This page intentionally left blank
Practical Clinical Oncology
Practical Clinical Oncology has been designed as a ha...
70 downloads
1619 Views
3MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
This page intentionally left blank
Practical Clinical Oncology
Practical Clinical Oncology has been designed as a handson review of all aspects of current practice in clinical oncology. The introductory section to the book provides background information on the four main treatment modalities – radiotherapy, chemotherapy, hormone therapy and biological therapy – plus research, oncological emergencies and palliative care. Subsequent chapters describe the diagnosis and treatment of malignancies, based on tumour site or type. Each chapter follows a standard template, making it easier for the readers to locate information quickly; multiple choice questions are also provided to enable readers to test their knowledge. With an emphasis on practical information that will be useful in day-to-day decision making and treatment, Practical Clinical Oncology is an invaluable resource on clinical care of the cancer patient for all trainees in clinical oncology, medical oncology, surgical oncology and palliative care, as well as for specialist nurses and radiographers. Louise Hanna is a Consultant Clinical Oncologist at the Velindre Cancer Centre, Velindre Hospital, Cardiff. Tom Crosby is a Consultant Clinical Oncologist at the Velindre Cancer Centre, Velindre Hospital, Cardiff. Fergus Macbeth is a Consultant Clinical Oncologist at the Velindre Cancer Centre, Velindre Hospital, Cardiff.
Practical Clinical Oncology Edited by
Louise Hanna Tom Crosby Fergus Macbeth
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521618168 © Cambridge University Press 2008 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2008
ISBN-13 978-0-511-37854-6
eBook (NetLibrary)
ISBN-13
paperback
978-0-521-61816-8
Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate. Every effort has been made to ensure that the information in this text, including drug and radiotherapy doses, is correct. However, the reader is strongly advised to consult the published product information and data sheets provided by manufacturers. The authors and publishers cannot accept any legal responsibility or liability for any errors or omissions which occur in this book or for the misuse or misapplication of the material therein
CONTENTS
List of contributors Preface Acknowledgements Abbreviations
page vii ix x xi
14 Anus
174
Richard Adams and Tom Crosby
15 Gastrointestinal stromal tumours
183
Kate Parker and Tom Crosby
1 Practical issues in cytotoxic chemotherapy usage
1
16 Breast
190
Nayyer Iqbal and Peter Barrett-Lee
Sian Evans and Philip Savage
17 Kidney 2 Biological treatments in cancer
13
214
Jason Lester and John Wagstaff
Rachel Jones and Robert Leonard
18 Bladder 3 Hormones in cancer
23
Jacinta Abraham and John Staffurth
4 Radiotherapy planning
19 Prostate 39
Andrew Tyler and Louise Hanna
5 Research in cancer Robert Hills
6 Oncological emergencies
70
7 Palliative care
83
8 Head and neck
93
9 Oesophagus
121
10 Stomach
132
11 Liver, gallbladder and biliary tract
141
12 Exocrine pancreas
151
13 Colon and rectum Richard Adams, Timothy Maughan and Tom Crosby
304
Philip Savage
28 Lung 159
296
Louise Hanna and Malcolm Adams
27 Gestational trophoblast tumours
Somnath Mukherjee and Tom Crosby
290
Louise Hanna and Malcolm Adams
26 Vulva
Somnath Mukherjee and Tom Crosby
278
Louise Hanna and Malcolm Adams
25 Vagina
Michael Button and Tom Crosby
267
Louise Hanna and Malcolm Adams
24 Cervix
Tom Crosby
257
Louise Hanna and Malcolm Adams
23 Body of the uterus
Laura Moss and Chris Gaffney
252
Jim Barber
22 Ovary
Simon Noble
240
Jim Barber and John Staffurth
21 Penis
Paul Shaw
231
Jim Barber and John Staffurth
20 Testis 55
222
Stephen Williams and Jim Barber
313
Fergus Macbeth and Carys Morgan
29 Mesothelioma
328
Louise Hanna and Fergus Macbeth
v
Contents
30 Soft tissue and bone tumours in adults
335
Owen Tilsley
31 The lymphomas and myeloma
347 370 382 395
Tom Crosby, Louise Hanna and Dafydd Roberts
35 Thyroid Laura Moss
vi
426
38 Cancer of unknown primary
442
Paul Shaw and Tom Crosby
Sankha Suvra Mitra
34 Melanoma
37 Cancer in children Sally Goodman
Sean Elyan
33 Skin cancer other than melanoma
418
Atul Kalhan and Aled Rees
Eve Gallop-Evans and Chris Poynton
32 Central nervous system
36 Neuroendocrine tumours
406
39 The use of radiotherapy in the treatment of benign conditions
449
Alison Brewster
Multiple choice questions Multiple choice answers Index
453 468 469
CONTRIBUTORS
Jacinta Abraham
Eve Gallop-Evans
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Malcolm Adams
Sally Goodman
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Consultant Clinical Oncologist, The Bristol Haematology and Oncology Centre, Bristol, UK.
Richard Adams Senior Lecturer in Clinical Oncology, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Louise Hanna Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Jim Barber Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Robert Hills Senior Lecturer in Translational Statistics, Department of Haematology, Cardiff University, Cardiff, UK.
Peter Barrett-Lee Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Alison Brewster Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Michael Button Specialist Registrar in Clinical Oncology, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Tom Crosby Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Sean Elyan Consultant Clinical Oncologist, Cheltenham General Hospital, Cheltenham, UK.
Nayyer Iqbal Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Rachel Jones Specialist Registrar in Medical Oncology, South West Wales Cancer Institute, Singleton Hospital, Sketty, Swansea, UK.
Atul Kalhan Specialist Registrar, Centre for Endocrine and Diabetes Sciences, University Hospital of Wales, Heath Park, Cardiff, UK.
Robert Leonard Professor of Medical Oncology, South West Wales Cancer Institute, Singleton Hospital, Sketty, Swansea, UK.
Sian Evans
Jason Lester
Chief Pharmacist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Chris Gaffney
Fergus Macbeth
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
vii
Contributors
Timothy Maughan
Aled Rees
Professor of Cancer Studies, Consultant Clinical
Senior Lecturer, Centre for Endocrine and Diabetes
Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Sciences, School of Medicine, Cardiff University, Heath Park, Cardiff, UK.
Sankha Suvra Mitra Consultant Clinical Oncologist, The Sussex Cancer Centre, Royal Sussex County Hospital, Brighton and Sussex University Hospitals NHS Trust, Brighton,
Dafydd Roberts Consultant Dermatologist, Singleton Hospital, Sketty, Swansea, UK.
Philip Savage
East Sussex, UK.
Consultant Medical Oncologist, Department of Health
Carys Morgan
Gestational Trophoblastic Tumour Unit, Charing Cross Hospital, London, UK.
Specialist Registrar in Clinical Oncology, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Paul Shaw
Laura Moss
University, Park Place, Cardiff, UK.
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
John Staffurth
Somnath Mukherjee Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Simon Noble Senior Lecturer and Honorary Consultant in Palliative Care, Royal Gwent Hospital, Newport, UK.
Kate Parker Specialist Registrar in Clinical Oncology, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Specialist Registrar in Clinical Oncology, Bobby Moore Clinical Research Fellow, School of Bioscience, Cardiff
Clinical Senior Lecturer in Oncology, Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Owen Tilsley Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
Andrew Tyler Medical Physicist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
John Wagstaff Consultant Medical Oncologist, Singleton Hospital, Sketty, Swansea, UK.
Chris Poynton
Stephen Williams
Senior Lecturer and Honorary Consultant Haematologist, University Hospital of Wales, Heath Park, Cardiff, UK.
Specialist Registrar in Clinical Oncology, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.
viii
PREFACE
This book is intended primarily for trainees in clinical oncology, but members of other professions such as medical oncology, surgery, palliative care, nursing and radiography will also find it useful. The book started life as a set of lecture notes from the Cardiff Annual FRCR Part II course but has since grown to include more topics than could possibly be covered during the three days of that course. Our approach in producing this volume has been to focus on practical suggestions appropriate to day-to-day decision making during the treatment of oncology patients. We are very grateful to our colleagues from Velindre and elsewhere, who are listed on page xi, for reviewing specific chapters and ensuring that the advice contained within is as widely applicable as possible. The first seven chapters cover ‘generic’ topics that provide background information on cancer treatments. These are chemotherapy, biological and hormonal treatments, radiotherapy planning, research, emergencies and palliative care. The chapters that follow each focus on a tumour site or tumour type. In this latter group, the chapter layout is fairly consistent to help the reader navigate through the book. Thus, each chapter begins with background information on tumour types, anatomy, incidence, epidemiology, risk factors and
aetiology. Next, there are sections on pathology, routes of spread and, where appropriate, screening. These are followed by clinical sections on presentation, investigations, treatment, and prognosis. Most of the chapters also discuss areas of current interest and clinical trials, reflecting the rapidly changing nature of clinical oncology where many areas of practice are open to debate. Where references are given, we have tried as much as possible to include the key publications that have influenced clinical practice. Towards the end of the book, there is a series of ‘single best answer’ multiple choice questions, which give the readers the opportunity to test their knowledge. In a book of this length, it is not possible to provide as much of the subject as would be found, for example, in the larger multivolume oncology textbooks. Nevertheless, an attempt has been made to give an overview of clinical oncology practice at the present time, which we hope will be of interest and benefit to trainees. The idea for writing this book came about several years ago when two of the editors (TC and LH) were studying for their FRCR part II examination. They have since become consultants in Velindre Hospital with FM and all three now teach on the Cardiff Annual FRCR part II course.
ix
ACKNOWLEDGEMENTS
We are very grateful to those who have helped by reviewing specific chapters: Dr K. Benstead, Dr G. Bertelli, Dr P. Blake, Dr N. Burnet, Mr A. Carter, Dr M. Evans, Dr S. Falk, Prof A. Fiander, Dr C. Gaffney, Dr J. Glees, Dr P. Keeley, Dr V. Khoo, Dr U. Mallick, Dr M. Marples, Prof M. Mason, Dr C. Nutting, Ms T. Perrett, Dr N. O’Rourke, Dr D. Sebag-Montefiore, Dr N. Somaiah, Prof R. Taylor, Prof H. Thomas, Dr A. Wardley, Dr S. Williams and Dr V. Wolstenholme. Our thanks also go to those who have helped with the preparation of this book: Jane Flewitt, Rhydian Maggs
x
and Lucy Wills of the Department of Medical Physics, Velindre Cancer Centre; Anne Cleves of the Cancer Research Wales Library; and the staff at Cambridge University Press, particularly Pauline Graham, Betty Fulford and Eleanor Umali who have always responded helpfully to our many queries. And, finally, we acknowledge our respective families for their enduring love and support during the preparation of this book.
ABBREVIATIONS
General 1D 2D 3D 5AC 5-FU 5-HIAA 5-HT3 5YDFS 5YOS 5YS αFP βhCG AAPM ABCSG ABL ACE ACh ACOSOG ACTH ACTION ADH ADT AF AGES AHT AI AIDS AIM HIGH AJCC AKT ALL ALM
ALMANAC one-dimensional two-dimensional three-dimensional MUC subtypes A and C 5-fluorouracil 5-hydroxy-indoleacetic acid 5-hydroxy-tryptamine 3 5-year disease-free survival 5-year overall survival 5-year survival alpha feto-protein beta human chorionic gonadotrophin American Association of Physicists in Medicine Austrian Breast and Colorectal Cancer Study Group Abelson protein tyrosine kinase anticholinesterase acetylcholine American College of Surgeons Oncology Group Adrenocorticotrophic hormone Adjuvant Chemotherapy in Ovarian Neoplasm antidiuretic hormone androgen deprivation therapy activating function age, grade, extent, size adjuvant hormone therapy aromatase inhibitor aquired immune deficiency syndrome Adjuvant Interferon in Melanoma HIGH risk American Joint Committee on Cancer thymoma viral proto-oncogene acute lymphoblastic leukaemia acral lentinginous melanoma
ALND AMES AML ANC APBI AP A–P AP-1 ApC APC APR APUD AR ARE ARNO ARSAC ASCO ASCT ASH ASTEC ASTRO ATAC ATD ATLAS ATP aTTom AUC b.d. BC BCG BCL2 BCR
Axillary Lymphatic Mapping Against Nodal Axillary Clearance axillary lymph node dissection age, metastases, extent, size acute myelocytic leukaemia axillary nodal clearance accelerated partial breast irradiation abdominoperineal anterior–posterior activator protein-1 antigen-presenting cell adenomatosis polyposis coli abdominoperineal resection amine precursor uptake and decarboxylation androgen receptor androgen response elements arimidex-nolvadex Administration of Radioactive Substances Advisory Committee American Society of Clinical Oncology autologous stem cell transplant American Society of Hematology A STudy in Endometrial Cancer American Society for Therapeutic Radiology and Oncology Arimidex, Tamoxifen, Alone or in Combination amino-terminal domain Adjuvant Tamoxifen Longer Against Shorter adenosine triphosphate adjuvant Tamoxifen – Treatment offer more area under curve twice a day British Columbia Bacille Calmette-Gu´erin B-cell CLL/lymphoma 2 breakpoint cluster region xi
Abbreviations
BCT BGND BIG BILCAP BIR BMD BMI BMT BNLI BOND BP BRAF BRCA BSA BSC BSO BTOG BTS CA CagA CAIX CALGB Cb CCLG CD CDH-1 CE CEA CgA CHART CHF CI CIN CIS CK CLA CLCG CLL CML cN CNS xii
breast conservation therapy bilateral groin node dissection Breast International Group adjuvant capecitabine in biliary tract cancer trial British Institute of Radiology bone mineral density body mass index bone marrow transplant British National Lymphoma Investigation Bowel Oncology with Cetuximab Antibody blood pressure v-raf murine sarcoma viral oncogene homolog B1 breast cancer gene body surface area best supportive care bilateral salpingo-oophorectomy British Thoracic Oncology Group British Thoracic Society cancer antigen cytotoxicity-associated immunodominant antigen carbonic anhydrase IX Cancer and Leukaemia Group B carboplatin Children’s Cancer and Leukaemia Group cluster of differentiation cadherin 1 conversion electron carcino-embryonic antigen chromogranin A continuous hyperfractionated accelerated radiotherapy congestive heart failure confidence interval cervical intraepithelial neoplasia carcinoma in situ cytokeratin common leukocyte antigen Children’s Leukaemia Cooperative Group chronic lymphocytic leukaemia chronic myelocytic leukaemia clinical lymph node stage central nervous system
COG COMS CONSORT COPD CR CRC CrCl Creat CRM CRP CRT CRUK CSF CSI CT CTA CTV CTZ CUP CVP CX CXR CYP1B1 D D2 DAHANCA DCC DCIS DDFS DDT DES DFS DHA DHT DLBCL dmax DMC DMSA DMSO DNA DOPA DRE DRR DTC DTIC DVH
Children’s Oncology Group of North America Collaborative Ocular Melanoma Study Consolidated Standards of Reporting Trials chronic obstructive pulmonary disease complete response colorectal cancer creatinine clearance serum creatinine circumferential resection margin c-reactive protein chemoradiotherapy Cancer Research UK cerebrospinal fluid craniospinal irradiation computed tomography clinical trials authorisation clinical target volume chemoreceptor trigger zone cancer of unknown primary central venous pressure characteristic X-ray photon chest X-ray cytochrome P450, family 1, subfamily B, polypeptide 1 docetaxel dopamine D2 Danish Head and Neck Cancer deleted in colon cancer ductal carcinoma in situ distant-disease-free survival dichloro-diphenyl-trichloroethane diethylstilboestrol disease-free survival dihydroxyandrostenedione 5α dihydrotestosterone diffuse large B-cell lymphoma depth of maximum dose Data Monitoring Committee dimercapto succinic acid dimethyl sulfoxide deoxyribonucleic acid dihydroxyphenylalanine digital rectal examination digitally reconstructed radiograph differentiated thyroid cancer dacarbazine dose volume histogram
Abbreviations
EAU EBC EBCTCG EBRT EBV ECG Echo ECOG EDTA EGF EGFR EIC ELISA ELND EM EMA ENET ENT EORTC EPO EPP ER ERB B1
ERBB2
ERCP ERE ERG ESPAC ESR ESTRO EU EUA EURAMOS
European Association of Urology early breast cancer Early Breast Cancer Trialists’ Collaborative Group external beam radiotherapy Epstein Barr virus electrocardiogram echocardiogram Eastern Cooperative Oncology Group ethylenediaminetetraacetic acid epidermal growth factor epidermal growth factor receptor extensive intraductal component enzyme-linked immunosorbent assay elective lymph node dissection electron microscopy epithelial membrane antigen European Neuroendocrine Tumor Network ear nose throat European Organisation for Research and Treatment of Cancer erythropoietin extrapleural pneumonectomy oestrogen receptor official symbol = EGFR; official name = epidermal growth factor receptor (erythroblastic leukemia viral (v-erb-b) oncogene homolog, avian) (alias = HER-2); official symbol = ERBB2; official name = v-erb-b2 erythroblastic leukemia viral oncogene homolog 2, neuro/glioblastoma derived oncogene homolog (avian) endoscopic retrograde cholangiopancreatogram oestrogen response element v-ets erythroblastosis virus E26 oncogene homolog (avian) European Study Group for Pancreatic Cancer erythocyte sedimentation rate European Society for Therapeutic Radiology and Oncology European Union examination under anaesthetic European and American Osteosarcoma Study Group
EUS EWS FA FAP FBC FDA FDG FIGO FinHer FISH FL FLI1 FLIPI FLT3 fms FNA FNAC FOB FRCR FSD FSH FT4 FTC GANT GBq GC G-CSF GCP GCT GELA GFR GHRH GHSG GI GIST GLI GnRH GO GOG GORD GP GTT GTV Gy H2
endoscopic ultrasound EWSR1; official name = Ewing sarcoma breakpoint region 1 folinic acid familial adenomatous polyposis full blood count Food and Drug Administration fluorodeoxyglucose Federation Internationale de Gynecologie et d’Obstetrique Finland Herceptin Trial fluorescent in situ hybridisation follicular lymphoma Friend leukemia virus integration 1 Follicular Lymphoma International Prognostic Index fms-related tyrosine kinase 3 peptide deformylase fine needle aspiration fine needle aspiration cytology faecal occult blood Fellow of the Royal College of Radiologists focus skin distance follicle-stimulating hormone free T4 follicular thyroid carcinoma gastrointestinal autonomic tumour giga-Becquerel germinal centre granulocyte-colony stimulating factor good clinical practice germ cell tumour Groupe d’ Etude des Lymphomes de l’ Adulte glomerular filtration rate growth hormone-releasing hormone German Hodgkin Study Group gastrointestinal gastrointestinal stromal tumour glioma-associated oncogene homolog gonadotropin-releasing hormone gastro-oesophageal Gynecologic Oncology Group gastro-oesophageal reflux disease General Practitioner gestational trophoblast tumour gross tumour volume Gray histamine H2 xiii
Abbreviations
HAART HAI HART HBF HBL HBV HCC HCG HCV HDC/ASCT HDCT HDR HDU HER-2 HERA HGG HH HHV HIFU HIV HL HMB HNPCC HPOA HPV HR HRT HTLV-1 IBCSG ICAM1 ICC ICG ICON ICRU IDC IDL IES IFNα IFNAR IFNGR IFRT IGCCC
xiv
highly active antiretroviral therapy hepatic arterial infusion hyperfractionated accelerated radiotherapy heterotopic bone formation hepatoblastoma hepatitis B virus hepatocellular carcinoma human chorionic gonadotropin hepatitis C virus high-dose chemotherapy with autologous stem cell transplant high-dose chemotherapy high dose rate high-dependency unit human epidermal growth factor receptor 2 Herceptin® Adjuvant high-grade glioma hedgehog human herpesvirus high-intensity focussed ultrasound human immunodeficiency virus Hodgkin lymphoma human melanoma black hereditary non-polyposis colorectal cancer hypertrophic pulmonary osteo-arthropathy human papilloma virus hazard ratio hormone replacement therapy human T-cell lymphotropic virus -1 International Breast Cancer Study Group intercellular adhesion molecule 1 interstitial cells of Cajal iodocyanine green International Collaborative Ovarian Neoplasm study International Commission on Radiation Units and Measurements invasive ductal carcinoma indirect laryngoscopy International Exemestane Study interferon alpha interferon (alpha and beta) receptor interferon gamma receptor involved-field radiotherapy International Germ Cell Consensus Classification
IGCCCG IGCN IGF-1 IGRT IHC IHD IJV IL-2 ILC ILT i.m. IM IMIG IMN IMP IMRT INTACT IPFS IPI IRS ISO ISS ITA ITT ITU IU i.v. IVC IVU JDG JVP KIT
KPS KRAS LACE LAK LCIS LD LDH LDL LDR LE
International Germ Cell Cancer Collaborative Group intratubular germ cell neoplasia insulin-like growth factor-1 image-guided radiation therapy immunohistochemistry ischaemic heart disease internal jugular vein interleukin-2 invasive lobular carcinoma intraluminal brachytherapy intramuscular internal margin International Mesothelioma Interest Group internal mammary node investigational medicinal product intensity-modulated radiation therapy Iressa NSCLC Trial Assessing Combination Treatment international prognostic factor score international prognostic index Intergroup Rhabdomyosarcoma Studies International Organisation for Standardisation international staging system Italian Tamoxifen Arimidex trial intention to treat intensive therapy unit international units intravenous inferior vena cava intravenous urogram jugulo-digastric jugulo-venous pressure kitten (v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog) Karnofsky performance status v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog Lung Adjuvant Cisplatin Evaluation lymphokine-activated killer lobular carcinoma in situ latissimus dorsi; lymphocyte depleted lactate dehydrogenase low-density lipoprotein low dose rate local excision
Abbreviations
LEEP LFT LGG LH LHRH LHRHa LLETZ LM LMM LN LOH LR LVI MAB MACIS
MAG 3 MAGE MAGIC MALT MAP MAPK MART MBC MBq MC MDM MDR MDT MEN MF MFH MGMT MGUS MHRA MIBG MInT MLC MM MR MRC
loop electro-excision procedure liver function test low-grade glioma luteinising hormone luteinising hormone releasing hormone luteinising hormone releasing hormone agonist large loop excision of the transformation zone lentigo maligna lentigo maligna melanoma lymph node loss of heterozygosity local recurrence lymphovascular invasion maximal androgen blockade metastases, age, completeness of surgery, invasion of extrathyroidal tissues, size mercaptoacetyltriglycerine melanoma antigen expression family Medical Research Council Adjuvant Gastric Infusional Chemotherapy mucosa-associated lymphoid tissue mitogen activated protein mitogen-activated protein kinase melanoma antigen recognised by T-cells metastatic breast cancer mega-Becquerel mixed cellularity multidisciplinary meeting medium dose rate multidisciplinary team multiple endocrine neoplasia mycosis fungoides malignant fibrous histiocytoma O6 methylguanine-DNA methyltransferase monoclonal gammopathy of undetermined significance Medicines and Healthcare Products Regulatory Authority meta-iodobenzylguanidine MabThera® International Trial multileaf collimator malignant melanoma magnetic resonance Medical Research Council
MRCP MRI mRNA MS MSCC MSH MSI MSU MTC MTD MTI mTOR MTT MTU MUC-1 MUGA scan MV MW MYC NAC NAHT NCCN NCCTG NCI NCICanada NCRI NCRN Nd-YAG NEAT NET NF NHL NHS NICE NIH n-myc
NNT NOS NPVMD
magnetic resonance cholangiopancreatogram magnetic resonance imaging messenger ribonucleic acid median survival malignant spinal cord compression DNA mismatch repair gene microsatellite instability midstream urine medullary thyroid carcinoma maximally tolerated dose malignant teratoma intermediate target of rapamycin malignant teratoma trophoblastic malignant teratoma undifferentiated mucin 1 multigated acquisition scan megavoltage molecular weight v-myc myelocytomatosis viral oncogene homolog nipple areola complex neoadjuvant hormone therapy National Comprehensive Cancer Network National Central Cancer Treatment Group National Cancer Institute National Cancer Institute of Canada National Cancer Research Institute National Cancer Research Network neodynium-doped yttrium-aluminium-garnet National Epirubicin Adjuvant Trial neuroendocrine tumour neurofibromatosis non-Hodgkin lymphoma National Health Service National Institute for Health and Clinical Excellence National Institutes of Health official symbol = NMI; official name = N-myc (and STAT) interactor (NMI) number needed to treat not otherwise specified non-pulmonary visceral metastatic disease xv
Abbreviations
NS NSAA NSABP NSAID N/saline NSCLC NSGCT NST NTRAC OAR OC OFS OR OS p450 PAI-1 Pap PCI PCNSL pCR PDA PDC PDD PDGF PDGFR PDN PDR PDT PEI PERCHE PET PFS PgR PI3K PICC PLD PLDH pM PMS pN PNET xvi
not significant non-steroidal antiandrogen National Surgical Adjuvant Breast and Bowel Project non-steroidal anti-inflammatory drug normal saline non-small-cell lung cancer non-seminomatous germ cell tumour no specific type National Translational Cancer Research Network organs at risk oesophageal cancer ovarian function suppression odds ratio overall survival cytochrome p450 plasminogen activator inhibitor type 1 Papanicolau prophylactic cranial irradiation primary central nervous system lymphoma pathological complete response poorly differentiated adenocarcinoma poorly differentiated carcinoma percentage depth dose platelet-derived growth factor platelet-derived growth factor receptor poorly differentiated neoplasm pulsed dose rate photodynamic therapy percutaneous ethanol injection Premenopausal Endocrine Responsive Chemotherapy Trial positron emission tomography progression-free survival progesterone receptor phosphatidyl inositol 3 kinase peripherally inserted central catheter post-transplantation lymphoproliferative disease pegylated liposomal doxorubicin hydrochloride pathological metastasis stage PMS1 postmeiotic segregation increased 1 pathological lymph node stage primitive neuroectodermal tumour
p.o. PORT PPE PR PR-A PR-B PRV PS PSA PSTT pT PTC
PTCH PTEN
PTH PTH-RP PTV PUVA PV PVC QA QART QLQ QOL R0 RAGE RAF RAS Rb RB1 RBE RCC RCR RCT REAL RECIST RFA rhTSH
per oral postoperative radiotherapy palmar-plantar erythrodysaesthesia partial response progesterone receptor A progesterone receptor B planning risk volume WHO performance status prostate-specific antigen placental site trophoblast tumour pathological tumour stage papillary thyroid carcinoma (in thyroid cancer); percutaneous transhepatic cholangiograph (in hepatobiliary cancer) patched gene phosphatase and tensin homolog (mutated in multiple advanced cancers 1) parathyroid hormone parathyroid hormone-related peptide planning target volume psoralen plus ultraviolet A per vagina poly(vinyl chloride) quality assurance Quality Assurance in Radiation Therapy quality of life questionnaire quality of life complete resection renal antigen expression family v-raf-1 murine leukaemia viral oncogene homolog rat sarcoma viral oncogene homolog retinoblastoma retinoblastoma 1 (including osteosarcoma) radiobiologically equivalent dose renal cell carcinoma Royal College of Radiologists randomised controlled trial revised European-American lymphoma response evaluation criteria in solid tumours radiofrequency ablation recombinant human thyroid-stimulating hormone
Abbreviations
RMI RMH RMS RPLND RR RT RTOG S-100 SAB SABCS SAE SCC SCF SCLC SCTBG SEER SERM SHH SI S–I SIADH SIGN SIOP SLN SLND SM SMA SLNB SOFT SPECT src SRH SSD SSG SSM SSP SSRS STNI SUSAR SUV SV 40
relative malignancy index Royal Marsden Hospital rhabdomyosarcoma retroperitoneal lymph node dissection response rate radiotherapy Radiation Therapy Oncology Group S-100 calcium-binding protein same as before San Antonio Breast Cancer Symposium serious adverse event squamous cell carcinoma supraclavicular fossa small-cell lung cancer Scottish Cancer Trials Breast Group Surveillance Epidemiology and End Results selective oestrogen receptor modulator sonic hedgehog sacro-iliac superior–inferior syndrome of inappropriate antidiuretic hormone Scottish Intercollegiate Guidelines Network Soci´et´e Internationale d’Oncologie P´ediatrique sentinel lymph node sentinel lymph node dissection setup margin smooth muscle actin sentinel lymph node biopsy Suppression of Ovarian Function Trial single photon emission computed tomography v-src sarcoma (Schmidt-Ruppin A-2) viral oncogene homolog stigmata of recent haemorrhage source-skin distance Scandanavian Sarcoma Group superficial spreading melanoma statutory sick pay somatostatin receptor scintigraphy subtotal nodal irradiation suspected unexpected serious adverse reaction standardised uptake value simian virus 40
SVC SVCO SWOG SXR T4 TACE TAH TB TBI TCC TD t.d.s. TEM TEXT TFT Tg TGF-β Tis TKI TLD TLS TME TMR TNFα TNM TP53 TPR TRAM TRH TRUS TSC TSE TSH TTF-1 TUR TURBT TURP TVS UC U+E UICC UK UKCCSG UKCTOCS UKNET
superior vena cava superior vena cava obstruction Southwest Oncology Group superficial X-ray thyroxine transarterial chemoembolisation total abdominal hysterectomy tuberculosis total-body irradiation transitional cell carcinoma teratoma differentiated three times a day transanal endoscopic microsurgery Tamoxifen and EXemestane Trial thyroid function test thyroglobulin transforming growth factor beta carcinoma in situ tyrosine kinase inhibitor thermoluminescence dosimetry tumour lysis syndrome total mesorectal excision tissue maximum ratio tumour necrosis factor alfa tumour nodes metastases tumour protein p53 tissue phantom ratio transverse rectus abdominus myocutaneous thyrotropin-releasing hormone transrectal ultrasound trial steering committee total skin electrons thyroid-stimulating hormone thyroid transcription factor 1 transurethral resection transurethral resection of bladder tumour transurethral resection of the prostate transvaginal ultrasound ulcerative colitits urea and electrolytes International Union Against Cancer United Kingdom United Kingdom Children’s Cancer Study Group UK Collaborative Trial of Ovarian Cancer Screening United Kingdom Neuroendocrine Society xvii
Abbreviations
ULN uPA US USA US Intergroup USPIO UV UVA UVB VAIN VATS VC VEGF VEGFR VHL VIN VIP VSIM VQ WAGR WBC WBS WHO WLE XRT
upper limit of normal urokinase-type plasminogen activator ultrasound scan United States of America United States Intergroup ultra-small-particle iron oxide ultraviolet ultraviolet A ultraviolet B vaginal intraepithelial neoplasia video-assisted thoroscopic surgery vomiting centre vascular endothelial growth factor vascular endothelial growth factor receptor Von Hippel-Lindau vulval intraepithelial neoplasia vasoactive intestinal peptide virtual simulation software ventilation perfusion Wilms’ tumour, aniridia, ambiguous genitalia and mental retardation white blood cell whole-body scan World Health Organisation wide local excision X-ray treatment
BEAM BEC BEP BOP Bu/Cy CAF CarboPEC CAP CAV C-BOP CG CHOEP
CHOP CMF CMV COJEC COPP CTD
Chemotherapy regimens ABVD ABVPP AC A→CMF AC→ paclitaxel ACVBP A-T-C AVG BEACOPP
xviii
doxorubicin, bleomycin, vinblastine, dacarbazine doxorubicin, bleomycin, vinblastine, procarbazine, prednisolone doxorubicin, cyclophosphamide doxorubicin, cyclophosphamide, methotrexate, 5-FU doxorubicin, cyclophosphamide then paclitaxel doxorubicin, cyclophosphamide, vindesine, bleomycin, prednisolone doxorubicin, paclitaxel, cyclophosphamide doxorubicin, vinblastine, gemcitabine bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisolone
CVAD CVP CYVADIC DAT DHAP EBVD EC ECF E→CMF ECX EMA-CO
carmustine, etoposide, cytarabine, melphalan bleomycin, etoposide, carboplatin bleomycin, etoposide, cisplatin bleomycin, vincristine, cisplatin busulphan, cyclophosphamide cyclophosphamide, doxorubicin, 5-FU carboplatin, etoposide, cyclophosphamide cyclophosphamide, doxorubicin, cisplatin cyclophosphamide, doxorubicin, vincristine carboplatin, bleomycin, vincristine, cisplatin cisplatin gemcitabine cyclophosphamide, doxorubicin, vincristine, etoposide, prednisolone cyclophosphamide, doxorubicin, vincristine, prednisolone cyclophosphamide, methotrexate, 5-FU cisplatin, methotrexate, vinblastine cisplatin, vincristine, carboplatin, etoposide, cyclophosphamide cyclophosphamide, vincristine, procarbazine, prednisolone cyclophosphamide, thalidomide, dexamethasone cyclophosphamide, vincristine, doxorubicin, dexamethasone cyclophosphamide, vincristine, prednisolone cyclophosphamide, vincristine, doxorubicin, dacarbazine daunorubicin, ara-C, thioguanine dexamethasone, cytarabine, cisplatin epirubicin, bleomycin, vinblastine, prednisolone epirubicin, cyclophosphamide epirubicin, cisplatin, 5-FU epirubicin, cyclophosphamide, methotrexate, 5-FU epirubicin, cisplatin, capecitabine etoposide, methotrexate, actinomycin-D – cyclophosphamide, vincristine
Abbreviations
EOF EOX EP EP-EMA ESHAP FAC FAMTX FC FEC FEC→ docetaxel FMD FOLFIRI FOLFOX GemCarbo Gem-cis GT HD-MTX Hyper-CVAD
ICE JEB MACOP-B
m-BACOD
MAP MIC MOPP MOPP-ABV
MVAC MVP NP
epirubicin, oxaliplatin, 5-FU epirubicin, oxaliplatin, capecitabine etoposide, cisplatin etoposide, cisplatin – etoposide, methotrexate, actinomycin-D etoposide, methylprednisolone, cytarabine, cisplatin 5-FU, doxorubicin, cyclophosphamide 5-FU, doxorubicin, methotrexate fludarabine, cyclophosphamide 5-FU, epirubicin, cyclophosphamide 5-FU, epirubicin, cyclophosphamide then docetaxel fludarabine, mitoxantrone, dexamethasone 5-FU, FA, irinotecan 5-FU, FA, oxaliplatin gemcitabine, carboplatin gemcitabine, cisplatin gemcitabine, paclitaxel high-dose methotrexate fractionated cyclophosphamide, vincristine, doxorubicin, dexamethasone ifosfamide, carboplatin, etoposide carboplatin, etoposide, bleomycin methotrexate, doxorubicin, cyclophosphamide, vincristine, prednisone, bleomycin methotrexate, bleomycin, doxorubicin, cyclophosphamide, vincristine, dexamethasone methotrexate, doxorubicin, cisplatin mitomycin, ifosfamide, cisplatin mechlorethamine, vincristine, procarbazine, prednisolone mechlorethamine, vincristine, procarbazine, prednisolone, doxorubicin, bleomycin, vinblastine methotrexate, vinblastine, doxorubicin, cisplatin mitomycin, vinblastine, cisplatin vinorelbine, cisplatin
OEPA PIAF PLADO ProMACEcytaBOM
R-CHOP
R-DHAP R-ICE R-CODOXM/IVAC
Stanford V
SuperPLADO TAC TIP VAD VIP XELOX Z-DEX
vincristine, etoposide, prednisolone, doxorubicin cisplatin, interferon, doxorubicin, 5-FU cisplatin, doxorubicin prednisone, methotrexate, doxorubicin, cyclophosphamide, etoposide, cytarabine, bleomycin, vincristine, and methotrexate rituximab, cyclophosphamide, doxorubicin, vincristine, prednisolone rituximab, dexamethasone, cytarabine, cisplatin rituximab, ifosfamide, carboplatin, etoposide rituximab, cyclophosphamide, vincristine, doxorubicin, cytarabine, methotrexate, etoposide, ifosfamide doxorubicin, vinblastine, vincristine, bleomycin, mechlorethamine, etoposide, prednisone PLADO + carboplatin docetaxel, doxorubicin, cyclophosphamide paclitaxel, ifosfamide, cisplatin vincristine, doxorubicin, dexamethasone etoposide, ifosfamide, cisplatin capecitabine, oxaliplatin idarubicin, dexamethasone
Radioisotopes 60
Co Cs 18 F 123 I 125 I 131 I 111 In 192 Ir 103 Pd 106 Ru 99m Tc 90 Y 137
cobalt-60 caesium-137 fluorine-18 iodine-123 iodine-125 iodine-131 indium-111 iridium-192 palladium-103 ruthenium-106 technetium-99m yttrium-90
xix
1
PRACTICAL ISSUES IN CYTOTOXIC CHEMOTHERAPY USAGE Sian Evans and Philip Savage
Introduction In this chapter, the principles of cytotoxic chemotherapy treatment and the appropriate use of anticancer drugs, including some of the new targeted drugs, will be discussed. It will not be possible to give a comprehensive description of every drug and regimen, and standard chemotherapy textbooks (e.g. Allwood et al., 2002; Summerhays and Daniels, 2003) or specialist websites (e.g. BC Cancer Agency, www.bccancer.bc.ca/default.htm) should be consulted for this information. However, this chapter should provide chemotherapy prescribers and administrators with enough information to discuss treatments with patients, to prescribe chemotherapy safely and to manage the common treatment-related side effects.
breast cancer, where the rate of relapse is higher when the dose intensity is reduced (Wood et al., 1994). Generally, the regimens used in these treatments have significant side effects including neutropenia, and the use of granulocyte-colony stimulating factor (G-CSF) may be required to keep treatment on schedule. However, because there is the clear intent of achieving either cure or, for adjuvant treatment, an increased chance of cure, these side effects and treatment-related risks are seen as acceptable temporary problems. In contrast, patients having palliative chemotherapy should benefit by experiencing an improved quality of life. An increase in overall survival is not usually the primary aim of treatment and very toxic treatments are not usually justified. In this case, maintaining dose intensity is not so important and dose reductions can be made to ensure that the patient safely tolerates the treatment.
Aims of chemotherapy treatment There are three main indications for the use of chemotherapy: r The management of patients with curable advanced malignancies including choriocarcinoma, testicular cancer, Hodgkin lymphoma and high-grade nonHodgkin lymphoma (NHL). r The preoperative or postoperative adjuvant treatment of localised malignancies, primarily breast cancer and colorectal cancer. r The treatment of patients with advanced incurable malignancies, where the primary aim is palliation and symptom control, sometimes without a major expectation of prolonging survival. Before starting a course of chemotherapy, both the prescriber and the patient should be clear about the aims of treatment. When chemotherapy is used curatively, it is essential to maintain the calculated dose and dosage schedule according to the treatment protocol. The importance of this has previously been shown for testicular cancer (Toner et al., 2001), for lymphoma (Lepage et al., 1993) and in the adjuvant treatment of
Cytotoxic chemotherapy: mode of action Cytotoxic chemotherapy drugs are systemic therapies that aim to kill or slow the growth of tumour cells while being relatively sparing to normal cells. The sensitivity of different tumour types to the actions of chemotherapy drugs varies widely among the cells of origin and across the range of drugs. In curable cancers, malignant cells can be many times more sensitive to cytotoxic drugs than the cells they have arisen from and, fortunately, more sensitive than the cells of the bone marrow. In the more common malignancies, tumour cells are generally more sensitive to cytotoxic drugs than are their parent cells, but they are insufficiently sensitive to achieve a cure. Alongside this major divide between the differing types of malignancy, there is a wide range of activity of the different chemotherapy drugs across the different tumours. Whereas the majority of chemotherapy drugs have been developed empirically, the mechanisms for the greater effectiveness of some drugs in some tumours
1
Sian Evans and Philip Savage
are now becoming better understood. For example, capecitabine is metabolised to the active drug fluorouracil by thymidine phosphorylase, which is present in greater concentrations in some types of tumour cells than in normal cells.
Cell-cycle specificity Historically, the mode of action of many chemotherapy drugs has been divided between those termed ‘cell-cycle specific’ and those that are ‘cell-cycle non-specific.’ The cycle-specific drugs, such as the antimetabolites (methotrexate, fluorouracil and gemcitabine), interact predominantly with cells that are actively synthesising new DNA in the S phase. These drugs are most effective in tumours with high mitotic indices and they produce greater cell kill if given in prolonged exposures, killing larger numbers as cells move through to the S phase. In contrast, the cell-cycle non-specific drugs interact with cells in all parts of the cycle and they can be cytotoxic to more slowly proliferating tumour cells. The common cell-cycle non-specific drugs such as the alkylating agents and the antitumour antibiotics have activity at all phases of the cell cycle, and cell killing is more closely linked to the total dose rather than to the duration of administration. Although the cell-cycle distinction has been of great value in developing drug protocols and combinations, modern research suggests that this distinction is relatively crude and in fact most drugs have actions against both dividing and resting cells. However, the distinction is of some use in anticipating the side effects of chemotherapy, where the extended use of cell-cyclespecific drugs can lead to major problems with neutropenia and mucosal damage.
Chemotherapy scheduling and regimens Combination chemotherapy regimens When cytotoxic drugs are developed and licensed, it is usually on the basis of data that have been obtained in trials using single agents. After this initial approval, new drugs are generally combined with others into combination chemotherapy schedules. Single-agent therapy does remain a part of palliative chemotherapy for several tumour types; however, the best results from chemotherapy are usually with a combination of drugs, each of which is active in the tumour type as a 2
single agent. In general, the principles of choosing combinations of chemotherapy are as follow: r Each drug is active against the tumour as a single agent. r There are no clinically important drug interactions between the agents. r Combinations should avoid drugs of the same class or those with similar modes of action. r The drugs should have different dose-limiting toxicities. The bleomycin, etoposide, cisplatin regimen used for the treatment of advanced testicular cancer gives one of the best examples of the benefits derived from combining chemotherapy agents. In this regimen, bleomycin, etoposide and cisplatin have all been shown to have significant activity as single agents, but their durations of response are generally short. The three drugs have different patterns of dose-limiting toxicities. Bleomycin carries a risk of pulmonary toxicity, but it causes minimal myelosuppression; the dose of cisplatin is limited by renal toxicity, but it also causes minimal myelosuppression; and etoposide is highly myelosuppressive, but it has no significant pulmonary or renal toxicity. By combining these drugs with their differing patterns of toxicity, each can be used at nearly the full single-agent dose. The impact of this approach to chemotherapy treatment has been central in producing an overall cure rate of greater than 80% for patients with advanced testicular cancer (Williams et al., 1987). The treatment of high-grade B-cell NHL provides a similar example of effective combination chemotherapy and the benefits of adding in a modern drug with a completely different mode of action. Since its introduction in the 1970s, cyclophosphamide, doxorubicin, vincristine, and prednisolone (CHOP) chemotherapy has been the standard of care for high-grade B-cell NHL. In this regimen, two myelosuppressive drugs, cyclophosphamide and doxorubicin, are combined with vincristine and prednisolone, which are non-myelosuppressive. Over the past 20 years, there have been several important trials that have tested more toxic and complex regimens against CHOP, with negative results (Fisher et al., 1993; Gordon et al., 1992). However, the more recent addition of an anti-CD20 monoclonal antibody, rituximab, to CHOP has brought an increase in the eventfree survival at 5 years from 29% to 47% (Feugier et al., 2005). Rituximab has a completely different mode of action from that of conventional cytotoxic drugs and it rarely causes significant side effects. The combination of rituximab with CHOP chemotherapy is now standard
Practical issues in cytotoxic chemotherapy usage
practice, and it represents the first major change to highgrade NHL management in 15 years. Another monoclonal antibody, trastuzumab, is bringing about exciting results in the treatment of breast cancer. Early results suggest that the addition of trastuzumab to conventional adjuvant chemotherapy results in an approximately 50% decrease in the risk of relapse measured at two years, and there is a realistic expectation that this will translate to higher overall survival figures as the trial data mature (Piccart-Gebhart et al., 2005; Romond et al., 2005).
Protocols and guidelines The introduction of peer-reviewed treatment policies within the National Health Service (NHS) has led each cancer network and NHS Trust that treats cancer to have formal protocols of their approved chemotherapy regimens. These regimens should be familiar to the health professionals who dispense and administer them, and ‘off-protocol’ regimens should not generally be prescribed unless there is good evidence in the research literature to do so.
Scheduling and administration of chemotherapy Generally, the scheduling and administration of chemotherapy follow the protocols used in the original clinical trials. Although the precise sequence of giving drugs in many regimens is relatively unimportant, there are some situations where drugs must be given in the correct order. The most frequent situation where this applies is the combination of paclitaxel and carboplatin in the treatment of patients with ovarian cancer. Carboplatin is a cell-cycle non-specific drug; thus, it would be suited to bolus administration in a single large dose. However, because of the risk of hypersensitivity, it is given as an infusion over a minimum of 30 minutes rather than as a bolus. It is also potentially unstable, especially in poly(vinyl choride) (PVC) containers, and so the final volume of the infusion is also critical. The myelosuppression nadir from carboplatin is between 14 and 21 days, which would indicate administration using a 28day cycle. Paclitaxel is cell-cycle specific and so ideally it should be given in multiple fractions over a prolonged period. The original phase III trials, which were completed before licensing, administered the drug over 24 hours.
However, this presents logistical problems; thus, more recent studies such as the International Collaborative Ovarian Neoplasm (ICON) 3 trial have used a 3-hour infusion (ICON Group, 2002). Attempts to reduce this to 1 hour have resulted in a number of hypersensitivity reactions, and most centres continue with a 3-hour infusion. The nadir of myelosuppression from paclitaxel occurs after 10 days, indicating a maximum cycle length of 21 days. Combining the two drugs presents a dilemma about the length of the cycle, but trials have shown that a 21-day cycle does not produce unacceptable myelosuppression and it removes the theoretical possibility of tumour growth attributed to suboptimal scheduling of paclitaxel. Paclitaxel also causes hypersensitivity reactions, mainly because of the need for a solubiliser to allow the drug to dissolve. It too has stability problems and it needs administration through non-PVC lines. Finally, in routine practice, paclitaxel is administered before carboplatin, with a premedication of a corticosteroid, antihistamine and H2 antagonist. When used in this sequence, the safety profile is consistent with that reported for single-agent use. In contrast, in studies when paclitaxel was given after carboplatin, patients showed a more profound myelosuppression and an approximately 20% decrease in paclitaxel clearance.
Calculation of doses Body surface area The ideal method of calculating a suitable dose of a cytotoxic drug would take into account its pharmacokinetic properties – the ability of the body to deliver the drug to its site of action and the subsequent metabolism and excretion. The dose of the drug could then be adjusted further depending on the actual toxicity seen in each patient. However, although more precise and individualised approaches to chemotherapy drug dosing are often advocated, routine cytotoxic chemotherapy doses continue to be calculated according to the patient’s body surface area (BSA) (Veal et al., 2003). There are several formulae for calculating BSA; the most commonly used is that of DuBois and DuBois, which interestingly dates from 1916 and was based on data from only eight adults and one child (DuBois and DuBois, 1916). Other formulae using both electronic and manual methods (nomograms and slide rules) are available, and so the method used may vary among centres but there is generally good correlation among the methods. 3
Sian Evans and Philip Savage
Dose capping Using the calculated BSA in large and obese patients may lead to relative overdosing of chemotherapy with an associated risk of excess toxicity. To place an upper limit on drug dose, dose capping is frequently used, and many institutions will use 2.2 m2 as an upper limit for curative and adjuvant treatments and 2 m2 for palliative treatments. However, care should be taken when prescribing for tall, non-obese individuals because there is a potential risk of underdosing if the BSA is capped at 2.2 m2 . An exception to the 2.2 m2 cap is with the use of vincristine; with this drug, the dose is usually capped at 2 mg, so although the dose is 1.4 mg/m2 in the CHOP regimen, all but the smallest patients receive a capped dose of 2 mg.
Area under the curve (AUC) dosage Among the commonly used chemotherapy drugs, carboplatin is the only one to have its dose calculated directly according to the renal function. This drug is excreted unchanged by the kidneys, and a formula has been developed (the Calvert equation) which is based on renal function (Calvert et al., 1989). The desired AUC (area under the curve of serum levels against time) is chosen, and the dose is calculated by the following formula: Dose (mg) = desired AUC × (GFR ml/min + 25) The GFR is the glomerular filtration rate. The Calvert equation was developed using the more accurate EDTA clearance as the measure of GFR. If the GFR is calculated from serum creatinine using the Cockcroft-Gault formula (see following text), a correction, such as multiplying by 1.1, should be applied.
Body weight dosing Body weight alone is not used often in calculating doses of cytotoxic drugs. However, some of the newer drugs, such as the monoclonal antibody trastuzumab, are calculated on body weight alone.
Flat dosing Bleomycin is the only commonly used cytotoxic drug for which a fixed dose is employed. In the treatment of testicular cancer, a fixed dose of 30 000 units is used irrespective of the patient’s BSA. 4
Pretreatment investigations and checks Before initiating chemotherapy Informed consent Information on individual drugs or regimens is usually available in a written form and should be used to supplement verbal information. In addition to an explanation of the purpose of treatment and the adverse effects of the drugs, patients should be given clear advice about monitoring for, and action on, suspected neutropenic fever and other serious adverse effects of chemotherapy. This must include 24-hour contact telephone numbers at the hospital.
Cardiotoxic drugs For patients on cardiotoxic drugs such as doxorubicin or the other anthracyclines, pretreatment cardiac assessment, such as a MUGA scan, is recommended, especially if the patient has a history of cardiac disease, is elderly, or has had previous anthracycline exposure or mediastinal radiotherapy. Repeated monitoring of cardiac function should be performed according to the protocol being followed, and care should be taken to avoid exceeding the lifetime dose recommendations if more treatment is given at a later date.
Renal function An accurate measurement of renal function, such as an EDTA clearance, is generally required for carboplatin, for which the dose is calculated according to the GFR. Accurate assessment is also recommended for treatments using cisplatin, because this drug is significantly nephrotoxic. If there is evidence of a significant decline in renal function, the cisplatin in the regimen can generally be amended by substituting carboplatin, usually at an AUC of 4 or 5. For other drugs and for continuing assessment of renal function, the creatinine clearance (CrCl) can be calculated from the serum creatinine using the Cockcroft–Gault formula as follows: F × (140 − age in years) × weight (kg) Serum creatinine(μM/l) F = 1.04 (females) or 1.23 (males)
CrCl =
Doses will need to be amended for drugs that are excreted renally if there is a reduced GFR. Information on the appropriate dose reduction in renal impairment will be found in most protocols or national guidelines, such as those available online from the North London Cancer Network (www.nlcn.nhs.uk/professional. php).
Practical issues in cytotoxic chemotherapy usage
Hepatic function Most drugs undergo some metabolism by the liver. The capacity of the liver to handle drugs, even when there is hepatic impairment, is large, and the need for dose reductions is relatively uncommon. However, some drugs, including doxorubicin, do need dose reductions in the presence of hepatic impairment, and the liver function tests, bilirubin, transaminases and alkaline phosphatase should be reviewed before treatment. Increases in the alkaline phosphatase, alone or accompanied by slight increases in transaminases, do not usually require dose reductions, but elevation of bilirubin, particularly if accompanied by increases in transaminases, usually requires the dose reduction of drugs that are metabolised in the liver. Irinotecan, which is excreted in the bile, has to be dose-reduced in the presence of elevated serum bilirubin. The treatment protocols in most units include advice on appropriate dose reductions; other sources of advice include the websites of the BC Cancer Agency and the North London Cancer Network.
Baseline assessments of tumour With the exception of adjuvant treatment, a baseline measurement and regular objective measurement of response are required to assess whether the patient is benefiting from chemotherapy. This can involve direct physical measurement of the tumour, radiological examination, biochemical tests or measurement of tumour markers, depending on the disease site. Where the principal aim is palliative, one should also monitor symptomatic benefit carefully and balance this against the treatment toxicity.
Central lines Patients who have chemotherapy through ambulatory infusion devices must have central access before treatment, using a Hickman® or PICC line. Patients with poor veins or those who are to receive multiday infusions will also benefit from central lines early on in their treatment, if possible. Although many patients do not experience any problems with these lines, studies have shown that up to 11% develop line-related thromboses, and 19%, line infections. Patients should therefore be monitored regularly for these problems (Minassian et al., 2000).
Height and weight To calculate BSA, height and weight measurements are needed. The patient’s body weight should be measured
before each course of chemotherapy, and again if there is reason to suspect that it has altered by more than 5%. Where the dose is calculated on the body weight alone, smaller changes in weight have a greater impact on the dose and, in this situation, the patient’s weight should be checked regularly; for example, every three months, or if the body weight is thought to have changed by more than 5%.
Before each cycle Full blood count The patient’s full blood count should be taken close to the actual day of administration, ideally on the day of treatment or the day before. Significant neutropenia or thrombocytopenia will mean a treatment delay. Patients who are anaemic rarely require a delay in chemotherapy and can be transfused if their haemoglobin level drops below 9.5 g/dl or if they develop symptoms of anaemia. Patients who are admitted with neutropenic fever or who have had more than one delay in treatment during a course of chemotherapy will require a dose reduction if receiving palliative treatment or support with growth factors if they are receiving curative or adjuvant treatment.
Biochemical, renal, liver and bone profile A full biochemical profile is required before treatment to ensure that there has been no significant change in renal or hepatic function due to either the treatment or the tumour. In response to deteriorating renal or hepatic function, some drugs may need dose reduction or a change to alternative therapies. Some patients who presented with disease-related hepatic or renal impairment may no longer need a dose reduction if these parameters return to normal with effective treatment. Some drugs have specific toxic effects that require monitoring. For example, cisplatin increases the renal excretion of potassium and magnesium and oral supplementation is required frequently. The typical practice in the United Kingdom (UK) is to add 20–40 mM of potassium and about 8 mM of magnesium per day to hydration fluids.
Tumour markers and other tests for response If a tumour secretes circulating tumour markers, measuring these can be a rapid, simple and economic method to monitor the response of the tumour to treatment. Tumour markers are most useful in gestational 5
Sian Evans and Philip Savage
choriocarcinoma, where human chorionic gonadotrophin (HCG) is constitutively produced by all tumours, and they provide an excellent method of following response to treatment. In advanced testicular cancers, approximately 60% of tumours make one or both of HCG and alpha feto-protein. These can be used to monitor response or to indicate a change to second-line therapy when the rate of fall is inappropriately slow (Toner et al., 1990). Other tumour markers are also used in the assessment of some other malignancies, such as cancer antigen 125 in ovarian cancer and prostate-specific antigen in metastatic prostate cancer.
Major toxicities and their management This section looks at practical issues of managing chemotherapy toxicities. The main toxicities of individual chemotherapy drugs are summarised in the Appendix (see p. 11).
Myelosuppression Neutropenic fever and neutropenic sepsis The most frequent serious complications of chemotherapy treatment are neutropenic fever and neutropenic sepsis. Although the nadir of neutropenia varies among drugs (e.g. patients on docetaxel can become neutropenic two to three days after treatment, and the nadir for mitomycin C is several weeks after treatment), the risk of neutropenia should be considered at all times during the course of chemotherapy treatment. Patients with neutropenic fever or neutropenic sepsis must be treated promptly using the local policies for the empirical treatment. The treatment plan with intravenous antibiotics will usually include an aminoglycoside and an antipseudomonal antibiotic or a cephalosporin and vancomycin. Patients with an apparent site of infection, such as chest infection, urinary tract infection or central-line sepsis, should also receive an antibiotic with appropriate additional cover. Most patients with neutropenic fever will initially appear well, apart from the pyrexia, and they will respond to treatment with intravenous or oral antibiotics. However, a number of patients can either present or rapidly become seriously unwell with hypotension, shock and end-organ failure. In these patients, rapid assessment and management are essential. Volume replacement with intravenous fluids must be initiated, and consideration should be given to transfer to a high6
dependency or intensive-care unit. All patients with neutropenic sepsis can deteriorate very quickly, and junior medical staff must be made aware of the need to contact more senior staff, if necessary, for advice on the management of such patients. Neutropenic fever and sepsis are also considered in Chapter 6 (see p. 77).
Primary and secondary prophylaxis Patients receiving palliative chemotherapy who are admitted with neutropenic fever, or those who have persistently low neutrophil counts without fever, should generally have dose reductions made to their treatment. In contrast, patients receiving curative (including adjuvant) treatment should receive G-CSF prophylaxis, as secondary prophylaxis, to prevent further episodes of neutropenia and to maintain dose intensity. More recently, there has been a move towards primary prophylaxis (i.e. from cycle 1) for patients on very myelosuppressive regimens, and the current ASH/ASCO guidelines recommend this approach for regimens with a greater-than-20% risk of neutropenic sepsis in the first cycle (Smith et al., 2006). However, this is not yet standard practice in the UK. Practical issues with the timing and administration of growth factors can be a problem, but this has been overcome with the introduction of pegylated G-CSF (Neulasta® ), which requires only a single administration, generally 24 hours after chemotherapy administration. Studies have demonstrated Neulasta® to have been more successful in preventing neutropenic sepsis than daily G-CSF and to have improved patient tolerability. Although Neulasta® is more expensive than daily G-CSF, it may be cost effective when compared to the costs of a hospital admission for neutropenic fever.
Prophylactic antibiotics The Significant trial showed a small, but definite, reduction in the incidence of admissions for neutropenic fever for patients with solid tumours who were given a prophylactic quinolone antibiotic (Cullen et al., 2005). The data from this study, and the increasing evidence that prophylactic growth factors are clinically and economically effective, are leading many centres to incorporate their use into routine protocols. In regimens that cause prolonged myelosuppression, particularly lymphoma regimens with long-term steroid administration, patients will benefit from the use of cotrimoxazole to reduce the risk of pneumocystis carinii infection.
Practical issues in cytotoxic chemotherapy usage
Prophylactic antibiotics are also considered in Chapter 6 (see p. 79).
Anaemia Some commonly used cytotoxic drugs, including cisplatin, cause a gradual reduction in haemoglobin levels over a course of treatment. This does not usually need a dose reduction or delay in treatment, but it can dramatically affect a patient’s quality of life. A blood transfusion is standard practice. Erythropoietin may be indicated, but it is expensive and under review by the National Institute for Health and Clinical Excellence (www.nice.org.uk). Erythropoietin is also considered in Chapter 2 (see p. 20).
wise fashion (Herrstedt et al., 2005). All chemotherapy units will have guidance on which level of antiemetics to use with each chemotherapy drug or regimen. Oral metoclopramide or domperidone are usually recommended for drugs of low emetogenic potential, such as bleomycin, vindesine or gemcitabine. However, the majority of drugs and regimens require more powerful antiemetics, generally a 5-HT3 antagonist and dexamethasone, on the day of treatment and for one to two days afterwards. For patients who continue to have problems, newer drugs, including aprepitant, a neurokinin-1 receptor antagonist, are recommended in addition to 5-HT3 antagonists and corticosteroids.
Delayed nausea and vomiting
Nausea and vomiting The problems of chemotherapy-associated nausea and vomiting have become far less since the introduction of the 5-HT3 antagonist drugs such as ondansetron and granisetron. Many new patients still expect nausea and vomiting to be a major problem, but they can be reassured that, with appropriate use of antiemetics, this is now rare.
The incidence of delayed nausea and vomiting is increased when there is poor control of the acute phase. The 5-HT3 antagonists are not generally effective in treating delayed nausea and vomiting. Cyclizine may be substituted for metoclopramide for the treatment of delayed nausea and vomiting or during prolonged oral regimens if metoclopramide is ineffective. Good antiemetic prophylaxis, and the use of aprepitant or palonosetron with subsequent cycles, may also prevent delayed nausea and vomiting.
Predisposing factors The predisposing factors for an increased risk of nausea and vomiting include the previous poor control of nausea/vomiting, a history of motion sickness, being of a younger age, being female and having a chronic, low alcohol intake (Gralla et al., 1999). These factors may require increased prophylaxis above those recommended here.
Anticipatory nausea and vomiting Anticipatory nausea and vomiting occur up to and during administration of chemotherapy, and are mainly due to the psychological effects associated with previous treatment. The management of this problem involves considering a benzodiazepine before treatment and/or on the previous evening. Lorazepam 0.5 to 1 mg sublingually or orally is the drug of choice, and it is well tolerated.
Acute nausea and vomiting Acute nausea and vomiting are defined as occurring up to 24 hours after chemotherapy administration. The drugs used in prevention depend on the emetogenic potential of the regimen, and they are used in a step-
Other dose-limiting toxicities Cardiotoxicity Cumulative cardiac toxicity is a problem associated with the anthracyclines, and treatment should remain generally within the standard guidelines on total lifetime dose. In the case of doxorubicin, this is 450 mg/m2 . Cardiac function should be monitored more closely in patients with cardiac problems or in patients who have received previous treatment with anthracyclines or mediastinal radiotherapy. A number of other drugs can also occasionally cause cardiotoxicity; fluorouracil is a good example that is linked with cardiac ischaemia and arrhythmias.
Renal toxicity Although renal function is monitored before each chemotherapy cycle in most regimens, particular attention is needed for drugs that are either renally toxic in themselves or are excreted by the kidneys. Cisplatin is the drug that is most frequently linked to renal toxicity. Appropriate treatment modifications should be made if there is a significant rise in the serum creatinine while on 7
Sian Evans and Philip Savage
therapy, pending a more accurate assessment of renal function such as an EDTA clearance or 24-hour CrCl. High-dose methotrexate can also cause renal toxicity, whereas cyclophosphamide and ifosfamide can cause both renal toxicity and haemorrhagic cystitis.
Diarrhoea Diarrhoea can be a dose-limiting toxicity with the fluoropyrimidines such as capecitabine and 5-FU. Diarrhoea can also occur as part of the cholinergic syndrome, which is seen with irinotecan. It is important to recognise the symptoms early and to start treatment with fluids, rehydration salts and loperamide.
Palmar–plantar erythrodysaesthesia (PPE) PPE is the dose-limiting toxicity with capecitabine and liposomal doxorubicin. Patients should be encouraged to use emollients, but effective prevention is difficult. A dose delay is usually required for grade 2 or greater PPE. Pyridoxine 50 mg three times a day is often used as treatment, but it is not clear how effective it is. Avoidance of heat can help to prevent PPE.
Other toxicities Alopecia Alopecia can be a very distressing side effect of chemotherapy treatment for some patients. The problem can be minimised to some extent by the use of scalp hypothermia in patients receiving bolus injections or short infusions of doxorubicin, epirubicin and docetaxel. However, many chemotherapy patients do develop significant alopecia and they need to be aware of this possibility. Arrangements should be offered for wigs and other cosmetic supports such as head scarves.
Fertility and foetal abnormalities (contraception) Chemotherapy can affect the patient’s fertility. The regimens used in the treatment of testicular cancer, Hodgkin lymphoma, and high-grade NHL tend to have a relatively modest impact on fertility, but it is routine good practice to offer sperm storage for men undergoing chemotherapy. The situation for women is less satisfactory because techniques for the preservation of oocytes or ovarian tissue are not yet reliable or widely used. Embryo storage is time-consuming and may cause a significant delay in starting treatment. Surprisingly, the incidence of foetal abnormalities born to patients who have previously completed 8
chemotherapy appears to be similar to that in the normal population. Patients are advised to defer pregnancy for 12 months after the completion of treatment but there is little evidence to say whether or not this is too cautious. In addition to the potential risk of foetal abnormalities, the risk of relapse needs to be taken into account when giving patients advice about the timing of future pregnancies. More detailed information on cancer treatment, chemotherapy and fertility is readily available (Lee et al., 2006).
Phlebitis and local reactions Phlebitis is a common problem with irritant drugs such as dacarbazine, the alkylating agents and vinca alkaloids. These drugs should always be given as bolus injections via fast-flowing drips or through a central line.
Administration issues Safe administration of chemotherapy In the UK, there are formal standards for the safe prescribing, dispensing and administration of chemotherapy. The standards, and methods of auditing these, vary in detail among the home countries, but every organisation that provides chemotherapy has the responsibility to maintain policies and procedures to ensure that these standards are met. Training, competency and adequate facilities are the keys to safely prescribing, dispensing, and administering chemotherapy. All healthcare staff involved in chemotherapy must be aware of their local policies.
Intrathecal chemotherapy There have been a number of fatalities from the inadvertent administration of vinca alkaloids by the intrathecal route. Intrathecal chemotherapy must be prescribed and administered only by trained specialist registrars, staff grades and consultants. All prescribers must be aware of the local arrangements for administering chemotherapy and they should refuse to undertake any procedure for which they have not been trained.
Extravasation A number of cytotoxic drugs are vesicant and, in the event of extravasation, they can cause local tissue necrosis. The patient may report pain on injection, but there may not be any obvious local reaction. There are a
Practical issues in cytotoxic chemotherapy usage
number of general and individual drug-specific measures for the treatment of suspected extravasation and the problem should be dealt with as an emergency (see Chapter 6, p. 80).
Hypersensitivity and anaphylaxis Certain cytotoxic drugs, and most monoclonal antibodies, can produce hypersensitivity reactions that can range from a ‘flu-like’ syndrome to anaphylactic shock. Platinums and taxanes are the cytotoxic drugs that are most commonly implicated. Premedication with corticosteroids, antihistamines and paracetamol may be recommended, depending on the expected reaction. Full resuscitation facilities must be available for patients receiving first doses of agents that are known to cause hypersensitivity. Generally, this involves administration in a hospital setting at a time when staffing levels and skill mix are appropriate to deal with emergencies. Subsequent doses may be given at other locations depending on the risk of future reactions. The detailed treatment of anaphylaxis will depend on local policy but will include epinephrine (adrenaline), and oxygen must be available (see Chapter 6, p. 79).
Oral chemotherapy and overcompliance With oral chemotherapy, overcompliance is often a problem when capecitabine is involved: patients may ignore the onset of dose-limiting adverse effects, such as diarrhoea, and then continue treatment and become dehydrated. Careful patient education is essential and treatment must not be started until the prescriber is sure that the patient fully understands how to take the medication, the use of supportive medication, the circumstances when treatment should be discontinued, and how to obtain help.
Summary
Cholinergic syndrome is seen with irinotecan, and it is characterised by flushing, sweating and diarrhoea. Treatment is with atropine, and premedication with this is recommended for subsequent cycles. Late-onset diarrhoea (more than 24 hours postinfusion) is also seen, and it is treated with loperamide, rehydration, and possibly antibiotics. Patients must be educated to recognise these symptoms and to take appropriate action.
The development of effective chemotherapy treatments has been one of the great successes of modern medicine. As a result, most patients with high-grade NHL, Hodgkin lymphoma or testicular cancer are treated with the expectation of cure. Alongside this, the adjuvant treatment of breast and bowel cancer has increased the overall cure rates of these illnesses by at least 10%. Although these drugs are highly toxic, their use in the modern multidisciplinary cancer centres, following established protocols and patterns of administration, results in relatively little unpredictable toxicity. Most additional information on the safe and effective use of the current drugs is readily available within each unit or on the expert websites. At present, the introduction of new targeted anticancer drugs such as the monoclonal antibodies and kinase inhibitors will be the main area of change in cancer treatment. It is likely that these drugs will be increasingly combined with existing therapies not only to produce enhanced results but also to produce differing patterns of side effects that will require ongoing education and training.
Oesophageal–pharyngeal syndrome
REFERENCES
Cholinergic syndrome
This is seen with oxaliplatin and, although rare, can be very distressing for the patient, who may confuse this symptom with respiratory or cardiac arrest. Avoidance of cold liquids and food and not exposing the body to sudden cold will usually prevent this syndrome, and patients can be reassured if they are warned in advance that this symptom is transient. Treatment is symptomatic; warm drinks often help in mild cases. Prolonging infusion times has been used successfully to reduce the rates of recurrence of oesophageal– pharyngeal syndrome.
Allwood, M., Stanley, A. and Wright, P. (2002). The Cytotoxics Handbook, 4th edn. Oxford: Radcliffe Medical Press. Calvert, A. H., Newell, D. R., Gumbrell, L. A. et al. (1989). Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J. Clin. Oncol., 7, 1748–56. Cullen, M., Steven, N., Billingham, L. et al. (2005). Antibacterial prophylaxis after chemotherapy for solid tumors and lymphomas. N. Engl. J. Med., 353, 988–98. DuBois, D. and DuBois, E. F. (1916). A formula to estimate the approximate surface area if height and weight be known. Arch. Intern. Med., 17, 863–71. Feugier, P., Van Hoof, A., Sebban, C. et al. (2005). Long-term results of the R-CHOP Study in the treatment of elderly patients with
9
Sian Evans and Philip Savage
diffuse large B-cell lymphoma: a study by the Groupe d’Etude des Lymphomes de l’Adulte. J. Clin. Oncol., 23, 4117–26. Fisher, R. I., Gaynor, E. R., Dahlberg, S. et al. (1993). Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin’s lymphoma. N. Engl. J. Med., 328, 1002–6. Gordon, L. I., Harrington, D., Andersen, J. et al. (1992). Comparison of a second-generation combination chemotherapeutic regimen (m-BACOD) with a standard regimen (CHOP) for advanced diffuse non-Hodgkin’s lymphoma. N. Engl. J. Med., 327, 1342–9. Gralla, R. J., Osoba, D., Kris, M. G. et al. (1999). Recommendations for the use of antiemetics: evidence-based, clinical practice guidelines. American Society of Clinical Oncology. J. Clin. Oncol., 17, 2971–94. Herrstedt, J., Aapro, M. S., Roila, F. et al. (2005). ESMO minimum clinical recommendations for prophylaxis of chemotherapy-induced nausea and vomiting (NV). Ann. Oncol., 16 (Suppl. 1), i77–9. ICON Group. (2002). Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide, doxorubicin, and cisplatin in women with ovarian cancer: the ICON3 randomised trial. Lancet, 360, 505–15. Lee, S. J., Schover, L. R., Partridge, A. H. et al. (2006). American Society of Clinical Oncology recommendations on fertility preservation in cancer patients. J. Clin. Oncol., 24, 2917–31. Lepage, E., Gisselbrecht, C., Haioun, C. et al. (1993). Prognostic significance of received relative dose intensity in non-Hodgkin’s lymphoma patients: application to LNH-87 protocol. The GELA (Groupe d’Etude des Lymphomes de l’Adulte). Ann. Oncol., 4, 651–6. Minassian, V. A., Sood, A. K., Lowe, P. et al. (2000). Longterm central venous access in gynecologic cancer patients. J. Am. Coll. Surg., 191, 403–9. Piccart-Gebhart, M. J., Procter, M., Leyland-Jones, B. et al. (2005). Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N. Engl. J. Med., 353, 1659–72.
10
Romond, E. H., Perez, E. A., Bryant, J. et al. (2005). Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N. Engl. J. Med., 353, 1673–84. Smith, T. J., Khatcheressian, J., Lyman, G. H. et al. (2006). 2006 update of recommendations for the use of white blood cell growth factors: an evidence-based clinical practice guideline. J. Clin. Oncol., 24, 3187–205. Summerhays, M. and Daniels, S. (2003). Practical Chemotherapy – A Multidisciplinary Guide. Oxford: Radcliffe Medical Press. Toner, G. C., Geller, N. L., Tan, C. et al. (1990). Serum tumor marker half-life during chemotherapy allows early prediction of complete response and survival in nonseminomatous germ cell tumors. Cancer Res., 50, 5904–10. Toner, G. C., Stockler, M. R., Boyer, M. J. et al. (2001). Comparison of two standard chemotherapy regimens for good-prognosis germ-cell tumours: a randomised trial. Australian and New Zealand Germ Cell Trial Group. Lancet, 357, 739–45. Veal, G. J., Coulthard, S. A. and Boddy, A. V. (2003). Chemotherapy individualization. Invest. New Drugs, 21, 149–56. Williams, S. D., Birch, R., Einhorn, L. H. et al. (1987). Treatment of disseminated germ-cell tumors with cisplatin, bleomycin, and either vinblastine or etoposide. N. Engl. J. Med., 316, 1435–40. Wood, W. C., Budman, D. R., Korzun, A. H. et al. (1994). Dose and dose intensity of adjuvant chemotherapy for stage II, node-positive breast carcinoma. N. Engl. J. Med., 330, 1253–9.
FURTHER READING Allwood, M., Stanley, A. and Wright, P. (2002). The Cytotoxics Handbook, 4th edn. Oxford: Radcliffe Medical Press. Solimando, D. A. (2005). Drug Information Handbook for Oncology, 5th edn. Hudson, OH: Lexi-Comp. Summerhays, M. and Daniels, S. (2003). Practical Chemotherapy – A Multidisciplinary Guide. Oxford: Radcliffe Medical Press.
Practical issues in cytotoxic chemotherapy usage
Appendix. Cancer treatment drugs and their major toxicities Emetogenic Drug/class
Myelosuppression
potential
Other major toxicities
Notes
Cyclophosphamide
++
Ifosfamide
++
++
Haemorrhagic cystitis.
May need mesna.
++
Haemorrhagic cystitis.
Needs mesna routinely to
Alkylating agents
Encephalopathy. Lomustine
+++ Delayed
++
myelosuppression.
Pulmonary.
prevent haemorrhagic cystitis. Capsules 40 mg.
Renal.
Anthracyclines Doxorubicin
+++
++
Cardiotoxic. Alopecia.
Vesicant, give by bolus injection or centrally.
Epirubicin
+++
++
As above.
As above.
Idarubicin
+++
++
As above.
As above. Oral form available.
Depends on schedule. Liposomal
+++
+
PPE. Cardiotoxic.
Exfoliant.
+++ Delayed
+
Renal.
Vesicant.
doxorubicin Mitomycin
myelosuppression. Mitozantrone
+++
+
Cardiotoxic.
Exfoliant.
Dacarbazine
+++
+
Hepatotoxic.
Vesicant.
+
+
PPE.
Tablets 500 mg and 150 mg.
Antimetabolites Capecitabine
Diarrhoea. Fludarabine
+++
+
If blood transfusion required by patient, need irradiated blood. Consider PCP prophylaxis.
Fluorouracil
+
+
Diarrhoea.
Gemcitabine
+
+
Influenza-like reactions
Methotrexate
+
+
Mucositis. Renal toxicity in Note drug interactions. Avoid in
PPE with long infusions.
after infusion. high dose.
patients with pleural effusions and/or ascites.
Vinca alkaloids and etoposide Vincristine
Minimal.
Minimal.
Neurotoxic. +++
Must be given by intravenous injection only. Vesicant. Max. dose 2 mg.
Vinblastine
+
Minimal.
Neurotoxic. ++
Must be given by intravenous injection only. Vesicant. Max. dose usually 10 mg. (continued )
11
Sian Evans and Philip Savage
Appendix. (Continued ) Emetogenic Drug/class
Myelosuppression
potential
Other major toxicities
Notes
Vinorelbine
+
Minimal.
+ Neurotoxic.
Must be given by intravenous injection only. Vesicant. Max. i.v. dose 60 mg. Oral form available.
Etoposide
++
+
Alopecia.
Oral form (large capsules) twice i.v. dose.
Platinums Cisplatin
+
+++
Nephrotoxic. Neurotoxic.
Exfoliant. Check renal function carefully; good pre- and posthydration essential.
Carboplatin
++ Especially
Oxaliplatin
++
++
Less nephrotoxic than
++
Neurotoxic.
platelets.
cisplatin.
Hypersensitivity reactions common. Exfoliant. Hypersensitivity reactions.
Taxanes Paclitaxel
++
+
Alopecia. Neurotoxic.
Vesicant. Hypersensitivity reactions – needs premedication.
Docetaxel
+++
+
Alopecia.
Vesicant. Hypersensitivity reactions – needs premedication.
Topoisomerase 1 inhibitors Irinotecan
++
++
Diarrhoea. Cholinergic
May need atropine.
syndrome. Topotecan
+++
+
Alopecia.
Exfoliant.
Other cytotoxics Bleomycin
No.
Minimal.
Pulmonary (fibrosis). Skin.
Dacarbazine
++
+++
Some alopecia.
Procarbazine
+
++
Vesicant. Weak MAOI; avoid alcohol. 50 mg capsules.
MAOI = monoamine oxidase inhibitor; PCP = pneumocystis carinii pneumonia; PPE = palmar–plantar erythrodysaesthesia.
12
2
BIOLOGICAL TREATMENTS IN CANCER Rachel Jones and Robert Leonard
Introduction Biological agents are becoming increasingly important in the management of cancer patients. For patients with some tumours – such as renal cancer, where chemotherapy is not very effective – biological agents are the main treatment after surgery. In other solid tumours, although chemotherapy has improved survival rates it has been at the expense of more toxicity. Increasing evidence supports the use of biological agents both as single agents and in combination with chemotherapy. Over the next decade, it is likely that biological therapies will significantly change the management of cancer patients. Because this is a rapidly changing field, the licensed indications for new drugs is also changing; a useful resource for up-to-date information on licensed drugs can be found in the electronic Medicines Compendium (www.medicines.org.uk/). This chapter will concentrate on five areas: r Cytokines. r Tyrosine kinase inhibitors. r Monoclonal antibodies. r Haemopoietic colony stimulating factors. r Vaccines.
Cytokines Cytokines are proteins that are secreted by haemopoietic and non-haemopoietic cells and which play an important role in the immune system. They work by binding to specific effector-cell surface receptors, thereby initiating a number of intracellular signalling cascades that change the functioning of the effector cell. This group of molecules is complex and increasing in number, and there are several different classification systems that have been used, based on the molecules’ presumed function, their cell of secretion, their target of action or their structure. There are five recognised major cytokine families:
r Type I cytokines (interleukins). r Type II cytokines (interferons). r The tumour necrosis factor family. r The immunoglobulin supergene family. r The chemokines. Although there are over a hundred cytokines, only interleukin-2 and interferon alpha have an established role in the management of cancer patients.
Interleukin-2 (IL-2) Background
r IL-2 was first identified by its action as a T-cell growth factor in 1976.
r It is released by T-helper cells in response to antigen presentation and interleukin-1, and it is involved in activation of both T-helper cells and cytotoxic T-cells. r In vitro incubation of lymphoid cells with recombinant IL-2 leads to the generation of lymphokineactivated killer (LAK) cells, which are able to lyse fresh tumour suspensions.
Renal cell cancer – metastatic disease In phase II trials of high-dose IL-2 there was a response rate of 14% (CR 5%, PR 9%; Fyfe et al., 1995). The median duration of response was 19 months for patients who achieved a partial response, but 4% of patients died of probable treatment-related toxicity. A systematic review of various IL-2 schedules showed that bolus i.v. and s.c. protocols had higher complete-response rates than continuous infusion, although it was difficult to draw firm conclusions because of the heterogeneity of the data. The duration of response was highest for i.v. bolus, and half the responding patients had responses that lasted for 36 months or more (Baaten et al., 2004). Currently, IL-2 is licensed for use in metastatic renal cell carcinoma. It should not be given to patients who are unlikely to respond, such as those with ECOG performance status of 2 or greater or those with ECOG performance status 1, with more than one site of metastatic
13
Rachel Jones and Robert Leonard
disease, who have relapsed within 12 months of diagnosis.
Other situations For patients with high-risk renal cell cancer, no effective adjuvant treatment is presently available. A randomised trial involving patients with resected locally advanced or metastatic renal cell carcinoma on one course of IL-2 or observation was closed early because an interim analysis found the disease-free survival end point could not be achieved, despite full accrual (Clark et al., 2003). In metastatic melanoma, a review of eight clinical trials, which included 270 patients treated with high-dose recombinant IL-2, showed an overall response rate of 16% (CR 6%, PR 10%; Atkins et al., 2000). The median duration of complete responses was at least 59 months. However, the toxicity associated with this regimen was severe and 2% died from sepsis.
Toxicity of IL-2 IL-2 is a highly toxic drug, and it is associated with capillary leak syndrome, hypotension, cardiac arrhythmias, pulmonary oedema, fever and death. In view of its significant toxicity, patients need to be selected carefully and only treated in centres with experienced clinicians and facilities adequate enough to provide haemodynamic support.
Interferons Interferons have antiviral, antiproliferative and immunomodulatory effects. There are three types: alpha, beta and gamma. Interferons alpha and beta are type 1 interferons, which bind to a cell-surface receptor complex called IFNAR, and interferon gamma is a type 2 interferon, which binds to the cell-surface receptor complex IFNGR. Of the three types, only interferon alpha has an accepted role in cancer management.
Interferon alpha (IFNα) Naturally occurring IFNα is produced by many cells, including T and B lymphocytes, fibroblasts, endothelial cells and osteoblasts. It acts by stimulating macrophages, natural killer cells and cytotoxic T lymphocytes.
Renal cell carcinoma – metastatic disease IFNα has a response rate of 14% in metastatic renal cell carcinoma, as shown by a review of more than 14
1000 patients (Wirth, 1993). The duration of remission was rarely over two years. Previous nephrectomy and Karnofsky performance status greater than or equal to 80 may be associated with better survival. Five risk factors are associated with short survival: low Karnofsky performance status, high LDH, low haemoglobin, high corrected serum calcium and an interval from diagnosis to initiation of therapy of less than 1 year (Motzer et al., 2002). In view of the toxicity associated with this treatment, these factors may help select patients most likely to benefit from IFNα treatment. In recurrent or metastatic renal cell cancer, IFNα-2a is licensed for use at a starting dose of 3 MU three times a week, building up to 18 MU three times a week if tolerated.
Melanoma – adjuvant setting The use of IFNα in the adjuvant management of melanoma is controversial because of conflicting survival results in trials that look at both high- and low-dose treatment. Because of the toxicity and duration of treatment, the need for dose modifications was frequently reported. Currently, high-dose IFNα has become a standard adjuvant treatment in the USA, but with no consistent overall survival data and the associated toxicity, it has not become a standard of care in the UK. Lowdose IFNα, however, is licensed for use in the UK as an adjuvant treatment for melanoma.
Melanoma – metastatic disease IFNα has been widely tested in the metastatic setting in melanoma. A review of three phase II trials showed a response rate of 22%, with a median time to disease progression of 3.9 months and overall survival of 11.3 months. The best responses were noted in soft tissue and small-volume cutaneous disease (Creagan et al., 1986).
Haematological malignancies IFNα has an established role in some haematological malignancies, particularly in hairy cell leukaemia and chronic myelogenous leukaemia.
Others In metastatic carcinoid, IFNα has been used with somatostatin analogues to improve the symptoms associated with hormone secretion in patients who have become resistant or refractory to somatostatin analogues alone. However, uncertainty exists regarding its effectiveness, and its use has not become widespread (see also Chapter 36, p. 423).
Biological treatments in cancer
Toxicity of IFNα The main toxicities associated with treatment are flulike symptoms, fatigue, depression, neutropenia and reversible hepatotoxicity. Patients may be unable to tolerate long-term treatment. Side effects associated with long-term therapy include hypothyroidism.
Cytokine combination therapy The use of combination biological therapies is the subject of current clinical trials. The recently closed MRC RE04 trial randomised patients with metastatic renal cell carcinoma between IFNα monotherapy and the combination of IFNα, IL-2, and 5-fluorouracil because of the promising results seen with the combination regimen (Atzpodien et al., 1993).
Tumour necrosis factor α (TNFα) TNFα has been studied in combination with melphalan in unresectable melanoma of the extremities. There does not appear to be a clear survival benefit from the addition of TNFα over melphalan alone. TNFα has minimal activity in advanced cancer and is highly toxic.
Tyrosine kinase inhibitors Tyrosine kinases are enzymes that can transfer phosphate groups from ATP to a tyrosine residue on a protein. They act on intracellular signalling pathways that are involved with promoting tumour growth, invasion, angiogenesis and resistance to apoptosis. The development of tyrosine kinase inhibitors has been an important breakthrough in cancer therapy. They are oral preparations, frequently with different, nonoverlapping toxicities from chemotherapy, which allow them to be combined with chemotherapy and radiotherapy or to be given as single agents. Tyrosine kinases can be divided into receptor tyrosine kinases and nonreceptor tyrosine kinases. In receptor tyrosine kinases, the tyrosine kinase itself forms an integral part of the receptor molecule, which spans the membrane. The binding of a ligand to the receptor can activate signalling pathways, such as the RAS-RAF MAP kinase pathway and the PI3K-AKT pathway, which lead to a change in activity of transcription factors and to DNA synthesis. The epidermal growth factor receptor (EGFR) is a receptor tyrosine kinase.
Non-receptor tyrosine kinases transmit intracellular signals via recognition of phosphorylated tyrosines by SH domains on other binding proteins. The BCRABL fusion protein, which is found in chronic myeloid leukaemia, is an example of how abnormal tyrosine kinase activity can cause malignancy. The gene for BCR on chromosome 22 fuses with the gene for the ABL protein (a tyrosine kinase) on chromosome 9. This translocation results in an abnormal ‘Philadelphia’ chromosome. Synthesis of the resultant fusion protein BCRABL is unregulated and the ABL tyrosine kinase activity, through further binding to growth factor-related binding protein, results in RAS activation. The action mechanism of tyrosine kinase inhibitors is to prevent autophosphorylation by competing with ATP to prevent its binding to the intracellular tyrosine kinase region. Table 2.1 shows some tyrosine kinase inhibitors in clinical use and in trials.
Imatinib Imatinib is an oral tyrosine kinase inhibitor which acts on more than one type of tyrosine kinase. It treats Philadelphia-positive CML by inhibiting the tyrosine kinase activity of the BCR-ABL fusion protein. In untreated, early chronic-phase Philadelphia-positive CML, major cytogenetic responses are seen in 80 to 90% of patients (Kantarjian et al., 2002). The recommended dose is 400 mg daily in chronic-phase CML and 600 mg daily in accelerated-phase or blast crisis. Imatinib has also been shown to be active against the KIT and PDGFR tyrosine kinases. Gastrointestinal stromal tumours are associated with a mutation in KIT which results in ligand-independent activation of the receptor. Previously these patients had limited options for treatment, but now several trials have shown consistent clinical responses (Demetri et al., 2002) resulting in imatinib being licensed for patients with KIT positive, unresectable and/or metastatic malignant gastrointestinal stromal tumours. The recommended dose is 400 mg daily (see also Chapter 15, p. 186).
Toxicity of imatinib Imatinib is generally well tolerated, although it can be associated with haematological toxicity, especially in the treatment of CML. Other toxicities include peripheral oedema, fluid retention, skin rash, nausea, abdominal pain, diarrhoea and fatigue. 15
Rachel Jones and Robert Leonard
Table 2.1. Summary of tyrosine kinase inhibitors in clinical use and trials. Name
Target
Tumour type
Imatinib
BCR-ABL/KIT
Chronic myelogenous leukaemia
Gefitinib
EGFR
Non-small-cell lung cancer
Gastrointestinal stromal tumour Erlotinib
EGFR
Non-small-cell lung cancer
Sunitinib
VEGFR/PDGFR/KIT/FLT3
Renal cell carcinoma
Sorafenib
BRAF/VEGFR
Renal cell carcinoma
Lapatinib
EGFR/HER-2
Breast cancer
Gastrointestinal stromal tumour Melanoma Non-small-cell lung cancer Colorectal cancer Canertinib
EGFR
Breast cancer Non-small-cell lung cancer Ovarian cancer
Vatalanib
VEGFR
Colorectal Prostate Renal Glioblastoma multiforme
Zactima
VEGFR/EGFR
Non-small-cell lung cancer
ZD6474 BCR = breakpoint cluster region; ABL = Abelson protein tyrosine kinase; KIT = kitten (v-kit: derived from feline sarcoma virus – abbreviated ‘kitten’; official name v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog); EGFR = epidermal growth factor receptor; VEGFR = vascular endothelial growth factor receptor; PDGFR = platelet-derived growth factor receptor; FLT = fms-related tyrosine kinase; BRAF = v-raf murine sarcoma viral oncogene homolog B1; HER-2 = human epidermal growth factor receptor 2. Adapted from Arora and Scholar, 2005.
EGFR tyrosine kinase inhibitors Gefitinib Gefitinib is an EGFR tyrosine kinase inhibitor that was initially studied in non-small-cell lung cancer (NSCLC). It is also being studied in other cancers, including breast and head and neck cancer. Gefinitib acts by reversibly inhibiting autophosphorylation, altering transcription factor activity, and shifting cells from S phase into G0/G1. Despite some promising early results (Fukuoka et al., 2003; Kris et al., 2002), recent studies in NSCLC have shown no benefit from adding gefitinib to chemotherapy (INTACT 1 and 2; Giaccone et al., 2002; Johnson et al., 2002). The main side effects are nausea, vomiting, diarrhoea and an acneiform rash. Gefitinib is not currently licensed for use in the UK. 16
Erlotinib Erlotinib is an oral reversible EGFR-targeted tyrosinekinase inhibitor that is highly selective for its receptor at nanomolar concentrations. It causes cell-cycle arrest in G1 phase and apoptosis. In progressive NSCLC after chemotherapy it gives a response rate of 12% with a median survival of 8.4 months (Perez-Soler, 2004). In this trial, the occurrence and severity of a rash was associated with improved survival. Erlotinib is currently licensed for the treatment of patients with locally advanced or metastatic NSCLC after failure of at least one prior chemotherapy regimen. The recommended dose is 150 mg daily. The most common dose-limiting toxicities associated with erlotinib are acneiform skin rash and diarrhoea, which may require a dose reduction.
Biological treatments in cancer
Lapatinib Lapatinib is an oral dual tyrosine-kinase inhibitor of EGFR and HER-2. It is active against many breast cancer cell lines in vitro, including trastuzumab-conditioned HER-2-positive cell lines. A phase III trial in patients with metastatic HER-2-positive breast cancer who had previously received anthracyclines, taxanes and trastuzumab compared a combination of lapatinib with capecitabine against capecitabine alone. The results showed a significant increase in response rate (22.5% versus 14.3%) and median progression-free survival (36.9 weeks versus 19.7 weeks) in favour of the combination (Geyer, conference presentation, ASCO, 2006). Further phase III trials are ongoing and are investigating lapatinib as both a single agent and in combination with chemotherapy and hormone therapy. The main toxicities of lapatinib are diarrhoea and skin rash.
Management of the skin rash with anti-EGFR agents Drugs that target the EGFR, including erlotinib, gefitinib, lapatinib and cetuximab, produce a characteristic rash as a side effect. The rash is characterised by inflammatory papules and pustules on the face, chest and back and may resemble folliculitis or an acneiform drug eruption. The development of a rash is one of the best predictive factors for disease response. The rash may respond to clindamycin phosphate 1% gel, a combination of clindamycin 1% and benzoyl peroxide 5% gel, or oral antibiotics such as tetracycline 250 mg q.d.s. or minocycline 100 mg b.d. Topical and systemic corticosteroids have produced variable responses. Patients with extensive or persistent skin involvement should be referred to a dermatologist.
Vascular endothelial growth factor receptor tyrosine kinase inhibitors Sunitinib Sunitinib is an oral multi-targeted kinase inhibitor which inhibits vascular endothelial growth factor receptors (VEGFR1, VEGFR 2 and VEGFR 3), platelet-derived growth factor receptors (PDGFRα and PDGFRβ), KIT and FLT3. In patients with metastatic renal cell cancer who have failed prior cytokine therapy, sunitinib gives a response rate of 33%, and 37% achieve stable disease for three months (Eskens, 2004). A recent phase III trial that compared sunitinib with IFNα as first-line therapy in metastatic renal cell cancer showed a statistically
significant improvement in progression-free survival (47.3 versus 24.9 weeks) and an objective response rate (24.8% versus 4.9%) in favour of sunitinib (Motzer et al., 2006). Sunitinib is also active against GIST. In a phase III study in refractory GIST patients, sunitinib delayed the median time to progression from 1.5 to 6 months and improved overall survival (Demetri et al., 2005). Sunitinib is licensed for (a) the management of unresectable and/or metastatic malignant GISTs that are resistant to imatinib and (b) the treatment of advanced and/or metastatic renal cell carcinoma following failure of IFNα or IL-2 therapy. The recommended dose of sunitinib is 50 mg taken orally once a day for four consecutive weeks, followed by a two-week rest period. The dose may be increased or decreased, based on the patient’s ability to tolerate it. The daily dose should neither exceed 87.5 mg nor be decreased below 37.5 mg. Significant toxicities have been associated with sunitinib, including left ventricular dysfunction, haemorrhagic events, hypertension, gastrointestinal toxicity including nausea and diarrhoea, and myelosuppression. Dermatological toxicity can include skin discolouration and depigmentation of the hair or skin.
Sorafenib Sorafenib is an oral multi-targeted kinase inhibitor that targets RAF kinase, VEGFR and PDGFR. In patients with advanced renal cell cancer who had received one previous systemic treatment, sorafenib significantly prolonged progression-free survival when compared with best supportive care (Escudier et al., 2005). Malignant melanoma can be associated with a mutation in BRAF, which is targeted by sorafenib, and studies are under way to determine whether sorafenib is effective. Sorafenib is indicated for the treatment of patients with advanced renal cell carcinoma who have failed previous treatment with IFNα- or IL-2-based therapy or who are considered unsuitable for such therapy. The recommended dose is 400 mg twice a day. Sorafenib has been associated with tolerable toxicity including hand and foot syndrome, skin rash, hypertension, diarrhoea, and fatigue.
Monoclonal antibodies Early attempts at using monoclonal antibodies were limited by the strong immune responses evoked by murine antibodies, but advances in genetic engineering have provided better tolerated preparations. 17
Rachel Jones and Robert Leonard
Table 2.2. Nomenclature of monoclonal antibodies. Type
Origin
Nomenclature
Murine
Mouse
“-omab”
Chimeric
65 to 90% human
“-ximab”
murine variable regions Humanised
95% human
“-zumab”
5% murine hypervariable region Human
Human
“-umab”
Adapted from Wick, 2004.
Classification and nomenclature Antibodies consist of four polypeptides: two heavy chains and two light chains which are joined together by disulphide bonds to form a ‘Y’ shape. Each chain has a variable amino acid sequence at its tip, which is the antigen binding site. The constant region (Fc) determines the mechanism that is used to destroy the antigen. Antibodies can be classified according to their constant region (e.g. IgG, IgM). Monoclonal antibodies are named according to their components (see Table 2.2). The following are tumour antigen characteristics that are ideal for targeting by antibodies (Harris, 2004): r The antigen should be expressed stably and homogeneously by tumour cells. r There should be limited expression of the antigen in normal tissues. r There should be minimal or no soluble form of the antigen (to avoid rapid clearance or extracellular binding of the antibody). r The antigen should be present on the cell surface to allow clear access for the antibody. The mechanisms of antibody action are as follow: r Tumour cells are killed by activation of the immune system through the Fc component, by either complement-mediated or antibody-dependent cell-mediated cytotoxicity. r Receptor protein kinase phosphorylation is inhibited by the antibody binding to the receptor or to the ligand.
Trastuzumab Trastuzumab is a humanised IgG1 antibody that targets the HER-2 protein, which is overexpressed in approx18
imately 15 to 20% of patients with breast cancer. It is active against HER-2-positive breast cancer and it has been approved by NICE in metastatic breast cancer, either in combination with paclitaxel or as a single agent, in those who have received prior chemotherapy (NICE, 2002). Trastuzumab is also being investigated as an adjuvant treatment for breast cancer following chemotherapy. Interim analysis of the HERA trial, in which patients were randomised to observation, 1 year or 2 years of trastuzumab, showed a significant improvement in disease-free survival, and the recent 2006 ASCO update confirmed an overall survival benefit in the one-year arm (I. Smith, conference presentation, ASCO, 2006). NICE has approved one year’s treatment with trastuzumab following adjuvant chemotherapy for patients with HER-2-positive tumours (+++ on IHC; NICE, 2006). In metastatic breast cancer, trastuzumab is indicated for the treatment of patients with HER-2-positive breast cancer: r As monotherapy in patients who have received at least two chemotherapy regimes (including at least an anthracycline and a taxane unless patients are unsuitable for these treatments). r In combination with paclitaxel in patients who have not received chemotherapy for their metastatic disease and for whom an anthracycline is not suitable. r In combination with docetaxel for patients who have not received chemotherapy for their metastatic disease. Toxicities associated with trastuzumab are an acute infusion reaction and hypersensitivity, including anaphylaxis. Approximately 10% of patients have deterioration in their left ventricular function and 2 to 4% develop significant heart failure. Trastuzumab is a cardiotoxic drug, and a baseline measurement of cardiac function such as a MUGA scan must be performed. Cardiac function then needs to be monitored every 3 to 6 months because of possible left ventricular impairment. Monitoring is especially important in the adjuvant setting.
Cetuximab Cetuximab is a chimeric IgG1 antibody that targets the EGFR (ERB B1). In patients with metastatic irinotecanrefractory colorectal cancer, the BOND study compared irinotecan and cetuximab with cetuximab alone. There was a significantly improved response rate and median
Biological treatments in cancer
time to progression for combination therapy (Cunningham et al., 2004). In metastatic colorectal cancer, cetuximab is currently licensed for use in combination with irinotecan for patients with EGFR-expressing tumours after failure of chemotherapy that included irinotecan. In patients with head and neck cancer, cetuximab improves the duration of locoregional control, progression-free survival and overall survival compared to radiotherapy alone (Bonner et al., 2006). It is licensed for head and neck tumours in combination with radiotherapy for locally advanced squamous cell carcinoma. A possible mechanism of action is through inhibition of the EGFR/TKI complex, which can become activated during radiotherapy treatment. Cetuximab is given once a week. The initial dose is 400 mg/m2 followed by 250 mg/m2 . The toxicities include an infusion reaction, nail disorders and an acneiform rash which may be associated with response. The management of cetuximab rash is discussed in a previous section (see p. 17).
Bevacizumab Bevacizumab is a recombinant humanised monoclonal IgG1 antibody targeted against VEGF. In metastatic colorectal cancer, adding bevacizumab to chemotherapy significantly increases progression-free survival but there is no evidence of improved overall survival (Hurwitz et al., 2004; Kabbinavar et al., 2003, 2005). Bevacizumab is licensed for use in combination with intravenous 5-FU/folinic acid or intravenous 5FU/folinic acid with irinotecan as a first-line treatment for patients with metastatic colorectal carcinoma. The recommended dose is 5 mg/kg body weight given every two weeks as an intravenous infusion. Dose reduction for toxicity may be required. The toxicities associated with bevacizumab are hypertension, arterial thromboembolism, haemorrhage, poor wound healing, proteinuria, and gastrointestinal perforation.
Rituximab Rituximab is a chimeric IgG1 antibody that acts against the CD20 antigen on B lymphocytes (Price and Sikora, 2002). It is licensed for use in NHL and has been approved by NICE as a first-line treatment in combination with CHOP chemotherapy in CD20-positive diffuse large B-cell lymphoma and as an option in combination with cyclophosphamide, vincristine and prednisolone
in follicular lymphoma. The toxicities associated with treatment are acute infusion reactions, tumour lysis syndrome and mucocutaneous toxicity.
Radiolabelled antibodies Some monoclonal antibodies are further modified through conjugation with radioisotopes to allow more accurate delivery of radiation with less toxicity to surrounding normal tissue. They are currently being assessed in clinical trials, but are not licensed for use in the UK.
Haemopoietic colony stimulating factors Granulocyte colony stimulating factor (G-CSF) Endogenous G-CSF induces the maturation of neutrophils. Recombinant preparations are available for clinical use, and they reduce the morbidity and mortality of neutropenia. Both short- and long-acting (pegylated) agents are available, which can reduce the risk of infections and can reduce treatment delays in patients treated with chemotherapy. Potential uses for G-CSF include the following: r Primary prophylaxis with the first cycle of chemotherapy. r Secondary prophylaxis in subsequent cycles following the first episode of febrile neutropenia. r Therapeutically, to reduce duration of neutropenia during acute infective episode. r To allow the delivery of dose-dense chemotherapy regimens (chemotherapy regimens in which the cycle length has been reduced). Several international guidelines have been published that review the use of these agents for the aforementioned indications. Recent international guidelines from the National Comprehensive Cancer Network (www.nccn.org) and ASCO (Ozer et al., 2000) are available but the reader is advised to be aware that local guidelines may differ for the use of these agents.
Primary and secondary prophylaxis This is discussed in Chapter 1 (see p. 6).
Febrile patients This is discussed in Chapter 6 (see p. 78). 19
Rachel Jones and Robert Leonard
Dose-density chemotherapy regimens The concept of dose-density chemotherapy is to deliver multiple, reduced-interval cycles of chemotherapy with G-CSF. Although there is evidence of a benefit with this approach in NHL (Pfreundschuh et al., 2004) and breast cancer (Citron et al., 2003), use of G-CSF for dosedensity chemotherapy regimens should be restricted to clinical trials.
Erythropoietin Most patients who are symptomatic from anaemia are treated with blood transfusions, although an alternative treatment is to administer recombinant human erythropoietin. Three recombinant erythropoietins are licensed for use in patients who have symptomatic anaemia in non-myeloid malignancies and who are receiving chemotherapy: r Epoetin alpha. r Epoetin beta. r The longer-acting darbepoetin alpha. The short-acting erythropoietins are licensed for treatment of symptomatic anaemia in patients receiving chemotherapy for solid tumours, malignant lymphoma, or multiple myeloma (epoetin beta is also licensed for CLL). Darbepoetin alpha is only licensed for use in patients receiving chemotherapy for solid tumours. All are licensed to be used in patients with a haemoglobin level less than 11 g/dl. A Cochrane review of 27 trials involving a total of 3287 patients showed that recombinant human erythropoietin significantly decreases the need for red-cell transfusions (Bohlius et al., 2004). However, it is not clear whether erythropoietin improves tumour response and overall survival. The use of erythropoietin is currently under review by NICE.
cific tumour-associated antigens that will lead to the death of the tumour cell. To achieve this aim: r The tumour antigen needs to be recognised on an antigen-presenting cell (ApC) by interaction with a cytotoxic T-cell receptor. r The ApCs need to be co-stimulated through members of the B7 family and various adhesion molecules (e.g. ICAM1). r Effector mechanisms need to be generated (e.g. secretion of cytokines causing tumour cell destruction). r Immune-suppressive factors produced by the tumour (e.g. TGF-β) need to be avoided. Tumour-associated antigens are usually proteins but they can be carbohydrates (e.g. gangliosides) and glycoproteins. They may differ from normal cells only in a quantitative manner (i.e. they are present in significantly greater amounts than in normal cells). Table 2.3 shows tumour-associated antigens that may be suitable targets for vaccination.
Antigen vaccines Viral-protein-based vaccines Viral proteins from cancer-causing viruses may be used as antigens to prevent infection by the virus. HPV virus types 16 and 18 are responsible for about 70% of cervical cancers. Two vaccines have been developed which use non-infectious virus-like particles of the major capsid protein L1. Cervarix® prevents infection by HPV types 16 and 18 and shows some cross-protection against other oncogenic HPV viruses, such as HPV 31 and 45 (Harper et al., 2006). Gardasil® prevents infection by HPV 16 and 18, and by the HPV types 6 and 11, which are responsible for genital warts (Munoz, 2006). It is hoped that HPV vaccines will have a major impact on the incidence of cervical cancer in the future. Discussions are ongoing about who, and when, to vaccinate.
Peptide-based vaccines and carbohydrate-based vaccines
Vaccines Vaccination strategies have been investigated as a treatment option in the management and prevention of cancer. A major recent advance in vaccine development for cancer prevention has been the development of two vaccines against human papilloma virus (HPV), the main cause of cervical cancer. The aim of vaccination against an established tumour is to stimulate a cellular immune response against spe20
Peptides derived from tumour-associated antigens or gangliosides (which are sialated glycolipid antigens) can be used as vaccines. Gangliosides are over expressed in melanoma.
Cellular vaccines Cellular vaccines can be based on whole cells, tumour lysates, or antigens that have been shed from the surface of tumour cells grown in vitro.
Biological treatments in cancer
Table 2.3. Examples of tumour-associated antigens that are potential targets for cancer vaccines. Tumour-associated antigen
Tumour
Melanoma antigen expression family (MAGE)
Melanoma, breast (normal testicular proteins)
Renal antigen expression family (RAGE)
Melanoma, kidney
Melanoma antigen recognised by T cells 1/Melan A (MART)
Melanoma
Caspase 8
Head and neck tumours (squamous cell carcinoma)
B catenin
Melanoma
TGF-β receptor II
Colorectal
MUC1
Breast, pancreatic, lung and colorectal
ERBB2/HER-2
Breast, ovarian and lung
αFP
Hepatocellular
Prostate specific antigen
Prostate
Gangliosides
Melanoma
CEA
Colon
TGF-β = transforming growth factor β; MUC1 = mucin 1; ERBB2/HER-2 = human epidermal growth factor receptor 2; αFP = alpha feto-protein; CEA = carcino-embryonic antigen. Modified from Espinoza-Delgado, 2002; Janeway et al., 2001; and Rosenberg et al., 2004.
Dendritic cell vaccines Dendritic cells are potent antigen-presenting cells and they can be used to present the tumour-associated antigen to increase the chances of an immune response to the vaccine. Their use is currently restricted to clinical trials.
established cancer is recommended for use outside a clinical trial. Vaccines are likely to play a more important role in cancer prevention as seen with emerging data on recombinant vaccines against HPV 16 and 18 to prevent cervical cancer.
REFERENCES
Other vaccines r DNA-based vaccines can be specific genes that encode the tumour-associated antigen cloned into a bacterial plasmid. r Other nucleotide sequences can be included that code for immune effectors to increase the immunological response to the vaccine. r Recombinant viruses (e.g. adenovirus) can be used as vectors for the tumour-associated antigen. r Heat shock proteins can promote the development of dendritic cells into mature ApCs. Interest exists in combining them with tumour-associated antigens.
Clinical data on vaccines A recent review of cancer vaccines by the National Cancer Institute (USA) showed that the objective response rate in 440 patients with established tumours was only 2.6% (Rosenberg et al., 2004). At present no vaccine for
Arora, A. and Scholar, E. M. (2005). Role of tyrosine kinase inhibitors in cancer therapy. J. Pharm. Exp. Ther., 315, 971–9. Atkins, M. B., Kunkel, L., Sznol, M. et al. (2000). High-dose recombinant interleukin-2 therapy in patients with metastatic melanoma: long-term survival update. Cancer J. Sci. Am., 6 (Suppl. 1), S11–14. Atzpodien, J., Kirchner, H., Hanninen, E. L. et al. (1993). Interleukin-2 in combination with interferon-alpha and 5-fluorouracil for metastatic renal cell cancer. Eur. J. Cancer, 29A (Suppl. 5), S6–8. Baaten, G., Voogd, A. C. and Wagstaff, J. (2004). A systematic review of the relation between interleukin-2 schedule and outcome in patients with metastatic renal cell cancer. Eur. J. Cancer, 40, 1127–44. Bohlius, J., Langensiepen, S., Schwarzer, G. et al. (2004). Erythropoietin for patients with malignant disease. The Cochrane Database of Systematic Reviews. Issue 3. Art. no. CD003407. Bonner, J. A., Harari, P. M., Giralt, J. et al. (2006). Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N. Engl. J. Med., 354, 567–78. Citron, M. L., Berry, D. A., Cirrincione, C. et al. (2003). Randomized trial of dose-dense versus conventionally scheduled and sequential versus concurrent combination chemotherapy as
21
Rachel Jones and Robert Leonard
postoperative adjuvant treatment of node-positive primary breast cancer: first report of Intergroup Trial C9741/Cancer and Leukemia Group B Trial 9741. J. Clin. Oncol., 21, 1431–9. Clark, J. I., Atkins, M. B., Urba, W. J. et al. (2003). Adjuvant high-dose bolus interleukin-2 for patients with high-risk renal cell carcinoma: a Cytokine Working Group randomized trial. J. Clin. Oncol., 21, 3133–40. Creagan, E. T., Ahmann, D. L., Frytak, S. et al. (1986). Phase II trials of recombinant leukocyte A interferon in disseminated malignant melanoma: results in 96 patients. Cancer Treat. Rep., 70, 619–24. Cunningham, D., Humblet, Y., Siena, S. et al. (2004). Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N. Engl. J. Med., 351, 337–45. Demetri, G. D., Von Mehren, M., Blanke, C. D. et al. (2002). Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N. Engl. J. Med., 347, 472–80. Demetri, G. D., Van Oosterom, A. T., Blackstein, M. et al. (2005). Phase III multicenter, randomized, double-blind, placebo controlled trial of SU11248 in patients following failure of imatinib for metastatic GIST. J. Clin. Oncol., 23, s308. Escudier, B., Szczylik, C., Eisen, T. et al. (2005). Randomized phase III trial of the Raf kinase and VEGFR inhibitor sorafenib (BAY 43–9006) in patients with advanced renal cell carcinoma (RCC). Proc. Am. Soc. Clin. Oncol., 23, 4510. Eskens, F. A. L. M. (2004). Angiogenesis inhibitors in clinical development; where are we now and where are we going? Br. J. Cancer, 90, 1–7. Espinoza-Delgado, I. (2002). Cancer vaccines. The Oncologist, 7 (Suppl. 3), 20–33. Fukuoka, M., Yano, S., Giaccone, G. et al. (2003). Multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small-cell lung cancer (The IDEAL 1 Trial) [corrected]. J. Clin. Oncol., 21, 2237–46. Fyfe, G., Fisher, R. I., Rosenberg, S. A. et al. (1995). Results of treatment of 255 patients with metastatic renal cell carcinoma who received high-dose recombinant interleukin-2 therapy. J. Clin. Oncol., 13, 688–96. Giaccone, G., Johnson, D. H., Manegold, C. et al. (2002). A phase III clinical trial of ZD 1829 (Iressa) in combination with gemcitabine and cisplatin in chemotherapy-naive patients with advanced non-small-cell lung cancer (INTACT 1). Ann. Oncol., 13, Abstr. 4. Harper, D. M., Franco, E. L., Wheeler, C. M. et al. (2006). Sustained efficacy of up to 4.5 years of a bivalent L1 virus-like particle vaccine against human papillomavirus types 16 and 18: follow-up from a randomised control trial. Lancet, 367, 1247–55. Harris, M. (2004). Monoclonal antibodies as therapeutic agents for cancer. Lancet Oncol., 5, 292–302. Hurwitz, H., Fehrenbacher, L., Novotny, W. et al. (2004). Bevacizumab plus irinotecan, fluorouracil and leucovorin for metastatic colorectal cancer. N. Engl. J. Med., 350, 2335–42. Janeway, C. A., Travers, P., Walport, M. et al. (2001). Immunobiology: The Immune System in Health and Disease, 5th edn. New York: Garland. Johnson, D. H., Herbst, R. S., Giaccone, G. et al. (2002). ZD 1839 (Iressa) in combination with paclitaxel and carboplatin in
22
chemotherapy-naive patients with advanced non-small-cell lung cancer (NSCLC): results from a phase III clinical trial (INTACT 2). Ann. Oncol., 13, 127. Kabbinavar, F., Hurwitz, H., Fehrenbacher, L. et al. (2003). Phase II randomized trial comparing bevacizumab plus fluorouracil (FU)/leucovorin (LV) with FU/LV alone in patients with metastatic colorectal cancer. J. Clin. Oncol., 21, 60–5. Kabbinavar, F., Schulz, J., McCleod, M. et al. (2005). Addition of bevacizumab to bolus fluorouracil and leucovorin in first line metastatic colorectal cancer: results of a randomized phase II trial. J. Clin. Oncol., 23, 3697–705. Kantarjian, H., Sawyers, C., Hochhaus, A. et al. (2002). Hematologic and cytogenetic responses to imatinib mesylate in chronic myelogenous leukemia. N. Engl. J. Med., 346, 645–52. Kris, M. G., Natale, R. B., Herbst, R. S. et al. (2002). A phase II trial of ZD 1839 (Iressa) in advanced non-small cell lung cancer patients (NSCLC) who had failed platinum- and docetaxel-based regimens (IDEAL 2). Proc. Am. Soc. Clin. Oncol., 21, Abstr. 1166. Motzer, R. J., Bacik, J., Murphy, B. A. et al. (2002). Interferon-alfa as a comparative treatment for clinical trials of new therapies against advanced renal cell carcinoma. J. Clin. Oncol., 20, 289–96. Motzer, R. J., Hutson, T. E., Tomczak, P. et al. (2006). Phase III randomized trial of sunitinib malate versus interferon-alfa as first line systemic therapy for patients with metastatic renal cell cancer. Proc. Am. Soc. Clin. Oncol., 24, LBA3. Munoz, N. (2006). A review of the phase III clinical data for the quadrivalent human papilloma vaccine (Gardasil® ). Eur. J. Obstet. Gynecol. Reprod. Biol., Aug. 18 [Epub ahead of print]. NICE. (2002). Guidance on the Use of Trastuzumab for the Treatment of Advanced Breast Cancer. Technology Appraisal Guidance No. 34. London: National Institute for Clinical Excellence. NICE. (2006). Trastuzumab for the Adjuvant Treatment of Early-Stage HER2 Positive Breast Cancer. Technology Appraisal Guidance No. 107. London: National Institute for Clinical Excellence. Ozer, H., Armitage, J. O., Bennett, C. L. et al. (2000). 2000 update of recommendations for the use of hematopoietic colony-stimulating factors: evidence-based, clinical practice guidelines. American Society of Clinical Oncology Growth Factors Expert Panel. J. Clin. Oncol., 18, 3558–85. Perez-Soler, R. (2004). The role of erlotinib (Tarceva, OSI 774) in the treatment of non-small cell lung cancer. Clin. Cancer Res., 10, 4238s–40s. Pfreundschuh, M., Trumper, L., Kloess, M. et al. (2004). Two-weekly or 3-weekly CHOP chemotherapy with or without etoposide for the treatment of young patients with good-prognosis (normal LDH) aggressive lymphomas: results of the NHL-B1 trial of the DSHNHL. Blood, 104, 626–33. Price, P. and Sikora, K. (2002). Treatment of Cancer. London: Chapman and Hall. Rosenberg, S. A., Yang, J. C. and Restifo, N. P. (2004). Cancer immunotherapy: moving beyond current vaccines. Nat. Med., 10, 909–15. Wick, J. (2004). What’s in a drug name? J. Am. Pharm. Assoc., 44, 12–14. Wirth, M. P. (1993). Immunotherapy for metastatic renal cell carcinoma. Urol. Clin. North Am., 20, 283–95.
3
HORMONES IN CANCER Jacinta Abraham and John Staffurth
Introduction Hormonal therapies are some of the oldest active systemic anticancer therapies in use today. In 1896 Beatson demonstrated that surgical oophorectomy resulted in tumour regression in some premenopausal women with metastatic breast cancer, and, by doing so, he was the first to identify a link between ovarian function and breast cancer. Substantial evidence now exists that hormones play a key role in both the cause and the outcome of several cancers. Although this is most clearly seen in breast and prostate cancer, other cancers that may exhibit hormone dependence include endometrial, ovarian and testicular cancers. Hormones are classified into two groups: r Non-steroidal hormones include peptides, polypeptides or derivatives of amino acids, and they generally act via cell-membrane-localised receptors which trigger second messengers within the cytoplasm. r Steroidal hormones, such as oestrogens, androgens and progestins, bind to intracellular receptors to mediate their action. This chapter will focus primarily on the steroidal hormones that are of particular importance in breast and prostate cancer: the oestrogens, progestins and androgens. It should be read in conjunction with the relevant site-specific chapters (Chapters 16 and 19). This chapter provides some background knowledge of the production and functioning of hormones and their receptors, which will help in the understanding of commonly used therapies. The aetiology of hormone-related cancers is discussed in the relevant site-specific chapters. Steroidal hormones have the potential to activate oncogenes or inactivate tumour-suppressor genes, which can produce a sequence of changes within the cell that ultimately leads to cancer. Frequently, the continued growth of the cancer is dependent upon the continued stimulation by hormones; in this situation, removing the hormonal stimulus causes the cancer to
regress. Anticancer hormone therapies work in a number of different ways, for example, by affecting hormone synthesis, metabolism or action, or by altering hormone receptor expression within the cell. A summary of the major historical landmarks in anticancer hormonal therapy since Beatson’s observation in 1896 is listed here: r Large doses of oestrogenic steroids inhibit gonadotropin secretion in women (Zondek, 1940). r Surgical orchidectomy results in the regression of advanced prostate cancer (Huggins and Hodges, 1941). r Adrenalectomy results in regression of advanced breast and prostate cancers (Huggins and Bergenstal, 1952). r Prolonged exposure to oestrogens or androgens, alone or in combination, leads to breast and prostate cancers in rats (Noble, 1980).
Hormone synthesis Oestrogen synthesis In premenopausal women, oestrogens are synthesised from cholesterol. Oestrogen synthesis takes place mainly in the granulosa cells of the ovaries; production is cyclical and is controlled by positive and negative feedback via the hypothalamic–pituitary–gonadal axis (see Figure 3.1). Inhibin is a polypeptide which is also produced by ovarian granulosa cells, and it inhibits follicle-stimulating hormone (FSH) release by pituitary gonadotrophs. In postmenopausal women, the main site of oestrogen synthesis is adipose tissue, via the enzyme aromatase, and production varies depending on environmental and genetic factors such as obesity. In men, 30% of plasma oestrogens originate from the testis and 70% arise from peripheral aromatisation of androstenedione and testosterone; testosterone is converted to oestradiol by the enzyme aromatase. Although only 0.5% of the total daily production of 23
Jacinta Abraham and John Staffurth
Negative Feedback
Hypothalamus
GnRH
GnRH
Negative Feedback
Hypothalamus CRH
CRH Pituitary FSH/LH
Pituitary FSH/LH
ACTH
Ovaries
Adrenals
Androgen conversion within ovarian granulosa cells Testosterone Aromatase Oestradiol
Androgen
Testes
Adrenals
Cholesterol
Cholesterol
Testosterone
Androstenedione
Prostate Peripheral conversion within adipose tissue/skin Androstenedione Aromatase Oestrone 17βOH dehydrogenase Oestradiol
Testosterone
ACTH
DHT
Androstenedione
Peripheral fat DHT
Androstenedione
AR activation and nuclear translocation
Figure 3.2. The hypothalamic–pituitary–gonadal axis in the male. GnRH = gonadotropin-releasing hormone; CRH = corticotropinreleasing hormone; FSH = follicle-stimulating hormone; LH = luteinising hormone; ACTH = adrenocorticotrophic hormone; DHT = 5α-dihydrotestosterone; AR = androgen receptor. Testosterone, DHT, androstenedione, and circulating oestrogens contribute to negative
Oestrogens Breast Endometrium Ovary Figure 3.1. The hypothalamic–pituitary–gonadal axis in the female. GnRH = gonadotropin-releasing hormone; CRH = corticotropin-releasing hormone; FSH = follicle-stimulating hormone; LH = luteinising hormone; ACTH = adrenocorticotrophic hormone.
testosterone is converted, oestradiol is a much more potent gonadotropin inhibitor than testosterone.
Androgen synthesis In men, the main circulating androgen is testosterone, 90% of which is produced by the Leydig cells of the testis (see Figure 3.2). The remaining circulating androgens, such as dihydroxyandrostenedione (DHA), androstenedione, and DHA sulphate, are produced in the adrenal cortex from cholesterol. All of these androgens may be 24
feedback.
metabolised within the prostate to the more biologically active 5α-dihydrotestosterone (DHT), which binds to the androgen receptor (AR) with a three- to five-times greater affinity than testosterone.
Hormone receptors Oestrogen receptor There are two oestrogen receptors (ERs) which are encoded by separate genes: r Oestrogen receptor alpha (ERα) is encoded by a gene on chromosome 6. r Oestrogen receptor beta (ERβ) is encoded on chromosome 14. The two receptors have a similar structure but they share only 47% amino acid homology. Not all ER-positive breast cancers respond to hormonal treatments, and there is some evidence that different expression of the two ERs in breast cancer cells affects hormone responsiveness and hormone resistance. For example, ERα is often markedly upregulated
Hormones in cancer
in breast cancer and it is the marker that can best predict responsiveness to hormonal treatments. But ERβ may modulate the ability of cells to respond to oestrogen, and its expression has been implicated in tamoxifen resistance (Miller et al., 2006).
Oestrogen receptor functioning To understand how antioestrogen treatments work, it is necessary to be aware of three important domains on the ER: activating functions 1 and 2 (AF1 and AF2) which carry out the effector functions of the receptor and the ligand binding domain. First, oestradiol binds to the ER in the nucleus via the ligand binding domain on AF2, and this causes dimerisation, phosphorylation, and a conformational shape change. This allows the ER to bind to DNA oestrogen response elements (EREs), which are upstream of oestrogen-responsive genes. Second, the transcriptional activity of the genes that contain EREs is modulated by AF1 and/or AF2. Third, co-regulator proteins can also influence ER-mediated transcription; depending on the ligand, the ER interacts with either co-repressors or co-activators to inhibit or enhance its transcriptional activity on target genes (Milano et al., 2006). Tamoxifen blocks AF2 when it competes with oestradiol for ER binding, but AF1 remains functional, which accounts for the oestrogenic activity seen in some tissues. Aromatase inhibitors block the production of oestrogens, and in doing so they effectively prevent the action of both AF1 and AF2.
Oestrogenic effects The following is a list of the effects of oestrogen: r Development of female secondary sexual characteristics. r Endometrial growth. r Bone formation. r Procoagulant. r Reduced low-density lipoprotein (LDL).
Progesterone receptor There is one gene for the progesterone receptor (PgR) but it exists as two isoforms, A (PR-A) and B (PR-B), which have different physiological functions. PgR is important in mammary gland development, and an excess production of PR-B is associated with breast cancer risk. However, PR-A can repress PR-B and ER and its expression is related to tamoxifen resistance (Hopp et al., 2004).
Main actions of progesterone Progesterone performs the following actions: r Causes endometrial secretory phase. r Increases the viscosity of cervical mucus. r Facilitates mammary gland development.
Receptor status in breast cancer and response to hormone therapy At present, it is standard practice to measure ER and PgR status in breast cancer, but it is not usual to distinguish between the subtypes. Approximately 75% of postmenopausal and 50% of premenopausal women with breast cancer will have hormone-receptor-positive tumours. Whether the cancer is sensitive to hormone treatment is related to the degree of ER and PgR positivity, but receptor status alone is not the only determinant of hormone responsiveness. r ER+/PgR+ disease confers the highest response rates of up to 70%. The response is less for ER+/PgR− cancers. r In contrast, ER−/PgR− cancers, with current testing, will have a response rate to hormone therapy of less than 5% and patients with these tumours are likely to derive a greater benefit from chemotherapy. r ER−/PgR+ cancers account for 5% of all breast cancers and they should be regarded as being hormone responsive because they will have response rates of up to 30%. r An ER-negative status may be detected in metastatic or recurrent disease even when the primary tumour is ER-positive. It is not clear if this represents a change in ER expression or if it reflects a heterogeneous primary tumour in which the ER-negative component has a survival advantage, allowing it to metastasise.
Androgen receptor The AR gene is on the long arm of the X chromosome, q11-13, and it has eight exons. In contrast to the situation in breast cancer, the measurement of AR in patients with prostate cancer is not performed routinely because both hormone-dependent and -independent prostate cancers possess functioning AR. AR status provides no prognostic or diagnostic value.
Androgen receptor functioning DHT binds to the AR via the ligand binding domain, and this binding results in separation from heat shock protein 90 (HSP-90; a protein that maintains the AR 25
Jacinta Abraham and John Staffurth
Table 3.1. Current IBCSG trials of ovarian function suppression (OFS) in premenopausal women Trial SOFT
Trial design
Trial question
Tamoxifen 5 years or
Does OFS add to endocrine therapy in
OFS + tamoxifen 5 years or
premenopausal women?
OFS + exemestane 5 years TEXT
OFS + tamoxifen 5 years or OFS + exemestane 5 years
Is an aromatase inhibitor superior to tamoxifen in premenopausal women treated with OFS?
May or may not have chemotherapy PERCHE
OFS + tamoxifen or exemestane 5 years or OFS + chemotherapy + tamoxifen or
Is chemotherapy necessary in premenopausal women who receive OFS and hormone therapy?
exemestane 5 years
in an inactive state), dimerisation to form a DHT–AR complex, phosphorylation, and a conformational shape change. This complex is translocated to the nucleus of prostate cells, which allows DNA binding of the AR to androgen response elements (AREs) upstream of androgen-regulated genes. The activity of genes containing AREs can be modulated by transcription factors via the highly variable amino-terminal domain (ATD).
Androgenic effects The effects of androgen include the following: r Virilisation and development of secondary sexual characteristics. r Development of prostate and seminal vesicles. r Increased muscle mass. r Increased bone density. r Libido and frequency of erection.
Breast cancer: ovarian function suppression (OFS) In premenopausal women, the most obvious and direct way to reduce oestrogen production is to suppress ovarian function, which can be achieved by chemical, radiotherapeutic or surgical treatments. Chemical ablation is achieved by using gonadotropin-releasing hormone (GnRH/LHRH) agonists such as goserelin, which is currently the treatment of choice, particularly in early breast cancer, because its effects are reversible. The hormonal effects following administration of goserelin may be observed by measuring plasma levels of luteinising hormone (LH), FSH, and oestradiol.
26
In advanced disease, all three techniques are essentially equivalent and they produce response rates of 25 to 30%, with a median response duration of 9 to 12 months. In early breast cancer, OFS has been shown to produce a survival benefit equivalent to that achieved by chemotherapy (EBCTG, 2005). The various roles of OFS, endocrine therapy and chemotherapy in early breast cancer are now the subject of clinical trials coordinated by the International Breast Cancer Study Group (IBCSG; see Chapter 16).
Breast cancer: other hormone therapies Table 3.2 summarises some of the hormonal therapies used in breast cancer.
Antioestrogens Tamoxifen Tamoxifen is the most extensively studied and widely used antioestrogen in the treatment of breast cancer; it has been the mainstay of endocrine therapy for breast cancer for many years. Tamoxifen is described as a selective oestrogen receptor modulator (SERM) because of its agonist and antagonist activities, which causes a varied effect on genes, cells, and tissues.
Mode of action of tamoxifen
r SERMs modulate oestrogen activity by binding to the ER, resulting in a conformational shape change. This alters the balance of co-activator and co-repressor complexes and affects the regulation of activation
Hormones in cancer
Table 3.2. Hormonal therapies in breast cancer Drug
Type
Dose/route
Mode of action
Tamoxifen
Antioestrogen
20 mg daily p.o.
Competes with oestradiol for ER
Anastrazole
Non-steroidal aromatase
1 mg daily p.o.
Competitive aromatase inhibition
Letrozole
Non-steroidal aromatase
2.5 mg daily p.o.
Competitive aromatase inhibition Irreversible aromatase inhibition
binding inhibitor inhibitor Exemestane
Steroidal aromatase inhibitor
25 mg daily p.o.
Fulvestrant
ER antagonist
250 mg (5 ml) monthly i.m.
Downregulation of the ER protein
Megestrol acetate
Progestin
80–160 mg daily p.o.
Cellular action not fully understood; downregulation of ovarian steroidogenesis
Aminoglutethimide
First generation aromatase
250 mg twice daily p.o.
inhibitor
Inhibits the aromatase enzyme; is given with hydrocortisone to prevent adrenal insufficiency; rarely used because of side effects including rash, dizziness, nausea, and drowsiness
ER = oestrogen receptor.
domain AF2. AF1 remains active, which accounts for the oestrogenic effects of tamoxifen. r Impaired transcription of oestrogen-dependent genes occurs in the breast. r Tamoxifen also reduces plasma levels of the potent cancer mitogen IGF1.
Clinical studies of tamoxifen
r Five years of postsurgical adjuvant tamoxifen reduces the annual recurrence rate by 41% and annual mortality rate by 34% in ER-positive early breast cancer (EBCTG, 2005). r Tamoxifen is beneficial in both pre- and postmenopausal women. r Further detailed data and discussion on clinical studies and treatment recommendations may be found in Chapter 16 (p. 203).
r Thrombotic events include deep vein thrombosis and pulmonary embolism. The risk is increased further when tamoxifen is co-administered with chemotherapy. Tamoxifen should be avoided in individuals at risk of thrombosis. r An increased risk of endometrial changes exists, including hyperplasia and polyps. This is caused by the oestrogen agonist effects of tamoxifen. There is a small but statistically significant increase in endometrial cancer. Women on tamoxifen with abnormal vaginal bleeding should be promptly evaluated. r The beneficial effects of the weak agonist effect seen with tamoxifen are a reduction in cholesterol levels, decreased cardiac morbidity, and increased bone mineral density in postmenopausal women.
Fulvestrant Adverse effects and beneficial effects of tamoxifen
r Tamoxifen shows increased symptoms when compared with placebo: hot flushes (64 versus 48%), vaginal discharge (30 versus 15%), irregular menses (25 versus 19%), and thrombotic events (1.7 versus 0.4%) (Fisher et al., 1989).
Fulvestrant is an ER antagonist that competitively binds and downregulates the ER protein; no oestrogen agonist effect has been demonstrated. It is currently licensed for use in postmenopausal hormone-positive advanced breast cancer that has progressed following antioestrogen therapy. Side effects such as gastrointestinal
27
Jacinta Abraham and John Staffurth
symptoms, headache, back pain, hot flushes and pharyngitis have been reported.
Aromatase inhibitors Types
r The first generation of AIs includes aminoglutethimide, which has been available for more than 20 years. It has significant side effects and is now rarely used. r Third-generation AIs are better tolerated and are used more commonly. r AIs are classified as type 1 steroidal inhibitors, which lead to irreversible aromatase inactivation (e.g. exemestane and formestane), and type 2 nonsteroidal competitive inhibitors such as anastrazole and letrozole.
Mode of action AIs act by inhibiting or inactivating the p450 enzyme aromatase, and they significantly reduce serum oestradiol. AIs are contraindicated in premenopausal women because ovarian suppression would lead to feedback on the hypothalamic–pituitary–gonadal axis, resulting in gonadal stimulation. When AIs are indicated in these patients, the AIs should be given only after ovarian function suppression.
Use in early postmenopausal breast cancer Several studies of early breast cancer have compared AIs with tamoxifen and have confirmed the safety and efficacy of AIs; all show that the use of an AI is associated with an increased rate of disease-free survival. These studies have led us to redefine the optimal hormonal strategy for breast cancer patients. These studies and the various strategies now available are discussed in Chapter 16 (see p. 203) and shown in Table 3.3. A greater understanding of primary resistance to tamoxifen (in particular, being able to predict who is going to have tamoxifen-resistant disease) and longer-term follow-up data may help identify the subgroups that might benefit most from initial or subsequent AIs.
Use in premenopausal breast cancer As previously mentioned, AIs are contraindicated in premenopausal women unless they are undergoing OFS. In early premenopausal breast cancer, the role of concurrent use of OFS and AI is the subject of ongoing clinical trials coordinated by BIG/IBCSG (see Table
28
3.1). In advanced disease, a role clearly exists for OFS and AI use in women who have advanced disease with receptor-positive tumours and who have progressed on tamoxifen. AIs alone should be used with caution in premenopausal women who have become amenorrhoeic after chemotherapy. A recent report has shown that AIs may promote recovery of ovarian function in some of these women, even up to the age of 50 and after many months of amenorrhoea (Smith et al., 2006). It is suggested that serial measurements of oestradiol, LH, and FSH levels should be performed; however, these measurements require validated sensitive immunoassays which may not be readily available in all centres.
Adverse effects of aromatase inhibitors The comparative studies of AIs and tamoxifen have provided excellent comparative toxicity data. Tamoxifen is associated more with thromboembolic events, vaginal discharge, and endometrial hyperplasia or cancers, whereas the aromatase inhibitors are more linked with osteoporosis, fractures, and arthralgia. AIs are associated with an increased risk of musculoskeletal events, including arthralgia, fractures, and reduced bone mineral density (BMD). The clinical relevance of this reduction in BMD and how it should be monitored while the patient receives an AI are matters of ongoing debate. The ATAC bone substudy identified that women who had normal BMD at the outset and were treated with an AI did not become osteopenic or osteoporotic after five years (Howell et al., 2005). Those with a reduced BMD at the start were more at risk of developing a further reduction. Monitoring should be performed for individuals who are at a high risk of osteoporosis (see list that follows), accompanied by strong advice that reinforces diet and lifestyle (i.e. high-calcium diet and smoking cessation). The role of bisphosphonates in early breast cancer patients at risk of osteoporosis is a subject of current clinical trials. Risk factors for osteoporosis include the following: r Radiographic evidence of osteopenia and/or vertebral deformity. r Loss of height, thoracic kyphosis. r Previous fragility fracture. r Prolonged corticosteroid use. r Premature menopause (< 45 years). r Prolonged secondary amenorrhoea (> 1 year). r Primary hypogonadism.
Hormones in cancer
Table 3.3. Summary of main studies of early breast cancer that compare AIs with tamoxifen Trial
Study design
Patients (n)
Median FU
DFS
Other
ATAC
Upfront anastrazole vs
9366
68 months
Improved for
No benefit for
tamoxifen for 5 years or
anastrazole
combination arm;
in combination
(HR 0.87,
benefit greater in
(included ER unknown
p = 0.01)
receptor-positive
and negative) BIG-98
Letrozole (L) vs tamoxifen
patients 8010
25.8 months
Improved for
(T); 5 years T or 5 years
letrozole (HR
L or 2 years T, 3 years L
0.81, p = 0.003)
No crossover results available
or 2 years L, 3 years T IES
Tamoxifen for 5 years vs
4742
37.4 months
Improved for
Overall survival
2–3 years tamoxifen
exemestane
benefit in the
and switch to 2–3 years
(HR 0.76,
ER+ve group for
exemestane
p < 0.0001)
switching to exemestane: HR 0.83, p = 0.05
ABCSG 8/ ARNO 95
Tamoxifen for 5 years vs
3224
28 months
Improved for
tamoxifen for 2 years
anastrozole
benefit in ARNO
and anastrozole for 3
(HR 0.60,
95 for switching to
years; randomisation
p = 0.0009)
anastrazole: n = 979, HR 0.53,
for switching at trial
p = 0.045
entry MA-17
Overall survival
Tamoxifen for 5 years
5187
30 months
Improved for
Overall survival
followed by letrozole
letrozole (HR
benefit in node
for 5 years or placebo
0.58, p = 0.001)
positive: HR 0.60, p = 0.04
for 5 years
DFS = disease-free survival; ER = oestrogen receptor; FU = follow-up; HR = hazard ratio; vs = versus. Adapted from Coombes et al., 2004; Goss et al., 2005; Howell et al., 2005; Jakesz et al., 2005; Thurlimann, 2005.
r Maternal history of osteoporosis. r Low body mass index (< 19 kg/m2 ). The rate of occurrence of thrombotic events is reduced when using AIs compared to that for tamoxifen use. The rates of hot flushes are equivalent.
Progestins The two most widely used progestins are megestrol acetate and medroxyprogesterone acetate. These compounds are thought to act directly via the PgR but they also have an indirect action on the ER. Their biochemical effects are complex and not well understood. Progestins decrease the circulating levels of gonadotropins, inter-
fere with steroid synthesis, and exhibit glucocorticoid activity, which can lead to corticosteroidal side effects and adrenocorticotrophic hormone suppression. The optimal progestin dose is controversial. High doses have been used in premenopausal women but this is also associated with increased side effects. The several side effects seen with their use include weight gain, fluid retention, breast tenderness, nausea, and hypertension.
Breast cancer: mechanisms of resistance Hormone-positive tumours may show resistance to endocrine therapy, either at first exposure (de novo) or after a period of time (acquired). Understanding
29
Jacinta Abraham and John Staffurth
the mechanisms of resistance will help to predict the likely response to specific treatments and will help in the development of new agents targeted at endocrineresistant pathways at the molecular level. Nearly all patients with advanced breast cancer who initially respond to hormone therapy will ultimately become hormone refractory. The development of resistance to tamoxifen does not necessarily predict resistance to an AI; patients initially responding to tamoxifen have a 50% response rate to AIs.
De Novo resistance A strongly ER-positive tumour is likely to respond better to tamoxifen than one with a weakly positive ER status. However, ER status alone does not accurately predict for intrinsic resistance. Possible mechanisms of resistance include the following: r Some of the actions of oestrogens appear to be mediated through transcription factors such as activator protein 1 (AP-1). Enhanced activation of AP-1 transcription factors has been associated with tamoxifen resistance in both human breast cancers and xenografts. r Expression of ERβ has been seen to activate AP-1regulated genes when bound to tamoxifen. Increased expression of ERβ is implicated in tamoxifen resistance. r ERs located in the membrane can activate growthfactor-signalling pathways; this bidirectional crosstalk between signalling pathways may be active in endocrine resistance, for example, binding of oestrogen, and even tamoxifen in membrane ER can activate the epidermal growth factor family receptors (Milano et al., 2006).
Acquired resistance Peptide growth factor receptor pathways such as EGFR and HER-2 become selectively upregulated in breast cancer cells that acquire resistance to tamoxifen during prolonged exposure, and they offer an alternative mechanism that stimulates cell growth. Enhanced expression of EGFR and subsequent downstream MAPK activation have been found in MCF-7 breast cancer cells that become resistant to tamoxifen over time. In addition, the effect of low levels of oestrogen stimulation induced by tamoxifen and AIs may result in sensitisation of the ER. 30
The agonist activity of tamoxifen is also an important mechanism of resistance; that is, tamoxifen may become an ER agonist in breast cancer cells. This concept is supported by the fact that withdrawal responses may occur, albeit rarely.
