhibits COX-293 and has been shown to reduce colon polyp formation in transgenic models.94 Curcumin has already entered clinical trials95 as have several other dietary constituents (including tea-catechins, the soy constituent genistein and resveratrol, a component of red grapes) that have been shown to undermine oncogenic mechanisms in preclinical models.
Chemoprevention trial design The design of clinical chemoprevention trials continues to evolve and has, until recently, been largely modelled on the methods of drug development used for cytotoxic agents. In phase I studies, the dose-related safety of agents is determined and focuses on toxicity and pharmacokinetics. A major concern with such studies is the choice of a dose to recommend for further testing. Given the lack of toxicity of such agents, a variety of biomarkers have been incorporated in several phase I studies, and endpoints have been chosen depending on modulation of these. The area of biomarker development remains a huge challenge in this field. In phase II trials, a randomized, placebo-control design is used to evaluate dose response and common toxicities that are likely to result from long-term administration. In some instances, a variety of dose levels are evaluated in these trials and biomarkers are incorporated. Such studies are often undertaken in patients with premalignant lesions or those cured of an initial cancer but who are at risk of developing a second primary tumour. If safety and efficacy are judged to be promising in these trials, the agent proceeds to randomized, prospective, large-scale phase III clinical trials. These are the ultimate tests of drug efficacy, measuring the incidence of new tumours as well as surrogate biomarkers, in relation to compliance and toxicity. Phase III trials are often very large, of long duration and of high cost.
ROLE OF SURGERY It is now possible to identify individuals with a variety of conditions or genetic predisposition who have an increased risk of developing malignancies. Surgery can be an effective means of preventing the development of cancer in a
Conclusions 1439
proportion of these patients by removing the target organ. For many years, the prime example of the value of surgery in preventing cancer followed the observation that the risk of malignancy in a testis was increased by 20-fold if it was undescended.96 Correction of an undescended testicle reduces this risk.97* Although correction alone may not always prevent malignant transformation, this allows the testis to be regularly examined for change. Orchidectomy is recommended for a non-palpable undescended testicle in the post-pubertal population.98 Thyroid malignancies can arise in the inherited conditions multiple endocrine neoplasia type 2 and familial medullary thyroid carcinoma. Mutations in the RET proto-oncogene have been found to play a key role in the development of medullary thyroid cancer and are associated with a high risk which justifies prophylactic thyroidectomy.99 When surgery is performed in a specialist centre shortly after identification of a RET proto-oncogene mutation, morbidity from the procedure is low, and cure rates have been reported to be 100 per cent.100 Surgery plays an important role in the prevention of colorectal cancer as risk factors for this disease are increasingly identified. Ulcerative colitis has long been associated with a predisposition to this disease and the risk rises with the duration of disease and the extent of involvement of the colon.101 Some studies have suggested that patients who have pancolitis for more than 20 years will have a risk of developing colorectal cancer of up to 35 per cent.102 Regular surveillance of such individuals is recommended with endoscopic biopsies being taken. The presence of dysplasia (particularly high grade) is associated with an increased risk of subsequent development of invasive malignancy. Prophylactic panproctocolectomy is undertaken is most centres, but no widespread agreement exists as to the optimal timing. Most groups would agree, however, that the risk of developing malignancy after 20 years of disease is sufficiently high to recommend this procedure even if no dysplasia is evident. Many centres would undertake this after 10 years of active disease, and the threshold for intervention will be lower if dysplastic lesions are identified. The risk of colorectal cancer appears to be eliminated by prophylactic surgery so long as no focus of malignancy is detected in the resected specimen.103* Familial adenomatous polyposis is one of the bestdescribed hereditary malignancies that is transmitted as an autosomal dominant condition. Almost 100 per cent of affected individuals will develop colon cancer by the age of 40.104 Prophylactic colectomy is advised during the teenage years. Unfortunately there is a high incidence of upper gastrointestinal tract polyposis and malignancy in patients who have undergone colorectal resection and regular upper gastrointestinal (GI) endoscopy must be undertaken.105,106 Hereditary non-polyposis colorectal cancer (HNPCC) has increasingly been recognized and accounts for approximately 5 per cent of all colorectal cancers107 It is inherited as an autosomal dominant condition and arises from germ-line mutations of DNA mismatch repair genes.
Gene carriers have a 90 per cent risk of developing malignancy. There appears to be a marked reduction in the risk if a regular screening programme is instituted, and it is recommended that this begin at the age of 20 years in identified individuals.108 Subtotal colectomy, retaining the rectum, is effective to reduce the risk of large bowel malignancy, although patients remain at risk of developing subsequent rectal cancer and other malignancies.109 Prophylactic mastectomy is increasingly considered for the prevention of breast cancer in patients found to be at high risk for this disease. BRCA1 and BRCA2 gene mutations are associated with up to an 85 per cent lifetime risk of developing breast cancer110 and options for reducing this risk include surveillance, the use of tamoxifen as chemoprevention and bilateral mastectomies. Unfortunately the effectiveness of intensive screening in this population has not been proven by prospective studies and the NSABP breast cancer prevention trial showed a benefit for tamoxifen in healthy BRCA2 carriers but no benefit for women with BRCA1 mutations.111 Prophylactic mastectomy is often used, but major concerns relate to the occurrence of breast cancers even after this procedure has been undertaken. This results from the fact that breast tissue can be found in areas of chest wall, axilla and abdomen, which are outside the usual surgical resection fields.112 No prospective randomized trials comparing prophylactic mastectomy with observation have been performed in BRCA1 and BRCA2 carriers but retrospective analysis from the Mayo Clinic113* showed a 90 per cent reduction in the risk of breast cancer following bilateral prophylactic mastectomy in women with a family history of breast cancer. At the current time, such surgery should only be offered after careful discussion with affected women who must balance the uncertain benefit of such surgery with the potential psychological morbidity. Abnormalities of the BRCA1 and BRCA2 genes are also associated with an elevated lifetime risk of developing ovarian carcinoma, and there is evidence that bilateral prophylactic oophorectomy reduces this risk. A retrospective analysis of women who had undergone this procedure compared with matched controls revealed a significant reduction in ovarian cancer incidence,114* and this is therefore a recommended procedure for women who carry either of these genes and do not wish to retain fertility.
CONCLUSIONS With increasing knowledge about the causes of cancer and identification of those at increased risk of developing malignancy, the potential to prevent or delay its onset is of huge importance. Surgery has a role in reducing the risk of breast, colon and thyroid tumours in small defined populations, and hormone manipulation has an impact on breast and prostate cancer incidence. Although diet is believed to be involved in the aetiology of many tumours, modifications to isolated factors in dietary intake have not
1440 Cancer prevention strategies
yet been definitively proven to alter cancer risk, although strong evidence points to the potential value of reducing fat (particularly saturated), increasing fibre and possibly increasing folate intake. Modifying diets is notoriously difficult and an approach of potential importance is the use of diet-derived agents that can be given on a long-term basis as ‘nutraceuticals’ in chemoprevention programmes. To date, no such interventions have been proven to reduce cancer risk, but several trials are ongoing in this relatively new field of research. It is hoped that the area of chemoprevention will develop to be able to offer possibilities of intervention for those increasing number of individuals who are found to have a genetic predisposition to developing malignancies. In parallel with these approaches, it is clearly essential to continue reducing exposure to cigarette smoke, the most important single identified cause of cancer. It has been estimated that tobacco causes approximately 30 per cent of all cancers, and reducing its use remains the most promising strategy of cancer prevention.
◆4
●5
●6
●7
8
9
10
KEY LEARNING POINTS ●
●
●
●
●
Approximately 60 per cent of all cancers are believed to be caused by lifestyle choices – predominantly diet and cigarette smoking – and are therefore potentially preventable. Surgery has a role in prevention of some forms of cancers in high-risk individuals – particularly colorectal cancer in those with familial adenomatous polyposis (FAP), breast cancer in carriers of susceptibility genes and testicular tumours in those with undescended testes. Chemoprevention involves the use of dietderived or pharmacological agents to prevent or delay carcinogenesis. Chemoprevention has been proven to be effective with the use of tamoxifen in breast cancer and finasteride in prostate cancer. The reduction of fat, and increase of fibre and folate in the diet appear to reduce the risk of malignancy – particularly of the colon.
REFERENCES Doll R, Peto R. The Causes of Cancer: quantitative estimates of avoidable risks of cancer in the United States today. J Natl Cancer Inst 1981; 66:1191–308. ●2 Peto R, Lopez AD, Boreham J, et al. Mortality from tobacco in developed countries: indirect estimation from national vital statistics. Lancet 1996; 39:1268–78. ●3 Doll R, Hill AB. Mortality in relation to smoking: ten years’ observation on British Doctors. BMJ 1964; 1:1399–410.
11 ●12
●13
●14
15
●16
●17
◆18
●1
◆19
20
Cincirpini PM, McClure JB. Smoking Cessation: recent developments in behavioral and pharmacologic interventions. Oncology 1998; 12:249–59. Ziegler RG, Hoover RN, Pike MC, et al. Migration patterns and breast cancer risk in Asian-American women. J Natl Cancer Inst 1993; 85:1819–27. Armstrong B, Doll R. Environmental factors and cancer incidence and mortality in different countries, with special reference to dietary practices. Int J Cancer 1975; 15:617–31. Buell P. Changing incidence of breast cancer in Japanese-American women. J Natl Cancer Inst 1973; 51:1479–83. Englyst HN, Bingham SA, Runswick SA, et al. Dietary fibre (non-starch polysaccharides) in fruit, vegetables and nuts. J Hum Nutr Diet 1988; 1:247–86. Bingham S. Mechanisms and experimental evidence relating to dietary fibre and starch to protection against large bowel cancer. Proc Nutr Soc 1990; 49:153–171. Boffa LC, Luption JR, Mariani MR et al. Modulation of colonic cell proliferation, histone acetylation and luminal short chain fatty acids by variation of dietary fibre (wheat bran) in rats. Cancer Res 1992; 52:5906–5912. Burkitt DP. Related disease-related cause. Lancet 1969; 2:1229–31. Howe GR, Benito E, Castelleto R, et al. Dietary intake of fiber and decreased risk of cancers of the colon and rectum: evidence from the combined analysis of 13 case-control studies J Natl Cancer Inst 1992; 84:1887–96. Kato I, Akhmedkhanov A, Koenig K, et al. Prospective study of diet and female colorectal cancer: The New York University Women’s Health Study. Nutr Cancer 1997; 28:276–81. Fuchs CS, Giovannucci EL, Colditz GA, et al. Dietary fiber and the risk of colorectal cancer and adenoma in women. N Engl J Med 1999; 340:169–76. Terry P, Giovannucci E, Michels KB et al. Fruit, vegetables, dietary fiber, and risk of colorectal cancer. J Natl Cancer Inst 2001a; 93:525–33. Bingham SA, Day NE, Luben R, et al. Dietary fibre in food and protection against colorectal cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC): an observational study. Lancet 2003; 361:1496–501. Michels KB, Fuchs CS, Giovannucci E, et al. Fiber intake and incidence of colorectal cancer among 76,947 women and 47,279 men. Cancer Epidemiol Biomarkers Prev 2005; 14:842–9. COMA Working Group on Diet and Cancer. Nutritional aspects of the development of cancer. UK Department of Health Report on Health and Social Subjects 1998 No 48. London: HMSO. Graham S, Marshall J, Mettlin C, et al. Diet in the epidemiology of breast cancer. Am J Epidemiol. 1982; 116:68–75 Rohan TE, Howe GR Friedenreich CM, et al. Dietry fiber, vitamins A, C & E, and risk of breast cancer: a cohort study. Cancer Causes Control 1993; 4:29–37.
References 1441
●21
22
23
●24
25
●26
●27
28
29
30
●31
32
●33
●34
●35
36
37
Bolterweck AA, van den Bradt PA, Goldbohm RA. Vitamins, carotenoids dietary fiber, and the risk of gastric cancer: results from a prospective study after 6.3 years of followup. Cancer 2000; 88:737–48. Welsch CW, Relationship between dietary fats and experimental mammary tumorigenesis: a review and critique. Cancer Res 1992; 52[Suppl]:2040S–2048S. Prentice RL, Sheppard L. Dietary fat and cancer: consistency of the epidemiological data and disease prevention that may follow from a practical reduction in fat consumption. Cancer Causes Control 1990; 1:81. Hunter DJ, Spiegelman D, Adami HO, et al. Cohort studies of fat intake and the risk of breast cancer: a pooled analysis. N Engl J Med 1996; 334:356–361. Chen J, Campbell TC, Junyao L, et al. Diet life-style and mortality in China – a study of the characteristics of 65 Chinese counties. Oxford, England: Oxford University Press, 1990. Howe GR, Hirohata T, Hislop TG et al. Dietary factors and risk of breast cancer: combined analysis of 12 case-control studies. J Natl Cancer Inst 1990; 82:561–9. Willett WC, Hunter DJ, Stampfer MJ, et al. Dietary fat and fiber in relation to risk of breast cancer: an 8-year followup. JAMA 1992; 268:2037–44. Giovannucci E, Stampfer MJ, Colditz GA, et al. A comparison of prospective and retrospective assessments of diet in the study of breast cancer. Am J Epidemiol 1993; 137:502–11. Smith-Warner SA, Spiegelman D, Adami HO, et al. Types of dietary fat and breast cancer: a pooled analysis of cohort studies. Int J Cancer 2001; 92:767–74. Holmes MD, Hunter DJ, Colditz GA, et al. Association of dietary intake of fat and fatty acids with risk of breast cancer. JAMA 1999; 281:914–20. Wolk A, Bergstrom R, Hunter D, et al. A prospective study of association of mono-unsaturated and other types of fat with risk of breast cancer. Arch Intern Med 1998; 158:41–45. Martin-Moreno JM, Willett WC, Gorgojo L, et al. Dietary fat, olive oil intake and breast cancer risk. Int J Cancer 1994; 58:774–80. Willett WC, Stampfer MJ, Colditz GA, et al. Relation of meat, fat and fiber intake to the risk of colon cancer in a prospective study among women. N Engl J Med; 1990; 323:1664–1672. Goldbohm RA, van den Brandt PA, van’t Veer P, et al. A prospective cohort study on the relation between meat consumption and the risk of colon cancer. Cancer Res 1994; 54:718–23. Thun MJ, Calle EE, Namboodiri MM, et al. Risk factors for fatal colon cancer in a large prospective study. J Natl Cancer Inst 1992; 84:1491–1500. Giovannucci E, Rimm EB, Stampfer MJ, et al. Intake of fat, meat and fiber in relation to risk of colon cancer in men. Cancer Res 1994; 54:2390–7. Norat T, Lukanova A, Ferrari P et al. Meat consumption and colorectal cancer risk: dose-response meta-analysis of epidemiological studies. Int J Cancer 2002; 98 :241–56.
◆38
39
●40
41
42
●43
●44
45
●46
◆47
◆48
◆49
●50
●51
◆52
53
●54
Graham S, Haughey B, Marshall J, et al. Diet in the epidemiology of carcinoma of the prostate gland. J Natl Cancer Inst 1983: 70:687–692. Schuurman AG, van den Brandt PA, Dorant E, et al. Association of energy and fat intake with prostate carcinoma risk. Cancer 1999; 86:1019–27. Mills PK, Beeson WL, Phillips RL et al. Cohort study of diet, lifestyle and prostate cancer in Adventist men. Cancer 1989; 64:598–604. Le Marchand L, Kolonel LN, Wilkens LR, et al. Animal fat consumption and prostate cancer: a prospective study in Hawaii. Epidemiology 1994; 5:276–82. Dragsted LO, Strube M, Larsen JC. Cancer protective factors in fruits and vegetables: biochemical and biological background. Pharmacol Toxicol 1993; 72[Suppl]:s116–35. Smith-Warner SA, Spiegelman D, Yaun SS, et al. Intake of fruit and vegetables and the risk of breast cancer: a pooled analysis of cohort studies. JAMA 2001; 285:769–76. Michels KB, Giovannucci E, Joshipura KJ, et al. A prospective study of fruit and vegetable consumption and colorectal cancer incidence. J Natl Cancer Inst 2000; 92: 1740–52. Kneller RW, McLaughlin JK, Bjelke E, et al. A cohort study of stomach cancer in a high-risk American population. Cancer 1991; 68:672–8. Feskanich D, Ziegler RG, Michaud DS, et al. Prospective study of fruit and vegetable consumption and risk of lung cancer among men and women. J Natl Cancer Inst 2000; 92: 1812–23. Peto R, Doll R, Buckley JD, et al. Can dietary beta carotene materially reduce human cancer rates: Nature 1981; 290:201–8. Lippman SM, Kessler JF, Meyskens FL Jr. Retinoids as preventive and therapeutic anticancer agents. Cancer Treat Rep 1987; 71:493–515. World Cancer Research Fund and American Institute for Cancer Research. Food, nutrition and prevention of cancer: a global perspective. Washington: American Institute for Cancer Research. 1997; Wiseman MJ (ed.), IARC Sci Publ. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med 1994; 330:1029–35. Omenn GS, Goodman GE, Thornquist MD, et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. N Engl J Med 1996; 334:1150–5. Giovannucci E. Epidemiological Studies of folate and colorectal neoplasias: a review. J Nutr 2002; 132:2350–5. Kim DH, Smith-Warner SA, Hunter DJ, et al. Pooled analysis of prospective cohort studies on folate and colorectal cancer. Pooling Project of Diet and Cancer Investigations. Am J Epidemiol 2001; [Suppl 153]:S118(abstr). Clark LC, Combs GF Jr, Turnball BW, et al. Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin. A randomised controlled trial. JAMA 1996; 276:1957–63.
1442 Cancer prevention strategies
●55
◆56
57
58 ◆59
◆60
61
62
●63
64
●65
66
67 ●68
●69
●70
71
72
Heinonen OP, Albanes D, Virtamo J, et al. Prostate cancer and supplementation with alpha-tocopherol and betacarotene: incidence and mortality in a controlled trial. J Natl Cancer Inst 1998; 90:440–6. Wattenberg LW. Chemoprophylaxis of carcinogenesis: a review. Cancer Res 1966; 26:1520–26. Sporn MB, Dunlop NM, Newton DL, Smith JM. Prevention of chemical carcinogenesis by vitamin A and its synthetic analogs (retinoids). Fed Proc 1976; 35:1332-8. Lippman SM, Hong WK. Commentary – Cancer prevention by delay. Clin Cancer Res 2002; 8:305–13. Vogelstein B, Fearon ER, Hamilton SR, et al. Genetic alterations during colorectal tumor development. N Engl J Med 1988; 319:525–32. Thun M, Henley S, Patrono C. Nonsteroidal antiinflammatory drugs as anticancer agents: mechanistic, pharmacologic, and clinical issues. J Natl Cancer Inst 2002; 94:252–66. Masferrer J, Leahy K, Koki A, et al. Antiangiogenic and antitumor activities of cyclooxygenase-2 inhibitors. Cancer Res 2000; 60:1306–11. Labayle D, Fischer D, Vielh P, et al. Sulindac causes regression of rectal polyps in familial adenomatous polyposis. Gastroenterology 1991; 101:635–9. Nugent K, Farmer K, Spigelman A, et al. Randomized controlled trial of the effect of sulindac on duodenal and rectal polyposis and cell proliferation in patients with familial adenomatous polyposis Br J Surg 1993; 80:1618–19. Giardiello FM, Yang VW, Hylind LM, et al. Primary chemoprevention of familial adenomatous polyposis with sulindac. N Engl J Med 2002; 346:1054–9. Steinbach G, Lynch P, Phillips R, et al. The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. N Engl J Med 2000; 342: 1946–52. Vanchieri C. Researchers Plan to Continue to Study COX-2 inhibitors in cancer treatment and prevention. J Natl Cancer InstI 2005; 97:552–3. Thun M, Namboodiri M, Heath CJ. Aspirin use and reduced risk of fatal colon cancer. N Engl J Med 1991; 325:1593–6. Sandler RS, Halabi S, Baron JA, et al. A randomized trial of aspirin to prevent colorectal adenomas in patients with previous colorectal cancer. N Engl J Med 2003; 348:883–90. Baron JA, Cole BF, Sandler RS, et al. A randomized trial of aspirin to prevent colorectal adenomas N Engl J Med 2003; 348:891–9. Giardiello FM, Yang VW, Hylind LM, et al. Primary chemoprevention of familial adenomatous polyposis with sulindac. N Engl J Med 2002; 346:1054–9. Ladenheim J, Garcia G, Titzer D, et al. Effect of sulindac on sporadic colonic polyps. Gastroenterology 1995; 108:1083–7. Sample D, Wargovich M, Fischer SM, et al. A dose-finding study of aspirin for chemoprevention utilizing rectal mucosal prostaglandin E(2) levels as a biomarker. Cancer Epidemiol Biomarkers Prev 2002; 11:275–9.
●73
74 75
76
77 78
●79
80
●81
82
83
●84
85
●86
●87
88
●89
Gann P, Manson J Glynn R, et al. Low-dose aspirin and incidence of colorectal tumors in a randomised trial. J Natl Cancer Inst 1993; 85:1220–4. Prescott SM, Fitzpatrick FA. Cyclooxygenase-2 and carcinogenesis. Biochim Biophys Acta 2000: 1470:M69–78 Wald NJ, Law MR. A strategy to reduce cardiovascular disease by more than 80%. Brit Med J 2003; 326: 1419–24. Friis S, Sorensen HT, McLaughlin JK, et al. A populationbased cohort study of the risk of colorectal and other cancers among users of low-dose aspirin Br J Cancer 2003; 88:684–8. Thun M, Namboodiri M, Calle E, et al. Aspirin use and risk of fatal cancer. Cancer Res 1993; 53:1322–7. Corley DA, Kerlikowski K, Verma R, et al. Protective association of aspirin/NSAIDs and esophageal cancer: a systematic review and meta-analysis. Gastroenterology 2003; 124:47–56. Wang W, Huang J, Zheng G, et al. Non-steroidal antiinflammatory drug use and the risk of gastric cancer: a systematic review and metaanalysis J Natl Cancer Inst 2003; 95:1784–91. Johnson TW, Anderson KE, Lazovich D, et al. Association of aspirin and nonsteroidal anti-inflammatory drug use with breast cancer. Cancer Epidemiol Biomarkers Prev 2002; 11:1586–91. Khuder S, Mutgi A. Breast Cancer and NSAID use: a metaanalysis Br J Cancer 2001; 84:1188–92. Akhmedkhanov A, Toniolo P, Zeleniuch-Jacquotte A, et al. Aspirin and epithelial ovarian cancer. Preventive Medicine 2001; 33:682–687. Sorensen HT, Friis S, Norgard B, et al. Risk of cancer in a large cohort of non-aspirin NSAID users: a population-based study. Br J Cancer 2003; 88:1687–92. Schreinemachers D, Everson R. Aspirin use and lung, colon and breast cancer incidence in a prospective study. Epidemiology 1994; 5:138–146. Langman MJS, Cheng KK, Gilman EA, et al. Effect of antiinflammatory drugs on overall risk of common cancer: casecontrol study in general practice database. BMJ 2000; 320:1642–6. Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 study. J Natl Cancer Inst 1998; 90:1371–1388. Cuzick J, Powles T, Veronesi U, et al. Overview of the main outcomes in breast-cancer prevention trials. Lancet 2003; 360:817–824. Decensi A, Robertson C, Rotmensz N, et al. Effect of tamoxifen and transdermal hormone replacement therapy on cardiovascular risk factors in a prevention trial. Italian Chemoprevention Group. Br J Cancer 1998; 78:572–8. Cummings SR, Eckert S, Kruegar KA, et al. The effect of raloxifene on risk of breast cancer in postmenopausal women: results from the MORE randomised trial. Multiple Outcomes of Raloxifene Evaluation. JAMA 1999; 281:2189–2197.
References 1443
●90
91
◆92
93
94
95
96 97
98
99
◆100
101
●102
Thompson IM, Goodman PJ, Tangen CM, et al. The influence of finasteride on the development of prostate cancer. N Engl J Med 2003; 349:215–24. Unger JM, Thompson IM, Le Blanc MC, et al. Estimated impact of the prostate prevention trial on population mortality. Cancer 2005; 103:1375–80. Manson MM. Inhibition of survival signalling by dietary polyphenols and indole-3-carbinol. Eur J Cancer 2005; 41:1842–53. Aggarwal BB, Kumar A, Bharti AC. Anticancer potential of curcumin in preclinical and clinical studies. Anticancer Res 2003; 23:363–98. Perkins S, Verschoyle RD, Hill KA, et al. Chemopreventive efficacy and pharmacokinetics of curcumin in the Min/+ mouse, a model of familial adenomatous polyposis. Cancer Epidem Biomar Prevent 2002; 11:535–40. Sharma RA, Euden S, Platton SL, et al, Phase I clinical trial of oral curcumin: biomarkers of systemic activity and compliance. Clin Cancer Res 2004; 10:6847–54. Gill B, Kogan S. Cryptorchidism: current concepts. Pediatr Clin North Am 1997; 44:1211–27. Pike MC, Chiluers C, Peckham MHJ. Effects of age at orchiopexy on risk of testicular cancer. Lancet 1986; 1:1246–8. Farrer JH, Walker AH, Rajfer J. Management of the postpubertal cryptorchid testis: a statistical review. J Urol 1985; 134:1071–6. Wells SA Jr, Chi DD, Toshima K, et al. Predictive DNA testing and prophylactic thyroidectomy in patients at risk for multiple endocrine neoplasia type 2A. Ann Surg 1994; 220:237–47. Brandi ML, Gagel RF, Angeli A, et al. Consensus: guidelines for diagnosis and therapy of MEN type 1 and type 2. J Clin Endocrinol Metab 2001; 86:5658–71. Lennard-Jones JE, Melville DM, Moison BC, et al. Precancer and cancer in extensive colitis findings among 401 patients over 22 years. Gut 1990; 31:800–6. Ekbom A, Helmick C, Zack M, et al. Ulcerative colitis and colorectal cancer: a population-based study. N Engl J Med 1990; 323:1228–33.
103 Lennard-Jones JE. Is colonoscopic cancer surveillance in ulcerative colitis essential for every patient? Eur J Cancer 1995; 31A:1178–82. 104 Peterson GM, Brensinger JD, Johnson KA, et al. Genetic testing and counseling for hereditary form of colorectal cancer. Cancer 1999; 86[Suppl]:2540–50. 105 Lindor NM, Greene MH. The cancer handbook of family cancer syndromes. Mayo familial cancer program. J Natl Cancer Inst 1998; 90:1039–71. 106 Spigelman Ad, Williams CB, Talbot IC, et al. Upper gastrointestinal cancer in patients with familial adenomatous polyposis. Lancet 1989; 2(8666): 783–5. ◆107 Lynch HT, Smyrk TC, Watson P, et al. Genetics, natural history, tumor spectrum, and pathology of hereditary nonpolyposis colorectal cancer: an updated review. Gastroenterology 1993; 104:1535–49. ●108 Jarvinen HJ, Mecklin JP, Sistonen P. Screening reduces colorectal cancer in hereditary nonpolyposis colorectal cancer families. Gastroenterology 1995; 108:1405–11. 109 Rodriguez-Bigas MA. Prophylactic colectomy for gene carriers in hereditary nonpolyposis colorectal cancer. Has the time come? Cancer 1996; 78:199–201. 110 Struweing JP, Hartge P, Waholder S, et al. The risk of cancer associated with specific mutation of BRCA1 and BRCA2 among Ashkenazic Jews. N Engl J Med 1997; 336:1401–8. ●111 King MC, Wieand S, Hale K, et al. Tamoxifen and breast cancer incidence among women with inherited mutations in BRCA-1 and BRCA-2. JAMA 2001; 286:2251–6. ●112 Eldar S, Meguid MM, Beatty JD. Cancer of the breast after prophylactic subcutaneous mastectomy. Am J Surg 1984; 148:692–3. ●113 Hartmann LC, Schaid DJ, Woods JE, et al. Efficacy of bilateral prophylactic mastectomy in women with a family history of breast cancer. N Engl J Med 1999; 340:77–84. ●114 Reubbeck TR, Lynch HT, Neuhausen SL, et al. Prophylactic oophorectomy in carriers of BRCA-1 and BRCA-2 mutations. N Engl J Med 2002; 346:1616–22.