Role of the progesterone receptor in hormone resistance In ER-positive disease, a PgR-negative status is linked to a reduced chance of response to tamoxifen but not necessarily to AI resistance. Tamoxifen resistance may be linked to growth factor receptor pathways such as HER-2. Excessive growth factor signalling or overexpression of HER-2 downregulates expression of the PgR gene. Crosstalk may also occur between the ER and growth factor receptor pathways such as HER-2, and so oestrogen or tamoxifen might stimulate these alternative pathways. Support for this hypothesis comes from the ATAC study in which the benefit of time to recurrence seen with anastrozole compared with tamoxifen was greater in the ER-positive/PgR-negative subgroup compared to the ER-positive/PgR-positive subgroup (Dowsett et al., 2005). This observation was not made in a subgroup analysis in the BIG 1–98 trial, and it remains a continuing area of research.
Breast cancer: areas of current interest Addition of fulvestrant in patients who relapse on AI The rationale is that patients who relapse on AIs may have tumours that are hypersensitive to the low levels of oestrogen observed in AI treatment. Fulvestrant, a downregulator of the ER, may be more effective in an environment of ‘low’ compared to ‘normal’ physiological postmenopausal oestradiol levels. This is being studied in the SOFEA trial.
Raloxifene Raloxifene is a nearly pure oestrogen antagonist and it has no stimulatory effects on the endometrium. Initial findings from the STAR breast cancer prevention study have shown that raloxifene is as effective as tamoxifen in reducing the incidence of invasive breast cancer in postmenopausal women and is associated with fewer thromboembolic events and endometrial cancers compared to tamoxifen.
Hormones in cancer
Table 3.4. Hormonal therapies currently used in prostate cancer Drug
Type
Dose/route
Mode of action
Goserelin
LHRH agonist
3.6 mg every 28 days or
Reduces pituitary production
10.8 mg every 3 months
of LH and FSH
s.c. Leuprorelin
LHRH agonist
3.75 mg every 28 days or 11.25 mg every 3 months
Reduces pituitary production of LH and FSH
s.c./i.m. Bicalutamide
Non-steroidal antiandrogen
50–150 mg daily p.o.
Competitive AR inhibition
Flutamide
Non-steroidal antiandrogen
250 mg t.d.s. p.o.
Competitive AR inhibition
Cyproterone acetate
Steroidal antiandrogen
100 mg t.d.s. p.o.
Reduces pituitary production
Diethylstilboestrol
Oestrogen
1–5 mg daily
Reduces pituitary production
Prednisolone
Corticosteroid
5–10 mg daily
Adrenal suppression
Dexamethasone
Corticosteroid
0.5–2 mg daily
Adrenal suppression
of LH and FSH of LH and FSH
AR = androgen receptor; FSH = follicle stimulating hormone; LH = luteinising hormone; LHRH = luteinising hormone-releasing hormone.
Non-hormonal approaches: signal transduction inhibitors EGFR expression in preclinical breast cancer models is associated with increased proliferation, resistance to apoptosis, and a poor outcome. There is interest in whether EGFR inhibitors may overcome hormone resistance.
Prostate cancer: hormonal therapies The hormonal therapies currently used in prostate cancer are summarised in Table 3.4.
Androgen deprivation therapy Androgen deprivation therapy (ADT) can be achieved by either castration or single agent bicalutamide (150 mg daily). Castration can be achieved surgically (subcapsular orchidectomy) or chemically with LHRH agonists (LHRHa), and it produces serum testosterone levels of less than 50 ng/ml. Alternatives in emergency situations, such as impending spinal cord compression, include diethylstilboestrol (DES; 3 to 5 mg daily), cyproterone acetate or ketoconazole. In advanced or recurrent prostate cancer, ADT produces responses in approximately 85% of patients, often with dramatic reversal of the clinical picture.
The median duration of response is approximately 18 months. There is evidence that the PSA doubling time before and the PSA nadir following ADT are of prognostic significance (D’Amico et al., 2006; Rodrigues et al., 2006). Eventually, and inevitably, patients will relapse. Trials investigating adjuvant therapies that might prolong the time to relapse have generally shown no effect or only minimal benefit. The ongoing MRC STAMPEDE study is investigating the benefits of zoledronic acid, docetaxel, and celecoxib as single agents and in various combinations in this group of patients.
LHRH agonists: goserelin and leuprorelin Mode of action LHRHa are synthetic analogues of the decapeptide LHRH which have increased affinity for the LHRH receptor and reduced susceptibility to protease degradation, which results in increased binding to the receptor. After an initial surge in LH and FSH and a rise in testosterone (unless temporary antiandrogens are used), downregulation of LHRH receptors and desensitisation of the gonadotroph to the normal pulses of LHRH occur. LH and FSH fall and castrate testosterone levels are reached in 10 to 20 days. Reduced serum testosterone leads to reduced expression of androgen-regulated gene products. Some cells undergo apoptosis (androgen dependent), some 31
Jacinta Abraham and John Staffurth
undergo cell-cycle arrest (androgen sensitive), and some are unaffected (androgen insensitive). Oral administration is ineffective because LHRHa are rapidly cleared by the liver, but subcutaneous or intramuscular delivery gives 94% bioavailability. LHRHa can be given as monthly or three-monthly depot injections. The temporary rise in testosterone (flare) should be prevented by giving three weeks of antiandrogens (bicalutamide 50 to 150 mg daily, flutamide 250 mg t.d.s. or cyproterone acetate 100 mg t.d.s.), starting at least one week before the first LHRHa injection.
Hormone therapy as an adjunct to radical radiotherapy Several studies have shown a benefit of the addi-
vanced or asymptomatic metastatic disease are incurable, and so the balance between quality and quantity of life is very important. They may not need immediate hormonal therapy. ADT has well-documented acute and late side effects that may be avoidable for a considerable duration in certain clinical situations. Deferred therapy is an option for patients with locally advanced disease who are not having radical therapy or for those with asymptomatic early metastatic disease (e.g. painless rib metastases only; Anonymous, 1997). Patients should remain under close supervision and be made aware of the symptoms of progressive disease. A similar argument exists for the use of single-agent bicalutamide as an alternative to castration in certain situations, although it is not licensed for single-agent use in metastatic disease.
tion of hormone therapy to radical radiotherapy (Table 3.5). Hormone therapy can be used either neoadjuvantly and concurrently with the radiotherapy (neoadjuvant hormone therapy, NAHT) or as an adjuvant following radiotherapy (adjuvant hormone therapy, AHT). Studies looking at these individual strategies have mainly included patients with intermediate- or high-risk disease and have shown a benefit to either approach. (See Chapter 9 for more information.) Neoadjuvant hormone therapy causes cytoreduction and may radiosensitise cells. Maximum cytoreduction occurs at approximately three months. In patients with bulky tumours, NAHT reduces the volume of rectum irradiated (Zelefsky and Harrison, 1997). Radiosensitisation would be expected to improve local control, which was shown only in RTOG 8610 in low-grade locally advanced tumours. The optimum duration of NAHT is unknown. TROG 96.01 has shown improved outcome in locally advanced tumours over six months compared to three months. NAHT has not been shown to benefit patients with small-volume good-risk localised disease. Three years of AHT improves five-year overall survival in patients with locally advanced disease who are treated with prostate and pelvic radiotherapy (Bolla et al., 2002). With regard to combined studies, the L101 and L200 studies showed no additional benefit to adding AHT in patients who had also received NAHT and prostate radiotherapy. RTOG 9202 investigated the added benefit of two years of AHT in addition to NAHT and prostate and pelvic radiotherapy in patients with locally advanced disease. Improved overall survival has thus far only been seen in high-grade tumours (Hanks et al., 2003), which suggests that men with locally advanced prostate cancer and a Gleason score of greater than or equal to 8 would benefit from a combined course of hormonal treatment.
Intermittent therapy in advanced disease The ratio-
Adjuvant hormonal therapy following surgery Two
nale for intermittent therapy is to allow androgendependent cells to repopulate the tumour, thereby delaying the development of androgen-resistant clones. It may also allow the patient some time free from acute side effects and may delay the development of late side effects of ADT. Two large studies are investigating the role of intermittent hormone therapy in patients with rising PSA (Intercontinental) and patients with metastatic disease (EORTC 30985).
studies have investigated the role of AHT following radical prostatectomy. One study showed no benefit in terms of overall survival for the addition of flutamide in patients with locally advanced, node-negative disease. At a median of 11.9 years, the second study showed improved overall survival with immediate ADT in patients with involved pelvic lymph nodes (HR 1.84; p = 0.04). Progression-free and cause-specific survival were also improved (Messing et al., 2006).
Clinical use Advanced disease LHRHa have been used for more than 15 years for the management of metastatic prostate cancer, as the medical alternative to orchidectomy. Trials show equivalent response rates and survival, although they have been too small to show true equivalence. However, LHRHa are reversible, and they are the preferred management approach (see Chapter 19).
Deferred therapy in locally advanced or asymptomatic metastatic disease Patients with locally ad-
32
Patient group
Locally advanced
T2–3
TROG 96.01
L101
T3–4 or N1 (some
RTOG 8531
T1–4, N0–1, M0
Early Prostate
T2–3
T2c-4
L200
RTOG 9202 ± 2 yrs of AHT
starting 2 months pre-RT
PPRT + 4 months of MAB
starting 3 months pre-RT
PrRT + 5 or 10 months of MAB
pre-RT
MAB starting 3 months
PrRT + 0, 3 or 10 months of
150 mg daily for 2–5 yrs
1554
325
161
1370
PrRT ± adjuvant bicalutamide
AHT or LHRHa at relapse
401
945
277
PPRT + concurrent and 3 yrs of
relapse) LHRHa
week of RT) or delayed (at
PPRT + immediate (in last
orchiectomy + PrRT
Orchiectomy vs PrRT vs
starting 3 months pre-RT
PrRT ± 3 months MAB 161
5.8 yrs
3.7 yrs
5 yrs
5.3 yrs
5.5 yrs
7.3 yrs
5 yrs
p < 0.0001)
AHT better (HR 0.75,
NAHT/AHT
No difference NAHT and
NAHT/AHT
No difference NAHT and
p = 0.002)
AHT better (HR 0.72,
p = 0.0001)
AHT better (HR 0.43,
(HR 0.77, p < 0.0001)
AHT improved 10 yrs bNED
(HR 0.59, p = 0.009)
Improved bNED for NAHT
(HR 0.66, p = 0.04)
OS advantage for high-grade tumours
unplanned subgroup analysis showed
AHT improved bNED, MDFS, and LDFS;
HT (HR 0.59 and 0.53, p = 0.009)
Both improved bNED compared to no
p = 0.003); no difference in OS
advanced disease (T3–4) (HR 0.58,
subgroup analysis for locally
Improved outcome significant on
62% (HR 0.59, p = 0.0002)
AHT improved CSS and OS; 5 yr OS 78 vs
p < 0.004)
with AHT; 10 yr OS 53 vs 38% (HR 0.76,
Improved LDFS, MDFS, CSS, and OS
orchiectomy
Delayed time to metastatic disease with
See section on combined studies, below
months
improved with 6 months NAHT vs 3
pre-RT
LDFS, MDFS, and CSS; each measure
3 and 6 months NAHT improved bNED,
low-grade tumours only
Reduced MDFS, bNED and CSS seen in
Other results/issues
p < 0.0001 for 6 months)
NAHT better (HR 0.65, p =
p = 0.004)
NAHT better at 8 yrs (HR 0.85,
Disease-free survival
0.0001 for 3 months; HR 0.56,
5.9 yrs
6.7 yrs
Median FU
RT + 0, 3, or 6 months of
818
456
n
MAB starting 2 or 5 months
Prostate and seminal vesicle
starting 2 months pre-RT
PPRT ± 4 months of MAB
Study design
versus; yrs = years. Adapted from Bolla et al., 1997; Denham et al., 2005; Fellows et al., 1992; Hanks et al., 2003; Laverdiere et al., 2004; Pilepich et al., 2001, 2005; Tyrrell et al., 2005.
patients; NAHT = neoadjuvant hormone therapy; OS = overall survival; PPRT = prostate and pelvic radiotherapy; PrRT = prostate radiotherapy; RT = radiotherapy; vs =
local disease-free survival; LHRH = luteinising hormone-releasing hormone; MAB = maximal androgen blockade; MDFS = metastatic disease-free survival; n = number of
AHT = adjuvant hormone therapy; bNED = biochemical disease-free survival; CSS = cause-specific survival; FU = follow-up; HR = hazard ratio; HT = hormone therapy; LDFS =
T2–3
L101
Combined studies
Cancer Trial
T1–4
EORTC 22863
pT3)
T2–4
MRC PR02
Adjuvant studies
Bulky T2–4 N0–1
RTOG 8610
Neoadjuvant studies
Trial
Table 3.5. Studies of neoadjuvant and/or adjuvant hormonal therapies in prostate cancer
Jacinta Abraham and John Staffurth
Adverse events The general symptoms of ADT are hot flushes, sweats, weight gain, and mood changes, especially emotional lability and depression. The depression can be profound and depressive psychoses have been reported. Patients become impotent and have decreased libido. They may notice tiredness, weakness, and reduced energy levels. Men with advanced prostate cancer have been shown to have reduced BMD at presentation compared with controls (40 versus 27%; Hussain et al., 2003). ADT also causes a 4 to 10% loss of BMD within the first year of therapy. There is an increased rate of all-site and hip fractures (13 versus 20% and 2 versus 4%, respectively) in prostate cancer patients treated with ADT compared to those treated without (Shahinian et al., 2005). Bisphosphonates can be used to prevent BMD loss in men who are starting ADT (Smith et al., 2003), although their effect on fracture rate has not been reported.
Non-steroidal antiandrogens: bicalutamide and flutamide Mode of action The modes of action of non-steroidal antiandrogens include the following: r Competitive inhibition of the AR prevents the binding of DHT in the prostate; binding results in a conformational shape change, altering the balance of coactivator and co-repressor complexes and reducing the effects of DHT. r Stimulation of the hypothalamus results in LH secretion, which leads to the production of testosterone, thus reducing many of the side effects associated with LHRHa, including impotence. r Bicalutamide can be used in combination with LHRHa (dose = 50 mg daily) for maximal androgen blockade (MAB) or as a single agent at 150 mg daily (see Chapter 19).
Clinical uses Single-agent bicalutamide 150 mg daily is less effective than surgical castration in patients with metastatic disease in terms of time to progression and overall survival; the outcome for patients with non-metastatic locally advanced disease was not statistically different (Iversen et al., 2000). MAB has been compared with medical and surgical castration alone in several trials and meta-analyses. 34
Overall, it seems that there is a small overall survival advantage of approximately 7 months (2 to 5% improvement in 5-year overall survival) when MAB is used (with bicalutamide as the preferred antiandrogen). The largest benefit may be for patients with minimal disease. However, advisory groups on both sides of the Atlantic have questioned the cost-effectiveness of MAB, and standard therapy remains LHRHa alone (Schmitt et al., 2000). The Early Prostate Cancer Trials have assessed the role of bicalutamide alone, or as an adjunct, in three parallel studies including 8113 patients with localised or locally advanced prostate cancer (McLeod et al., 2006). At a median follow-up of 7.4 years, in men with localised disease, there is no benefit from early or adjuvant bicalutamide, and there is a trend towards decreased survival in patients otherwise undergoing watchful waiting (HR 1.16 [95% CI 0.99 to 1.37]; p = 0.07). In patients with locally advanced disease who received radiotherapy (n = 305), bicalutamide significantly reduced the risk of disease progression by 44% (HR 0.56 [95% CI 0.40 to 0.78]; p < 0.001) and death by 35% (HR 0.65 [95% CI 0.44 to 0.93]; p = 0.03; See and Tyrrell, 2006). Bicalutamide is not currently licensed for use in localised disease.
Adverse events Approximately 70% of men suffer breast pain and gynaecomastia; 10% report hot flushes, asthenia, rash, and impotence; 6% report alopecia and weight gain (Wirth et al., 2004). The breast-related side effects can be most effectively managed by using adjuvant tamoxifen 10 to 20 mg daily, although the long-term effects on the cancer and the patient are not yet known. Breast bud irradiation and anastrazole are less-effective alternatives. Tamoxifen may also have a role in patients who have already developed gynaecomastia (Boccardo et al., 2005; Perdona et al., 2005). There are well-documented beneficial effects of bicalutamide on BMD (presumably due to increased testosterone and oestradiol levels), but unfortunately bicalutamide cannot be used to reverse the loss of BMD seen with previous use of LHRHa.
Antiandrogen withdrawal Withdrawing antiandrogen therapy has been reported to give response rates of up to 30% in patients who have been responding to antiandrogen therapy. This response
Hormones in cancer
is thought to be due to mutation in the AR, which results in stimulation by the antiandrogen rather than inhibition. Re-challenging with an alternative antiandrogen may result in further responses (e.g. bicalutamide following flutamide or nilutamide following bicalutamide).
Pharmacokinetics DES can be given orally and undergoes first-pass metabolism within the liver to form its active metabolites.
Clinical use
Cyproterone acetate is licensed for use in advanced prostate cancer and for suppression of tumour flare with initial LHRHa.
DES (3 to 5 mg daily) can be used as a single agent to achieve rapid castrate testosterone levels. It can also be used in combination with LHRHa at 1 mg for hormone-refractory disease. Transcutaneous application may bypass the first-pass metabolism and thus reduce the liver, cardiac, and thromboembolic side effects (being assessed in the ongoing MRC PATCH study).
Mode of action
Adverse events
The biochemical effects of steroidal antiandrogens are multiple, complex, and not well defined: r Antiandrogenic. r Reduced LHRH and LH release. r Reduced steroid synthesis leading directly to reduced testosterone levels. r Interference with the intracellular binding of DHT to the androgen receptor. r Progestational and glucocorticoid activities. The main clinical use of steroidal antiandrogens is to prevent LHRHa-induced tumour flare.
DES causes sodium retention with oedema and can precipitate congestive cardiac failure. It can cause thromboembolic disease. Hepatotoxicity can occur with jaundice, hepatitis and abnormal liver function tests. It should be given with aspirin 75 mg daily or full anticoagulation to prevent the thromboembolic disease. Gynaecomastia occurs, which is prevented with the use of preDES radiotherapy (7 to 10 Gy). Nausea and impotence also occur.
Steroidal antiandrogens: cyproterone acetate, megestrol acetate, and medroxyprogesterone acetate
Adverse events Fluid retention can occur which can precipitate congestive cardiac failure and cause ankle oedema. Steroidal antiandrogens are also associated with an increased rate of thromboembolic disease. Their hepatotoxicity includes jaundice, hepatitis, and abnormal liver function tests. Reduced libido and erectile dysfunction also occur.
Exogenous oestrogens: diethylstilboestrol and ethinyloestradiol Diethylstilboestrol (DES) is the only oestrogen that is used in standard clinical practice.
Glucocorticoids: prednisolone or dexamethasone Mode of action The mode of action of glucocorticoids involves suppression of adrenal androgen production: r Response rates to prednisolone/prednisone (5 to 20 mg daily) and hydrocortisone (30 mg daily) are in the order of 30% (PSA decrease of at least 50%), with a median duration of response of about 2 months. r Dexamethasone (0.5 to 2 mg daily) is highly active, with response rates of up to 80%, and up to 30% of patients may get an 80% reduction in PSA. The median time to progression is approximately 7 months. PSA responders have a median survival of approximately 15 months.
Mode of action The exact mode of action is not understood, but it is likely that DES functions in multiple ways to reduce androgenic activity: r Suppression of adrenal androgen production. r Reduced LH/FSH production from the pituitary. r Reduced testicular production of testosterone.
Adverse effects Typical glucocorticoid toxicities include proximal myopathy, cushingoid facies, thin skin with easy bruising, hypertension, mental changes including hyperactivity and confusion, and increased risk of osteoporosis. The lowest effective dose should be prescribed. 35
Jacinta Abraham and John Staffurth
Prostate: mechanisms of resistance The underlying mechanism for hormone resistance is still largely unknown. Because the adrenal glands produce 5 to 10% of circulating androgens, the majority of successful hormonal manoeuvres used in LHRHarefractory disease reduce adrenal androgen production. However, even with MAB, patients relapse. After ADT some cells undergo apoptosis (androgen dependent) and some undergo cell-cycle arrest (androgen sensitive) while others are unaffected (androgen insensitive). It is thought that before treatment with ADT there is a predominance of androgen-sensitive cells, but after ADT the less-dependent cells grow more and become the dominant clones. It has been postulated that the malignant cells are prostate cancer stem cells, whose progeny are primarily androgen sensitive in the presence of androgens. But in the absence of androgens or with low circulating levels of androgens, the progeny become increasingly androgen resistant. An alternative model is that androgen-sensitive cells initially enter cell-cycle arrest, but over time they develop molecular pathways that overcome the low/absent circulating androgens. These include the following: r AR mutation – increased activation by low levels of androgens or low-potency androgens. r Upregulated levels of AR mRNA and protein – this may also alter AR antagonists to give them some agonist function. r Continuously activated AR in the absence of androgens. r Upregulation of the ARE in the absence of activated AR (e.g. via upregulated transcription factors such as those in the EGFR pathway). r Upregulated survival pathways (i.e. reduced apoptosis, e.g. via BCL2 or survivin).
Prostate cancer: areas of current interest New LHRH antagonists: degarelix, abarelix, and cetrorelix Degarelix, abarelix, and cetrorelix lead to immediate competitive inhibition of LHRH receptors, reduced LH and FSH levels within 8 to 24 hours, and castrate testosterone levels within one week, in 75% of men. There is no testosterone surge (or flare), and there is a more rapid fall in PSA than occurs with LHRHa. They are given as 36
a subcutaneous injection every 4 weeks because they are not orally bioavailable. Studies are under way in both hormone-refractory and initial hormone therapy scenarios (Weckermann and Harzmann, 2004). Toxicity includes histamine-mediated allergic reactions. Presently these new LHRH antagonists are only licensed in the United States for patients unable or unwilling to be treated with LHRHa and willing to accept the risk of histamine-mediate allergic reactions. They are Food and Drug Administration (FDA) approved on a voluntary risk management programme.
Selective CYP1B1 inhibitors: abiraterone acetate Abiraterone acetate is a steroidal inhibitor of a key adrenal enzyme involved in the androgen production pathway (Attard et al., 2005). It has good bioavailability at doses greater than 200 mg, but there is a fairly large interpatient variability. Studies are underway, primarily in hormone-refractory prostate cancer. Toxicities include mild mood variations, flushing attacks, testicular discomfort, and headaches.
Non-hormonal approaches In patients with hormone-refractory disease, several approaches are being investigated, including signal transduction inhibitors, inhibition of upregulated antiapoptotic pathways, and immunotherapy. Active agents may be effective in earlier stages of disease.
REFERENCES Anonymous. (1997). Immediate versus deferred treatment for advanced prostatic cancer: initial results of the Medical Research Council Trial. The Medical Research Council Prostate Cancer Working Party Investigators Group. Br. J. Urol., 79, 235–46. Attard, G., Belldegrun, A. S. and de Bono, J. S. (2005). Selective blockade of androgenic steroid synthesis by novel lyase inhibitors as a therapeutic strategy for treating metastatic prostate cancer. B. J. U. Int., 96, 1241–6. Beatson, G. T. (1896). On the treatment of inoperable cases of carcinoma of the mamma. Suggestions for a new method of treatment with illustrative cases. Lancet, ii, 104–7. Boccardo, F., Rubagotti, A., Battaglia, M. et al. (2005). Evaluation of tamoxifen and anastrozole in the prevention of gynecomastia and breast pain induced by bicalutamide monotherapy of prostate cancer. J. Clin. Oncol., 23, 808–15. Bolla, M., Gonzalez, D., Warde, P. et al. (1997). Improved survival in patients with locally advanced prostate cancer treated with radiotherapy and goserelin. N. Engl. J. Med., 337, 295–300.
Hormones in cancer
Bolla, M., Collette, L., Blank, L. et al. (2002). Long-term results with immediate androgen suppression and external irradiation in patients with locally advanced prostate cancer (an EORTC study): a phase III randomised trial. Lancet, 360, 103–6. Coombes, R. C., Hall, E., Gibson, L. J. et al. (2004). A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. N. Engl. J. Med., 350, 1081–92. D’Amico, A. V., Kantoff, P., Loffredo, M. et al. (2006). Predictors of mortality after prostate-specific antigen failure. Int. J. Radiat. Oncol. Biol. Phys. 65, 656–60. Denham, J. W., Steigler, A. and Lamb, D. S. (2005). Short-term androgen deprivation and radiotherapy for locally advanced prostate cancer: results from the Trans-Tasman Radiation Oncology Group 96.01 randomised controlled trial. Lancet Oncol., 6, 841–50. Dowsett, M., Cuzick, J., Wale, C. et al. (2005). Retrospective analysis of time to recurrence in the ATAC trial according to hormone receptor status: an hypothesis-generating study. J. Clin. Oncol., 23, 7512–17. EBCTG. (2005). Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet, 365, 1687–717. Fellows, G. J., Clark, P. B., Beynon, L. L. et al. (1992). Treatment of advanced localised prostatic cancer by orchiectomy, radiotherapy, or combined treatment. A Medical Research Council Study. Urological Cancer Working Party – Subgroup on Prostatic Cancer. Br. J. Urol., 70, 304–9. Fisher, B., Costantino, J., Redmond, C. et al. (1989). A randomized clinical trial evaluating tamoxifen in the treatment of patients with node-negative breast cancer who have oestrogenreceptor-positive tumors. N. Engl. J. Med., 320, 479–84. Goss, P. E., Ingle, J. N., Martino, S. et al. (2005). Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: Updated findings from NCIC CTG MA.17. J. Nat. Cancer Inst., 97, 1262–71. Hanks, G. E., Pajak, T. F., Porter, A. et al. (2003). Phase III trial of long-term adjuvant androgen deprivation after neoadjuvant hormonal cytoreduction and radiotherapy in locally advanced carcinoma of the prostate: the Radiation Therapy Oncology Group Protocol 92–02. J. Clin. Oncol., 21, 3972–8. Hopp, T. A., Weiss, H. L., Hilsenbeck, S. G. et al. (2004). Breast cancer patients with progesterone receptor PR-A-rich tumors have poorer disease-free survival states. Clin. Cancer Res., 10, 2751–60. Howell, A., Cuzick, J., Baum, M. et al. (2005). Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years’ adjuvant treatment for breast cancer. Lancet, 365, 60–2. Huggins, C. and Bergenstal, D. M. (1952). Inhibition of human mammary and prostatic cancers by adrenalectomy. Cancer Res., 12, 134–41. Huggins, C. and Hodges, C. V. (1941). Studies on prostate cancer. The effects of castration, of oestrogen and of androgen injection on serum phosphatases in carcinoma of the prostate. Cancer Res., 1, 293–7. Hussain, S. A., Weston, R., Stephenson, R. N. et al. (2003). Immediate dual energy X-ray absorptiometry reveals a high incidence of osteoporosis in patients with advanced prostate
cancer before hormonal manipulation. B. J. U. Int., 92, 690–4. Iversen, P., Tyrrell, C. J., Kaisary, A. V. et al. (2000). Bicalutamide monotherapy compared with castration in patients with nonmetastatic locally advanced prostate cancer: 6.3 years of followup. J. Urol., 164, 1579–82. Jakesz, R., Jonat, W., Gnant, M. et al. (2005). Switching of post-menopausal women with endocrine-responsive early breast cancer to anastrozole after 2 years’ adjuvant tamoxifen: combined results of ABCSG trial 8 and ARNO 95 trial. Lancet, 366, 455–62. Laverdiere, J., Nabid, A., De Bedoya, L. D. et al. (2004). The efficacy and sequencing of a short course of androgen suppression on freedom from biochemical failure when administered with radiation therapy for T2-T3 prostate cancer. J. Urol., 171, 1137–40. McLeod, D. G., Iversen, P., See, W. A. et al. (2006). Bicalutamide 150 mg plus standard care vs standard care alone for early prostate cancer. B. J. U. Int., 97, 247–54. Messing, E. M., Manola, J., Yao, J. et al. (2006). Immediate versus deferred androgen deprivation treatment in patients with node-positive prostate cancer after radical prostatectomy and pelvic lymphadenectomy. Lancet Oncol., 7, 472–9. Milano, A., Dal Lago, L., Sotiriou, C. et al. (2006). What clinicians need to know about antioestrogen resistance in breast cancer therapy. Eur. J. Cancer, 42, 2692–705. Miller, W. R., Anderson, T. J., Dixon, J. M. et al. (2006). Oestrogen receptor beta and neoadjuvant therapy with tamoxifen: prediction of response and effects of treatment. Br. J. Cancer, 94, 1333–8. Noble, R. L. (1980). Production of Nb rat carcinoma of the dorsal prostate and response of estrogen-dependent transplants to sex hormones and tamoxifen. Cancer Res., 40, 3547–50. Perdona, S., Autorino, R., De Placido, S. et al. (2005). Efficacy of tamoxifen and radiotherapy for prevention and treatment of gynaecomastia and breast pain caused by bicalutamide in prostate cancer: a randomised controlled trial. Lancet Oncol., 6, 295–300. Pilepich, M. V., Winter, K., John, M. J. et al. (2001). Phase III radiation therapy oncology group (RTOG) trial 86–10 of androgen deprivation adjuvant to definitive radiotherapy in locally advanced carcinoma of the prostate. Int. J. Radiat. Oncol. Biol. Phys., 50,1243–52. Pilepich, M. V., Winter, K., Lawton, C. A. et al. (2005). Androgen suppression adjuvant to definitive radiotherapy in prostate carcinoma – long-term results of phase III RTOG 85–31. Int. J. Radiat. Oncol. Biol. Phys., 61, 1285–90. Rodrigues, N. A., Chen, M. H., Catalona, W. J. et al. (2006). Predictors of mortality after androgen-deprivation therapy in patients with rapidly rising prostate-specific antigen levels after local therapy for prostate cancer. Cancer, 107, 514–20. Schmitt, B., Bennett, C., Seidenfeld, J. et al. (2000). Maximal androgen blockade for advanced prostate cancer. Cochrane Database Syst. Rev., 2, CD001526. See, W. A. and Tyrrell C. J. (2006). The addition of bicalutamide 150 mg to radiotherapy significantly improves overall survival in men with locally advanced prostate cancer. J. Cancer Res. Clin. Oncol. 132 (Suppl. 13), 7–16.
37
Jacinta Abraham and John Staffurth
Shahinian, V. B., Kuo, Y. F., Freeman, J. L. et al. (2005). Risk of fracture after androgen deprivation for prostate cancer. N. Engl. J. Med., 352, 154–64. Smith, I. E., Dowsett, M., Yap, Y. S. et al. (2006). Adjuvant aromatase inhibitors for early breast cancer after chemotherapy-induced amenorrhoea: caution and suggested guidelines. J. Clin. Oncol., 24, 2444–7. Smith, M. R., Eastham, J., Gleason, D. M. et al. (2003). Randomized controlled trial of zoledronic acid to prevent bone loss in men receiving androgen deprivation therapy for nonmetastatic prostate cancer. J. Urol., 169, 2008–12. Thurlimann, B. on behalf of the BIG1–98 Collaborative Group. (2005). A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. N. Engl. J. Med., 353, 2747–57. Tyrrell, C. J., Payne, H., See, W. A. et al. (2005). Bicalutamide (‘Casodex’) 150 mg as adjuvant to radiotherapy in patients with localised or locally advanced prostate cancer: results
38
from the randomised Early Prostate Cancer Programme. Radiother. Oncol., 76, 4–10. Weckermann, D. and Harzmann R. (2004). Hormone therapy in prostate cancer: LHRH antagonists versus LHRH analogues. Eur. J. Urol. 46, 279–83; discussion 283–4. Wirth, M. P., See, W. A., McLeod, D. G. et al. (2004). Bicalutamide 150 mg in addition to standard care in patients with localized or locally advanced prostate cancer: results from the second analysis of the early prostate cancer program at median followup of 5.4 years. J. Urol., 172 (5 Pt. 1) 1865–70. Zelefsky, M. J. and Harrison A. (1997). Neoadjuvant androgen ablation prior to radiotherapy for prostate cancer: reducing the potential morbidity of therapy. Urology, 49 (Suppl. 3A), 38–45. Zondek, B. (1940). Effect of prolonged administration of oestrogen on uterus and anterior pituitary of human beings. J. A. M. A., 114, 1850–54.
4
RADIOTHERAPY PLANNING Andrew Tyler and Louise Hanna
Introduction It is important to understand the basic techniques of radiotherapy planning because these will help when developing complex plans. These techniques, which are taught at the FRCR Part I level (Royal College of Radiologists), are used by the treatment centre’s Physics Department for checking the validity of calculations before they are applied to patients, and anyone interpreting plans will need to be familiar with them to know whether it is worth adjusting treatment plans during review. There are several useful reviews of radiotherapy physics in the literature. One by Shiu and Mellenberg (2001) includes sections on isodose planning. Another, by Purdy (2000), provides a perspective on future directions in 3D treatment planning. Radiotherapy is a rapidly developing field and it is important to ensure that new methods satisfy safety and effectiveness requirements before being adopted as routine treatments. This chapter will deal with the general principles of developing isodose plans that are suitable for treatment and will use specific examples to highlight particular points. It will focus mainly on external beam radiotherapy with megavoltage photons; there will be shorter sections on the use of electrons and brachytherapy. Some aspects of radiotherapy planning using lower energy (kilovoltage) photons will be discussed in Chapter 33 (see p. 382).
Treatment planning overview Radiotherapy planning can be divided into stages as follows: r Patient preparation, position and immobilisation. r Localisation method (e.g. orthogonal films, CT scanning, and image co-registration). r Definition of target volumes and organs at risk. r Radiotherapy technique, including beam arrangements, beam energy, size and shape, weighting, wedges, and production of isodose plan.
r Prescription, including number of phases, dose, energy, and fractionation.
r Verification (i.e. checking the geometrical set-up of the treatment). This can take place before treatment in the simulator or CT simulator, and / or during treatment using portal imaging. Quality assurance (QA) is essential and it is of paramount importance for each member of the planning team to be familiar with the overall process.
Patient position and immobilisation For radiotherapy to be effective it must be delivered accurately, and it is important to understand the methods of immobilisation and the inevitable uncertainties in the delivery of treatment. Patients must be positioned to allow the optimum delivery of the radiotherapy while maintaining comfort. Variables include whether the patient is prone or supine, and the position of the limbs or the neck. Radiotherapy treatments that require unusual treatment positions include total-body irradiation, when the patient may lie in the lateral position with the arms in front of the chest to provide lung compensation, and total skin electron treatment, when the patient may stand up, adopting the stance of a cross-country skier (one arm and the opposite leg forward, the other arm and leg back). A wide variety of different immobilisation devices exists. Which is most appropriate depends on the position of the tumour, the need to protect adjacent structures, and the treatment intent. The following are some commonly used devices: r Simple head, knee or ankle supports which help to stabilise the patient. For patients who do not have very mobile superficial tissues, treatments can be delivered to within ± 5 mm, providing that a careful set-up, suitable reference points, and calibrated couch movements are all used. r Chest board with cross bar to fix the arms above the head (e.g. for breast treatments). 39
Andrew Tyler and Louise Hanna
r Vacuum bags offer customised support (e.g. for limbs when it is necessary to support the limb in a position that allows access for beams). r Vacuum-formed shells, moulded thermoplastic shells (heated in warm water) or Perspex® shells formed from individualised plaster moulds. By using wellfitted vacuum-formed shells for head and neck treatments, inaccuracies in set up can be reduced to within 3 mm. However, vacuum-formed shells for other anatomical sites do not usually give this reliability. r Complete fixation devices such as stereotactic frames with mouth bites give set up accuracy to within about 1 mm. These also require high-level QA on treatment machines. It is important to be certain that the immobilisation device is effective. For example, one study looked at the use of a customised immobilisation system (Vacfix® cushion) for the treatment of the prostate, and the authors concluded that the treatment accuracy was not improved when compared with the conventional technique (using standard ankle stocks; Nutting et al., 2000b).
Imaging the patient Imaging the patient is essential to delineate gross tumour volume (GTV), to design the clinical target volume (CTV) and planning target volume (PTV), and to outline critical structures in accordance with ICRU reports 50 and 62 (ICRU, 1993, 1999).
CT simulation A CT simulator is one of the mainstays of 3D conformal planning. It is typically a wide-bore CT scanner with virtual simulation software (VSIM) and radiotherapy accessories (e.g. flat couch, mounts for immobilisation devices, laser positioning, etc.). It provides precise imaging and positioning information for the treatment planning process. Automatic electronic data transfer to a treatment planning system is required to enable the treatment calculation to be performed. The wide bore, although theoretically providing a slight reduction in image quality, allows patients to be scanned in the treatment position and for reference marks, from which treatments can be set up, to be identified securely. CT scans are taken over a few seconds, whereas a treatment fraction can take longer to deliver. Therefore, normal patient motion, such as breathing, may be “frozen” at any point in the cycle on the CT scan and the expansion of the CTV to PTV must take this into account. 40
The planning CT scan is usually taken using a slice interval of 1.5 to 5 mm. The narrower the slice width, the better will be the resolution in the digital reconstructed radiograph (DRR). It is important to include the proposed target volume and the organs at risk in the scan to allow dose volume histograms (DVHs) to be calculated. Administration of intravenous contrast helps to distinguish tumour from normal tissues. One example of this is during planning for lung cancer. There has been debate whether the use of contrast alters the dosimetry, but Lees et al. (2005) have found it makes little difference in dose computation. For some tumour sites, such as the brain, MRI scans are better than CT scans for delineating the GTV, but there is software by which the ‘diagnostic’ MRI scan can be co-registered with the planning CT scan to aid in the planning process. MRI/CT co-registration is most accurate for tumours that are close to bony structures with characteristic features, for example near the skull base. It is less helpful for tumours that can move easily relative to bony landmarks or for those not close to bony features. Because of this, MRI/CT co-registration is of limited value for some sarcomas. It is important to make sure that not only the position and scaling but also the rotations have been faithfully applied. This is easier in dual-modality machines, such as PET/CT, in which the patient is scanned in the same position. Corrections for MRI distortions must also be validated carefully.
Nuclear medicine Although radiopharmaceutical imaging tends to be used for diagnostic purposes, it is increasingly being used to inform the treatment planning process, especially with the advent of PET/CT where the automatic co-registration of the images can be used as a direct input into the planning process. Applications currently in development include lung, lymphoma, and colorectal cancers.
‘Conventional’ radiotherapy planning Conventional radiotherapy planning has been largely superseded by CT planning. The target volumes are delineated on orthogonal X-ray images and then transferred onto an outline of the patient taken at one or more levels. Alternatively, for some treatments using simple beam arrangements such as a parallel opposed pair of fields, the field borders themselves are defined on a radiograph or simulator image. These borders represent the PTV with a small margin to take account of the beam penumbra. Tissue definition, particularly
Radiotherapy planning
for soft tissues, is not clear when using orthogonal films.
Defining the target volumes Standards The publications ICRU 50 and ICRU 62 (ICRU, 1993, 1999) will be referred to throughout this text. Anyone taking part in the planning process should be familiar with the terms defined within. As a brief recap, the following terms are used to ensure accurate treatment records are kept (ICRU, 1993, 1999): r Gross tumour volume (GTV) – the demonstrable tumour. r Clinical target volume (CTV) – the GTV and/or suspected subclinical tumour. r Planning target volume (PTV) – the CTV and a margin for uncertainties, which may be systematic or random, and are taken into account by r the internal margin (IM), which is added to account for involuntary changes in the organs that surround the CTV. r the set-up margin (SM), which accounts for the uncertainties and lack of reproducibility in setting up the patient day by day. r Treated volume – the volume enclosed by an isodose surface, selected and specified by the radiation oncologist as being appropriate to achieve the purpose of treatment (e.g. tumour eradication, palliation). r Irradiated volume – the tissue volume that receives a dose that is considered significant in relation to normal tissue tolerance. r Organs at risk (OARs) – critical normal structures the radiation sensitivity of which may significantly affect treatment planning and/or prescribed dose. r Planning organ at risk volume (PRV) – the organs at risk with a suitable margin to account for movements and uncertainties in set-up (see p. 42). Figure 4.1 shows tumour and target volumes as defined in ICRU 50 (ICRU, 1993). The target volume is delineated with all the relevant clinical information available including medical history, pretreatment diagnostic and staging investigation results, scan images, operation notes and pathology reports. The three volumes, GTV, CTV and PTV, can be defined, together with the OAR and PRV. There can be more than one OAR. Any reduction in the treatment margins spares more normal tissues and introduces an opportunity for considering dose escalation for the tumour. When mar-
GTV CTV PTV Treated volume Irradiated volume Figure 4.1. International Commission on Radiation Units volume definitions. GTV = gross tumour volume; CTV = clinical target volume; PTV = planning target volume. Redrawn with permission from ICRU, 1993: ICRU Report 50. Prescribing, Recording, and Reporting Photon Beam Therapy. Bethesda, Maryland: International Commission on Radiation Units and Measurements.
gins are smaller, it becomes more difficult to be sure that the tumour is being treated. Although the treatment machine delivers radiation with an accuracy supported by elaborate QA, there are many other uncertainties in the planning process, including our inability to define microscopic disease.
Delineating the GTV Delineating the GTV involves outlining the demonstrable tumour. However, considerable variation between GTVs has been observed among different radiation oncologists (Logue et al., 1998; Weiss and Hess, 2003). Discussion with the multidisciplinary team, dedicated radiotherapy planning meetings and involvement of a specialist radiologist can facilitate this process. For example, in the head and neck, the tumour volumes may shrink during radiotherapy.
Growth from GTV to CTV By definition, it is not possible to identify subclinical disease by imaging or clinical examination. Individual patient variation will add to the uncertainty (e.g. the internal mammary chain lymph nodes lie between 1 and 6 cm deep and up to 4 cm from the midline). Some organs are more difficult to delineate than others, but other imaging techniques such as MRI may be helpful. Guidelines have been produced for CTV definition in high-grade gliomas, head and neck, and lung tumours (Apisarnthanarax et al., 2006; Gregoire et al., 2000; Jansen et al., 2000; Senan et al., 2004). 41
Andrew Tyler and Louise Hanna
Forming the PTV The dose delivered to the CTV or OARs will not be the same as that shown on the plan if they are not in exactly the same place during all the treatments. Internal movements are significant in the thorax, abdomen, and pelvis. For example, the prostate and seminal vesicles can move with changes in the volume of the bladder or rectum, up to a centimetre each way anterior–posterior (A–P) and the CTV of tumours in the lung, particularly those nearest to the diaphragm, will move during breathing. r Changes in the position or size of the CTV (both during and between fractions) must be taken into account. These natural variations are inevitable, but the size of the variation can to some extent be estimated and accounted for. These form the IM. r Uncertainties of patient set-up, alignment of beams, and so forth also need to be taken into account; these make up the SM. r The combination of these two margins around the CTV forms the PTV (ICRU, 1999). r The uncertainties associated with these margins have been estimated (BIR, 2003) and methods of calculation have been introduced that ensure that 95% of the prescribed dose covers the CTV in 90% of cases. The BIR (2003) document uses the concepts of systematic and random errors, and, as part of the document, some key examples by clinical site provide excellent guidance. r Options such as image-guided radiation therapy (IGRT) and respiratory gating can be employed to reduce the required margins. Techniques to reduce tumour movement have been used in thorax treatments such as breath hold, active breathing control, or depressing the chest to reduce the movement. Lung tumour motion can be in the order of 2 cm during a normal breathing cycle, depending on the lobe of the lung affected (Seppenwoolde et al., 2002). r Automatic growth tools on the planning computer may help, but the mechanism should be understood before using them and they should always be used with care.
Delineating organs at risk OARs are normal structures with a high sensitivity to radiation and it is important that they are delineated on the plan. Some, such as the spinal cord, may be described as being ‘serial’ with a high ‘relative seriality.’ 42
Others such as the lung may be described as ‘parallel’ and have a low ‘relative seriality’ (ICRU, 1999). Within some organs (e.g. the heart), there may be a combination of serial and parallel structures. The function of serial organs may be seriously affected if even a small portion is irradiated above a tolerance dose. The effect of radiation on the function of parallel organs is more dependent on the volume irradiated. For a serial organ, the accuracy of treatment planning and delivery are important to ensure that tolerance is not exceeded. In a similar manner to the construction of the PTV, a margin must be added to make the planning organ at risk volume (PRV). In practice, this applies most commonly to the spinal cord.
Principles of the isodose plan: photons Basic beam data It is important that the concepts of basic beam data are understood. These are described briefly here, but in detail in other texts (e.g. Khan, 2003; Williams and Thwaites, 2000). These topics are all covered in part 1 of the FRCR course (RCR) and are discussed in ICRU 24 (ICRU, 1976). Beam divergence. The width of the radiation beam increases linearly with distance from the treatment head. Beam penumbra. At the edge of the radiation beam, the dose reduces over the distance of some millimetres. The penumbra is the distance between the 80 and 20% isodoses. In general it will be necessary to use a beam that is wider than the PTV to allow the 95% isodose to encompass the PTV. Percentage depth dose. The dose is expressed as a percentage of the maximum dose deposited by the beam. This maximum dose occurs at a depth dmax . With megavoltage beams the dose is deposited by secondary electrons, which travel primarily in a forward direction. As a result there is buildup of dose below the skin surface before dmax is reached. Buildup changes also occur at interfaces between tissues of differing densities. In some centres the tissue phantom ratio (TPR) or tissue maximum ratio (TMR) is used instead of the percentage depth dose because each is easier to use with isocentric calculations. Radial profile. The radial profile of a megavoltage beam changes with depth due to the differential hardening across the beam. This means that, if an asymmetric beam is created, the shape of the profile may also be noticeably asymmetric.
Radiotherapy planning
Changes with distance. As the distance from the treatment machine to the patient increases, the crosssectional area of the beam increases because of divergence. The inverse-square law causes the intensity of the radiation beam to decrease but the percentage depth doses below dmax to increase. Changes with field size. As the field size increases, the central axis receives more radiation per monitor unit because of increased scatter from the machine head and within the patient.
Beam arrangements Typical beam arrangements are as follow: r A single beam is used for superficial tumours such as skin cancers, or for tumours that are not at or near the midplane (e.g. spinal tumours having palliative treatment). r For two beams, r opposing beams are used for palliative treatments or for treatments in sites that have a relatively small separation, such as head and neck, breast and limbs; r non-opposing beams, for example, are set at right angles for the floor of mouth and maxillary antrum. Wedges are needed to reduce the inhomogeneities that arise in this situation. If the external contour of the patient is irregular in all directions around the central axis of the beam then external tissue compensators or ‘field-in-field’ techniques using multileaf collimators (MLCs) will probably be required. r Combinations of beams are usually used for radical treatments in sites such as the chest, pelvis and brain, where the use of multiple beams spares adjacent normal tissues. It is also important to understand beam weighting so that the relative proportion of dose delivered to the tumour from each beam can be adjusted (Williams and Thwaites, 2000).
exceeded. In prostate planning, an anterior beam is combined with two wedged lateral fields. Use of lateral fields, rather than posterior obliques, allows beam shaping to be more effective in reducing the dose to the rectum. In head and neck treatments, the exit beams should avoid such critical structures as the contralateral eye and spinal cord. Treatment plans can frequently be improved with the use of non-coplanar beams, which can improve the dose distribution in the PTV and reduce the dose to organs at risk. Non-coplanar beam arrangements can be achieved easily by linear accelerators, by rotating the table and the gantry, but it is important to ensure that the beam direction can be achieved (e.g. there is no collision between the treatment head and the couch or the patient). Noncoplanar beams are particularly useful in central nervous system (CNS) and head and neck treatments.
Isodose shapes for combinations of beams The directions of the beams will influence the shape of the treated volume. By tracing the edges of the beams, it is possible to identify the general shape of the high-dose volume, provided that the beams are well balanced and that there is a homogeneous dose distribution across the PTV. In Figure 4.2, the treated volume in the plane shown will be approximately the hexagon enclosed by the beam edges. It is important to identify potential regions of hot and cold dose. The global hot spot in a balanced multibeam treatment is likely to lie close to the centre of the treated volume.
Beam 2
Choosing beam directions The choice of beam directions will depend on the treatment intent and the locations of the PTV and OARs. For example, with radical oesophagus treatments, delivering the dose entirely by anterior and posterior parallelopposed fields would exceed spinal cord tolerance. Similarly, using oblique laterals instead of the posterior beam would cause an excessive dose to the lungs. A two-phase technique that combines both of these approaches can prevent tolerance doses from being
Beam 1
Beam 3
Figure 4.2. Beam edges creating the edge of the treated volume. In the scenario presented here there are three beams (beams 1, 2 and 3). The shaded area represents the volume that received the highest radiation dose, and it is formed by the edges of the beams.
43
Andrew Tyler and Louise Hanna
There are occasions when it may be difficult to prevent the dose to the PTV from dipping low. Examples include the following: r The superior and inferior (S–I) ends of a coplanar plan because of the lack of dose scattering into the volume from outside the beams. r When a wedged pair is used (e.g. with an anterior and a lateral field), it is common for the postero-medial aspect to be underdosed. It is possible to top up the beam with a small amount of dose from the contralateral side, but this obviously has implications for the dose to normal tissue on that side. Outside of the treated volume, the greatest doses for this type of plan will lie underneath the thin ends of the wedges. Use of a field-in-field technique, where regions of underdose are topped up with small beamlets, can help to reduce inhomogeneities of dose.
Beam modifications: shape, wedges, and tissue compensators Beam shaping Shaping the beam around the PTV spares more normal tissue. For example, randomised trials of prostate treatments have shown that the use of conformal radiotherapy reduces rectal toxicity, and this has led to studies of dose escalation with a view to improving local control (Dearnaley et al., 1999). Traditionally, low-meltingpoint alloy blocks have been shaped individually for each field for megavoltage treatments, but in many cases field shaping can be achieved with MLCs. Making low-melting-point alloy blocks requires time and handling of heavy and hazardous substances. The conformity achieved by MLCs is limited by the width of the leaves and their direction of movement. A potential problem is leakage between the abutted leaf ends; this can be reduced if the collimator jaws are positioned as far as possible over the leaf abutments (Klein et al., 1995).
Wedges Wedges are used in two situations where there is an unwanted dose gradient that is relatively uniform in one direction: r To act as a tissue compensator for a sloping external surface (see discussion of surface obliquity that follows). r To even out the dose when two or more beams intersect, in order to avoid hot and cold spots within the treated volume. 44
Surface obliquity Surface obliquity will cause isodoses to tilt in the same directions as the surface. If the obliquity occurs only in one direction, it can be corrected for by placing a wedge in the beam. If it occurs in both the S–I and A–P directions, then it may be necessary to consider using a custom-built tissue compensator. Tissue compensators are used most frequently in head and neck treatments. They can also be used in other situations such as 3D radiotherapy of the breast, although they may be needed less often in the future as intensity-modulated radiation therapy (IMRT) develops.
Calculation cycle Everybody involved in a patient calculation should know the whole calculation process and where each factor is applied in their own department, even if many of the factors are applied by the planning computer programs. The method of calculation varies from centre to centre, depending on how the factors are defined, and it is not possible to go through each variation in this chapter. Because of this, it is important to be familiar with the local method before starting any calculations involving patients. Factors (and their relative values) that need to be understood follow (Williams and Thwaites, 2000): r Percentage depth dose (PDD) or tissue maximum ratio (TMR). r Inverse-square law. r Wedge factor. r Field size (output) factor. r Accessory factors (e.g. trays).
Using a 3D planning system Production of the isodose plan With a 3D planning system, the target volumes and PRVs can be delineated individually on each slice of a crosssectional imaging study such as a CT scan. Software can allow co-registration of images from other modalities such as MRI or PET. After identifying the relevant volumes, it is then possible to simulate the positions of treatment beams and shielding and to use algorithms to calculate and display the resulting isodose curves. The respective doses to the PTV and PRVs can be displayed in the form of a DVH, which is a line graph showing the proportion of the volume that receives at least a specified dose of radiation. DVHs are a convenient
Radiotherapy planning
way of comparing different planning solutions for the same tumour in order to select the best dose distribution. Data from the planning process can be used to create a ‘beam’s eye view’ (BEV) image of the treatment fields, which can be superimposed on the digital reconstructed radiograph and used for verification purposes.
Selection of the ICRU reference point The recommendation for reporting doses is based on the ICRU reference point. According to ICRU 50 (ICRU, 1993), the reference point is selected according to the following criteria: r The point should be clinically relevant and representative of the dose throughout the PTV. r It should be easy to define in a clear and unambiguous way. r It should be selected where the dose can be accurately determined. r It should be selected in a region where there is no steep dose gradient. In practice the ICRU reference point is often at the isocentre or at the intersection of beams, but sometimes this is not possible, in which case it should be selected to be in a place where dose specification is considered to be meaningful.
Inhomogeneities of density When organs are of different densities, there will be changes in the penetration of the beam. For example, the dose in a central organ, such as the oesophagus, from a beam passing through an aerated lung can increase by as much as 15% when a correction is applied. This is because in a well-aerated lung, the extra penetration of the beam is usually slightly greater than 2% per centimetre of the lung traversed. In contrast, bones tend to have little effect on the dose; dense bone tends to lie in thin layers, and the core of long bones is almost unit density. Artificial hips are very high density and they can create particular problems during planning. In addition to the artefacts seen on CT imaging, the planning algorithm may not be able to correct properly for the veryhigh-density material. It is best to avoid passing beams through metal objects if possible. Treatment planning systems will make corrections for changes of density pixel by pixel, by using the CT values. Nearly all planning systems will estimate these effects, but some of the so-called 3D planning systems
still use 1D corrections, which means that only the primary beam is corrected. This can lead to uncertainties in dose when adjacent structures have different densities, because changes from scatter contributions will be significant. Many systems are using a 2D correction which does reduce some of this uncertainty, but routine clinical use of a full 3D correction as given by Monte Carlo simulation is still some way off because of the very large computing power required.
Doses to organs at risk (OARs) Outside the PTV, significant volumes of normal tissue are irradiated to a high dose, and they would benefit from receiving a smaller dose. The OAR dose often limits the highest dose that can be delivered to the tumour. In general, it is best to try to avoid letting any part of the beam pass through an OAR, and skilful use of the beam edge and beam direction can often ensure that the beam passes through the PTV but avoids an OAR. However, it is not always necessary to avoid an OAR completely. For example, in the thorax it may be appropriate to pass a posterior beam through the spinal cord to reduce lung dose and meet the appropriate DVH constraint. If one or more beams exit through an OAR, it is important to check the tolerance doses carefully. In addition to choosing appropriate beam directions, it is also possible to use wedges to reduce the doses in sensitive structures. For example, in Figure 4.3, a steep wedge angle on the lateral beam allows extra dose to be given from the anterior beam, and this can be used to spare a structure such as the contralateral eye or brain, but obviously at a cost of dose to the deeper posterior structures which could include the brain stem.
Matching adjacent beams in complex treatments Sometimes beams need to be adjacent to each other, for example, in combined breast and supraclavicular fossa radiotherapy, or craniospinal axis treatments. When two diverging beams are adjacent to each other, there will be an overlap and a potential gap, depending on the depth at which the beams meet. Various methods have been used to minimise any variations in dose that occur when beams are matched together. However, care must be taken because, although it is possible to produce what looks like a perfect plan with perfect junctions between the beams, patients can move between treatments even when they are in an immobilisation shell. 45
Andrew Tyler and Louise Hanna
Anterior beam
immobilisation and the use of accurate and stable marks for beam setup. Otherwise, movements may require online megavoltage portal imaging or use of a kilovoltage CT imager attached to the linac gantry. Ultrasound-based guidance can also been used for prostate treatments.
WA = 15°
Lateral beam WA = 60°
Validity of the setup
Isodose (%) 95 70 50 40 30
Figure 4.3. Drawing dose away from sensitive structures using wedges. A steep wedge angle on the lateral beam allows extra dose to be given from the anterior beam and can be used to spare a structure such as the contralateral eye or brain, but obviously at a cost of dose to the deeper structures posterior to the target volume, which could include the brain stem. WA = wedge angle.
For craniospinal treatments, the junction can be moved twice during treatment to spread the effect of any uncertainty in positioning. Other ways of minimising dose variations at junctions include half-beam blocking or couch rotations, both of which can compensate for the effect of beam divergence, but the possibility of patient movement during treatment must still be considered. When treating neck nodes, it is sometimes necessary to have an electron field next to a photon field, which can pose even more difficulties because the shapes of the isodoses are completely different. One solution has been to increase the number of photon fields with the intention of trying to avoid using the electron fields. This type of treatment is a prime case for complex IMRT.
Verification Verification of the planned treatment The treatment that is actually delivered will differ from that which has been planned because of setup errors and internal movements during and between fractions. If possible, it is best to minimise these effects by careful 46
Indexed positioning, using identical simulation and treatment couches, has been developed to aid in quick and repeatable setups. Although portal imaging helps to verify the setup in relation to bony landmarks, the organ of interest may not show up on the image, and internal organs may move during the course of treatment. A prostate gland can move more than 5 mm in 7 minutes; therefore, the position of the prostate can be significantly different from the time of scan to the time of treatment (Padhani et al., 1999). Implanted markers, such as gold grains (e.g. in the prostate) which can be monitored on the portal image, have been shown to improve the accuracy of treatment setup (Shimizu et al., 2000). The treatment position can be verified before treatment by using a simulator or CT simulator, and/or during treatment by using portal imaging. By taking serial images it is possible to judge whether a setup error is systematic or random. Correction of setup errors by moving treatment fields should only be carried out for systematic errors. Techniques such as adjusting the setup at the time of the radiotherapy fraction using information from a CT scanner attached to the gantry will become increasingly important in overcoming these problems. A useful review can be found in the literature (Langen and Jones, 2001). Hornick et al. (1998) have modified treatment couches to automatically compensate for the variations found in rotational and tilt orientations.
Verification of treatment dose If the CTV does not receive the intended dose then the risk of recurrence is likely to increase. For example, as beam conformity increases, there may be a greater risk of a recurrence at the edge of the treatment field. It will become increasingly important to identify the position of recurrence within the PTV in conformal radiotherapy. At present, thermoluminescence dosimetry (TLD) is the only practical method of in vivo dosimetry but it is usually only used at the skin surface. Work is
Radiotherapy planning
currently under way to evaluate electronic portal imaging devices to measure exit doses, which may provide some verification of the dose received from each portal.
Intensity-modulated radiation therapy (IMRT) IMRT involves modulating the intensity profile of the radiation beams to improve spatial agreement (conformity) between the resulting dose distribution and the tumour target volume (Bortfeld, 2006). Typically multiple treatment fields are used. In fact, the idea of beam modulation is not new: tissue compensators and wedges have been used for many years. IMRT delivers the variations of intensity across the beam either in ‘step-and-shoot’ mode, in which segments are delivered before moving the MLCs to the next position, or ‘dynamic’ mode, in which the MLC shape is continuously adjusted during the beam portal delivery. In IMRT there is an extra step in the treatment planning process. The idealised plan is passed to a sequencer which sets up practical MLC positions, depending on whether the treatment will be delivered in step-and-shoot or dynamic mode or by tomotherapy. This sequence is then re-planned to show the treated dose. In step-and-shoot mode, the intensity-modulated beam is delivered as a series of discrete fields, each with a small number of monitor units. The dynamic or slidingwindow method also achieves intensity modulation with MLCs, but here the leaves move across the field continuously during the irradiation. Serial tomotherapy involves a linear accelerator beam that moves around a patient and delivers the treatment in narrow ‘slices.’ In serial tomotherapy, the couch is moved in between treatment slices, whereas in helical tomotherapy the couch moves continuously during treatment and the volume is treated in a single spiral (Beavis, 2004; Hong et al., 2005). Because it achieves greater dose homogeneity and conformality, IMRT lends itself to the treatment of irregularly shaped PTVs, such as concave volumes. IMRT has the potential to improve tumour control and reduce normal tissue toxicity. Most experience to date has been with head and neck and prostate radiotherapy, but IMRT is also used in other sites, including CNS, lung, gynaecological, breast and gastrointestinal treatments (Coles et al., 2005a, 2005b; Guerrero Urbano and Nutting, 2004a, 2004b).
IMRT does tend to cause an increase in the low-dose bath of radiation in adjacent tissues, due to the long beam-on times, and this should be considered in the treatment decision. Useful reviews have been written (McNair et al., 2003; Nutting et al., 2000a); the review by Nutting et al. includes clinical results. Modern treatments have been recognized as being more sensitive to geometric uncertainties than conventional treatments because of the high dose gradients involved (Purdy, 2002).
Inverse planning For inverse planning, the radiation oncologist specifies the required dose distribution in the PVT and OAR. A computer algorithm is then used to calculate the beam modulations that will best fulfil the requirement.
Forward planning Forward planning is a simpler form of IMRT planning. The beam intensity is modulated by adding smaller beams with different weightings to a conformal therapy beam, which can be achieved with standard treatment planning software, and treatment can be performed on any machine without the need for IMRT sequencing software.
Quality assurance of treatment planning systems Reducing treatment errors To treat patients safely, technical quality is essential but not enough: QA of the clinical process is also required. Systems such as ‘Quality Assurance in Radiotherapy’ (QART) should be put in place to ensure overall control of the activities in the radiotherapy department (Kehoe and Rugg, 1999). The new management standard (ISO 9000: 2000) ensures that secure working processes are in place and that these are continually reviewed and improvements are incorporated. All staff involved in the process of giving radiotherapy to patients must be fully involved in the QA and management system. Increasing the complexity of treatments does not necessarily result in more treatment errors; automated data transfer and automated setups have considerably decreased errors without increasing treatment times. However, increasing the complexity of manual treatments does lead to more errors. 47
Andrew Tyler and Louise Hanna
Modern 3D treatment planning systems provide a wealth of information which was previously not available, such as the density of internal body organs and DVHs, but using this information also needs careful QA to ensure that the data are valid (Purdy and Harms, 1998). New tests and phantoms have been designed to investigate the non-dosimetric functions of these systems. Commissioning 3D treatment planning systems involves a large number of quality checks, including testing the reliability of the transfer of body outlines, of the values for tissue density and of the projection of reconstructions. ‘TG53’ is an important document produced by a topic group of the AAPM Radiation Therapy Committee which describes all the tests to be considered when commissioning a treatment planning system (Fraass et al., 1998). Care needs to be taken when introducing what seem to be simpler treatments such as replacing physical wedges with dynamic ones. Implementation into the planning systems is complicated and entails significant QA before use in a clinical setting. Other tools such as the DVH used in 3D planning can cause errors if they are not used correctly. For example, the number of dose points used to form the DVH may be low and this will give a false volume, or the computer may be unable to evaluate the dose within a volume where there is a high dose gradient. Also, it may not be appropriate to define some organs (e.g. hollow structures like the rectum) simply by delineating their outer surface. It is expected that the dose under MLCs will increase because of leaf penetration and interleaf leakage, but if the backup collimators can be used, then the jaws will provide extra shielding and reduce the peripheral dose at these points. Verification systems have helped the treatment process but, to reduce the risk of errors, further automation of data transfer is essential.
Critical analysis of treatment plans Using the basic principles discussed earlier, the quality of a treatment plan can be reviewed and decisions can be made about whether to improve the distribution or whether a compromise must be made between delivery of the dose to the PTV and that to the adjacent OARs. Three principles, backed up by ICRU 50, provide the basics for this analysis:
48
Homogenous dose to the PTV. When the dose is prescribed at the ICRU reference point (typically the crosspoint of the beams), the highest dose in the PTV is usually considerably less than the +7% maximum described in ICRU 50. However, it may be more difficult to ensure that the PTV is covered by the 95% isodose while not irradiating too much normal tissue. Organs at risk. The plan should be designed to avoid overtreating any dose-limiting organ, but sometimes a compromise may have to be made, which will limit the overall delivery of dose to the PTV. Other normal tissues. The high dose volume in the PTV should fall off rapidly in normal tissue. When the correct choice of energy and beams is used, the doses in normal tissue are not usually a problem, but there will always be higher doses at the entry points of the beams. The exit sites of all beams should be checked carefully to ensure that the dose to sensitive tissues is at a safe level.
Examples of plans that require improvement Figure 4.4 shows two different plans for a radical radiotherapy treatment for carcinoma of the larynx. In example (a), two lateral beams are used and the PTV dose homogeneity is not ideal, because of the oblique skin surface. The solution, shown in example (b), is to use fairly small-angled wedged beams (∼15◦ ) to act as tissue compensators. However, care must be taken here because any changes in the patient’s contour in the S–I direction will not be accounted for, and a full compensator may be needed. Figure 4.5 shows two different plans for radical radiotherapy to the prostate. In plan (a), the 95% isodose lies considerably outside the lateral margins of the PTV. The solution to this problem is to make the anterior beam narrower, as shown in plan (b). Note that the DVHs for the PTV and rectum in both plans are identical. This shows that it is important to look at both the plan and the DVHs. This three-field plan requires 60◦ wedges on the lateral beams to compensate for the dose gradient caused by the anterior beam. The plan also illustrates the importance of using the correct beam direction. Using lateral beams rather than posterior obliques allows for greater sparing of the rectum, because the beam edge is used to try to provide the maximum dose gradient between the prostate and the rectum.
Radiotherapy planning
Isodose % 95 70 60 50 35
% 110 105 102 100 95 70 20
(a) (a)
Isodose % 95 70 60 50 35
% 102 100 95 70 20
(b) 100
(b)
90
Figure 4.4. Two different plans for a radical radiotherapy treatment for
80
carcinoma of the larynx (see text for further explanation).
70
Wide anterior beam Conformal plan
Figure 4.6 shows two plans for radical radiotherapy for carcinoma of the bronchus. In example (a), all three beams pass through the spinal cord, and cord tolerance is exceeded. A solution is shown in example (b) where the posterior beam has been moved to a posterior oblique direction and the anterior oblique beam has been moved more laterally. As a result only the exit beam from the anterior direction passes through the spinal cord and the dose is now within tolerance. Note also that, by exiting through the cord, the anterior beam avoids the lung. Part (c) shows the DVH for spinal cord is significantly better for plan (b), and although the lung DVH is worse for plan (b) it is still well within acceptable limits (V20 of less than 35%).
Electrons Electron beam isodoses have a shape that is significantly different from that of photon beams. The fairly homogenous dose distribution close to the surface and rapid fall-
Volume (%)
PTV
60 50 40
Rectum
30 20 10 0 0
10
20
30
40
50
60
70
80
90
100
110
Dose (%)
(c) Figure 4.5. Two different plans for radical radiotherapy to the prostate. The DVHs for both plans are shown in (c) (see text for further explanation).
off at depth make electron doses ideal for treating superficial tumours. The reporting of electron doses should follow ICRU 71 (ICRU, 2004) in a way similar to that for photons. Figure 4.7 shows an electron isodose for an 18-MeV electron beam.
49
Andrew Tyler and Louise Hanna
Isodose (%) 10 20 50 70 80 90 95 100
6 Off-axis distance (inline) (cm)
Isodose % 95 80 60 30 20
4 2 0 -2 -4 -6 0
(a)
15
Figure 4.7. Isodoses for a 6 × 6cm 18-MeV electron beam. Note how the higher isodoses become narrower with depth, while the lower isodoses become wider.
100 Original plan Cord-sparing plan
Isodose (%) 10 20 50 70 80 90 95 100 102
8 Off-axis distance (crossline) (cm)
(b)
6 4
Water
2 0 -2 Al -4 -6 -8
PTV 80
0
70
Volume (%)
10 Depth (cm)
Isodose % 95 80 60 30 20
90
5
5 Depth (cm)
10
Cord PRV
60
Figure 4.8. Isodoses for a 10 × 10cm 12-MeV electron beam with an
50
aluminium (Al) insert. Note the hot spot. The isodoses have been
40
normalised to the dmax of a beam with no insert.
30 Combined Lungs
20
r Bolus is often required to increase the dose at the
10 0 0
10
20
30
40
50
60
70
80
90
100
110
120
Dose (%)
(c) Figure 4.6. Two plans for radical radiotherapy for carcinoma of the bronchus. PRV = planning organ at risk volume; PTV = planning target volume (see text for further explanation).
Care is required when using electrons:
r High-dose treatment isodoses tend to pinch in, especially with small fields.
r Low-dose isodoses tend to spread out and can impinge on local OARs. 50
skin surface. The thickness of the bolus will need to be taken into account when deciding on the electron energy. The uncertainties are much greater than in photon beam treatments: r Skin apposition rather than point placement of beams is used for setup (cf. isocentric setups for photons). r Non-standard air gaps between the applicator and patient will affect the scattered dose. r Oblique surfaces will significantly alter the isodoses. r Inhomogeneities within the patient such as air gaps or high-density structures may cause unexpected hot and cold spots as demonstrated by Figure 4.8, where an aluminium sheet has been placed in the phantom.
Radiotherapy planning
Table 4.1. Properties of some radionuclides used in brachytherapy. Approximate Isotope
Average photon
half-life
Disintegration
energy (keV)
Types of source
Caesium-137 (137 Cs)
30.2 years
β, γ
662
Needles, tubes, afterloading
Iridium-192 (192 Ir)
74 days
β, γ
380
Wires, pins, plaque, afterloading
Cobalt-60 (60 Co)
5.3 years
β, γ
1250
Needles, tubes, afterloading Seeds
Iodine-125 (125 I)
60 days
CE, CX, γ
28
Palladium-103 (103 Pd)
16.9 days
CX
20–23 (mean 21)
Seeds
Ruthenium-106 (106 Ru)
374 days
β
3541
Plaque
CE = conversion electron; CX = characteristic X-ray photon.
Brachytherapy Introduction Brachytherapy is radiation treatment given by placing a radioactive source near or in the target (usually a malignant tumour). It is possible to deliver a high dose of radiation to the target while sparing adjacent normal tissues; therefore, brachytherapy is a highly conformal type of therapy. It is also often accelerated radiotherapy. Tumours suitable for brachytherapy must be accessible, of moderate size, and able to be delineated. Brachytherapy treatments are described according to the position of the radioactive source in relation to the body: r Interstitial – in the target (may be permanent, as in iodine seeds, or temporary, as in iridium wires). r Intracavitary – in a body cavity. r Intraluminal – into a lumen. r Intravascular – into an artery. r Surface applications. For radiation protection purposes, non-radioactive guides are placed first whenever possible, followed by ‘afterloading’ of the radioactive source. Afterloading may be either manual or remote, but there is an increasing trend towards the use of remote afterloading. This section will describe the broad principles of brachytherapy. For more detail, the reader is referred to more comprehensive texts (see Further Reading).
Isotopes The properties of some radionuclides used in brachytherapy are shown in Table 4.1.
Dose rates The International Commission on Radiation Units and Measurements report 38 (ICRU, 1985) classified radiation dose rates at the point or surface where the dose is prescribed into low, medium and high, which have important differences in their radiobiological effects: r Low dose rate (LDR): 0.4 to 2 Gy/h. N. B. dose may need to be modified if it exceeds 1 Gy/h. r Medium dose rate (MDR): 2 to 12 Gy/h. r High dose rate (HDR): > 12 Gy/h. Thus, as the dose rate increases, it is necessary to reduce the total dose and fractionate the radiotherapy to avoid unacceptable late tissue damage.
Dosimetry The radiation dose delivered to the tissues is primarily determined by the activity of the source, the length of time that the sources are in place, and the inversesquare law (dose is proportional to the inverse square of the distance from the source). Traditional methods of dosimetry use rules to govern the configuration of radiation sources and methods to calculate the dose received by the tissues.
Paris system The Paris system was based on experience in the use of iridium-192 wires (Pierquin et al., 1978). Rules for the Paris system are as follow: r The linear activity of the sources must be uniform along the full length of the source and identical for each source. r The radioactive sources must be linear, parallel and straight, and their centres must be in the same plane 51
Andrew Tyler and Louise Hanna
r r r r r r
perpendicular to the direction of each source, which is called the central plane. In an implant all the sources must be separated equally from each other, but the source separation may vary from one implant to another. The separation between the sources may vary between implants, but the separation should be 0.8 to 2 cm. For volume implants the array of the source intersections in the central plane should be either in an equilateral triangle or a square configuration. The average length of the source should be 25 to 30% longer than the target volume, per uncrossed end. The dose is calculated as follows: The basal dose rate is calculated in the central plane of the implant and is the arithmetic mean of the local minimum dose rates. The reference dose rate is calculated as 85% of the basal dose rate and is the dose rate used for the tumour prescription and calculation time of the implant. The isodose surface that corresponds to the reference dose rate is called the reference isodose and this must encompass the PTV. For this to occur, the radiation sources must be longer than the PTV, unless the ends are crossed as in iridium hairpins. The volume enclosed in the reference isodose is known as the treated volume.
Manchester interstitial system The Manchester interstitial system was based on experience in the use of radium sources and is also known as the Paterson–Parker system (Meredith, 1967). Its rules governed the distribution of radiation sources for different configurations including planar implants, volume implants and moulds. For further details see Meredith (1967).
Manchester system for gynaecological brachytherapy The applicators are configured as a central uterine tube and two vaginal ovoids, with packing or a rectal retractor to push the rectum away from the high-dose volume. The resulting isodose distribution is a pear-shaped volume encompassing the uterus, cervix, paracervical tissues, and upper vagina. The radiation dose is specified at point A, which is 2 cm lateral to the centre of the uterine canal and 2 cm above the lateral vaginal fornices (this latter level is sometimes described as the level of the cervical os or the level of the flange at the base of the intrauterine tube). Point B is 5 cm lateral 52
to the midline at the level of point A. Other reference points are specified by ICRU 38 (ICRU, 1985), including the bladder and rectal points, which can be defined on the anterior and lateral radiographs. Reporting doses to these points allows comparison between centres. However, new imaging techniques, computerised treatment planning and optimisation open new prospects for individualised treatment and reporting dose to OARs (P¨otter et al., 2006). Isotopes used include caesium-137 for low- and medium-dose-rate treatments or cobalt-60 and iridium-192 for high-dose-rate treatments. Remote afterloading is typically used.
Brachytherapy procedures Indications for brachytherapy include the following: r Primary radical treatment (e.g. small tumours of tongue, floor of mouth, prostate). r Treatment in combination with external beam radiotherapy. r Treatment in combination with surgery (e.g. intraoperative sarcoma treatments). r Re-irradiation within a previously treated volume (e.g. second primary tumours or locally recurrent tumours). r Palliative treatments (e.g. bronchus, oesophagus). r Benign conditions (e.g. keloids). In the UK, the most widespread use of brachytherapy is for gynaecological cancer, head and neck cancer, and, increasingly, prostate cancer. Brachytherapy for ocular melanoma is discussed in Chapter 34 (see p. 403). Practical steps in brachytherapy: r The treatment is preplanned to determine the tumour volume, the target volume, the technique and the number/size of radiation sources required. An appropriate radionuclide is selected. r The implant procedure itself takes place under general or local anaesthetic. r With afterloading techniques, the guides for the active sources are placed first. r The position of the sources is verified (e.g. using orthogonal X-rays, CT, MRI, ultrasound). r The treatment time is calculated. r The active sources are then placed. r Once the implant is complete, the radiation sources are removed, followed by the applicators. ICRU 58 recommends that the brachytherapy treatment be reported in a standardised way, including information on target volume, description of sources
Radiotherapy planning
and technique, dose prescribed and a description of the high- and low-dose volumes (ICRU, 1997).
Quality assurance and radiation protection in brachytherapy Local rules and systems of work must be drawn up that enable the centre to comply with current legislation. These rules include source storage and preparation, written systems of work for staff entering controlled radiation areas and contingency plans for emergency situations such as source sticking. On receipt of a radiation source, checks should be carried out that include independent measurements of the activity of the source and tests for leakage. The quality of calculation methods needs to be properly ensured and a regular quality control programme should be in place. For computerised systems, this is especially important so that the transfer of patient image data and the validity of the data and calculations are verified. Any equipment used to measure the actual delivered dose needs to be calibrated against a recognised standard.
Areas of current interest in brachytherapy Optimisation Use of cross-sectional imaging, computer planning systems, and treatment machines with moveable stepping sources allows the radiation oncologist to define the target volume, obtain dose volume histograms and achieve the best dose distribution. The dwell times for the source can be varied at different positions, enabling the dose to conform more accurately to the target volume. But it is still important to take great care in positioning the brachytherapy sources, because optimisation cannot make a bad insertion good.
Pulsed dose rate Pulsed dose rate (PDR) is a hyperfractionated form of brachytherapy that is under evaluation; it uses remote afterloading and a single high-activity source that is stepped through dwell positions within the catheters. It combines the advantages of afterloading and the ability to optimise the treatment in a multiple-fraction highdose-rate treatment and it may provide an alternative to low-dose-rate brachytherapy, although differences in relative biological effectiveness must be taken into account.
REFERENCES Apisarnthanarax, S., Elliott, D. D., El-Nagger, A. K. et al. (2006). Determining optimal clinical target volume margins in head-and-neck cancer based on microscopic extracapsular extension of metastatic neck nodes. Int. J. Radiat. Oncol. Biol. Phys., 64, 678–83. Beavis, A. W. (2004). Is tomotherapy the future of IMRT? Br. J. Radiol., 77, 285–95. BIR. (2003). Geometric Uncertainties in Radiotherapy. Defining the Planning Target Volume. Prepared by a Working Party of the British Institute of Radiology. London: British Institute of Radiology. Bortfeld, T. (2006). IMRT: a review and preview. Phys. Med. Biol., 51, R363–79. Coles, C. E., Moody, A. M., Wilson, C. B. et al. (2005a). Reduction of radiotherapy-induced late complications in early breast cancer: the role of intensity-modulated radiation therapy and partial breast irradiation. Part I – normal tissue complications. Clin. Oncol. (R. Coll. Radiol.), 17, 16–24. Coles, C. E., Moody, A. M., Wilson, C. B. et al. (2005b). Reduction of radiotherapy-induced late complications in early breast cancer: the role of intensity-modulated radiation therapy and partial breast irradiation. Part II – radiotherapy strategies to reduce radiation-induced late effects. Clin. Oncol. (R. Coll. Radiol.), 17, 98–110. Dearnaley, D. P., Khoo, V. S., Norman, A. R. et al. (1999). Comparison of radiation side-effects of conformal and conventional radiotherapy in prostate cancer: a randomised trial. Lancet, 353, 267–72. Fraass, B., Doppke, K., Hunt, M. et al. (1998). American Association of Physicists in Medicine Radiation Therapy Committee Task Group 53: quality assurance for clinical radiotherapy treatment planning. Med. Phys., 25, 1773–829. Gregoire, V., Coche, E., Cosnard, G. et al. (2000). Selection and delineation of lymph node target volumes in head and neck conformal radiotherapy. Proposal for standardizing terminology and procedure based on the surgical experience. Radiother. Oncol., 56, 135–50. Guerrero Urbano, M. T. and Nutting, C. M. (2004a). Clinical use of intensity-modulated radiotherapy: part I. Br. J. Radiol., 77, 88–96. Guerrero Urbano, M. T. and Nutting, C. M. (2004b). Clinical use of intensity-modulated radiotherapy: part II. Br. J. Radiol., 77, 177–82. Hong, T. S., Ritter, M. A., Tome, W. A. et al. (2005). Intensitymodulated radiation therapy; emerging cancer treatment technology. Br. J. Cancer, 92, 1819–24. Hornick, D. C., Litzenberg, D. W., Lam, K. L. et al. (1998). A tilt and roll device for automated correction of rotational setup errors. Med. Phys., 25, 1739–40. ICRU. (1976). ICRU Report 24. Determination of Absorbed Dose in a Patient Irradiated by Beams of X or Gamma rays in Radiotherapy Procedures. Bethesda, Maryland: International Commission on Radiation Units and Measurements. ICRU. (1985). ICRU Report 38. Dose and Volume Specification for Reporting Intracavitary Therapy in Gynaecology. Bethesda, Maryland: International Commission on Radiation Units and Measurements.
53
Andrew Tyler and Louise Hanna
ICRU. (1993). ICRU Report 50. Prescribing, Recording, and Reporting Photon Beam Therapy. Bethesda, Maryland: International Commission on Radiation Units and Measurements. ICRU. (1997). ICRU Report 58. Dose and Volume Specification in Interstitial Brachytherapy. Bethesda, Maryland: International Commission on Radiation Units and Measurements. ICRU. (1999). ICRU Report 62. Prescribing, Recording, and Reporting Photon Beam Therapy (Supplement to ICRU Report 50). Bethesda, Maryland: International Commission on Radiation Units and Measurements. ICRU. (2004). ICRU Report 71. Prescribing, Recording, and Reporting Electron Beam Therapy. Journal of the ICRU, 4 (No. 1). Oxford: Oxford University Press. Jansen, E. P. M., Dewit, L. G. H., van Herk, M. et al. (2000). Target volumes in radiotherapy for high-grade malignant glioma of the brain. Radiother. Oncol, 56, 151–6. Kehoe, T. and Rugg, L. J. (1999). From technical quality assurance of radiotherapy to a comprehensive quality of service management system. Radiother. Oncol., 51, 281–90, Khan, F. M. (2003). The Physics of Radiation Therapy, 3rd edn. Philadelphia: Lippincott, Williams and Wilkins. Klein, E. E., Harms, W. B., Low, D. A. et al. (1995). Clinical implementation of a commercial multileaf collimator: dosimetry, networking, simulation and quality assurance. Int. J. Radiat. Oncol. Biol. Phys., 33, 1195–208. Langen, K. M. and Jones, D. T. L. (2001). Organ motion and its management. Int. J. Radiat. Oncol. Biol. Phys., 50, 265–78. Lees, J., Holloway, L., Fuller, M. et al. (2005). Effect of intravenous contrast on treatment planning system dose calculations in the lung. Australas. Phys. Eng. Sci. Med. 28, 190–5. Logue, J. P., Sharrock, C. L., Cowan, R. A. et al. (1998). Clinical variability of target volume description in conformal radiotherapy planning. Int. J. Radiat. Oncol. Biol. Phys., 41, 929–31. McNair, H. A., Adams, E. J., Clark, C. H. et al. (2003). Implementation of IMRT in the radiotherapy department. Br. J. Radiol., 76, 850–6. Meredith, W. J. (1967). Radium dosage: the Manchester system/compiled from articles, by Ralston Paterson (and others). Edinburgh: Livingstone. Nutting, C., Dearnaley, D. P. and Webb, S. (2000a). Intensity modulated radiation therapy: a clinical review. Br. J. Radiol., 73, 459–69. Nutting, C. M., Khoo, V. S., Walker, V. et al. (2000b). A randomized study of the use of a customized immobilization system in the treatment of prostate cancer with conformal radiotherapy. Radiother. Oncol., 54, 1–9. Padhani, A. R., Khoo, V. S., Suckling, J. et al. (1999). Evaluating the effect of rectal distension and rectal movement on prostate gland position using cine MRI. Int. J. Radiat. Oncol. Biol. Phys., 44, 525–33. Pierquin, B., Dutreix, A., Paine, C. H. et al. (1978). The Paris system in interstitial radiation therapy. Acta Radiol. Oncol. Radiat. Phys. Biol., 17, 33–48.
54
P¨otter, R., Haie-Meder, C., Van Limbergen, E. et al. (2006). Recommendations from gynaecological (GYN) GEC ESTRO working group (II): concepts and terms in 3D image-based treatment planning in cervix cancer brachytherapy-3D dose volume parameters and aspects of 3D image-based anatomy, radiation physics, radiobiology. Radiother. Oncol., 78, 67–77. Purdy, J. A. (2000). Future directions in 3-D treatment planning and delivery: a physicist’s perspective. Int. J. Radiat. Oncol. Biol. Phys., 46, 3–6. Purdy, J. A. (2002). Dose-volume specification: new challenges with intensity-modulated radiation therapy. Semin. Radiat. Oncol., 12, 199–209. Purdy, J. A. and Harms, W. B. (1998). Quality assurance for 3D conformal radiation therapy. Strahlenther. Onkol., 174, (Suppl. 2), 2–7. Senan, S., Chapet, O., Lagerwaard, F. J. et al. (2004). Defining target volumes for non-small cell lung carcinoma. Semin. Radiat. Oncol., 14, 308–14. Seppenwoolde, Y., Shirato, H., Kitamura, K. et al. (2002). Precise and real-time measurement of 3D tumor motion in lung due to breathing and heartbeat, measured during radiotherapy. Int. J. Radiat. Oncol. Biol. Phys., 53, 822–34. Shimizu, S., Shirato, H., Kitamura, K. et al. (2000). Use of an implanted marker and real-time tracking of the marker for the positioning of prostate and bladder cancers. Int. J. Radiat. Oncol. Biol. Phys., 48, 1591–7. Shiu, A. S. and Mellenberg, D. E. (2001). General practice of radiation oncology physics in the 21st century: AAPM Monograph no. 26. Madison, Wisconsin: Medical Physics Publishing. Weiss, E., and Hess, C. F. (2003). The impact of gross tumor volume (GTV) and clinical target volume (CTV) definition on the total accuracy in radiotherapy theoretical aspects and practical experiences. Strahlenther. Onkol., 179, 21–30. Williams, J. R. and Thwaites, D. I. (2000). Radiotherapy Physics in Practice. Oxford: Oxford University Press.
FURTHER READING Bomford, C. K. and Kunkler, I. H. (2003). Walter and Miller’s Textbook of Radiotherapy. Radiation Physics, Therapy and Oncology, 6th edn. Edinburgh: Churchill Livingstone. Dobbs, J., Barrett, A. and Ash, D. (1999). Practical Radiotherapy Planning, 3rd edn. New York: Arnold. Gerbaulet, A., P¨otter, R., Mazeron, J.-J. et al. (2002). The GEC ESTRO Handbook of Brachytherapy. Brussels: European Society for Therapeutic Radiology and Oncology. Hoskin, P. (2006). Radiotherapy in Practice. External Beam Therapy. Oxford: Oxford University Press. Hoskin, P. and Coyle, C. (2005). Radiotherapy in Practice. Brachytherapy. Oxford: Oxford University Press. Joslin, C. A. F., Flynn, A. and Hall, E. J. (2001). Principles and Practice of Brachytherapy: Using Afterloading Systems. London: Arnold.
5
RESEARCH IN CANCER Robert Hills
Introduction It is the responsibility of clinicians to provide the best possible care for their patients. However, this simple statement masks a much more complex issue. How does one know precisely what the best care is for a particular patient? In particular, how does one balance the likely benefits and risks for a particular course of treatment? A new drug may appear promising, but can one really be sure that it represents a real improvement on current practice? Generally speaking, unless the action of a particular treatment is both immediate and breathtaking (such as insulin for diabetic coma), we cannot be absolutely certain which treatment is best for which people. Historical comparisons, or other database-dependent methods, can prove misleading. What is required is a method that will provide reliable, convincing evidence that can be used to inform future practice. Fortunately, there is such a tool: the randomised controlled trial (RCT). At its heart are two principles. First, through randomisation, any differences between patients receiving one treatment and those receiving another are purely down to chance; therefore, if a sufficiently large difference is detected, then it must be due to the only factor that is systematically different between the two groups, namely the treatment. Second, with large numbers of patients, it becomes easier to detect smaller treatment effects and to conclude that any differences are not the result of chance. This, the statistical aspect of RCTs, is effectively a formalisation of common sense. If one tosses a coin 10 times and gets 6 heads and 4 tails, it is not out of the ordinary; but if one saw 6000 heads and 4000 tails from 10 000 tosses, then one would be concerned that the coin may be biased. The proportion of heads is the same, but larger numbers give stronger evidence of an unfair coin. This chapter will concentrate on obtaining reliable evidence on the efficacy (whether the treatment works under ideal conditions, usually in a highly selected population) and effectiveness (whether a treatment will be beneficial in a real-life setting) of treatments for cancer.
In particular, it will look at the factors that constitute a successful clinical trial, how the ideas can be extended to look at the weight of evidence provided by a number of clinical trials (meta-analysis) and how additional laboratory studies can help assess more modern targeted therapies. Much excellent literature (e.g. Altman, 1991; Duley and Farrell, 2002; Pocock, 1996) has already been devoted to the theory of the RCT (and much that is less excellent). Likewise, medical statistics has been well covered in a number of books and articles (Altman et al., 2000; Swinscow and Campbell, 2002). This chapter will concentrate on the underlying principles, with particular emphasis on their applications in cancer. This chapter is intended to provide useful information for researchers planning to conduct an RCT, and also provides useful pointers for clinicians wanting to critically appraise a trial for reliability.
Clinical trials Introduction The path taken by a new treatment before it comes into common use is typically a long one. Initially, a compound may well be developed based on an association seen in a laboratory and backed up by early research involving animals. But, although this avenue may identify promising treatments, it does not tell us the proper dose to give for treatment (or even if it is safe at all to use in humans), whether promising laboratory results translate into activity in humans, and, most important, whether the treatment gives better results when used in clinical practice compared to current best care. Increasingly, too, health care providers are interested in determining whether or not new treatments represent sufficiently good value for money. Thus, a new treatment (typically a drug) will go through four stages of development, and at each stage a clinical trial will be run to ascertain that the treatment still remains sufficiently promising to proceed: Phase I represents the first trials performed in humans; it determines that the treatment is safe 55
Robert Hills
and identifies the target dose. Such trials can involve either healthy volunteers or patients with the condition in question. The treatment dose starts off quite low, at a point where no real toxicity is expected, and is increased until the maximally tolerated dose (MTD), defined according to the proportion of patients experiencing significant toxicity, is found. A number of different designs for phase I trials have been proposed, ranging from the traditional Fibonacci design to more sophisticated techniques based on modelling the predicted toxicity levels based on all available data (Storer, 1989; Storer and DeMets, 1987). Phase II trials identify whether or not the treatment, at a safe dose, demonstrates sufficient evidence of activity. These studies can be considered as small screening studies, typically with a few dozen patients, and weed out treatments that are unlikely to be of any benefit in clinical practice. There is little point in going to the considerable effort and expense of setting up a large RCT if there is only a small likelihood of success. Typically, such trials use short-term surrogate end points, such as tumour shrinkage or remission induction, which are correlated with more clinically relevant end points such as overall survival. Historically, phase II studies were uncontrolled studies, where all enrolled patients received the study drug. But recently, a randomised design has increasingly been used. Such trials are still relatively small, and tend to use shortterm surrogate end points, but can form the basis of a larger phase III study if the treatment appears sufficiently promising. Phase III trials tend to be relatively large and randomised. These identify whether the treatment, given either in addition to or instead of current therapy, gives improved clinical outcomes. The phase III RCT is dealt with in more detail in the next section. After a new drug has been identified and demonstrated to provide worthwhile benefit in a phase III trial, it is studied further in phase IV trials. The entirety of the phase I to III process may only have meant that a few hundred patients have been given the drug, and that rare or long-term side effects may not have become apparent. Phase IV trials allow the study of the drug in standard clinical care in a large number of patients to identify any important safety issues, and to look at the long-term balance of risks and benefits. This four-stage process of treatment development is not set in stone; indeed, for many non-pharmaceutical 56
treatments, the earlier phases of development may not be directly relevant. Also, some of the phases may be performed simultaneously. For example, identifying the ideal dose for a new drug may not be a factor of toxicity alone: similar efficacy may be obtained at drug doses lower than the MTD, with a corresponding reduction in toxicity levels. Thus, it may make sense to perform a phase I/II trial in which both activity and toxicity are considered simultaneously. Likewise, a phase II/III trial could be designed in which a small, randomised pilot is assessed and the decision of whether to proceed to a full-scale large RCT is based on preliminary results.
Randomised controlled trials The main idea underlying the RCT is the need to distinguish between effects that are moderate, but still clinically meaningful, and those that are too small to be of any real clinical interest (Yusuf et al., 1984). Often, early reports of a treatment tend to be extremely positive and raise the possibility of major clinical advances. Yet, only a few treatments in current practice tend to work overwhelmingly well; the history of improvements in outcomes in cancer is one of incremental progress, and many of the drugs taken for granted today (e.g. tamoxifen in breast cancer) produce only moderate benefits, yet, owing to the worldwide prevalence of the conditions being treated, save many hundreds of thousands of lives annually. The actual likelihood of a new intervention having a big treatment effect, or being vastly superior to an existing therapy, is fairly low. It is more realistic to expect a moderate difference between interventions, or moderate effect compared with placebo. Therefore, it is important to provide reliable evidence about moderate benefits and to be able to reduce as far as possible any systematic biases that might affect the results. For this reason, non-randomised studies do not typically provide robust enough evidence. Case studies, in particular, may be seriously misleading. A potentially serious risk of selection bias exists: it is impossible to tell whether only the most promising patients are selected for the new treatment, and without a comparator group, the true scale of any ‘breakthrough’ cannot be gauged. Similarly, historically controlled studies, in which care involving the new treatment is compared to previous outcome rates before the new care was introduced, can provide unreliable estimates of a treatment’s effectiveness. For example, a comparison of today’s Alevel results with those of 1980 show an improved pass rate. Students sitting A levels in 1980 received free school
Research in cancer
Box 5.1 SAB in acute myeloid leukaemia. When comparing SAB with DAT treatment for acute myeloid leukaemia, the following results were obtained (Wheatley, 2002). Number of Treatment
Period
patients
CR rate
Induction
Resistant
death
disease
DAT
1984–90
167
47%
30%
23%
SAB
1990–98
284
61%
15%
24%
0.006
0.00007
0.2
p-value
DAT = daunorubicin + Ara-C + thioguanine; SAB = same as before (i.e. DAT).
milk while at primary school; those sitting them today do not. So, are we to conclude that free milk makes you less intelligent? More seriously, the general improvement in cancer outcomes over time means that one cannot be sure if any improvements are due to the drug, to the generally improving prognosis, to differences in case mix over time or to a combination of all three. In acute myeloid leukaemia, a historically controlled study of standard DAT therapy for patients over age 60 against the SAB regimen showed a statistically significant benefit for SAB (Wheatley, 2002; Box 5.1). This appears promising until one realises that SAB stands for ‘same as before’; in other words, the improvement has nothing to do with the treatment, but is likely instead to be a result of better supportive care. Likewise, if clinicians are more likely to give a treatment to patients with more severe symptoms, then those who receive therapy are of worse prognosis than those who don’t. A straight comparison could therefore show that the treatment appeared worse (Green and Byar, 1984). The only way to reduce such selection biases as much as possible is to randomise patients: allocate treatments in a way that produces equivalent groups and precludes the chance that the next treatment allocation can be predicted. Allocation by date of birth, for example, may not produce equivalent groups because it is easy to predict what a person’s treatment is going to be and then make a decision on whether to enter the trial based on this knowledge. A number of ways of allocating patients can produce equivalent groups (Altman and Bland, 1999, 2005). The simplest method is to use so-called simple randomisation in which the chance of receiving a given treatment is the same, irrespective of previous entries into the trial. Simple randomisation can be achieved using a random number list, a computer-generated random number or even tossing a coin. This method, however, can lead to chance imbalances in the numbers allocated to dif-
ferent treatment groups, so some trials use permuted block randomisation in which the number of patients in each group is required to be in balance at various stages through the trial (e.g. after every 4 or 6 patients). Some more sophisticated methods also ensure balance across important prognostic factors. It is possible to extend permuted block randomisation to this scenario or to use allocation algorithms such as minimisation, which generally require the use of a computer. In all except simple randomisation, where previous allocations do not influence the next, it is important to ensure that there are no patterns that could allow the next allocation to be predicted. (For example, in permuted block randomisation with a block length of 2, after an even number of patients the treatments will be in balance – in a single-centre trial, the allocation of every other patient could be predicted precisely). It is a fallacy that strict random allocation is required: far more important is concealment of the treatment allocation until the patient is irreversibly committed to the trial. Furthermore, patients who did not receive their preferred treatment cannot be re-randomised. Clinicians can and do attempt to subvert some randomisation approaches, and care needs to be taken to ensure that this is made as difficult as possible (Schulz, 1995). Box 5.2 gives some thoughts on good practice in randomisation.
Choice of end points Generally speaking, the choice of a suitable end point in cancer is less tricky than for many other chronic conditions. In many cases, the main aim is to prolong life, so the primary measured outcome in a trial is mortality. However, in certain instances, the outcome measure may be different. For example, in conditions that are relatively rare and in which outcomes are already very good (e.g. acute promyelocytic leukaemia), a trial assessing mortality may have to both be very large and run for a 57
Robert Hills
Box 5.2 Good practice in randomisation.
r Use third-party randomisation. An independent randomisation service with a degree of separation between clinician and patient provides more security. Envelopes containing the treatment allocation may appear an attractive option but need to be policed to stop clinicians from opening several envelopes until they find the preferred treatment. A secure computer database held at a central trials office is a good way to ensure allocation concealment. r Collect all important prognostic factors prior to randomisation. The recorded value of a prognostic factor may change or be influenced by treatment allocation. r Balance prognostic groups by stratification or minimisation. Similar numbers of different types of patient are ensured to be in each group. At its simplest, this could be achieved by a randomisation list for each category, but for more than a few variables this becomes cumbersome and a computerised system is preferable and allows simultaneous minimisation across many variables.
long time, making such a study impractical. In these circumstances, it may be more relevant to assess quality of life – can one achieve as-good outcomes with less toxicity? This approach has been used in the NCRI AML15 trial (www.aml15.bham.ac.uk). Similarly, the primary aim of some trials may not be to directly improve mortality but to reduce the incidence of adverse events, as in the SIGNIFICANT trial of prophylactic antibiotics, which measured the number of febrile episodes (Cullen et al., 2005). The choice of outcome measure influences both the type of statistical test that is being performed and the size a trial needs to be. In choosing a suitable outcome measure, it is important to choose a measure that is of clinical relevance and that can be used to help guide future practice. One may consider outcomes relevant to patients, clinicians and to organizations such as the National Institute for Health and Clinical Excellence (NICE). Generally speaking, the simplest outcomes are often the best. It is important, however, not to try and collect too many outcome measures: the trial will then lack focus, and, by chance, one outcome may indicate that one particular treatment is better, and for another 58
outcome the reverse is the case. Given enough outcomes, it is usually possible to find a significant result for one of them even in trials of the most unpromising treatments. Therefore, it is crucial to identify a small number of primary and secondary outcomes, which are considered the most important. These may be clinical outcomes obtained from notes or patient-centred outcomes collected through questionnaires. If a questionnaire is being used, then it is worthwhile to ensure that it is validated – to make sure that the results are meaningful and reproducible and, if possible, to define how big a difference is relevant. Sometimes, the real outcome of interest occurs a long time in the future. Because it is impractical to wait many years for an answer, it is tempting instead to use a surrogate outcome, which predicts this future outcome (e.g. recurrence to predict survival benefit). However, this approach can be fraught with difficulties. For example, there may be a difference in recurrence-free survival, but this difference is not necessarily translated into a significant survival benefit (Specht et al., 1998).
Blinding To create comparable groups for analysis, it is important that there is no foreknowledge of treatment allocation. This concept is known as allocation concealment. However, even after randomisation there are occasions when knowledge of which treatment group a patient is in can influence their treatment or their outcomes. For example, one treatment may require additional followup visits, so that supportive care or additional treatments could be provided preferentially to one group or another. Alternatively, if outcomes are dependent on clinician or patient rating of health state (e.g. in qualityof-life trials) then knowledge of a patient’s treatment may influence the perception of health status (especially if one option is to offer no treatment). There are a number of ways of combating this problem. Standardised treatment protocols can be introduced, so that all patients receive equivalent supportive care and clinician contact. Also, the use of objective outcome measures, such as mortality, tends to minimise the effect of knowledge of treatment allocation on outcome. Another alternative is to blind the trial treatments. Trials are typically referred to as either singleblind or double-blind. In the former, either the patient or the clinician/assessor is unaware of the treatment allocation; generally, the person recording the outcome is the one who needs to be blinded. In double-blind
Research in cancer
trials, both the patient and the clinician/assessor are unaware of which treatment was given. This way, any beliefs about the intervention, negative or positive, should not be expected to influence the outcome. The use of a matching placebo can ensure blinding; alternatively, pharmacists can make up drug solutions that are indistinguishable irrespective of the treatments actually contained. For other trials, some degree of blinding can be achieved with imagination: in trials of surgical techniques, the use of sham surgery has been advocated, although sometimes this will require the participant’s acceptance of an additional, clinically unnecessary procedure. It has been argued that sham procedures are methodologically necessary to produce valid results; therefore, as long as the participant is informed that they will receive either a real or a sham intervention and that the sham procedure will be indistinguishable from the real treatment under investigation, there is no deception involved (Miller and Kaptchuk, 2004).
Choice of subjects, and how many In a randomised trial, because the different treatment groups are equivalent, it follows that any observed differences must be the result either of chance or of a genuine difference between the treatments. The probability that observed differences are the result only of chance differences between groups is given by the pvalue, which tells you how often a trial of an ineffective treatment would be expected to produce this type of result. If the p-value is sufficiently small, then one can conclude that the data are inconsistent with the treatment being ineffective, rather in the same way that a jury starts with the presumption of innocence and asks whether there is enough evidence to overturn it beyond reasonable doubt. Alternatively, one can determine an estimate of the size of the treatment effect, together with a measure of a likely range, based on the natural variability between patients. Clearly, to detect moderate differences, one must discriminate between the signal and the ambient noise: although one cannot reduce differences between people, larger numbers enable one to improve the signalto-noise ratio. However, for relatively uncommon conditions, the number of patients required in a study to detect moderate treatment effects may perhaps be more than it is feasible to recruit. The smaller the treatment effect one wishes to detect, the larger the number of patients one needs to recruit. To have a 90% chance of detecting whether a treatment increases the proportion
of patients entering remission from 50 to 75% requires only about 150 participants (Machin et al., 1997). But, to detect an improvement from 50 to 60%, which would still be worthwhile, a study needs about 1000 patients; to detect a 5% improvement would require about 4000 participants. A corollary of this is that a non-significant result does not automatically mean that there is no treatment difference: the trial could simply have been too small to detect a moderate, but worthwhile, treatment effect (Altman and Bland, 1995). The use of effect sizes and confidence intervals overcomes this problem. The 95% confidence interval contains the true treatment effect 95% of the time; only if this does not include a clinically relevant treatment effect is it fair to conclude that a worthwhile difference is unlikely. The number of patients required to answer a question reliably depends on a number of factors. First, the sample size depends on the type of outcome being measured: dichotomous outcomes (e.g. entered remission, experienced grade 3 or 4 toxicity), continuous outcomes (e.g. quality-of-life scores, size of a laboratory measure) and time-to-event outcomes (e.g. survival time, duration of remission) all need different calculations. Second, the size of treatment effect one wishes to detect needs to be specified (e.g. from 50 to 60% entering remission, a difference of 10 points on a scale with a standard deviation of 25, etc.). Finally, it is important to specify the level at which statistical testing will be performed (typically set at a significance level of p ≤ 0.05) and how certain one wants to be of detecting a treatment effect if it really is there. This latter concept is known as the power and is typically set at 80 or 90%, representing a 1-in-5 or 1-in-10 chance of missing a real treatment effect. The smaller the effect one wishes to detect and the smaller the significance level, or the higher the power, the larger the number of patients required. It may seem attractive to aim to detect a large treatment effect, or to set the power low, but this means that one runs the risk of the estimate of treatment effect to be sufficiently imprecise as to include both no treatment effect and a clinically relevant effect, thus making the trial inconclusive. When considering the required number of patients, it is also important to determine how these patients are likely to be found. Extending a trial over a number of centres will increase the relevant pool of potential participants. However, it is important to identify precisely who is eligible for the trial. Patients who could potentially be at risk from one of the trial interventions need to be excluded. However, when deciding which patients 59
Robert Hills
to include in a potential trial, it is possible to have selection criteria that are too rigid and exclusive. Not only does restricting the population too much reduce the pool of potential participants, making it more difficult to achieve target recruitment, but the results of the trial may not be immediately generalisable to other important groups of patients. For example, age can be applied as an exclusion criterion on arbitrary grounds. For this reason, it has been argued that it is preferable to avoid prescriptive inclusion and exclusion criteria and aim for a representative sample from which generalisable results can be obtained (Collins et al., 1996). For a given patient, there may be uncertainty about the relative merits of different courses of action. Presented with the same circumstances, clinicians may have differences in opinion. The ethical imperative then would be to try to reduce uncertainty by contributing to the evidence, and the most appropriate way of achieving this would be to recruit the patient into a well-designed clinical trial. It would be unethical for a patient to have their treatment chosen at random if either they or their doctor are certain about what treatment they prefer. However, randomisation can be considered when both doctor and patient are uncertain about which treatment is preferable. No further restrictions, other than diagnostic and safety-related criteria, would then be applied, allowing a wide range of patients to be recruited. This sort of wide, pragmatic entry condition has an added benefit. If a wide range of patients is randomised, it is possible to examine, albeit cautiously, whether different groups of participants respond differently to the treatment. Of course, it is important to collect data to identify precisely what type of patient has been entered, but only by running large-scale randomised trials with wide entry criteria can questions of which treatment is best for which patient be answered with any confidence.
Collecting data As pointed out earlier, it is important to choose a small number of primary and secondary outcomes to assess the effectiveness of a new treatment. It is also important to determine exactly how these data will be collected. Can, for example, national records (e.g. of mortality) be used to determine outcomes? If so, then this means that one is not reliant on clinicians completing forms, and there can be (virtually) complete ascertainment of the primary outcome measures. Alternatively, it may be preferable to ask clinicians or patients to complete 60
forms at predetermined time points. In this way, additional patient information can be captured. It is often tempting to include extra ‘nice-to-know’ data, perhaps, with a view to using them for analyses, additional to the main paper. However, before becoming committed to extra work, not only for the statisticians and data managers but also for clinicians, one needs to be sure that the data are really relevant. Data that can be collected only on a minority of patients are unlikely to influence the trial’s conclusions. If it is unclear how, or indeed whether, data are to be analysed, they should not be collected. Collecting data that will not be analysed puts an extra burden, without any ultimate purpose, on participants and clinicians. Keeping data collection to a minimum also increases the likelihood of it being collected. Data should be collected on simple, well-designed forms. It is helpful for trial organisers to liaise with the statistician and the person designing the database for storing information to ensure that forms are clear and unambiguous and that data are coded wherever possible because coded answers are easier to analyse than free text fields, which may require extensive recoding before analysis. Forms that are well designed and as short as possible increase the likelihood of completion (Edwards et al., 2002). To obtain good follow-up: r Keep it short – don’t ask for unnecessary data. r Keep it simple – don’t add unnecessary extra tests. r Keep it seldom – don’t repeat assessments too many times.
Analysing data: the intention-to-treat principle Once patients are randomised into a trial, they should remain in the analyses, even if they stop taking the treatment or, indeed, never even receive it. The main idea behind randomisation is the creation of equivalent groups of patients; if patients are left out of the analysis, the groups cease to be equivalent. Intention-totreat (ITT) analysis analyses every patient according to the treatment to which they were allocated rather than what they received. In cases in which there are protocol deviations, then ITT analyses tend to be conservative: they underestimate the treatment effect. So, if one sees a difference, then one can be reasonably sure that it really is there. Other sorts of analysis may look attractive because they compare what actually happened, not what was meant to happen. But, as can be seen in Box 5.3, these other analyses can lead to misleading results,
Research in cancer
Box 5.3 Intention to treat.
Consider a trial of chemotherapy with associated toxicity. The trial is performed in two types of patients, who are recruited in equal numbers: ‘fit’ patients who have 70% survival, and ‘unfit’ patients who have a 30% survival probability. These survival probabilities are unaffected by the new treatment. Because the treatment is toxic but potentially curative, some clinicians are not happy to give unfit patients the more-toxic treatment, and they give the less-toxic existing therapy instead. Likewise, fit patients who draw control treatment may receive the new therapy because they are thought to be fit enough to withstand the side effects. In the trial, 20% of each group receives the opposite chemotherapy. It is possible derive two tables of the proportions of patients surviving. First, for the ITT analysis, Allocated new treatment
Allocated control treatment
Fit
50%
50%
Unfit
50%
50%
Survival ratea
50%
50%
And second, for the as-treated analysis, Receiving new treatment
Receiving control treatment
Fit
60%
40%
Unfit
40%
60%
Survival ratea
54%
46%
a
The survival rate is calculated as 70% times the proportion who are ‘fit’ plus
30% times the proportion who are ‘unfit’ for each group.
Whereas an ITT analysis correctly shows that the new chemotherapy is no better in prolonging survival (and, hence, it is more toxic and less attractive than the original option), the as-treated analysis produces an 8% survival benefit, which may be enough to change clinical practice. The difference arises because fewer poor-risk patients receive the new treatment. This sort of analysis tends to show an artificial benefit for the treatment received by the better-risk patients. making ineffective therapies appear worthwhile. What ITT really measures is the effect of introducing the policy of giving the treatment: what is the effect of introducing a new treatment in the real world? Inevitably, there will be circumstances in which patients who have committed to the trial do not complete treatment and follow-up. Sometimes this is treatment-related, for example, if side effects are distressing or if there is no perceived benefit from the treatment. These drop-outs are not random and may appear more frequently in one arm of the study than another, so it is crucial to note the reason for non-compliance or withdrawal, especially if no further follow-up information is obtainable. Failure to do this will introduce a systematic difference between the groups (known as attrition bias). Ideally, however, patients who withdraw from treatment should still be encouraged to contribute to data collection; it is a common misconception that
a deviation from the protocol necessitates withdrawal. In cancer trials in which the outcome is mortality, this is less of a problem because simple follow-up should be available from clinicians or from national records, but it can cause serious problems in quality-of-life studies. For such studies, where direct patient contact may be required, a meta-analysis of methods of improving the response rate to postal questionnaires identified a number of successful strategies (Edwards et al., 2002), such as a precontact before the questionnaire is sent, provision of a prepaid reply envelope and using a short, interesting questionnaire, in addition to obvious monetary incentives.
Perils of subgroups A clinical trial sets out to answer whether a particular treatment is effective, on average, for a wide range of 61
Robert Hills
patients. Although a trial can give some evidence on what treatment to give an individual patient, it does so by considering the population as a whole, and not each separate individual. It is tempting to try and identify subgroups of patients that benefit from an intervention. However, note that a significant result (p < 0.05) will be seen by chance 5% of the time. So, in a trial of an ineffective treatment, if one looks at 20 different subgroups of patients, there is likely to be on average a significant result in one subgroup. It is laughable to say that, for people born on one particular day of the month, putting brown sugar in their coffee is any better (or worse!) in controlling febrile neutropenia than using white sugar. However, it is possible that, if one runs a large RCT and ‘data dredges’ through data from enough subgroups, one will get this kind of spurious result, which will appear purely by chance. If one looks hard enough, it is likely that one will find a statistically significant result – but it is also possible that the result is a fluke. There are statistical techniques to test whether observed effects are real (Assmann et al., 2000), but trials that look at subgroups usually need to be much bigger. As a rule of thumb, to detect a difference reliably within subgroups of the same size as a treatment effect that you have decided is clinically meaningful would require four times as many trial participants as it would to detect an overall effect. Subgroup analyses are therefore typically designed to detect qualitative differences in treatment effects (treatment works in one group, but not in another) as opposed to smaller, quantitative differences in effect size (treatment works in both groups, but gives greater benefit in one group than another). If there are subgroups in which there are legitimate reasons for anticipating a different response to treatment (e.g. oestrogen receptor status in trials of hormonal treatments for breast cancer), then these should be specified in the trial protocol in advance, together with a justification of the underlying mechanism and the influence it may have on the outcome. In any case, one should not conclude, simply because a treatment provides a significant effect in one subgroup and not in another, that the treatment works only for one subgroup of patients. (There are a number of examples of this erroneous reasoning in literature: Cascinelli, 1994; Cascinelli et al., 1994; Creutzig et al., 2001; Wheatley and Hills, 2001.) When significant subgroup differences exist, they should always be viewed in the context of the overall findings. It is unlikely that, if the overall result is positive, a subgroup finding of significant harm is to be relied on. Again, such qualitative effects need to 62
Table 5.1. A typical 2 × 2 factorial design. One-quarter of patients receives both treatments A and B, one-quarter receives A only, one-quarter B only and one-quarter receives control. Drug B
Control
Drug A
A+B
A
Control
B
Control
be viewed with scepticism unless there is some plausible, prespecified reason why some subgroups should behave differently.
Additional topics A common theme in the preceding sections is the need to provide reliable evidence that is likely to influence future clinical practice. This means providing evidence on patients similar to those one would meet in everyday clinical practice and ensuring that the level of evidence is sufficiently strong to enable the making of a decision on treatment. The necessary corollary of this is that trials need to be large. With only a finite population of patients from which to choose, recruiting enough participants can be difficult. For this reason, there are a number of techniques that can help trials be run more efficiently. In particular, two design features are widely used in cancer clinical trials so that advances can be made as rapidly as possible. The first is the factorial design. Factorial designs allow more than one research question to be answered simultaneously. If two different drugs are available and no interactions are expected between them, then if both drugs were shown to work, it would be reasonable to give them both together. To test both treatments, one can randomise patients in a 2 × 2 factorial trial design to receive drug A or control, in addition to receiving either drug B or control (Table 5.1). Then, to assess the effect of treatment A, one can compare patients receiving A + B with patients receiving B, and those receiving A with control. Because a similar procedure can be done to assess treatment B, every patient is contributing to two comparisons. Thus, two research questions can be answered at once. This method is more complicated if there is evidence of interaction between the treatments, but it is only in a factorial design that such interactions can reliably be investigated.
Research in cancer
Box 5.4 The AML14 trial.
The AML14 trial was intended for patients with AML over the age of 60 and ran in two parts depending on whether patients were deemed fit for intensive chemotherapy. For those receiving intensive treatment, the design was a 3 × 2 factorial design for induction, with a further randomisation to either 3 or 4 courses of treatment, giving a total of 12 treatment options (3 × 2 × 2). Two courses of induction treatment Daunorubicin 50 mg/m2
Ara-C 200 mg/m2 b.d.
and treatment
three courses
vs
vs
lasting in total for
vs
Daunorubicin 35 mg/m2
Ara-C 100 mg/m2 b.d.
with
four courses
vs Daunorubicin 35 mg/m2 + PSC-833
Factorial designs are common in cancer. For example, the MRC/NCRI trials in acute myeloid leukaemia (AML) typically use factorial designs (e.g. the AML14 trial; see Box 5.4; Burnett et al., 2005). Other methods can be used to ensure that trials recruit a large, clinically representative group of patients. One such approach is the pragmatic design adopted by the MRC QUASAR trial of chemotherapy for colorectal cancer, which considered the effectiveness of 5-FU/folinic acid–based chemotherapy as well as two different doses of folinic acid and the effect of adding levamisole to chemotherapy (QUASAR Collaborative Group, 2000). To recruit widely, clinicians were able to specify if a patient had a clear indication for chemotherapy (in which case patients were randomised in a 2 × 2 factorial design, between two doses of folinic acid and the addition or not of levamisole), or an uncertain indication, in which case they were randomised either to receive chemotherapy (randomised in the same 2 × 2 design) or not. The trial recruited more than 7000 participants to help provide reliable evidence, not only on whether to give 5-FU/FA chemotherapy but also to provide evidence on the dose of folinic acid and the role of levamisole. In rarer conditions, the challenge is perhaps a little different. In such conditions, a large trial may take many years to complete, meaning that many treatments compete for the same small pool of patients. Therefore, it is important to identify as efficiently as possible those treatments that are likely to be the most promising. For conditions with poor prognosis, one possible way forward is provided by the current NCRI AML16 trial of non-intensive therapy for older patients with AML (see www.aml16.bham.ac.uk). Here, because relatively large improvements are required from expensive treatments to make them worthwhile, treatments are tested
on a small number of patients and those treatments that show no sign of benefit are dropped. This ‘pick a winner’ approach should enable useless treatments to be discarded early, with only a small chance of inadvertently discarding a good option.
Ethical considerations Introduction Most clinical research can only be carried out with the cooperation of patients who enter clinical trials, agree to be treated according to study protocols and who are followed up and their information used to assess the value of a new treatment. It is therefore paramount that the rights and well-being of participants are considered. In particular, because treatments are usually not without risks, it is important that patients are aware of any potential risks, and that there is not already good evidence that one or the other treatment is better. There is a considerable body of literature, and a lively debate, on what constitutes good clinical practice (GCP) in clinical trials, as well as a growing body of guidance and legislation, including the data protection act, various handbooks on GCP, and the recent European Union directive. One clear area of debate is how to strike the correct balance between protecting the interests of participants and excessive bureaucracy, which may tend to reduce the number and size of clinical trials (Burman et al., 2001).
Study protocol Perhaps the most important way of achieving goodquality clinical research is by having a well-written trial 63
Robert Hills
protocol. Such a protocol is not only a justification for the research being performed, it is the trial ‘bible.’ It should contain all relevant information and instructions about the trial, such as treatment schedules and reporting methods. During the course of a trial, protocols may be modified as new information becomes available, or for more prosaic, practical reasons, such as alterations in drug supply, doses or route of administration. The protocol needs to be updated with this information and all investigators provided with a copy of the new protocol. For this reason, it is sometimes helpful to store copies of all trial documentation on a website so that clinicians can download all the necessary information to keep their own site files up to date.
Informed consent All participants in a clinical trial have the right to be adequately informed about the benefits and risks involved in participating in a clinical trial. Participation in a clinical trial must be voluntary, and informed consent of participants needs to be taken. Information about the trial, and participation in it, is typically given using a patient information sheet, which explains the research, the possible treatments, what data are being collected and how data will be used. For studies that include the taking of tissue samples, consent needs to be obtained for this too. Generally speaking, potential participants need to be given time to digest the information they have been given and to ask any questions they may have; thus, the taking of participants’ consent involves a long period of time. Strategies may need to be considered to allow participants the time they need (e.g. ensuring that a research nurse is present during clinics to help with the process and provide a point of contact for participants).
Good clinical practice and ethics committees The aims of GCP in clinical trials are twofold: first, to protect the participants and, second, to provide assurance that the trial has been carried out to a sufficiently high standard that results are credible. Various aspects of GCP ensure that the trial is, and remains, ethical; that treatments are safe and used appropriately; that data collection is adequate and reliable; and that data are stored securely. Although these aims can all be agreed on, the precise way in which they are achieved is still a matter for debate, and a number of guidelines on GCP exist, from the UK MRC Guidelines (available from www.mrc.ac.uk), aimed at academic trialists, 64
to the International Committee on Harmonisation (ICH) guidelines adopted by industry-sponsored trials (see www.ich.org). These guidelines differ in the extent to which, for example, onsite monitoring, or source data verification is required, but both stress the importance of having well-written protocols, independent Trial Steering Committees (TSC) and set roles and responsibilities of the different people taking part in the study. All trials require ethical approval, which involves submitting the study protocol and supporting documentation, such as patient information sheets, to a Research Ethics Committee. The ethics application process has been streamlined, and paperwork for ethics committee applications can be found centrally at www.nres .npsa.nhs.uk. The ethics committee will then assess the protocol and make a judgement. Additionally, each centre is required to undergo a site-specific assessment to ensure that the centre, and the people who will be taking part in the trial, are capable of supporting the protocol. Also, trials in the EU that involve an investigational medicinal product (IMP) are now subject to the 2001 EU Directive, which was transposed into UK law on 1 May 2004. Trials now require a sponsor, and all trials involving IMPs require clinical trial authorisation from the relevant medicines agency (see http://eudract .emea.eu.int). Another implication of the directive is that trials require standard operating procedures to ensure consistency across the trial and to provide standards against which trials can be audited. Perhaps one of the most important consequences of the EU Directive is that the reporting of adverse events has now been standardised. All serious adverse events (SAEs) need to be reported to the sponsor, who must determine whether or not the SAE is expected. Suspected unexpected serious adverse reactions need to be reported to the ethics committee, and to the Medicines and Healthcare Products Regulatory Agency. Although the aims of the EU directive are laudable, its arrival has not been met with uniform acceptance. There are concerns, particularly in the academic clinical trials community, that the additional work it will entail will reduce the number, size and quality of RCTs. In particular, an increasing stress on source data verification may use considerable resources for little obvious gain. It is argued that central monitoring can detect anomalies and errors in data more cost-effectively than onsite monitors (Buyse et al., 1999) and that occasionally monitors in centres have failed to detect problems identified using central checks (Enserink, 1996). Clearly, it is not
Research in cancer
an easy task to strike the right balance in the monitoring and regulation of trials. However, since the coming of the EU directive, a number of helpful web sites have been set up to guide trialists through the increasing maze of regulation, notably the Clinical Trials Toolkit (www.ct-toolkit.ac.uk).
Data monitoring committees: is the trial still ethical? As discussed earlier, a clinical trial only remains ethical while the question being addressed remains relevant. If there is good evidence that one treatment is better than another, then clearly, one group of patients will be receiving care that is known to be suboptimal, making the trial unethical. In some clinical trials, the strength of evidence in favour of one treatment over another may be such that an answer is known well before the scheduled end of the trial. Alternatively, the safety profile of one treatment may make it unacceptable. In both cases, there is a pressing argument to abandon the clinical trial in the face of such strong evidence. This is one of the roles of the Data Monitoring Committee (DMC). This group should be independent of the trial organisers, and the job of its members is to look at the accumulating data from the trial in light of other external evidence (e.g. other trial reports and new references). If they are convinced that the evidence is sufficiently strong and likely to change clinical practice were it revealed to clinicians and participants, then they can recommend closing the trial. The definition of ‘sufficiently strong’ is one that needs to be determined before the trial starts. There have been a number of different statistical stopping rules proposed (Ellenberg et al., 2002), but all of them realise that periodically looking at the data increases the chance of a chance positive result, and they therefore require a stronger level of evidence than p < 0.05. Some stopping rules set an extreme level of significance (such as p ≤ 0.002) throughout the trial, whereas others adapt depending on how close the trial is to completion. Increasingly as well, trials can be stopped for futility when there is strong evidence that the treatments being compared are not materially different. Alternatively, trials can be closed because of safety concerns if the incidence of adverse events in one group is unacceptably high, or, on occasion, if study recruitment is so slow that a reliable answer is unlikely. DMCs have another useful purpose: because interim analyses need to be performed, the preparation of a DMC report (which needs
to be kept secret from the trial organisers, clinicians, and participants) can identify problems with data collection methods or compliance with treatment.
Research networks As we have seen, to provide reliable results, clinical trials in cancer tend to need to be large. This poses a challenge for researchers: how does one recruit enough patients? One option would be to recruit over an extended period, but by the time the results are published, the question may no longer be relevant. A better approach is to foster the creation of a collaborative group to enable widespread recruitment. For example, the NCRI/MRC trials in AML recruit widely both in the UK and abroad. Typically, up to 200 clinicians take part in these trials, and annual recruitment is about 1000 patients, out of an incident population of around 2000. Such commitment to recruitment enables trials to recruit and report quickly, thereby advancing knowledge and practice. Recent initiatives in the UK have recognised the importance of fostering collaboration, and a number of clinical research networks have recently been introduced. In particular the National Cancer Research Network (NCRN) was set up in 2001 to improve the level of recruitment to trials and other well-designed research. The NCRN has set up 34 cancer research networks and employs data managers, nurses and other staff who can provide help and support to clinical trials. These important resources are available for trials that have been funded through the MRC and CRUK, as well as for other trials that have been formally reviewed and adopted as part of the NCRN trials portfolio. Other NCRN initiatives include the development of a computer infrastructure for clinical trials management and the identification of areas where systematic reviews of current evidence are required.
Health economics There is an increasing need for research to provide evidence of cost-effectiveness as well as clinical effectiveness. This is something specifically required by NICE both in approving individual interventions for the NHS as part of the appraisal process and in its clinical guidelines programme. Therefore, clinical trials often now include health economic outcomes, or a parallel study may be set up in a subset of patients or participating centres. Typically, this will involve trying to identify and record all the contacts that study participants have with 65
Robert Hills
the health services during the trial. These can then be costed, and the information used in combination with the effectiveness outcomes to provide information on cost-effectiveness. Quality-of-life data may also be collected using a standardised instrument and these data may be used to provide a cost-utility estimate, such as the ‘quality-adjusted life-year.’
Translational research There is an increasing amount of laboratory cancer research being carried out and with it comes the need to determine whether novel findings in the laboratory can translate into testable hypotheses in patients and ultimately improved outcomes. This ‘bench-to-bedside’ approach should lead ultimately to the development of targeted therapies for a wide range of conditions. This approach can be challenging. Trials need to be designed to collect additional disease markers to identify patients who are likely to benefit from targeted therapies, and a concentration on specific subgroups of patients brings with it the problem of recruiting enough patients to provide reliable and meaningful results. Ethically, participants in trials will need to know what samples will be taken, how these will be stored, and how results from any samples will be used. Additionally, gene array technology has dramatically increased the number of markers available for analysis and is a source of hope for identifying prognostic markers that can then be used to help develop targeted therapies. Of course, the challenge, when there are a large number of potential markers and combinations of markers, is to identify which markers are genuinely prognostic and which are merely artefactual. This new area of laboratory research has led to the development of methods for testing which avenues are the most promising for future research. In the UK, the National Translational Cancer Research Network is supporting infrastructure to help facilitate translational research in cancer.
Synthesising research results: systematic reviews and meta-analysis When deciding the best treatment for a particular patient, it is important to consider all the available research evidence. Whereas one RCT may show a significant benefit for a new treatment, others may not; therefore, it is important to be able to put trial results in context. Additionally, new trials need to be designed 66
with reference to the findings of a systematic review of previous research. This will reduce the chance of repeating previous mistakes and will also stop unnecessary (and therefore unethical) trials from being conducted when there is already enough evidence about the effectiveness of a given treatment. As the editors of the Lancet have said: ‘Unnecessary and badly presented clinical research injures volunteers and patients as surely as any other form of bad medicine, as well as wasting resources and abusing the trust placed in investigators by their trial participants’ (Young and Horton, 2005). The journal now requires authors to include a clear summary of previous research findings. Of course, to summarise previous research findings, the previous research must be in the public domain. In the past, trials with negative or inconclusive results were considered uninteresting and were much less likely to be published (so-called ‘file drawer’ or publication bias; Dickersin et al., 1987). Such problems have led to the setting up of a number of online journals where such results can be published and put into the public domain. Otherwise, a search of the published literature would tend only to identify trials that showed significant benefits, giving an unduly optimistic view of a treatment’s effectiveness. Once relevant research has been identified and systematically reviewed for its quality, it is important to be able to synthesise and summarise the results. In most cases this may just be a narrative summary that describes and comments on all the available evidence; but it may be possible to carry out a statistical synthesis. Meta-analysis is a powerful tool for doing this: it provides quantitative estimates of a treatment’s effectiveness by combining data from a number of RCTs. One advantage of doing such an analysis is that it can identify important treatment benefits which previous trials had been too small to detect reliably. One such example comes from the Early Breast Cancer Trialists’ Collaborative Group meta-analysis of tamoxifen in breast cancer (Early Breast Cancer Trialists Collaborative Group, 1990). Of the 28 randomised trials of tamoxifen versus no treatment analysed in the first cycle of the overview, only 4 showed a statistically significant survival benefit. Yet, a meta-analysis of their findings demonstrated a highly statistically significant (p < 10−6 ) 16% proportional reduction in mortality achieved with tamoxifen. Previous trials had been too small to detect such an improvement reliably. Only by using a meta-analysis was it possible to demonstrate the utility of a treatment that has since annually saved many thousands of lives.
Research in cancer
Box 5.5 Trials in context.
r When developing the research question, conduct a systematic review or identify a relevant review done by someone else. r Learn from the achievements (and mistakes) of past trials but read critically. r Discuss the findings of your study in the context of an updated systematic review. There are, however, dangers in carrying out an inappropriate meta-analysis. If there is a large amount of heterogeneity between trials, then merely reporting the total treatment effect can give misleading results. For this reason it is important to explore sources of heterogeneity (e.g. different end points, doses, dosing regimens, or different patient groups) to determine whether such heterogeneity materially affects the results of the meta-analysis and the estimate of effectiveness. A meta-analysis or other review is only as good as the data that go into it and needs to be continually updated in light of new results. After a new clinical trial has been completed, the data need to be looked at in relation to previous trials and the impact of the results discussed. This is best achieved by adding the trial into the existing meta-analysis and interpreting any effect that it has on the overall result. It is, after all, the totality of trial data that provides the best evidence, and overemphasis on any particular trial, even the one you conducted, can give misleading findings. Box 5.5 gives guidelines for putting results in context.
Assessing published reports of research All the ideas presented in this chapter apply equally well to designing one’s own clinical trial as they do to assessing other people’s research. The obvious questions that one needs to ask when reading research are whether the results appear believable and whether the methodology is sound. Few pieces of research are without some methodological flaws (Altman, 1994), but do these flaws invalidate the findings? In assessing the merits of different treatments we have already seen that different types of study carry different weight when making evidence-based decisions on health care. There is a generally accepted order of precedence (Guyatt et al., 1995): 1. Systematic reviews and meta-analyses. 2. RCTs with definitive results (based on confidence intervals). 3. RCTs with non-definitive results.
4. 5. 6. 7.
Cohort studies. Case-control studies. Cross-sectional studies. Case reports. As can be seen, for results of an assessment of a treatment to be reliable, a study needs to be randomised and be of sufficient size to provide reliable evidence. Results need to be reported fairly, with no undue emphasis on particular subgroups or end points. Faced with a paper, it can sometimes be quite difficult to check whether the report matches the original design of the study. However, a number of journals now allow (and encourage) trialists to publish their study protocols online, allowing readers to make a direct comparison between the results obtained and the originally proposed methodology. Additionally, some of the larger medical journals require trials to be reported according to the CONSORT guidance (Moher et al., 2001). Perhaps the most important aspect of the CONSORT guidance, and the one that will be most familiar, is the requirement to include in a paper a flowchart showing the journey of the patient population through the trial, showing how many patients did not receive their treatment according to the protocol and how many patients dropped out from (or were lost to) follow-up. However, the CONSORT statement makes a number of other recommendations, including the use of confidence intervals rather than merely p-values, and reporting of eligibility criteria, adverse events, the method of randomisation and the generalisability of findings. However, even with perfect reporting, not all randomised trials are of equal quality (and, hence, equal reliability). The trial report may demonstrate that the randomisation sequence could have been subverted, but CONSORT is not a device for improving the design of trials; and a poorly designed trial may well lead to misleading results. There are many methods of assessing the quality of trials, from simple checklists to quality scores. Meta-analyses can then be performed with reference to these quality scores to determine whether the quality of the trial affects the result. A recent monograph (Moher et al., 1999) concluded that more research was needed in this area to determine the effect that trial quality, and different assessments of trial quality, have on systematic reviews and, ultimately, evidence-based practice.
Conclusions Without proper evidence, evidence-based practice is impossible. In assessing the effects of new treatments, it 67
Robert Hills
is impossible to overstate the importance of RCTs, and meta-analyses of such trials. But randomisation alone is not enough. Trials need to be well designed, use appropriate end points and be properly analysed before their results can be fed into clinical practice. This chapter has identified important aspects of researching new treatments in cancer, and applying these guidelines, together with a healthy scepticism and one’s innate common sense, should help produce better research and more reliable interpretations of research. It is often said that learning from one’s mistakes is fruitful. It is also true that learning from others’ mistakes is less painful, so critical reading of existing research is a very good way of starting to design one’s own trial.
REFERENCES Altman, D. G. (1991). Statistical Methods for Medical Research, 2nd edn. London: Chapman and Hall. Altman, D. G. (1994). The scandal of poor medical research. B. M. J., 308, 283–4. Altman, D. and Bland, J. M. (1995). Absence of evidence is not evidence of absence. B. M. J., 311, 485. Altman, D. and Bland, J. M. (1999). How to randomise. B. M. J., 319, 703–4. Altman, D. and Bland, J. M. (2005). Treatment allocation by minimisation. B. M. J., 330, 843. Altman, D. G., Machin, D., Bryant, T. N. et al. (2000). Statistics with Confidence. London: BMJ Books. Assmann, S. F., Pocock, S., Enos, L. E. et al. (2000). Subgroup analysis and other (mis)uses of baseline data in clinical trials. Lancet, 355, 1064–9. Burman, W. J., Reves, R. R., Cohn, D. L. et al. (2001). Breaking the camel’s back: multicenter clinical trials and local institutional review boards. Ann. Intern. Med., 134, 152–7. Burnett, A. K., Milligan, D. W., Prentice, A. G. et al. (2005). Modification or dose or treatment duration has no impact on outcome of AML in older patients: preliminary results of the UKNCRI AML14 trial. Blood, 106 (162A), Abstr. 543. Buyse, M., George, S. L., Evans, S. et al. (1999). The role of biostatistics in the prevention, detection and treatment of fraud in clinical trials. Statist. Med., 18, 3435–51. Cascinelli, N. (1994). Adjuvant interferon in melanoma – reply. Lancet, 343, 1499 (letter). Cascinelli, N., Bufalino, R., Morabito, A. et al. (1994). Results of adjuvant interferon study in WHO melanoma programme. Lancet, 343, 913–4. Collins, R., Peto, R., Gray, R. et al. (1996). Large-scale evidence: trials and overviews. In Oxford Textbook of Medicine, Vol. 1, ed. D. Weatherall, J. G. G. Ledingham and D. A. Warrell, 3rd edn. Oxford: Oxford University Press. Creutzig, U., Ritter, J., Zimmermann, M. et al. (2001). Idarubicin improves blast cell clearance during induction therapy in children with AML: results of study AML-BFM 93. Leukemia, 15, 348–54.
68
Cullen, M., Steven, N., Billingham, L. et al. (2005). Antibacterial prophylaxis after chemotherapy for solid tumors and lymphomas. N. Engl. J. Med., 353, 988–98. Dickersin, K., Chan, S., Chalmers, T. C. et al. (1987). Publication bias and clinical trials. Contr. Clin. Trials, 8, 343–53. Duley, L. and Farrell, B. (2002). Clinical Trials. London: BMJ Books. Early Breast Cancer Trialists Collaborative Group. (1990). Treatment of Early Breast Cancer: Volume 1, Worldwide Evidence 1985–90. Oxford: Oxford University Press. Edwards, P., Roberts, I., Clarke, M. et al. (2002). Increasing response rates to postal questionnaires: systematic review. B. M. J., 324, 1183. Ellenberg, S., Fleming, T. and DeMets, D. (2002). Data Monitoring Committees in Clinical Trials: A Practical Perspective. Chichester: Wiley. Enserink, M. (1996). Clinical trials: fraud and ethics charges hit stroke drug trial. Science, 274, 2004–5. Green, S. B. and Byar, D. P. (1984). Using observational data from registries to compare treatments – the fallacy of omnimetrics. Stat. Med., 3, 361–70. Guyatt, G. H., Sackett, D. L., Sinclair, J. C. et al. (1995). Users’ guides to the medical literature IX: A method for grading healthcare recommendations. J. A. M. A., 274, 1800–4. Machin, D., Campbell, M. J., Fayers, P. M. et al. (1997). Sample Size Tables for Clinical Research, 2nd edn. Oxford: Blackwell. Miller, F. G. and Kaptchuk, T. J. (2004). Sham procedures and the ethics of clinical trials. J. R. Soc. Med., 97, 576–8. Moher, D., Cook, D. J., Jadad, A. R. et al. (1999). Assessing the quality of reports of randomised trials: implications for the conduct of meta-analyses. Health Technol. Assess., 3 (12), i–iv, 1–98. Moher, D., Schulz, K. F. and Altman, D. G. for the CONSORT group. (2001). The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials. Lancet, 357, 1191–4. Pocock, S. J. (1996). Clinical Trials: A Practical Approach. London: John Wiley and Sons. QUASAR Collaborative Group. (2000). Comparison of fluorouracil with additional levamisole, higher-dose folinic acid, or both, as adjuvant chemotherapy for colorectal cancer: a randomised trial. Lancet, 355, 1588–96. Schulz, K. F. (1995). Subverting randomization in controlled trials. J. A. M. A., 274, 1456–8. Specht, L., Gray, R. G., Clarke, M. J. et al. for the International Hodgkins Disease Collaborative Group. (1998). Influence of more extensive radiotherapy and adjuvant chemotherapy on long-term outcome of early-stage Hodgkin’s disease: a meta-analysis of 23 randomised trials involving 3,888 patients. J. Clin. Oncol., 16, 830–43. Storer, B. E. (1989). Design and analysis of Phase I clinical trials. Biometrics, 45, 925–37. Storer, B. and DeMets, D. (1987). Current Phase I/II designs: are they adequate? J. Clin. Res. Drug Devel., 1, 121–30. Swinscow, T. D. V. and Campbell, M. J. (2002). Statistics at Square One. London: BMJ Books. Wheatley, K. (2002) SAB: a promising new treatment for AML in the elderly? Br. J. Haematol., 118, 432–3. Wheatley, K. and Hills, R. K. (2001). Inappropriate reporting and interpretation of subgroups in the AML-BFM 93 study. Leukemia, 15, 1803–4.
Research in cancer
Young, C. and Horton, R. (2005). Putting clinical trials into context. Lancet, 366, 107–8. Yusuf, S., Collins, R. and Peto, R. (1984). Why do we need some large, simple randomized trials? Stat. Med., 3, 409–22.
FURTHER READING There are many excellent books and articles on different aspects of clinical trials, notably those by Altman (1991), Assmann et al. (2000), Collins et al. (1996), Duley and Farrell (2002), and Yusuf et al. (1984). Additionally, the following references are valuable resources. Chalmers, I. (1993). The Cochrane Collaboration: preparing, maintaining and disseminating systematic reviews of the effects of health care. Ann. N. Y. Acad. Sci., 703, 156–63.
The DAMOCLES Study Group. (2005). A proposed charter for clinical trial data monitoring committees: helping them do their job well. Lancet, 365, 711–22. Greenhalgh, T. (2006). How to Read a Paper, 3rd edn. Oxford: Blackwell. Peto, R. (1987). Why do we need systematic overviews of randomized trials? Stat. Med., 6, 233–44. Peto, R., Pike, M. C., Armitage, P. et al. (1976). Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Part I: introduction and design. Br. J. Cancer, 34, 585–612. Peto, R., Pike, M. C., Armitage, P. et al. (1977). Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Part II: analysis and examples. Br. J. Cancer, 35, 1–39.
69
6
ONCOLOGICAL EMERGENCIES Paul Shaw
Introduction
Treatment overview
An oncological emergency is an acute medical problem related to cancer or its treatment, which may result in serious morbidity or mortality if not treated quickly. It may be secondary to a structural/obstructive, metabolic or treatment-related complication (Cervantes and Chirivella, 2004). The emergency may be the first manifestation of malignant disease, particularly for superior vena cava obstruction (SVCO) and malignant spinal cord compression (MSCC). Spinal cord compression is the commonest neurological complication of cancer, occurring in approximately 5% of all cancer patients. Around 10% of advanced solid tumours give rise to malignant hypercalcaemia.
As with any acute medical emergency, resuscitation measures may be needed to ensure that airway, breathing, and circulation are maintained. Adequate hydration, oxygen, and monitoring of fluid balance are important in patients with sepsis or tumour lysis syndrome. Steroids are used in patients with SVCO and suspected spinal cord compression, although the evidence base supporting their use is poor. Mannitol infusions may be needed for severe symptomatic raised intracranial pressure that does not respond to steroids. Pain, breathlessness, and distress should be treated as priorities, especially in patients presenting with end-stage cancer and an oncological emergency. The WHO pain ladder is a suitable framework to guide appropriate analgesic use. Some seriously ill patients may need to be transferred to a high-dependency unit (HDU) or intensive therapy unit (ITU), especially those with a treatable malignancy and a good prognosis and those who develop complications of curative chemotherapy. Liaison with specialist colleagues at an early stage is recommended. An oncological emergency, like any other emergency, requires prompt assessment and action by appropriately experienced staff. Sometimes the emergency situation may be predictable, in which case a previously agreed plan of action will be helpful. Thought must be given to the appropriateness and value of investigations and treatment because some patients will be in the terminal phase with progressive and treatment-refractory disease. End-of-life care should be instituted rapidly as a priority to relieve unnecessary distress.
Types of emergency Metabolic emergencies include the following: r Hypercalcaemia. r Syndrome of inappropriate antidiuretic hormone (SIADH). Structural/obstructive emergencies include r MSCC and cauda equina compression. r SVCO. r Raised intracranial pressure. r Acute airway obstruction. r Bleeding. r Urinary obstruction. r Cardiac tamponade. r Pain: this has been named the ‘fifth vital sign’ following pulse, blood pressure, temperature and respiration; when pain is present it should evoke an immediate response. Treatment of pain is considered in Chapter 7. Treatment-related emergencies include r Neutropenic fever/sepsis. r Anaphylaxis related to a chemotherapeutic agent. r Tumour lysis syndrome. r Extravasation of a chemotherapeutic agent. 70
Metabolic emergencies Hypercalcaemia Definition Hypercalcaemia occurs when corrected calcium is greater than 2.6 mM/l. It is caused by direct bone destruction by bone metastases or, in patients without
Oncological emergencies
bone metastases, by circulating factors such as parathyroid hormone-related peptide (PTH-RP). PTH-RP is associated particularly with lung, breast, renal, head and neck carcinomas, myeloma and lymphoma. Patients with cancer may also have incidental primary hyperparathyroidism.
Presentation Hypercalcaemia occurs in 10% of patients with advanced solid tumours; symptoms are fatigue, anorexia, nausea, vomiting, confusion, abdominal pain/constipation, polyuria and polydipsia. If left untreated it leads to somnolence, coma and death.
Investigations Serum calcium level, electrolytes and renal biochemistry.
Treatment if corrected calcium is less than 3.0 mM/l Asymptomatic patients with a corrected calcium of less than 3.0 mM/l who are about to have chemotherapy or radiotherapy should be rehydrated, kept mobile, and monitored regularly. Patients who are symptomatic or who are expected to have a slow response to anticancer treatment should be treated as follows.
Treatment if corrected calcium is greater than or equal to 3.0 mM/l or patient is symptomatic Fluid replacement: r Give at least 3 l of sodium chloride 0.9% in 24 hours. r Stop thiazide diuretics. r Furosemide should not be given until dehydration has been treated, but it does increase calcium excretion. Bisphosphonates: r Zoledronic acid has a fast onset and a long duration of action. r Ensure adequate hydration, then give a dose of 4 mg zoledronic acid in at least 50 ml of either sodium choride 0.9% or glucose 5% over 15 minutes. r Monitor renal function and serum calcium, phosphate and potassium. r Side effects of bisphosphonates include gastrointestinal upset, flu-like symptoms and exacerbation of metastatic bone pain. Osteonecrosis of the mandible can occur after chronic use. r Hypocalcaemia occurs in 50% of patients but it rarely causes symptoms because of a compensatory increase in PTH levels secondary to the decreased calcium levels.
Other drugs used less frequently for hypercalcaemia include the following: r Calcitonin. Salcatonin has a very rapid onset and may be used in patients with dangerously high serum calcium levels, regardless of fluid status. Tachyphylaxis develops, so use of salcatonin is limited. r Gallium nitrate. r Mithramycin.
Syndrome of inappropriate antidiuretic hormone (SIADH) SIADH is caused by excess levels of antidiuretic hormone (ADH), which leads to failure to excrete dilute urine, and so water retention and low serum sodium levels occur. It is most commonly associated with smallcell lung cancer (SCLC). It occasionally occurs in a wide variety of other cancers and can also be caused by chest infections, hypothyroidism and drugs such as antidepressants, ACE inhibitors, cyclophosphamide and cisplatin.
Presentation Patients are often asymptomatic but may experience fatigue, lethargy, nausea, anorexia, muscle cramps, depression and behavioural changes; it may also be an incidental finding. If the serum sodium is less than 110 mM, somnolence, depressed deep tendon reflexes, pseudobulbar palsy, seizure, coma and death may occur.
Investigation Hyponatraemia and reduced plasma osmolarity occur in the presence of inappropriately concentrated urine (urine osmolarity > 500 mM/kg). Renal failure, hypothyroidism and adrenal insufficiency are excluded by checking biochemistry, TFTs and a short synacthen test.
Treatment
r Fluid restriction to 0.5 to 1 l/day usually results in symptomatic and biochemical improvement.
r Demeclocycline (600 to 1200 mg/day) should be given if the patient does not respond to fluid restriction alone. Demeclocycline has a diuretic effect by altering the effect of ADH on renal fluid resorption. r The underlying malignancy should be treated to reverse the cause of electrolyte imbalance. r In an emergency situation, such as altered consciousness or fitting, cautious administration of 71
Paul Shaw
intravenous 1.8% sodium chloride may be considered. However, a rapid rise in sodium may cause osmotic demyelination syndrome and, therefore, the infusion should be reduced or stopped as soon as the patient’s neurological condition improves.
Structural/obstructive emergencies Malignant spinal cord and cauda equina compression This condition is caused by pressure from tumour (growing directly between vertebral bodies or growing from bone metastases) or a collapsed vertebral body on the spinal cord or cauda equina. This may then cause paraparesis, or paraplegia, loss of sensation and bladder or bowel dysfunction. It is particularly associated with breast cancer, lung cancer and prostate cancer.
Presentation Symptoms and signs will depend on the level of compression. Because the spinal cord ends at about the L1 level, compression above this point will give an upper motor neurone pattern of weakness and below this point, a lower motor neurone pattern. Muscle weakness often occurs before sensory loss or autonomic dysfunction (impotence, urinary or faecal incontinence or retention). Approximately 60% of patients will have pain, which may be radicular. Asymptomatic cord compression is estimated to occur in one-third of patients with prostate cancer and bone metastases (Bayley et al., 2001).
Investigation An MRI of the whole spine is performed to detect the level of compression. Multiple levels of compression may be found. Intravenous contrast improves the detection of intradural and intramedullary tumours. In patients who have not been previously diagnosed with cancer it is important to establish a tissue diagnosis if possible (e.g. by needle biopsy or during surgical decompression).
Initial treatment Initial treatment includes the following: r Dexamethasone 16 mg is given daily in divided doses with proton pump inhibitor protection, reducing the dose after a few days to avoid toxicity. r The patient should be nursed in the supine position if there is any possibility of spinal instability.
72
r Patients require a multidisciplinary team approach including specialist physiotherapy and nursing care.
Surgery In selected patients with a single area of MSCC from a solid tumour, the benefit of immediate circumferential decompression of the spinal cord followed by 10 × 3 Gy fractions of radiotherapy has recently been shown to be superior to radiotherapy alone (Patchell et al., 2005). This study was closed early after the accrual of 123 patients when the surgical arm was superior at interim analysis, with the percentage of patients retaining the ability to walk after surgery being greater than those receiving radiotherapy alone (84 versus 57%; p = 0.001). In addition 10 out of 16 paraplegic patients gained the ability to walk following surgery as opposed to only 3 of 16 in the radiotherapy-only group. It is important that patients would need to be carefully selected for this approach. In this study patients were included if they had a good performance status, predicted survival greater than 3 months and had not been paraplegic for more than 48 hours.
Role of radiotherapy Radiotherapy is the most commonly used treatment of MSCC. Patients should receive radiotherapy as soon as is practically possible after arrival in the oncology unit. Referrals occur most frequently on Friday afternoons, as confirmed by a retrospective review of 443 patients treated over 10 years (Poortmans et al., 2001). A radiotherapy technique is as follows: r Patient preparation, positioning and immobilisation: patient lies prone or supine, with arms alongside. Polystyrene knee support and/or head support may be used. r Localisation and target volume: simulator films are taken at the level of spinal cord compression. After cross-sectional imaging, the target volume includes the level of compression and one vertebral body above and below this level, usually ensuring that the inferior and superior limits cross an intervertebral space. Typically the field borders are defined at the time of simulation. The centre of the field is in the midline along the spinus processes and has a usual width of 8 cm. r Plan: a single posterior field is used most often. But upper cervical cord compression can be treated using opposed lateral fields to avoid having the exit beam pass through the throat and mouth. A typical field
Oncological emergencies
length for a posterior treatment field would be 10 to 15 cm. r Dose, fractionation and energy: for patients with metastatic disease, typical doses include 20 Gy in 5 daily fractions or 30 Gy in 10 daily fractions given as either an applied dose or prescribed at the depth of the spinal cord (which is determined using the MRI scan) using 6-MV photons. If opposed lateral beams are used then the same dose may be given but it is prescribed to the ICRU reference point (centre of the intersecting beams). For patients with primary tumours, such as solitary plasmacytoma of bone, a higher dose may be required with the aim of achieving a cure (see Chapter 31, p. 366).
Areas of current interest A recent retrospective study of radiotherapy dose has compared 8 Gy in a single fraction, 20 Gy in 5 fractions over 1 week, 30 Gy in 10 fractions over 2 weeks, and 40 Gy in 20 fractions over 4 weeks in patients with non-small-cell lung cancer (NSCLC; Rades et al., 2006). The functional outcome was equivalent for short-course regimens (8 and 20 Gy) as well as for long-course treatment, and the authors concluded that 8 Gy in a single fraction is an appropriate dose in patients with NSCLC who generally have a poor prognosis. The outcome was related to the development time of motor deficit prior to radiotherapy (> 14 days being better than a shorter time interval). Overall there was an improvement in motor function in 14% of patients, no change in 54%, and deterioration in 32%.
Superior vena caval obstruction (SVCO) SVCO is caused by compression, invasion or occasionally intraluminal thrombus. It is usually associated with SCLC, NSCLC and lymphoma (NHL more frequently than HL), which account for more than 90% of cases; thymoma and germ cell cancer are rarer causes.
Presentation The onset is typically insidious over weeks and results in compensatory collateral venous channels in the territory of the SVC. The symptoms are worse on bending forwards and they include neck and face swelling, conjunctival suffusion, headache, nasal congestion, epistaxis, dizziness and syncope. Examination reveals raised jugulo-venous pressure, venous collaterals, arm oedema and plethora. Occasionally there is
sudden occlusion and the patient becomes acutely unwell.
Investigation SVCO is rarely an acute emergency, unless there is associated airway obstruction and stridor, and so there is usually time for further investigation aimed at making a histological diagnosis (Ostler et al., 1997). A chest X-ray often shows a widened mediastinum. A CT scan provides information about both the site and the cause of the obstruction, by distinguishing external compression from intravascular thrombosis. It can also give staging information about the underlying tumour. An attempt should be made to establish a histological diagnosis before starting treatment, either by CT-guided fine needle aspiration or biopsy or by bronchoscopy, sputum cytology or biopsy of enlarged neck nodes. Sometimes mediastinoscopy or thoracoscopy may be required (Ostler et al., 1997).
Treatment Treatment includes the following: r Initial management: sit the patient up, giving oxygen as required and steroids (e.g. dexamethasone 12 to 16 mg daily in divided doses), with proton pump inhibitor protection. r A stent can be a useful holding measure in patients who require urgent symptom relief, and this will often allow time to establish a tissue diagnosis. r Thrombosis should be appropriately treated if present. r Specific treatment is tailored to the individual underlying disease. r Chemotherapy is indicated for patients with chemosensitive tumours such as lymphoma, germ cell tumour or SCLC. r Radiotherapy is the mainstay of treatment for patients with other solid tumours. A Cochrane systematic review concluded that chemotherapy and radiotherapy were equally effective at relieving SVCO secondary to lung cancer, whereas stent insertion provided higher rates of response more rapidly (Rowell and Gleeson, 2001). It seems appropriate that highly responsive tumours (NHL, HD, SCLC) should be treated with chemotherapy first, whereas patients with NSCLC and severe SVCO should get a stent first. A radiotherapy technique is as follows: r Patient preparation positioning and immobilisation: the patient lies supine, with a headrest and arms by
73
Paul Shaw
his or her sides. If the patient is unable to lie flat, the sitting position is used, with arms by sides. r Localisation and target volume: simulated treatment fields or, where available, CT planning is used with the target volume to cover the superior vena cava, tumour and mediastinum. r Dose, energy and fractionation: palliative fractionation regimens may be used (e.g. 10 Gy in single fraction or 16 Gy in two fractions) if the patient’s performance status and prognosis are poor and if the field size is limited to 12 × 12 cm. This may result in quick symptomatic relief with minimal visits to the radiotherapy department. Patients with localised disease and better performance status may benefit from a higher dose (e.g. 36 Gy in 12 fractions over 2.5 weeks) in the hope that the tumour and symptomatic control will be prolonged. Occasionally radiotherapy is given with radical intent in appropriately selected patients such as those with NSCLC whose disease can be encompassed in a radical volume. In these patients, neoadjuvant chemotherapy should also be considered.
Raised intracranial pressure Raised intracranial pressure results from the spaceoccupying effect of intracranial tumours, which are most often metastatic.
Presentation Patients present with headache, nausea, vomiting, visual disturbance, seizure, ataxia and changes in personality and behaviour. Clinical findings may include visual field loss, loss of spontaneous retinal venous pulsations, papilloedema and sixth cranial nerve palsy. There may be a reduced consciousness level, slow pulse rate and raised blood pressure. Herniation of the cerebral peduncle can result in hemiplegia. Focal signs may present depending on the site of the tumour.
and possibly other sites of metastatic disease (see Chapter 38, p. 442).
Treatment Treatment is as follows: r Give dexamethasone 12 to 16 mg daily in divided doses with proton pump inhibitor cover. r If severe (rapidly falling Glasgow Coma Score or moribund state), give mannitol 0.5 to 1 g/kg i.v. over 15 minutes. r Avoid fluid overload. r Provide analgesia for headache (paracetamol with or without NSAID or stronger analgesia as documented in WHO pain ladder). r For specific anticancer treatment, if a single metastasis is found, neurosurgery plus radiotherapy may be appropriate. If there are multiple metastases, patients who are most likely to benefit from whole-brain radiotherapy are those who are mobile and have had a good symptomatic benefit from steroids. A randomised controlled trial has shown that 12 Gy in 2 fractions on consecutive days is not inferior to 30 Gy in 10 fractions over 2 weeks in patients with symptomatic cerebral metastases and poor performance status who need treatment (Priestman et al., 1996).
Special case: obstructive hydrocephalus Patients presenting de novo with tumours causing obstruction to the flow of cerebrospinal fluid and resulting in obstructive hydrocephalus should be considered for ventricular-peritoneal shunt insertion. Immediate resuscitation with mannitol and steroids should be instituted.
Acute airway obstruction Acute airway obstruction is blockage of the main-stem bronchi, carina, trachea or larynx and is commonly caused by direct tumour extension from lung cancer or head and neck cancer.
Investigation A CT scan of the brain will detect most brain metastases. MRI has a greater sensitivity for detecting small metastases, meningeal disease and fourth ventricle obstruction. In most patients who have not been previously diagnosed with cancer, efforts should be made to obtain a histological diagnosis, especially if there is a single brain tumour. Such cases should be discussed with the neurosurgical team. A biopsy might not be appropriate in frail patients who have multiple cerebral metastases 74
Presentation The patient presents with dyspnoea and stridor.
Investigation Patients with upper airway obstruction should have direct visualisation by laryngoscopy or bronchoscopy according to the level of obstruction. Clinical clues as to the level of obstruction include the presence of neck swelling and stridor with upper obstruction or
Oncological emergencies
Table 6.1. Endobronchial intervention in lung cancer
Bleeding
Lung pathology
Endoscopic intervention
Bleeding is more likely to occur in patients treated with anticoagulants.
Bleeding from central airway
Argon plasma coagulator
Examples
tumour Intraluminal tumour
Nd-YAG laser Electrocautery Argon plasma coagulator Cryotherapy Photo-dynamic therapy
Intramural tumour
HDR endobronchial
Extrinsic compression/
Airway stent
brachytherapy (192 Ir) airway wall destruction Adapted from Freitag (2004).
monophonic wheeze (on auscultation) with lower airway obstruction. Chest X-ray or CT scan of neck and thorax or both should be considered.
Treatment Treatment is as follows: r Heliox (79% He, 21% O2 ) contains helium, which has a lower density and therefore lower specific gravity than oxygen, nitrogen or air. As a result, during turbulent flow, the flow velocity will be higher when heliox is used. This reduces the work required to breathe when the upper airway is obstructed. r If the upper airway is acutely compromised, emergency tracheostomy or endotracheal intubation may be required. r Specific intervention should be designed to diagnose and treat the obstruction.
Interventional bronchoscopy For the majority of patients, external beam radiotherapy is all that is required. However, Table 6.1 shows the options available (adapted from Freitag, 2004), which can be combined with chemotherapy and/or radiotherapy. These endobronchial treatments have been reviewed by Morris et al. (2002), who concluded that ‘good to excellent short term palliation’ may be achieved.
External beam radiotherapy Patients are usually treated with 20 Gy in 5 fractions to reduce the chance of larger single fractions increasing oedema.
Massive haemoptysis is most commonly associated with lung cancer but is also associated with endobronchial metastases from carcinoid, breast, kidney, sarcoma and colon cancers. It is defined as expectoration of more than 100 ml of blood in a single episode during 24 to 48 hours. It may be associated with respiratory difficulty and can lead to rapid deterioration with airway obstruction, anaemia and hypovolaemic shock. It may be associated with coagulation disorder, thrombocytopenia, or fungal infection. In haematemesis, approximately 2 to 5% of upper GI bleeding is related to malignancy. Even in patients with cancer, haematemesis may be caused by benign disease (peptic ulcer disease, oesophagitis, gastritis, duodenitis). Mallory-Weiss tears may be secondary to vomiting induced by chemotherapy, renal failure or advanced malignancy (Palmer, 2004). With the move from surgical treatment towards chemotherapy with or without radiotherapy for the treatment of primary gastric lymphoma (commonest extranodal site of NHL), the incidence of GI haemorrhage is estimated at 5% (Maisey et al., 2004). Oesophagogastric tumours rarely present with acute GI haemorrhage. The Rockall score is a risk assessment tool for GI haemorrhage (Rockall et al., 1996: see Table 6.2). Patients with a Rockall score of six or more have a predicted mortality of around 50%. Multivariate analysis identifies shock, age, co-morbidity and specific endoscopic findings as independent variables predicting re-bleeding and death (Palmer, 2004). Haematuria may occur with malignant tumours involving the genitourinary tract but it occurs most commonly in renal, bladder and prostate cancer. Patients may present with asymptomatic haematuria, associated symptoms related to the underlying cancer, or severe pain caused by clot retention.
Assessment and investigation Secure the airway, breathing, and circulation first. Then r Perform a full blood count, clotting screen and renal and liver profile. r Perform a CT scan or endoscopy (e.g. bronchoscopy, upper or lower GI endoscopy, cystoscopy) to establish diagnosis and identify the site of bleeding. These tests also provide prognostic information in the case 75
Paul Shaw
Table 6.2. The Rockall scoring system Variable
Score 0
Score 1
Age (years)
< 60
60–79
≥ 80
–
Shock
None
Pulse > 100 bpm,
Pulse > 100 bpm, systolic BP
–
Co-morbidity
None
–
normal BP
Score 2
Score 3
< 100 mmHg Cardiac failure, ischaemic
Renal failure, liver failure,
heart disease, other major
disseminated malignancy
co-morbidity Diagnosis
Mallory-Weiss tear,
All other
no lesion seen,
Malignancy of upper GI tract
–
Blood in upper GI tract,
–
diagnoses
no SRH Major SRH
None
–
adherent blood clot, visible or spurting vessel BP = blood pressure; bpm = beats per minute; SRH = stigmata of recent haemorrhage. Three clinical variables (age, shock and co-morbidity) and two endoscopic variables (diagnosis and major SRH) are each categorised as shown in the table. A score of 0 to 3 points is awarded for each category, giving a maximum total score of 11. Patients with a Rockall score of 6 or more have a predicted mortality of around 50%. Adapted from Rockall et al. (1996).
of upper GI bleeding to direct the appropriate level of care (Palmer, 2004). r Perform a urine or sputum microscopy and culture as appropriate.
Treatment For patients who require active resuscitation/intervention: r Secure airway, breathing and circulation. r Use fluid resuscitation with normal saline or colloid to restore blood pressure and urine output, which can be monitored by measuring the CVP. r Patients with haemoglobin below 10 g/dl should receive a blood transfusion. r Recognise and treat underlying renal impairment, and cardiovascular or cerebrovascular disease, because co-morbid conditions can decompensate in the presence of acute haemorrhage. r Consider tranexamic acid (with caution in haematuria because of risk of clot retention) or specific measures for a bleeding disorder (e.g. platelet transfusion, vitamin K, fresh frozen plasma). r Actively bleeding and shocked patients should be managed in a high-dependency unit assuming this is appropriate for the individual patient. Site-specific interventions include the following: r Haemoptysis – bronchoscopy/radiotherapy as discussed earlier. 76
r Haematemesis options: r Drug therapy (e.g. proton pump inhibitors, somatostatin).
r Endoscopic therapy (e.g. direct injection of adrenaline into bleeding ulcers, effective in 90%), fibrin glue and human thrombin (Palmer, 2004) and heat and mechanical devices. r Radiotherapy to the tumour bed, single fraction of 8 Gy or 20 Gy in five fractions over 1 week. r Haematuria options: r Radiotherapy to the prostate or bladder with a single fraction of 8 Gy (or a planned volume for radical treatment). r Cystoscopy with electrocautery/laser. r Renal artery embolism for renal tumours bleeding into the urogenital tract. For patients in the terminal phase of advanced malignancy who experience massive and uncontrollable bleeding such as carotid blow-out or massive haemoptysis, intravenous midazolam and diamorphine provides rapid sedation and palliation.
Urinary obstruction This is associated with urological or gynaecological tumours, especially carcinoma of prostate or cervix. Recurrent rectal cancer or pelvic metastases may result
Oncological emergencies
in bilateral ureteric dilatation and hydronephrosis. Constipation is a reversible cause.
apy to the pericardium could be considered: 30 Gy in 10 fractions over 2 weeks.
Presentation The patient may be asymptomatic or present with flank pain, anuria and raised creatinine. Partial obstruction may present with alternating polyuria and oliguria. Urinary tract infection may occur because of the obstruction.
Investigation A renal tract ultrasound may show bilateral hydronephrosis. A CT scan of the pelvis can be used to identify the site and cause of obstruction, especially a retroperitoneal or pelvic mass.
Treatment The decision of whether to place stents should take into account the patient’s performance status, stage of disease, and chance of response to anticancer treatment. Ureteric stents may be placed under local anaesthetic or, if this is not feasible, by percutaneous nephrostomy. Pain and urinary tract infection should be treated.
Cardiac tamponade Increased intrapericardial pressure occurs due to excess pericardial fluid, which reduces cardiac filling and leads to circulatory compromise. Cardiac tamponade is most commonly associated with a malignant pericardial effusion from lung cancer, ovarian cancer and primary cardiac tumours.
Presentation Two-thirds of patients are asymptomatic. Symptoms include breathlessness, chest pain, orthopnoea and weakness. There are signs of haemodynamic compromise – raised JVP, tachycardia, hypotension, increased pulsus paradoxus and oedema.
Investigation Two-dimensional echocardiography should be used to diagnose the effusion, assess the haemodynamic impact and assist with obtaining fluid for cytological examination.
Treatment Pericardiocentesis and a pericardial window should be performed if fluid re-accumulates. In patients for whom a pericardial window is not thought suitable, radiother-
Treatment-related emergencies Neutropenic fever/sepsis Neutropenic fever/sepsis may occur after almost any chemotherapy regimen but is mainly associated with cancers that have been treated with intensive myelosuppressive regimens (e.g. lymphoma or leukaemia). Neutropenia with septic shock is a very serious problem and needs to be managed as an acute emergency. The more common occurrence of febrile neutropenia without shock must still be treated promptly, but it is more easily managed with oral or intravenous antibiotics and close observation.
Risk of complications Patients can be stratified into low or high risk of serious complication. Factors such as solid tumour, short predicted duration of neutropenia, good performance status and absence of clinical signs of serious infection are considered to be predictive of a low risk. A low-risk patient may sometimes be treated as an outpatient assuming they are supported, have a GP and a telephone and live near a hospital (discussed subsequently).
Presentation The patient presents with depressed neutrophil cell count (< 1.5 × 109 /1 neutrophils) and recent cytotoxic therapy in the presence of raised temperature (> 37.5◦ C). The effect of sepsis-induced vasodilation producing hypotension can rapidly cause end-organ (renal, hepatic, cerebral) damage that can be fatal.
Initial assessment Identify immediately any patients with septic shock who have low blood pressure, decreased urine output, increased serum lactate or a change in mental status. In cases of septic shock, instigate prompt resuscitative measures which include possible transfer to ITU/HDU. Appropriate early communication with intensive care specialists can be very helpful in selection of patients for intensive organ support. Septic shock is sepsis with hypotension, despite adequate fluid resuscitation, along with the presence of perfusion abnormalities that may include lactic acidosis, oliguria or an acute alteration in mental status. 77
Paul Shaw
Hypotension is a systolic blood pressure of less than 90 mmHg or a reduction of greater than 40 mmHg from baseline in the absence of other causes.
Initial management of septic shock Immediate measures include the following: r Intravenous access with medium- to wide-bore cannula. r Start 1000 ml N/saline or 500 ml colloid over 30 minutes. r If there is evidence of hypoxia, give oxygen (24% if COPD or previous bleomycin). r Investigate serum lactate, FBC, electrolytes, urea, creatinine and LFTs, clotting, glucose, line cultures, blood cultures, blood gases and MSU. r Give intravenous antibiotics according to local hospital policy. r Record vital signs every 15 minutes. r Initiate a chest X-ray if signs suggest a chest infection. r If serum lactate is greater than 4 mM/l consider transfer to ITU/HDU. After 30 minutes, if the patient is still hypotensive, start 2 l N/saline or colloid over 1 hour (but use caution in cardiac disease) and consider transfer to ITU/HDU. If there is evidence of renal impairment, monitor urine output via indwelling catheter. After 1 hour, urine output should be 0.5 to 1 ml/kg per hour. If output remains below this level, consider transfer to ITU.
Febrile neutropenia Patients may be grouped into low-risk and high-risk categories to decide the appropriate management. In all cases the local policy should be followed. Investigate the following: r Full blood count and differential. r Electrolytes, renal and liver biochemistry. r Serum lactate. r Blood culture from indwelling venous catheter and peripheral site. r Urine microscopy and culture. r Sputum culture. r Swabs of any discharging sites potentially culpable. r Chest X-ray may be helpful. For use of antibiotics in low-risk patients, the following criteria must be met: r Neutropenia is expected to last fewer than 7 days. r Absolute neutrophil count is not expected to fall below 0.1 × 109 /l. 78
r The patient is not involved in any current or recent antibiotic treatment.
r The patient is not unwell and there are no signs of shock, vomiting, major organ involvement or other co-morbidity. r No highly immunosuppressive therapy is involved. r The patient has a committed live-in caregiver. r The patient has a telephone at home. r The patient lives less than 30 minutes travel time from a cancer centre. r The patient is registered with GP surgery or health centre. In patients without an allergy to penicillin, start with oral ciprofloxacin 750 mg b.d. and co-amoxiclav 625 mg t.d.s. for 7 days. For those with a penicillin allergy, use ciprofloxacin 750 mg p.o. b.d. and clarithromycin 500 mg p.o. b.d. for 7 days. Patients should take the first dose under supervision and can then be discharged with an information sheet. Any patient who does not fulfil the low-risk criteria should be treated as a high-risk patient. For patients without an allergy to penicillin, start with a combination such as intravenous piperacillin 4.5 g t.d.s. and gentamicin loading dose of 6 mg/kg. For patients with a penicillin allergy, use teicoplanin and gentamicin. For those with specific localising signs, additional antibiotics may be added; for example, clarithromycin for chest infection and teicoplanin for a line infection. It is important to follow local guidelines and consult with the bacteriologist. If the fever settles within 48 to 72 hours, the intravenous antibiotics should be continued for an additional 24 hours. Oral antibiotics are then given for 5 days, similar to the low-risk protocol. If the fever persists beyond 48 to 72 hours, patients who have a central venous catheter and have not yet started teicoplanin should do so, consideration should be given to removing the catheter, and the microbiology department should be contacted to discuss test results and possible antifungal therapy. If a pathogen has not been isolated, additional cultures and serology are needed, along with a CT scan of the chest and bronchoalveolar lavage. The use of amphotericin B or antiviral agents or non-infectious causes of fever should also be considered.
Use of G-CSF For patients with febrile neutropenia and a high risk of complications, the use of G-CSF should be considered. Such patients include those with the following:
Oncological emergencies
r Profound neutropenia (ANC < 0.1 × 109 /l). r Prolonged neutropenia (> 10 days). r Pneumonia. r Hypotension. r Multiorgan dysfunction. r Uncontrolled primary disease. r Invasive fungal infections. r Age > 65 years. r Those who were hospital inpatients at the time of developing the fever. G-CSF is not usually used in afebrile patients with neutropenia or in patients with uncomplicated neutropenic fever such as those with none of the features just described.
phoblastic leukaemia. When first used, taxanes resulted in major hypersensitivity reactions in 30% of patients, with 40% suffering mild symptoms. These reactions were associated with fast infusion rates and had a tendency to occur after the second infusion, usually in the first few minutes of treatment and with resolution 15 to 20 minutes after stopping the infusion. Carboplatin sensitivity is unpredictable and reactions may occur following a prolonged course of treatment – at any time during the infusion or indeed days following its administration. Monoclonal antibodies such as rituximab and cetuximab may cause a cytokine release syndrome.
Presentation Prophylactic antibiotics A randomised double-blind placebo controlled trial in patients receiving cyclic chemotherapy for solid tumours and lymphoma and who are at risk of neutropenia (< 500 neutrophils per mm3 ) compared the use of prophylactic levofloxacin or placebo for 7 days during the period of neutropenia. Levofloxacin significantly reduced the incidence of clinically documented infection (3.5 versus 7.9%; p < 0.001) and hospitalisation for the treatment of neutropenic infection (15.7 versus 21.6%; p < 0.004) with few adverse effects (Cullen et al., 2005). A recent meta-analysis of trials of prophylactic antibiotics in neutropenic patients has shown a decrease in the risk of death with their use compared to placebo or no treatment (RR = 0.67; CI 0.55 to 0.81). Fluoroquinolone prophylaxis reduced the risk for all-cause mortality (RR = 0.52; CI = 0.35 to 0.77) as well as infection-related mortality, fever, clinically documented infections and microbiologically documented infections. The authors concluded that antibiotic prophylaxis, preferably with a fluoroquinolone, should be considered for neutropenic patients (Gafter-Gvili et al., 2005). Patients receiving chemotherapy for solid tumour or lymphoma and who are at risk of bacterial infection and severe neutropenia (< 500 neutrophils per mm3 ) without G-CSF support should be considered possible candidates for prophylactic levofloxacin.
Anaphylaxis related to anticancer drugs Anaphylaxis is associated particularly with paclitaxel, carboplatin and docetaxel. L-asparaginase may cause anaphylaxis in 10% of patients treated for acute lym-
Patients may present with agitation, hypotension, bronchospasm, and rash. Angioedema and urticaria, abdominal pain, rash, chest tightness, laryngeal oedema and tongue swelling also occur.
Prevention Prophylactic steroids and antihistamines reduce the incidence of hypersensitivity reactions to taxanes and carboplatin.
Treatment Treat for anaphylaxis as follows: r Disconnect the drug. r Secure airway, breathing and circulation. r Give oxygen, lie the patient flat and elevate the patient’s legs if hypotensive. r If there is stridor, wheeze, respiratory distress or clinical signs of shock then give adrenaline (epinephrine; 1:1000 solution) 0.5 ml i.m. and repeat after 5 minutes if there is no improvement. r Give 10 mg i.v. chlorpheniramine. r For all severe or recurrent reactions give 200 mg i.v. hydrocortisone. r If shock fails to respond to drug measures, give 1 to 2 l of i.v. crystalloid. r Most deaths due to anaphylaxis are associated with giving adrenaline too late.
Tumour lysis syndrome (TLS) Tumour lysis syndrome is caused by sudden tumour necrosis either due to treatment or occurring spontaneously. It causes metabolic abnormalities, particularly hyperkalaemia, hyperuricaemia, hyperphosphataemia and secondary hypocalcaemia. 79
Paul Shaw
TLS is associated with chemosensitive, bulky tumours such as high-grade lymphoma, acute leukaemia and Burkitt lymphoma. It is rarely seen in low-grade lymphomas or solid tumours. Pre-existing renal failure may be a contributory factor. Patients with lymphoma who have a raised LDH (> 1500 IU/l) are likely to have a high tumour burden and are at increased risk.
Presentation Patients present with the following: r Non-specific symptoms of weakness, nausea, vomiting, myalgia and dark urine. r Electrolyte imbalance, which can result in arrhythmias, neuromuscular irritability, seizure and death. Arrhythmias are a common cause of death if left untreated. r Renal failure, which occurs frequently, secondary to hyperuricaemia.
Investigation
r Check levels of serum electrolytes Na+ , K+ , PO4 −3 , and Ca+2 ; uric acid; acid/base balance; and renal function. r Monitor ECG. r The typical biochemical picture is of hyperuricaemia, hyperkalaemia, hyperphosphataemia, hypocalcaemia, lactic acidosis and renal failure.
Prevention
r Early diagnosis requires a high level of suspicion. In r
r r r
patients at particular risk, routine uric acid and electrolyte measurements are sensible. Hydrate the patient with at least 3 l of normal saline per 24 hours to maintain urine output greater than 100 ml per hour with or without diuretic to maintain urate clearance. Administer oral sodium bicarbonate to alkalise urine and prevent urate nephropathy in acidic conditions. Give allopurinol 300 mg/day to prevent hyperuricaemia. Unfortunately, despite these treatments, 14% of patients will still develop renal problems (Bessmertny et al., 2005).
sulphonate), calcium gluconate, sodium bicarbonate to correct acidosis and dextrose/insulin injection. r Therapy for hyperphosphataemia and hypocalcaemia involves oral phosphate binders (aluminium hydroxide 30 ml q.d.s.) and calcium gluconate (10 ml i.v. injection). r Therapy for hyperuricaemia involves sodium bicarbonate to maintain urine pH > 7.0 and allopurinol 600 to 800 mg/day. r When severe TLS develops, intensive care support and continuous monitoring are necessary. r Renal dialysis is required if hyperphosphataemia, symptomatic hypocalcaemia, persistent hyperkalaemia, hyperuricaemia and anuria/oligouria, acidosis or volume overload develops. Rasburicase is the first recombinant uricolytic agent (urate oxidase) and it exerts its action on uric acid, metabolising it to allantoin, which is five to ten times more soluble in urine than uric acid. It has a role in the prevention and treatment of TLS and is licensed for use immediately before and during the start of chemotherapy. The advantage of recombinant urate oxidase is its rapid onset of action. The safety and efficacy of rasburicase is currently being assessed and compared with allopurinol in a phase III clinical trial (Rampello et al., 2006).
Extravasation of chemotherapy Extravasation is leakage of intravenous drugs from a vein into the surrounding tissue.
Presentation Extravasation may present during the administration of chemotherapy or later, with pain and swelling at the site of the intravenous cannula. Most commonly, it causes pain and localised tissue inflammation. More seriously, it can result in ulceration, necrosis, sloughing of the skin, damage to underlying structures and permanent disability. Table 6.3 shows cytotoxic agents classified according to the type of reaction they produce. The severity of an extravasation will depend on the infusion site, the concentration and volume of the chemotherapy drug and the treatment given for the extravasation. Table 6.4 shows the risk factors for extravasation.
Treatment Treatment
r Aggressive hydration is required. r Therapy for hyperkalaemia includes cation exchangeresins binding potassium (sodium polystyrene
80
Experience from case reports and small series has resulted in the publication of guidelines for the prevention and treatment of extravasation (Goolsby and Lombardo, 2006). There are some general principles for
Oncological emergencies
Table 6.3. Cytotoxic agents classified according to the
Table 6.5. Antidotes for extravasation
type of reaction they typically produce Drug Class of drug
Definition
Examples
Vesicant
Capable of causing
Anthracyclines
Irritant
Topical DMSO 50–99% Topical hydrocortisone cream
pain,
Vinca alkaloids
inflammation,
Paclitaxel
Mitomycin
As for anthracyclines
blistering,
Streptozocin
Vinca alkaloids
Infiltrate the site with hyaluronidase
necrosis
Mechlorethamine
(1500 units of hyaluronidase in
Oxaliplatin
2 ml of water for injection, or 0.9%
Mitomycin
sodium chloride) using 0.2 ml
Platinum
injections over and around the
Capable of causing irritation and inflammation
Non-irritant
Cold compress
affected area
compounds
Warm compress
Etoposide Irinotecan
Non-irritant
Anthracyclines
Antidote/treatment
Paclitaxel
Infiltrate site with mixture of
Topotecan
hydrocortisone and
Flurouracil
chlorpheniramine followed by hyaluronidase
Cyclophosphamide
Warm compression alternating with
Ifosfamide
topical antihistamine cream
Bleomycin Fludarabine Gemcitabine Methotrexate Adapted from Goolsby and Lombardo (2006).
Cisplatin
Infiltrate site with 3% isotonic sodium thiosulphate, aspirate back, then give hyaluronidase Topical hydrocortisone cream and a warm compress
DMSO = dimethyl sulfoxide. For more details see the National Extravasation Information Service
Table 6.4. Risk factors for extravasation Factor
Description
Vein physiology
Fragile, small, sclerosed
Pharmacological
Duration and chemotherapy dosage exposure to tissue
Physiological
SVCO, lymphoedema, peripheral
Radiotherapeutic
Previous local radiotherapy,
Mechanical
Needle insertion technique, multiple
neuropathy, phlebitis radiation recall reactions venepuncture sites Adapted from Goolsby and Lombardo (2006).
the management of extravasation, together with specific measures for each chemotherapy drug. In the UK, the ‘Green Card’ reporting system is in place to collect and analyse data on extravasation. It is coordinated through the National Extravasation Information Service, based at St. Chad’s Unit, City Hospital in Birmingham. The
(www.extravasation.org.uk) or Allwood and Stanley (2002).
web address is www.extravasation.org.uk and it provides information on treatment of extravasation. In addition many centres have developed local protocols. Initial management should include the following: r Stop infusion, disconnect tubing, but leave i.v. cannula in situ. r Attempt aspiration of vesicant and administer antidote (see Table 6.5) if appropriate. r Keep limb elevated with either cold or warm compression as indicated (see Table 6.5 for examples). r Provide full documentation and monitoring (consider using a photograph record). r Estimate the amount of extravasated drug. r Consider immediate surgical opinion in cases of anthracycline extravasation or when conservative treatment fails to improve symptoms or tissue damage occurs. 81
Paul Shaw
REFERENCES Allwood, M. and Stanley, A. (2002). The Cytotoxics Handbook. Oxford: Radcliffe Medical Press. Bayley, A., Milosevic, M., Blend, R. et al. (2001). A prospective study of factors predicting clinically occult spinal cord compression in patients with metastatic prostate carcinoma. Cancer, 92, 303–10. Bessmertny, O., Robitaille, L. M. and Cairo, M. S. (2005). Rasburicase: a new approach for preventing and/or treating tumour lysis syndrome. Curr. Pharm. Des., 11, 4177–85. Cervantes, A. and Chirivella, L. (2004). Oncological emergencies. Ann. Oncol., 15 (Suppl. 4), S299–306. Cullen, M., Steven, N., Billingham, L. et al. (2005). Antibacterial prophylaxis after chemotherapy for solid tumours and lymphomas. N. Engl. J. Med., 353, 988–98. Freitag, L. (2004). Interventional endoscopic treatment. Lung Cancer, 45 (Suppl. 2), S235–8. Gafter-Gvili, A., Fraser, A., Paul, M. et al. (2005). Meta-analysis: antibiotic prophylaxis reduces mortality in neutropenic patients. Ann. Intern. Med., 142, 979–95. Goolsby, T. V. and Lombardo, F. A. (2006). Extravasation of chemotherapeutic agents: prevention and treatment. Semin. Oncol., 33, 139–43. Maisey, N., Norman, A., Prior, Y. et al. (2004). Chemotherapy for primary gastric lymphoma: does in-patient observation prevent complications? Clin. Oncol. (R. Coll. Radiol.), 16, 48–52. Morris, C. D., Budde, J. M., Godette, K. D. et al. (2002). Palliative management of malignant airway obstruction. Ann. Thorac. Surg., 74, 1928–32.
82
Ostler, P. J., Clarke, D. P., Watkinson, A. F. et al. (1997). Superior vena cava obstruction; a modern management strategy. Clin. Oncol. (R. Coll. Radiol.), 9, 83–9. Palmer, K. (2004). Management of haematemesis and melaena. Postgrad. Med. J., 80, 399–404. Patchell, R. A., Tibbs, P. A., Regine, W. F. et al. (2005). Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomised trial. Lancet, 366, 643–8. Poortmans, P., Vulto, A. and Raaijmakers, E. (2001). Always on a Friday? Time pattern of referral for spinal cord compression. Acta Oncol., 40, 88–91. Priestman, T. J., Dunn, J., Brada, M. et al. (1996). Final results of the Royal College of Radiologists’ trial comparing two different radiotherapy schedules in the treatment of cerebral metastases. Clin. Oncol. (R. Coll. Radiol.), 8, 308–15. Rades, D., Stalpers, L. J., Schulte, R. et al. (2006). Defining the appropriate radiotherapy regimen for metastatic spinal cord compression in non-small cell lung cancer patients. Eur. J. Cancer, 42, 1052–6. Rampello, E., Fricia, T. and Malaguarnera, M. (2006). The management of tumour lysis syndrome. Nat. Clin. Pract. Oncol., 3, 438–47. Rockall, T. A., Logan, R. F., Devlin, H. B. et al. (1996). Risk assessment after acute upper gastrointestinal haemorrhage. Gut, 38, 316–21. Rowell, N. P. and Gleeson, F. V. (2001). Steroids, radiotherapy, chemotherapy and stents for superior vena caval obstruction in carcinoma of the bronchus. Cochrane Database Syst. Rev., 4, CD001316.
7
PALLIATIVE CARE Simon Noble
Introduction Changing role of palliative care in oncology The World Health Organisation (WHO) defines palliative care as ‘the active total care of patients whose disease is not responsive to curative treatment, where control of pain, of other symptoms and of psychological, social and spiritual problems is paramount with the achievement of the best possible quality of life for patients and their families as the goal’ (World Health Organisation, 1990). Palliative care should now be considered an integral part of service planning and care delivery in oncology. This chapter covers common problems in symptom control, communication, ethical decision making, and the financial difficulties of patients with advanced cancer.
Changing model of palliative care In the old model of the cancer journey (Figure 7.1(a)), palliative care services would only be involved at the end of life when no further oncological or supportive treatments were available. This was a ‘terminal care’ service for those clearly at the end of life. But, the symptoms that were being controlled occur not only at the end of life but also at different degrees throughout the cancer journey. The new model (Figure 7.1(b)) attempts to dovetail palliative care with active treatment, gradually increasing its involvement as active treatment becomes less appropriate. Some patients, such as those with carcinoma of the pancreas, are likely to need palliative care input early in their illness.
Pain control Pain is an unpleasant sensory and emotional experience associated with actual or potential tissue damage. It is experienced by up to 70% of patients with advanced cancer and of these, one-third have a single pain, one-
third have two pains and one-third have three or more pains (Royal College of Physicians, 2000). Pain may be r Related to the cancer itself (e.g. metastatic bone pain). r Treatment related (e.g. neuropathy secondary to chemotherapy). r Related to cancer and debility (e.g. constipation). r Unrelated to the cancer (due to another co-existing condition).
Pharmacological methods The WHO analgesic ladder (World Health Organisation, 1990) is very useful in managing cancer pain. It can alleviate pain in more than 80% of patients by starting at a level most appropriate to the patient’s pain and increasing in steps until adequate analgesia is achieved (Table 7.1). Adjuvant analgesics may be used at any point depending on the type of pain the patient is experiencing; an adjuvant analgesic is a drug whose primary indication is for something other than pain but which has an analgesic effect for certain types of pain (Table 7.2). When starting someone on an opioid, it is important to prescribe regular laxatives and a suitable antiemetic (such as haloperidol 1.5 mg nocte). Patients should also be written up for adequate breakthrough analgesia, which amounts to one-sixth of the total dose of morphine prescribed for 24 hours (Fallon and McConnell, 2006; see Tables 7.3 and 7.4 for breakthrough doses and conversions to other opiates).
Opioid toxicity Morphine is glucuronidated in the liver and then excreted through the kidneys. Renal impairment and/or dehydration may lead to accumulation of opioid metabolites and opioid toxicity, which commonly results in cognitive impairment and confusion (Hall and Sykes, 2004). Other causes of confusion such as hyponatraemia, cerebral disease, infection and hypercalcaemia should be considered and excluded. The other features of opioid toxicity include the following: 83
Simon Noble
Table 7.1. The World Health Organisation (WHO) analgesic ladder Step
Severity of pain
Analgesia
Examples
Step 1
Mild
Non-opioid ± adjuvant
Paracetamol
Step 2
Mild to moderate
Weak opioid
Codeine
± non-opioid
Dihydrocodeine
± adjuvant Step 3
Moderate to severe
Strong opioid
Morphine
± non-opioid
Fentanyl
± adjuvant
Oxycodone Hydromorphone
Anti-cancer treatments
Terminal care
Table 7.2. Adjuvant analgesics Drug type Non-steroidal
Diagnosis
(a)
Death
Indication – examples Bone metastases
anti-inflammatory
Liver capsule pain
drugs (e.g. diclofenac
Soft tissue infiltration
50 mg t.d.s.) Corticosteroids
Anti-cancer treatments Palliative care
Soft tissue infiltration
8–16 mg daily)
Liver capsule pain Nerve compression
Grief Anticonvulsants/
Diagnosis
(b)
antidepressants (e.g.
Death
gabapentin 100–300 mg
Figure 7.1. (a) Traditional model of palliative care showing set point at
nocte, amitriptyline
which active treatment ceases and comfort-only treatment begins.
25 mg nocte)
(b) Current model of palliative care showing gradual introduction of
Bisphosphonates
increasing palliative care as active treatment becomes less appropriate.
Raised intracranial pressure
(e.g. dexamethasone
Nerve compression or infiltration Paraneoplastic neuropathies
Malignant bone pain
(e.g. pamidronate, zoledronate)
r Pinpoint pupils. r Myoclonus/metabolic flap. r Visual hallucinations. Patients typically see dark spots in the periphery of their vision and may think they have seen animals run under the bed. r Drowsiness (severe toxicity). r Respiratory depression (severe toxicity).
r The needs and views of the patient and carers. r Experience of the referring physician. r Availability of the procedure. r Available expertise (e.g. nursing staff) to manage the intervention. Table 7.5 shows non-pharmacological methods of pain control.
Non-pharmacological methods Wherever possible, non-pharmacological methods of analgesia should be considered. There are many options available and their use is likely to depend on:
84
Antiemetic prescribing Numerous neurotransmitter receptors are involved in transmitting the impulses connected with nausea and
Palliative care
Table 7.3. Morphine doses and conversion Breakthrough dose of morphine = total dose of morphine in 24 hours, divided by 6 (e.g. for someone on MST 60 mg b.d., total 24-hour dose = 60 mg × 2 = 120 mg): Breakthrough dose = {60 mg × 2} ÷ 6 = 120 mg ÷ 6 = 20 mg p.r.n. Converting oral morphine to subcutaneous diamorphine via syringe driver Take the total 24-hour dose of morphine and divide by 3 (subcutaneous diamorphine is three times as potent as oral morphine), e.g. for someone on MST 60 mg b.d.: Subcutaneous diamorphine dose = {60 mg × 2} ÷ 3 = 40 mg diamorphine s.c. over 24 hours
Table 7.4. Morphine potencies Oxycodone is twice as potent as oral morphine (e.g. 10 mg oxynorm = 20 mg oromorph) Tramadol is stronger than you think! 100 mg tramadol
clopramide may cause restlessness and Parkinsonism). r Consider non-pharmacological measures such as relaxation, acupuncture, and providing small frequent meals that do not have extremes of taste or smell.
q.d.s. = 80 mg oral morphine = 40 mg MST b.d. Co-codamol 30/500 2 tabs. q.d.s = 24 mg morphine
vomiting to the vomiting centre (VC) and chemoreceptor trigger zone (CTZ) in the midbrain (Mannix, 2006). Chemical triggers (such as drugs, metabolites and toxins) are detected at the CTZ, whereas the VC receives input from stretch receptors on nerve terminals (e.g. from the liver capsule being stretched by metastases or bowel dilatation due to obstruction; Baines, 1997) as well as input from higher mental centres (e.g. pain, fear memory) and integrates these with input from the CTZ. Different antiemetics block different receptors and the choice of antiemetic should be guided by the underlying cause of vomiting (Glare et al., 2004; Rhodes and McDaniel, 2001; Tables 7.6 and 7.7). When managing a patient with nausea and vomiting: r Identify possible causes. r Consider probable pathways and neurotransmitters involved to choose the most appropriate antiemetic. r Use an antiemetic regularly and titrate the dose, considering alternatives to the oral route if necessary. r When using combinations of drugs, remember potential interactions (e.g. metoclopramide and cyclizine have an antagonistic effect; haloperidol and meto-
Constipation Constipation is a common symptom facing cancer patients and should be prevented because it can cause: r Abdominal pain and colic. r Intestinal obstruction. r Confusion. r Urinary retention. r Overflow diarrhoea. Consider and, if appropriate, treat the cause of the constipation: r Drugs (e.g. opioids, tricyclics, antacids, phenothiazines). r Dehydration. r Abdominal wall muscle paresis (e.g. from spinal cord compression). r Hypercalcaemia. r Primary or secondary bowel disease. General measures such as increasing fluid intake, high-fibre diet and improved mobility should be introduced. Various laxatives are available and should be chosen for either bowel stimulation or stool softening. Typical starting doses are given but should be titrated up to achieve appropriate results: r Senna (stimulant) 7.5 mg p.o. once a day. r Bisacodyl (stimulant) 5 mg p.o. once a day. r Sodium docusate (softener) 200 mg p.o. twice a day. r Lactulose (osmotic) 10 ml p.o. twice a day. r Magnesium hydroxide (softener) 10 ml p.o. twice a day.
85
Simon Noble
Table 7.5. Non-pharmacological methods of pain control for cancer patients Procedure
Indication
Other important issues
Radiotherapy
Commonly used for pain from bone metastases
Transcutaneous electrical nerve
Useful for opiate-sparing Need for transfer to oncology department
Used as an adjuvant to other analgesics
Requires expertise of physiotherapist
For neuropathic pain in clearly
Anaesthetic involvement required
stimulation Nerve blocks
identifiable nerve distribution Epidural/intrathecal analgesia
For complex cord-related pains and plexopathies
Anaesthetic involvement essential Ongoing care difficult if an indwelling catheter is used
Orthopaedic stabilisation
Useful for incident pain (pain on
Orthopaedic opinion required
movement), e.g. vertebral disease or prophylaxis of fracture in long bones with metastases Vertebroplasty
Useful for stabilisation and analgesia of
Radiological expertise required
pain related to vertebral metastatic disease
Table 7.6. Different antiemetics Drug
Mechanism of action
Clinical uses
Metoclopramide
Prokinetic, weak D2 (dopamine receptor) antagonist
Gastric outflow obstruction Upper GI bleed Liver metastases Carcinoma head of pancreas
Haloperidol
Good central antiemetic effect (D2 antagonist)
Metabolic causes of nausea: Hypercalcaemia Uraemia Antibiotics Opioids
Cyclizine
ACh (acetylcholine)/H2 (histamine) – central acting
Central causes Raised intracranial pressure Motion sickness/vestibular problems
Ondansetron or granisetron
5-HT3 antagonist
Chemotherapy-induced nausea (Very constipating)
r Co-danthramer (softener and stimulant) caps or suspension twice a day. The choice of laxatives varies in different units. It is our practice to use a combination of softener and stimulant in advanced cancer patients. Start with a combination of senna and magnesium hydroxide, with a typical dose of 10 ml of each, twice a day, although the 86
choice of laxative should be based on individual patients.
Important points Co-danthramer can cause a sore perianal rash and colours the urine orange. It is not recommended in bedbound patients, especially those with catheters.
Palliative care
Table 7.7. Choices of antiemetics Cause of vomiting
Choice of antiemetic
Drug/toxin/metabolic
Haloperidol, levomepromazine
Radiotherapy
Haloperidol
Chemotherapy
Ondansetron, dexamethasone,
Bowel obstruction
Cyclizine, hyoscine
metoclopramide butylbromide, octreotide, corticosteroids Delayed gastric
Metoclopramide, domperidone
emptying Raised intracranial
Cyclizine, dexamethasone
pressure
Docusate acts as a softener at 200 mg twice a day but has softener and stimulant properties at 200 mg three times a day.
Diarrhoea Consider and treat the causes of diarrhoea. Causes of diarrhoea include: r Clostridium difficile r Radiotherapy. r Drugs (e.g. NSAIDs, laxatives). r Acquired lactose intolerance. Do not forget that opioids cause constipation and that diarrhoea may occur after decreasing the dose or converting to an alternative opioid such as fentanyl. For symptomatic control of diarrhoea consider the following: r Codeine 30 mg p.o. four times a day. r Loperamide 2 mg (1 capsule) p.o. four times a day. Increase dose to 16 mg daily although higher doses have been reported. r Octreotide 300 μg s.c. in 24 hours; up to 1200 μg in 24 hours.
Anorexia and dysphagia Dysphagia occurs in more than 20% of patients with advanced cancer. The causes are often complex but may involve alterations in oro-oesophageal anatomy, oral mucosa, paresis of cranial nerves IX, X and XI, and higher cortical functioning (the cerebral cortex is involved in the control of swallowing). A careful history and exami-
nation to find reversible causes of dysphagia is essential. The following causes of dysphagia in advanced cancer should always be considered: r Obstruction in the oropharynx. r External compression. r Upper motor neurone damage (e.g. cerebral metastases). r Cranial nerve involvement (e.g. metastases at the base of the skull). r Mucosal inflammation (e.g. radiotherapy, infection). r Dry mucosa (e.g. antimuscarinic drugs). The anorexia/cachexia syndrome is very common in cancer patients and it results in decreased appetite, weight loss, decreased energy levels and weakness. The pathophysiology is complex and involves alterations of metabolic, neurohormonal and anabolic mechanisms. Its management should include treating the reversible causes of anorexia, such as nausea, and using appetite stimulants. Corticosteroids and progestagens have been used but there is no strong evidence that their use improves lean body mass or outcome. Nasogastric feeding may be used for carefully selected patients. The decision to offer nutritional support should be a multidisciplinary one involving the patient, their carers and the dietician. As yet, there are no studies to suggest that enteral feeding alters the clinical outcome or quality of life for these patients.
Pleural effusions Malignant pleural effusions occur in up to 15% of cancer patients, with lung and breast cancers accounting for more than two-thirds of cases. When the primary tumour is chemosensitive, patients with malignant pleural effusions may be treated with systemic chemotherapy. However, the mainstay of treatment is symptomatic drainage. Several options are available: r Repeated thoracocentesis. r Drainage and chemical pleurodesis. r Chronic indwelling pleural catheter. r Pleuroperitoneal shunt. The first two are the most commonly used. However, before embarking upon a potentially dangerous procedure, consider the following questions: 1. Does the effusion account for the patient’s symptoms? 2. Is the effusion recurrent? 3. Is the patient well enough for the procedure? 4. Is the lung likely to re-expand? 87
Simon Noble
Ascites Up to 50% of patients with malignancy will develop ascites, especially those with cancer of the ovary, gastrointestinal system or breast. Ascites is associated with a poor prognosis and it is likely to recur. Whenever possible, the cause of the ascites should be established because it may be due to peritoneal seeding, intrahepatic disease or lymphatic obstruction from retroperitoneal disease. The main medical treatment is spironolactone (100 to 400 mg/day) with a loop diuretic such as furosemide (40 to 80 mg/day). Although this management may delay or prevent the need for paracentesis, it must be done with care because depleting the intravascular volume may precipitate renal failure. Paracentesis has been shown to reduce hospital stay when compared to medical treatment for large-volume ascites. Draining the ascites relieves the symptoms but it may precipitate hypovolaemia. Therefore, the fluid should be drained slowly over several hours, clamping the drain for 1 hour after every litre. The use of albumin replacement therapy is not recommended in patients with advanced cancer because it has little sustained effect on long-term albumin levels and there is an increased risk of pulmonary oedema.
Psychological symptoms Anxiety Anxiety is apprehension of danger and dread accompanied by restlessness, tension, tachycardia and dyspnoea without a clearly identifiable cause (Cathcart, 2006). Anxiety may be severe and may also make other symptoms such as pain, breathlessness and nausea worse. It is important to address the underlying factors such as pain or drug withdrawal (including alcohol or nicotine). If nicotine withdrawal is suspected, nicotine patches may be used. Psychological treatment, such as cognitive behavioural therapy, can help patients to identify the thoughts that cause them distress and find ways of challenging them before the anxiety gets worse (Greer et al., 1992; Meyer and Mark, 1995). Drug treatment includes the use of benzodiazepines (e.g. diazepam 1 to 5 mg orally) as required. Selective serotonin reuptake inhibitors may be used to relieve panic attacks if benzodiazepines are ineffective and tricyclic antidepressants in low doses can also be used. 88
Complementary therapies such as aromatherapy may also help the patient (Jackson and Lipman, 2006).
Depression Depression is a temporary mental state or chronic mental disorder characterised by feelings of sadness, loneliness, despair, low self-esteem and self-reproach. Although as many as 69% of patients with advanced cancer may experience depression, they are often not treated adequately (Barraclough, 1997; Endicott, 1984; Lloyd-Williams et al., 1999). Many of the symptoms (e.g. fatigue and loss of energy) usually used to diagnose depression are very common in cancer patients and so diagnosis can be difficult. Also, thoughts of death may be realistic and a basis for appropriate planning. A number of substitute symptoms have therefore been suggested for the diagnosis of depression in patients with advanced malignant disease: social withdrawal, pessimism and lack of reactivity. The single question ‘Are you depressed?’ has been found to have a high sensitivity, specificity and positive predictive value in some studies but not in others (Lloyd-Williams et al., 2003).
Confusion and delirium Delirium is characterised by global cerebral dysfunction and by a lowered level of consciousness, combined with disturbed attention, thinking, memory, psychomotor behaviour, emotion and sleep–wake cycle (Lawlor et al., 2000). Delirium is present in up to 42% of patients with advanced cancer. It may often complicate care at the end of life because patients are often unable to give a history or accurately communicate their symptoms. Furthermore, delirium generates distress for patients and their families and impedes communication between them. Managing patients with delirium in the palliative care setting should follow a stepwise approach: 1. Identify the underlying cause and decide whether it is reversible. 2. Identify precipitating and perpetuating factors (e.g. hypercalcaemia or infection) and decide whether they are treatable. 3. Use environmental strategies such as reassurance, re-orientation, the presence of familiar staff and family, limit staff changes and reduce the amount of noise stimulation.
Palliative care
4. Benzodiazepines can be used if other options have failed, particularly when agitation and restlessness are prominent. Antipsychotic agents such as haloperidol and levomepromazine can be used, especially when sedation is needed. Elderly patients should receive reduced drug doses because benzodiazepines carry a greater risk of paradoxical agitation. Risperidone and olanzapine should be avoided in older patients with cerebrovascular disease (Breitbart et al., 1996).
Syringe drivers Syringe drivers are not just for dying patients. They can also be used for patients who: r Are unable to swallow. r Are nauseated. r Have poor absorption because of bowel stasis, pancreatic dysfunction or hypoalbuminaemia, which can cause oedema of the bowel mucosa. r Are too ill to take medicines. Syringe drivers can be used to control pain, nausea, restlessness and colic. It is important that, when combinations are used, the drugs are compatible and do not crystallise. The following commonly used drugs can be mixed together in a syringe driver: r Diamorphine. r Haloperidol. r Cyclizine. r Hyoscine hydrobromide. r Midazolam. r Levomepromazine. r Metoclopramide. A useful website for checking drug compatibilities in syringe drivers is www.pallmed.net.
Care in the last 48 hours of life Once the team has recognised that a patient is dying, the following should be done: r Stop all unnecessary medicines. r Ensure you write up p.r.n. medicines for all symptomatic eventualities: r Pain – s.c. diamorphine or morphine. r Agitation – s.c. midazolam. r Death rattle – s.c. hyoscine hydrobromide. r Nausea – s.c. cyclizine. r If the patient is on regular analgesics/antiemetics, convert to syringe driver. r Communicate regularly with the patient’s family.
Communication issues The need to communicate sensitive and often upsetting information is not exclusive to palliative medicine. It is an integral part of all health care professionals’ practice, and breaking bad news and handling difficult questions occur commonly in the day-to-day practice of the oncologist. Whether communication occurs within a busy outpatient clinic or on an inpatient ward round, there are several key principles that should form the framework of any difficult consultation. The following ten-step approach can be used as a guide to breaking bad news. 1. Preparation. r Know all facts before the meeting. r Who does the patient want to be present? r Ensure that another health professional (such as a nurse) is present. r Ensure that there is no interruption and you leave your pager with someone. r Ensure that there are enough chairs for everyone. 2. What does the patient know? r Assess what the patient understands is happening at the moment. r Ask for the the patient’s narrative of events. This will give you an idea of how much the patient knows, and his or her preferred vocabulary, etc. 3. Is more information wanted? r Test the waters. r Be aware that it may be frightening for a patient to ask for more information. r Ask, ‘Would you like me to explain a bit more?’ 4. Give a warning shot. r Perhaps start by saying ‘I’m afraid it looks rather serious.’ r Allow time for the patient to respond. 5. Allow denial. r Denial is a powerful coping mechanism. r Denial allows the patient to control the amount of information. 6. Explain (if requested). r Narrow the information gap. r Give the information slowly, step by step. r The details may not be remembered but the way you do it will be. 7. Listen to the patient’s concerns. r Allow time for the patient to express his or her feelings. r Perhaps ask, ‘What are your main concerns at the moment?’ 89
Simon Noble
8. Encourage the patient to express his or her feelings r This is important because it shows your empathy. r Perhaps say, ‘This must all have come as a bit of a shock for you.’ 9. Summary and plan. r Summarise concerns, plan treatment and foster hope. When someone is given devastating news, it is worsened by the thought that there is nothing that can be done. False reassurance is inappropriate, but offering hope in the form of an active approach to symptom control, goal setting and making realistic plans offers some control in a distressing situation. Some clinicians will make a record (written or recorded) of the consultation for the patient to keep and review at a later occasion. 10. Offer your availability. r The patient may need further information because he or she may not have remembered all the details. r The patient may need continuing support because psychological adjustment may take weeks or months. Following are a number of mistakes commonly made during the consultation: r Not checking prior knowledge. r Using jargon. r Giving too much information too quickly. r Not allowing for silence. r Showing discomfort when the patient is distressed.
Not checking prior knowledge One of the commonest mistakes is not checking what the patient already understands about his or her condition. Do not assume anything. Just because one of your colleagues has documented that the patient has been fully informed about the diagnosis and prognosis does not mean he or she has understood or retained the information. Studies have shown that patients can recall only 30% of what is said to them in a consultation and so it is always important that you check beforehand what they know.
Using jargon When we feel uncomfortable in a consultation, we tend to use more and more technical words to re-establish our status and importance. This commonly happens when patients become upset or angry. Jargon puts up a barrier between us and the patient, reducing the likelihood of clear communication.
90
Not allowing for silence If there is an uncomfortable pause, we may feel that we have to fill the silence with more information or an inappropriate comment – this should always be avoided. Silence is a valuable tool in communicating. It helps by showing that you are listening and by giving the patient time to assimilate the news, to react to it and to ask questions. As a patient once explained, ‘When you give me bad news, you will pause and all you can hear is silence. Well, all I hear is noise; it’s just that it is internal. When you speak you are interrupting me getting my head together.’
Discomfort with patient distress BAD NEWS IS BAD NEWS! YOU CAN’T TURN IT INTO GOOD NEWS! By definition, bad news is going to upset the receiver. If a patient cries, he or she is expressing an appropriate emotion in response to devastating news, and we need to be comfortable with that. The problem is that we may feel guilty that we have made them cry and may feel compelled to say something to ‘make it better.’ What happens in practice is that we end up giving patients false hope or reassurance, which often makes it harder for them to come to terms with their illness later on. Sometimes it will pave the way for a highly complicated bereavement.
Giving too much information too quickly The main task in breaking bad news is taking a patient from a point where he or she knows nothing to a point where he or she understands everything, if he or she wants to. The skill is in slowing down the transition of information so that the patient can handle it. Too much too soon will lead to psychological disturbance and possibly denial. The ‘warning shot’ technique helps test the water and find out how much the patient will allow you to tell. Sometimes the patient will make it quite clear that they do not want any more information; this wish should be respected. When talking to relatives, the principles of communication are the same as talking with patients. However, it is important to acknowledge that their agenda or concerns may differ from those of the patient. Anger and distress are frequently encountered and should be handled sensitively and calmly. Listening to and addressing each of their concerns is essential, although your duty of care must always remain with the patient. Don’t forget that you must respect the patient’s right to
Palliative care
confidentiality and discuss his or her care only if given permission to do so.
Financial considerations The impact of advanced cancer is not limited to the physical and emotional. Patients and families commonly experience financial difficulties, especially when the patient or main carer is self-employed. The role of the social worker as part of the multidisciplinary team is important in ensuring that financial aspects of the patient’s life are addressed. Simple things like accessing critical illness payments and pension plans, and arranging mortgage payment holidays, can all contribute to making life more manageable. Most patients will be entitled to financial assistance in the form of benefits. In assessing someone for benefits there are several things that must be known: r The patient’s age (whether he or she is younger or older than 65). r The prognosis of the patient’s condition (i.e. whether his or her prognosis is less than 6 months). r Whether the patient was working and paid National Insurance contributions before he or she became ill. Attendance allowance is paid to individuals over the age of 65 who need help looking after themselves; it pays for carers to ‘attend’ on the patient. The disability living allowance is paid to people under the age of 65 who need help looking after themselves. Incapacity benefit is for people who have worked and had previously made National Insurance contributions. They usually claim it once statutory sick pay (SSP) has ended or if they do not qualify for SSP. Those who have not paid enough National Insurance and who do not qualify for incapacity benefit are entitled to claim income support. The carer’s allowance is not claimable by the patient. Someone, a partner for example, who is looking after the patient can claim it. The DS-1500 is commonly claimed by patients with terminal disease. It is a way of claiming disability benefits for patients under ‘special rules,’ usually when the patient has a terminal condition with a prognosis of less than 6 months.
Ethical decision making The management of cancer patients is not solely guided by evidence-based guidelines. Sometimes decisions will
need to be made for which there is no clear-cut answer. In weighing up the pros and cons of a decision, it is useful to base it around the following ethical domains: r Autonomy. r Beneficence. r Non-maleficence. r Justice.
Autonomy Autonomy gives the patient the opportunity to make known his or her wishes and to make an informed decision. For a patient to do this, he or she needs the ‘capacity’ to make such a decision. For a patient to have capacity, he or she must be able to: r Take in information. r Understand the information. r Retain it. r Make a decision based on the given information. If a patient has capacity to make a decision, teams should respect his or her wishes. However, doctors are not obliged to give a treatment against their better clinical judgment if they feel it is futile.
Beneficence Professionals should make decisions with the intention of doing what is most likely to benefit the patient – the intention to do good.
Non-maleficence Conversely, non-maleficence dictates that when making a decision it should be done with the intention of doing no harm.
Justice Justice guides us to do what is fair, not only for the patient in question but also for other patients as a whole. It also ensures that we practice within what is legal within our society.
Areas of current interest The main areas of development in palliative care lie with its place in service delivery. It is now considered an integral part of cancer care, and all patients should be able to access palliative care if it is needed.
91
Simon Noble
Palliative care teams have an increasing involvement in managing the late effects of curative treatment such as postmastectomy pain or postchemotherapy symptoms such as peripheral neuropathy. There continues to be an interest in the field of opioids’ rotation, by which switching the type of opioids (e.g. morphine, hydromorphone, oxycodone) may improve pain control without an increase in the dose.
Ongoing research There is an increasing amount of research into the complexities of neuropathic pain and the dorsal horn remodelling that occurs secondary to cancer. Currently, several large clinical trials are looking into the management of the anorexia/cachexia syndrome, breathlessness, and fatigue.
REFERENCES Baines, M. J. (1997). ABC of palliative care: nausea, vomiting, and intestinal obstruction. B. M. J., 315, 1148–50. Barraclough, J. (1997). ABC of palliative care: depression, anxiety, and confusion. B. M. J., 315, 1365–8. Breitbart, W., Marotta, R., Platt, M. M. et al. (1996). A double-blind trial of haloperidol, chlorpromazine, and lorazepam in the treatment of delirium in hospitalized AIDS patients. Am. J. Psych., 153, 231–7. Cathcart, F. (2006). Psychological distress in patients with advanced cancer. Clin. Med., 6, 148–50. Endicott, J. (1984). Measurement of depression in patients with cancer. Cancer, 53, 2243–9. Fallon, M. and McConnell, S. (2006). The principles of cancer pain management. Clin. Med., 6, 136–9. Glare, P., Pereira, G., Kristjanson, L. J. et al.(2004). Systematic review of the efficacy of antiemetics in the treatment of
92
nausea in patients with far-advanced cancer. Support. Care Cancer, 12, 432–40. Greer, S., Moorey, S., Baruch, J. D. et al. (1992). Adjuvant psychological therapy for patients with cancer: a prospective randomized trial. B. M. J., 304, 675–80. Hall, E. J. and Sykes, N. P. (2004). Analgesia for patients with advanced disease: 1. Postgrad. Med. J., 80, 148–54. Jackson, K. C. and Lipman, A. G. (2006). Drug therapy for anxiety in palliative care (Cochrane review). The Cochrane Library, issue 3. Chichester, UK: John Wiley and Sons. Lawlor, P. G., Fainsinger, R. L. and Bruera, E. D. (2000). Delirium at the end of life: critical issues in clinical practice and research. J. A. M. A., 284, 2427–9. Lloyd-Williams, M., Friedman, T. and Rudd, N. (1999). A survey of antidepressant prescribing in the terminally ill. Pall. Med., 13, 243–8. Lloyd-Williams, M., Spiller, J. and Ward, J. (2003). Which depression screening tools should be used in palliative care? Pall. Med., 17, 40–3. Mannix, K. (2006). Palliation of nausea and vomiting in malignancy. Clin. Med., 6, 144–7. Meyer, T. J. and Mark, M. M. (1995). Effects of psychosocial interventions with adult cancer patients: a meta-analysis of randomized experiments. Health Psychol., 14, 101–8. Rhodes, V. A. and McDaniel, R. W. (2001). Nausea, vomiting, and retching: complex problems in palliative care. CA Cancer J. Clin., 51, 232–48. Erratum in: CA Cancer J. Clin., 51, 320. Royal College of Physicians. (2000). Principles of Pain Control in Palliative Care for Adults. Working Party Report. London: Royal College of Physicians. World Health Organisation. (1990). Cancer Pain Relief and Palliative Care. WHO Technical Report Series, 804. Geneva: World Health Organisation.
FURTHER READING Watson, M., Lucas, C., Hoy, A. et al. (2005). Oxford Handbook of Palliative Care. Oxford: Oxford Medical University Press.
8
HEAD AND NECK Laura Moss and Chris Gaffney
Introduction It is not possible in a book with this format to discuss the treatment details for each tumour type, stage and subsite. This chapter highlights treatment principles and controversies. This chapter covers the anatomical sites of the lip, oral cavity, pharynx, larynx, nose, paranasal sinuses, salivary glands, middle ear and squamous carcinoma of unknown primary. The first section considers general features of head and neck cancer. Subsequent sections focus on specific tumour sites.
General features of head and neck tumours Types of head and neck tumour The majority of head and neck cancers are squamous carcinomas, which account for 90% of all cases. The range of tumours is shown in Table 8.1.
Incidence and epidemiology of head and neck tumours r Approximately 8000 new cases occur per year in Engr r r r r
land and Wales (CRUK, 2004). Approximately 3000 deaths occur per year in England and Wales (CRUK, 2004). The UK annual incidence is approximately 14/100 000 cases per year (CRUK, 2004). Regional variation in the UK is 8 per 100 000 in the Thames region compared to 13 to 15 per 100 000 in Wales and northwest England (NICE, 2004). Head and neck tumours are more common in males than females (male-to-female ratio of 2:1), except for salivary gland tumours. Incidence increases from middle age, with peak incidence occurring in the 60- to 75-age group.
Risk factors and aetiology for head and neck tumours Prolonged exposure to environmental factors plays a key role. Smoking of cigarettes and cigars is related to cancers of the oral cavity, larynx and pharynx. Smoking of pipes is involved particularly in carcinoma of the lip. Tobacco chewing is also a risk factor (e.g. in Southeast Asia). Betel quid, which is chewed, is a mixture of betel nut and leaf, lime, catechu and tobacco. Buccal mucosa cancer may develop at the site of maximum contact. r Alcohol acts as both an independent risk factor and synergistically with smoking. The risk varies among tumour subsites: the association is strongest for pharyngeal cancer. r Wood dust causes adenocarcinoma of nose and paranasal sinuses. r Epstein Barr Virus (EBV) causes nasopharyngeal carcinoma. r Nitrosamines cause nasopharyngeal carcinoma. r Previous radiotherapy increases the risk of carcinoma. r Sun exposure causes lip cancer.
Anatomy The anatomical relations of the head and neck regions are discussed in the various subsections. The cervical lymph nodes are described as follows (AJCC, 2002): r Level I: Ia = submental, Ib = submandibular nodes. Boundaries include anterior and posterior bellies of digastric, mandible and hyoid. r Level II: upper deep cervical, extending from skull base to hyoid. r Level III: mid deep cervical, extending from hyoid to cricoid. r Level IV: lower deep cervical, extending from cricoid to clavicle. r Level V: posterior cervical, bounded by sternomastoid, trapezius and clavicle. 93
Laura Moss and Chris Gaffney
Table 8.1. Range of head and neck tumours Type of tumour
Examples
Benign
Glomus tumour Pleomorphic adenoma Pseudotumour Fibroma Papilloma, inverted papilloma Granuloma Haemangioma Juvenile angiofibroma Lymphangioma Angioma Adenoma Chondroma Ameloblastoma
Malignant primary
Of microscopic features, approximately 90% are squamous carcinomas arising from mucosal surfaces. Severe dysplasia or carcinoma in situ has a 15 to 30% risk of progression to frank malignancy.
Spread Head and neck cancer can spread: r Locally to involve local soft tissues, cartilage, bone, and peripheral nerves. r Via lymphatics to cervical lymph nodes. Remember that the usual pattern of lymphatic drainage may be disrupted by previous surgery or radiotherapy. Bilateral nodal disease is most common with nasopharynx, tongue base and supraglottic primaries. r Via the bloodstream, which is less common and usually occurs late, to the lung, bones and liver.
Squamous carcinoma Adenocarcinoma Neuroendocrine including olfactory neuroblastoma Mucoepidermoid, adenoid cystic and acinic cell carcinoma (salivary gland) Ameloblastic carcinoma, malignant
Screening for head and neck cancer There is no evidence that screening the general population or high-risk groups leads to a reduction in mortality or morbidity (British Association of Otorhinolaryngologists – Head and Neck Surgeons, 2002).
ameloblastoma Soft tissue sarcoma including Kaposi’s Osteosarcoma Lymphoma: NHL and Hodgkin lymphoma – usually in lymphoid tissue of Waldeyer’s ring Extramedullary plasmacytoma Mucosal melanoma Basal cell carcinoma Merkel cell tumour Malignant secondary
Metastatic spread from renal cell, thyroid, breast, lung, prostate or gastrointestinal carcinoma
r Level VI: paratracheal, pretracheal, extending from the hyoid to the suprasternal notch and laterally to the carotid sheath. r Level VII: inferior to suprasternal notch in the superior mediastinum.
Pathology Cancers typically arise on mucosal surfaces and may be exophytic or ulcerative. 94
Clinical presentation of head and neck cancers Symptoms depend on the anatomical site of primary disease and tumour stage. Symptoms related to the primary site include the following: r Persistent mucosal ulcer – oral cavity, oropharynx. r Dysphonia – larynx, hypopharynx. r Dyspnoea, stridor – larynx, hypopharynx. r Dysphagia, odynophagia and aspiration – larynx, hypopharynx. r Otalgia/referred pain – pharyngeal tumours. r Weight loss due to dysphagia or odynophagia or cachexia. r Anorexia. r Epistaxis, nasal obstruction, nasal discharge, nasal regurgitation – nose, nasopharynx, paranasal sinuses. r Hearing impairment – nasopharynx, middle ear. r Cranial nerve palsies – VII with parotid gland; infraorbital nerve with advanced maxillary sinus; II, III, IV and VI if cavernous sinus involved. r Cervical node enlargement. r Neck/facial mass lesion.
Head and neck
Symptoms of metastatic disease (rarely the presenting symptom) include the following: r Haemoptysis. r Bone pain. r Hypercalcaemia (the primary lesion may produce PTH-like protein or may be due to metastatic disease). Examination techniques for head and neck cancer are as follow: r Inspect and palpate the mucosal surfaces of the oral cavity (remove dentures) and inspect the pharynx. r Look for tongue and soft palate movement. r Look for trismus (inability to open the jaw due to parapharyngeal space invasion). r Examine the nose and ear if there are symptoms of nasal obstruction, discharge or regurgitation, epistaxis, reduced hearing or ear discharge. r Check laryngeal movement over the prevertebral fascia (loss of ‘crepitus’ indicates a locally advanced laryngeal tumour). r Carry out indirect laryngoscopy or flexible nasendoscopy. If it is difficult to obtain an adequate view because of gagging, consider local anaesthetic spray. If you are in a clinical examination and it is not possible to see the larynx well under examination conditions, tell the examiner what you are looking for: mucosal ulceration, asymmetry, pooling of saliva in the pyriform fossae and vocal cord movement (i.e. whether normal, reduced or fixed). r Carefully palpate the cervical lymph nodes (sensitivity and specificity for tumour involvement is 60 to 70%).
r CT or MRI to assess primary site (size, site, local extension, presence and extent of cartilage involvement, synchronous primary lesion) and cervical nodes (size, level, number and architecture). Note that dental amalgam produces an artefact that obscures CT findings in the oral cavity.
Staging for distant metastases
r CT of the thorax is used to look for synchronous primary tumours, mediastinal lymphadenopathy, or pulmonary metastases. CXR lacks sensitivity. The consensus guideline standards for audit include CT thorax in 90% of cases prior to treatment planning (British Association of Otorhinolaryngologists – Head and Neck Surgeons, 2002). r US/CT liver may be indicated if there is locally advanced disease or if there are abnormal liver function tests. r An isotope bone scan may be indicated to assess bone pain but is not ideal for assessing mandibular invasion. r 18 FDG-PET is indicated for assessment of possible tumour recurrence where conventional imaging may fail to differentiate between posttreatment fibrosis/ oedema and recurrent malignancy. It may also be considered for cancers of unknown primary when determining the site of origin may influence management (Intercollegiate Standing Committee on Nuclear Medicine, 2003).
Pretreatment assessment
Investigation and staging in head and neck cancer: general considerations Diagnostic and staging investigations will depend on the primary site and associated symptoms. The following is a list of commonly used investigations for the head and neck cancer patient.
Local staging
r Flexible or rigid nasendoscopy (depending on the
r Dental assessment with an orthopantomogram. This may also demonstrate mandibular invasion in advanced oral cavity carcinoma. r Nutritional assessment. Elective placement of a gastrostomy may be needed if there has been significant weight loss before diagnosis or if there are plans to carry out either an extensive resection resulting in impaired tongue mobility or swallowing function or irradiation of a large volume of mucosa. r Speech and swallow assessment.
anatomical site) in the clinic.
r Examination under anaesthesia, which may include panendoscopy, microlaryngoscopy, direct laryngoscopy, oesophagoscopy, bronchoscopy and biopsy depending on the symptoms, clinical findings and radiology. r US neck; fine needle aspiration cytology (FNAC) of suspicious cervical nodes. r US parotid with possible FNAC.
Treatment overview: general aspects in head and neck cancer The choice of treatment modality depends on tumour site, disease stage, pathological subtype, performance status, co-morbidity, patient’s wishes, functional outcome, cosmetic result and survival rate. Assessment in a multidisciplinary clinic is essential. 95
Laura Moss and Chris Gaffney
It is necessary to decide on a treatment modality for the primary site as well as the need for elective neck treatment in the absence of clinically apparent cervical nodes. Is resection possible with primary closure or is reconstruction of the defect required? The decision whether to proceed with surgery or radiotherapy will depend on the following: r Cure rate: is the survival outcome better with one modality compared to the other? r Operability of the disease: is there skull-base involvement, encasement of carotid vessels and so on that would make the disease unresectable? r Patient co-morbidity: is the patient at risk from a prolonged general anaesthetic? r Functional outcome: will physiological functions of speech, swallowing and breathing be better with one of the options? r Cosmetic result/disfigurement/altered body image. r Patient preference. Consider nutritional support: will nasogastric tube or gastrostomy/jejunostomy be required? Consider introduction to a patient previously treated in the same manner. This is often arranged prior to laryngectomy operations.
Elective nodal treatment Elective nodal treatment is usually performed when the risk of occult node involvement is greater than 20% (British Association of Otorhinolaryngologists – Head and Neck Surgeons, 2002), which is generally accepted practice but there is no strong evidence. The risk of occult node involvement is greater than 20% in most sites (except lip, early glottis, lower alveolar ridge). If the patient is undergoing microvascular reconstruction and entry to the neck is required then nodal dissection is usually performed. Elective nodal radiotherapy (RT) instead of neck dissection with 50 Gy in 25 fractions can achieve local control rates of 90 to 95%.
Role of surgery in head and neck cancer Surgery for individual subsites is discussed in the relevant sections.
Surgery for lymph node areas Lymph node dissection types are as follow: r Radical, involving removal of the spinal accessory nerve, internal jugular vein, sternomastoid, as well as lymph nodes levels I to V. 96
r Modified radical, which preserves one or more of the aforementioned structures.
r Selective neck dissection, which involves removal of specified lymph node groups. For pN designation following selective neck dissection six or more lymph nodes are needed in the resection. r Extended radical neck dissection, which removes one or more additional lymphatic or non-lymphatic structures.
Role of radiotherapy in head and neck cancer: general aspects Radiotherapy planning issues in general: patient positioning (The chapter does not discuss in detail CT-planned conformal radiotherapy or intensity-modulated radiation therapy.) Patient positioning: r The patient’s position should be comfortable, reproducible, and such that it is possible to acquire planning images and verify treatment delivery. An extended position should be used for nasopharyngeal and parotid tumours, and a straight position for larynx, oral cavity and oropharynx and paranasal sinuses. If posterior neck electrons are to be added, however, the spine is to be kept in a straight position. r Consider a mouth bite for oral cavity and paranasal sinus treatments. Beam arrangement and energy: r Opposed lateral photon fields with the patient supine are used to treat the majority of oral cavity, laryngeal and pharyngeal cancers. r Choosing the angle of orientation of lateral fields is important when taking cervical spine curvature and shoulder position into account. r When deciding the position of the junction between lateral and anterior neck fields, consider where the high-risk areas lie and try to avoid the junction there. The lower the junction and, hence, the longer the lateral fields, the greater the mucosal surface included and the acute mucosal reaction. r If the posterior neck nodes are involved in the lower neck, an anterior and posterior field arrangement may be needed to provide adequate dose coverage. r For posterior neck electron fields, the electron beam energy is chosen following assessment of depth of nodal groups at risk and spinal cord depth. Measurements are taken from cross-sectional imaging or from the transverse patient outline. The electron beam
Head and neck
r
r r r
energy is usually 9 to 12 MeV if the patient has no palpable posterior nodes and has an average neck size. If higher-energy electrons are used, great care must be taken to calculate dose to the spinal cord from the two photon phases and the electron beam. Shielding: If central spinal shielding in the anterior neck field is omitted, then a. Check the distance from the skin surface/stoma to the spinal cord to ensure spinal cord tolerance is not exceeded. b. The anterior field beam divergence at depth will contribute dose to the spinal cord if the lateral and anterior beams are matched at the skin surface. Therefore, insert a small shielding block in the inferior/posterior corner of lateral portals to prevent excessive dose to the cervical spinal cord. Alternative methods include half-beam blocking or angling the match line so that the angle exceeds the divergence of the anterior field. Portals are shaped with Cerrobend® blocks or multileaf collimators. Leaves are often 10 or 3 mm. Dose: For the anterior neck field, is the dose prescription point going to be at the skin surface or at 2 to 3 cm depth where the lymph nodes are? Consider the proportion of the dose from phases 1 and 2 when large lateral fields are required: because spinal cord tolerance is less than the dose required to treat head and neck cancers radically, the full dose cannot be given in a single phase. When deciding on the relative doses between the phases you must stipulate the acceptable dose to the spinal cord from the phase 1 photon field to remain within cord tolerance.
General indications for postoperative radiotherapy in head and neck cancer Primary site factors: r Positive margin.* r Close margins (≤ 5 mm).** r Poorly differentiated. r Lymphovascular invasion. r Perineural invasion.** r T3 or T4 disease. r CIS or dysplasia at resection margins. r Invasion of adjacent soft tissue.** r Multicentric primary. Lymph node factors: r Extra capsular spread.* r ≥ 2 nodes.**
r >1 lymph node level.** r Single node > 3 cm.** (A single asterisk indicates high-risk factors for locoregional recurrence; a double asterisk indicates intermediate risk factors. A combination of two or more intermediate factors would also be classified as high risk.)
Management of acute radiotherapy reactions For management of skin reactions, r Avoid perfumed soap and toiletries. r Avoid wet shaving. r Use aqueous cream for erythematous and dry skin. r Use hydrocolloids like Intrasite® gel or Geliperm® for moist desquamation. r Heat and moisture exchanger base plates will need to be avoided if there is peristomal skin reaction. r Avoid sun exposure. For oral mucositis, r Avoid spicy foods, alcohol, very hot or cold foods/ liquids and smoking. r Perform regular mouth washing (e.g. normal saline) and dental hygiene. r Use benzydamine hydrochloride mouthwash (diluted with water to reduce pain on initial contact with mucosa). r Use topical barriers with carmellose (e.g. Orabase® ) or with povidone and sodium hyaluronate (e.g. Gelclair® ). r There is conflicting evidence with regard to antibacterial rinses/lozenges. r Treat oral candidiasis. For odynophagia, r Follow a soft diet. r Use raspberry mucilage. r Use systemic analgesia (e.g. NSAID, compound codeine preparation, morphine sulphate solution; alcohol-free preparations are better tolerated) if severe. r Use sucralfate suspension. For xerostomia: Xerostomia is the subjective sensation of mouth dryness. There is a poor correlation between subjective reports of xerostomia and gland function. r Take frequent sips of water and maintain adequate daily fluid intake. r Sugar-free gum or pastilles may stimulate saliva flow if residual function is present. r Avoid medications that may exacerbate the situation (e.g. anticholinergics, antispasmodics, antipsychotics, antidepressants, diuretics, statins, opioids, NSAIDs or antihypertensives). 97
Laura Moss and Chris Gaffney
r Use saliva stimulants containing a combination of citr r
r r
ric and mallic acids (e.g. SST® , Salivix® ). Use artificial saliva preparations (e.g. Biot`ene Oralbalance® gel). Use pilocarpine 5 to 10 mg t.d.s. Some studies report increased salivary flow but this often fails to translate into improved quality of life. There are a significant number of contraindications and side effects. Follow scrupulous oral hygiene. Undergo regular dental review.
Radiotherapy fractionation Altering conventional fractionation regimens aims to improve the therapeutic index and to increase dose intensity (high total dose in a short overall treatment time) and may be achieved by increasing the total dose (via hyperfractionation, i.e. decreased dose per fraction) and/or decreasing overall treatment time (via acceleration). Decreasing the dose per fraction can protect late-responding tissue. Acceleration aims to overcome tumour cell repopulation. Randomised studies fall into three types: 1. Increased total dose, same overall treatment time (i.e. hyperfractionation), e.g. EORTC 22791: 1.15 Gy twice daily up to 80.5 Gy. This regimen showed a significant improvement in local control when compared to conventional fractionation (Horiot et al., 1992). 2. Same total dose, decreased overall treatment time (i.e. moderate acceleration), e.g. DAHANCA (six fractions per week). This treatment showed increased acute toxicity but a significant improvement in tumour control probability (Overgaard et al., 2003). 3. Decreased total dose and markedly decreased overall treatment time (i.e. very accelerated), e.g. CHART (54 Gy in 1.8 weeks). This treatment showed a marked increase in acute toxicity and decreased late morbidity (Dische et al., 1997). A meta-analysis of randomised studies (>6500 patients) showed a 3% improvement in survival and a 6% improvement in locoregional control for modified fractionation. Hyperfractionation showed the greatest benefit, with a 9% improvement in survival at 5 years (Bourhis et al., 2002).
Brachytherapy Services should be cancer network based. There will often be a need for cross-network supra-regional ser98
vice provision for highly specialised treatments (Royal College of Radiologists, 2001).
Role of chemotherapy in head and neck cancer: general aspects Neoadjuvant or induction chemotherapy The rationale for neoadjuvant chemotherapy is that drug delivery is better in well-vascularised untreated tumours, treatment-na¨ıve patients are far more tolerant of the associated treatment toxicities, and the potential exists for eradication of microscopic metastatic disease. Trial data have shown an increase in local control when neoadjuvant chemotherapy is combined with surgery or radiotherapy with some trials also showing an increase in survival. A meta-analysis showed a non-significant 2% survival benefit when looking at various chemotherapy regimens (Pignon et al., 2000) but a further analysis of nearly 2500 patients receiving only platinum and 5fluorouracil regimens showed a 16% improvement in relative survival and a 5% absolute survival benefit at 5 years (Monnerat et al., 2002).
Concurrent chemoradiotherapy A meta-analysis of 63 randomised trials of chemotherapy used in different clinical settings with more than 10 000 patients showed a survival benefit of 4% at 5 years. This improvement increased to 8% at 5 years with concomitant chemotherapy use (Pignon et al., 2000). This meta-analysis only addressed concurrent chemotherapy with conventionally fractionated radiotherapy. The use of concurrent chemotherapy is not proven with accelerated or hyperfractionated regimes. There is a significant increase in acute toxicity, especially mucositis; therefore, it is important to select patients with good performance status and limited comorbidity. In nasopharyngeal cancer management, this treatment approach has shown a reduction in locoregional recurrence and distant metastases as well as an increase in overall survival.
Postoperative concurrent chemoradiotherapy Despite the use of surgery and postoperative radiotherapy, high-risk squamous carcinoma (e.g. positive margins and extracapsular lymph node spread) of the head and neck frequently recurs in the locoregional area. Two phase III trials (Cooper et al., 2004; Bernier et al., 2004) reported by the RTOG and EORTC respectively, which looked at the addition of concurrent chemotherapy to
Head and neck
radiotherapy, demonstrated an improvement in locoregional control, with the EORTC study also reporting an increase in overall survival (53 versus 40% at 5 years).
Palliative chemotherapy Untreated metastatic or recurrent squamous cell carcinoma has a median survival of 4 months. Although cisplatin-containing chemotherapy has shown a survival advantage in the region of 8 to 10 weeks when compared to best supportive care, chemotherapy is more often used in this setting to palliate symptoms and restore function rather than to prolong life. Response rates in the order of 20 to 30% are seen, with most responses being partial and often short lived. However, even partial response can be associated with a significant improvement in quality of life when these patients are often dealing with cosmetically disfiguring disease and restricted breathing, swallowing and speech.
Rehabilitation following radical treatment for head and neck cancer Many surgical procedures may result in significantly altered physical appearance and physiological function. For example, following a laryngectomy, the patient may experience loss of sense of smell, impairment of the Valsalva manoeuvre and ability to strain and loss of humidification and air warming. r Social and psychological support are very important. r Communication can be facilitated by surgical voice restoration with speech valves in a tracheooesophageal fistula, electronic larynx, oesophageal voice and written communication aids (e.g. light writers, palm top aids, text messaging and e-mail). Speech and language therapy input can help in the long term for both voice and swallowing. Access to videofluoroscopy is important to assess swallowing function. r Long-term nutritional assessment and supplementation are often required. r Dental follow-up is required. r Prosthetics include dentures, dental implants and prostheses (nose, ear, eye).
Follow-up after radical treatment for head and neck cancer Aims of follow-up include the following: r Early detection of residual or recurrent disease with a view to salvage treatment. r Detection of second primary.
r Psychological support, including for modification of smoking and alcohol behaviour.
r Late morbidity and quality-of-life assessment. The duration of follow-up is usually 5 years (90% recur within first 2 years; NICE, 2004).
Prognosis in head and neck cancer The status of the cervical nodes may be the single most important prognostic factor. The prognosis is affected by the number of nodes, number of nodal levels involved and the presence of extracapsular spread. Survival is significantly worse when metastases involve lymph nodes beyond the first echelon of lymphatic drainage (British Association of Otolaryngologists – Head and Neck Surgeons, 2002).
Areas of current interest in head and neck cancer Intensity-modulated radiation therapy (IMRT) IMRT is a type of conformal therapy that confers a high radiation dose to the target volume and a low dose to sensitive structures. The intensity of the beam is varied across the treatment field, allowing a single accelerator beam to function as many smaller beams. A highly conformal plan can usually be achieved using five or seven equally spaced fields. The dose is delivered to the target through a greater amount of surrounding tissue. Although the volume of tissue exposed to a low radiation dose is greater, the volume of tissue exposed to high dose is less. Patient and organ movement are of great concern. The aim is to treat different targets simultaneously with different total doses and fraction sizes. IMRT is usually inverse planned; that is, treatment planning starts with a description of the desired result, unlike conventional treatment planning in which the computer system simulates dose distribution from a treatment planner’s defined set of beams and beam weights. Collimation can either be static, with a ‘stepand-shoot’ method or dynamic by changing beam shape during radiation delivery. The aims of IMRT are to minimise acute and late morbidity and to allow the possibility of dose escalation with improved local tumour control. Areas of clinical interest include (1) conforming target volumes to be able to treat radically at-risk posterior lymph node groups while avoiding excessive spinal cord irradiation (i.e. the ability to deliver horse-shoe-shaped 99
Laura Moss and Chris Gaffney
dose distribution) and (2) use of parotid-sparing techniques to reduce xerostomia.
Epidermal growth factor receptor (EGFR) signalling EGFR is a transmembrane protein with tyrosine kinase activity that transduces signals from the epithelial cell surface to the intracellular domain. Aberrant signalling via EGFR plays an important role in the carcinogenesis of squamous cell carcinoma of the head and neck, and tumours tend to express high levels of EGFR. A high degree of EGFR expression correlates with a poorer clinical outcome. Cetuximab, an anti-EGFR monoclonal antibody, in combination with radical radiotherapy in locally advanced squamous carcinoma of the head and neck, has shown promise in increasing median survival (54 months versus 28 months in those receiving radiotherapy alone; p = 0.02) in a phase III study with more than 400 patients (Bonner et al., 2004). However, in this study cetuximab was not compared to concurrent chemoradiotherapy.
Current clinical trials At the time of this writing, there were two trials registered with the National Cancer Research Network looking at treatment for head and neck cancer (www.ncrn.org.uk, accessed September 2006): EaStER Feasibility study – early glottic cancer: endoscopic excision or radiotherapy, feasibility study. PARSPORT – a multicentre randomised study of parotid-sparing IMRT in patients with head and neck cancer.
often present when advanced. Dysphonia can also arise from recurrent laryngeal nerve involvement. Patients can also present with dyspnoea, stridor, pain due to cartilage invasion, non-specific sore throat (especially when swallowing) and neck mass.
Investigation and staging Clinical investigation uses IDL or nasendoscopy (to look for laryngeal mucosal appearance – symmetry, exophytic or endophytic disease, vocal cord movement, salivary pooling, oedema, other ENT primary sites) and examination of the neck for lymphadenopathy/direct tumour extension. Laryngoscopy/microlaryngoscopy and biopsy are used to assess the site and extent of disease and other primary lesions. CT or MRI are used to assess the primary site (extra laryngeal spread, e.g. cartilage involvement, preepiglottic space involvement) and cervical nodes. Movement artefact may impair MRI images. Coronal MRI images can be particularly helpful if partial laryngeal surgery is planned. MRI is more sensitive, but less specific, compared with CT for cartilage involvement. Comparing CXR and CT of the thorax, a normal CXR does not ‘exclude’ metastases or synchronous primary because of its low sensitivity. However, the rate of detection of second primaries and unexpected lung metastases is quite low and some clinicians do not feel that CT imaging can be justified in this case.
Staging classification The staging classification for carcinomas of the larynx is shown in Table 8.2.
Carcinoma of the larynx Anatomy of the larynx Subsites of the larynx include the r Supraglottis: epiglottis, aryepiglottic folds, arytenoids and false cords. r Glottis: vocal cords and anterior and posterior commissures. r Subglottis: lower border of the glottis to lower border of cricoid cartilage.
Clinical presentation Glottic cancers often present at an early stage because of dysphonia (hoarse voice), whereas supraglottic cancers 100
Treatment overview for squamous carcinomas of the larynx Glottis It is important to know whether there is supraglottic or subglottic extension and whether the commissures are involved. In stage Tis (in situ disease), the entire mucosal abnormality needs to be examined to exclude invasive disease; therefore, an excision biopsy should be performed. If there are rapid or multiple recurrences, further endoscopic excision or radiotherapy should be considered. The situation may arise in which the lesion under review is clinically progressing but histology fails to confirm
Head and neck
Table 8.2. Tumour and nodal TNM stages for carcinoma of the larynx Stage
Description
Glottis T1
Limited to vocal cord(s), normal mobility: (a) one cord (b) both cords
T2
Supraglottis, subglottis, impaired cord mobility
T3
Cord fixation, paraglottic space, thyroid cartilage erosion
T4a
Through thyroid cartilage; trachea, soft tissues of neck; deep/extrinsic muscle of tongue, strap
T4b
Prevertebral space, mediastinal structures, carotid artery
muscles, thyroid, oesophagus
Supraglottis T1
One subsite, normal mobility
T2
Mucosa of > 1 adjacent subsite of supraglottis or glottis or adjacent region outside the
T3
Cord fixation or invades postcricoid area, pre-epiglottic tissues, paraglottic space, thyroid
T4a
Through thyroid cartilage; trachea, soft tissues of neck; deep/extrinsic muscle of tongue, strap
T4b
Prevertebral space, mediastinal structures, carotid artery
supraglottis; without fixation cartilage erosion muscles, thyroid, oesophagus
Subglottis T1
Limited to subglottis
T2
Extends to vocal cord(s) with normal/impaired mobility
T3
Cord fixation
T4a
Through cricoid or thyroid cartilage; trachea, deep/extrinsic muscle of tongue, strap muscles,
T4b
Prevertebral space, mediastinal structures, carotid artery
thyroid, oesophagus
Nodal stages: all subsites N1
Ipsilateral single ≤ 3 cm
N2a
Ipsilateral single > 3 to 6 cm
N2b
Ipsilateral multiple ≤ 6 cm
N2c
Bilateral, contralateral ≤ 6 cm
N3
> 6 cm
Adapted from UICC (2002).
invasion. In these cases it is often better to treat as for invasive disease rather than continue to observe. In stage T1 to T2, N0, give radical radiotherapy to the larynx only (there is a low incidence of occult nodes due to sparse lymphatics; therefore, there is no need for elective nodal RT) or conservation surgery (need to preserve more than half of the free margin of contralateral cord). If there is extensive superficial T2 disease, there will be a better functional outcome with radiotherapy. Stage T3 is a heterogeneous group. Treatment options include:
r Primary radiotherapy and salvage surgery. r Total laryngectomy and postoperative RT. r Concurrent chemoradiotherapy. If the tumour is bulky and causing respiratory compromise or stridor, consider surgical debulking before starting radical RT in order to maintain an adequate airway throughout. Stage T4 requires total laryngectomy and neck dissection and postoperative RT. If the patient is medically unfit for surgery or declines surgery then proceed with radiotherapy with or without chemotherapy. 101
Laura Moss and Chris Gaffney
Supraglottis There is a high incidence of clinical and occult cervical involvement, therefore, consider elective bilateral neck RT, especially if there is a midline tumour.
also reduces the extent of the lateral neck skin reaction. The gantry angle used is usually 45 to 60◦ in order to spare the spinal cord. This technique is more difficult if the volume is not parallel to the couch top.
Subglottis
Advanced glottis: radiotherapy
Patients often present with late-stage disease. A total laryngectomy and postoperative radiotherapy are usually required. If paratracheal nodes are positive, include the mediastinum in the radiation field.
The PTV depends on the disease extent, the nodal stage, and the operation status. If there is a T3 or T4 glottic or transglottic tumour, both sides of the neck should be irradiated. Use parallel-opposed lateral fields in a two-phase technique to stay within spinal cord tolerance. The posterior neck nodes (i.e. those overlying the spinal cord) are treated with electron fields matched at the skin surface to the phase 2 photon fields. To select the appropriate electron beam energy, refer to the patient’s physics plan and measure the depth of the spinal cord from the skin surface (or measure from CT/MRI images). Choose an electron energy that will cover the depth of the nodal areas sufficiently but without exceeding cord tolerance. The typical beam energy is 9 to 12 MeV. A lower anterior neck field may need to be added. Superiorly, the anterior neck field is matched to the lower edge of the lateral photon fields; inferiorly, the clavicular heads are covered and the lateral field edge lies at the junction of the medial two-thirds and the lateral third of the clavicle. If possible, shield the lung apices and the spinal cord. If a tracheostomy is present and the surrounding area is thought to be at risk, the central shielding will need to be omitted. In these cases remember to add shielding to the posterior/inferior aspect of the lateral photon field to prevent overdosing the spinal cord as a result of beam divergence from the anterior field.
Radiotherapy techniques for squamous carcinoma of the larynx General considerations When treating squamous cell carcinoma of the larynx with radiotherapy, take the following into account: r Immobilisation/beam direction shell. r The patient lies supine. r The cervical spine is straight. r Orthogonal films or planning CT are performed. r Mark any nodes, relevant scars and the stoma. r No tongue bite is required. The contour change can be compensated for by wedges (early glottic tumours with small lateralopposed fields) or missing tissue compensators when there is a complex contour and variable separations throughout the volume.
Early glottis: radiotherapy
r Centre the volume at the level of the vocal cords (i.e.
1 cm below the palpable thyroid cartilage promontory). r Include the entire thyroid and cricoid cartilages. r Margins must allow for movement with respiration and subclinical disease. r Prophylactic nodal RT is not required. r Typical volume sizes: T1 tumours, 4 to 5 cm; T2 tumours, 5 to 7 cm. r Use parallel-opposed lateral photon beams (6 MV) or anterior-oblique wedged fields. If the anterior commissure is involved, either use bolus anteriorly or ‘underwedge’ to give the anterior part of the treatment volume between 5 and 10% higher dose than the posterior part. If the patient is obese or has a short neck, the shoulders may obstruct the lateral fields. Therefore, consider using two anterior-oblique wedged fields. The PTV outline will need to be drawn on a central contour if an orthogonal planning technique is used. This technique 102
Supraglottis: radiotherapy There is a high incidence of involvement of the upper deep cervical lymph nodes, even for stage N0 clinically. For stage T1 to T2, N0, treat the primary tumour, preepiglottic space and upper deep cervical lymph nodes. Shield the oral cavity and salivary glands as much as possible without compromising the nodal areas. If the posterior border of the disease lies close to the spinal cord then a two-phase technique will be needed. For node-positive disease, bilateral cervical nodes and SCF nodes should be included in the radiation field. A two-phase parallel-opposed lateral photon field technique should be used with direct electron fields for the posterior neck and a direct anterior neck photon field (as described earlier).
Head and neck
Subglottis: radiotherapy The target volume includes the primary site, pre- and paratracheal nodes, lower deep cervical nodes and superior mediastinum. Problems arise with the changing body contour (superior–inferior) and beam obstruction by the shoulders. 3D conformal planning is likely to give the best tumour coverage and it allows beam selection and appropriate shielding to be added, taking contour changes and shoulder position into account.
Table 8.3. Five-year survival for carcinoma of the larynx Stage
Relative survival for larynx (%)
1
80
2
66
3
55
4
37
Adapted from AJCC (2002).
Set-up Some centres use a 100 cm FSD for orthogonally planned volumes, whereas others use an isocentric technique.
Carcinoma of the oral cavity
Radiotherapy dose for squamous carcinoma of the larynx
Subsites include r Lip (some classify this area as skin); more than 90% of cases occur on the lower lip. r Floor of mouth. r Anterior two-thirds of tongue. r Buccal mucosa. r Hard palate. r Upper and lower alveolus and gingiva. r Retromolar trigone.
Lateral photon fields should be prescribed to the ICRU 50 reference point. Posterior neck electron fields can be prescribed as given doses or as the dose at dmax . An anterior neck photon field can be prescribed as a given dose on the skin surface or at a depth of 2 to 3 cm. Primary radiotherapy doses are as follows: r For early glottis, common fractionation schedules are r 55 Gy in 20 fractions over 4 weeks. r 60 Gy in 25 fractions over 5 weeks. r 66 Gy in 30 to 33 fractions over 6 to 6.5 weeks. r For other sites and stages, r 64 to 66 Gy in 30 to 33 fractions over 6 to 6.5 weeks to sites of clinical disease. r 50 Gy in 25 fractions over 5 weeks as elective nodal irradiation. r For postoperative radiotherapy, r 60 Gy in 30 fractions over 6 weeks to residual disease and high-risk sites. r 50 Gy in 25 fractions over 5 weeks as elective nodal irradiation.
Follow-up Most recurrences happen within 2 years. It is important to have a speech and language therapist available for rehabilitation and to support surgical voice restoration.
Prognosis Five-year relative survival for squamous carcinoma of the larynx is shown in Table 8.3.
Anatomy of the oral cavity
Investigation and staging Tumour assessment involves: r Direct and bimanual palpation: tumour size, depth of invasion and extension to other subsites. r Examining range of tongue movement and extent of tongue protrusion. r Examining for trismus. r Cervical node palpation. Cross-sectional imaging may underestimate the extent of gingival and hard palate involvement. For Tstaging, MRI is more accurate for small primary lesions and when dental amalgam is present because of the artefacts caused by the amalgam on CT. MRI of the tongue may show whether the ipsilateral neurovascular bundle is involved or if there is crossing of the midline and submucosal involvement in adjacent areas. These findings may directly affect the surgical approach with respect to suitability for partial glossectomy. MRI images taken in the follow-up period may show non-specific high signal in postoperative or postradiotherapy areas that may mimic recurrence. Early mandibular invasion may be difficult to detect clinically and radiologically. A combination of CT, MRI, 103
Laura Moss and Chris Gaffney
Table 8.4. Tumour and nodal TNM classification of oral cavity cancer Stage
Description
T1
≤ 2 cm
T2
> 2 to 4 cm
T3
> 4 cm
T4a (lip)
Through cortical bone, inferior
T4a (oral cavity)
Through cortical bone, deep/extrinsic
alveolar nerve, floor of mouth, skin muscle of tongue, maxillary sinus, skin T4b
Masticator space, pterygoid plates, skull base, internal carotid artery
N1
Ipsilateral, single ≤ 3 cm
N2a
Ipsilateral single > 3 to 6 cm
N2b
Ipsilateral multiple ≤ 6 cm
N2c
Bilateral, contralateral ≤ 6 cm
N3
> 6 cm
Adapted from UICC (2002).
and plain film radiography may be needed to assess for mandibular invasion. Isotope bone scanning is often not helpful for detecting mandibular invasion.
Staging classification The TNM summary for carcinomas of the oral cavity is shown in Table 8.4.
Treatment overview for oral cavity cancer Early stage T1 or small T2 oral cancer Radiotherapy and surgery are equally effective but functional outcome may be better after RT. RT techniques include EBRT and interstitial RT. (Avoid brachytherapy if there is involvement of the tongue tip or the adjacent periosteum.) Brachytherapy is usually only used for small superficial tumours with less than a 10% risk of lymph node metastases. Small tumours of the tongue and floor of the mouth may be suitable for a hairpin implant technique whereas thicker tumours would need the plastic loop technique because the separation between the limbs of the hairpin is only 12 mm. Suitable buccal mucosa tumours can often be encompassed by a single plane of plastic tube implants. 104
With regard to surgery, frozen section evaluation of margins is advisable to ensure adequate resection. Temporary tracheostomy may be required if the airway could be compromised because of perioperative bleeding or oedema. Other techniques in use (but which lack evidence) include photodynamic therapy, Mohs’ micrographic surgery and transoral laser excision.
Larger T2, T3 or T4 oral cancers
r Combination therapy is used. r Surgery is preferred if the periosteum is involved. Primary closure of the surgical defect is the preferred method but, if functional compromise is likely, local flap or free flap reconstruction may be needed. The choice of flap depends on the type and size of defect, the impact of the type of reconstruction on function, and the donor site morbidity. r RT is preferred if the oral commissures are involved.
Clinically node-negative oral cancer The incidence of occult lymph node metastases is 30 to 35%. Tongue lesions may metastasise to level II and III nodes rather than level I. Level IV nodes are also at significant risk.
Lip cancer Of T1 and T2 lesions, 5 to 10% are node positive. Surgical excision is generally the preferred option provided the commissures are not involved and the functional and cosmetic outcome is acceptable. A frozen section is used to assess margins: r For one-third or less of the lip, close primarily. r For one- to two-thirds of the lower lip, use local flap reconstruction. r More than two-thirds of the lower lip usually requires a microvascular free flap tissue transfer or a regional pedicled flap to reconstruct. Radiotherapy is used when patients are medically inoperable or when the expected functional or cosmetic outcome is deemed unsatisfactory. Radiotherapy techniques include external beam (kilovoltage or electron) radiotherapy, brachytherapy or a combination of the two. Gingival shielding helps to reduce mucositis.
Radical radiotherapy for oral cavity carcinoma Position and immobilisation
r The patient should lie supine.
Head and neck
r The cervical spine should be kept straight. r Use of a tongue bite keeps lips apart, depresses and reduces the mobility of the tongue and can spare mucosal reactions. A tongue bite is often poorly tolerated when oral mucositis has developed.
Table 8.5. Five-year relative survival for oral cavity and lip cancers Stage
Lip survival (%)
Oral cavity survival (%)
1
83
68
Lip
2
73
53
For stage T1 to T2, N0, the target volume is the primary plus a 2 cm margin. Elective nodal radiotherapy is not indicated because there is a very low risk of occult nodal metastases. Treat with electrons or orthovoltage and use a lead cut-out and shielding intraorally. Some tumours may be suitable for interstitial brachytherapy. A template technique using rigid needles implanted along the horizontal axis of the lip is used. A single plane may be used for superficial tumours but, for a deeper tumour, three or more sources may be needed. For a large primary, the target volume includes the primary and local nodes. If nodes are positive, treat all nodes.
3
62
41
4
47
27
Tongue TNM staging does not take tumour depth into account. For early disease, TI and T2 control rates are comparable with surgery or RT. The target volume includes the primary tumour with a 2 cm margin and ipsilateral submandibular and upper deep cervical nodes if there is well-lateralised disease (because spread to ipsilateral nodes is common but spread to contralateral nodes uncommon). If the primary tumour approaches the midline or if there is nodepositive disease, treat bilateral nodes using opposed lateral fields. It is possible for the first site of nodal metastases to arise in the junctional node between levels III and IV. The beam arrangement is usually an anterior and lateral wedged field. Aim to spare the contralateral oral cavity mucosa, the parotid, and the spinal cord. The anterior field will, however, result in an acute lip reaction.
Floor of mouth For anterior T1 and small T2 lesions, r Use surgery or RT (EBRT or brachytherapy). If an early tumour extends on to the ventral aspect of the oral tongue, RT may be preferable because surgery could result in functional morbidity. r Two anterior wedged fields spare the oral mucosa and parotids more than opposed lateral fields.
Adapted from AJCC (2002).
r There should be a 2 cm margin around the primary tumour in all directions (i.e. the target volume includes the anterior floor of mouth, oral tongue, anterior lower alveolus, and anterior submandibular triangle). r Any occult nodes are usually contained in the first echelon nodes (i.e. submental and submandibular). r It may be possible to irradiate a small anterior tumour with fields lying anterior to the parotids, thereby reducing the risk of xerostomia. For other sites and larger T2 to T4 tumours, the target volume is the primary disease and bilateral cervical nodes.
Buccal mucosa The target volume is the primary tumour and ipsilateral submandibular lymph nodes. Wedged anterior and lateral fields are used.
Lower alveolus The target volume includes the ipsilateral mandible, submental and submandibular lymph nodes. A lateral wedged and oblique–anterior field arrangement allows maximum sparing of the oral mucosa and lip.
Retromolar trigone The target volume includes the primary tumour and ipsilateral upper deep cervical nodes, including the pterygoid fossa. Anterior and posterior oblique wedged fields spare the contralateral parotid and mucosa. If lymph nodes are clinically positive, add an ipsilateral anterior neck field to treat the rest of the cervical node levels.
Prognosis for oral cavity cancer The 5-year survival is shown in Table 8.5. 105
Laura Moss and Chris Gaffney
Carcinoma of the oropharynx
Table 8.6. Tumour and nodal TNM summary of staging for oropharyngeal carcinomas
Anatomy of the oropharynx The oropharynx extends from the palate to hyoid. The subsites include: r Tonsil (fossae and pillars; 60% of oropharyngeal cases). r Base of tongue (25%) and vallecula. r Inferior surface of soft palate (10%) and uvula. r Posterior pharyngeal wall.
Stage
Description
T1
≤ 2 cm
T2
> 2 to 4 cm
T3
> 4 cm
T4a
Larynx, deep/extrinsic muscle of tongue, medial
T4b
Lateral pterygoid muscle, pterygoid plates, lateral
pterygoid, hard palate, mandible nasopharynx, skull base, carotid artery
Clinical presentation Patients present with the following: r Sore throat. r Painful swallowing. r Referred otalgia. r Trismus. r Impaired sensation over the anterior chin due to involvement of the inferior alveolar nerve. r Altered sensation over the lateral tongue due to involvement of the lingual nerve in the infratemporal fossa. r Impaired tongue movement, including protrusion. r Altered speech (nasal quality when tongue and soft palate are not moving normally). r Neck mass.
Investigation and staging It is usually possible to visualise the disease clinically by using a combination of direct inspection, mirror examination, and flexible endoscopy to assess the posterior and inferior extent. For EUA and biopsy, assess local extent for staging and determine whether it crosses the midline and whether there is any tongue base involvement because these two factors are important when considering surgical resection and likelihood of contralateral lymph node involvement. CT is better for imaging cortical bone. Oropharyngeal tumours have a higher incidence of distant metastases and it is therefore important to assess for sites of distant disease. MRI obtains superior discrimination of tumour from muscle and other soft tissue and it is therefore often better for staging the primary tumour. Gadolinium enhancement may demonstrate invasion along nerves. 106
N1
Ipsilateral single ≤ 3 cm
N2a
Ipsilateral single > 3 to 6 cm
N2b
Ipsilateral multiple ≤ 6 cm
N2c
Bilateral, contralateral ≤ 6 cm
N3
> 6 cm
Adapted from UICC (2002).
Sixty percent of patients present with involved lymph nodes but contralateral spread is relatively uncommon unless the tumour involves or crosses the midline; then the risk of bilateral or contralateral lymph nodes is 15%.
Staging classification Table 8.6 shows the TNM classification of oropharygeal cancers.
Surgical treatments Tonsil For the tonsil, T1 or T2 tumours can be treated by transoral resection. T3 or T4 lesions usually require resection with reconstruction of the resulting defect. To allow adequate exposure of the tumour site, mandibulotomy is often required. Reconstruction of the defect is usually by means of a radial forearm free flap or pectoralis major myocutaneous flap. If the mandible needs to be resected, it is important to try and preserve the rim because this results in a far better functional result than a segmental resection.
Tongue base T1 disease can often be resected with little morbidity but more advanced disease is difficult to manage surgically because of the significant functional morbidity that results.
Head and neck
Large tumours, for example, would require a total glossectomy and, because of the significant risk of aspiration following this procedure, a laryngectomy is usually needed as well.
Soft palate Tumours tend to arise on the free edge and present relatively early (T1 or T2). Resection of these can be effective, but for larger tumours problems arise with surgical treatment because of the functional morbidity associated with reconstruction.
field on the simulator or by CT planning for a PTV to include the primary site and nodal groups in continuity. If the tonsillar mass extends medially to involve the soft palate and approaches the uvula or extends inferiorly to infiltrate the tongue base, then the GTV and margins are increased and opposed lateral portals
Radical radiotherapy for oropharynx cancers Except for well-lateralised tumours of the tonsil, most situations require the use of parallel-opposed beams. If the tumour is clinically node negative, the firststation nodes (levels II to IV) are irradiated because of the high risk of microscopic involvement. If there is no evidence of nodal disease upon clinical examination, then involvement of nodal levels I and V is rare (approximately 1%). If malignant lymphadenopathy is apparent upon clinical examination, then the risk of level I nodes being involved is approximately 12% (level V nodes, 10%).
(a)
Tonsil Tonsillar fossa tumours present with more advanced stages than those of the tonsillar pillar or soft palate. The overall risk of lymph node involvement is high: 60 to 75% for fossa tumours and 45% for pillar tumours. The risk of contralateral nodes is generally low: 10% for fossa, 5% for pillar. Small (1 to 2 cm) tumours of the anterior pillar or tonsil with limited spread to the adjacent soft palate or retromolar trigone have a small risk of spread to ipsilateral lymph nodes and almost no risk of contralateral nodal metastases. Tumours in the posterior tonsillar pillar are more likely to metastasise to the spinal accessory and upper posterior triangle nodes. The target volume for node-negative T1 or T2 welllateralised primary tumours is the primary tumour plus 2 cm margin and first echelon ipsilateral nodes. A wedged-pair technique allows relative sparing of the contralateral parotid gland. The anterior pillar is approximately 4 cm deep to the external surface of the face in most patients. If there is node-positive lateralised T1 or T2 disease, one would usually include all ipsilateral neck nodes in the target volume by adding an ipsilateral anterior neck
(b) Figure 8.1. Radiation fields for postoperative radiotherapy for a stage T3 N2b carcinoma of the tonsil treated with parallel-opposed lateral fields. (a) Phase 1: note how the fields overlap the spinal cord. (b) Phase 2: note how the posterior borders of the fields are anterior to the spinal cord. An electron field was matched posteriorly on each side to ensure coverage of the upper posterior neck nodes. The lower cervical nodes were treated with an anterior photon field (not shown).
107
Laura Moss and Chris Gaffney
are used. Figure 8.1 shows the radiation fields for a postoperative radiotherapy treatment for a T3 N2b carcinoma of the tonsil. Opposed lateral fields were used.
Soft palate For T1 or T2 cancers, treat with surgery or RT. Local control rates are 80 to 90%. RT often results in a better functional result. (Surgery may result in problems with nasal regurgitation.) For T3 or T4 cancers, treat with surgery and RT. With RT alone the local control rate is approximately 30 to 50%.
Posterior pharyngeal wall First-echelon nodes are the retropharyngeal nodes and nodes at levels II and III. The risk of LN metastases is 25% for T1 and greater than 75% for T4.
Base of tongue The lymphatics course downward towards the hyoid bone where they pierce the pharyngeal wall and drain to level II. The jugulo-digastric node (the largest of the level II nodes) is frequently the first node to be involved, followed by levels III and IV. NB disruption of normal lymphatic channels by the presence of a tumour or surgery may result in aberrant patterns of spread to levels I and V or to contralateral nodes. For T1 cancers, equivalent local control and survival rates are obtained for surgery or RT. The choice of treatment modality is usually based on likely morbidity and RT is often favoured. For large T2, T3 and T4 tumours the treatment options are as follow: r EBRT and/or brachytherapy. r Glossolaryngectomy and postoperative RT if there is a midline tumour (high functional morbidity). r Brachytherapy to the primary and/or EBRT and surgery to the neck. r Chemoradiotherapy Even if there is a clinically node-negative tumour, consider elective neck RT (NB propensity for bilateral spread). For T1 or T2 cancers, treat levels I to III. For T3 or T4 cancers, treat levels I to IV.
Table 8.7. Five-year relative survival of oropharyngeal cancer Stage
Survival – all subsites combined (%)
1
57
2
54
3
43
4
30
Adapted from AJCC (2002).
Carcinoma of the hypopharynx Anatomy of the hypopharynx The hypopharynx extends postero-laterally in relation to the larynx from the aryepiglottic fold at the level of the hyoid to the lower level of the cricoid cartilage. The subsites are: r The posterior pharyngeal wall, which is bounded superiorly by the floor of the vallecula and inferiorly by the inferior border of the cricoid cartilage. r The pyriform fossae, which extend from the pharyngoepiglottic fold to the upper end of the oesophagus at the lower border of the cricoid cartilage. The medial extent includes the aryepiglottic folds and the arytenoid and cricoid cartilages. The lateral border is the inner surface of the thyroid cartilage. r The postcricoid, which extends from the level of the arytenoid cartilages and connecting folds to the inferior border of cricoid cartilage.
Clinical presentation Carcinoma of the hypopharynx usually presents late. Symptoms include: r Neck mass. r Pain with swallowing. r Sensation of something in the throat. r Excess salivation. r Voice changes (muffled and husky rather than hoarse). r Frequent coughing bouts due to inefficient closure of the glottis during swallowing. r Referred otalgia.
Prognosis
Investigation and staging
Table 8.7 shows the 5-year survival of oropharygeal cancer.
With indirect laryngoscopy, it may be possible to visualise tumours arising in the pyriform fossae and
108
Head and neck
posterior pharyngeal wall but it is difficult to see the postcricoid area. However, it may be possible to see displacement of the larynx anteriorly, saliva pooling or vocal cord oedema. Neck palpation should be used for cervical nodes and to assess laryngeal crepitus (absence of crepitus signifies laryngeal involvement). Swelling as a result of direct extension of the primary disease into neck tissue may be palpable and it will move with swallowing, unlike lymphadenopathy. Lateral soft tissue plain film may demonstrate widening of the prevertebral space, anterior displacement of the larynx or thyroid cartilage destruction. Microlaryngoscopy/oesophagoscopy/bronchoscopy and biopsy are essential to assess the local extent, especially inferiorly (invasion of the trachea, extension to the oesophagus, etc.). CT is better at detecting the presence of thyroid cartilage involvement, whereas MRI shows soft tissue detail. Cancer of the hypopharynx has the highest incidence of distant metastases relative to other head and neck sites (lung 80%, mediastinal LN 35%, liver 30%, and bone 30%). If there is palpable neck disease, extranodal spread, or more than three lymph nodes, there is a higher risk of distant metastases. Therefore, consider lung, liver or bone imaging. Two-thirds of patients are lymph node positive at presentation (especially levels II to IV; levels I and V are rarely involved but level VI may be involved if there is apical pyriform fossa disease or a postcricoid primary). Occult lymph node metastases are found in 40% of patients after neck dissection for cN0 disease.
Staging classification The TNM classification for carcinomas of the hypopharynx is shown in Table 8.8.
Treatment overview of hypopharyngeal carcinoma There are few comparative data on different treatment modalities and it is often difficult to interpret outcome as the reports use different outcome measures (survival, quality of life, functional outcome).
Stage T1 Stage T1 is uncommon. Treatment options include:
Table 8.8. Tumour and nodal TNM classification for carcinomas of the hypopharynx Stage
Description
T1
≤ 2 cm and limited to one subsite
T2
> 2 to 4 cm or more than one subsite
T3
> 4 cm or with hemilarynx fixation
T4a
Thyroid/cricoid cartilage, hyoid bone, thyroid gland, oesophagus, central compartment soft tissue
T4b
Prevertebral fascia, carotid artery, mediastinal structures
N1
Ipsilateral single ≤ 3 cm
N2a
Ipsilateral single > 3 to 6 cm
N2b
Ipsilateral multiple ≤ 6 cm
N2c
Bilateral, contralateral ≤ 6 cm
N3
> 6 cm
Adapted from UICC (2002).
r Conservation surgery (e.g. endoscopic pharyngectomy if a small lesion is easily accessible through the operating laryngoscope). r Partial pharyngolaryngectomy, which is indicated for early stage (T1 or T2) tumours in patients who can tolerate some degree of chronic aspiration. Half the larynx and half the hypopharynx are resected and the hyoid bone, thyroid ala, arytenoid, epiglottis, aryepiglottic fold, arytenoid eminence and false cord are removed on the affected side. It is difficult to predict postoperative speech and swallowing function. r Radical radiotherapy. There is not enough research evidence to make valid comparisons among these options.
Stages T2 to T4 Optimal local control probably results from surgery and postoperative radiotherapy. Surgery for T3 or T4 disease typically requires laryngectomy as part of the procedure. Relative contraindications to surgery include invasion of the prevertebral fascia and carotid involvement.
N0 disease If the patient is being treated surgically then selective lymph node dissection is performed to include levels II to IV. If there is involvement of the postcricoid region or the apex of the pyriform fossa then level VI nodes are 109
Laura Moss and Chris Gaffney
also removed. If there is oropharyngeal extension then level I nodes are also removed. If there is a central tumour then a bilateral neck dissection is recommended. Primary radical radiotherapy with concurrent chemotherapy where possible is indicated for early lesions and in medically inoperable cases. If the treatment is for early disease then the first-station nodes are included.
Table 8.9. Five-year relative survival for cancers of the hypopharynx Stage
Survival – all subsites combined (%)
1
41
2
36
3
36
4
20
Adapted from AJCC (2002).
Surgical treatments Endoscopic resection of the early posterior wall and pyriform fossa tumours can result in low morbidity and good functional results. Partial pharyngectomy with or without partial laryngectomy may be suitable in selected cases. More advanced disease requires pharyngolaryngectomy with reconstructon of the resulting defect with a free jejunal transfer or gastric pull-up.
Radical radiotherapy for hypopharyngeal carcinoma
cervical oesophagus and mediastinum) and to avoid the shoulders. In practice this means that the beams need to be angled inferiorly, which may be achieved by a floor twist. It may be necessary to add an anterior mediastinal field. With 3D conformal planning there is greater ability to cover target volume adequately because non-coplanar planning is possible. Significant contour changes are more easily appreciated and more complex beam shielding is possible. The DVH and beam’s-eye-view facilities also help planning and verification.
Position and immobilisation
r An immobilisation/beam direction shell is required. r The patient lies supine. r The cervical spine is normally straight, unless there is a pyriform fossa primary with a large neck mass, in which case it may be preferable to have the chin extended in order to move the oral cavity and mandible out of the field.
Potential problems
r Bulky nodes may overlie the spinal cord. r There may be involvement of mediastinal nodes, nodes adjacent to the skull base and nodes in the submastoid region. In these circumstances, a large volume needs to be treated and there will be a significant contour change in the superior–inferior direction as well as in the anterior–posterior direction. r The shoulders may obstruct beam entry. r It may not be possible to treat with a radical dose if there is large-volume bulky disease.
Target volume Add a 2 cm margin to the GTV in the superior–inferior direction to allow for subclinical disease. With orthogonal film planning, a coronal technique is often needed to treat the inferior aspect of PTV (the 110
Prognosis for cancer of the hypopharynx The 5-year survival for cancers of the hypophaynx is shown in Table 8.9.
Carcinoma of the nasopharynx Types of tumour affecting the nasopharynx The WHO classification divides the commonest tumours into three types (Shanmugaratnam and Sobin, 1978): r Type 1, keratinising squamous cell carcinoma. r Type 2, non-keratinising carcinoma. r Type 3, undifferentiated, associated with a lymphoid infiltrate, hence the term lymphoepithelioma. Other tumour types include lymphoma, plasmacytoma, and adenocarcinoma.
Anatomy of the nasopharynx The postero-superior wall extends from the level of the junction of the hard and soft palate to the base of the ¨ skull. The lateral wall includes the fossa of Rosenmuller. The inferior wall consists of the superior surface of the soft palate. Most tumours occur in the lateral wall or roof.
Head and neck
Incidence and epidemiology Carcinoma of the nasopharynx is relatively rare in the Western world; the UK incidence is 0.4 per 100 000. The incidence in southern China is approximately 50 times higher.
Risk factors and aetiology There are multiple risk factors, including diet, viral agents and genetic predisposition. In endemic areas, recognised risk factors are the high intake of salt-cured fish and meat with the associated release of volatile nitrosamines during the cooking process and EBV.
Table 8.10. Tumour and nodal TNM classification for nasopharyngeal carcinomas Stage
Description
T1
Limited to nasopharynx
T2a
Soft tissue extension to oropharynx/nasal cavity
T2b
Tumour with parapharyngeal extension
T3
Bony structures, paranasal sinuses
T4
Intracranial, cranial nerves, infratemporal fossa,
without parapharyngeal extension
hypopharynx, orbit, masticator space N1
Unilateral node(s) ≤ 6 cm, above supraclavicular
N2
Bilateral node(s) ≤ 6 cm, above supraclavicular
N3a
> 6 cm
N3b
In supraclavicular fossa
fossa fossa
Clinical presentation Symptoms are rarely related directly to the nasopharynx itself and are more likely to be due to local or lymph node spread of the tumour: r Nasal blockage. r Conductive deafness, tinnitus. r Cranial nerve palsies. r Cervical lymphadenopathy (often in the posterior triangle). r 80% of patients present with palpable cervical lymph nodes: 50% are bilateral.
Diagnosis and staging r The nose should be examined by anterior rhinoscopy and the postnasal space should be examined by posterior rhinoscopy or via nasendoscopy. r The tympanic membrane should be assessed for middle ear aeration which, if abnormal, could indicate eustachian tube dysfunction or obstruction. r The hearing should be examined for conductive deafness. r The cranial nerves should be assessed.
Investigations Investigation involves direct endoscopy with biopsy under anaesthetic. If the patient has presented only with cervical lymphadenopathy and there is no obvious nasopharyngeal lesion, then blind biopsies should be taken from the nasopharynx. A US of neck nodes with FNAC should be performed. CT or MRI should be used to assess the extent of primary disease. MRI is more accurate for evaluating
Adapted from UICC (2002).
the primary. The presence of deep tumour infiltration appears to be of more prognostic significance than the volume of the tumour. CT is useful for assessing the skull base for cortical bone erosion whereas MRI can often demonstrate involvement of the marrow space. Imaging of lung, mediastinum, bone and liver should be considered to assess sites of distant metastases. The first-echelon nodes are the parapharyngeal and retropharyngeal nodes. Bilateral nodes do not appear to influence prognosis. Radical radiotherapy treatment results in significant mucosal toxicity and enteral feeding is often required; therefore, consider PEG placement before starting treatment.
Staging classification Table 8.10 shows the TNM classification for carcinoma of the nasopharynx.
Treatment overview for nasopharyngeal carcinoma Surgery Apart from obtaining tissue for histological verification, surgery has no place in the initial management of nasopharyngeal carcinoma. Surgery may be 111
Laura Moss and Chris Gaffney
considered in patients who have partial responses in cervical nodes, nodal relapse or local recurrence at the primary site. Relapse in the neck nodes without a recurrence at the primary site is rare.
r The field margins are individually tailored to take into
Radiotherapy is the primary treatment of choice, usually with chemotherapy, for both the primary site and neck nodes. The technique is discussed later.
account the primary location, the size of the tumour, and any nodal involvement. r The volume needs to encompass the nasopharynx, the floor of sphenoid sinus, the clivus, the pterygoid fossa, the parapharyngeal space, the retropharyngeal nodes, the cervical nodes and the SCF. The extent of oropharyngeal and nasal cavity coverage depends on the individual case.
Chemotherapy
Problems in radiotherapy planning
Radiotherapy
Chemotherapy is commonly used, but there is conflicting evidence and there are conflicting opinions and techniques. The concerns relate to increased morbidity (especially mucositis) and some positive trials that have shown a benefit to chemotherapy have had poor outcomes in the standard RT arm. Patients need to be of good performance status and have good renal function. Commonly used chemotherapy strategies include the following: r Two cycles of neoadjuvant cisplatin and 5-fluorouracil (5-FU) prior to radical EBRT. EBRT may then be given with concurrent cisplatin (100 mg/m2 , days 1, 22 and 43). This often achieves a good partial/complete clinical response after two neoadjuvant cycles. r Concurrent chemoradiotherapy only (cisplatin 100 mg/m2 , days 1, 22 and 43). r Concurrent cisplatin chemoradiotherapy followed by 3 cycles of adjuvant cisplatin and 5-FU at 28-day intervals (Al-Sarraf et al., 1998). Prechemotherapy investigations are a baseline assessment of GFR; audiogram; renal, liver and bone profiles; and FBC.
Radical radiotherapy for carcinoma of the nasopharynx In the UK, nasopharyngeal carcinoma is usually treated with EBRT, but intracavitary boosts can be used in specialist centres, such as an additional 10 Gy boost prescribed to a depth of 1 cm. It is important that the catheters are positioned up against the tumour in the vault of the nasopharynx because there is a tendency for the catheters to fall downward.
r It is impossible to treat an entire volume in continuity to a radical dose in a single phase.
r There is no ideal technique. The fields covering the
r r r
r r
primary and neck nodes overlap and there are difficulties matching photon and electron fields. This may result in cold spots over areas of potential disease and hot spots over critical structures. There is a large target volume with extensive mucosal coverage. Advanced disease may extend into cranial fossae; orbits and normal tissue tolerances may be compromised. It is necessary to decide whether to treat the neck nodes with anterior and posterior fields or to treat with an anterior field only. This decision will depend on the levels of involved nodes and their bulk. A decision needs to be made whether to use orthogonal films or CT for planning. To cover potential microscopic disease, the volume must include the skull base, the floor of the middle cranial fossa, the posterior half of the nasal cavity and orbit, the sphenoidal and posterior ethmoidal sinuses, and the parapharyngeal space.
Positioning
r The patient lies supine. r The neck is extended but the cervical spine is kept straight.
r Cervical nodes are marked with wire if orthogonal planning is used.
r It may be useful to mark the lateral orbital margin with wire for orthogonal planning.
r The target volume is the primary disease with a 2 to 3 cm margin, the skull base and the retropharyngeal and cervical lymph nodes bilaterally.
General issues in radiotherapy planning
r The primary site and bilateral neck need to be treated because there are extensive lymphatics and these cross the midline.