This page intentionally left blank
Index Note: The alphabetical order is letter by letter. Page references in bold refer to tables or boxes; page numbers in italic refer to figures
17-AAG 894 abdominal radiotherapy, immobilization 1237–8, 1237 abdomino-perineal resection 675 ABO blood antigen 622 abortions, mistaken for hydatidiform mole 811 abscess, percutaneous drainage 151 ABT-737 95 ABVD regimen 1032, 1033–4, 1033, 1034, 1036, 1038 ABV regimen 1032, 1033, 1033 acanthosis nigricans 507, 1203 access to healthcare 20, 1359 ACE regimen 522 achalasia, oesophageal cancer risk 539 ACNU 300 acromegaly 443–5 ACTH, see adrenocorticotrophic hormone ACTHoma 451 actinic keratosis 836, 839–40 actinomycin D 87 activity 87 Ewing’s tumour 1131, 1131–2 gestational trophoblastic tumours 823–4, 824, 825 non-melanoma skin cancer 852 primary bone tumours 915, 918 rhabdomyosarcoma 326–7 soft tissue sarcoma 954 toxicity 87 Active Breathing Coordinator 1273 activism 1362 acute lymphoblastic leukaemia (ALL) classifications 991–2, 991 clinical features 991 CNS relapse 993–4 epidemiology 990–1 molecular pathogenesis 992–3 ocular metastases 329 paediatric 1121–3 incidence 1119 infant 1123 mature B-cell 1123 prognostic factors 992, 993 treatment general principles 993 Ph-positive disease 994–5 post-remission 994
relapsed disease 995 remission induction 993–4 stem cell transplantation 995, 1101 acute myeloid leukaemia 1 (AML1) gene 984 acute myeloid leukaemia (AML) 981–2 blood films 982, 982, 983, 983 classifications 982–4, 983, 1123 clinical features 982 epidemiology and aetiology 982 molecular pathogenesis 984, 985 ocular metastases 329 paediatric 1123–4 prognostic factors 983, 984–5 risk after chemotherapy, GTD 828 second primary neoplasms 930 stem cell transplantation 1101 therapy-related 982 treatment acute promyelocytic leukaemia 988–9 general principles 985–6 investigational 987–8 patient older than 60 years 986–7 patient under 60 years 986 relapse 990 stem cell transplantation 989–90 acute promyelocytic leukaemia (APL) 103, 984 treatment 988–9 ACVBP regimen 1050, 1052 Addison’s disease, pigmentation 1204 adeno-associated viruses (AAVs), as viral vectors 270, 272 adenoma colorectal 668, 673 MEN-1 syndrome 456 parathyroid 430, 449–50 pituitary 443 adenomatous polyposis coli (APC) 963 adenoviruses genome structure 270, 271, 272–3 oncolytic 272–3 vaccines 177 as viral vectors 270–1 ADEPT (antibody-directed enzyme prodrug therapy) 229, 686 adolescent cancers 1118 acute lymphoblastic leukaemia 1123
Ewing’s tumour 1130–3 adoptive cell therapy 178, 194–8, 894–5, 1108 cell expansion 196–7 cell selection 197 cell survival 197 EBV-specific T cells 195 genetically-modified cells 196, 197 lymphokine-activated killer (LAK) cells 195 malignant melanoma 894–5 strategy 196, 197 tumour-infiltrating lymphocytes (TILs) 195–6 adrenalectomy, bilateral 446–7 adrenal gland metastases 506 adrenal hyperplasia 456 adrenocortical tumours 457–8 adrenocorticotrophic hormone (ACTH) 439, 446, 1201 adriamycin bladder cancer 624, 630, 630, 632 gastric cancer 611, 612 multiple myeloma 1083, 1084, 1085 soft tissue sarcoma 954, 956–7, 961 uterine sarcoma 772 Adult Co-Morbidity Evaluation-27 (ACE-27) 1418 adult T-cell leukaemia/lymphoma (ATLL) 863, 1010, 1055–6 advance directives 1299–300 advice, giving 1307 aflatoxin 560 AFP, see alpha-fetoprotein AFR gene 851 Africa head and neck tumours 388, 389 hepatocellular carcinoma 559, 559, 562 life expectancy 5, 5 see also named countries African-Americans 573, 731, 906, 1074 African races 906 Afro-Caribbeans 1073 AG-013736 894 age at diagnosis of cancer 15 brain tumours 288 colorectal cancer 666–7, 667 and communication 1308
1446 Index
age (contd.) Hodgkin’s lymphoma 1028 and multiple myeloma 1073 ovarian cancer 748 and prostate cancer 645 and renal cell cancer 729, 729, 730 ageing populations 20 agitation, in terminal disease 1302 AIDS Clinical Trials Group, staging of Kaposi’s sarcoma 1170, 1170 AIDS-related cancers, see HIV/AIDSrelated malignancies AIM chemotherapy 956 air, MRI characteristics 118 airway obstruction, stent placement 153, 155, 1190, 1191 AKT protein 29 albumin, serum 1081, 1081, 1297 alcohol consumption, in cancer aetiology 346, 540 alcohol injection, see ethanol injection aldosterone 457 alemtuzumab (Campath-1H) 232, 860, 1008, 1014, 1088, 1100, 1106 alkaline phosphatase (ALP) 837 lung cancer 508 multiple myeloma 1077, 1078 placental 710 alkylalkane sulphonates 83 alkylating agents 81–3 discovery 81 leukaemia risk 982 mechanism of action 81 multiple myeloma 1084–5, 1089–90 secondary tumour induction 930 see also individual drugs ALL, see acute lymphoblastic leukaemia allelic loss gall-bladder cancer 593 see also loss of heterozygosity (LOH) all-trans retinoic acid (ATRA) 102 acute myeloid leukaemia 987, 988, 989 Kaposi’s sarcoma 1173 paediatric AML 1123 17-allylaminogeldanamycin (17-AAG) 1003 ALP, see alkaline phosphatase α:β ratios 60–1 alpha-fetoprotein (AFP) brain tumours 313 hepatocellular carcinoma 561, 562 paediatric germ cell tumours 1138 paediatric liver tumours 1127, 1128 testicular germ-cell tumours 709, 710, 710, 724 α -methylacyl-CoA-racemase (AMACR) 648 alphaviruses 270, 272 recombinant, vaccines 177 alteplase 1205 ambulatory care, child cancer patient 1155 ambulatory care centres 1354, 1362
ameloblastoma 399 amenorrhoea 442, 1038–9 American Indians 592 Ames test 25 amifostine 49 aminoglutethimide 1201 5-aminolevulinic acid (ALA) 295 amitriptyline hydrochloride 1296 AML, see acute myeloid leukaemia amputation primary bone tumours 908, 915 soft tissue sarcoma 939, 943, 953 Amsterdam Criteria 669, 669, 1327 amyloidosis, myeloma 1077 amytriptyline 1083 anaemia 1204 chronic in advanced cancer 1298 multiple myeloma 1077, 1078, 1081, 1083 and radiotherapy outcomes 354–5 and response to radiation 49 anal cancer adenocarcinoma 695, 701 brachytherapy 1219, 1226 chemo-radiotherapy trials 1283–4 diagnosis and staging 696–7, 697 epidemiology 696 extra-pelvic metastases 700–1 HIV infection 696, 702, 1179 homosexual men (HIV-negative) 1179 lymph node metastases 700 management anal canal tumour 698–701, 698, 699 peri-anal tumour 701–2 radiotherapy 699–700 pathology 695–6 presentation 696 risk factors 696 surgery 701 analgesia, see pain control analgesics COX-2 inhibitors 1291 NSAIDs 1291–2 opioids 1291 parenteral 1302 WHO ladder 1291 anal intra-epithelial neoplasia (AIN) 695, 696, 1179 incidence 696 analogues 79 anal sphincter, preservation in colorectal cancer surgery 674 anaplastic large cell lymphoma (ALCL) 1055 paediatric 1142 anastellin 254 anastomoses, colorectal cancer surgery 674 anastrozole 473, 475, 771 anatomical knowledge 1427–8 anatomic anomalies, pancreatico-biliary duct union 585, 592, 593
androgen-deprivation therapy 654, 656, 657–8 androgen receptor gene 1323 aneuploidy, endometrial cancer 765 angiofibroma, juvenile 398 angiogenesis 252–4 breast cancer 255 and drug resistance 92 endogenous inhibition 254 evaluation of markers 255 history of research 251 imaging 116–17, 255–6, 261–2 ladder cancer 626 molecular basis 253–4, 253 as prognostic/predictive marker 255 as therapeutic target 100–1, 256–62, 278–9, 1283 in AML 987–8 brain tumours 301 colorectal cancer 256 inhibitors 256–8, 256, 658, 894 monoclonal antibodies 31, 228–9 myelodysplastic syndromes 1013 receptor tyrosine kinase inhibitors 31 vascular disrupting agents 258–62 and tumour hypoxia 252 angiogenesis therapies, as therapeutic target, Kaposi’s sarcoma 1173 angiogenic switch 252 angiography CT 116, 117, 1205 magnetic resonance 588 angioimmunoblastic T-cell lymphoma 1055 angiomatosis, bacillary/epithelioid 1170 angiopoetin-1 (Ang-1) 254 Ann Arbour staging system 1048, 1048, 1140 anorexia–cachexia syndrome 1197–8 anosmia 368 anthracyclines 86–7 ALL 993 breast cancer 474 cardiac toxicity 1195–6, 1417–18 liposomal delivery 86–7 paediatric AML 1123 anthropometry, child patient 1149 antibiotics, prophylactic 1205 antibody-dependent cellular cytotoxicity (ADCC) 227–9 antibody-directed enzyme pro-drug therapy (ADEPT) 229, 686 anti-cholinergic drugs 1296–7 anti-coagulation 1204, 1205 anticonvulsants 295, 302, 581, 1294 anti-depressants depression 1300–1 neuropathic pain 1293–4 antidepressants, tri-cyclic 581 anti-diuretic hormone, syndrome of inappropriate secretion (SIADH) 1202–3
Index 1447
anti-emetic drugs 1148, 1296–7 antifolate agents, pemetrexed 520 anti-folates 84–5 anti-g250/calX antibody, renal cell cancer 742, 742 antigenic drift 179 antihistamines 1299 anti-oestrogen agents non-steroidal 771 see also tamoxifen anti-sense gene therapy, malignant melanoma 895 antisense oligonucleotides 276–7, 277, 988, 1009 anti-thymocyte globulin (ATG) 230 anus anatomy 695 conservation in vulval cancer treatment 804 lymphatic drainage 695 anxiety 1294 aortic stenting 163 APC (polyposis) gene 96, 458, 670, 963, 1322 aplidine 90 apocrine adenocarcinoma, vulva 805 apolizumab 1009 apoptosis molecular basis 27–8, 27 regulation 95–6 as therapeutic target 988, 1283 appendiceal carcinoid tumours 452, 454 appetite stimulation 1298 APUD concept 438 apurinic/apyrimidinic endocuclease (APE1/redox factor-1) 252 AQ4N 92, 262, 355 Arabian Gulf states 6, 6 Ara-C, see cytarabine Ara-CHOMP regimen 334 ARCON (accelerated radiotherapy, carbogen and nicotinamide) 48, 354, 1283, 1285 Argentina, hepatocellular carcinoma 559 aromatase inhibitors 469, 1410 breast cancer 473, 475, 475 mode of action 473, 476 safety 477 uterine cancers 771, 773 aromatic amines 615 aromatic hydrocarbons 982 arresten 254 arsenic 930 chronic ingestion 839, 840, 841 arsenic trioxide (ATO) 984, 988, 989, 1013, 1124 arterioportography, CT 130 arteriovenous malformations, uterus 818, 819 asbestos exposure 347 ascites 1199, 1297 ovarian cancer 134, 134
ascorbic acid 603 Ashkenazi Jewish groups 749 Asia, bladder cancer 614 Asians, multiple myeloma 1074 Askin tumour 1130 L-asparaginase, ALL 994 Aspergillus flavus 560 Aspergillus infections 1108–9 aspiration, head and neck cancer 351, 368 aspirin cancer prevention/risk reduction 540, 667, 1330, 1437 in MI, ISIS-2 trial 1341, 1342, 1346–7, 1346 association studies 1323 astroblastoma 307 astrocytoma 302–3 anaplastic (grade III) 304, 315 diffuse (grade II) 303–4 glioblastoma (grade IV) 304–6, 315 incidence 287 molecular biology 291 paediatric 1119, 1143, 1144 pilocytic (grade 1) 288, 303 spinal 315 subependymal giant-cell 307 ataxia telangiectasia 26, 28, 40, 470, 1120, 1140, 1321 atlizumab 1178 ATM gene 28, 646 atmospheric pollution 503 atomic bombs 409, 1074, 1418 atropine 683, 1219 atypical mole syndrome 880 atypical teratoid rhabdoid tumours (ATRhTs) 1147–8 Auer rods 983, 983 Aurora kinases 101 inhibitors 33, 101, 1002 Australia, hepatocellular carcinoma 559 Austria, prostate cancer screening 646 autoimmunity in cancer 1201, 1201 in immunotherapy 895 autonomic neuropathy 507 Avastin, see bevacizumab AVE8062A (AC7700) 259, 260–1 axial flaps 350 axillary adenopathy 472–3, 489 axillary node dissection 472–3, 1373 azacytidine 104, 1013 azathioprine 869 aziridinyl drugs 83 see also mitomycin C; thiotepa BAALC gene 984 bacille Calmette–Guèrin (BCG) 193 adverse effects 624–5 bladder cancer 624–5, 627 contraindications 625 malignant melanoma 888–9 small-cell lung cancer 526
bacteria, gene therapy vectors 274 bacterial infections 1205 post stem cell transplant 1108 bacterial vaccines 177 bad news, breaking 1308, 1310–11, 1311 BAGCOPP regimen 1035 Bak 278 Bangladesh 345 barium studies 112–13, 130, 132, 673 Barrett’s oesophagus 540 Bartholin’s gland carcinoma 805 Bartonella henselae 1170 basal cell carcinoma (BCC) aetiology 836, 840–1 clinical features 841–2 cystic 843 diagnosis 844 eyelid 334, 847 incidence 840 linear 844 lip 375 management 844–5 follow-up 850 non-surgical 847–50 surgical 845–7 morphoeic (sclerosing) 843, 843 nodular 842–3, 842 nose 392–3 pigmented 843, 843 staging 844 superficial 844 vulva 805 basal cell naevus (Gorlin’s) syndrome 40, 288, 758, 836–8, 1146, 1321, 1329, 1331 basaloid squamous carcinoma 399 basic fibroblast growth factor (bFGF) 1169 basiliximab 230 BAT (B-mode Acquisition and Targeting) device 1270, 1270 Bax 278 BCD regimen 915 BCG, see bacille Calmette–Guèrin bcl-2 gene 208, 278, 586 antisense therapy 658, 988, 1009 NHL 1057 bcl-2 protein 95, 96, 96, 1057, 1059 bcl-6 gene 1174 BCNU, see carmustine Bcr/abl antigen 170, 173 BCR-ABL fusion gene, CML 997–8, 998 BEACOPP chemotherapy, Hodgkin’s lymphoma 1034–5, 1034, 1036, 1038 BEAM regimen 1037, 1102 Bec2 526 Beckwith–Wiedemann syndrome 458, 811, 1120, 1127 BEC regimen 314 Belarus 409 Bellini’s duct carcinoma 730, 731 Bence Jones proteins 1073, 1077
1448 Index
benzene 982, 1074 benzocaine lozenges 1295 benzodiazepines 1294, 1302 benzydamine mouthwash 1295 BEP regimen 314, 724 ovarian tumours 757, 758 paediatric germ-cell tumours 1138 testicular germ-cell tumours 716–17, 722–3 toxicity 722–3 bereavement 1302–3 within at-risk family 1329 β1-glycoprotein, gestational trophoblastic tumours 812 β2-microglobulin 1081, 1081 β-carotene 346, 505, 1436 β-catenin 96, 963 β-tubulin mutations 94 betel nut chewing 346, 376, 539 Bethesda Guidelines 669, 669, 1327–8 bevacizumab 228, 251, 256–7, 256, 988 breast cancer 241 carcinoids 454 colorectal cancer 240, 680, 683, 683 lung cancer 521 melanoma 894 non-small-cell lung cancer 240–1 renal cell cancer 741–2 toxicities 241–2 bexarotene 858, 858, 861 bias, avoidance in clinical studies 1340–2 bicalutamide 657–8 Bifidobacterium 274 bilateral inferior petrosal sinus sampling (BIPSS) 446 bile duct cancer, see cholangiocarcinoma biliary-enteric anastomosis 586 biliary-enteric bypass 591–2 biliary obstruction cholangiocarcinoma 587 pancreatic cancer 576 percutaneous drainage/stenting 151, 153–4, 156, 576, 590, 1199, 1199 biliary tree, rhabdomyosarcoma 1125 biochemotherapy malignant melanoma 891–2 renal cell cancer 739–40 biological therapies head and neck cancer 369–70 multiple myeloma 1090 prostate cancer 658 see also immunotherapy; moleculartargeted therapies; monoclonal antibodies biomarkers 10 in chemo-prevention trials 1438 biopsy musculoskelatal 149–50, 149 percutaneous abdominal 147–9, 148 percutaneous transthoracic 147, 148 soft tissue sarcoma 933–5 bioreductive drugs 355
Birt–Hogg–Dube syndrome 732 bismuth-213 216–17 Bismuth classification 589, 590 bisphosphonates bone pain 1293 breast cancer 481, 491 hypercalcaemia 1202, 1300 multiple myeloma 1082, 1089 prostate cancer 656, 658 bladder flaccidity after radical hysterectomy 790 radiotherapy late damage 791 bladder cancer advanced disease 631–3 aetiology 615–16 biological markers 621–2, 622 carcinoma in-situ 617, 624, 625 chemo-radiotherapy trials 1283, 1284–5 chemotherapy 1426 intra-vesical 622–6, 623 systemic 625 clinical presentation 617–18 costs of treatment 1355 diagnosis and staging 133, 618 epidemiology 133, 614 genetics 616, 616 imaging 133–4, 619–20 invasive 617 chemotherapy 630–1, 630 prognostic factors 625–6 radiotherapy 627–9, 628, 629–30 surgery 626–7 metastases 631 mortality rates 614, 631 palliative treatments 632–3 radiotherapy 625, 627–30, 1272 case selection 627–8 CT planning 628 improving efficacy of 628–9 patient immobilization 1237–8 planning 1239–41, 1240 prognosis following 628 screening 620 staging 618–19, 619 superficial tumours 616–17 management 620–5, 623 risk of recurrence 621–2, 621 surgery bladder preservation 627, 629, 1284–5 invasive cancer 626–7 superficial low-grade tumours 620–1 transurethral resection (TURBT) 1285 surveillance and recurrence 133–4 bladder replacement, orthotopic 626–7 blast cells (myeloblasts) 982, 982, 983 bleomycin 87 anal cancer 700 Hodgkin’s lymphoma 1035 lung-fibrosis induction 1191, 1192 malignant pleural effusion 1194
non-melanoma skin cancer 852 oesophageal cancer 552 paediatric germ cell tumours 1138 primary bone tumours 915 testicular germ-cell tumours 715–17, 722–3 toxicity 87, 722–3 see also BEP regimen Blom–Singer valve 364 blood, MRI characteristics 118 blood group, and gestational trophoblastic disease 813 blood loss, see haemorrhage blood transfusion, radiotherapy response/efficacy 49 Bloom’s syndrome 982 Bloom syndrome 1321 BMS-181174 355 BOADICEA model 1324 body mass index (BMI) renal cell carcinoma risk 732–3 see also obesity bone, MRI characteristics 118 bone marrow, stromal cells 1075 bone marrow transplant, see stem cell transplantation bone metastases breast cancer 481, 491 prevention 1202 endometrial cancer 771 lung cancer 506, 514 malignant melanoma 890 multiple myeloma 1075–6, 1077, 1078, 1080, 1089 pain 1292, 1293 prostate cancer 648–9, 650, 658 renal cell cancer 733, 734 treatment 736 bone mineral density, prostate cancer 658 bone morphogenetic protein 648–9 recombinant (rhBMP) 351 bone scan, prostate cancer 651 bone tumours aetiology 906 axial skeleton 909–10, 919 biopsy 149–50, 149, 907 chemotherapy 911, 911 chondrosarcoma 921 classification 905, 906 clinical presentation 907 Ewing’s tumours 917–20, 1119, 1119 incidence 905–6 lymphoma 920–1 malignant fibrous histiocytoma 906, 908, 920 metastatic, see bone metastases osteosarcoma 911–17 aetiology 906 clinical presentation 912–13 incidence 905 investigations 913–14, 913, 914 staging 908
Index 1449
treatment 915–17, 916 paediatric 1119, 1119, 1130–3 preoperative embolization 159, 160 radiotherapy 910–11, 910 secondary, following cancer therapy 930 staging 907, 908 surgery 908–10, 909 bony defects, reconstruction 350, 351 boron neutron capture therapy (BNCT) 298 Borrelia burgdorferi 1062 bortezomib 96, 96, 102, 256, 258, 987 multiple myeloma 1090 Bosnia–Herzegovina 345 bosutinib 1001 bowel obstruction 1198–9, 1198, 1297 stenting 153, 155, 675, 1198 Bowen’s disease 839–40, 1203 brachial plexus, radiation tolerance 42 brachytherapy 52, 1353 advantages of low-dose rate 52 brain tumours 298 cervical cancer 786–8 adverse effects 791 high dose rate (HDR) 793 cervical lymph nodes 375 characteristics of 1210–11 cholangiocarcinoma 591, 592 clinical indications anal canal/rectal cancer 1226 breast cancer 1227 oral cancers 1219, 1220–1, 1225–6 penile cancer 1226–7 prostate cancer 1227–30 skin cancer 1226 colorectal cancer 687 complications 1220–1, 1229–30 dose and dose-rate considerations 1221–2, 1222 dosimetry 1211–16 computer calculations 1225 in vivo 1222–5, 1224, 1225 endobronchial 513–14 endometrial cancer 767–8, 772 eye tumours 326, 327, 332, 332 head and neck cancer 363 high-dose rate 52, 549, 549, 552 history of development 1210 implant techniques 1217–18 iridium wire checking prior to use 1225 ordering 1218 physical properties 1211 lip carcinoma 376 ‘Manchester’ system 1210 oesophageal cancer 549, 552 oral cavity tumours 1219, 1220–1, 1225–6 oral tumours 380 post-operative care 1220 preparation of patient 1218–19 pre-planning 1216–17
prostate cancer 655, 655 radiation safety 1211 radiobiological modelling 64–5, 69–70 soft tissue sarcoma 943, 945–6, 953 tonsillar tumours 381 vaginal cancer 798 wound healing 945, 954 BRAF gene 879, 894 braf protein 410 Bragg peak 950, 951 brain radiation injury 359, 391 radiation tolerance 42 radiotherapy ALL 994, 1122 metastases 528 whole brain (WBRT) 893, 1177, 1200 brain imaging MRI 119, 119 MRS 119–20, 120 brain metastases 293, 315–16, 316, 316 breast cancer 491–2 chemotherapy 894 gestational trophoblastic disease 817, 826 lung cancer 506, 514 malignant melanoma 893–4 MRI imaging 119 multiple 316, 316 radiotherapy 528 raised intracranial pressure 1200 recursive partitoning analysis (RPA) classification 893–4 renal cell carcinoma 733 solitary 316 brain stem, glioma 307–8 brain tumours aetiology 288–9 and age 288 associated genetic syndromes 288, 289 astrocytoma 302–6 chemotherapy 300–1 chordoma 311 classification 289–90, 289 craniopharyngioma 314 ependymoma 307 epidemiology 123, 287–8, 287 gene therapy 301 glioma, radio-immunotherapy 218 haemangiopericytoma 311 imaging 124–5, 293–4, 294 immunology 291–2 local environment 291 medulloblastoma 308–9, 309 meningioma 309–11 metastatic, see brain metastases molecular biology 290–1, 290, 291 oligodendroma 306–7, 306 paediatric 1143–8 classification 1143, 1143 glial 1143–5 primitive neuroectodermal (PNET) 1145–8
pineal region 287, 287, 312–14 presentation 292–3, 293 primary lymphoma 287, 293, 311–12 radiotherapy 295–8, 296, 297 conformal 1257 normal tissue reactions 298–300 planning 1242, 1242, 1243 raised intracranial pressure 1200 rare neuro-epithelial tumours 307–8 spread 292 support of patient/family 302 surgery 295 BRCA1/BRCA2 genes 470, 646, 1318–20, 1319 cancer prophylaxis 1329, 1330, 1439 cancer risk assessment 1323–4, 1324 cancer risk perception 1324–6, 1325 fallopian tube cancer 758 ovarian cancer risk 749 pancreatic cancer 574 targeted drugs 97 breakthrough pain 1292 breast cancer audit of mammography and pathology reporting 1371 axillary adenopathy 472–3, 489 axillary node clearance 472–3 arm lymphoedema after 1373 bilateral 487 biological therapies 480–1, 481 biopsy 147, 150 bisphosphonate therapy 481, 491 bone metastases, prevention 1202 brachytherapy 1227 brain metastases 491–2 cancer ‘lottery’ 1378 chemotherapy early disease 474, 477–80, 477, 478, 479, 479 locally-advanced disease 490 metastatic disease 491 diagnosis and staging 128–9 drug resistance 98 ductal carcinoma in-situ (DCIS) 471 endometrial cancer risk 764 epidemiology 128, 468 familial genetic heterogeneity 1318 genetics clinic referral guidelines 1327 management of at-risk individual 1333–4, 1333 prophylactic measures 1329, 1330, 1439 risk assessment 1323–4, 1324 risk perception 1324–6, 1325 follow-up 129, 493, 1374 gene expression profiling 30 genetics 469–70 Her-2/neu-positive 98, 237–8, 1424 historical treatment 3–4 in HIV 1181
1450 Index
breast cancer (contd.) hormone replacement in 492–3 hormone therapy adverse effects 477 clinical trials overviews 1347–8, 1348 combined with chemotherapy 480 early disease 473–7, 474 molecular basis of response 476–7, 476 imaging 128–30 CT 129 mammography 128–9 MRI 129–30 PET 130 inflammatory 489 late effects of therapy 1417 lobular carcinoma in-situ (LCIS) 471 locally-advanced 489–90 loco-regional recurrence 490 males 488 metastases, ocular 334 metastatic complications 491–2 treatment 490–1 monoclonal antibody therapy bevacizumab 240, 241 trastuzumab 238–9 oestrogen receptors (ER) 470, 473, 475, 476–7, 476, 480 other associated primary tumours 289 outcomes 1355–6, 1355, 1378 Paget’s disease 488 pathology and prognostic indices 470–1, 471 post-mastectomy lymphangiosarcoma 930 in pregnancy 487–8 presentation and investigation 469 prevention 469, 1330 hormone manipulation 1437–8 surgery 1439 radio-immunotherapy 219–20 radiotherapy 481–6, 482, 483, 484, 485, 486, 487 bone metastases 491 immobilization devices 1236, 1236 induced tumours 929 new/complex methods 493 planning 1241 risk factors 468–9, 828, 1434, 1435 screening 9, 469 surgery early disease 471–3 locally advanced disease 489–90 treatment costs 1355, 1356 tumour angiogenesis 255 Breast Cancer Linkage Consortium 1318, 1323, 1324 breast reconstruction 472 breathlessness, advanced disease 1294–5 breath test, lung cancer screening 505
Breslow thickness 882, 883 vulval melanoma 804–5 British Association of Surgical Oncology 129 bromocriptine 440–1, 445, 447, 448 5-bromo-deoxyuridine 1282 bronchial cancer, see lung cancer bronchial carcinoids 454 bronchial obstruction 153, 155, 1190, 1190, 1191 broncho-oesophageal fistula 541, 550 bronchoscopy, oesophageal cancer 542 bryostatin 90 buccal mucosa carcinoma 378–9 bupivacaine epidural/spinal 1294 wound lavage 1299 burden of cancer, global 4, 4 Burkitt’s lymphoma 1052, 1053–4 endemic 1141 HIV-associated 1174 non-endemic 1141 ocular metastases 329 paediatric 1120, 1140, 1141–2 burn-out, doctor 1309–10 busulfan 83 pulmonary toxicity 1192–3 stem cell transplantation 1014, 1102 ‘butterfly glioma’ 292, 292 CA 15.3 751 CA19-9 (carbohydrate antigen 19-9) 576, 588, 594, 673, 751 CA 72.4 751 CA 125 endometrial cancer 765, 772 ovarian cancer 134, 135, 751, 752, 755 cabergoline 440, 442, 443, 445 cachexia 1197–8 cachexia-anorexia syndrome 1298 Caelyx 86–7 Cajal, interstitial cells of 960 calcitonin serum levels 413, 413, 414, 451, 507 therapy in hypercalcaemia 1202, 1300 calcium MRI characteristics 118 serum, see hypocalcaemia; hypercalcaemia calcium channel blockers 1294 calcium-sensing receptor (CaSR) 449 calf thymic extract 869 calicheamicin 230 Calman–Hine report 1359, 1379–80, 1379 Calvados 345, 380 and oropharyngeal cancer 345, 380 Cameron report 1380, 1382–3, 1383 Campath-1H (alemtuzumab) 232, 860, 1008, 1014, 1088, 1100, 1106 Campbell Report 1380 camptothecin 89 Canada, breast cancer outcomes 1355
Canadian National Breast Screening Study 1434 canarypox virus, vaccine 177 cancer centres 1403 cancer journey 1377, 1378, 1385 improving 1399, 1400, 1401 cancer networks 1381, 1382, 1398–9 commissioning cancer services 1399 dividends of 1399 management teams 1398, 1399 network service plans 1398 workforce strategy 1398 cancer pre-disposition genes 1318–22, 1319–20 cancer risks 1323 assessment 1323–4, 1323, 1324 perception 1324–5 genetic testing 1330–4, 1331–2 identification approaches 1322–3 inheritance 1320, 1322 management of known/suspected carriers 1326–8, 1327–8 mechanisms of action 1320, 1322 penetrance 1318, 1323 phenotypic features 1322–3 see also individual genes cancer risk assessment 14 cancer services access to 20, 1359 centralization 1354, 1360, 1373, 1403 demand for 1352 national investment plan 1363–4 new Europe 1356–8 palliative care 1290 patient expectations 1352, 1363 rehabilitation 1360 staged model 1353 technological change 1353 UK, see UK cancer services USA 1353–6 Cancer Services Collaborative Improvement Partnership 1401 Cancer and Steroid Hormone (CASH/CLAUS) study 1323 cancer surgery 10 future of 10, 10 history of 3–4 cancer-testis antigens 170 Candida infections oral 359, 1295 post stem cell transplant 1108–9 cannabinoids 1297, 1298 canstatin 254 Canvaxin vaccine 888–9, 888, 892 capecitabine cholangiocarcinoma 591, 592 colonic carcinoma 682 GEP endocrine tumours 452 mechanism and pharmacology 85 pancreatic cancer 580 rectal cancer 678 CAPIRI chemotherapy 683
Index 1451
CAPOX regimen 684 carbamazepine 1083 carbogen 47, 48, 354, 1283 carbon dioxide laser basal cell carcinoma 846–7 vulval intra-epithelial neoplasia 801 carboplatin adverse effects 792 bladder cancer 632 brain tumours 300 combination with radiotherapy 1285 endometrial cancer 771 fallopian tube cancer 759 gall-bladder cancer 595 head and neck cancer 358 lung cancer 516, 517, 519, 520, 520, 522–3 mechanism of action 84 oesophageal cancer 552 ovarian tumours 753, 754, 757 paediatric germ-cell tumours 1138 paediatric osteosarcoma 1134–5 testicular germ-cell tumours 716, 722 toxicity 84 see also chemotherapy regimens containing carboplatin carboxylesterase 94 carcinoembryonic antigen (CEA) 170, 1190 antibody 686 radio-immunotherapy 218, 219 biliary tract cancer 588, 594 colorectal cancer 131, 667, 672, 673, 685 pancreatic cancer 576 carcinogenesis 1436 first evidence 25 carcinogens chemotherapy agents 615 gastric cancer aetiology 603 head and neck cancer aetiology 346, 346, 347 hepatocellular carcinoma aetiology 560 industrial chemicals 503, 585, 615, 930 tobacco smoking 615 carcinoid syndrome 453 carcinoid tumours 452–4 appendiceal 452, 454 atypical 453, 454 bronchial 454 colon 454 IFN-α therapy 192 liver metastases 562, 567 treatment 160–2, 161, 567–8 lung 510 carcinoma in-situ bladder 617, 624, 625 breast 471 larynx 385 testis 707, 707 cardiac arrhythmias 1196 cardiac tamponade 1195, 1195 cardiac tumours 837
‘caretaker’ genes 1322 carcinogenesis, lung cancer 504, 504 carmustine (BCNU) 300, 1192, 1193 malignant melanoma 890–1, 892 multiple myeloma 1085 stem cell transplantation 1102 topical 857–8 Carney complex 409, 444, 447, 458 Caroli’s disease 585 carotenoids 349 dietary 1436 carotid artery stenting 163, 164 carotid-body tumours 397 caspases 95 Castleman’s disease 1178 cataract, radiation-induced 332, 362–3 catecholamines neuroblastoma 1129, 1130 phaeochromocytoma 430, 455 plasma levels 455 urinary 430, 1129, 1130 catheters, intravenous access 1205 cauda equina lesions 492 causes of cancer 7, 7, 8, 15 cavernous haemangioma, ocular 330 cavography 9 CAV regimen 522, 526 CBOP-BEP regimen 716 CCNU (lomustine) 300, 306, 1144, 1146 CD4+ lymphocytes 168, 168, 173 count and anal cancer treatment 702 CD8+ T cells 173 CD20 230–1 monoclonal antibodies 230–2 see also rituximab CD25, monoclonal antibodies 230 CD30 antibodies 1040 CD33 monoclonal antibody 987 monoclonal antibody (gemtuzumab) 1123 monoclonal antibody (HuM195) 216–17 CD34 1100 CD52 232 monoclonal antibody, see alemtuzumab (CMAPATH-1H) CDE regimen 1175, 1176 CDKN2A gene 26, 291, 880 CDKN2 gene 574, 593 celecoxib 258, 881, 1437 cell cycle 26–7, 75–6 as chemotherapy target 76, 76, 90–1, 91, 101 molecular control 27, 27 and radiosensitivity 45, 356 selective chemotherapy agents 301, 302, 1282 and tumour growth 76 cell fate, control 27–8, 27 cell loss, head and neck cancer 352–3, 352, 353
cell membrane, drug transport 92–3, 92 cell signalling, see signal transduction cellular adhesion molecules bladder cancer 626 colorectal cancer 670–1 multiple myeloma 1075 cellular adoptive therapy, see adoptive cell therapy centralization of cancer care 1354, 1360, 1373, 1403 central nervous system (CNS) involvement in renal cell carcinoma 733 radiotherapy damage 359, 391 raised intracranial pressure 1200 central nervous system (CNS) tumours aetiology 288–9 associated genetic syndromes 288, 289 classification 289–90, 289 epidemiology 287–8, 287 immunology 291–2 local environment 291 lymphoma 137, 1176–8 metastases lung cancer 506, 514 melanoma 893–4 radiotherapy 893–4 surgical resection 893 testicular germ-cell tumours 717 molecular biology 290–1, 290, 291 paediatric 1142–3 clinical presentation 1143 ependymomas 1119, 1144–5 glial tumours 1143–4 incidence 1119 primitive neuroectodermal tumours 1145–8 presentation 292–3, 293 primary lymphoma 1052–3 radio-immunotherapy 218 spinal 314–15 see also brain tumours central venous catheter, migration/extraction 156, 158, 158 c-erbB1 gene 470, 476, 480 c-erbB2 (Her-2) gene 95, 98, 447, 504, 504, 586, 604, 616, 765, 916 cerebellar degeneration, subacute 507 cerebellar syndrome 308 cerebral herniation 1200 cerebral toxoplasmosis 1177 cervical cancer 7 aetiology 778–9 audit of management 1372 cerebral metastases 119, 119 chemo-radiotherapy trials 1285–6 clinical features 780–1 haemorrhage 789–90 histological types 780 and prognosis 784 HIV-positive women 1179–81 incidence 777–8 incidental finding at hysterectomy 789
1452 Index
cervical cancer (contd.) investigations 781 micro-invasive disease 780, 785 MRI endoluminal 122 lymph node imaging 121 in pregnancy 789 pre-invasive lesion, see cervical intraepithelial neoplasia (CIN) radiotherapy, vaginal cancer risk 796 recurrence 790 screening 781, 792 spread of 780 staging and prognostic factors 781–4, 782, 782, 783 treatment chemotherapy 789 complications 790–2 costs 1355 potential developments 792–3 radiotherapy 785–9 cervical glandular intra-epithelial neoplasia (CGIN) 777 cervical intra-epithelial neoplasia (CIN) HIV-positive women 1180 incidence 777–8 investigations 781 pathology 780 prevention (vaccination) 784 therapeutic vaccination 177 treatment 784–5 cervical lymphadenopathy child 1139 head and neck tumours 373–5 imaging 125, 126, 126 inoperable 375 occult primary tumour 373, 373, 374 radiotherapy 374–5 surgical dissection 374, 375 thyroid cancer 413–14, 416–17, 428 tonsillar tumours 380, 381 cervical smear test 9 cervix anatomy 779, 779 lymphatic drainage 779 transformation zone 778, 778 tumour invasion in endometrial cancer 136 cetuximab 46–7, 95, 228, 234–7, 366, 580, 1286 colorectal cancer 236–7, 680, 683, 683, 684, 686 combined with cisplatin and radiotherapy 236 combined with radiotherapy 235 head and neck cancer 234–6 indicators of response to 236–7 CHART trial 366, 367 Chediak–Higashi syndrome 1203 chemo-embolization 162 hepatocellular carcinoma 564–5, 565 chemo-prevention 1436–8
clinical trial design 1438 cyclo-oxygenase inhibitors 1436–7 diet-derived agents 1438 head and neck cancer 348–9, 349 hormones as targets 1437–8 lung cancer 504–5 retinoids 103 chemo-radiotherapy 49–51 clinical experience 50–1 clinical studies anal cancer 1283–4 bladder cancer 1284–5 cervical cancer 1285–6 head and neck cancer 1286 goal of 1280 mechanisms of enhanced tumour control 49–50 mechanisms of interactions 1281–2 targeted molecular therapies 1283 ‘chemoreduction’ 323 chemotherapy accelerated 524 adverse effects, myelosuppression 792 anti-angiogenesis 258 attitudes of patients and health professionals 1311–12, 1311 brain tumours 300–1 carcinogenic 615 cell cycle dysregulation 76, 76, 90–1, 91, 101, 1282 changes in 1353 classification 81 alkylating agents 81–3 anti-metabolites 84–6 anti-tumour antibiotics 86–7 marine organism-derived agents 90 plant-derived agents 87–9 platinum analogues 79, 83–4 topoisomerase I and II inhibitors 89, 94–5 combination regimens 78 complications, neutropenia 1205–6 convection-enhanced delivery (CED) 301 dose intensity 77, 78 dose-response effects 78 future of 13, 16–17 high dose Hodgkin’s lymphoma 1036, 1037 small-cell lung cancer 525 history of use 4 hypoxically-activated agents 92 individualized 97 intra-arterial 567, 612, 916 intra-peritoneal 611–12, 754, 1199 isolated limb infusion 889–90 isolated limb perfusion 889, 952–3 late effects 828 secondary tumours 930, 1151 leukaemia risk 982 limitations as sole therapy 1280–1 liposomal delivery 366
novel agents 1360–2 apoptosis pathway targets 101–2 costs 1361–2 differentiation agents 102–3 epigenetic therapies 103–4 signal transduction inhibitors 97–101, 228–9, 366 telomerase-targeted 104–5 ‘Packer’ regimen 309 pro-drugs 82 radio-sensitizing agents 1282–3 regional infusion 158 resistance cell hypoxia 353–5 retinoblastoma 323 route of administration 77–8 scheduling 78–9 success of 4, 4 toxicity cardiac 86, 1152, 1195–6, 1417–18 child cancer patient 1152–3 neurological 792 pulmonary 1152–3, 1192–3 renal 791–2, 792, 1152 tumour cell kill 76–7 see also drug development Chernobyl nuclear accident 409, 410, 1074 chest pain, lung cancer 506, 506 chest wall, lymphoma 137 chest X-ray lung cancer screening 505 testicular germ-cell cancer 711, 711 Child-Pugh score 563, 563 children communication with 1313 genetic testing 1334 palliative care 1290 see also infant; paediatric cancers Chile 592, 593 chimeric immune receptors 196 chimney sweeps 25 China breast cancer mortality 1435 head and neck cancer 345 hepatocellular carcinoma 559, 559, 560, 562 nasopharyngeal cancer 388 chlamydial infection, orbital lymphoma 336 Chlamydia psitaci 1061–2 chlorambucil 82, 1007 pulmonary toxicity 1192–3 chlorhexidine 1295 chloromethylether 503 chlorophenol exposure 930 ChlVPP/EVA hybrid therapy 1035, 1036 ChlVPP regimen 1140 CHOEP regimen 1050 cholangiocarcinoma 585–92 aetiology 585–6 anatomical relations 586–7 clinical presentation 587
Index 1453
diagnosis 588–9, 588, 589 differential diagnosis 587 future trends 595 incidence 585 intra-hepatic 561, 585, 587 molecular biology and genetics 586 natural history 587 palliation 591–2 pathology 587 percutaneous biliary drainage 153–4, 156 resectability rates 591 staging 586, 586, 589, 590 treatment 590–1 cholangiography, percutaneous transhepatic (PTC) 153–4, 156, 588–9, 589, 594 cholangitis 587 primary sclerosing (PSC) 585, 587 cholecystectomy extended 595 gall-bladder cancer 594–5 malignancy risk 586 choledochal cysts 585, 586 cholestasis, cholangiocarcinoma 587, 590 cholesterol intake, pancreatic cancer 574 choline 119–20, 120, 133 chondrosarcoma 905, 906, 908, 921, 950 CHOP regimen 860, 1050–2, 1058, 1061, 1141–2, 1175, 1176 chordoma 311 choriocarcinoma 313, 709, 712, 713, 717 brain metastases 1200 gestational clinical features 816–17 epidemiology 813 infantile 817 lung metastases 817, 818, 1190–1, 1192 pathogenesis 811–12 spinal cord compression 1200 testicular 709 choroidal naevi 330 choroid-plexus tumours 308 chororetinopathy, radiation 332 chromophobe renal carcinoma 730–1, 730 chronic disease, cancer as 20 chronic lymphocytic leukaemia (CLL) clinical features 1005, 1006 epidemiology 1005 molecular pathogenesis 1005 monoclonal antibody therapies 231, 232 renal failure 1196–7, 1196 staging and prognostic factors 1006, 1007 T-cell 1009 treatment 1006–9 chronic myeloid leukaemia (CML) clinical features 996, 997 epidemiology 996 molecular pathogenesis 996–9, 997–9 prognostic and predictive factors 999
treatment advance phase disease 1004 general principles 999–1000 imatinib mesylate 98, 1000–1 immunotherapy 1002 interferon-α 1002 investigational 1002–3 relapsed disease 1004 stem cell transplantation 1003–4, 1009 chylothorax 1193, 1194 chylous ascites 1199 chylous fistula 351 cigarette smoking cancer aetiology 8, 615, 667, 732, 1440 cervix 778–9, 790 head and neck cancers 346, 346, 372, 388 lung 503 oesophagus 540 oral cancer 379 pancreas 573 renal 732 carcinogens 615 cessation 1362 advice 1306–7 effects of 503 UK national cancer plan 1386–7 in children 503 control/reduction of 6–7, 15, 347–8, 504 passive (indirect) 503 reverse 379 ciliary body, benign tumours 330 cimetidine 1299 circadian rhythms, and tumour growth 356 cirrhosis 585 in hepatocellular carcinoma 561, 562, 562, 563 liver resection 564 cisplatin adverse effects, myelosuppression 792 anal cancer 700 bladder cancer 630, 630, 631 brain tumours 300, 309 cervical cancer 792 combination with radiotherapy 1284 combined with IL-2/IFN-α 891–2 gastric cancer 610–11, 612 head and neck cancer 358 intraperitoneal 84 malignant melanoma 890–1 mechanism of action 83 metastatic carcinoid tumours 568 non-melanoma skin cancer 852 non-small cell lung cancer 515–16, 516, 517, 518, 519, 520, 520 oesophageal cancer 552–3 osteosarcoma 915–16, 916, 1135–6 ovarian cancer epithelial 753, 754 germ cell tumours 757
paediatric CNS tumours 1146 paediatric germ-cell tumours 1139 resistance 96–7 small-cell lung cancer 522–3 soft tissue sarcoma 961 spectrum of activity 84 testicular germ-cell tumours 715–16, 721–2, 724 thyroid tumours 451 toxicity 83–4, 722–3 hypomagnesaemia 1203 neurological 792 renal 791–2, 792 cisplatin-etoposide, GEP endocrine tumours 452 citrate, prostate cancer 120 c-kit gene 388, 960, 962 cladiribine 1062 Clark’s level 882, 883 Claus model 1323 clear cell carcinoma renal 730, 730 vaginal 799 clear cell sarcoma 928 kidney 1137 clinical audit 1366 Clinical Genetics Department 1326 referral guidelines 1326, 1327–8 Clinical Oncology Information Network (COIN) 1368, 1369 clinical target volume (CTV) 1243, 1244 clinical trials avoidance of bias 1340–2 avoidance of random errors 1342–5 chemo-radiotherapy 1283–6 data-dependent subgroup analyses 1341–2 economic and quality-of-life assessments 1343 entry procedures (‘uncertainty principle)’ 1343–5, 1344 European Carotid Surgery Trial (ECST) 1343–4, 1344 expectations of 1338, 1339 factorial design 1348–9, 1349 ISIS-2 1341–2, 1342, 1346–7, 1346, 1348 ISIS-4 1349–50, 1350 large advantages of 1339, 1343, 1349–50, 1350 relevance to clinical practice 1350 LIMIT-2 1349–50, 1350 magnesium infusion in MI 1349–50, 1350 molecular targeted therapies 30, 34–5 non-randomized or small-sized 1339 outcome assessment 1341 patient numbers 1339 patient recruitment 1314 phase I 80 phase II 80–1 phase III 81, 1410–11, 1413
1454 Index
clinical trials (contd.) randomization 1340–1, 1342–3 small refuted by mega-trial 1349–50, 1350 systematic overviews antiplatelet drugs 1347, 1347 reliability of 1345–6 targeted therapies, surrogate endpoints 1410 treatment allocation 1340–1 UK co-ordination of 1384–5 clitoris 799 clodronate 481, 1089, 1202, 1293, 1300 clofarabine 987, 988, 995 clondronate, prostate cancer 656, 658 Clonorchis sinensis 585 Cloquet’s node 885–6 cloretazine 987 Clostridium 274 clubbing 507 c-Met gene 586 C-met proto-oncogene 100 CMF regimen 82 CML, see chronic myeloid leukaemia CMV regimen 630–2, 630 c-myc gene 447, 470, 1053, 1120, 1174 coaxial introducer 147 cobalt-60 1233 Cockayne’s syndrome 40 co-danthramer/co-danthrusate 1297 codeine phosphate 790, 791, 1292, 1295 Codman’s triangle 912, 913 CODOX-M/IVAC chemotherapy 1054 coeliac lymph nodes 587 coeliac plexus block 161, 162, 581, 592, 1294 coffee drinking 574 cognitive changes brain tumours 293 cranial irradiation 528 colectomy, prophylactic 1439 colitis, ulcerative 668–9, 1439 collecting duct (Bellini’s) carcinoma 730, 731 collimator, multi-leaf (MLC) 1260, 1260, 1265–7, 1265, 1266 colon, carcinoids 454 colon cancer, hereditary, guidelines for genetics referral 1327 colon GISTs 960 colonography, CT 132 colonoscopy 130, 132 polyp removal 674 screening 9, 673 colon transposition 382 colorectal cancer aetiology 666–7, 667 angiogenesis inhibition 256 associated conditions 668–70, 669 chemotherapy 682–5 resistance 95, 96 diagnosis and staging 130–1, 131, 667
epidemiology 130 follow-up 685–6 future developments in treatment 686–7 incidence 666 and dietary factors 1434 liver metastases 567 local recurrence 675 metastatic 676, 683–5 molecular pathogenesis 670–1 monoclonal antibody therapy 236–7, 240 NHS guidelines 687–8 pathology of invasive cancer 671–2 patient experience 1359 patient pathway 1399, 1400 prevention 667, 1330, 1437 surgical 1439 radiotherapy 676–82 advances in 686–7 colon cancer 679–82 complications 680–2 intraoperative brachytherapy 687 rectal cancer 676–9, 679 recurrence 131 risk after treatment of gestational trophoblastic disease 828 high risk groups 131–2 in inflammatory bowel disease 1439 screening 131–2, 673–4 staging 671–2, 672–3 surgery 674–6 surveillance and recurrence 131 survival 666 treatment costs 1355 colorectal polyps 667–8, 670 adenomatous 668, 670 hyperplastic 668 treatment 674 colostomy anal cancer 698, 701 colorectal cancer 674, 675 colpectomy (vaginectomy) 798 colposcopy cervical intra-epithelial neoplasia 781 vaginal cancer 797 columella tumours 392 combined modality therapy, HIVassociated anal cancer 1179 combretastatin A4 phosphate (CA4P) 233, 259, 260, 260, 261 ‘commando’ operation 374, 381 Commission for Health Improvement/National Audit Office Report 1396–7, 1397 Common Terminology Criteria Adverse Effects (CTCAE) 1416 communication assessing quality of life 1314 bad news 1308, 1310–11, 1311 basic skills 1309 burn-out 1309–10
child cancer patient 1154–5 with children 1313 definition 1305 doctor’s perspective 1306–7 element of good 1308–10 empathy 1309 engendering hope 1312 importance of 1305–6 improvement/skills training 1306, 1307, 1308 informed consent 1313 language 1307, 1309 non-verbal 1308 patient and family 1291 patient’s perspective 1307–8 patient’s prognosis 1312–13 published audit 1374 racial issues 1315 and screening for cancer 1314 sexual problems 1314–15 skills training 1307 therapeutic dialogue 1313 treatment decisions 1189 see also speech co-morbidity 1418 complaints, patients’ 1306, 1309 complement-dependent cytotoxicity 228 computed tomography (CT) advantages 115 angiography 116, 117, 1205 applications bladder cancer 133 brain tumours 124–5, 293, 296, 296 colorectal cancer 130 endometrial cancer 135 head and neck cancers 125, 126 lung cancer 126–8, 127, 128, 129 lung cancer screening 128 non-Hodgkin’s lymphoma 137, 138 prostate cancer 132 arterioportography (CTA) 130 bladder cancer 619–20 brain tumours, surgical guidance 295 cholangiocarcinoma 588 colonography 132 colorectal cancer diagnosis 667 conformal radiotherapy planning 1256 contrast media 115 conventional single-slice 115 functional 116–17, 117 fusion with other imaging modalities 116 MRI 304 PET 123–4, 125, 126, 127, 128, 129 gastric cancer 605 hepatocellular carcinoma 561 multidetector 116, 116, 117 multiple myeloma 1078 oesophageal cancer 541, 542 ovarian cancer 752 pancreatic cancer 576 post processing 115
Index 1455
primary bone tumours 914 principles 114–15 prostate cancer 650 radiotherapy, brain tumours 1242, 1242, 1243 radiotherapy planning 628, 1238, 1239–41, 1239–41 bone tumours 910–11, 910 colorectal cancer 681 prostate cancer 654, 654 simulator 1241–2 spiral (helical) 115–16, 115 testicular germ-cell tumours 711–12, 711 thyroid 413, 414 computer-aided diagnosis (CAD) 116 computers decision-making 1363 medical audit 1368 radiotherapy planning 1240, 1245–9 conformal radiotherapy (CFRT) basic principles 1255–6 brain tumours 1257 clinical benefit 1255 concept and rationale 1254–5, 1255 prostate cancer 1256, 1257–8, 1258 soft tissue sarcoma 950 treatment planning evaluation 1258–9, 1258, 1259 optimization algorithms and class solutions 1257–8, 1258 quality assurance 1259–60, 1260 target volume definition 1256–7, 1256 confusional states, advanced disease 1301 congenital abnormalities, and paediatric cancer 1119–20, 1121 congestive heart failure, bevacizumab therapy 242 conjunctiva benign tumours 330–1 lymphoma 336–7 malignant melanoma 335–6 squamous cell carcinoma, HIV-positive patient 1181 conjunctival intra-epithelial neoplasia (CIN) 335 Conn’s syndrome 457, 458 consciousness, altered, brain tumours 293 consent child cancer patient 1155 informed 1313 constipation 1297 consumerism, in medicine 14, 18–19 contraceptives, oral 748, 813, 827 Co-operative Ewing’s Sarcoma Study (CESS) 1131–2 co-ordination problems, brain tumours 293 COPP regimen 1033–4, 1034, 1035, 1036, 1038 cord blood, stem cells 1099
CO regimen 823–4, 824 corticosteroids advanced cancer 1294, 1295, 1296, 1297, 1298, 1300 ALL 993 anorexia-cachexia syndrome 1197–8 brain tumours 292, 295, 299, 301–2 in gastrointestinal obstruction 1198 multiple myeloma 1083, 1084 raised intracranial pressure 1200 spinal cord compression 1200 stem cell transplantation 1109 topical, mycosis fungoides 857 corticotroph adenomas, silent 443 corticotrophic releasing hormone (CRH) 446 cortisol, urinary free 1201 cortisol, serum, Cushing’s disease 446 costs of cancer care 20–1, 1403 see also economics of cancer Cotswold Staging Classification (Hodgkin’s lymphoma) 1030, 1031 cough advanced disease 1295 lung cancer 506, 506 Courvoisier’s sign 575 Cowden’s syndrome 26, 29, 288, 409, 413, 1321, 1329, 1333 cranial irradiation acute lymphoblastic leukaemia 994 adverse effects 359, 362, 528 lung cancer metastases 528 paediatric ALL 1122 whole brain 893, 1177, 1200 cranial nerve palsies 389, 397 cranial nerves, radiation injury 359, 362 craniopharyngioma 314, 448 craniotomy 295 C-reactive protein (CRP) 162, 1078, 1081 creatinine clearance, chemotherapy 791–2, 792 Croatia 345 Crooke’s cell changes 446 cryopreservation, ovary/semen 1038–9 cryotherapy 840 basal cell carcinoma 845–6 contraindications 846 non-melanoma skin cancer 852 prostate cancer 655 retinoblastoma 325 tongue tumours 377 CTD regimen 1084, 1089–90 CTLA4 (cytotoxic T-cell lymphocyte A4) antibodies 892–3 curcumin 346, 1438 curettage, non-melanoma skin cancer 845, 852 Cushing’s disease (pituitary-dependent Cushing’s syndrome) 443, 446–7 Cushing’s syndrome 457, 458, 507, 1201 cutaneous lymphoma B-cell 864–5
T-cell 853–64 WHO/EORTC classification 853, 854, 855 C-VAMP chemotherapy 1084 CVB regimen 1102 CVE regimen 523, 523 CXCR4 841 Cyberknife system 1275, 1275 cyclin D1 gene 449, 470 cyclin-dependent kinases (CDKs) 27, 101, 447 inhibitors 101 cyclin D expression 574, 615, 1061 cyclizine 1296, 1297 cyclo-oxygenase-2 (COX-2) 616, 625 over-expression in cancers 1436 cyclo-oxygenase inhibitors (coxibs) 349, 625, 658, 1291, 1293, 1436–7, 1438 cyclophosphamide 309 ALL 993, 994 carcinogenic effects 615 cardiac toxicity 1195–6 CLL 1007–9 Ewing’s tumour 1131, 1131–2 gestational trophoblastic tumours 823–4, 824 malignant melanoma 892 mechanism of action 82 myeloma 1083, 1084, 1085, 1089–90 non-small cell lung cancer 517 ovarian cancer 753, 757 primary bone tumours 915, 918 pulmonary toxicity 1192–3 soft tissue sarcomas 954–6, 955, 961 stem cell transplantation 1102 toxicity 82 see also chemotherapy regimens containing cyclophosphamide cyclosporin 323, 1109 toxicity 1104, 1106 cyproheptadine 447 cyproterone acetate 657 cystectomy 133 avoidance in bladder cancer 1284–5 partial 627 radical 626–7 cystitis haemorrhagic after stem cell transplantation 1104 radiation 790, 1419 cystoscopy 133, 618 cytarabine leukaemias 986, 988, 990, 994, 1000, 1001 mechanism of action 86 myelodysplastic syndromes 1012 ocular tumour 334 radioprotection 1281 stem cell transplantation 1102 toxicity 86 cytochrome P450 enzymes 92, 94, 586
1456 Index
cytogenetic abnormalities 1322 ALL 992–3 AML 984 CML 996–9 desmoid tumours 962–3 Ewing’s sarcoma 906, 1130–1 follicular lymphoma 1057 Merkel-cell carcinoma 398 multiple myeloma 1076, 1081 mycosis fungoides 857 myelodysplastic syndromes 1011 rhabdomyosarcoma 1125 soft tissue sarcoma 927–8 cytogenetic analysis 1322 cytokine genes 280 cytokines 188–94 brain tumours 292 cachexia induction 1197–8 combined with monoclonal antibodies 231 families 188 immunosuppressive 179, 187 interleukins 189–90, 193–4 local administration intra-lesional 194 isolated limb perfusion 194 lymphocyte interactions 188, 188 myeloma pathology 1075 renal cell cancer 736–40 role in HIV-related Kaposi’s sarcoma 1169 T helper 1 (Th1) 168, 168 tumour necrosis factor-α (TNF-α) 190, 194 in vaccine strategies 194 see also named cytokines cytomegalovirus (CMV) infection, stem cell transplant 1108 cytoreductive gene therapy 277–80 cytoreductive surgery ovarian cancer 752, 753 renal cell cancer 735 cytosine arabinoside, see cytarabine cytosine deaminase (CD) genes 278 CYVADIC regimen 954, 956, 961 Czech Republic 1356, 1357 dacarbazine (DTIC) 83 malignant melanoma 890, 894 soft tissue sarcomas 952, 954–6, 955, 961 dactinomycin, see actinomycin D dantron 1297 Dartmouth regimen 890–1 dasatinib 995, 1001, 1002 daunorubicin 86 ALL 993 AML 986 liposomal 86–7 Kaposi’s sarcoma 1171–2, 1172 DaunoXome 86 da Vinci master-slave 1430, 1430
DCC (deleted in colorectal cancer) gene 670–1 D-dimers 1204 deafness 362, 368 death 1301–2 child cancer patient 1154 place of 1291 stigma of 1306 death rattle 1302 decision-making computer aided 1363 head and neck cancer 343 patient-centred 1313 radiotherapy 1234 treatment plans 1189 decitabine 104, 1013 decoy receptors 257 defibrotide 1106 DeGramont chemotherapy, modified 683 dehydration 1302 delirium 1301 delto-pectoral flap 350 demeclocycline 1203 demyelination, after brain irradiation 299 dendritic cells (DCs) 168, 168, 174, 895 dendritic cell vaccines 174–6 Denekamp, Juliana 258 denileukin deftitox 860, 1008 Denmark, hepatocellular carcinoma 559 dental care prior to oral brachytherapy 1220–1 radiotherapy for head and neck cancer 361, 361 deoxynucleoside analogues, AML 988 depression 1300–1 cancer patient 1307 depsipeptide 987, 1013 dermatitis, exfoliative 1204 dermatitis herpetiformis 1204 dermatomyositis 40, 507, 1203 desferiprone 1012–13 des-γ-carboxyprothrombin protein (PIVKA-II) 562 desmoid tumours 962–6 aetiology and pathogenesis 962–3 epidemiology 962 natural history 964 pathology 962 presentation and diagnosis 963–4 treatment 964–6 desmopressin (DDAVP) 449 developed countries, Hodgkin’s lymphoma 1028 developing countries head and neck cancer 345 hepatocellular carcinoma 559, 559 Hodgkin’s lymphoma 1028 life expectancy 5, 5 retinoblastoma 321 Dexa-BEAM regimen 1037 dexamethasone 1297, 1298 anorexia-cachexia syndrome 1198
gastrointestinal obstruction 1198 multiple myeloma 1084, 1089–90 raised intracranial pressure 1200 spinal metastases 492 dexrazoxane 1196, 1417–18 dextrose infusion 1203 DHAP regimen 1053 diabetes insipidus 323, 329, 449, 450, 868, 869 diabetes mellitus 574, 575, 580 diagnosis of cancer 10, 1423–4 average age at 15 communication with patient 1308, 1310–11 future innovations 16, 16 patients’ reactions 1307, 1308 diamorphine 1302 diarrhoea 1297 after pelvic radiotherapy 790, 791, 1418 carcinoid syndrome 453 in stem cell transplantation 1105–6, 1106 diazepam 1294 diazoxide 452 diclofenac 1291–2 didanosine 1175 diet and cancer prevention 7, 7, 505, 1198, 1330, 1438 and cancer risk 8, 15, 1433–6, 1439–40 breast cancer 1434, 1435 colorectal cancer 667, 1434, 1435 endometrial cancer 764 head and neck cancer 346 nasopharyngeal cancer 388 oesophageal cancer 539, 540 pancreatic cancer 573–4 prostate cancer 645, 646–7, 1435 thyroid cancer 409, 410–11 fat 1434–5 fibre 1433–4 iodine 409, 411–12 UK national cancer plan targets 1386–7 see also nutrition; nutritional support diethylstilbestrol (DES) 796 differentiation antigens 170 diffuse idiopathic skeletal hyperostosis (DISH) 838 diffuse large B-cell lymphoma (DLBCL) 1048–53 cutaneous 865 epidemiology 1048 expression profiling 29–30 histological subtypes 1048–9 HIV-associated 1054 management algorithm 1051 paediatric 1141–2 prognostic factors 1049–50, 1049 radio-immunotherapy 207 treatment advanced-stage disease 1050 central nervous system disease 1052
Index 1457
in elderly patients 1052 limited-stage disease 1050–2 relapsed disease 1053, 1053 diffuse neuroendocrine system (DNES) 438 diffusion tensor imaging 293 digital rectal examination (DRE) 649–50 dihydrocodeine 1292 dihydropyridimine dehydrogenase (DPD) 85, 685 dilatation 153–6 oesophagus 153, 553–4 venous 154, 156, 157 see also stenting diltiazem 1294 dimercapto succinic acid, pentavalent (VDMSA) 414 dioxins 930 diplopia 368 DISC-HSV (disabled infectious single cycle-herpes simplex virus) 177 disseminated intravascular coagulation (DIC) 1197, 1199 diuretics 1082, 1202, 1203, 1297 dl1520 virus (ONYX-015) 273, 273 DLBCL, see diffuse large B-cell lymphoma DMXAA (5, 6-dimethylxanthenone-4acetic acid) 258, 259, 259 DNA naked injection 274 structure and replication 242 DNA alkylators see alkylating agents DNA damage, ultraviolet light 879 DNA flow cytometry, soft tissue sarcomas 937–8 DNA methylation 103–4, 1013 DNA methyltransferase inhibitors 104 DNA-repair 26 enhancement in drug resistance 96–7 nucleotide excision repair 838 ovarian cancer 749 paediatric cancers 1120 docetaxel bladder cancer 632 carcinoid tumours 453 gastric cancer 611 mechanism of action 88 non-small cell lung cancer 517, 518, 519, 520, 520 ovarian cancer 754 pharmacokinetics 88 prostate cancer 656, 658 soft tissue sarcoma 961 toxicity 88–9 dolastatin 10 90 domiciliary care, child cancer patient 1155 domperidone 1298 donor, stem cell graft 1099–101 donor lymphocyte infusions 195, 1004 dopamine agonists, pituitary tumours 440, 442, 443, 448 Doppler effect 113