112
Phase 1
r Large opposing lateral photon fields are used to cover the entire target volume.
Head and neck
r Shield the brain stem, optic chiasm and the anterior third of the orbit. (The optic nerves exit the orbits just beneath the anterior clinoids and the brain stem lies immediately posterior to the clivus). Request TLDs. r The lower cervical lymph nodes and SCF nodes are usually treated with a direct anterior field. r By placing the junction between the upper lateral and the lower anterior fields at the level of the hyoid, it may be possible to spare much of the larynx and pharynx and, hence, reduce the acute mucosal reactions at these sites by having a central shielding strip in the anterior beam. r Take care not to have the junction of the lateral and anterior photon fields over palpable/radiologically visible disease. (Some argue that it is acceptable to have a junction through enlarged nodes because the dose at the junction of two diverging photon portals is more likely to be hot than cold.) For subsequent phases (usually two- or three-phase techniques), the volume can be divided into two as follow: 1. The primary tumour, including nasopharynx, parapharyngeal space, skull base, posterior third of the orbit, and any extension into the nasal cavity or oropharynx. NB. It may be necessary to add an additional anterior photon field to give adequate dose coverage to any nasal cavity disease extension. 2. The neck nodes. When deciding on the optimal technique to treat cervical node disease, the degree of nodal involvement and the position of the nodes must be taken into account. Options include: i. Opposed lateral photon beams with electrons to boost the posterior nodal groups so as to remain within spinal cord tolerance. ii. Anterior and posterior opposed photon fields. iii. Anterior photon field only. TLDs should be requested to estimate dose to lens. Dose is 66 Gy in 33 fractions over 6.5 weeks. The separate contributions from phase 1 and phase 2 differ among centres and the local protocol should be followed.
Palliative treatments For metastatic disease cisplatin and 5-FU may provide significant symptom palliation. Clinical responses tend to be rapid but there is potential for rapid relapse when treatment stops.
Table 8.11. Five-year relative survival for nasopharynx cancer Stage
Survival (%)
1
63
2
52
3
56
4
39
Adapted from AJCC (2002).
Prognosis Table 8.11 shows the 5-year relative survival for cancers of the nasopharynx.
Carcinoma of the maxillary antrum (and other paranasal sinuses) Anatomy of the maxillary sinus The maxillary antrum is related superiorly to the floor of the orbit. The medial wall separates it from the nasal cavity, and the anterior wall lies deep to the cheek. The floor of the antrum is the alveolar process of the maxilla and the hard palate. The pterygoid plates and pterygopalatine fossa lie posteriorly. ¨ Ohngrens line connects the medial canthus of the eye to the angle of the mandible. This line is used to divide the maxillary antrum into the antero-inferior portion (which is associated with earlier presentation and good prognosis) and the supero-posterior portion (which is associated with early involvement of adjacent critical structures and a worse prognosis).
Pathology Squamous carcinoma is the most common tumour type and is associated with nickel exposure in some cases. These tumours arise most commonly in the maxillary sinus, followed by the nasal cavity, ethmoid sinus and sphenoid sinus. Adenocarcinomas are associated with hardwood dust exposure. Adenoid cystic carcinoma can demonstrate widespread perineural infiltration.
Clinical presentation Carcinoma of the maxillary antrum often presents at a locally advanced stage. Symptoms are: 113
Laura Moss and Chris Gaffney
r Epistaxis or nasal blockage. r Cheek swelling. r Poorly fitting dentures or loose dentition. r Impaired vision (due to abnormal ocular movement or proptosis). r Facial or cheek pain. r Trismus (due to pterygoid fossa involvement from posterior extension of the disease). Approximately 15% of patients are LN positive on presentation.
Table 8.12. Tumour and nodal TNM summary for carcinomas of the maxillary sinus Stage
Description
T1
Confined to mucosa
T2
Bone erosion/destruction, hard palate, middle
T3
Posterior bony wall maxillary sinus,
nasal meatus subcutaneous tissues, floor/medial wall of orbit, pterygoid fossa, ethmoid sinus
Investigation and staging Clinical examination involves examining the gingivobuccal gutter and palate, looking for cheek swelling and infraorbital nerve involvement, examining the nasal cavity and examining the range of jaw movement. CT scanning is good for assessing bone detail and determining whether the cancer is confined within the boundaries of the sinus or is eroding through bone. Inflamed mucosa and retained secretions have a similar appearance on CT and can be mistaken for tumours. MRI is better than CT at distinguishing tumours from retained secretions and surrounding soft tissues. It is also helpful in assessing the presence of skull base invasion, soft tissue intracranial extension or perineural spread. Biopsy should be performed via the least-invasive method (e.g. endoscopic biopsy via the nose), which will allow excellent visualisation and minimum alteration to the tumour and surrounding anatomy; biopsy is associated with low morbidity.
Staging classification Table 8.12 shows the TNM staging for carcinoma of the maxillary sinus.
Treatment overview Because this disease typically presents at a locally advanced stage, combined modality treatment is usually preferred if the patient is of good performance status with limited co-morbidity and can manage with the functional alterations that result from surgical resection.
Surgical treatments There are a variety of different maxillectomy operations (medial, subtotal, total or extended). Maxillectomy may 114
T4a
Anterior orbit, cheek skin, pterygoid plates, infratemporal fossa, cribriform plate, sphenoid/frontal sinus
T4b
Orbital apex, dura, brain, middle cranial fossa, cranial nerves other than V2, nasopharynx, clivus
N1
Ipsilateral single ≤ 3 cm
N2a
Ipsilateral single > 3 to 6 cm
N2b
Ipsilateral multiple ≤ 6 cm
N2c
Bilateral, contralateral ≤ 6 cm
N3
> 6 cm
Adapted from UICC (2002).
need to be combined with orbital exenteration or craniofacial resection depending on the extent of disease. The resulting defect needs to be filled with an obturator or a free flap.
Radical radiotherapy for carcinoma of the maxillary sinus Radiotherapy technique An immobilisation shell should be used. The complex shape of the target volume and surrounding critical structures makes 3D conformal RT desirable. The use of a mouth bite to depress the tongue should be considered. The target volume includes the ethmoid sinuses, nasal cavity, hard palate, gingivobuccal sinus, pterygoid fossa and cheek, as well as the maxillary antrum. Anterior and lateral wedged fields are often used. A contralateral lateral field is used to top up the dose to the postero-medial aspect of the volume. It is important to avoid the optic chiasm and hypothalamus and beam exit through the contralateral eye. Figure 8.2 shows a radiotherapy plan for a locally advanced carcinoma of
Head and neck
Table 8.13. Five-year relative survival for cancers of the maxillary sinus Stage
Survival (%)
1
60
2
50
3
46
4
31
superior boundary is the ethmoid sinus and the inferior boundary is the hard palate. The lateral wall is also the medial wall of the maxillary sinus and includes the turbinates and nasolacrimal duct. The sphenoid sinus lies posterior and superior and the nasopharynx lies directly posterior.
Pathology Squamous carcinoma is most common. Others include adenocarcinoma, mucosal melanoma and olfactory neuroblastoma (a neuroendocrine tumour that arises from the superior aspect of the nasal cavity).
Adapted from AJCC (2002).
Isodose % 95 70 50 20
Spread Lymphatic drainage is to the submandibular and upper deep cervical nodes.
Clinical presentation
Figure 8.2. A CT-planned radical radiotherapy treatment for a locally advanced carcinoma of the maxillary antrum showing the anterior and two lateral beams. This image shows the isodoses at one level only, and it is important to study the whole plan to make sure the dose to
Nasal cavity cancers often do not present until they reach a significant size. Symptoms are: r Nasal obstruction or discharge. r Epistaxis. r Local pain. r Sinusitis. r Gingival/palatal swelling, loose teeth and poorly fitting dentures if the disease extends inferiorly. r Proptosis and diplopia if there is superior extension into the orbit. r Trismus from pterygoid muscle involvement may result from lateral extension.
critical structures, such as the eyes, optic chiasm and hypothalamus, is as low as possible.
Investigation and staging the maxillary antrum in which 3D conformal planning has been used.
Cross-sectional imaging is required to determine the posterior extent of disease.
Staging classification Prognosis The 5-year survival for maxillary sinus cancers is shown in Table 8.13.
Carcinoma of the nose and nasal cavity Anatomy of the nasal cavity The nasal cavity is the most superior part of the upper airway. It is divided in the midline by the septum. The
The TNM summary is shown in Table 8.14.
Surgical treatments Lateral rhinotomy can be used to treat cancers involving the nasal septum and lateral nasal wall that extend up to, but not including, the anterior skull base. Craniofacial resection can be used to excise olfactory neuroblastoma and the procedure can be extended to 115
Laura Moss and Chris Gaffney
Table 8.14. Tumour and nodal TNM classification for nasal cavity and ethmoid sinus Stage
Description
T1
1 subsite
T2
2 subsites or adjacent nasoethmoidal site
T3
Medial wall/floor of orbit, maxillary sinus, palate,
T4a
Anterior orbit, skin of nose/cheek, anterior
cribriform plate cranial fossa (minimal), pterygoid plates, sphenoid/frontal sinuses T4b
Orbital apex, dura, brain, middle cranial fossa, cranial nerves other than V2, nasopharynx,
external nose to enclose the irregular shape. It is important not to exceed the tolerance doses of the optic nerve and chiasm, pituitary and brain. Consideration should be given to placing bolus in the nasal cavity to reduce dose inhomogeneity. In the postoperative situation following total rhinectomy, the following problems may occur: r There may be dose inhomogeneity in the air cavity with electron treatments. r Megavoltage photons may cause problems because of the dose at depth and the proximity of brain/brain stem. Consideration should be given to using tissue equivalent bolus in the air cavity.
clivus N1
Ipsilateral single ≤ 3 cm
N2a
Ipsilateral single > 3 to 6 cm
N2b
Ipsilateral multiple ≤ 6 cm
N2c
Bilateral, contralateral ≤ 6 cm
N3
> 6 cm
Adapted from UICC (2002).
include a maxillectomy or orbital exenteration depending on the extent of the disease. Total rhinectomy is required for extensive tumours of the cartilaginous nasal skeleton and dorsum of the nose. It is often difficult to reconstruct the nose with a flap in a way that leads to a satisfactory cosmetic result, and a prosthesis usually offers a better rehabilitation option.
Radical radiotherapy for nose and nasal cavity carcinomas Radiotherapy technique The primary site alone is usually irradiated. The lymphatic drainage routes are to the submandibular and upper deep cervical nodes. (The area is not routinely irradiated if the disease is confined to the external nose because there is only an approximately 15% risk of involvement.) If the cancer is positioned anteriorly, a single anterior field with shielding to the eyes (inverted ‘T’ shape) will be sufficient. The field extends to the nasal bridge, midway to the lip and 1 cm lateral to the ala nasi. If there is surface disease then bolus may be required. If the cancer is positioned posteriorly in the nose, an anterior-oblique wedged pair or anterior and opposed laterals will be required. Wax bolus is placed over the 116
Salivary gland Introduction The major salivary glands are the paired parotids, submandibular, and sublingual glands. The minor salivary glands are small and are found beneath the mucosa of the upper aerodigestive tract. The highest concentration is found on the hard palate, followed by the nasal cavity and oral cavity. Tumours are most common in the parotid but, because only 20% of these are malignant, the incidence of malignancy is highest in the submandibular (>50% malignant), sublingual and minor glands. Salivary gland tumours account for 7% of epithelial head and neck cancers. The incidence is approximately 1 in 100 000.
Anatomy of the parotid gland r The superficial lobe lies between the masseter (anterior) and mastoid process (posterior).
r The deep lobe is posterior to the mandibular ramus extending medially, anterior to the styloid process, and adjacent to the parapharyngeal space and the lateral pterygoid muscle. r The plane separating the two lobes is formed by the facial nerve. r Lymphatic drainage is to the preauricular and intraparotid nodes.
Pathology There is a wide range of histological types and clinical behaviours. Malignant tumour types include the following:
Head and neck
r Adenoid cystic carcinoma is the most common type; it r r r r r r r r r
is more common in minor glands than major salivary glands. Mucoepidermoid (the most common major salivary gland tumour: 4 to 9%). Acinic cell carcinoma (accounts for 3% of parotid tumours). Adenocarcinoma. Lymphoma. Salivary duct carcinoma. Sebaceous carcinoma. Carcinoma arising in a pleomorphic adenoma. Squamous carcinoma. Small-cell and undifferentiated carcinoma.
Table 8.15. Tumour and nodal TNM classification for carcinomas of the salivary glands Stage
Description
T1
≤ 2 cm, without extraparenchymal extension
T2
> 2 to 4 cm, without extraparenchymal extension
T3
> 4 cm and/or extraparenchymal extension
T4a
Skin, mandible, ear canal, facial nerve
T4b
Skull, pterygoid plates, carotid artery
N1
Ipsilateral single ≤ 3 cm
N2
Ipsilateral single > 3 to 6 cm Ipsilateral multiple ≤ 6 cm Bilateral, contralateral ≤ 6 cm
N3
Clinical presentation Clinically it may be difficult to distinguish between benign and malignant tumours. However, pain, rapid growth, facial nerve involvement, an immobile tumour, trismus and nodal involvement all suggest malignancy.
> 6 cm
Adapted from UICC (2002).
be better treated with radical radiotherapy. An example would be a minor salivary gland tumour in a sinus abutting the skull base.
Investigation and staging Fine needle aspiration cytology is reported to be capable of distinguishing between benign and malignant tumours in 90% of cases. It is important to avoid open biopsy of major salivary gland tumours because of the risk of tumour spillage and seeding.
Imaging Use US and image-guided FNA. In a CT scan, dental amalgam may produce artefacts and, therefore, MRI may be a better choice. It is important to determine whether the tumour is arising in the superficial or deep lobe, its relationship with the facial nerve and whether there is any extraglandular extension.
Staging classification The TNM classification summary is shown in Table 8.15.
Treatment overview The extent of the tumour is the single most important factor affecting the choice of treatment modality. The tumour location is also important; for example, a borderline resectable tumour in an inaccessible area may
Surgery Superficial parotidectomy can be performed, provided there is no involvement of the deep lobe or the facial nerve. Total parotidectomy is indicated if the deep lobe is involved. The facial nerve can be preserved if there is no direct involvement. Total parotidectomy is also indicated for adenoid cystic tumours because of the propensity for perineural spread. Neck dissection is indicated if there is clinical or radiological evidence of node involvement, if the tumour is locally advanced and if the tumour type is high grade or adenocarcinoma, squamous or undifferentiated.
Postoperative radiotherapy of the parotid gland Indications for postoperative radiotherapy Indications for RT include the following: r Inoperable tumours. r Positive or close resection margins. r High-grade type. r Advanced stage: skin or bone involvement; extensive perineural spread. r Residual disease. r Positive nodes. 117
Laura Moss and Chris Gaffney
r Recurrence. r Adenoid cystic type. r Close proximity to facial nerve where the nerve has been preserved.
r For pleomorphic adenoma, consideration should be given to postoperative radiotherapy if there is incomplete excision or spillage at operation or after reexcision of locally recurrent tumours. It is important to weigh the risk of RT for this benign disease in young patients against the risk of recurrence and risk of late morbidity including second malignancy.
Table 8.16. Five-year relative survival for major salivary glands combined Stage
Survival (%)
1
86
2
66
3
53
4
32
Adapted from AJCC (2002).
Radiotherapy technique The entire parotid bed should be included, extending above the zygomatic arch for malignant tumours (if adenoid cystic, extend to include the facial nerve route and skull base) to below the angle of the mandible. Anteriorly, the masseter should be included, and posteriorly, the volume should extend to the mastoid process. The parapharyngeal space should be included if irradiating a malignant tumour. 3D conformal RT has advantages over 2D planning because complex volume shapes are possible and it helps avoid surrounding critical structures. The field arrangement is of anterior and posterior oblique wedged fields to treat the parotid or parotid bed in the postoperative setting. The beam angles are selected to avoid the contralateral parotid and spinal cord. An anterior ipsilateral lower neck field is matched onto the oblique fields if there are involved nodes or a high risk of occult lymph nodes. The dose is administered as a single phase. Figure 8.3 shows a radiotherapy plan for a postoperative treatment following parotidectomy for a high-grade mucoepidermoid carcinoma.
Prognosis The 5-year survival for malignant tumours is shown in Table 8.16.
Isodose % 100 95 70 50 20
Figure 8.3. A CT-planned postoperative radiotherapy treatment following excision of a high-grade mucoepidermoid cancer of the parotid. Bolus was used because in this case there was concern about tumour seeding in the skin. The volume marked on the contralateral side is the other parotid gland.
Spread Spread occurs medially to the inner ear and facial nerve, anteriorly into the parotid, superiorly to the middle cranial fossa, posteriorly to the mastoid and laterally to the ear canal.
Clinical examination
Carcinoma of the middle ear
r Examine the ear, parotid, mastoid area and postnasal
Anatomy of the ear
r Examine cranial nerve function. r Palpate the cervical and facial nodes.
The external ear canal is 2.5 cm long. The outer portion is cartilaginous and the inner portion is bony. The middle ear lies between the tympanic membrane and the inner ear. It contains the ossicles and semicircular canals and it communicates posteriorly with the mastoid air space. 118
space.
Investigation and staging r Perform examination and biopsy under general anaesthesia.
Head and neck
r CT is the investigation of choice for assessing the temporal bone. r MRI is more useful than CT for defining the soft tissue extent.
Staging classification No staging system is accepted by UICC or AJCC.
Treatment overview Surgery and postoperative RT give the best chance of local control.
Surgical treatments The required procedure depends on the local extent. It may entail lateral temporal bone resection (removal of osseous and cartilaginous external auditory canal, tympanic membrane, malleus and incus) with mastoidectomy, parotidectomy and neck dissection.
Radical radiotherapy for carcinoma of the middle ear Primary radical radiotherapy is seldom used as the sole treatment modality. Postoperative radiotherapy is indicated for situations where there are close or positive resection margins and perineural invasion. These findings are common because of the limitations on the margins of surgical resection. As a result, postoperative radiotherapy is indicated in most cases. The volume includes the pre- and postauricular nodes and the mastoid process and is triangular in shape. It is an ideal anatomical site for 3D conformal therapy.
Prognosis The 5-year survival is approximately 30%.
Cervical lymphadenopathy from an unknown primary Introduction This condition occurs when squamous carcinoma is identified in cervical lymph nodes but there is no upper aerodigestive tract primary site evident. It accounts for 5% of head and neck cancers. In 40 to 50% of cases,
upper aerodigestive tract examination will identify the primary lesion.
Clinical presentation The patient presents with a neck mass.
Investigation and staging Panendoscopy and biopsies are undertaken from any suspicious areas plus the postnasal space, the tonsils (or bilateral tonsillectomy) and the tongue base. A tonsil or tongue base tumour is found in approximately a third of cases. CT and/or MRI should be used to look for potential primary sites. 18 FDG-PET imaging may also be used to detect a primary site. The optimal timing is debated: some advocate PET imaging before EUA and blind biopsies to avoid the delay between biopsy and imaging (which is recommended to reduce the chances of a falsepositive result caused by increased uptake at biopsy sites) and to target subsequent biopsies based on the imaging findings. Others recommend PET only if the EUA and biopsies fail to detect the primary site because of limited resource availability, and an EUA with targeted biopsies of potential mucosal primary sites may detect the primary lesion.
Staging classification There is no specific TNM staging system. In practice the neck nodes are staged in the same way as other head and neck cancer subsites (excluding the nasopharynx).
Treatment overview for cervical lymphadenopathy of unknown primary If FNAC of the presenting neck mass is inconclusive or negative for malignancy, then an excision biopsy will be required. Consider doing a frozen section at the time of the excision biopsy as well as panendoscopy of the upper aerodigestive tract to try and locate a primary lesion. If the cervical node FNAC is positive for squamous carcinoma, then a neck dissection is indicated. If the pathology from the neck dissection specimen reports N1 disease, then observation only is required, unless there are adverse prognostic factors present (e.g. extracapsular spread). If pathology reveals N2 or N3 disease, then proceed with postoperative RT. It is important to decide 119
Laura Moss and Chris Gaffney
whether to treat the ipsilateral neck only or to extend the treatment volume to potential occult mucosal primary sites. The evidence for elective mucosal irradiation is conflicting. It is advocated by some on the basis that there is a reduced incidence of a primary lesion developing if all the potential mucosal primary sites are irradiated along with the nodal drainage areas. However, only a small percentage of patients actually develop a primary tumour. This may occur many years after the RT and raises the possibility that a second primary may have occurred rather than a late appearance of the original primary tumour. Elective mucosal irradiation has not been shown to increase survival but does have potential for severe morbidity.
REFERENCES AJCC. (2002). AJCC Cancer Staging Manual, ed. F. Greene et al., 6th edn. New York: Springer. Al-Sarraf, M., LeBlanc, M., Giri, P. et al. (1998). Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: phase III randomized intergroup study 0099. J. Clin. Oncol., 16, 1310–17. Bernier, J., Domenge, C., Ozsahin, M. et al. (2004). Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. (EORTC Trial 22931). N. Engl. J. Med., 350, 1945–52. Bonner, J. A., Giralt, J., Harari, P. M. et al. (2004). Cetuximab prolongs survival in patients with locoregionally advanced squamous cell carcinoma of head and neck: a phase III study of high dose radiation therapy with or without cetuximab. J. Clin. Oncol., ASCO Annual Meeting Proceedings (Post-Meeting Edition). Vol. 22 (14 Suppl.), 5507. Bourhis, J., Syz, N., Overgaard, J. et al. (2002). Conventional vs modified fractionated radiotherapy. Meta-analysis of radiotherapy in head and neck sqamous carcinoma: a meta-analysis based on individual patient data. Int. J. Radiat. Oncol. Biol. Phys., 54 (Suppl.), 71–2. British Association of Otorhinolaryngologists – Head and Neck Surgeons. (2002). Effective Head and Neck Cancer
120
Management: Third Consensus Document. London: Royal College of Surgeons. Cooper, J., Pajak, T., Forastiere, A. et al. (2004). Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N. Engl. J. Med., 350, 1937–44. CRUK. (2004) CancerStats Monograph 2004: Cancer Incidence, Survival and Mortality in the UK and EU, ed. J. R. Toms. London: CRUK. Dische, S., Saunders, M., Barrett, A. et al. (1997). A randomised multicentre trial of CHART versus conventional radiotherapy in head and neck cancer. Radiother. Oncol., 44, 123–36. Horiot, J. C., Le Fur, R., N’Guyen, T. et al. (1992). Hyperfractionation versus conventional fractionation in oropharyngeal carcinoma: final analysis of a randomized trial of the EORTC cooperative group of radiotherapy. Radiother. Oncol., 25, 231–41. Intercollegiate Standing Committee on Nuclear Medicine. (2003). Positron Emission Tomography: A Strategy for Provision in the UK. London: Royal College of Physicians. Monnerat, C., Faivre, S., Temam, S. et al. (2002). End points for new agents in induction chemotherapy for locally advanced head and neck cancers. Ann. Oncol., 13, 995–1006. NICE. (2004). Improving Outcomes in Head and Neck Cancers. London: NICE. Overgaard, J., Hansen, H., Specht, L. et al. (2003). Five compared with six fractions per week of conventional radiotherapy of squamous-cell carcinoma of head and neck: DAHANCA 6 and 7 randomised controlled trial. Lancet, 362, 933–40. Pignon, J., Bourhis, J., Domenge, C. et al. (2000). Chemotherapy added to locoregional treatment for head and neck squamous-cell carcinoma: three meta-analyses of updated individual data. MACH-NC Collaborative Group MetaAnalysis of chemotherapy on head and neck cancer. Lancet, 355, 949–55. Royal College of Radiologists. (2001). The Role and Development of Brachytherapy Services in the UK. London: Royal College Radiologists. Shanmugaratnam, K. and Sobin, L. H. (1978). International Histological Classification of Tumours, No 19: Histological Typing of Upper Respiratory Tract Tumors. Geneva: World Health Organization, p. 32. UICC. (2002). TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss.
9
OESOPHAGUS Tom Crosby
Introduction
Anatomy
In the past few decades there has been a dramatic increase in the incidence of adenocarcinoma of the oesophagus, predominantly in the lower oesophagus and gastro-oesophageal junction. This trend has been noted across most patient populations worldwide but is most noticeable in the younger, white male population, where it would appear to be related to gastrooesophageal reflux disease (GORD) and to be less strongly associated with alcohol and smoking. Meanwhile, the incidence of squamous cancer worldwide has remained steady or fallen slightly, though there are large geographical variations. The majority of patients present with symptoms of locally advanced or metastatic disease, which limits survival from any treatment. Combined modality therapy is increasingly used in these cases. Chemoradiation is more effective than radiotherapy alone and there is some evidence that preoperative chemotherapy is superior to surgery alone. There is a continuing controversy about the exact role of surgery combined with chemoradiotherapy and which one should be used as the primary therapy. Clinical trials are currently trying to define the optimum radiotherapy and chemotherapy regimens.
The oesophagus is usually measured from the central incisors (usually from 15 to 40 cm) at endoscopy. It is often divided into sections, which are defined in Table 9.2. The sternal notch is at 18 cm and the carina is usually at 25 cm but this can vary significantly among patients. The gastro-oesophageal (GO) junction is usually at about 40 cm. Siewert describes a system to classify tumours involving the GO junction. Type I tumours are predominantly oesophageal and type III predominantly gastric. In type II tumours the disease equally straddles the junction.
Types of oesophageal tumour The types of oesophageal tumours are shown in Table 9.1. The oesophagus is a relatively common site for a second primary cancer. For instance, following successful treatment for head and neck cancer, 4% of patients per year develop a second primary, 30% of which are oesophageal, especially squamous cancers. This is likely to be related to a ‘field change’ effect on the mucosa from common aetiological agents, although it has been suggested that tumours may be seeded at the time of enteral feeding.
Incidence and epidemiology Tumour incidence in the UK is 11 per 100 000, with approximately 7000 new cases and approximately 6000 deaths per year in the UK. There is a large worldwide variation in incidence. Squamous cell carcinoma occurs more often in Iran, China and Transkei (South Africa), with rates as high as 100 per 100 000. In the past three decades, there has been a 3.5-fold increase in men and a two-fold increase in women in the incidence of adenocarcinoma of the lower oesophagus. During the same time period, there has also been an increase in occurrence of adenocarcinoma of the gastric cardia. The incidence of oesophageal cancer increases with age, although there are also increasing numbers of cases occurring in the young, particularly among male Caucasians.
Carcinoma of the oesophagus Risk factors and aetiology The increase in adenocarcinomas (about 10% per year) of the middle and lower oesophagus appears to be associated more with reflux of acid (gastric and possible bile) and less with alcohol intake and smoking. The risk increases with age and is more common in men, 121
Tom Crosby
Table 9.1. Types of oesophageal tumour Type
Examples
Benign
Leiomyoma Haemangioma Adenoma
Malignant primary
Adenocarcinoma (65%) Squamous (∼25%) Others (∼5%) Small cell Lymphoma BCC Melanoma Leiomyosarcoma Kaposi’s sarcoma Adenoid cystic Mucoepidermoid
in 3.5%, severe; 1.5% will have long-segment (> 3 cm) Barrett’s oesophagus. Over a period of 2 to 5 years, 0.35 and 0.12% of these patients will develop low- and high-grade dysplastic change, respectively. Therefore, in patients with Barrett’s oesophagus, there is only about a 1% lifetime risk of developing an adenocarcinoma. Significant problems exist with interobserver variation between pathologists when classifying dysplastic changes, and severe dysplasia is synonymous with carcinoma in situ. Other related conditions include: r Achalasia. r Tylosis palmaris. r Coeliac disease. r Plummer-Vinson syndrome. Apart from GORD and Barrett’s oesophagus, the aforementioned conditions are primarily associated with squamous cell carcinoma.
Gastro-intestinal stromal tumour Carcinoid Malignant secondary
Lung Breast Melanoma
except for those cancers occurring in the upper third of the oesophagus, which are associated with PlummerVinson syndrome. Risk factors are as follow.
Physical/chemical Physical and chemical causes of oesophageal cancer include: r GORD. r Alcohol. r Smoking. r Corrosives. r Reduced dietary vitamin C. r Malnutrition (e.g. zinc deficiency). Possible infective causes may be associated with r Helicobacter pylori. r Human papilloma virus. r Fungally infected cereals.
Pathology Adenocarcinomas occurring in the lower oesophagus usually arise on the background of Barrett’s oesophagus, a columnar epithelialisation of the native squamous mucosa. The sequence from metaplasia through degrees of dysplasia to invasive adenocarcinoma is associated with genetic changes such as loss of TP53 function, LOH of the Rb gene, overexpression of cyclins D1 and E, and inactivation of p16 and p27. Amplification of MYC and K- and H-RAS occur late in the transition to adenocarcinoma. A high proportion of adenocarcinomas have foci of gastric- or intestinal-type lining in the immediate vicinity of the tumour and elsewhere in the oesophagus. Pathological features are shown in Table 9.3.
Spread Spread of primary tumour Oesophageal tumours are usually locally advanced at presentation because there is no peri-oesophageal serosa to inhibit their growth. They spread circumferentially and longitudinally along submucosal and perineural pathways to form skip lesions of up to 5 to 6 cm from the primary tumour. Infiltration into mediastinal structures occurs, most frequently into the trachea, aorta, pleura, diaphragm and vertebrae.
Associated conditions An associated condition is Barrett’s oesophagus. Of people over 30 years old, 10% have symptomatic reflux disease. In 10% of those with the disease, it will be mild and 122
Lymphatic spread The first-station lymph nodes (N1) of the oesophagus are supraclavicular; upper, middle and lower
Oesophagus
Table 9.2. Parts of the oesophagus Correlation with Part of oesophagus
Anatomical description
Distance from incisors
vertebral bodies
Cervical oesophagus
Starts below cricopharyngeus
15–18 cm
C6 – T2/3
to thoracic inlet Thoracic oesophagus Upper
To tracheal bifurcation
18–24 cm
T3 ∼ T4/5
Middle
To halfway to GO junction
24–32 cm
T5 ∼ T8
Lower
To GO junction
32–40 cm
T8 ∼ T10
GO = gastro-oesophageal.
Table 9.3. Pathological features of carcinoma of the oesophagus Description Macroscopic features
Squamous cell Plaque-like lesion or elevation of
para-oesophageal; right and left paratracheal; aortopulmonary; subcarinal; diaphragmatic; paracardial; left gastric; common hepatic; splenic artery and coeliac. In addition to regional node spread, the rich submucosal plexus means that upper-third tumours can spread to coeliac nodes and lower-third to supraclavicular and deep cervical nodes in up to 30% of cases. This spread is classified as distant or M1b disease.
mucosa evolving into polypoid fungating lesion or necrotic
Haematogenous spread
ulceration
The most frequent sites of haematogenous metastatic spread are the liver and lungs. Other sites are becoming more common with the increase in adenocarcinomas, such as bone, brain and skin.
Adenocarcinoma Often nodular and multicentric, arising in Barrett’s mucosa. Similar growth to above progressing to circumferential
Clinical presentation
mass Microscopic features
Squamous cell Usually moderately to well differentiated with typical features of squamous cancer (i.e. small cells with/without keratinisation) Adenocarcinoma Usually moderately to poorly differentiated, divided into intestinal, diffuse or adenosquamous type. Intestinal type shows well-formed glands lined by
Typically, patients present late with progressive dysphagia, often solids to liquids over 3 to 6 months. The dysphagia grading system is as follows: r Grade 1, patient has difficulty with some foods such as bread and meat. r Grade 2, patient is able to eat a soft diet. r Grade 3, patient only manages a liquid diet. r Grade 4, complete dysphagia (including saliva). About 50% of patients experience pain (odynophagia) with or without signs or symptoms of aspiration or weight loss. Patients often point to the sternal notch if there is an upper-level obstruction and to the epigastrium if there is a lower-level obstruction.
malignant cells whereas the diffuse type is composed of mucin-producing neoplastic cells
Investigation and staging A full blood count, biochemical profile and check CEA and CA19–9 tumour markers should be performed. 123
Tom Crosby
Diagnostic endoscopy with or without a barium swallow. A barium swallow is sometimes done first but if it is abnormal, an endoscopy is still needed to identify the start of the tumour, measure its length (if the scope will pass) and take multiple biopsies. Endoscopic ultrasound (EUS) is good for detecting the primary and mediastinal lymph nodes; there is an 80 to 95% concordance with pathology. The disease length defined by EUS is frequently longer than the tumour length measured at endoscopy because both submucosal and nodal disease above and/or below the primary tumour may be identified. CT scanning will show the primary tumour as thickening but will often underestimate the tumour length and the presence of nodal disease. MRI is not used routinely but may be useful for determining infiltration into the aorta or trachea. Consider bronchoscopy for tumours above the carina or where there are signs suggestive of T4 disease. Laparoscopy should be considered for lower oesophageal lesions that extend below the diaphragm. PET scanning may have a role in the staging and treatment of oesophageal cancer. It may identify more widespread disease and so prevent unnecessary surgery in up to 20% of cases, but this needs confirmation in larger studies (Luketich et al., 1997). A PET scan can show nodal disease with up to 85% accuracy (Kobori et al., 1999). It may increase radiotherapy target volumes in up to 10% of cases but should not lead to reduced fields because of the significant false-negative rate. It may also have a future role in assessing response to chemoradiation.
Staging classification The TNM staging classification is shown in Table 9.4. The commonest stage at presentation is T3 N1 disease with or without distant metastases.
Treatment overview For patients with tumours that appear resectable and who are sufficiently fit, surgery remains the mainstay of therapy in the UK. Surgery alone may be used for patients with early disease (T1 or T2, N0) or who are not fit for neoadjuvant therapy. Two cycles of neoadjuvant cisplatin and 5-fluorouracil (5-FU) should be considered standard for fit patients with no metastases (Medical Research Council Oesophageal Cancer Working Group, 2002). 124
Table 9.4. The TNM staging classification for carcinoma of the oesophagus Stage
Description
T1
Confined to mucosa
T2
Tumour spread through submucosa into muscle
T3
Tumour extends through muscle layer
T4
Tumour infiltrates into surrounding anatomical
layer
structures N1
Involvement of regional nodes
M1a
Involvement of coeliac nodes from lower-thoracic tumours or supraclavicular/deep cervical nodes from upper thoracic tumours (i.e. still potentially encompassable in radiation portal)
M1b
Dissemination to non-regional lymph nodes or haematogenously to distant sites
Adapted from UICC (2002).
In the UK, definitive chemoradiation is used for patients with no metastases and who are unfit for surgery, and radiotherapy alone is used for patients unfit for chemotherapy. Patients with tumours of the upper third are often treated with non-surgical therapy. Palliative chemotherapy, radiotherapy (external beam/intraluminal), laser therapy, stent and supportive care alone are all treatment options for patients with advanced disease. Palliative surgical resection should be avoided where possible because there is good evidence that quality of life will never be regained when patients relapse within 2 years of their resection (Blazeby et al., 2000).
Early oesophageal cancer (T1 or T2, N0) Early oesophageal cancer is more common in countries such as Japan, where there are surveillance and screening programmes. Subdivision of this stage is important. Because there is a rich lymphovascular plexus in the submucosa, the incidence of node-positive disease is 16 to 36% for patients with T2 disease. Surgical resection of these cases is the treatment of choice but chemoradiation may play a role if the patient is not fit enough. Photodynamic therapy has been used, either alone or in combination with external beam XRT. As well as being useful for palliation in early inoperable disease, photodynamic therapy can achieve long-term control
Oesophagus
when used alone in approximately 40% of cases. It has also been given in addition to external beam radiotherapy.
Locally advanced oesophageal cancer (T3 or T4 and/or N1) Management depends on stage, performance status, co-morbidity and patient preference. There have been no relevant randomised trials of surgery compared to radiotherapy and caution is needed in interpreting comparisons of historical data. Radiotherapy patients have often been turned down for surgery and surgical staging may result in stage ‘migration.’
Advanced oesophageal cancer (M1) Approximately 60% of patients are not fit for radical therapy, either because of the extent of their disease or because of co-morbidities and poor performance status.
Treatment of non-metastatic carcinoma of the oesophagus
is an ongoing debate about the optimal surgical technique for operable tumours between the two- or threestage, extended lymphadenectomy and the transhiatal approach, although most specialist upper GI surgeons in the UK perform the former.
Postoperative adjuvant therapy Despite improvements in perioperative care, the outcome from surgery for patients with oesophageal cancer remains poor and only 10 to 30% are alive 5 years after diagnosis. In an attempt to improve these results, trials of adjuvant therapy have been undertaken. There is no evidence that chemotherapy given after surgical resection improves overall survival. In addition, neither pre- nor postoperative radiotherapy, when used alone, improves survival and neither is routinely recommended. Adjuvant therapy may improve local control in patients with R1 (microscopic residual) and patients with circumferential margin involvement with low nodal involvement (e.g. fewer than three nodes involved), although the evidence for this is not strong.
Surgery
Preoperative chemotherapy
Surgery is the mainstay of treatment for most patients with oesophageal cancer who have technically resectable disease and are medically fit. All patients should be considered for (neo)adjuvant therapy. Curative surgery is possible in about 25 to 50% of cases; rates vary among countries. It should be performed by specialist upper GI surgeons working within specialist multidisciplinary teams, who should achieve a postoperative mortality rate of 5 to 10%. Transthoracic two-stage (Ivor Lewis) surgery involves laparotomy and coeliac lymph node dissection. Right thoracotomy is performed for mobilisation and resection and mediastinal lymphadenectomy, along with intrathoracic anastomosis. En-bloc dissection involves thoracoabdominal resection, a radical resection that can include stomach and spleen. Total thoracic, three-stage (McKeown) surgery is as per Ivor Lewis above, but with neck exploration, dissection, and anastomosis. The transhiatal approach involves resection by an abdominal laparotomy with a cervical anastomosis. Serious morbidities may arise as a result of moving patients during operations; mediastinal trauma, respiratory problems due to single lung ventilation during surgery and postoperative anastomotic leaks. There
The results from a US Intergroup study (Kelsen et al., 1998) and the UK MRC OE02 trial (Medical Research Council Oesophageal Cancer Working Group, 2002) are conflicting. The former study randomised patients to three cycles of cisplatin and 5-FU before and after surgery (if a response was seen) or surgery alone and found no significant survival difference (2-year survival of 35 versus 37%, p = NS). The UK MRC study randomised 802 patients to two cycles of preoperative cisplatin and 5-FU or surgery alone and found a survival advantage at 2 years for patients who had preoperative therapy (43 versus 34%; HR = 0.79, 95% CI = 0.67–0.93, p = 0.004). These conflicting data are difficult to explain but the results of the much larger UK trial are supported by a Cochrane review that includes nine additional RCTs showing a small survival advantage that becomes statistically significant some years after treatment but is mitigated somewhat by increased toxicity (Malthaner and Fenlon, 2004). Two cycles of cisplatin (80 mg/m2 , day 1) and 5-FU (1000 mg/m2 , days 2 to 5), given before surgery, is now the accepted standard treatment in the UK. This regimen is the standard arm in the current MRC OE05 trial (see below), in which it is being compared with four cycles of ECX chemotherapy (epirubicin, cisplatin and capecitabine). 125
Tom Crosby
Preoperative chemoradiotherapy Two randomised trials have shown better survival rates for chemoradiation (CRT) given 4 to 6 weeks before radical surgery compared to surgery alone. Walsh et al. described a schedule with a relatively high dose per fraction of radiotherapy together with cisplatin/5-FU chemotherapy (Walsh et al., 1996). Although there was an improvement in survival, the validity of the outcome has been questioned because of relatively poor outcome in the surgery-only arm and possible imbalances between the treatment groups. Urba et al. used a hyperfractionated schedule but, perhaps because of an underpowered trial design, the improved outcome was not statistically significant (Urba et al., 2001). The benefit has only been reported in patients with adenocarcinoma. In both trials, preoperative CRT was associated with about a 30% long-term survival rate. In a larger multicentre French study of squamous cancer only, an improvement in cancer-specific survival did not translate into an overall survival benefit (Bosset et al., 1997); this was a result of an increase in postoperative surgical mortality which has also been seen in other studies (leading some to be cautious about using this approach). Together with the lack of consistent benefit, perhaps due to the regimens tested, means that triple modality therapy should only be used in centres that have experience in this complicated therapy and ideally as part of clinical trials. Patients who achieve a pathological complete response (pCR) after CRT (30 to 50%) have a better prognosis and such patients may not benefit from surgery. However, it is not yet possible to reliably identify such patients preoperatively. We know that even in 41% of patients who are shown by endoscopic mucosal biopsy to be clear of disease, residual disease will be found at surgery (Bates et al., 1996). pCR is associated with higher doses of radiotherapy, 5-FU and cisplatin, lower age, and a shorter treatment time (Geh et al., 2000).
Definitive chemoradiotherapy Definitive chemoradiotherapy has been widely used to treat patients with oesophageal cancer, although the data available are predominantly for squamous cell oesophageal cancer (Bedenne et al., 2002; Cooper et al., 1999; Stahl et al., 1996; Wong and Malthaner, 2005). In the UK, patients who are inoperable for medical reasons, in whom a complete R0 is unlikely or who decline surgery, should be considered for definitive CRT if their disease can be encompassed in a radical radiotherapy volume. Studies consistently show a 5-year survival of 126
about 30%, similar to that seen in surgical series (Crosby et al., 2004; Denham et al., 2003). Concurrent chemoradiation has been shown in a randomised trial and a subsequent systematic review to be significantly better than radiotherapy alone, although at the expense of increased toxicity (Cooper et al., 1999). Using radical radiotherapy as monotherapy for carefully selected patients, long-term survival can be achieved in about 20% of cases (Sykes et al., 1998). Other retrospective studies have found survival rates more consistently between 5 and 10% (Earlam and CunhaMelo, 1980) and the randomised trial of chemoradiation compared to radiotherapy alone had no survivors beyond 3 years with radiotherapy alone (Cooper et al., 1999).
Radiotherapy and chemoradiotherapy technique As a practical guide for radiotherapy planning, the following are definitions of oesophageal thirds: r Upper third, cervical oesophagus; 15 to 18 cm. r Middle third, slightly arbitrary upper limit; 18 to 31 cm. r Lower third, to GO junction; 31 to 40 cm. Patients with upper-third tumours can be treated in the same way as patients with a head and neck cancer such as postcricoid carcinoma. A shell is needed for planning, and anterior and anterior oblique fields are used as part of combined modality therapy. This technique will not be discussed further here. The radiotherapy technique for the mid- and lower thirds of the oesophagus includes patient preparation, positioning and immobilisation. The patients are planned with their arms above the head, to avoid obstruction of the posterior oblique fields, in both phases to avoid movement between phases. With regard to localisation and target volume, using all diagnostic data available (endoscopy, EUS, CT, barium swallow), treatment is localised using CT planning, ideally with 3D conformal capabilities. Anterior and two lateral tattoos are used with a central radio-opaque marker. Axial slices, at ≤0.5 cm intervals, are taken and the upper and lower borders determined. The GTV is defined on each slice and the PTV is grown manually or automatically. In the longitudinal direction, the PTV should be grown manually along the axis of the oesophagus and nodal regions because of the inclination of the target volume. The composite target area is drawn on the central slice. Normal structures (lungs, heart and spinal cord) are added in the same way. An outline is taken through the central volume.
Oesophagus
r The GTV should encompass the most proximal and distal disease extent identified on any diagnostic procedure (CT, EUS or PET). r Disease localisation can be difficult at or around the GO junction on a CT scan and this can be made easier by the use of a barium swallow. r The tumour length defined by EUS is nearly always longer than that seen at endoscopy reflecting submucosal spread. Also the disease length is often longer again due to the lymphadenopathy lying more proximal or distal to the tumour. Treatment can be delivered with a single three- or four-field plan throughout treatment or using a twophase technique. Anterior–posterior fields in the first phase reduce the radiation doses to the lung and ensure adequate coverage of the posterior mediastinum, whereas the three-field second phase spares the spinal cord and heart. When EUS has been used to stage the disease, the target volume stays the same for both phases. It is important to know the spinal cord dose that will be received during the second phase so that you can determine when to change to the three-field plan from phase 1. The aim is to treat the tumour and ‘normal’ oesophagus, to allow for submucosal spread, and to treat the mediastinal lymph nodes. The clinical target volume can be described r Superiorly and inferiorly as the EUS-defined extent of the tumour (primary or nodal) with a 2 cm margin along the axis of the oesophagus. r Laterally and anteriorly as the 1 cm margin around EUS-defined tumour. r Posteriorly as the 0.5 to 1 cm margin around EUSdefined tumour. Another 1 cm is added in all directions to create the PTV. An alternative to this is to use a reducing-field technique (5 cm for phase 1 and 2.5 to 3 cm for phase 2), particularly if EUS is not available. For lower-third tumours, if the tumour involves the GO junction (i.e. Siewert types 1 and 2), the inferior margins that define the PTV should be 3 cm below the EUS-defined GTV. However, the inferior margin should be grown manually along the pathway of the draining lymph node stations (i.e. along the lesser curve to include the paracardial and left gastric lymph nodes with or without the common hepatic, coeliac, and splenic arteries). Planning is important: inhomogeneity can occur because of changing body contour, the position of the oesophagus along its length and the different tissues through which radiation must pass to the PTV. During a
single phase or the second phase of the two-phase technique, the posterior oblique fields are wedged, usually 15 to 30o , with the thick end located posteriorly. Occasionally the anterior field can be wedged (∼15o ) in the superior–inferior direction. The plane of treatment may be inclined, requiring head rotations. The plan should be verified in the simulator using barium before starting treatment and at least one portal image should be taken in the first three fractions of each phase on the linear accelerator and then weekly throughout treatment thereafter. Typical radiation doses: r If radiation is the sole treatment, a dose of 60 to 64 Gy (2 Gy per fraction) should be delivered to the target volume. r If chemoradiotherapy is the sole curative therapy, reduce the dose to 50 Gy in 25 fractions. r If chemoradiotherapy is preoperative, give 45 Gy in 25 fractions. The dose, treating each field daily Monday to Friday, is prescribed to the ICRU 50 reference point, usually the point of intersection of the central axes. The PTV minimum should be no less than 95% and the PTV maximum should be no more than 107%, corresponding approximately to a 10% variation of the ICRU-defined minimum and maximum of the dose prescribed to the ICRU 50 reference point. No point outside the PTV should receive greater than 105%. In a two-phase technique, give about 50 to 60% of the dose in phase 1. r In the first phase, give 26 to 30 Gy in 13 to 15 fractions. r In the second phase, give 20 to 24 Gy in 10 to 12 fractions. r To avoid excessive spinal cord dose, plan phase 2 at the same time as phase 1, because the cord will receive 20 to 50% of the dose prescribed in phase 2. r Use 10 MV photons. Recommendations for normal tissue tolerances are as follow: r Combined lungs, V20 ≤ 25%. r Heart, V40 ≤ 30%. r Spinal cord, V45 = 0%. Figure 9.1 shows a 3D conformal radiotherapy plan for carcinoma of the oesophagus.
Concurrent chemotherapy Cisplatin and 5-FU are the agents most commonly used concurrently with RT. They both have reasonable singleagent activity and are potent radiosensitisers. The commonest regimen is probably four three-weekly cycles, in 127
Tom Crosby
Isodose % 95 70 50 20 10
Isodose % 95 70 50 20 10
(c)
(a) 100
PTV
90 Cord PRV 80
Volume (%)
70 60 Heart 50 40 Liver
30
Combined Lungs
20 10 0 0
10
20
30
40
50
60
70
80
90
100
110
Dose (%)
(b)
(d)
Figure 9.1. A 3D conformal plan for radical radiotherapy for carcinoma of the oesophagus. In this plan, a single-phase treatment has been used with four fields: (a) transverse section showing how the dose outside the PTV is spread between critical structures; (b) sagittal section showing the oblique angle of the PTV relative to the horizontal; (c) coronal section showing how the isodoses have been shaped to conform to the PTV, which has been achieved by the use of multileaf collimators; and (d) DVHs for the PTV and organs at risk. PRV = planning organ at risk volume; PTV = planning target volume.
which radiotherapy is given in cycles 3 and 4. This regimen delivers the same number of planned chemotherapy cycles as used in the RTOG-85–01 study (Cooper et al., 1999) and, because the non-concurrent chemotherapy is given in a neoadjuvant phase, there is time for careful radiotherapy planning and improving the patient’s dysphagia before the radiotherapy starts. Side effects from radiotherapy are shown in Table 9.5.
Treatment of recurrent carcinoma of the oesophagus The prognosis of patients with recurrent disease is very poor. The pattern of recurrent disease varies depend128
ing on the initial local therapy given. After surgery, the majority of patients who relapse will do so at distant sites and they should be considered for palliative chemotherapy (as discussed in the next section), depending on their level of fitness. After definitive chemoradiation, the tumour most commonly recurs locally and endoscopic placement of a stent as described in the next section is frequently used to relieve dysphagia.
Palliative treatments and treatment of metastatic carcinoma of the oesophagus Surgery Surgical resection has no role in the palliation of patients with oesophageal cancer.
Oesophagus
Table 9.5. Side effects of radiotherapy to the oesophagus Side effect
Management
Acute Tiredness
General advice for fatigue: explanation, goal setting, moderate exercise if able
Mucositis
Mucilage
Myelosuppression
Maintain haemoglobin above
Pneumonitis
Usually self-limiting, reducing
12 g/dl course steroids if severe Late Benign stricture
Endoscopic evaluation, biopsy, and dilatation; avoid stent if possible
Pulmonary fibrosis
Medical management
Pericarditis
Medical management
Ischaemic heart disease
Medical management
Tracheo-oesophageal
Endoscopic placement of
fistula
quality of life than other combinations (Webb et al., 1997).
Endoscopic treatment Stents In experienced hands, endoscopic stent insertion is usually successful and provides good palliation of dysphagia. Expandable metal stents provide a wider lumen, do not require dilatation to be inserted endoscopically, and are less likely to move than plastic ones. They are, however, more expensive, associated with more pain, and have not been shown to improve survival or quality of life.
Endoscopic laser-thermal Nd-YAG or photodynamic therapy (PDT) Both of these interventions need two or three sessions to provide worthwhile benefit and often need to be repeated every 4 to 8 weeks. Laser therapy is better for exophytic, short lesions of the mid or lower oesophagus. PDT is better for submucosal, flat, for infiltrating tumours and for repeated treatments, but there may be problems related to the UV sensitiser.
covered stent
Dilatation Dilators rarely provide more than a few days of relief from malignant dysphagia.
Alcohol injection Alcohol injection can be good for tumour overgrowth of stents and bleeding tumours.
Palliative chemotherapy For patients with advanced GO cancer, chemotherapy has been shown to improve median survival and quality of life. In those who do respond, dysphagia usually improves after about 10 days. The benefits of second-line therapy are less certain. Responses can be seen with taxanes and with irinotecan-based combination therapy. The following is an example of a chemotherapy regimen using epirubicin, cisplatin and infusional 5-FU (ECF): r Epirubicin 50 mg/m2 day 1. r Cisplatin 60 mg/m2 day 1. r 5-FU 200 mg/m2 per day, days 1 to 21 (or capecitabine, 625 mg/m2 orally b.d., days 1 to 21). r Antiemetics include r Dexamethasone 8 mg i.v. and 4 mg orally b.d., days 3 to 7. r 5-HT3 antagonist i.v. r Metoclopramide 10 mg q.d.s. p.r.n. for 10 days. This regimen is associated with response rates of about 45%, and with significantly better survival and
Palliative radiotherapy Both external beam radiotherapy (EBRT) and intraluminal brachytherapy (ILT) are effective in relieving dysphagia. EBRT has been shown to be effective in 60 to 80% of cases, although the benefits are not maximal until 4 to 6 weeks after treatment. It may make dysphagia worse at first because of mucositis. It is most useful for patients with minimal dysphagia and a dose that will control the local disease for the majority of their remaining life should be used. The dose is 30 Gy in 10 fractions over 2 weeks with anterior–posterior fields or 40 Gy in 15 fractions over 3 weeks either alone or in combination with chemotherapy. Shorter fractionation schedules, such as 20 Gy in 5 fractions over 1 week, may not be effective for long enough. ILT has the advantage of requiring only a single treatment with fewer systemic side effects and its effect may be seen earlier. The dose is 1500 cGy at 1 cm, with highdose-rate Microselectron® . 129
Tom Crosby
Special clinical situations Small-cell carcinoma of the oesophagus Small-cell carcinoma of the oesophagus accounts for 0.8 to 2.4% of all cases of OC. Like pulmonary small-cell carcinoma, it is likely that the cell of origin is a pluripotential stem cell. This may be why the tumours may be mixed with keratin and mucin production, together with small-cell features. Treatment should be similar to that for patients with pulmonary small-cell carcinoma, with combined chemotherapy and radiotherapy. It has been suggested recently that patients with localised disease may benefit from surgery following chemotherapy and radiotherapy because, in those who have had a resection, a significant proportion have had a residual viable tumour (Medgyesy et al., 2000). But any possible advantage must be balanced against the risks of major surgery in patients with a high chance of treatment failure because of metastatic disease and a relatively poor prognosis. The overall median survival is approximately 12 months for patients with limited disease at presentation.
Prognosis of carcinoma of the oesophagus Results from selected series For radical surgery, there is a 50% 5-year survival for early tumour (T1 or T2, N0) and 20% 5-year survival for locally advanced tumours. Radical radiotherapy results in 20% 5-year survival, and radical chemoradiotherapy in 30% 5-year survival. In advanced disease, MS is 5 months without treatment versus 11 months with chemotherapy. All cases show ∼7% survival at 5 years.
Areas of current interest Dose of radiotherapy In the RTOG-85–01 study, which showed a benefit for chemoradiation over radiotherapy alone, there was still a 45% local failure rate. Therefore, INT 0123 was designed to see whether a higher dose of XRT (64.8 Gy) would reduce this finding compared to a modified standard arm (50.4 Gy). There was no difference in the two arms and so the trial closed early. There were nine deaths (9%) in the high-dose arm, although these did not occur during the high-dose therapy. Two-year survival in the high-dose arm was 24% compared to 33% in the standard arm (Minsky et al., 2002). However, an overview of trials of chemoradiation has suggested that higher doses are associated with better outcomes (Geh et al., 2000). 130
Selective surgery A group in Canada, following administration of definitive chemoradiation (three cycles cisplatin/5-FU and XRT 50 Gy in 25 fractions starting with cycle 2), only offered surgery to those who have either biopsy-positive disease or 75% or lower rate of regression on CT or with localised recurrence. This option of selective surgery is the subject of an ongoing EORTC trial.
Current trials in oesophageal cancer The MRC OE05 is a UK multicentre trial that compares the benefit of the chemotherapy schedule from OE02 with four cycles of ECX (epirubicin, cisplatin and capecitabine) in 1300 patients with adenocarcinoma of the oesophagus.
Recent important trials in oesophageal cancer Intergroup 0113 (Kelsen et al., 1998) is a 400-patient trial that compared three cycles of preoperative cisplatin/5FU with the option for postoperative chemotherapy if there was evidence of a response with surgery alone. There was no difference in outcome (35 versus 37% 2-year OS). The MRC 0E02 trial (Medical Research Council Oesophageal Cancer Working Group, 2002) studied 802 patients in the UK with operable oesophageal cancer who were randomised to receive surgery with or without two cycles of cisplatin/5-FU chemotherapy. Survival at 2 years was improved from 34 to 43% in favour of preoperative therapy. The RTOG-85–01 study (Cooper et al., 1999) involved randomised (1986–90) and non-randomised (1990–1) segments: 129 patients in the randomised and 73 in the follow-on study, which compared radiotherapy alone (64 Gy in 32 fractions over 6.5 weeks) with chemoradiotherapy (50 Gy in 25 fractions + cisplatin/5-FU; four cycles, two concomitant with XRT). Of patients who received CRT, 30% survived 3 years, whereas there was no long-term survivor in the RT-only arm of the study.
REFERENCES Bates, B. A., Detterbeck, F. C., Bernard, S. A. et al. (1996). Concurrent radiation therapy and chemotherapy followed by esophagectomy for localized esophageal carcinoma. J. Clin. Oncol., 14, 156–63. Bedenne, L., Michel, P., Bouche, O. et al. (2002). Randomized phase III trial in locally advanced esophageal cancer:
Oesophagus
radiochemotherapy followed by surgery versus radiochemotherapy alone (FFCD 9102). Proc. Am. Soc. Clin. Oncol., Abstr 519. Blazeby, J., Farndon, J., Donovan, J. et al. (2000). A prospective longitudinal study examining the quality of life of patients with esophageal carcinoma. Cancer, 88, 1781–7. Bosset, J. F., Gignoux, M., Triboulet, J. P. et al. (1997). Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N. Engl. J. Med., 337, 161–7. Cooper, J. S., Guo, M. D., Herskovic, A. et al. (1999). Chemoradiotherapy of locally advanced esophageal cancer: long term follow-up of a prospective randomized trial (RTOG 85–01). J. A. M. A., 281, 1623–7. Crosby, T. D., Brewster, A. E., Borley, A. et al. (2004). Definitive chemoradiation in patients with inoperable oesophageal carcinoma. Br. J. Cancer., 90, 70–5. Denham, J. W., Steigler, A., Kilmurray, J. et al. (2003). Relapse patterns after chemoradiation for carcinoma of the oesophagus. Clin. Oncol. (R. Coll. Radiol.), 15, 98–108. Earlam, R. and Cunha-Melo, J. R. (1980). Oesophageal squamous cell carcinoma: II. A critical view of radiotherapy. Br. J. Surg., 67, 457–61. Geh, J., Bond, S., Bentzen, S. et al. (2000). Preoperative chemoradiotherapy in esophageal cancer: evidence of dose response. Proc. Am. Soc. Clin. Oncol., Abstr 958. Kelsen, D. P., Ginsberg, R., Pajak, T. F. et al. (1998). Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N. Engl. J. Med., 339, 1979–84. Kobori, O., Kirihara, Y., Kosaka, N. et al. (1999). Positron emission tomography of esophageal carcinoma using 11 C-choline and 18 F-flurodeoxyglucose: a novel method of preoperative lymphnode staging. Cancer, 86, 1638–48. Luketich, J. D., Schauer, P. R., Meltzer, C. C. et al. (1997). Role of positron emission tomography in staging esophageal cancer. Ann. Thoracic Surg., 64, 765–9. Malthaner, R. and Fenlon, D. (2004). Preoperative chemotherapy for resectable thoracic esophageal cancer (Cochrane Review). In The Cochrane Library, Issue 3. Oxford: Update Software.
Medgyesy, C. D., Wolff, R. A., Putnam, J. B., Jr. et al. (2000). Small cell carcinoma of the esophagus: the University of Texas M. D. Anderson Cancer Center experience and literature review. Cancer, 88, 262–7. Medical Research Council Oesophageal Cancer Working Group. (2002). Surgical resection with or without preoperative chemotherapy in oesophageal cancer: a randomised controlled trial. Lancet, 359, 1727–33. Minsky, B. D., Pajak, T. F., Ginsberg, R. J. et al. (2002). INT 0123 (Radiation Therapy Oncology Group 94–05) phase III trial of combined-modality therapy for esophageal cancer: high-dose versus standard-dose radiation therapy. J. Clin. Oncol., 20, 1167–74. Stahl, M., Wilke, H., Fink, U. et al. (1996). Combined preoperative chemotherapy and radiotherapy in patients with locally advanced esophageal cancer: Interim analysis of phase II trial. J. Clin. Oncol., 14, 829–37. Sykes, A. J., Burt, P. A., Slevin, N. J. et al. (1998). Radical radiotherapy for carcinoma of the oesophagus: an effective alternative to surgery. Radiother. Oncol., 48, 15–21. UICC. (2002). In TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 60–4. Urba, S. G., Orringer, M. B., Turrisi, A. et al. (2001). Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J. Clin. Oncol., 19, 305–13. Walsh, T., Noonan, N., Hollywood, D. et al. (1996). A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N. Engl. J. Med., 335, 462–7. Webb, A., Cunningham, D., Scarffe, J. H. et al. (1997). Randomized trial comparing epirubicin, cisplatin, and fluorouracil versus fluorouracil, doxorubicin, and methotrexate in advanced esophagogastric cancer. J. Clin. Oncol., 15, 261–7. Wong, R. and Malthaner, R. (2005). Combined chemotherapy and radiotherapy (without surgery) compared with radiotherapy alone in localized carcinoma of the esophagus (Cochrane Review). In The Cochrane Library, Issue 2. Oxford: Update Software.
131
10
STOMACH Michael Button and Tom Crosby
Introduction There has been a steady decline in the incidence of gastric cancer in most countries in the world in the past 50 years. However, gastric cancer remains a major health problem: it is the sixth most common malignancy in the UK and the second most common cause of cancerrelated death worldwide. Despite the decline in the cancer that was previously most common, the distal type of cancer, there has been a rapid rise in cancers affecting the gastro-oesophageal junction and cardia, particularly among young white people, which reflects changes in aetiological factors. The only current curative treatment is surgery, but in the UK most patients present late with locally advanced or metastatic disease. Only 25 to 40% of cases are amenable to potentially curative surgery, and, even in these cases, local recurrence may occur in up to 50% and the 5-year survival is 30 to 40%. Because of this and because response rates to combination chemotherapy are 40 to 50% in patients with advanced disease, adjuvant therapy is being used increasingly. Perioperative chemotherapy is used most commonly in the UK, whereas postoperative chemoradiotherapy is standard treatment in the USA.
Types of tumour The types of tumour that affect the stomach are shown in Table 10.1. Adenocarcinoma accounts for 95% of all malignant tumours.
Anatomy The stomach begins at the gastro-oesophageal junction and ends at the pylorus. It is anatomically defined in three parts: the proximal fundus (cardia), the body, and the distal pylorus (antrum). It is covered anteriorly by the peritoneum of the greater sac and posteriorly by the peritoneum of the lesser sac. Proximally, it abuts the diaphragm on the left and the left lobe of the liver on the 132
right. Other adjacent organs (and, therefore, potential sites of direct invasion) are the spleen, the left adrenal gland, the superior portion of the left kidney, the pancreas, and the transverse colon. The vascular supply of the stomach comes from the coeliac axis via the left gastric, right gastric and gastroepiploic arteries (from the common hepatic artery) and the left gastro-epiploic and short gastric arteries (from the splenic artery). The coeliac axis originates at or below the pedicle of T12 in 75% of people and at or above the pedicle of L1 in 25% (Kao et al., 1993). Lymphatic drainage follows the vascular supply, mostly draining into the coeliac nodal area, although a rich lymphatic plexus complicates drainage routes. The nodal stations are shown in Figure 10.1.
Incidence and epidemiology The UK annual incidence of stomach cancer is 15 to 20/100 000; approximately 10 000 new cases per year; approximately 6500 deaths occur per year. The incidence is falling for endemic distal intestinal-type tumours (see p. 133) that are related to environmental factors, but it is rising for proximal intestinal-type tumours that are related to gastro-oesophageal reflux disease (GORD). Proximal tumours now account for 50% of gastric cancer diagnoses in the UK. In some series, oesophageal cancer is now more common than gastric cancer, but this may reflect diagnostic variations in the classification of tumours around the gastro-oesophageal junction. There is a high incidence of stomach cancer in Japan (>90 in 100 000), South America and Eastern Europe, and a low incidence in the USA, Israel and Kuwait. This difference is probably due to exposure to environmental factors in early life. Over several generations, the risk in migrant populations changes toward that of the host country. The peak incidence age of stomach cancer is 65 years, and the male-to-female ratio is 3:2.
Stomach
Table 10.1. Types of stomach tumour Type
Examples
Benign
Inflammatory fibroid polyp Adenoma (sessile/pedunculated polyps) Leiomyoma Adenoid cystic tumour Hamartoma Carcinomas
Malignant primary
Adenocarcinoma: diffuse/intestinal types Squamous Small cell Others Lymphoma (most common site of primary GI lymphoma) Carcinoid tumour GI stromal tumour Malignant secondary
Rare, involvement of stomach by lobular carcinoma breast
GI = gastrointestinal.
1 12 5
8
9
2
7 10 3
11 4
13 6
or nitrates, smoking, low socioeconomic status (which may be related to the previous factors) and radiation exposure at a young age. Helicobacter pylori infection increases cancer risk by three to six times, especially for intestinal-type distal carcinoma; the decreasing incidence may be related to improved treatment of H. pylori. H. pylori has been classified as a class I carcinogen for gastric cancer. Expression of the Cag A virulence factor further increases the risk. With regard to inflammation, Barrett’s oesophagitis (related to obesity, smoking and GORD) is associated with proximal gastric and gastro-oesophageal cancers. Atrophic gastritis (secondary to H. pylori) is associated with distal disease. There is a seven-fold increase in the incidence of malignancy in the 5 years after diagnosis of a benign gastric ulcer and a three-fold increase in incidence following the diagnosis of pernicious anaemia. Genetic risk factors include, for example, CDH1 mutation (E-cadherin), which carries a risk of stomach cancer of up to 80% by the age of 80 (V. Blair, conference presentation, ASCO Gastrointestinal Cancers Symposium virtual meeting, 2006, www.asco.org); inherited cancer syndromes such as FAP, HNPCC, BRCA2 and Li Fraumeni syndrome; and the patient having blood type A. Of benign adenomas, 10 to 20% of tumours more than 2 cm in size transform into carcinoma. Factors that protect against stomach cancer include the use of aspirin or NSAIDs, diets rich in fruit and vegetables or vitamin C and the patient having blood type O.
14
Figure 10.1. Japanese nodal stations as described by Japanese Gastric Cancer Association (1998). 1 = right cardiac; 2 = left cardiac; 3 = lesser curvature; 4 = greater curvature; 5 = suprapyloric; 6 = infrapyloric; 7 = left gastric artery; 8 = common hepatic artery; 9 = coeliac artery; 10 = splenic hilus; 11 = splenic artery; 12 = hepatoduodenal ligament; 13 = retropancreatic; 14 = mesenteric root.
Carcinoma of the stomach Risk factors Environmental risk factors include diets low in vitamins A and C, diets high in salty or smoked foods
Pathology In the past, 50% of stomach tumours started in the pyloric region, 25% in the body and 25% in the cardia; tumours in the lesser curve were three or four times more frequent that those in the greater curve. However, the frequency of proximal tumours is increasing. The two histological variants described in the Lauren classification, intestinal and diffuse, are both mucin-secreting adenocarcinomas (see Tables 10.2 and 10.3 for pathological features). Environmental factors are thought to be more important in the aetiology of the intestinal variant, and this tumour may arise in a multistage process from chronic active gastritis through gastric atrophy, intestinal metaplasia and dysplasia to frank malignancy (Correa, 1995). Genetic factors are more associated with the diffuse type. However, H. pylori infection is associated with both histological types. 133
Michael Button and Tom Crosby
Table 10.2. Pathological features of intestinal-type
Table 10.3. Pathological features of diffuse-type
stomach cancers
stomach cancers
Features
Description
Features
Description
Macroscopic
Usually exophytic, producing nodular,
Macroscopic
Usually endophytic; growth penetrates
Microscopic
polypoid or fungating masses which
the stomach wall and spreads laterally
often ulcerate; often metastasise to the
producing marked thickening
liver
(‘leather bottle stomach’ or linitus
Glandular formation predominates with
plastica) with or without ulceration;
varying degrees of differentiation;
may include serosal or lymph node
associated with intestinal metaplasia in nearly 100% of cases
involvement Microscopic
Diffusely infiltrative with sheets of cells (often ‘signet ring’ due to nuclear compression by mucin); often poorly
The appearance of gastric cancers can also vary. The Borrman classification divides tumours into five types: type I, polypoid or fungating; type II, ulcerated with elevated borders; type III, ulcerated and invading the gastric wall; type IV, diffusely infiltrating (linitus plastica); and type V, unclassifiable (Borrman, 1926). Immunohistochemistry with cytokeratin markers CK7 and 20 and mucin (MUC-1, -2, -5AC and -6) are often positive in oesophago-gastric cancers. It is not clear whether differences in the patterns of staining could be used to differentiate between tumours of the antrum, cardia, and those related to Barrett’s oesophagitis.
Spread Gastric cancers usually spread either by direct extension, the lymphatic system, the vascular system or by the transperitoneal route. Thirty percent of patients have liver involvement at presentation, and around 60% have lymph node involvement (for discussion, see Smalley et al., 2002). Gastric cancer spreads locally to contiguous structures (as described earlier in the anatomy section) or to the anterior abdominal wall. Once local invasion has occurred, there may be haematogenous or lymphatic spread. In lymphatic spread, the rich submucosal lymphatic plexus helps intramural spread, including proximally into the oesophagus, but not usually into the duodenum (‘duodenal block’). The initial draining lymph nodes sit on the greater and lesser curves but other lymph node groups are often involved (including the SCF). The patterns of involvement are difficult to predict because of the rich lymphatic network. Spread via the falciform liga134
differentiated; minimal gland formation; not associated with intestinal metaplasia
ment can produce subcutaneous, periumbilical tumour deposits (Sister Mary Joseph’s nodules). In haematogenous spread, venous drainage is initially to the liver (involved in 30% of cases). Systemic metastases are not as common; lung, bone and brain metastases occur in decreasing frequency. Transcoelomic spread involves peritoneal dissemination after extension through the serosal surface of the stomach, to the ovaries (Krukenberg tumours), rectum or rectal shelf (Blumer’s tumour).
Clinical presentation Symptoms of gastric cancer include anorexia, weight loss, epigastric discomfort, early satiety, dysphagia, vomiting and bleeding (haemetemesis or melaena, 10%). Patients may have a poor performance status because of their disease. Examination findings include epigastric mass, SCF lymph nodes, weight loss (indicator of a poor prognosis), and signs of metastases.
Investigation and staging A full patient history should be taken and directed clinical examination should be performed. FBC (anaemia) and biochemistry (albumin, liver function tests, renal function if chemotherapy contemplated) tests should also be performed. Endoscopy can be highly sensitive
Stomach
Table 10.4. AJCC Cancer Staging Manual 6th Edition for
Table 10.5. Stage groupings for carcinoma of the
carcinoma of the stomach (applies to carcinomas only)
stomach
Stage
Description
Stage
Description
TX
Primary tumour cannot be assessed
Stage 0
Tis N0 M0
T0
No evidence of primary tumour
Stage IA
T1 N0 M0
Tis
Carcinoma in situ: intraepithelial tumour without
Stage IB
T1 N1 M0
invasion of the lamina propria T1
Tumour invades lamina propria or submucosa
T2
Tumour invades muscularis propria or subserosa
T2a
Tumour invades muscularis propria
T2b
Tumour invades subserosa
T3
Tumour penetrates serosa (visceral peritoneum)
T2a/b N0 M0 Stage II
T2a/b N1 M0 T3 N0 M0 Stage IIIA
T2a/b N2 M T3 N1 M0
without invasion of adjacent structures T4
T1 N2 M0
T4 N0 M0
Tumour invades adjacent structures Stage IIIB
T3 N2 M0
Stage IV
T4 N1–3 M0
NX
Regional lymph nodes cannot be assessed
N0
No regional lymph nodes
N1
Metastases in 1–6 regional lymph nodes
T1–3 N3 M0
N2
Metastases in 7–15 regional lymph nodes
Any T
N3
Metastases > 15 regional lymph nodes
MX
Distant metastases cannot be assessed
M0
No distant metastases
M1
Distant metastases
Adapted from AJCC (2002).
in experienced hands but may miss diffuse or intramucosal tumours. A barium swallow may be more sensitive for detecting linitis plastica, which can be missed endoscopically but may show characteristic radiological features. CT scan of the chest/abdomen/pelvis should be performed to assess the primary tumour, lymph nodes and metastases. Endoscopic ultrasound (EUS) is used for assessing the depth of penetration and local lymph node involvement, although EUS is less sensitive in the assessment of stomach cancer than in the assessment of oesophageal tumours. Laparoscopy assesses the peritoneum, liver capsule and mobility of the stomach and is considered a standard investigation before surgical resection. Other investigations should be performed as clinically indicated (e.g. bone scan). CEA has low sensitivity (raised in only 30%) and, therefore, is not clinically useful. PET is experimental at present but may be useful for lymph node staging and in assessing response to neoadjuvant chemotherapy. The aim is to accurately stage patients and to reduce the rate of ‘open-and-close’ laparotomy.
Any N M1
Adapted from AJCC (2002).
Stage classification The AJCC classification and stage groupings are shown in Tables 10.4 and 10.5, respectively. It is important to note the total number of nodes examined. Tumours are also graded: r GX – grade cannot be assessed. r G1 – well differentiated. r G2 – moderately differentiated. r G3 – poorly differentiated. r G4 – undifferentiated.
Treatment overview Surgery is the mainstay of radical treatment, and the type of resection is tailored to the site of the tumour in the stomach. Debate exists about the optimal extent of lymph node dissection because of which there are large geographical variations in treatment. Following the publication of the Intergroup 0116 randomised controlled trial, the use of adjuvant chemoradiotherapy has become standard in the USA, but not in Europe. Combination chemotherapy is effective at improving symptoms and offers a survival benefit with improved quality of life in the palliative setting. Extending the benefit of chemotherapy to the adjuvant setting has proved challenging, although progress is being made with the 135
Michael Button and Tom Crosby
recently presented UK MRC ST02 (MAGIC) trial results, which showed a significant survival benefit from the use of perioperative chemotherapy.
Table 10.6. Patterns of failure following surgery for stomach cancer Incidence (%)
Surgery
Pattern of failure
Radical surgery Complete resection is the only curative treatment (Jansen et al., 2005). However, more than two-thirds of patients diagnosed with gastric cancer will have unresectable disease or metastases at presentation. Of those whose disease is resectable, local and/or regional lymph node recurrence occurs in 54% at second-look laparotomy (Gunderson and Sosin, 1982) and in up to 90% at autopsy (Lim et al., 2005). In the UK, only 5% of patients present with T1 disease. Total gastrectomy may not be necessary as long as free resection margins can be obtained with a subtotal resection. Distal tumours can be treated by partial gastrectomy (if a 6 cm proximal clearance can be achieved). However, proximal tumours usually require total gastrectomy, because more limited resections tend to produce worse functional outcomes and may result in a
(a)
(b)
Figure 10.2. Surgical scheme for a total gastrectomy: (a) before and (b) after, with a Roux-en-Y reconstruction.
(a)
(b)
Figure 10.3. Surgical scheme for a Billroth II partial gastrectomy: (a) before and (b) after.
136
Clinical Reoperation
Autopsy
Locoregional
38
67
80–93
Gastric bed
21
54
52–68
Anastomosis/
25
26
54–60
8
42
52
23
41
30–50
stumps Abdominal wound Lymph nodes Peritoneal seeding
5
Localised
19
Diffuse
22
Distant metastases
52
22
49
From Smalley et al. (2002).
higher rate of local recurrence. Surgical schemes for total and partial gastrectomy are shown in Figures 10.2 and 10.3, respectively. There is controversy about the required extent of surgical resection. D1 resection includes removal of the perigastric nodes within 3 cm of the tumour. D2 resection involves a more extensive lymph node dissection, with removal of lymph nodes around the left gastric artery, hepatic artery, splenic hilum and also a splenectomy and distal pancreatectomy. D2 resection is more common in Japan, whereas the more limited D1 resection is more common in Western centres. Outcomes from more extensive D2 resections performed in Western centres are worse than those obtained in Japan with greater postoperative morbidity and mortality (up to 10% postoperatively). These differences may be related to surgical experience, to the surgery itself, to stage migration, to the fact that patients in Japan are often younger and fitter or even to a difference in disease biology. At least three randomised trials and a Cochrane review (McCulloch et al., 2003) have not shown an advantage for more extensive lymph node resection. The majority of specialist upper GI surgeons in the UK perform a modified D2 resection for suitably fit patients. Despite extensive surgery, locoregional recurrence is common. As a result, the role of adjuvant treatments has been extensively investigated. Patterns of failure after surgery for stomach cancer are shown in Table 10.6.
Stomach
Chemotherapy Adjuvant chemotherapy Despite gastric carcinomas being the most chemosensitive of gastrointestinal carcinomas, with high response rates seen in phase II studies, prospective randomised controlled trials of postoperative adjuvant chemotherapy have not shown a significant effect on survival (Shimada and Ajani, 1999). Several meta-analyses have shown that, if there is benefit, it is small and does not seem to justify routine use of adjuvant chemotherapy (Earle and Maroun, 1999; Hallisey et al., 1994; Hermans et al., 1993). Adjuvant chemotherapy should not be routinely offered to patients with gastric cancer except as part of a clinical trial.
Perioperative chemotherapy The MRC ST02 ‘MAGIC’ trial (Cunningham et al., 2006a) included 503 patients of WHO performance status 0 or 1 with stage II or greater adenocarcinoma of the stomach, lower oesophagus or gastro-oesophageal junction. The extent of resection was not specified and was left to the discretion of the surgeon. Trial accrual was slow and it took over 10 years to complete the study. A protocol change to include patients with adenocarcinoma of the lower-third oesophagus was made toward the end of the study. The results showed that perioperative chemotherapy (three cycles of pre- and three cycles of postoperative epirubicin, cisplatin and continuous infusional 5-FU) improved survival compared to surgery alone, with a higher overall survival (hazard ratio for death = 0.75; 95% CI = 0.60–0.93, p = 0.009; 5-year survival rate, 36 versus 23%) and progression-free survival (hazard ratio for progression = 0.66; 95% CI = 0.53–0.81, p ≤ 0.001). Resected tumours were smaller and of less advanced T-stage after chemotherapy, and only 10% of patients had local recurrence alone, suggesting this regimen was effective at improving local as well as distant control. This chemotherapy regimen (with capecitabine replacing infusional 5-FU) will be the standard arm of the next NCRN multicentred RCT of adjuvant therapy. However, some centres have yet to adopt perioperative chemotherapy as a standard because of concerns regarding the aforementioned trial and the relative poor overall survival for patients in this study.
Palliative chemotherapy Randomised trials comparing palliative chemotherapy with best supportive care have shown a significant
improvement in median survival (Glimelius et al., 1997; Murad et al., 1993; Pyrhonen et al., 1995), from about 3 to 10 months; it has also been shown to improve quality of life and to be cost effective (Glimelius et al., 1997). However, it can be toxic and take up a significant amount of the patient’s remaining life and so the decision about whether it is appropriate for an individual patient needs to be made carefully. It is not clear what combination regimen is the most effective. 5-FU is active as a single agent and is included in most combination regimens. A commonly used regimen in the UK and Europe is ECF (see following discussion), which has response rates of 40 to 50%. A randomised trial has shown it to be more effective than FAMTX (5-FU, adriamycin and methotrexate; Webb et al., 1997). Some newer combination regimens give higher response rates, but the median survival in phase III trials is still generally less than 12 months (Ajani, 2005). Recently published phase III trials have investigated various chemotherapy regimens for advanced disease (Ajani, 2005), all of which are 5-FU based and give similar median overall survival and 2-year survival rates. The TAX-325 trial (the largest to date) compared cisplatin/5-FU with docetaxel/cisplatin/5-FU in 457 chemotherapy-na¨ıve patients and found higher response rates and survival in the docetaxel-containing arm (RR of 37 versus 25%, median OS 9.2 versus 8.6 months, 2-year survival 18 versus 9%) at the cost of greater toxicity (Moiseyenko et al., 2005). Further interesting data have come from the REAL II trial, a phase III trial that evaluates the potential roles of oxaliplatin and capecitabine in chemotherapyna¨ıve patients of ECOG performance status 0 to 2 with histologically proven oesophago-gastric cancer. Of the patients studied, 75% were metastatic and 25% locally advanced; 85% had adenocarcinoma and tumours were evenly divided between oesophagus, GOJ and stomach. The four-arm trial evaluated epirubicin (E), cisplatin (C), oxaliplatin (O), 5-FU (F) and capecitabine (X) in the following combinations: ECF, ECX, EOF and EOX, with doses of X increasing from 500 mg/m2 (continuous daily dose) to 625 mg/m2 after analysis of fluoropyrimidine toxicity suggested that this would be tolerable (Sumpter et al., 2005). This study, presented in abstract form in 2006, has demonstrated equivalence among the treatment arms, suggesting that the more-convenient oral capecitabine may safely replace infusional 5-FU in this combination (Cunningham et al., 2006b). The role of oxaliplatin awaits further 137
Michael Button and Tom Crosby
analysis of toxicity, health, economic, and quality-of-life data. ECF regimen: r Epirubicin 50 mg/m2 , day 1. r Cisplatin 60 mg/m2 , day 1. r 5-FU 200 mg/m2 per day, days 1 to 21 (or capecitabine 625 mg/m2 p.o. b.d., days 1 to 21). r Antiemetics: r Dexamethasone 8 mg i.v. then 4 mg p.o. b.d. for 3 days. r 5-HT3 antagonist i.v. r Metoclopramide 10 mg q.d.s. p.r.n. for 10 days.
Second-line palliative chemotherapy It is not clear whether second-line chemotherapy at relapse following a response to previous chemotherapy is worthwhile. It should only be considered as part of a clinical trial and will be the subject of a UK NCRN trial comparing docetaxel with supportive care. Secondline chemotherapy after a failure to respond to first-line treatment is rarely effective.
Radiotherapy and chemoradiotherapy Adjuvant chemoradiotherapy The very high local recurrence rates after radical surgery mean that adjuvant radiotherapy is an important option. The US Intergroup Study 0116 (Macdonald et al., 2001) showed significant improvements in rates of both disease-free and overall survival for patients treated with adjuvant radiotherapy (45 Gy in 5 weeks) and concomitant 5-FU/folinic acid compared to surgery alone. As a result, adjuvant chemoradiotherapy has become a standard therapy in the USA for high-risk patients. However, only 10% of patients had extensive (D2) resection, 30% did not complete the chemoradiation because of toxicity, and more than 30% of the radiotherapy plans had significant errors. This treatment has not been adopted as a standard therapy in the UK or large parts of Europe. Its benefit following more extensive surgery and following preoperative chemotherapy is not known and it should only be carried out in specialist centres with appropriate experience. Despite its occasional use in practice, the principles of radiotherapy planning of gastric tumours should be known. If it is given as an adjuvant treatment, radiotherapy should start within 10 weeks of surgery when possible.
138
External beam: technique Patient preparation involves nutritional support, which is very important. Patients should be consuming over 1500 kcal/day before starting treatment; however, this is often difficult when more than three-quarters of the stomach has been resected. Differential renal function should be measured because one kidney often receives an abovetolerance dose of radiation. The patient should be immobilised and positioned supine with arms raised and supported. The planning CT should extend from the sternal notch to L3 to generate adequate dose–volume histograms for organs at risk. The target volume is defined using pre- and postoperative CT scans (ideally, diagnostic quality with oral and i.v. contrast as well as a non-contrast planning scan), a pathology report, an operative note and ideally a personal discussion with the surgeon to identify the areas at highest risk of recurrence. The placing of radio-opaque clips can help. The treatment volume will depend on tumour factors (e.g. proximal or distal location, stage and sites of involved lymph nodes) and the type of operation (total or partial gastrectomy) but needs to cover the tumour bed, gastric remnant (if applicable), anastomosis, duodenal stump (distal resection margin), and regional lymphatics (varying depending on those most at risk). Therefore, with respect to the target volume, the tumour bed, anastomosis, and residual gastric remnant should be adequately covered. The nodal areas at risk include the gastric and gastro-epiploic (usually resected with primary); coeliac nodes, porta hepatis, subpyloric, gastroduodenal, splenic-suprapancreatic and retropanceaticoduodenal nodes. (For a more detailed discussion about this treatment technique, see Smalley et al., 2002.) Margins of 1 cm around the CTV should usually be adequate to generate a PTV that accounts for setup errors and organ motion. In the Macdonald Intergroup 0116 study, anterior– posterior fields were used throughout, giving a dose to the cord of approximately 48 Gy and higher doses to the small bowel. CT planning and individually tailored field arrangements may reduce the dose to organs outside the PTV. The typical field size (superior–inferior extent) is T8/9 or T9/10 interspace to L1/2 interspace (for proximal tumours) or L3/4 interspace (distal tumours), a maximum of 20 cm × 20 cm. For verification, all patients should have a simulator check film prior to starting treatment. Portal images
Stomach
Table 10.7. Side effects of chemoradiotherapy for gastric cancer Side effect
Comments
Acute Nausea
Antiemetics including 5-HT3 if
Fatigue
Advice re management of
Diarrhoea
Low residue diet, loperamide as
Myelosuppression
Weekly FBC
required fatigue, check haemoglobin required
Late Myelopathy
The risks of these late sequelae
trolling pain or bleeding. Treatment is usually well tolerated but can cause short-lived nausea and vomiting, abdominal cramps, or diarrhoea, many of which are prevented by the routine use of 5-HT3 antagonists.
Other palliative treatments Endoscopic laser photo-coagulation can be useful for controlling bleeding or for debulking a large, obstructing tumour. Expandable metal stents may relieve dysphagia due to gastro-oesophageal tumours. Occasionally, palliative surgical bypass may be required for gastric outlet obstruction or even gastrectomy for uncontrollable bleeding. A coeliac plexus block may palliate severe pain.
should be kept to a minimum with careful planning
Prognosis
Malabsorption Radiation enteritis
should be obtained on at least the first 2 days of treatment and thereafter once weekly. With regard to dose, fractionation and energy, isocentric treatment should be given with a minimum energy of 6 MV and source-axis distance of 100 cm, giving 45 Gy in 25 fractions over 5 weeks to the entire tumour bed, anastomoses and regional lymph nodes in a single phase, treating all fields each day. MLCs, wedges and dual asymmetric collimators are recommended. Doselimiting structures include: r Kidneys – at least three-quarters of one kidney should receive ≤ 20 Gy. r Heart – no more than 30% should receive > 40 Gy. r Liver – no more than 60% should receive > 30 Gy. r Spinal cord – no part should receive > 45 Gy.
Concurrent chemotherapy A regimen of 5-FU/folinic acid was used in the Intergroup Study 0116 but this regimen is probably suboptimal; infusional 5-FU, oral fluoropyrimidines or a combination regimen may be better, and clinical trials are under way to confirm this. Table 10.7 shows the toxicity from chemoradiotherapy.
Palliative radiotherapy Palliative radiotherapy (30 Gy in 10 fractions or an 8 Gy single fraction, given by antero-posterior fields, simulated with barium contrast) can be very effective in con-
Because patients usually present late, the overall 5-year survival is less than 10%, but it is as high as 70% for early (T1) tumours. The median survival for patients with unresectable (or untreated metastatic) disease is around 4 months. However, even in patients with resectable disease, locoregional recurrence and distant metastases are common (see Table 10.6).
Prognostic factors Prognostic factors include tumour stage at presentation, resectability, morphology (diffuse types have worse prognosis) and poor tumour differentiation.
REFERENCES Ajani, J. A. (2005). Evolving chemotherapy for advanced gastric cancer. Oncologist, 10 (Suppl. 3), 49–58. AJCC. (2002). AJCC Cancer Staging Manual, ed. F. L. Green et al., 6th edn. New York: Springer. Borrman, R. (1926). Geschwulste des magens und duodenums. In Handbuch der Speziellen Pathogischen Antomie und Histologie, Ed. F. Henske and O. Lubarsch. Berlin: Julius Springer, IV-L864–71. Correa, P. (1995). Helicobacter pylori and gastric carcinogenesis. Am. J. Surg. Path., 19 (Suppl. 1), S37–43. Cunningham, D., Allum, W. H., Stenning, S. P. et al. (2006a). Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N. Engl. J. Med., 355, 76–7. Cunningham, D., Rao, S., Starling, N. et al. (2006b). Randomised multicentre phase III study comparing capecitabine with fluorouracil and oxaliplatin with cisplatin in patients with advanced oesophagogastric (OG) cancer: The REAL 2 trial. J. Clin. Oncol., 2006 ASCO Meeting Proceedings Part I, Vol. 24, No. 18S (June 20 Suppl.), LBA4017.
139
Michael Button and Tom Crosby
Earle, C. C. and Maroun, J. A. (1999). Adjuvant chemotherapy after curative resection for gastric cancer in non-Asian patients: revisiting a meta-analysis of randomized trials. Eur. J. Cancer, 35, 1059–64. Glimelius, B., Ekstrom, K., Hoffman, K. et al. (1997). Randomized comparison between chemotherapy plus best supportive care with best supportive care in advanced gastric cancer. Ann. Oncol., 8, 163–8. Gunderson, L. L. and Sosin, H. (1982). Adenocarcinoma of the stomach: areas of failure in a re-operation series (second or symptomatic look) clinicopathologic correlation and implications for adjuvant therapy. Int. J. Rad. Oncol. Biol. Phys., 8, 1–11. Hallisey, M. T., Dunn, J. A., Ward, L. C. et al. (1994). The second British Stomach Cancer Group trial of adjuvant radiotherapy or chemotherapy in resectable gastric cancer: five year follow-up. Lancet, 343, 1309–12. Hermans, J., Bonenkamp, J. J., Boon, M. C. et al. (1993). Adjuvant therapy after curative resection for gastric cancer: meta-analysis of randomized trials. J. Clin. Oncol., 11, 1441–7. Jansen, E., Boot, H., Verheij, M. et al. (2005). Optimal locoregional treatment in gastric cancer. J. Clin. Oncol., 23, 4509–17. Japanese Gastric Cancer Association. (1998). Japanese classification of gastric carcinoma – 2nd English edition. Gastric Cancer, 1, 10–24. Kao, G., Whittington, R. and Coia, L. (1993). Anatomy of the celiac axis and superior mesenteric artery and its significance in radiation therapy. Int. J. Rad. Oncol. Biol. Phys., 25, 131–4. Lim, L., Michael, M., Mann, G. B. et al. (2005). Adjuvant therapy in gastric cancer.J. Clin. Oncol., 23, 6220–32. Macdonald, J. S., Smalley, S. R., Benedetti, J. et al. (2001). Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastro-esophageal junction. N. Engl. J. Med., 345, 725–30.
140
McCulloch, P., Nita, M. E., Kazi, H. et al. (2003). Extended versus limited lymph nodes dissection technique for adenocarcinoma of the stomach. Cochrane Database Syst. Rev., 18, CD001964. Moiseyenko, V., Ajani, J., Tjulandin, S. et al. (2005). Final results of a randomised phase III trial (TAX 325) comparing docetaxel (T) combined with cisplatin (C) and 5-fluorouracil (F) to CF in patients (pts) with metastatic gastric adenocarcinoma (MGA). J. Clin. Oncol., 23, 308s. Murad, A. M., Santiago, F. F., Petroianu, A. et al. (1993). Modified therapy with 5-fluorouracil, doxorubicin, and methotrexate in advanced gastric cancer. Cancer, 72, 37–41. Pyrhonen, S., Kuitunen, T., Nyandoto, P. et al. (1995). Randomised comparison of fluorouracil, epidoxorubicin and methotrexate (FEMTX) plus supportive care with supportive care alone in patients with non-resectable gastric cancer. Br. J. Cancer, 71, 587–91. Shimada, K. and Ajani, J. A. (1999). Adjuvant therapy for gastric carcinoma patients in the past 15 years: a review of Western and Oriental trials. Cancer, 86, 1657–68. Smalley, S., Gunderson, L., Tepper, J. et al. (2002). Gastric surgical adjuvant radiotherapy consensus report: rationale and treatment implementation. Int. J. Radiat. Oncol. Biol. Phys., 52, 283–93. Sumpter, K., Harper-Wynne, C., Cunningham, D. et al. (2005). Report of two protocol planned interim analyses in a randomised multicentre phase III study comparing capecitabine with fluorouracil and oxaliplatin with cisplatin in patients with advanced oesophagogastric cancer receiving ECF. Br. J. Cancer, 92, 1976–83. Webb, A., Cunningham, D., Scarffe, J. H. et al. (1997). Randomized trial comparing epirubicin, cisplatin, and fluorouracil versus fluorouracil, doxorubicin, and methotrexate in advanced esophagogastric cancer. J. Clin. Oncol., 15, 261–7.
11
LIVER , GALLBLADDER AND BILIARY TRACT Somnath Mukherjee and Tom Crosby
Introduction Primary liver cancer is one of the most common cancers worldwide, and it predominantly affects people in developing countries. It is often associated with chronic liver infections and it is more common in males. Patients usually present with advanced disease and treatment options are influenced, and often limited, by co-morbidities, especially poor function of the rest of the liver. Tumours of the gallbladder and biliary tract are relatively rare. Patients often present late with symptoms of biliary obstruction which, together with cholangitis, is a common cause of morbidity and death and is the main target for palliative therapies. Gallbladder and biliary tract tumours are moderately chemosensitive. Cytological or histological confirmation of disease is often difficult, and specialist multidisciplinary teams of expert radiologists and pathologists should be involved in the diagnosis and staging. Radical surgery should only be carried out by tertiary surgical teams, and patients should be managed, whenever possible, within clinical trials.
Tumours of the liver The types of tumour that affect the liver are shown in Table 11.1.
The division between the lateral and medial segments is the gallbladder and IVC and not the falciform ligament. The right part of the liver is made up anteriorly of segments V and VIII (inferior and superior, respectively) and posteriorly of segments VI and VII (inferior and superior, respectively). The latter is the right lateral surface as seen on a CT scan. Contrast in the portal, hepatic, and inferior caval veins on CT allows distinct segmental definition.
Hepatocellular carcinoma Incidence and epidemiology Primary hepatocellular carcinoma (HCC) is the eighth most common solid tumour worldwide. Although the incidence is low in the UK and other developed countries, it is rising because of the increase in viral hepatitis. The UK annual incidence rate is 1.8 in 100 000 with 3000 new cases reported per year. The number of deaths is sometimes slightly higher than the incidence rate; this anomaly is probably due to the poor long-term survival and misdiagnosis of secondary malignancies as the primary disease. The male-to-female ratio is 3:1. Areas of the world that are at high risk include East and Southeast Asia and Sub-Saharan Africa. Countries such as Taiwan, China, Malaysia, Japan and those in middle Africa have the highest incidence rates (up to 36 in 100 000) because of the high incidence of hepatitis.
Anatomy of the liver The liver is divided into right and left lobes by the falciform ligament, but more importantly, in terms of surgical resection, a segmental division can be made and seen with imaging based on the relationship to the hepatic and portal veins. There are four segments in both the left and the right liver. The left liver consists of the caudate lobe (segment I), lateral segments II and III (superior and inferior lateral, respectively, seen extending to the left surface on a CT scan), and medial segment IV.
Risk factors and aetiology The main causative factors in developing HCC appear to be both chronic liver-cell injury and inflammation; both factors can be additive or synergistic. For instance, chronic hepatitis C virus (HCV) infection is more commonly associated with HCC in chronic alcohol drinkers, and aflatoxin interacts with hepatitis B virus (HBV) infection to increase the risk of HCC threefold.
141
Somnath Mukherjee and Tom Crosby
Table 11.1. The range of tumours affecting the liver
Table 11.2. Pathological features of hepatocellular carcinoma
Type
Examples
Benign
Haemangioma Focal nodular hyperplasia
Features
Description
Macroscopic
Yellowish-white nodules with areas of
Nodular regenerative hyperplasia
haemorrhage or necrosis;
Hepatic adenoma
background features of cirrhosis
Hepatic cystadenoma
Microscopic
Lipoma
clear cytoplasm cf. renal cell
Hamartoma
carcinoma
Bile duct cystadenoma Malignant primary
Large cells similar to hepatocytes with
In the fibrolamellar variant the tumour forms cords with collagen strands
Hepatocellular (conventional, fibrolamellar) Cholangiocarcinoma (intrahepatic) Mixed hepatocellular-cholangiocarcinoma Hepatoblastoma
Pathology
Hepatic angiosarcoma Malignant secondary
Any tumour Adenocarcinoma (e.g. bowel, breast, lung) Sarcoma Lymphoma
Tumours may appear as a large solitary lesion (30%) or they may be multicentric (60%). Even what appear to be solitary tumours may have multiple satellite lesions around a central tumour. The pathological features of HCC are shown in Table 11.2.
Carcinoid
Clinical presentation
Infective HCC is associated with HBV and HCV infection. Chronic hepatitis B and C, with or without cirrhosis infection, multiplies the risk of HCC by 100. In Europe the main factor in developing HCC is HCV infection.
Inflammatory HCC is also associated with other causes of cirrhosis such as hereditary haemochromatosis, Wilson’s disease and type 1 glycogen storage disease.
Chemical injury Exposure to a number of chemical agents increases the risk of HCC. Alcohol is the most common; others include nitrites, hydrocarbons, solvents and polyvinyl chloride (PVC), which is particularly associated with hepatic angiosarcoma. Aflatoxins, hepatotoxic agents produced by the fungi Aspergillus flavus and A. parasiticus, are causative factors in Africa and Asia. 142
HCC is usually asymptomatic and patients present only rarely with metastatic disease. HCC is often found incidentally in someone with chronic liver disease (hepatitis or cirrhosis) and it may be picked up on imaging or with a sudden rise in alphafetoprotein (αFP), although as many as 30% of tumours do not produce αFP. Patients may present with decompensation of previously stable chronic liver failure with features such as ascites, jaundice, anorexia, GI bleeding, weight loss and encephalopathy. Occasionally, patients may have paraneoplastic syndromes such as hypoglycaemia, hypercalcaemia, erythrocytosis and ectopic gonadotrophic or adrenocorticotrophic hormone. Metastases to lung, bones and adrenal and lymph nodes are rare.
Diagnostic and staging investigations In addition to diagnosis and staging, investigations are aimed at assessing the patient’s suitability for treatment, particularly, the level of function of the ‘normal’ or nonmalignant liver:
Liver, gallbladder and biliary tract
r Medical and family histories should be noted (prer r r r
r
vious liver disease, drug exposure, symptoms) and examination (liver failure) should be performed. A full blood count should be performed, along with renal, liver and bone profiles. HBV and HCV serology and αFP levels should be checked; blood clotting tests should be performed and serum albumin level taken. The indocyanine green retention rate at 15 minutes determines the adequacy of liver function for surgery. A biopsy should only be done if patients are inoperable. For patients in whom surgery may be possible, biopsy is not performed so that possible tumour seeding can be avoided. Ultrasound, arterial phase CT, MRI and CT angiography all play a role and such specialist investigations should only be carried out by a specialist multidisciplinary team.
Table 11.3. TNM staging classification of hepatocellular carcinoma Stage
Description
T1
Solitary tumour without evidence of
T2
Solitary tumour with evidence of
vascular invasion vascular invasion; or multiple tumours, none more than 5 cm in greatest dimension T3
Multiple tumours more than 5 cm or tumour involving a major branch of the portal or hepatic vein(s)
T4
Tumour(s) with direct invasion of adjacent organs other than the gallbladder or with perforation of visceral peritoneum
NX
Regional lymph nodes cannot be
N0
No regional lymph node metastasis
N1
Regional lymph node metastasis
assessed
Staging classification The TNM and stage groupings are shown in Tables 11.3 and 11.4, respectively.
Treatment overview For patients who are fit enough and have resectable tumours (T1 to T3, N0, M0 and some T4), surgery is the treatment of choice. Patients who are unfit for surgery, or who have unresectable tumours, may be suitable for palliative systemic treatment, although response rates are very poor.
MX
Distant metastasis cannot be assessed
M0
No distant metastasis
M1
Distant metastasis
Adapted from UICC (2002).
Table 11.4. Stage groupings for hepatocellular carcinoma Stage
Surgery Local resection Local resection of small isolated tumours is possible but, because hepatitis or chronic liver-cell injury or cirrhosis is multicentric, recurrence is often seen. Local resection is the treatment of choice for patients with non-cirrhotic livers and for patients with Child’s A cirrhosis who have a small solitary tumour and minimal portal hypertension (hepatic venous pressure gradient less than 10 mmHg).
Liver transplantation for HCC The risk of multicentric recurrence is significantly reduced by the use of liver transplantation (Llovet et al., 1999). Risk factors for recurrence are the presence of HBV infection, multiple tumours, tumour size and vascular invasion (Mazzaferro et al., 1996).
Description
I
T1
N0
M0
II
T2
N0
M0
IIIA
T3
N0
M0
IIIB
T4
N0
M0
IIIC
Any T
N1
M0
IV
Any T
Any N
M1
Adapted from UICC (2002).
r Transplantation treats hepatoma as well as the underlying problem (cirrhosis).
r The selection (Milan) criteria for transplantation require one tumour between 2 and 5 cm or two or three tumours, each 3 cm or smaller; however, selected patients with HCC larger than 5 cm in size have been 143
Somnath Mukherjee and Tom Crosby
treated with multimodal therapy (chemoembolisation, surgery, systemic chemotherapy) with a report of an impressive 5-year survival rate of 44% (Roayaie et al., 2002). r A shortcoming of transplantation is a lack of donors.
Local non-surgical therapy Transarterial chemoembolisation (TACE), hepatic arterial chemotherapy, radiofrequency ablation, ethanol injection and conformal radiotherapy are all options for the palliation of patients with tumours 3 to 5 cm in diameter. There have been reports of survival similar to that in surgical studies and use of these procedures often reflects local experience and expertise. However, it is not clear from the literature whether or not local non-surgical therapy should be the standard approach for inoperable patients, perhaps because patient populations who have received this treatment have been heterogeneous. TACE combines intra-arterial chemotherapy, often doxorubicin and lipiodol (an oily iodised contrast agent administered manually), exploiting the fact that HCCs are highly vascular and predominantly take their blood supply from the hepatic arteries. A meta-analysis that included seven randomised trials of arterial embolisation has shown that TACE is more effective than conservative therapy or 5-FU chemotherapy and shows a significant improvement in 2-year survival (OR = 0.53, 95% CI = 0.32–0.89) and a median survival of more than 2 years (Llovet et al., 2003). Radiofrequency ablation (RFA) has been shown to be more effective than percutaneous ethanol injection, and for tumours up to 2 cm in diameter, an 85% complete response has been reported. For tumours less than 5 cm in size, the overall survival with laparoscopic RFA is comparable to that following surgical resection, although it is associated with higher incidence of local recurrence. Percutaneous RFA is associated with needle-track seeding, which may preclude its use.
Radiotherapy External beam radiotherapy Radiotherapy has not been used extensively to treat HCC, but the advent of conformal and intensitymodulated radiotherapy may change this. It is important to remember that there is significant movement of the liver (1 to 2 cm) in the superior–inferior direction with respiration, and that the tolerance of both normal 144
and diseased liver to radiotherapy is limited. Conformal radiotherapy may have a role in conjunction with chemoembolisation (Seong et al., 1999).
Targeted radiotherapy Radioactive microspheres containing yttrium-90 given by intra-arterial infusion have shown activity in unresectable HCC. A total radiation dose of 120 Gy resulted in better survival than lower doses of radiation (55.9 and 26.2 Gy, respectively; Lau et al., 1994). A technetium lung scan should be done first to rule out significant shunting to the lungs, which contraindicates yttrium-90 treatment.
Systemic therapy HCC is a chemoresistant tumour. The following strategies can also be implemented.
Adjuvant therapy The role of postoperative adjuvant therapy is not well established. The main site of relapse is the residual liver itself, and one randomised study has shown a lower recurrence rate and prolonged survival by giving postoperative intra-arterial iodine-131-labelled lipiodol (Lau et al., 1999). Interferon in patients with hepatitis C in addition to the use of oral retinoids have also been associated with reduced tumour recurrence (Muto et al., 1996; Yoshida et al., 2004).
Palliative single-agent doxorubicin Although doxorubicin has never been shown to be better than supportive care alone, it is the most tested agent both by itself and in combination with other agents and it is frequently used as the comparator ‘standard’ arm in trials of novel therapies. It has response rates of 15 to 25%.
Combination treatments Combination therapy has been reported to give a higher response rate but no improvement in survival. Combinations with biological agents such as interferon, for example in the PIAF regimen (cisplatin, interferon, doxorubicin, and 5-FU), again have given response rates of up to 26% but have significant toxicity and unproven survival benefits (Yeo et al., 2005). However, this regimen rendered 9 of 13 unresectable tumours resectable (Leung et al., 1999) and, in an updated analysis, the authors reported 8 pathological complete responses in 15 patients who underwent surgery (Lau et al., 2001).
Liver, gallbladder and biliary tract
Table 11.5. The Child-Pugh grading system for cirrhosis. Scores of 1 to 3 are given for clinical investigations and features and summed: Child class A = score of 5 to 6; Child class B = score of 7 to 9; Child class C = score greater than 9 Score
Bilirubin (μmol/l)
Albumin (g/l)
Prothrombin time (s)
1
≤ 34
> 35
≤4
None
None
2
34–50
28–35
4–6
1–2
Mild
3
> 50
≤ 28
>6
3–4
Severe
a
Hepatic encephalopathy
Ascites
Encephalopathy score determined by degree of consciousness, intellectual/personality impairment, neurological signs and EEG changes. Adapted from Pugh et al. (1973).
Tamoxifen Despite the fact that 30% of tumours express oestrogen receptors, randomised trials have not shown any benefit from the use of tamoxifen.
Newer agents Newer agents that exhibit activity in HCC include gemcitabine (RR = 18%; Yang et al., 2000), capecitabine (RR = 13%; Lozano et al., 2000), erlotinib (RR = 10%; Philip et al., 2004), irinotecan (RR = 7%; O’Reilly et al., 2001), and bevacuzimab (Schwartz et al., 2004).
Prognosis Surgical series The 5-year survival (5YS) rates for surgery are as follows: r Stage I – 60%. r Stage II – 45%. r Stage III – 20%. r Stage IV – 10%. The 5-year survival after resection is 41 to 74% (Fong et al., 1999; Llovet et al., 1999), whereas the 5-year survival after transplantation (following Milan criteria) is 70 to 80% (Bismuth et al., 1999; Llovet et al., 1999).
Non-surgical series The overall 5-year survival for non-surgical series is 10% or less. The prognosis of patients with HCC is significantly influenced by the severity of cirrhosis present in the non-malignant liver. This can be graded according to the Child-Pugh classification (see Table 11.5); for example, patients with Child’s C cirrhosis with HCC have a median survival of approximately 3 months. This system is also used for patient selection and stratification
in clinical trials and to help compare outcomes from such studies.
Current trials The UK NCRN trial Hep-1 is a randomised clinical trial that evaluates the benefits of doxorubicin chemoembolisation compared to systemic doxorubicin in patients with unresectable, advanced HCC.
Angiosarcoma of the liver Angiosarcomas of the liver are rare tumours but they are notable because of their association with exposure to PVC monomers. They occur 10 to 20 years after exposure and not always in the most heavily exposed individuals. They have also been described in patients who received thorotrast (a diagnostic radioactive contrast agent used between 1930 and 1950). Patients present with a large, often painful, hepatic mass. Treatment consists of surgical resection, where possible, or palliative anthracycline chemotherapy.
Carcinoma of the gallbladder and biliary tree Types of tumour The types of tumour affecting the gallbladder and biliary tree are shown in Table 11.6; more than 80% are adenocarcinomas.
Incidence and epidemiology In England and Wales, 1200 new cases are reported each year. The 1-year survival is 22%, and the 5-year survival is 5 to 10%; the male-to-female ratios are 2:3 and 1:1 145
Somnath Mukherjee and Tom Crosby
Table 11.6. Types of tumour affecting the gallbladder
Table 11.7. Bismuth classification for perihilar
and biliary tree
tumours
Anatomical site
Examples
Carcinoma gallbladder
Adenocarcinoma Adenosquamous Squamous cell Small cell
Carcinoma biliary tree
Type
Description
Type I
Tumour is below confluence of right and left
Type II
Confined to the confluence
Type III
Extension into right or left hepatic ducts
Type IV
Extension into right and left hepatic ducts or
ducts
Cholangiocarcinoma
multicentric
Nodular Diffuse
Adapted from Bismuth and Corlette (1975).
Papillary
for gallbladder cancer and cholangiocarcinoma, respectively. The cases are distributed as follow: r Gallbladder cancer, 40%. r Cholangiocarcinoma (intra- and extrahepatic), 43%. r Periampullary, 13%. r Others, 4%.
Risk factors and aetiology
Left hepatic duct Right hepatic duct Common hepatic duct
Upper third including hilus
Common bile duct Gallbladder
Middle third
Carcinoma of the gallbladder Risk factors for gallbladder cancer include: r Obesity. r Gallstones (especially those larger than 3 cm). r Chronic typhoid and paratyphoid carriers. r Polyps (greater than 1 cm in diameter). r Ulcerative colitis.
Duodenum
Lower third
Pancreas
Figure 11.1. Locations of the upper, middle and lower thirds of the extrahepatic biliary tree (as described in Chamberlain and Blumgart,
Cholangiocarcinoma
2000). Perihilar tumours are classified further according to the Bismuth
Risk factors for cholangiocarcinoma include r Primary sclerosing cholangitis (lifetime risk of 10%), which can be associated with ulcerative colitis. r Clonorchis sinensis infestation. r Polycystic liver disease, gallstones (somewhat weaker correlation than with carcinoma of gallbladder). r Choledochal cyst. r Caroli’s disease, a rare congenital condition of multiple saccular dilatations of the intrahepatic bile ducts. r Chemical carcinogens (aflatoxin, vinyl chloride, methylene chloride, thorotrast).
classification (see Table 11.7).
Anatomy The anatomy of the gallbladder and main bile ducts is shown in Figure 11.1. The Bismuth-Corlette classification is shown in Table 11.7. 146
Clinical presentation The most common presentation of patients with cholangiocarcinoma is obstructive jaundice (gallbladder, bile duct). Patients also present with fluctuating jaundice (periampullary carcinoma) and constitutional symptoms such as weight loss (> 50% of cases), anorexia, and fatigability. Hepatomegaly and pain in the right upper quadrant (especially for hilar tumours and carcinoma of the gallbladder) also occur. A palpable, nontender gallbladder can indicate an obstructive tumour below the level of the cystic duct (Courvoisier’s law), also gallbladder carcinoma. Symptoms and signs of metastasis include ascites and pleural effusion. Carcinoma
Liver, gallbladder and biliary tract
of the gallbladder can also present as acute or chronic cholecystitis; incidental diagnosis of the cancer is made at surgery.
Clinical examination LFT: obstructive picture US shows dilated bile duct
Investigations Blood tests Following a full blood count, renal and liver profiles and test for CA19–9 tumour marker, LFT derangement consistent with obstructive jaundice (raised alkaline phosphatase, gamma-glutamyl transferase, bilirubin) and persistent elevation of CA19–9 despite decompression of the biliary tree suggest biliary duct/pancreatic malignancy. CA19–9 is elevated in up to 85% of patients with cholangiocarcinoma.
Proximal obstruction
Distal obstruction
PTC Cytology ± stent
ERCP / MRCP Cytology ± stent
Figure 11.2. Investigation strategy for a patient with obstructive jaundice. ERCP = endoscopic retrograde cholangiopancreatogram; LFT = liver function tests; MRCP = magnetic resonance cholangiopancreatogram; PTC = percutaneous transhepatic
Imaging Ultrasonography is the first-line investigation for suspected biliary tract obstruction; it is used to find intrahepatic duct dilatation. Extrahepatic ducts may be dilated in carcinoma of the distal bile duct/pancreatic head tumours. Ultrasonography is useful in ruling out choledocholithiasis. A CT scan is useful in assessing the extent of local disease and regional lymphadenopathy (gallbladder, intrahepatic and perihilar cholangiocarcinoma) and to rule out metastatic disease. It can establish resectability in approximately 60% of cases. ERCP and PTC are used to image the intra- and extrahepatic bile ducts and to localise the site of obstruction. PTC is useful for proximal tumours (perihilar cholangiocarcinomas), whereas ERCP is most useful for distal bile duct tumours. Although invasive, the procedures allow for obtaining a cytological diagnosis (positive in 30%), biopsy, and insertion of endobiliary stents. EUS is useful for imaging the gallbladder, distal extrahepatic bile duct and lymph nodes, and it facilitates a guided biopsy. Laparoscopy rules out peritoneal metastases before curative surgery. MRI and magnetic resonance cholangiopancreatogram (MRCP) are non-invasive investigations that are used to evaluate the biliary tree in suspected cholangiocarcinoma.
Advantages The advantages of imaging are that the procedures are non-invasive, they show the extent of duct involvement (as in MRCP), they indicate local staging (invasion of
cholangiograph; US = ultrasound.
adjacent liver, vessels and lymph nodes) and they detect liver metastasis.
Disadvantage A disadvantage is that imaging does not give histological/cytological diagnosis. PET scanning is under investigation and has been shown to detect cholangiocarcinomas that are more than 1 cm in diameter and to detect metastases (not detected by other imaging) in 30% of cases (Anderson et al., 2004). Figure 11.2 shows an investigation strategy for a patient with obstructive jaundice.
Treatment overview Fewer than 20% of patients have resectable tumours at the time of diagnosis and there is a high risk of recurrence after surgery. The success of adjuvant treatment is unproven and treatment should only be considered in the context of a clinical trial. Radiotherapy and chemoradiation, both as adjuvant and primary therapy, should be regarded as investigational and are not widely practised in the UK. Palliative chemotherapy is an option for fitter patients (performance status 0 to 2) with incurable disease although there is no randomised trial showing a survival advantage of palliative chemotherapy over that of active supportive care. A plastic or metallic stent may be useful in palliating obstructive jaundice, a common feature of the disease. Given these uncertainties about management, patients should always be considered for clinical trials. 147
Somnath Mukherjee and Tom Crosby
Hepaticojejunostomy is usually possible for tumours below the first division of the left or right main duct with uninvolved vessels. There is a high risk of local recurrence at the porta hepatis and of distant metastases. Surgery is an option for perihilar tumours: r Bismuth classifications I and II require en bloc resection of the bile duct, gallbladder, lymph nodes and Roux-en-Y hepaticojejunostomy. r Bismuth III is as above, plus right or left hemihepatectomy. r Bismuth IV is as Bismuth I and II, plus extended right or left hemi-hepatectomy. Surgery for the lower third involves a pancreaticoduodenectomy. Intrahepatic cholangiocarcinoma requires a resection of involved segments/lobes.
gemcitabine, oxaliplatin, docetaxel, interferon and the novel tumour antibiotic rebeccamycin. The combination of 5-FU and mitomycin has been reported to give a response rate of 26% (Chen et al., 2001) and one randomised phase II trial showed superior survival and response rate for the addition of cisplatin to 5-FU (median survival 7.8 versus 5.3 months; 19 versus 7%; Mitry et al., 2002). Gemcitabine monotherapy is associated with response rates of 22 to 45% and a median survival of 6 to 20 months. The addition of cisplatin may enhance the efficacy of gemcitabine and this question is being addressed in the ABC 02 study, which is randomising patients with inoperable biliary tract carcinoma to single-agent gemcitabine with or without cisplatin. The combination of gemcitabine and oxaliplatin has demonstrated a response rate of 35.5% (MaindraultGoebel et al., 2003). Oral capecitabine has also shown activity in biliary tract tumours, either as a single agent or in combination with cisplatin (Kim et al., 2003), gemcitabine (Cho et al., 2005) and mitomycin (Kornek et al., 2004). Acceptable treatment options include the following: 1. 5-FU (modified by leucovorin) with or without mitomycin C. Although most trials have used bolus 5-FU, an infusional regimen through a central access device is more commonly used in the UK. 2. Cisplatin and 5-FU. 3. Gemcitabine 1000 mg/m2 , weeks 1 to 7, then 1 week off; subsequent cycles weekly for 3 out of 4 weeks.
Chemotherapy
Photodynamic therapy
Adjuvant chemotherapy
Photodynamic therapy involves injection of a photosensitiser followed by direct endoscopic illumination of the tumour using light of a specific wavelength, which causes activation of the photosensitising compound, release of oxygen free radicals and tumour death. A pilot study has shown a drop in bilirubin levels and an increase in survival rates that compares favourably to historical series (Ortner, 2001).
Surgery Radical surgery for carcinoma of the gallbladder Carcinoma of the gallbladder is found incidentally in fewer than 1% of cases of routine cholecystectomies for cholelithiasis; carcinoma in situ and tumours limited to the muscle layer may be effectively treated by cholecystectomy alone. Advanced stages require radical cholecystectomy, which involves nodal dissection and may entail excision of adjacent liver tissue. Laparoscopic procedures are contraindicated in patients with known carcinoma because of the high risk of needletrack seeding.
Radical surgery for carcinoma of the bile duct
One randomised trial has shown there is no survival advantage obtained from giving adjuvant chemotherapy using infusional 5-FU and mitomycin followed by a prolonged course of oral fluoropyrimidine over surgery alone (Takada et al., 2002). Patients should be considered for entry into clinical trials like BILCAP, which is evaluating the role of adjuvant capecitabine in patients with completely resected biliary tract cancer.
Palliative chemotherapy There is no evidence from randomised trials that chemotherapy is superior to supportive care alone. Single-agent 5-FU (modified by leucovorin) has been reported to produce response rates of up to 32% (Choi et al., 2000). Other drugs that seem to be active in cholangiocarcinoma include epirubicin, mitomycin, 148
Stent insertion Biliary obstruction and cholangitis are common causes of morbidity and death. Plastic stents are placed if surgery is planned or if the prognosis is poor. Metallic stents have a lower rate of complications (e.g. from infection or blockage) but are more expensive. Patients should be prescribed ursodeoxycholic acid in
Liver, gallbladder and biliary tract
an attempt to keep the stent open. If the obstruction cannot be relieved endoscopically from below, PTC can be attempted with a view to external biliary drainage and thereafter by placing the stent internally across the stricture.
Prognosis Prognosis after surgery Lymph node metastasis, perineural invasion, positive resection margin and perihilar tumours are all associated with a poorer prognosis. In one series a 5-year survival rate of 43% was reported for patients who have resected intrahepatic cholangiocarcinoma (Nakagohri et al., 2003).
Ongoing trials in biliary tract cancer In the ABC 02 study, patients with advanced inoperable biliary tract cancers are randomised to single-agent gemcitabine or a gemcitabine and cisplatin combination. In the BILCAP trial, patients with operable biliary tract cancer are randomised to surgery alone or to surgery and adjuvant capecitabine chemotherapy. The PHOTOSTENT 2 trial is for patients with inoperable biliary tract cancers; randomisation is between biliary stenting with or without photodynamic therapy. On progression, patients are eligible for the ABC 02 study. SWOG 0202 is a phase II study of gemcitabine and capecitabine in patients with unresectable locally advanced or metastatic gallbladder cancer or cholangiocarcinoma.
REFERENCES Anderson, C. D., Rice, M., Pinson, C. W. et al. (2004). Fluorodeoxyglucose PET imaging in the evaluation of gallbladder carcinoma and cholangiocarcinoma. J. Gastrointest. Surg., 8, 90–7. Bismuth, H. and Corlette, M. B. (1975). Intrahepatic cholangioenteric anastomosis in carcinoma of the hilus of the liver. Surg. Gynecol. Obstet., 140, 170–8. Bismuth, H., Majno, P. E. and Adam, R. (1999). Liver transplantation for hepatocellular carcinoma. Semin. Liver Dis., 19, 311–22. Chamberlain, R. S. and Blumgart, L. H. (2000). Hilar cholangiocarcinoma: a review and commentary. Ann. Surg. Oncol., 7, 55–66. Chen, J. S., Lin, Y. C., Jan, Y. Y. et al. (2001). Mitomycin C with weekly 24-h infusion of high-dose 5-fluorouracil and leucovorin in patients with biliary tract and periampullar carcinomas. Anticancer Drugs, 12, 339–43.
Cho, J. Y., Paik, Y. H., Chang, Y. S. et al. (2005). Capecitabine combined with gemcitabine (CapGem) as first-line treatment in patients with advanced/metastatic biliary tract carcinoma. Cancer, 104, 2753–8. Choi, C. W., Choi, I. K., Seo, J. H. et al. (2000). Effects of 5-fluorouracil and leucovorin in the treatment of pancreatic-biliary tract adenocarcinomas. Am. J. Clin. Oncol., 23, 425–8. Fong, Y., Sun, R. L., Jarnagin, W. et al. (1999). An analysis of 412 cases of hepatocellular carcinoma at a Western center. Ann. Surg., 229, 790–800. Kim, T. W., Chang, H. M., Kang, H. J. et al. (2003). Phase II study of capecitabine plus cisplatin as first-line chemotherapy in advanced biliary cancer. Ann. Oncol, 14, 1115–20. Kornek, G. V., Schuell, B., Laengle, F. et al. (2004). Mitomycin C in combination with capecitabine or biweekly high-dose gemcitabine in patients with advanced biliary tract cancer: a randomised phase II trial. Ann. Oncol., 15, 478–83. Lau, W. Y., Leung, T. W., Ho, S. K. et al. (1999). Adjuvant intra-arterial iodine-131-labelled lipiodol for resectable hepatocellular carcinoma: a prospective randomised trial. Lancet, 353, 797–801. Lau, W. Y., Leung, T. W., Ho, S. et al. (1994). Treatment of inoperable hepatocellular carcinoma with intrahepatic arterial yttrium-90 microspheres: a phase I and II study. Br. J. Cancer, 70, 994–9. Lau, W. Y., Leung, T. W., Lai, B. S. et al. (2001). Pre-operative systemic chemoimmunotherapy and sequential resection for unresectable hepatocellular carcinoma. Ann. Surg., 233, 236–41. Leung, T. W. T., Patt, Y. Z., Lau, W. Y. et al. (1999). Complete pathological remission is possible with systemic combination chemotherapy for inoperable hepatocellular carcinoma. Clin. Cancer Res., 5, 1676–81. Llovet, J. M., Fuster, J. and Bruix, J. (1999). Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation. Hepatology, 30, 1434–40. Llovet, J. M., Burroughs, A. and Bruix, J. (2003). Hepatocellular carcinoma. Lancet, 362, 1907–17. Lozano, R. D., Patt, Y. Z., Hassan, M. M. et al. (2000). Oral capecitabine (Xeloda) for the treatment of hepatobiliary cancers (hepatocellular carcinoma, cholangiocarcinoma and gallbladder cancer). Proc. Am. Soc. Clin. Oncol., 19, Abstr. 1025. Maindrault-Goebel, F., Selle, F., Rosmorduc, O. et al. (2003). A phase II study of gemcitabine and oxaliplatin (GEMOX) in advanced biliary adenocarcinoma (ABA). Final results. Proc. Am. Soc. Clin. Oncol., 22, Abstr. 1178. Mazzaferro, V., Regalia, E., Doci, R. et al. (1996). Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N. Engl. J. Med., 334, 693–9. Mitry, E., van Custem, E., Van Laethem, J. et al. (2002). A randomised phase II trial of weekly high-dose 5FU (HD-FU) with and without folinic acid (FA) and cisplatin (P) in patients (pts) with advanced biliary tract carcinoma: the EORTC 40955 trial. Proc. Am. Soc. Clin. Oncol., 21, Abstr. 696. Muto, Y., Moriwaki, H., Ninomiya, M. et al. (1996). Prevention of second primary tumors by an acyclic retinoid, polyprenoic
149
Somnath Mukherjee and Tom Crosby
acid, in patients with hepatocellular carcinoma. N. Engl. J. Med., 334, 1561–7. Nakagohri, T., Asano, T., Kinoshita, H. et al. (2003). Aggressive surgical resection for hilar-invasive and peripheral intrahepatic cholangiocarcinoma. World J. Surg, 27, 289–93. O’Reilly, E. M., Stuart, K. E., Sanz-Altamira, P. M. et al. (2001). A phase II study of irinotecan in patients with advanced hepatocellular carcinoma. Cancer, 91, 101–5. Ortner, M. (2001). Photodynamic therapy for cholangiocarcinoma. J. Hepatobil. Pancreat. Surg., 8, 137–9. Philip, P. A., Mahoney, M., Thomas, J. et al. (2004). Phase II trial of erlotinib (OSI-774) in patients with hepatocellular or biliary cancer. J. Clin. Oncol., 2004 ASCO Meeting Proceedings (Post-Meeting Edition), 22 (no. 14S July 15 Suppl.), 4025. Pugh, R. N. H., Murray-Lyon, I. M., Dawson, J. L. et al. (1973). Transection of the oesophagus for bleeding oesophageal varices. Br. J. Surg., 60, 646–9. Roayaie, S., Frischer, J. S., Emre, S. H. et al. (2002). Long-term results with multimodal adjuvant therapy and liver transplantation for the treatment of hepatocellular carcinomas larger than 5 centimetres. Ann. Surg., 235, 533–9. Schwartz, J. D., Schwartz, M. V., Goldman, J. et al. (2004). Bevacizumab in hepatocellular carcinoma in patients without metastasis and without invasion of the portal vein. J. Clin.
150
Oncol., 2004 ASCO Meeting Proceedings (Post-Meeting Edition), 22 (14S July 15 Suppl.), 4088. Seong, J., Keum, K. C., Han, K. H. et al. (1999). Combined transcatheter arterial chemoembolization and local radiotherapy of unresectable hepatocellular carcinoma. Int. J. Radiat. Oncol. Biol. Phys., 43, 393–7. Takada, T., Amano, H., Yasuda, H. et al. (2002). Is postoperative adjuvant chemotherapy useful for gallbladder carcinoma? A phase III multicenter prospective randomized controlled trial in patients with resected pancreaticobiliary carcinoma. Cancer, 95, 1685–95. UICC. (2002). TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 81–3. Yang, T. S., Lin, Y. C., Chen, J. S., et al. (2000). Phase II study of gemcitabine in patients with advanced hepatocellular carcinoma. Cancer, 89, 750–6. Yeo, W., Mok, T. S., Zee, B. et al. (2005). A randomized phase III study of doxorubicin versus cisplatin/interferon alpha-2b/doxorubicin/fluorouracil (PIAF) combination chemotherapy for unresectable hepatocellular carcinoma. J. Natl. Cancer Inst., 97, 1532–8. Yoshida, H., Tateishi, R., Arakawa, Y. et al. (2004). Benefits of interferon therapy in hepatocellular carcinoma prevention for individual patients with chronic hepatitis C. Gut, 53, 425–30.
12
EXOCRINE PANCREAS Somnath Mukherjee and Tom Crosby
Introduction Nearly 7000 new cases of pancreatic carcinoma are diagnosed in the UK each year and nearly as many patients die from the disease. The major risk factors include smoking, diet and a previous total gastrectomy. There is also a positive association between long-standing diabetes and pancreatic cancer. Surgery is the only curative option, but fewer than 20% of patients are suitable for this approach. Gemcitabine-based chemotherapy is the mainstay of treatment for fit patients with advanced or metastatic cancers. Chemoradiation is an option for locally advanced disease. There has been growing interest in combining gemcitabine with radiotherapy (including newer radiotherapy techniques like IMRT), biological agents (cetuximab, erlotinib and bevacizumab), and other chemotherapy agents (oxaliplatin and capecitabine).
Types of pancreatic tumour Benign cysts can be congenital and they arise from anomalous development of the pancreatic ducts. Pseudocysts are loculated collections of fluid arising from necrosis, inflammation or haemorrhage. They are often solitary, can measure 5 to 10 cm and are often found adjacent to the pancreas in the region of the tail. Benign cystadenomas occur in elderly women and are found incidentally at autopsy or during other investigations. Microcystic and papillary-cystic variants are found in younger women. The types of pancreatic tumour are shown in Table 12.1.
Incidence and epidemiology There are about 7000 new cases of pancreatic cancer in the UK each year and almost as many die from the disease. The annual incidence is 8.5 in 100 000; peak incidence occurs for men in their eighties and for women in their nineties.
Pancreatic cancer is the sixth leading cause of cancer death overall: the sixth most common cause of death in men (4% of cancer mortality) and the fifth most common cause of death in women (5% of cancer mortality). There is a male preponderance toward the disease (male-to-female ratio of 4:3). Most cases of the disease (80%) occur in the head of the pancreas.
Carcinoma of the exocrine pancreas Risk factors and aetiology Three percent of pancreatic cancers may be inherited. A Westernised diet (rich in protein and carbohydrates and poor in fruit and fibre) increases the risk, and cigarette smoking doubles the risk. Cancer family syndromes include inherited chronic pancreatitis, inherited diabetes mellitus and ataxia telangiectasia syndrome. Toxic chemical risk factors are 2-naphthylamine, benzidine, and DDT exposure. Other risk factors include long-standing diabetes mellitus, total gastrectomy (two to five times the risk) and pernicious anaemia.
Pathology The pathological features of carcinoma of the pancreas are shown in Table 12.2.
Anatomy The pancreas is a retroperitoneal structure (peritoneum on the anterior surface only) and it lies in front of the first and second lumbar vertebrae. It has four parts: head (including uncinate process), neck, body and tail. The pancreatic duct and common bile duct pass through the head to open in the ampulla of Vater; invasion or compression of these structures leads to obstructive jaundice and pancreatic insufficiency. Lymphatic drainage involves the pancreaticoduodenal, suprapancreatic, pyloric and pancreaticosplenic 151
Somnath Mukherjee and Tom Crosby
Table 12.1. Types of pancreatic tumour
Table 12.2. Pathological features of pancreatic cancers
Type
Examples
Features
Description
Congenital cysts
Macroscopic
White, scirrhous, infiltrative margins
Pseudocysts
Microscopic
Clearly recognisable glandular structures,
Benign tumours
Serous cystadenoma Pancreatic cysts Papillary-cystic/microcystic Primary tumours
Adenocarcinoma
of exocrine
Ductal
pancreas
Acinar
may be mucinous, occasional adenosquamous growth pattern Occasionally arise in cysts (cystadenocarcinoma) or from acinar cells (acinar cell carcinoma).
Anaplastic Cystadenocarcinoma Adenoacanthoma Squamous cell carcinoma Sarcoma Solid and papillary neoplasms Neuroendocrine
Islet cell tumour
tumours of
Gastrinoma
pancreas (see
Glucagonoma
Chapter 36)
VIPoma Carcinoid Somatostatinoma
Metastasis
the liver and the spinal cord are likely to be included in the radiation fields when planning radiotherapy to the pancreas.
Rare (breast, lung, melanoma, non-Hodgkin lymphoma) More frequently metastases to retropancreatic lymph nodes which mimic pancreatic tumour
Clinical presentation Pain (epigastric for head lesions, left upper quadrant for tail lesions), anorexia and weight loss are the most commonly presented symptoms. Back pain relieved by leaning forward is due to infiltration of the retroperitoneal structures and usually suggests inoperability. Metastatic disease (liver, peritoneum, lung, rarely bone) is common at presentation. Clinical examination should also include careful palpation for supraclavicular lymphadenopathy (Virchow’s node).
Clinical features of pancreatic cancer nodes, which drain to coeliac and superior mesenteric nodes. Vagus and splanchic nerves form the coeliac and superior mesenteric plexus; posterior extension of a tumour can involve the first and second coeliac ganglia, leading to back pain. Vessels in close proximity to the pancreas include the superior mesenteric vessels, the portal vein, the splenic vein and the coeliac artery and its branches. Tumour extension to these structures is often a sign of inoperability and can lead to splenic or portal vein thrombosis. In relation to the surrounding organs, the head of the pancreas lies within the C-loop of the duodenum and in close proximity to the stomach and jejunum. The tail abuts the hilum of the spleen and the head is close to the right kidney. The aforementioned structures along with 152
Symptoms and signs from primary tumour and local spread include: r Gastric outlet obstruction (duodenal spread). r Obstructive jaundice. r Cholangitis. r Fatty diarrhoea (obstruction of bile duct by primary or malignant lymphadenopathy). r Back pain (spread to retroperitoneum and coeliac plexus). r New-onset diabetes mellitus. r Oesophageal varices (portal vein thrombosis). r Altered bowel habit (infiltration of colon). Symptoms and signs from metastatic disease include: r Jaundice (extensive liver metastases). r Abdominal pain and ascites. r Blumer’s shelf (peritoneal metastasis in pouch of Douglas, which can be palpated rectally).
Exocrine pancreas
r Shortness of breath (pulmonary metastasis, pleural effusion). r Virchow’s node (malignant left supraclavicular node). Paraneoplastic features include: r Migratory thrombophlebitis (Trousseau’s sign). r Weber-Christian disease (subcutaneous fat necrosis, polyarthralgia, eosinophilia), which is associated with acinar cell tumour. r Dermatomyositis/polymyositis. Constitutional symptoms include fatigue, weight loss, and anorexia.
Investigation and staging Investigations are directed towards diagnosis and staging of the tumour.
Blood tests A full blood count, liver function tests, urea and electrolytes are checked. A coagulation profile is necessary for patients who present with obstructive jaundice. Levels of the CA19–9 tumour marker are raised in 70% of patients with pancreatic cancer. A low initial level and a fall in marker with therapy are associated with a better prognosis. The CA19–9 level needs to be interpreted with caution in patients with obstructive jaundice because the jaundice can cause high levels of the marker.
imaging to assess vascular involvement. EUS is also useful for obtaining a biopsy, especially in patients with a small tumour in whom a standard ultrasound (US) or CT-guided biopsy may be difficult.
Further staging and diagnostic procedures For patients who are otherwise suitable for radical surgery, preoperative laparoscopy may find unsuspected peritoneal metastases. Intraoperative US may detect small liver metastases not seen on conventional imaging. In the past, pancreatic lesions have often been treated as ‘cancers’ based on a radiological diagnosis in the presence of raised tumour markers. Ideally, a histological diagnosis is needed before treatment: it may be radiologically difficult to distinguish chronic pancreatitis from carcinoma, and CA19–9 may be spuriously elevated in the presence of obstructive jaundice. Histologically benign disease is found in up to 10% of radical resections. Rarer histological variants like neuroendocrine tumours can be missed if a histological diagnosis is not established. CT, US or EUS-guided FNA are safe procedures, and needle-track seeding is extremely rare.
Staging classification The TNM definitions and staging classification are shown in Tables 12.3 and 12.4, respectively.
Imaging A CT scan is usually used for local staging and to exclude metastatic disease. A dual-phase helical CT scan (arterial phase to show the pancreas and venous phase to look for liver metastasis) has been reported to have a 79% positive predictive value and a 96% negative predictive value in predicting tumour resectibility. 3D reconstruction may give additional information about the involvement of vascular structures. MRI has been shown to be equivalent to a good-quality CT scan but may be limited by cost and availability. The endoscopic retrograde cholangiopancreaticogram is useful in patients presenting with obstructive jaundice. The ‘double-duct’ sign (occlusion of common bile duct and pancreatic duct) is almost diagnostic of pancreatic cancer; brushings obtained during the procedure can establish a malignant diagnosis (although the yield is low) and an endobiliary stent can be placed to relieve biliary obstruction. Endoscopic ultrasound (EUS) is a relatively new test which can be used before surgery along with standard
Treatment overview The prognosis of pancreatic cancer is poor. Patients with localised disease should be discussed at a multidisciplinary meeting that includes a hepato-biliarypancreatic surgeon, a radiologist, an oncologist and a pathologist to decide the most appropriate management. About 20% of patients are resectable at diagnosis with a median survival of about 20 months (Neoptolemos et al., 2004). For patients who are inoperable, stage (locally advanced or metastatic) and performance status are important factors guiding management and overall prognosis. For patients with locally advanced non-metastatic cancer, both chemotherapy alone and chemoradiation are acceptable options with a median survival of about 10 months. Patients with metastatic disease have a poorer prognosis: a median survival of approximately 6 months and only 20% surviving 12 months from diagnosis (Burris et al., 1997). Chemotherapy is appropriate 153
Somnath Mukherjee and Tom Crosby
Table 12.3. TNM definitions of carcinoma of the pancreas Stage
Description
TX
Primary tumour cannot be assessed
T0
No evidence of primary tumour
Tis
Carcinoma in situ
T1
Tumour limited to the pancreas, ≤ 2 cm in greatest dimension
T2
Table 12.4. Stage groups for carcinoma of the pancreas Stage
Description
0
Tis, N0, M0
IA
T1, N0, M0
IB
T2, N0, M0
IIA
T3, N0, M0
IIB
T1, N1, M0
Tumour limited to the pancreas, > 2 cm in
T2, N1, M0
greatest dimension T3
without involvement of the coeliac axis or the superior mesenteric artery T4
T3, N1, M0
Tumour extends beyond the pancreas but III
T4, any N, M0
IV
Any T, any N, M1
Tumour involves the coeliac axis or the superior mesenteric artery (unresectable primary
Adapted from UICC (2002).
tumour) NX
Regional lymph nodes cannot be assessed
N0
No regional lymph node metastasis
N1
Regional lymph node metastasis
MX
Distant metastasis cannot be assessed
M0
No distant metastasis
M1
Distant metastasis
Adapted from UICC (2002).
for patients with WHO performance status 0 to 2. Nutritional support and close collaboration with the palliative care team is needed to ensure good symptom control. Endobiliary stenting (for obstructive jaundice), palliative radiotherapy and a coeliac plexus nerve block (for control of local pain due to retroperitoneal infiltration) are useful measures.
Perioperative mortality and morbidity is inversely related to a surgeon’s case load (Birkmeyer et al., 1999; Neoptolemos et al., 1997) and patients should therefore be diagnosed and treated by a specialist surgical team (NHS Executive, 2001). The surgical complications include delayed gastric emptying, pancreatic fistula, sepsis, haemorrhage, malabsorption and diabetes mellitus.