Doppler ultrasound 113, 126, 751–2 dose intensity 77, 78 Down’s syndrome 40, 706, 707, 982, 1120, 1121 risk of acute myeloid leukaemia 1124 doxazosin 455 doxorubicin administration and dose 86 ALL 993, 994 AML 986 bladder cancer 623 cardiac toxicity 1195–6, 1417–18 endometrial cancer 771 Ewing’s tumour 1131, 1131–2 gastric cancer 611 hepatocellular carcinoma 566 Hodgkin’s lymphoma 1032, 1033 islet cell tumours 568 liposomal 86–7, 432, 861 Kaposi’s sarcoma 1171–2, 1172 non-melanoma skin cancer 852 non-small cell lung cancer 517 ovarian cancer 753 paediatric osteosarcoma 1134–6 pancreatic cancer 579 primary bone tumours 915–16, 916, 918 soft tissue sarcomas 952, 954–6, 961, 962 thyroid cancer 428–9, 432, 451 toxicities 86 see also chemotherapy regimens containing doxorubicin drinks, high-temperature 539 drooling 1296 drug development 11, 11–13 building regulatory assets 1412–13 bulk drug and manufacturing 1408 controlling 1413 costs of 11, 18, 1411, 1413 formulation 80, 1408 future of 13, 13, 16–17 impact of translational science 1409–10 molecular-targeted agents 30, 34–5 multidisciplinary process 1407–8 oncology guidelines 1409 paediatric drugs 1119 phases 12–13, 12 post-marketing approval 1412 regulatory issues 1411–12 screening of agents 79–80 side effects reporting 1411 toxicology and pharmacological studies 80, 1409 see also clinical trials drug resistance 77, 90–7 altered target expression 94–5 control of intracellular drug concentration 92–3 cytokinetic 90–1, 91 drug metabolism 94 enhanced DNA repair 96–7 future directions 97 multi-drug 93
tumour cell apoptosis 95–6 tumour microenvironment 91–2 dry mouth (xerostomia) 1295–6 DTIC, see dacarbazine Dukes staging, colorectal cancer 672, 673 Durie/Salmon staging system 1081 dyes, photosensitive 295 dye-stuffs 615 dynamic contrast enhancement-MRI (DCE-MRI) 34–5 dysembryoplastic neuro-epithelial tumour 308 dysgerminomas 757 dyspepsia, gastric cancer 603–4 dysphagia 1296 head and neck cancer 368 lung cancer 506, 506 oesophageal cancer 540, 550, 552 palliation 554–5, 1296 secondary to radiotherapy 425, 514 dysplasia-associated lesion/mass (DALM) 668–9 dyspnoea advanced cancer 1294–5, 1300 lung cancer 506, 506 parenchymal lung disease 1190–1 E1A gene 273 E7070 101 EAP regimen, gastric cancer 611, 612 ear tumours 395–7, 846 Eastern Europe 345, 614 EBVP regimen 1032 E-cadherin 626 E-cadherin gene (CDH-1) 603 ECF regimen 610, 612 ECMV regimen 523, 523 economics of cancer cost of new therapies 1360–2 new European states 1356–8 USA 1353–6 ecteinascidin-743 (ET-743/trabectidin) 90, 961 education, child with cancer 1153, 1155 Edwards’ syndrome 811 effusions malignant pleural 137, 492, 1193, 1194 pericardial 492 EGFR gene 470, 476 Egypt 615 EICESS 92 trial 1132–3 elderly patients communication with 1308 DLBCL treatment 1052 lung cancer treatment 519–20 electrocautery, bladder tumours 621 electrodissection, basal cell carcinoma 845 electrolarynx 364 electron-beam therapy, pinnal tumours 397 electronic portal imaging devices (EPIDs) 1270
1458 Index
e-mail communication, patients 1359 EMA regimen 823–4, 824, 825 embolism, tumour, gestational trophoblastic tumours 825–6, 826 embolization, vascular 158–62 agents used 158 bone tumours 159, 160 hepatic artery 160–2, 161, 454, 564–5 portal vein 590 renal carcinoma 159–60, 160 emergency situations 1299–300 endobronchial radiotherapy 513–14 endocrine deficiency, radiation-induced 361–2 endodermal sinus tumours, vagina 799 endoluminal imaging, MRI 122 endometrial cancer aetiology 764–5 chemotherapy 771 co-morbidity 135 diagnosis and staging 135–6, 136 epidemiology 135, 764 fertility preservation 136 follow-up 772 future treatment prospects 772 hormonal therapy 771 investigation 766 palliative treatment 771 pathology 765–6 radiotherapy 767–71 risk factors 477 staging 766, 766 surgery 766–7 surveillance and recurrence 136 symptoms 766 treatment of recurrence 772 young patients 771–2 endoscopic mucosal resection (EMR) 550 endoscopic retrograde cholangiopancreatography (ERCP) 153, 576, 588–9, 589, 594 endoscopic ultrasound (EUS) 113 cholangiocarcinoma 588, 589 gastric cancer 605 pancreatic tumours 576 endostar (rh-endostatin) 258 endostatin 254, 256, 257–8 recombinant (endostar) 258 endothelin receptor antagonists 658 Enneking staging system 1134, 1134 enucleation ocular melanoma 333, 333 retinoblastoma 322 ENV gene 271 environmental exposures bladder cancer 615 cholangiocarcinoma 585 head and neck cancers 346–7, 346, 393 multiple myeloma 1074 renal cell carcinoma risk 733 soft tissue sarcomas 929–30 environmental factors, head and neck cancers 346, 346
Enzastaurin 301 EORTC study 656 ependymoma brain 307 myxopapillary 307, 315 paediatric 1119, 1144–5 spinal 315 EphA4 tyrosine kinase receptor 857 ephrin-B2 254 ephrin-B4 254 ephrin ligands (A/B) 254 EPIC study 7, 1434 epidemiology of cancer 2–4 and wealth 5–6, 6, 6 epidermal growth factor (EGF) 647 epidermal growth factor receptor (EGFR) 228, 233, 233 antibodies colorectal cancer 680, 683, 683, 685 pancreatic cancer 580 bladder cancer 615, 622 and drug resistance 95 expression in tumours 46 glioblastoma 290 inhibition, monoclonal antibodies 1286 inhibitors 95, 454 lung cancer therapy 520–1 inhibitors/antibodies 46–7 ligands 233, 233 monoclonal antibodies 32 monoclonal antibodies targeting 228–9, 233–9, 234 overexpression in gastric cancer 604 targeted therapies 1283 tyrosine kinase inhibitors 31–2, 99, 348, 366 epidermodysplasia verruciformis 836 epidermolysis bullosa dystrophica 836 epidophyllotoxins 300, 312 epigenetic therapies 103–4 epilepsy 302 epinephrine:norepinephrine ratio 430 epirubicin 86 bladder cancer 622, 623, 624 gastric cancer 611 soft tissue sarcoma 955, 956 epithelial-cell adhesion molecule (EGP-2) 276 epithelioid angiomatosis 1170 epithelioid trophoblastic tumours (ETT) 812, 817 epitheloid haemangioendothelioma 562 epitope dominance 174 EPOCH regimen 1054, 1175, 1176 epratuzumab 1009 Epstein-Barr virus (EBV) 7, 1168 adoptive immunotherapy 195 CNS tumours 288, 289 and gastric cancer 603 HIV-associated Hodgkin’s disease 1179 leukaemias 991 lymphoma 1028, 1045 Burkitt’s 1141
HIV-related 1173–4 nasopharyngeal cancer 347, 388, 391 oral hairy leucoplakia 376 equality, health/health care 14, 15, 20 ERCP, see endoscopic retrograde cholangiopancreatography erectile dysfunction 653 ERG gene 984 ergometrine 820 ergot alkaloids 440–1 erlotinib 95, 99, 256, 257, 580, 592 non-small cell lung cancer 521 renal cell cancer 741–2 erythema gyratum repens 1203 erythrocyte sedimentation rate (ESR) 1078 erythroplasia 377 erythropoietin (EPO) 49, 354–5, 372, 1012–13, 1083, 1204 ESHAP regimen 1053 estramustine 658 etamsylate 1298 etanidazole 354 ethanol injection coeliac plexus block 162, 162 liver tumours 152, 564, 565–6 ethics 1313 ethmoid sinus anatomy 392 tumours 394, 395, 398 ethnic factors gastric cancer 602 genetic cancers 1323 hydatidiform mole 812 inherited cancers 1330 multiple myeloma 1073–4 ovarian cancer 749 paediatric cancers 1119 pancreatic cancer 573, 573 primary bone tumours 906 etianidazole 48 etoposide AML 986 Ewing’s tumour 1131, 1132 gastric cancer 611, 612 gestational trophoblastic tumours 823–4, 824, 825 Langerhans cell histiocytosis 869 lung cancer 516, 517 mechanism of action 89 metastatic carcinoid tumours 568 oesophageal cancer 552 paediatric germ cell tumours 1138–9 paediatric glioma 1144 paediatric osteosarcoma 1134 small-cell lung cancer 521–3, 522 stem cell transplantation 1102 toxicity 89 see also BEP regimen; JOE regimen EUROCARE study 1379 EUROCARE 2 study 1380 European Carotid Surgery Trial (ECST) 1343–4, 1344
Index 1459
European Medicines Agency (EMEA) 1409, 1412 European Organisation for Research and Treatment of Cancer (EORTC) 621 European Osteosarcoma Intergroup study 1135 European Prospective Investigation into Cancer and Nutrition (EPIC) Study 7, 1434 European Randomised Trial of Screening for Prostate Cancer (ERSPC) 646 European Union, new member states 1356–8 Eusol 802 everolimus 454 evidence-based medicine, see clinical trials Ewing’s tumours 917–20 aetiology 906 chemotherapy 918–19 clinical features and diagnosis 917–18 extra-osseous 926, 928, 931 incidence 905–6 investigations 918 paediatric incidence 1119, 1119 management 1130–3 prognostic factors 918–19 radiotherapy 919–20 staging 908 surgery 919 treatment decisions 918 EWS-FLI-1 fusion gene 928 EWS gene 1130–1 excision cross-complementing 1 (ERCC1) gene 96–7 exemestane 473 exenteration orbit 336 pelvic 675 vaginal 798 exosomes 175, 178 extended field radiotherapy (EF-RT) 1031–2, 1033 external auditory canal 396 extracellular matrix (ECM) 291 extra-corporeal photophoresis 861 eye choroidal metastases 506 enucleation 322, 333, 333 radiotherapy 323–5 injury 324, 324, 327, 332, 359, 362–3, 847, 848 tolerance 42 tumours, see ocular tumours eyelid tumours adenocarcinoma 334 basal cell carcinoma 334, 847 benign 330–1 melanoma 335, 336 squamous cell carcinoma 335 FA, see folinic acid facial oedema 351
facial transplants 352 faecal occult blood test 132, 673 fallopian tube cancer 758–9 familial adenomatous polyposis (FAP) 574, 668, 670, 962, 963, 1127, 1329 genetics 670 prevention of cancer 1437 screening and management 1331 surgical resection 674 see also Gardner’s syndrome familial atypical mole/malignant melanoma syndrome (FAMMM) 574 familial breast-ovarian cancer syndrome 1/2 1319 familial clustering 1318 familial disease, myeloma 1074 familial isolated hyperparathyroidism (FIHP) 449 familial medullary thyroid carcinoma (FMTC) 412, 412, 430, 450 familial polyposis coli (Gardner’s syndrome) 458 family bereavement 1302–3 communication with 1291 support 1300 family history bladder cancer 615 breast cancer 468 gestational trophoblastic disease 813 Hodgkin’s lymphoma 1028 malignant melanoma 880, 881 ovarian cancer 748, 749 pancreatic cancer 574 prostate cancer 646 testicular cancer 706, 707 family trees 1325, 1329, 1329 FAM regimen 611, 612 FAMTX regimen 610–11, 612 Fanconi anaemia 40, 982, 1120, 1121, 1203, 1321 FAP, see familial adenomatous polyposis Far East cholangiocarcinoma 585 head and neck cancers 345 hepatocellular carcinoma 559, 559, 560, 562 hydatidiform mole incidence 812 nasopharyngeal cancer 388, 389 see also named countries farnesyl transferase (FT) inhibitors 32, 99–100, 580, 987, 1013 Fas ligand (FasL) 278, 279 fast neutron therapy 51, 354, 958 fat, dietary 645, 667, 764, 1434–5 fatigue 1298 FC receptors (FcRs) 227–8 Federation of Gynaecology and Obstetrics (FIGO) 112 feminization 458 femoral head, avascular necrosis 791 fentanyl 1292
fertility after endometrial cancer 136, 771 after Hodgkin’s lymphoma 1038–9 following gestational trophoblastic disease 828 paediatric cancer survivors 1152 fibrates 881 fibre, dietary 574, 667, 1433–4 fibrinolytic agents 1194, 1205 in MI, clinical trials 1344–5, 1345, 1348–9, 1349 fibroblast growth factor (FGF) 647 acidic 622 fibroblasts prostatic stroma 647 radiosensitivity 40–1 fibrosarcoma 925 paediatric 1119 fibrosing alveolitis 503 fibrosis mediastinal 550 radiation 361, 391, 514, 791, 1220 subcutaneous 945, 945, 1372 submucous 377 fibulin 254 fiducials 1274 field cancerization 347 FIGO, see International Federation of Gynaecology and Obstetrics (FIGO) finasteride 647, 1438 fine needle aspiration cytology (FNAC) breast lesion 147 cervical lymph nodes 125 gall-bladder cancer 594 lymph nodes 884 soft tissue sarcoma 934 thyroid 413, 414 fine-needle biopsy, liver 562 finger clubbing 507 fish, salted 388 fistula after cervical cancer surgery 790 broncho-oesophageal 541, 550 chylous 351 rectovaginal 798 tracheo-bronchial 541 tracheo-oesophageal 153 vaginal 701 vesico-vaginal 798 FKHR transcription factor 1125 flap reconstruction 350–1, 377 flavonoids 258 flavopiridol 101, 1009 flexible sigmoidoscopy, colorectal cancer screening 673 Fli-1 gene 1130 flow cytometry, CLL 1007 FLT3 gene 984, 993 FLT3 receptor tyrosine kinase, inhibitors 988 fluconazole 1295
1460 Index
fludarabine 861, 895 adverse effects 1007 leukaemias 986, 1007, 1124 NHL 1059, 1061, 1062 resistance to 232 fluid-retention syndrome 88 fluids, advanced disease 1302 fluorescence in situ hybridization (FISH) 962 fluoropyrimidines 85 see also 5-fluorouracil; capecitabine 5-fluorouracil (5-FU) anal cancer 698–9, 699, 700–1, 1179 bladder cancer 630, 631 cardiac toxicity 1196 colorectal cancer 256, 677, 678, 682–5, 683–4 combination with radiotherapy 1283–4 endocrine tumours 452 gall-bladder cancer 595 gastric cancer 610–11, 610 head and neck cancer 358 islet cell tumours 568 mechanism and pharmacology 85, 684 non-melanoma skin cancer 852 oesophageal cancer 552, 553 pancreatic cancer 577, 578, 579 topical 840, 849 toxicity 85, 684 vulval cancer 804 fluoxetine 1301 flushing 1204 flutamide 657, 658 FNAC, see fine needle aspiration cytology folate deficiency 1204 FOLFIRI regimen 683, 684 FOLFOX regimen 683–4 FOLFOX4 regimen 240, 256 folic acid 84 dietary intake 1436 folinic acid (FA) adverse effects 684 colorectal cancer 677, 682–5, 683 gall-bladder cancer 595 gestational trophoblastic tumours 823–4, 823, 824, 828 HIV-associated primary CNS lymphoma 1177–8 pancreatic cancer 577 follicle centre lymphoma (FCL) 336 follicle stimulating hormone (FSH) 439 follicular lymphoma epidemiology 1056–7 histological grading 1057 management algorithm 1060 prognostic factors 1057 treatment newly diagnosed disease 1057–9 radio-immunotherapy 206–7 relapsed disease 1059, 1061 follow-up breast cancer 129, 493, 1374
colorectal cancer 685–6 effectiveness, audit 1374 gestational trophoblastic disease 821, 827, 827 long-term 1419 ovarian cancer 755 protocols 1431 survivors of cancer 1417, 1419, 1431 testicular cancer 1374 Food and Drug Administration (FDA) 1411, 1412 foot, soft tissue sarcoma 946, 946, 953 foreign body, intravascular 156, 158, 158 foscarnet 1108 fotemustine 894 fowlpox, vaccines 177 fractures, radiotherapy-induced 911 fragile histidine triad (FHIT), lung cancer 504, 504 France breast cancer outcomes 1355 nasopharyngeal cancer 345 Frank model 1324 freckling, severe 838, 839 free fibular flap 351, 381 French-American-British (FAB) group leukaemia classifications 981 ALL 991–2, 991, 1121 AML 983–4, 983 myelodysplastic syndromes classification 1011–12, 1012 frontal sinuses 392 frozen shoulder 351 fruit and vegetable consumption 505, 539, 540, 1435–6 and colorectal cancer 667 and gastric cancer 603 and pancreatic cancer 574 fumarate hydratase gene 732 functional loss, spinal tumours 314 funding of cancer care 14, 20–1 fungal infections, post stem cell transplant 1108–9 furosemide 1202, 1203, 1297 fusion genes 927–8, 984, 992–3, 996–9 fusion proteins 860 future of cancer care barriers to innovation 17–18 delivery services 16, 21 diagnostics 13–14, 16, 16 drug treatments 13, 13, 16–17 economics 14 patient experience 18–19 prevention and screening 15 professional reconfiguration 19 gabapentin 1083, 1294 gadobenate dimeglumine 121 gadoxetate 121 GAG gene 271 galectin-3 gene 447 gall bladder, calcification/porcelain 592
gall-bladder cancer aetiology and risk factors 592–3 clinical presentation 594 differential diagnosis 594 epidemiology 592 future trends 595 investigations 594 metastasis and natural history 593 molecular biology and genetics 593 pathology 593 staging 593, 594 treatment 594–5 gallium nitrate 1202, 1300 gallium scan 123 gallstones, gall-bladder cancer risk 592 gamma camera 122 gamma knife radiosurgery 333, 448, 893 gamma values, radiotherapy 42–3 ganciclovir 895, 1108 gangliocytoma 308 ganglioglioma 308 ganglioside GM2 antigen 888, 889 Gardner’s syndrome (familial polyposis coli) 40, 409, 413, 458, 670, 962–4, 1127, 1203 genetics 927 garlic 667 gastrectomy 606–8 gastric cancer aetiology 603 chemo-prevention 1437 chemotherapy 610–11 adjuvant 607, 611–12 in advanced disease 610–11, 610 neoadjuvant 612 resistance 605 diagnosis 603–4 and dietary factors 1434 incidence 602–3, 602 molecular biology 604–5 pathology 604 prevention 7 radiotherapy 608–10 staging 605–6, 606 surgical treatment 606–8 gastric lymphomas 1061–2 gastric outlet obstruction 153 gastric transposition 382 gastrinoma 451 gastro-entero-pancreatic (GEP) endocrine tumours 451–2 gastrointestinal stromal tumours (GISTs) 98, 925, 929 treatment 960 gastrointestinal tract lymphoma 137, 1061–2 obstruction 1198–9, 1198 advanced cancer 1297 stent placement 153, 155, 675, 1198 perforation 242 radiation tolerance 42
Index 1461
gastro-oesophageal reflux disease (GERD) 539, 540 gastrostomy head and neck cancer 350 percutaneous 152–3, 153, 1197, 1198 GATA-1 gene 1124 ‘gatekeeper’ genes 1322 G-CSF, see granulocyte colony-stimulating factor gefitinib 34, 95, 99, 348, 366, 454, 458, 658 non-small cell lung cancer 520–1 gemcitabine 1283 cholangiocarcinoma 592 Hodgkin’s lymphoma 1035 mechanism of action 86 non-small cell lung cancer 517, 518, 519, 520, 520 pancreatic cancer 578, 579–80, 579 primary cutaneous T-cell lymphoma 861 renal cell cancer 736 soft tissue sarcomas 961 toxicity 86 gemcitabine/cisplatin (GC) 632, 632 gemfibrozil 861 gemtuzumab ozogamicin 230, 987, 989, 1123 gender differences bladder cancer incidence 614 floor of mouth tumour 378 head and neck cancer 344–5 lung cancer incidence and mortality 502, 502, 503 microprolactinomas 442 oesophageal cancer 539 oropharyngeal tumours 379–80 renal cell cancer 729, 729, 730 gene-directed prodrug therapy 277–8, 278 gene expression profiling 29–30, 97 breast cancer 30 multiple myeloma 1081 soft tissue sarcomas 929 gene gun immunization 176 General Medical Council, on communication with patient 1305 general practitioner (GP), referrals 1372 genes, causing cancer, see cancer predisposition genes gene silencing 276–7, 277 gene therapy 18, 105 brain tumours 301 chronic lymphocytic leukaemia 1008–9 clinical trials 269, 278, 279 colorectal cancers 686 corrective 276–7, 277 cytoreductive 277–80 head and neck cancer 369–70 immunomodulatory 279–80 indications 269, 269 interventional radiology 163 malignant melanoma 895 non-viral vectors 274
transciptional targeting 275, 275 transductional targeting 275–6 viral vectors 269–70, 270 adeno-associated virus 272 adenoviruses 270–1, 270, 271 alphaviruses 272 herpes simplex virus 270, 272 oncolytic 272–4, 273 retroviruses 270, 271–2, 271 genetic counselling 1334 genetic profiling 1424 genetics clinics 1328–9 genetic testing 1330–4, 1331–2 children 1334 social implications 1334 thyroid cancer 430 genistein 346, 1438 genome-wide expression profiling, NHL 1049–50 genomic imprinting 813 genomic technology 13 geographical variations 1433 bladder cancer 614 cholangiocarcinoma 585 gall-bladder cancer 592 gastric cancer incidence 602–3 head and neck cancers 345 hepatocellular carcinoma 559, 559 hydatidiform mole 812 lung cancer 502 non-Hodgkin’s lymphoma 1045, 1046 paediatric cancers 1119 hepatocellular carcinoma 1127 prostate cancer 645 testicular germ-cell tumours 706 xeroderma pigmentosum 838 GEP tumours, see gastro-entero-pancreatic (GEP) endocrine tumours Germany 559 germ-cell tumours 706 central nervous system, incidence 287 cytogenetics 707, 1138 extra-gonadal 712, 713, 724–5, 724 lung metastases 1190–1, 1191, 1193 ovarian 750, 756–8 paediatric 1138–9 incidence 1119 pathology 707–8 pinea 313–14 retroperitoneal 725 testicular, see testicular germ-cell tumours ureteric obstruction 1196 germinoma 313, 1138–9 gestational trophoblast, normal 809, 809 gestational trophoblastic disease (GTD) chemotherapy indications for 821–2, 821 regimens 823–4 complications acute 825–6 long-term 828
contraceptive advice 827 drug-resistant disease 818, 825 epidemiology and aetiology 812–14 epithelioid trophoblastic tumours 812, 817 first described 809 follow-up 827, 827 hydatidiform mole complete 810–11 evacuation 820–1, 820 partial 811 investigation genetic analysis 819–20 imaging 817–19, 818–19 placental-site trophoblastic tumour 812, 813, 826–7 prognostic scoring and staging 822–3, 822 giant-cell astrocytoma 307 gigantism 443–5 GISTs, see gastrointestinal stromal tumours Glasgow Coma scale, children 1150 Gleason score 1428, 1429 and selection for brachytherapy 1229 Gleason score (prostate cancer) 647–8, 648, 651, 652 gliadel 300 glioblastoma 304–6 imaging 294, 294 incidence 287 molecular biology 291 molecular therapy 301, 302 treatment and outcomes 296, 297, 305–6, 305, 306 tumour heterogeneity 291 glioblastoma multiforme 218, 670, 1144 glioma 288 brachytherapy 298 brain-stem 307–8 characteristic pattern 292, 292 chemotherapy 300, 301 imaging 120, 293–4, 294 molecular biology 290–1, 290 optic nerve 308, 328–9 paediatric 1143–4 radio-immunotherapy 218 gliomatosis cerebri 308 Glivec 301, 302 glomerulonephritis, membranous 507 glomus vagale tumours 397 glossectomy partial 377 total 351, 380 glucagonoma 451 glutathione reductase gene 841 glutathione-S-transferase 647 glycopyrronium 1296, 1302 glycosylation, altered 171 GM-CSF, see granulocyte-macrophage colony-stimulating factor GNAS gene 458
1462 Index
goitre endemic 409, 412 nodular 412–13, 413 gold-grain implant brachytherapy 326, 327 Goldie-Coldman hypothesis 77, 90 gonadotrophin-releasing hormone (GnRH) 443, 473, 771 Gorlin’s (basal cell naevus) syndrome 758, 836–8, 1146, 1321, 1329 screening/management of at-risk individual 1331 gossypol 458 gp100 antigen 170, 176 graft-versus-host disease (GvHD) 195, 1062 grading 1109, 1110 propylaxis 230, 1104 risk factors 1109, 1109 graft-versus-leukaemia (GVL) effect 195 graft-versus-tumour (GVT) effect 1062 granisetron 1297, 1299 granulocyte colony-stimulating factor (G-CSF) AML 986 bladder cancer 632 and gestational trophoblastic disease 814 lung cancer chemotherapy 525 myelodysplastic syndromes 1012 in neutropenic sepsis 1205–6 osteosarcoma therapy 916, 916 soft tissue sarcoma therapy 956, 962 granulocyte-macrophage colonystimulating factor (GM-CSF) biological effectsl 190 gene therapy 280 melanoma cells transfectted with 895 soft tissue sarcoma therapy 962 therapeutic potential 190 granulomatous slack skin disease 862 granulosa cell tumours, ovary 758 GRFoma 451 groin lymph nodes, see inguinal lymph nodes Group d’Oncologie Radiotherapié Tet et Cou (GORTEC) 1416–17 growth, child cancer patient 1151–2 growth factors expressed by Kaposi’s sarcoma cells 1169 gestational trophoblastic disease 814 haemopoietic 1012 lung cancer 510 prostate cancer pathogenesis 647 growth hormone (GH) deficiency 449 pituitary tumours 444–5, 444 gum tumours 379 Gynaecologic Oncology Group (GOG) 135 gynaecomastia 507, 709 H19 gene 813 HAART, see highly active antiretroviral treatment (HAART)
haemangioendothelioma 335 haemangioma, eye 330, 331 haemangiopericytoma 311, 335 haematoporphyrin 332 haematoxylin and eosin (H&E) staining 882 haematuria 733 bladder cancer 617–18 management in advanced disease 632–3 screening 620 cervical cancer treatment 791 haemoglobin levels multiple myeloma 1081 and radiotherapy 49 cervical cancer 788 head and neck cancer 354–5, 372 haemopoietic growth factors 525, 525 haemoptysis advanced cancer 1295 lung cancer 513–14 haemorrhage cervical cancer 789–90 chronic 1298 gestational trophoblastic disease 825 head and neck cancer 368 massive 1300 haemospermia 650 hairy cell leukaemia (HCL) 1010 halogenated purines/pyramidines 1282 haloperidol 1297 hand, soft tissue sarcoma 946, 953, 953 HCC, see hepatocellular carcinoma hCHK2 mutations 927 head, immobilization for radiotherapy 1235–6, 1236 headache 389, 455 brain tumours 292–3, 293 management 302 head and neck cancer aetiology 346–7 buccal mucosa 378–9 cervical lymph nodes 373–5, 373 chemo-radiotherapy trials 1283, 1286 chemotherapy 358, 359, 364–6, 365 costs of treatment 1355 developing world 345 esthesio-neuroblastoma 398 gene therapy 273 imaging 125–6, 126 incidence 344–5, 345 juvenile angiofibroma 398 larynx 383–8 lip 375–6, 376 management difficulties 343–4, 343–4 melanoma 885 Merkel-cell carcinoma 398–9 metastatic 365 monoclonal antibody therapy 234–6 mortality 344–5, 345 nasal/para-nasal sinuses 346, 392–5, 398 nasopharynx 388–92 neck dissection 350, 374, 374, 375
radical 351, 374, 374, 1373 novel therapies 369–70 oral cavity 376–8, 376 outcomes 345–6 palliative care 367–9, 368 paraganglioma 397–8 patient assessment 372–3 preventative strategies 347–8, 348, 349 prognostic factors 370, 372 quality of life 367 radiobiology 353–6 radiotherapy 352–64 brachytherapy 363 complications 356, 359–63 fractionation 44, 47, 354, 355–6, 357–8 gamma values 42–3 immobilization 1235–6, 1236 intensity-modulated 1268 patient advice 359–60, 360 planning 1241, 1242, 1242 postoperative 363–4 practical aspects 363 re-treatment 364 rare and unusual 397–400 rehabilitation 364, 364 second primary tumours 347 service organization 344 staging 370–2, 371–2 surgery, principles 349–52 syncope 389 tumour kinetics 352–3, 352, 353 healthcare professions reconfiguration 16, 19 training 14 health education, smoking cessation 504 health information, access to 1358–9, 1363 Health Maintenance Organization (HMO) 1355 health taxes 15 health technology assessment (HTA) 1366 health transition, global 5 hearing loss 362, 391 heart, radiation tolerance 42 heart attack, see myocardial infarction heat shock protein-gp96 peptide complex 892 heat shock proteins (Hsps) 102 inhibitors 102, 1003 vaccines 172–3 hedgehogs 254 Helicobacter spp. 592 Helicobacter pylori 7, 603, 1061 helium ion therapy 332–3 hemiglossectomy 377 heparin 1204, 1205 hepatectomy, see liver resection hepatic artery chemotherapy infusion 567 embolization 160–2, 161, 454, 564–5 hepatico-jejunostomy, Roux-loop 595 hepatic sarcoma 930 hepatitis, radiation 566
Index 1463
hepatitis-B (HBV) 7, 170, 559, 560 hepatitis-C (HCV) 559, 560, 585 hepatoblastoma, paediatric 1127 hepatocellular carcinoma (HCC) aetiology and risk factors 560, 560 clinical presentation 561 diagnosis 561–2 epidemiology 559–60 histology 562 imaging 114 natural history 562, 562 paediatric 1120, 1127–8 screening 562 staging and prognosis 562–3, 563 treatment 563 local curative 564–6 systemic 566–7 hepatocyte growth factor (HGF) 100, 647, 1075, 1076 hepatolithiasis 585 hepatomegaly 454, 561, 587 hepato-renal syndrome 1197 Her-2/neu (c-erbB2) 95, 98, 447, 504, 504, 586, 604, 616, 765, 916 herbicides, phenoxy 930 herceptin, see trastuzumab hereditary cancer syndromes 1318–20, 1319–20 clinial examination 1328–9 CNS tumours 288, 289 Li-Fraumeni syndrome 927 non-melanoma skin cancer 836–9 renal cell cancer 731–2, 732 soft tissue sarcomas 927 tumour suppressor genes 26 hereditary non-polyposis colorectal cancer (HNPCC) 26, 574, 669, 1319, 1322 clinical criteria 669, 669 clinical genetics referral guidelines 1327–8 endometrial cancer 765 screening 673–4, 1332, 1439 herpes saimiri virus 1168 herpes simplex virus (HSV) 347, 1295 DISC-HSV 177 as gene vector 270, 272 oncolytic 273–4, 273 thydmidine kinase gene/ganciclovir (GCV) system 278 vulval cancer 799 herpes zoster 1204 hexamethylmelamine 83 high-intensity focused ultrasound (HIFU) 114, 655–6 highly-active anti-retroviral therapy (HAART) 702, 1054, 1167, 1170, 1175–6, 1180 hilar adenopathy 1191 Hippocrates 3 Hirschsprung disease 1128 hirsuitism 1204 histiocytoma, malignant fibrous 906, 908, 920
histiocytosis, Langerhans cell 867–9 histiocytosis X 329 histological diagnosis 1424 histone deacetylases (HDACs) 987 inhibitors 104, 987, 1003, 1013 history of cancer treatment 3–4 HIV/AIDS infection 5, 7, 376, 707, 1298 cancer risk 503 deaths from 1167 HAART 702, 1054, 1167, 1170, 1175–6, 1180 HIV/AIDS-related malignancies anal cancer 696, 702, 1179 Castleman’s disease 1178 central nervous system 311, 311 cervical cancer 779, 1179–81 Hodgkin’s disease 1178–9 Kaposi’s sarcoma 193, 1168–73 leiomyosarcoma in child 1181 non-Hodgkin’s lymphoma 1053, 1054 primary CNS 1176–8 systemic 1173–6 non-Kaposi’s skin tumours 1181 primary effusion lymphoma 1176 squamous cell carcinoma of conjunctiva 1181 testicular germ cell tumours 1181 hMLH1 gene 749 hMSH2/hMSH6 genes 749 HNPCC, see hereditary non-polyposis colorectal cancer hoarseness laryngeal cancer 383–4 lung cancer 506 Hodgkin and Reed-Sternberg cells 1027, 1028, 1139 Hodgkin’s lymphoma aetiology 1028 clinical presentation 1029, 1029, 1030 diagnosis 1029–30 elderly patients 1038 epidemiology 1028 experimental therapies 1039–40 fertility 1038–9 first-line treatment 1027, 1031–8, 1032 HIV/AIDs-related 1178–9 long-term management/follow-up 1039, 1039 mixed cell (MC) 1028 nodular lymphocyte-predominant (NLPHL) 1028–9, 1028, 1032–3 nodular sclerosing (NS) 1028 paediatric 1119, 1119, 1139–40, 1417–18 pathophysiology 1028–9 prognostic factors and treatment choice 1030–1 relapsed/refractory 1027, 1036–8 renal impairment 1196 staging classification 1030, 1031, 1140 stem cell transplantation 1037, 1038, 1101 thyroid 431 Hodgkin, Thomas 1027
holistic care models 1363 home-based care 1357 Homer Wright rosettes 398 homoharringtonine (HHT) 1001 homozygosity analysis 813 honey 802 Hong Kong nasopharyngeal cancer 345, 388, 391, 392 paediatric hepatocellular carcinoma 1127 hope, cancer patient 1311–12 hormonal therapy breast cancer 473–7, 474 combined with chemotherapy 480 ‘early’ disease 1347–8, 1348 neuroendocrine tumours 568 prevention of cancers 1437–8 prostate cancer 654, 656, 657–8, 1280 hormone production ectopic 1201 eutopic 1201 hormone replacement therapy (HRT) after radical hysterectomy 792 breast cancer patient 492–3 and endometrial cancer 764, 771 menopause 748 orchidectomy 724 pituitary tumours 449 Horner’s syndrome 506 hospital admissions 1354, 1362 hospital volume 544, 1354, 1429 Ho system 389 Hounsfield, G.N. 114 Hounsfield units 114–15 HOX11 gene 993 HRAS1-VNTR gene 670 H-ras gene 99, 447, 604, 615, 851 HRPT2 gene 449 5-HT3 antagonists 1297, 1299 HuM195 anti-CD33 monoclonal antibody 216, 217 human anti-mouse antibody reactions, in radio-immunotherapy 208, 209 human chorionic gonadotrophin (hCG) 313 assays 814–15 ectopic secretion 1201 gestational trophoblastic disease 811, 815, 816 hydatidiform mole, post-evacuation levels 820, 820, 821 molecular background and function 814 paediatric germ cell tumours 1138 testicular germ-cell tumours 709, 710, 710, 717 human genome, knowledge of 1361 Human Genome Project 4, 8, 13, 97 human herpes virus-8 (HHV-8) 1054 genome 1168, 1168 molecular virology 1168–9 role in HIV-associated lymphoma 1174 role in Kaposi’s sarcoma 1169
1464 Index
human leukocyte antigens (HLA) compatibility of stem cell donor/recipient 1100–1 expression in tumours 179 human papillomavirus (HPV) 7, 170 anal cancers 696 cervical cancer/CIN 778, 1180 head and neck cancers 347, 394 non-melanoma skin cancer 836 oncogenic properties 1180 type 16 696 type 18 696 vaccines 784, 1181 vaginal carcinoma 796 vulval cancer 799 human placental lactogen (hPL) 812 human T-cell leukaemia virus type 1 (HTLV-1) 1055–6 human T-cell lymphotrophic virus (HTLV) 991 HuMax-CD20 antibody 231–2 Hungary 1356, 1357 Hüüthle-cell carcinoma 410, 411 hycanthone 579 hydatidiform mole 809 clinical features 815–16, 816 complete 810–11, 810 contraception following 827 epidemiology and aetiology 812–14 evacuation 820, 820 follow-up 821 gestational trophoblastic tumour after 814, 818, 818, 821 incidence and ethnic origin 812 invasive 811, 816 partial 810, 811 pregnancies mistaken for 811 twin pregnancies 816, 821 hydration, advanced cancer 1295, 1297, 1302 hydrocephalus, secondary to chemotherapy 323 hydrocortisone, replacement in pituitary tumours 449 hydromorphone 1292 hydronephrosis 1196 hydroxybutyrate dehydrogenase 710 5-hydroxyindole acetic acid (5-HIAA) 453, 568 11-α-hydroxylase 458 5-hydroxytryptamine (5-HT/serotonin) 453, 567 hyoscine 1296, 1297, 1299, 1302 hyperbaric oxygen therapy 48, 354, 1419 hypercalcaemia 1201–2 causes 1201–2 hyperparathyroidism 430 management 1202, 1300 metastatic breast cancer 492 multiple myeloma 1077, 1081, 1082 parathyroid adenoma/carcinoma 450 prostate cancer 650
symptoms 1202 hypercholesterolaemia 446 hypercortisolaemia 446–7 hyper-CVAD regimen 994, 996 hyperdiploidy 992, 993 hyperparathyroidism 430 primary 449–50 hyperparathyroidism–jaw tumour syndrome (HPT–JT) 449 hyperprolactinaemia 443 functionless pituitary tumours 443 macroprolactinoma 439–40 microprolactinoma 442 hypertension phaeochromocytoma 455 and renal cell carcinoma 732–3 hyperthermia 325, 332, 369 hyperthyroidism 1201 hypertrichosis lanuginosa (malignant down) 1203 hypertrophic pulmonary osteoarthropathy (HPOA) 1194 hypoglycaemia 1203 hypokalaemia 1298 hypomagnesaemia 1203 hypomethylating agents 1013 hyponatraemia 1202–3, 1202 hypopharynx anatomy 381 tumours 381–3, 381 hypothalamo-pituitary dysfunction 391 hypothalamus, glioma 1144 hypothyroidism 362, 861 hypoxaemia, chronic 397 hypoxia, see tumour hypoxia hypoxia-inducible factor (HIF) 252, 262, 731, 732 inhibitors 262 hypoxically-activated drugs 92 hypoxic cell sensitizers 48, 354, 1282, 1285 hysterectomy 134, 135 endometrial cancer 766–7 incidental finding of cervical cancer 789 radical in cervical cancer, complications 785, 790, 792 total abdominal 752 IAP (inhibitors of apoptosis proteins) 95 ibandronate 481, 491, 1202 ibritumomab tiuxetan 229 ICE regimen 82, 1053, 1053 idarubicin 86, 986, 1084 IFL regimen 256 ifosfamide Ewing’s tumour 1131, 1131–2 mechanism of action 82 NHL 1052 non-small cell lung cancer 517, 518 ovarian germ cell tumours 757 paediatric germ cell tumours 1139 paediatric osteosarcoma 1134 paediatric rhabdomyosarcoma 1125
primary bone tumours 915, 918 soft tissue sarcoma 952, 955, 956 metastatic 961–2 testicular germ-cell tumours 722 toxicity 82 ileal conduit 626 ileostomy 674 ileus 681 iliac lymph nodes 886 image-guided radiotherapy (IGRT) basic principles 1269–70 concept and rationale 1268–9 quality assurance 1275 strategies adaptive 1272 breath-hold gated 1272–3 predictive 1272 respiratory gated methods 1273–4 tracking systems 1274–5, 1274, 1275 imaging 10 future of 16 imaging treatment response 139–40 technological advances 1424–5 see also individual imaging modalities imatinib mesylate 388 combination with other agents 1001 in GISTs 960 in leukaemias 988, 994–5, 1000–1 mechanisms of action 98 resistance to 1001 in sarcomas 962 side effects 1000 in small-cell lung cancer 526 imiquimod 840, 849–50 imiquimod cream 801 immobilization, radiotherapy 946, 946, 1235–8, 1236, 1237 immune-modulation, gene therapy 279–80 immune reconstitution inflammatory syndrome (IRIS) 1170 immune response, to cancer cells 171 immunoediting, cancer 167 immunoglobulin heavy chain (IgH) genes 1075 immunological tolerance 171 immunosuppression 1167 brain tumours 311 cancer risk 503 and cervical cancer 779 and CNS tumours 289 infections 1205–6 lymphoma 1053 and non-melanoma skin cancer 836, 842 immunosurveillance theory 167, 186 immunotherapy 167–8, 658 AIDS-related Kaposi’s sarcome 1171 antigen-specific 168 autoimmunity onset 895 brain tumours 301 cutaneous T-cell lymphomas 860 Hodgkin’s lymphoma 1039–40 knock-out mouse models 186
Index 1465
leukaemias ALL 996 CLL 1008–9 CML 1002 limitations of 178 malignant melanoma 891–2, 894–5 non-specific 168 renal cell cancer 736–40, 737 small cell lung cancer 526–7 tumour escape 178–9 see also adoptive cell therapy; vaccines immunotoxins 229 imprinted genes, gestational trophoblastic disease 813 IMRT, see intensity modulated radiotherapy incidence of cancer global 4–5, 4 and wealth 5–6, 6, 6 incidentaloma 457 incontinence faecal 701 urinary 653 India 346, 376 indium-111-labelled monoclonal antibodies 135 indium-111-octreotide scanning 414, 451, 452 indium-111-pentreotide 452 indomethacin 881 industrial chemicals bladder cancer aetiology 615 cholangiocarcinoma aetiology 585 lung cancer aetiology 503 soft tissue sarcoma 930 inequality, health/health care 14, 15, 20 infant acute lymphoblastic leukaemia 1123 choriocarcinoma 817 Langerhans cell histiocytosis 868 neuroblastoma 1128–30 infections 1190, 1205–6 in advanced disease 1299–300 after embolization therapy 162 in cancer aetiology 7, 7, 8, 170 bladder cancer 615 cervical cancer 778, 1180 hepatitis B/C viruses 559, 560 leukaemia 991 MALT lymphomas 1061–2 multiple myeloma 1083 NHL 1045 spontaneous tumour regression 168 stem cell transplant recipient 1107–9 see also named infectious agents inferior vena cava (IVC) involvement in renal cell carcinoma 733, 735 leiomyosarcoma 958, 959 infertility, colorectal cancer therapy 681 inflammatory atrophy, prostate gland 647
inflammatory bowel disease 668–9, 1439 inflammatory cancer, breast 489 information child cancer patient 1155 patient 1358–9, 1363 information technology 10 NHS national system 1385 informed consent 1313 infusion pump, analgesia 1294, 1302 inguinal lymph nodes anal cancer 700 dissection 802 melanoma 885–6 vulval cancer 801, 802, 803 inguino-femoral lymph nodes 700 inheritance, cancer predisposition genes 1320, 1322 inherited cancer patients’ risk perception 1324–5 risk assessment 1323–4, 1323, 1324 inherited syndromes, renal cell cancer 731–2, 732 INI1/hSNF5 gene 1148 innovation, barriers to 17–18, 18 instrument dials, luminous 906 insulin 1203 insulin growth factor-1 (IGF-1) 1075 pituitary tumours 444–5 insulin-like growth factor (IGF-1) 647 insulin-like growth factor II (IGF-II), gene, GTT 813 insulinoma 451, 1203 insulin resistance 1198 α-2-integrin, bladder cancer 626 integrin inhibitors 31 integrin receptors 1169 intensity-modulated radiotherapy (IMRT) 123, 1418 anal cancer 700 basic principles 1260–2, 1261 clinical experience 1268 concept and rationale 1260 delivery methods 1264–7, 1265, 1266, 1267 implementation 1268 prostate cancer 656 quality assurance 1267–8 soft tissue sarcoma 949–50, 950 treatment planning 1262–4 intensive care, stem cell transplant recipient 1106–7 interferon (IFN) 188, 190, 191–3 combined with rituximab 231 multiple myeloma 1088 recombinant, small-cell lung cancer 523, 524 interferon-α (IFN-α) 191–3, 451 AIDS-related Kaposi’s sarcoma 1171 biology 190, 191 carcinoid tumours 453 clinical potential 190, 191–3 CML 1000, 1002
combined with cytotixic chemotherapy 740 combined with cytotoxic chemotherapy 891–2 combined with interleukin-2 (IL-2) 739 cutaneous T-cell lymphomas 858, 858, 860 gene therapy 280 head and neck cancer prevention 349 liver metastatic carcinoid disease 568 malignant melanoma 886–7, 887, 891–2 non-Hodgkin’s lymphoma 1055, 1056 pegylated forms 887, 1171 renal cell cancer therapy 736–7, 737, 739–40, 741 toxicities 737–8 interferon-β (IFN-β) 190, 191 interferon-γ (IFN-γ) 190, 191, 952–3 interleukin-1α (IL-1α) 189, 1075 interleukin-1β (IL-1β) 1075, 1076 interleukin-2 (IL-2) 193–4 biological role 189 combined with cytotixic chemotherapy 740 combined with interferon-α (IFN-α) 739 combined with rituximab 231 gene therapy 280 high dose therapy 737, 739 malignant melanoma therapy 891 renal cell cancer therapy 737, 738–9, 738 therapeutic potential 189, 193–4, 195 toxicity 738–9 interleukin-3β (IL-3β) 189 interleukin-4 (IL-4) 189 interleukin-6 (IL-6) 189 antibody 1090 cachexia induction 1197–8 HHV8-encoded homologue 1178 multiple myeloma 1075, 1076, 1081 receptor blocking agents 1178 receptor expression 1173 interleukin-7 (IL-7) 189 interleukin-10 (IL-10) 179, 189 interleukin-11 (IL-11) 189 interleukin-12 (IL-12) 189, 231 interleukin-15 (IL-15) 189 interleukin-18 (IL-18) 189 interleukin-21 (IL-21) 190 internal iliac lymph nodes 680 International Breast MRI Multicenter Consortium 129–30 International Committee for Harmonisation (ICH) 1412 International Federation of Gynaecology and Obstetrics (FIGO) endometrial cancer staging 766, 766 gestational trophoblastic tumours staging 822–3, 822 vaginal cancer staging 797, 797 vulval cancer staging 800, 800 International Prognostic Index 1049, 1049
1466 Index
internet 1363 interventional radiology biopsy procedures 147–50 definition 146 dilatation techniques 153–6 extraction techniques 156, 158, 158 and gene therapy 163 infusion techniques 158 percutaneous puncture and drainage 150–1, 151 percutaneous tumour ablation 151–2, 151, 152 range of procedures 146, 146 vascular embolization 158–62 venous sampling 150, 150 vertebroplasty 162, 163 intracranial pressure (ICP) raised 298–9, 302, 308 management 1200 intra-epithelial neoplasia anus 695 cervical (CIN) 777–8, 780, 781, 784–5 vagina (VAIN) 796–8 vulva (VIN) 800, 801 intussusception 1141 inverse planning 1257 inverting papilloma (Ringhertz tumour) 394 involved field radiotherapy (IFRT) 1031, 1033, 1034, 1035 iodine-123-meta-iodobenzylguanidine (MIBG) scintigraphy 451, 455 iodine-123-meta-iodobenzylguanidine (MIBG) therapy 451, 452, 455 iodine-125 brachytherapy 332, 655 iodine-131 imaging 414 iodine-131-tositumomab 207, 229 characteristics 211 radio-immunotherapy 207–10, 212–13, 1058 iodine, dietary 409, 411–12 5-iodo-deoxypyrimidine 1282 Ireland, smoking ban 504 iridium-192 implant, head and neck cancer 363, 375, 376 physical properties 1211 irinotecan (CPT-11) brain tumours 300–1 colorectal cancer 683, 684, 685 combined with 5-FU and FA 684 mechanism of action 89, 684 non-small cell lung cancer 517 paediatric CNS tumours 1147 resistance 94–5 toxicity 89 IRIS, see immune reconstitution inflammatory syndrome iris, melanoma 331 iron chelators 1012–13 ischio-rectal fossa 675 ISIS-2 trial 1341–2, 1342, 1346–7, 1346, 1348
ISIS-4 trial 1349–50, 1350 islet-cell tumours 451–2 chemotherapy 568 isolated limb infusion (ILI) 889–90 isolated limb perfusion (ILP) cytokines 194 malignant melanoma 194, 889 soft tissue sarcoma 952–3 Italy breast cancer outcomes 1355 hepatocellular carcinoma 559 itch, see pruritus IVE regimen 1053 JAK2 gene 1011 Janus kinase (JAK) 191 Janus kinase (JAK)/STAT pathway 857, 1075 Japan breast cancer outcomes 1355 gall-bladder cancer 592, 593 gastric cancer 602, 606 hepatocellular carcinoma 559 Japanese migrants 287 jaundice obstructive biliary-enteric bypass 591–2 cholangiocarcinoma 587, 588, 591–2 management 1199, 1199 pancreatic cancer 575, 576 stent placement 153–4, 156, 591, 592, 1199, 1199 post stem cell transplantation 1106 jaw cysts 837 JEB regimen 1138 jejunal grafts 351, 382 Jewish ethnic groups 287, 749 JOE regimen 323 John B Harvey 4 joint stiffness, after radiotherapy 945, 945 JunB 857 JunD transcription factor 456 Kadish staging system 398 Kaposi’s sarcoma 177 clinical features 1169–70 differential diagnosis 1170 epidemiology 1168 evidence for definition as true malignancy 1169 interferon therapy 193 novel therapies 1172–3 pathogenesis 1168–9 staging 1170, 1170 treatment 1170–2 treatment response evaluation 1171, 1172 Kenilworth model 1326 keratinocyte growth factor (KGF), human recombinant 1105 keratoacanthoma 851, 1329 ketamine 1294 ketoconazole 446, 447, 458, 1201, 1295
ketorolac trometamol 1292 keyhole limpet haemocyanin (KLH) 888 Ki-67, bladder cancer 626 Ki-67 levels, gestational trophoblastic tumours 812 Ki-367 antigen 447 kidney non-Hodkin’s lymphoma 137, 138 radiation tolerance 42 kidney tumours benign 731 see also renal cell cancer kilovoltage X-ray imagers 1270–1, 1271 kinase inhibitors 99 KIT transmembrane receptor 98 Klatskin tumours 586 Klinefelter’s syndrome 724, 1120, 1138 knock-out mouse models 186 Knudson’s two-hit hypothesis 26, 1322 Koch pouch 626 K-ras antigens, mutant peptide vaccines 173 K-ras gene 670, 984 biliary tract cancers 586, 593 endometrial cancer 765 lung cancer 504, 504 pancreatic cancer 574 Kupffer cells, imaging 121–2 kyphoplasty 17 labia majora 799 labia minora 799 lacrimal gland tumours benign 331 malignant 336 lactate dehydrogenase (LDH) 508 malignant melanoma 884 multiple myeloma 1078, 1081 non-Hodgkin’s lymphoma 1049, 1049 pleural effusion 1193 soft tissue sarcoma 931 testicular germ-cell tumours 710, 710 Lahey scoring system (Langerhans cell histiocytosis ) 868, 869 Lambert-Eaton syndrome 507 laminectomy 1200 lamotrigene 302 Langerhans cell histiocytosis 867–9 language 1307, 1309 see also communication lanreotide 451, 568 laparoscopy 1429–30 cervical cancer 792–3 cholangiocarcinoma, diagnosis 589 cholecystectomy 594 colorectal cancer 674–5 oesophageal cancer 542 pancreatic tumours 576 prostatectomy 653 lapatinib 34, 99 laryngeal cancer 346, 383–8 aetiological factors 346, 347, 388 assessment 384
Index 1467
cetuximab therapy 235 clinical featues 383–4 conservation of larynx 365–6 geographical variation 345 glottic carcinoma 385–7 incidence 344 laryngectomy 387–8 radiotherapy 355, 384–5, 387 staging 383, 383 subglottic tumours 387 supraglottic tumours 384 verrucous carcinoma 384 laryngectomy 380 following radiotherapy for laryngeal cancer 387–8 rehabilitation/speech restoration after 351, 364, 367 laryngoscopy, direct 384 laser therapy basal cell carcinoma 846–7 bladder cancer 620–1 bronchial obstruction 1190, 1190 colorectal cancer recurrence 675 gastric cancer 608 gastrointestinal obstruction 1198 laryngeal cancer 385 lung cancer 514 malignant melanoma 890 ocular melanoma 331–2 oesophageal cancer 554 vulval intra-epithelial neoplasia 801 Late Effects of Normal Tissue scales (LENT SOMA) 1416 late effects of therapy brachytherapy 1220 cardiac morbidity 1417–18 child cancer patient 1150–3 clinical studies 1419 data collection 1416–17 defined 1415 fertility 1152 grading and reporting 1415–16 management 1418–19 neurocognitive 528, 1153 organ toxicity 1152–3 radiotherapy 790–1 second malignancies 1418 child cancer survivor 1151 latissimus dorsi flap 381 laxatives 1297 LDH, see lactate dehydrogenase leiomyoma, eye 330 leiomyosarcoma aetiology 928 HIV-positive child 1181 relative frequency 925 retroperitoneal 958–9, 959 uterine 772, 961 vulva 805 lenalidomide 742, 1008–9, 1013, 1090 lens, radiation injury 359, 362–3 lentigines 838, 839 lentigo maligna 885
lentigo maligna melanoma 881, 885 lentiviruses 270, 272 Leser-Trélat sign 750 letrozole 473, 475, 475 leucoplakia 376 oral hairy 376–7 leucovorin (folic acid) 84, 256 leukaemias acute promyelocytic (APLs) 103 adult T-cell 1010 classification 981 hairy cell 1010 L1210 murine model 76 large granular lymphocytic 1010 paediatric 1121 acute lymphoblastic leukaemia 1121–3 incidence 1119 radio-immunotherapy 215–17 levamisole, colonic carcinoma 682 levator ani muscle, resection 675 LeVeen shunt 1199 levitracitam 302 levomepromazine 1297, 1301, 1302 Leydig-cell tumours ovary 758 testis 725 L’hermitte’s sign 514, 550 LHRH agonists 469 prostate cancer 656, 657–8 Libya 838, 839 lichen sclerosus 799 life expectancy 5 and wealth 5 lifestyle changes, cancer prevention 1330 lifestyles 15 Li-Fraumeni syndrome 26, 288, 457–8, 470, 906, 1120, 1133, 1319 cancer genetics referral guidelines 1327 sarcoma development 927, 928 limb oedema, cervical cancer treatment 791 limb salvage primary bone tumours 908, 915, 1134 soft tissue sarcoma 939, 941, 941, 952–3 limb weakness, brain tumours 293 LIMIT-2 trial 1349–50, 1350 linear accelerators (LINACs) 1245, 1254 imaging systems 1270–1, 1271 modulating (tracking unit) 1264 linkage analysis 813, 1322 lip, squamous cell carcinoma 839, 850, 852 lip carcinoma 375–6, 376 lipid-lowering drugs, and melanoma incidence 881 lipiodol 158 lipomas 933–4 lipoplexes 274 liposarcoma 925, 928, 931, 932, 933, 958 diagnosis 933–4 retroperitoneal 958–9, 958 liposomes 366 chemotherapy 86–7
cutaneous T-cell lymphoma 861 Kaposi’s sarcoma 1171–2, 1172 gene vectors 274, 274 vaccine delivery 177 lip tumours, brachytherapy 1226 Listeria monocytogenes, recombinant vaccines 177 lisuride 440 lithium carbonate 422 liver involvement in cholangiocarcinoma 587 involvement in non-Hodgkin’s lymphoma 137 MR imaging 120–2 percutaneous biopsy 149 radiation therapy 566–7 radiation tolerance 42 resection, see hepatectomy segmental anatomy 560, 560, 561 sinusoidal obstructive syndrome 1106 tumour embolization 160–2, 161, 454, 564–5 tumours, see hepatocellular carcinoma; liver metastases liver fluke infestation 585, 586 liver metastases carcinoid tumours 562, 567–8 colorectal cancer 130, 676, 685 down-staging 567 incidence 559–60 lung cancer 506 neuroendocrine tumours 567–8 percutaneous ablation 151–2, 151 surgical resection 567 symptoms and presentation 561 ultrasound imaging 114, 114 liver resection cholangiocarcinoma 590–1 gall-bladder cancer 595 hepatocellular carcinoma 564 metastases 567, 676 standard surgical procedures 560, 561 liver transplantation cholangiocarcinoma 591 hepatocellular carcinoma 564, 564 paediatric liver tumours 1127 local anaesthetics nebulized 1295 use in pain control 1294 LOD score 1322 lomustine (CCNU) 300, 306, 452 loperamide 790, 791, 1297 lorazepam 1294 loss of heterozygosity (LOH) 17p 290–1 adrenocortical tumours 458 bladder cancer 616 chromosome 2p 444 chromosome 17p 444, 458 CNS tumours 290–1, 306 colorectal cancer 670 lung cancer 504 MEN syndrome 456
1468 Index
loss of heterozygosity (LOH) (contd.) osteogenic sarcoma 1133 paediatric CNS tumours 1146 paediatric hepatocellular carcinoma 1127 pituitary tumours 444 Rb gene 906, 1133, 1134 see also allelic loss lovastatin 446 lower urinary tract symptoms (LUTS) 649 low molecular weight heparin 1204 lumiliximab 1009 luminous dial painters 906 lung biopsy, percutaneous 147, 148 lung cancer adenocarcinoma 510, 511 aetiology 503–4, 504 brain metastases 528 bronchial obstruction, management 1190, 1191 clinical features 505–6, 506 epidemiology 1354 HIV-positive patients 1181 imaging 126–8, 127, 128 incidence 502, 502 metastases, extra-thoracic 506 metastatic, ocular deposits 334 mixed cell type 510 mortality costs 1353 non-small-cell (NSCLC) advanced disease treatment 517–18 aetiology 503 bevacizumab therapy 240–1 chemotherapy 99, 514–20, 515, 516, 517, 519 cytokine expression 179 genetic factors 504 molecular targeted therapy 520–1 radiotherapy 512–14, 513, 514 staging and prognosis 127, 128, 508, 510 surgical treatment 511–12, 512 treatment of elderly patients 519–20 paraneoplastic syndromes 506, 507 pathology 508–10, 511 population screening 128 primary prevention 504–5 prognostic factors 508 radiotherapy immobilization 1236 planning 1242–3 risk, and diet 1436 screening 10, 505 small-cell (SCLC) ACTH syndrome 1201 aetiology 503 chemotherapy 521–7 combined-modality treatment 527–9, 527 genetic factors 504, 504 interferon therapy 192–3 pathology 510–11
radio-immunotherapy 218 staging 126–7, 129, 508 surgical adjuvant therapy 528–9 squamous cell 509–10 staging 126–7, 127, 128, 507–8, 508, 509 surveillance/recurrence 127–8 SVCO 1194–5 treatment costs 1355 lung fibrosis drug-induced 1191, 1193 radiation-induced 1193, 1193 lung infiltration, diffuse 1190–2, 1192, 1193 lung metastases choriocarcinoma 817, 1190–1, 1192 colorectal cancer 676, 685 germ cell tumours 1190–1, 1191, 1193 gestational trophoblastic disease 817, 818 management 825–6, 826 malignant melanoma 890 osteosarcoma 913, 914, 916–17 paediatric osteosarcoma 1134 renal cell cancer 733, 736 soft tissue sarcomas 960–1 testicular germ-cell tumours 711, 711, 712 lungs lymphoma 137 radiation tolerance 42 luteinizing hormone (LH) 439 lycopenes 647 lymphadenopathy imaging 120, 121, 126, 136 infectious 1030 reactive 1030 lymph node dissection 1425–6 gastric cancer 607–8 malignant melanoma 885–6 neck, see neck dissection lymphocele, pelvic 790 lymphoedema advanced disease 1299 arm in breast cancer 1373 prostate cancer 650 lymphokine-activated killer (LAK) cells 195 lymphoma adult T-cell 1010 bone 908, 920–1 central nervous system 287, 293, 311–12, 311 costs of treatment 1355 lymphoplasmacytic/lymphoplasmacytoi d (LPL) 336 MALT type 329, 336–7, 411, 412, 430–1, 431, 1061–2 ocular 333–4 primary cutaneous 853 B-cell 864–5 T-cell 853–64 WHO/EORTC classification 853, 854, 855
primary effusion (PEL) 1176 thyroid 411, 412, 414 management 430–1, 431 see also Hodgkin’s lymphoma; nonHodgkin’s lymphoma lymphomatoid papulosis (LyP) 863–4 lymphotactin 190 Lynch syndrome 749, 765 M344 antigen 622 McCune–Albright syndrome 444, 458 macrophage colony-stimulating factor (MCSF) 190 biological effects 190 therapeutic potential 190 macrophage-inhibitory protein 1α 1076 MAGE-1 gene 168 MAGE 292 ‘magic bullet’ hypothesis 226 magic roundabout (Robo4) 254 magnesium deficiency 1298 infusion in MI, clinical trials 1349–50, 1350 serum 1203 magnesium sulphate 1203 magnetic resonance angiography (MRA) 588 magnetic resonance cholangiopancreatography (MRCP) 576, 588, 588 magnetic resonance imaging (MRI) 117–22 advantages 118 applications bladder cancer 133, 620 brain tumours 124–5, 293–4, 294 breast cancer 129–30 cholangiocarcinoma 588 colorectal cancer 130–1, 131, 667, 671 endometrial cancer 136, 136 hepatocellular carcinoma 561 multiple myeloma 1078, 1080 nasopharyngeal cancer 389 non-Hodgkin’s lymphoma 137 ovarian cancer 134, 752 primary bone tumours 913, 914, 915 prostate cancer 132–3, 132, 650–1 soft tissue sarcomas 931–2, 932 spinal cord 314, 315 testicular germ-cell tumours 712 thyroid 413, 414–15 assessment of antiangiogenic therapy 261–2 DCE 261–2 diffusion-weighted 119, 119 disadvantages 118 in drug development 34–5 dynamic 620 dynamic contrast enhancement (DCE-MRI) 34–5 endoluminal 122
Index 1469
endovaginal coil 797 FLAIR 303, 304 fusion with CT 116 gadolinium-DTPA enhancement 620 lymphotropic nanoparticle 1425 perfusion 118–19 principles 117–18 radiotherapy plannng 1242, 1242 safety 122 tissue characteristics 118, 118 tissue-specific contrast agents 120–2, 121 magnetic resonance lymphography 120, 121, 126, 136 magnetic resonance spectroscopy (MRS) 119–20, 120 in drug development 35 prostate cancer 133 soft tissue sarcoma 932 MAID regimen 952, 956 malabsorption 791, 1198 malaria infection 991 malignant fibrous histiocytoma (MFH) aetiology 928, 929 bone 906, 908, 920 relative frequency 925 treatment 950 malignant peripheral nerve sheath tumour (MPNST) 925, 927 malignant teratoma intermediate (MTI) 708 malignant teratoma trophoblastic (MTT) 709 malnutrition, protein energy 1148, 1149 mammography 128–9 after breast cancer 493 breast cancer screening 469 post-treatment surveillance of breast cancer 129 reporting, audit 1371 managed clinical networks (MCN) Northern Ireland 1383–4, 1384 Scotland 1380, 1381–2, 1382, 1402–3 ‘Manchester’ system 766–8, 1210, 1324 mandible, tumour invasion 378 mantle cell lymphoma 1061 MAPK pathway 100 MAP regimen 1135–6 marginal zone lymphoma 1061–2 margins, resection non-melanoma skin cancer 848–9, 852 primary bone tumours 908, 909 soft tissue sarcoma 940–1, 940 marimastat 527 marine organisms, drugs derived from 90 Marjolin’s ulcer 836 MART-1 antigen 170 mastectomy breast reconstruction 472 developments in 10 early breast cancer 472 lymphangiosarcoma 930
prophylactic 1329, 1330, 1439 mathematical modelling, see radiobiological modelling matrix metalloproteinases (MMPs) 254, 526 inhibitors 31, 526–7, 580 maxillary sinus tumours 392, 394–5, 394 MCF regimen 610–11 MDM2 gene 928–9 MDS, see myelodysplastic syndromes meat consumption 645, 667, 1435 mechlorethamine (nitrogen mustard) 81, 82, 857, 869 MEC hybrid therapy 1035, 1036 mediastinal fibrosis 550 mediastinal lymphadenopathy 127 mediastinal mass biopsy 147, 150 germ-cell tumour 724 Hodgkin’s lymphoma 1029, 1030 superior vena caval obstruction 154, 157 medical audit audit of 1368 vs clinical audit 1366 vs clinical research 1366 computerized clinical information systems 1368 definition 1365–6 examples 1370 cervical cancer outcome 1372 diagnosis of cancer 1371 follow-up effectiveness 1374 geographical variation in mortality 1374–5 head and neck cancer treatment 1373 lymphoedema in breast cancer 1373 organization nof services 1370–1 palliative care 1373–4 pancreatic cancer management 1373 pathology reporting 1371 patient information needs 1374 radiotherapy fractionation practice 1371–2 radiotherapy outcomes 1365 radiotherapy waiting times 1372, 1379 resource allocation 1370 soft tissue sarcoma management 1373 testicular cancer treatment 1372–3 individual case review 1369 meetings 1367–8 methodology 1366–8 oncological 1368–9 patient group analysis 1369–70 purpose 1366 Medical Research Council (MRC) bladder cancer trials 621, 621 STAMPEDE study 656, 659 medical workforce, see healthcare professions medroxyprogesterone acetate (MPA) 737, 737, 771, 1198, 1298