Palliative surgery In some cases of biliary and/or gastrointestinal obstruction, palliative bypass surgery is required, which entails a choledochojejunostomy and/or gastrojejunostomy.
Chemotherapy and chemoradiation Postoperative adjuvant chemotherapy
Surgery Radical surgery Pancreatico-duodenectomy (Whipple’s procedure) is the standard operation for tumours in the head of the pancreas with an operative mortality of 1 to 16%. Pylorus-preserving pancreatico-duodenectomy may be as effective but can result in delayed gastric emptying (Lin et al., 2005). Patients with cancer of the body or tail present late and, therefore, are less likely to have resectable disease. Distal pancreatectomy is the surgery of choice and total pancreatectomy may be necessary to achieve clear margins for large tumours. 154
Postoperative adjuvant chemotherapy (5-FU with folinic acid, days 1 to 5, every 4 weeks for six cycles) is standard practice in the UK, based on the results of the ESPAC 1 study, which randomised patients to no adjuvant therapy, adjuvant chemotherapy alone, adjuvant chemoradiation or adjuvant chemoradiation followed by chemotherapy. Patients on adjuvant chemotherapy had a significant survival advantage (median survival 20.1 months compared to 15.5 months; p = 0.0009) and chemoradiation was found to be detrimental (median survival 15.9 months compared to 17.9 months; p = 0.05) (Neoptolemos et al., 2004). The radiotherapy arms of the study have been criticised for using a low-dose
Exocrine pancreas
split-course technique, lacking a detailed protocol and providing inadequate quality assurance. The early results from the CONKO 001 study, which randomised patients to gemcitabine or no adjuvant therapy, were presented in ASCO 2005 (Neuhaus et al., 2005). The median disease-free survival in the two arms was 14.2 and 7.5 months, respectively (p ≤ 0.001). ESPAC 3, which randomised patients between 5-FU with folinic acid and gemcitabine, has completed accrual and results are awaited.
Postoperative adjuvant chemoradiation Adjuvant chemoradiation is widely practised in the USA. However, there is only one randomised trial, which includes only 49 patients (Kalser and Ellenberg, 1985), to support its use and this approach is not favoured in Europe or the UK. In an EORTC trial, 218 patients with pancreatic head and periampullary tumours were randomised to either adjuvant chemoradiotherapy (split-course) or surgery alone. There was a small but not statistically significant improvement in survival with combined modality therapy in patients with pancreatic cancer (median survival 24.5 months compared to 19.0 months; 2-year survival 34% compared to 26%; p = 0.099). The trial was underpowered, used what would be considered a suboptimal dose and lacked quality assurance (Klinkenbijl et al., 1999).
Chemotherapy and chemoradiotherapy for locally advanced disease Chemotherapy, alone or in combination with radiotherapy, is an acceptable treatment option for patients with locally advanced non-metastatic disease. Trials comparing radiotherapy and chemotherapy (5-FU) to either treatment alone favour the use of combined-modality therapy, which has a median survival of about 10 months (Gastrointestinal Tumour Study Group, 1988; Moertel et al., 1981). However, such trials were carried out in the 1980s and the reported outcome from recent trials in patients treated with gemcitabine-based chemotherapy alone is similar but with less toxicity (Cunningham et al., 2005; Louvet et al., 2005). The additional benefit of chemoradiation in patients receiving gemcitabine-based chemotherapy is unknown and should be evaluated formally in a randomised study. Gemcitabine is a potent radiosensitiser and there has been a growing interest in its use in chemoradiation. One randomised study showed gemcitabine to be superior to 5FU-based chemoradiation
(Li et al., 2003). Acceptable treatment options are as follow: 1. Gemcitabine weeks 1 to 7, then 1 week off; subsequent cycles weekly, 3 out of 4 weeks. 2. Gemcitabine and capecitabine combination: gemcitabine 1000 mg/m2 weekly and capecitabine 1660 mg/m2 daily, 3 out of 4 weeks. 3. Radiotherapy (45 to 50.4 Gy in 25 to 28 fractions) with infusional 5-FU (200 mg/m2 per day).
Technique for chemoradiotherapy for locally advanced disease As a preplanning measure, patients should have an EDTA clearance and functioning renogram to assess total and individual renal function. CT localisation should be performed with the patient in the supine position with arms raised; intravenous contrast should be used. The planning target volume is the gross tumour volume (GTV) with a 2 cm margin. Three or four coplanar fields and anterior–posterior beams are used; lateral fields are angled anteriorly to reduce the dose to the kidneys. Dose, energy, fractionation are as follow: give 45 to 50.4 Gy in 25 to 28 fractions to the ICRU 50 reference point, 1.8 Gy per fraction, using 10 MV photons, treating all fields daily, Monday to Friday. Maximum and minimum allowable dose within the PTV should be 107 and 95%, respectively, of the dose prescribed at ICRU 50. Portal beam imaging should be taken during the first three days of treatment, and then weekly thereafter, for the purpose of verification. DVHs for critical structures at risk should be obtained and the dose to the critical structures should be considered before accepting the final plan. The following are tolerance doses for organs at risk: r Small bowel – no more than 10% to receive > 45 Gy; the maximum dose to the small bowel is 50 Gy. r Spinal cord – maximum of 45 Gy. r Kidney – one fully functioning kidney should be excluded from treatment if possible; the equivalent of at least 67% of one functional kidney must receive less than 20 Gy. r Liver – 60% of the liver should receive no more than 30 Gy, and the whole liver no more than 20 Gy. Concurrent chemotherapy additionally involves infusional 5-FU at 200 mg/m2 per day throughout radiotherapy. Toxicity of chemoradiotherapy include gastrointestinal: anorexia, nausea and vomiting, dehydration, 155
Somnath Mukherjee and Tom Crosby
gastrointestinal bleeding (may occur a few weeks after the end of chemoradiation) and duodenal stenosis. Haematological toxicity includes anaemia, leucopenia, or thrombocytopenia (especially with gemcitabinebased chemoradiation). General care for patient toxicity should include prophylactic antiemetics (according to local hospital policy) and weekly clinical review.
Palliative chemotherapy for metastatic or inoperable disease Of patients with pancreatic cancer, 80% have locally advanced, inoperable or metastatic disease at presentation. Since its recommendation by NICE in May 2001, single-agent gemcitabine has been the drug of choice for patients with a Karnofsky performance score of 50 or more. One randomised trial (Burris et al., 1997) showed that gemcitabine was more effective than bolus 5-FU in terms of median survival (5.6 versus 4.4 months; p = 0.0025), progression-free survival (2.3 versus 0.9 months; p = 0.0002), and 1-year overall survival (18 versus 2%; p = 0.0002). Trials comparing gemcitabine alone with gemcitabine in combination with a variety of drugs (irinotecan, pemetrexed, exatecan, 5-FU, oxaliplatin, marimastat and tipifarnib) have all failed to show a survival advantage; trials have, however, shown an increase in median survival when gemcitabine is used in combination with erlotinib (6.37 versus 5.91 months; p = 0.025; Moore, 2005) and capecitabine (7.4 versus 6 months; p = 0.026; Cunningham et al., 2005). Acceptable treatment options are as follow: 1. Gemcitabine 1000 mg/m2 days 1, 8 and 15, every 4 weeks or once weekly for weeks 1 to 7, then 1 week off with subsequent cycles weekly for 3 out of every 4 weeks. 2. Gemcitabine as above combined with capecitabine 1660 mg/m2 daily, 3 out of every 4 weeks. 3. Gemcitabine and erlotinib combination (erlotinib is licensed for use but currently not available in the NHS).
Radiotherapy Palliative radiotherapy A short course of radiotherapy may be used to palliate local symptoms such as pain in patients with metastatic disease or in symptomatic patients with locally advanced disease not suitable for chemoradiation.
156
Technique for palliative radiotherapy CT localisation in a CT simulator as described earlier is ideal. However, conventional simulation with the patient supine and given oral contrast is acceptable and may be more practical. The target volume for CT simulation is a GTV with a 2 to 3 cm margin. For conventional simulation (head of pancreas), the target volume is delineated by the superior border or middle of the T11 vertebra; the inferior border of the L3 vertebra; the left border, 2 cm to the left of the vertebral-body edge; and the right border, 5 cm to the right of the vertebral-body edge. Anterior–posterior parallel opposed fields are used. Dose and fractionation involve the administration of 20 Gy in 5 fractions (4 Gy per fraction) or 30 Gy in 10 fractions.
Endoscopic relief of obstruction Obstructive jaundice is a major cause of morbidity in patients with pancreatic cancer. Plastic stents are placed if surgery is planned or if the prognosis is poor. Metallic stents have a lower complication rate (infection, blockage) but are more expensive. Patients should be given ursodeoxycholic acid to help keep the stent patent.
Areas of current interest New chemotherapeutic agents under evaluation Oxaliplatin, capecitabine, irinotecan, orathecin (oral topoisomerase 1 inhibitor) and docetaxel have demonstrated activity in the treatment of pancreatic cancer. Apart from capecitabine, none of the other drugs (alone or in combination) has resulted in better overall survival than single-agent gemcitabine (Cunningham, et al., 2005); a combination of oxaliplatin and 5-FU has a survival advantage compared to active support as a second-line therapy in patients progressing on gemcitabine.
Fixed dose-rate gemcitabine The intracellular effect of gemcitabine is limited by its intracellular phosphorylation. One trial has shown that a regimen using a lower dose rate infusion (1500 mg/m2 given at 10 mg/min) has greater activity than one that uses a higher dose rate (2200 mg/m2 given over 30 minutes) but with more toxicity (Tempero et al., 2003).
Exocrine pancreas
REFERENCES
Table 12.5. Ongoing/planned trials in pancreatic cancer Study group
Trial arms
Phase
CALGB
Gemcitabine ± bevacizumab
3
SWOG
Gemcitabine ± cetuximab
3
ECOG
Docetaxel/irinotecan ±
2
cetuximab University of Chicago NCRN (UK)
Gemcitabine + bevacizumab +
2
(cetuximab or erlotinib) TELOVAX – gemcitabine ±
3
telomerase vaccine
Novel agents Pancreatic cancer cell lines express RAS mutations and mutations in EGF receptors and overexpress receptors for VEGF ligands and secrete VEGF-A. Tipafarnib (a farnasyl transferase inhibitor), cetuximab and erlotinib (EGFR antagonists), and bevacizumab (VEGF antagonist) have all shown activity in phase I and II studies. Erlotinib is licensed for first-line use in conjunction with gemcitabine in advanced pancreatic cancer.
Gemcitabine-based chemoradiation Several phase I and II studies and one phase III study (Li et al., 2003) have shown that gemcitabine-based chemoradiation is a feasible treatment. Both low-dose gemcitabine (200 to 400 mg/m2 weekly, 40 to 50 mg/m2 twice weekly) with full-dose radiation and full-dose gemcitabine with low-dose radiation (39 Gy) have been reported to have acceptable toxicity. The toxicity of such regimens depends on the size of the radiotherapy field as well as the dose and schedule of gemcitabine. A median survival of 6 to 16 months has been reported.
Intensity-modulated radiation therapy (IMRT) IMRT delivers a radiation dose more conformally than conventional 2D or 3D radiotherapy. Compared to conventional radiation, IMRT reduces the mean dose to the liver, kidneys, stomach and small bowel without compromising local control (Milano et al., 2004).
Ongoing/planned trials in pancreatic cancer A list of ongoing or upcoming trials is shown in Table 12.5.
Birkmeyer, J. D., Finlayson, S. R., Tosteson A. N. et al. (1999). Effect of hospital volume on in-hospital mortality with pancreaticoduodenectomy. Surgery, 125, 250–6. Burris, H. A. III, Moore, M. J., Anderson, J. et al. (1997). Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial. J. Clin. Oncol., 15, 2403–13. Cunningham, D., Chau, I., Stoken, D. et al. (2005). Phase III randomised comparison of gemcitabine (GEM) versus gemcitabine plus capecitabine (GEM-CAP) in patients with advanced pancreatic cancer. Proc. ECCO, 13, Abstr. PS11. Gastrointestinal Tumor Study Group. (1988). Treatment of locally unresectable carcinoma of the pancreas – comparision of combined-modality therapy (chemotherapy plus radiotherapy) to chemotherapy alone. J. Natl. Cancer Inst., 80, 751–5. Kalser, M. H. and Ellenberg, S. S. (1985). Pancreatic cancer. Adjuvant combined radiation and chemotherapy following curative resection. Arch. Surg., 120, 899–903. Klinkenbijl, J. H., Jeekel, J., Sahmoud, T. et al. (1999). Adjuvant radiotherapy and 5-fluorouracil after curative resection of cancer of the pancreas and periampullary region: phase III trial of the EORTC gastrointestinal tract cancer cooperative group. Ann. Surg., 230, 776–82. Li, C. P., Chao, Y., Chi, K. H. et al. (2003). Concurrent chemoradiotherapy treatment of locally advanced pancreatic cancer: gemcitabine versus 5-fluorouracil, a randomized controlled study. Int. J. Radiat. Oncol. Biol. Phys., 57, 98–104. Lin, P. W., Shan, Y. S., Lin, Y. J. et al. (2005). Pancreaticoduodenectomy for pancreatic head cancer: PPPD versus Whipple procedure. Hepatogastroenterology, 52, 1601–4. Louvet, C., Labianca, R., Hammel, P. et al. (2005). Gemcitabine in combination with oxaliplatin compared with gemcitabine alone in locally advanced or metastatic pancreatic cancer: results of a GERCOR and GISCAD phase III trial. J. Clin. Oncol., 23, 3509–16. Milano, M., Chmura, S. J., Garofalo, M. C. et al. (2004). Intensity-modulated radiotherapy in treatment of pancreatic and bile duct malignancies: toxicity and clinical outcome. Int. J. Radiat. Oncol. Biol. Phys., 59, 445–53. Moertel, C. G., Frytak, S., Hahn, R. G. et al. (1981). Therapy of locally unresectable pancreatic carcinoma; a randomized comparison of high dose (6000 rads) radiation alone, moderate dose radiation (4000 rads + 5-fluorouracil), and high dose radiation + 5-fluorouracil: The Gastrointestinal Tumor Study Group. Cancer, 48, 1705–10. Moore, M. J. (2005). Brief communication: a new combination in the treatment of advanced pancreatic cancer. Semin. Oncol., 32 (6 Suppl. 8), 5–6. Neoptolemos, J. P., Russell, R. C., Bramhall, S. et al. (1997). Low mortality following resection for pancreatic and periampullary tumours in 1026 patients: UK survey of specialist pancreatic units. UK Pancreatic Cancer Group. Br. J. Surg., 84, 1370–6.
157
Somnath Mukherjee and Tom Crosby
Neoptolemos, J. P., Stocken, D. D., Friess, H. et al. (2004). A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N. Engl. J. Med., 350, 1200–10. Neuhaus, P., Oettle, H., Post, S. et al. (2005). A randomised, prospective, multicenter, phase III trial of adjuvant chemotherapy with gemcitabine vs. observation in patients with resected pancreatic cancer. J. Clin. Oncol. ASCO Annual Meeting Proceedings, 23 (16 Suppl. I), 4013. NHS Executive. (2001). Guidance on Commissioning Cancer Services – Improving Outcomes in Upper Gastro-intestinal
158
Cancers. NHS Executive Catalogue Number 23943. London: Department of Health. Tempero, M., Plunkett, W., Ruiz Van Haperen, V. et al. (2003). Randomized phase II comparison of dose-intense gemcitabine thirty-minute infusion and fixed dose rate infusion in patients with pancreatic adenocarcinoma, J. Clin. Oncol., 21, 3402–8. UICC. (2002). TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 93–6.
13
COLON AND RECTUM Richard Adams, Timothy Maughan and Tom Crosby
Introduction Colorectal cancer (CRC) is second in incidence in Europe only to lung cancer, and it causes around 204 000 deaths each year. The aetiology of CRC is still unclear, but the eight- to ten-fold higher incidence in the developed world compared to that in the developing world suggests environmental causes. Around 15 to 20% of CRCs are of familial origin. Screening is currently being adopted in the UK with the roll-out of a programme of faecal occult blood (FOB) testing, followed by colonoscopy if FOB testing is positive. Many other countries are also considering such a programme. Surgery is the only curative treatment and total mesorectal excision (TME) is now well established as the best way of managing rectal carcinoma. But the last 10 years have also seen a rapid increase in the use of preoperative radiotherapy, of neoadjuvant and adjuvant chemotherapy, and of new agents for advanced disease, with small but incremental improvements in outcome. Targeted therapy such as epithelial growth factor receptor (EGFR) inhibitors and vascular endothelial growth factor inhibitors have been tested in patients with advanced CRC, but the role of these therapies in routine management has not yet been established.
Types of colorectal tumours The range of tumours that affect the colon and rectum is shown in Table 13.1.
Incidence and epidemiology The annual incidence of CRC is 54 per 100 000 in the UK, and 35 000 new cases are reported per year (CRUK; http://info.cancerresearchuk.org, accessed September 2006). More than 16 100 deaths (50% of rectal origin) occur per year in the UK, making CRC the second most common cause of cancer death. CRC accounts for
around 13% of non-skin cancers, and 98% of all cancers in the large intestine are adenocarcinomas. There is a higher incidence of CRC in affluent Western countries and a lower incidence in Japan, Africa and South America. The peak incidence of CRC occurs at ages 60 to 70 years. The male-to-female ratios for colonic and rectal tumours are 2:3 and 2:1, respectively. In general, tumours of the colon outnumber those of the rectum by 3 to 2.
Carcinoma of the colon and rectum Risk factors and aetiology Family history A family history of CRC, particularly in first-degree relatives younger than age 40, is associated with an increased risk, and genetic causes probably account for around 15% of all CRCs. There are two well-recognised inherited cancer syndromes: hereditary non-polyposis colorectal cancer (HNPCC) and familial adenomatous polyposis (FAP). HNPCC (Lynch syndrome) is an autosomal dominant condition involving defective mismatch-repair genes, which allow gene mutations to accumulate. HNPCC probably accounts for approximately 2% of colorectal cancers, and affected gene carriers have an 80% lifetime risk of CRC. HNPCC carries an associated increased risk of endometrial, ovarian, gastric, pancreatic and renal malignancies and has a propensity for right-sided tumours, which are often mucin-producing and less aggressive. HNPCC is divided into two types: r The Lynch I classification describes CRC that is colon site-specific, right-sided, involves multiple tumours and is more often mucinous. r Lynch II CRC involves increased risk of carcinoma of the colon and endometrium, multiple tumours and an early age of onset. The clinical diagnosis of HNPCC is made using Modified Amsterdam Criteria (Vasen et al., 1999) and is then 159
Richard Adams, Timothy Maughan and Tom Crosby
Table 13.1. The range of tumours that affect the colon and rectum Type
Examples
Non-neoplastic
Benign polyps
tumours (no
Hyperplastic polyps
malignant potential)
Hamartomatous polyps: Juvenile polyps and Peutz-Jegher polyps
r Attenuated FAP involves fewer adenomas and a slightly lower risk of colonic cancer.
Conditions associated with colorectal cancer Ulcerative colitis (UC) and, to a lesser extent, Crohn’s colitis are associated with CRC. The risk is greatest in patients who develop early onset UC with a large proportion of the colon affected. These cancers are often multifocal, flat, infiltrating and poorly differentiated.
Inflammatory polyps Lymphoid polyps Neoplastic epithelial tumours
Benign polyps Tubular adenoma Tubulo-villous adenoma Villous adenoma Malignant polyps Adenocarcinoma Carcinoid tumour Anal zone carcinoma
Mesenchymal tumours
Benign Leiomyoma Lipoma Neuroma Angioma Malignant Leiomyosarcoma Liposarcoma Malignant spindle-cell tumour
Other factors Epidemiological studies on the cause of CRC now provide compelling evidence for environmental factors. There is a higher incidence in affluent Western countries and a lower incidence in Southeast Asia, Africa, and much of South America, which is presumably due to dietary factors. Other risk factors include smoking, which results in an increased risk of adenomas but with no proven increase in CRC. Increased body mass index, between 23 and 30 kg/m2 , causes a linear increase in the risk of colon cancer. Possible dietary risk factors include a diet low in indigestible starch, high in refined carbohydrates and fat content, and a decreased intake of protective micronutrients. Increased fruit and vegetable intake results in a protective effect for colon and other cancers as suggested by case-control studies, but cohort studies in the USA provide less evidence for this finding.
Kaposi’s sarcoma Other
Lymphoma
Screening and prevention Primary prevention confirmed by laboratory testing for MSH1 and 2 and PMS1 and 2 mutations. Familial adenomatous polyposis (FAP) is an autosomal dominant condition characterised by defects in the APC gene on chromosome 5. FAP accounts for fewer than 1% of all CRCs. It often manifests as a carpet of adenomatous polyps throughout the large intestine; 90% of affected individuals will have an invasive cancer by the age of 45. There are three variants of FAP: r Gardner’s syndrome involves colorectal adenomatous polyps, osteomas and desmoid tumours. r Turcot’s syndrome involves colonic polyps associated with CNS tumours, including ependymomas and medulloblastoma. 160
The EPIC (European Prospective Investigation into Cancer and Nutrition) study has concluded that diets deficient in fibre are associated with CRC and a doubling of fibre intake could reduce the risk of CRC by 40%. More recent results from this study have suggested that diets high in fish and low in red meat content may protect against CRC (Bingham et al., 2003; Jacobs et al., 1998; Key et al., 2002).
Chemoprevention There is currently no evidence that any drugs are effective in preventing CRC.
Secondary prevention Strategies for secondary prevention are reviewed in a recent article by Agrawal and Syngal (2005).
Colon and rectum
Faecal occult blood (FOB) There have been three large population-based studies in people between the ages of 50 and 80, including one from Nottingham. In the Nottingham trial, there was a 15% reduction in cumulative CRC mortality in the screened group; however, the false-positive rate was high, at around 90%. FOB testing has been launched in areas of the UK with a plan for general phased roll-out in late 2006 (Autier et al., 2003; Hardcastle et al., 1996).
Normal epithelium 5q-mutation / loss FAP Hyperproliferative epithelium DNA hypomethylation Early adenoma
Sigmoidoscopy or colonoscopy Flexible sigmoidoscopy is both sensitive and specific, but it will only pick up abnormalities in the rectum and sigmoid colon, which account for approximately 65% of colorectal carcinomas. Colonoscopy is better at picking up proximal lesions but is more expensive and incurs higher morbidity. These procedures also have a preventative role because polyps, which might be premalignant, may be removed. Polypectomy via sigmoidoscopy or colonoscopy is generally performed on at-risk patients (i.e. those with a positive family history). When used as part of a screening programme, polypectomy has shown a 59% reduction in mortality from bowel cancer.
12p mutation KRAS Intermediate adenoma 18q loss ?DCC Late adenoma 17p loss TP53 Carcinoma
Metastases
Strategies for patients with inflammatory bowel disease Screening with regular colonoscopies is offered to high-risk patients and prophylactic panproctocolectomy is appropriate in selected cases.
Figure 13.1. Adapted Vogelstein model of carcinogenesis for CRC. DCC = deleted in colon cancer; DNA = deoxyribonucleic acid; FAP = familial adenomatous polyposis; KRAS official name = v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog; TP53 = tumour protein p53. See
Pathology The adenoma-carcinoma sequence Vogelstein described a model of transition from normal bowel, through adenomas, to carcinoma (Fearon and Vogelstein, 1990). This multistage concept of carcinogenesis includes mutation or deletion of the tumour suppressor gene adenomatous polyposis coli (APC), followed at some stage by mutation in the KRAS protooncogene. Then the deleted in colon cancer (DCC) tumour suppressor gene probably helps to initiate metastatic potential and, eventually, TP53 gene mutation and DNA microsatellite instability (MSI). Vogelstein’s model of carcinogenesis is shown in Figure 13.1 and it defines the pathway to chromosomal instability. A separate pathway characterised by MSI occurs in HNPCC along with an acquired loss of DNA mismatch repair (Rowan et al., 2005). The risk of cancer varies with the size and number of polyps:
Fearon and Vogelstein (1990).
r Polyps ≤ 1 cm carry a ≤ 1% risk of invasive malignancy. r Polyps between 1 and 2 cm carry a 10% risk. r Polyps greater than 2 cm carry a 50% risk. r 3% of CRCs are multicentric. Features associated with an increased risk of malignant potential within colonic polyps are shown in Table 13.2.
Morphology The morphology of colorectal adenocarcinoma is shown in Table 13.3.
Spread Carcinoma of the colon and rectum can spread to adjacent structures such as the small or large bowel, the bladder, the uterus and so on. Transcoelomic spread occurs with diffuse peritoneal disease. 161
Richard Adams, Timothy Maughan and Tom Crosby
Table 13.2. Features associated with an increased risk
Table 13.3. Pathological features of colorectal
of malignant potential within colonic polyps
adenocarcinoma
Risk category
Feature
Features
Description
High risk
Large size (> 1.5 cm)
Macroscopic
Polypoid, fungating, ulcerative or diffusely infiltrating
Sessile or flat
Proximal tumours often grow as polypoid
Severe dysplasia present
or fungating masses extending along
Squamous metaplasia present
one wall of the caecum or colon
Villous architecture present
Carcinomas in the distal colon tend to be
Multiple polyps Low risk
annular, stenosing tumours
Small size (≤ 1 cm) Pedunculated
Microscopic
Well, moderately or poorly differentiated
Mild dysplasia present
Spectrum from tall columnar cells akin to
No metaplasia present
those found in benign polyps but with
Tubular architecture present
evident invasion to frankly anaplastic
Single polyp
carcinomas Strong desmoplastic reactions are common, giving macroscopic firmness
Regional lymph nodes are involved in 40 to 70% of cases at operation; the lymph node chain usually follows the supplying blood vessels (for rectal cancers these will be pararectal lymph nodes, nodes at the bifurcation of the inferior mesenteric artery, hypogastric nodes and presacral nodes). Haematogenous spread is most often to the liver, lungs and bone, occurring in that order of frequency, but many other sites are possible, including the brain. In 25 to 30% of patients at presentation, the tumour will have spread either locally or to distant sites and will be unsuitable for radical treatment.
Clinical presentation
Many tumours produce mucin; rarer mucinous tumours produce sheets/pools of mucin, which carry a poor prognosis 10% of adenocarcinomas of the colon have neuroendocrine differentiation; small-cell undifferentiated tumours are of neuroendocrine origin A few have signet ring appearances more commonly found in stomach or breast carcinomas Distal colon tumours occasionally have foci of squamous differentiation and are known as adenosquamous carcinomas
Symptoms and signs from the primary tumour Symptoms that occur more commonly in rightsided lesions include unexplained anaemia, ill-defined abdominal pain, an abdominal mass, weight loss and rectal bleeding with altered blood. More common in left-sided lesions are a change in bowel habit, namely diarrhoea or constipation, obstruction, rectal bleeding, tenesmus and mucoid discharge. Rectal examination can identify 75% of rectal tumours.
Criteria for referral to a specialist Symptoms that require urgent referral to a specialist have been reviewed (NHS Executive, 2000). They include, for all ages: r A palpable right-sided, abdominal mass. 162
r A palpable rectal mass. r Rectal bleeding with a change in bowel habit to more frequent defecation or looser stools (or both) that is persistent over 6 weeks. r Iron-deficiency anaemia (haemoglobin concentration ≤ 11 g/dl in men or ≤ 10 g/dl in postmenopausal women) without obvious cause. Additional urgent symptoms in patients over age 60 include: r Persistent rectal bleeding without anal symptoms. r A change of bowel habit to more frequent defecation or looser stools (or both), without rectal bleeding, and persistent for 6 weeks.
Colon and rectum
Table 13.4. Dukes’, TNM and AJCC staging classifications shown with approximate 5-year survival by stage Dukes’ stage (modified)
A
Five-year Description
Stage (AJCC)
TNM
survival (%)
In situ
0
Tis N0 M0
100%
Cancer confined to submucosa or muscularis
I
T1 N0 M0
95%
propria but not through it B(1)
Into but not beyond muscularis propria; no LN
B(2)
Through the muscularis propria with no nodes
C(1)
Nodes positive but not the apical node
II
spread
T2 N0 M0
90–95%
T3 N0 M0
70–80%
T4 N0 M0
involved III
Any T,
40%
N1–2, M0 C(2)
Apical node positive
D
Metastatic
IV
Any T,
5%
Any N, M1
Investigation Staging investigations A full blood count and liver function tests should be performed; 65% of the liver can be involved before an abnormality is detected. CEA is raised in 85% of patients with CRC. Higher CEA values are associated with a worse prognosis when measured preoperatively but they do not change clinical management. Colonoscopy or double-contrast barium enema are used because 5% of lesions are synchronous. Biopsy is essential because of the wide differential diagnosis: inflammatory bowel disease/colitis (ischaemic or inflammatory), infective diarrhoea, diverticular mass or abscess. A CT scan of the thorax and abdomen or CXR and US of liver can be performed. Pelvic MRI should be performed for patients with rectal tumours, to assess the likely circumferential resection margin (CRM) and the need for preoperative therapy (see discussion that follows). Endo-anal US gives good information on size, penetration and nodal involvement of low rectal tumours and may be appropriate, especially if local excision of a rectal T1 lesion is contemplated. A PET scan is only relevant when resection of liver metastases is considered.
be accurate in over 90% of cases (can miss microscopic T3 disease). Once again, nodal assessment is relatively poor (accurate in 50 to 80%). Recent studies have suggested that MRI can be used to identify the mesorectal fascia (the only investigation able to visualise this plane used in TME dissection) and to predict clearance at the CRM. The MERCURY study showed concordance between MRI and histological findings in the spread of tumour beyond the lamina propria and, therefore, the ability to predict the threat of tumour to the CRM (MERCURY Study Group, 2006).
Pathological staging of colorectal cancer Dukes’ staging has undergone various modifications and was originally used to describe rectal cancers, but it is now also accepted for colonic cancers and is shown in Table 13.4. The TNM classification is shown in Table 13.5.
Treatment for non-metastatic disease: colonic carcinoma The general principle of non-metastatic colon cancer management is to obtain staging information, resect the tumour if possible and then consider the patient for adjuvant (or palliative) therapy.
Local staging of rectal cancer Digital rectal examination (DRE) can be accurate for T staging in up to 80% of cases, but it is poor at assessing nodal disease. EUS is very user-dependent but it can
Surgery Surgery is an essential part of curative treatment for colon cancer. Whenever possible, patients should be 163
Richard Adams, Timothy Maughan and Tom Crosby
Table 13.5. TNM classifications of CRC
Dukes’ C and to selected patients with Dukes’ B colon cancer. See the Appendix for a list of different adjuvant chemotherapy treatments (p. 173).
Stage
Definition
T1
Tumour invades submucosa
T2
Tumour invades muscularis propria
Fluoropyrimidines All adjuvant regimens are based on
T3
Tumour invades through muscularis propria and
fluoropyrimidines. In the 1990s the following guidelines were established: r Low-dose folinic acid (FA) is just as effective as highdose FA (Anonymous, 2000). r Levamisole is ineffective (Haller et al., 2005). r When using a standard 5-FU and FA regimen, there is no advantage to extending the duration of therapy past 6 months (O’Connell et al., 1998). r Infusional (de Gramont) chemotherapy is no more effective than bolus 5-FU/FA in the adjuvant setting. r All 5-FU-based regimens are associated with a risk of toxicity, notably diarrhoea, stomatitis and leucopenia. r The weekly bolus 5-FU and FA regimen is less toxic than the conventional 5-day 4-week (Mayo) regimen, but in a non-randomised comparison in the QUASAR trial the 5-FU/FA regimen showed no reduction in effectiveness (Kerr et al., 2000) and has become the standard bolus 5-FU regimen in the UK. r Neutropenia is more prevalent in patients over age 70 but they are just as likely to complete treatment as younger patients and they have a similar survival benefit from treatment (Lichtman, 2004).
into subserosa or into non-peritonealised pericoloic or perirectal tissue T4
Tumour directly invades other organs/structures
N1
1–3 lymph nodes involved
N2
4 lymph nodes involved (12 should be examined)
M1
Presence of metastases
and/or perforates visceral peritoneum
Adapted from UICC (2002).
operated on by the specialist team. This means that patients presenting as an emergency with obstruction or perforation should be resuscitated and first have a defunctioning colostomy, followed by elective removal of the tumour. Stenting an obstructing left-sided colonic or rectal tumour can also be useful in an emergency (Athreya et al., 2006). Definitive surgery involves removing the appropriate bowel segment with its mesentery, vascular pedicle and draining lymph nodes.
Adjuvant chemotherapy for colon cancer T1 or T2, N0 tumours Adjuvant chemotherapy has shown no benefit in patients with T1 or T2, N0 tumours.
T3 or node-positive tumours Two recent metaanalyses have shown that prolonged use of a 5FU-based regimen for longer than 3 months can improve survival in colorectal cancer. For patients with node-positive colon cancer, the benefit is about 6% at 5 years (range 2 to 10%; NHS Executive, 1997). For patients with node-negative colon cancer, the increase in overall survival is 1 to 5% after a median follow-up of 4.2 years (Gray et al., 2004), although it remains controversial whether all such patients should be offered adjuvant therapy. There are certain poor-risk features identifiable in node-negative patients that may be used to select those suitable for adjuvant treatment – serosal involvement, perforated tumours, extramural vascular invasion and involvement of the CRM – although these selection criteria have not yet been prospectively verified. Current standard practice in the UK is to offer adjuvant chemotherapy to patients with 164
Oral fluoropyrimidines Capecitabine is as effective as and less toxic than the Mayo clinic regimen (X-ACT trial; Twelves et al., 2005) and is useful in patients with poor venous access. Another oral fluoropyrimidine, tegafururacil, also has a proven role in adjuvant therapy for rectal cancers.
Oxaliplatin as adjuvant Trials comparing 5-FU and FA plus oxaliplatin versus 5-FU and FA alone have both shown an improvement in disease-free survival at 3 years of about 6% (MOSAIC and NSABP C-07 trials; Andre et al., 2004; Wolmark et al., 2005), but with a greater risk of neuropathy. Oxaliplatin with a fluoropyrimidine has therefore become the standard adjuvant therapy in patients who are fit to receive this more toxic regimen and whose risk of relapse is high, such as those with T4, heavily node-positive cancers.
Irinotecan as adjuvant Three trials have shown no benefit from adding irinotecan to adjuvant 5-FU and
Colon and rectum
FA in colon cancer (Saltz et al., 2004; Van Cutsem et al., 2005; Ychou et al., 2005).
Adjuvant online Data from clinical trials in colon cancer can be used to help estimate the benefit of adjuvant chemotherapy attributable to oxaliplatin plus fluoropyrimidine combination therapy through the ‘adjuvant on-line’ service (adjuvantonline, password required; https://www.adjuvantonline.com/colon.jsp).
the mesorectum, and it reduces the risk of an involved resection margin (defined as ≤ 1 mm; MacFarlane et al., 1993). In the pre-TME era, even when the surgeon felt macroscopic clearance had been achieved, 36% of patients were CRM +ve and had a 64% chance of local recurrence. The patients with CRM −ve surgery had a local recurrence rate of only a 9% (Adam et al., 1994).
Early stage rectal cancer Treatment for non-metastatic disease: rectal carcinoma Patients whose tumours are predicted to be likely to recur after surgery alone can be selected for preoperative therapies. The risk of local recurrence following surgery is related to the CRM and the nodal status. Based on the results of the MERCURY study described earlier, MRI can be used to identify three prognostic groups based on stage, the predicted relationship of the tumour to the CRM, and the lymph node status. The risk categories are not universally agreed but can be considered to be r Favourable: rT1 to rT3a (minimal extension into rectum), N0. r Unfavourable: > rT3a (significant mesorectal contamination), or N +ve with margin not at risk. r Advanced: T4 or CRM ≤ 1 mm. Retrospective series have shown that low rectal cancers with poor prognostic features (G3, LVI +ve, T2, R1 resection) do worse with local excision (LE) or abdomino-perineal resection (APR), which is consistent with the findings of phase III clinical trials such as the Dutch Rectal Cancer Trial and CR07.
Surgery Surgery is technically more difficult for rectal than colonic tumours because the pelvis is narrow, especially in the male. Very low tumours require an APR. For tumours that are higher, the fact that they do not tend to spread distally means an anterior resection is possible. Anterior resection is made easier by the use of a staple gun to form the anastomosis. Low anterior resection is associated with increased morbidity including bladder and sexual dysfunction and increased bowel frequency than for patients with higher rectal tumours. TME has become a standard procedure for rectal cancer surgery. This technique, which has been advocated for years by Heald, involves sharp dissection of
A local excision is an alternative to APR in low rectal cancers that are unlikely to have lymph node metastases. In practice this procedure is for well-differentiated T1 tumours: r Low risk (≤ 10% chance of nodal disease) – T1, G1. r Intermediate risk (10 to 20%) – T2, G2. r High risk (> 30%) – G3, T2 or T3, lymphovascular invasion. The most common approach to local excision is transanal endoscopic microsurgery (TEM). In experienced surgical centres local excision for low-risk T1 tumours can be associated with excellent results (Willett, 1999). Very few data exist that support the use of postoperative chemoradiotherapy after local excision alone. Patients with adverse risk factors after local excision should be considered for further resection.
Adjuvant chemotherapy for rectal cancer The clinical benefit of adjuvant chemotherapy alone for patients with rectal cancer has been difficult to prove because many trials have assessed the combination of chemotherapy and radiotherapy together. However, the meta-analysis of trials of adjuvant chemotherapy for CRC showed a greater benefit in rectal cancer than in colon cancer (OR for mortality = 0.64; 95% CI = 0.48– 0.85) and estimated the benefit to be a 9% increase in survival at 5 years for rectal cancer patients (Dube et al., 1997). It is currently recommended that patients with rectal cancer be given the same adjuvant systemic therapy as those with colon cancer, stage for stage, but this is still the subject of ongoing clinical trials.
Radiotherapy and chemoradiotherapy in rectal cancer There is good evidence for the use of either shortcourse preoperative radiotherapy (25 Gy in 5 fractions) or longer-course chemoradiotherapy (45 to 50.4 Gy in 25 to 28 fractions), both given to a planned three- or four-field volume. 165
Richard Adams, Timothy Maughan and Tom Crosby
A recent German study compared long-course preand postoperative chemoradiotherapy and showed an improvement in local control rate (13 versus 6%) in those given preoperative treatment (Sauer et al., 2004). There was also more acute and late grade 3 or 4 toxicity in the postoperative group (acute, 40 versus 27%; late, 24 versus 14%). Preoperative radiotherapy schedules have become standard practice in the UK.
Preoperative radiotherapy and chemoradiotherapy An overview of preoperative radiotherapy trials showed a significant reduction in the odds of death or local recurrence (OR = 0.74, p = 0.03 and OR = 0.49, p ≤ 0.001, respectively; Camma et al., 2000). A metaanalysis published in 2001 showed that local recurrence is reduced from 17 to 10%, whereas for trials in which a radiobiologically equivalent dose (RBE) of more than 30 Gy was given, recurrence was reduced from 20 to 9.8% (Colorectal Cancer Collaborative Group, 2001). Within this group, early trials used a two-field technique and were associated with a risk of cardiovascular mortality; this effect was not seen in later trials, which used a threeor four-field technique. For those trials that employed higher radiation doses (RBE > 30), there was a marginal survival advantage (56.5 versus 58.9%; p = 0.04). Long-term follow-up (median 13 years) was provided by the Swedish Rectal Cancer Trial, which showed an 8% improvement in overall survival (38 versus 30%; p = 0.008) and a reduction in local recurrence from 26 to 9% (p = 0.001; Folkesson et al., 2005), but there was an increased risk of short-term morbidity (perineal wound infection/non-healing) and late morbidity (increased stool frequency and incontinence of loose stools, urgency and emptying difficulties; Birgisson et al., 2005). The Dutch TME trial looked at preoperative shortcourse radiotherapy in more than 1800 patients undergoing TME surgery. There was no survival benefit, but the local recurrence rate was significantly lower in patients who had radiotherapy (11.4 versus 5.8%; Marijnen et al., 2005). The MRC CR07 trial randomised 1350 rectal cancer patients to preoperative radiotherapy (25 Gy in 5 fractions) followed by surgery, or surgery followed by selective postoperative chemoradiotherapy for CRM-positive tumours (95% had TME). This trial showed 10% CRM positivity with a 60-day postoperative mortality of 1.3%. The local relapse rates reduced from 17 to 5% overall at 5 years in favour of preoperative radiotherapy, and the effect was seen in patients with CRM −ve tumours. Node-positive patients who 166
had CRM-negative tumours and who did not receive postoperative radiotherapy had a high local relapse rate of 17% at 5 years. Such patients should be considered for postoperative chemoradiotherapy if no neoadjuvant therapy has been given (Sebag-Montefiore et al., 2006).
Postoperative radiotherapy A meta-analysis of postoperative radiotherapy trials in rectal cancer indicates an effect on local recurrence, but the benefit is smaller than for preoperative radiotherapy (18.6 versus 13.3%). Once again there was no significant effect on either rectal-cancer-specific or overall survival (Colorectal Cancer Collaborative Group, 2001). In the UK, postoperative radiotherapy is reserved for patients who are found to have CRM involvement at surgery and who did not receive preoperative radiotherapy. The role of postoperative radiotherapy in node-positive patients is less certain.
Practicalities of radiotherapy for rectal cancer Patient position and immobilisation are as follows for the external beam radiotherapy technique. The patient lies prone with toes together, heels apart, head resting on the hands, and a comfortably full bladder to displace the small bowel. Either CT simulation or orthogonal plain X-ray films are taken, A–P and lateral, with localisation marks (tattoos posterior and two laterals) and rulers. Barium may be instilled into the rectum to assist in localisation and a radio-opaque marker placed at the anal margin. Oral contrast Gastrografin may be taken orally 1 hour before simulation to delineate small bowel volume within the field. The target volume includes the primary tumour, local lymphatics and the presacral region: r The superior border is defined as the superior aspect of the sacrum or 3 cm above the upper limit of the macroscopic tumour, whichever is highest. r The inferior border is 3 to 5 cm below the lower limit of macroscopic tumour. The anal canal is only included in low rectal tumours and perineal scar in postoperative AP resections. r The lateral borders are the pelvic side walls and internal iliac regions, 1 cm lateral to the pelvic brim or 3 cm to the most lateral portion of macroscopic disease, whichever is the most lateral. r The posterior border encompasses the presacral nodes and sacral hollow. r The anterior border is 2 to 3 cm anterior to the macroscopic tumour or anastomosis but at least to an area 2 cm anterior to the sacral promontory in a horizontal
Colon and rectum
plane. In women, the anterior border should include the posterior vaginal wall. Practically this includes the posterior two-thirds of the femoral heads or comes to the posterior symphysis pubis in T4 lesions. Lead shielding may be applied to the posterior sacrum and anterior to the anal margin (if not a low tumour). An outline of the patient is taken, if not CT simulated, and a plan is produced by the Department of Physics. The plan is a three-field plan with one posterior and two lateral fields. Typically 45◦ wedges are used on the two lateral fields. Field sizes are typically 12 to 15 cm (S–I) × 14 to 17 cm for the posterior fields, and 12 to 15 cm (S–I) × 10 to 12 cm for the lateral fields. Dose, energy and fractionation procedures are as follow: r For preoperative short-course radiotherapy without chemotherapy, the patient receives 25 Gy in 5 fractions over 1 week, prescribed to the ICRU reference point, using 6 to 10 MV photons, followed by surgery at around 1 week; r Or, for preoperative long-course chemoradiotherapy, the patient receives 45 Gy in 25 fractions over 5 weeks, prescribed to the ICRU reference point, using 6 to 10 MV photons, followed by surgery at 6 weeks. r The postoperative dose is 45 to 50 Gy in 25 fractions over 5 weeks. r In case of residual disease in postoperative treatment, a small-volume boost of 10 to 15 Gy can be given in 5 to 7 fractions. r The total dose should be 55 to 60 Gy over 6.5 to 7 weeks to a patient with macroscopic disease. r Inoperable or recurrent disease can be dosed with 45 to 50 Gy in 25 fractions; consider a boost as just mentioned to maximise local control. Be careful if patients have adhesions, inflammatory bowel disease, diverticular disease or diabetes mellitus. The indication for treatment should be reviewed and treatment fields should be kept as small as possible to reduce toxicity. Lasers should be used to align the patient; treat all fields isocentrically each day, set to the ICRU 50 reference point. Portal beam imaging should be taken on the first day of treatment. Concurrent chemotherapy in combination with longcourse radiotherapy involves continuous infusion of 5-FU at 200 mg/m2 per day throughout radiotherapy treatment; alternatively, dose for days 1 to 5 and days 29 to 34: should be 5-FU 350 mg/m2 per day i.v. bolus and FA 20 mg/m2 per day i.v. bolus.
Though an unlicensed indication, there is also evidence from phase II and ongoing phase III studies for the use of capecitabine at a dose of 825 mg/m2 b.d., 7 days per week for the 5 weeks of radiotherapy. When explaining the procedure to the patient undergoing radiotherapy or chemoradiotherapy, give written information and refer the patient to a specialist nurse counsellor if one is available. Explain the procedure and its possible side effects: r Acute: the patient may experience tiredness, diarrhoea, and cystitis, and patients with low tumours may get a severe and painful perineal reaction. Myelosuppression is rare. r Late: menopause (in females) and infertility/sterility, impotence, bowel dysfunction and urge incontinence may occur. There is a small risk of long-term bladder effects and risk of small bowel obstruction. The following information should be used to support the patient: r For diarrhoea, give imodium and advise a low-residue diet. r For cystitis, send MSU, treat the infection, and encourage the patient to drink fluids. r For skin reaction, apply aqueous cream topically b.d., or 1% hydrocortisone cream if severe; apply Instillagel® to areas of moist desquamation (low rectal tumours). r For anaemia, red cell transfusion can be performed (keep Hb > 10.0 g/dl). r For neutropenia, treat the patient as per local neutropenic sepsis protocol. r For nausea, give the patient metoclopramide.
Treatment for advanced/inoperable local disease: colon and rectal carcinoma Inoperable primary disease Inoperable primary disease is most common in the rectum. There is no evidence that palliative resection improves survival. In some patients, clinically unresectable tumours may become operable after radiotherapy. Radiotherapy with or without chemotherapy can give useful palliation of pain and control of local disease. In a patient with an inoperable obstructing rectal cancer, a defunctioning colostomy may provide useful palliation but will not relieve tenesmus, bleeding or mucous discharge. Trans-anal tumour ablation using laser, electrocoagulation or resectional techniques may provide better palliation and should be considered in selected cases. 167
Richard Adams, Timothy Maughan and Tom Crosby
Palliative radiotherapy Patients with rectal cancer who are medically unfit for surgery, and who have symptoms of local pain, discharge, or bleeding, may be treated with palliative external beam therapy to the pelvis, either as a single fraction of 8 Gy or fractionated to 36 Gy in six fractions over 6 weeks. Fitter patients may benefit from a higher dose of 45 Gy in 25 fractions. Palliative radiotherapy is also beneficial for bony metastases or local bony erosion (8 Gy in a single fraction). Patients with fixed caecal tumours may benefit from local radiotherapy (8 Gy in a single fraction), particularly if they have bleeding and recurrent anaemia.
Endoscopic therapy Stenting of colonic and high rectal tumours is feasible and effective either for immediate relief before radical surgery or as a purely palliative procedure. Mid and low rectal tumours can be complicated by stent migration and tenesmus.
Palliative surgery Even in the palliative setting, excision of a symptomatic tumour and re-anastomosis is preferable to a palliative bypass because excision is associated with a better survival rate and quality of life. However, this may not be possible. In very unfit and frail patients some surgeons will attempt to debulk rectal tumours trans-anally, a procedure which has little morbidity.
Locoregional recurrence The 3-year survival of patients with locoregional recurrence of CRC is about 10%. Treatment with chemoradiotherapy should be considered if it has not already been given. Some evidence exists that resection of locally recurrent cancer may improve survival but this has not been proved in a randomised trial. Preoperative chemoradiation before surgical resection of recurrent disease has increased resectability rates by up to 60%. Symptom relief occurs in around 70 to 80% of patients treated with radiotherapy but the median duration of relief is only 3 months.
Treatment for metastatic disease: colonic and rectal carcinoma Surgery for isolated liver metastases The liver and lung are the most common sites for metastatic CRC. Systemic treatment is usually the only 168
option although, for a few patients, surgical excision of metastases or in situ destructive therapy may be feasible. Around 15 to 20% of patients with liver metastases may benefit from primary resection and, with careful patient selection, hepatectomy can achieve a 5-year survival of 40 to 50%. A retrospective review of 2040 patients with metachronous isolated hepatic metastases showed that those who underwent resection had a mean survival of 31 months (projected 5-year survival of 26%) whereas those patients who did not undergo resection had a mean survival of 11 months (projected 5-year survival of 2%; p ≤ 0.0001; Wade et al., 1996). The survival rate is worse for patients presenting with metastatic disease that is synchronous with their primary tumours (compared with a prolonged progression-free interval), those with a heavy positive lymph node burden with the primary tumour, those with a higher number of liver metastases, and those with a higher CEA (Fong and Salo, 1999). Patients should not generally be considered for surgery if there is disease outside the liver, all the hepatic veins are involved or not enough viable liver tissue will be left after resection. Some patients with solitary pulmonary metastases may also be suitable for resection. Case series of selected patients have reported long-term survival, even in patients with liver metastases undergoing staged removal of metastases. Such patients should be selected on an individual basis with involvement of the specialist lung multidisciplinary team.
Neoadjuvant chemotherapy for metastatic disease Non-randomised evidence supports the use of preoperative chemotherapy prior to resection in patients with potentially operable liver metastases but which are considered inoperable at presentation (Adam et al., 2004; Bismuth et al., 1996). Progression of disease while the patient is on chemotherapy is a poor prognostic factor.
In situ destructive therapies for liver metastases Interstitial laser ablation, cryotherapy and radiofrequency ablation have been used for isolated colorectal liver metastases, but there is no evidence to support routine use of these therapies.
Palliative chemotherapy A number of agents have demonstrated activity in metastatic CRC: the fluoropyrimidines including 5-FU, oral prodrugs capecitabine and combination UFT; the
Colon and rectum
chemotherapy agents oxaliplatin, irinotecan, raltitrexed, and mitomycin; and the biological agents bevacizumab, cetuximab and panitumumab. Median overall survival has improved, from 6 to 8 months for patients managed with supportive care alone, to 10 to 14 months for those treated with fluoropyrimidines alone, to 22 to 24 months for patients receiving combination regimens including irinotecan, oxaliplatin and bevacizumab.
Palliative chemotherapy: fluoropyrimidines A meta-analysis of five trials of palliative chemotherapy has shown an improved survival with chemotherapy compared with best supportive care (p = 0.0002). The evidence suggests that, for patients with advanced disease, early chemotherapy before clinical deterioration improves survival by 3 to 6 months without any adverse impact on quality of life (NHS Executive, 1997). Fluoropyrimidine monotherapy is used for patients who are unfit for combination chemotherapy or who have low-volume, slowly progressive metastatic disease. However, the standard therapy for fit patients is now combination chemotherapy with a fluoropyrimidine and either oxaliplatin or irinotecan (NICE, 2005). The optimal way of delivering 5-FU is by infusion over a minimum of 24 hours. The oral prodrug capecitabine is also approved for monotherapy and has been shown to be less toxic than the Mayo Clinic regimen, which is now no longer used. The use of capecitabine has not been compared to optimal infusional 5-FU regimens. The optimum duration of treatment for chemotherapy is not known. A standard approach in the UK is to give chemotherapy for 12 weeks followed by a break, and then start again when there is evidence of progressive disease. This practise is based on evidence from the MRC CR06 study (Maughan et al., 2002) and the OPTIMOX studies (Tournigand et al., 2006). The MRC CR06 trial showed a non-statistically significant trend towards superior results in patients who receive intermittent rather than continuous chemotherapy. This is being further explored in the MRC COIN and AEROMC04 trials. Experience is required in deciding whether patients should be given chemotherapy. Poor performance status, low serum albumin, high alkaline phosphatase and liver involvement are independent predictors of progression; low serum albumin, high γ -glutamyl transferase and high CEA predict for poor survival. Performance status is a particularly strong indicator. In a meta-analysis of patients treated in trials of 5-FUbased chemotherapy, median survivals were 4, 10 and
14 months for patients with ECOG performance status scores of 2, 1 and 0, respectively (Thirion et al., 1999).
Palliative chemotherapy: irinotecan and oxaliplatin There is evidence to support the use of irinotecan and oxaliplatin as a first-line treatment in combination with 5-FU. NICE guidance 93, issued in August 2005, states that ‘irinotecan and oxaliplatin are recommended as possible treatments for people with advanced colorectal cancer, if they are used in these ways: r Irinotecan is given with 5-fluorouracil and folinic acid to people who have not had chemotherapy for advanced colorectal cancer before, or given on its own to people who have already had chemotherapy. r Oxaliplatin is always given with 5-fluorouracil and folinic acid.’ Meanwhile, the same guidance stated that ‘raltitrexed is not recommended for people with advanced colorectal cancer, unless they are taking part in a clinical trial’ (NICE, 2005). The sequencing of chemotherapy has been investigated in a number of trials, including the MRC CR08 (FOCUS) trial (Maughan, 2005; Tournigand et al., 2006). There is no evidence that any particular sequence of oxaliplatin, irinotecan, and fluoropyrimidine is superior to another.
Regional chemotherapy Hepatic arterial infusion of chemotherapy, especially 5FU in metastatic disease or via portal vein infusion for adjuvant therapy, were both evaluated in the 1990s but have not been shown to be effective (James et al., 2003).
Monoclonal antibodies At least two monoclonal antibodies are effective when given with standard chemotherapy regimens in metastatic CRC: bevacizumab and cetuximab. Bevacizumab is a monoclonal antibody for use against the vascular endothelial growth factor ligand and is licensed for first-line metastatic use. One trial of 813 patients randomised to first-line irinotecan and fluoropyrimidine with or without bevacizumab found that response rates increased from 34.8 to 44.8% and median survival increased from 15.6 to 20.3 months in the firstline setting (Hurwitz et al., 2004). Cetuximab is an EGFR monoclonal antibody licensed for use with irinotecan in second- or subsequent line treatment of metastatic CRC. When used as a combination therapy, cetuximab plus irinotecan has been found to double the response 169
Richard Adams, Timothy Maughan and Tom Crosby
rate, increasing time to disease progression and trending towards a survival advantage (Cunningham et al., 2004). In the UK, which has yet to adopt either of these monoclonal antibodies as standard therapy, it is hoped that trials such as the large (> 2400 patients) phase III COIN trial will give further evidence for the combination of oxaliplatin, 5-FU and cetuximab, because small phase II trials that used this combination have shown response rates of up to 81% (Van Cutsem et al., 2004). Panitumumab, a human anti-EGFR monoclonal antibody has been licensed as monotherapy for chemorefractory disease.
Follow-up for colorectal carcinoma after radical therapy There is controversy over the intensity of follow-up that should be used after resection of a primary colorectal carcinoma and this question is being addressed in an ongoing UK trial (FACS) that looks at options of intensive surveillance with hospital clinical and radiological surveillance to GP-based follow-up. An overview by Renehan et al., which included 5 trials but only 1342 patients, showed a significant benefit from intensive follow-up, generally every 3 months for the first year and subsequently spreading out to every 6 months or annually (Renehan et al., 2002). The evidence pointed towards a reduction in cancer-related mortality of 9 to 13% with an average earlier pickup rate of metastatic disease of 8.5 months.
Prognosis of colorectal carcinoma The 5-year survival by stage is shown in Table 13.4.
Areas of current interest No consistent predictive molecular or chromosomal biomarkers has been found to date for prognostic or therapeutic gain. For colon cancer, ongoing trials in the UK and abroad are evaluating the additional benefit of adding biological therapies to chemotherapy, including QUASAR 2 (capecitabine ± bevacizumab), AVANT (FOLFOX ± bevacizumab) and FFCD-PETACC-8 (combination chemotherapy ± cetuximab). In adjuvant therapy for colon cancer, oral fluoropyrimidines probably exert a toxicity profile that differs slightly compared to weekly bolus 5-FU and FA. Toxicity, cost and patient preference must be taken into account 170
when considering which therapy is most appropriate; this is being further examined in the UK PACT trial. For adjuvant chemotherapy for rectal cancer, the ongoing Cancer Research UK-sponsored CHRONICLE trial will attempt to define the benefit of adjuvant chemotherapy in those patients who have received neoadjuvant chemoradiotherapy. New biological therapies are being studied to treat metastatic disease. Another area of interest is in evaluating the role of ultra-small-particle iron oxide–assisted MRI, which may enhance the positive predictive value of nodal involvement in rectal cancer.
Current clinical trials All NCRN trials can be found at www.ncrn.org.uk/ portfolio/dbase.asp? A list of global trials is available at www.cancer.gov/clinicaltrials. The CAPP2 study is a randomised controlled trial of colorectal polyp and cancer prevention using aspirin and resistant starch in carriers of HNPCC (Lynch syndrome). The CHRONICLE study tests adjuvant chemotherapy (capecitabine/oxaliplatin) in patients with completely resected locally advanced rectal cancer who had received preoperative long-course chemoradiotherapy. The COIN trial is a recently launched trial that compares continued chemotherapy plus the EGFR antibody cetuximab versus intermittent chemotherapy with standard palliative combination treatment using oxaliplatin and a fluoropyrimidine in first-line treatment of patients with metastatic CRC. FACS is a randomised controlled trial that assesses the cost-effectiveness of intensive versus no scheduled follow-up in patients who have undergone resection for CRC with curative intent. The FOCUS 2 trial is a randomised trial to assess the role of irinotecan and oxaliplatin in advanced incurable CRC for less fit patients. Genetic Factors in Colorectal Cancer studies the role of genetic factors in clinical outcome for CRC patients. NSCCG stands for the National Study of Colorectal Cancer Genetics. OxaliCap-RT integrates intravenous oxaliplatin plus oral capecitabine with pelvic radiation for rectal cancer. PACT, or Patient Preferences in Adjuvant Colorectal Cancer Therapy, is a randomised crossover clinical trial comparing bolus fluorouracil/leucovorin to capecitabine as treatment for moderate- to high-risk resected CRC.
Colon and rectum
PICCOLO is a phase II study that will look at irinotecan (Ir) versus Ir plus panitumumab versus Ir plus cyclosporin in fluorouracil-resistant advanced CRC. The QUASAR 2 trial studies postoperative adjuvant chemotherapy capecitabine plus bevacizumab in colon cancer. RICE is a phase I/II study of radiotherapy, irinotecan, and capecitabine, followed by excision for locally advanced rectal cancer. SIGGAR 1 involves CT colonography, colonoscopy or barium enema for the diagnosis of CRC in older symptomatic patients. The Wales Polyp study is a screening study that looks at genes and multiple colorectal adenomas. XERXES is a phase I/II study that looks at cetuximab and capecitabine in preoperative chemoradiotherapy for rectal cancer.
REFERENCES Adam, I. J., Mohamdee, M. O., Martin, I. G. et al. (1994). Role of circumferential margin involvement in the local recurrence of rectal cancer. Lancet, 344, 707–11. Adam, R., Delvart, V., Pascal, G. et al. (2004). Rescue surgery for unresectable colorectal liver metastases downstaged by chemotherapy: a model to predict long-term survival. Ann. Surg., 240, 644–57. Agrawal, J. and Syngal, S. (2005). Colon cancer screening strategies. Curr. Opin. Gastroenterol., 21, 59–63. Andre, T., Boni, C., Mounedji-Boudiaf, L. et al. (2004). Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N. Engl. J. Med., 350, 2343–51. Anonymous. (2000). Comparison of fluorouracil with additional levamisole, higher-dose folinic acid, or both, as adjuvant chemotherapy for colorectal cancer: a randomised trial. QUASAR Collaborative Group. Lancet, 355, 1588–96. Athreya, S., Moss, J., Urquhart, G. et al. (2006). Colorectal stenting for colonic obstruction: the indications, complications, effectiveness and outcome – 5-year review. Eur. J. Radiol., 60, 91–4. Autier, P., Boyle, P., Buyse, M. et al. (2003). Is FOB screening really the answer for lowering mortality in colorectal cancer? Recent Results Cancer Res., 163, 254–63. Bingham, S. A., Day, N. E., Luben, R. et al. (2003). Dietary fibre in food and protection against colorectal cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC): an observational study. Lancet, 361, 1496–501; Erratum in Lancet, 362, 1000. Birgisson, H., Pahlman, L., Gunnarsson, U. et al. (2005). Adverse effects of preoperative radiation therapy for rectal cancer: long-term follow-up of the Swedish Rectal Cancer Trial. J. Clin. Oncol., 23, 8697–705. Bismuth, H., Adam, R., Levi, F. et al. (1996). Resection of nonresectable liver metastases from colorectal cancer after neoadjuvant chemotherapy. Ann. Surg., 224, 509–20.
Camma, C., Giunta, M., Fiorica, F. et al. (2000). Preoperative radiotherapy for resectable rectal cancer: a meta-analysis. J.A.M.A., 284, 1008–15. Colorectal Cancer Collaborative Group. (2001). Adjuvant radiotherapy for rectal cancer: a systematic overview of 8507 patients from 22 randomised trials. Lancet, 358, 1291–304. Cunningham, D., Humblet, Y., Siena, S. et al. (2004). Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N. Engl. J. Med., 351, 337–45. Dube, S., Heyen, F., and Jenicek, M. (1997). Adjuvant chemotherapy in colorectal carcinoma: results of a meta-analysis. Dis. Colon Rectum, 40, 35–41. Fearon, E. R. and Vogelstein, B. (1990). A genetic model for colorectal tumorigenesis. Cell, 61, 759–67. Folkesson, J., Birgisson, H., Pahlman, L. et al. (2005). Swedish Rectal Cancer Trial: long lasting benefits from radiotherapy on survival and local recurrence rate. J. Clin. Oncol., 23, 5644–50. Fong, Y. and Salo, J. (1999). Surgical therapy of hepatic colorectal metastasis. Semin. Oncol., 26, 514–23. Gray, R. G., Barnwell, J., Hills, R. et al. (2004). QUASAR: a randomised study of adjuvant chemotherapy (CT) vs. observation including 3238 colorectal cancer patients. J. Clin. Oncol., 2004 ASCO Annual Meeting Proceedings (Post-Meeting Edition), Vol. 22, No. 14S (July 15 Suppl.), 3501. Haller, D. G., Catalano, P. J., Macdonald, J. S. et al. (2005). Phase III study of fluorouracil, leucovorin and levamisole in high-risk stage II and III colon cancer: final report of Intergroup 0089. J. Clin. Oncol., 23, 8671–8. Hardcastle, J. D., Chamberlain, J. O., Robinson, M. H. et al. (1996). Randomised controlled trial of faecal-occult-blood screening for colorectal cancer. Lancet, 348, 1472–7. Hurwitz, H., Fehrenbacher, L., Novotny, W. et al. (2004). Bevacizumab plus irinotecan, fluorouracil and leucovorin for metastatic colorectal cancer. N. Engl. J. Med, 350, 2335–42. Jacobs, D. R., Jr., Marquart, L., Slavin, J. et al. (1998). Whole-grain intake and cancer: an expanded review and meta-analysis. Nutr. Cancer, 30, 85–96. James, R. D., Donaldson, D., Gray, R. et al. (2003). Randomised clinical trial of adjuvant radiotherapy and 5-fluorouracil infusion in colorectal cancer (AXIS). Br. J. Surg. 90, 1200–12. Kerr, D. J., Gray, R., McConkey, C. et al. (2000). Adjuvant chemotherapy with 5-fluorouracil, L-folinic acid and levamisole for patients with colorectal cancer: non-randomised comparison of weekly versus four-weekly schedules – less pain, same gain. QUASAR Colorectal Cancer Study Group. Ann. Oncol., 11, 947–55. Key, T. J., Allen, N. E., Spencer, E. A. et al. (2002). The effect of diet on risk of cancer. Lancet, 360, 861–8. Lichtman, S. M. (2004). Chemotherapy in the elderly. Semin. Oncol., 31, 160–74. MacFarlane, J. K., Ryall, R. D. and Heald, R. J. (1993). Mesorectal excision for rectal cancer. Lancet, 341, 457–60. Marijnen, C., Peeters, K., Putter, H. et al. (2005). Long term results, toxicity and quality of life in the TME trial. Proc. Am. Soc. Clin. Oncol. Gastrointestinal Cancer Symposium, Abstr. 166. Maughan, T. S., James, R. D., Kerr, D. J. et al. (2002). Comparison of survival, palliation and quality of life with three chemotherapy
171
Richard Adams, Timothy Maughan and Tom Crosby
regimens in metastatic colorectal cancer: a multicentre randomised trial. Lancet, 359, 1555–63. Maughan, T. (2005). Fluorouracil (FU), oxaliplatin, (OX), CPT-11 (irinotecan, Ir) use and sequencing in advanced colorectal cancer (ACRC): the UK MRC FOCUS (CR08) trial. Proc. Am. Soc. Clin. Oncol. Gastrointestinal Cancers Symposium, Abstr. 165. MERCURY Study Group. (2006). Diagnostic accuracy of preoperative magnetic resonance imaging in predicting curative resection of rectal cancer: prospective observational study. B.M.J., 333, 779. NHS Executive. (1997). Cancer Guidance Sub-group of the Clinical Outcomes Group. Guidance Improving Outcomes in Colorectal Cancer: The Research Evidence. London: Department of Health. NHS Executive. (2000). Referral Guidelines for Suspected Cancer. London: Department of Health. NICE. (2005). Technology Appraisal 93. Colorectal Cancer (advanced) Irinotecan, Oxaliplatin and Raltitrexed (Review): Guidance. London: NICE. O’Connell, M. J., Laurie, J. A., Kahn, M. et al. (1998). Prospectively randomised trial of postoperative adjuvant chemotherapy in patients with high-risk colon cancer. J. Clin. Oncol., 16, 295–300. Renehan, A. G., Egger, M., Saunders, M. P. et al. (2002). Impact on survival of intensive follow up after curative resection for colorectal cancer: systematic review and meta-analysis of randomised trials. B.M.J., 324, 813. Rowan, A., Halford, S., Gaasenbeek, M. et al. (2005). Refining molecular analysis in the pathways of colorectal carcinogenesis. Clin. Gastro. Hepatol., 3, 1115–23. Saltz, L. B., Niedzwiecki, D., Hollis, D. et al. (2004). Irinotecan plus fluorouracil/leucovorin (IFL) versus fluorouracil/leucovorin alone (FL) in stage III colon cancer (intergroup trial CALGB C89803). J. Clin. Oncol., 2004 ASCO Annual Meeting Proceedings (Post-Meeting Edition), Vol. 22, No, 18S (July 15 Suppl.), 3500. Sauer, R., Becker, H., Hohenberger, W. et al. (2004). Preoperative versus postoperative chemoradiotherapy for rectal cancer. N. Engl. J. Med., 351, 1731–40. Sebag-Montefiore, D., Steele, R., Quirke, P. et al. (2006). Routine short-course pre-op radiotherapy or selective post-op chemoradiotherapy for resectable rectal cancer? Preliminary results of the MRC CR07 randomised trial. J. Clin. Oncol., 2006 ASCO Meeting Proceedings Part I, Vol. 24, No. 18S (June 20 Suppl.), 3511. Thirion, P., Wolmark, N., Haddad, E. et al. (1999). Survival impact of chemotherapy in patients with colorectal metastases confined
172
to the liver: a re-analysis of 1458 non-operable patients randomised in 22 trials and 4 meta-analyses. Meta-Analysis Group in Cancer. Ann. Oncol., 10, 1317–20. Tournigand, C., Cervantes, A., Figer, A. et al. (2006). OPTIMOX1: a randomised study of FOLFOX4 or FOLFOX7 with oxaliplatin in a stop-and-go fashion in advanced colorectal cancer – a GERCOR study. J. Clin. Oncol., 24, 394–400. Twelves, C., Wong, A., Nowacki, M. P. et al. (2005). Capecitabine as adjuvant treatment for stage III colon cancer. N. Engl. J. Med., 352, 2696–704. UICC. (2002). TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 72–6. Van Cutsem, E., Labianca, R., Hossfeld, G. et al. (2005). Randomised phase III trial comparing infused irinotecan/5-fluorouracil (5-FU)/folinic acid (IF) versus 5-FU/FA (F) in stage III colon cancer patients (pts). (PETACC 3). J. Clin. Oncol., 2005 ASCO Meeting Proceedings, Vol. 23, No. 16S, Part I of II (June 1 Suppl.), 8. Van Cutsem, E. J. D., Tabernero, J., Dianz-Rubio, E. et al. (2004). An international phase II study of cetuximab in combination with oxaliplatin/5-fluorouracil (5-FU)/folinic acid (FA) (FOLFOX4) in the first-line treatment of patients with metastatic colorectal cancer (CRC) expressing epidermal growth factor receptor. Ann. Oncol., 15 (Suppl. 3), 339. Vasen, H. F., Watson, P., Mecklin, J. P. et al. (1999). New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative group on HNPCC. Gastroenterology, 116, 1453–6. Wade, T. P., Virgo, K. S., Li, M. J. et al. (1996). Outcomes after detection of metastatic carcinoma of the colon and rectum in a national hospital system. J. Am. Coll. Surg., 182, 353–61. Willett, C. G. (1999). Technical advances in the treatment of patients with rectal cancer. Int. J. Radiat. Oncol. Biol. Phys., 45, 1107–8. Wolmark, N., Wieand, H. S., Kuebler, J. P. et al. (2005). A phase III trial comparing FULV to FULV + oxaliplatin in stage II or III carcinoma of the colon: results of NSABP Protocol C-07. J. Clin. Oncol., ASCO Annual Meeting Proceedings, 23 (16 Suppl. I), 3500. Ychou, M., Raoul, J., Douillard, J. et al. (2005). A phase II randomized trial of LV5FU2+CPT-11 vs. LV5FU2 alone in adjuvant high risk colon cancer (FNCLCC Accord02/FFCD9802). J. Clin. Oncol., ASCO Annual Meeting Proceedings, Vol. 23, No. 16S, Part I of II (June 1 Suppl.), 3502.
Colon and rectum
Appendix. Colorectal chemotherapy regimens Dose of Regimen
Mode
Doses of chemotherapy
leucovorin
Duration
Interval
Use
Mayo
Bolus
5-FU 425 mg/m2
20 mg/m2
Day 1–5
4 weeks
1st line
QUASAR
Bolus
5-FU 370 mg/m2
20 mg/m2
30 weeks
1 week
Adjuvant
de Gramont
Bolus/infusion
(5-FU 400 mg/m2 bolus then
200 mg/m2
2 days
14 days
1st line
175 mg
2 days
14 days
1st line
-
12 weeks
-
1st line
600 mg/m2 over 22 hours) D1 and 2 Modified de
Bolus/infusion
2.8 g/m2 over 46 hours
Gramont Lokich
5-FU 400 mg/m2 bolus then
Continuous
5-FU 300 mg/m2 per day
infusion Capecitabine
Oral
1250 mg/m2 b.d.
-
14 days
3 weeks
1st line
Oxaliplatin
Bolus/infusion
5-FU 400 mg/m2 bolus then
175 mg
2 days
14 days
1st line with
modified de
2.4 g/m2 over 46 hours plus
potentially
Gramont
oxaliplatin 85 mg/m2
resectable liver mets
Irinotecan modified
Bolus/infusion
400 mg/m2 bolus, 2.4 g/m2
175 mg
2 days
14 days
2nd line
over 46 hours plus
de Gramont
irinotecan 180 mg/m2 Irinotecan
Short infusion
350 mg/m2
-
1 day
3 weeks
2nd line
XELOX
Oral/infusion
Capecitabine 1000 mg/m2
-
2 weeks
3 weeks
Unlicensed
(capecitabine and
p.o. b.d. days 1–14 plus
oxaliplatin)
oxaliplatin 130 mg/m2 over
metastatic
2 hours
173
14
ANUS Richard Adams and Tom Crosby
Introduction
Incidence and epidemiology
Anal cancer, like carcinoma of the cervix, is strongly associated with human papilloma virus (HPV) infection. In the past few decades, treatment has swung away dramatically from primary surgery to definitive chemoradiotherapy, with the evident physical and psychosocial benefits of organ preservation. Tumours of lower stage T1 or T2, N0, have an excellent outcome with the chemoradiotherapy approach. Unfortunately, many patients present with T3, T4 or N ≥ 1 disease and they have a significant risk of both locoregional and subsequent distant failure. Half of patients still die within 5 years of treatment. Therefore, efforts are ongoing to improve outcomes via better local and systemic therapy.
The annual incidence of anal cancer in the UK is approximately 1.5 in 100 000. There are approximately 720 new cases reported per year in England (National Statistics, 2005), which constitute approximately 10% of anorectal tumours; approximately 200 deaths occur per year from this cancer. There has been a slight but steady increase in incidence in Western countries; wide geographical variation in incidence mirrors the relative risk of HPV infection (e.g. high incidence in areas of Brazil) with similar high rates of penile, vulval and cervical cancer. Anal-canal tumours occur more often in women (male-to-female ratio of 1:2.5) but cancer of the anal margin is slightly more common in men. The peak incidence age for anal cancer is 60 to 65 years but there is a bimodal distribution that includes a younger group, ages 35 to 40 years.
Types of anal tumour Types of anal tumour are shown in Table 14.1.
Anatomy The anal canal extends from the rectum to the junction of the hair-bearing skin of the perianal region. It is 3 to 4 cm long and its walls are kept in apposition by the sphincter muscles, except during defaecation. The anal margin is the perianal skin immediately adjacent to the distal limit of the anal canal. The anal verge is the lower end of the anal canal. The tumours can be divided into three types: 1. Anal-margin tumours are usually small and well differentiated and are more common in men. 2. Anal-canal tumours are more common than those at the anal margin. They are more common in women (male-to-female ratio ∼1:2.5), are often moderately or poorly differentiated, and carry a worse prognosis. 3. Dual-component tumours have both anal-margin and -canal components and it is not possible to define from where the tumour originated. These comprise the majority of tumours. 174
Carcinoma of the anus Risk factors and aetiology Anal cancer was previously believed to be associated with chronic inflammation from haemorrhoids, fistulae, and fissures, but there is no evidence to support this or an association with ulcerative colitis. A case-control study performed between 1978 and 1985 established a link in women between anal cancer and genital warts (Daling et al., 1987). Men with anal cancer were more likely to be single or to have had homosexual relationships or to have practiced anal-receptive intercourse, as well as to have had sexually transmitted diseases (Daling et al., 1982). The sexually transmitted agent implicated in anal cancer is HPV. The presence of the virus, especially types 16 and 18, can be found in more than 90% of cases (Tilston, 1997). HPV positivity or negativity does not appear to affect survival. Other factors for disease include having multiple sexual partners; having other sexually transmitted viral and
Anus
Table 14.1. Types of anal tumour Type Benign
Examples Condylomata Conditions that may cause
Table 14.2. Pathological features of squamous carcinoma of the anus Features
Description
Macroscopic
Nodular or plaque-like when small, but
Microscopic
Canal tumours tend to be composed of
swelling/ulceration Haemorrhoids
ulcerative and infiltrative when larger
Fissure
small cells with basaloid features and
Fistula
non-keratinisation and HPV +ve
Abscess
∼100%. Perianal tumours are usually
Crohn’s disease Primary malignant
large-cell, keratinising and HPV +ve in
Anal canal
about two-thirds of cases
Squamous cell carcinoma: Large-cell keratinising Large-cell non-keratinising (transitional) Basaloid Adenocarcinoma Small-cell carcinoma Malignant melanoma Lymphoma Carcinoid
of squamous carcinoma, which may also be keratinising and non-keratinising. Because the natural histories and prognoses for all these subtypes are similar, they may be collectively described as epidermoid tumours. The pathological features of squamous carcinoma of the anus are shown in Table 14.2.
GIST Undifferentiated carcinoma Anal margin Squamous cell carcinoma Basal cell carcinoma Bowen’s disease Paget’s disease Secondary malignant
Metastatic spread from other tumours: case reports only
Anal intraepithelial neoplasia (AIN) Anal intraepithelial neoplasia is graded from one to three as in CIN (cf. carcinoma of the cervix) and is considered precancerous (the precursor of squamous cell carcinoma of the anus). The tumour is usually flat or raised, and ulceration is more likely to represent invasive disease.
Direct spread from rectum, cervix, vagina, etc. GIST = gastrointestinal stromal tumour.
bacterial infections; a history of cancer or intraepithelial neoplasia of the cervix, vagina or vulva; cigarette smoking; and iatrogenic immunosuppression for organ transplant. And, although anal cancer is not an AIDSdefining malignancy, an increased incidence of anal cancers is associated with HIV/AIDS.
Pathology Squamous carcinomas represent 90% of anal cancers. Cloacogenic, basaloid and transitional are all variants
Spread There are abundant lymphatic channels in the anus. Overall, approximately 25% of patients have palpable inguinal nodes at presentation. However, of these, only about 50% have malignant lymphadenopathy cytologically confirmed upon fine needle aspiration (FNA). A negative FNA does not exclude malignancy; despite this fact, excision biopsy is rarely performed in clinical practice.
Spread direct from primary Cancer spread from the primary site is upward, submucosally to the rectum and bladder; laterally, into the ischio-rectal fossa and sphincter muscles; in women, to the vagina/urethra; and in men, to the prostate and downward to the perianal skin. 175
Richard Adams and Tom Crosby
Table 14.3. TNM classification of carcinomas of the
Table 14.4. Stage grouping for carcinomas of the anal
anal canal
canal
Stage
Definition
Stage
Description
T1
2 cm or less
Stage 0
Tis
N0
M0
T2
Greater than 2 cm, not more than 5 cm
Stage I
T1
N0
M0
T3
Greater than 5 cm
T4
Of any size invading adjacent organs (e.g. vagina,
Stage II
T2
N0
M0
T3
N0
M0
urethra, bladder); involvement of sphincter muscles alone is not T4 N1
Metastasis in perirectal lymph node(s)
N2
Metastasis in unilateral internal iliac and/or
Stage IIIA
Stage IIIB
unilateral inguinal lymph node(s) N3
Metastasis in perirectal and inguinal lymph nodes and/or bilateral internal iliac and/or bilateral inguinal lymph nodes
M1
Distant metastasis present
Stage IV
T1–3
N1
M0
T4
N0
M0 M0
T4
N1
Any T
N2
M0
Any T
N3
M0
Any T
Any N
M1
Adapted from UICC (2002).
Adapted from UICC (2002).
Investigation and staging
Lymphatic spread Low anal tumours, anal verge tumours and anal-margin tumours spread to perirectal nodes, followed by the inguinal nodes, and then to external iliac vessels and the common iliac/para-aortic chain. Mid and upper anal-canal tumours spread via the internal iliac nodes to the pelvis, including the hypogastric and obturator nodes and not infrequently to paraaortic/retroperitoneal nodes.
Metastatic spread Haematogenous metastatic spread is to the liver, less frequently to the lungs and bones, and, rarely, to the brain.
Examination without anaesthetic is often adequate to stage local disease (in ∼80% of patients). In patients with extensive tumour or excessive pain and tenderness, examination under anaesthetic (EUA) may be required. Biopsy of both the primary tumour and suspicious nodes in the inguinal region should be performed or, more commonly, FNA, but the false-negative rate is higher with FNA. A full blood count, biochemical profile and, if risk factors are present, counselling and an HIV test should be performed. A chest X-ray, computed tomography (CT) of the abdomen and pelvis, and MRI can be performed. MRI is more effective at imaging the primary tumour and perirectal disease. Further investigations should be performed as clinically or biochemically indicated.
Stage classification Clinical presentation The tumour occurs as a lump or mass, either found by the patient on wiping or causing the patient discomfort. Bleeding, discharge and anal discomfort occur in about 50% of patients and about 25% are aware of a mass. The non-specific nature of the tumour can lead to a reporting delay of up to 6 months in one-third of patients. A tumour is occasionally found during an investigation of malignant inguinal lymphadenopathy. It is rare for anal cancers to present with metastases outside the pelvis. 176
TNM classification and stage groupings The TNM classification and stage grouping are shown in Tables 14.3 and 14.4, respectively. The staging applies to anal-canal tumours; anal-margin tumours are staged in the same way as skin tumours.
Treatment overview Radiotherapy replaced surgery as the initial treatment in the 1980s. Evidence that concurrent chemoradiotherapy (CRT) improves local disease control compared to
Anus
radiotherapy alone emerged in the 1990s. Current studies are under way to find the most effective chemoradiotherapy regimen. Abdominoperineal resection (APR) is used for patients who are unsuitable for CRT, for patients with persistent/recurrent local disease or for patients who have unmanageable late toxicity after CRT. NICE service guidance on colorectal cancer suggests that anal cancer should be managed by a network-based specialist team that includes an oncologist and a surgeon and that has established links with gynaecological and plastic surgical services. Within each radiotherapy facility, no more than two clinical oncologists should take responsibility for the care of patients with anal cancer and these should be core members of the anal cancer MDT (NICE, 2004). If the tumour recurs following CRT, the only curative salvage therapy is APR.
Radical surgery for anal carcinoma Surgery for primary tumour Well-differentiated margin tumours less than 2 cm in diameter may be locally excised if margins of 5 mm or larger are possible. The incidence of nodal disease is less than 5%. APR is now rarely used as a primary treatment. However, if radiotherapy is contraindicated – because the patient has previously had radiotherapy to the pelvis, there is a functioning transplanted kidney present in the pelvis, or there is already loss of continence due to the tumour – an end colostomy may be appropriate. The management of high-grade AIN is uncertain. Where possible this type of cancer should be treated by local excision, and APR should be reserved for symptomatic uncontrolled multifocal disease. Adjuvant radiotherapy or CRT should be considered for patients who have undergone an excisional biopsy for an invasive tumour with positive margins, treating to a dose of 30 Gy (Hu et al., 1999).
Radical radiotherapy and chemoradiotherapy Nigro et al. (1983), from the Wayne State University Cancer Center in Michigan, delivered a preoperative CRT schedule. On examination of the surgical specimen, 7 out of 12 patients had no viable tumour (Nigro et al., 1974, 1983). The dose schedule was 30 Gy in 15 fractions with two cycles of chemotherapy (5-FU and mitomycin C, two cycles, 4 weeks apart). Simlarly, Cummings et al.
(1984) described a cohort of patients treated with CRT and compared them to a similar group of patients treated with radiation alone. The radiation given was 45 to 50 Gy in 25 fractions. Marked acute toxicity was seen when radiotherapy was given with 5-FU and mitomycin C (two cycles of 5-FU, 1 gm/m2 and mitomycin C, 10 mg/m2 ). A 4-week gap was therefore introduced between the two treatment phases. At 6 months, local control was achieved in 60% of the patients treated with radiotherapy alone, as opposed to 94% treated with continuous-phase chemoradiation and 93% in the splitcourse CRT group. In 1996, the UKCCR published a randomised control trial of 585 patients (ACT I) that compared radiotherapy and CRT (Anal Cancer Trial Working Party, 1996). ACT I was a very successful trial, and ∼30% of all patients with anal cancer in the UK took part. Local failure occurred in 39% of patients in the CRT arm versus 61% who underwent radiotherapy alone. Anal cancer deaths were reduced by 29%, but overall survival was not significantly different. Of patients who failed CRT, 50% were salvaged with surgery as opposed to 60% of patients who failed after radiotherapy alone. Chemoradiotherapy was associated with significantly greater toxicity. The dose schedule was 45 Gy in 25 or 20 fractions with two cycles of 5-FU and mitomycin C following a 6-week break for assessment. If there was less than a 50% response, patients went on to an APR. If there was greater than a 50% response, patients received a boost, either with brachytherapy 10 Gy/day with iridium-192 (25 Gy) or EBRT 15 Gy in six fractions. The radiotherapy technique was as follows: the superior border was the bottom of the S–I joints and the inferior border covered the tumour by at least 3 cm. It was recommended that the inguinal nodes were included with the use of large parallel-opposed fields, although this was not mandatory. The chemotherapy regimen was 5-FU, 1 g/m2 on days 1 to 4 and mitomycin C, 12 mg/m2 on day 1. The second cycle, given during the last week of radiotherapy, consisted of 5-FU alone. Two further trials in the USA and Europe confirmed the superiority of local control of CRT over radiotherapy alone, albeit at the expense of increased acute toxicity (Bartelink et al.,1997; Flam et al., 1996).
Radiotherapy technique Consider the use of a temporary defunctioning colostomy if the patient has a large obstructing tumour or incontinence. The patient is simulated and treated 177
Richard Adams and Tom Crosby
prone, with hands under the head. Orthogonal films are taken. It is useful to instil a little barium in the anus and rectum and to use a radio-opaque marker (wire) to localise the anal verge. The bladder should be comfortably full. Palpable lymph nodes should be marked with wire. Bolus is applied to the natal cleft in all analmargin tumours and anal-canal tumours that extend to within 2 cm of the anal verge. The GTV should be marked on the film before the phase 1 and phase 2 field borders are defined. The PTV should be individually adapted; it includes the known tumour and involved lymph nodes with at least a 3 cm margin and also treats potential lymph node involvement within the true pelvis and up to the common iliac nodes in higher tumours using a shrinking field technique. This technique was developed under the rationale that 30 Gy of CRT should be adequate to treat microscopic disease and 50 Gy is required to treat gross tumour (Melcher and Sebag-Montefiore, 2003). Phase 1 and 2 volumes as described next are extracted from the ACT II protocol, which is the current UKCCCR anal cancer trial (see p. 181). The target volume for phase 1 includes inguinal nodes, the true pelvis and the known tumour with large anterior–posterior parallel-opposed fields. The following fields should be expanded to ensure coverage of the GTV plus 3 cm: r Superior border – 2 cm above the inferior aspect of the S–I joints, unless pelvic lymph nodes or the primary tumour extend to within 3 cm, in which case the border is recommended to extend 3 cm above the upper limit of macroscopic disease. r Inferior border – 3 cm below the anal margin, or 3 cm below macroscopic disease for marginal tumours. r Lateral borders – lateral to the femoral heads to cover inguinal nodes completely. Large parallel-opposed fields are used to treat the phase 1 target volume. Field sizes are typically 14 to 16 cm (S–I) × 17 to 20 cm for large opposed fields. The phase 2 target volume is to treat macroscopic disease with at least a 3 cm margin. Treatment is dependent on the presence or absence of significant lymphadenopathy determined by CT or MRI imaging. Planning using orthogonal films or CT planning and radio-opaque markers is recommended (for rectal contrast and an anal verge marker). In the absence of lymph node disease, a three- to four-portal plan should be used for all anal-canal tumours. Anal-margin tumours may be treated with a single direct field with a 3 cm margin. If lymph node disease is evident, anterior–posterior 178
parallel-opposed fields are used to include the GTV plus a margin of 3 cm. For phases 1 and 2, the verification technique requires that simulator check films taken before treatment, and portal check films be taken on the set weekly and for the first three fractions of each phase. Dose, fractionation and energy for phase 1 involves 30.6 Gy in 17 fractions to the ICRU reference point (usually midpoint) for microscopic and macroscopic disease. For phase 2 give 19.8 Gy in 11 fractions to the ICRU reference point for macroscopic disease. The total dose should be 50.4 Gy in 28 fractions over 5.5 weeks to the ICRU 50 reference point, usually the point of the beam intersection using 10 MV photons. Unscheduled interruptions should be avoided. The side effects of radiotherapy are shown in Table 14.5.
Chemotherapy regimen in combination with radiotherapy The dose regimen is 5-FU 1000 mg/m2 per day by continuous 24-hour infusion on days 1 to 4 and days 29 to 32. Mitomycin C, 12 mg/m2 , is given by i.v. bolus on day 1 (max. single dose 20 mg). 5-FU should be commenced at least 1 hour before the first fraction of radiotherapy. If GFR is 50 ml/minute or less (confirmed by EDTA clearance), omit mitomycin C. Some patients, such as small elderly women, may have a spuriously low GFR as estimated by using the Cockcroft-Gault formula. Weekly FBC should be obtained, and prophylactic antibiotics (ciprofloxacin 250 mg b.d.) should be given throughout treatment.
Implant/brachytherapy boost Implantation alone may be suitable as a boost for T1 and T2 tumours if they: r Lie below the anal-rectal ring. r Occupy less than 50% of the circumference of the canal. r Are 1 cm or less in thickness. r Have no nodal involvement. When brachytherapy is used alone it is common to give a dose in the region of 55 to 60 Gy to the PTV. If used in combination with external beam radiotherapy, a dose of 45 Gy in 25 fractions to the pelvic field including inguinal lymph nodes and a brachytherapy boost of 25 Gy are used. Brachytherapy should only be done in specialist centres with a skilled operator. There is a high risk of dose inhomogeneity with implants in this region, which may lead to radionecrosis.
Anus
Table 14.5. Side effects of radiotherapy for anal cancer Affect
Management
Acute side effects Lethargy
Advice re: coping with fatigue (e.g. pacing, goal setting)
Cutaneous toxicity from dry, usually to
Expose to air; use hydrocortisone 1% cream Occasionally, 2nd Skin® application is difficult in this area; hair dryer blowing
moist, desquamation between buttocks. Haematological toxicity
cold can be soothing FBC once a week, delay if neutrophil count ≤ 1.0 × 109 /l or platelet count ≤ 50 × 109 /l
Cystitis
Alkalinisation of urine; drinking fluids; potassium citrate/cranberry juice
Diarrhoea, tenesmus
Loperamide/codeine/low-residue diet; delay RT if diarrhoea > 7 times per day and requiring parenteral support, or incontinence; postpone until under control
Crampy abdominal discomfort
Suspend treatment if guarding or significant tenderness
Late side effects Sterility
HRT for premenopausal females Offer sperm banking to males MUST BE DISCUSSED AT CONSENT
Anal stenosis
AP resection if failure to control conservatively
Faecal incontinence
AP resection if failure to control conservatively
Telangiectasia of the rectum/bladder
Surgery if failure to control conservatively
leading to bleeding Bowel fistula
Small bowel resection if severe, often multiple loops affected
Chronic diarrhoea
Loperamide
Technique for brachytherapy The patient is prepared for theatre with twice-daily phosphate enemas. Codeine is sometimes used over several days before treatment. In theatre the patient lies in the lithotomy position and is catheterised. It is essential to have details of the initial tumour, because the tumour is likely to have responded substantially after the first phase of therapy. EUA is performed to decide on the arrangement of the implant, usually between five and seven hollow applicators, separated by 1 cm and not more than two-thirds circumference of the anal canal, and should be 30% longer than the tumour length. Suture a circular or horseshoe template to the perianal skin. Put in tubes with guide needles, and place index finger into the anus and to guide needles subcutaneously. Once satisfied with the position, tighten the screw to keep the needles in place. Apply a plastic obturator to prevent coning of needles at the apex.
Dose with high-dose-rate iridium-192, 15 Gy in two fractions, 6 hours apart (alternatively 16.5 Gy in three fractions) or at a low dose rate of 20 to 25 Gy at 0.5 Gy/ hour. Provide adequate analgesia for the patient on ward, including diamorphine subcutaneous infusion. The second treatment, and removal of tubes, is done without anaesthetic but is usually tolerated reasonably well.
Treatment for locally recurrent anal carcinoma Formal restaging is essential; localised disease can be considered for radical resection (APR). In a report from a single centre in the UK involving 254 patients with radiotherapy alone or with concurrent chemotherapy, there were 99 local disease failures, all but 5 being within 3 years of primary therapy (Renehan et al., 2005). Age, dose and stage were significantly associated with 179
Richard Adams and Tom Crosby
recurrence. Of 73 patients who underwent surgical salvage the 3- and 5-year survival rates were 55 and 29%, respectively. Recurrent disease in the inguinal nodes (in patients previously treated with CRT to this site and in whom there is local control of the primary) can be treated with block dissection. If the malignant lymphadenopathy is mobile/non-fixed, this treatment provides local control in about 80% of patients. The procedure is associated with a significant risk of lymphoedema, which is increased by pre- or postoperative radiation.
Palliative treatment for anal carcinoma Palliative radiotherapy/chemoradiotherapy Radiotherapy alone can be useful for patients who are unfit to receive chemotherapy. Useful palliation and even long-term disease control can be achieved with radiotherapy with a dose of 45 to 54 Gy in 25 to 30 fractions. Fungating inguinal nodes can be treated with 30 Gy in ten fractions, using 4 to 6 MV photons with bolus or, in more frail patients, 6 Gy per fraction weekly for 5 to 6 weeks. Effective palliation/local control can also be achieved with a low-dose CRT regimen consisting of 30 Gy to the GTV plus a 3 cm margin with concomitant 5-FU (Charnley et al., 2005).
Palliative chemotherapy Haematogenous metastatic disease tends to occur late, but it can be rapidly progressive. Combination chemotherapy regimens are often used, such as cisplatin with 5-FU or mitomycin C with 5-FU, giving treatment not previously used during primary CRT. However, the evidence for their use comes from studies in metastatic carcinoma of the cervix rather than anal cancer.
Follow-up after radical treatment Clinically assess the patient at 6 weeks after the end of CRT and subsequently every 1 to 3 months; EUA is reserved for patients who find clinical examination too uncomfortable or for those in whom local disease remains an issue. Consider a radiotherapy boost/ implant if there is less than a 50% response or consider salvage surgery with APR after restaging. If possible one should avoid a large biopsy in a previously irradiated area because of the risk of radionecrosis. However, it is 180
essential to confirm persistent or recurrent disease if an APR is being considered.
Special problems Anal cancer in patients with HIV/AIDS Patients with HIV/AIDS are at increased risk of developing anal cancer. All patients who are HIV +ve and have anal cancer (including those with CD4 count > 200 cells/ mm3 ) should be on highly active antiretroviral therapy (HAART). Patients treated with radiotherapy with or without chemotherapy are at increased risk of mucosal and cutaneous toxicity. Such patients should be monitored closely. Patients with established AIDS or a CD4 count ≤ 200 cells/mm3 appear more likely to suffer local and haematological toxicity and it would seem appropriate to optimise retroviral therapy (HAART) prior to treatment. Consideration of a modified chemotherapy dose regimen, reduced PTV or a reduced radiotherapy dose should be considered for this group.
Adenocarcinoma Although long-term stoma-free disease control has been described following radiotherapy or CRT, patients should be treated as if they have a low rectal adenocarcinoma and have an APR. But in an attempt to reduce involvement of the circumferential margin and the risk of subsequent local recurrence, preoperative radiotherapy with or without chemotherapy is frequently given.
Prognosis Survival Radiotherapy and radical surgery have similar survival and local control rates but radiotherapy has the advantage of being sphincter-sparing. For anal-canal tumours, the 5-year survival rate with radical CRT 65%, and the 5-year local control rate and rate of colostomy-free survival is 75%. For anal-margin tumours, the rates are 80 and 85%, respectively. The average time of survival after diagnosis with metastatic disease is 8 to 12 months.
Prognostic factors The principal prognostic factor of anal cancer is disease stage (i.e. TNM). Distant metastases clearly have the greatest impact on survival, whereas the T stage determines the chance of local control. Node-positive patients have a 10 to 20% lower 5-year survival rate than node-negative patients.
Anus
Patients with confirmed primary epidermoid anal cancer. Staged and biopsied by EUA and CT scan
GFR > 50 ml/min Randomise
5-FU and MMC + Radiotherapy
5-FU and MMC + Radiotherapy
5-FU and CDDP + Radiotherapy
5-FU and CDDP + Radiotherapy
No maintenance therapy
Maintenance therapy of 5-FU and CCCP
No maintenance therapy
Maintenance therapy of 5-FU and CDDP
Figure 14.1. The randomisation schedule of the ACT II trial. 5-FU = 5-fluorouracil; CDDP = cisplatin; CT = computed tomography; EUA = examination under anaesthetic; GFR = glomerular filtration rate; MMC = mitomycin C. Adapted from the ACT II protocol (available from www.ncrn.org.uk/).
The prognosis in females is generally better than in males. Ki-67, mitotic rate, TP53 mutation, and DNA aneuploidy are also factors affecting prognosis, but it is uncertain whether these are independent of the stage of the tumour.
Areas of current interest Adjuvant therapy has no proven role, although patients with adenocarcinoma of the anus may receive this treatment in line with rectal cancers. The ACT II study is investigating the role of ‘maintenance therapy’ for two cycles after CRT in squamous cell cancers. Radiotherapy dose, adjuvant/neoadjuvant therapy and the role of cisplatin (ACT II) and capecitabine in anal cancer are currently of interest. Treatment differences between early (T1 to 2, N0) and locally advanced (T3 to 4 or N ≥ 1) disease and treatment in the elderly are also areas of interest.
Current clinical trials The current UKCCCR trial (ACT II) has been extended due to excellent recruitment and now includes more than 600 patients with anal cancer; this trial addresses the issue of whether mitomycin C (MMC) or cisplatin with 5-FU and radiotherapy produce a greater complete
remission rate. ACT II also looks at the role of maintenance therapy after CRT using two cycles of cisplatin and 5-FU. See Figure 14.1 for the ACT II randomisation schedule. The EORTC is running a phase II/III randomised study (EORTC 22011) that looks at radiation therapy, mitomycin C and either 5-FU or cisplatin in treating patients with locally advanced anal cancer.
REFERENCES Anal Cancer Trial Working Party. (1996). Epidermoid anal cancer: results from the UKCCCR randomised trial of radiotherapy alone versus radiotherapy, 5-fluorouracil and mitomycin. Lancet, 348, 1049–54. Bartelink, H., Roelofsen, F., Eschwege, F. et al. (1997). Concomitant radiotherapy and chemotherapy is superior to radiotherapy alone in the treatment of locally advanced anal cancer: results of a phase III randomized trial of the European Organization for Research and Treatment of Cancer Radiotherapy and Gastrointestinal Cooperative Groups. J. Clin. Oncol., 15, 2040–9. Charnley, N., Choudhury, A., Chesser, P. et al. (2005). Effective treatment of anal cancer in the elderly with low-dose chemoradiotherapy. Br. J. Cancer, 92, 1221–5. Cummings, B., Keane, T., Thomas, G. et al. (1984). Results and toxicity of the treatment of anal canal carcinoma by radiation therapy or radiation therapy and chemotherapy. Cancer, 54, 2062–8.
181
Richard Adams and Tom Crosby
Daling, J. R., Weiss, N. S., Klopfenstein, L. L. et. al. (1982). Correlates of homosexual behavior and the incidence of anal cancer. J. A. M. A., 247, 1988–90. Daling, J. R., Weiss, N. S., Hislop, T. G. et al. (1987). Sexual practices, sexually transmitted diseases and the incidence of anal cancer. N. Engl. J. Med., 317, 973–7. Flam, M., John, M., Pajak, T. F. et al. (1996). Role of mitomycin in combination with fluorouracil and radiotherapy, and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: results of a phase III randomized intergroup study. J. Clin. Oncol., 14, 2527–39. Hu, K., Minsky, B. D., Cohen, A. M. et. al. (1999). 30 Gy may be an adequate dose in patients with anal cancer treated with excisional biopsy followed by combined modality therapy. J. Surg. Oncol., 70, 71–7. Melcher, A. A. and Sebag-Montefiore, D. (2003). Concurrent chemoradiotherapy for squamous cell carcinoma of the anus using a shrinking field radiotherapy technique without a boost. Br. J. Cancer, 88, 1352–7.
182
National Statistics. (2005). Series MB1 no. 34. Cancer Statistics Registrations. Registrations of Cancer Diagnosed in 2003, England. London: Office for National Statistics. NICE. (2004). Guidance on Cancer Services. Improving Outcomes in Colorectal Cancers. Manual Update. London: NICE. Nigro, N. D., Vaitkevicius, V. K. and Considine, B., Jr. (1974). Combined therapy for cancer of the anal canal: a preliminary report. Dis. Colon Rectum, 17, 354–6. Nigro, N. D., Seydel, H. G., Considine, B. et al. (1983). Combined preoperative radiation and chemotherapy for squamous cell carcinoma of the anal canal. Cancer, 51, 1826–9. Renehan, A. G., Saunders, M. P., Schofield, P. F. et al. (2005). Patterns of local disease failure and outcome after salvage surgery in patients with anal cancer. Br. J. Surg., 92, 605–14. Tilston, P. (1997). Anal human papilloma virus and anal cancer. J. Clin. Pathol., 50, 625–34. UICC. (2002). TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 77–80.
15
GASTROINTESTINAL STROMAL TUMOURS Kate Parker and Tom Crosby
Introduction Gastrointestinal stromal tumours (GISTs) are rare mesenchymal tumours that can occur anywhere in the gastrointestinal tract. GISTs have been difficult to diagnose in the past which, along with their rarity, makes their incidence hard to estimate. In the past, there was also very little to offer GIST patients in the way of treatment, and the prognosis was extremely poor. However, recent understanding of the molecular pathology involved in GISTs has made diagnosis more accurate and effective treatments are now available with molecular-targeted therapy. As research continues, it is likely that more molecular-targeted therapies will become available for this condition. GISTs originate from the interstitial cells of Cajal (ICC) – pacemaker cells that control gut motility. The tumours are diagnosed by a combination of morphological features and immunohistochemistry staining. Oncogenesis appears to be related to dysregulation of the proto-oncogene KIT, a growth-factor receptor tyrosine kinase (de Silva and Reid, 2003).
Types of tumour Of gastrointestinal mesenchymal tumours, GIST is the most common. The gastrointestinal autonomic tumour (GANT) is a phenotypic variant of GIST with neuroendocrine differentiation. True smooth muscle tumours also occur, but Schwann cell tumours are rare. Of GISTs, 60 to 70% express CD34, suggesting GISTs are an entity distinct from smooth muscle tumours. The majority (95%) of GISTs stain positively with CD117, an antibody that recognises an extracellular epitope of KIT (Corless et al., 2004). The most common mutation, which is found in 70% of GISTs, is in exon 11 of KIT. Ten percent have a mutation in exon 9, and rare mutations are found in exons 13 and 17. Of tumours that are morphologically similar to GISTs, 5% are CD117 negative. Approximately 30% of these have mutations
in the platelet-derived growth factor receptor alpha (PDGFRα; Kindblom et al., 2002).
Incidence and epidemiology The annual incidence in the UK is between 10 and 20 per million; it is difficult to obtain an accurate count because of the cancer’s rarity and because of previous difficulty in diagnosis. There are 600 to 1200 new cases reported per year in England and 30 to 60 new cases per year in Wales. GISTs have become more reliably diagnosable since KIT expression has become detectable by immunohistochemistry, which may change the perceived incidence (Novartis, 2003). GIST occurrence is rare before age 40; the median age at diagnosis is 50 to 60 years. Some studies show a slight male predominance but most show an equal gender distribution of the disease. GISTs represent 1% or fewer of all primary tumours of the GI tract and 5% of soft tissue sarcomas (Duffaud and Blay, 2003). The 5-year survival is roughly 50%, dropping to 35% at 10 years (note that these figures are before imatinib treatment). See Kindblom et al. (2002) for more information on incidence and epidemiology.
Risk factors and aetiology Because the tumour is rare, the aetiology is largely unknown. There are rare familial forms of GIST, which may be due to germline mutation in the genes for KIT or PDGFRα (Joensuu et al., 2002), including a rare syndrome – the Carney triad – and a possible association with neurofibromatosis type 1.
Histogenesis and pathology Histogenesis GISTs are thought to originate from the precursor of the ICC. GISTs and ICC share morphological features 183
Kate Parker and Tom Crosby
Table 15.1. Immunohistochemical features of GIST
Table 15.2. The molecular classification of GISTs
Immunohistochemical feature
Frequency (%)
Type
Description
CD117 (KIT) positive
90%
KIT (95%)
Exon 11: 70% of patients
CD34 positive
60–70%
Exon 9: 10% of patients
SMA positive
30–40%
Exons 13 and 17: rare (2% of patients)
S-100 protein positive
5%
PDGFRα
Exon 18: uncommon (6% of patients)
CD = cluster of differentiation; KIT = kitten (v-kit
(∼1.5%)
Hardy-Zuckerman 4 feline sarcoma viral oncogene
Wild type
Molecular aetiology unclear
homolog); S-100 = S-100 calcium binding protein; SMA =
Familial
Germline KIT or PDGFRα mutation
smooth muscle actin. Adapted from Fletcher et al. (2002).
Exons 12 and 14: rare (1.4% of patients)
GIST Paediatric
KIT and PDGFRα mutations rare
KIT = kitten (v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog);
and contain a similar form of embryonic smooth muscle myosin. They also both stain positive for CD117 (KIT). The main role of the ICC is as pacemaker cells that control gut motility. KIT is a transmembrane receptor tyrosine kinase. The ligand involved is the stem cell factor, which is also known as the mast cell growth factor. Ligand binding causes dimerisation of adjacent KIT molecules, which activates the intracellular signalling cascades affecting cell proliferation, adhesion, and differentiation, often via the signal transduction intermediates AKT and MAPK. Mutations in KIT often cause a change in these signals, which results in a gain of function – constitutive activation (Heinrich et al., 2003). Mutation in KIT is important but not felt to be solely sufficient to produce malignant transformation in a GIST. Studies have reported associations among mutation, histological appearance, and anatomical location of GISTs (Kindblom et al., 2002) and between mutation and prognosis (Fletcher et al., 2002).
Pathology
PDGFR = platelet-derived growth factor receptor. Adapted from Corless et al. (2004).
Microscopic appearance Microscopically, GISTs can occur as three different types: r Spindle cell (70%) – relatively uniform eosinophilic cells arranged in short fascicles or whorls. r Epithelioid (20%) – rounded cells with variably eosinophilic or clear cytoplasm. r Mixed – areas of each of the aforementioned types, either mingled together or with an abrupt transition between them.
Immunohistochemistry Table 15.1 shows the immunohistochemical features of GISTs. Metastatic deposits from other tumours can also be positive for CD117, including melanoma, angiosarcoma, and seminoma.
Macroscopic features
Molecular classification
GISTs are located in the stomach (60 to 70%); small intestine (20 to 30%); colon, rectum, and oesophagus (5 to 15%); and various other locations (omentum, mesentery, retroperitoneum, ≤ 5%). Tumours are 2 to 30 cm in diameter at the time of diagnosis but may be 1 cm or smaller if found incidentally. Most GISTs are submucosal, grow endophytically, and are often well circumscribed with a whorled, fibroid-like appearance. Larger lesions may display cystic degeneration or necrosis.
The molecular classification of GISTs is shown in Table 15.2.
184
Spread Modes of GIST spread involve spread by the primary tumour to involve contiguous structures. Spread to lymph nodes is rare even when there is metastatic disease, but it may occur following surgery. Metastatic spread commonly includes the liver and peritoneal
Gastrointestinal stromal tumours
cavity. Lung, bone or subcutaneous sites are involved, but rarely. Of all GISTs, 40 to 90% recur postoperatively, either locally or with metastatic disease.
Clinical presentation The patient initially presents typically with an intraabdominal tumour or as an emergency with intestinal haemorrhage or obstruction. Symptoms include gastrointestinal bleeding (50%), abdominal pain (20 to 50%), gastrointestinal obstruction (10 to 30%), tiredness, and general malaise. GIST is an incidental finding and patients are asymptomatic in 20 to 30% of cases.
Investigation and staging CT (or MRI) of the thorax, abdomen and pelvis should be used to look for metastases. Consider assessment of endocrine and αFP and βhCG levels to rule out adrenal tumours and teratomas, depending on the clinical context. Percutaneous biopsy is not advised, because there is a risk of necrotic tumour leakage from the biopsy site. PET scanning and endoscopic ultrasound may add useful information for the management of some patients.
Stage classification Malignant GIST lesions are not classified using TNM nomenclature. GIST should be classified as localised or metastatic. Lymph node involvement is classified as metastatic.
Treatment Treatment overview Treatment involves staging investigations and discussion at a multidisciplinary team (MDT) meeting to decide operability. If the tumour is operable, then it should be resected following standard oncological principles; if the tumour is not operable, then a biopsy should be performed to confirm the diagnosis. Histopathological review by an experienced pathologist and MDT review should be done to confirm whether the disease is CD117-positive GIST and to assess the tumour’s aggressiveness. If the tumour is fully resected, then follow-up should be done according to assessed risk (see discussion that
follows) and the patient should be considered for entry into adjuvant trials. If the tumour is inoperable or residual disease is present, then start imatinib therapy (if the patient is fit enough). Assess patient response at 3-month intervals and discuss the CT scan in an MDT meeting. Progressive disease should also be discussed at the MDT meeting (see the following section for options). In the case of stable disease or patient response, continue giving imatinib until the disease progresses or becomes operable.
Surgery Surgical resection remains the best hope for a cure. It should be discussed within the relevant MDT (both the upper GI team and sarcoma team; NICE, 2006) and performed by a surgeon who is fully trained and experienced in cancer surgery in the relevant body area. The primary aim is complete resection while avoiding tumour rupture. The approach depends on the presentation. Some evidence exists that preoperative imatinib may be beneficial and trials are likely to open in this area at some point (Eisenberg and Judson, 2004), but currently it should be noted that the role of imatinib in localised disease, and the role of surgery in metastatic disease, is experimental.
Tumours larger than 2 cm Wide local excision is required for tumours larger than 2 cm, with a margin of 1 to 2 cm. When adjacent organs are involved, en bloc dissection is necessary with the help of appropriate specialist surgeons. For tumours of the bowel, surgery will amount to a formal segmental resection with the accompanying mesentery. Since GISTs rarely metastasise to nodes, extended lymphadenectomy is not required.
Small tumours The management of asymptomatic small (≤ 2 cm) tumours is controversial. A proportion of these have a very low risk of malignant transformation and a firm diagnosis is not always made. If these tumours are not resected, they should be re-imaged at 6 and 12 months with a CT scan or endoscopic ultrasound.
Tumours arising outside GI tract The tumour should be dissected from adjacent organs and then excised with a margin of normal tissue. The aim is macroscopic removal of the intact tumour mass regardless of its size. 185
Kate Parker and Tom Crosby
Multiple tumours Patients with more than one primary GIST should be referred to the cancer genetics service for advice, and each tumour should be treated as appropriate.
Emergency presentations Patients may present with an obstruction and need an emergency laparotomy. If the tumour cannot be excised, a bypass procedure should be undertaken and the tumour should be biopsied. Patients then require an appropriate staging scan and an MDT discussion. Patients on imatinib can present with acute gastrointestinal haemorrhage due to tumour response. These patients should be managed intensively because the prognosis of patients on imatinib may be reasonable, with good quality of life.
Advanced disease Surgery may have a role in excision of recurrent disease, debulking of advanced disease, excision of disease fol¨ lowing response to imatinib (Bumming et al., 2003), and procedures for palliation of symptoms.
Radiotherapy GIST is a radioresistant tumour; therefore, there is no role for radiotherapy in primary treatment. Radiotherapy could be considered as a palliative measure in advanced disease (Eisenberg and Judson, 2004).
Cytotoxic chemotherapy GISTs are resistant to conventional cytotoxic chemotherapy but do respond to targeted therapy as outlined in the next section (Eisenberg and Judson, 2004).
Imatinib Indications Indications for use of imatinib are a confirmed KIT(CD117-)positive GIST (Verweij et al., 2003b) and inoperable recurrent or metastatic disease confirmed by the MDT.
Baseline investigations The disease should be assessed with a CT scan to allow accurate monitoring. Informed patient consent is needed (including advice to the patient on avoiding
186
pregnancy and breast feeding). The patient’s medication should be checked for drugs that interact via the p450 system. The patient’s performance status and cardiac status should be assessed and baseline liver and renal function tests and FBC should be performed. The patient’s weight should also be checked.
Treatment The patient should be given imatinib 400 mg p.o. once a day continuously (van Oosterom et al., 2001; Verweij et al., 2003a), which should be taken with a large glass of water (to avoid gastric irritation). The patient should also avoid caffeine and grapefruit for 1 hour before and after receiving the dose and should avoid lying down for 1 hour afterward.
Monitoring Two weeks after starting imatinib, check LFT and FBC, assess toxicity, measure the patient’s weight (to determine fluid retention), and perform a physical and symptomatic assessment of response. Repeat this procedure at 4- to 6-week intervals. Repeat a CT or PET scan after 3 months and discuss the results at an MDT meeting to assess the patient’s response. If there is evidence of response or stable disease then continue treatment according to toxicity parameters. Repeat the assessment with CT every 6 months or earlier if clinically indicated. See van Oosterom et al. (2001) for more information on patient monitoring.
Confirmed disease progression NICE guidance dictates that imatinib should be discontinued when the disease progresses (NICE, 2004). Upon stopping treatment with imatinib, it is important to watch for tumour flare (which suggests that, even with progressive disease, there are some cell populations responsive to imatinib). In patients who have initially responded to imatinib but then become resistant, dose escalation should be discussed at the MDT meeting (dose escalation is controversial; one study showed a 33% response to dose escalation but the median duration of response was only 3 months; Zalcberg et al., 2004). In view of the tumour flare phenomenon and research showing an increase in FDG metabolism of GIST deposits after stopping imatinib treatment, there is some evidence that therapy should continue despite disease relapse. Consider a second-line treatment (e.g. sunitinib).
Gastrointestinal stromal tumours
Sometimes nodules of resistant disease grow in previously controlled metastases. Surgery may be considered for these ‘tumour-resistant clones’ but the benefit of this approach is uncertain.
Assessing response using contrast-enhanced CT A response to therapy can be seen on a CT scan as early as 1 month after treatment starts. GISTs can decrease in size but may also increase in size (because of haemorrhage or myxoid degeneration). The standard RECIST criteria for assessing tumour response are therefore inadequate as the sole method to assess response (Silberman and Joensuu, 2002). As well as changing in size, the tumour can become more homogeneous and hypoattenuated on CT, and tumour vessels and solid enhancing nodules may disappear. Therefore, it is important that the CT scans used to assess response are reported by radiologists experienced in assessing response in GIST, and that they are discussed in the MDT meeting. Criteria to help predict response on CT, which include size and number of tumours, degree and extent of enhancement, change in Hounsfield unit, presence or absence of tumour vessels, and the presence or absence of solid nodules within the tumours, have been evaluated and were found to be better than RECIST criteria for assessing the response of GISTs to imatinib. These new criteria may make the determination of response assessment via CT more accurate.
Role of PET Metabolic responses to imatinib can be reported as early as 24 hours after treatment and changes in FDG-PET images at 1 week have been reported compared to those seen on a CT scan at 2 months (Stroobants et al., 2003). However, FDG-PET is not routinely available in many UK centres and international response criteria for FDGPET have not been standardised. In addition, 20% of GISTs do not take up appreciable FDG at baseline and so cannot be monitored in this way; for these tumours, CT remains the routine form of radiological assessment.
Toxicities Imatinib toxicities include nausea and vomiting (level 1 antiemetics are usually enough to help); diarrhoea and dyspepsia; and oedema, often periorbital (occasionally pleural effusions, ascites). Diuretic therapy is helpful but, if oedema is severe, imatinib treatment should be stopped. Abnormal LFTs should be checked and treat-
ment withheld if levels are three times the upper limit of normal. If a skin rash occurs, it is usually mild and selflimiting, but it can be severe. Occasionally neutropenia can occur (monitor FBC). Gastrointestinal bleeding also occurs, often due to response and tumour regression (van Oosterom et al., 2002).
Special clinical situations Positive surgical resection margins Residual disease should be excised whenever possible. Imatinib is only currently licensed for use in recurrent or metastatic disease that is inoperable.
KIT-negative GIST The histological diagnosis should be confirmed by an experienced pathologist. A proportion of KIT-negative GIST patients still respond to imatinib, and although the mechanism is poorly understood it may be explained by mutational analysis (Blanke et al., 2001). However, the NICE guidance says that imatinib should only be used for KIT-positive GIST. Treatment options should be discussed on a case-by-case basis at the appropriate MDT meeting.
Adjuvant therapy Currently there is no evidence for adjuvant use of imatinib.
Palliation Imatinib is used in the palliative setting as described earlier in the chapter. Where imatinib is inappropriate, or has been discontinued because of progressive disease, patient management involves active supportive care that includes palliation of symptoms. Chemotherapy and radiotherapy are unlikely to be helpful.
Follow-up Routine follow-up after surgery depends on the risk of recurrence (see the prognostic factors in Table 15.3). All patients should have a CT scan at 3 months. Patients with very-low-risk tumours need an annual review and no further CT scan unless clinically indicated. For patients with low-risk tumours, repeat a CT scan at 12 months. Perform an annual review thereafter and a CT scan only if indicated clinically.
187
Kate Parker and Tom Crosby
Table 15.3. Prognostic factors for GIST Mitotic count per 50 Risk group
Size
high-powered fields
Very low risk
< 2 cm
<5
Low risk
2–5 cm
<5
Intermediate risk
< 5 cm
6–10
High risk
5–10 cm
<5
> 5 cm
>5
> 10
Any rate
Any size
> 10
Adapted from Fletcher et al. (2002).
For patients with intermediate-risk tumours, repeat a CT scan at 9 months and then annually for 5 years in addition to an annual clinical review. For patients with high-risk tumours, perform a CT scan every 6 months for 3 years, then annually over the next 5 years. A clinical review should be performed every 6 months. GIST can recur many years after a curative resection.
Prognosis The 5-year survival rate is about 50% and the 10-year survival is about 35%. These figures are for patients treated before imatinib was available and before improvements in diagnostic accuracy by the introduction of CD117 immunohistochemistry. The response rate to imatinib is initially between 80 and 90% (which is derived from a very small number of complete responses, around 67% partial responses and the remainder of patients with stable disease; Benjamin et al., 2003; Demetri et al., 2006; Verweij et al., 2004). The median time to progression on imatinib is around 24 months (Von Mehren et al., 2002), and the median time to achieving CR or PR is 13 weeks (Demetri et al., 2002). The response to imatinib can be predicted in part according to mutational analysis; in one study exon 11 mutation was associated with an 84% partial response rate, compared to 48% for exon 9 and 0% for PDGFRα mutation. Prognostic factors for GIST are shown in Table 15.3. 188
Areas of current interest and clinical trials Clinical trials Trials looking at the role of adjuvant imatinib are currently open and recruiting in both Europe (EORTC and SSG) and the USA (ACOSOG): EORTC 62024 aims for 400 patients to have 400 mg imatinib daily for 2 years following R0 or R1 resection for non-metastatic GIST. ACOSOG Z9001 is a trial involving 732 patients receiving 400 mg imatinib daily for 1 year following R0 or R1 resection for GIST. The Scandinavian trial SSGXVII involves 240 patients on 400 mg imatinib daily following R1 surgery or for a high-risk group for either 1 or 3 years. The Radiation Therapy Oncology Group phase II trial investigates the use of neoadjuvant imatinib to debulk tumours prior to surgery or to make inoperable tumours resectable. Trials are needed to address the following questions: r What is the role of dose escalation of imatinib in patients who initially respond and then progress (median time to progression is 2 years)? A proposed trial involves imatinib 800 mg versus sunitinib. r What is the role of other new molecular therapies? A proposed trial involves imatinib ± bevacizumab. r What is the role of neoadjuvant imatinib? r What is the role of imatinib in KIT-negative tumours?
New drugs: other targeted inhibitors of KIT activity Sunitinib Sunitinib is a small-molecule inhibitor of all VEGFRs, PDGFRα, PDGFRβ, FLT3 and KIT that is being investigated for use in imatinib-resistant patients. A phase III trial has recently been closed early, due to efficacy end points being reached, with a significant benefit for sunitinib in progression-free survival of 6.3 months compared to 1.5 months in the placebo arm of the study (Faivre et al., 2006).
Dasatinib Dasatinib is a synthetic small-molecule inhibitor of tyrosine kinases including KIT, PDGFR, BCR-ABL and SRC. It is active when given orally, and it may have activity in imatinib-resistant patients.
REFERENCES Benjamin, R. S., Ranking, C., Fletcher, C. et al. (2003). Phase III dose-randomized study of imatinib mesylate (STI571) for
Gastrointestinal stromal tumours
GIST: Intergroup S0033 early results. Proc. Am. Soc. Clin. Oncol., Abstr. 3271. Blanke, C. D., Eisenberg, B. L. and Heinrich, M. D. (2001). Evaluation of the safety and the efficacy of an oral molecularly-targeted therapy, STI571, in patients with unresectable or metastatic gastrointestinal stromal tumours (GISTs) expressing c-KIT (CD117). Curr. Treat. Options Oncol., 2, 485–91. ¨ Bumming, P., Andersson, J., Meis-Kindblom, J. M. et al. (2003). Neoadjuvant, adjuvant and palliative treatment of gastrointestinal stromal tumours (GIST) with imatinib: a centre-based study of 17 patients. Br. J. Cancer, 89, 460–4. Corless, C. L., Fletcher, J. A. and Heinrich, M. C. (2004). Biology of gastrointestinal stromal tumors. J. Clin. Oncol., 22, 3813– 25. Demetri, G. D., von Mehren, M., Blanke, C. D. et al. (2002). Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N. Engl. J. Med., 347, 472–80. Demetri, G., van Oosterom, A. T. and Garrett, C. (2006). Improved survival and sustained clinical benefit with SU11248 (SU) in pts with GIST after failure of imatinib mesylate (IM) therapy in a phase III trial. Proc. Am. Soc. Clin. Oncol. Gastrointestinal Cancer Symposium, Abstr. 8. de Silva, C. and Reid, R. (2003). Gastrointestinal stromal tumours (GIST): C-kit mutations, CD117 expression, differential diagnosis and targeted cancer therapy with Imatinib. Pathol. Oncol. Res., 9, 13–19. Duffaud, F. and Blay, J.-Y. (2003). Gastrointestinal stromal tumors: biology and treatment. Oncology, 65, 187–97. Eisenberg, B. L. and Judson, I. (2004). Surgery and imatinib in the management of GIST: emerging approaches to adjuvant and neoadjuvant therapy. Ann. Surg. Oncol., 11, 465–75. Faivre, S., Delbaldo, C., Vera, K. et al. (2006). Safety, pharmacokinetic, and antitumour activity of SU11248, a novel oral multitarget tyrosine kinase inhibitor, in patients with cancer. J. Clin. Oncol., 24, 4–5. Fletcher, C. D., Berman, J. J., Corless, C. et al. (2002). Diagnosis of gastrointestinal stromal tumors. A consensus approach. Hum. Pathol., 33, 459–65. Heinrich, M. C., Corless, C. L., Demitri, G. D. et al. (2003). Kinase mutations and imatinib response in patients with metastatic gastrointestinal stromal tumor. J. Clin. Oncol., 21, 4342–9. Joensuu, H., Fletcher, C., Dimitrijevic, S. et al. (2002). Management of malignant gastrointestinal stromal tumours. Lancet Oncol., 3, 655–64. ¨ Kindblom, L. G., Meis-Kindblom, J., Bumming, P. et al. (2002). Incidence, prevalence, phenotype and biologic spectrum of gastrointestinal stromal cell tumors (GIST) – a population-based study of 600 cases. Ann. Oncol., 13 (Suppl. 5), 157, Abstr. 5770.
NICE. (2004). Technology Appraisal 86. Imatinib for the Treatment of Unresectable and/or Metastatic Gastrointestinal Stromal Tumours. London: NICE. NICE. (2006). Guidance on Cancer Services. Improving Outcomes for People with Sarcoma. The Manual. London: NICE. r in Novartis. (2003). Open, randomized, phase II study of Glivec patients with unresectable or metastatic malignant gastrointestinal stromal tumors expressing c-kit. Report Data on File, Novartis UK Ltd., Study No STI51B2222, 15 December. Silberman, S. and Joensuu, H. (2002). Overview of issues related to Imatinib therapy of advanced gastrointestinal stromal tumors: a discussion amongst experts. Eur. J. Cancer, 38 (Suppl. 5), S66–9. Stroobants, S., Goeminne, J., Seegers, M. et al. (2003). 18FDG-Positron Emission tomography for the early prediction of response in advanced soft tissue sarcoma treated with imatinib mesylate. Eur. J. Cancer, 39, 2012–20. van Oosterom, A. T., Judson, I., Verweij, J. et al. (2001). Safety and efficacy of Imatinib (STI571) in metastatic gastrointestinal stromal tumours: a phase I study. Lancet, 358, 1421–3. van Oosterom, A. T., Judson, I., Verweij, J. et al. (2002). Update of phase I study of imatinib (STI571) in advanced soft tissue sarcomas and gastrointestinal stromal tumors: a report of the EORTC Soft Tissue and Bone Sarcoma Group. Eur. J. Cancer, 38 (Suppl. 5), S83–7. Verweij, J., Casali, P. G., Zalcberg, J. et al. (2003a). Early efficacy comparison of two doses of Imatinib for the treatment of advanced gastrointestinal stromal tumors (GIST): interim results of a randomised phase III trial from the EORTC-STBSG, ISG and AGITG. Proc. Am. Soc. Clin. Oncol., Abstr. 3272. Verweij, J., van Oosterom, A., Blay, J. Y. et.al. (2003b). Imatinib mesylate (STI-571 Glivec, Gleevec) is an active agent for gastrointestinal stromal tumours but does not yield responses in other soft tissue sarcomas that are unselected for molecular target: results form an EORTC Soft Tissue and Bone Sarcoma Group phase II study. Eur. J. Cancer, 39, 2006–11. Verweij, J., Casali, P. G., Zalcberg, J. et al. (2004). Progression-free survival in gastrointestinal stromal tumours with high-dose imatinib: randomized trial. Lancet, 364, 1127–34. Von Mehren, M., Blanke, C., Joensuu, H. et al. (2002). High incidence of durable responses induced by Imatinib mesylate (Gleevec) in patients with resectable and metastatic gastrointestinal stromal tumors (GISTs). Proc. Am. Soc. Clin. Oncol., Abstr. 1608. Zalcberg, J., Verweij, J., Casali, P. G. et al. (2004). Outcome of patients with advanced gastro-intestinal stromal tumours (GIST) crossing over to a daily imatinib dose of 800 mg (HD) after progression on 400 mg – an international, intergroup study of the EORTC, ISG and AGITG. Proc. Am. Soc. Clin. Oncol., Abstr. 9004.
189
16
BREAST Nayyer Iqbal and Peter Barrett-Lee
Incidence and epidemiology Introduction The lifetime risk of breast cancer in women is one in nine. Breast cancer most commonly presents as a lump in the breast, but the use of screening has also allowed very early cancers to be diagnosed before they can be detected clinically. The management of breast cancer has changed dramatically. In the past, patients were treated with radical mastectomy, whereas today, patients are usually treated with breast-conserving techniques – wide local excision and radiotherapy. The past few decades have also seen the wider use and development of systemic therapies: hormonal treatments, chemotherapy and, recently, the monoclonal antibody trastuzumab. As a result, there has been a steady fall in the mortality from breast cancer.
Types of tumour affecting the breast Table 16.1 shows the range of tumours that can affect the breast.
Anatomy The female breast extends from the 2nd to the 6th rib, and it is made up of 15 to 20 lobes which radiate out from the nipple. The nipple is surrounded by the areola. Each breast is divided into a central portion and four quadrants. The upper outer quadrant also contains the axillary tail. The lymphatic drainage from the breast is primarily to the axillary lymph nodes but also to the internal mammary nodes, which lie in the thorax alongside the internal thoracic artery. A few lymphatic channels also communicate with those in the opposite breast and in the abdominal wall. In the male and prepubertal female, the nipple and areola are small, and the breast tissue does not usually extend beyond the areola. 190
Breast cancer is the most common malignancy in women in the UK. More than 40 000 women in the UK develop breast cancer each year (which accounts for 29% of all cancer in women). There are approximately 125 cases of male breast cancer each year in the UK. There is a wide geographic variation in the ageadjusted incidence of breast cancer; occurrence is fivefold higher in Western countries, such as those in North America and Northern Europe, compared to Asian and African countries such as Japan, China, India and Zimbawe. There also appears to be a significant geographic variation in breast cancer mortality; within Europe, the highest mortality is in Eastern followed by Western Europe, the lowest in the Nordic countries (Coleman et al., 2003), although the validity of the comparisons made has been questioned and the reasons for any real differences are unknown. The breast cancer mortality rate has fallen since 1990, with one of the highest rates of decline occurring in the UK (Coleman et al., 2003), and this is believed to be due to screening, improved treatment (improved surgical techniques, adjuvant hormone treatment, increased use of better adjuvant chemotherapy) and the MDT (Thomson et al., 2004).
Risk factors and aetiology Breast cancer occurs predominantly in women and is rare among men; 0.5% of all breast cancers occur in men (0.7% of all male cancers). Breast cancer is also rare among women younger than age 20. The incidence increases with age, doubling every 10 years until menopause, when the rate of increase slows. The risk of disease is 1 in 39 before age 50, 1 in 29 before age 60 and 1 in 12 by age 80. Reproductive factors include a younger age at menarche (if menarche is before the age 11, relative risk [(RR)] = 3.), late natural menopause (if the menopause occurs after age 54, RR = 2) and late age at first birth
Breast
Table 16.1. Tumours that affect the breast Type
Examples
Benign
Fibroadenoma Solitary cyst Intraduct papilloma Adenomas Duct ectasia Epithelial hyperplasia (without atypia) Sclerosing adenosis Radial scar Complex sclerosing lesion Atypical ductal hyperplasia Atypical lobular hyperplasia
Malignant primary
Carcinoma in situ Tis (ductal carcinoma in situ) Tis (lobular carcinoma in situ) Tis (Paget) Invasive carcinoma Invasive ductal carcinoma NOS or NST (75% of cases) Invasive lobular carcinoma (10%) Medullary carcinoma (5%) Tubular carcinoma Mucinous (colloid) carcinoma Cribriform carcinoma Papillary carcinoma Adenoid cystic carcinoma Apocrine carcinoma Secretory carcinoma Squamous cell carcinoma Inflammatory carcinoma Paget’s carcinoma Metaplastic carcinoma Others Phyllodes tumour Sarcomas Vascular tumours Lymphoma
Malignant secondary
Melanoma Lung Ovary Kidney Stomach Thyroid (medullary) Rhabdomyosarcoma (alveolar)
NOS = not otherwise specified; NST = no special type.
(if the first pregnancy is after age 40, RR = 3). Nulliparity has a RR of 3 compared to high parity. Women who undergo bilateral oophorectomy before the age of 35 have a reduced risk (RR = 0.6) compared to women who undergo natural menopause. In benign proliferative breast disease, for atypical epithelial hyperplasia alone, RR = 4. If atypical hyperplasia is found in a woman with a family history of breast cancer in a first-degree relative, RR = 9. Women with palpable cysts, complex fibroadenomas, duct papillomas, sclerosing adenosis, and moderate or florid epithelial hyperplasia may have a slightly higher risk of breast cancer (1.5 to 3 times) than women without these changes. When younger women are exposed to ionising radiation to the chest (e.g. for Hodgkin lymphoma) then RR = 3. For exogenous hormone administration involving hormone replacement therapy (HRT), the Million Women study showed that HRT users were more likely to develop breast cancer than non-users (adjusted relative risk 1.66 [95% CI 1.58–1.75]; p < 0.0001) (Beral and Million Women Study Collaborators, 2003). The magnitude of risk was greater for patients taking oestrogen and progestin in combination than for oestrogen alone (p < 0.0001). This study has been criticised, however, because of its selective population (i.e. women who accepted an invitation for mammographic screening). The HABITS trial (hormonal replacement after breast cancer – is it safe?) showed that the use of HRT in the early years after diagnosis can increase the rate of breast cancer recurrences. In this trial, 434 women were randomised to HRT or no HRT; 26 women in the HRT group developed a new breast cancer event compared with only 7 women in the non-HRT group. The trial was stopped early, after a median follow-up of 2.1 years, because the results suggested an unacceptably high risk for women with previous breast cancer who took HRT (Holmberg et al., 2004). The data for oral contraceptives (OCs) are less consistent than those for HRT. In a large meta-analysis, use of OC within 10 years was associated with a slightly increased risk of having breast cancer diagnosed. The relative risk of breast cancer in current users was 1.24 (Anonymous, 1996). However, the breast cancers that developed were of favourable prognosis and did not affect the disease mortality. If diethylstilbestrol is used during pregnancy, RR = 2. Dietary factors in breast cancer include obesity and alcohol intake. Obesity increases breast cancer risk 191
Nayyer Iqbal and Peter Barrett-Lee
in postmenopausal women by around 50%, probably by increasing the serum concentration of oestradiol and decreasing the serum concentration of sex hormone binding globulin, causing a substantial increase in bio-available oestradiol. Conversely, obesity in premenopausal women reduces breast cancer risk. A moderate alcohol intakes increases breast cancer risk by about 10 to 20%. A high intake of saturated fat or smoking are not yet proven to be risk factors. In socio-economic groups I and II, RR = 2. With regard to inherited susceptibility, the RR is 2 or more when breast cancer occurs in a first-degree relative at a younger age. There are referral criteria for referral from primary care to secondary care and from secondary care to tertiary care (NICE, 2006c). Up to 10% of breast cancer cases in Western countries are thought to result from a genetic predisposition. The types of genetic damage may be activation and overexpression of proto-oncogenes (EGFR [c-erb b1], ERBB2 [c-erb b2, neu or HER-2], BCL2, MYC) or inactivation of tumoursuppressor genes (RB1, TP53 [p53] and NME1 [NM23]). Breast cancer may result from a mutation in one or more critical genes: r The BRCA1 gene is located on chromosome 17q21. BRCA1 mutations account for 2% of breast cancers and increase the chances of breast cancer by 35 to 85%. They are associated with early onset breast cancer. The lifetime risk of ovarian cancer is up to 60%. BRCA1 tumours tend to have aggressive pathologic features. They are typically ER, PgR and HER-2/neu-negative, often with a basal-like phenotype (Sorlie et al., 2003). r The BRCA2 gene is located on a region of chromosome arm13q. The BRCA2 mutation accounts for 1% of breast cancer, and it increases the chance of breast cancer by 20 to 60%. The gene is associated with a 6% lifetime risk of male breast cancer. It is also associated with elevated risk for the development of other cancers, such as prostate, pancreas and bladder cancer, and non-Hodgkin lymphoma. BRCA2 tumours typically express ER and PgR and tend to be of higher grade with less tubule formation. r The TP53 gene is located on chromosome 17p13. An inherited mutation in TP53 (Li-Fraumeni syndrome) is associated with a 50% lifetime risk of developing breast cancer. r The androgen-receptor gene is found on the Y chromosome. Mutation of this gene has been associated with several cases of male breast cancer.
192
Ataxia-telangiectasia (A-T) is a rare autosomal recessive disorder that renders homozygous individuals more vulnerable to cancer because of an elevated risk of mutation. A-T heterozygotes are about 0.5 to 1% of the general population and are probably breast-cancer prone. Cowden’s syndrome (PTEN) is an autosomal dominant condition. Half of affected females have fibrocystic disease of the breast and breast cancer, of which onethird are bilateral. Other syndromes with increased risk of breast cancer include Muir-Torre syndrome and Peutz-Jeghers syndrome. When a patient has previously had breast cancer in one breast, the risk of contralateral cancer is estimated at 0.5 to 1.0% per year. However, the risk is much higher if the first breast cancer occurred at a younger age and in a known gene carrier.
Pathology Almost all breast carcinomas are adenocarcinomas arising from the epithelial cells that line the terminal duct lobular unit (TDLU). The anatomical distribution of breast tumours includes the upper outer quadrant (50%), upper inner quadrant (10%), lower inner quadrant (10%), lower outer quadrant (10%) and central (20%). Diffuse tumours comprise 3%, and 5% are multicentric (i.e. in different quadrants). Multifocal means two or more foci of disease exist in the same quadrant, although these terms are often interchanged. Of the different locations in the breast, tumours in the upper outer quadrant carry the best prognosis. The Nottingham combined histological grading of breast carcinoma (Elston-Ellis modification of the Scarff-Bloom-Richardson system) is based on: r Degree of tubule formation – > 75% gives a score of 1, 10 to 75% scores 2 and < 10% scores 3. r Nuclear pleomorphism – mild, moderate or severe give a score of 1, 2 or 3, respectively. r Mitotic index, which depends on the number of mitoses per ten high-powered fields, scores 1 to 3. The grades are as follow: r Grade I = total score = 3 to 5 points. r Grade II = total score = 6 to 7 points. r Grade III = total score = 8 to 9 points. The oestrogen receptor (ER) and progesterone receptor (PgR) status can be established via immunohistochemistry (IHC) and a semiquantitative measure is
Breast
provided by the ‘H-score.’ About 20% of patients are true ER negative by these criteria (i.e. the H-score is zero). Tumours are ER positive in more than two thirds of postmenopausal patients but in fewer than half of those who are premenopausal. PgRs can be identified in some breast cancers; their presence depends on an intact ER pathway. About 20% of ER-negative tumours are PgR-positive and possible reasons for their ER-negative status include a very low-level expression of ER or a false-negative result. ER-negative PgR-positive tumours account for roughly 5% of all tumours and are likely to be hormone responsive. HER-2/neu expression can be assessed by IHC with an antibody that recognises cell surface receptors (the result is scored –, 1+, 2+ or 3+), by measuring the number of gene copies with fluorescent in situ hybridisation (FISH) or by the level of circulating receptor protein. HER-2/neu positivity is established by either an IHC score of 3+ or a FISH amplification of 2.1 or greater. An IHC of 2+ should be confirmed by FISH.