medullary cell carcinoma kidneys 731 thyroid 411, 412, 412, 450–1, 456 medulloblastoma diagnosis and treatment 309, 309 incidence 287 molecular biology 291, 291 paediatric 1145–7 presentation 308 medulloepithelioma 327, 328 megestrol acetate 1197–8, 1298, 1299 meibomian cysts 334 Meigs syndrome 758 Melacine vaccine 888, 888, 892 melamines 83 melanin 836 melanoma-associated antigens 170 melanoma, malignant acral lentiginous 881 adjuvant therapy 886–9 adoptive cell therapy 196 aetiology 879–80 chemotherapy 886–7, 894 cytokine therapy 191–4 diagnosis and investigations 881, 884–5 familial 26, 1319 screening 1333 immunotherapy 891–2, 894–5 incidence 879 intra-ocular 327, 331–3, 895–6 loco-regional recurrence 889–90 metastases 890–3 brain 893–4 mortality 879 mucosal 399 nodular 881 novel therapies 894–5 ocular adnexa 335–6 pathogenesis 879 pathology 881–2 primary prevention 880–1 radiotherapy 890, 894–5 sarcoma associated with 928 sentinel lymph node biopsy 1426 staging and prognosis 882–4, 882, 883, 884 surgical treatment brain metastases 893 lymph node dissection 885–6 primary lesion 885, 885 treatment costs 1355 vaccines 172, 173, 176, 177, 887–9, 888, 892–3 vagina 799 vulva 804–5 melanosis generalized 1204 primary acquired (PAM) 335–6 melphalan 82, 952 malignant melanoma 889 multiple myeloma 1081, 1083, 1085, 1085, 1089
1470 Index
melphalan (contd.) paediatric osteosarcoma 1134 in stem cell transplantation 1102 membrane attack complex (MAC) 228 MEN-1 gene 449, 1320 MEN see multiple endocrine neoplasia bladder cancer incidence 614 breast cancer 488 head and neck cancer 344–5 lung cancer incidence and mortality 502, 502, 503 microprolactinomas 442 meningeal disease, testicular germ-cell tumours 717 menin gene 444, 447, 456 meningioma 309–11, 310 histological variants 310 imaging 293 incidence 287 molecular biology 291, 291 orbit 335 presentation 309–10 treatment 310–11 meningitis, carcinomatous 492 menopause, age of 748 mercaptopurine 86, 94, 869, 994 Merkel cell carcinoma 398–9, 866–7 aetiology 866 clinical features 866 pathology 867 treatment 867 mesna, soft tissue sarcoma 956, 961 mesulergine 440 meta-analyses, see systematic overviews metabonomics 29 methadone 1292–3, 1295 methotrexate adverse effects 791–2 ALL 994 bladder cancer 625, 630, 630, 631–2, 632 bone tumours 915–16, 916 gestational trophoblastic tumours 823–4, 823, 824, 828 head and neck cancer 366 HIV-associated lymphoma 1177–8 intrathecal 323, 994 Langerhans cell histiocytosis 869 mechanism of action 84 NHL 1052 non-melanoma skin cancer 852 non-small-cell lung cancer 517 ocular lymphoma 334 oesophageal cancer 552 paediatric CNS tumours 1146 pediatric osteosarcoma 1134–6 primary CNS lymphoma 312 soft tissue sarcoma 961 stem cell transplantation 1109 toxicities 823, 1193 5-methoxypsoralen (5-MOP) 858 methyl-CCNU 579
methylphenidate 1301 methyl sodium succinate 329 metoclopramide 1297, 1298 met oncogene 732, 1320 metyrapone 446, 447, 458, 1201 MGMT gene 306 MGUS, see monoclonal gammopathy of undetermined significance MIBG scan 123 MIC regimen 519 microangiopathyic haemolytic anaemia 1104 microglial cells, phagocytic 291–2 microsatellite instability (MSI), colonic cancers 685 microspheres 175, 177 micro-vessel density (MVD) 626 microvessels, tumour 255 midazolam 1301, 1302 middle ear tumours 395–6 migrants 645, 1433 Milan criteria 564, 564, 565 MIMic (Multi-leaf Intensity-Modulating Collimator) 1266, 1267 miniBEAM regimen 1053 mini-chromosome maintenance (MCM) proteins 618 minimally invasive surgery 1429–31 renal cell cancer 1425 see also laparoscopy mismatch-repair genes 1322 misonidazole 48, 1282 misoprostol 1293 mithramycin 1202 mitogen-activated protein (MAP) kinases 254, 410 mitomycin C (MMC) anal cancer 698–9, 699, 700–1 bladder cancer 622–3, 623 combination with radiotherapy 1283–4, 1285 conjunctival epithelial neoplasms 335 gastric cancer 610–11 head and neck cancer 355, 358 HIV-associated anal cancer 1179 mechanisms of action 83, 95 non-small-cell lung cancer 517, 518, 519 oesophageal cancer 552 pancreatic cancer 579 soft tissue sarcoma 961 toxicity 83 vulval cancer 804 mitotane 446, 447, 458, 1201 mitoxantrone 87, 658, 986, 1194 mixed-lineage leukaemia (MLL) gene 984 MMC, see mitomycin-C MMR gene 670 Moh’s micrographic surgery 846, 852 molecular biology 23–4 carcinogenesis 25–9 central dogma 24–5 future impact 35–6 gene expression profiling 29–30
molecular-targeted therapies 11–12, 17, 30, 31–3, 1283, 1409–10 acute myeloid leukaemia 987–8, 990 clinical study endpoints 30, 34–5 CML 996 drivers of 17 marketed 18 pancreatic cancer 580 moles (naevi) 881 monoclonal antibodies 203 anti-EGFR 580, 742, 742, 1286 anti-VEGF 31, 228, 240–2, 251, 256–7, 256, 521, 741–2, 742 combined with cytokine therapy 231 coupled to toxins 229 development of 226–7 haematologic malignancies 230–3, 987, 996, 1008 ALL 996 AML 987, 989 CLL 1009 cutaneous T-cell lymphomas 860 Hodgkin’s lymphoma 1039–40 mechanisms antibody-based cellular cytotoxicity 227–8 complement-dependent cytotoxicity 228 drug-conjugated 229 radio-immunotherapy 229 signal transduction manipulation 98–9, 228–9 radiolabelled 123, 135 solid tumours 233–42 breast cancer 480–1, 481 colorectal cancer 680, 683, 686 multiple myeloma 1090 renal cell cancer 741–2, 742 vascular targeting 228–9, 233 monoclonal gammopathy of undetermined significance (MGUS) 1074, 1080, 1090 mons pubis 799 MOPP regimen 82, 1032, 1033, 1033, 1035, 1139–40 Mormons, Utah 615 morphine 302, 1291, 1292 mortality of cancer, UK 1379, 1385 mouse double minute 2 (MDM2) 290 mouth problems, terminal disease 1295–6 mouth tumours, see oral cavity tumours mouthwashes 1295 M-protein (paraprotein), serum MGUS 1090 multiple myeloma 1073, 1078, 1080, 1080, 1081 MRA, see magnetic resonance angiography MRCP see magnetic resonance cholangiopancreatography see magnetic resonance cholangiopancreatography
Index 1471
MRSA 1205 MS-275 987 MTHFR gene 670 M-TOR inhibitors 33, 454 MTS2 gene 593 Mucin-1 (MUC-1) gene 275 mucins 171 mucin secretion, cholangiocarcinoma 587 mucosa-associated lymphoid tissue (MALT) 412, 430–1, 431 mucosa-associated lymphoid tissue (MALT) lymphoma 336–7, 1061–2 mucosal flaps 350 mucositis radiation 1221 radiation/chemotherapy induced 356, 366, 367 in stem cell transplantation 1104–5, 1105 Muir-Torre syndrome 669, 1319, 1329 multidisciplinary team communication 1315 defined 1378 multidisciplinary team approach 1423, 1428–9 colorectal cancer 687 drug development 1407–8 head and neck cancer 343–4, 344 oesophageal cancer care 541 soft tissue sarcomas 926 testicular cancer 1427 UK cancer services 1378–9, 1385 multi-drug resistance 93 inhibitors 458, 987 multi-drug resistant gene (MD-1) 605 multi-leaf collimators 1260, 1260, 1265–7, 1265, 1266 multiple endocrine neoplasia (MEN) 456–7 overlap 456 thyroid tumours 412, 413 type 1 444, 447, 449, 456, 1320, 1332 type 2 (MEN-2) 456–7, 1120, 1320, 1334, 1439 patient screening and management 1331–2 type 2a 412, 412, 449, 450, 451 type 2b 412, 412, 413, 414, 430, 450 multiple myeloma aetiology 1074 anaemia 1077, 1078, 1081, 1083 asymptomatic 1077–8, 1080, 1082 bone disease 1075–6, 1077 management 1082–3, 1089 chemotherapy 1082, 1083–5 clinical features and complications 1076–8 course of disease 1081–2 diagnosis 1080–1, 1081 high dose therapy and stem cell transplantation 1083, 1086–8, 1086, 1089
hypercalcaemia 1082 incidence 1073–4 infection 1083 investigations 1078, 1079, 1080 maintenance therapy 1088–9 MGUS 1074, 1090 myeloma-related organ or tissue impairment (ROTI) 1080, 1081 new treatment approaches 1090 pain management 1082–3 pathogenesis and biology 1074–6 radiotherapy 1085–6 refractory and relapsed disease 1089–90 renal impairment 1077, 1082 staging and prognostic factors 1081, 1081 mumps virus 273 Munchausen syndrome 1329 Munchausen syndrome by proxy 1329 muromonab (OKT3) 230 muscle fibrosis, post-radiation 361 muscle flaps 351 mustard gas 4 MVAC regimen 630–2, 630 MV-CEA virus 273 MVP regimen 516, 518 MYC gene 993 Mycobacterium avium 1175 Mycobacterium vaccae, as immunological adjuvant in lung cancer 526 mycophenolate mofetil (MMF) 1103 mycosis fungoides clinical staging 855 clinicopathological features 856–7, 856 erythrodermic 856, 856, 859, 861–2 folliculotropic 862 molecular pathogenesis 857 plaque stage 856, 856 prognosis 853, 861–2 treatment 232, 857–61 chemotherapy 860–1 immunotherapy 860 PUVA therapy 858, 862 radiotherapy 858–60 topical 857–8 tumours 856, 856 variants/subtypes 862 myelitis, radiation-induced 299, 514 myeloablation, stem cell transplantation 1103 myelodysplastic syndromes (MDS) classifications 1011–12, 1012 definition 1011 epidemiology 1011 risk assessment 1012 treatment 1012–14 myeloma Bence Jones only 1078 see also multiple myeloma ‘myeloma kidney’ 1077 myeloma-related organ or tissue impairment (ROTI) 1080, 1081 myelopathy, radiation-induced 362
myelosuppression aafter chemotherapy 792 chemotherapy 78, 82 radio-immunotherapy 208–9 MYH polyposis 1319 myocardial infarction (MI) clinical trials antiplatlet therapy 1341, 1342, 1346–7, 1346, 1348–9, 1349 fibrinolytic therapy 1344–5, 1345 magnesium infusion 1349–50, 1350 treatment costs 1356 myocutaneous flaps 350 N-acetyl acetate (NAA) 119–20, 120 N-acetylcysteine (NAC) 348, 349 N-acetyl transferase (NAT) 586 naevi conjunctival 331 ocular 330 skin 881 naevocellular naevi 331 naevus of Ota 331 naloxone 1299 NALP7 gene 813 nanoparticles 177, 366 NAO, see National Audit Office 2-naphthylamine 615 nasal cavity anatomy 392 tumours 346, 393–5, 398 nasogastric tube suction 1198 nasopharyngeal cancer 368, 388–92, 1243–4 clinical presentation 389 distribution 345 management 389–91 prognostic factors 391 recurrence 391 rhabdomyosarcoma 1125 staging 389, 390 treatment failure 391–2, 392 nasopharynx, anatomy 388–9 National Audit Office (NAO) 1385 joint CHI report on cancer services 1396, 1397 report on patients’ experience 1397–9, 1398 National Cancer Institute (NCI), Common Toxicity Criteria (NCI CTC) 1416 national cancer investment plans 1363–4 national cancer plans (UK) 1380–5 progress against targets 1385, 1386–96 National Cancer Research Institute (NCRI) 1384–5 National Cancer Research Network (NCRN) 1385 National Health Service (NHS) colorectal cancer guidelines 687–8 national information technology system 1385 see also UK cancer services
1472 Index
National Institute for Health and Clinical Excellence (NICE) 1362, 1399, 1412 guidance on anti-emetic drugs 1296 guidelines for cancer geneticist referrals 1326, 1327–8 National Wilm’s Tumour Study Group (NWTSG) 1137 Native Americans, pancreatic cancer 573 natural killer (NK) cells 227 nausea advanced disease 1296–7 after stem cell transplantation 1105 in bowel obstruction 1198–9 brain tumours 302 causes 1296 management 1148, 1296–7 paediatric cancer patient 1148 NCAM 671 neck dissection head and neck cancer 350, 374, 374, 375 head and neck tumours, hypopharynx 382 radical 351, 374, 374 audit of outomes 1373 thyroid cancer 428 tonsillar tumours 381 neck irradiation, and thyroid cancer 409 neck nodes, see cervical lymphadenopathy necrolytic migratory erythema 1203 need, definitions 1352 nelarabine 995 Nelson’s syndrome 446, 447 neodymium-Yag (Nd-Yag) laser bladder cancer 620–1 bronchial carcinoma 1190, 1190 colorectal cancer 675 gastrointestinal obstruction 1198 nephrectomy cytoreductive 735 laparoscopic 1425, 1429 partial 735, 1427, 1427 radical 734–5 renal cell cancer 1425 Wilm’s tumour 1137 nephroblastoma, see Wilms’ tumour nephrostomy, percutaneous 150–1, 151, 162 nephrotic syndrome 507, 1196 neural cell adhesion molecule (N-CAM) 1075 neural crest tumours 1128–30 neuroblastoma incidence 1119 olfactory (esthesio-neuroblastoma) 398 orbital metastases 329 paediatric 1128–30 tumour cell vaccine 194 neurocytoma, central 308 neuroendocrine tumours 567–8 neurofibroma, orbit 331 neurofibromatosis 1203
type 1 (NF1) 288, 454–5, 457, 927, 1120, 1321, 1329, 1332 type 2 (NF2) 288, 1321, 1329, 1332 neurological complications chemotherapy 792 management in child patient 1149–50 neurolysis, coeliac plexus 161, 162, 581, 592, 1294 neuroregulins 233 neurotoxicity cranial irradiation 528 drugs 88, 792 neutron therapy 51, 354, 958 neutropenia chemotherapy 525, 1205–6 radio-immunotherapy 209 sepsis 1205–6 NF1 gene 927 NHL, see non-Hodgkin’s lymphoma nickel refining 346, 347 nicotinamide 48, 354 nifedipine 1294, 1296 Nigeria, paediatric cancers 1119 nilotinib 995, 1001, 1002 nilutamide 658 nimorazole 48, 354, 1282 nitrites, dietary 603 nitrogen mustard (mechlorethamine) 81, 82, 857, 869 nitrosamines, dietary 346, 388, 586 nitrosureas 82–3, 300, 305–6, 312 Nixon, President R 1353 NMDA receptor antagonists 1294 N-methyltriazines 83 N-nitroso compounds 539, 603 nomograms 1428, 1429 non-Hodgkin’s lymphoma (NHL) 136–7 aetiology 1045 bone 920–1 bone marrow disease 137–8 Burkitt’s 1052, 1053–4 classification 1045, 1047–8, 1047 criteria for treatment response 138–9, 138 diagnosis and staging 137–8, 138 epidemiology 1045, 1046–7 extra-nodal disease 137–8, 138 follicular 206–7, 1057–61 HIV/AIDS-related 1054 primary CNS 1176–8 systemic 1173–6 MALT type 329, 336–7, 411, 412, 430–1, 431, 1061–2 mantle cell 1061 marginal zone 1061–2 monoclonal antibody therapies 230–1 nodal disease 137 paediatric 1140–2 anaplastic large cell 1142 Burkitt’s lymphoma 1141–2 incidence 1119, 1140 precursor T- and B-cell lymphoblastic lymphoma 1142
presentation 1141 peripheral T-cell 1054–6 anaplastic large cell 1055 angioimmunoblastic T-cell 1055 HTLV-1 associated adult T-cell leukaemia/lymphoma 1055–6 primary CNS 1052 radio-immunotherapy 205–8 comparison of radio-isotopes 211, 212–13, 214 dosimetry 214 integration into treatment algorithms 214–15, 215 myeloablative prior to SCT 210–11 secondary myelodysplastic syndrome 1014 staging 1048, 1048 stem cell transplantation 1101 allogenic 1056, 1062–3 autologous 1050, 1053, 1055, 1058, 1059, 1061 SVCO 1194–5 testicular 706 thyroid 412 vaccination trials 175 see also diffuse large B-cell lymphoma (DLBCL); primary T-cell lymphoma non-melanoma skin cancer aetiology 836 familial syndromes 836–9 pre-malignant conditions 839–40 see also named cancers non-steroidal anti-inflammatory drugs (NSAIDs) 1291–2 adverse effects 1293 bladder cancer 625 bone pain 1293 prevention of cancer 1436–7 protection against cancer 540 North Africa 388, 389 Northern Ireland cancer services 1378 Campbell report 1380 reorganization 1383–4, 1384 Norway, breast cancer outcomes 1355 nose, skin tumours 392–3, 393 NOTCH1 gene 993 Notch signalling 254 Nottingham Index 471 N-RAS gene 984 nuclear factor-κB (NF-κB) 96, 96, 179 inhibitors 257, 258 nuclear medicine 122–4 nucleoside analogues 988, 1059, 1061, 1062 nucleotide excision repair (NER) 96–7, 838 nurse-led biopsy survey, audit 1371 nurses 19 nutrition 1197–8 anti-tumour diets 1198
Index 1473
assessment in child patient 1149 head and neck cancer 360 oral brachytherapy 1220 pancreatic cancer 580–1 see also diet nutritional support gastrostomy tube placement 152–3, 153 paediatric cancer patient 1148–9, 1152 NVB regimen 519 obatoclax (GX15-070) 1009 obesity, and cancer risk 469, 540, 574, 667, 732–3, 1435 oblimersen sodium 988, 1009 occult tumours, cervical lymphadenopathy 373, 373, 384 occupational exposures head and neck cancers 346–7, 346 nasal cavity and para-nasal sinus tumours 393 see also environmental exposures octreotide 1297 in bowel obstruction 1198–9 long-acting 445 neuroendocrine tumours 568 pituitary tumours 445, 447 short-acting 445, 568 slow-release 568 octreotide scanning 123 pituitary tumours 447 thyroid cancer 413, 451 ocular adnexal tumours benign 330–1 malignant 334–7 epithelial 334–5 lymphoma 336–7 melanocytic 335–6 secondary 337–8 ocular tumours adult 329–30 benign 330 metastatic 329 primary malignant 331–4 melanoma 895–6 paediatric histiocytosis X 329 medulloepithelioma 327, 328 optic nerve glioma 328–9 orbital rhabdomyosarcoma 325–7 retinoblastoma 320–5 secondary 329 oculodermal melanocytosis (naevus of Ota) 331 odontogenic keratocysts 837 oedema after radiotherapy 945, 945, 1299 facial 351 lower limb 791 malignant transformation 930 oesophageal cancer adenocarcinoma 539–40, 545, 552–3 aetiology 539–40
chemo-prevention 1437 chemotherapy 544–6 advanced disease 552–3 clinical presentation 540–1 costs of treatment 1355 diagnosis and staging 541–2, 542, 543 endoscopic treatment 550, 553–4 epidemiology 539 management advanced disease 551–5 localized (operable) disease 542–50 locally advanced disease 550–1 palliation of dysphagia 554–5 multidisciplinary care 541 radiotherapy 547–9, 548, 549 advanced disease 551–2 brachytherapy 549, 549, 552 complications 549–50 rare histological types 540 squamous cell carcinoma 539–40, 552–3 stenting 153, 154, 554–5, 554 surgery 543–4, 546–7 lymphadenectomy 544 oesophagectomy 543–4, 546–7 oesophagitis, radiation 425, 514, 549 oesophagogastrectomy 608 oesophagus dilatation/stent placement 153, 154, 553–4, 554–5, 554 post-radiation stricture 550 radiation tolerance 42 reconstruction 543 ulceration 1296 oestrogen deficiency in prolactinoma 442 and dietary fibre 1434 and endometrial cancer 764 replacement, pituitary tumours 449 therapy, juvenile angiofibroma 398 and thyroid cancer 409 oestrogen receptors, breast tumours 470, 473, 475, 476–7, 476, 480 Ohngren’s line 392 OKT3 (muromonab) 230 olfactory neuroblastoma 398 oligodendroma 291, 291 oligonucleotides, antisense 276–7, 277, 988, 1009 olive oil 1435 ‘omics’ technologies 23, 29 On-board Imager 1271, 1271 onco-foetal antigens 169, 170 oncogenes defined 1320 discovery 25 gene therapy targeting 276–7 lung cancer 504, 504 oncogenesis 23 onco-viral antigens 170 ONYX-015 273, 273, 370 oophorectomy
ovarian tumours 752, 757 prophylactic 749, 1439 o’p’-DDD, see mitotane opioid antagonists 1299 opioids atypical 1292 fears about use of 1291 higher-dose 1292 low-dose 1292 strong drugs 1292–3 switching 1292–3 Opisthorchis viverrini 585 OPSI syndrome 1029 optic chiasm glioma, paediatric 1144 optic nerve glioma 308, 328–9 radiation injury 332, 359, 363 radiation tolerance 42 oral cavity tumours 376–7, 376 aetiology 376 brachytherapy 1219, 1220–1, 1225–6 buccal mucosa 378–9 cetuximab therapy 235 floor of mouth 378 gums 379 mobile tongue 377–8 pre-malignant conditions 376–7 retromolar trigone 379 temporary tracheostomy 350 tongue, posterior third 380 oral contraceptives 748, 813, 827 orbit benign tumours 330–1 exenteration 336 malignant tumours 336–7 rhabdomyosarcoma 325–7 rhabdomyosarcoma 1125 orchidectomy prostate cancer 657 testicular germ-cell tumours 713 Orfit shell 295–6, 296, 1235 Organization for Economic Co-operation and Development (OECD), breast cancer outcomes study 1355–6, 1355 organ preservation 1427 bladder cancer 1284–5 renal cell cancer 1427, 1427 oropharyngeal cancer 346, 379–80, 379 cetuximab therapy 235 temporary tracheostomy 350 osseomyocutaneous flaps 350 osteitis pubis 802 osteoarthropathy hypertrophic pulmonary 1194 pulmonary hypertrophic 507 osteoclastoma needle biopsy 149 preoperative embolization 159 osteonecrosis of the jaw (ONJ) 1089 osteoporosis myeloma 1078 risk in prolactinoma 442
1474 Index
osteoprotegerin (OPG) 649, 1074, 1076 osteosarcoma 905 clinical presentation 912–13 epidemiology 905–6 incidence 905 investigations 913–14, 913, 914 paediatric 1119, 1119, 1133–6 Paget’s 917 staging 908 subtypes 911–12, 917 high-grade central 911–12, 912 low-grade central 912 parosteal 917 surface/periosteal 912, 913, 914, 917 treatment chemotherapy 915–16, 916 of metastases 916–17 radiotherapy 916 surgery 915 outcomes of cancer 3 improvement 1423 prediction 1428 provider volume/surgical specialization 1429 see also late effects of therapy; survival of cancer ovarian cancer aetiology and risk factors 748–9 costs of treatment 1355 diagnosis and staging 134, 134, 750–2, 751 epidemiology 134, 748 follow-up protocol 755 germ cell 756–8 histological classification 750, 750 low malignant potential 756 natural history 749–50 population screening 135 prognosis and survival 748, 756 prophylaxis 1330, 1439 radio-immunotherapy 219 recurrent 755 risk factors 468, 469 screening 752 sex cord stromal tumours 750, 758 surveillance and recurrence 134–5 treatment 752–5 surgical 752, 753 tumour markers 751, 752, 755 ovary cryopreservation 1038–9 radiation tolerance 42 suppression in breast cancer 473 oxaliplatin 84 colorectal cancer 678, 682–5, 683–4 mechanism of action 84, 684 pancreatic cancer 580 side effects 685 oxazaphosphorines 82 see also cyclophosphamide; ifosfamide Oxi4503 261 oxycodone 1292
oxygen hyperbaric therapy 354 and tumour radio-esponsiveness 47–8 tumour tissue 252 oxygen levels, head and neck tumours 353–5 p16 gene 504, 504, 586, 851 p21 gene 574 p53 gene 26, 169, 277, 447, 470 biliary tract cancers 586, 593 bladder cancer 615, 622 breast cancer 489 cervical cancer 778, 784 colorectal cance 670–1 endometrial cancer 765 gastric cancer 604–5 gene therapy 895 germ-line mutations 927, 928 head and neck cancer 347, 369–70 HIV-associated lymphoma 1174 lung cancer 504, 504 manipulation in cancer therapy 102 non-melanoma skin cancer 836, 841 ovarian cancer 749 paediatric cancers 1120 pancreatic cancer 574 primary bone tumours 906 soft tissue sarcoma 928–9 targeted therapies 1283 thyroid cancer 410, 432 wild-type therapy 277, 369–70, 928 p53 protein 28 p57kip2 gene 813 packaging cells 270 paclitaxel bladder cancer 632 endometrial cancer 771 gestational trophoblastic tumours 825 Kaposi’s sarcoma 1172 mechanism of action 88 non-small-cell lung cancer 517, 518, 519, 520 ovarian cancer 754, 755 pharmacokinetics 88 resistance 94 testicular germ cell tumours 722 toxicities 88 paediatric cancers acute lymphoblastic leukaemia 1121–3 acute myeloid leukaemia 1123–4 central nervous system 288, 1142–3 atypical teratoid rhabdoid tumours 1147–8 clinical presentation 1143 ependymomas 1119, 1144–5 glial tumours 1143–4 primitive neuroectodermal tumours 1145–8 characteristics 1118 cure/survival rates 1354 epidemiology 1119–21
Ewing’s tumour 1130–3 germ cell 1138–9 heritable factors 1120 histological features 1119 Hodgkin’s disease 1139–40 late effects of treatment 1150–3 fertility and sex hormone function 1152 growth 1151–2 neurocognitive 1153 organ toxicity 1152–3 second tumours 1151 leiomyosarcoma (HIV-positive child) 1181 liver 1127–8 multi-centre collaborative approach 1118 neuroblastoma 1128–30 non-Hodgkin’s lymphoma 1140–2 osteogenic sarcoma 1133–6 palliative care 1154, 1290 psychological aspects 1154–5 quality of life 1153–4 renal 1136–7 rhabdomyosarcoma 1125–7 stratification 1118–19 supportive care 1148–50, 1154–5 survivors 1358 Pagetoid reticulosis 862 Paget’s disease 906, 917 breast 488 extra-mammary 865–6 nipple 1203 vulva 799, 800, 805 pain bone 1292, 1293 breakthrough 1292 incident 1292 neuropathic 1293–4 spinal tumours 314 pain control 1291–4 adjuvant therapy 1293–4 brain tumours 302 child cancer patient 1154 coeliac plexus block 161, 162, 581, 592, 1294 head and neck cancer 367–9, 368 infusion pumps 1294, 1302 medical audit 1373–4 multiple myeloma 1082–3 non-oral drug delivery 1302 pancreatic cancer 581 vertebroplasty 162, 163 see also analgesics paladium-103 brachytherapy 655 palliative care at home 1357 bladder cancer 632–3 cachexia-anorexia syndrome 1298 children 1290 cholangiocarcinoma 591–2 death 1301
Index 1475
emergencies 1299–300 endometrial cancer 771 gastrointestinal problems 1295–8 ascites 1297–8 constipation 1297 dysphagia 554–5, 1296 intestinal obstruction 1297 mouth 1295 nausea and vomiting 1296–7 head and neck cancer 367–9, 368 lung cancer, radiotherapy 513–14, 514 lymphoedema 1299 management planning 1290–1 oesophageal cancer 554–5 paediatric cancer patient 1154 pain management 1291–4 audit 1373–4 pancreatic cancer 579–81 psychological care 1300–1 radiotherapy 1293 rehabilitation 1300 respiratory symptoms 1294–5 skin problems 1299 social support 1301 specialist services 1290 spiritual 1301 staff support 1302 sweating 1299 symptom control 1291, 1302 palmar/plantar pits 837 pamidronate 1082, 1089, 1293 breast cancer 481, 491 hypercalcaemia 1202, 1300 Pancoast syndrome 506 pancreatectomy 1354 pancreatic cancer aetiology 573–4 anatomical location 574 chemotherapy and chemo-radiotherapy 577–80, 578 clinical presentation 575 coeliac axis block 161, 162, 581, 592, 1294 costs of treatment 1355 genetics 574 incidence 573, 573 investigation and diagnosis 576 lymphatic spread 574 management of related problems 580–1 molecular therapies 580 natural history/prognosis 575–6 nutrition 1198 obstructive jaundice 1199, 1199 pain management 581 palliative chemotherapy 579–80 pathology and staging 575, 575 specialization of treatment 1373 surgery 576–7 pancreatico-biliary ductal union, anomalous 585, 592, 593 pancreatico-duodenectomy 577, 591, 595, 1198
pancreatitis, chronic 574 pancytopenia, myelodysplastic syndromes 1011 panitumumab 237, 580 panproctocolectomy, prophylactic 1439 papillary carcinoma, renal 730, 730 papilloma bladder 616–17 eyelid 330–1 para-aortic nodes, metastases, endometrial cancer 766 paracetamol 1292 parafibromin 449 paraganglioma 397–8, 454 paranasal sinus tumours 346, 394–5, 394, 398 paraneoplastic syndromes 1189 cutaneous 1203–4, 1203, 1204 hypertrophic pulmonary osteoarthropathy 1194 lung cancer 506, 507 nasopharyngeal cancer 389 nervous system 1201, 1201 ovarian cancer 750 renal cell carcinoma 733 paraprotein, serum MGUS 1090 multiple myeloma 1073, 1078, 1080, 1080, 1081 plasmacytoma 399 parasitic infestations, bile duct cancer 585 parathyroid gland 449, 456 adenoma/carcinoma 430, 449–50, 456 hyperplasia 430, 449 parathyroid hormone (PTH) 449–50, 507 parathyroid hormone-related peptide (PTHrP) 1201–2 parents, support of 1155 pArf gene 928–9 Parinaud’s syndrome 312, 314 Paris system 1211, 1212, 1212 parotid gland, radiation 375 parotitis, radiation-induced 360 paroxetine 1299 PARP (poly [ADP-ribose] polymerase) 97 parvoviruses 272 Patched gene 836, 838, 841, 841, 851 patient-centred care 1377 patient expectations 1352, 1363 head and neck cancer treatment 367 patient experience 18–19, 1358–60, 1362, 1385 improving 1399, 1400, 1401 patient information 1358–9, 1361, 1363 stem cell transplantation 1111 patients, as consumers of health care 14, 18–19 patient support, head and neck cancer 369 patient support organizations 1315 PAX3-FKHR gene 928 PAX3 transcription factor 1125 PAX7-FKHR gene 928
PAX7 transcription factor 1125 PAX8 transcription factor 410 PCAE regimen 525 PCNU 300 PCV regimen 301, 306, 307 peau d’orange 489 pectoralis major flap 381 pegaptanib sodium 257 pegvisomant 445 PEI regimen 1139 pelvic lymphadenectomy 785 pelvic lymphadenopathy 700 pelvic lymphocele 790 pelvic radiotherapy complications 791, 1418–19 CT planning 1239–40, 1239, 1240 immobilization of patient 1237–8, 1237 pelvis, Ewing’s sarcoma 919, 920 pemetrexed 85, 520 penile cancer 1425–6 brachytherapy 1226–7 sentinel node biopsy 1426 pentagastrin testing 430 pentostatin 861 pentoxyfylline 48 peptide spreading 175 peptide vaccines 173–4 percutaneous ethanol injection (PEI) 152, 564, 565–6 percutaneous gastrostomy 152–3, 153, 1197, 1198 percutaneous transhepatic cholangiography (PTC) 153–4, 156, 588–9, 589, 594 PE regimen 526 performance status, lung cancer prognosis 508 pergolide 440 peri-anal cancer diagnosis and staging 697, 697 management 701–2 pericardial effusion 492 pericardial tumour 1195, 1195 perichondritis, ear cartilage 397 peripheral blood progenitor cells (PBPCs) 1086 peripherally inserted central catheters (PICC) 13 peripheral nerve tumours, orbit 331 peripheral neuropathy, sensorimotor 507 peripheral T-cell lymphomas anaplastic large cell 1055 angioimmunoblastic T-cell 1055 extra-nodal NK/T cell 1056 hepatosplenic T-cell 1056 HTLV-associated 1055–6 unspecified 1056 peritoneal cavity, chemotherapy 611–12, 754, 1199 peritoneo-venous shunts 1297–8 Perspex shell 1235–6, 1235, 1241 pertuzumab (2C4) 228, 239–40