Lymph nodes To obtain an adequate assessment of nodal staging, histopathological assessment of the excised lymph node is required: only 70% of involved nodes are detectable clinically or radiologically. Involvement of axillary nodes occurs in up to 50% of symptomatic breast cancer and in 10 to 20% of screen-detected breast cancer. More than 90% of women with metastases to the internal mammary nodes have axillary node involvement and, in the remaining 5 to 10%, most have tumours involving the medial half of the breast. Internal mammary nodes have evidence of tumour in 26% of patients with inner quadrant tumours and 15% with outer quadrant tumours. In patients with axillary-node-positive breast cancer, histological sampling of the internal mammary chain reveals occult disease in 20 to 50% of cases.
Spread Routes of cancer spread include direct local spread to the chest wall, lymphatic spread and haemotogenous spread. Lymphatic spread first involves the regional lymph nodes: the ipsilateral axillary (levels I, II and III), infraclavicular, internal mammary and supraclavicular nodes. Any other lymph node metastasis is coded as a distant metastasis. Tumours from any site in the breast can spread to any nodal site. Internal mammary nodes
are more likely to be involved in patients who also have axillary nodal disease. Haematogenous spread moves to bone, lung, liver, brain, skin and so forth.
Screening and prevention Screening The aim of screening is to reduce mortality from breast cancer by detecting and treating it at an earlier stage. In the UK the current target screening age is 50 to 70 years. Women below the age of 50 are not invited for screening because the incidence of cancer is lower and the density of the breast tissue makes cancers more difficult to detect. Women over 70 years are still entitled to breast screening, although they will not be routinely invited to attend. There should be two views on the mammogram: mediolateral oblique and cranio-caudal. Taking two views rather than one increases the detection rate by 43%. During the initial prevalence screen, 10% of women are expected to be recalled because of some mammographic abnormality; after additional mammographic views, clinical assessment and ultrasound, 1.5% will require diagnostic biopsy and about 0.5% will have cancer. Mammography is less effective in women under the age of 35 because the breast is relatively radiodense. With modern screening a radiation dose of less than 1.5 mGy is standard. More information about the UK breast cancer screening programme can be found at www.cancerscreening.nhs.uk/. Women at very high risk of breast cancer, such as those with germ-line mutations in BRCA1 or 2, may opt for breast surveillance from an early age (e.g. 25 to 30 years). Options include monthly breast selfexamination, annual clinical examination and annual mammographic screening with ultrasound.
Prevention Currently there is no proven role for breast cancer prevention in the general population. For women at high risk of breast cancer, options include chemoprevention and prophylactic surgery. Tamoxifen is known to reduce the incidence of breast cancer, but its use is associated with potentially serious adverse effects such as endometrial cancer and thromboembolic disease. Raloxifene, another oestrogen antagonist with fewer endometrial side effects, and aromatase inhibitors are under investigation. Prophylactic mastectomy dramatically reduces the risk of breast cancer.
193
Nayyer Iqbal and Peter Barrett-Lee
Clinical presentation
Investigations
Breast lump
Triple assessment
A malignant breast nodule is usually solitary, unilateral, solid, hard, irregular, non-mobile and non-tender. A painful lump or lumpiness, breast distortion or pain alone should be taken seriously but are less likely to be associated with an underlying malignancy.
After a complete history and thorough clinical examination, triple assessment is performed, which includes clinical examination (score 1 to 5), imaging by bilateral mammography (score R1 to R5) or ultrasonography (score U1 to U5) and cytology (fine needle aspiration cytology) or histology (Tru-cut® biopsy). Core biopsy has the advantage of providing a histological diagnosis and can differentiate between invasive and in situ carcinoma. The ER, PR and HER-2 status can also be tested. However, to be accurate, HER-2 testing should be performed on the surgical sample. Cytology is reported as follows: r C1 – inadequate. r C2 – benign. r C3 – suspicious probably benign. r C4 – suspicious probably malignant. r C5 – malignant epithelial cells. Biopsy is reported as follows: r B1 – normal breast tissue, no lesion to account for imaging findings is present (i.e. probably missed). r B2 – benign. r B3 – benign but may be associated with more malignant lesions nearby (i.e. diagnostic excision is necessary, e.g. atypical ductal hyperplasia or papillary lesions). r B4 – suspicious, probably malignant, but core is crushed or there is too little tumour present (insufficient evidence for malignant diagnosis). r B5 – malignant tumour present, either DCIS or invasive carcinoma. A provisional type and grade is usually given if possible.
Skin changes The skin may show thickening, redness, dimpling, and/or inflammation. Skin dimpling or a change in contour is present in up to 25% of patients with breast cancer. A diffuse dimpling or peau d’orange (like orange peel) is caused by infiltration of the tumour into the subcutaneous lymphatic channels.
Nipple changes Thickening and loss of elasticity, causing flattening or inversion of the nipple, is suspicious if it is a new finding in a breast. A persistent scaly or eczema-like lesion may be an indication of Paget’s disease. Spontaneous discharge in patients older than 50 years of age is likely to be caused by carcinoma.
Regional disease Malignant axillary lymphadenopathy is usually caused by the presence of an ipsilateral breast cancer. In the absence of evidence of another primary source of disease, patients should be managed the same as they would be for the equivalent stage of primary breast cancer where the primary has been locally resected (see Chapter 38, p. 442).
Metastatic disease Signs or symptoms from metastatic disease, most commonly affecting bone, liver, lung, brain and skin, are sometimes the presenting feature of breast cancer.
Screening Breast cancer can present as a result of a primary screening programme in the general population, during follow-up from a previously treated breast cancer or during screening carried out because of a strong family history. 194
Other tests A full blood count, liver function tests and serum calcium level should be taken. Patients with T1 and T2 primary breast tumours have an incidence of metastatic disease of only 2% and ‘routine’ staging for asymptomatic patients is not indicated (SIGN, 2005). In patients with advanced (T3/T4) disease there is an incidence of metastatic disease of 15 to 20%; staging investigations are performed for these patients if it will affect management, including chest Xray, liver ultrasound and bone scan. Even for patients with advanced disease, a ‘routine’ CT scan is not indicated (RCR, 2006). For patients with breathlessness, a CXR is performed first, followed by a CT scan if the CXR is normal and lymphangitis is suspected (RCR, 2006).
Breast
Table 16.2. The TNM classification of breast cancer T Stage
Description
Tis
Tis (DCIS), Tis (LCIS), Tis (Paget)
T1
≤ 2 cm
T1mic
≤ 0.1 cm
T1a
> 0.1 to 0.5 cm
T1b
> 0.5 to 1 cm
T1c
> 1 to 2 cm
T2
> 2 to 5 cm
T3
> 5 cm
T4
a
Chest wall /skin a
T4a
Chest wall
T4b
Skin oedema (including peau d’orange),
T4c
Both 4a and 4b
T4d
Inflammatory carcinoma
ulceration, satellite skin nodules
a
Chest wall includes ribs, intercostal muscles and serratus anterior muscle but not pectoral muscle. Dimpling of skin, nipple retraction, or other skin changes, except those in T4b and T4d, may occur in T1, T2 or T3 without affecting the classification. DCIS = ductal carcinoma in situ; LCIS = lobular carcinoma in situ. Adapted from UICC (2002).
Bone marrow aspiration is performed if there is an unexplained cytopenia or a leucoerythroblastic blood smear. The blood CA15–3, CA27–29 and CEA tumour marker levels can be useful to follow response to treatment in advanced disease.
Staging
sive breast cancer in the ipsilateral breast within 10 years of diagnosis. Because of this, DCIS requires local therapy. The most common mammographic finding in DCIS is branching microcalcification localised to a small region of breast. DCIS accounts for 20 to 25% of all cancer detected by screening. HER-2/neu amplification and TP53 mutation are often found. The prognosis is associated with size, grade, distance to the resection margin and age.
Treatment options in DCIS Wide local excision alone may be an option in very small (< 5 mm), unicentric, low-grade DCIS of the solid, cribriform, or papillary subtypes. Wide local excision with adjuvant whole-breast radiotherapy is the standard treatment for most women with DCIS. Radiotherapy reduces the relative risk of local failure by approximately one-half. The overall survival is not affected. The use of a radiation boost applied to the tumour bed may be considered to maximise local control, especially in patients up to 50 years of age. Mastectomy with or without reconstruction is recommended in patients with widespread disease (disease in two or more quadrants) or in cases in which surgical margins free of carcinoma cannot be obtained. A scoring index for DCIS (Van Nuys Prognostic Index) is shown in Table 16.5, and a suggested management strategy based on the scoring system is shown in Table 16.6 (Silverstein, 2003). However, there is not a consensus on the management of this condition. If a mastectomy is not done, there is a proven risk reduction of recurrent disease (both malignant and premalignant) in patients treated with radiotherapy and/or tamoxifen (in ER-positive cases).
Role of axillary nodal surgery
The staging of breast cancer is determined by the American Joint Committee on Cancer (AJCC) and is a clinical and pathologic staging system based on the TNM classification (UICC, 2002). Table 16.2 shows the T staging for breast cancer, Table 16.3 shows the N and M staging for breast cancer, and Table 16.4 shows the staging groupings for breast cancer.
Axillary dissection or sampling is not recommended in cases of pure DCIS. Sentinel node biopsy is generally not needed in cases of DCIS, unless invasive cancer cannot be definitively excluded, for example when microcalcifications are incompletely removed or if there is extensive microcalcification or multicentric disease and a mastectomy is planned.
Treatment: non-invasive carcinoma of the breast (Tis)
Role of adjuvant radiotherapy
Ductal carcinoma in situ (DCIS) DCIS is a true premalignant condition. Between 30 and 50% of women with untreated DCIS will develop inva-
Radiotherapy has been shown to reduce local recurrence by two- to four-fold after breast-conservative surgery compared with BCT alone but has no effect on overall survival (Bijker et al., 2005; Fisher et al., 1991, 1998). 195
Nayyer Iqbal and Peter Barrett-Lee
Table 16.3. N and M staging of breast cancer N stage
Description
pN stage
Description
N1
Movable axillary
PN1mi
Micrometastasis > 0.2 mm ≤ 2 mm
pN1a
1–3 axillary nodes
pN1b
Internal mammary nodes with microscopic metastasis by sentinel
pN1c
1–3 axillary nodes and internal mammary nodes with microscopic
node biopsy but not clinically apparent metastasis by sentinel node biopsy but not clinically apparent N2a
Fixed axillary
pN2a
4–9 axillary nodes
N2b
Internal mammary
pN2b
Internal mammary nodes, clinically apparent, without axillary nodes
clinically apparent N3a
Infraclavicular
pN3a
≥ 10 axillary nodes or infraclavicular node(s)
N3b
Internal mammary and
pN3b
Internal mammary nodes, clinically apparent, with axillary node(s) or > 3 axillary nodes and internal mammary nodes with microscopic
axillary
metastasis by sentinel node biopsy but not clinically apparent N3c
Supraclavicular
pN3c
M stage
Description
MX
Distant metastasis cannot be assessed
M0
No distant metastasis
M1
Distant metastasis
Supraclavicular
Adapted from UICC (2002).
Role of adjuvant tamoxifen
Table 16.4. Stage groupings for breast cancer Stage
Description
Stage 0
Tis
N0
M0
Stage I
T1a
N0
M0
Stage IIA
T0
N1
M0
T1a
N1
M0
T2
N0
M0
T2
N1
M0
T3
N0
M0
T0
N2
M0
T1a
N2
M0
T2
N2
M0
T3
N1, N2
M0
Stage IIIB
T4
N0, N1, N2
M0
Stage IIIC
Any T
N3
M0
Stage IV
Any T
Any N
M1
Stage IIB
Stage IIIA
a
T1 includes T1mic < 0.1 cm microinvasion. Adapted from UICC (2002).
196
The benefit of tamoxifen in patients with clear resection margins after breast conservation therapy (BCT) and radiotherapy appears small and it is not recommended routinely (Fisher et al., 1999). Ongoing studies are evaluating the role of aromatase inhibitors (AIs) in ER-positive DCIS. The role of tamoxifen for DCIS in the absence of RT is also being evaluated, in the US Intergroup trial RTOG 9804.
Lobular carcinoma in situ (LCIS) LCIS is not thought to be a true premalignant condition but rather a marker of increased risk of breast cancer. It is associated with approximately a 30% lifetime risk of developing an invasive carcinoma. The invasive cancer is usually ductal and may be present in the same or the opposite breast. LCIS is found predominantly in premenopausal women; it is almost always multicentric in the breast and bilateral in about one-third of cases. LCIS cells are commonly ER-positive, whereas the overexpression of
Breast
Table 16.5. A scoring system for DCIS
Table 16.6. Management of DCIS
Score components
Van Nuys
Definition
Marginsa
Prognostic Index
Ten-year local
Score
recurrence rate
Management
1
≥ 10 mm
2
1–9 mm
4–6
3%
3
< 1 mm
7–9
27%
WLE + RT
10–12
66%
Mastectomy
Histological subtype
WLE
1
Grades 1–2, no necrosis
RT = radiotherapy; WLE = wide local excision. Adapted
2
Grades 1–2, necrosis present
from Silverstein (2003).
3
Grade 3
Size 1
≤ 1.5 cm
2
1.6–4 cm
3
> 4 cm
Breast-conserving surgery Breast conservation ther-
Age 1
> 60 years
2
40–60 years
3
< 40 years
a
and areola and it is usually combined with an axillary node dissection (see discussion that follows).
The margin in DCIS is considered negative if it is > 10 mm; it is inadequate if it is < 1 mm. There is no uniform consensus if the margin is 1 to 10 mm. Adapted from Silverstein (2003).
ERBB2 and TP53 are uncommon. LCIS is not clinically palpable or detectable by mammography but is identified incidentally in about 1% of benign breast biopsies. For treatment of LCIS, observation alone is the preferred option because the risk of developing invasive cancer is low (21% over 15 years). The histology of the invasive cancer tends to be favourable. Bilateral mastectomy with or without reconstruction is preferred in special circumstances (such as a strong family history of invasive breast cancer or nongenetic predisposition) because the risk of invasive breast cancer after a diagnosis of LCIS is equal in both breasts.
Management of early breast cancer Surgery in breast cancer Primary disease Modified radical mastectomy Modified radical mastectomy involves removal of the entire breast, nipple
apy (BCT) consists of wide local excision (WLE) and postoperative RT. Systematic reviews have shown similar outcomes for BCT and mastectomy (EBCTCG, 1995; Fisher et al., 2002a; Morris et al., 1997; Veronesi et al., 2002). The NSABP B-06 trial (Fisher et al., 2002a) randomised 1851 women with invasive breast cancer and negative margins to modified radical mastectomy, or lumpectomy plus axillary dissection and RT, or lumpectomy plus axillary dissection alone. There was no significant difference in the OS or DFS at 20 years but there was a significant difference in local recurrence (LR) at 20 years between lumpectomy plus RT and lumpectomy alone (14.3% versus 39.2%; p < 0.001). These results were confirmed in the meta-analysis performed by the EBCTCG (EBCTCG, 1995). The most important surgical factor influencing LR is the completeness of excision. Current practice aims to achieve microscopically disease-free margins of at least 1 mm and it is unacceptable to have tumour cells at the surgical margin without further excision. The acceptable size of the margin is influenced by which margin is the closest and the need to minimise the degree of mutilation (e.g. the deep margin or skin margin cannot be easily increased). In cases of a positive margin, the patient should have a further excision to achieve a clear margin, but if this would lead to an unacceptable cosmetic outcome, or the patient chooses, a mastectomy should be done. It may be a reasonable option to treat selected cases with BCT using a microscopically focally positive margin in the absence of extensive intraductal tumour by RT together with a boost. The absolute contraindications for BCT and RT include:
197
Nayyer Iqbal and Peter Barrett-Lee
r Previous RT to the breast or chest wall. r RT during pregnancy. r Appearance of diffuse suspicious or malignant microcalcifications.
r Widespread disease that cannot be removed by WLE through a single incision to achieve clear margins and a satisfactory cosmetic result. r Positive pathologic margin. Relative contraindications include: r Active connective tissue disease involving the skin (especially scleroderma and lupus). r Tumour larger than 5 cm. r Focally positive margin.
Axillary surgery Until recently, axillary dissection (levels 1 to 3) was the preferred technique for all women with invasive breast cancer and at least 10 lymph nodes were needed for pathologic evaluation to accurately stage the axilla (NIH Consensus Conference, 1991). Axillary node clearance involves clearing the axillary contents from the volume bounded by the axillary skin laterally; latissimus dorsi, teres major and subscapularis posteriorly; the lower border of the axillary vein superiorly; pectoralis muscles anteriorly; and the chest wall medially. The levels of axillary nodes are defined as: level 1, inferolateral to pectoralis minor; level 2, posterior to pectoralis minor; and level 3, superomedial to petoralis minor. In reality, these nodes are in continuity with each other but the concept of axillary node levels is useful when discussing the extent of axillary node surgery. Some centres have chosen to use axillary nodal sampling, involving removal of approximately four level 1 nodes. Although this causes less morbidity, further surgery or postoperative radiotherapy would be needed if involved nodes are found. Increased awareness and screening have resulted in more women being diagnosed with early stage breast cancer and 60 to 80% of patients are node-negative. Lymph node dissection is considered overtreatment and is associated with significant morbidity, such as axillary pain and numbness, decreased range of arm movement, and chronic lymphoedema. Therefore, sentinel lymph node biopsy (SLNB) has emerged as a minimally invasive procedure that accurately assesses axillary node status (Cody, 1999; Mansel et al., unpublished data; Veronesi et al., 2003). The SLN (or nodes) is the first node in the regional lymphatic basin to which a tumour drains. Localisation of SLN is possible by injecting blue dye and a radioactive colloid tracer around the tumour (peritumoural), into 198
the dermis (subdermal) or under the nipple (subareolar), but only peritumoural injections map accurately to the internal mammary lymph nodes. The first nodal station or SLN are detectable either as visible blue nodes or as radioactive nodes detected by a hand probe. The combination of blue dye and radioisotope gives better results than either agent alone. Localisation rates of over 95% and false-negative rates of less than 5% are possible after an initial learning curve. Compared to conventional axillary dissection, at 6 months, overall arm morbidity is less and quality of life is better. However, it is important that SLNB should meet all the following criteria: r It should be performed by an experienced surgical team. r The node should be clinically negative at the time of diagnosis. r The patient should have no previous chemotherapy or hormonal therapy.
Breast reconstruction Breast reconstruction should be offered to women with breast cancer who are undergoing mastectomy. Immediate reconstruction has economic benefits, produces better results, and reduces psychosocial morbidity compared with delayed reconstruction. There is no absolute contraindication for immediate reconstruction. Relative contraindications include ischaemic heart disease, obesity, diabetes, steroid treatment, smoking and metastatic disease. Delayed reconstruction should be considered when there is a concern about tumour clearance or when postoperative RT is planned. Different choices exist for breast reconstruction procedures: r Myocutaneous flap reconstruction, using the latissimus dorsi (LD) muscle or the transverse rectus abdominus myocutaneous flap (TRAM). r An extended LD reconstruction combined with a skinsparing mastectomy, which may or may not require an implant. r Insertion of a breast prosthesis. A tissue expander is gradually inflated. When it is an adequate size it is removed and replaced by a permanent prosthesis which often contains silicone. r The LD muscle and fat without overlying skin can also be used to fill defects in the breast following WLE for large tumours. The reconstruction is done as a second procedure after the histology of the WLE with a complete excision.
Breast
Role of adjuvant radiotherapy
Isodose % 100 95 70 10
Post-BCT radiotherapy Adjuvant whole-breast RT reduces the risk of local recurrence by two-thirds (EBCTCG, 1995; Fyles et al., 2004; Morris et al., 1997; NIH Consensus Conference, 1991; Veronesi et al., 2002). Partial-breast radiation is still investigational. The EBCTCG (2005) has shown that adjuvant RT gives significantly better control of local recurrence with little effect on breast cancer mortality during the first 5 years. During the subsequent decade there is a lower breast cancer mortality rate but in previous analyses this has been balanced by an excess of deaths due to cardiac disease. This has not been seen in recent RT trials using modern treatment techniques. The magnitude of the benefit of RT is greater in node-positive patients. It has been shown in a randomised trial that chemotherapy and tamoxifen are not appropriate substitutes for RT (NSABP B-06, NSABP B-21 and the Scottish trial; Fisher et al., 2002b; Forrest et al., 1996; Wolmark et al., 2000). However, the use of systemic therapy may add to the local control given by RT. It remains uncertain whether RT may be safely omitted for women over 70 years of age with clear margins and who also receive adjuvant endocrine treatment, which is the subject of an ongoing study (Hughes et al., 2004). RT is usually given after the patient has finished adjuvant chemotherapy. Radiation given concurrently with anthracycline-based regimens or taxanes is not recommended. Although some centres avoid the use of concurrent tamoxifen with RT because of theoretical concerns that residual tumour cells may be put into growth arrest and therefore be resistant to radiation, trials have not substantiated this concern (Ahn et al., 2005; Harris et al., 2005). A retrospective review has shown increased risk of LR with a delay in RT (Huang et al., 2003).
Radiotherapy technique: postoperative radiotherapy to breast The patient is positioned supine on an angled ‘breast board’ or flat, depending on the method of localisation (a breast board has the advantage that the sternum is more horizontal, but it may not fit in a narrowbore CT scanner). One or both arms are abducted. For patients with very large breasts, a prone position is useful. Conventional orthogonal X-ray films or CT scanning is used. For postoperative RT to the breast, there is no GTV, the CTV is the glandular breast tissue, and the PTV
Figure 16.1. 3D isocentric radiotherapy plan for postoperative carcinoma of the breast. In addition to the two main tangential fields, small ‘filler fields’ have been added to improve the dose distribution.
is the CTV with a 1 cm margin, usually allowing 5 mm skin sparing. For conventional planning with orthogonal films, glandular breast tissue cannot be visualised, and a target volume is marked with approximate field borders: r Medial – midline. r Lateral – mid-axillary line. r Superior – suprasternal notch. r Inferior – 1 cm below the breast tissue. r Deep – incorporating a maximum of 2 to 3 cm of lung. A plan is provided by the department of physics. Usually two tangential parallel-opposed fields are used. Determining the optimal dose is a matter of ongoing clinical trials. Doses are prescribed to the ICRU reference point. Doses in current use include: r 50 Gy in 25 fractions over 5 weeks. r 40 Gy in 15 fractions over 3 weeks. r 42.5 Gy in 16 fractions over 3.5 weeks. Figure 16.1 shows an RT plan for a postoperative treatment for carcinoma of the breast.
Postmastectomy radiotherapy Although several individual trials have shown a survival advantage for using RT after mastectomy (Overgaard et al., 2004; Ragaz et al., 2005), it is not considered necessary in patients who have a low risk of recurrence. Indications for postmastectomy RT are if the tumour is stage T3 or T4, if four or more lymph nodes are involved, and if there is a positive excision margin (< 1 mm). Ongoing clinical studies will determine the role of this treatment in patients with a lower risk of local recurrence, such as those with one to three axillary nodes involved, multifocal disease and tumours with 199
Nayyer Iqbal and Peter Barrett-Lee
extensive lymphovascular infiltration. A high-grade tumour and young patient age are not sole indications for treatment. The technique for chest wall RT following mastectomy is similar to that for breast RT. The main difference is the clinical target volume. It is difficult to define what this should be but it is usually the chest wall excluding ribs and muscle. In practice, the field borders and doses are similar to those defined for breast RT.
Role of a radiation boost to the breast The role of a radiation boost given after whole-breast RT with external beam RT or a brachytherapy boost is uncertain and is not routinely recommended. It has been associated with a slight improvement in local control but needs to be balanced against a higher morbidity, especially in patients whose risk of recurrence is low. One study, however, showed a significantly reduced risk of LR, especially in patients younger than age 50 who received a boost of 16 Gy to the tumour bed. The 5-year actuarial rates of LR were 7.3% (95% CI 6.8–7.6%) and 4.3% (95% CI 3.8–4.7%) between two groups (p < 0.001; Bartelink et al., 2001). A boost of 10 to 16 Gy is therefore considered in younger patients, particularly in those in whom there is concern regarding the adequacy of surgical clearance.
Radiotherapy to the regional lymphatics RT to the ipsilateral supraclavicular fossa is associated with a reduced risk of disease failure at this site and is used in patients with the highest risk of disease such as those with more than four positive axillary lymph nodes. One anterior field is used, giving the same applied dose as that used for the breast/chest wall therapy. Axillary RT is not necessary after a complete axillary clearance. The lymph node failure rates are very low, and combined treatment to the axilla is associated with a much higher risk of lymphoedema (38.3% after axillary clearance and RT compared with 7.4% after axillary clearance alone; p < 0.001; Fisher et al., 1980; Kissin et al., 1986). RT should only be considered if axillary clearance is not performed (patient refuses or is medically unfit) or for patients with extensive extranodal tumour spread. When the axilla is irradiated, most of the dose is given through an anterior field that also treats the supraclavicular nodes. Delivery of 50 Gy from the anterior field results in a dose of 35 to 37 Gy to the midline axillary structures. A posterior photon field is then used to 200
supplement the mid axilla to the desired target dose. Because of the morbidity of axillary irradiation and because recurrence in the axilla is rare, the usual dose is 40 Gy; higher doses are only used in patients who have extensive axillary soft tissue disease or who have not had axillary dissection. RT to the internal mammary nodes (IMNs) is controversial because recurrence at this site is very rare and most patients at risk receive adjuvant systemic therapy. The IMNs are difficult to treat because their exact position is often uncertain, and the RT fields that include them irradiate more normal tissue, possibly increasing the risk of cardiac complications (Arriagada et al., 1988; Le et al., 1990). The technique for RT to the axilla and supraclavicular fossa depends on whether the breast or chest wall is also treated. Matching diverging field borders can be difficult and requires a gap at the skin surface. The use of half-beam blocking can help to reduce dose inhomogeneities. The position of the lymph nodes varies among patients, and there is a need to develop ways of identifying their position. With conventional localisation, the approximate field borders for the anterior field are as follow: r Medial – 1 cm lateral to the midline. r Lateral – for supraclavicular fossa only, the junction between the medial two-thirds and lateral third of the clavicle; for SCF and the axilla, cover the humeral head. r Inferior – cover the lower border of clavicle. r Superior – 3 cm above the clavicle; the C7/T1 junction. The posterior field for the axilla, if used, is angled to be in line with the rib cage. Doses used are 50 Gy in 25 fractions over 5 weeks and 40 Gy in 15 fractions over 3 weeks.
Other radiotherapy issues Three-dimensional planning using IMRT usually produces a more homogenous dose distribution across the breast. Less irradiation of normal tissue may lead to less morbidity, and ongoing trials will determine whether this justifies the extra time and expense involved in this planning process. Partial-breast irradiation can be done by CT planning, MammoSite® , brachytherapy or intraoperative RT. In accelerated partial-breast irradiation (APBI), RT is only given to the part of the breast closest to the site of the excised tumour. However, because of the lack of data from phase III trials on long-term safety, efficacy and issues about treatment of regional nodes, APBI should
Breast
Table 16.7. Choice of adjuvant treatment in breast cancer Treatment if endocrine
Treatment if endocrine
Risk category
Treatment if endocrine responsive
response uncertain
non-responsive
Low risk
ET or Nil
ET or Nil
Not applicable
Intermediate risk
ET alone or CT → ET (CT + ET)
CT → ET (CT + ET)
CT
High risk
CT → ET (CT + ET)
CT → ET (CT + ET)
CT
CT = chemotherapy; ET = endocrine therapy. Adapted from Goldhirsch et al. (2005).
only be used as part of a clinical trial (Orecchia, 2005; Wallner et al., 2004).
r A high number of involved lymph nodes. r High tumour levels of urokinase-type plasminogen activator inhibitor type 1.
Adjuvant treatment: general For the adjuvant treatment in early breast cancer, international, national and local guidelines should be followed. There are three international guidelines: 1. St. Gallen guidelines were started in 1978 and have been updated every 2 years since 2001; the last update was in January 2005. 2. The National Comprehensive Cancer Network (NCCN) guidelines were started in 1996 and are updated every year. 3. The National Institutes of Health (NIH) guidelines were started in 1980 and are updated every 5 years. In January 2005, the 9th St. Gallen made a fundamental change in the algorithm for the selection of adjuvant systemic therapy for early breast cancer (Goldhirsch et al., 2005). The important points to consider are: r Whether the tumour is endocrine-responsive, endocrine non-responsive or the endocrine response is uncertain. r The menopausal status. r The risk category – low, intermediate or high. r The potential benefit from the different types of adjuvant treatment – chemotherapy, endocrine therapy, trastuzumab and radiotherapy. r Patient preference, comorbidities and performance status.
Endocrine responsiveness In endocrine-responsive breast cancer, the cells express steroid hormone receptors (ER, PgR). The following occur when endocrine response is uncertain: r Low levels of steroid hormone receptor (<10% of cells positive). r Lack of PgR. r HER-2 overexpression.
r Increased proliferation marker. Endocrine-non-responsive breast cancer cells have no detectable expression of steroid hormone receptors.
Risk categories The patient is at low risk if the cancer is node-negative and all of the following features apply: r pT ≤ 2 cm. r Grade 1. r Absence of peritumoral vascular invasion. r HER-2/neu gene neither overexpressed nor amplified. r Age ≥ 35. The patient is at intermediate risk if the cancer is node-negative and at least one of the following features applies: r pT > 2 cm. r Grade 2 to 3. r Presence of peritumoral vascular invasion. r HER-2/neu gene overexpressed or amplified. r Age < 35. r The cancer is node-positive (one to three involved nodes) and HER-2/neu gene is neither overexpressed nor amplified. The patient is considered to be at high risk when the cancer is node-positive, with one to three involved nodes, and the HER-2/neu gene is overexpressed or amplified, or node-positive with four or more involved nodes. The choice of systemic treatment (2005) is shown in Table 16.7.
Role of adjuvant chemotherapy The benefit of adjuvant chemotherapy in terms of OS and DFS depends on the endocrine responsiveness, age, the aforementioned risk categories and the type of 201
Nayyer Iqbal and Peter Barrett-Lee
chemotherapy planned. The relative and absolute benefits can be estimated using systems such as Adjuvant! Online (www.adjuvantonline.com) and must be discussed with patients thoroughly at the time of diagnosis and staging. It is important to understand that the predicted benefits of therapy are estimates and the confidence intervals around such estimates are not usually given. Clearly these will be wider when fewer trial data are available. There is no consensus about how individual chemotherapy drugs should be sequenced (SIGN, 2005). In general, the higher the degree of endocrine responsiveness, the lower the likely benefit from adding chemotherapy. Also, the younger the patient, the higher the likely benefit from adding chemotherapy to other adjuvant treatments. Adjuvant treatment in women over age 70 should be individualised, taking comorbidities into account. The Oncotype DX Recurrence ScoreTM (RS) has been developed in an attempt to further inform the decision regarding the benefit of chemotherapy in node-negative and ER-positive patients for whom genomic information is available. The RS is based on the expression profile from a select panel of 21 genes in the patient’s tumour tissue based on retrospective analysis of two trials (NSABP B-14 and B-20). However, further prospective validation is needed (Paik et al., 2004).
Non-anthracycline non-taxane-containing regimens Chemotherapy with cyclophosphamide, methotrexate and 5-FU (CMF) repeated every 28 days for six cycles (cyclophosphamide 100 mg/m2 orally days 1 to 14, methotrexate 40 mg/ m2 i.v. days 1 and 8, and 5-FU 600 mg/m2 days 1 and 8) reduces the risk of recurrence and death from breast cancer (Bonadonna et al., 1995, 2005). In its most recent publication, the EBCTCG has shown that women receiving chemotherapy had a relative reduction in the risk of breast cancer recurrence by 23% and death by 17% in any year compared with the control group (EBCTCG, 2005).
Anthracycline, non-taxane-containing regimens Examples of non-taxane-containing regimens of anthracycline include: r FAC (5-FU, doxorubicin, cyclophosphamide). r FEC (5-FU, epirubicin, cyclophosphamide). r AC (doxorubicin, cyclophosphamide). r EC (epirubicin, cyclophosphamide). 202
r A→CMF (doxorubicin then cyclophosphamide, methotrexate, 5-FU).
r E→CMF (epirubicin then cyclophosphamide, methotrexate, 5-FU). These regimens are preferred over CMF for use in node-positive disease and HER-2-positive cases. A number of individual studies showed a small survival advantage for anthracycline regimens over that of CMF (Fisher et al., 1990, 2001; Poole et al., 2003). The Oxford overview compared a number of anthracyclinebased regimens with CMF and showed an 11% further reduction in the odds of recurrence (p = 0.006) and a 16% further reduction in the odds of death (p = 0.02; EBCTCG, 2005). Some examples of anthracycline-containing combinations are as follow: r Canadian CEF (5-FU 500 mg/m2 i.v. days 1 and 8, epirubicin 60 mg/m2 i.v. days 1 and 8, cyclophosphamide 75 mg/m2 orally days 1 to 14) repeated every 28 days for six cycles (Levine et al., 1998). r French FEC (5-FU 500 mg/m2 i.v. day 1, epirubicin 100 mg/ m2 i.v. day 1, cyclophosphamide 500 mg/m2 i.v. day 1) repeated every 21 days for six cycles (French Adjuvant Study Group, 2001). r FEC (5-FU 600 mg/m2 i.v. day 1, epirubicin 60 mg/m2 i.v. day 1, cyclophosphamide 600 mg/m2 i.v. day 1) repeated every 21 days for six cycles (Budman et al., 1998). It is not clear which, if any, of these combinations is the most effective.
Anthracycline- and taxane-containing regimens Examples of regimens that use anthracycline and taxane include: r AC→paclitaxel (doxorubicin, cyclophosphamide, then paclitaxel). r TAC (docetaxel, doxorubicin, cyclophosphamide). r FEC→docetaxel (5-FU, epirubicin, cyclophosphamide, then docetaxel). Results from a number of studies have found a modest improvement in DFS but not a consistently meaningful improvement in OS with anthracyline- and taxanecontaining regimens (Buzdar et al., 2002; Henderson et al., 2003; Mamounas et al., 2003; Roche et al., 2004). The BCIRG-001 trial showed that TAC (docetaxel 75 mg/m2 , doxorubicin 50 mg/m2 and cyclophosphamide 500 mg/m2 ) reduced the relative risk of recurrence by 28% and death by 30% compared to FAC (5FU 500 mg/m2 , doxorubicin 50 mg/m2 and cyclophosphamide 500 mg/m2 ; Martin et al., 2005). The 5-year
Breast
DFS was 75 versus 68% and that for OS was 87 versus 81%. On the basis of this trial, NICE has approved the use of TAC as an adjuvant therapy for women with nodepositive breast cancer (NICE, 2006a). It should be noted that this arm of the study was associated with a high incidence of neutropenic sepsis despite routine use of G-CSF. A large UK multicentre study, TACT, randomised more than 4000 patients between the control arm (FEC or E-CMF) and the experimental arm (four cycles of FEC followed by four cycles of docetaxel 100 mg/m2 ). The results of this study are awaited. Another UK study, TANGO, randomised patients to paclitaxel with or without gemcitabine (GT) after four cycles of epirubicin with cyclophosphamide (all every 3 weeks). Again the results are awaited.
Tamoxifen should be given sequentially after adjuvant chemotherapy (Albain et al., 2004a). If tamoxifen is contraindicated, ovarian function suppression (OFS) with goserelin is an acceptable alternative. Tamoxifen should be avoided during pregnancy owing to its teratogenicity. OFS with tamoxifen appears to be as effective as CMF chemotherapy, but in patients receiving chemotherapy and tamoxifen it is not clear if OFS has an additional benefit. However, ongoing trials such as SOFT, TEXT, and PERCHE are addressing the issues of OFS, AIs and chemotherapy in endocrine-responsive breast cancer in young women (Dellapasqua et al., 2005; Pritchard, 2005; see Chapter 3, p. 26). OFS may be discussed individually, especially for very young patients, or for patients of any age who are at high risk, if chemotherapy did not induce OFS.
Dose-dense regimens Examples of dose-dense regimens include dose-dense AC→paclitaxel (doxorubicin, cyclophosphamide, then paclitaxel) and dose-dense A-T-C (doxorubicin, paclitaxel, cyclophosphamide). ‘Dose dense’ describes an increased dose intensity achieved by dosing at more frequent intervals than normal, and using growth factor support to achieve fortnightly cycles. It is not yet clear whether this regimen is more effective than standard regimens (Citron et al., 2003).
Role of adjuvant endocrine treatment Tamoxifen is a selective oestrogen receptor modulator and blocks the action of oestrogen by binding to one of the activating regions of the ER. Tamoxifen is useful in the adjuvant treatment of breast cancer in both pre- and postmenopausal women. The EBCTCG has shown that tamoxifen, when given to women with ERpositive breast cancer over 5 years, reduces the annual recurrence rate by 41% and annual mortality rate by 34% (EBCTCG, 2005). Five years of tamoxifen is more effective than two, and the recently closed trials aTTOM and ATLAS have addressed the question of continuing tamoxifen beyond 5 years.
Premenopausal patients Premenopausal patients whose tumours are not shown to have absent ERs or PgRs should be considered for adjuvant endocrine therapy (SIGN, 2005). They should receive tamoxifen as a standard adjuvant treatment for 5 years.
Postmenopausal patients Oestrogen receptors have two activation domains: AF1 and AF-2. Tamoxifen blocks AF-2 but not AF-1 so there remains oestrogenic activity in some tissues. AIs exert their antioestrogenic effects through inhibition of the aromatase enzyme, which converts circulating androgens to oestrogen. AIs are active in endocrineresponsive tumours and may be beneficial in cases of tamoxifen resistance. Anastrozole and letrozole are nonsteroidal competitive inhibitors of aromatase whereas exemestane is a steroid-irreversible AI and inhibits aromatisation in vivo by 98%. Because of the results of recently reported RCTs, there are now several options for postmenopausal women who require adjuvant endocrine therapy: r In women with low-risk disease or with comorbidities such as osteoporosis or heart disease, tamoxifen should be given for 5 years. r Upfront use of an AI (anastrozole, letrozole) alone for 5 years is based on the results from RCTs. The ATAC trial with 9366 patients (Howell et al., 2005a) showed that 5 years of anastrozole significantly increases DFS compared to tamoxifen and the benefit appears to increase with longer follow-up. But no benefit in OS has yet been seen. At 68-month follow-up, there were fewer recurrences with anastrozole (HR 0.87; p = 0.01). The benefit was largest in those with PgRnegative tumours. Joint, muscle and bone pain, especially bone fractures, were more frequent with anastrozole, whereas gynaecological and vascular events occurred more frequently with tamoxifen. The BIG 1– 98 Trial with 8010 patients (Thurlimann et al., 2005) 203
Nayyer Iqbal and Peter Barrett-Lee
showed that letrozole significantly increases DFS (HR 0.81; p = 0.003) compared to tamoxifen. There was no increased benefit in ER-positive and PgR-negative patients. Again there was no difference in OS. In addition to the aforementioned side effects, cardiovascular events were more common in patients receiving letrozole. r The patient could switch to an AI (exemestane, anastrozole) after 2 to 5 years of tamoxifen to complete 5 years of therapy, based on two RCTs. The IES trial (4742 patients) tested a switch to exemestane after 2 to 3 years of tamoxifen (Coombes et al., 2004). At 3 years there was a highly significant benefit in disease-free survival in the exemestane arm (HR 0.76; p = 0.0001). However, exemestane was associated with higher incidence of arthralgia (5.4 versus 3.6%) and diarrhoea (4.3 versus 2.3%). The ABCSG trial 8 and ARNO 95 trial (Jakesz et al., 2005) – the joint analysis of the Austrian and German trials of 3224 patients – showed that anastrozole treatment after 2 years of tamoxifen gave a significantly better relapse-free survival compared with continuing the patient on tamoxifen (HR 0.60; p = 0.0009). r Another option is to switch the patient to an AI (letrozole) after 5 years of tamoxifen (extended adjuvant treatment). The MA-17 trial (Goss et al., 2005) compared letrozole with placebo after completion of about 5 years of tamoxifen in 5187 women. Letrozole improved DFS over placebo (HR 0.58; p = 0.001). Overall, in postmenopausal women with breast cancer, tamoxifen remains the initial choice of adjuvant treatment unless there are relative contraindications to its use. Consideration should be given to switching to an AI after 2 to 3 years or after 5 years of tamoxifen (SIGN, 2005). This recommendation is supported by the results of a recent meta-analysis that shows that postmenopausal patients who switch to an AI after 2 or 3 years of tamoxifen had improved disease-free survival (HR 0.59 [95% CI 0.48–0.74]; p < 0.0001) and overall survival (HR 0.71 [95% CI 0.52–0.98]; p = 0.04; Jonat et al., 2006). NICE has recently published a guidance for hormonal therapies for the adjuvant treatment of early ER-positive breast cancer. The AIs (anastrozole, letrozole, exemestane) are recommended as options within their licensed indications. The choice of treatment should be made after a discussion between the responsible clinician and the patient about the risks and benefits of each option (NICE, 2006d).
204
Role of adjuvant trastuzumab Trastuzumab is a chimeric antibody (95% human and 5% murine) developed against the HER-2/neu transmembrane epidermal growth factor receptor. The results of five reported RCTs that used trastuzumab as an adjuvant therapy for early breast cancer have shown a dramatic improvement in recurrence-free survival and a 50% overall reduction in the risk of recurrence when trastuzumab was given in combination with or following chemotherapy as shown in Table 16.8 (Joensuu et al., 2006; Perez et al., 2005; Piccart-Gebhart, 2005; Romond et al., 2005; Slamon et al., 2005). The benefit was seen in all analysed subgroups. The risk of cardiac dysfunction associated with adjuvant trastuzumab is acceptable, predictable, avoidable and reversible. The risk is related to both age and baseline left ventricular ejection fraction, but it is not known whether sequential or anthracycline-free regimens present a lower risk of cardiac events. RT can be given before or concurrent with trastuzumab. The optimum duration of treatment with trastuzumab may be known when the results of the 2-year arm of the HERA trial are available. Trastuzumab has now been recommended by NICE for treatment of patients with HER-2-positive tumours in the adjuvant setting (NICE, 2006b).
Management of locally advanced breast cancer Locally advanced breast cancer is defined as inoperable non-metastatic disease and occurs in the following stages: r Stage IIIA (T0–3, N2, M0). r Stage IIIB (T4, N0–2, M0). r Stage IIIC (any T, N3).
Management Women who are fit enough receive primary chemotherapy, which may allow for breast-conserving surgery rather than mastectomy (Hanrahan et al., 2005) and is possible in up to 80% of suitable patients. It should be clear at the outset whether breast conservation is intended, because conservation is not appropriate for patients with central tumours when the cosmetic outcome will be unacceptable, with multifocal tumours, or with inflammatory breast cancer. The rate of radiological complete response (CR) is best assessed using
Breast
Table 16.8. Summary of the trastuzumab trials Trial
Median FU
(no. of patients)
(years)
NSABP B-31
2
(2030) NCCTG N9831
2
(3505)
HERA
1
(5090)
BCIRG 006
2
(3222)
Fin Her
3
(1010)
Cardiac Arms
HR
CHF (%)
death (n)
AC (×4) → P (×4)
1
0.8
1
AC (×4) → P (×4) + Hw (×12) → Hw (×40)
0.48
4.1
0
AC → Pw (×12)
1
0.3
1
AC → Pw (×12) → Hw (×52)
0.87
2.5
1
AC → Pw (×12) + Hw (×12) → Hw (40)
0.64
3.5
0
Chemo
1
0
1
Chemo → H3w × 1 year
0.54
0.6
0
Chemo → H3w × 2 year
n/a
n/a
n/a
AC (×4) → D (×4)
1
0.3
0
AC (×4) → D (×4) + Hw (×12) → H3w (×13)
0.49
1.6
0
DCarbo (×6) + Hw (×19) → H3w (×11)
0.61
0.4
0
D (×3) or Vw (×8) → CEF (×3)
1
1
0
D (×3) + Hw (×9) or V (×8) + Hw (×9) → CEF (×3)
0.42
0
0
A = doxorubicin; C = cyclophosphamide; Carbo = carboplatin; Chemo = standard chemotherapy; CHF = congestive heart failure; D = docetaxel; E = epirubicin; F = 5-FU; H = trastuzumab; HR = hazard ratio for death; P = paclitaxel; V = vinorelbine; w = weekly. Adapted from Joensuu et al. (2006), Perez et al. (2005), Piccart-Gebhart (2005), Romond et al. (2005) and Slamon et al. (2005).
ultrasound ( ± MRI) and is higher than the rate of pathological CR. In the case of radiological CR, RT may be used instead of surgery but it is associated with a higher LR rate. Therefore, treatment is neoadjuvant chemotherapy (usually anthracycline-based ± taxane) followed by: BCT plus ALND plus RT with or without hormone therapy (which is not suitable in inflammatory breast cancer); mastectomy plus ALND with or without RT, hormone or delayed breast reconstruction; or definitive RT to the breast, axilla and SCF, with or without hormone. If there is no response to neoadjuvant chemotherapy, then alternative additional chemotherapy and/or preoperative RT should be considered. Adjuvant hormone treatment in postmenopausal women should be with an AI. Patients with HER-2positive tumours should be considered for trastuzumab. Neoadjuvant endocrine therapy has been evaluated recently in less fit/older patients with endocrinesensitive disease who are unlikely to tolerate chemotherapy. Letrozole is licensed for neoadjuvant therapy (followed by surgery and/or RT) in postmenopausal
women with large operable or locally advanced ERpositive breast cancer. Letrozole gives a higher response rate than tamoxifen in patients with HER-2-positive tumours (Bhatnagar, 2006). Letrozole or anastrozole treatment results in more patients able to undergo BCT, the complete excision rate is better, and the rates of local recurrence are low.
Management of metastatic breast cancer Consideration should be given to repeating ER, PgR and HER-2 assays if biopsy material is available from a metastasis, because the receptor status of the metastasis can be different from that of the primary tumour. The aim of treatment should be to improve survival and to improve quality of life with minimal toxicity.
Primary hormone therapy Hormone therapy alone is often given if the patient has ER/PgR-positive disease, bone/soft tissue disease only, or asymptomatic and minimal visceral disease.
205
Nayyer Iqbal and Peter Barrett-Lee
In postmenopausal women, the use of AIs produces better results than tamoxifen (Mauri et al., 2006). In premenopausal women who have had previous antioestrogen therapy and who are within 1 year of antioestrogen exposure, the preferred second-line therapy is ovarian ablation (by surgery, RT or LHRH agonists) with or without an antioestrogen. In premenopausal women without previous exposure to an antioestrogen, initial treatment with an antioestrogen with or without LHRH agonists is preferred.
Second-line hormone therapy Postmenopausal women whose breast cancer responds to an endocrine manoeuvre should receive additional endocrine therapies for second-line and subsequent therapy as outlined here. After second-line hormonal therapy, little evidence exists to assist in selecting the optimal sequence of hormonal therapy. Options are: r Non-steroidal AIs (anastrozole and letrozole). r Steroidal AI (exemestane). r Pure antioestrogen (e.g. fulvestrant, which lacks the oestrogen agonistic activity of tamoxifen and is a monthly gluteal intramuscular injection. It is at least as effective as anastrozole in patients whose disease progressed on previous endocrine therapy and has fewer side effects [Howell et al., 2005b]). r Progestin (megestrol acetate). r Androgens (fluoxymesterone). r High oestrogen (ethinyl oestradiol). Premenopausal women with ER-positive disease should have ovarian ablation/suppression and then follow postmenopausal guidelines. Patients who develop endocrine-resistant disease, and are suitably fit, should be considered for chemotherapy and/or trastuzumab if the disease is HER-2 positive.
Primary chemotherapy Primary chemotherapy should be considered for suitably fit patients particularly if they have ER/PgRnegative, symptomatic visceral or hormone-refractory disease. Combination chemotherapy is better than sequential single agents but is associated with greater toxicity; sequential single-agent treatments are still considered appropriate. It would be normal to start with an anthracycline if it is not already being used as an adjuvant treat206
ment followed by a taxane single agent or in combination. The choice of regimen depends on the firstline history of received adjuvant treatment. Suitable regimens include FEC, FAC, AC, EC, doxorubicin plus docetaxel/paclitaxel, CMF, docetaxel plus capecitabine, and gemcitabine plus paclitaxel (Leonard and Howell, 2000; O’Shaughnessy et al., 2002). Single-agent treatments include docetaxel, paclitaxel, albumin-bound paclitaxel, capecitabine, vinorelbine, doxorubicin, epirubicin, pegylated liposomal doxorubicin and gemcitabine. Taxanes can be given weekly or every 3 weeks. Other active agents are cisplatin, carboplatin, etoposide, vinblastine and continuous-infusion 5-FU. A regimen of bevacizumab (a humanised monoclonal antibody against VEGF receptor) plus paclitaxel gives a better DFS and OS than paclitaxel alone (Miller et al., 2005). This regimen may be suitable as a first-line chemotherapy for women with recurrent breast cancer but it will need to be reviewed by regulatory bodies such as NICE before decisions to use this combination are possible. Patients should receive appropriate endocrine maintenance treatment between courses of palliative chemotherapy, unless or until endocrine resistance develops.
Targeted therapy Trastuzumab with or without chemotherapy, with or without endocrine therapy, should be considered if the tumour is HER-2-positive and if there is symptomatic visceral or hormone-refractory disease (Albain et al., 2004b; Cobleigh et al., 1999; Marty et al., 2005; Slamon et al., 2001). NICE recommends the use of trastuzumab in patients with HER-2-positive advanced disease in combination with taxotere after anthracycline use or as a single agent after the use of a taxane (NICE, 2002). Single-agent trastuzumab in HER-2-positive metastatic disease gives a 35% clinical response rate as a firstline treatment. There is evidence that trastuzumab plus chemotherapy may have a better PFS and response rate. Combinations are: r Trastuzumab + docetaxel. r Trastuzumab + paclitaxel. r Trastuzumab + vinorelbine. r Trastuzumab + capecitabine. In this situation, trastuzumab is continued as a single agent after a period of combined treatment with a cytotoxic agent. Treatment with trastuzumab should
Breast
be stopped in patients who develop progressive extracerebral disease. Patients on trastuzumab have a relatively high rate of isolated relapse in the CNS because trastuzumab does not effectively cross the blood–brain barrier. Such patients should be considered for radiotherapy and/or further chemotherapy. Their prognosis appears to be much better than for HER-2-negative patients with CNS metastases.
Use of bisphosphonates Women with bone metastases, especially lytic ones, should be given a bisphosphonate, which reduces the frequency of skeletal-related events and relieves pain (Brown and Coleman, 2002). Pamidronate, clodronate, ibandronate (intravenous or oral) and zoledronic acid (intravenous) have all been shown to be effective in patients with breast cancer. Intravenous zoledronic acid is one of the most powerful inhibitors of osteoclast activity and has been shown to be better than pamidronate for controlling malignant hypercalcaemia and skeletalrelated events. However, side effects including flulike symptoms, renal toxicity and abnormal calcium metabolism need to be monitored to decide whether necessary doses should be reduced or alternative bisphosphonates should be used. The patient’s dental hygiene should also be assessed before treatment because there are increasing cases of osteonecrosis of the jaw being reported in patients treated with zoledronic acid. Patients also need oral calcium and vitamin D supplementation.
Other treatments Patients with metastatic breast cancer develop a number of localised problems that can be treated with local RT, surgery or regional chemotherapy (e.g. intrathecal methotrexate for leptomeningeal carcinomatosis).
Management of recurrent disease Half of the patients with locally recurrent disease have evidence of visceral metastasis at presentation, either clinically obvious or on restaging. Locoregional recurrence may present as chest wall, supraclavicular or axillary recurrence. Patients with a locoregional recurrence can still be cured but treating them may be difficult. Those with an isolated chest wall recurrence, particularly after a long disease-free interval, should be treated intensively and,
although the DFS at 5 years may be 25 to 30%, half of the patients will still be alive after 5 years. Management depends on previous treatment: r For patients who had previous BCT, mastectomy is the usual treatment (or further WLE in highly selected cases). r For patients who had previous mastectomy, a complete surgical resection with a clear margin should be aimed for followed by RT (chest wall and areas of lymphatic spread) if it was not given previously. Giving RT with hyperthermia is also a promising approach. After surgery and RT, chemotherapy and endocrine treatment should be considered. Sometimes neoadjuvant chemotherapy or hormone therapy is given prior to surgical excision.
Special clinical situations Pregnancy and breast cancer Breast cancer during pregnancy is uncommon (Eedarapalli and Jain, 2006). Patients may be diagnosed late and often have involved axillary lymph nodes, and larger tumours, which are more likely to be poorly differentiated, ER/PgR-negative and HER-2-positive (slightly, 30%). Late diagnosis may be related to the endocrine effects of pregnancy or missed diagnosis caused by the tumour being difficult to detect within swollen breast tissue. It is not clear that there is any difference in prognosis, stage for stage, compared to women who are not pregnant. Possible investigations include a mammogram, which can be safely performed with appropriate shielding of the foetus; an ultrasound scan of the breast; a chest X-ray with shielding; and an ultrasound scan of the liver. There should be very close liaison with the obstetrician who is monitoring foetal growth and development. Management depends on the timing of the diagnosis. During the first trimester the options are to continue pregnancy and treat with mastectomy and ALND. Adjuvant chemotherapy can begin in the second trimester. Chemotherapy cannot be given during the first trimester because of the high risk of foetal malformations (whereas the risk is only 1.3% in the second and third trimesters). Taxane, trastuzumab (risk of oligohydramnios), adjuvant RT, and adjuvant endocrine treatment can only be given postpartum. Termination, followed by standard treatment, is the other option. 207
Nayyer Iqbal and Peter Barrett-Lee
During the second trimester and early third trimester, the options are mastectomy or BCT plus ALND followed by adjuvant chemotherapy. Chemotherapy should not be given later than 35 weeks to avoid haematological complications at delivery. Anthracyclines and alkylating agents can be used (e.g. in FAC chemotherapy), but methotrexate should be avoided. Adjuvant trastuzumab (risk of oligohydramnios), adjuvant RT and adjuvant endocrine treatment can only be given postpartum. Another option is neoadjuvant chemotherapy followed by postpartum mastectomy or BCT plus ALND with or without adjuvant trastuzumab (risk of oligohydramnios), with or without adjuvant RT and adjuvant endocrine treatment. During the late third trimester, mastectomy or BCT plus ALND can be performed followed by postpartum adjuvant chemotherapy, adjuvant trastuzumab, adjuvant RT, and adjuvant endocrine treatment as indicated.
If the involved NAC is associated with underlying invasive breast cancer, options include: r Mastectomy + axillary staging. r Excision of the breast tumour and excision of the NAC with whole-breast RT. Appropriate adjuvant systemic therapy depends on the histology of the underlying breast cancer.
Male breast cancer The median age of occurrence of breast cancer in men is in the sixth decade; it is usually ER-positive. LCIS is not found in men, and infiltrating lobular carcinoma is unusual. Local recurrence occurs in about 20% of patients. Standard treatment involves modified radical mastectomy with or without adjuvant RT, with or without adjuvant chemotherapy, with or without adjuvant endocrine treatment stage for stage, similar to treatment for female breast cancer.
Paget’s disease Paget’s disease is a rare manifestation of breast cancer (1 to 4% of all breast cancers) characterised by the presence of neoplastic Paget’s cells in the epidermis of the nipple areola complex (NAC). The patient presents with eczema of the areola, bleeding, ulceration and itching of the nipple. Diagnosis is often delayed. The disease is usually unilateral and occurs most frequently in the fifth or sixth decade. Paget’s disease is usually (more than 90%) associated with an underlying breast cancer, which may not be adjacent to the NAC, and about 50% of underlying tumours are palpable. For palpable tumours 90 to 94% are invasive cancers; if not palpable, 66 to 68% are DCIS. Examination and imaging of the breast should be performed and MRI considered if no abnormality is found. Full-thickness skin biopsy of the involved NAC should be performed. If there is only involvement of the NAC with no evidence of underlying malignancy, management options include: r Mastectomy + axillary staging. r Excision of a NAC with whole-breast RT. If the involved NAC is associated with underlying DCIS, options include: r Mastectomy ± axillary staging. r Excision of DCIS + excision of the NAC with wholebreast RT.
208
Bilateral breast cancer Of patients who present with breast cancer, 1% may present with bilateral primaries in the breast. In patients with unilateral breast cancers, the annual incidence of a contralateral primary is about 0.75%.
Inflammatory breast cancer Inflammatory breast cancer accounts for 3% of all breast cancers. It is a rare but virulent form of breast cancer with poor prognosis. It is characterised by diffuse, rapid (within 3 months), brawny indurations of the skin with an erysipeloid edge. The skin changes are caused by extensive involvement of dermal lymphatics. A skin biopsy usually shows dermal lymphatic invasion but a skin biopsy is not essential to establish the diagnosis. Half the patients with inflammatory breast cancer have an underlying mass. Mammograms, ultrasound and, sometimes MRI of the breast, may be required. The aforementioned staging investigations should be carried out to look for metastases. Inflammatory breast cancers tend to be poorly differentiated, often ER- and PgR-negative and more likely to have HER-2 overexpression. Management involves neoadjuvant chemotherapy (16% complete response, 45% partial response) followed by mastectomy and ALND plus adjuvant RT with or
Breast
without adjuvant hormone, with or without adjuvant trastuzumab. In cases of complete response, radical RT may be discussed as an option. In cases of progressive disease, alternative chemotherapy, hormone therapy, and/or RT should be considered.
Prognosis Prognostic factors Regional lymph node status The status of the regional lymph nodes is the most important prognostic factor and is directly related to survival and the best predictor of systemic micrometastases. Although not routinely evaluated, the presence of positive internal mammary nodes is associated with a worse prognosis.
than cancer of no special type. The inflammatory type has a poor prognosis. An extensive intraduct component (EIC) is defined as an infiltrating ductal cancer in which greater than 25% of the tumour volume is DCIS and the DCIS extends beyond the invasive cancer into surrounding normal breast parenchyma. EIC is associated with increased local recurrence after BCT.
Lymphatic or vascular invasion Lymphatic or vascular invasion is present in 25% of breast cancer patients and is associated with a doubling of the rate of local recurrence and high risk of short-term systemic relapse.
HER-2 oncogene
The tumour size directly correlates with 10-year survival, involvement of four or more positive lymph nodes, and the risk of metastasis.
Of breast cancer patients, 20% overexpress the HER2 oncogene. Gene amplification or overexpression of HER-2 carries an adverse prognosis; it is also a predictive factor and patients are more likely to respond to anthracycline-based chemotherapy and to be resistant to tamoxifen.
Tumour grade
Other markers of poor prognosis
The grade of the tumour is an important predictor of both disease-free and overall survival.
Other indicators of poor prognosis include the following: r High proliferation rates (measured by fraction of cells in S-phase, Ki67 and MIB-1 monoclonal antibodies, or bromodeoxyruridine). r Aneuploid as opposed to diploid tumours. r TP53 accumulation, especially in the presence of abnormal expression of other markers such as pglycoprotein, BCL2 and p21 WAF1. r Proteases and second-messenger systems – high concentrations of cyclic AMP, and the presence of cathepsin D, cathepsin B and matrix metalloproteinase 2.
Tumour size
Age Younger women (< 35 years) have a poorer prognosis than older patients with cancer of an equivalent stage. Younger women are at a higher risk of relapse with a relative risk of 1.6 for distant metastases. The rate of local recurrence after 5 years is 17% in patients under age 35 compared with 6% in those over 50.
Hormone receptor status Patients with ER-positive tumours live longer than those with ER-negative tumours and they are more likely to respond to hormonal treatment. Patients with ER- and PgR-negative tumours have a higher complete response rate (> 30%) following neoadjuvant chemotherapy but survival time is shorter (Colleoni et al., 2004).
Histological type Histological type is one of the best predictors of longterm survival. Many of the so-called special types of invasive breast carcinoma (invasive tubular, cribriform, mucinous, papillary, microinvasive, adenoid cystic and medullary) are associated with a much better prognosis
Prognosis The 5-year survival according to stage of the disease is shown in Table 16.9. The Nottingham prognostic index (NI) is calculated as follows: NI = (0.2 × size) + lymph node stage + grade For lymph node stage, score 1 (if N0), score 2 (if one to three LNs are positive), score 3 (if four or more LNs are positive). For grading, score 1 (for grade 1), score 2 (for grade 2) and score 3 (for grade 3).
209
Nayyer Iqbal and Peter Barrett-Lee
Table 16.9. Five-year survival according to stage of
Table 16.10. Ten-year survival according to the
disease
prognostic index
Stage
Survival at 5 years
Prognostic group
NI scoring
Ten-year survival (%)
I
84%
Excellent
≤ 2.4
94
II
71%
Good
2.5 to ≤ 3.4
83
III
48%
Moderate I
3.5 to ≤ 4.4
70
IV
18%
Moderate II
4.5 to ≤ 5.4
51
Poor
> 5.4
19
Adapted from Miller et al. (1994).
NI = Nottingham prognostic index; see Blamey (1996).
The 10-year survival according to the Nottingham prognostic index is shown in Table 16.10.
Ongoing clinical trials Details of current trials in breast cancer can be found at www.ncrn.org.uk. FAST is a prospective randomised clinical trial testing 5.7 and 6.0 Gy fractions of whole-breast RT in terms of late normal tissue responses and tumour control. DCIS II is a randomised trial (currently being set up) that will test observation (no RT) against RT in women with low-risk completely excised ER-positive DCIS of the breast on adjuvant endocrine therapy. BBC is a British breast cancer study of risk factors for contralateral breast cancer. POSH is a prospective study of outcomes in sporadic versus hereditary breast cancer. AZURE attempts to answer whether adjuvant zoledronic acid reduces recurrence in patients with highrisk, localised breast cancer. ZICE is a comparative evaluation of zoledronate (intravenous) versus ibandronate (oral). The BISMARK trial explores the cost-effective use of bisphosphonates in metastatic bone disease using a comparison of bone-marker-directed zoledronic acid therapy versus a standard schedule. TACT 2 is a trial of accelerated adjuvant chemotherapy with capecitabine in early breast cancer. DEVA is a multicentre randomised trial of sequential epirubicin and docetaxel versus epirubicin in nodepositive postmenopausal breast cancer patients. SOFEA is a study of fulvestrant with or without concomitant anastrozole versus exemestane following progression on non-steroidal AIs. TEXT is a phase III trial evaluating the role of exemestane plus GnRH analogue as an adjuvant therapy 210
for premenopausal women with endocrine-responsive breast cancer. HOT is a randomised phase II trial of hyperbaric oxygen therapy for patients with chronic arm lymphoedema following RT for early breast cancer. DIETCOMPLYF explores the role of diet, complementary treatment and lifestyle in breast cancer survival – a study of phytoestrogens in breast cancer.
REFERENCES Ahn, P. H., Vu, H. T., Lannin, D. et al. (2005). Sequence of radiotherapy with tamoxifen in conservatively managed breast cancer does not affect local relapse rates. J. Clin. Oncol., 23, 17–23. Albain, K., Barlow, W., O’Malley, F. et al. (2004a). Concurrent (CAFT) versus sequential (CAF-T) chemo hormonal therapy (cyclophosphamide, doxorubicin, 5-fluorouracil, tamoxifen) versus T alone for postmenopausal, node-positive, estrogen (ER) and/or progesterone (PgR) receptor-positive breast cancer: mature outcomes and new biologic correlates on phase III intergroup trial 0100 (SWOG-8814). Breast Cancer Res. Treat., 88 (Suppl. 1), A-37. Albain, K. S., Nag, S., Calderillo-Ruiz, G. et al. (2004b). Global phase III study of gemcitabine plus paclitaxel (GT) vs. paclitaxel (T) as frontline therapy for metastatic breast cancer (MBC): first report of overall survival. Proc. Am. Soc. Clin. Oncol., 23 (5), Abstr. 510. Anonymous. (1996). Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53 297 women with breast cancer and 100 239 women without breast cancer from 54 epidemiological studies. Lancet, 347, 1713–27. Arriagada, R., Le, M. C., Mouriesse, H. et al. (1988). Long-term effect of internal mammary chain treatment. Results of a multivariate analysis of 1195 patients with operable breast cancer and positive axillary nodes. Radiother. Oncol., 11, 213–22. Bartelink, H., Horiot, J.-C., Poortmans, P. et al. (2001). Recurrence rates after treatment of breast cancer with standard radiotherapy with or without additional radiation. N. Engl. J. Med., 345, 1378–87.
Breast
Beral, V. and Million Women Study Collaborators. (2003). Breast cancer and hormone replacement therapy in the Million Women Study. Lancet, 362, 419–27. Bhatnagar, A. S. (2006). Review of the development of letrozole and its use in advanced breast cancer and in the neoadjuvant setting. Breast, 15 (Suppl. 1), S3–13. Bijker, N., Meijnen, P. H., Bogaerts, J. et al. (2005). Radiotherapy in breast-conserving treatment for ductal carcinoma in situ (DCIS): ten-year results of European Organisation for Research and Treatment of Cancer (EORTC) randomized trial 10853. SABCS 2005, San Anton, Abstr. 7. Blamey, R. W. (1996). The design and clinical use of the Nottingham Prognostic Index in breast cancer. Breast, 5, 156–7. Bonadonna, G., Valagussa, P., Moliterni, A. et al. (1995). Adjuvant cyclophosphamide, methotrexate, and fluorouracil in node positive breast cancer; the results of 20 years of follow-up. N. Engl. J. Med., 332, 901–6. Bonadonna, G., Moliterni, A., Zambetti, M. et al. (2005). 30 years’ follow-up of randomised studies of adjuvant CMF in operable breast cancer: cohort study. B.M. J., 330, 217. Brown, J. E. and Coleman, R. E. (2002). The present and future role of bisphosphonates in the management of patients with breast cancer. Breast Cancer Res., 4, 24–9. Budman, D. R., Berry, D., Cirrincione, C. T. et al. (1998). Dose and dose intensity as determinants of outcome in the adjuvant treatment of breast cancer. The Cancer and Leukemia Group B. J. Natl. Cancer Inst., 90, 1205–11. Buzdar, A. U., Singletary, S. E., Valero, V. et al. (2002). Evaluation of paclitaxel in adjuvant chemotherapy for patients with operable breast cancer: preliminary data of a prospective randomized trial. Clin. Cancer Res., 8, 1073–9. Citron, M. L., Berry, D. A., Cirrincione, C. et al. (2003). Randomized trial of dose-dense versus conventionally scheduled and sequential versus concurrent combination chemotherapy as postoperative adjuvant treatment of node-positive primary breast cancer: first report of Intergroup Trial C9741/Cancer and Leukemia Group B Trial 9741. J. Clin. Oncol., 21, 1431–9. Cobleigh, M. A.,Vogel, C. L., Tripathy, D. et al. (1999). Multinational study of the efficacy and safety of humanized anti-HER2 monoclonal antibody in women who have HER2overexpressing metastatic breast cancer that has progressed after chemotherapy for metastatic disease. J. Clin. Oncol., 17, 2639–48. Cody, H. S. III. (1999). Sentinel lymph-node mapping in breast cancer. Oncology (Williston Park), 13, 25–34. Coleman, M. P., Gatta, G., Verdecchia, A. et al. (2003). EUROCARE-3 summary: cancer survival in Europe at the end of the 20th Century. Ann. Oncol., 14 (Suppl. 5), v128–49. Colleoni, M., Viale, G., Zahrieh, D. et al. (2004). Chemotherapy is more effective in patients with breast cancer not expressing steroid hormone receptors: a study of preoperative treatment. Clin. Cancer Res., 10, 6622–8. Coombes, R. C., Hall, E., Gibson, L. J. et al. (2004). A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. N. Engl. J. Med., 350, 1081–92. Dellapasqua, S., Colleoni, M., Gelber, R. D. et al. (2005). Adjuvant endocrine therapy for premenopausal women with early breast cancer. J. Clin. Oncol., 23, 1736–50.
EBCTCG. (1995). Effects of radiotherapy and surgery in early breast cancer. An overview of the randomized trials. N. Engl. J. Med., 333, 1444–55. EBCTCG. (2005). Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet, 365, 1687–717. Eedarapalli, P. and Jain, S. (2006). Breast cancer in pregnancy. J. Obstet. Gynaecol., 26, 1–4. Fisher, B., Montague, E., Redmond, C. et al. (1980). Findings from NSABP Protocol No. B-04-comparison of radical mastectomy with alternative treatments for primary breast cancer. I. Radiation compliance and its relation to treatment outcome. Cancer, 46, 1–13. Fisher, B., Brown, A. M., Dimitrov, N. V. et al. (1990). Two months of doxorubicin-cyclophosphamide with and without interval reinduction therapy compared with 6 months of cyclophosphamide, methotrexate, and fluorouracil in positive-node breast cancer patients with tamoxifennonresponsive tumors: results from the National Surgical Adjuvant Breast and Bowel Project B-15. J. Clin. Oncol., 8, 1483–96. Fisher, B., Dignam, J., Wolmark, N. et al. (1998). Lumpectomy and radiation therapy for the treatment of intraductal breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-17. J. Clin. Oncol., 16, 441–52. Fisher, B., Dignam, J., Wolmark, N. et al. (1999). Tamoxifen in treatment of intraductal breast cancer: National Surgical Adjuvant Breast and Bowel Project B-24 randomised controlled trial. Lancet, 353, 1993–2000. Fisher, B., Anderson, S., Tan-Chui, E. et al. (2001). Tamoxifen and chemotherapy for axillary node-negative, estrogen receptor-negative breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-23. J. Clin. Oncol., 19, 931–42. Fisher, B., Anderson, S., Bryant, J. et al. (2002a). Twenty-year follow up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N. Engl. J. Med., 347, 1233–41. Fisher, B., Bryant, J., Dignam, J. J. et al. (2002b). Tamoxifen, radiation therapy, or both for prevention of ipsilateral breast tumor recurrence after lumpectomy in women with invasive breast cancers of one centimeter or less. J. Clin. Oncol., 20, 4141–9. Fisher, E. R., Leeming, R., Anderson, S. et al. (1991). Conservative management of intraductal carcinoma (DCIS) of the breast. Collaborating NSABP investigators. J. Surg. Oncol., 47, 139–47. Forrest, A. P., Stewart, H. J., Everington, D. et al. (1996). Randomised controlled trial of conservation therapy for breast cancer: 6-year analysis of the Scottish trial. Lancet, 348, 708–13. French Adjuvant Study Group. (2001). Benefit of a high-dose epirubicin regimen in adjuvant chemotherapy for node positive breast cancer patients with poor prognostic factors: 5-year follow-up results of French Adjuvant Study Group 05 randomized trial. J. Clin. Oncol., 19, 602–11. Fyles, A. W., McCready, D. R., Manchul, L. A. et al. (2004). Tamoxifen with or without breast irradiation in women 50 years of age or older with early breast cancer. N. Engl. J. Med., 351, 963–70.
211
Nayyer Iqbal and Peter Barrett-Lee
Goldhirsch, A., Glick, J. H., Gelber, R. D. et al. (2005). Meeting highlights: international expert consensus on the primary therapy of early breast cancer 2005. Ann. Oncol., 16, 1569–83. Goss, P. E., Ingle, J. N., Martino, S. et al. (2005). Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: updated findings from NCIC CTG MA.17. J. Nat. Cancer Inst., 97, 1262–71. Hanrahan, E. O., Hennessy, B. T. and Valero, V. (2005). Neoadjuvant systemic therapy for breast cancer: an overview and review of recent clinical trials. Expert Opin. Pharmacother., 6, 1477–91. Harris, E. E., Christensen, V. J., Hwang, W. T. et al. (2005). Impact of concurrent versus sequential tamoxifen with radiation therapy in early-stage breast cancer patients undergoing breast conservation treatment. J. Clin. Oncol., 23, 11–16. Henderson, I. C., Berry, D. A., Demetri, G. D. et al. (2003). Improved outcomes from adding sequential paclitaxel but not from escalating Doxorubicin dose in an adjuvant chemotherapy regimen for patients with node-positive primary breast cancer. J. Clin. Oncol., 21, 976–83. Holmberg, L., Anderson, H. and HABITS Steering and Data Monitoring Committees. (2004). HABITS (hormonal replacement therapy after breast cancer – is it safe?), a randomised comparison: trial stopped. Lancet, 363, 453–5. Howell, A., Cuzick, J., Baum, M. et al. (2005a). Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years’ adjuvant treatment for breast cancer. Lancet, 365, 60–2. Howell, A., Pippen, J., Elledge, R. M. et al. (2005b). Fulvestrant versus anastrozole for the treatment of advanced breast carcinoma: a prospectively planned combined survival analysis of two multicenter trials. Cancer, 104, 236–9. Huang, J., Barbera, M., Brouwers, M. et al. (2003). Does delay in starting treatment affect the outcomes of radiotherapy? A systematic review. J. Clin. Oncol., 21, 555–63. Hughes, K. S., Schnaper, L. A., Berry, D. et al. (2004). Lumpectomy plus tamoxifen with or without irradiation in women 70 years of age or older with early breast cancer. N. Engl. J. Med., 351, 971–7. Jakesz, R., Jonat, W., Gnant, M. et al. (2005). Switching of postmenopausal women with endocrine-responsive early breast cancer to anastrozole after 2 years’ adjuvant tamoxifen: combined results of ABCSG trial 8 and ARNO 95 trial. Lancet, 366, 455–62. Joensuu, H., Kellokumpu-Lehtinen, P. L., Bono, P. et al. (2006). Adjuvant docetaxel or vinorelbine with or without trastuzumab for breast cancer. N. Engl. J. Med., 354, 809–20. Jonat, W., Gnant, M., Boccardo, F. et al. (2006). Effectiveness of switching from adjuvant tamoxifen to anastrozole in postmenopausal women with hormone-sensitive early-stage breast cancer: a meta-analysis. Lancet Oncol., 7, 991–6. Kissin, M. W., Querci della Rovere, G., Easton, D. et al. (1986). Risk of lymphoedema following the treatment of breast cancer. Br. J. Surg., 73, 580–4. Le, M. G., Arriagada, R., de Vathaire, F. et al. (1990). Can internal mammary chain treatment decrease the risk of death for patients with medial breast cancers and positive axillary lymph nodes? Cancer, 66, 2313–18.
212
Leonard, R. C. and Howell, A. (2000). A systematic review of docetaxel, paclitaxel, and vinorelbine in the treatment of advanced breast cancer. Adv. Breast Cancer, 2, 1–3. Levine, M. N., Bramwell, V. H., Pritchard, K. I. et al. (1998). Randomized trial of intensive cyclophosphamide, epirubicin, and fluorouracil chemotherapy compared with cyclophosphamide, methotrexate, and fluorouracil in premenopausal women with node-positive breast cancer. J. Clin. Oncol., 16, 2651–8. Mamounas, E. P., Bryant, J., Lembersky, B. C. et al. (2003). Paclitaxel following doxorubicin/cyclophosphamide as adjuvant chemotherapy for node-positive breast cancer: results from NSABP B-28. Proc. Am. Soc. Clin. Oncol., 22 (4), Abstr. 12. Mansel, R. E., Fallowfield, L., Goyal, A. et al. (unpublished data). Sentinel node biopsy versus standard axillary treatment in breast cancer: results of the randomized multicentre ALMANAC trial. Martin, M., Pienkowski, T., Mackey, J. et al. (2005). Adjuvant docetaxel for node-positive breast cancer. N. Engl. J. Med., 352, 2302–13. Marty, M., Cognetti, F., Maraninchi, D. et al. (2005). Randomized phase II trial of the efficacy and safety of trastuzumab combined with docetaxel in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer administered as first line treatment: the M77001 Study Group. J. Clin. Oncol., 23, 4265–74. Mauri, D., Pavlidis, N., Polyzos, N. P. et al. (2006). Survival with aromatase inhibitors and inactivators versus standard hormonal therapy in advanced breast cancer: a meta-analysis. J. Natl. Cancer Inst., 98, 1285–91. Miller, K. D., Wang, M., Gralow, J. et al. (2005). A randomized phase III trial of paclitaxel plus bevacizumab as first-line therapy for locally recurrent or metastatic breast cancer: a trial coordinated by Eastern Cooperative Oncology Group (E2100). SABCS 2005, Abstr. 3. Miller, W. R., Ellis, I. O., Sainsbury, J. R. C. et al. (1994). ABC of breast diseases. Prognostic factors. B.M. J., 309, 1573–6. Morris, A. D., Morris, R. D., Wilson, J. F. et al. (1997). Breast-conserving therapy vs mastectomy in early-stage breast cancer: a meta-analysis of 10-year survival. Cancer J. Sci. Am., 3, 6–12. NICE. (2002). Guidance on the Use of Trastuzumab for the Treatment of Advanced Breast Cancer. Technology Appraisal Guidance 34. London: NICE. NICE. (2006a). Docetaxel for the Adjuvant Treatment of Early Node-Positive Breast Cancer. Part review of NICE Technology Appraisal Guidance 30. NICE Technology Appraisal Guidance 109. London: National Institute for Health and Clinical Excellence. NICE. (2006b). Trastuzumab for the Adjuvant Treatment of Early-Stage HER2-Positive Breast Cancer. NICE Technology Appraisal Guidance 107. London: National Institute for Health and Clinical Excellence. NICE. (2006c). Familial Breast Cancer. NICE Clinical Guideline. www.nice.org.uk/guidance/cg41. London: National Institute for Health and Clinical Excellence. NICE. (2006d). Hormonal Therapies for the Adjuvant Treatment of Early Oestrogen-Receptor-Positive Breast Cancer. London: National Institute for Health and Clinical Excellence.
Breast
NIH Consensus Conference. (1991). Treatment of early-stage breast cancer. J.A.M. A., 265, 391–5. Orecchia, R. (2005). Intraoperative RT to the breast. Breast, 14 (Suppl.), S8. O’Shaughnessy, J., Miles, D., Vukelja, S. et al. (2002). Superior survival with capecitabine plus docetaxel combination chemotherapy in anthracycline-pretreated patients with advanced breast cancer: phase III trial results. J. Clin. Oncol., 20, 2812–23. Overgaard, M., Nielsen, H. M. and Overgaard, J. (2004). Is the benefit of post mastectomy irradiation limited to patients with 4 or more positive nodes, as recommended in international consensus reports? A subgroup analysis of the DBCG 82 b and c randomized trials. ESTRO 2004, Amsterdam, Abstr. 33. Paik, S., Shak, S., Tang, G. et al. (2004). A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N. Engl. J. Med., 351, 2817–26. Perez, E. A., Suman, V. J., Davidson, N. et al. (2005). Further analysis of NCCTG-N9831. Presented at the ASCO Annual Meeting, Orlando, FL (available at www.asco.org.virtualmeeting). Piccart-Gebhart, M. J. (2005). HERA trial. Presented at the ASCO Annual Meeting, Orlando, FL (available at www.asco.org.virtualmeeting). Poole, C. J., Earl, H. M., Dunn, J. A. et al. (2003). NEAT and SCTBG BR9601 phase III adjuvant breast trials show a significant relapse-free and overall survival advantage for sequential ECMF. Proc. ASCO, Abstr. 13. Pritchard, K. (2005). Adjuvant endocrine therapies for pre/perimenopausal women. Breast, 14 (Suppl.), S9. Ragaz, J., Olivotto, I. A., Spinelli, J. J. et al. (2005). Locoregional radiation therapy in patients with high-risk breast cancer receiving adjuvant chemotherapy: 20-year results of the British Columbia randomized trial J. Natl. Cancer Inst., 97, 116–26. RCR. (2006). Recommendations for Cross-Sectional Imaging in Cancer Management. London: Royal College of Radiologists. Roche, H., Fumoleau, P., Spielmann, M. et al. (2004). Five years analysis of the PACS 01 trial: 6 cycles of FEC100 vs 3 cycles of FEC100 followed by 3 cycles of docetaxel (D) for the adjuvant treatment of node positive breast cancer. Breast Cancer Res. Treat., 89, S16. Romond, E., Perez, E. A., Bryant, J. et al. (2005). Joint analysis of NSAPB-B-31 and NCCTG-N9831. Presented at the ASCO Annual Meeting, Orlando, FL (available at www.asco.org.virtualmeeting).
SIGN. (2005). Management of Breast Cancer in Women. A National Clinical Guideline. No. 84. Edinburgh: Scottish Intercollegiate Guidelines Network. Silverstein, M. J. (2003). The University of Southern California/Van Nuys prognostic index for ductal carcinoma in situ of the breast. Am. J. Surg., 186, 337–43. Slamon, D., Eiermann, W., Robert, N. et al. (2005). Phase III randomized trial comparing doxorubicin and cyclophosphamide followed by docetaxel (ACT) with doxorubicin and cyclophosphamide followed by docetaxel and trastuzumab (ACTH) with docetaxel, carboplatin and trastuzumab (TCH) in HER2 positive early breast cancer patients: BCIRG 006 study, presented at SABCS, 2005, Abstr. 1. Slamon, D. J., Leyland-Jones, B., Shak, S. et al. (2001). Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that over expresses HER2. N. Engl. J. Med., 344, 783–92. Sorlie, T., Tibshirani, R., Parker, J. et al. (2003). Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc. Natl. Acad. Sci. USA, 100, 8418–23. Thomson, C. S., Brewster, D. H., Dewar, J. A. et al. (2004). Improvements in survival for women with breast cancer in Scotland between 1987 and 1993: impact of earlier diagnosis and changes in treatment. Eur. J. Cancer, 40, 743–53. Thurlimann, B., Keshaviah, A., Coates, A. S. et al. (2005). A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. N. Engl. J. Med., 353, 2747–57. UICC. (2002). TNM Classification of Malignant Tumours, 6th edn., ed. L. H. Sobin and Ch. Wittekind. New York: Wiley-Liss, pp. 131–41. Veronesi, U., Cascinelli, N., Mariani, L. et al. (2002). Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. N. Engl. J. Med., 347, 1227–32. Veronesi, U., Paganelli, G., Viale, G. et al. (2003). A randomized comparison of sentinel-node biopsy with routine axillary dissection in breast cancer. N. Engl. J. Med., 349, 546–53. Wallner, P., Arthur, D., Bartelink, H. et al. (2004). Workshop on partial breast irradiation: state of the art and the science. J. Natl. Cancer Inst., 96, 175–84. Wolmark, N., Dignam, J., Margolese, R. et al. (2000). The role of radiotherapy and tamoxifen in the management of node-negative breast cancer -1.0 cm treated with lumpectomy: preliminary results of NSABP protocol B-21. Proc. Ann. Meet. Am. Soc. Clin. Oncol. 19, A271. Abstr.
213
17
KIDNEY Jason Lester and John Wagstaff
Introduction
Incidence and epidemiology
Cancer of the kidney represents 3% of adult malignancies. There are about 6500 new cases per year diagnosed in England and Wales, occurring most commonly between the ages of 50 and 70 years. Men are more frequently affected than women. More than 30% of patients present with metastatic disease. The majority of malignant tumours are adenocarcinomas, arising from the proximal renal tubular epithelium. These tumours were previously called hypernephroma because it was believed that they originated from adrenal rests, but they are correctly termed renal cell carcinoma (RCC). Transitional cell carcinoma of the renal pelvis accounts for 5% of all renal malignancies, and is covered separately at the end of the chapter.
Representing 3% of adult malignancies, approximately 6500 new cases of kidney cancer are diagnosed per year in the UK (Cancer Research UK; info.cancerresearchuk. org, accessed September, 2006), resulting in approximately 3500 deaths. Unlike some other cancers, the death rate is still rising. Kidney cancer occurs most commonly in people 50 to 80 years of age. Men are more frequently affected than women; the male-to-female ratio is 5:3. It is the 10th most common cancer in men and 15th most common cancer in women in the UK. Disease incidence has increased over the past 30 years, not only because of an increase in incidental findings.
Renal cell carcinoma Risk factors and aetiology
Types of kidney tumour Kidney tumours can be benign, malignant primary or metastatic. Benign tumours include cysts (simple, complex, multiple), inflammatory (infection, infarction), adenoma, and oncocytoma. Malignant primary tumours include RCC, lymphoma, sarcoma and renal pelvis tumours (5% of malignant renal cancers arise from the renal pelvis, and more than 90% of these are transitional cell carcinoma).
Anatomy The kidneys are retroperitoneal structures that lie between the 11th rib and the transverse process of the 3rd lumbar vertebral body, each approximately 11 cm in length, the right lying slightly lower than the left. Each kidney is surrounded by perinephric fat which in turn is covered by Gerota’s fascia. The right kidney abuts the liver and stomach and the left, the spleen, stomach, and pancreas. The lymphatics drain along the renal vessels, on the right draining to paracaval and aortocaval nodes, and on the left to the para-aortic region. 214
Risk factors for RCC include smoking (up to one-third of cancers may be due to smoking, and there seems to be a dose-response effect); obesity, especially in women; and use of phenacetin analgesics. Acquired cystic kidney disease, which occurs in nearly half of patients on dialysis, is also a factor. The risk is up to 30 times greater in dialysis patients with cystic changes compared to the general population. Also thorotrast, which was used as a contrast agent in the 1920s, is associated with hepatocellular carcinoma and cholangiocarcinoma as well as RCC. Occupational factors include leather tanning, shoe working and asbestos exposure. Genetic risk factors for RCC include Von HippelLindau disease, tuberous sclerosis, and adult polycystic disease. In Von Hippel-Lindau disease, the gene involved is on the short arm of chromosome 3; this disease occurs in 1 in 36 000 births and is associated with clear cell adenocarcinomas, which are often multiple and bilateral. There are two subtypes: type I, without phaeochromocytoma, and type II, with phaeochromocytoma. Tuberous sclerosis and adult polycystic disease both involve autosomal dominant inheritance; their
Kidney
Table 17.1. The Heidelberg classification of renal cell tumours Type
Examples
Benign
Metanephric adenoma and metanephric adenofibroma Papillary adenoma Renal oncocytoma
Malignant
Common or conventional (clear cell) RCC: most common form of RCC, comprising 70–75% of cases Papillary (formerly chromophilic) carcinoma: 10–15% of cases Chromophobe carcinoma: 2–5% of cases Collecting duct carcinoma: 1% of cases RCC, unclassified Sarcomatoid differentiation can occur
because it varies among tumours and it does not correlate well with prognosis. Grade I tumours demonstrate small, uniform nuclei without nucleoli. Multinucleated giant tumour cells are seen only in grade IV tumours, which occasionally may demonstrate spindling and severe nuclear anaplasia resembling a sarcoma. When tumour heterogeneity is present, the highest grade is always assigned.
Cytogenetics Histological differentiation of benign from malignant primary renal tumours can be difficult. Cytogenetic analysis can be a useful aid. For example, clear cell carcinoma is characterised by loss of part of the short arm of chromosome 3. Regions that are frequently lost are 3p12–14, 3p21 and 3p25. Other aberrations include trisomy of chromosome 5 (Infobiogen; www.infobiogen .fr/services/chromcancer/Tumors/RenalCellCarcinID 5021.html, accessed September 2006).
in both clear cell and papillary variants of renal cancer; it was previously considered to be a separate entity but is not RCC = renal cell carcinoma. Adapted from Kovacs et al. (1997).
disease incidences are 1 in 10 000 and 1 in 1000, respectively.
Pathology Classification of renal cell tumours The major morphological classifications discriminate eight subtypes of primary renal cell tumours that are related to the basic cell types of the nephron from which they are derived. The Heidelberg classification is shown in Table 17.1 (Kovacs et al., 1997). The pathological features of RCC are shown in Table 17.2.
Grade The Fuhrman nuclear grading system is widely used (Fuhrman et al., 1982). Several studies involving large numbers of patients have shown good correlation with survival (Murphy et al., 1994). Four grades are recognised based on nuclear size, nuclear contour, and the presence of nucleoli. Mitotic activity is not considered
Spread Renal cancer can spread locally, via lymphatics, and via the blood. Local spread is to the adrenal gland, renal vein, inferior vena cava, Gerota’s fascia, and perinephric tissue. It spreads to lymph nodes at the renal hilum, abdominal para-aortic, and paracaval regions, and via the blood to lung, bone, soft tissues, the liver, the CNS, and skin.
Clinical presentation The tumour may remain clinically silent until metastases develop. The classic triad of pain, flank mass, and haematuria occurs in only 19% of patients and predicts a poor prognosis. However, 50% of patients will have two of these symptoms. Other symptoms include fever, sweats, weight loss, malaise, and those secondary to metastatic spread such as bone pain. A varicocele can occur in up to 2% of males due to obstruction of the testicular vein. This will be left-sided because of the different anatomy of the left and right renal veins.
Paraneoplastic syndromes Paraneoplastic syndromes include hypercalcaemia (PTH-related peptide), polycythaemia (EPO-like molecule), hypertension (renin), and hepatic dysfunction (unknown mechanism). 215
Jason Lester and John Wagstaff
Table 17.2. Pathological features of renal cell carcinoma Features
Description
Macroscopic
Clear cell carcinoma typically arises from the cortex; solid, lobulated and yellow; the average size at diagnosis is 7 cm; a pseudocapsule is often present; haemorrhage, necrosis and/or calcification are common; the tumour may contain single or multiple fluid-filled cysts
Microscopic
Clear cell carcinoma: clear cytoplasm, condensed hyperchromatic nuclei Papillary carcinoma: centrally located small nuclei and ground glass eosinophilic staining cytoplasm; papillary architecture; there are types I and II with different molecular defects Chromophobe carcinoma: polygonal tumour cells with a transparent, reticulated cytoplasm; cytoplasm crowded with glycogen deposits Collecting duct carcinoma: tubular growth pattern; basophilic cytoplasm, anaplastic nuclei
Investigation and staging Differentiating benign from malignant tumours There are no reliable criteria to distinguish benign renal adenomas from RCC. In 1950, a postmortem study reported that metastases were rare when renal tumours were less than 3 cm in size, and these tumours should be considered benign adenomas (Bell, 1950). With the advent of CT scans, an increasing number of small tumours that had already metastasised were reported; therefore, tumour size is no longer considered a reliable criterion to distinguish RCC (Aso and Homma, 1992). In addition, histological features of adenoma and RCC overlap, and so a biopsy may be misleading. Abdominal ultrasound and CT detect incidental renal cysts in a significant proportion of people. Postmortem results have shown that approximately 50 percent of people over age 50 have one or more renal cysts, and other studies indicate that almost one-third have at least one renal cyst that is identifiable on CT (Tada et al., 1983). In 1986, Bosniak created a four-part classification of cystic renal masses that are found on CT scans (Bosniak, 1986). The system uses Hounsfield units to categorise these lesions in order of increasing probability of malignancy. This classification system has been shown to be useful in separating tumours requiring surgery from those that can be safely followed up.
pulmonary metastases. MRI is useful for imaging the vena cava. A bone scan, full blood count and a biochemical profile, including serum calcium and alkaline phosphatase, are also useful. A renogram (DMSA or MAG 3) should be performed if renal impairment is involved.
Staging classification The regional lymph nodes are the hilar, abdominal paraaortic, and paracaval nodes. Tables 17.3 and 17.4 show the TNM classification and stage groupings for RCC.
Treatment overview Localised or locally advanced resectable disease (stages I to III) Surgical resection is the treatment of choice where appropriate. Adjuvant treatment is under investigation and should not be used routinely outside clinical trials.
Inoperable or metastatic disease Treatment for inoperable or metastatic RCC is individualised. Consider palliative nephrectomy, metastectomy, cytokine therapy or palliative radiotherapy.
Surgery
Staging investigations
Radical nephrectomy
Renal ultrasound can be useful in evaluating questionable cystic renal tumours if CT imaging is inconclusive. Contrast-enhanced CT scans of the thorax, abdomen and pelvis are used to look at perirenal extension, renal vein/caval involvement, lymph node enlargement and
Surgical resection is the only potentially curative therapy for RCC. Radical nephrectomy includes removal of the kidney, adrenal gland and perirenal fat within Gerota’s fascia, with or without a regional lymph node dissection. Surgery can be extended to remove the
216
Kidney
Table 17.3. TNM staging of renal cell carcinoma
Table 17.4. Stage grouping of renal cell cancer
Stage
Description
Stage
Description
T1
≤ 7 cm, confined to the kidney
I
T1 N0 M0
T1a
≤ 4 cm
II
T2 N0 M0
T1b
> 4 cm and ≤ 7 cm
T2
> 7 cm limited to the kidney
III
T3 N0 M0 or
T3
Extends into major veins or directly invades adrenal gland or perinephric
T1–3 N1 M0 IV
Any T N2 M0 or
tissues but not beyond Gerota’s fascia T3a
Any T Any N M1
Directly invades adrenal gland or perinephric tissues but not beyond
T4 N0–1 M0 or
Adapted from UICC (2002).
Gerota’s fascia T3b
Grossly extends into renal vein or vena
T3c
Tumour grossly extends into vena cava or
T4
Directly invades beyond Gerota’s fascia
N0
No regional nodal metastases
N1
Single regional node
N2
More than one regional node
Palliative nephrectomy
M0
No distant metastasis
M1
Distant metastasis
Palliative nephrectomy can be performed if the burden of metastatic disease is small and the patient is fit. It may alleviate pain and bleeding and may improve paraneoplastic syndromes such as hypercalcaemia. Spontaneous regression of metastases following surgery occurs in fewer than 1% of cases (Montie et al., 1977). Two randomised trials have examined the role of nephrectomy in patients who were subsequently treated with interferon for metastatic disease (Flanigan et al., 2001; Mickisch et al., 2001). Both of these trials showed significantly better survival for patients undergoing nephrectomy. Nephrectomy should, therefore, be performed in all patients who are to undergo immunotherapy, provided that their performance status is adequate.
cava or its wall below the diaphragm its wall above the diaphragm
Adapted from UICC (2002).
tumour or tumour thrombus from the infra- and supradiaphragmatic vena cava and, rarely, the right atrium; cardiopulmonary bypass is required for the latter. In patients with bilateral stage I tumours (concurrent or subsequent), bilateral partial nephrectomy or unilateral partial nephrectomy with contralateral radical nephrectomy, when technically feasible, may be a preferred alternative to bilateral nephrectomy with dialysis or transplantation. Increasing evidence suggests that a partial nephrectomy is curative in selected cases (tumours < 4 cm in size).
Laparoscopic (partial) nephrectomy Laparoscopic nephrectomy is a less invasive procedure. Laparoscopic surgery tends to incur less morbidity and is associated with a shorter recovery time and less blood loss. The need for pain medications is reduced, but operating room time and costs are higher. Disadvantages include concerns about spillage, limited experience, and technical difficulties in defining surgical margins. A transperitoneal or retroperitoneal approach may be used. Studies of laparoscopic nephrectomy have now
demonstrated clearly that the local recurrence rate (i.e. secondary to spillage) is no lower with this approach, and survival is equivalent to an open operation. Many believe that this approach should be used wherever possible.
Metastectomy Solitary metastases can be resected, and long-term survival has been reported. As expected, patients with a long disease-free interval following surgery survive longer than those presenting with a solitary metastasis synchronous with a primary lesion (O’Dea et al., 1978).
Radiotherapy Preoperative radiotherapy Studies of preoperative radiotherapy have not shown any evidence of benefit. It is not used in routine clinical practice. 217
Jason Lester and John Wagstaff
Adjuvant radiotherapy Two old randomised trials showed no benefit from postoperative radiotherapy (Finney, 1973; Kjaer et al., 1987). Fatal liver damage was reported and the complication rates were high. These trials predated conformal therapy, however. Adjuvant radiotherapy should not be used in routine clinical practice.
Radiotherapy for metastases Palliative radiotherapy can alleviate the symptoms of metastatic disease, particularly bone pain. Bone lesions from kidney cancer are lytic and can be aggressive. An 8 Gy single fraction or 20 Gy in five fractions are suitable dose fractionations.
Palliative radiotherapy to the primary tumour Radiotherapy plays no role in the curative treatment of RCC. It can be used to control bleeding, but results are often disappointing when it is used for pain relief.
Palliative radiotherapy technique The patient lies supine on the X-ray simulator couch with arms by sides, knees supported by a polystyrene wedge and without further immobilisation. In defining the target volume, intravenous contrast is given to outline the kidney. Treatment field edges are positioned to cover part or all of the affected kidney. (Make sure you are treating the correct kidney.) For large tumours, CT planning allows more accurate delineation of the disease and can reduce the dose to normal tissues. Use parallel-opposed anterior and posterior fields, 6 to 10 MV photons, on a linear accelerator. The dose is 20 Gy in five fractions over 1 week prescribed to the midplane. Side effects of radiotherapy include tiredness, nausea (give prophylactic antiemetics) and diarrhoea.
Biological treatment Adjuvant immunotherapy There is no evidence that adjuvant interferon α (IFNα) or interleukin-2 (IL-2) monotherapy improves survival following potentially curative radical nephrectomy (Clark et al., 2003; Messing et al., 2003). The EORTC 30955 trial has randomised high-risk postoperative patients (T3+ and/or N1+) to the Atzpodien regimen (see discussion later in this chapter) or to surveillance. This study is closed to recruitment and results are awaited. HYDRA is an ongoing NCRN study of similar design.
218
Adjuvant therapy is not currently used in routine clinical practice.
Adjuvant vaccine therapy A German trial randomised 558 radical nephrectomy patients to autologous renal tumour cell vaccine or surveillance and reported a modest improvement in progression-free, but not overall, survival with the vaccine (Jocham et al., 2004). This expensive treatment (18 000 Euros) is not widely used in routine clinical practice.
Palliative cytokine therapy: interferon α At present, cytokine therapies are the basis of current systemic treatment, but results are disappointing, showing a limited antitumour effect. An overview, published in the Cochrane library (Coppin et al., 2005), demonstrated that immunotherapy (interferon and/or IL-2) produced an objective response rate (RR) of 12.9% (CR 3.6%). This was compared with an RR of 2.5% in ten non-immunotherapy arms and 4.3% in two placebo arms. The median survival was 13.3 months, which was a 3.8-month improvement compared to non-immunotherapy study arms. In four studies (644 patients) the odds ratio for death at 1 year was significantly reduced by interferon therapy (OR 0.56; 95% CI 0.4–0.77). In this analysis the addition of anything to single-agent interferon did not significantly affect the outcome. Several prognostic scoring systems have been developed that can help select patients for treatment with immunotherapy (Mekhail et al., 2005). Five factors (Table 17.5) were used to divide patients into three prognostic groups: favourable (19% of patients), intermediate (70%) and poor (11%). The median survival of these groups was 28.6, 14.6 and 4.5 months, respectively. The addition of two other prognostic variables (prior radiotherapy and none or one versus two or more metastatic sites) to these initial five factors divided the patients into favourable (37%), intermediate (35%) and poor (28%) with median survival times of 26, 14.4 and 7.3 months, respectively. Based on these numbers, interferon therapy should not be offered to patients in the poor prognostic group. These prognostic groupings have been used to select patients for clinical trials of the new targeted agents that are discussed later in this chapter. No definite dose-response relationship has been proven for interferon therapy. Because tachyphylaxis usually develops, a widely used protocol is to begin with 3 million units subcutaneously three times weekly,
Kidney
Table 17.5. Prognostic variables for renal cell cancer Factors
Poor prognosis
KPS
< 80
Time from diagnosis to
< 12
treatment with interferon α (months) Haemoglobin
< lower limit of laboratory’s reference range
Lactate dehydrogenase
> 1.5× upper limit of laboratory’s reference range
Corrected serum calcium (mg/dl)
> 10.0 (equivalent to > 2.5 mM/l)
KPS = Karnofsky performance status. Risk groups are defined as follows: favourable, no poor prognostic factors present; intermediate, one or two poor prognostic factors present; poor, more than two poor prognostic
IX (CAIX) have an almost 40% chance of long-term survival and possible cure. High-dose bolus IL-2 is the only therapy that has been shown to cure metastatic renal carcinoma. Patients who are young, have an excellent performance status, and whose tumours stain positively for CAIX should be referred for this therapy.
Palliative combination cytokine therapy To date, there is no evidence that IFNα/IL-2 combinations with or without conventional chemotherapy show a survival advantage over monotherapy. High response rates have been reported with the Atzpodien regimen (IFNα, IL-2 and 5-FU), but the treatment is associated with significant toxicity and two other phase II trials failed to confirm the high response rate. A small randomised trial has shown a survival advantage of using the Atzpodien regimen compared to tamoxifen in metastatic RCC (Atzpodien et al., 2001). This regimen is currently being compared with single agent IFNα in a randomised trial in advanced RCC (REO4).
factors present. Adapted from Mekhail et al. (2005).
Chemotherapy escalating to 6 and then 9 million units at fortnigtly intervals. Treatment should be continued until progressive disease is documented. The median time to response is 6 months. Side effects include influenza-like symptoms, lethargy and malaise, nausea, anorexia and depression. Rarely, suicidal behaviour; cardiovascular, renal and hepatotoxicity; hypersensitivity reactions; and hypertriglyceridaemia can occur. Auto-immune thyroid disease also occurs frequently, but this has been shown to be a good phenomenon because its presence predicts a better outcome.
Palliative cytokine therapy: interleukin-2 In the Cochrane overview, IL-2 seems to incur a response rate and median survival similar to that of IFN-α. High-dose therapy is associated with higher complete response rates than IFN (7 to 8%). These responses have been shown to be very durable, with 80 to 85% of patients being alive after 10 years and probably cured of their disease. Side effects include flu-like symptoms, nausea, skin rashes, diarrhoea, capillary-leak syndrome, renal impairment, bone marrow suppression and, rarely, cardiac toxicity and confusion. One recent study (Atkins et al., 2005) has shown that patients whose tumours stain positively for the expression of carbonic anhydrase
RCC is chemoresistant. Conventional chemotherapy has not been shown to give a survival advantage, and response rates are poor (Yagoda et al., 1995). It should not be used in routine clinical practice outside a clinical trial.
Hormonal treatments Palliative progesterones Progesterones were used for many years based on responses seen in animal models and a lack of any other effective systemic therapy. Response rates are 0 to 2%, with no proven clinical benefit (Kjaer, 1988). Patients may, however, have increased appetite and an improved feeling of well-being when on these agents. This treatment is no longer recommended.
Palliative tamoxifen Tamoxifen may have a low level of activity in high doses (100 mg/m2 per day), but it is not used in routine clinical practice in the UK.
Other treatments In patients who are not candidates for surgery, arterial embolisation or radiofrequency ablation can provide
219
Jason Lester and John Wagstaff
palliation. The beneficial effect on longer-term disease control and palliation is uncertain.
Prognosis Prognostic factors Survival is dependent on the stage and grade of the tumour. Following radical nephrectomy, renal vein involvement alone does not seem to adversely affect 5-year survival but does reduce 10-year survival. In metastatic disease, increased survival is associated with good PS, absence of weight loss, presence of only pulmonary metastasis, removal of the primary tumour and a long disease-free interval between nephrectomy and the appearance of metastases.
Overall survival Five-year survival by stage is r I – 60 to 90%. r II – 50 to 70%. r III – 20 to 60%. r IV – 0 to 20%. and 5-year survival (M0 disease) by Furhman grade is r I – 90%. r II – 90%. r III– 55%. r IV – 33%.
Areas of current interest Tyrosine kinase inhibitors (TKI) Sunitinib (SU011248, Pfizer) and sorafenib (BAY 43– 9006, Bayer) are oral multitargeted TKIs which show promise in RCC. They have both direct antitumour effects and antiangiogenic activity, with inhibition of VEGF and PDGF receptors. Phase II studies using sunitinib in patients failing prior cytokine therapy have shown overall response rates of up to 40% (Motzer et al., 2006b). The drug appeared to be well tolerated and side effects included fatigue, diarrhoea, nausea and stomatitis. A phase III trial randomising patients to sunitinib or IFN has shown superiority in terms of response rate, progression-free and overall survival (Motzer et al., 2006a). Sorafenib has been tested in a phase II randomised placebo-controlled trial in patients who had failed at least one prior systemic therapy (Escudier et al., 2005). Progression-free survival was significantly prolonged compared to placebo (24 versus 12 weeks; p < 0.000001) and overall survival data is awaited. Side effects are simi220
lar to those of sunitinib. Both of these agents are licensed in the UK for use after cytokine failure but have not yet been evaluated by NICE. Temsirolimus is an mTOR inhibitor that has been tested in a randomised trial comparing temsirolimus alone, IFN alone and the combination as a first-line therapy in patients with a poor prognosis according to the Motzer scoring system (Hudes et al., 2006). Median survival was 7 months for IFN alone and 11 months for the temsirolimus-alone arm (p = 0.0069). There was no statistically significant difference in overall survival between the two-drug combination and IFN alone. Temsirolimus has been licensed as an orphan drug in the European Union.
Ongoing clinical trials HYDRA is a phase III trial that tests adjuvant IL-2, IFNα, and 5-fluorouracil (5-FU) in patients with a high risk of relapse (T3b to T4 or N1/2 or positive margins or vascular invasion) after surgical treatment for RCC. REO4 is a phase III trial involving IFNα versus IFNα, IL-2 and 5-FU in metastatic RCC.
Carcinoma of the renal pelvis Renal pelvis carcinoma is a relatively rare tumour, constituting 5% of all renal tumours; more than 90% are transitional cell carcinomas (TCCs). Squamous cell carcinoma, which is usually associated with chronic inflammation or infection of the renal pelvis, accounts for the majority of the remaining malignant tumours. It is more common in men (male-to-female ratio of 2:1) and usually occurs between the ages of 50 and 70. The major risk factor is smoking. Other factors include phenacetin use, urban residence, work in the dye and textile industry, chronic inflammation and Balkan nephropathy. TCC is often multifocal, and up to 50% of patients with TCC may also develop bladder cancer. The majority of patients present with frank haematuria. An IVU will often demonstrate a filling defect in the collecting system, and CT or MRI is used to assess local extent and distant spread. Ureteroscopy allows direct visualisation and biopsy of the majority of tumours. Localised/locally advanced disease can be treated by radical nephrectomy including total removal of the ipsilateral ureter. Early stage low-grade cancers are often treated with a more conservative local surgical excision. Follow-up with urine cytology and cystoureteroscopy is recommended, because up to 50% of patients will
Kidney
develop another renal tract cancer. The patient MUST be advised to stop smoking, which will greatly reduce the risk of subsequent cancers developing. REFERENCES Aso, Y. and Homma, Y. (1992). A survey of incidental renal cell carcinoma in Japan. J. Urol., 147, 340–3. Atkins, M., Regan, M., McDermott, D. et al. (2005). Carbonic anhydrase IX expression predicts outcome of interleukin 2 therapy for renal cancer. Clin. Cancer Res., 11, 3714–21. Atzpodien, J., Kirchner, H., Illiger, H. J. et al. (2001). IL-2 in combination with IFN-alpha and 5-FU versus tamoxifen in metastatic renal cell carcinoma: long-term results of a controlled randomized clinical trial. Br. J. Cancer, 85, 1130–6. Bell, E. T. (1950). Renal Diseases, 2nd edn. Philadelphia, PA: Lea and Febiger. Bosniak, M. A. (1986). The current radiological approach to renal cysts. Radiology, 158, 1–10. Clark, J., Atkins, M., Urba,W. et al. (2003). Adjuvant high-dose bolus interleukin-2 for patients with high-risk renal cell carcinoma: a cytokine working group randomized trial. J. Clin. Oncol., 21, 3133–40. Coppin, C., Porzsolt, F., Awa, A. et al. (2005). Immunotherapy for advanced renal cell cancer. Cochrane Database Syst. Rev., 1, CD001425. Escudier, B., Szczylik, C., Eisen, T. et al. (2005). Randomized phase III trial of the Raf kinase and VEGFR inhibitor sorafenib (BAY 43–9006) in patients with advanced renal cell carcinoma (RCC). Proc. Am. Soc. Clin. Oncol., 23, 380s. Finney, R. (1973). Radiotherapy in the treatment of hypernephroma: a clinical trial. Br. J. Urol., 45, 26–40. Flanigan, R. C., Salmon, S. E., Blumenstein, B. A. et al. (2001). Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N. Engl. J. Med., 345, 1655–9. Fuhrman, S. A., Lasky, L. C., Limas, C. et al. (1982). Prognostic significance of morphological parameters in renal cell carcinoma. Am. J. Surg. Pathol., 6, 655–63. Hudes, G., Carducci, M., Tomczak, P. et al. (2006). A phase 3, randomized, 3-arm study of temsirolimus (TEMSR) or interferon-alpha (IFN) or the combination of TEMSR + IFN in the treatment of first-line, poor-risk patients with advanced renal cell carcinoma (adv RCC). J. Clin. Oncol., 24 (18 Suppl.), LBA4. Jocham, D., Richter, A., Hoffmann, L. et al. (2004). Adjuvant autologous renal tumour cell vaccine and risk of tumour progression in patients with renal-cell carcinoma after radical nephrectomy: phase III, randomised controlled trial. Lancet, 363, 594–9.
Kjaer, M. (1988). The role of medroxyprogesterone acetate (MPA) in the treatment of renal adenocarcinoma. Cancer Treat. Rev., 15, 195–209. Kjaer, M., Frederiksen, P. L. and Engelholm, S. A. et al. (1987). Postoperative radiotherapy in stage II and III renal adenocarcinoma: a randomised trial by the Copenhagen Renal Cancer Study Group. Int. J. Radiat. Oncol. Biol. Phys., 13, 665–72. Kovacs, G., Akhtar, M., Beckwith, B. J. et al. (1997). The Heidelberg classification of renal cell tumours. J. Pathol., 183, 131–3. Mekhail, T. M., Abou-Jawde, R. M., Boumerhi, G. et al. (2005). Validation and extension of the Memorial Sloan-Kettering prognostic factors model for survival in patients with previously untreated metastatic renal cell carcinoma. J. Clin. Oncol., 23, 832–41. Messing, E., Manola, J., Wilding, G. et al. (2003). Phase III study of interferon alfa-NL as adjuvant treatment for resectable renal cell carcinoma: an Eastern Cooperative Oncology Group/Intergroup trial. J. Clin. Oncol., 21, 1214–22. Mickisch, G. H., Garin, A., van Poppel, H. et al. (2001). Radical nephrectomy plus interferon-alfa-based immunotherapy compared with interferon-alfa alone in metastatic renal-cell carcinoma: a randomised trial. Lancet, 358, 966–70. Montie, J. E., Stewart, B. H., Straffon, R. A. et al. (1977). The role of adjunctive nephrectomy in patients with metastatic renal cell carcinoma. J. Urol., 117, 272–5. Motzer, R. J., Hutson, T. E., Tomczak, P. et al. (2006a). Phase III randomised trial of sunitinib maleate (SU11248) versus interferon-alfa (IFN-a) as first-line systemic therapy for patients with metastatic renal cell carcinoma (mRCC). J. Clin. Oncol., 24 (18 Suppl.), LBA3. Motzer, R. J., Michaelson, M. D., Redman, B. G. et al. (2006b). Activity of SU11248, a multitargeted inhibitor of vascular endothelial growth factor receptor and platelet-derived growth factor receptor, in patients with metastatic renal cell carcinoma. J. Clin. Oncol., 24, 16–24. Murphy, W. M., Beckwith, J. B. and Farrow, G. M. (1994). Tumors of the kidney, bladder, and related urinary structures. In: Atlas of Tumor Pathology, 3rd edn. Bethesda, MD: Armed Force Institute of Pathology. O’Dea, M. J., Zincke, H., Utz, D. C. et al. (1978). The treatment of renal cell carcinoma with solitary metastasis. J. Urol., 120, 540–2. Tada, S., Yamagishi, J., Kobayashi, H. et al. (1983). The incidence of simple renal cyst by computed tomography. Clin. Radiol., 34, 437–9. UICC. (2002). In TNM Classification of Malignant Tumours, 6th edn. Ed. L. H. Sobin and Ch. Wittekind. New York: Wiley-Liss, pp. 193–5. Yagoda, A., Abi-Rached, B., Petrylak, D. et al. (1995). Chemotherapy for advanced renal-cell carcinoma: 1983–1993. Semin. Oncol., 22, 42–60.
221
18
BLADDER Stephen Williams and Jim Barber
Introduction Bladder cancer is an important cause of morbidity and mortality and has a high incidence rate in industrialised countries. Each year in the UK, around 10 000 people develop a bladder cancer, and nearly half of those affected will die from their disease. Bladder cancers have a wide range of biological behaviours, from superficial tumours, which can be treated by local resection, to highly aggressive and infiltrative tumours. The treatment of invasive disease has traditionally been radical radiotherapy that has been given in an effort to avoid cystectomy. Although radiotherapy is still an important part of radical and palliative treatment, evidence is emerging that combined treatment modalities may improve outcomes.
connects with the urethra, and in men the bladder neck rests on the prostate gland. The ureters join the bladder at the base. When empty, the bladder has two inferolateral surfaces and one superior surface, and when full, it becomes ovoid. In women, the bladder is related posteriorly to the uterus and vagina, and in men it is separated from the rectum by the seminal vesicles, the vasa deferentia, and the rectovesical fascia.
Carcinoma of the bladder Risk factors and aetiology
Bladder cancer is the fourth most common cancer in men (6% of cancer cases) and the eighth most common cancer in women (2.5% of cases). The incidence is 17 per 100 000 of the population (CRUK; www.cancerresearchuk.org/). Because of recurrences, the prevalence in men is second only to prostate cancer. Bladder cancer is more common in white people. There has been a steady rise in the occurrence of bladder tumours over the past 20 years (33% in the USA). The peak age of incidence is 69 to 71 years.
The majority of transitional cell bladder cancers are related to a history of smoking (probably because of the excretion of carcinogens such as 4-aminobiphenyl), which increases the risk about four-fold. More important, continuing to smoke increases the progression rate of non-invasive cancers. Stopping smoking reduces the risk of developing cancer but it takes many years for the risk to return to normal. Historically, transitional cell cancers were associated with exposure to various industrial chemicals including analine dyes, 2-naphthylamine, 4-aminobiphenyl, 4-nitrobiphthol, benzidine, 2-amino-1-naphthol and acrolein (rubber dye). Health and safety procedures should now prevent exposure to these chemicals. Conditions associated with ‘squamous metaplasia’ can lead to the development of squamous cell carcinomas. These include chronic urinary stasis (e.g. longterm catheter, bladder stones or paraplegia) and, in the subtropics, chronic infection with schistosomiasis – a major public health problem in the Middle East, where squamous cell carcinomas of the bladder are relatively common.
Anatomy
Screening
The shape and size of the bladder change depending on how much urine it contains. It lies in the anterior pelvis behind the pubic bone and its apex is joined to the umbilicus via the urachus. The neck of the bladder
Screening for asymptomatic haematuria is not recommended because its positive predictive value is too low (0.5%) to warrant mass screening. However, routine screening for microscopic haematuria may be indicated
Types of bladder tumour The types of bladder tumour that can occur are shown in Table 18.1.
Incidence and epidemiology
222
Bladder
Table 18.1. Types of bladder tumour Type Benign
Examples Inflammatory plaques Transitional cell papilloma Inverted papilloma Leiomyoma
Malignant primary
Carcinomas Transitional cell carcinoma (90%) Squamous cell carcinoma (approx. 5%) Adenocarcinoma Mixed Small-cell carcinoma
invaded the muscle layer have a high chance of local recurrence, even of metastasising, and must be treated aggressively. Overall at least half these patients will die of their cancers. ‘Carcinoma in situ’ is a natural precursor of highgrade cancers, many of them arising from such areas. On cystoscopy, carcinoma in situ appears as flat erythematous patches and often leads to irritating urinary tract symptoms. Histologically the epithelium loses its orderly structure and may be thickened; cytologically cells are similar to G3 TCC but without any evidence of invasion. This is a condition that must be taken seriously, because it will progress quickly to a muscleinvading cancer over a period of a few years.
Others
Malignant secondary
Sarcoma
Other cancers
Lymphoma
TCCs commonly contain components of squamous cell carcinoma, adenocarcinoma, small-cell or sarcomatoid elements. Such tumours are normally treated in the same way as TCCs, but it may be difficult to distinguish between a TCC with squamous differentiation and a true ‘pure’ squamous cell carcinoma. Pure squamous cell cancers are rare in the absence of a predisposing factor for squamous metaplasia. Small-cell (oat cell) carcinoma is well recognised and should be managed in the same way as small-cell carcinoma of the lung. Adenocarcinomas are extremely rare. They are occasionally seen in the dome of the bladder, where they are thought to originate from a persistent urachus, but they may also occur around the trigone (possibly originating from cystic glandularis).
Direct spread (e.g. from prostate, cervix, vagina) Distant spread from tumours at other sites
for populations exposed to bladder carcinogens, including heavy smokers.
Pathology Urothelial tumours Urothelial tumours are derived from the transitional cell epithelium and are termed transitional cell carcinomas (TCCs). They are most common in the base of the bladder, and multiple tumours are frequent (up to 40%). Apart from some benign conditions such as the bladder papilloma, they fall into two broad groups (with some overlap), classified by their behaviour: tumours with low or high malignant potential. A TCC of low malignant potential is likely to be papillary and superficial. Histologically these tumours are of low grade (G1 to 2) with mild cellular dysplasia. As they grow larger, they may become muscle invasive, but this is a relatively rare event. A conservative approach with local resection is the usual treatment. Death from cancer is relatively rare. A TCC of high malignant potential is more often solid or flat and is likely to grow both into and outside the muscle wall of the bladder. Histologically these tumours are high-grade cancers (G3). Even tumours that have not
Clinical presentation Patients usually present with haematuria and all cases of haematuria should be referred urgently as suspected cancers to a urologist for further investigation. Most urologists run rapid-access haematuria clinics, where prompt and efficient investigations can be performed. Even if an obvious cause is present (e.g. minimal haematuria associated with a proven urinary tract infection in a female patient), a co-existent cancer may be present. Patients with bladder cancer may also complain of urgency, dysuria and increased urinary frequency.
Investigation Investigations for haematuria of unknown cause It is clearly important to screen the whole of the urinary tract for tumours, stones and any structural 223
Stephen Williams and Jim Barber
abnormalities. Urinalysis should be performed for cytology and culture, but caution is needed in attributing significant haematuria to a urinary tract infection. A flexible cystoscopy is generally required to diagnose bladder tumours, together with imaging of the upper urinary tracts by an ultrasound of kidneys and bladder and an IVU. A CT scan of the abdomen and pelvis is an alternative but this is not standard practice in the UK.
Further investigations if bladder tumour is confirmed The first stage in management is to perform transurethral resection (TURBT), usually under general anaesthetic, using a resectoscope. TURBT includes resection of all visible tumour (including muscle in the specimen if possible) and a separate resection biopsy from the border of the resected area and the tumour base. After a successful resection, there should be no mass to feel on EUA but the results of bimanual examination must be considered in the context of the other staging investigations. The TURBT will give information about the grade of cancer and the depth of invasion. Ideally there should be some muscle in the biopsy, and the pathologist will be able to comment on the presence or absence of muscle invasion. Local staging should be completed by MRI if available because it is agreed that MRI scanning is better than CT for assessing the extent of the primary tumour. However, CT scanning may give similar information about lymph node enlargement. Concordance across cystoscopic findings, the pathology, and the radiological findings increases the confidence in the results of staging, but sometimes the findings may not agree. For example, a large solid tumour may be apparently deeply invasive at operation but there may be no muscle invasion on the biopsy. In other cases, a tumour may appear small but, on imaging, can be growing outside the bladder. MDT review is then particularly important for deciding the stage of the tumour. For patients being considered for radical treatment, staging investigations involve a CT scan of the chest and abdomen and a bone scan. Lung metastases may be the first radiological sign of metastatic disease, and bone metastases may be found in around 5% of cases. PET scanning is of limited use because of interference from urinary excretion of contrast in the bladder. 224
Table 18.2. TNM classification for carcinoma of the bladder Stage
Description
Tis
Carcinoma in situ
Ta
Noninvasive papillary tumour
T1
Subepithelial connective tissue
T2a
Inner half of muscularis
T2b
Outer half of muscularis
T3a
Perivesical tissues (micro)
T3b
Perivesical tissues (macro)
T4a
Prostate, uterus or vagina
T4b
Pelvic or abdominal wall
N0
Nodes free of tumour
N1
Single node positive (< 2 cm)
N2
Single node positive (2–5 cm) or
N3
Node positive (> 5 cm)
M0
No distant metastases
M1
Distant metastases
multiple nodes positive (< 5 cm)
Adapted from UICC (2002).
Staging Tumours originate in the bladder epithelium and infiltrate deeply into the muscle layers, penetrating through the bladder wall into perivesical fat and adjacent organs. Lymphatic spread is first to the hypogastric, obturator, internal, external and common iliac lymph nodes. The risk of lymph node metastases is proportional to the depth of tumour invasion (i.e. 20% for lamina propria invasion, 30% for superficial muscle invasion and 60% with full-thickness muscle invasion). The staging system has been changed several times over the past few years, and so care must be taken when comparing published results that use the old systems. The TNM staging classification for carcinomas of the bladder is shown in Table 18.2.
Treatment overview: superficial TCC of the bladder Of urothelial bladder tumours, 70% are superficial; that is, the tumour is confined to the bladder mucosa and tumour growth has not been detected beyond the lamina propria. All superficial bladder tumours have a tendency to recur, and multiple or recurrent tumours have a higher
Bladder
risk (Parmar et al., 1989). Although recurrence can be dealt with by a repeat TUR, adjuvant therapy should be considered to prevent the risk of progression to muscleinvasive disease and a potentially fatal cancer, as follows.
Low risk of progression (mostly TCCs of low malignant potential) pTa or pT1, G1–2 A single dose of intravesical mitomycin C (20 mg) within 6 hours of resection significantly decreases recurrences (Sylvester et al., 2004) and is standard practice in the UK, followed by regular cystoscopic follow-up.
High risk of progression (mostly TCCs of high malignant potential) pTa–pT1, G3; CIS; multiple recurrences of G2 tumours Consider adjuvant intravesical BCG therapy (Harland and the UK NCRI Bladder Clinical Studies Group, 2005; Shelley et al., 2001). There is evidence that intravesical BCG is somewhat more effective than mitomycin, albeit considerably more toxic (Shelley et al., 2003). If there are further high-grade recurrences or resistant CIS the risk of progression to muscle-invasive disease is high and early cystectomy should be considered. Although it is not standard therapy, cystectomy is sometimes performed for patients with T1 G3 disease. Even with this approach, long-term cancer control rates are only 90%, because some patients have metastases. In the UK this approach would usually be reserved for those patients that fail BCG therapy. Radiotherapy is used less today as a method of avoiding cystectomy in patients with recurrent pT1 G3 disease, particularly if there is co-existent carcinoma in situ. A randomised trial of radiotherapy (MRC BS06 study) failed to show any clear benefit for radiotherapy in this situation (Harland and the UK NCRI Bladder Clinical Studies Group, 2005), but radiotherapy remains an option for patients who have failed intravesical therapy and who are unfit for cystectomy.
BCG therapy BCG is an attenuated mycobacterium used for vaccinations. The exact mechanism of its effect is unknown, but it contacts tumour cells by a fibronectin attachment protein which causes internalisation into the tumour cells. The aim of treatment is to deliver 10 000 000 organisms per instillation in 50 ml of normal saline 2 to 4 weeks post-TURBT. The solution should be retained in the bladder for about 2 hours. Immunosuppression, frank
haematuria and bacterial infection are contraindications. Usually around six weekly treatments are given. The side effects of BCG therapy are considerably worse than with mitomycin and include dysuria, urgency and frequency in most patients, lasting several days. In rare cases, systemic BCG infection can occur (pyrexia of > 38.5◦ C for > 24 hours or a pyrexia of > 39.5◦ C), which would require antituberculous antimicrobial therapy.
Treatment of invasive bladder cancer: management of the primary For patients with tumours that invade muscle (T2 and above) there are high recurrence and metastasis rates following TURBT alone, and radical treatment with surgery or radiotherapy is advised. Until the early 1990s, in the UK, Canada and elsewhere, a policy of radical radiotherapy and surgical salvage was pursued. This strategy results in about 30% of patients being longterm survivors, a substantial proportion with their bladders intact. Radiotherapy fails to give local control of the tumour in up to half the patients, but the majority of patients with local failure develop metastases at the same time, which means that salvage cystectomy is only needed for a few patients. This approach was justified for many years because of the high mortality and morbidity rates for radical cystectomy and by the evidence from one small randomised trial from the Institute of Urology (London). This trial showed little difference in the survival outcome between cystectomy and bladder preservation. However, a metaanalysis of this trial and three other similar trials from around the same period shows that this conclusion was flawed: the trials do indeed show a substantial and significant benefit for elective surgical management (Shelley et al., 2004). To what extent these data can be extrapolated to present-day practice is debated, but the best results reported recently have been achieved by using neoadjuvant chemotherapy combined with cystectomy and pelvic nodal clearance. Five-year survival rates of just less than 50% have been recorded even in multicentre studies. Unfortunately, this is a relatively arduous and toxic treatment and many patients presenting with bladder cancer are on the borderline of fitness for this approach. On the other hand, outcomes similar to those after surgery have been reported following the use of neoadjuvant chemotherapy and chemoradiotherapy 225
Stephen Williams and Jim Barber
regimens (Shipley et al., 2005), also with 5-year survival at around 50%. It appears that patients who respond well to initial neoadjuvant chemotherapy may be good candidates for bladder preservation with radiotherapy or chemoradiotherapy and a randomised study (SPARE) will be starting shortly in the UK to test this hypothesis.
Patient fitness and attitude It is important to decide at an early stage whether patients are fit enough for a cystectomy and/or chemotherapy treatment, and whether they are prepared to have such an operation. Because bladder cancer is so strongly associated with smoking, patients often have co-existent cardiovascular and pulmonary disease and they may also have poor renal function due to obstructive uropathy. Hydronephrosis is often seen on CT scanning; if it is found, the patient’s renal function must be carefully evaluated. For patients with a GFR below 60 ml/min, the risks and benefits of surgery and chemotherapy should be weighed up carefully and the patient’s preference considered. Patients tolerate neoadjuvant chemotherapy poorly if their GFR is much below 50 ml/min.
Radical cystectomy Radical cystectomy consists of removal of the bladder and distal ureters and neighbouring organs, such as the prostate and seminal vesicles in men and the uterus and adnexa in women. Removal of the urethra is indicated if the bladder neck is involved in women or the prostate is involved in men. The operative mortality of radical cystectomy has decreased substantially over the past 20 years and is about 3% in experienced centres. There are three options for surgery: ileal conduit, continent urinary diversion or orthotopic neobladder. An ileal conduit is the quickest and simplest method but requires the patient to wear a urostomy bag. The ileal conduit is made from a 15- to 25-cm-long segment of ileum. One end of the conduit is closed with stitches or staples, and the other end is brought to the surface of the abdomen to form a stoma. The ureters are implanted into the closed end of the conduit. Some degree of hydronephrosis is common, but overall the complication rate is low. For continent urinary diversion, the continent urinary pouch is made from approximately 2 feet of ileum. The pouch is emptied by inserting a soft silicone catheter into the stoma four to eight times a day, and so a stoma is still necessary. 226
The orthtropic neobladder, also known as ‘continent orthotopic urinary reconstruction,’ appeals especially to younger patients who wish to avoid a stoma on the abdomen and/or wearing a urostomy bag. The reservoir is constructed from small bowel and is anastomosed to the top of the urethra; the patient empties the bladder by abdominal straining or clean intermittent catheterisation. In many cases those with neobladders learn how to completely empty their bladders and do not require intermittent catheterisation; however, this is a learned response and not guaranteed. Disadvantages include nocturnal leakage and problems with voiding requiring intermittent self-catheterisation.
Role of radical radiotherapy: a bladder preservation strategy The role of radiotherapy in bladder cancer has been reviewed (Logue and McBain, 2005).
Applications Younger patients with small tumours, no ureteric obstruction, a complete resection at TURBT, or a complete response with chemotherapy have a good relative outcome after radiotherapy. However, today, radiotherapy is more often given to elderly patients or those considered to be unfit for cystectomy or to patients with inoperable tumours.
Setup With the patient in a supine position, skin tattoos are placed anteriorly over the pubic symphysis and laterally over the iliac crests to prevent lateral rotation. CT scanning should be used to simulate the patient with an empty bladder. On the other hand, if nodal irradiation is considered, the patient should have a full bladder during the pelvic treatment so that the small bowel is pushed out of the field. Target volume Standard practice is to treat the whole bladder with a 1 to 2 cm margin without making any attempt to treat the pelvic lymph nodes (see section on nodal treatment). Important information about the site and spread of the tumour should be available both from radiological investigation (CT or MRI) and from the cystoscopy and EUA. For tumours with an extravesical component, a wider margin should be allowed around those areas. Standard margins may be inadequate to allow for bladder motion in many patients. Patients particularly at risk are those with a large, dilated rectum, large tumours or tumours in the dome of the bladder (the bladder moves more superiorly). Drawing a wider
Bladder
Isodose % 95 60 50
whether extending the lymph node dissection proximally is useful. The long term survival rates of patients who have microscopic nodal involvement (around 25% 5-year survival) support the use of nodal surgery. For patients treated with bladder preservation, it is standard practice in the UK to avoid nodal irradiation, despite the high risk of nodal disease, because the benefits are unclear.
Clinically node positive Figure 18.1. A radiotherapy plan for treating bladder cancer.
margin should be considered for these patients but is often impractical because of the size of the resulting fields.
Use of CRT/IMRT 3D planning with the use of conformal blocking or an MLC is strongly recommended, because it allows a significant reduction in the amount of small bowel that is irradiated and may also reduce the rectal dose. Unfortunately, dose escalation is probably not practical with the standard whole-bladder technique, because doses are limited by the tolerance of the bladder itself. A more sophisticated approach to irradiation of the bladder is needed in which a higher dose of radiation is given to the tumour with relative sparing of uninvolved areas: a move away from the ‘whole-bladder treatment.’ For example, in the BC2001 trial, the whole bladder is included in a first phase followed by a boost to the tumour alone. IMRT may be applicable but has not yet been investigated.
Beams One anterior and two posterior oblique wedged fields are used, depending on the bladder contour. Figure 18.1 shows a radiotherapy plan for treating bladder cancer. For the nodal irradiation technique (not often performed) please refer to the prostate chapter (Chapter 19, see p. 231). Doses Single-phase treatments involve 64 Gy in 32 fractions over 6.5 weeks or 55 Gy in 20 fractions over 4 weeks.
Treatment of invasive bladder cancer: management of the nodes Clinically node negative Radical cystectomy generally includes elective dissection of the regional pelvic lymph nodes, but it is not clear
Most patients with macroscopic nodal disease relapse with metastases, but there are undoubtedly a few longterm survivors after radical cystectomy and nodal clearance. Before starting treatment a careful search for metastatic disease should be carried out. Patients with good performance status and who show nodal enlargement on MRI can be considered for radical therapy with neoadjuvant chemotherapy followed by a cystectomy/node clearance if a good response is obtained. For many patients, radical therapy is not possible. Palliative chemotherapy is considered if the patient is fit enough and, in the event of a good response, is followed by consolidation radiotherapy treatment to the bladder and possibly nodal areas. It is unlikely that macroscopic nodal disease can be cured with the relatively modest radiotherapy doses currently possible in this situation and radiotherapy is frequently palliative.
Combined modality treatment (chemoradiotherapy) Concurrent chemoradiotherapy may have a role in the treatment of bladder cancer (Sherwood et al., 2005). Results comparable to cystectomy have been achieved with intensive bladder preserving regimens using neoadjuvant chemotherapy (gemcitabine and cisplatin) and concurrent chemoradiotherapy with cisplatin (Shipley et al., 2005). Importantly, these regimens have also included pelvic nodal irradiation. However, the effectiveness of concurrent chemoradiation is uncertain, because only one small prospective randomised controlled phase III trial has been published. This trial shows a 19% improvement in local control with concurrent cisplatin but no significant improvement in overall survival, progression-free survival, or the distant metastasis rate (Coppin et al., 1996). The only other randomised trial of concurrent chemoradiation is the UK BC2001 study, in which a combination of 5-fluorouracil and mitomycin C is used concurrently with radiotherapy. The regimen (based on the ‘Nigro’ anal cancer regimen) showed good results in a small 227
Stephen Williams and Jim Barber
phase II study but it will be several years before the results of BC2001 are available.
Chemotherapy: neoadjuvant and adjuvant It is now clear from several randomised trials and two meta-analyses that the addition of neoadjuvant chemotherapy to either cystectomy or radiotherapy provides a modest overall survival benefit of around 5% (Advanced Bladder Cancer Meta-analysis Collaboration, 2003, 2005; Grossman et al., 2003; International Collaboration of Trialists, 1999; McLaren, 2005; Winquist et al., 2004). Although this outcome is similar to the survival benefit from adjuvant therapy in other cancers, the toxicity of this treatment is at the limit of what can be tolerated for an adjuvant therapy for most patients. In fit patients with good renal function, however, its use should be encouraged.
Adjuvant or neoadjuvant? Neoadjuvant chemotherapy is generally favoured because it is more practicable. The recovery period after cystectomy can be very long and it may be weeks or months before it is possible to give chemotherapy. Moreover, surgery may be easier in patients who have a good response to chemotherapy. Despite this, some clinicians feel that the delay in definitive local treatment might be detrimental and, therefore, they favour an adjuvant approach. Unfortunately, the benefits of adjuvant chemotherapy are unlikely to be quantified in the near future because a meta-analysis of adjuvant chemotherapy trials was inconclusive, because of the small numbers of patients and an EORTC trial answering this question is recruiting poorly. The same regimens are in use for palliative and adjuvant chemotherapy.
Palliative chemotherapy A few patients with a relatively slow-growing TCC may achieve long remissions (sometimes multiple remissions) with combination chemotherapy treatment. Sadly, for the majority, the results of palliative chemotherapy are disappointing; median survival is around 12 months. Many patients are elderly or in relatively poor general health and so the decision to give chemotherapy needs a careful clinical assessment and time to discuss the pros and cons of the treatment with the patient. Patients with good performance status and renal clearance above 60 ml/min tolerate chemotherapy 228
better. Ureteric obstruction is common and stent insertion should be considered first if there is obstruction present. Alternatively, in patients with poor renal function, carboplatin can be substituted for cisplatin, but response rates are consistently inferior (Petrioli et al., 1996). For a variety of reasons, the incidence of thromboembolic disease in these patients is extremely high, and consideration should be given to prophylactic anticoagulation.
Chemotherapy regimens Combination regimens containing cisplatin appear more effective than single-agent cisplatin, and there is a consensus that CMV and MVAC have similar efficacy and toxicity. Both of these regimens are decreasing in use, because randomised data show that in metastatic disease Gem-Cis (cisplatin 80 mg/m2 day 1, gemcitabine 1000 mg/m2 day 1 and 8) provides a rate of survival similar to that of CMV but with considerably less toxicity (von der Maase et al., 2000). Many centres have adopted Gem-Cis as a standard regimen for palliative treatment. Its use is also increasing in the adjuvant setting, despite the lack of evidence. Accelerated MVAC (given with GCSF support) is considered an alternative to Gem-Cis and has similar toxicity (Sternberg et al., 2001). Phase II trials of combination regimens including taxanes with cisplatin have shown promising results but randomised data are awaited. Large-scale trials with novel agents have not yet been performed for bladder cancer.
Palliative radiotherapy Palliative radiotherapy may be highly effective at temporarily relieving haematuria and urinary symptoms in locally advanced and metastatic bladder cancer. It is also an option for elderly or unfit patients with clinically localised disease. A randomised trial has shown similar palliation rates with a short, hypofractionated treatment (21 Gy in 3 fractions) compared to a higher-dose palliative treatment (35 Gy in 10 fractions; Duchesne et al., 2000). The target volume may vary widely among patients and so the technique must be determined on a caseby-case basis. Treatment limited to the bladder only may improve urinary symptoms and has the least toxicity, particularly when delivered with a radiotherapy plan. Treatment to the whole pelvis with parallel fields
Bladder
permits the inclusion of pelvic lymph nodes but can lead to severe rectal toxicity. Cystograms are rarely used today because of easy access to CT planning and/or simulation. A dose/fractionation schedule of 35 Gy in 10 fractions over 2 weeks or 30 to 36 Gy in 5 to 6 fractions weekly can be delivered using a CT plan but this dose would be considered on the borderline of tolerance if parallel fields are used. If treatment to the whole pelvis is required with parallel fields, then 30 Gy in 10 fractions is better tolerated by the patient. If a smaller volume, such as the bladder only, is being treated, low dose hypofractionated treatment is effective (e.g. 21 Gy in 3 fractions over 1 week). Even a single fraction of 10 Gy may provide relief from haematuria in a patient with poor performance status.
Areas of current interest and current clinical trials Areas of current interest Areas of current interest include the following: r The use of BCG maintenance treatment. r The use of combined modality treatment. r The role of new chemotherapy drugs (e.g. gemcitabine, paclitaxel and vinflunine). r The role of adjuvant chemotherapy. r Dose escalation in radiotherapy. r The use of imaging to overcome the effects of bladder movement during radiotherapy (Muren et al., 2004). r The use of novel drugs such as anti-EGFR agents (Alonzi and Hoskin, 2005).