1476 Index
pesticides 1073 Peutz-Jeghers syndrome 26, 758, 1204, 1329 P-glycoprotein expression 987 phaeochromocytoma 430, 454–6 associated genetic disorders 457 association with MEN-2 456 diagnosis 455 treatment 455 pharmacists 19 pharyngeal tumours 382–3 pharyngo-laryngectomy 382 pharyngolaryngo-oesophagectomy 351, 544 pharynx, reconstruction 351 phenacetin 615 phenothiazine 1302 phenoxyacetic acid 930 phenoxy herbicides 930 phenytoin 94 Philadelphia chromosome 98, 992, 993, 994–5, 996–7, 997 phosphatidylinositol-3-kinase (PI3K) pathway 28–9, 28, 290, 1075 inhibitors 1003 phosphoinositol-4, 5-diphosphate (PIP2) 28–9 photocoagulation ocular melanoma 331–2 retinoblastoma 325 photodynamic therapy (PDT) basal cell carcinoma 849 cholangiocarcinoma 592 head and neck cancer 369 non-melanoma skin cancer 840 ocular metastases 334 oesophageal cancer 550, 554 prostate cancer 656 Photofrin 369 photophoresis, extra-corporeal 861 photosensitization 332, 369 phototherapy 51, 858, 862, 880 ‘physaliferous’ cells 311 Physicians’ Health Study 1437 PIAF regimen 566 pilocarpine 1296 pineal tumours 287, 287, 312–14 pinna, tumours 396–7 piroxican 881 pituitary gland, radiation 361–2 pituitary/placental glycoprotein hormones 814 pituitary tumours 438 ACTH-secreting 439, 446–7 adenoma 456 carcinoma 447–8 chemotherapy 449 classification 439 functionless 443 GH secreting (somatotrophinoma) 443–5, 444 hormone replacement therapy 449
posterior pituitary 439 prolactin-secreting 439–43 radiotherapy 448–9 treatment objectives 439 place of death 1291 child cancer patient 1154 placenta, normal 809, 812 placental alkaline phosphatase (PLAP) 313 placental growth factor-1 (PLGF-1) 253 placental-site trophoblastic tumour (PSTT) 812 follow-up 827 incidence 813 management 826–7 plant phytoestrogens 1330 plasma cells 1073, 1075 plasmacytoma extramedullary 1091 head and neck 399–400 solitary of bone 1090–1 platelet-derived growth factor (PDGF) 254 platelet-derived growth factor receptor (PDGFR) 290, 894 platelet endothelial cell adhesion molecule (PECAM 1) 1075 platinum analogues 79, 83–4 brain tumours 300, 309, 314 mechanisms of action 83 ovarian cancer 753, 753, 754, 755 resistance 96–7 spectrum of activity 84 toxicity 83–4 see also named drugs PLCO (Prostate, Lung, Colon and Ovary) programme 135, 646 PLCO study (US) 135 pleural effusion 137, 492, 1193–4, 1194 pleural fluid, examination 1193 pleurodesis 1194 Plummer–Vinson’s syndrome 539 Pneumocystis carinii, see Pneumocystis jirovecii Pneumocystis jirovecii 1108, 1190 pneumonia, prophylaxix 1175, 1176 pneumonectomy 512, 1193 pneumonia syndrome, stem cell transplant 1106 pneumonitis, radiation 514, 550 pneumothorax 147 podophyllin 89 podophyllotoxin 89 POEMS syndrome 1078, 1091 Poland 345, 1356 POL gene 271 pollution, atmospheric 503 POLO (predictive organ localization) 1272 poly-lactive-co-glycolide microspheres 177 polymyositis 507 polyphenols 881 polyplexes 274 polypoid dysplasia, colorectal 668–9 polyps
colorectal 667–8, 670 adenomatous 668 hyperplastic 668 treatment 674 gall-bladder 594 POMB ACE regimen 716 porfiromycin 355 porphyria cutanea tarda 1204 portal vein embolization (PVE) 590 tumour invasion 577 porto-caval (Winslow) node 137 positron emission tomography (PET) 10, 123–4, 1424–5 advantages 123 brain tumours 124–5, 294 cholangiocarcinoma 589 colorectal cancer 131, 685, 686, 687 in drug development 35 endometrial cancer 136 fluorine-18 fluorothymidine (18F-FLT) 130 fluorodeoxyglucose (FDG) 123 fusion with CT 123–4, 125, 126, 127, 128, 129 future developments 124 gastric cancer 605 head and neck tumours 125, 126 Hodgkin’s lymphoma 1027, 1035–6 indications, breast cancer 130 limitations 123 lung cancer 127, 128 malignant melanoma 884–5 myeloma 1080 non-Hodgkin’s lymphoma 139 oesophageal cancer 542 ovarian cancer 135 pancreatic cancer 576 phaemochromocytoma 455 pituitary tumours 447 principles 123 in radiotherapy planning 1242–3 soft tissue sarcoma 932–3, 933 testicular germ-cell tumours 712 thyroid cancer 414 post-cricoid tumours 382 post-embolization syndrome (PES) 162 posterior fossa syndrome 302 post-nasal sinus tumours, radiotherapy 1241 pouch of Douglas 796 poverty 5–6, 6, 6 poxviruses, vaccines 177 PPARγ (peroxisome proliferator-activated receptor γ) 410 PPomas 451 prednisolone multiple myeloma 1081, 1085, 1085 prostate cancer 658 see also CHOP regimen; R-CHOP regimen pregnancy
Index 1477
breast cancer 487–8 cervical cancer 789 desmoid tumours 963 microprolactinoma 442 molar, see hydatidiform mole ovarian cancer risk 748 twin hydatidiform mole 816, 821 pre-sacral lymph nodes 680 presentation of cancer 10 President’s Cancer Panel (USA) 1358, 1359 prevention of cancer 6–8, 1361, 1424 breast cancer 469, 1330 dietary factors 7, 7, 505, 1198, 1330, 1438 future 15 head and neck cancer 346, 347–8, 348, 349 hormone manipulation 1437–8 infection control 7 inherited cancers 1329, 1330 lifestyle changes 1330 lung cancer 504–5 melanoma 880–1 prostate cancer 646–7, 1438 surgery 749, 1438–9 targeted programmes 7–8 tobacco smoking control 6–7, 15 UK national cancer plan targets 1386–9 see also chemo-prevention; vaccines primary acquired melanosis (PAM) 335–6 primary care, UK 1401–2 primary care trusts (PCTs) 1398, 1399, 1401–2 primary central nervous system lymphoma (PCNSL) 1052 primary cutaneous CD30+ lymphoproliferative disorders 863–4 primary effusion lymphoma (PEL) 1176 primary sclerosing cholangitis (PSC) 585, 587 primitive neuro-ectodermal tumours (PNETs) 308–9, 1119 procarbazine 83, 300, 312, 1146 proctitis, radiation 681, 682, 1230, 1372, 1419 pro-drug gene therapy 277–8, 278, 895 pro-drugs 82, 92, 94, 995 progestational agents 1298 progesterone receptors, endometrial cancer 765–6 progestogens anorexia-cachexia syndrome 1197 endometrial cancer 771 prognosis discussion of 1312–13 patients’ perception of 1311–12 prokinetic drugs 1298 prolactinomas 439–43, 456 macroprolactinomas 439–41, 441 microprolactinomas 441–3 pregnancy 442 proliferating cell nuclear antigen (PCNA), bladder cancer 626 prolylhydroxylases (PHDs) 252
prolymphocytic leukaemia 1010 promoters 24 tissue- and tumour-specific 275, 275 promyelocytic leukaemia (PML) gene 984 proptosis 327, 328, 329 prostate cancer aetiology and pathogenesis 645–7, 1435 biological therapies 658 bone metastases 648–9, 650, 658 brachytherapy 1227–30 clinical features 649–50 combined modality therapy 1280 costs of treatment 1355 diagnosis 132–3, 132, 650–1 epidemiology 132, 645, 646 hereditary 1319, 1320, 1323, 1327 in HIV 1181 hormone therapy 654, 656, 657–8, 1423 lymph node spread 648, 648 management active monitoring 653 locally-advanced disease 656–7 metastatic disease 657–8 novel therapies 655–6 organ-confined disease 651–6 principles 651, 652 recurrent disease 657 surgery 653–4 outcome prediction 1428, 1429 pathology and staging 647–9, 648, 648, 649 patient experience 1359 radio-immunotherapy 220 radiotherapy 654–5, 654, 655, 657 α:β ratios 43 conformal 1256, 1257–8, 1258 image guidance 1270 intensity-modulated 1268 patient immobilization 1237 planning 1239–40, 1240, 1242 screening and prevention 9, 645, 646–7, 1424, 1438 surveillance and recurrence 133 tumour markers 646, 650, 651, 653, 657 vaccine trials 172, 176 Prostate Cancer Trialists’ Collaborative Group (PCTCG) 658 prostatectomy, radical 653, 657 prostate membrane-specific antigen (PSMA) 650 prostate specific antigen (PSA) 650, 1424 after treatment of cancer 1230 diagnosis and screening 645, 646, 650, 651, 653 doubling time 650, 657 levels after prostatectomy 657 prostate stem-cell antigen (PSCA) 650 prostatic acid phosphatase 650 prostatic intra-epithelial neoplasia (PIN) 647 prostatic rhabdomyosarcoma 1125 prostatic stromal fibroblasts 647
prostheses, limb amputation 1134 protease inhibitors 1170 proteasome inhibitors 258 ProtecT study 646, 651–2 protein energy malnutrition 1148, 1149 protein kinase A, regulatory subunit 1 (PRKAR1A) gene 444, 447, 458 protein kinase C 254 proteins, translation 24–5 proteinuria 241 proteomics 29, 752 proton-beam therapy 51, 332–3, 448, 950–2, 951 proton-pump inhibitors 1293, 1296 proto-oncogenes 25 pruritus 1199, 1204 advanced disease 1299 management 1299 vulval tumours 801, 805 PSC833 987 pseudo-capsule 940 pseudocoxarthritis 838 Pseudomonas aeruginosa 1205 psoralen and ultraviolet A radiation (PUVA) therapy 858, 862, 880 psychological aspects, childhood cancer 1154–5 psychological care 1300–1 breast cancer 492 psychological support, breast cancer patient 492 PTEN gene 26, 29, 290, 749, 765 PTEN hamartoma tumour syndrome (Cowden Syndrome) 1321, 1329, 1333 PTK 787 301 ptosis, painful 336 public health campaigns malignant melanoma 880–1 smoking cessation 504 pulmonary embolism 1204–5 pulmonary hypertrophic osteoarthropathy 507 pulmonary toxicity, child cancer patient 1152–3 purine analogues 86, 986, 1008 PUVA therapy 858, 862, 880 PVB regimen 715 pyriform fossa tumours 381–2 pyrimidine analogues 86 pyrimidines, halogenated 1282 quality assurance conformal radiotherapy planning 1259–60, 1260 radiotherapy 1249 quality of life assessment 1314 cancer survivors 1356 head and neck cancer 367 lung cancer 508 paediatric cancer patient 1153–4
1478 Index
QUASAR study 1343 quinagolide 440, 442 racial factors, see ethnic factors racial issues 1315 radial forearm flap 351, 377 radiation dose planning 1245–8 reporting/recording 1248–9 radiation exposure 1418 and bladder cancer 615 lung cancer 503 multiple myeloma 1074 paediatric cancer 1120 primary bone tumours 906 soft tissue sarcoma aetiology 929–30 and thyroid cancer 408–9 radiation injury CNS 298–300, 309, 359, 391 oesophagus 549–50 pneumonitis 550 skin 836, 848, 849 radical surgery hysterectomy 785, 790, 792 neck dissection 351, 374, 374, 1373 penile cancer 1425–6 renal cell cancer 1425 vulvectomy 801 radioactive scleral plaque therapy 324–5, 332–3 radiobiological modelling advanced approaches BED inter-comparisons 69 brachytherapy dose gradients 69–70 closely spaced fractions 68–9 concomitant chemo-radiotherapy 70–1 cost-based treatment comparisons 70 estimation of tumour cure probabilities from BED 67 high linear energy transfer radiations 71–2 incorporating cell-loss factors 68 optimization 67–8 palliative therapy 71 serum tumour markers 71 tissue sub-lethal damage repair 69 tumour volume changes 70 volume effect and normal tissues 70 general models addition of external-beam and focal methods 66 amelioration of incorrect treatment dose 65–6 biologically effective dose (BED) 59–60 brachytherapy 64–5 effect of unscheduled treatment interruptions 62–4 fractionation sensitivies (α:β ratios) 60–1 tumour repopulation 61–2 linear-quadratic (LQ) formulation 59
methods of presenting results 64 software packages for bio-effect planning 72 types of model 59 uses of 58 radiobiology, head and neck cancer 353–6 radiofrequency ablation 151–2 liver tumours 151–2, 151, 564, 565–6 renal cancer 152, 152, 735 radiographic markers 1274 radiography, conventional 112–13 digital 113 lung cancer screening 505 multiple myeloma 1078 primary bone tumours 913 soft tissue sarcoma 931 testicular germ-cell tumours 711, 711 radio-immunotherapy 229 antigen targeting 203–4 clinical perspectives 220–1 clinical trials 205–8 definition 203 dosimetry 214 duration of response 214 integration into treatment practice 214–15 in leukaemias 215–17 in non-Hodgkin’s lymphoma 205–15, 1053, 1058 radio-isotopes bismuth-213 216–17 comparison of 211, 211, 214 iodine-131 207–8 yttrium-90 205–7 in solid tumours 217–20 breast 219–20 CNS 218 gastrointestinal 218–19 toxicity and safety 208–10 radioiodine therapy 418–21 complications 423–4 dosimetry 422–3 radionecrosis 299, 361 radionuclide bone scans 123, 129, 133, 734, 1080 radionuclide imaging ovarian cancer 135 thyroid cancer 413, 414 radionuclide therapy carcinoid tumours 453 GEP endocrine tumours 452 liver tumours 566–7 radio-protective agents 49, 1281 radio-responsiveness and anaemia 49 dose-response curves 42–3, 42 early and late reactions 41 tumour clonogenic cell burden 52 and tumour hypoxia 47–9 radio-sensitivity 40, 790 and cell cycle 45 extreme 40–1 Gorlin’s syndrome 838
low-dose hyper-radiosensitivity 43 organs/normal tissue tolerance 41–2, 42 treatment planning 41 tumour-cell burden 52 radio-sensitizing agents 354, 1282–3 radiosurgery brain metastases, malignant melanoma 893 ocular melanoma 333 pituitary tumours 447, 448 radiotherapy 481–6, 482, 483, 484, 485, 486, 487 ARCON 48, 354 beam direction shell 296, 296 boron neutron capture therapy (BNCT) 298 changes in 1353 chemical sensitizing agents 48 combination with chemotherapy 49–51 combination with surgery 51 complications 911 acute reactions 359–60, 359 avascular necrosis of femoral heads 791 cystitis 790 eye 324 fibrosis 791, 1191–2, 1193 gastrointestinal tract 681–2, 681, 790 head and neck irradiation 356–63 lymphoedema 791, 945, 945, 1299 mucositis 356 skin 681, 790 vaginal treatment 796, 798 vulval therapy 802 wound healing 944–5, 945, 953–4 see also late effects cranial 528 craniospinal 1139 dose-response curves 1255, 1255 dosimetry precision 52–3 electron beam 1245 extended field radiotherapy (EF-RT) 1031–2, 1033 fractionation accelerated 47, 628–9 α:β ratios 60–1 head and neck cancers 44, 47, 354, 355–6, 357–8 lung cancer 512–13 rationale for 43, 45–7 future of 11, 11 history of 4 hyperfractionation 43, 47, 528, 686–7 image-guided 655 basic principles 1269–70 concept and rationale 1268–9 imaging systems 1270–1 quality assurance 1275 strategies 1271–5 immobilization devices 946, 946, 1235–8, 1236, 1237 intensity-modulated (IMRT) 609–10, 656, 700, 1418
Index 1479
involved field (IFRT) 1031, 1033, 1034, 1035 late effects 360–3, 681, 681, 790–1, 945, 945, 1415 grading and reporting 1415–16 oesophageal radiation 550 sarcoma formation 929–30 second tumours 1133, 1151 lateral beam 323–4 limitations as sole therapy 1280–1 medical audit 1365 fractionation practice 1371–2 late effects 1372 organization of services 1370–1 waiting times 1372, 1379 neutron therapy 51, 354, 958 normal tissue damage 47 organ motion 1244 pituitary 448–9 planning 1234, 1234 computerized dose planning 1245–9 CT 681, 1239–41, 1239–41 image registration 1243 MRI 1242, 1242 organs at-risk 1244–5 patient position/immobilization 1235–8, 1239 PET 687 target localization 1238–43, 1269 target volume 363, 1243–5, 1256–7 tumour and target volume 1243–5 ultrasound 1242 verification of treatment 1249 predicting demand for 1403 predictive assays 356 proton-beam 51, 332–3, 448, 950–2, 951 quality assurance 1249 resistance 352–3, 353–5, 353 role in cancer 1234 stereotaxy 298 superficial 847–8, 1246 technological advances 1254 therapeutic ratio 1255, 1255 treatment decisions 1233–4 treatment gaps 62–4, 355 tumour cell repopulation 46–7, 355 waiting times 1372, 1379, 1401 wedge-pair technique 326, 326 whole-brain 893, 1177, 1200 whole-neuraxis 297, 298, 308, 313 see also conformal radiotherapy (CFRT); intensity-modulated radiotherapy (IMRT); radiobiological modelling radon exposure 503 raf kinases 100 inhibitors 894 RAF-MEK-ERK signalling 851 ralitrexed, colorectal cancer 684 raloxifene 1438 raltitrexed 85 random errors, avoidance in clinical studies 1342–5 randomization
avoidance of biases 1340–1 avoidance of random errors 1342–5 mega-trials 1343 rapamycin (sirolimus) 1003 ras oncogenes 99–100, 749 RAS pathway 851 ras protein 410 Ras/Raf pathway 879, 894 R-CHOP regimen 207, 430, 1050, 1058, 1176, 1176 REACT study 1417 REAL (Revised European American classification of Lymphoid neoplasms) 1140–1 rebeccamycin analogue 592 receptor activator of nuclear factor-κB ligand (RANK-L) 1074, 1076 RECIST, see Response Evaluation Criteria in Solid Tumours reconstructive surgery breast 472 head and neck cancer 350–1, 381, 382 oesophagus 543 primary bone tumours 909, 909, 915, 919 tongue tumours 377 vaginal surgery 798 rectal bleeding after cervical cancer treatment 791 after prostate brachytherapy 1230 rectal cancer brachytherapy 1226 diagnosis and staging 122, 130–1, 131 GISTs 960 low 675, 676 NHS guidelines 687–8 radiotherapy 676–9, 679 staging 671 surgery 674–5, 687 anus preservation 678–9 see also colorectal cancer rectal examination cervical cancer 781 digital (DRE) 649–50 recto-urethral fistula 655 rectovaginal fistula 798 rectus abdominis flap 351 red cell aplasia 1204 red meat consumption 1435 redox factor-1 (Ref1/apurinic/apyrimidinic endocuclease) 252 Reed, Dorothy 1027 Reed-Sternberg cells 1027, 1028, 1139 Reese-Ellsworth classification (retinoblastoma) 321, 322 regional cancer networks 1429 rehabilitation 1300, 1360 head and neck cancer 364, 364 laryngectomy 351, 364 religion 1301 renal cell cancer clinical evaluation and staging 117, 733–4, 734
clinical features 733 collecting duct (Bellini’s duct) 731 conformal radiotherapy 1258, 1259 epidemiology 729, 729, 730 hereditary syndromes 731–2, 732 histopathology 730–1, 730 immunotherapy 736–40, 737 IVC tumour thrombus 1425, 1426 localized disease 734–5 medullary carcinoma 731 metastatic disease non-surgical management 736–41 patterns of spread 733 surgery 735–6 novel treatment agents 741–3 percutaneous tumour ablation 152, 152, 735 preoperative embolization 159–60, 160 prognostic factors 729, 729, 740–1, 741 radiotherapy 736 relapsed disease 735 risk factors 732–3 surgery 1425, 1427 tumour imaging 117 renal-cell carcinoma cytokine therapy 192–3 interferon-α therapy 192 interleukin-2 193 radio-immunotherapy 220 vaccine trials 172, 173, 175 renal drainage, percutaneous 150–1, 151 renal impairment anti-inflammatory drugs 1293 direct tumour involvement 1196, 1196 multiple myeloma 1077 stem cell transplant recipeint 1104 ureteric obstruction 1196 renal toxicity, chemotherapy 791–2, 792, 1152 renal tumours benign 731 clear cell sarcoma 1137 rhabdoid tumour 1137 Wilms’ (nephroblastoma) 731, 1119, 1119, 1136–7 see also renal cell cancer reovirus 273 respiratory failure 825–6, 1190–2 Response Evaluation Criteria in Solid Tumours (RECIST) 139–40 resveratrol 1438 reteplase 1205 retina malignant tumours 333 radiation injury 363 radiotherapy 323 retinoblastoma assessment of child 321 classification 321, 322 genetics 320–1 incidence 1119 management of at-risk individual/family 1332
1480 Index
retinoblastoma (contd.) metastatic 322–3 orbital recurrence 322 paediatric, risk of osteosarcoma 1133 presentation 321 second neoplasms 321, 927, 930 treatment aims 321–2 chemotherapy 323, 323 cryotherapy 325 external-beam radiotherapy 323–4 hyperthermia 325 photocoagulation 325 radioactive scleral plaque therapy 324–5 retinoblastoma (Rb/Rb1) gene 26 bladder cancer 615, 622 cervical cancer 778 loss of heterozygosity 504, 1133, 1134 lung cancer 504, 504 mechnism of action 1320, 1320, 1322 multiple myeloma 1076 osteosarcoma 906 ovarian cancer 749 paediatric cancers 1120, 1133 pituitary carcinoma 447 retinoblastoma 320–1 soft tissue sarcoma aetiology 929 retinoic acid receptor α gene (RARα) 984 retinoic acid receptors (RARs) 102–3, 348, 1173 retinoids 102–3, 348, 349, 625 adverse effects 861 contraindications 861 cutaneous T-cell lymphomas 861 myelodysplastic syndromes 1012 non-melanoma skin cancer treatment 838 novel 861 systemic 839 topical 858 Kaposi’s sarcoma 1173 retinoid X-receptor (RXR) 348, 349 retinoid X receptors (RXRs) 102–3, 1173 RET oncogene 412, 430, 432, 456, 1320, 1320 retromolar trigone tumour 379 retropancreatic lymph nodes 587 retroperitoneal lymph nodes 137, 138, 711 dissection 1427–8, 1428 retroperitoneum mass, biopsy 148, 149 primary germ-cell tumours 725 sarcoma 931, 958–60, 958, 959 retroviral vectors 270, 271–2, 271 retroviruses, tumour vaccines 177 reverse transcriptase polymerase chain reaction (RT-PCR) 884 rhabdoid tumour, kidney 1137 rhabdomyosarcoma 925, 931 aetiology 928 alveolar 928, 1125
embryonal (ERMS) 1125, 1126 orbit 325–7 paediatric 1119, 1119, 1125–7 vagina (sarcoma botryoides) 799 vulva 805 rhenium-188 216 ribonucleotide reductase (RR) 273–4 Ringertz tumour 394 rituximab 230–1 acute lymphALL 996 CLL 1008 combination with cytokines 231 HIV-associated Castleman’s disease 1178 Hodgkin’s lymphoma 1033 non-Hodgkin’s lymphoma 1050, 1057–8, 1057, 1061 follicular 1057–8, 1057, 1059 HIV-associated 1176 paediatric 1142 R-CHOP regimen 207, 430 stem cell transplant conditioning 230 thyroid lymphoma 430 RNA small inhibitory (siRNA) 276–7 transcription 24 RNA-induced silencing complex (RISC) 277, 277 RNA polymerases 24 robotic surgery 10, 1430–1, 1430 Royal College of Radiologists 129 Clinical Oncology Information Network (COIN) 1368, 1369 survey of resource allocation in clinical oncology 1370 Royal Marsden Hospital NHS Foundation, genetics referral guidelines 1326, 1327–8 RTOG 85-31 study 656 RTOG 86-10 study 656 rubber industry 615 Rumania 559 Russian Federation 345 Rye classification 1139 St Jude staging system 1141, 1141 saline, nebulized 1295 saliva artificial 1295–6 radiotherapy effects 361 salivary glands effects of radiation 359, 360–1 radiation effects 375 Salmonella infections 592 Salmonella typhimurium 274 salpingo-oophorectomy bilateral 134, 135 ovarian cancer 752, 757 prophylactic 749 salpino-oophorectomy, bilateral, endometrial cancer 766–7 salted fish 388
salt fish consumption 346 sarcoma aetiology 906 alveolar soft part 928 clear cell 928 defined 925 epitheloid 946 hepatic 930 spindle cell 956 synovial 925, 931 uterus 772–3 vulva 805 see also osteosarcoma; soft-tissue sarcoma sarcoma botryoides 799 scar tissue, squamous cell carcinoma arising from 851 scFvL19 233 Schistosoma haematobium 615 schistosomiasis 7 schwannoma, malignant 335 scleral plaque therapy 324–5, 332–3 SCL gene 993 scopolamine 1219 Scotland Acute Service Review 1381 cancer services 1377 reorganization 1381–2, 1382 report on 1380 managed clinical networks 1381–2, 1382, 1402–3 national cancer plan 1380 implementation 1402 screening 8 advantages and disadvantages 8–9, 8 assessing the benefits of programmes 9–10 bladder cancer 620 breast cancer 469 cervical cytology 781, 792 colorectal cancer 131–2, 673–4 communication with patients 1314 costs of 1354 future of 15 guidelines 9 hepatocellular carcinoma 562 kit for 10 lead-time bias 8 length bias 8 lung cancer 128, 505 neuroblastoma 1130 ovarian cancer 135, 752 programme development 9 prostate cancer 645, 646, 1424 selection bias 8–9 sensitivity 8 specificity 8 success of 1354–5 UK national cancer plan targets 1386–9 scrotal cancer 25 SCT, see stem cell transplantation seborrhoeic keratosis 331, 750, 1203
Index 1481
secondary myelodysplastic syndrome (MDS) 1111 second primary tumours 1418 chemotherapy-related 1151 childhood cancer survivors 1133, 1151 radiotherapy-related 1133, 1151 retinoblastoma 321, 927, 930 sedation 1301, 1302 seizures brain tumours 292, 293 management 302 selective serotonin reuptake inhibitors (SSSRIs) 1294, 1300–1 selenium 647, 1436 ‘self’ antigens 175 self-management, late effects of therapy 1418–19 semen, cryopreservation 1038 seminoma mediastinal 724 testis 706, 708 imaging 711–12, 711 management 718–21 tumour markers 710, 710 Semliki forest virus (SFV) 272 semustine (MeCCNU) 300 Senegal 559, 559, 562 sensory disturbance, brain tumours 293 sentinel lymph node biopsy (SLNB) 1426 breast cancer 472 malignant melanoma 882, 883–4 penile cancer 1426 vulval cancer 802 Sertoli-cell tumour ovary 758 testis 725 sestamibi scanning, parathyroid tumours 450 sex chromosome abnormalities 1120 sex-cord stromal tumours ovary 750, 758 testis 725 sex hormones paediatric cancer survivors 1152 and thyroid cancer 409 sexual behaviour anal cancer 696 cervical neoplasia 778 sexual functioning, after prostate cancer 653, 1230 sexuality, cancer patient 367, 792, 1314–15 Sézary syndrome (SS) 862–3 clinical features 862 diagnosis 862–3 molecular pathogenesis 857 prognosis 862, 863 treatment 858, 863 SIADH, see syndrome of inappropriate ADH secretion siblings, child cancer patient 1155 sickle cell disease/trait 731 Siemens Artiste system 1271, 1271
sigmoidoscopy 132, 673 signal transducers and activators of transcription 1/2 (STAT1/2) 191 signal transduction 28–9, 28 inhibitors 97–101, 228–9, 366 silent corticotroph adenomas 443 simulator, radiotherapy 1238–9 simulator CT scanner 1241 simvastatin 446 Sinbis virus (SIN) 272 single nucleotide polymorphisms (SNPs) colon cancer 670 DNA repair genes 96–7 single photon emission computed tomography (SPECT) 122, 135 brain tumour 294 parathyroid tumours 450 thallium-201 1177 sinuses, para-nasal anatomy 392 tumours 346, 394–5, 394, 398 sinusoidal obstruction syndrome (SOS) 1106 Sipple’s syndrome, see multiple endocrine neoplasia 2 (MEN-2) sirolimus (rapamycin) 1003 Sister Mary Joseph nodule 575, 750 skin, radiation injury 836, 848, 849 skin cancers adnexal 835 aetiology 836 audit of nurse-led biopsy role 1371 brachytherapy 1226 children 1120 epidemiology 835 familial syndromes 836–9 Merkel-cell carcinoma 398–9 nose 392–3, 393 see also melanoma, malignant; nonmelanoma skin cancer and named skin cancers skin flaps 350, 351 skin lesions familial cancer syndromes 1329 malignancy associated 1203–4, 1203, 1204 radiotherapy 681, 790 skin tattoos, radiotherapy planning 1237, 1237, 1238, 1239, 1239 slits 254 small cell cancer anus 701 oesophagus 540 prostate 648 small inhibitory RNAs (siRNA) 276–7 small round blue cell tumours 1130 SMF regimen 579 Smokeline 348 smoking, see cigarette smoking smooth muscle tumours eye 330 leiomyosarcoma, HIV-positive child 1181
snuff dipping 376 social changes, and cancer care 1357 social insurance 1363 social problems, of cancer patients 1360 socio-economic factors 14, 15 cancer survival 1374–5 and cervical cancer 777 survival of cancer 1356 sodium, serum 1202–3, 1202 sodium-iodide symporter 432 sodium porfimer 592 sodium valproate 447 soft tissue fibrosis 361, 391 soft tissue sarcomas 925 aetiology 927–30 anatomical sites 926, 926, 931 benign tumours 926 biopsy 933–5 chemotherapy 952–3 adjuvant 954–7, 955 in metastatic disease 961–2 neoadjuvant 952–3, 956 clinical evaluation 930–4 desmoid tumours 962–6 Ewing’s 926, 928, 931 functional outcome 957 hand and foot 946, 953, 953 histological subtypes 925, 925 incidence 925–6 isolated limb perfusion, cytokines 194 liposarcoma 931, 932, 933 local control 926 radiotherapy 942–3, 943 surgical resection 926, 939, 940, 940, 941 local recurrence, treatment 957 lymph node involvement 931 management audit 1373 metastases rates 938–9, 939 treatment 960–2 multidisciplinary teams 926 paediatric, incidence 1119 pathology 934–8 radiotherapy 939, 941–4 adjuvant 944–5 alone 957–8 brachytherapy 945–6 IMRT 949–50, 950 neoadjuvant 942, 944 proton beam 950–2, 951 treatment planning 946–7, 947, 950 volume and dose 947–9 retroperitoneal 931, 958–60, 958, 959 rhabdomyosarcoma, paediatric 1119, 1119, 1125–7 second malignancies 1418 shoulder 951 staging 938–9, 938, 939 surgery 926, 939–41 limb-sparing 939, 941, 941, 952–3 re-resection 941
1482 Index