Current trials See www.ncrn.org.uk/ for information regarding ongoing trials. BC2001 is a 2 × 2 factorial randomised phase III study comparing standard versus whole-bladder radiotherapy with tumour boost with or without synchronous chemotherapy in muscle-invasive bladder cancer. BCON is a multicentre randomised trial of radical radiotherapy with carbogen in locally advanced bladder cancer. EORTC 30986 is a randomised phase II/III study assessing gemcitabine/carboplatin with methotrexate/ carboplatin/vinblastine in previously untreated patients with urothelial cancer who are ineligible for cisplatin-based chemotherapy. EORTC 30994 is a randomised phase III trial comparing immediate versus deferred chemotherapy after
radical cystectomy in patients with pT3-pT4 and/or N+ M0 TCC of the bladder. ODMIT C is a multicentre, prospective, randomised study of the value of a single dose of intravesical mitomycin in preventing the development of bladder tumours following nephroureterectomy for TCC of the upper urinary tract.
REFERENCES Advanced Bladder Cancer Meta-analysis Collaboration. (2003). Neoadjuvant chemotherapy in invasive bladder cancer: a systemic review and meta-analysis. Lancet, 361, 1927–34. Advanced Bladder Cancer Meta-analysis Collaboration. (2005). Adjuvant chemotherapy in invasive bladder cancer: a systematic review and meta-analysis of individual patient data. Eur. Urol., 48, 189–99; discussion 199–201. Alonzi, R. and Hoskin, P. (2005). Novel therapies in bladder cancer. Clin. Oncol. (R. Coll. Radiol.), 17, 524–38. Coppin, C. M., Gospodarowicz, M. K., James, K. et al. (1996). Improved local control of invasive bladder cancer by concurrent cisplatin and preoperative or definitive radiation. J. Clin. Oncol., 14, 2901–7. Duchesne, G. M., Bolger, J. J., Griffiths, G. O. et al. (2000). A randomized trial of hypo-fractionated schedules of palliative radiotherapy in the management of bladder carcinoma: results of Medical Research Council trial BA09. Int. J. Radiat. Oncol. Biol. Phys., 47, 379–88. Grossman, H. B., Natale, R. B., Tangen, C. M. et al. (2003). Neoadjuvant chemotherapy plus cystectomy alone for locally advanced bladder cancer. N. Engl. J. Med., 349, 859–66. Harland, S. J. and the UK NCRI Bladder Clinical Studies Group. (2005). A randomised trial of radical radiotherapy in pT1G3 NXM0 bladder cancer (MRC BS06). J. Clin. Oncol., 23, (16 Suppl. I), 4505. International Collaboration of Trialists. (1999). Neoadjuvant cisplatin, methotrexate, and vinblastine chemotherapy for muscle invasive bladder cancer: a randomised controlled trial. Lancet, 354, 533–40. Logue, J. and McBain, C. A. (2005). Radiation therapy for muscle-invasive bladder cancer: treatment planning and delivery. Clin. Oncol. (R. Coll. Radiol.), 17, 508–13. McLaren, D. B. (2005). Neoadjuvant chemotherapy in transitional-cell carcinoma of the bladder. Clin. Oncol., 17, 503–7. Muren, L. P., Smaaland, R. and Dahl, O. (2004). Conformal radiotherapy of urinary bladder cancer. Radiother. Oncol., 73, 387–98. Parmar, M. K., Freedman, L. S., Hargreave, T. B. et al. (1989). Prognostic factors for recurrence and follow-up policies in the treatment of superficial bladder cancer: report from the British Medical Research Council Subgroup on Superficial Bladder Cancer (Urological Cancer working party). J. Urol., 142, 284–8. Petrioli, R., Frediani, B., Manganelli, A. et al. (1996). Comparison between a cisplatin-containing regimen and a carboplatincontaining regimen for recurrent or metastatic bladder cancer patients. A randomized phase II study. Cancer, 77, 344–51.
229
Stephen Williams and Jim Barber
Shelley, M. D., Barber, J., Wilt, T. et al. (2004). A meta-analysis of randomised trials suggests a survival benefit for combined radiotherapy and radical cystectomy compared with radical radiotherapy for invasive bladder cancer: are these data relevant to modern practice? Clin. Oncol. (R. Coll. Radiol.), 16, 166–71. Shelley, M. D., Court, J. B., Kynaston H. et al. (2003). Intravesical bacillus Calmette-Guerin versus mitomycin C for Ta and T1 bladder cancer. Cochrane Database Syst. Rev., (3), CD003231. Shelley, M. D., Kynaston, H., Court, J. et al. (2001). A systematic review of intravesical bacillus Calmette-Guerin plus transurethral resection vs transurethral resection alone in Ta and T1 bladder cancer. B. J. U. Int., 88, 209–16. Shipley, W. U., Zietman, A. L., Kaufman, D. S. et al. (2005). Selective bladder preservation by trimodality therapy for patients with muscularis propria-invasive bladder cancer and who are cystectomy candidates – the Massachusetts General Hospital and Radiation Therapy Oncology Group experiences. Semin. Radiat. Oncol., 15, 36–41. Sternberg, C. N., de Mulder, P. H., Schornagel, J. H. et al. (2001). Randomised phase III trial of high dose intensity methotrexate, vinblastine, doxorubicin and cisplatin (MVAC)
230
chemotherapy and recombinant human granulocyte colony-stimulating factor versus classic MVAC in advanced urothelial tract tumours: European Organisation for Research and Treatment of Cancer Protocol no. 30924. J. Clin. Oncol., 19, 2638–46. Sylvester, R. J., Oosterlinck, W. and van der Meijden, A. P. (2004). A single immediate postoperative instillation of chemotherapy decreases the risk of recurrence in patients with stage Ta T1 bladder cancer: a meta-analysis of published results of randomized clinical trials. J. Urol., 171 (6 Pt. 1), 2186–90. UICC. (2002). TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind. New York: Wiley-Liss, pp. 199–202. von der Maase, H., Hansen, S. W., Roberts, J. T. et al. (2000). Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin and cisplatin in advanced or metastatic bladder cancer: results of a large randomised multinational multicentre phase III study. J. Clin. Oncol., 18, 3068–77. Winquist, E., Kirchner, T. S., Segal, R. et al. (2004). Neoadjuvant chemotherapy for transitional cell carcinoma of the bladder: a systemic review and meta-analysis. J. Urol., 171 (2 Pt. 1), 561–9.
19
PROSTATE Jim Barber and John Staffurth
Introduction
Anatomy
The biological behaviour of prostate cancer varies widely. Many tumours are found incidentally, whereas others cause signs and symptoms early on and may progress rapidly to disseminated disease. The incidence of disease is increasing, probably as a result of greater detection, but it is not clear whether earlier diagnosis will lead to longer survival. For men with early stage prostate cancer, the best treatment is not known and there are options of observation, surgery, external beam radiotherapy or brachytherapy. For patients with advanced disease, the main treatment is hormonal, in the form of androgen deprivation. Attempts are currently under way to improve the therapeutic ratio of radiotherapy by using new techniques such as intensitymodulated radiation therapy (IMRT).
The prostate gland is in the low pelvis, behind the symphysis pubis and in front of the rectum. Laterally it is related to the anterior fibres of the levator ani muscles. It is surrounded by a pseudocapsule. It is roughly conical in shape, and its upper surface, the ‘base,’ is in contact with the bladder. The prostate is divided into five lobes: anterior, posterior, median and two lateral; it is also divided into three zones: central, transitional and peripheral. Most prostate cancers arise in the peripheral zone. The seminal vesicles lie above and behind the prostate, between the bladder and rectum, and their ducts enter the base of the prostate.
Carcinoma of the prostate Risk factors and aetiology
Range of tumours Types of tumours of the prostate are shown in Table 19.1.
Incidence and epidemiology The incidence of prostate cancer in the UK is 87 in 100 000 per year; approximately 30 000 cases occur annually, with approximately 10 000 deaths in the UK. Prostate cancer is the second most common cause of cancer death in men. Disease incidence is increasing, which is attributed to increased disease screening. The peak incidence age is 70 to 75 years. The geographical distribution varies; the highest incidence occurs in Western countries and the lowest incidence is in Asia. Age-specific mortality rates have fallen slightly in both the USA and the UK, perhaps related to better treatment rather than screening because the UK has no organised screening programme.
Family history is an important risk factor for disease. The relative risk doubles if one first-degree relative is diagnosed before age 70 and is four times higher if two relatives are diagnosed and if one is under age 65. Genetic susceptibility also plays a role: the risk is doubled in Afro-Caribbean men, who may experience tumours that are more aggressive. A diet rich in animal fat and proteins may also increase the risk of disease.
Pathology The pathological features of adenocarcinoma of the prostate are shown in Table 19.2. Important stage information is obtained from systematic biopsies. A high proportion of positive biopsy cores (6/8) and the length of tumour in any individual core (e.g. > 10 mm) are highly associated with pathological T3a disease.
231
Jim Barber and John Staffurth
Table 19.1. The range of tumours that occur in the
Table 19.2. Pathological features of adenocarcinoma
prostate
of the prostate
Type
Examples
Features
Description
Benign
Benign enlargement (nodular hyperplasia)
Macroscopic
Typically arise in periphery of gland
Malignant
Carcinoma
posteriorly and form a firm, gritty
Adenocarcinoma (> 95%)
nodule
primary
Transitional carcinoma, squamous
secondary
Microscopic appearances are graded according to the Gleason system, which
(<5%)
is based on degree of glandular dif-
Others
ferentiation and growth pattern within
Lymphoma, sarcoma, carcinosarcoma,
the tumour; the Gleason ‘patterns’ are
carcinoid Malignant
Microscopic
carcinoma or small-cell carcinoma
allocated a number from 1 to 5, ranging
Direct spread from bladder or rectum
from very well to very poorly differen-
Metastatic spread from other primary
tiated; for each tumour, two scores will
sites is rare
be assigned: one for the predominant pattern and one for any secondary pattern, although patterns that occupy
Rare tumours Transitional cell carcinomas (TCCs) may arise in the prostatic urethra or ducts and should be managed in a similar way to TCC elsewhere. Small-cell carcinomas of the prostate behave in the same way as small-cell tumours of the lung and should be treated accordingly.
Spread Prostate carcinoma can spread locally to involve seminal vesicles and the base of the bladder. Spread to the rectum is inhibited by Denonvillier’s fascia. It can also spread via lymphatics to pelvic and para-aortic lymph nodes or via the blood, most often to bone, especially spine, femur, pelvis and ribs. Spread to other organs such as liver and lung is uncommon.
less than 5% are often not included; the Gleason ‘grade’ (or sum score) is obtained by adding together the two scores, but if only one grade is seen, this is doubled (i.e. single pattern 3 would become Gleason grade 3 + 3 = 6) In the UK, guidelines restrict the use of Gleason patterns 1 and 2 to TURP chips only For needle biopsies, the Gleason sum scores are Gleason 6 (3 + 3, well differentiated), Gleason 7 (3 + 4, moderately differentiated) and Gleason 8 or above (e.g. 4 + 4, 3 + 5, 4 + 5, poorly differentiated); occasionally a third (‘tertiary’) pattern is seen, but there is as yet no international consensus of the
Screening
implications of this occurrence. For very small tumours of 1 mm or less in
Screening trials have thus far been inconclusive. Screening results in approximately a four-fold increase in the incidence of prostate cancer, and many patients may be overtreated. Prostate specific antigen (PSA) screening may not be sensitive enough to detect the most aggressive tumours, because they may produce little PSA. Screening from age 50 via annual PSA and digital rectal examination (DRE) is recommended in the USA but not in the UK. The decision whether or not to have PSA measured should be an informed personal choice, but many men have PSA tests done as part of ‘routine’ medical checks. 232
diameter, the diagnosis can be confirmed by immunohistochemical staining; the myoepithelial layer of cells surrounding normal glands is absent around malignant glands; this layer is difficult to see with standard haematoxylin and eosin staining but shows clearly with high-molecularweight cytokeratin stains
Prostate
Clinical presentation Local symptoms Early prostate cancer rarely produces any symptoms, but many men present with lower urinary tract symptoms due to benign prostatic hypertrophy unrelated to their cancers. Locally advanced tumours sometimes give lower urinary tract symptoms, occasionally haematuria, and rarely perineal pain.
Table 19.3. TNM T-stage classification of adenocarcinoma of the prostate TNM stage
Description
T1
No tumour palpable or visible by imaging
T1a
≤ 5% tumour in TUR chips
T1b
> 5% tumour in TUR chips
T1c
Diagnosed by TRUS biopsy
Lymph node spread or metastatic disease
T2
Palpable tumour confined to the prostate
Most men presenting with metastatic disease have bone metastases that cause local pain or spinal cord compression but no urinary symptoms. Some patients (< 20%) have predominantly nodal metastatic disease which can lead to lower-body oedema. Lung and liver metastases are uncommon and are normally related to very-high-grade disease (> Gleason 8). Brain metastases are virtually unknown, but extradural disease that compresses the spinal cord may occur in patients with advanced disease.
T2a
Tumour involves 50% of 1 lobe or less
T2b
Tumour involves >50% of 1 lobe
T2c
Tumour involves both lobes
T3
Tumour extends through prostatic capsule
T3a
Extracapsular extension
T3b
Seminal vesicle extension
T4
Tumour is fixed or invades adjacent structures such as bladder neck, rectum, pelvic wall
TRUS = transrectal ultrasound; TUR = transurethral resection. Adapted from UICC (2002).
Investigation and staging Transrectal ultrasound (TRUS)-guided systematic sampling (‘octant biopsies’) of the peripheral zone is usually performed. Four biopsies are taken from each side: three from the mid-lobar peripheral zone of the prostate and an additional biopsy of the lateral peripheral zone. Each core should be clearly labelled so that correlation can be made with a DRE and MRI scans. This technique is more likely to detect a cancer than finger- or TRUS-guided biopsy. MRI clearly shows the peripheral zone of the prostate on a T2-weighted image, and tumours may be visible as a low-signal region in an area of intense high signal (on the T2-weighted image, the high-signal region is normal tissue). After a TRUS biopsy, blood may also appear as a low-signal area; because of this, staging scans following a biopsy should be delayed for 4 to 6 weeks. Extracapsular involvement, seminal vesicle invasion and nodal disease can be identified, as well as small bone metastases that may not be seen on a bone scan. MRI has a low specificity for small-volume nodal disease (e.g. 0.5 to 1.5 cm in size), and node sampling is sometimes performed for indeterminate lesions. CT scans give little information on the structure of the prostate, although they may be useful for nodal staging. Routine bone scans are not required for low-risk patients (PSA ≤ 10 ng/ml, T ≤ 2 and Gleason score ≤
6). Indeed, a positive bone scan is very uncommon in any patient presenting with a PSA level of 20 ng/ml or less. However, some clinicians regard a bone scan as good practice for all patients. These rules do not apply to patients who are relapsing after radical therapy or progressing on androgen deprivation therapy (ADT); their risk of having metastatic disease should be assessed individually. The role of PET scanning has yet to be fully evaluated. Local-stage information is provided by a combination of DRE, biopsy information, MRI scanning and pathology, especially in patients who have had radical surgery. The greater the agreement among these different methods, the more reliable the staging.
Staging classification The TNM T-staging classification for adenocarcinoma of prostate cancer is shown in Table 19.3. Note that, in the TNM system, T1a/b tumours are rarely diagnosed today because so few TURPs are performed. The stage groupings for T2 disease have, confusingly, been changed twice in the past 10 years (indeed, within the UK, the 1997 UICC staging is still widely used). Determining the distinction between T2 and T3 disease on DRE takes skill and experience and 233
Jim Barber and John Staffurth
may not be highly reproducible between observers. In early T3 disease, ‘extracapsular extension’ is felt as a distortion of the normal prostate shape, which is important because it identifies patients who are on the borderline for surgical treatment.
Treatment overview Most specialist MDTs in the UK aim to give patients with clinically localised prostate cancer a choice of treatment.
Disease confined to prostate Prostate-confined disease refers to tumours that are stage T2 or smaller (often called ‘organ confined,’ although this should only be a pathological designation). Choices include radical prostatectomy, interstitial brachytherapy or external beam radiotherapy (possibly with adjuvant hormonal therapy). There are also some new therapies, such as cryotherapy and high-intensity focussed ultrasound, which are currently experimental. Although some of the best results are reported for surgery, no definitive evidence exists that any one of these options is better in terms of cancer control. Retrospective surveys of quality of life have shown broadly similar results across the different treatments.
Locally advanced disease For stage T3 tumours and larger, patients should be offered external beam radiotherapy with neoadjuvant hormone therapy with or without adjuvant hormone therapy (see Chapter 19, p. 237).
Observation alone (active surveillance, watchful waiting) Observation alone can be undertaken with two different approaches, reflected in the intensity of follow-up, watchful waiting and active surveillance. Watchful waiting is ideal for patients who are unlikely to develop symptoms from prostate cancer during their lifetime; therefore, the morbidity of treatment can be avoided. Patient selection is important because the results of one randomised trial show a small but significant improvement in overall survival (p = 0.04) in patients treated immediately with prostatectomy (Bill-Axelson et al., 2005). The same trial also showed that the patients in the prostatectomy arm had reduced 10-year risks of local progression, dis-
234
tant metastases and prostate cancer deaths (HR = 0.33, 0.60 and 0.56, respectively). Ideal candidates for this approach are elderly patients (> 70 to 75 years) or younger patients with serious co-morbidities and good-prognosis tumours. The decision to treat a patient on progression is generally only taken if the patient becomes symptomatic and treatment is with palliative intent, such as with ADT. Active surveillance is an intensive approach for patients who are potentially curable but whose disease characteristics are so good that they are considered to have a relatively high chance of not developing progressive disease throughout their lifespan. It is important, however, that these patients are closely monitored; if they display evidence of progressive disease, a decision to treat with radical intent at the earliest opportunity is likely to be taken because the aim will be a cure. Surveillance involves sequential PSA testing and DRE. At diagnosis, a ‘PSA history’ that extends back over several months may be available. A PSA doubling time (PSAdt) can be estimated by inspection of a plot of PSA value against time on a log-linear plot [PSAdt = log 2 × time / (log PSA2 − log PSA)]. (The observed PSA values are called PSA and PSA2.) A PSAdt of less than 6 months would be regarded as rapid, with a high risk of metastatic disease. A PSAdt of more than 2 years indicates slow disease progression, and monitoring is a more reasonable approach. Having established a reliable baseline and estimated the PSAdt, it should not be necessary to perform a test more than every 6 months. Most patients will develop slow (and clinically insignificant) rises in PSA, which are entirely predictable from their PSA history, but a few patients have a very variable PSA level and monitoring may be difficult. Monitoring continues as long as the patient wants, provided that there is no evidence of clinical progression and that the PSA level rises no faster than expected.
Role of surgery Radical prostatectomy Radical prostatectomy is only an option for patients with T2 tumours or smaller and is an operation that should only be carried out by a specialised surgical team. The approach may be retropubic, perineal or laparoscopic; most experience is with the retropubic approach. The operation is difficult because of the rich venous plexus around the prostate and the relative lack of tissue,
Prostate
particularly around the apex, to allow for clearance of tumour with wide margins. Around 25% of patients may have positive margins after surgery. Intraoperative frozen-section analysis of pelvic nodes can spare patients with positive nodes an unnecessary operation, because these patients probably cannot be cured by surgery. Radical prostatectomy carries particular advantages: good published outcomes, immediate treatment, rapid access to prognostic information from pathological samples and decrease in PSA following surgery. Disadvantages of radical prostatectomy include high rates of erectile dysfunction (overall, at least 50%), depending on whether nerve-sparing operations can be performed, and urinary morbidity (with a small proportion of patients suffering persistent urinary incontinence). Based on the results from radiotherapy studies, it is possible that patients with high-grade cancers might benefit from immediate postoperative adjuvant hormone therapy. However, there is no evidence to support this finding. Following surgery, the PSA level should fall to undetectable levels (usually < 0.1 ng/ml). The persistence of PSA may represent either a small volume of residual benign prostate or residual cancer, but a rising PSA level suggests residual localised or metastatic disease. A variety of strategies have been used to deal with this situation, including the use of ‘salvage’ radiotherapy and hormone therapy. Factors associated with a poor outcome after salvage radiotherapy include a Gleason score of 8 to 10, short time to failure, rapid PSA doubling time (less than 10 months), high PSA and absence of risk factors for local recurrence such as margin positivity. Because radiotherapy is a local treatment, markers of systemic failure are also likely to be associated with a poor outcome. Several retrospective studies have also identified that PSA level at the time of radiotherapy can be used to divide patients into different prognostic groups, with cut-offs found at 0.5 to 2.7 ng/ml. Thus, absolute PSA thresholds for the appropriateness of salvage radiotherapy are hard to define, and it should be considered a continuum – the higher the PSA, the lower the chance of long-term success with radiotherapy. Irradiation of patients who have a PSA level greater than 2.0 ng/ml is unlikely to be successful. Finally, it is worth noting that normal prostate tissue can re-grow and that not all patients with detectable PSA after prostatectomy will experience recurrent disease.
Patients who have pathologically positive margins appear to have significantly better disease-free survival end points when treated with immediate ‘adjuvant’ postoperative radiotherapy (Bolla et al., 2005) and this approach should be considered. In the future, ‘supersensitive’ PSA assays may help in selecting patients for postoperative treatment.
Brachytherapy Brachytherapy is an option for patients with T2 or smaller tumours. Permanent interstitial implantation with transperineal insertion of iodine-125 seeds for patients with T2 or smaller tumours is now available in some UK centres. Good results, comparable to those for prostatectomy, are achieved in good-prognosis patients, but there is less experience in patients with Gleason grades of 7 and above. There is no evidence that local control is better than with external beam radiotherapy, but the treatment is attractive because very high doses of radiation (e.g. 140 to 145 Gy) can be given very accurately to the prostate. The procedure is an outpatient one, usually with only two visits to hospital. Brachytherapy treatment is done under anaesthetic, and the prostate is imaged with TRUS. It is not practical to implant patients who have T3 disease, those who have had a previous TURP (unacceptable rates of urinary morbidity) or patients with a large (50 ml) prostate because of ‘pubic arch interference.’ The implantation is performed through a coordinate grid mounted on the same gantry as the TRUS. Distribution of the sources is achieved by a combination of seeds and spacers introduced in each needle.
Morbidity from brachytherapy Acute morbidity from brachytherapy involves urinary pain, frequency and obstruction. The bowel reaction is not as severe as that for external beam radiotherapy. Late morbidity includes urinary morbidity and impotence (rates similar to those with external beam). The urethral dose should be kept below 125 Gy if possible.
High-dose-rate afterloading brachytherapy High-dose-rate afterloading brachytherapy can be performed with a Microselectron® machine using a technique very similar to that just described. Although the invasive nature of the technique limits it to one or two fractions a few hours apart, it is a practical method of
235
Jim Barber and John Staffurth
delivering a radiation boost in a highly conformal fashion.
External beam radiotherapy Radical external beam treatment Radical external beam radiotherapy is a treatment option for patients with T2 or smaller tumours and is standard treatment for patients with T3 or larger tumours. At standard doses, there is low toxicity from external beam radiotherapy to the prostate alone using conformal therapy, providing large pelvic fields are avoided (Dearnaley et al., 1999). There is increasing evidence that suggests an improvement in biochemical disease-free survival with higher doses of radiation, but no survival benefit so far (Pollack et al., 2000). Most oncologists in the UK are currently using only a modest escalation in dose over that used 10 years ago. There are diminishing returns from the highest doses (>78 Gy) because margins must be decreased to maintain patient tolerance and the risk of missing the tumour increases. The radiotherapy technique involves the following. r Patient preparation, positioning and immobilisation. Unwanted patient movement can be limited by using an immobilisation device (e.g. knee support). Normally the patient is advised to hold a comfortably full bladder for treatment. r Localisation. Outlining the prostate on CT is made easier by using information from MRI scanning, particularly to identify the apex (most inferior portion) of the prostate and seminal vesicles. A description of margins and doses follows. r Plan. A three-field plan with an anterior and two lateral beams adequately treats the prostate. Use of two lateral fields rather than posterior obliques allows greater sparing of the rectum. The minimum amount of rectum should be included in the high-dose volume by inspection of the plan and DVH. Less than 25% of the rectal volume should receive more than 95% of the dose and less than 50% of the volume more than 50% of the dose. r Verification. Regular electronic portal imaging for quality assurance, for example daily for the first 3 days and then weekly, should be routine, but it will only document setup variation related to bony anatomy. Methods for ‘on treatment’ localisation of the prostate include ultrasound, gold seed markers implanted in the prostate, and inflatable rectal balloons, but none is standard in the UK.
236
Please note that doses are often reported as a ‘target minimum’, rather than according to ICRU 50, and many regimens use 1.8 Gy fractions; be sure to specify fraction size and the dose specification point. The doses quoted here are specified to ICRU 50 and may be up to 5% higher than target minimum doses. The GTV is usually the whole prostate. The dose that can be prescribed depends on the target volume as follows. The standard dose, for which conformal blocking is preferred but not mandatory, with 1 cm margins around the GTV, is 55 Gy in 20 fractions of 2.75 Gy or 64 to 70 Gy in 2 Gy fractions. A Canadian trial showed little difference between these two regimens (Lukka et al., 2005). The medium-dose escalation, for which conformal blocking is essential, with margins no larger than 1 cm, is 70 to 74 Gy in 2 Gy fractions. The high-dose escalation, for which conformal blocking is essential, requires a high degree of precision in treatment delivery and quality assurance: consider reducing the margins from 1 cm to 0.5 cm after 56 Gy or using a single-phase margin of < 1 cm (e.g. 7 to 8 mm). The high-dose regimen is 74 to 78 Gy in 2 Gy fractions. For doses much higher than 78 Gy in 2 Gy fractions, morbidity becomes unacceptable with conventional planning. Seminal vesicles can be included in treatment up to 56 Gy in 28 fractions if the patient is at greater than low risk (if PSA + 10[Grade − 6] < 15). If the seminal vesicles lie in a favourable position (not curving too far posteriorly), it is possible to give a higher dose without increasing the rectal dose. If the seminal vesicles are macroscopically involved on an MRI scan, then as much of the seminal vesicles as possible should be given the full dose of radiation. Pelvic treatment, rarely practised and controversial in the UK, is widely practised in the US and Europe. The RTOG 9413 trial (Roach et al., 2003) suggested that a combination of concurrent hormonal therapy and pelvic radiation for patients at high risk of pelvic nodal spread gave better biochemical survival than prostate radiation alone. The risk of nodal involvement is high (20%) in patients with high-grade disease (Gleason 8 or higher) and PSA level greater than 20, or can be estimated using the Roach formula (RISK = 2/3 PSA + 10[Grade − 6]). It is convenient to plan a two-phase treatment including the pelvis when using CT planning so that conformal blocking to the rectum can be
Prostate
performed throughout the treatment to the portion of the rectum lying in the high-dose volume. This treatment is a compromise because some lymph nodes that lie more posteriorly in the pelvis may be shielded. Doses of 50 Gy to the pelvis with a boost of 20 Gy to the prostate are associated with unacceptable toxicity in this approach. For phase 1, give 45 Gy in 25 fractions to the prostate and pelvis. The PTV is the whole pelvis up to approximately S2/3 with conformal blocking of the rectum. Phase 2 involves only the prostate; give 24 Gy in 12 fractions. Further dose escalation is possible but could be limited by an increased rectal dose from the larger phase 1 dose. Seminal vesicles can be included subject to rectal doses being acceptable.
Postoperative adjuvant radiotherapy The most appropriate target volume is currently uncertain. The anastomosis should be included and the volume guided by the use of preoperative MRI scans and pathological information about positive margins. The dose is 60 to 64 Gy in 30 to 32 fractions or 55 Gy in 20 fractions.
Follow-up after radiotherapy The definition of biochemical failure after radiotherapy using serial estimations of PSA has been controversial, but the ASTRO consensus defines it as three successive increases in PSA above a nadir value, with the date of recurrence being the midpoint between the nadir value and the first rise. PSA testing gives advance information (sometimes many years’ warning) about relapse after radiotherapy. In patients who have received radiotherapy alone without hormone therapy, a stable PSA 4 to 5 years later probably indicates a cure. Cure rates are highest in those patients with a PSA nadir of less than 0.5 ng/ml. However, the original ASTRO definition was realised to be unworkable if neoadjuvant ADT had been used, because the PSA level often rises sequentially above the hormone-induced nadir even in patients who do not eventually relapse. A recent comparison of three methods for determining biochemical relapse (ASTRO, modified ASTRO, and nadir + 2 ng/ml) showed that the nadir + 2 ng/ml definition was the best predictor of sustained, true, biochemical and clinical failure (Buyyounouski et al., 2005). The treatment of patients who develop PSA relapse following RT can be guided by the rate of rise of PSA or
PSAdt and by the results of restaging with a bone scan. Immediate hormone therapy is advised for patients with metastatic disease or a PSAdt of less than 6 months.
Principles of hormonal therapy (see Chapter 3) Surgical castration by subcapsular orchidectomy or medical castration by LHRH analogue provides the same biochemical result, which is a castrate level of testosterone. With LHRH analogues, patients should be prescribed an antiandrogen to prevent tumour flare (see Chapter 3, p. 32). Bicalutamide and flutamide are non-steroidal antiandrogens (NSAAs) that result in a normal or high testosterone level and high gonadotrophin levels. The advantage is less erectile dysfunction and less loss of libido, but the resulting gynaecomastia can be distressing. It can be partly prevented by giving low-dose radiotherapy (either 21 Gy in 3 fractions or 10 Gy in a single fraction to the breasts) and/or by prescribing tamoxifen (see Chapter 3, p. 34). LHRH analogues give disease control and survival rates in metastatic disease equivalent to those of surgical castration (80% overall response rates). NSAAs such as bicalutamide, when used on their own, are less effective. The combination of LHRH analogue and NSAA (‘maximal androgen blockade’) has shown no significant benefit over LHRH analogue alone in several meta-analyses. The use of NSAAs should be limited to second-line hormonal therapy (in combination with LHRH analogue) or for patients unable to tolerate an LHRH analogue.
Adjuvant and neoadjuvant hormonal therapy There have been a number of trials of neoadjuvant and adjuvant therapy (see Chapter 3, p. 32 and Table 3.5, p. 33). Several of these trials have shown a survival benefit for prolonged hormone therapy in patients with high-grade tumours. Patients with high-grade tumours should be offered treatment with an LHRH analogue for 3 years or longer. For patients with Gleason grades 6 and 7, the effect on survival of a short period of neoadjuvant hormone therapy (3 to 6 months) combined with radiotherapy is uncertain, although this treatment is widely used in the UK and USA. There appears to be a significant disease-free survival benefit, and there are reductions in the planned target volume, which may reduce radiation
237
Jim Barber and John Staffurth
toxicity. Any potential benefit in tumour control must be offset against considerable morbidity including impotence, sweats and flushes, depression and a long-term effect on bone density. Although morbidity may be less with bicalutamide, the trial data are immature (Wirth et al., 2004) and there is no apparent survival benefit; bicalutamide cannot currently be recommended for adjuvant therapy.
Palliative treatments Hormone treatments for metastatic disease Patients who present with metastatic disease should be started on hormonal therapy because the majority (80%) respond well. When the patient presents with a high PSA (>100 ng/ml), bone metastases and a palpable tumour, a biopsy may not be needed but may give prognostic information and be necessary for entry into clinical trial protocols. Patients are treated with an LHRHa, but they must be started on an antiandrogen first to prevent the tumour flare that can occur with LHRHa. Cyproterone acetate 100 mg t.d.s is the standard antiandrogen, but bicalutamide 150 mg o.d. can be used as an alternative in patients with cardiac disease or hypertension, in whom cardiac failure can occur with cyproterone acetate. The PSA falls to a low level (in many cases < 0.1 ng/ml) and the median duration of biochemical response is around 18 months. Clinical relapse often occurs after about 2 years but there is wide variation. Few patients die before 2 years, but few live more than 5 years. Occasionally patients present as an emergency with spinal cord compression before having any hormone therapy. The chance of substantial recovery is high with hormonal treatment alone. It is very important to achieve castrate testosterone levels as an emergency, either by orchidectomy, DES or cyproterone, in combination with high-dose corticosteroids. In most cases surgical intervention to the spine can be deferred at this stage unless there is neurological deterioration. Radiotherapy to the spine is often given, although it is uncertain whether it provides any additional benefit over hormone therapy alone. Patients who have prolonged responses to primary hormone therapy have a good chance of responding to second-line hormone therapy such as bicalutamide 50 mg. Corticosteroids can produce dramatic responses partly because of their adrenal suppressive action, but overall the response rate is similar to that for NSAAs. Prednisolone 10 mg is well tolerated. Dexamethasone at 1.5 to 2 mg is sometimes highly effective but, when 238
given for prolonged periods, causes serious steroidrelated morbidity. For more details, see Chapter 3 (p. 35).
Palliative radiotherapy Bone metastases usually respond well to single fractions of external beam radiotherapy. Strontium-89 therapy (e.g. 150 MBq) gives longer lasting pain relief for multiple sites of disease but no survival benefit. An alternative is extended-field or hemi-body treatments, which are well tolerated if 5-HT3 antagonist antiemetics are used. Treatment can be hazardous in patients who have poor bone marrow reserve because it may cause prolonged thrombocytopenia.
Cytotoxic chemotherapy The use of cytotoxic chemotherapy should be an informed individual choice. A combination of docetaxel (75 mg/m2 every 3 weeks) and prednisolone appears to confer a 2 to 3 month survival benefit compared to a combination of mitoxantrone and prednisolone (Tannock et al., 2004). Use of up to ten cycles of docetaxel has just been approved by NICE for patients with hormone-refractory disease and a Karnofsky performance status of 70 or more. Some patients can achieve a substantial palliative benefit, but many have insufficient bone marrow reserve, or are too unwell, to tolerate cytotoxic chemotherapy. The role of docetaxel given with other systemic therapies is being investigated in the ‘STAMPEDE’ study.
Bisphosphonates The role of bisphosphonates in the prevention of ‘skeletal-related’ events in prostate cancer is less clearcut than in breast cancer, because pathological fractures are relatively uncommon and most men with painful bone metastases respond very well to external beam radiotherapy. In addition, pamidronate appears to have little or no activity in metastatic disease, and zoledronate only weak activity (Saad et al., 2004). Patients with advanced prostate cancer have an increased rate of osteoporosis, which may be exacerbated by the use of long-term ADT, which is associated with both a reduction in bone mineral density and an increased fracture risk. There is no standard approach to monitoring these men; baseline and serial (e.g. biennial) bone densitometry for men starting long-term ADT, with an appropriate therapeutic intervention for patients found to be osteoporotic, seems sensible. Use of bisphosphonates in this setting has been shown to
Prostate
improve bone mineral density but its effect on fracture rate has not yet been reported.
Prognosis Prognostic groupings for localised prostate cancer (N0 M0) Many groups have developed their own definitions and groupings for evaluating prostate cancer. A good prognosis involves a PSA < 10, a Gleason grade of 6 or lower and a tumour that is stage T2 or less (death rate from prostate cancer over 10 years is < 10%). An intermediate prognosis involves any of the following: PSA 10 to 20, Gleason 7 or stage T3 (up to 10 to 20% death rate over 10 years. A poor prognosis is defined by any Gleason score of 8 or higher and a PSA > 20 (resulting in a > 20% chance of death over 10 years).
Areas of current interest Role of intensity modulated radiation therapy (IMRT) IMRT is widely used in the USA for the treatment of prostate cancer. It is not clear to what extent IMRT can improve the outcome for prostate-only treatments, but it could provide better coverage of seminal vesicles and pelvic nodes while sparing the rectum. A reduction in rectal toxicity may improve patient outcomes, but longer follow-up is required to show whether there is an effect on survival. The use of IMRT should be considered experimental.
Ongoing clinical trials RT01 is a dose escalation study (64 versus 74 Gy), for which results have not yet been reported. PR07 is a trial involving long-term hormone therapy with or without radiotherapy for locally advanced disease; this study has recently closed. The PROTECT study is an ongoing trial that randomises screen-detected prostate cancer patients to prostatectomy, radical radiotherapy or surveillance. Both the RTOG and EORTC are conducting randomised trials of hormone therapy in early stage dis-
ease, but it may be many years before these results are available. The CHIP study is an ongoing trial comparing 74 Gy with high-dose hypofractionated treatment. In the ‘STAMPEDE Study,’ patients commencing on hormone therapy are randomised to a variety of options including immediate docetaxel; this trial is commencing shortly.
REFERENCES Bill-Axelson, A., Holmberg, L., Ruutu, M. et al. (2005). Radical prostatectomy versus watchful waiting in early prostate cancer. N. Engl. J. Med., 352, 1977–84. Bolla, M., van Poppel, H., Collette, L., et al. (2005). Postoperative radiotherapy after radical prostatectomy: a randomised controlled trial (EORTC trial 22911). Lancet, 366, 572–8. Buyyounouski, M. K., Hanlon, A. L., Eisenberg, D. F. et al. (2005). Defining biochemical failure after radiotherapy with and without androgen deprivation for prostate cancer. Int. J. Radiat. Oncol. Biol. Phys., 63, 1455–62. Dearnaley, D. P., Khoo, V. S., Norman, A. R. et al. (1999). Comparison of radiation side-effects of conformal and conventional radiotherapy in prostate cancer: a randomised trial. Lancet, 353, 267–72. Lukka, H., Hayter, C., Julian J. A. et al. (2005). Randomized trial comparing two fractionation schedules for patients with localized prostate cancer. J. Clin. Oncol., 23. 6132–8. Pollack, A., Zagars, G. K., Smith, L. G. et al. (2000). Preliminary results of a randomized radiotherapy dose-escalation study comparing 70 Gy with 78 Gy for prostate cancer. J. Clin. Oncol., 18, 3904–11. Roach, M. III, DeSilvio, M., Lawton, C. et al. (2003). Phase III trial comparing whole-pelvic versus prostate-only radiotherapy and neoadjuvant versus adjuvant combined androgen suppression: Radiation Therapy Oncology Group 9413. J. Clin. Oncol., 21 1904–11. Saad, F., Gleason, D. M., Murray, R. et al. (2004). Long-term efficacy of zoledronic acid for the prevention of skeletal complications in patients with metastatic hormone-refractory prostate cancer. J. Natl. Cancer Inst., 96, 879–82. Tannock, I. F., de Wit, R., Berry, W. R. et al. (2004). Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N. Engl. J. Med., 351, 1502–12. UICC. (2002). TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 184–7. Wirth, M. P., See, W. A., McLeod, D. G. et al. (2004). Bicalutamide 150 mg in addition to standard care in patients with localized or locally advanced prostate cancer: results from the second analysis of the early prostate cancer program at median followup of 5.4 years. J. Urol., 172, 1865–70.
239
20
TESTIS Jim Barber and John Staffurth
Introduction The treatment of testicular cancer is a success story for oncology, with high cure rates even for patients with advanced metastatic disease. The management has changed little over the past 20 years, there is a high degree of consensus about treatment, and standard protocols are well developed. Nonetheless, patients with testicular cancer are best managed by specialised multidisciplinary teams.
Range of cancers The range of testicular cancers is shown in Table 20.1.
Germ cell tumours Incidence and epidemiology The annual incidence of germ cell tumours in the UK is approximately 800 cases. Germ cell tumours are the most common malignant tumours that occur in men between the ages of 25 and 35; there is a second peak between the ages of 55 to 65 years. The reason for the rapid rise in incidence in the Western world is not known. Theories range from the impact of environmental oestrogens, related to oral contraceptive use, to the increased scrotal temperatures resulting from the use of disposable diapers in infancy.
Risk factors and aetiology Both genetic and epigenetic factors probably affect the development of testicular cancer. There is certainly a genetic component, with around 2% of cases reporting an affected first-degree relative, and a ten-fold increased relative risk in a brother of an affected relative. However, a range of conditions associated with subnormal testicular development, such as testicular maldescent, Klinefelter’s syndrome, Down’s syndrome and subfertility, are also associated with a higher risk of cancer. 240
Testicular maldescent (cryptorchidism) is associated with an approximately ten-fold increased risk of a testicular tumour. Orchidopexy, if done when the child is young enough (probably before 2 years of age), partially reduces this risk. The mechanism of carcinogenesis is unknown, but it is likely that a loss of inhibitory feedback to the pituitary gland from the abnormal germinal epithelium may predispose a person to cancer by continued hormonal stimulation of the germ cells. It is standard practice to perform orchidopexy on maldescended testes because this is thought to reduce their malignant potential.
Screening There is no evidence that population screening for testicular cancer reduces mortality, which would be expected in a rare disease with effective treatment available. Despite periodic health education campaigns directed towards young males, the US Preventive Services Task Force recently concluded that screening was more likely to do harm than good. Testicular cancer patients should all be taught testicular self-examination, because they run a particularly high risk of second, contralateral cancers.
Pathology and classification of germ cell neoplasms Germ cell neoplasms can be classified into the following: r Intratubular germ cell neoplasia – abnormal germ cells within the seminiferous tubules which resemble germ cells at early stages of differentiation. r Teratoma or ‘non-seminoma’ – macroscopically firm nodular tumour with areas of haemorrhage and necrosis. For microscopic features see Table 20.2. r Seminoma, which can be classical or spermatocytic. Classical seminomas have fibrous septa, lymphocytic infiltrate, typical large round cells with distinct cell borders, clear cytoplasm and large nuclei and prominent nucleoli. They may co-exist with non-seminoma
Testis
Table 20.1. Range of testicular cancers Type
Examples
Germ cell tumours
Seminoma Teratoma (non-seminomatous germ cell tumour) Mixed tumour (seminoma and teratoma) Spermatocytic seminoma
Non-germ cell tumours
Sex cord/gonadal stromal tumours; Leydig cell tumours, sertoli cell tumours Haemopoietic tumours (lymphoma, leukaemic infiltrate) Metastatic tumours Mesenchymal tumours (embryonal
develop between the testicular/spermatic cord lymphatics and local lymphatics in the groin. In these patients, the pelvic lymph nodes may be at higher risk of spread. Theoretically, this should not affect patients who had contralateral surgery only. The presence of lymphovascular invasion is a helpful indicator of the risk of metastatic disease in patients with teratoma. In stage 1 disease, the risk of relapse is approximately 50% in the presence of vascular invasion but only 20% in the absence of vascular invasion. However, in patients with seminoma, vascular invasion is less helpful as an indicator. True vascular invasion is rare but is sometimes seen as an artefact of slide preparation; tumour size (> 4 cm) and involvement of the rete testis appear better indicators of a higher risk of relapse. Table 20.3 shows patterns of spread for seminoma and teratoma.
rhabdomyosarcoma, leiomyosarcoma, etc.) Benign conditions
Epidermoid cyst
which could be confused with tumours
(designated a mixed tumour), but ‘pure seminomas’ may nonetheless contain a few syncytiotrophoblastic elements. Spermatocytic seminomas have a superficial resemblance to classical seminoma but are a distant relation. There are three types of cells with different nuclear size: large, small and intermediate. Some nuclei may exhibit a presence of nuclear thread-like chromatin. Table 20.2 shows the subgroups of teratoma/nonseminoma. It is important for those treating patients with testicular cancer to be aware of the differences between the WHO and British systems, which are highlighted in this table.
Clinical presentation The majority of patients present with a painless testicular swelling which is sometimes discovered after trauma. On examination, the mass normally appears to be in the testis. It is useful to know that extratesticular swellings are usually benign epididymal cysts or represent epididymitis, whereas tumours in the epididymis are extremely rare. Breast enlargement is rare and usually associated with a high βhCG level. Patients with advanced disease can present with nonspecific symptoms such as fatigue and weight loss, associated with back pain. They may present with dyspnoea due to lung metastases or associated pulmonary emboli. A para-aortic mass may cause ureteric obstruction and renal failure (more often in patients with seminoma rather than with teratoma). Mediastinal germ cell tumours may present with classic signs and symptoms of superior vena caval obstruction.
Investigation and staging Pattern of spread The ‘first station’ lymph nodes for tumours of the testis are the inter-aortocaval nodes for right-sided tumours (located in the midline) and the left para-aortic nodes for left-sided tumours (located near the left renal hilum). In patients who have had ipsilateral inguino-scrotal surgery in the past (e.g. orchidopexy as a child, but probably not vasectomy), anastomotic lymphatics can
Tumour markers Serum tumour markers are important in staging and in monitoring therapy. βhCG arises from syncytiotrophoblastic elements and is raised in 10 to 20% of patients with seminoma and around 35% of those with teratoma. Most stage 1 seminoma patients have normal markers, although marginal increases in βhCG (5 to 10 iu/l) are common and levels tend to fall to less than 1 iu/l after 241
Jim Barber and John Staffurth
Table 20.2. British and American (WHO) systems of classification of teratoma/‘non-seminoma’ British
American/WHO
Microscopic appearance
Teratoma differentiated
Mature teratoma; immature teratoma
Differentiated tissue only
Malignant teratoma
Teratoma with embryonal carcinoma and/or yolk
Undifferentiated tumour and/or yolk sac
intermediate Malignant teratoma
sac tumour
tumour with differentiated tissue present
Embryonal carcinoma
Undifferentiated malignant teratoma only
undifferentiated Malignant teratoma trophoblastic
(may include yolk sac elements) Choriocarcinoma (implies that this is the only element present, high risk for brain metastases)
orchidectomy. Levels of βhCG much over 100 iu/l are normally associated with teratoma but occasionally are seen in patients with seminoma containing syncytiotrophoblastic elements. Massive increases in βhCG may indicate metastatic choriocarcinoma. αFP arises from yolk sac elements. It is raised in around 60% of patients who have teratoma, but not in patients with seminoma. Borderline elevation and/or fluctuation of αFP levels (up to around 20 ng/ml) is common, and in many patients no obvious cause can be found. Repeating the measurement of αFP levels every week can be diagnostic because patients with active cancer generally have exponentially rising αFP levels. LDH is required for allocation of prognostic grouping. A baseline hormone profile (testosterone, etc.) may be useful.
Other investigations When a patient is suspected of having a testicular cancer, initial investigation should consist of testicular ultrasound and a check for serum tumour markers. These should be measured preoperatively in all patients. If a tumour is confirmed, most patients should undergo a high inguinal orchidectomy because a scrotal incision may result in disruption to the normal lymphatic drainage and predispose the patient to involvement of the iliac or inguinal nodes. However, in patients with very high marker levels, or in those with symptoms suggestive of metastatic disease, the orchidectomy may be deferred until after chemotherapy treatment. CT scanning of the chest, abdomen and pelvis should be performed in all patients, before surgery if possible; MRI of the brain should be considered if there is choriocarcinoma or the patient is in a poor prognostic group. 242
Malignant trophoblast (syncytiotrophoblast plus cytotrophoblast)
Staging/prognostic grouping of testicular tumours For practical purposes, most clinicians make use of the excellent Royal Marsden Hospital staging system for early disease (stage I teratoma and seminoma, stage II seminoma), but decision making in stage II teratoma and disease that is stage III and above is made by using prognostic groupings according to the International Germ Cell Consensus Classification (IGCCCG, 1997). Technically speaking, prognostic group allocation is determined from the nadir levels of αFP and βhCG after orchidectomy, not from the preoperative levels. The stage groupings and prognostic categories are shown in Tables 20.4 and 20.5, respectively.
Management of intratubular germ cell neoplasia (IGCN) Intratubular germ cell neoplasia is frequently found alongside germ cell tumours and in the contralateral testis in around 5% of cases. It is the accepted precursor of testicular cancer, and some studies suggest a 50% progression rate to cancer within a few years. The risk of contralateral IGCN is high in patients with testicular atrophy (Harland et al., 1998). There is probably an association with testicular microlithiasis although, more often than not, the conditions occur independently. Some experts advocate routine biopsy of the contralateral testicle at the time of the orchidectomy, whereas others perform a biopsy only for high-risk patients (i.e. with a testicular volume < 15 ml). If IGCN is found, it can be eliminated by treatment with radiotherapy, for example, 20 Gy in 10 fractions;
Testis
Table 20.3. Modes of spread for seminoma and teratoma Mode of spread
Seminoma
Teratoma (non-seminoma)
Local (invasion of local
Rete testis (signifies increased risk of nodal
Rete testis as for seminoma
non-testicular structures is rare) Lymphatic
involvement) Highly predictable in a stepwise manner direct to para-aortic nodes and then to pelvic and
Similar to seminoma but less predictable
mediastinal nodes Blood borne (N. B., incidence is
Relatively uncommon
Common in lungs; rare in liver,
50% in the presence of vascular
brain and bone but this signifies
invasion)
a poor prognostic group
however, such treatment may cause infertility. Many IGCN patients already have borderline hypogonadism, and radiotherapy (even at 20 Gy) may further disrupt Leydig cell function. The quality of life may be poor in patients who are on testosterone replacement therapy because of the lack of high-quality depot preparations. Side effects of testosterone replacement include acne, polycythaemia, fluid retention, sleep apnoea and testosterone-induced prostate growth. An alternative to biopsy and radiotherapy is a combination of testicular self-examination and annual ultrasonic follow-up. Ultimately, the decision must rest with the patient.
Treatment overview: teratoma and seminoma Stage I disease Seminoma is exquisitely sensitive to both radiotherapy and chemotherapy, so adjuvant therapy is commonly used in stage I disease, even when the relapse risk is low. In stage I teratoma the threshold for offering adjuvant therapy is much higher, because the chemotherapy regimen (BEP) is more toxic, and many patients are managed solely with surveillance. Because teratomas produce tumour markers more commonly than seminomas, surveillance is easier in patients with teratoma.
Treatment of stage I and II seminoma Stage I seminoma Adjuvant radiotherapy Around 15 to 20% of patients with stage I disease will relapse after orchidectomy, primarily and predictably within the para-aortic nodes. Adjuvant radiotherapy to the para-aortic nodes has been practised for over 50 years and gives excellent cure rates. This success must be weighed against the toxicity of the treatment; although well tolerated by many, a few patients develop prolonged fatigue and there are concerns over the chances of developing a second malignancy from the treatment. The estimation of the increased risk of second malignancy is complex and requires case-control studies because there are no available mature data from randomised trials. Most of the published data are for relatively short follow-up (15 years), but even so there is a well-documented increased risk (∼ two- to three-fold) of second malignancies within the radiation portals (stomach, pancreas, bladder and kidney; Bokemeyer and Schmoll, 1995). This radiation risk is a serious concern because of the large volume of tissue irradiated (containing many vital organs, bowel, pancreas, stomach, etc.) and the relative youth of the patients; moreover, around 85% of the patients have no need of adjuvant treatment and would have received unnecessary radiation.
Surveillance Patient relapse can be detected by reguMetastatic disease Patients with seminoma may be treated with radiotherapy up to stage II, but all other patients with metastatic disease should receive combination chemotherapy.
lar CT scanning, and treatment is then confined to those who relapse. Because relapses are seldom detected by tumour markers or CXR, CT scanning is essential; it is usually done every 3 to 6 months in the first year, and 243
Jim Barber and John Staffurth
Table 20.4. Stage groupings for testicular tumours Royal Marsden Hospital stage/IGCCC prognostic group
Description
0
In situ carcinoma
I
Tumour confined to testis, normal staging scans αFP/βhCG must fall to normal Can be further divided into high-risk (LVI positive) and low-risk (LVI negative)
IM
Persistently raised tumour markers after orchidectomy (uncommon with better
II (NB usually seminoma only)
Infradiaphragmatic lymph node involvement
quality CT scanning) IIa < 2 cm IIb 2–5 cm IIc > 5 cm III
Supradiaphragmatic lymph node involvement IIIa < 2 cm IIIb 2 – 5 cm IIIc > 5 cm
IV
Extranodal metastases IVa < 2 cm IVb 2–5 cm IVc > 5 cm Lung metastases: L1 < 3 lesion; L2 > 3 lesions; L3 > 3 lesions > 2 cm H+ liver; B+ brain; N+ neck node; M+ mediastinum
αFP = alpha feto-protein; βhCG = beta human chorionic gonadotrophin; CT = computed tomography; IGCCC = International Germ Cell Consensus Classification; LVI = lymphovascular invasion.
then rapidly scaled down to annual scanning. Although this approach is not recommended in the NICE guidance outside clinical trials, it is an internationally used strategy, and the overall survival from the three largest surveillance series is 99.3%. MRI surveillance of para-aortic nodal disease, when combined with CXR and tumour markers, is an alternative using less radiation exposure.
Adjuvant
chemotherapy Adjuvant chemotherapy with a single cycle of carboplatin AUC 7 gives 2-year tumour control and survival rates equivalent to adjuvant para-aortic radiotherapy (Oliver et al., 2005). There is a reduction in the incidence of contralateral germ cell tumours in the short term. There are also uncertainties about the risk of late relapse and about the long-term effect of the drug on fertility because the follow-up time of the trials is too short, which has led to widely differing policies on sperm banking in these patients.
244
So which option? Many oncologists in the UK use a single cycle of carboplatin, but this practice is not universally accepted. Despite the risk of second malignancy, radiotherapy is still widely used in the USA and there is a wide variation of practices throughout the world. The pattern of relapse with adjuvant carboplatin is quite different from that after radiotherapy; in patients having radiotherapy the relapses are mainly in the chest and pelvis, whereas in those treated with carboplatin the relapses tend to occur in the para-aortic region. This finding suggests that the treatments are to some extent complementary and gives a rationale for combining them in the treatment of stage II seminoma.
Stage II seminoma Traditionally, patients with stage IIa/IIb seminoma are treated with ‘dogleg’ radiotherapy with a boost for bulky disease. Treatment failure is considerably more likely than in stage I disease, and around 20% of patients will
Testis
Table 20.5. Prognostic categories for testicular tumours Prognostic group
Description
‘Good-prognosis’
All of the following:
metastatic 95%
αFP < 1000 ng/ml
cure
βhCG < 5000 iu/l LDH < 1.5× normal Testicular primary site
‘Intermediate-
Any of the following:
prognosis’
αFP 1000–10 000 ng/ml,
metastatic 80%
βhCG 5000–50 000 iu/l
cure
LDH 1.5–10× normal Primary site – retroperitoneal teratoma, or any non-testicular seminoma site
‘Poor-prognosis’
Any of the following:
metastatic 50%
αFP > 10 000 ng/ml
cure
βhCG > 50 000 iu/l LDH > 10× normal Mediastinal teratoma primary site Liver/brain/bone metastases (‘non-pulmonary visceral metastatic disease’)
αFP = alpha feto-protein; βhCG = beta human chorionic gonadotrophin; LDH = lactate dehydrogenase. Adapted from IGCCCG (1997).
relapse despite this treatment. A single course of carboplatin before radiotherapy is effective in reducing the relapse rate to around 5% (Patterson et al., 2001). The optimal management of this group of patients has not been fully defined (von der Maase, 2001). The optimum size of the radiation field has not been determined in clinical trials; para-aortic strip nodal irradiation alone may be adequate, given the effectiveness of carboplatin in reducing relapse rates in clinically uninvolved nodal areas (see following text). Combination chemotherapy (with EP; see the section on chemotherapy) is often used to treat patients with stage IIc (bulky) seminoma. The advantages of better cancer control rates must be offset against the greatly increased morbidity and a small chance of treatmentrelated death. The location of the tumour is also important in making the treatment decision. For example, even quite large, centrally sited inter-aortocaval nodes
(often associated with right-sided tumours) can be treated with radiotherapy and a single cycle of carboplatin, whereas left-sided nodes may overlie the left kidney and require combination chemotherapy to avoid irradiation of a large portion of the kidney.
Technique of radiotherapy in seminoma: strips and doglegs For adjuvant treatment it is generally accepted that in most cases only 20 Gy in 10 fractions (rather than 30 Gy or more) is required to sterilise microscopic seminoma and that a para-aortic field alone (rather than a dogleg) is adequate (Fossa et al., 1999). The reduction in dose probably reduces the acute toxicity, and using a smaller treatment volume probably reduces the chance of second malignancies. Whereas a dose of 20 Gy is now well accepted for stage I disease, a dose of 30 Gy in 2 Gy fractions (perhaps with a boost of 5 Gy in 2 fractions to any bulky areas) is still considered appropriate for stage IIa/b disease. It would be reasonable to reduce the dose slightly (e.g. omitting the boost) if a single cycle of carboplatin AUC 7 is given in addition to the radiotherapy.
Patients with a low risk of microscopic pelvic disease: para-aortic field only Practice varies among centres, from using rectangular fields planned using orthogonal films combined with an IVU, to using full custom blocks or MLC to shield the kidneys accurately. Most centres now have access to a CT planning scan or CT simulator to delineate the volume, which is usually best delivered with parallel fields. This then allows most of both kidneys to be shielded easily. The standard target volume for stage I seminoma patients is the para-aortic nodes from around T11 to L5. Care must be taken, particularly for left-sided tumours, to ensure that the field extends to the medial border of the left kidney. The lateral borders of the target volume lie around 4 to 4.5 cm from the midline.
Patients with a higher risk of microscopic pelvic disease: dogleg field Patients with pathological paraaortic nodes (stage II) and those who have a history of ipsilateral inguino-scrotal surgery are considered to be at a higher risk of microscopic pelvic disease. Pathological para-aortic nodes must be identified on the planning scan (if done without contrast enhancement then with the aid of diagnostic imaging). For leftsided nodes, there is a fine balance between allowing sufficient lateral margins and including too much of the left kidney. Most pathological nodes lie within the 245
Jim Barber and John Staffurth
traditional 8 to 9 cm field width, but occasionally eccentrically located nodes may occur outside this range. The dogleg field is similar to the para-aortic field but extends inferiorly to include the ipsilateral iliac nodes to the level of the obturator foramen. The lateral border of the field normally extends to the pelvic side wall, and medially the bladder and central pelvic contents can usually be shielded. The addition of the dogleg field increases the treatment volume considerably, and so the risk of second malignancy is likely to be much higher than with a para-aortic field alone. Scrotal irradiation, using kilovoltage radiation and lead shielding over the contralateral testicle, is occasionally used for patients who have had an extensive tumour in the spermatic cord or had scrotal violation during surgery. However, with so much uncertainty about the site of residual disease, chemotherapy may be preferred in these patients.
Toxicity of radiotherapy Most patients experience severe nausea but this usually can be controlled by the use of 5-HT3 antagonist drugs. Mild tiredness is common, but occasionally it can be prolonged – months or even years. A dogleg field may deliver around 40 cGy of scattered radiation to the scrotum throughout a course of treatment, depending on the field size and machine used (probably a lot more than with a para-aortic strip alone). Even at this dose, there is no doubt that sperm production is affected; there is a well-documented drop in sperm counts to subfertile levels and sperm banking should be considered. Patients should be also advised to avoid conception for 6 to 12 months to avoid the possibility of a teratogenic effect if the sperm count has not actually fallen to subfertile levels. The risk of second cancers is as discussed earlier.
Treatment of stage I teratoma Patients without lymphovascular invasion: low risk of relapse The chance of relapse in patients without lymphovascular invasion is less than 20% and many of these patients have positive tumour markers at diagnosis; they are best managed by surveillance. The risks of adjuvant chemotherapy and/or surgery outweigh the benefits, and most patients who relapse can be cured with systemic chemotherapy. Tumour markers should return to normal levels after surgery and weekly marker samples should be taken 246
until this occurs. Monthly visits are recommended initially but after the first 6 to 12 months the interval between visits can be extended. A typical schedule may be every 2 months in year two, every 3 months in year three, every 4 to 6 months in years four and five, and then annually. Tumour-marker estimations should be done at all visits and a CXR is usually taken, although this rarely identifies first relapse. Follow-up CT scanning is done at around 3 months and at 1 year; the value of extra CT scanning is debated. The recently reported MRC trial TE08 randomised 400 patients between surveillance programs with either two or five CT scans during the initial 2 years of follow-up. Preliminary results, reported at ASCO in 2006, support the use of the two-CT-scan protocol (Mead et al., 2006).
Patients with lymphovascular invasion: high risk of relapse If the tumour shows lymphovascular invasion, the chances of a relapse are high – around 45%. There are three ways of managing these patients: r Surgical staging (with two cycles of BEP for pN+ disease). In the USA (but not in the UK), ‘staging’ retroperitoneal lymph node dissections are routinely performed for stage I patients. Patients with positive para-aortic nodes at surgery (pathological stage II disease) may be cured with surgery alone, but there is a high relapse rate. Therefore, most receive adjuvant chemotherapy (two cycles of BEP), which has been shown to reduce recurrence but not improve overall survival (Williams et al., 1987). There is a strong feeling outside the USA that staging surgery adds morbidity but does little to affect the long-term outcome. r Adjuvant chemotherapy (two cycles of BEP for all) for all patients at a high risk of recurrence has been shown to reduce the rate of relapse to approximately 2% (Cullen et al., 1996). Although this practice is highly effective and is standard management in many centres in the UK, adjuvant chemotherapy has a distinct morbidity and even a mortality risk. r Surveillance (with 3 × BEP for those who relapse). Surveillance is a reasonable practice, provided the patients are monitored carefully, particularly in the first 6 to 12 months. Some patients are not psychologically suited to this approach and it must be explained carefully. For those who relapse, the majority would be in the good-prognosis metastatic group and would therefore receive 3× BEP. Surveillance is a more attractive option than it was a few years ago now that salvage therapy is usually
Testis
limited to 3× BEP rather than 4× BEP. The TE22 study attempted to address the use of FDG-PET in these high-risk patients. The trial was closed by the Independent Data Monitoring Committee after 111 of a planned 135 patients were recruited because 33 of 87 (37%) PET-negative patients had relapsed within 1 year. Of all patients registered for the study, 21% were PET positive at screening (Huddart et al., 2006). Therefore, PET scanning is not useful for predicting which patients will relapse on surveillance.
Patients with borderline nodal enlargement Sometimes, a CT scan shows para-aortic nodes that are larger than normally expected but not quite large enough to meet formal radiological size criteria for enlargement. In this situation, it can be difficult to determine whether or not the lymph nodes are affected by cancer. Nodes are more likely to be pathological if they are in the expected location for the side of tumour and if the patient is in the high-risk group. Experience to date has not shown any benefit from doing a PET scan, but early repeat CT scanning after approximately 3 months, combined with regular marker estimations, usually provide enough information to decide whether the nodes are pathologically involved.
Treatment of metastatic disease: BEP chemotherapy Most patients with metastases should be cured, but the outcome depends mostly on the patient’s prognostic group. Patients with good-prognosis disease (including those with stage IM disease and most of those relapsing after surveillance) are adequately treated by three cycles of BEP (bleomycin, etoposide, cisplatin), whereas four cycles would be recommended for all other patients. The BEP regimen has been extensively tested over the past 20 years and is currently the best regimen for testicular cancer. Attempts to reduce the toxicity of BEP have been unsuccessful. Bleomycin causes potentially fatal pneumonitis, but its omission reduces cure rates by several percent. Substituting carboplatin (‘BEC’) for cisplatin (‘BEP’) reduces the renal toxicity and makes the regimen more acceptable but is also less effective. There have also been several attempts to increase the effectiveness of BEP, particularly in patients with poorprognosis disease. Several trials of intensive schedules (e.g. BOP/VIP, VIP, high-dose BEP) have simply resulted in more toxicity and similar overall survival
rates. High-dose chemotherapy with stem cell support is widely practised yet should probably be considered experimental and be confined to clinical trials (see section on high-dose chemotherapy [HDCT]). For patients with metastatic seminoma, there is less evidence that bleomycin is essential, and it is often left out of the regimen. The standard regimen is either three or four cycles of EP but the three-cycle regimen has not been tested in randomised trials. It has been shown that single-agent carboplatin (four cycles) is less effective than four cycles of EP (Horwich et al., 2000), but for elderly, unfit patients, or those in renal failure, carboplatin alone can deliver acceptable cure rates with few treatment-related deaths. Confusion about BEP may occur because there are so many different ‘recipes.’ For example, the 3- and 5day BEP schedules have been shown to give the same cure rates in good-prognosis disease (but have not been tested in the other groups), but at the price of slightly increased toxicity (gastrointestinal and tinnitus) for the 3-day version. Many centres in the UK use the 3-day BEP schedule to save on bed occupancy, but 5-day BEP can be safely given with patients attending the hospital during the day only, providing they have adequate hydration and there is the option to stay in the hospital if required. There is also a ‘low-dose’ BEP (etoposide total dose 360 mg/m2 per cycle). Table 20.6 shows two versions of BEP chemotherapy.
Practical management of BEP Cisplatin Renal damage is rare if adequate hydration is given with magnesium replacement. Most patients have good renal function, but GFR should be measured by EDTA clearance before starting chemotherapy. For a patient with good renal function, a GFR estimated before each cycle using the Cockcroft–Gault formula is adequate. Severe vascular toxicity (myocardial infarction and stroke) has been reported in young men receiving BEP and has been attributed to the cisplatin component; the mechanism is unknown but may be reversible vascular spasm. A switch to carboplatin should be considered if this occurs.
Bleomycin lung Bleomycin may cause a potentially fatal pneumonitis, which is partially dose related and is rarely seen at cumulative doses of less than 270 000 IU. The risk increases sharply at doses above 360 000 IU. The risks of giving bleomycin outweigh the benefits in patients who are at high risk of bleomycin toxicity such as those aged over 40 years, those with poor renal 247
Jim Barber and John Staffurth
Table 20.6. Two versions of BEP chemotherapy Regimen
Description
Standard 5-day BEP
Weekly bleomycin 30 000 IU, i.v. (D2, 8, 15) should be given regardless of blood count Total dose of bleomycin 270 000 to 360 000 IU Days 1–5 cisplatin 20 mg/m2 Days 1–5 etoposide 100 mg/m2
Standard 3-day BEP
Weekly bleomycin 30 000 IU, i.v. (D2, 8, 15) should be given regardless of blood count Total dose of bleomycin 270 000 to 360 000 IU Days 1 and 2 cisplatin 50 mg/m2 Days 1–3 etoposide 165 mg/m2
function (GFR < 80 ml/minute; O’Sullivan et al., 2003) and smokers. A clinical respiratory assessment should be made before each treatment, especially before cycles three and four, when bleomycin toxicity is more common. A history of gradually increasing shortness of breath and a dry cough are worrying but non-specific symptoms. Monitoring lung function tests during chemotherapy is sometimes recommended, but test interpretation is difficult because lung function deteriorates in most patients during chemotherapy. Thromboembolic disease is a common complication in hospitalised patients with advanced disease. For patients who present with significant dyspnoea (and no evidence of infection), the possibility of a pulmonary embolism or bleomycin lung should be considered. A CT scan of the thorax can help to diagnose bleomycin lung by identifying the diffuse patchy fibrosis/exudates typical of pneumonitis, combined with a VQ scan or pulmonary angiogram to exclude thromboembolic disease.
Residual mass postchemotherapy: teratoma When a teratoma metastasises, it may differentiate into mature teratoma (particularly if TD is present in the orchidectomy specimen). Following chemotherapy, the malignant elements of the tumour may be completely cleared but residual mature teratoma may remain, because it is not sensitive to chemotherapy. Therefore, it is common practice to continue to observe a residual mass. Mature teratoma has the potential to grow slowly over time and may undergo subsequent malignant change, and it is likely that malignant change
248
within mature teratoma is a major cause of late relapse following chemotherapy. Residual para-aortic disease may be resected by retroperitoneal lymph node dissection (RPLND; Hendry et al., 1993). RPLND is a major operation, with a significant risk of infection and pulmonary embolus, and recovery may take several weeks. The operation is highly specialised and may require vascular expertise to deal with residual disease tightly adherent to major blood vessels. Potency is usually normal afterwards but patients may experience retrograde ejaculation and lose fertility as a result. When selecting patients for RPLND, it is useful to know that small residual lymph nodes (< 1 cm) may resolve completely over a period of several months and usually represent necrotic residues. However, any persisting masses much larger than this, or indeed persisting nodes of any size, are likely to contain mature teratoma. FDG-PET may be more accurate than CT in identifying residual malignancy but is not always reliable (De Wit et al., 2006). Surgical practice varies widely. Some surgeons recommend that all patients should have RPLND (Fossa et al., 1989) after chemotherapy whether or not there are radiological abnormalities, but, with this approach, many patients will have unnecessary operations. Other surgeons maintain that small lesions (up to 3 cm) can be safely monitored, but this monitoring would have to be life-long. But there is consensus in the UK that residual mature teratoma should be resected if possible, and improving cure rates have been partially attributed to an increased use of postchemotherapy RPLND.
Testis
A small proportion (around 10%) of resected specimens contain residual cancer (i.e. frankly malignant cells), although the pathological identification of malignant elements can be difficult. Residual viable malignancy is associated with a high incidence of relapse, and further chemotherapy (possibly second-line) should be considered.
Residual mass postchemotherapy: seminoma Residual masses left after chemotherapy for seminoma are not normally resected; they are often extremely vascular, difficult to remove and rarely contain active cancer. Assuming the diagnosis of pure seminoma was correct, mature teratoma should not complicate the issue. PET scanning may be helpful in identifying residual cancer but, as with teratoma, is not reliable (De Wit et al., 2006). Patients with residual masses following chemotherapy are usually managed with observation, but those at high risk of relapse (e.g. those receiving second-line chemotherapy) might be considered for radiotherapy or high-dose chemotherapy with stem cell support.
Fertility and hormonal issues Patients should be offered the option of sperm banking before combination chemotherapy. But for those patients who receive a single cycle of carboplatin or para-aortic radiotherapy, sperm banking may not be necessary because the effect of these treatments on long-term fertility is probably small but not precisely known. Approximately 50% of men requiring chemotherapy for metastatic disease will have low sperm counts before chemotherapy. Of those with normal counts, about 50% will have regained normal counts 2 years after chemotherapy and 80% at 5 years (Lampe et al., 1997). There is a high incidence of hypogonadism among testicular cancer patients because of the association of testicular cancer with testicular atrophy and subfertility. A hormonal assessment should be made part of followup and replacement therapy should be considered.
Relapsed disease Relapse is most common in patients presenting with extragonadal primary tumour sites and in those who do not achieve complete remission, either biochemically or radiologically. Unfortunately, these patients have the poorest results with salvage second-line chemotherapy.
On the other hand, a patient with a testicular primary tumour who has had a complete response to chemotherapy and relapses neither too early (> 6 months) nor too late (< 2 years) has around a 50% chance of durable remission or cure with salvage chemotherapy. Second-line chemotherapy usually involves cisplatin with one or two new drugs not previously used. If the patient has previously received bleomycin, it should not form part of the relapse regimen because of the toxicity risk. For patients who received BEP the first time around, suitable regimens would be VIP (VP16 [etoposide], ifosfamide, cisplatin) or TIP (paclitaxel, ifosfamide, cisplatin). High-dose chemotherapy with peripheral stem cell support could be considered an option (see below) and residual masses, if operable, should be resected promptly. Patients who relapse late (arbitrarily, at more than 2 years) form a distinct subgroup. Histology may show an adult solid tumour (probably arising from residual mature teratoma) rather than a germ cell tumour, and these may be marker negative. Such recurrences do not respond well to standard chemotherapy regimens. Surgery should be considered initially, particularly if there is a potentially operable tumour at a single site. For patients with metastatic disease, however, there is normally little choice but to try second-line chemotherapy and consider surgery afterward.
High-dose chemotherapy with peripheral stem cell support HDCT has been used to treat relapsed patients and it is now being tested as a first-line treatment in poorprognosis patients. Escalation of the cisplatin dose is limited by renal toxicity, but carboplatin can be given at an AUC of up to 20 with stem cell support. There is a relatively low risk of treatment-related death in appropriately experienced centres, but whether this approach improves the outcomes is not yet known. For relapsed patients, HDCT appears to confer around a 10% absolute improvement in 5-year diseasefree and overall survival compared to matched historical controls. However, a multicentre European randomised trial using a single cycle of HDCT (CarboPEC) after three cycles of induction chemotherapy with VIP gave identical results as giving four cycles of VIP. Therefore, this approach is not justified outside a clinical trial. Established factors associated with a poor outcome with HDCT include absolute refractory disease, progressive
249
Jim Barber and John Staffurth
disease prior to HDCT, a mediastinal primary site and βhCG greater than 1000 IU/l. For first-line treatment of the high-risk patient, there are three randomised trials currently under way to test the effectiveness of this approach in Europe and the USA but all are recruiting slowly. The lack of enthusiasm for transplantation has stimulated the development of the NCRI TE23 study, in which a more conventional but intensive multidrug combination (C-BOP/BEP) will be compared to the use of BEP alone.
Palliative treatments Patients who are not thought to be curable with secondline chemotherapy, or who relapse after second-line chemotherapy, may be suitable for palliative therapy in addition to best supportive care. Germ cell cancers occasionally are slow growing, and are chronically recurring. Multiple remissions can be seen with repeated chemotherapy treatment. It is worthwhile considering surgical debulking of disease and even palliative radiotherapy in selected cases and long remissions may be seen.
Follow-up Late recurrences are rare in patients with testicular germ cell tumours and follow-up to detect recurrence may not be needed after 5 years, except in those presenting with metastatic NSGCTs (Shahidi et al., 2002).
Special clinical situations: spermatocytic seminoma Spermatocytic seminoma is said to account for fewer than 5% of seminomas and typically occurs in older men. It bears some histological resemblance to, and can be confused with, seminoma. Although it originates from germ cells, there is a dispute as to whether it should be classified as a germ cell tumour, because it is not descended from germ cell carcinoma in situ. It has no ovarian counterpart and is not associated with the usual risk factors for testicular cancer. Anaplastic transformation is reported, which can be confused with embryonal carcinoma, but does not appear to change the prognosis. Spermatocytic seminoma is usually benign and slow growing, but can occasionally grow very large. Metastatic disease is virtually unknown, but there are
250
occasional reports of sarcomatous transformation associated with the development of metastatic disease.
Sex cord stromal tumours Sex cord stromal tumours are similar to their ovarian counterparts. The Sertoli cells provide support for spermatogenesis, and the Leydig cells secrete testosterone. They are usually classified as benign, but there are a few case reports of metastases occurring. Unfortunately, there are no well-defined pathological criteria that can predict aggressive behaviour. Most oncologists in the UK take a conservative approach to management of these tumours. An orchidectomy and a staging CT scan should be performed. In the unlikely event of para-aortic nodal disease being found, surgery could be considered, with a node dissection if technically possible, and a cure attempted. Close follow-up can cause much unnecessary anxiety in patients who essentially have benign tumours of little significance. It would be reasonable to monitor testosterone and gonadotrophin levels in these patients, although, because these levels have virtually never been estimated preoperatively, it is not generally known whether the tumours secrete hormones or not. For the very occasional patient with inoperable or metastatic disease, cytotoxic chemotherapy could be considered. However, there is no established regimen with documented activity, and it should be recognised that BEP chemotherapy is unlikely to give significant responses.
Ongoing clinical trials In set-up TE21 is a phase II/III randomised trial of BEP versus paclitaxel-BEP in intermediate-risk metastatic germ cell tumours (EORTC 30983). TE23 is a randomised phase II trial of intensive induction chemotherapy (C-BOP/BEP) and standard BEP chemotherapy in poor-prognosis male germ cell tumours.
Open GE301 (ISRCTN11143611) is an international study for the salvage treatment of germ cell tumours that
Testis
compares IPE (ifosfamide, cisplatin and etoposide; four cycles) with IPE (three cycles) plus carboPEC (carboplatin, etoposide and cyclophosphamide; one cycle).
Closed, awaiting results TE08 is a randomised trial of two CT scans versus five CT scans in the surveillance of patients with stage I teratoma of the testis (preliminary results: Mead et al., 2006). TE22 is a study of 18-FDG PET in the prediction of relapse in patients with a clinical stage I nonseminomatous germ cell tumour (preliminary results: Huddart et al., 2006). TIP is a phase II study of paclitaxel, cisplatin and ifosfamide used as induction therapy in the treatment of patients relapsing after BEP (bleomycin, etoposide, cisplatin) chemotherapy for patients with metastatic germ cell tumours.
REFERENCES Bokemeyer, C. and Schmoll, H. J. (1995). Treatment of testicular cancer and the development of secondary malignancies. J. Clin. Oncol., 13, 283–92. Cullen, M. H., Stenning, S. P., Parkinson, M. C. et al. (1996). Short-course adjuvant chemotherapy in high-risk stage I nonseminomatous germ cell tumors of the testis: a Medical Research Council report. J. Clin. Oncol., 14, 1106–13. De Wit, M., Hartmann, M., Brenner, W. et al. (2006). [18F]-FDG-PET in germ cell tumors following chemotherapy: results of the German multicenter trial. J. Clin. Oncol., 24, (18 Suppl.), 4521. Fossa, S. D., Ous, S., Lien, H. H. et al. (1989). Post-chemotherapy lymph node histology in radiologically normal patients with metastatic nonseminomatous testicular cancer. J. Urol., 141, 557–9. Fossa, S. D., Horwich, A., Russell, J. M. et al. (1999). Optimal planning target volume for stage I testicular seminoma: A Medical Research Council randomized trial. J. Clin. Oncol., 17, 1146. Harland, S. J., Cook, P. A., Fossa, S. D. et al. (1998). Intratubular
germ cell neoplasia of the contralateral testis in testicular cancer: defining a high risk group. J. Urol., 160, 1353–7. Hendry, W. F., A’Hern, R. P., Hetherington, J. W. et al. (1993). Para-aortic lymphadenectomy after chemotherapy for metastatic non-seminomatous germ cell tumours: prognostic value and therapeutic benefit. Br. J. Urol., 71, 208–13. Horwich, A., Oliver, R. T., Wilkinson, P. M. et al. (2000). A medical research council randomized trial of single agent carboplatin versus etoposide and cisplatin for advanced metastatic seminoma. Br. J. Cancer, 83, 1623–9. Huddart, R., O’Doherty, M., Padhani, A. et al. (2006). A prospective study of 18 FDG PET in the prediction of relapse in patients with high risk clinical stage I (CS1) non-seminomatous germ cell cancer (NSGCT): MRC study TE22. J. Clin. Oncol., 24, (18 Suppl.), 4520. IGCCCG. (1997). International Germ Cell Consensus Classification: a prognostic factor-based staging system for metastatic germ cell cancers. J. Clin. Oncol., 15, 594–603. Lampe, H., Horwich, A., Norman, A. et al. (1997). Fertility after chemotherapy for testicular germ cell cancers. J. Clin. Oncol., 15, 239–45. Mead, G. M., Rustin, G. J., Stenning, S. P. et al. (2006). Medical Research Council trial of 2 versus 5 CT scans in the surveillance of patients with stage I non-seminomatous germ cell tumours of the testis. J. Clin. Oncol., 24 (18 Suppl.), 4519. Oliver, R. T., Mason, M., Mead, G. M. et al. (2005). Radiotherapy versus single-dose carboplatin in adjuvant treatment of stage I seminoma: a randomised trial. Lancet, 366, 293–300. O’Sullivan, J. M., Huddart, R. A., Norman, A. R. et al. (2003). Predicting the risk of bleomycin lung toxicity in patients with germ-cell tumours. Ann. Oncol., 14, 91–6. Patterson, H., Norman, A. R., Mitra, S. S. et al. (2001). Combination carboplatin and radiotherapy in the management of stage II testicular seminoma: comparison with radiotherapy treatment alone. Radiother. Oncol., 59, 5–11. Shahidi, M., Norman, A. R., Dearnaley, D. P. et al. (2002). Late recurrences in 1263 men with testicular germ cell tumors. Multivariate analysis of risk factors and implications for management. Cancer, 95, 520–30. von der Maase, H. (2001). Do we have a new standard of treatment for patients with seminoma stage IIA and stage IIB? Radiother. Oncol., 59, 1–3. Williams, S. D., Birch, R., Einhorn, L. H. et al. (1987). Treatment of disseminated germ-cell tumors with cisplatin, bleomycin, and either vinblastine or etoposide. N. Engl. J. Med., 316,1435–40.
251
21
PENIS Jim Barber
Range of tumours The range of tumours affecting the penis is shown in Table 21.1.
Incidence and epidemiology Penile cancer is relatively uncommon with around 350 cases annually in the UK, although in Asia, Africa and South America (e.g. Uganda, Brazil) the disease is more common. The peak incidence is in men over age 70 but a substantial number of cases occur in much younger men.
Risk factors and aetiology Penile cancer appears to be less consistently associated with HPV than is cervical carcinoma. Most studies have only identified HPV subtypes 16 and 18 in around 50% of cases of classic squamous cell tumours, although there is a more consistent association with carcinoma in situ and the verrucous variant of squamous cell carcinoma (Dillner et al., 2000). Circumcision at birth reduces the risk considerably, and this leads to a theory that carcinoma of the penis is related to poor hygiene, possibly to the accumulation of smegma. Phimosis is thought to increase the risk, and also makes early detection much more difficult; tumours are more likely to present late.
Pathology A number of premalignant lesions are recognised. Balanitis xerotica obliterans is the penile equivalent of lichen planus and is associated with a small risk of invasive cancer. Intraepithelial carcinoma presents in a variety of shapes, from nodules to erythematous plaques. It is designated erythroplasia of Queyrat if it appears on the glans or prepuce, or Bowen’s disease if it involves the skin. The risk of transformation to invasive squamous cell carcinoma (SCC) is around 30% for carcinoma in situ on the glans. 252
A range of invasive squamous cancers occur, rather similar to those seen in the anal or vulval region: classic, well to poorly differentiated squamous cell cancers, a basaloid variant and a number of verrucous tumours (e.g. giant condylomata, Bushke Lowenstein tumour). The classic squamous cell tumours behave in a fashion similar to that of squamous tumours elsewhere. The risk of local recurrence and metastatic spread is highly dependent on the stage and grade of the tumour, and most deaths from penile cancer result from the more aggressive G3 lesions. Penile cancer usually metastasises quite systematically through the local lymphatics, first to the inguinal nodes (bilateral disease is quite frequent) and then to the pelvic nodes. Distant metastases are relatively uncommon without previous inguinal metastases. Verrucous tumours may exhibit features of viral change (koilocytosis). Their pattern of invasion is to advance on a broad front, in a destructive rather than invasive fashion, and they are thought to metastasise only rarely. There are case reports of anaplastic transformation, which may be spontaneous but sometimes appears to be related to radiotherapy treatment. The current literature (though sparse) would suggest that this is a relatively rare event, probably occurring in fewer than 5% of cases. Occasionally, the penis is involved with Kaposi’s sarcoma, basal cell carcinomas or melanoma. It can also be involved by metastases from other sites. Both bladder and prostate cancer can invade the penis locally but the diagnosis rarely causes any difficulty.
Diagnosis and staging Review by a specialist MDT is recommended at the outset. For small tumours an attempt at local excision is a reasonable first step. Larger tumours should, if possible, be biopsied and fully staged before starting the patient on definitive local therapy. Traditional staging of penile cancer simply consists of a clinical examination of the primary and of inguinal
Penis
Table 21.1. Range of tumours affecting the penis
Table 21.2. The TNM T and N stages for carcinoma of the penis
Type of tumour
Examples
Benign
Condylomata acuminata Bowenoid papulosis Leiomyoma Balanitis xerotica obliterans (premalignant) Carcinoma in situ (erythroplasia of Queyrat, Bowen’s disease)
Malignant primary
Squamous carcinoma Verrucous carcinoma (very low risk of metastasis) Others Basal cell carcinoma Kaposi’s sarcoma Melanoma
Stage
Description
Tis
Carcinoma in situ
Ta
Tumour is non-invasive carcinoma
T1
Tumour invades subepithelial connective
T2
Tumour invades corpus spongiosum or
T3
Tumour invades the urethra or prostate gland
T4
Tumour invades other adjacent structures
N0
No nodal metastases
N1
Single unilateral nodal metastases
N2
Multiple superficial nodal metastases
N3
Multiple deep nodal metastases
tissue corpora cavernosa
Adapted from UICC (2002).
Sarcoma Malignant
e.g. from bladder, prostate
secondary
which is leading to changes in the way penile cancer is managed.
lymph nodes. However, MRI scanning provides supplementary information for both the primary because it can detect invasion of the corpora cavernosa and the nodal regions. In particular, if patients have inguinal nodal enlargement, spread to nodal groups in the pelvis or elsewhere may be detected. Although distant metastases are rare, it would be reasonable to complete staging with CT scanning of the chest and abdomen if there is clinical or radiological evidence of nodal disease.
Staging The TNM staging classification for penile carcinoma is shown in Table 21.2.
Treatment of carcinoma in situ For isolated, well-defined tumours, local excision seems a sensible choice; it results in high rates of local control and allows pathological analysis of the whole lesion so that invasive cancer can be excluded. Sometimes, carcinoma in situ may be patchy and poorly defined, and excision might involve substantial mutilation with a need for reconstruction. In these circumstances, high rates of local control have been achieved with radiotherapy (McLean et al., 1993), and there are several alternatives, including laser therapy, PDT, 5-FU cream and cryotherapy. Recurrence is common, and progression to invasive cancer is very possible. Careful follow-up in a multidisciplinary clinic is advised.
Treatment overview
Invasive squamous cell carcinoma: treatment of the primary
The treatment of patients with penile cancer is multimodal and may involve a urologist, an andrologist, a plastic surgeon, an oncologist, a specialist dermatology pathologist and a specialist nurse. NICE guidance recommends accumulating this expertise in just a few highly specialised supraregional MDTs around the UK,
Surgical options must be individualised for each patient depending on the location, grade and size of the tumour relative to the size of the penis. For smaller, low-risk tumours of low grade (up to T1, G2), a penis-preserving strategy should be considered. Options include local excision with reconstruction, 253
Jim Barber
external beam radiotherapy, brachytherapy or laser excision. For all high-grade (G3) tumours and larger ones (T2 or greater), local recurrence is more frequent. Clear margins of around 10 mm are needed for G3 tumours, and partial amputation is considered standard therapy. Techniques of penile conservation, radiotherapy and surgical reconstruction should be discussed.
Role of radiotherapy External beam radiotherapy External bean radiotherapy gives somewhat lower local control rates than penectomy (around 60%; Azrif et al., 2006, Sarin et al., 1997), but local failures can be dealt with via salvage surgery. Although the distress of penile amputation can be avoided, the morbidity of treatment is substantial and it should not be seen as an easy option for the elderly or unfit patient, for whom surgery may be quicker and easier. A painful acute urethral reaction combined with moist desquamation may occur during treatment. In the longer term, there may be painful fibrosis, telangiectasia and urethral stricture, which may require intermittent dilatation. Despite all this, there is a proportion of patients who have excellent cosmetic results and maintain full potency after radical radiotherapy, although how high this proportion is has not been precisely reported in the literature.
Radiotherapy techniques For small tumours, it is theoretically possible to use kilovoltage therapy or electron therapy with a lead cut out. However, tumours small enough for this technique are rarely seen, and they may well be more conveniently dealt with by local surgical excision. It is helpful for the patient to undergo circumcision before the treatment, because a phimosis may occur after radiotherapy (which would make visual follow-up impossible). For treatment of the whole penile shaft, a wax immobilisation block should be used: this holds the penis in a vertical position and ensures a high dose out to the surface of the skin or glans. The wax block is in two halves which are clipped together for treatment. Some centres keep a number of standard sized blocks, whereas others make customised blocks for each patient. It is important to achieve the full dose of radiotherapy at the skin surface; depending on the fit of the wax mould, it may be necessary to insert moist gauze in between to avoid any air gaps. There is a variety of techniques to ensure that the penis does not slip down within the 254
block; it may be ‘stepped’ at the level of the glans, or r Treatment is with 4 to 6 held up with gauze Tubigrip. MV photons. The tolerance dose of the penis is around 55 Gy in 20 fractions over 4 weeks or 64 Gy in 32 fractions over 6.5 weeks.
Interstitial implantation brachytherapy Slightly higher rates of local control have been reported for interstitial implantation compared to external beam radiotherapy (Kiltie et al., 2000). However, most reported series of external beam radiotherapy have included larger tumours unsuitable for implantation and so it is not clearly established that implantation is more effective. Brachytherapy does not give any significant advantage over external radiotherapy in reducing the treatment volume, because the treatment volume (the whole penile shaft) is about the same. Low-dose-rate brachytherapy over several days may reduce the risk of late tissue damage but this theoretical advantage is lost if high-dose-rate afterloaded brachytherapy is used. Implantation is usually performed with a two-plane iridium implant and a template to either side to ensure that the sources are kept parallel. It is essential to perform circumcision first, and the patient is catheterised throughout the treatment. Around 65 Gy to the 85% reference isodose is given over 6 to 7 days.
Management of the nodes Patients with minimally involved groin nodes (micrometastatic disease) have a high cure rate of around 70 to 80%, whereas there are few survivors when extranodal extension or pelvic nodal disease is involved. Survival is also poor when salvage surgery is performed for recurrent nodal disease (Kiltie et al., 2000), which suggests that it is worthwhile to attempt to resect nodal metastases at the earliest opportunity. Unfortunately, the morbidity of this procedure is considerable: there is a high risk of lymphoedema, infection, seroma and thromboembolic complications.
Clinically and radiologically negative groin High-risk patients should be considered for bilateral elective lymph node dissection. Unfortunately, the exact proportion of patients with a clinically ‘negative’ groin who go on to develop pathological nodal disease is rather uncertain. European Association of Urology guidance recommends elective nodal dissection for patients of high grade, or stage T2 of any grade, but
Penis
this advice is based on the assumption that the risk of nodal involvement is around 70%. In a recent series from the UK, the risk for similarly stratified patients was only around 20% (Hegarty et al., 2006). Clearly, at this level of risk, many patients would undergo unnecessary nodal dissections and so this is a controversial area. For low-risk patients, observation is a reasonable strategy. Regular clinical examination (initially at least monthly or bimonthly) could be supplemented with sequential pelvic MRI studies to increase the sensitivity of clinical nodal examination. The investigation of nodal disease may be improved by the relatively new technique of dynamic sentinel node biopsy. Isosulphan blue and radiolabelled colloid are injected into the tumour, and the sentinel node is located with a gamma probe and by visual identification of the dye. The node is sent for frozen-section examination and, if the section is positive, a full nodal dissection is performed. If it is negative the patients are just kept under observation (Tanis et al., 2002). The sensitivity for detecting nodal disease is around 80%. The initial experience from using this approach appears promising, but it is still considered to be experimental. As in other pelvic cancers, MRI scanning can be refined by the use of ultrasmall superparamagnetic iron oxide particles (e.g. ferumoxtran), which are taken up by normal lymph nodes but not those affected by tumour. Initial results in patients with penile cancer appear promising but the technique is not currently available in the UK.
Patients with clinically involved groin nodes Not all enlarged inguinal nodes contain cancer, because reactive nodes are common, particularly if there is infection. It is often recommended that enlarged inguinal nodes be re-evaluated after 4 to 6 weeks of antibiotic therapy, but a decision about node management can normally be made without waiting so long. Some patients with high-risk tumours should be considered for elective nodal dissection in any case, and a combination of clinical examination, MRI scanning and FNA will lead to a positive diagnosis of nodal disease in most of the other patients. Bilateral radical inguinal node dissections should be done. A more conservative operation (the modified Catalona dissection) rather than the standard radical inguinal node dissection is considered adequate for elective dissections (i.e. of clinically uninvolved nodes),
without the need for sartorius muscle transfer to cover the vessels. Some surgeons advocate a further extended pelvic nodal clearance if there are multiple inguinal lymph nodes involved but this is controversial. Patients who present with enlarged nodes during follow-up normally have recurrent disease and reactive nodal enlargement is less common. If the recurrence occurs late (after 2 years) and only on one side, it is reasonable to only carry out a unilateral dissection rather than the standard bilateral procedure.
Management of patients following surgery with nodal disease For patients who have had a complete resection of nodal disease, observation alone is appropriate. Those patients who have positive margins or an extensive extracapsular tumour are at high risk of local recurrence. For these patients, there is no standard option; however, it would be usual practice to treat the areas at risk to radiation tolerance (50 to 60 Gy in 2 Gy fractions). There is interest in the use of concurrent chemoradiotherapy using, for example, weekly cisplatin (40 mg/m2 ).
Mangement of the patient with locally advanced or unresectable groin disease Uncontrolled local disease is very unpleasant, involving fungation of local disease, chronic infections, severe lymphoedema, skin breakdown, skin metastases and thromboembolic disease. There is little in the literature about how best to manage these patients. Patients who have unresectable groin disease and those with pelvic nodal metastases should be considered for treatment with chemoradiotherapy if they are fit enough. Experience from cervical and anal cancer suggests that chemoradiation may be better than radiotherapy alone, and some clinicians have used chemoradiation with concurrent cisplatin and/or 5-FU to try to get local control of the cancer. For example, 45 to 50 Gy can be delivered in 25 fractions to the groins and pelvis with parallel fields, with weekly cisplatin (40 mg/m2 ), and a further 20 Gy in 10 fractions delivered to sites of bulky disease, but because penile cancer is an uncommon condition, there are few data on the results of this approach to treatment. If a patient displays a good response to this treatment, surgery may be considered. 255
Jim Barber
Role of chemotherapy Adjuvant chemotherapy Because of the relatively small number of cases, it is very unlikely that adjuvant chemotherapy will ever be adequately tested in a randomised trial. Experience in other cancers suggests that adjuvant chemotherapy may confer a small survival benefit, but it cannot be recommended as routine practice.
Palliative chemotherapy Patient responses have been reported with regimens similar to those effective in other SCCs, for example, a combination of cisplatin and 5-FU. There is no evidence that one regimen is more effective than any other. Palliative chemotherapy in metastatic disease is an option, but, because advanced penile cancer is often predominantly a locoregional problem, patients might benefit more from focussing aggressive local treatment on the pelvis and groins, perhaps with chemoradiotherapy.
Dillner, J., Meijer, C. J., von Krogh, G. et al. (2000). Epidemiology of human papillomavirus infection. Scand. J. Urol. Nephrol. Suppl., 205, 194–200. Hegarty, P., Minhas, S., Freeman, A. et al. (2006). A prospective study of 100 cases of penile cancer managed according to European Association of Urology guidelines. B. J. U. Int., 98, 526–31. Kiltie, A. E., Elwell, C., Close, H. J. et al. (2000). Iridium-192 implantation for node-negative carcinoma of the penis: the Cookridge Hospital experience. Clin. Oncol. (R. Coll. Radiol.), 12, 25–31. McLean, M., Akl, A. M., Warde, P. et al. (1993). The results of primary radiation therapy in the management of squamous cell carcinoma of the penis. Int. J. Radiat. Oncol. Biol. Phys., 25, 623–8. Sarin, R., Norman, A. R., Steel, G. G. et al. (1997). Treatment results and prognostic factors in 101 men treated for squamous carcinoma of the penis. Int. J. Radiat. Oncol. Biol. Phys., 38, 713–22. Tanis, P. J., Lont, A. P., Meinhardt, W. et al. (2002). Dynamic sentinel node biopsy for penile cancer: reliability of a staging technique. J. Urol., 168, 76–80. UICC. (2002). TNM Classification of Malignant Tumours. Ed. L. H. Sobin and Ch. Wittekind. New York: Wiley-Liss, pp. 181–3.
REFERENCES
FURTHER READING
Azrif, M., Logue, J. P., Swindell, R. et al. (2006). External-beam radiotherapy in T1–2 N0 penile carcinoma. Clin. Oncol. (R. Coll. Radiol.), 18, 320–5.
European Association of Urology Guidelines on management of penile cancer, www.uroweb.org/files/uploaded files/guidelines/ 22891 Penile Cancer.pdf
256
22
OVARY Louise Hanna and Malcolm Adams
Introduction Ovarian cancer is the most common cause of death from gynaecological malignancy in the Western world. It has been named a ‘silent killer’ because of its lack of symptoms during early stages. Around 90% of ovarian cancers arise from the epithelium. Two-thirds of patients present late, with stage III or IV disease, with increasing abdominal symptoms including ascites. Typically, treatment depends on a combination of surgery and chemotherapy. Improvements in surgical techniques and chemotherapy agents have resulted in a modest increase in the 5-year survival over the past 30 years although, even now, two-thirds of women die from their disease.
Types of tumour affecting the ovary The WHO classification of tumours of the ovary defines broad categories of ovarian tumours (WHO classification, 2003): r Surface epithelial-stromal tumours. r Sex cord-stromal tumours. r Germ cell tumours. r Tumours of the rete ovarii. r Miscellaneous tumours. r Lymphomas and haematopoietic tumours. r Secondary tumours. Surface epithelial-stromal tumours are classified as benign, borderline or malignant. The subtypes are serous; mucinous; endometrioid, including malignant ¨ mixed mullerian tumour (carcinosarcoma); clear cell; transitional cell; squamous cell; mixed; and undifferentiated or unclassified. Sex cord-stromal tumours are classified as granulosastromal cell tumours (including granulosa cell tumours and theca-fibroma tumours), sertoli-stromal cell tumours, sex-cord stromal tumours of mixed or unclassified cell types, and steroid cell tumours.
Germ cell tumours are classified as primitive germ cell tumours (including dysgerminoma, yolk sac tumour, and embryonal carcinoma), biphasic or triphasic teratomas (including immature teratoma and mature teratoma), and monodermal teratoma (composed of a single type of tissue and includes struma ovarii, which is composed of thyroid cells).
Incidence and epidemiology The annual incidence of ovarian cancer in the UK is 18.2 in 100 000 women. Approximately 6800 cases are reported per year in the UK (National Statistics, 2005). Disease incidence shows a slightly increasing trend. Epithelial ovarian cancer is predominantly a disease of the Western world. The incidence of epithelial ovarian carcinoma rises steeply after the age of 50, the greatest number of cases being between the ages of 55 and 74. Borderline tumours or hereditary cancers, particularly those associated with the BRCA1 gene, occur at a younger age.
Epithelial-stromal ovarian cancers Risk factors and aetiology Aetiology of ovarian carcinoma Epithelial ovarian carcinoma is thought to arise from the surface epithelium of the ovary or from entrapped epithelial cells in inclusion cysts. There are several current hypotheses that attempt to explain the aetiology of ovarian cancer, such as the incessant ovulation hypothesis, which suggests that cancers arise through repeated trauma to epithelial cells during ovulation and, therefore, factors that suppress ovulation will be protective. However, none of the hypotheses completely explain all the data from epidemiology, and there is a need for a greater understanding of the pathogenesis of ovarian cancer to develop new strategies for prevention (Modugno, 2003).
257
Louise Hanna and Malcolm Adams
Table 22.1. Pathological features of epithelial-stromal tumours Type
Macroscopic features
Microscopic features
Benign
Commonly smooth-walled and cystic
Regular architecture with uniform cells, rare mitoses
Borderline
Frequently cystic or solid, but without
Some features of malignancy such as irregular architecture,
haemorrhage or necrosis
multilayering of cells and mitotic activity, but basement membrane remains intact
Malignant
Solid and/or cystic with areas of
solid areas; cells range from well differentiated to poorly
with tumour extending into the
differentiated, with nuclear and cytoplasmic atypia and
peritoneal cavity
frequent mitoses
Risk factors and protective factors for ovarian cancer Family history
r Having a first degree relative with breast or ovarian cancer doubles the risk.
r Having a relative with cancer of the stomach, intestine r
r
r
r
or lung or lymphoma also increases the risk of ovarian cancer. Germline mutations in either the BRCA1 or BRCA2 genes are inherited in an autosomal dominant fashion, and they significantly increase the risk of breast and ovarian cancer in affected individuals. Carriers of a germline mutation in BRCA1 have an overall risk for breast cancer of 56 to 68% and for ovarian cancer of 16 to 39% (Antoniou et al., 2003; Risch et al., 2001; Struewing et al., 1997). Carriers of a germline mutation in BRCA2 have an overall risk for breast cancer of 45 to 54% and for ovarian cancer of 11 to 16% (Antoniou et al., 2003; Struewing et al., 1997). Women with hereditary non-polyposis colorectal cancer syndrome have a 40 to 60% lifetime risk for colon cancer, a 40 to 60% lifetime risk for endometrial cancer and a 12% lifetime risk for ovarian cancer (Lu et al., 2005).
Other risk factors and protective factors There is increasing evidence from case-controlled and cohort studies that several factors affect the risk of ovarian cancer. These factors have been reviewed (Hanna and Adams, 2006) and are summarised in the following list. These factors are important when considering possible strategies to reduce the risk of ovarian cancer in highrisk women: 258
Variable architecture from infiltrative glands to poorly defined
haemorrhage and necrosis, frequently
r Protective factors – pregnancy, breast feeding, use of the oral contraceptive pill, tubal ligation and hysterectomy. r Possible protective factors – late menarche, early menopause, exercise, outdoor lifestyle (possibly linked to vitamin D), diet rich in fruit and vegetables (e.g. tomatoes and carrots, which contain carotenoids). r Risk factors – infertility, high socioeconomic status. r Possible risk factors – long term HRT, obesity, occupational exposure (e.g. industrial chemicals, organic dusts, asbestos).
Pathology of epithelial-stromal ovarian tumours Tumours are classified as benign, borderline or malignant (see Table 22.1). Malignant epithelial-stromal tumours account for 90% of ovarian cancers. They arise from the surface epithelium or its derivatives such as inclusion cysts and range in size from sub-centimetre size to more than 20 cm in diameter. Cells resemble distinctive types of epithelium, for example, fallopian tube (serous), endocervix or intestine (mucinous), endometrium (endometrioid), urothelium (transitional cell), clear ¨ cells and squamous epithelium. The mixed mullerian tumour (carcinosarcoma) has malignant epithelial and stromal elements.
Spread The predominant modes of spread of ovarian cancer are local and peritoneal. Local spread is to the fallopian tubes, uterus, vagina, bowel, bladder and pelvic side
Ovary
wall. Peritoneal spread is to any peritoneal surface, particularly omentum, paracolic gutters, bowel mesentery, and the undersurface of the right diaphragm. Spread to other mesothelial cavities includes pleural effusion and pericardial effusion; lymph node spread involves pelvic lymph nodes and para-aortic lymph nodes. Haematogenous spread occurs to the liver (rare), bone and lung (very rare).
Screening and prevention In theory it should be possible to improve the survival of ovarian cancer patients by detecting and treating the disease at an early stage. Early treatment is hindered by a lack of detailed knowledge of the natural history of ovarian cancer and the absence of a known precancerous lesion. Any screening test would require a low number of false-positive tests, because the diagnostic and therapeutic intervention is invasive (i.e. laparotomy). Women at high risk of ovarian cancer, such as those with a strong family history or proven BRCA1 or BRCA2 mutation carriers, may be offered surveillance using transvaginal ultrasound (TVS) and serum CA125 levels, although this approach is unproven. For the general population, data will be available in the next few years from the UKCTOCS study of TVS and CA125, which recently completed recruitment of 200 000 women. Strategies for the prevention of ovarian cancer for high-risk women include prophylactic salpingo-oophorectomy, which dramatically reduces the risk of ovarian cancer. Women undergoing this procedure have a small subsequent risk of papillary serous peritoneal cancer (Meeuwissen et al., 2005).
Clinical presentation Ovarian cancer is known as a ‘silent killer’ because symptoms may be absent or vague and non-specific in early stages. The most common symptoms of ovarian cancer arise because of peritoneal spread of the tumour and include abdominal pain, bloating, abdominal swelling (usually due to ascites), nausea, anorexia, vague bowel or bladder symptoms and weight loss. Other symptoms include back pain, shoulder tip pain, peripheral oedema, shortness of breath (pleural or pericardial effusion) or, rarely, symptoms from metastatic disease such as jaundice. Ovarian cancer only rarely causes neurological symp-
toms due to paraneoplastic cerebellar degeneration, neuropathy, or brain metastases. Clinical signs include a pelvic mass, abdominal mass, ascites and pleural effusion.
Investigation and staging Initial tests for patients presenting with only a pelvic mass include TVS, serum CA125 and αFP and βhCG (if the patient is less than 40 years old) to exclude germ cell tumour. For patients with a pelvic mass only, the Relative Malignancy Index (RMI) can predict the chances of malignancy. The RMI is the product of the serum CA125 level, the ultrasound score (characteristic features being multilocular cyst, solid areas, bilateral lesions, ascites and intra-abdominal metastases; no characteristic features = 0, one characteristic feature = 1, more than one characteristic feature = 4), and the patient’s menopausal status (premenopausal = 1, postmenopausal = 4). An RMI greater than 200 has a sensitivity of 74 to 80% and a specificity of 89 to 92% for ovarian cancer, and patients should be referred to a centre that has experience in ovarian cancer surgery (SIGN, 2003). Such patients should undergo further staging investigations (e.g. CT scan of the abdomen and pelvis and CXR). Tests for patients with clinically advanced disease include FBC, renal, liver and bone profiles; serum CA125; CT scan of the abdomen and pelvis; CXR; and cytology from an ascites tap if the patient is symptomatic.
Staging classification The most commonly used staging classification is that of the Federation Internationale de Gynecologie et d’Obstetrique (FIGO; Benedet et al., 2000). It is shown in Table 22.2.
Treatment overview The standard treatment approach in the UK is an initial surgical approach for accurate surgical staging and to attempt optimal debulking. Following surgery, almost all patients receive adjuvant platinum-based chemotherapy. The exceptions are patients who have very-good-prognosis tumours (such as stage IA grade 1 tumours) and who have had complete surgical staging, when only clinical follow-up is justified. 259
Louise Hanna and Malcolm Adams
Table 22.2. Federation Internationale de Gynecologie et d’Obstetrique (FIGO) staging of ovarian cancer Stage
Description
I
Tumour confined to the ovaries
IA
Limited to one ovary, capsule intact, no tumour on ovarian surface, no malignant cells in ascites or peritoneal
IB
Tumour limited to both ovaries, capsule intact, no tumour on ovarian surface, no malignant cells in ascites or
IC
Tumour limited to one or both ovaries with any of the following: capsule ruptured, tumour on ovarian surface,
washings peritoneal washings malignant cells in ascites or peritoneal washings II
Tumour involves one or both ovaries with pelvic extension
IIA
Extension and/or implants in uterus and/or tubes; no malignant cells in ascites or peritoneal washings
IIB
Extension to other pelvic organ: no malignant cells in ascites or peritoneal washings
IIC
Pelvic extension (IIA or IIB) with malignant cells in ascites or peritoneal washings
III
Tumour involves one or both ovaries with microscopically confirmed peritoneal metastasis outside the pelvis
IIIA
Microscopic peritoneal metastasis beyond pelvis
IIIB
Macroscopic peritoneal metastasis beyond pelvis 2 cm or less in greatest dimension
IIIC
Peritoneal metastasis beyond pelvis more than 2 cm in greatest dimension and/or regional lymph node
IV
Distant metastasis
and/or regional lymph node metastasis
metastasis
Adapted from Benedet et al. (2000).
Some patients are not considered well enough for initial surgery, such as those with a significant pleural effusion or hypoproteinaemia. Neoadjuvant chemotherapy followed by interval debulking surgery is an option for these patients. Conservative surgery with unilateral oophorectomy and wedge biopsy of the contralateral ovary is an option for patients wishing to retain fertility and who have good-prognosis tumours (e.g. borderline tumours, germ cell tumours and stage IA epithelial cancers).
Surgery Surgery is used to obtain histology, accurately stage the tumour, and treat the disease by complete removal or debulking; the role of surgery has been reviewed in a recent national guideline (SIGN, 2003). Surgery should be carried out by a specialist gynaecological oncologist surgeon. For patients with advanced disease, a recent meta-analysis shows that maximal cytoreduction is associated with improved overall survival (Bristow et al., 2002). 260
Surgical procedure A midline incision is performed to allow adequate disease assessment. Ascites or peritoneal washings are sent for cytological analysis. A thorough inspection of the peritoneal surfaces is undertaken and followed by total abdominal hysterectomy, bilateral salpingooophorectomy and infra-colic omentectomy. If it is not possible to remove the tumour fully, deposits are debulked as much as possible, including bowel resection if necessary and appropriate. There are differing views over the extent of random peritoneal biopsies that are required to stage the disease fully. There is no proven therapeutic value for lymphadenectomy.
Fertility-sparing options In women who wish to retain their fertility, the affected ovary alone may be removed, leaving the contralateral ovary and uterus intact. A synchronous tumour in the contralateral ovary or uterus should be excluded as far as possible with wedge resection of the ovary and endometrial curettings/hysteroscopy.
Ovary
Table 22.3. Summary of paclitaxel trials Trial
FIGO stage
Chemotherapy
OS (m)
GOG 111
III and IV
TP vs
38
PCyclo
24
TP vs
36
PCyclo
26
TP vs
26.6
T vs
26.0
P
30.2
(McGuire et al., 1996) OV10
IIB–IV
(Piccart et al., 2000) GOG132
III and IV
(Muggia et al., 2000)
ICON 3
80% III and IV
TP vs
36.1
(ICON Group, 2002)
20% I and II
CAP or Carbo
35.4
A = doxorubicin, C = Cyclo = cyclophosphamide, Carbo = carboplatin, OS(m) = overall survival in months, P = cisplatin, T = paclitaxel.
Interval debulking If maximum cytoreduction has been achieved during initial surgery, then there is no benefit from a further laparotomy. However, interval debulking may be appropriate for patients in whom initial debulking was unsuccessful and who have not progressed after three to four cycles of chemotherapy. Such decisions are individualised and should be made within the context of the specialist MDT.
Chemotherapy The role of postoperative chemotherapy has been the subject of randomised trials over the past two decades. The Advanced Ovarian Cancer Trialists Group found a benefit from platinum-based compared to non-platinum regimens, and that carboplatin and cisplatin were equally effective (Anonymous, 1991). There was a trend towards a benefit for platinum-containing combination chemotherapy rather than single-agent platinum; however, the subsequent ICON 2 study showed that single-agent carboplatin was as effective as CAP (cyclophosphamide, doxorubicin and cisplatin; Anonymous, 1998). For stage I disease, the ICON 1 and the ACTION trials (Trimbos et al., 2003) showed a 7% survival benefit in patients who received platinum-based chemotherapy. As a result many UK oncologists would advise single-agent carboplatin for patients with stage I ovarian cancer, with the exception of patients with completely resected stage IA grade 1 tumours, who have an excellent prognosis without adjuvant treatment.
More recently for advanced disease, there have been four pivotal trials studying the use of paclitaxel in addition to platinum-based chemotherapy. Whereas the GOG 111 and OV10 trials showed a highly significant benefit to adding paclitaxel, the GOG 132 and ICON 3 trials did not. There has been much debate over the reasons why these trials have shown inconsistent results. The trials are summarised in Table 22.3. Because of these inconsistent results, the role of paclitaxel has been reviewed in NICE guidance no. 55 (NICE, 2003), which emphasises that women should be involved in making choices about their treatment. It recommends that paclitaxel in combination with platinum-based chemotherapy or platinum-based chemotherapy alone should be offered as alternatives for first-line chemotherapy in the treatment of ovarian cancer. The risks and benefits should be made available and the decision made after discussion. The NICE guidance does not specify any particular stage of disease but emphasises that factors to be taken into account include the side-effect profile of the treatments, the stage of disease, the extent of surgical treatment and the diseaserelated performance status.
Single-agent carboplatin for ovarian cancer
r Pretreatment investigations include information from surgery, imaging, pathology, FBC, U + E, creatinine, LFT and estimate of renal function (e.g. EDTA clearance). r The options should be discussed according to NICE guidance, and written information given about side 261
Louise Hanna and Malcolm Adams
effects and precautions (e.g. monitoring temperature). r A chemotherapy regimen is carboplatin AUC × 5 to 6 i.v.i. over half an hour (AUC [area under curve] = GFR + 25), repeated every 21 days, for a total of six cycles. r Antiemetics include dexamethasone and metoclopramide. r Side effects include nausea (treat with antiemetics), febrile neutropenia (pretreatment education and treat with antibiotics according to local neutropenia/septic shock policy), anaemia (give red cell transfusion) and thrombocytopenia (give platelet transfusion).
Paclitaxel and carboplatin for ovarian cancer
r Pretreatment investigations include information from
r
r r r
surgery, imaging, pathology, FBC, U + E, creatinine, LFT and estimate of renal function (e.g. EDTA clearance). The options should be discussed according to NICE guidance. Written information should be given, including a description of relevant side effects (e.g. monitoring temperature and the management of alopecia). A chemotherapy regimen is paclitaxel (175 mg/m2 over 3 hours) and carboplatin (AUC × 5 to 6 over half an hour), repeated every 21 days for a total of six cycles. Patient premedication includes dexamethasone, chlorpheniramine and ranitidine to reduce the risk of hypersensitivity reactions. Side effects, in addition to those listed for single-agent carboplatin, include alopecia (offer use of scalp cooling, wig, head scarf ), peripheral neuropathy (consider stopping paclitaxel if persistent), and muscle and joint aches (provide simple analgesics).
Recurrent disease The majority of patients with recurrent disease will have already received platinum-based chemotherapy, either alone or in combination with paclitaxel. The ICON 4 study randomised women with platinum-sensitive disease to paclitaxel plus platinum-based chemotherapy or platinum-based chemotherapy alone. Overall, there was a 7% improvement in survival at 2 years in women receiving paclitaxel and platinum, including those women who had already received paclitaxel (Parmar et al., 2003). The NICE Technology Appraisal no. 91 (NICE, 2005) addresses the issues surrounding fur262
Table 22.4. NICE descriptions of cancer types following platinum-based chemotherapy Category
Description
Platinum-sensitive
Relapse more than 12 months
ovarian cancer
after response to platinum-based treatment
Partially platinumsensitive ovarian cancer Platinum-resistant
Relapses 6–12 months after response to platinum-based treatment Relapses less than 6 months
ovarian cancer
after response to
Platinum-refractory
Does not respond to
platinum-based treatment ovarian cancer
platinum-based treatment
Adapted from NICE (2005).
ther chemotherapy and divides disease into categories as shown in Table 22.4. In patients with platinum-sensitive or partially platinum-sensitive ovarian cancer, further platinumbased treatment is recommended (with the option of adding in paclitaxel), except in women who are allergic to platinum-based compounds. Single-agent paclitaxel may be given as a second-line treatment in women with platinum-resistant or refractory disease. Pegylated liposomal doxorubicin hydrochloride (PLDH) is an option for women with partially platinumsensitive, platinum-resistant and platinum-refractory ovarian cancer. Topotecan is only recommended as a second-line treatment for women with platinumresistant and platinum-refractory ovarian cancer for whom paclitaxel and PLDH are considered inappropriate.
PLDH and topotecan For PLDH the response rate is 28% in patients with platinum-sensitive disease, falling to 12.3% in patients with platinum-refractory disease (Muggia and Hamilton, 2001). For topotecan the overall response rates are 19 to 33% in patients with platinum-sensitive disease, 14 to 18% in patients with platinum-resistant disease, and 5 to 11% in patients with platinum-refractory disease (Herzog, 2002). The dose-limiting toxicity with PLDH is frequently rash including palmar-plantar erythrodysaesthesia. With topotecan, myelosuppression, especially neutropenia, is common. A recently
Ovary
published randomised trial showed a significant survival advantage for PLDH compared to topotecan (median survival 63.6 versus 57.0 weeks; Gordon et al., 2004).
Allergy to platinum chemotherapy Patients may develop an allergy to carboplatin and experience anaphylaxis during the infusion. Prompt treatment with chlorpheniramine, hydrocortisone, with or without adrenaline, is required. Carboplatin should not be re-administered, but patients may tolerate cisplatin chemotherapy, which is usually given with extra steroid cover as a precaution.
Paclitaxel neuropathy Patients may develop a peripheral neuropathy because of paclitaxel. If the neuropathy is debilitating and persistent during treatment, paclitaxel should be stopped. A balance needs to be struck between the debilitation and benefit.
Other treatments for palliation Patients with ovarian cancer may experience symptoms of recurrent ascites, pleural effusion or bowel obstruction, particularly when their disease becomes chemotherapy resistant.
Ascites Patients present with symptoms of abdominal distension and discomfort, shortness of breath and poor appetite. Paracentesis may be done as a day case procedure and an abdominal ultrasound may be helpful to identify the largest pocket of fluid. Patients with intractable ascites may benefit from a Denver peritovenous shunt.
Pleural effusion Patients present with shortness of breath and/or chest discomfort. Pleural aspiration (removal of approximately 1.5 l of fluid) provides rapid relief and may be done as a day case. Recurrence may be prevented by pleurodesis, which requires a chest drain to allow drainage of fluid to ‘dryness.’
Bowel obstruction Bowel obstruction may be acute or subacute. Patients present with colicky abdominal pain, constipation and vomiting. The characteristic X-ray changes may not be apparent in the early stages. Vomiting caused by bowel
obstruction may respond to cyclizine, hyoscine butylbromide, or octreotide.
Pelvic pain or PV bleeding Patients with local pelvic symptoms may benefit from palliative radiotherapy to the whole pelvis (e.g. 20 to 30 Gy in 5 to 10 fractions over 1 to 2 weeks). This treatment is usually limited to patients whose disease is confined to the pelvis and in whom chemotherapy is not appropriate.
Hormones The response rate to tamoxifen is low, at approximately 10%, but it may be beneficial in patients whose disease has a long natural history. Nevertheless, tamoxifen is an attractive option for some patients because of its low toxicity profile.
Prognosis The overall 5-year survival for ovarian cancer is about 35%, reflecting the typical late presentation of this disease. There is a significant age trend: patients aged 15 to 39 have a 5-year survival of around 80% compared with fewer than 20% for patients aged over 80 years (Cancer Research UK, Statistics, cancer facts and figures, www.cancerresearchuk.org/aboutcancer/statistics/, accessed September 2006). Similarly, there is a significant trend associated with FIGO stage: more than 75% of patients presenting with stage I disease typically survive 5 years, compared to less than 20% of patients with stage IV disease.
Areas of current interest Areas of current interest in ovarian cancer include the following: r Ovarian cancer screening. r The timing of surgery. r When to retreat in previously treated patients whose CA125 tumour marker rises, and the role of CA125 in follow-up. r Intraperitoneal chemotherapy, which has received much interest since the publication of the GOG 158 and GOG 172 studies, which showed a benefit in optimally debulked patients who receive intraperitoneal chemotherapy (Runowicz, 2006). r The role of other new chemotherapeutic agents (e.g. gemcitabine). 263
Louise Hanna and Malcolm Adams
Ongoing clinical trials At the time of writing the trials listed here for established ovarian cancer were open to recruitment and registered with the National Cancer Research Network (www.ncrn.org.uk, accessed September 2006). The CHORUS study is a randomised feasibility trial to determine the impact of timing of surgery and chemotherapy in newly diagnosed patients with advanced epithelial ovarian, primary peritoneal or fallopian tube carcinoma. OV05 is a randomised trial in relapsed ovarian cancer, comparing early treatment based on CA125 levels alone versus delayed treatment based on conventional clinical indicators. EORTC 55971 is a phase II randomised study of neoadjuvant chemotherapy followed by interval debulking surgery versus upfront cytoreductive surgery followed by chemotherapy with or without interval debulking surgery in patients with stage IIIC or IV epithelial ovarian cancer. SCOTROC 4 is a prospective multicentre randomised trial of carboplatin flat dosing versus intrapatient dose escalation in first-line chemotherapy of ovarian, fallopian tube and primary peritoneal cancers. SCOTROC 5 is a feasibility study of sequential carboplatin followed by paclitaxel and gemcitabine as a firstline chemotherapy for stage IC to IV ovarian, fallopian tube and primary peritoneal carcinomas. The BIBF1120 trial is a randomised placebocontrolled phase II study of continuous maintenance treatment with BIBF1120 following chemotherapy in relapsed ovarian cancer. BIBF1120 is an inhibitor of VEGF and VEGF-2. The CA125 doubling time study investigates the use of change in CA125 doubling time to detect the activity of cytostatic agents in women with ovarian carcinoma. The DNA methylation study looks at DNA methylation as a predictor for response and progression-free survival in patients with ovarian cancer. Early relapsed ovarian cancer (OII) is a phase II study of carboplatin and gemcitabine chemotherapy in patients with advanced ovarian cancer that was resistant or refractory to previous platinum chemotherapy.
Borderline ovarian tumours Borderline ovarian tumours are epithelial tumours of low malignant potential characterised by a lack of 264
stromal invasion in the ovary. Typically, they affect women at a younger age than does invasive cancer. Most borderline tumours are either serous or mucinous; approximately half of all cases are serous and onethird are mucinous. Most patients with borderline tumours present with early stage disease; however, stage III disease can occur. The mainstay of treatment is surgery, with maximal cytoreduction, but there is a role for conservative surgery in young women with early or localised disease who wish to retain their fertility. Women treated conservatively require close follow-up because the contralateral ovary may become affected. The role of adjuvant chemotherapy remains to be defined, with no proven benefit shown. Recurrent disease, however, may respond to platinum-based chemotherapy. The prognosis is significantly better than for invasive cancer; the 5-year survivals for stage I and stage III disease are up to 99% and 55 to 75%, respectively (WHO classification, 2003). For patients with stage III disease, the prognosis is worse if the peritoneal implants are invasive rather than growing on the surface of the peritoneum (Longacre et al., 2005).
Pseudomyxoma peritonei Pseudomyxoma peritonei refers to the condition involving abundant mucinous ascites in the pelvis and abdominal cavity, surrounded by fibrous tissue. The associated tumour cells may be benign, borderline or malignant. The finding of associated mucinous ovarian tumours usually indicates metastatic disease from the appendix or elsewhere in the GI tract rather than primary ovarian disease. The optimal management consists of removal of the tumour and complex peritonectomy, which is frequently combined with intraperitoneal chemotherapy (Moran and Cecil, 2003). The long-term prognosis is poor.
Granulosa cell tumour of the ovary Granulosa cell tumours account for less than 5% of all ovarian tumours. There are two distinct histological types. The first is the juvenile granulosa cell tumour, which accounts for 5% of granulosa cell tumours, occurs up to the age of 30, and nearly always presents in stage I. The second type is adult granulosa cell, which occurs from middle to old age and accounts for around 95% of cases. Presentation may be with non-specific abdominal or pelvic symptoms, vaginal bleeding (due to endometrial
Ovary
hyperplasia or adenocarcinoma, and associated with excess endogenous oestrogen produced by tumour cells), or acute tumour rupture and haemoperitoneum (due to the vascular nature of the tumour). Most patients present with stage I disease, and the treatment for young women is conservative fertilitysparing surgery (unilateral salpingo-oophorectomy). Older women are treated with total abdominal hysterectomy, bilateral salpingo-oophorectomy and infra-colic omentectomy. The role of adjuvant treatment has yet to be defined, and randomised trials are lacking for this uncommon disease. Most oncologists reserve further treatment until relapse. Responses to platinum-based chemotherapy regimens have been reported, including combinations of platinum, etoposide and bleomycin, or platinum, vinblastine and bleomycin. Radiotherapy may also have a role in delaying the progression of inoperable disease. There have also been responses reported to hormonal therapy, such as progestagens or gonadorelin analogues. The most important prognostic factor is stage of disease; age, tumour rupture and amount of residual disease have also been reported as significant. The survival for stage I disease is reported to be around 90% at 5 years, whereas for those with advanced-stage disease it is around 30%. Relapses have been reported 20 years after the original presentation and, for this reason, prolonged follow-up is recommended.
Ovarian germ cell tumours Ovarian germ cell tumours are a diverse group of tumours. The majority are mature teratomas (most commonly benign cystic tumours, also known as dermoid cysts), which have a peak incidence around the age of 30 years. Germ cell tumours are classified as follows: r Primitive germ cell tumours, including dysgerminoma and yolk sac tumour (also called endodermal sinus tumour); yolk sac tumour is characteristically positive for αFP. r Biphasic or triphasic tumours, including immature teratoma and mature teratoma. Immature teratoma is graded (1 to 3) according to the amount of immature neuroepithelial tissue present. r Monodermal teratoma including struma ovarii, a benign mature teratoma composed of thyroid tissue.
Malignant germ cell tumours The malignant germ cell tumours account for less than 5% of all ovarian cancers and are most common in women under the age of 20, the peak age being around 18 years old. The majority of malignant tumours are unilateral, and patients typically present with pain and a pelvic mass. Investigations must include αFP and βhCG. Surgical management involves unilateral oophorectomy only, in the majority of cases. More radical surgery should be avoided because fertility can usually be preserved without compromising the chance of a cure. Many would advise a programme of postoperative surveillance for patients with stage IA dysgerminoma or grade 1 stage I immature teratoma, with regular and frequent clinical, biochemical and radiological assessment. For others and those who relapse on surveillance, chemotherapy with bleomycin, etoposide and cisplatin (BEP) is frequently curative. The overall rate of survival for patients with malignant germ cell tumours is around 90%. A series involving 59 patients with metastatic ovarian germ cell tumours treated with chemotherapy had a 3-year survival of 87.8%, with no relapses occurring more than 3 years after treatment (Bower et al., 1996).
REFERENCES Anonymous. (1991). Chemotherapy in advanced ovarian cancer: an overview of randomised clinical trials. B. M. J., 303, 884–93. Anonymous. (1998). ICON2: randomised trial of single-agent carboplatin against three-drug combination of CAP (cyclophosphamide, doxorubicin and cisplatin) in women with ovarian cancer. Lancet, 352, 1571–6. Antoniou, A., Pharoah, P. D., Narod, S. et al. (2003). Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case series unselected for family history: a combined analysis of 22 studies. Am. J. Hum. Genet., 72, 1117–30. Benedet, J. L., Bender, H., Jones, H. 3rd et al. (2000). FIGO staging classifications and clinical practice guidelines of gynaecologic cancers. Int. J. Gynecol. Obstet., 70, 209–62. Bower, M., Fife, K., Holden, L. et al. (1996). Chemotherapy for ovarian germ cell tumours. Eur. J. Cancer., 32A, 593–7. Bristow, R. E., Tomacruz, R. S., Armstrong, D. K. et al. (2002). Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. J. Clin. Oncol., 20, 1248–59. Gordon, A. N., Tonda, M., Sun, S. et al. (2004). Long term survival advantage for women treated with pegylated liposomal doxorubicin compared with topotecan in a phase 3 randomized study of recurrent and refractory epithelial ovarian cancer. Gynecol. Oncol., 95, 1–8. Hanna, L. and Adams, M. (2006). Prevention of ovarian cancer. Best Pract. Res. Clin. Obstet. Gynaecol., 20, 339–62.
265
Louise Hanna and Malcolm Adams
Herzog, T. J. (2002). Update on the role of topotecan in the treatment of recurrent ovarian cancer. Oncologist, 7 (Suppl. 5), 3–10. ICON Group. (2002). Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide, doxorubicin, and cisplatin in women with ovarian cancer: the ICON 3 randomised trial. Lancet, 360, 505–15. Longacre, T. A., McKenney, J. K., Tazelaar, H. D. et al. (2005). Ovarian serous tumors of low malignant potential (borderline tumors): outcome-based study of 276 patients with long-term (> or = 5-year) follow-up. Am. J. Surg. Pathol., 29, 707–23. Lu, K. H., Dinh, M., Kohlmann, W. et al. (2005). Gynecologic cancer as a “sentinal cancer” for women with hereditary nonpolyposis colorectal cancer syndrome. Obstet. Gynecol., 105, 569–74. McGuire, W. P., Hoskins, W. J., Brady, M. F. et al. (1996). Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N. Engl. J. Med., 334, 1–6. Meeuwissen, P. A., Seynaeve, C., Brekelmans, C. T. et al. (2005). Outcome of surveillance and prophylactic salpingooophorectomy in asymptomatic women at high risk for ovarian cancer. Gynecol. Oncol., 97, 476–82. Modugno, F. (2003). Ovarian cancer and high-risk women: implications for prevention, screening, and early detection. Gynecol. Oncol., 91, 15–31. Moran, B. J. and Cecil, T. D. (2003). The etiology, clinical presentation, and management of pseudomyxoma peritonei. Surg. Oncol. Clin. N. Am., 12, 585–603. Muggia, F. and Hamilton, A. (2001). Phase II data on Caelyx® in ovarian cancer. Eur. J. Cancer., 37 (Suppl. 9), S15–18. Muggia, F. M., Braly, P. S., Brady, M. F. et al. (2000). Phase III randomized study of cisplatin versus paclitaxel versus cisplatin and paclitaxel in patients with suboptimal stage III or IV ovarian cancer: a Gynecologic Oncology Group study. J. Clin. Oncol., 18, 106–15. National Statistics. (2005). In Cancer Statistics Registrations, ed. M. Gautrey, M. Sheldreke and N. Cooper. Series MB1 no. 33. London: Office for National Statistics, p. 17.
266
NICE. (2003). Technology Appraisal Guidance – No. 55. Guidance of the Use of Paclitaxel in the Treatment of Ovarian Cancer. London: National Institute for Clinical Excellence. NICE. (2005). Technology Appraisal Guidance – No. 91. Paclitaxel, Pegylated Liposomal Doxorubicin Hydrochloride and Topotecan for Second-line or Subsequent Treatment of Advanced Ovarian Cancer. Review of Technology Appraisal Guidance 28, 45 and 55. London: National Institute for Clinical Excellence. Parmar, M. K., Ledermann, J. A., Colombo, N. et al. (2003). Paclitaxel plus platinum-based chemotherapy versus conventional platinum-based chemotherapy in women with relapsed ovarian cancer: the ICON4/AGO-OVAR-2.2 trial. Lancet, 361, 2099–106. Piccart, M. J., Bertelsen, K., James, K. et al. (2000). Randomized intergroup trial of cisplatin-paclitaxel versus cisplatin-cyclophosphamide in women with advanced epithelial ovarian cancer: three-year results. J. Natl. Cancer Inst., 92, 699–708. Risch, H. A., McLaughlin, J. R., Cole, D. E. et al. (2001). Prevalence and penetrance of germline BRCA1 and BRCA2 mutations in a population series of 649 women with ovarian cancer. Am. J. Hum. Genet., 68, 700–10. Runowicz, C. (2006). Should patients with ovarian cancer receive intraperitoneal chemotherapy following initial cytoreductive surgery? Nat. Clin. Pract. Oncol., 3, 416–7. SIGN. (2003). Guideline 75. Epithelial Ovarian Cancer. Edinburgh: Scottish Intercollegiate Guidelines Network. Struewing, J. P., Hartge, P., Wacholder, S. et al. (1997). The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N. Engl. J. Med., 336, 1401–8. Trimbos, J. B., Parmar, M., Vergote, I. et al. (2003). International Collaborative Ovarian Neoplasm trial 1 and Adjuvant ChemoTherapy In Ovarian Neoplasm trial: two parallel randomized phase III trials of adjuvant chemotherapy in patients with early-stage ovarian carcinoma. J. Natl. Cancer. Inst., 95, 105–12. WHO Classification. (2003). In World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of the Breast and Female Genital Organs, ed. A. Tavassoli and P. Devilee. Lyon: IARC Press, Chap. 4.
23
BODY OF THE UTERUS Louise Hanna and Malcolm Adams
Endometrial carcinoma Introduction
Risk factors and aetiology
The most common tumour affecting the body of the uterus is endometrial adenocarcinoma. The major risk factor for disease is unopposed oestrogen stimulation of the endometrium, which is associated with obesity; because of this, endometrial cancer is more common among women in developed countries. Most patients present with stage I disease and have a good prognosis when treated with a combination of surgery and selective postoperative radiotherapy. Other tumours affecting the body of the uterus include the uterine sarcomas, a group of tumours that may arise from the endometrium or the myometrium. These are aggressive tumours but treatment may be curative for early stage disease. Gestational trophoblastic tumours are discussed in Chapter 27.
Approximately 80% of endometrial carcinomas are of the endometrioid type and they arise against a background of unopposed oestrogen stimulation, which may be endogenous or exogenous.
Types of tumour affecting the uterus Approximately 90% of endometrial cancers are carcinomas, and approximately 90% of these are adenocarcinomas. Types of uterine tumour are shown in Table 23.1.
Incidence and epidemiology The annual incidence of uterine cancer is 14.9 in 100 000 (CRUK National Statistics; see info.cancerresearchuk. org, accessed September, 2006). In 2002 there were 5600 new cases of uterine cancer diagnosed in the UK. Uterine cancer accounts for 4% of all female malignancies. The disease is more common in the Western world than in developing countries and more common in women with high socioeconomic status and nulliparity. The high incidence has been linked with increasing levels of obesity and physical inactivity (Schouten et al., 2004). Endometrial carcinoma occurs typically in the postmenopausal age group, and the median age is 60 years.
Factors increasing risk Factors that increase the risk of endometrial carcinoma include increasing age, obesity, long-term exposure to unopposed oestrogens, genetic factors and atypical endometrial hyperplasia (reviewed by Amant et al., 2005b). Obesity, sometimes in association with diabetes and hypertension, causes high levels of unopposed endogenous oestrogen via conversion of androstenedione to oestrone in peripheral fat. Exogenous oestrogens include oestrogen-only hormone replacement therapy, which should not be prescribed in women with a native uterus, or hormonereplacement therapy with less than 12 to 14 days of progestagens. Long-term tamoxifen, which is used in the treatment and prevention of breast cancer, has a weak oestrogenic effect on the uterus. Endogenous estrogens include those secreted by granulosa cell tumours. Polycystic ovary syndrome, increasing years of menstruation, nulliparity and infertility are also associated with endometrial cancer. Genetic factors for endometrial carcinoma include a positive family history of endometrial, breast or colorectal cancer in a first-degree relative. An uncommon autosomal dominant genetic cause is hereditary non-polyposis colorectal cancer (HNPCC, Lynch type II), which is associated with colorectal, pancreatic, endometrial, breast and ovarian cancer. Atypical endometrial hyperplasia appears to be a premalignant phase for endometrioid adenocarcinoma (Kurman et al., 1985).
267
Louise Hanna and Malcolm Adams
Table 23.1. Types of uterine tumour
Table 23.2. Pathological features of endometrial carcinoma
Type of Description
tumour
Examples
Benign
Endometrial polyp
Macroscopic
Leiomyoma (fibroid)
features
Malignant
Endometrial carcinoma (90% of cancers)
Microscopic
primary
Uterine sarcomas:
features
¨ Mixed mullerian tumour (carcinosarcoma)
Often form a friable papillary or polypoid mass protruding into the uterine cavity Endometrioid type tumours, which account for about 80% of endometrial carcinomas, may be graded 1 to 3 on the basis of glandular formation,
Endometrial stromal sarcoma
non-squamous solid areas and
Leiomyosarcoma
nuclear atypia (WHO classification,
Others
2003); grade 1 represents well-formed
Lymphoma
glands with no more than 5% solid
Gestational trophoblastic diseases
non-squamous areas and no striking
Malignant
Direct spread from:
cytological atypia; mucinous tumours
secondary
Ovary
are nearly all grade 1
Rectum
Type II tumours
Bladder
Serous carcinoma has papillae, covered
Cervix
with highly pleomorphic tumour cells,
Vagina
frequent mitoses and necrosis; the
Metastatic spread (e.g. from breast)
probable precursor lesion is endometrial intraepithelial carcinoma (Ambros et al., 1995) Clear cell carcinoma has clear,
Factors decreasing risk
glycogen-filled cells and hobnail cells
Factors that decrease the risk of endometrial cancer include grand multiparity, oral contraceptive pill use, physical activity and a diet including some phytooestrogens.
with highly pleomorphic nuclei, which project into lumens and papillary spaces
Pathology Endometrial carcinomas are classified as follows (WHO classification, 2003): r Endometrial adenocarcinoma (including the variants with squamous differentiation, villoglandular, secretory and ciliated-cell). r Mucinous carcinoma. r Serous carcinoma. r Clear-cell carcinoma. r Mixed-cell adenocarcinoma. r Squamous-cell carcinoma. r Transitional-cell carcinoma. r Small-cell carcinoma. r Undifferentiated carcinoma. r Others. Endometrial carcinomas can be divided into type I and type II (Bokhman, 1983). Type I cancers com268
prise endometrioid and mucinous types, which are oestrogen-dependent tumours and frequently associated with atypical endometrial hyperplasia. Type II cancers lack an association with oestrogen stimulation and are characterised by aggressive behaviour. The two main type II carcinomas are serous and clear cell. Typical pathological features of endometrial carcinoma are shown in Table 23.2. Many endometrial carcinomas, particularly endometrioid carcinomas, express oestrogen and progesterone receptors. Other molecular characteristics include loss of function in tumour-suppressor pathways, in particular the PTEN pathway in type I cancers and the TP53 pathway in type II cancers.
Body of the uterus
Spread The regional lymph nodes are the pelvic and the paraaortic nodes. The main routes of lymphatic spread in the pelvis are to the utero-ovarian, parametrial, presacral, hypogastric, external iliac and common iliac nodes (Benedet et al., 2000). Endometrial cancer can spread locally through the myometrium and to the serosal surface of the uterus, to the cervix, the parametria, the fallopian tubes, the vagina, the bladder and the rectum. It can spread to lymph nodes, particularly pelvic nodes, para-aortic nodes and mediastinal nodes. Blood-borne cancer spread occurs to the lung, bone and liver, and peritoneal spread to peritoneal surfaces can occur.