soft tissue sarcomas (contd.) thigh 950, 950, 951, 951, 954 unresectable/locally advanced 958 solitary plasmacytoma of bone (SPB) 1090–1 somatostatin analogues 568 bowel obstruction 1198–9 carcinoid tumours 453 octreotide 1198–9 pituitary tumours 443, 445, 447–8 somatostatinomas 451 somatotrophinomas 443–5 somnolence, after CNS irradiation 299 Sonic Hedgehog (SHH) 291, 841, 841 sorafenib 100, 256, 257, 742–3, 742, 894 Soros Foundation 1357 South Africa 6, 6, 839 South East Cancer Network (SCAN) 1402 Southwest Oncology Group study 658 Soviet republic, former 6, 6 Spain 559 SPC2996 1009 specialization in cancer care 1354, 1360, 1378, 1380, 1429 SPECT, see single photon emission computed tomography speech disturbance, brain tumours 293 speech restoration, post-laryngectomy 351, 364, 367 sphenoid sinus anatomy 392 tumours 395 spinal cord radiation injury 359, 362, 550 radiation tolerance 42, 299 spinal cord compression 1199–201 breast cancer 492 causes 1199 intramedullary tumour 315 lung cancer metastases 514 malignant 1374 management 1200 prostate cancer 650, 657 symptoms 1199–200 testicular germ-cell tumours 717 spinal tumours metastatic 733, 736 primary 314–15 spindle cell sarcoma 956 spirituality 1312 spiritual support 1301 spironolactone 1297 splanchnic nerves, thoracic, ablation 581 spleen, lymphoma 137 splenic marginal zone lymphoma (SMZL) 1062 Sprouty (Spry) 254 sputum cytology, lung cancer screening 505 sputum retention 1295 sqaumous cell carcinoma head and neck (HNSCC) 234
monoclonal antibody therapy 234–6 squamous cell carcinoma (SCC) anus 695, 702 basaloid 399 bladder 617 conjunctiva 1181 eyelid 335 head and neck chemo-radiotherapy 1286 growth kinetics 352–3, 352, 352, 353 lip 375–6, 839, 850, 852 lung 509–10 middle ear 395–6 oesophagus 539–40, 552–3 skin 850–2 aetiology 836, 850 arising from scar tissue 851 clinical features 850 vulva 800 squamous intra-epithelial lesion (SIL) 1179, 1180 Src family kinases 100 stability genes 26 staging of cancer 1424–5 STAMPEDE study 656, 659 Stanford V regimen, Hodgkin’s lymphoma 1032, 1033, 1034, 1035, 1036 Staphylococcus aureus, methicillin resistant (MRSA) 1205 Staphylococcus infections, post stem cell transplant 1108 STAR trial 1330 statins 446, 881 steatorrhoea 580 stem cells, sources 1098–9 stem cell transplantation (SCT) allogenic 1038, 1056, 1062–3 autologous Hodgkin’s lymphoma 1037 leukaemias 986, 990, 995, 1004 multiple myeloma 1083, 1086–8, 1086, 1089 non-Hodgkin’s lymphoma 1050, 1053, 1055, 1058, 1059, 1061 causes of death after 1105 complications 1104–7, 1105, 1106 infectious 1107–9 late 1111 pulmonary 1106–7 conditioning/preparative regimens 230, 1014, 1100, 1102–4 myeloablative 1103 radio-immunotherapy 216 reduced intensity 989, 1062, 1088, 1103–4 curative potential 1098 determinants of engraftment 1100–1 donor availability 1099 donor HLA compatibility 1100–1 graft T-cell depletion 1100 graft-versus-host disease 1062, 1109–11, 1109
grading 1109, 1110 prophylaxis 230, 1104 risk factors 1109, 1109 graft-versus-tumour (GVT) effect 1062 indications 1101–2 ALL 995, 1101, 1122 AML 989–90, 1101, 1124 CLL 1009 CML 1003–4, 1101 gestational trophoblastic tumours 825 Hodgkin’s lymphoma 1037, 1038 multiple myeloma 1083, 1086–8, 1086, 1089 myelodysplastic syndromes (MDS) 1014 non-Hodgkin’s lymphoma 1050, 1053, 1055, 1056, 1058, 1059, 1061, 1062–3 renal cell cancer 740 monitoring of chimerism and relapse 1104 number of procedures performed 1099 patient selection 1102 programme requirements and accreditation 1111–12 stenting 153–6 airway 153, 155, 1190, 1191 aorta 163 bile ducts 153–4, 156, 576, 590, 591, 592, 1199, 1199 carotid artery 163, 164 gastrointestinal tract 153, 155, 675, 1198 oesophagus 153, 154, 554–5, 554 superior vena cava 154, 157, 1195 venous 154, 156, 157 stents metal expandable 154, 156, 554–5, 591, 675 plastic 591 stereotactic biopsy, brain 295 stereotactic radiosurgery (SRS) 298 brain metastases 893 fractionated 298 ocular melanoma 333 pituitary tumours 447, 448 Sternberg, Carl 1027 Stewart–Treves syndrome 930 stilbestrol 657 stomach, radiation tolerance 42 stomach cancer, see gastric cancer Streptococcus pneumoniae 1205 streptokinase, in MI, clinical trials 1348–9, 1349 streptozotocin 300, 568, 579 stridor 384, 1295 stroma derived factor-1 (SDF-1) 1075 SU5416 988 subcutaneous fibrosis, after radiotherapy 945, 945 suberoylanilide hydroxamic acid (SAHA) 987, 1013
Index 1483
submucous fibrosis 377 subtotal nodal irradiation, Hodgkin’s lymphoma 1031, 1033, 1034 succinic dehydrogenase genes 397 sucralfate 1299 suicide gene therapy 269–78, 278, 301 colorectal cancers 686 thyroid cancer 432 sulindac 1437 sun exposure, melanoma 880–1 sunitinib 256, 257, 453–4, 742, 742 sunscreens 880 superior mesenteric vein, tumour invasion 577 superior vena caval obstruction (SVCO) 154, 157, 506, 506, 1194–5, 1300 superior vena cava (SVC), stenting 154, 157, 1195 support breast cancer patient 492 parents of child cancer patient 1155 supportive care brain tumours 302 child cancer patient 1148–50, 1154–5 support organizations 1315 supraclavicular lymph nodes 587 suramin 458 surgery, prophylactic 1424, 1438–9 surgical oncology 1423 anatomical knowledge 1427–8 minimally invasive surgery 1429–30 multimodal team-based approach 1423, 1427, 1428–9 organ and function preservation 1427 provider volume:outcome relationships 544, 1354, 1429 robotic surgery 1430–1 Surveillance, Epidemiology and End Results (SEER) registry 573, 652, 696, 930, 1354 survival of cancer 3, 1358 breast cancer 1378 improvement 1423 socio-demographic factors 1356, 1374–5 UK versus European figures 1379 see also late effects of therapy survivin 95 survivor guilt syndrome 1334 survivors of cancer 1358 follow-up 1417, 1419, 1431 quality of life 1356 SVCO, see superior vena caval obstruction swallowing, after head and neck cancer surgery 350, 351 sweating 455, 1299 Sweden, breast cancer outcomes 1355 Syed-Neblett applicator 798 symptom control advanced cancer 1291, 1302 child cancer patient 1148, 1154 syncope, head and neck cancers 368, 389 syndecan 1 1075
syndrome of inappropriate ADH secretion (SIADH) 507, 1201–2 syndromes rare genetic with increased cancer risk 1321 see also heritary cancer syndromes; paraneoplastic syndromes Synergy 1271 synovial sarcoma 925, 931 systematic overviews (meta-analyses) antiplatlet drugs 1347, 1347 reliability of 1345–6 SYT-SSX fusion gene 927 5-T4 antigen 604 Taiwan hepatocellular carcinoma 559, 559, 562 nasopharyngeal cancer 388 tamoxifen adverse effects 477 breast cancer prevention 469, 1330, 1437–8 combination with chemotherapy 480 in ‘early’ breast cancer 473, 474, 475–6, 475, 475, 1347–8, 1348 and endometrial cancer 764–5, 771 malignant melanoma 891, 892 mechanism of action 476, 476 occult breast cancer 373 and thyroid cancer 409 tanning beds 880 targeted therapies 1283 tariquidar 458 taste disturbance, radiotherapy 361 tattoos, radiotherapy planning 1237, 1237, 1238, 1239, 1239 tauromustine (TCNU) 300 TAX 327 trial 658 taxanes 87–9 adverse effects, myelosuppression 792 bladder cancer 632 carcinoids 453 endometrial cancer 771 Kaposi’s sarcoma 1172 mechanism of action 88, 1282 non-small-cell lung cancer 517, 518 oesophageal cancer 552 ovarian cancer 754, 755 pharmacokinetics 88 prostate cancer 658 resistance to 94 toxicity 88–9, 366, 792 see also named taxanes T-BEP regimen 716 TBI, see total body irradiation T-cell chronic lymphocytic leukaemia (T-CLL) 1009 T cells 167 chimeric immune receptors 196 depletion in stem cell graft 1100 Epstein-Barr specific 195 genetically-engineered 196
regulatory (Tregs) 171 tumour tolerance 171 technetium-99 bone scanning 651 technetium-99m sestamibi scintigraphy 450 teeth care in head and neck radiotherapy 361, 361 care prior to oral brachytherapy 1220–1 teicoplanin 1205 TEL-AMLI fusion gene 992–3 telomerase, as therapeutic target 104–5 temozolomide 300, 306, 307, 312, 890, 894 temporal bone, radiotherapy damage 359 temporal lobe necrosis 359, 362 temsirulimus 454, 742, 742 tenacsin 292 tenesmus 1294 teniposide 519 teratoma brain 313 differentiated 708 extra-gonadal 712, 713, 724–5, 724 ‘growing teratoma syndrome’ 717, 717 malignant teratoma intermediate 708 malignant teratoma trophoblastic 709 malignant teratoma undifferentiated 708 ovary 757–8 paediatric 1138–9 testis 706 clinical presentation 712–13 imaging 711–12, 711 management 713–18, 714 pathology 708–9 staging 709, 709 tumour markers 709–10, 710 terminal care home-based 1357 see also palliative care testicular cancer audit of treatment 1372–3 follow-up effectiveness 1374 multimodal surgical oncology 1427 outcomes 1423 retroperitoneal lymph node dissection 1427–8, 1428 tumour markers 1425 testicular germ-cell tumours aetiology 706–7, 707 clinical presentation 712–13, 713 imaging 711–12, 711 incidence 706 lung metastases 1190–1, 1191, 1193 management non-seminomatous (teratoma) 713–18, 714 seminoma 708, 718–21 pathogenesis 707 pathology 707–8, 708 serum tumour markers 709–10, 710, 717
1484 Index
testicular germ-cell tumours (contd.) sex-cord stromal tumours 725 staging 709 metastatic teratoma 714, 715 yolk-sac tumour 709 testicular maldescent 706, 707 testicular tumours, HIV-positive men 1181 testosterone, replacement in pituitary tumours 449 testosterone replacement 724 tetracycline 1194 TFE-ASPL fusion gene 928 Thailand, cholangiocarcinoma 585 thalassaemia 586 thalidomide 256, 257, 301, 894, 1055, 1298, 1299 chronic lymphocytic leukaemia (CLL) 1008–9 Kaposi’s sarcoma 1173 multiple myeloma 1084, 1085, 1088–9, 1089–90 myelodysplastic syndromes 1013 renal cell cancer 742 thallium-201 SPECT 294, 294, 1177 thallium scanning 123, 414 thermotherapy, transpupillary 332 thigh, soft tissue sarcoma 950, 950, 951, 951, 954 thioetpa, paediatric glioma 1144 6-thioguanine (6-TG) 86, 986 thiotepa 83, 622, 623 thorax, immobilization 1236, 1236 3G4 233 thrombocytopenia 209, 525 thromboembolism bevacizumab therapy 241–2 pulmonary 1204–5 venous 1204 vulval cancer surgery 802 thrombospondin-1 (TSP-1) 254, 626 thymidine phosphorylase (TP) 684 thymidylate synthetase (TS) 84, 85, 94 inhibitors 684, 685 thymus, involvement in NHL 137 thyrocytes, effects of oestrogen 409 thyroglobulin (Tg) 413, 416, 420, 420 thyroid cancer aetiology 408–9 anaplastic 411–12, 426–7, 427 chemotherapy 426 diagnostic evaluation 412–15, 413 differentiated endocrine treatment 417–18 prognosis 415, 415, 416, 417 surgery 416–17 epidemiology 408 follicular 408, 410–11, 411 follow-up 420, 421 future prospects 431–2 iodine-negative disease 432 lymphoma 411, 412, 414 management 430–1, 431
medullary cell 411, 450–1, 456 familial 412, 412, 430 gene therapy 432 management 428–30, 429 metastatic 421–2, 428 mortality 408 papillary 408, 410, 411 pathogenesis 409–10 pathology and natural history 410–12 radio-iodine ablation 418–21 complications 423–4 dosimetry 422–3 radiotherapy (external beam) 424–6, 425, 426 rare tumours 412 surgery 416 thyroidectomy 416, 419 thyroid function tests 421 thyroid gland, radiation damage 362 thyroid nodules 412–13, 413 thyroid replacement, pituitary tumours 449 thyroid stimulating hormone (TSH) 209, 417–18 after thyroidectomy 419 recombinant (rhTSH) 420–1 thyroid transcription factor-1 (TTF-1) 510 thyroxine 418, 449 timeline of cancer 3–4, 4 tinnitus 397 tipifarnib (zarnestra) 987, 1013 TIP regimen 722 tirapazamine 48, 92, 262, 355, 519, 1282 tissue engineering 351 tissue necrosis, brachytherapy 1220 tissue plasminogen activator 1205 TLS-CHOP fusion gene 928 TNM staging anal cancer 696–7, 697 bile duct cancer 586 bladder cancer 618–19, 619 colorectal cancer 672, 672–3 gall-bladder cancer 593 gastric cancer 605–6, 606 hepatocellular carcinoma 562–3, 563 lung cancer 507–8, 508, 509 oesophageal cancer 542, 543 pancreatic cancer 575, 575 prostate cancer 649, 649 renal cell carcinoma 733, 734 soft tissue sarcoma 938–9, 938 testicular germ-cell tumours 709, 709 tobacco ban 504 smokeless 376 use and cancer risk 346, 375, 376, 379 see also cigarette smoking tocopherols 349 tolerance, immunological 171 Toll-like receptors 849 tomotherapy 1266, 1267 tongue tumours
brachytherapy 1220, 1225–6 mobile tongue 377–8 posterior third of tongue 380 reconstructive surgery 377 tonsillar tumours 374, 380–1, 381 topogram 1239, 1239 topo-isomerase II 480 topo-isomerase inhibitors 89, 94–5 in brain tumours 300–1 leukaemia risk 982 see also irinotecan (CPT-11) topotecan 517, 526, 986, 1126 mechanism of action 89 toxicity 89 tositumomab 229 see also iodine-131-tositumomab total body irradiation (TBI) 1102–3, 1102 total mesorectal excision (TME) 674 total skin electron beam therapy 857, 858–60 toxicology testing 80, 1409 toxoplasmosis, cerebral in HIV infection 1177 TP53 gene 457 mechanism of action 1322 trabectidin (ET-743) 961 trachea, stent placement 153, 155 tracheo-bronchial fistula 541 tracheo-oesophageal fistula 153 tracheostomy 350 TRAIL 101, 278 tramadol 1292 tranexamic acid 789–90, 1298, 1299 transarterial chemoembolization (TACE) 565 transcription 2 transcriptomics 29 transcutaneous electrical nerve stimulation (TENS) 1154 transforming growth factor-α (TGF- α) 647 transforming growth factor-β (TGF- β) 179 bladder cancer 622 Kaposi’s sarcoma 1169 transfusions, red cell 1012 transient abnormal myelopoiesis (TAM) 1124 transition-cell carcinoma (TCCs), prostate 648 translation 24–5 transpupillary thermotherapy 332 transrectal ultrasound (TRUS) 132, 650, 797 transurethral resection of bladder tumour (TURBT) 620, 1285 transurethral ultrasound (TUUS) 620 transvaginal ultrasound (TVUS) 134, 135, 751, 758 transverse rectus abdominis myocutaneous (TRAM) flap 472 trastuzumab 228, 238–9, 480–1, 481, 1424
Index 1485
cardiac toxicity 1196 development 33 mechanism 98 trauma, in cancer aetiology 930, 963 traztuzumab 95 treatment planning 1189 treatment response, imaging 138–40, 138 treosulfan 83 tretinoin 838 tricyclic antidepressants 1293–4, 1300–1 tri-iodothyronine (T3) 418 trismus 361, 391 trisomy chromosome 8 928, 962–3, 1131 chromosome 13q 1076 chromosome 20 962 Troisier’s sign 575 Trousseau’s sign/syndrome 575, 750 troxacitabine 988 Tru-Cut needle biopsy, soft tissue sarcoma 933 truth-telling 1310 T10 regimen 915 tuberous sclerosis 26, 288, 731–2, 1120 tubulin-binding agents 258–9 tumour ablation ethanol injection 152, 564, 565–6 liver tumours/metastases 151–2, 151, 564, 565–6 radiofrequency 151–2, 151, 564, 565–6, 735 radio-iodine 418–24 renal tumours 152, 152, 735 tumour-associated antigens 168–71, 186 alteration in expression 179 cancer-testis 169, 170 classification 169–70, 169 colorectal cancer 686 differentiation 169, 170 idiotypic 171 lung cancer 510 onco-foetal 169, 170 onco-viral 169, 170 over-expressed antigens/mutated antigens 169, 170 post-translationally altered 171 ‘self’ 175 tumour-specific unique 169, 170–1 tumour-associated antigens (TAA), brain tumours 292 tumour-cell burden, clonogenic 52 tumour cells repopulation 46–7, 1281 accelerated 355 sub-lethal damage inhibition of repair 1282 repair 45, 68 tumour embolism, pulmonary 825–6, 826 tumour escape, mechanisms 178–9 tumour ‘evolution’ 90, 187, 187 tumour growth and circadian rhythms 356
doubling times 45–6, 46, 76 kinetics 75–7, 352–3, 352, 352, 353 tumour hypoxia 47–9, 91–2, 1282 assessment 49 brain tumours 291 and chemo/radiotherapy resistance 353–5 overcoming 47–8 and prognosis in solid tumours 255 targeting 262 and tumour angiogenesis 252 tumour-infiltrating lymphocytes (TILs) 195–6 tumour lysates 175 tumour lysis syndrome 1197 tumour markers 1425 biliary tract cancer 588, 594 bladder cancer 621–2, 622 CA 125 134 CA 125 765 carcinoembryonic antigen (CEA) 170, 1190 endometrial cancer 765 human chorionic gonadotrophin (hCG) 815 lung cancer 510–11 ovarian cancer 134, 135, 751, 752, 755 pancreatic cancer 576 prostate cancer 646, 650, 651, 653, 657 testicular tumours 709–10, 710, 717, 1425 tumour necrosis, induction 48 tumour necrosis factor-α (TNF- α) 1075, 1076 biological effects 190 cachexia 1197 isolated limb perfusion 952–3 malignant melanoma 889 therapeutic potential 190, 194 tumour suppressor genes 25–6, 26, 27–8, 277, 1320, 1322 gene therapy 277 in hereditary cancer syndromes 26 and HPV 778, 784 inactivation in paediatric cancer 1120 lung cancer 504, 504 manipulation in cancer therapy 102 pancreatic cancer 574 squamous cell carcinoma of skin 851 tumour thrombus, renal cell cancer 1425, 1426 tumstatin 254 Turcot’s syndrome 288, 409, 670, 1319, 1329 turmeric 1438 Turner’s syndrome 811, 1120 2C4, see pertuzumab tylosis 539, 1203 tyrosinase 169–70, 170, 884 tyrosine kinase inhibitors 98, 228–9, 1424 carcinoid tumours 453–4 EGFR 31–2, 99, 348, 366, 520–1
multi-targeted 742–3, 742 pancreatic cancer 580 PD166326 1003 in renal cell cancer 742–3, 742 second generation 1002 dasatinib 1002 nilotinib 995, 1002 third generation 1002–3 see also imatinib mesylate U-01 101 ubiquitin 95 ubiquitin-proteosome system 102 UK breast cancer outcomes 1355 hepatocellular carcinoma 559 lung cancer epidemiology 502, 502 UK cancer services 1359 Calman-Hine report 1379–80, 1379 cancer networks 1377, 1381, 1382, 1398–9 commissioning cancer services 1399 dividends of 1399 management teams 1398, 1399 network service service plans 1398 workforce strategy 1398 cancer survival figures 1379, 1385 clinical research 1384–5 clinical standards monitoring 1384 Commission for Health Improvement/Audit Commission Report 1396–7, 1397 costs and state investment 1403 crisis 1378 future directions 1403–4 modernization 1399, 1400, 1401 national cancer plans 1380–1 progress against targets 1385, 1386–96 patient experience 1377, 1378, 1385 pattern of organization 1377, 1403–4 reorganization effectiveness of 1385–92 England 1381, 1382 Northern Ireland 1383–4, 1384 primary care 1401–2 Scotland 1381–2, 1382 Wales 1382–3 specialization 1378, 1380 whole system approach 1400 UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) 135 UK-ET trials 1132 Ukraine 345 ulcerative colitis 668–9 ultra-small particles of iron oxide (USPIO) 120, 126, 136, 137 ultrasonography 113–14 biopsy guidance 147, 150 breast 129, 1242 cholangiocarcinoma 588, 589 contrast agents 113–14
1486 Index
ultrasonography (contd.) Doppler 113, 126, 751–2 endo-oesophageal 541–2 head and neck cancers 125 high-intensity focused (HIFU) 114, 655–6 intracavity 113 liver lesions 114, 114 ovarian cancer 751–2 pancreatic tumours 576 principles of 113 radiotherapy planning 1242, 1270, 1270 thyroid gland 413 transrectal (TRUS) 132, 650, 797 transurethral (TUUS) 620 transvaginal (TVS) 751, 758 transvaginal (TVUS) 134, 135, 766 ultraviolet light exposure 836, 841, 879, 880 umbilical cord blood 1099 United States (USA), hepatocellular carcinoma 559 uracil and tegafur (UFT)-based chemotherapy 515–16, 516 urate nephropathy 1196–7 ureteric stenosis 791 urinary cytology 618 urinary tract obstruction 1196 urinary diversion 150–1, 151 urine analysis bladder cancer 618 renal cell carcinoma 733 urogram, intravenous 112 urothelial (transitional) cell carcinoma 615 USA breast cancer outcomes 1355 economics of cancer 1353–6, 1353 President’s Cancer Panel (USA) 1358, 1359 prostate cancer screening 646 War on Cancer 1353 US Intergroup Ewing’s Sarcoma Study (IESS) 1131 US Nurses’ Health Study 1435 uterus sarcoma 772–3 see also endometrial cancer; gestational trophoblastic disease (GTD) uveal tumours, malignant 327, 330, 331–3 VACA regimen 1131–2 VAC chemotherapy 327 vaccines 7, 171 adjuvants 174 autologous cell, renal cell cancer 740 chronic lymphocytic leukaemia (CLL) 1009 chronic myeloid leukaemia (CML) 1002 clinical study limitations 179–80 colorectal cancer 686 combined with cytokines 194 delivery systems 177–8
dendritic-cell 174–6 DNA 176 GM-CSF transfected cells 895 GM-ganlioside-keyhole limpet hemocyanin) 192 heat-shock proteins 172–3, 892 human papillomavirus (HPV) 1181 malignant melanoma 887–9, 888, 892–3, 895 peptide 173–4 recombinant viral/bacterial 177 T-cell response 174 whole-cell 172 vaccinia viruses 175, 177 VAC regimen 757, 918, 1131–2 vacuolating cytotoxin (VacA) 603 VAD regimen 1083, 1084 vagina anatomy 796–7 changes after pelvic radiotherapy 792 radiotherapy, complications 796, 798 reconstructive surgery 798 vaginal bleeding 780, 815, 817, 825 vaginal cancer aetiology 796 clear cell adenocarcinoma 799 clinical staging 797, 797 diagnosis and assessment 797 endodermal sinus 799 epidemiology 796 melanoma 799 natural history 797 pathology 797 rhabdomyosarcoma 799 treatment 798 vaginal discharge 780 vaginal fistula 701 vaginal intra-epithelial neoplasia (VAIN) 796, 797 diagnosis 797 natural history 797 pathology 797 treatment 798 vaginectomy (colpectomy) 798 VAIA regimen 1132 valproic acid 987 VAMP regimen 1084 van Ness rotation plasty 1134 Van Nuys Prognostic Index 471, 471 VAPEC-B therapy, Hodgkin’s lymphoma 1035, 1036 vascular disrupting agents (VDAs) 256, 256, 258–62 ligand-directed 259 pharmacodynamic assessment 261–2 small molecule 258–60, 259 vascular embolization 158–62 hepatic artery 160–2, 161, 454, 564–5 palliative 160–2, 161 portal vein 590 renal artery 159–60, 160
vascular endothelial growth factor receptor (VEGFR) 253–4, 1283 antibody, see bevazicumab inhibitors 521, 894 vascular endothelial growth factor (VEGF) 179, 647 in AML 987–8 discovery 251 monoclonal antibodies 31, 228, 240–2, 251, 741–2, 742 myeloma 1075 renal tumours 731, 732 roles of 253 tyrosine kinase inhibitors 731, 742, 742 vascular ‘hot spots’ 255 vascular sarcomas 925 vascular tumours, eye 330, 331, 335 vasculogenesis 252 vasculogenic mimicry 252 vasoactive drugs, tumour hypoxia 48 vasoactive intestinal peptide (VIP) 507 vastus lateralis 940 vatalanib (PTK787/ZK222584) 257 VEEP regimen 1140 vegetable consumption, see fruit and vegetable consumption VEGF-trap 257 VEGRF, glioblastoma 304 VeIP regimen 722 vena caval filter 154, 156 Venezuelan equine encephalitis virus (VEE) 272 venous catheters access 1205 extraction of migrated 156, 158, 158 peripherally inserted central (PICC) 158 tunnelled external 158 venous sampling 150, 150 venous thromboembolism, bevacizumab therapy 241–2 venous thrombosis, diagnosis and prevention 1204 ventilation-perfusion (V/Q) scan, pulmonary embolism 1205 verrucous carcinoma, vulva 805 vertebrae, percutaneous biopsy 149–50, 149 vertebroplasty, percutaneous 162, 163 vesico-vaginal fistula 798 VHL gene 95–6, 731, 732, 741 VICE regimen 522 vinblastine 87 bladder cancer 630–2, 630, 632 Hodgkin’s lymphoma 1032, 1033 intralesional, Kaposi’s sarcoma 1171 lung cancer 517, 518 non-melanoma skin cancer 852 testicular germ cell tumours 715 vinca alkaloids 87 vincristine 87 ALL 993 CNS tumours 301, 306, 307, 309, 1146
Index 1487
Ewing’s tumour 918, 1131–2 gestational trophoblastic tumours 823–4, 824 multiple myeloma 1084, 1085 non-melanoma skin cancer 852 orbital rhabdomyosarcoma 326–7 ovarian germ cell tumours 757 soft tissue sarcomas 954–6, 955 see also chemotherapy regimens containing vincristine vindesine 87, 451, 517, 518 vinorelbine 516, 516, 517, 518, 519, 520, 520 vinyl chloride 930 VIPomas 451 VIP regimen 717 viral infections and cervical neoplasia 778 head and neck cancer 347 Hodgkin’s lymphoma aetiology 1028 nasopharyngeal cancer 345, 388 post stem cell transplant 1108 viral vectors (gene therapy) 269–74, 270, 271 oncolytic 272–4, 273 Virchow’s node 587 virilization 457, 458 virus-like particles (VIPS) 177–8, 1181 visceral flaps, pedicled 351 visual disturbance 293 visual loss, radiotherapy 394 vitamin A 102, 346, 625 vitamin B12 absorption 791 vitamin C 346 vitamin D, prostate cancer 646 vitamin D analogues 103 vitamin D receptor 449 vitamin E 1436 VMCP regimen 1085 vocal cords fixation in laryngeal cancer 383 Teflon injection 506 voice changes, laryngeal cancer 383–4 Voices of a Broken System 1359 vomiting advanced disease 1296–7 after stem cell transplantation 1105 in bowel obstruction 1198–9 causes 1296 management 1296–7 paediatric cancer patient 1148 von Hippel-Lindau syndrome 330, 454, 457, 1120, 1321 brain tumours 288 patient screening and management 1331 renal tumour 152, 731, 732 von Hippel-Lindau (VHL) gene 95–6, 731, 732, 741 von Hippel-Lindau (VHL) protein 252 VP-16 see etoposide brain tumours 300
vulva anatomy 799 lymphatic drainage 799–800 vulval cancer Bartholin’s gland carcinoma 805 basal cell carcinoma 805 diagnosis and assessment 800–1 epdiemiology 799 melanoma 804–5 natural history 800 Paget’s disease 799, 800, 805 pathology 800 sarcoma 805 staging 800 treatment 801–4 verrucous carcinoma 805 vulval intra-epithelial neoplasia (VIN) 800–1 vulvectomy, radical 801 WAGR syndrome (Wilms-aniridiagenitourinary abnormalitiesretardation syndrome) 1120, 1322 waiting times radiotherapy 1372, 1379, 1401 referral to treatment 1372, 1385, 1401 Wales cancer services 1377 reorganization 1382–3 report 1380 warfarin 1204, 1205 pulmonary embolism 1205 War On Cancer (USA) 1353 ‘watchful waiting’, prostate cancer 653 water retention, in SIADH 1202–3 wealth 5–6, 6, 6 Weber-Christian disease 1203 weight gain 1298 weight loss 1197–8 advanced disease 1298 head and neck cancer 368 pancreatic cancer 575 radiotherapy 360 Wermer’s syndrome, see multiple endocrine neoplasia 1 (MEN-1) Westermark sign 1205 Western diet 7 Western Europe, bladder cancer 614 Whipple’s procedure 577 Whipple’s triad 451 WHO, see World Health Organization whole-brain radiotherapy (WBRT) 893, 1177, 1200 Wilms-aniridia-genitourinary abnormalities-retardation (WAGR) syndrome 1120 Wilm’s tumour 26, 731, 1321, 1322, 1332, 1424 epidemiology 1119, 1119, 1136 genetics 1120 pathology 1136 presentation 1136
staging 1136, 1137 treatment 1137 Winslow node 137 Wiskott-Aldrich syndrome 1121, 1140 women bladder cancer incidence 614 lung cancer incidence and mortality 502, 502, 503 soft tissue sarcoma 926 Women’s Interagency HIV Study (WIHS) 1180 Woringer-Kolopp disease 862, 862 World Bank 5 World Cancer Report 1362 World Health Organization, classification of cutaneous lymphoma 853, 854, 855 World Health Organization (WHO) anal cancer classification 695 analgesic ladder 1291 brain tumour classification 289–90, 289, 310 endocrine tumour classification 447, 451 gastric tumour classification 606 gestational trophoblastic disease classification 809 health report 5 Hodgkin’s lymphoma classification 1028–9, 1028 leukaemia classifications 981 ALL 991, 991 AML 983–4, 984, 1123 myelodysplastic syndrome classification 1011, 1012 nasopharyngeal carcinoma classification 388 non-Hodgkin’s lymphoma classification 1045, 1047, 1047 paediatric CNS tumour classification 1143, 1143 pancreatic tumour classification 575 prostate cancer grading 648 response to treatment criteria 139 wortmannin 1003 wound complications 944–5, 945, 953–4, 954 wound healing, chronic 836 wounds, fungating 1299 xanthoastrocytoma, pleomorphic 307 xeroderma pigmentosum 26, 838–9, 1321 xerosis 838, 839 xerostomia 360–1, 367, 375, 1295–6 X-linked inheritance 1320 X-ray film 112 X-rays 112–13 advantages/disadvantages of 112 lung cancer screening 505 multiple myeloma 1078 primary bone tumours 913 soft tissue sarcoma 931
1488 Index
X-rays (contd.) testicular germ-cell cancer 711, 711 X-ray volumetric imaging 1249
yttrium-90-ibritumomab tiuxuetan 1053, 1058 yttrium-90 therapy 432, 452, 566–7
yolk-sac tumours 313, 709 yttrium-90-ibritumomab tiuxetan 205–7, 229 clinical experience 205–7 clinical trials 212 toxicity and safety 208–9
zarnestra (tipifarnib) 987, 1013 ZD6126 259, 260, 261 ZD6474 257 Z-Dex therapy 1084 zidovudine 1056 Zimbabwe 559
zoledronic acid breast cancer 481, 491 hypercalcaemia 1202 multiple myeloma 1082, 1089 prostate cancer 258, 658 side effects 1089 Zollinger–Ellison syndrome 451 zosuqidar (LY335979) 987