Clinical presentation The usual presenting symptom is post-menopausal bleeding. This early sign allows prompt investigation and it means that most endometrial carcinomas are curable. Any woman with postmenopausal bleeding should be investigated because there is a 15% risk of endometrial carcinoma or hyperplasia (Gredmark et al., 1995). Most patients with endometrial adenocarcinoma present with stage I disease. A population-based study in Norway found 81% stage I, 11% stage II, 6% stage III and 2% stage IV disease (Abeler and Kjorstad, 1991). About 8% of endometrial carcinomas are associated with a simultaneous ovarian carcinoma of the same histology. These tumours are generally regarded as independent tumours (Eifel et al., 1982). Clinical features of the primary tumour are postmenopausal bleeding (usual presentation), vaginal discharge, other abnormal bleeding (intermenstrual, menorrhagia, postcoital) and a pelvic mass. Clinical features of local spread of the primary or lymphadenopathy are pelvic pain, rarely renal failure (ureteric obstruction), haematuria, bowel symptoms (e.g. constipation, rectal bleeding), and back pain (from para-aortic nodes). Metastatic disease is sometimes detected incidentally on a chest X-ray, or causes shortness of breath, bone pain and, rarely, cachexia and jaundice.
Screening There is no screening in use for the general population. Screening has a possible role in high-risk groups, such as
women taking tamoxifen or those with HNPCC. Screening involves a combination of ultrasound, endometrial biopsy and serum CA125. There is no evidence of a clinical benefit.
Investigation and staging Diagnostic investigations Transvaginal ultrasound can detect a thickened endometrium; the endometrium is abnormal if it is greater than 4–5 mm in postmenopausal women who do not take hormone replacement therapy (Gupta et al., 2002). A hysteroscopy may be performed as an outpatient procedure under local anaesthesia and allows inspection of the uterine cavity. Pipelle biopsy, an outpatient procedure, involves taking samples of endometrium through a long plastic tube. Examination under anaesthetic and curettage are indicated if the patient is unable to tolerate an outpatient procedure: r Anaesthetic assessment is important (especially for patients who are obese, hypertensive, diabetic or have heart disease). r Inspection of vulva, vagina, cervix, bimanual palpation. r Dilatation of the cervix, with or without hysteroscopy. r Curettings from endocervical canal and uterine body. r Cystoscopy or sigmoidoscopy if extension to bladder or rectum is suspected. Pregnancy status should be checked in premenopausal women.
Staging investigations Although endometrial cancer is staged surgically, it is essential to gain as much information about the extent of disease as possible to enable individualisation of treatment: r Imaging of the pelvis and abdomen is used to look at the depth of myometrial invasion, involvement of the cervix and possible lymph node enlargement. MRI scanning is superior to transvaginal ultrasound or CT scanning in detecting the depth of myometrial invasion (Kim et al., 1995). r Chest X-ray. r Full blood count. r Biochemical profile. r Serum CA125 may be useful for follow-up. r Other investigations such as a bone scan are undertaken if clinically or biochemically indicated. 269
Louise Hanna and Malcolm Adams
Table 23.3. Federation Internationale de Gynecologie et d’Obstetrique (FIGO) staging of endometrial cancer FIGO Stage
Description
IA
Limited to endometrium
IB
Invades up to less than half of myometrium
IC
Invades to more than one-half of
IIA
Endocervical glandular involvement only
myometrium IIB
Cervical stromal invasion
IIIA
Tumour involves serosa and/or adnexa (direct extension or metastasis) and/or cancer cells in ascites or peritoneal washings
IIIB
Vaginal involvement (direct extension or
IIIC
Metastasis to pelvic and/or para-aortic
IVA
Tumour invades bladder mucosa and/or
IVB
Distant metastasis (excluding metastasis
metastasis) lymph nodes bowel mucosa to vagina, pelvic serosa or adnexa, including metastasis to intra-abdominal lymph nodes other than para-aortic and/or inguinal nodes) Adapted from Benedet et al. (2000).
Staging The most commonly used staging system is that described by the Federation Internationale de Gynecologie et d’Obstetrique (FIGO). It is shown in Table 23.3.
Treatment overview Stage I Standard primary treatment is total abdominal hysterectomy (TAH) and bilateral salpingo-oophorectomy (BSO) and peritoneal washings, followed by selective radiotherapy for high-risk cases. The role of lymphadenectomy has recently been the subject of a randomised trial, the ASTEC study. There was no evidence that lymphadenectomy improves overall or disease-free survival; the 3-year survival was 86% for patients who had lymphadenectomy and 87% for those who did not (H. Kitchener, conference presentation, British Gynaecological Cancer Society Annual Meeting, 2005). 270
Laparoscopic or vaginal hysterectomy is an option for very obese patients who are unable to undergo an abdominal procedure. Patients who are medically unfit for hysterectomy also have the option of primary radiotherapy. Patients with grade 1, stage I tumours, who are unfit for surgery or radiotherapy, may respond to intrauterine progesterone, although there are conflicting reports of the usefulness of this approach (Dhar et al., 2005; Montz et al., 2002). Fertility-sparing options are considered later in the chapter.
Stage II Patients with stage II disease are at a greater risk of occult lymph node involvement. If cervical involvement is identified preoperatively, many centres advocate a radical hysterectomy followed by postoperative radiotherapy, although this approach is the subject of debate. The precise role of radiotherapy after radical hysterectomy is unclear, however, particularly if the lymph nodes are pathologically uninvolved.
Stages III to IVA Patients with cancer at stages III to IVA are generally unsuitable for surgery because the disease extends outside the uterus. Wherever possible, radical radiotherapy is given. If radical treatment is not possible (because of disease extent or comorbidity), palliative procedures, including radiotherapy, chemotherapy or hormonal treatment, are indicated.
Stage IVB Patients with metastatic disease require palliative treatments, which are aimed at improving the quality of life. A symptomatic benefit may be achieved with measures such as radiotherapy, chemotherapy or hormonal treatments. Occasionally, long-term disease control can be achieved with hormonal treatments for metastatic lowgrade endometrioid cancers.
Surgical treatments Total abdominal hysterectomy and bilateral salpingo-oophorectomy TAH and BSO make up the standard treatment for patients with stage I endometrial cancer. A midline laparotomy incision is performed and peritoneal washings are taken. The peritoneal surfaces are inspected carefully. The uterus, fallopian tubes and ovaries are
Body of the uterus
removed, the latter to exclude an ovarian metastasis or a synchronous ovarian cancer.
Laparoscopic hysterectomy The laparoscopic hysterectomy technique is appealing because postoperative recovery is quicker, hospital stays are shorter and wound complications are fewer. A recent randomised trial of laparoscopy versus laparotomy in endometrial cancer showed equivalent overall survival and disease-free survival in both groups (Tozzi et al., 2005).
Radical hysterectomy and lymphadenectomy As discussed earlier, the ASTEC study showed no benefit from lymphadenectomy in stage I endometrial cancer. For women who are known to have stage II disease prior to surgery, many centres would advocate radical (Wertheim-type) hysterectomy. The uterus, cervix, upper vagina and parametria are removed and pelvic lymphadenectomy is undertaken. In endometrial cancer, bilateral oophorectomy is also indicated because of the risk of spread to the ovary or a synchronous ovarian cancer.
Radiotherapy Definitive radical radiotherapy Radical radiotherapy is a standard treatment for patients with stage III endometrial cancer who are able to tolerate the procedure. Radical radiotherapy is also given as a primary treatment for patients with stage I or II disease who are medically unfit or decline surgery. Treatment involves external beam pelvic radiotherapy followed by brachytherapy. In selected patients with no adverse prognostic features, brachytherapy alone may be considered; however, the techniques carry some uncertainty as to whether the uterine cavity is fully treated.
External beam technique The patient lies supine on the couch with arms by sides, knees supported by a polystyrene wedge, and without further immobilisation. The bladder should be comfortably full. Orthogonal plain X-ray films are taken, A–P and lateral, with localisation marks (tattoos anterior and two laterals) and rulers. A barium-soaked tampon or vaginal ‘stock’ is placed in the vagina. For large tumours, CT planning is preferable to ensure the uterus is included in the target volume. The clinical target volume includes the primary tumour with a margin, uterus, cervix,
parametria, adnexae and regional nodes. A PTV may be grown around the CTV, or field borders marked as follows. It is important to remember that the field border is approximately 4 to 5 mm outside of the 95% isodose: r Superior border – between L5 and S1 vertebrae. r Inferior border – 2 cm below the inferior extent of the tumour (no higher than the lower border of the obturator foramina). r Lateral borders – 1.5 to 2 cm outside the bony pelvic side wall. r Posterior border – lower margin of the S2 vertebral body, approximately 2 cm in front of the sacral hollow or 2 cm behind the tumour on the CT planning scan. r Anterior border – through the symphysis pubis or 2 cm in front of the tumour on the CT planning scan. For planning, an outline of the patient is taken, and a plan is produced by the Physics Department. A four-field plan adequately covers the pelvis, with an anterior, a posterior, and two lateral fields. There is equal weighting from each beam. Field sizes are typically 12 to 15 cm (S–I) × 14 to 17 cm for the anterior and posterior fields, and 12 to 15 cm (S–I) × 10 to 12 cm for the lateral fields. With regard to dose energy and fractionation, the dose is prescribed to the ICRU reference point (intersection of beams), using 6 to 10 MV photons. Each field is treated daily, 5 days a week. The dose schedule is 40 to 50.4 Gy in 1.8 to 2 Gy per fraction prescribed to the ICRU reference point. Typical fractionation regimens are as follows: r 40 Gy in 20 fractions over 4 weeks. r 45 Gy in 25 fractions over 5 weeks. r 50.4 Gy in 28 fractions over 5.5 weeks. Verification is performed via portal beam imaging taken on the first 3 days of treatment and weekly thereafter.
Brachytherapy Typically a general anaesthetic is required for brachytherapy (or appropriate local or regional anaesthesia). A thorough anaesthetic assessment is required because many women in this group will have been turned down for definitive surgery because of comorbidity. A variety of techniques has been described, including a single-line source, an intra-uterine source in combination with vaginal ovoids and packing methods such as Heyman’s capsules. The uterine cavity is frequently longer than in patients with cervical cancer. There is no consensus over which dose specification points should be used. CT or MRI scanning with the applicators and the use of a computerised planning system can optimise the dose distribution. For a comprehensive review 271
Louise Hanna and Malcolm Adams
Table 23.4. Toxicity from pelvic radiotherapy for endometrial cancer Effect
Management
Acute effects Tiredness
Advice (e.g. goal setting, prioritising activity, self-pacing, treat associated
Anaemia
Red cell transfusion
Diarrhoea
Loperamide, low-residue diet
Proctitis
Rectal steroids or sucralfate
anxiety/depression)
Cystitis
Exclude infection, maintain fluid intake
Skin reaction
Aqueous cream or 1% hydrocortisone cream if severe
Nausea
Antiemetics (e.g. phenothiazines, metoclopramide, steroids, 5-HT3 antagonists, benzodiazepines)
Late effects Bowel stricture
Stool softeners, dietary advice; may progress to subacute or acute bowel
Bowel perforation
Emergency assessment, laparotomy if indicated
Rectal bleeding
Replace blood loss, proctoscopy, colonoscopy, surgery if indicated,
Haemorrhagic cystitis
Urine microscopy, renal function, cystoscopy, site-specific treatment for
Reduced bladder capacity, detrusor instability
Pelvic floor exercises, antimuscarinics, augmentation cystoplasty
Vaginal shortening and narrowing
Lubricants, vaginal dilators
obstruction
treatment directed to source of bleeding bleeding
of brachytherapy in endometrial cancer, see ESTRO (2002). For a single uterine tube with vaginal ovoids, ‘body loading’ refers to a variation of the classical Manchester isodose distribution, but with greater weighting in the uterine fundus to widen the isodoses at this level. The dose varies depending on the relative contributions from external beam and brachytherapy, and a detailed discussion is beyond the scope of this book. Some centres advocate the use of a central shield to reduce the external beam dose to the primary tumour, thus allowing a greater contribution from brachytherapy. Examples of brachytherapy doses include the following. r Following 45 Gy external beam radiotherapy, without a central shield: – For an intra-uterine tube with two vaginal ovoids: r The low-dose-rate dose is 25 to 30 Gy to point A in a single fraction with body loading. r The high-dose-rate dose is 21 to 24 Gy to point A in three to four fractions with body loading. – For Heyman’s capsule technique: r The low-dose-rate dose is 35 Gy in a single fraction. 272
r
The high-dose-rate dose is 30 Gy in six fractions.
r For brachytherapy alone, without external beam:
– The low-dose-rate dose is 70 to 80 Gy to the outer contour of the uterus in two fractions. – The high-dose-rate dose is 30 Gy in six fractions to the outer contour of the uterus, repeated 2 to 3 weeks later.
Explanation to patient Explain the procedure to the patient, including possible side effects. Give written information and allow the patient sufficient time to make a decision. Obtain informed consent, refer to a specialist nurse counsellor.
Toxicity of radiotherapy For a comprehensive review, see Faithfull and Wells (2003). Some of the toxicities from pelvic radiotherapy are shown in Table 23.4. Although premature menopause and infertility are also side effects of pelvic radiotherapy, most patients undergoing this treatment are already postmenopausal.
Postoperative radiotherapy The role of postoperative radiotherapy has been the subject of much debate. Three randomised trials have
Body of the uterus
demonstrated that postoperative radiotherapy reduces the risk of pelvic and vaginal recurrences but gives no benefit in overall survival (Aalders et al., 1980; Creutzberg et al., 2000; Keys et al., 2004). As a result the authors have concluded that postoperative radiotherapy is best confined to patients with risk factors for recurrence. Preliminary reports from the ASTEC study and a meta-analysis show no difference in overall survival between patients receiving or not receiving adjuvant external beam radiotherapy (EBRT). The incidence of isolated vaginal or pelvic recurrence at 5 years was 4% in those receiving EBRT and 7% in those who did not (p = 0.038; J. Orton, conference presentation, ASCO 2007). Risk factors for recurrence include stages IC, II and III, and grade 3 tumours, and postoperative radiotherapy should be considered for these patients. The precise roles of external beam and brachytherapy have yet to be defined, however, and are the subject of ongoing debate.
External beam technique: postoperative radiotherapy The patient lies supine on the couch with arms by sides, knees supported by a polystyrene wedge and without further immobilisation. The bladder should be comfortably full. Orthogonal plain X-ray films are taken, A–P and lateral, with localisation marks (tattoos anterior and two laterals) and rulers. A barium-soaked tampon or vaginal ‘stock’ is placed in the vagina. The CTV includes the operative bed and the regional lymph nodes. As for definitive radiotherapy, the PTV can be grown individually or field borders can be marked as described earlier. A four-field plan adequately covers the pelvis, with an anterior, a posterior, and two lateral fields. There is equal weighting from each beam. Field sizes are typically 12 to 15 cm (S–I) × 14 to 17 cm for the anterior and posterior fields and 12 to 15 cm (S–I) × 10 to 12 cm for the lateral fields. Dose-fractionation guidelines have been published by the Royal College of Radiologists (Board of the Faculty of Clinical Oncology, The Royal College of Radiologists, 2006). Acceptable treatment regimens are as follow: r 45 to 46 Gy in 1.8 to 2 Gy per fraction over 4.5 to 5 weeks. r 40 to 46 Gy in 20 to 25 daily fractions over 4 to 5 weeks. r Energy – 6 to 10 MV photons.
Portal beam imaging is taken for verification on the first 3 days of treatment and weekly thereafter.
Brachytherapy technique No anaesthetic is required. The procedure is carried out with a vaginal cylinder (obturator or ‘Dobbie’) and prior examination of the patient to determine which diameter tube should be used (typically 2 to 3 cm). Alternatively a pair of vaginal ovoids is used. Doses vary across centres and depend on the amount of external beam radiotherapy given. The local protocol should always be followed. Acceptable doses after external beam therapy include: r High-dose rate – 15 Gy in five daily fractions over 5 days to the top 3 to 5 cm, 0.5 cm from the surface of the obturator, or 8 Gy in two fractions, 3 to 5 days apart. r Low-dose rate: 20 to 25 Gy to the vaginal vault as a single fraction See earlier text for an explanation of side effects and toxicity discussion (p. 272).
Chemotherapy Adjuvant chemotherapy The role of chemotherapy as an adjuvant treatment is unproven but there is current interest in adjuvant chemotherapy for high-risk early stage and locally advanced disease.
Chemotherapy in advanced disease Chemotherapy may provide palliation in advanced disease. Single-agent carboplatin is well tolerated and has a response rate of around 30%, which is comparable with more toxic combinations such as cisplatin and doxorubicin. There is current interest in the combination of carboplatin and paclitaxel, which has been reported to have response rates of 63 to 87% (Akram et al., 2005; Michener et al., 2005).
Hormonal therapy In the adjuvant setting, a meta-analysis has shown that hormonal treatments do not improve overall survival and are associated with unacceptable cardiac morbidity and mortality (Martin-Hirsch et al., 1996). There is current interest in the use of progestagens alone in grade 1 endometrial carcinoma as a fertility-sparing treatment. A recent review of 81 patients reported a 76% initial response rate to 273
Louise Hanna and Malcolm Adams
progestagen therapy. Twenty of the responding patients became pregnant at least once after completing treatment (Ramirez et al., 2004). For patients with advanced, metastatic or recurrent endometrial cancer not amenable to surgery, progestagen treatment may provide useful palliation. Response rates of around 15 to 25% are achieved with medroxyprogesterone acetate. Responses are most likely to occur in patients with grade 1 tumours and in patients with a long history. Use of other hormonal treatments such as aromatase inhibitors, or the addition of tamoxifen to progestagens, is under investigation (Kieser and Oza, 2005).
Recurrent and metastatic disease Selected patients with local recurrence may still be suitable for radical treatments, either radiotherapy (if not already given) or surgery (e.g. pelvic exenteration). One study randomised high-risk stage I patients to surgery with or without radiotherapy. In the no-radiotherapy group, 15% developed a locoregional relapse and most of these underwent radical radiotherapy, achieving a 5year survival of 65% (Creutzberg et al., 2003). Para-aortic nodal disease may cause severe back pain and can be palliated with an additional para-aortic radiotherapy field. Metastatic disease may respond to hormones or chemotherapy as mentioned earlier.
r Stage 3 – 44%. r Stage 4 – 16%. r Overall – 73%. Areas of current interest Fertility-sparing options in young women Fewer than 5% of patients present at an age younger than 40 years. These patients are often nulliparous and they may wish to preserve their fertility. Progestagen treatment in selected cases, such as in patients with grade 1 stage I disease, has the potential to preserve fertility and allow subsequent pregnancies; however, close observation is required because of the risk of recurrence (Niwa et al., 2005).
Chemotherapy in high-risk endometrial cancer The role of adjuvant chemotherapy or concurrent chemoradiotherapy in high-risk early stage or locally advanced endometrial cancer is currently under investigation. A recent, very small study showed a highly significant result in favour of chemotherapy with cisplatin, doxorubin and cyclophosphamide, with an 88.5% progression-free survival compared with 50% in those receiving no chemotherapy (Aoki et al., 2004).
New chemotherapy agents
Prognosis
The topoisomerase inhibitor topotecan has shown some activity in previously treated women with endometrial cancer. Holloway (2003) reported a 10% response rate, with 55% disease stabilisation.
Prognostic factors
Hormonal treatments
Adverse prognostic factors increasing include depth of myometrial invasion, adenocarcinoma grade 3, nonendometrioid type, particularly clear cell, serous and adenosquamous and lymph node metastases.
Prognosis Type 1 tumours represent 65% of endometrial carcinomas and the 5-year survival is 85.6%, whereas type 2 tumours represent 35% of endometrial carcinoma. Type 2 tumours tend to be poorly differentiated, with deep invasion of the myometrium. The 5-year survival is 58.8% (Bokhman, 1983). Serous carcinoma and clear cell carcinoma are associated with a poor prognosis because 40% of tumours are metastatic at presentation. The overall 5-year survival percentage figures are as follow (FIGO, 1994): r Stage 1 – 86%. r Stage 2 – 66%. 274
Gonadotropin-releasing hormone analogues, selective oestrogen receptor modulators, aromatase inhibitors, and intra-uterine progestagens are all potential treatments for endometrial cancer (Lai and Huang, 2006).
Molecular pathways New avenues for treatment may be found by exploring the precise molecular pathways involved in tumour growth, such as oestrogen- and progesterone-related pathways and the molecular differences between type I and type II tumours.
Follow-up Most recurrences happen within 2 years of treatment, and the role of follow-up beyond 2 years has been debated. Advocates for follow-up highlight the possibility of detecting a potentially curable asymp-
Body of the uterus
tomatic central recurrence and the need to monitor outcomes. The counterargument is that there is no firm evidence that follow-up alters overall survival.
Ongoing clinical trials At the time of writing there were two open clinical trials for endometrial cancer registered with the National Cancer Research Network (www.ncrn.org.uk, accessed September 2006). EORTC 55984 is a phase III randomised study of doxorubicin and cisplatin with or without paclitaxel in patients with locally advanced, metastatic and/or relapsed endometrial cancer. EORTC 55991 is a phase II randomised study of adjuvant radiation with or without chemotherapy in highrisk endometrial carcinoma.
Uterine sarcomas Introduction Uterine sarcomas are uncommon tumours, which account for fewer than 5% of malignant uterine tumours. The age range of women affected spans from early adulthood to the postmenopausal age. Presenting symptoms include vaginal bleeding, pelvic pain, pelvic mass or an incidental finding at hysterectomy. Uterine sarcomas are defined as homologous if the sarcomatous element is derived from elements found normally in the uterus, and heterologous if the sarcomatous element is derived from elements not normally found in the uterus. Uterine sarcomas can be described according to nuclear and cytoplasmic appearances and the mitotic index (number of mitoses per 10 high-power field). The mode of spread is local to the surrounding pelvic organs, lymphatic to lymph nodes, through the peritoneum to peritoneal surfaces, or via the blood to lung or bone. The staging system generally adopted for uterine sarcomas is the FIGO staging for endometrial cancers, although the subgroups are often omitted. There are three distinct types of sarcoma of the uterus: r Leiomyosarcoma (approximately 50% of uterine sarcomas). r Mixed mullerian ¨ tumour (approximately 30% of uterine sarcomas). r Endometrial stromal sarcoma (approximately 20% of uterine sarcomas). The prognosis for uterine sarcoma is unfavourable, with a 30% overall 5-year survival (Livi et al., 2004).
Types of uterine sarcoma Leiomyosarcoma The mean age of patients with leiomyosarcoma is 55.3 years (Nordal et al., 1993). Its benign counterpart is the leiomyoma or fibroid, but malignant change in a fibroid is very rare and only about 5 to 10% of leiomyosarcomas arise in a pre-existing fibroid. Leiomyosarcoma is a highly malignant tumour with an overall survival rate from 15 to 25%; patients with stage I and II tumours have a 5-year survival of 40 to 70% (WHO classification, 2003).
Mixed mullerian tumour ¨ ¨ The mixed mullerian tumour is also called carcinosarcoma. The median age of presenting patients is 65 years (WHO classification, 2003). The tumour is composed of malignant epithelial cells and malignant stromal cells. Occasionally the epithelial component is benign, which is called adenosarcoma. It is accepted ¨ that mixed mullerian tumours have an epithelial origin and represent metaplastic carcinomas (WHO classification, 2003). However, the pattern of spread and prognosis are different from high-grade endometrial ¨ carcinoma; mixed mullerian tumours are more likely to spread to lymph nodes and have a poorer prognosis (Amant et al., 2005a). The 3-year survival is poor at 22% (Livi et al., 2004).
Endometrial stromal sarcoma The mean age of patients is 55.3 years (Nordal et al., 1993). Endometrial stromal sarcoma is classified as either low grade or high grade; the latter is also known as undifferentiated endometrial or uterine sarcoma (WHO classification, 2003). Low-grade tumours frequently express oestrogen receptors and progesterone receptors, whereas undifferentiated endometrial sarcomas do not. One series reported 54% of endometrial stromal sarcomas to be of low-grade type at the time of diagnosis. The overall 5-year survival was 67%, but the 5-year survival of patients with grade 3 or 4 disease was only 33% (Nordal et al., 1993).
Treatment of uterine sarcomas The main treatment of uterine sarcomas is surgery. Postoperative radiotherapy reduces the risk of a local relapse, but it is not thought to influence overall survival (Hornback et al., 1986). Patients with advanced or metastatic disease may respond to palliative treatments. For example, patients with metastatic endometrial 275
Louise Hanna and Malcolm Adams
stromal sarcomas may benefit from progestagens. Modest responses may be seen with palliative chemotherapy (Kanjeekal et al., 2005). Response rates are typically in the order of 15 to 30% for regimens based on dox¨ orubicin or ifosfamide. Patients with a mixed mullerian tumour may respond to platinum-based chemotherapy.
REFERENCES Aalders, J., Abeler, V., Kolstad, P. et al. (1980). Postoperative external irradiation and prognostic parameters in stage I endometrial carcinoma: clinical and histopathological study of 540 patients. Obstet. Gynecol., 56, 419–27. Abeler, V. M. and Kjorstad, K. E. (1991). Endometrial adenocarcinoma in Norway. A study of a total population. Cancer, 67, 3093–103. Akram, T., Maseelall, P. and Fanning, J. (2005). Carboplatin and paclitaxel for the treatment of advanced or recurrent endometrial cancer. Am. J. Obstet. Gynecol., 192, 1365–7. Amant, F., Cadron, I., Fuso, L. et al. (2005a). Endometrial carcinosarcomas have a different prognosis and pattern of spread compared to high-risk epithelial endometrial cancer. Gynecol. Oncol., 98, 274–80. Amant, F., Moerman, P., Neven, P. et al. (2005b). Endometrial cancer. Lancet, 366, 491–505. Ambros, R. A., Sherman, M. E., Zahn, C. M. et al. (1995). Endometrial intraepithelial carcinoma: a distinctive lesion specifically associated with tumours displaying serous differentiation. Human Pathol., 26, 1260–7. Aoki, Y., Watanabe, M., Amikura, T. et al. (2004). Adjuvant chemotherapy as treatment of high-risk stage I and II endometrial cancer. Gynecol. Oncol., 94, 333–9. Benedet, J. L., Bender, H., Jones, H. 3rd et al. (2000). FIGO staging classifications and clinical practice guidelines in the management of gynecologic cancers. Int. J. Gynecol. Obstet., 70, 209–62. Board of the Faculty of Clinical Oncology, The Royal College of Radiologists. (2006). Radiotherapy Dose-Fractionation. London: Royal College of Radiologists, p. 36. Bokhman, J. V. (1983). Two pathogenetic types of endometrial carcinoma. Gynecol. Oncol., 15, 10–17. Creutzberg, C. L., van Putten, W. L., Koper, P. C. et al. (2000). Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. Lancet, 355, 1404–11. Creutzberg, C. L., van Putten, W. L., Koper, P. C. et al. (2003). Survival after relapse in patients with endometrial cancer: results from a randomized trial. Gynecol. Oncol., 89, 201–9. Dhar, K. K., NeedhiRajan, T., Koslowski, M. et al. (2005). Is levonorgestrel intrauterine system effective for treatment of early endometrial cancer? Report of four cases and review of the literature. Gynecol. Oncol., 97, 924–7. Eifel, P., Hendrickson, M., Ross, J. et al. (1982). Simultaneous presentation of carcinoma involving the ovary and the uterine corpus. Cancer, 50, 163–70.
276
ESTRO. (2002). The GEC ESTRO Handbook of Brachytherapy, ed. A. Gerbaulet, R. Potter, J.-J. Mazeron et al. Brussels: ESTRO, Chap. 15, pp. 365–401. Faithfull, S. and Wells, M. (2003). Supportive Care in Radiotherapy. London: Churchill Livingstone. FIGO. (1994). FIGO Annual Report on the Results of Treatment in Gynaecologic Cancer: Cancer of the Corpus Uteri 13040 Patients Treated 1987–1989, ed. F. Pettersson. Stockholm, Sweden, 22, 169–99. Gredmark, T., Kvint, S., Havel, G. et al. (1995). Histopathological findings in women with postmenopausal bleeding. Br. J. Obstet. Gynaecol., 102, 133–6. Gupta, J. K., Chien, P. F., Voit, D. et al. (2002). Ultrasonographic endometrial thickness for diagnosing endometrial pathology in women with postmenopausal bleeding: a meta-analysis. Acta Obstet. Gynecol. Scand., 81, 799–816. Holloway, R. W. (2003). Treatment options for endometrial cancer: experience with topotecan. Gynecol. Oncol., 90, S28–33. Hornback, N. B., Omura, G. and Major, F. J. (1986). Observations on the use of adjuvant radiation therapy in patients with stage I and II uterine sarcoma. Int. J. Radiat. Oncol. Biol. Phys., 12, 2127–30. Kanjeekal, S., Chambers, A., Fung, M. F. et al. (2005). Systemic therapy for advanced uterine sarcomas: a systematic review of the literature. Gynecol. Oncol., 97, 624–37. Keys, H. M., Roberts, J. A., Brunetto, V. L. et al. (2004). A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinoma: a Gynecologic Oncology Group study. Gynecol. Oncol. . 92, 744–51. Kieser, K. and Oza, A. M. (2005). What’s new in systemic therapy for endometrial cancer. Curr. Opin. Oncol., 17, 500–4. Kim, S. H., Kim, H. D., Song, Y. S. et al. (1995). Detection of deep myometrial invasion in endometrial carcinoma: comparison of transvaginal ultrasound, CT and MRI. J. Comput. Assist. Tomogr., 19, 766–72. Kurman, R. J., Kaminski, P. F. and Norris H. J. (1985). The behaviour of endometrial hyperplasia. A long-term study of ‘untreated’ hyperplasia in 170 patients. Cancer, 56, 403–12. Lai, C. H. and Huang, H. J. (2006). The role of hormones for the treatment of endometrial hyperplasia and endometrial cancer. Curr. Opin. Obstet. Gynecol., 18, 29–34. Livi, L., Andreopoulou, E., Shah, N. et al. (2004). Treatment of uterine sarcoma at the Royal Marsden Hospital from 1974 to 1988. Clin. Oncol. (R. Coll. Radiol.), 16, 261–8. Martin-Hirsch, P. L., Lilford, R. J. and Jarvis, G. J. (1996). Adjuvant progestagen therapy for the treatment of endometrial cancer: review and meta-analysis of published randomised controlled trials. Eur. J. Obstet. Gynecol. Reprod. Biol., 65, 201–7. Michener, C. M., Peterson, G., Kulp, B. et al. (2005). Carboplatin plus paclitaxel in the treatment of advanced or recurrent endometrial carcinoma. J. Cancer Res. Clin. Oncol., 131, 581–4. Montz, F. J., Bristow, R. E., Bovicelli, A. et al. (2002). Intrauterine progesterone treatment of early endometrial cancer. Am. J. Obstet. Gynecol., 186, 651–7. Niwa, K., Tagami, K., Lian, Z. et al. (2005). Outcome of fertility-preserving treatment in young women with endometrial carcinomas. Br. J. Obstet. Gynecol., 112, 317–20.
Body of the uterus
Nordal, R. N., Kjorstad, K. E., Stenwigh, A. E. et al. (1993). Leiomyosarcoma (LMS) and endometrial stromal sarcoma (ESS) of the uterus. A survey of patients treated in the Norwegian Radium Hospital 1976–1985. Int. J. Gynecol. Cancer, 3, 110–15. Ramirez, P. T., Frumovitz, M., Bodurka, D. C. et al. (2004). Hormonal therapy for the management of grade 1 endometrial adenocarcinoma: a literature review. Gynecol. Oncol., 95, 133–8. Schouten, L. J., Goldbohm, R. A. and van den Brandt, P. A. (2004). Anthropometry, physical activity, and endometrial cancer risk:
results from the Netherlands Cohort Study. J. Natl. Cancer. Inst., 96, 1635–8. Tozzi, R., Malur, S., Koehler, C. et al. (2005). Laparoscopy versus laparotomy in endometrial cancer: first analysis of survival of a randomized prospective study. J. Minim. Invasive Gynecol., 12, 130–6. WHO classification. (2003). In World Health Organization Classification of Tumours: Pathology and genetics of tumours of the breast and female genital organs, ed. A. Tavassoli and P. Devilee. Lyon: IARC Press, Chap. 4.
277
24
CERVIX Louise Hanna and Malcolm Adams
Introduction
Incidence and epidemiology
Cervical cancer is the most common cause of death from female malignancy worldwide. Overall it causes more than 273 000 deaths per year, accounting for 9% of all female cancer deaths. The incidence is highest in developing countries (Ferlay et al., 2004). The major risk factor is persistent human papilloma virus (HPV) infection, particularly types 16 and 18. In the UK the incidence of invasive disease has fallen as a result of cervical screening, and the mortality rates are 60% lower than they were 30 years ago. For patients presenting with very early stage cancers (stages IA1 to IB1), surgery is the mainstay of treatment. For patients presenting with later-stage disease (IB2 to IVA), the recent standard treatment has become concurrent radiotherapy with cisplatin-based chemotherapy. The prognosis is strongly related to the stage of disease at presentation. There is major interest in the prospect of cervical cancer prevention via the development of vaccines against HPV infection.
The annual incidence of cervical cancer in the UK is approximately 9 in 100 000 cases (Cancer Research UK; http.//info.cancerresearchuk.org, accessed 2006). It is the 12th most common cancer in females, and represents 2% of all female cancers. Approximately 2500 new cases of cervical cancer are diagnosed annually in the UK, from which approximately 1000 deaths occur annually. The peak incidence of disease occurs in ages 40 to 45 years. The incidence has fallen over the past 10 years because of effective screening. There is a large worldwide geographical variation in disease occurrence, with the highest incidence in developing countries.
Types of cervical tumour Cervical tumours can be benign, malignant primary or malignant secondary. The range of tumours is shown in Table 24.1.
Anatomy The cervix is approximately 2.5 cm long and it is situated in the pelvis at the lower end of the uterus. The lower part of the cervix projects into the vagina. The bladder lies anteriorly, and the pouch of Douglas (which may contain small bowel) and the rectum, posteriorly. The parametria lie laterally within the broad ligaments and they contain the ureters (1 to 2 cm from the cervix) and the uterine arteries. The cervix is hollow and communicates superiorly with the body of the uterus and inferiorly with the vagina. 278
Carcinoma of the cervix Risk factors and aetiology The major risk factor for carcinoma of the cervix is human papilloma virus (HPV) infection (Helmerhorst and Meijer, 2002). Although most women are exposed to HPV, it is the presence of persistent infection that leads to malignant change. Types 16, 18 and 31 confer the highest risk, with the viral proteins E6 and E7 being responsible for malignant transformation. Herpes simplex virus type 2 may act as a co-factor (Smith et al., 2002). Many of the other risk factors may reflect exposure to HPV: r Early age at first pregnancy. r Multiparity. r Multiple sexual partners. r Partner who has had multiple sexual partners. r Use of an oral contraceptive pill (Moreno et al., 2002). r Cigarette smoking. r Low social class. r Immunocompromise (AIDS defining illness). r Diethylstilboestrol exposure in utero (clear cell carcinoma, as in vagina). The premalignant phase of squamous cervical cancer is cervical intraepithelial neoplasia (CIN).
Cervix
Table 24.1. The range of cervical tumours Type
Examples
Benign
Endocervical polyp
Malignant
Carcinomas
primary
Squamous carcinoma (90%)
Table 24.2. Treatment following cervical smears Result of cervical smear
Action
Normal
Repeat in 3–5 years
Inadequate
Repeat; refer for colposcopy
Borderline
Repeat; refer for colposcopy
after 3 inadequate
Adenocarcinoma
after 3 borderline
Adenosquamous carcinoma
squamous or 1 borderline
Undifferentiated carcinoma Clear cell (DES exposure) Small cell
glandular Mild
Refer for colposcopy
Moderate
Refer for colposcopy within 4
Severe
Refer for colposcopy within 4
Possible invasion
Refer for colposcopy urgently
Glandular neoplasia
Refer for colposcopy urgently
Abnormal cervix or
Refer for gynaecological
Transitional cell Others
weeks
Carcinoid Lymphoma
weeks
Melanoma Sarcoma
within 2 weeks
Rhabdomyosarcoma (esp. paediatric) Malignant secondary
Direct spread from
within 2 weeks
Endometrium
symptoms of cervical
Bladder
cancer
examination
Ovary Vagina
Adapted from NHS Cancer Screening Programme (2004).
Peritoneum Metastatic spread from other tumours E.g. breast, rarely thyroid
Cervical screening Cervical screening detects abnormal cells at the dyskaryotic or precancerous stage (i.e. CIN). Screening was set up in the UK in the mid-1960s, but uptake by the general public was low. In 1988 the Department of Health set up the NHS Cervical Screening Programme, which incorporated a call-recall system and a requirement to meet quality standards. As a result the proportion of women screened rose from 45% in 1988–9 to 83.7% in 1999–2000 and deaths from cervical cancer have fallen by 7% every year. The protective effect for adenocarcinoma appears to be less (Boon et al., 1987). All women between the ages of 25 and 64 are eligible for screening every 3 to 5 years. The traditional method of performing a cervical or Papanicolau (Pap) smear is with a spatula. A scraping of cells is spread onto a glass slide and then processed in a laboratory and examined by a cytologist. The results are classified according to the British Society for Clin-
ical Cytology as ‘negative, borderline, mild, moderate, severe [dyskaryosis], ?glandular neoplasia, ?invasive, or inadequate.’ Newer techniques involving liquidbased cytology are up to 12% more sensitive and result in fewer inadequate samples. A sample of cells is obtained using a brush, which is then rinsed or broken off in a vial of preservative fluid. In the laboratory, cellular debris is removed and a thin layer of cervical cells is deposited onto a microscope slide. The National Institute for Health and Clinical Excellence recommended that liquid-based cytology be used as the primary means of processing samples in the cervical screening programme in England and Wales (NICE, 2003). Table 24.2 shows the management for different cervical smear results, as recommended by the NHS Cancer Screening Programme (2004). Colposcopy involves examination of the cervix with a low-power microscope. Acetic acid or iodine is used to stain areas of dyskaryosis white or yellow, respectively. Abnormal areas are removed using LLETZ (large loop excision of the transformation zone), LEEP (loop electro-excision procedure) or, when microinvasion is suspected, knife cone biopsy. Follow-up is essential because up to 5% of cases require further excision. 279
Louise Hanna and Malcolm Adams
Table 24.3. Pathological features of invasive cervical carcinoma Features
Description
Macroscopic
Microinvasive cancer is usually clinically undetectable; clinically detectable cancers are typically fungating, producing an obvious mass, but may be ulcerating or infiltrative
Microscopic
90% are squamous carcinomas; may be keratinising or non-keratinising; graded according to degree of
Examination may reveal a cervix that is enlarged or ulcerated and sometimes fixed (immobile). Clinical features of local spread from the primary or involved pelvic lymph nodes include renal failure (ureteric obstruction), frequency and dysuria, pelvic pain, bladder outflow obstruction, a change in bowel habit, rectal bleeding, haematuria, urine incontinence (vesico-vaginal fistula), faecal incontinence (rectovaginal fistula), deep pelvic pain or lymphoedema of the legs. Clinical features of metastatic spread include bone pain, weight loss, anorexia or cachexia; and rarely jaundice or dyspnoea.
pleomorphism and mitoses; in microinvasive cancer, small finger-like
Investigation and staging
processes penetrate the basement membrane.
Pathology Typically, carcinoma of the cervix originates at the squamous-columnar junction. The pre-invasive lesion is CIN, which has three grades. CIN is graded on the basis of cellular and nuclear atypia and mitoses. CIN 3 is carcinoma in situ. Invasion of the basement membrane indicates progression to true cervical cancer. In general, progression from CIN to invasive cancer takes 10 to 12 years. About 30% of women with CIN 3 develop invasive cancer. The pathological features of cervical cancer are shown in Table 24.3. Immunohistochemical stains can be useful in distinguishing between lineages of a neoplasm; for example, reactivity to anticytokeratin antibodies suggests an epithelial origin, and reactivity to antibodies against mucin suggests adenocarcinoma.
Examination under anaesthetic and biopsy give both diagnostic and staging information. Following a thorough inspection of the vulva, vagina and cervix, the cervix is palpated bimanually to estimate size, position and mobility. Parametrial invasion is best assessed via rectal examination. A cervical biopsy is taken, and cystoscopy, urine cytology and sigmoidoscopy are performed. Computed tomography (CT) and/or magnetic resonance imaging (MRI) of the abdomen and pelvis have superseded the use of an intravenous urogram. MRI is better for assessing the primary tumour, but either can be used to assess lymph node areas. MRI is essential for imaging pregnant women with cervical carcinoma. Other investigations include chest X-ray, a full blood count and a biochemical profile. EDTA clearance or estimation of creatinine clearance (e.g. Cockcroft-Gault formula, as described in Chapter 1, p. 4) should be checked if chemoradiotherapy is planned. Other investigations, such as a bone scan, are undertaken if they are clinically or biochemically indicated. Pregnancy status should be checked in premenopausal women.
Spread Cervical carcinoma can spread locally to involve contiguous structures (parametrium, vagina, ureters, bladder, rectum or peritoneum); via lymphatics to pelvic, para-aortic and mediastinal lymph nodes; or via the blood to lungs, bone and liver.
Staging classification The staging classification of cervical cancer is shown in Table 24.4. The first-station lymph nodes are the parametrial, internal and external iliac, presacral and common iliac (Benedet et al., 2000).
Clinical presentation
Treatment overview
Symptoms of the primary tumour include postcoital bleeding, intermenstrual bleeding, persistent vaginal discharge, postmenopausal bleeding and dyspareunia.
Stage IA1
280
Stage IA1 cervical cancer represents microinvasive disease. The risk of lymph node metastases is so low that
Cervix
Table 24.4. Federation Internationale de Gynecologie et d’Obstetrique (FIGO) staging of cervical cancer FIGO stage
Description
0
Carcinoma in situ
I
Confined to uterus
IA
Diagnosed only by microscopy
IA1
Depth ≤ 3 mm, horizontal spread ≤ 7 mm
IA2
Depth > 3–5 mm, horizontal spread ≤ 7
IB
Clinically visible or microscopic lesion
IB1
4 cm or less in greatest dimension
IB2
More than 4 cm in greatest dimension
II
Invades beyond uterus but not pelvic
mm greater than 1A2
wall or lower third of vagina IIA
Without parametrial invasion
IIB
With parametrial invasion
III
Extends to pelvic wall and/or lower third of vagina and/or hydronephrosis or non-functioning kidney
IIIA
Lower third of vagina, no extension to
IIIB
Pelvic wall and/or causes hydronephrosis
IVA
Invades mucosa of bladder or rectum
IVB
Distant metastasis
pelvic wall or non-functioning kidney and/or extends beyond true pelvis
Adapted from Benedet et al. (2000).
simple hysterectomy with or without oophorectomy is enough (Sevin et al., 1992). Knife cone biopsy alone may be sufficient in selected patients who wish to retain fertility. Intracavitary brachytherapy alone is needed for patients who are unfit for surgery.
Stage IA2 The risk of lymph node metastasis has been reported as 7.4% (Buckley et al., 1996). Typically radical hysterectomy with lymph node dissection is carried out, although there is interest in more conservative approaches (e.g. local excision of the cervical tumour with pelvic lymphadenectomy). Radical trachelectomy (see later discussion) is an option for patients who wish to preserve fertility but is limited to patients with tumours up to 2 cm with no adverse prognostic factors. Radical radiotherapy is needed for patients who are medically unfit.
Stage IB1 and small-volume stage IIA Patients with low-risk stage IB tumours (less than 4 cm) and small-volume stage IIA tumours may be treated with radical hysterectomy with preservation of the ovaries and lymphadenectomy, or radical radiotherapy. Radical trachelectomy is an option for patients who wish to maintain fertility, and as aforementioned is limited to patients with tumours up to 2 cm diameter and no adverse prognostic factors. Postoperative cisplatinbased chemoradiotherapy is given in high-risk cases (e.g. those with positive lymph nodes, involved resection margins, lymphovascular invasion or occult stage IIB disease; Peters et al., 2000).
Stage IB2 to IVA In 1999, five randomised clinical trials of concurrent platinum-based chemotherapy showed an overall survival benefit in patients with stages IB2 to IVA disease (Keys et al., 1999; Morris et al., 1999; Peters et al., 2000; Rose et al., 1999, Whitney et al., 1999). The risk of death from cervical cancer was reduced by 30 to 50% when cisplatin-based chemotherapy was added to radiotherapy. Consequently, in the same year, the National Cancer Institute issued a clinical alert, which stated that ‘strong consideration be given to the incorporation of concurrent cisplatin-based chemotherapy with radiation therapy in women who require radiation therapy for treatment of cervical cancer.’ These trials led to a metaanalysis that confirmed the benefit (Green et al., 2001). A subsequent trial performed by the National Cancer Institute of Canada demonstrated no survival advantage for concurrent platinum-based chemoradiation in stages IB to IVA cervical cancer patients (Pearcey et al., 2002). These conflicting results and a number of small randomised controlled trials with insufficient statistical power led to a further systematic review and metaanalysis (Lukka et al., 2002). This meta-analysis looked at eight randomised trials of cisplatin-based concurrent chemotherapy and radiotherapy versus radiotherapy alone. The relative risk of death (RR) for the whole group was 0.74 (95% CI 0.64–0.86) in favour of concurrent chemoradiotherapy. For patients with locally advanced disease the RR was 0.78 (95% CI 0.67–0.9) and for those with high-risk early stage disease the RR was 0.56 (95% CI 0.41–0.77). Thus, the benefit of concurrent cisplatin-based chemotherapy and radiotherapy in this setting was confirmed, with the greatest apparent benefit being in women with high-risk early stage disease. 281
Louise Hanna and Malcolm Adams
Stage IVB
Pelvic exenteration
Patients with metastatic disease require palliative treatment aimed at improving their quality of life.
Pelvic exenteration is a major operation that is usually performed on patients for whom more conservative approaches have failed. It involves the formation of one or more stomata (ileal conduit for anterior exenteration, colostomy for posterior exenteration, both for total exenteration). Patient selection is important because the procedure is both physically and psychologically demanding.
Recurrent disease Central pelvic recurrent disease following radical chemoradiotherapy can occasionally be cured by surgery. Evaluation with MRI scanning and a careful search for metastatic disease are required. Typically surgery involves pelvic exenteration (anterior, posterior or total) with the formation of one or two stomata (ileal conduit or colostomy). Recurrent disease at the pelvic side wall cannot be cured surgically, and treatment is palliative and aimed at maintaining the patient’s quality of life.
Surgical treatments Simple hysterectomy Patients with stage IA1 disease can have a simple hysterectomy, and no parametrial or lymph node dissection is required. Postmenopausal patients should also have a BSO. The cuff of the vagina typically involves fornices only.
Radical hysterectomy The term radical hysterectomy covers a range of operations that are usually carried out on patients with stages IA2 to IB1 tumours. The uterus and tubes are removed together with the primary tumour and a 1 cm margin (i.e. the parametria are removed). A cuff of the vagina is also removed and there is pelvic lymph node dissection. Premenopausal patients do not need to have their ovaries removed. The original technique was described by Wertheim.
Radical vaginal trachelectomy and laparoscopic lymphadenectomy Radical vaginal trachelectomy and laparoscopic lymphadenectomy have been used since the 1990s as a fertility-sparing technique for women with early stage cervical cancer (e.g. stages IA2 or IB1). The cervix, parametria, and upper vagina are removed and the vault of the vagina is sutured onto the lower uterine segment; this procedure can be done via a vaginal approach. A lymphadenectomy is also performed because of the risk of nodal disease and may be done laparoscopically. A recently reported series described 33 live births out of 50 pregnancies in 72 patients treated by radical vaginal trachelectomy (Plante et al., 2005). 282
Palliative surgery The most common palliative operation is the formation of a stoma to relieve symptoms of fistulae.
Radiotherapy Radical radiotherapy and chemoradiotherapy Radiotherapy alone is used for medically unfit patients with stages IA2 and IB1 disease or for patients who decline surgery. For patients with very small tumours with negative lymph nodes found on imaging, most or all of the treatment may be delivered by brachytherapy rather than by external beam. The standard radical treatment for patients with stages IB2 to IVA disease is a course of external beam radiotherapy with concurrent cisplatin-based chemotherapy, followed by brachytherapy. Radical radiotherapy alone is reserved for medically unfit patients.
Radiation dose and dose rate A variety of different radiation doses are in use throughout the UK and worldwide for external beam and brachytherapy. Much of the early data on outcomes for cervical cancer came from series incorporating low-dose-rate (LDR) brachytherapy with radium sources. Comparison of LDR treatments between centres is facilitated by ICRU 38 (International Commission on Radiological Units and Measurements, 1985). This report specifies the 60 Gy envelope for treatments using LDR brachytherapy and other reference points (e.g. bladder and rectum). More recently there has been a trend towards more extensive use of medium-dose-rate (MDR) and high-dose-rate (HDR) brachytherapy. Use of HDR brachytherapy for cervical cancer has recently become more widespread for reasons including convenience for the patient, radiation protection and machine availability. Table 24.5 shows a comparison of LDR and HDR brachytherapy. A recent report by the National Institute for Health and Clinical Excellence (NICE, 2006) looked at evidence for efficacy and safety of HDR brachytherapy. No significant difference in overall survival was found in
Cervix
Table 24.5. Comparison of low- and high-dose-rate brachytherapy for cervical carcinoma
Radio isotopes
Low dose rate
High dose rate
Caesium-137
Cobalt-60 Iridium-192
Time taken
Days
Convenience
Requires hospitalisation
Day case procedure
Radiobiology
Typically single procedure
Requires fractionation
Geometry
Longer treatment – applicators could move
Fixed position – reduces applicator movement
Thromboembolism
Greater risk
Lesser risk
randomised trials that compared HDR with either LDR or MDR brachytherapy. In terms of safety, serious complications including grade 3 or 4 toxicity occurred in 2 to 6% of patients in reported series of HDR brachytherapy. The NICE report concludes that the evidence on safety and efficacy of HDR brachytherapy for carcinoma of the cervix appears adequate to support its use, provided that the normal arrangements are in place for consent, audit, and clinical governance. In terms of the doses of radiation prescribed, the relative doses of external beam and brachytherapy depend on the initial volume of disease, the ability to displace the bladder and rectum, the degree of tumour regression during pelvic irradiation, and the institutional preference (Nag et al., 2000). Doses for external beam and brachytherapy vary across centres and the local protocol should always be followed. Current typical dose ranges for external beam radiotherapy in the UK are around 40 to 50.4 Gy in 1.8 to 2 Gy per fraction. LDR and MDR dose ranges are 25 to 30 Gy and 20 to 22 Gy, respectively, in one or two insertions. For HDR brachytherapy, doses range from 14 Gy in two fractions to 24 Gy in four fractions. There are no published UK guidelines regarding doses for HDR brachytherapy in cervical cancer, but there are guidelines written by the American Brachytherapy Society (Nag et al., 2000): they recommend treating point A to at least a total LDR equivalent of 80 to 85 Gy for early stage disease and 85 to 90 Gy for advanced-stage disease (bearing in mind that, as the dose rate increases, it is necessary to reduce the total radiation dose and fractionate to avoid unacceptable late tissue complications). Some dose and fractionation schemes for combining external beam with HDR brachytherapy are suggested, but the authors advise that these schemes have not been thoroughly tested. Doses in Europe also vary depending on
Minutes
stage of disease, dose rate, and institution; for a full discussion of techniques see Gerbaulet et al. (2002). Some centres use a pelvic-side-wall boost to increase the dose to the parametria if the tumour is stage IIB or greater or if the pelvic lymph nodes are enlarged. The following description outlines the salient points of a typical radical chemoradiotherapy treatment.
External beam radiotherapy technique The patient lies supine on a couch, head on a low headrest, with arms by sides and knees supported by a polystyrene wedge. The bladder should be comfortably full. Orthogonal plain X-ray films are taken, A–P and lateral, with localisation marks (tattoos anterior and two lateral) and rulers. A barium-soaked tampon or vaginal ‘stock’ should be placed in the vagina. Alternatively a planning CT scan is undertaken with localisation markers. The clinical target volume includes the tumour and uterus with a margin for microscopic spread and regional lymph nodes, upper vagina and other potential sites of spread such as the parametria. A PTV may be grown individually around the CTV or using field borders marked as follows. It is important to remember that the field border is approximately 4 to 5 mm outside of the 95% isodose: r Superior border – between L5 and S1 vertebrae. Alternatively, to cover the common iliac lymph nodes, between L4 and L5 vertebrae or CT planned to the level of the aortic bifurcation. This larger volume will result in more bowel toxicity. r Inferior border – 2 cm below the lower extent of the clinical tumour (no higher than the lower border of the obturator foramina). r Lateral borders – 1.5 to 2 cm outside the bony pelvic side wall. 283
Louise Hanna and Malcolm Adams
r Posterior border – lower margin of the S2 vertebral body, approximately 2 cm in front of the sacral hollow or 2 cm behind the tumour on a CT planning scan. r Anterior border – through the symphysis pubis or 2 cm in front of the tumour on the CT planning scan. An outline of the patient is taken, and a plan is produced by the Physics Department. A four-field plan covers the volume adequately, with an anterior, a posterior and two lateral fields and equal weighting from each beam. Field sizes are typically 12 to 15 cm (S–I) × 14 to 17 cm for the anterior and posterior fields, and 12 to 15 cm (S–I) × 10 to 12 cm for the lateral fields. Some units use wedged lateral fields and a lower weighting from the posterior field to reduce the radiation dose to the rectum. Portal beam imaging is taken for verification on the first 3 days of treatment and weekly thereafter. The dose is prescribed to the ICRU reference point (intersection of beams) using 6 to 10 MV photons. Each field is treated daily, 5 days a week. Carcinoma of the cervix is placed in category 1, which means any unscheduled breaks in treatment should be corrected for (Royal College of Radiologists, 2002). Typical doses for external beam radiotherapy (to be followed by appropriate brachytherapy) are as follow: r 40 Gy in 20 fractions over 4 weeks. r 45 Gy in 25 fractions over 5 weeks. r 50.4 Gy in 28 fractions over 5.5 weeks.
Concurrent chemotherapy Cisplatin is given weekly at a dose of 40 mg/m2 during the external beam treatment only (typically five treatments), with pre- and posthydration. Chemotherapy is given approximately 1 hour before radiotherapy. Chemotherapy should not be given to patients with poor renal function or poor performance status.
Brachytherapy Either a general anaesthetic or appropriate regional or local anaesthesia is required. A prior anaesthetic assessment reduces the risk of cancellation. HDR treatments may be done as a day case and treatment is given in a specialised brachytherapy suite, which allows remote monitoring of the patient. Patients undergoing LDR treatments need to be admitted and nursed in a specialised room. Modern afterloading techniques allow adequate radiation protection of staff. With the patient in the lithotomy position, a urinary catheter with 7 ml radio-opaque contrast is inserted, and the bladder is emptied. Examination under anaesthetic is performed to assess the tumour, and then a uter284
balloon 7 cm3
x
bladder reference point intrauterine sources
intravaginal sources
vaginal posterior wall 0.5 cm rectal reference point
Figure 24.1. Reference points for bladder and rectum during gynaecological brachytherapy. Redrawn with permission from International Commission on Radiological Units and Measurements (1985).
ine sound is passed to measure the uterine cavity length before dilating the cervix and inserting the intrauterine tube. Two ovoids are placed in the lateral vaginal fornices. Vaginal packing or a rectal shield is placed and the applicators are secured in place. Orthogonal films are taken. The classical Manchester system of dosimetry specifies a line source inserted into the uterus with two ovoids placed in the lateral vaginal fornices. This procedure creates a pear-shaped volume around the cervix, uterus, upper vagina and immediate parametria. The dose is prescribed to Manchester point A, defined as 2 cm above the lateral vaginal fornices and 2 cm lateral to the central uterine tube. Point B is 5 cm lateral to the midline. The ICRU 38 bladder point is the posterior surface of the bladder balloon, and the rectal point is 5 mm behind the posterior vaginal wall at the level of the lower end of the intrauterine source, as shown in Figure 24.1. The dose is 21 Gy given in three fractions with HDR brachytherapy (see aforementioned discussion of other doses and dose rates). The rectal dose should be less than two-thirds of the dose to point A. Figure 24.2 shows radiographs from a brachytherapy treatment for carcinoma of the cervix.
Explanation to patient Explain the procedure and side effects to the patient. Give written information and allow the patient sufficient time to make a decision. Obtain informed consent and refer the patient to a specialist nurse counsellor.
Cervix
Toxicity of chemoradiotherapy For a comprehensive review of radiotherapy toxicity, see Faithfull and Wells (2003). Important toxicities resulting from chemoradiotherapy for cervical carcinoma are shown in Table 24.6. The national audit of the management and outcome of carcinoma of the cervix treated with radiotherapy in 1993 found the crude rate of late severe complications to be 6.1% at 5 years (Denton et al., 2000).
Para-aortic irradiation
(a)
Occasionally patients with small-volume para-aortic nodes can achieve long-term survival if para-aortic nodal irradiation is added to radical pelvic radiotherapy. Cunningham et al. (1991) reported a series of patients with early stage cervical carcinoma, who were treated with radical hysterectomy, and who had histologically proven para-aortic lymph node metastases. Of the 21 patients treated with extended field radiotherapy, 48% survived for more than 5 years. The patient is treated in the supine position, with the arms by the sides. Anterior and posterior opposed fields are used to avoid the kidneys. The radiation fields are chosen to encompass the extent of disease, usually 8 cm wide; the upper border is the T12 to L1 junction, and the lower border is chosen to match the pelvic field with a suitable gap (e.g. midway down the L5 vertebra: discuss with the Physics Department). A dose of 45 Gy in 25 fractions over 5 weeks prescribed to the midplane with 6 to 10 MV photons may be used, but the dose should be reduced if the treatment intent is palliation.
Postoperative radiotherapy and chemoradiotherapy
(b) Figure 24.2. Radiographs from an intracavitary brachytherapy treatment for carcinoma of the cervix: (a) Anterior view. (b) Lateral view. Note the contrast-filled balloon of the urinary catheter, the intrauterine tube and the lateral tubes (the ovoids can be seen faintly on the anterior view). The rod that is posterior to the intrauterine tube and most clearly seen on the lateral view is the stem for a rectal retractor (the retractor itself is not visible). There is a dosimeter in the rectum, which is again seen most clearly on the lateral view. Note also the left ureteric stent.
Stages IA2 and IB1 patients treated with surgery who are subsequently found to have high-risk factors (involved lymph nodes, involved surgical resection margin, or disease extending outside the cervix) are treated with postoperative chemoradiotherapy (or radiotherapy alone for medically unfit patients). External beam radiotherapy is given, followed by brachytherapy with a vaginal obturator. The pelvis can be adequately treated by four fields: an anterior, a posterior and two lateral fields. The patient is treated in the supine position. The target volume is the operative bed and the regional lymph nodes and field arrangement is the same as that for definitive radiotherapy. Typical doses for external beam radiotherapy (to be followed by appropriate brachytherapy treatment) are as follow: r 40 Gy in 20 fractions over 4 weeks. 285
Louise Hanna and Malcolm Adams
Table 24.6. Toxicities from chemoradiotherapy for cervical carcinoma Effect
Management
Acute effects Tiredness
Advice about fatigue, goal setting; treat anxiety and depression;
Anaemia
Red cell transfusion to keep Hb ≥ 12 g/dl
Myelosuppression/neutropenic sepsis
Patient education, weekly review and FBC, prompt assessment
Diarrhoea
Loperamide, low-residue diet
appropriate rest; moderate exercise if able
and treatment of fever Proctitis
Rectal steroids or sucralfate
Cystitis
Exclude infection, maintain fluid intake, give potassium citrate
Skin reaction
Aqueous cream or 1% hydrocortisone cream if severe
Nausea
Antiemetics (e.g. phenothiazines, metoclopramide, steroids,
Peripheral neuropathy/renal impairment
Clinical monitoring; modify cisplatin dose as necessary
Hypocalcaemia/hypomagnesaemia
Supplementation
mixture 10 ml three times daily
5-HT3 antagonists, benzodiazepines)
Late effects Bowel stricture (may progress to subacute or acute
Stool softeners, dietary advice
bowel obstruction) Bowel perforation
Emergency assessment, laparotomy if indicated
Rectal bleeding
Replace blood loss, proctoscopy, colonoscopy, surgery if
Haemorrhagic cystitis
Urine microscopy, renal function, cystoscopy, site-specific
Reduced bladder capacity, detrusor instability
Pelvic floor exercises, antimuscarinics, augmentation cystoplasty
Vaginal shortening and narrowing
Lubricants, vaginal dilators
indicated, treatment directed to source of bleeding treatment for bleeding
Menopause
Hormone replacement therapy
Infertility
Adoption, surrogacy
r 45 Gy in 25 fractions over 5 weeks. r 50.4 Gy in 28 fractions over 5.5 weeks.
Chemotherapy
Weekly cisplatin-based chemotherapy is given as discussed earlier, during the external beam treatment. Brachytherapy is given with a 2 to 3 cm diameter vaginal cylinder. HDR prescriptions at 0.5 cm depth include 11 Gy in two fractions, 12 Gy in three fractions or 15 Gy in five fractions. An LDR prescription at 0.5 cm depth is 20 Gy. An alternative technique is to use vaginal ovoids.
Responses to palliative chemotherapy are generally disappointing but agents such as cisplatin (alone or in combination with 5-fluorouracil or methotrexate and bleomycin), carboplatin or paclitaxel may produce useful responses in about 30% of patients.
Palliative radiotherapy
Cervical cancer in pregnancy
Medically unfit patients not suitable for radical radiotherapy may benefit from palliative external beam treatment to the pelvis (e.g. 8 Gy in a single fraction or 20 to 30 Gy, fractionated). Palliative radiotherapy is also beneficial for patients with painful para-aortic nodes or bony metastases.
Treatment of pregnant patients depends on the gestational age. In the first and second trimesters it is usual to recommend termination of the pregnancy followed by treatment appropriate to the stage of cancer. In the third trimester, it is more common to wait until the foetus is viable before delivery and definitive
286
Special clinical problems
Cervix
treatment. If the cancer is at a relatively early stage, a caesarean section can be combined with a radical hysterectomy (Hopkins and Morley, 1992; Monk and Montz, 1992).
Table 24.7. Survival for carcinoma of the cervix FIGO stage
Approximate 5-year survival
IA
100%
Small-cell carcinoma of the cervix
IB
70–90%
Small-cell carcinoma of the cervix is a rare and aggressive tumour with a poor prognosis. Information on treatment and prognosis comes mainly from reported case series. In general terms, most patients receive chemotherapy similar to that prescribed for small-cell lung cancer, with the addition of radical pelvic radiotherapy in those with apparently localised disease. For the few patients who appear to present with small-volume early stage disease, surgery may play a role. Delaloge et al. (2000) reported ten patients with small-cell carcinoma of the cervix. Treatment was individualised and consisted of surgery, radiotherapy, chemotherapy or a combination of modalities. Chemotherapy included cisplatin and etoposide. Four of the ten patients had metastases at presentation or developed metastases during initial treatment. Eight out of ten patients relapsed within 16 months and died of their disease within 29 months. Two patients were alive and without disease at 13 and 53 months. The incidence of brain metastases differs among reported series but would appear to be less than that seen for small-cell carcinoma of the lung.
II
45–80%
III
30–65%
IV
5–20%
Prognosis Prognostic factors Several prognostic factors have been identified for cervical cancer (International Union Against Cancer, 1995). Adverse tumour-related prognostic factors in cervical cancer include increased tumour bulk, the presence of lymph node metastases, lymphovascular space invasion, increased cancer stage and adenocarcinoma. Patient-related factors include anaemia and poor performance status. Adverse treatment-related factors include a positive surgical resection margin, a long radiation treatment and no intracavitary brachytherapy treatment.
Five-year survival Combination chemoradiotherapy is a relatively recent development and so most reported series of 5-year survival relate to radiotherapy treatment alone or to surgery for early stage disease. Approximate figures are shown in Table 24.7.
Selected patients with local recurrence who are treated with pelvic exenteration may achieve a 50% 5-year survival rate.
Areas of current interest Radiation dose and dose rate for radical radiotherapy and chemoradiotherapy are current topics of interest. Hypoxia is a common occurrence in tumours and can lead to resistance to ionising radiation. Anaemia can contribute to hypoxia, and anaemia commonly occurs in cancer patients either due to the cancer itself or because of treatment. Evidence is emerging that correction of anaemia may enhance the radiosensitivity and chemosensitivity of solid tumours, and the optimal haemoglobin level in patients with gynaecological malignancy has been estimated at between 12 and 14 g/dl (Vaupel et al., 2002). Administration of recombinant human erythropoietin may enhance the effectiveness of radiotherapy and chemotherapy by increasing haemoglobin levels in cancer patients with malignancy (Harrison and Blackwell, 2004). Intensity-modulated radiation therapy (IMRT) is a novel approach to treatment planning, which conforms the prescription dose to the shape of the target in three dimensions. This targeted approach leads to the possibilities of normal tissue sparing or dose escalation to the tumour. Ongoing clinical studies of IMRT are looking at tumour control and toxicity (Salama et al., 2004). Modern brachytherapy techniques and advances in imaging and dosimetry may also allow greater specification and optimisation of doses to the tumour and critical structures (Kirisits et al., 2005). The prospect of cervical cancer prevention is an exciting development, heralded by the development of vaccines. A recent report of a vaccine against HPV types 6, 11, 16 and 18 (responsible for 70% of cervical cancers and 90% of genital warts) in 277 women has shown a 90% reduction in the incidence of persistent 287
Louise Hanna and Malcolm Adams
infection (Villa et al., 2005). A further study of another vaccine against HPV types 16 and 18 combined with an AS04 adjuvant has shown 100% efficacy against persistent HPV infection in 1113 women (Harper et al., 2004). To be effective, vaccination needs to be given in adolescence, before persistent HPV infection. Vaccination will initially have the effect of reducing the incidence of premalignant changes followed by a long-term effect on the incidence of invasive carcinoma. As with other cancers, there is current interest in the use of monoclonal antibodies alone or in combination with chemotherapy and radiotherapy as a treatment for cervical cancer.
Ongoing clinical trials At the time of writing there were three open trials registered with the National Cancer Research Network looking at established cervical cancer (National Cancer Research Network; www.ncrn.org.uk, accessed September 2006). CxII is a phase II study of weekly neoadjuvant chemotherapy followed by radical chemoradiation for locally advanced cervical cancer. SCOTCERV is a phase II study of docetaxel and gemcitabine as second-line chemotherapy in cervical cancer. CTCR-CE 01 studies dynamic contrast-enhanced MRI in combination with tumour molecular profiling as predictors of radiation response in cervical cancer.
REFERENCES Benedet, J. L., Bender, H., Jones, H. 3rd et al. (2000). FIGO staging classifications and clinical practice guidelines in the management of gynaecologic cancers. Int. J. Gynecol. Obstet., 70, 209–62. Boon, M. E., de Graaff Guilloud, J. C. et al. (1987). Efficacy of screening for cervical squamous and adenocarcinoma: the Dutch experience. Cancer, 59, 862–6. Buckley, S. L., Tritz, D. M., Van Le, L. et al. (1996). Lymph node metastases and prognosis in patients with stage IA2 cervical cancer. Gynecol. Oncol., 63, 4–9. Cunningham, M. J., Dunton, C. J., Corn, B. et al. (1991). Extended-field radiation therapy in early-stage cervical carcinoma: survival and complications. Gynecol. Oncol., 43, 51–4. Delaloge, S., Pautier, P., Kerbrat, P. et al. (2000). Neuroendocrine small cell carcinoma of the uterine cervix: what disease? What treatment? Report of ten cases and a review of the literature. Clin. Oncol. (R. Coll. Radiol.), 12, 357–62. Denton, A. S., Bond, S. J., Matthews, S. et al. (2000). National audit of the management and outcome of carcinoma of the cervix
288
treated with radiotherapy in 1993. Clin. Oncol. (R. Coll. Radiol.), 12, 347–53. Faithfull, S. and Wells, M. (2003). Supportive Care in Radiotherapy. London: Churchill Livingstone. Ferlay, J., Bray, F., Pisani, P. et al. (2004). GLOBOCAN 2002. Cancer Incidence, Mortality and Prevalence Worldwide. IARC CancerBase No. 5, Version 2.0. Lyon: IARC Press. Gerbaulet, A., Potter, R. and Haie-Meder, C. (2002). Cervix cancer. In The GEC ESTRO handbook of brachytherapy, ed. A. Gerbaulet, R. Potter, J.-J. Mazeron et al. Brussels: ESTRO, pp. 301–64. Green, J. A., Kirwan, J. M., Tierney, J. F. et al. (2001). Survival and recurrence after concomitant chemotherapy and radiotherapy for cancer of the uterine cervix: a systematic review and meta-analysis. Lancet, 358, 781–6. Harper, D. M., Franco, E. L., Wheeler, C. et al. (2004). Efficacy of a bivalent L1 virus-like particle vaccine in prevention of infection with human papillomavirus types 16 and 18 in young women: a randomised controlled trial. Lancet, 364, 1757–65. Harrison, L. and Blackwell, K. (2004). Hypoxia and anaemia: factors in decreased sensitivity to radiation therapy and chemotherapy? Oncologist, 9 (suppl. 5), 31–40. Helmerhorst, T. J. and Meijer, C. J. (2002). Cervical cancer should be considered as a rare complication of oncogenic HPV infection rather than a STD. Int. J. Gynecol. Cancer, 12, 235–6. Hopkins, M. P. and Morley, G. W. (1992). The prognosis and management of cervical cancer associated with pregnancy. Obstet. Gynecol., 80, 9–13. International Commission on Radiological Units and Measurements. (1985). Dose and Volume Specification for Intra-cavitary Therapy in Gynaecology. Report no. 38. Bethesda, MD: International Commission on Radiological Units and Measurements. International Union Against Cancer. (1995). Prognostic Factors in Cancer, ed. P. Hermanek, M. K. Gospodarowicz, D. E. Henson et al. Berlin: Springer-Verlag. Keys, H. M., Bundy, B. N. and Stehman, F. B. (1999). Cisplatin, radiation and adjuvant hysterectomy compared with radiation and adjuvant hysterectomy for bulky stage 1B cervical carcinoma. N. Engl. J. Med., 340, 1154–61. Kirisits, C., Potter, R., Lang, S. et al. (2005). Dose and volume parameters for MRI-based treatment planning in intracavitary brachytherapy for cervical cancer. Int. J. Radiat. Oncol. Biol. Phys., 62, 901–11. Lukka, H., Hirte, H., Fyles, A. et al. (2002). Concurrent cisplatin-based chemotherapy plus radiotherapy for cervical cancer – a meta-analysis. Clin. Oncol. (R. Coll. Radiolol.), 14, 203–12. Monk, B. J. and Montz, F. J. (1992). Invasive cervical cancer complicating intrauterine pregnancy: treatment with radical hysterectomy. Obstet. Gynecol., 80, 199–203. Moreno, V., Bosch, F. X., Munoz, N. et al. (2002). Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet, 359, 1085–92. Morris, M., Eifel, P., Lu, J. et al. (1999). Pelvic radiation with concurrent chemotherapy compared with pelvic and para-aortic radiation for high-risk cervical cancer. N. Engl. J. Med., 340, 1137–43.
Cervix
Nag, S., Erickson, B., Thomadsen, B. et al. (2000). The American Brachytherapy Society recommendations for high-dose-rate brachytherapy for carcinoma of the cervix. Int. J. Radiat. Oncol. Biol. Phys., 48, 201–11. NHS Cancer Screening Programme. (2004). Colposcopy and Programme Management for the NHS Cervical Screening Programme. NHSCSP publication no. 20. Sheffield: NHS Cancer Screening Programmes. NICE. (2003). Guidance on the Use of Liquid-Based Cytology for Cervical Screening, Technology Appraisal no. 69. London: National Institute for Clinical Excellence. NICE. (2006). High Dose Rate Brachytherapy for Carcinoma of the Cervix. Interventional Procedure Guidance 160. London: National Institute for Health and Clinical Excellence. Pearcey, R., Brundage, M., Drouin, P. et al. (2002). Phase III trial comparing radical radiotherapy with and without cisplatin chemotherapy in patients with advanced squamous cell cancer of the cervix. J. Clin. Oncol., 20, 966–72. Peters, W. A. 3rd, Liu, P. Y., Barrett, R. J. 2nd et al. (2000). Concurrent chemotherapy and pelvic radiation therapy compared with pelvic radiation therapy alone as adjuvant therapy after radical surgery in high-risk early-stage cancer of the cervix. J. Clin. Oncol., 18, 1606–13. Plante, M., Renaud, M. C., Hoskins, I. A. et al. (2005). Vaginal radical trachelectomy: a valuable fertility-preserving option in the management of early-stage cervical cancer. A series of 50 pregnancies and review of the literature. Gynecol. Oncol., 98, 3–10.
Rose, P. G., Bundy, B. N., Watkins, E. B. et al. (1999). Concurrent cisplatin-based radiotherapy and chemotherapy for locally advanced cervical cancer. N. Engl. J. Med., 340, 1144–53. Royal College of Radiologists. (2002). Guidelines for the Management of the Unscheduled Interruption or Prolongation of a Radical Course of Radiotherapy. London: The Royal College of Radiologists. Salama, J. K., Roeske, J. C., Mehta, N. et al. (2004). Intensity-modulated radiation therapy in gynecologic malignancies. Curr. Treat. Options Oncol., 5, 97–108. Sevin, B. U., Nadji, M., Averette, H. E. et al. (1992). Microinvasive carcinoma of the cervix. Cancer, 70, 2121–8. Smith, J. S., Herrero, R., Bosetti, C. et al. (2002). Herpes simplex virus-2 as a human papillomavirus cofactor in the etiology of invasive cervical cancer. J. Natl. Cancer Inst., 94, 1604–13. Vaupel, P., Thews, O., Mayer, A. et al. (2002). Oxygenation status of gynaecologic tumors: what is the optimal haemoglobin level? Strahlenther. Onkol., 178, 727–31. Villa, L. L., Costa, R. L., Petta, C. A. et al. (2005). Prophylactic quadrivalent human papillomavirus (types 6, 11, 16, and 18) L1 virus-like particle vaccine in young women: a randomised double-blind placebo-controlled multicentre phase II efficacy trial. Lancet Oncol., 6, 271–8. Whitney, C. W., Sause, W., Bundy, B. N. et al. (1999). Randomised comparison of fluorouracil plus cisplatin versus hydroxyurea as an adjunct to radiation therapy in stage IIB/IVA carcinoma of the cervix with negative para-aortic lymph nodes: a Gynecologic Oncology and South West Oncology Group Study. J. Clin. Oncol., 17, 1339–48.
289
25
VAGINA Louise Hanna and Malcolm Adams
Introduction Primary carcinoma of the vagina is a rare condition that mainly affects older women. One of the main risk factors is persistent human papilloma virus (HPV) infection. Treatments are individualised, and treatment decisions are based on factors that include the site, size and stage of the tumour, and which adjacent structures are involved.
Tumours affecting the vagina The most common malignant tumours affecting the vagina are tumours that have spread from adjacent structures (cervix and vulva). Table 25.1 shows the range of tumours that can affect the vagina (adapted from WHO Classification, 2003).
mately 1 to 2% of all gynaecological malignancy. The mortality-to-incidence ratio is 0.53 (National Statistics, 2005). Incidence increases with age, with peak incidence occurring in women over 80 years of age (National Statistics, 2005).
Carcinoma of the vagina Risk factors and aetiology Risk factors for vaginal cancer include increased age, persistent infection with HPV types 16 and 18, vaginal intraepithelial neoplasia (VAIN), previous abnormal cervical cytology or cervical cancer, chronic vaginal trauma (e.g. procidentia), immunosuppression and exposure to diethylstilboestrol in utero (associated with clear cell adenocarcinoma in patients under the age of 40).
Anatomy The vagina is a muscular tube about 8 cm long, and it extends upwards and backwards from the vulva to the uterus. The apex of the vagina, into which the cervix projects, is divided into four fornices: anterior, posterior, and two lateral. The relations of the vagina from superior to inferior are as follows: r Anterior – bladder, urethra. r Posterior – pouch of Douglas, rectum, perineal body (separates lower vagina from anus). r Lateral – ureter, pelvic floor and perineal muscles. The lymphatic drainage from the upper two-thirds is to the pelvic nodes, and from the lower third to the inguinal nodes.
Pathology According to the International Federation of Gynaecology and Obstetrics (FIGO) system, any vaginal tumour involving the cervix or vulva should be classed as cervical or vulval, respectively. More than 80% of vaginal cancers are squamous carcinomas. Of the remaining tumours, the most common are adenocarcinomas (including clear cell carcinoma, which is associated with diethylstilboestrol exposure in utero), melanoma and sarcoma botryoides. For squamous carcinoma, the typical precursor lesion is VAIN, which is associated with persistent HPV infection. Most tumours occur in the upper third of the vagina. Table 25.2 shows the histological features of squamous carcinoma of the vagina.
Incidence and epidemiology Vaginal cancer is rare; the annual disease incidence in the UK is 0.7 in 100 000 women. Approximately 180 new cases are diagnosed per year in England (National Statistics, 2005), and vaginal cancer accounts for approxi290
Spread Vaginal carcinoma spreads locally to paravaginal tissues and the pelvic side wall; the rectum and anus; and the bladder, urethra and ureter. Disease spreads via
Vagina
Table 25.1. Tumours affecting the vagina Type
Examples
Benign
Squamous tumours (condyloma acuminatum, squamous papilloma, fibroepithelial polyp) ¨ Glandular tumours (mullerian papilloma, adenoma) Mesenchymal tumours (leiomyoma, genital rhabdomyoma, deep angiomyxoma, postoperative spindle cell nodule) Benign mixed tumours Melanocytic naevus and blue naevus Dermoid cyst
Malignant primary
Squamous carcinoma (keratinising, non-keratinising, basaloid, verrucous, warty) Adenocarcinoma (clear cell, endometrioid, mucinous, mesonephric) Adenosquamous carcinoma Other tumours (carcinoid, small cell, undifferentiated, adenoid cystic, adenoid squamous) Sarcomas (sarcoma botryoides, leiomyosarcoma, undifferentiated, low-grade endometrioid stromal sarcoma) ¨ tumour, adenosarcoma) Malignant mixed tumours (e.g. carcinosarcoma = malignant mixed mullerian Malignant melanoma Yolk sac tumour Ewing tumour/PNET Lymphoma and leukaemia
Malignant secondary
Direct spread (from cervix, vulva, rectum) Metastasis (from endometrium, breast, ovary)
PNET = primitive neuroectodermal tumour. Adapted from WHO classification (2003).
lymphatics to pelvic nodes (from the upper two-thirds), inguinal and femoral nodes (from the lower third) and para-aortic nodes, and via the blood to the liver, lung and brain.
Clinical presentation Primary tumours exhibit the following symptoms and signs: vaginal bleeding, vaginal discharge, pain/ dyspareunia or fistula (vesico-vaginal or recto-vaginal). Symptoms and signs from lymph node or distant spread include an inguinal lymph node mass, lymphoedema of a lower limb, pelvic pain or, uncommonly, jaundice, bone pain and shortness of breath.
presence of fixity or involvement of adjacent structures. Biopsy should be conducted, as well as cystoscopy and proctoscopy. Pregnancy status should be checked in premenopausal women.
Further investigations for staging A CT or MRI scan of the pelvis and abdomen, looking at the extent of the primary tumour, pelvic and paraaortic lymph nodes and/or hydronephrosis should be performed, as well as a chest X-ray, full blood count and biochemical profile. Other investigations, such as a bone scan, are undertaken if there is a high degree of clinical suspicion.
Staging classification Investigation and staging Diagnostic investigations Examination under anaesthetic involves inspection of the vulva, vagina and cervix. The size and location of primary tumours should be documented, including the
The regional nodes for the upper two-thirds of the vagina are the pelvic nodes. The regional nodes for the lower third of the vagina are the inguinal and femoral nodes. Table 25.3 shows the staging classification for carcinoma of the vagina. 291
Louise Hanna and Malcolm Adams
Table 25.2. Histological features of squamous vaginal
Table 25.3. Staging classification for carcinoma of the
carcinoma
vagina
Features Macroscopic
Description
TNM
FIGO
stage
stage
Description
and constricting, polypoid, sessile,
Tis
0
Carcinoma in situ
indurated or fungating
T1
I
Tumour confined to vagina
T2
II
Tumour invades paravaginal tissues
May be exophytic, ulcerative, annular
May occur anywhere in the vagina Size varies from microscopic to more
but does not extend to pelvic side
than 10 cm Microscopic
Typically moderately differentiated and non-keratinising
wall T3
III
Tumour extends to pelvic side wall
T4
IVA
Tumour invades mucosa of bladder or rectum and/or extends beyond
Variants: spindle cell, warty, verrucous
the true pelvis Adapted from WHO classification (2003).
N0
No regional lymph node metastasis
N1
Regional lymph node metastasis
M0
Treatment overview Carcinoma of the vagina is rare, and there have been few randomised trials of treatment; most of the available information about treatment and outcome is from small retrospective studies. Treatment is individualised and it depends on the location of the primary tumour within the vagina and the extent of spread. Surgical treatments, if considered, are best for tumours affecting the central pelvis, because tumours that have spread to the pelvic side wall are not resectable. Most patients are treated with radiotherapy. Approximately 60% of patients present with stage I or II tumours. The following is an overview of the most commonly used treatment options by stage.
Stage I Treatment options for stage I tumours include r Surgery ± postoperative radiotherapy. r Radiotherapy – brachytherapy alone. r Radiotherapy – combined external beam plus brachytherapy.
Stages II and III Treatment options for stage II and III tumours include r Radiotherapy – combined external beam plus brachytherapy. r Radiotherapy – external beam alone. r Surgery ± postoperative radiotherapy (e.g. for small stage II tumours). 292
M1
No distant metastasis IVB
Distant metastasis
Adapted from UICC (2002).
Stage IVa Treatment options for stage IVa tumours include: r Radiotherapy – external beam treatment alone. r Radiotherapy – combined external beam plus brachytherapy. r Surgery ± postoperative radiotherapy (e.g. for central tumours with fistulae).
Stage IVb Treatment options for stage IVb tumours include palliative radiotherapy, chemotherapy, and symptom control.
Surgery The role of surgery in invasive carcinoma of the vagina has been reviewed by Tjalma et al. (2001). Their results were based on a policy of surgery with curative intent for patients with FIGO stage I, small stage II and selected stage IV tumours (i.e. patients for whom radical surgery with pelvic exenteration was possible). Treatment was individualised, and patients received postoperative radiotherapy if there were positive resection margins or involved lymph nodes. The 5-year survival for the squamous cancers was 74%. The authors concluded that selected patients have good survival and local control if treated with surgery and selective postoperative radiotherapy.
Vagina
The surgical techniques used were local excision, radical hysterectomy and/or vaginectomy, and pelvic exenteration.
Radiotherapy Frank et al. (2005) published guidelines for radiotherapy treatment for squamous vaginal carcinoma, based on their large series of 193 patients. Salient points are as follow: r Treatment is individualised, and it depends on tumour size, site and extent of spread. r Because of the risk of microscopic nodal disease, initial external beam radiotherapy (EBRT) is recommended, except for very small stage I tumours. r After initial EBRT, the patient is re-evaluated to determine how best to give a boost to the primary tumour; options include intracavitary therapy, interstitial therapy, or further EBRT. r There is an increasing use of combined, concurrent chemoradiotherapy. Mock et al. (2003) and Nanavati et al. (1993) reported that high-dose-rate (HDR) brachytherapy appears to be well tolerated and that it is as effective as low-dose-rate (LDR).
Radical radiotherapy Radiotherapy technique: carcinoma of the vagina (phase I, external beam) The patient lies supine with a low head rest, with the knees supported and the arms across the chest. A barium-soaked tampon is placed in the vagina and a radioopaque marker is placed at the introitus. Rulers are placed, and orthogonal X-ray films are taken. Localisation tattoos are placed, one anteriorly and two laterally. Alternatively, CT planning can be used. The clinical target volume is the primary tumour (GTV) with a 1 to 2 cm margin, plus the vagina and the regional lymph nodes. For disease affecting the upper two-thirds of the vagina, typical field borders are as follows: r Superior border – L5/S1 junction. r Inferior border – 3 cm below either the primary tumour or the introitus. r Lateral borders – 1 to 2 cm outside the bony pelvic side wall. For disease affecting the lower third of the vagina, typical field borders are as follows:
r Lateral borders include the inguinal and femoral nodes. The plan includes an anterior–posterior parallelopposed pair of fields or a four-field pelvic brick, depending on which nodal groups require treatment. Dose and fractionation is 45 Gy in 25 fractions over 5 weeks with 10 MV photons, prescribed to the midplane (ICRU reference point).
Radiotherapy technique: carcinoma of the vagina (phase II, boost) Typical scenarios for a radiotherapy boost are (1) intracavitary treatment with an intrauterine tube, (2) intracavitary treatment with a vaginal tube, (3) interstitial treatment and (4) an external beam boost. (1) For intracavitary treatment with an intrauterine tube, treatment indication is a tumour at the upper third of the vagina, and intact uterus. The technique involves an intrauterine tube and ovoids, similar to the procedure for carcinoma of the cervix. An LDR schedule is 25 to 30 Gy in a single fraction to point A; the HDR schedule is typically 21 to 24 Gy in three to four fractions to point A. (2) An indication for an intracavitary treatment with a vaginal tube is a very superficial tumour in the lower two-thirds of the vagina. The technique involves placement of a vaginal cylinder. An LDR schedule is 20 to 25 Gy in a single insertion prescribed 0.5 cm from the surface of the applicator. There are very few reported series for HDR (e.g. 21 Gy in three fractions or 20 Gy in four fractions prescribed 0.5 cm from the surface of the applicator (Mock et al., 2003; Nanavati et al., 1993). (3) An indication for interstitial treatment is a tumour in the distal two-thirds of the lateral or anterior wall. The technique involves, for example, iridium hair pins or wires. An LDR schedule is 20 to 30 Gy prescribed to the 85% reference isodose. (4) An indication for use of the external beam boost is a deeply infiltrating tumour (e.g. tumours involving the recto-vaginal septum or bladder). A conformal technique with CT or MRI planning to minimise the dose to adjacent structures is used. Doses include 18 to 24 Gy in 10 to 12 fractions.
Complications of radiotherapy Chyle et al. (1996) reported a 13% serious complication rate with radiotherapy. In this series, complications included rectal ulceration, stricture, proctitis, vaginal ulceration or necrosis, small bowel obstruction, recto-vaginal fistula, urethral stricture, vesico-vaginal fistula, haemorrhagic cystitis, vesico-peritoneal fistula, 293
Louise Hanna and Malcolm Adams
Table 25.4. Prognosis for vaginal cancer by stage Stage
Five-year survival
I
48–94%
II
29–80%
III
0–67%
IV
0–44%
Adapted from Tjalma et al. (2001).
vesico-cutaneous fistula, osteonecrosis of the pubis, and fatal pulmonary embolism.
Follow-up Most locoregional recurrences and distant failures occur within the first 2 years after treatment. The following is a suggested scheme for clinical follow-up: r Three-monthly for the first year. r Four-monthly for the second year. r Six-monthly for the third and fourth years. r Then annually for 1 year.
Poor prognostic factors include tumour in the lower vagina, adenocarcinoma (non-clear cell), increased tumour bulk and increased FIGO stage (Chyle et al., 1996).
Areas of current interest Chemoradiotherapy Perez et al. (1999) commented on the high level of distant metastases in carcinoma of the vagina; they concluded that more effective treatments require the addition of systemic therapy. As with cervical and vulval cancer, there is interest in concurrent chemoradiation. Dalrymple et al. (2004) reported a series of 14 patients with predominantly early stage disease who were treated with primary chemoradiotherapy using 5fluorouracil (5-FU) alone, or 5-FU with either cisplatin or mitomycin C, and the authors concluded that this was an effective treatment for vaginal cancer. Clearly, for this rare condition, obtaining evidence for the role of chemoradiotherapy is going to be difficult. For the foreseeable future, treatment decisions are likely to be based on reported series and extrapolation from the experience in other carcinomas of the ano-genital area.
Recurrent disease The predominant site of relapse in radically treated patients is locoregional. Patients previously treated with radiotherapy, who develop a central recurrence in the pelvis, can sometimes be cured with pelvic exenteration. Patients previously treated with surgery alone, who develop a localised recurrence in the pelvis, can sometimes be cured with radical radiotherapy.
Palliative treatments Typical scenarios in patients with incurable locoregional recurrences include pelvic pain, fistula, bleeding, infection, lymphoedema, and thromboembolic disease. A multidisciplinary approach is required. Interventions that provide a benefit include platinum-based palliative chemotherapy, palliative radiotherapy, defunctioning stoma, lymphoedema stockings and adequate pain control.
Prognosis Table 25.4 shows the 5-year survival for different stages of vaginal cancer (Tjalma et al., 2001). 294
Ongoing clinical trials At the time of writing there were no clinical trials in vaginal cancer registered with the National Cancer Research Network (National Cancer Research Network; www.ncrn.org.uk, accessed September 2006).
REFERENCES Chyle, V., Zagars, G. K., Wheeler, J. A. et al. (1996). Definitive radiotherapy for carcinoma of the vagina: outcome and prognostic factors. Int. J. Radiat. Oncol. Biol. Phys., 35, 891–905. Dalrymple, J. L., Russell, A. H., Lee, S. W. et al. (2004). Chemoradiation for primary invasive squamous carcinoma of the vagina. Int. J. Gynecol. Cancer, 14, 110–7. Frank, S. J., Jhingran, A., Levenback, C. et al. (2005). Definitive radiation therapy for squamous cell carcinoma of the vagina. Int. J. Radiat. Oncol. Biol. Phys., 62, 138–47. Mock, U., Kucera, H., Fellner, C. et al. (2003). High-dose-rate (HDR) brachytherapy with or without external beam radiotherapy in the treatment of primary vaginal carcinoma: long-term results and side effects. Int. J. Radiat. Oncol. Biol. Phys., 56, 950–7. Nanavati, P. J., Fanning, J., Hilgers, R. D. et al. (1993). High-dose-rate brachytherapy in primary stage I and II vaginal cancer. Gynecol. Oncol., 51, 67–71. National Statistics (2005). In Cancer Statistics and Registrations, Series MB1 no. 34, London: Office for National Statistics.
Vagina
Perez, C. A., Grigsby, P. W., Garipagaoglu, M. et al. (1999). Factors affecting long-term outcome of irradiation in carcinoma of the vagina. Int. J. Radiat. Oncol. Biol. Phys., 44, 37–45. Tjalma, W. A. A., Monaghan, J. M., de Barros Lopes, A. et al. (2001). The role of surgery in invasive squamous carcinoma of the vagina. Gynecol. Oncol., 81, 360–5.
UICC. (2002). In TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York, Wiley-Liss, pp. 150–3. WHO classification. (2003). In World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of the Breast and Female Genital Organs, ed. A. Tavassoli and P. Devilee. Lyon: IARC Press, Chap. 6.
295
26
VULVA Louise Hanna and Malcolm Adams
Introduction Carcinoma of the vulva is an uncommon disease. Approximately 75% of patients with vulval cancer are cured, but effective management requires the expertise of a multidisciplinary team to support patients through the physical and psychosexual morbidity that is associated with radical treatment.
Range of vulval tumours Table 26.1 shows the large range of benign and malignant tumours which may affect the vulva (adapted from WHO, 2003).
age of 70. The highest incidence occurs in underdeveloped countries.
Carcinoma of the vulva Risk factors and aetiology Squamous carcinoma is the most common malignant tumour of the vulva. Risk factors include human papilloma virus (HPV), especially types 16 and 18; HPV types 6 and 11, which are associated with verrucous carcinoma; cigarette smoking; immunosuppression (e.g. renal allograft); and a history of genital warts.
Pathology
Anatomy The vulva is the name given to the female external genitalia. The anatomical subsites are as follow: r Mons pubis, the rounded hair-bearing region in front of the pubis. r Labia majora, the hair-bearing skin extending from the mons pubis. r Labia minora, the non-hair-bearing folds of skin that meet posteriorly at the fourchette. r Clitoris, situated in the midline at the anterior ends of the labia minora. r Vestibule, the triangular-shaped skin between the labia minora.
Precursor lesions for squamous carcinoma of the vulva include the following: r Vulval intraepithelial neoplasia (VIN) associated with HPV. r VIN not associated with HPV. r Lichen sclerosus (6% risk of cancer). r Chronic granulomatous vulval disease. Table 26.2 shows the pathological features of squamous carcinoma of the vulva. Molecular abnormalities in vulval cancer include a disruption of the PTEN and TP53 pathways. TP53 can be inactivated through binding of the HPV E6 protein.
Spread
Incidence and epidemiology The annual incidence of vulval cancer in the UK is 3.5 in 100 000 women (National Statistics, 2005). Approximately 880 cases are diagnosed in England each year. Vulval cancer makes up about 5% of all gynaecological cancers. Cancer mortality is approximately one-quarter of patients diagnosed. Disease incidence increases with age; presentation is very rare in women under the age of 30, with peak incidence occurring in women over the 296
Approximately 30% of patients with operable vulval cancers have involved inguinal lymph nodes at the time of surgery. The major route of spread is locoregional, and frequently, patients who die from their disease have no clinical evidence of distant spread. Vulval cancer can spread locally, via lymphatics or via the bloodstream. Local spread can occur to the perineum, urethra, vagina, anus, bladder, rectum or pubic bone. Spread can occur via lymph nodes to inguinal nodes, femoral nodes
Vulva
Table 26.1. The range of vulval tumours Type
Examples
Benign
Squamous tumours (e.g. condylomata acuminata, keratoacanthoma) Glandular tumours (e.g. adenoma, Paget’s disease [n.b. 10–20% invasive]); arising from anogenital mammary-like glands (e.g. papillary hydradenoma) Soft tissue tumours (e.g. leiomyoma, granular cell tumour) Melanocytic naevus
Malignant primary
Squamous carcinoma, 85%; various types including keratinising, non-keratinising, verrucous Basal cell carcinoma Adenocarcinoma (e.g. arising from Bartholin gland; anogenital mammary-like gland; skene gland – the skene gland is the female equivalent of the prostate gland) Skin appendage tumour: malignant sweat gland tumour; sebaceous carcinoma Malignant soft tissue tumour (e.g. sarcoma botryoides, leiomyosarcoma) Malignant melanoma Lymphoma and leukaemia Melanoma
Malignant secondary
Direct spread from pelvic organs Metastasis from distant primary
Adapted from WHO (2003).
and pelvic nodes. Spread can occasionally occur via the blood stream to the liver, lung and bone.
Clinical presentation The predominant symptoms of carcinoma of the vulva are those of locoregional spread. If metastases become clinically apparent, it is usually late in the course of the disease. Symptoms and signs from the primary tumour include an ulcer or lump, discharge, bleeding, pain, odour or pruritus vulvae. Symptoms and signs from lymph node spread include an inguinal mass, lymphoedema of a lower limb or pelvic pain from lymphadenopathy. Symptoms and signs from metastatic disease are rare, but include jaundice, bone pain or pleural effusion.
Investigation and staging Diagnostic confirmation of disease involves examination under anaesthetic, inspection of the size and location of the primary and documentation of urethral, vaginal or anal involvement or fixity. This is combined with biopsy, cystoscopy and proctoscopy, a cervical smear if
one has not already been done, and FNA of clinically involved lymph nodes. Staging investigations include the following: r CT or MRI of the pelvis and abdomen, looking for pelvic and para-aortic lymph nodes or hydronephrosis. Large primary tumours may also be visible. r Chest X-ray. r Full blood count. r Biochemical profile. r Other investigations, such as a bone scan, are undertaken if there is a high clinical index of suspicion. Pregnancy status should be checked in premenopausal women.
Staging classification The regional nodes are the inguinal and femoral nodes. Tables 26.3 and 26.4 show the TNM classification and FIGO staging groups, respectively.
Treatment overview Vulval cancer should be managed by a multidisciplinary team that has sufficient surgical expertise to undertake radical procedures and the appropriate resources to 297
Louise Hanna and Malcolm Adams
Table 26.2. Pathological features of squamous carcinoma of the vulva Features
Description
Macroscopic
Ulcer, nodule, macule or pedunculated mass
Microscopic
Table 26.3. TNM classification of vulval cancer Stage
Description
T1
Tumour confined to vulva or vulva and
T1a
Tumour confined to vulva or vulva and
perineum, 2 cm or less in greatest dimension
Invasive neoplasm composed of
perineum, 2 cm or less in greatest dimension
malignant squamous cells of
and with stromal invasion of no greater than 1.0 mm
several morphological variants:
r Keratinising with squamous pearls r Non-keratinising r Basaloid with nests of immature basal
T1b
perineum, 2 cm or less in greatest dimension and with stromal invasion of greater than 1.0 mm
type cells
r Warty
(condylomatous) with cellular
T2
features of HPV
dimension T3
Tumour invades any of the following: lower
T4
Tumour invades any of the following: bladder
inflammatory infiltrate
r Kerato-acanthoma-like Variant with tumour giant cells is highly
urethra, vagina, anus
mucosa, rectal mucosa, upper urethral
aggressive, may resemble malignant
mucosa or is fixed to pubic bone
melanoma Adapted from WHO (2003).
manage the psychosexual impact of major surgery. Vulval cancer is an uncommon disease, and little evidence is available from randomised trials. Treatment decisions are frequently made on the basis of cohort data or phase II studies.
Stage IA Stage IA tumours require only local excision because of the negligible risk of lymph node metastases in tumours with 1 mm or smaller depth of invasion.
Tumour confined to vulva or vulva and perineum, more than 2 cm in greatest
r Verrucous, highly differentiated with rare mitotic figures and prominent
Tumour confined to vulva or vulva and
N0
No regional lymph node metastasis
N1
Unilateral regional lymph node metastasis
N2
Bilateral regional lymph node metastasis
M0
No distant metastasis
M1
Distant metastasis including pelvic lymph node metastasis
Adapted from UICC (2002).
locally advanced disease may benefit from preoperative or primary radiotherapy with or without concurrent chemotherapy.
Stage IVA Stages IB to II Stage IB to stage II tumours should be treated with radical vulvectomy and bilateral groin node dissection (BGND) because of the risk of lymph node metastases. Exceptions are very-well-lateralised tumours, which may be treated with modified radical vulvectomy and unilateral groin node dissection (Stehman et al., 1992).
Options for treatment of stage IVA tumours include radical vulvectomy and pelvic exenteration, or preoperative or primary radiotherapy with or without concurrent chemotherapy.
Stage IVB Treatment of stage IVB tumours is palliative, aimed at controlling symptoms.
Stage III Treatment for stage III tumours is individualised for this heterogeneous group of patients. Some patients will be suitable for radical vulvectomy and BGND, with or without postoperative radiotherapy. Patients with very 298
Surgery The standard treatment for vulval cancer is surgery. Treatment is individualised and aims to limit the
Vulva
Table 26.4. FIGO stage groupings of vulval cancer Stage
Description
0
Carcinoma in situ, intraepithelial neoplasia grade III
I
Tumours ≤ 2 cm confined to the vulva or
IA
Tumours ≤ 2 cm confined to the vulva or
perineum, no nodal metastasis perineum and with stromal invasion ≤ 1.0 mm, no nodal metastasis IB
Tumours ≤ 2 cm confined to the vulva or perineum and with stromal invasion > 1.0 mm, no nodal metastasis
II
Tumour confined to the vulva and/or perineum; > 2 cm in greatest dimension; no nodal metastasis
III
Tumour of any size with adjacent spread to the lower urethra and/or the vagina, or the
use of separate groin incisions in a ‘triple incision’ operation (Hacker et al. 1981) is associated with a shorter hospital stay and less blood loss, and nowadays is more common. A margin of 2 to 3 cm should be taken around the macroscopic disease. Complications include wound disruption and infection, haemorrhage, leg oedema, femoral nerve damage and altered sexual function or body image.
Radical vulvectomy, bilateral groin node dissection and pelvic exenteration This very radical procedure involves the formation of one or two stomas. Rigorous preoperative work-up and counselling are required to ensure that patients are likely to tolerate the procedure physically and psychologically. As a result of data from phase II clinical trials, patients are more often initially treated with radiotherapy or chemotherapy in the hope of reducing the extent of surgery required.
anus and/or unilateral regional lymph node metastasis IVA
Tumour invades any of the following: upper urethra, bladder mucosa, rectal mucosa, pelvic bone and/or bilateral regional node metastases
IVB
Any distant metastasis, including pelvic lymph nodes
Adapted from Benedet et al. (2000).
physical and psychological morbidity of major surgery, without compromising the chance of cure.
Local excision Local excision is suitable for T1a tumours. The remaining vulva must be normal and surgical margins should be 1 cm or greater.
Hemivulvectomy and unilateral groin node dissection Hemivulvectomy and unilateral groin node dissection are suitable for well-lateralised T1b tumours with no lymphovascular space invasion.
Radical vulvectomy and bilateral groin node dissection Radical vulvectomy and BGND are classically performed via a butterfly incision with en bloc removal of the groin nodes (inguinal and femoral). However, the
Radiotherapy and chemoradiotherapy Postoperative radiotherapy to the nodes Homesley et al. (1986) performed a randomised trial comparing postoperative radiotherapy with pelvic node dissection for patients who had undergone radical vulvectomy and BGND. There was a significant survival benefit in favour of radiotherapy, especially in patients with two or more positive groin nodes. In patients with unilateral nodal disease, whether to irradiate both sides postoperatively is a matter of debate. A radiotherapy technique for postoperative radiotherapy to the groin and pelvic nodes is as follows. The patient lies supine on the couch with arms by the sides, knees supported by a polystyrene wedge, without further immobilisation. An A–P X-ray film is taken or, alternatively, CT planning is used. The clinical target volume includes the inguinal and pelvic nodes. The field borders are as follow: r Superior border – mid-S–I joints. r Inferior border – to cover the lower border of the obturator foramina. r Lateral borders – to cover the groin nodes (approximately through the long axis of the femoral shaft). This radiotherapy technique uses parallel-opposed anterior and posterior fields, with a dose fractionation schedule of 45 Gy in 25 fractions over 5 weeks to the midplane (ICRU reference point) using 6 to 10 MV photons. Portal beam imaging is taken on the first 3 days of treatment and weekly thereafter for verification. 299
Louise Hanna and Malcolm Adams
Postoperative radiotherapy to the primary site The risk of local recurrence has been reported as zero for a resection margin of 8 mm or larger and 48% for a resection margin of less than 8 mm (Heaps et al., 1990). Faul et al. (1997) reported a non-randomised study of adjuvant radiotherapy for vulval cancer. Patients with resection margins of less than 8 mm who had adjuvant radiotherapy had a local recurrence rate of only 16% compared with 58% of patients who did not have adjuvant radiotherapy. Patients with a resection margin of 8 mm or smaller, who are unsuitable for further surgery, are likely to benefit from postoperative radiotherapy to a dose of 45 to 50 Gy.
Preoperative radiotherapy or chemoradiotherapy Primary tumours that involve the anus, rectum, rectovaginal septum, proximal urethra or bladder are uncommon but pose a particular problem because primary surgical clearance requires exenteration and radical vulvectomy, which may be associated with significant mortality and morbidity. Not all of the phase II studies of radiotherapy alone (e.g. Hacker et al., 1984) or more recently of concurrent chemoradiotherapy (Landoni et al., 1996; Lupi et al., 1996; Moore et al., 1998; Scheistroen and Trope, 1993) have shown beneficial responses for locally advanced vulval cancers. Overall, there is no proven survival benefit, and the potential exists for severe morbidity in these patients. Therefore, preoperative chemoradiotherapy should be performed only in the context of a clinical trial.
Primary radiotherapy or chemoradiotherapy for inoperable vulval cancer Data from small phase II studies suggest that chemoradiation should be considered as a therapeutic option in patients who are not suitable for surgery because of the extent of disease or co-existing medical disorders (e.g. Cunningham et al., 1997; Koh et al., 1993; Wahlen et al., 1995). However, caution is needed when dealing with frail, often elderly, patients in view of the potential toxicity of such treatment. Treatment is given in two phases, with the first delivering 45 Gy in 25 fractions to the primary tumour plus nodes. Consideration should then be given to surgical removal of residual disease if possible, or to a second phase of radiotherapy with electrons or brachytherapy to a total dose of 60 to 65 Gy in 1.8 to 2.0 Gy fractions (Board of the Faculty of Clinical Oncology, Royal College of Radiologists, 2006). 300
A technique for primary radiotherapy for advanced vulval cancer is as follows. The patient lies supine on the couch with arms by sides and knees supported by a polystyrene wedge. For phase I localisation and target volume definition, an A–P X-ray film is taken. The clinical target volume includes the primary tumour with a 2 cm margin, together with the groin and pelvic nodes. Typical field borders are as follows: r Superior border – sacral promontory. r Inferior border – 2 to 3 cm below the vulva. r Lateral borders – to cover groin nodes (approximately through the long axis of the femoral shaft). For phase II, the clinical target volume includes areas of known gross disease with a 2 cm margin. The radiotherapy technique, for phase I, involves parallel-opposed anterior and posterior fields, with bolus to sites of disease that are infiltrating skin. For phase II, if electrons are used, the energy is determined by the depth of the tumour, with bolus as necessary. The dose, energy and fractionation for phase I is 45 Gy in 25 fractions to the midplane (ICRU reference point) using 6 to 10 MV photons. For phase II, give a boost of, for example, 18 Gy in 10 fractions with electrons to areas of known disease (or surgery/brachytherapy). Portal beam images are taken on the first 3 days of treatment and weekly thereafter for verification. Concurrent chemotherapy involves giving cisplatin 50 mg/m2 days 1 and 29, and fluorouracil 1000 mg/m2 every 24 hours for days 1 to 4 and 29 to 32, or cisplatin 40 mg/m2 weekly for 5 weeks. When explaining the procedure to the patient, give written information and refer the patient to a specialist nurse counsellor. Explain the procedure plus side effects: r Acute side effects include tiredness, painful moist desquamation of the vulva, diarrhoea, cystitis, stomatitis, myelosuppression or anaemia. r Late side effects include menopause and infertility if relevant. There is a small risk of long-term bowel or bladder late radiation effects. Vaginal shortening may also occur.
Toxicity of chemoradiotherapy Chemoradiotherapy toxicity is shown in Table 26.5.
Chemotherapy Responses to palliative chemotherapy with cisplatin and 5-fluorouracil for vulval cancer are observed but
Vulva
Table 26.5. Toxicity of chemoradiotherapy for carcinoma of the vulva Toxicity
Management
Acute Anaemia
Weekly monitoring of FBC, red cell transfusion
Diarrhoea
Low-residue diet, loperamide
Skin erythema and desquamation
1% hydrocortisone cream or Intrasite® gel
Fatigue
Activity pacing, goal setting, stress management
Myelosuppression/
Weekly FBC, advice to monitor temperature daily, treat sepsis according to
Break from treatment
neutropenic sepsis
local protocol
Nausea and vomiting
Dexamethasone and metoclopramide with chemotherapy
Peripheral neuropathy/renal impairment
Clinical monitoring and cisplatin dose modification
Late Menopause
Hormone replacement therapy
Infertility
Advice re adoption, surrogacy as appropriate
Severe radiation toxicity to bowel or bladder
Surgical referral
(risk 3–10%) Vaginal shortening and narrowing
Use of vaginal dilators, lubricants
they are frequently short-lived. Typically, this combination requires inpatient admission. An outpatient regimen using a combination of bleomycin, methotrexate and CCNU was reported by Wagenaar et al. (2001) to have a 56% response rate. The progression-free survival was 4.8 months and the median survival was 7.8 months.
Recurrent disease If the disease recurs in the vulva following limited surgery, it can often be removed with a more radical excision or pelvic exenteration. Alternatively, chemoradiotherapy has been reported to give long-term disease control in 8 out of 15 patients (Thomas et al., 1989). Recurrences in the inguinal nodes following surgery may respond to chemoradiotherapy. Surgery for groin recurrences following radiotherapy has a high complication rate.
Palliative treatments Patients with recurrent and progressive vulval cancer frequently have uncontrolled locoregional disease. Pain, infection, bleeding and lymphoedema may be encountered, which require a multidisciplinary team approach.
Palliative treatments are considered in more detail in Chapter 7.
Prognosis Prognostic factors Poor prognostic factors include a tumour diameter greater than 2.5 cm; increased depth of invasion, lymphovascular space invasion and a greater extent of lymph node involvement.
Prognosis Overall mortality is approximately one-quarter of those patients diagnosed. Five-year survival rates by stage group are shown in Table 26.6 (Homesley et al., 1991; Podratz et al., 1983).
Areas of current interest Areas of current interest in vulval cancer include chemoradiation, sentinel node biopsy and VIN treatments. There is interest in whether the accurate determination of inguinal node status via sentinel node biopsy could reduce the need for radical surgery without compromising patient survival. In their review of potential techniques, Selman et al. (2005) found that sentinel 301
Louise Hanna and Malcolm Adams
Table 26.6. Five-year survival of vulval carcinoma by stage grouping Five-year FIGO stage
Survival
I
90–98%
II
81–85%
III
68–74%
Comments
Large variation within stage III depending on number of nodes involved and tumour size (e.g. three or more positive groin nodes, 5YS = 29%)
IV
20–31%
5YS = 5-year survival. Adapted from Podratz et al. (1983) and Homesley et al. (1991).
node biopsy with 99m Tc-labelled nanocolloid was the most promising test under investigation. There is also interest in the use of topical treatments for VIN, such as imiquimod, an immune stimulant, or cidofovir, an antiviral agent, and the use of photodynamic therapy.
Ongoing clinical trials At the time of writing, the following trials were registered with the National Cancer Research Network (www.ncrn.org.uk, accessed August 2006). EORTC 559985 is a phase II clinical trial that looks at paclitaxel as a single agent in locally advanced and/or metastatic or recurrent vulval cancer that is not amenable for surgery and/or radiotherapy. RT3 VIN is a randomised phase II multicentre trial of topical treatment in women with VIN.
REFERENCES Benedet, J.L., Bender, H., Jones, H. 3rd et al. (2000). FIGO staging classifications and clinical practice guidelines in the management of gynaecologic cancers. Int. J. Gyneco. Obstet., 70, 209–62. Board of the Faculty of Clinical Oncology, Royal College of Radiologists. (2006). Radiotherapy Dose-Fractionation. London: Royal College of Radiologists. Cunningham, M.J., Goyer, R.P., Gibbons, S.K. et al. (1997). Primary radiation, cisplatin and 5-fluorouracil for advanced squamous carcinoma of the vulva. Gynecol. Oncol., 66, 258–61.
302
Faul, C.M., Mirmow, D., Huang, Q. et al. (1997). Adjuvant radiation for vulvar carcinoma: improved local control. Int. J. Radiat. Oncol. Biol. Phys., 38, 381–9. Hacker, N.F., Leuchter, R.S., Berek, J.S. et al. (1981). Radical vulvectomy and bilateral inguinal lymphadenectomy through separate groin incisions. Obstet. Gynecol., 58, 574–9. Hacker, N.F., Berek, J.S., Juillard, G.J. et al. (1984). Preoperative radiation therapy for locally advanced vulvar cancer. Cancer, 54, 2056–61. Heaps, J.M., Fu, Y.S., Montz, F.J. et al. (1990). Surgical-pathologic variables predictive of local recurrence in squamous cell carcinoma of the vulva. Gynecol. Oncol., 38, 309–14. Homesley, H.D., Bundy, B.N., Sedlis, A. et al. (1986). Radiation therapy versus pelvic node resection for carcinoma of the vulva with positive groin nodes. Obstet. Gynecol. 68, 733–40. Homesley, H.D., Bundy, B.N., Sedlis, A. et al. (1991). Assessment of current International Federation of Gynecology and Obstetrics staging of vulvar carcinoma relative to prognostic factors for survival (a Gynecologic Oncology Group study). Am. J. Obstet. Gynecol., 164, 997–1003. Koh, W.J., Wallace, H.J. 3rd, Greer, B.E. et al. (1993). Combined radiotherapy and chemotherapy in the management of local-regionally advanced vulvar cancer. Int. J. Radiat. Oncol. Biol. Phys., 26, 809–16. Landoni, F., Maneo, A., Zanetta, G. et al. (1996). Concurrent preoperative chemotherapy with 5 fluorouracil and mitomycin C and radiotherapy (FUMIR) followed by limited surgery in locally advanced and recurrent vulvar carcinoma. Gynecol. Oncol., 61, 321–7. Lupi, G., Raspagliesi, F., Zucali, R. et al. (1996). Combined preoperative chemoradiotherapy followed by radical surgery in locally advanced vulvar cancer. A pilot study. Cancer, 77, 1472–8. Moore, D.H., Thomas, G.M., Montana, G.S. et al. (1998). Preoperative chemoradiation for advanced vulvar cancer: a phase II study of the Gynecologic Oncology Group. Int. Radiat. Oncol. Biol. Phys., 42, 79–85. National Statistics. (2005). In Cancer Statistics and Registrations, series MB1 no. 34, London: Office for National Statistics, p. 64. Podratz, K.C., Symmonds, R.E., Taylor, W.F. et al. (1983). Carcinoma of the vulva: analysis of treatment and survival. Obstet. Gynecol., 61, 63–74. Scheistroen, M. and Trope, K. (1993). Combined bleomycin and irradiation in preoperative treatment of advanced squamous cell carcinoma of the vulva. Acta Oncol.. 32, 657–61. Selman, T.J., Luesley, D.M., Acheson, N. et al. (2005). A systematic review of the accuracy of diagnostic tests for inguinal lymph node status in vulvar cancer. Gynecol. Oncol., 99, 206–14. Stehman, F.B., Bundy, B.N., Dvoretsky, P.M. et al. (1992). Early stage I carcinoma of the vulva treated with ipsilateral superficial inguinal lymphadenectomy and modified radical hemivulvectomy: a prospective study of the Gynecologic Oncology Group. Obstet. Gynecol., 79, 490–7. Thomas, G., Dembo, A., DePetrillo, A. et al. (1989). Concurrent radiation and chemotherapy in vulvar carcinoma. Gynecol. Oncol., 34, 263–7.
Vulva
UICC. (2002). In TNM Classification of Malignant Tumours, Ed. L.H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss. Wagenaar, H.C., Colombo, N., Vergote, I. et al. (2001). Bleomycin, methotrexate, and CCNU in locally advanced or recurrent, inoperable, squamous-cell carcinoma of the vulva: an EORTC Gynaecological Cancer Cooperative Group Study. European Organization for Research and Treatment of Cancer. Gynecol. Oncol., 81, 348–54.
Wahlen, S.A., Slater, J.D., Wagner, R.J. et al. (1995). Concurrent radiation therapy and chemotherapy in the treatment of primary squamous cell carcinoma of the vulva. Cancer, 75, 2289–94. WHO. (2003). In World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of the Breast and Female Genital Organs, ed. F.A. Tavassoli and P. Devilee. Lyon: IARC Press, Chap. 7.
303
27
GESTATIONAL TROPHOBLAST TUMOURS Philip Savage
Introduction Gestational trophoblast tumours (GTTs) are a group of rare but highly curable diseases. The diagnoses form a spectrum from the rarely malignant partial molar pregnancy through to the aggressive malignancies of choriocarcinoma and placental site trophoblast tumour. Fortunately, with effective treatment, nearly all patients can be cured, even those presenting with advanced metastatic disease. To provide an optimal service for this rare malignancy, the Department of Health in the UK directly funds a national human chorionic gonadotrophin (HCG) surveillance programme for all molar pregnancy patients as well as a follow-up service for treated GTT patients. The medical treatment of GTT in the UK is also centralised; there are two centres, one at Weston Park Hospital in Sheffield and the other at Charing Cross Hospital in London. Although the main clinical care for GTTs is centralised, it remains important for all UK oncologists to have a working knowledge of GTT because choriocarcinoma and placental site trophoblast tumour (PSTT) can occur as malignancies of unknown primary with a wide range of presentations. In these patients, early involvement from the local oncology team in achieving the correct diagnosis can be lifesaving. The pattern of GTT management in other countries varies; some countries have established centralised care, whereas, in many other countries, patients are treated locally by doctors who are likely to only see a case every few years. This chapter outlines the pathology, natural history, and management of GTT based on the experience of the UK GTT service at Charing Cross Hospital in London.
and the malignant forms: invasive mole, choriocarcinoma and PSTT, as shown in Table 27.1. All forms of GTT constitutively produce HCG, although abnormalities of the molecule from some tumour cells can lead to false-negative results in some types of simple HCG assays (Cole et al., 2001).
Premalignant forms of GTT The premalignant forms of GTT – partial and complete molar pregnancies – occur at a frequency of 0.2 to 1.5 per 1000 live births (Smith and Kim, 2003). Although partial or complete molar pregnancies are both unable to produce a viable foetus, they differ significantly in their genetic make-up and risk of malignant change.
Partial mole In a partial molar pregnancy, the trophoblast cells are triploid with two sets of paternal and one set of maternal chromosomes. In the early stages of pregnancy, a partial mole can resemble a normal conception when viewed via ultrasound. However, the embryo, if present, usually dies by week 8 or 9. The pathology from a partial mole frequently shows only focal changes and is usually far less florid than in a complete mole. As a result, the diagnosis of a partial mole may be missed; however, the risk of malignant change is less than 1 in 100 so partial moles rarely require additional treatment. The most frequent clinical presentation of a partial mole is bleeding in the first trimester or as a chance finding on a routine ultrasound. The gynaecological management is by suction or medical evacuation and, despite the low risk of malignant change, it is recommended that all such patients undergo HCG follow-up and monitoring.
Types of gestational trophoblast tumour
Complete mole
Gestational trophoblast tumours are divided into the premalignant partial and complete molar pregnancies
In a complete molar pregnancy, the trophoblast cells’ genetic material is entirely of male origin, following
304
Gestational trophoblast tumours
Table 27.1. Types of gestational trophoblast tumour Premalignant
Partial hydatidiform mole
Malignant
Invasive mole
Complete hydatidiform mole Choriocarcinoma Placental site trophoblast tumour
the loss of the nucleus from the ovum at some point in development or fertilisation. Most commonly, the chromosome count is 46XX, which results from a single sperm duplicating within an ovum that lacks the maternal chromosomes. Less frequently, the genotype can be 46XY, which occurs when an empty ovum is fertilised concurrently by two sperm. In the UK, the majority of complete moles are diagnosed in the first trimester following abnormal bleeding or from the first routine ultrasound, which characteristically demonstrates a complex echogenic intrauterine mass with numerous cystic spaces. The management is by suction evacuation and the histology demonstrates oedematous villi, although the characteristic textbook ‘bunch of grapes’ appearance is now rarely seen because this develops in the second trimester. Macroscopically, there is no foetal tissue, although microscopically, some embryonic cells can be found. In contrast to a partial mole, where malignant change only rarely occurs, chemotherapy for persistent disease is required in 10 to 15% of patients with complete molar pregnancies.
The international data on molar pregnancy incidence have shown considerable variation, with rates of up to 40.2 per 1000 in the Far East. It is interesting that updated data suggest that the incidence has now fallen to 2.3 per 1000, close to the European rates (Kim et al., 2004). It is presently unclear whether these changes are a result of improvements in diagnostic accuracy or exogenous factors such as changes to Western diet.
Risk factors for molar pregnancy By definition, gestational tumours can only occur as a result of fertilisation of an ovum. Two main groups of people have an enhanced risk of developing a molar pregnancy at conception: women at the extremes of the reproductive ages have an enhanced risk; girls under the age of 15 have a risk approximately 20 times higher than those aged 20 to 40, whereas women over age 45 have a several-hundred-fold higher risk (Sebire et al., 2002). The association is much stronger with complete molar pregnancy and the partial mole incidence is relatively constant across all age groups. The other group at increased risk of developing a molar pregnancy are women who have previously been diagnosed with a molar pregnancy. The reported risks are approximately 1 in 55 for patients with one previous molar pregnancy and 1 in 10 for patients with two (Sebire et al., 2003). Worldwide, a few cases of familial molar pregnancy have been described in which the risk of repeat molar pregnancy is very high. However, these familial cases are extremely rare, and it is a highly unlikely diagnosis for a woman with a single or even a repeat molar pregnancy.
Incidence of molar pregnancy The number of patients registered for follow-up after a molar pregnancy in England and Wales in 2004 was 1374. It is probable that this is a modest underestimation of the actual numbers because there are likely to be other cases that are reported as spontaneous miscarriage, are not diagnosed at termination or are never registered. However, set against the 639 721 live births, this gives an overall incidence of 1 molar pregnancy for every 465 live births. The relative numbers of partial and complete molar pregnancies are an area of debate because the initial diagnoses of many cases have changed after expert review (Paradinas, 1998). However, the current figures suggest that partial and complete molar pregnancies occur in roughly equal numbers (Sebire et al., 2002).
Screening after molar pregnancy In the majority of patients with molar pregnancy, the abnormal trophoblast cells are unable to sustain their growth indefinitely and, as the cells die off, the HCG levels return to normal and no further treatment is required. However, in approximately 10% of complete moles and 0.5% of partial moles, the abnormal trophoblast cells can continue to grow, invade locally and metastasise. At present, there is no accurate way of predicting which patients with molar pregnancies will develop persistent, invasive or metastatic disease. However, because the trophoblast cells constitutively produce HCG, monitoring the HCG level can provide an accurate assessment of regression or growth. 305
Philip Savage
The UK has had a national HCG-based screening service in place for over 30 years. All patients diagnosed with a molar pregnancy are registered, and samples are sent for HCG measurement every 2 weeks. The analysis of the results allows the careful monitoring of patients with regressing disease as the HCG falls and allows early identification of patients who require treatment before the development of any major problems. The diagnosis of the change from a premalignant form of GTT to a malignant form is usually made clinically, based on the clinical assessment and, in particular, the pattern of change of the HCG level. In contrast, information from a further biopsy in these patients is rarely of clinical value, and biopsy should be avoided because these highly vascular tumours can bleed very heavily. At Charing Cross Hospital, patients who require chemotherapy treatment after a molar pregnancy are generally referred to as having persistent trophoblast disease. Although the condition is scientifically a malignancy, we try to avoid using the word ‘cancer’ with these patients, particularly as they have a near 100% cure rate.
gests that a second evacuation is rarely of benefit in patients if the HCG level is higher than 5000 IU/l. Concerning these patients, it is now our preference to move directly to chemotherapy treatment (Pezeshki et al., 2004; Savage and Seckl, 2005).
Malignant forms of GTT Invasive mole An invasive mole generally arises from a previous complete mole and is characterised by invasion of the myometrium, which can lead to perforation of the uterus. Histologically, an invasive mole appears similar to a complete mole but is characterised by the ability to invade the myometrium if left untreated. With the introduction of routine ultrasound, the early evacuation of complete moles, and effective HCG surveillance, the diagnosis of invasive moles has become very rare in the UK.
Choriocarcinoma Indications for further treatment of patients with molar pregnancy Retrospective analysis of patients from the surveillance programme has allowed the identification of the following major indications for the treatment of patients undergoing post-molar pregnancy surveillance: r Brain, liver, gastrointestinal metastases or lung metastases larger than 2 cm on chest X-ray. r Histological evidence of choriocarcinoma. r Heavy PV bleeding or GI/intraperitoneal bleeding. r Pulmonary, vulval or vaginal metastases, unless the HCG level is falling. r Rising HCG level in two consecutive serum samples. r Level of HCG greater than 20 000 IU/l more than 4 weeks after evacuation. r Level of HCG plateaus in three consecutive serum samples. r Raised HCG level 6 months after evacuation. Whereas a small minority of these patients may be cured by a second uterine evacuation, the majority of patients will need chemotherapy treatment. Historically, we have suggested a cut-off point of an HCG value of 20 000 IU/l for patients with disease limited to the uterus and for whom a further evacuation is considered. However, analysis of the updated UK data from both Sheffield and Charing Cross sug-
306
Histologically and clinically, choriocarcinoma is obviously malignant and it presents the most frequent cause of emergency medical problems in the management of trophoblast disease. Choriocarcinoma may follow a complete molar pregnancy and can also occur after a term pregnancy, with an estimated frequency of 1 case per 50 000 live births. A biopsy is not essential, may be hazardous in each of these settings, and is generally avoided if a clinical diagnosis can be safely made. When pathology is available, the characteristic findings of choriocarcinoma show sheets of syncytiotrophoblast or cytotrophoblast cells with haemorrhage, necrosis and intravascular growth. Genetically choriocarcinoma can show a range of gross abnormalities but without any specific characteristic patterns. Despite the differing genetic make-up of persistent trophoblast disease after a molar pregnancy and choriocarcinoma occurring after term pregnancy, the treatment and response are determined by the overall prognostic scores rather than by the specific histology. Patients with choriocarcinoma can present with bleeding from the disease locally in the uterus or with a wide variety of symptoms from distant metastases, with the lungs, central nervous system and liver being the most frequent sites of distant disease.
Gestational trophoblast tumours
Placental site trophoblast tumour (PSTT) Placental site trophoblast tumours arise from the intermediate trophoblast cells and are the least common type of gestational trophoblast disease in that they form fewer than 2% of all cases, with an average of only one or two cases in the UK each year. Most frequently, PSTT follows a normal pregnancy but it can also occur after a non-molar abortion or a molar pregnancy. The clinical presentation of PSTT can range from slow-growing disease limited to the uterus to more rapidly growing metastatic disease that is similar in behaviour to choriocarcinoma. The average interval between the previous pregnancy and presentation of PSTT is 3.4 years and the most frequent presentations are bleeding and amenorrhoea. In PSTT, the HCG level, though elevated, is characteristically lower for the volume of disease than in the other types of GTT (Papadopoulos et al., 2002).
Investigations in GTT For the majority of patients with persistent trophoblast disease following a recent molar pregnancy, the only investigations performed before chemotherapy treatment are a Doppler ultrasound of the pelvis and a chest X-ray. These tests allow the formal exclusion of a new pregnancy as the cause of the HCG elevation, a measurement of the uterine tumour size, and an indication of the presence of any pulmonary metastases. The information is used as part of the prognostic scoring system that determines the intensity of the initial chemotherapy treatment; the presence of pulmonary metastases on the chest X-ray is an indication that prophylactic intrathecal chemotherapy with methotrexate should be considered. In contrast, patients presenting with choriocarcinoma or PSTT are fully staged with CT scans of the thorax, abdomen and pelvis and an MRI scan of the brain. These patients frequently have non-pulmonary metastases and the presence of CNS or hepatic disease alters the choice of initial chemotherapy treatment.
Staging classification and prognostic classification The FIGO staging classification for GTT is shown in Table 27.2. However, the value of full anatomical staging in optimising the treatment in GTT is relatively limited.
Table 27.2. FIGO staging of gestational trophoblast tumours. Stage
Description
Stage I
Tumour confined to the uterus
Stage II
Tumour extends outside of the uterus but is limited to the genital structures (adnexa, vagina, broad ligament)
Stage III
Tumour extends to the lungs with or
Stage IV
Tumour involves all other metastatic
without genital tract involvement sites From FIGO Oncology Committee (2002).
At Charing Cross and a number of other treatment centres, the FIGO prognostic scoring system as shown in Table 27.3 is more frequently used and is of more practical value (FIGO Oncology Committee, 2002). In this system, a number of prognostic factors are scored including the patient’s age, prior pregnancy, HCG level and number and sites of metastases; the total score produced is used to place patients into either low-risk or high-risk prognostic and treatment groups. For patients who fall into the low-risk prognostic group, treatment is generally commenced using relatively gentle chemotherapy with intramuscular methotrexate and folinic acid. In the high-risk prognostic group, historical data show that only about 10% of patients would be cured with single-agent methotrexate chemotherapy and treatment is usually started using the EMA-CO regimen (see the next section).
Chemotherapy treatment Low-risk disease management The most widely used standard therapy for patients with low-risk trophoblast disease is intramuscular methotrexate given with oral folinic acid rescue, as shown in Table 27.4. At Charing Cross Hospital, the first course of treatment is given as an inpatient procedure, with the following courses administered closer to home. However, patients with an initial pretreatment HCG of greater than 1000 IU/l often stay as inpatients for 3 weeks because they have a higher risk of bleeding as the larger tumours shrink rapidly with the initial chemotherapy.
307
Philip Savage
Table 27.3. The FIGO prognostic scoring system.a Prognostic score Prognostic factor
0
1
2
4
Age
<40
≥40
—
Antecedent pregnancy
Mole
Abortion
Term
Months from index
<4
4–6
7 to < 13
≥13
<1000
1000–10 000
10 000–100 000
>100 000
pregnancy Pretreatment HCG (IU/l) Largest tumour size
<3 cm
3 to < 5 cm
≥5 cm
—
Site of metastases
Lung
Spleen, kidney
Gastrointestinal
Brain, liver
Number of metastases
—
1–4
5–8
>8
Previous chemotherapy
—
—
Single agent
Two or more drugs
a
The total prognostic score is calculated by adding up individual scores for each prognostic factor. Total score of 6 or less =
low risk; a score of 7 or more = high risk. From FIGO Oncology Committee (2002).
The low-risk methotrexate chemotherapy treatment is usually well tolerated without major toxicity. Methotrexate does not cause alopecia or significant nausea, and myelosuppression is rare. The most common side effects are pleural inflammation, mucositis and hepatic toxicity, but each of these occurs only rarely. For the low-risk patients with lung metastases on their chest X-ray, we add central nervous system prophylaxis with intrathecal methotrexate administered on three occasions, 2 weeks apart. All patients have their HCG levels checked twice a week while on treatment, and, following HCG normalisation, treatment is continued for another three cycles (6 weeks) to ensure the eradication of any residual disease present below the level of serological detection. Approximately two-thirds of the low-risk patients are successfully treated with methotrexate alone; however, for the other third of patients who have an inadequate response to methotrexate, as shown by an HCG plateau or increase, treatment is changed to second-line therapy. For second-line therapy, treatment is changed to either single-agent actinomycin D at 0.5 mg for days 1 to 5 every 2 weeks or EMA-CO combination chemotherapy dependant on the HCG level at the time of change. Currently, we use a level of 300 IU/l as the upper limit for changing to actinomycin D. Figures 27.1(a) and 27.1(b) show the treatment graphs of two low-risk patients who required chemotherapy following complete molar pregnancies. The first patient 308
(Fig. 27.1(a)), is rapidly cured solely with methotrexate treatment, whereas the second patient (Fig. 27.1(b)), despite an initial response to methotrexate, requires a change to EMA-CO chemotherapy to successfully complete treatment. The overall survival of the low-risk treatment group is nearly 100% and the stepwise introduction of more toxic chemotherapy as necessary minimises the longterm toxicity from treatment in the majority of patients (McNeish et al., 2002).
High-risk disease management Historical data from before the introduction of the modern multiagent chemotherapy schedules showed that only 10% of the high-risk prognostic group of patients would be cured with single-agent therapy (Bagshawe et al., 1989). Fortunately, the introduction of combination chemotherapy treatments in the 1970s transformed this situation, and modern series show cure rates of 86% for high-risk patients using EMA-CO chemotherapy (Bower et al., 1997; Newlands et al., 1991). The drug combination of etoposide/methotrexate/actinomycin-D and cyclophosphamide/vincristine gives a dose-intense treatment with the five chemotherapy agents, delivered as two groups 1 week apart, as shown in Table 27.4. However, these drugs are fairly myelosuppressive and support with G-CSF injections is frequently helpful in keeping the treatment on schedule. Fortunately,
Gestational trophoblast tumours
Table 27.4. Chemotherapy regimens for trophoblast disease Regimen
Description
Methotrexate/
Day 1, methotrexate 50 mg i.m. at noon
folinic acid
Day 2, folinic acid 30 mg p.o. at 6 p.m. Day 3, methotrexate 50 mg i.m. at noon Day 4, folinic acid 30 mg p.o. at 6 p.m. Day 5, methotrexate 50 mg i.m. at noon Day 6, folinic acid 30 mg p.o. at 6 p.m. Day 7, methotrexate 50 mg i.m. at noon Day 8, folinic acid 30 mg p.o. at 6 p.m.
EMA-CO
who present with trophoblast disease have cerebral metastases at the time of diagnosis. A review of patients with CNS metastases treated between 1981 and 2000 indicated that 86% of the 39 patients were cured with chemotherapy treatment (Newlands et al., 2002). For patients with cerebral metastases, the EMA-CO chemotherapy is modified to contain a higher dose of methotrexate, which enhances penetration into the CNS. This modification is combined with an intrathecal methotrexate administration given the same week as the CO treatment. Table 27.4 shows chemotherapy regimens for trophoblast disease.
Week 1 Day 1, actinomycin-D 0.5 mg i.v., etoposide 100 mg/m2 i.v., methotrexate 300 mg/m2 i.v. Day 2, actinomycin-D 0.5 mg i.v., etoposide 100 mg/m2 i.v., folinic acid 15 mg p.o. 12-hourly × 4 doses (starting 24 hrs after commencing methotrexate) Week 2 Day 8, vincristine 1.4 mg/m2 (max. 2 mg) i.v., cyclophosphamide 600 mg/m2 i.v.
serious toxicity is rare with EMA-CO and the majority of patients tolerate treatment well. Similar to that of lowrisk patients, treatment is continued for 6 weeks after the normalisation of the HCG. Figure 27.1(c) shows the treatment graph of a high-risk patient who presented 3 months after the birth of her daughter and was successfully treated with EMA-CO chemotherapy. Of the high-risk patients treated with EMA-CO, approximately one in five develop a resistance to this drug combination and require a change in second-line drug treatment. Here, the EP-EMA regimen is generally used, in which cisplatin and a further dose of etoposide replace the vincristine and cyclophosphamide. EP-EMA chemotherapy, which may need to be combined with surgery on some occasions, produces a cure rate of 90% in this relatively rare group of patients (Newlands et al., 2000). In contrast to most other malignancies in which cerebral metastases are associated with a poor prognosis, trophoblast patients with CNS disease can be routinely cured of their disease. Approximately 4% of the patients
Management of placental site trophoblast tumour (PSTT) Placental site trophoblast disease is very rare, and it is treated differently from the more common types of trophoblast disease. The management is dependent on the disease status; if the disease is limited to the uterus, hysterectomy is often curative. In disseminated disease, treatment with chemotherapy using the EP-EMA regimen, continued for 8 weeks after the normalisation of the HCG level, is recommended. Following this, a hysterectomy is recommended because viable disease can persist in the uterus despite the HCG level falling to normal values. The current data for PSTT patients treated at Charing Cross Hospital show a 100% cure rate for those presenting within 4 years of pregnancy, but a poorer prognosis for those presenting after a longer interval (Papadopoulos et al., 2002).
Risk of relapse and late treatment complications After the HCG level has fallen to normal, the outlook is very good for patients with trophoblast disease. At this point, the risk of relapse is less than 5% for patients treated for low-risk disease and only 3% for high-risk patients treated with the EMA-CO regimen. Usually these recurrences happen within the first 12 months after treatment but can occur up to 6 years later. Fortunately, trophoblast tumours remain highly curable even at relapse and a recent analysis indicated that 100% of patients who were originally in the low-risk category can be cured on relapse, with a cure rate of 85% for 309
Philip Savage
100000
100000
10000
10000 1000
Jan
Feb
Mar
Stop RX
MTXFA
Dec
Treatment
1 MTXFA
1 MTXFA
10
MTXFA
10
MTXFA
100
Treatment
100
Apr
May
Date
Nov
Dec
Jan
Feb
Mar
Stop RX
1000
EMA CO MTX intrathecal EMA CO MTX intrathecal EMA CO EMA CO EMA MTX intrathecal CO EMA CO EMA CO EMA 1day CO EMA 1day CO
HCG(S)
1000000
HCG(S)
1000000
Apr
May
Date
(a)
(c)
1000000 100000
HCG(S)
10000 1000 100 10
Mar
Apr
CO EMA CO EMA CO EMA CO EMA 1day CO EMA 1day CO EMA 1day Stop RX
MTXFA
MTXFA
MTXFA
MTXFA
Treatment
1
May
Jun
Jul
Aug
Date
(b) Figure 27.1. Serum HCG graphs for three patients during treatment for gestational trophoblastic disease. (a) A patient who was cured rapidly by methotrexate treatment alone. (b) A patient who responded initially to methotrexate but then required a change to EMA-CO chemotherapy to successfully complete treatment. (c) A high-risk patient who was treated successfully with EMA-CO chemotherapy. CO = cyclophosphamide, vincristine; EMA = etoposide, methotrexate, actinomycin-D; MTXFA = methotrexate and folinic acid; HCG(S) = serum HCG (IU/l); RX = treatment.
those initially presenting with high-risk disease (Powles et al., 2006).
Subsequent fertility Fertility is usually maintained after either low- or highrisk chemotherapy treatment and regular periods restart within 6 months of completing chemotherapy. However, there is some toxicity: the chemotherapy treatment brings menopause forward by approximately 1 year for low-risk methotrexate and 5 years for high-risk EMA-CO (Bower et al., 1998). We recommend that pregnancy be avoided for 12 months after the end of treatment to minimise any pos310
sible damaging effects on developing oocytes and to minimise the confusion over disease relapse from the HCG produced in pregnancy. Many patients are surprised and delighted to hear that, despite the exposure to cytotoxic chemotherapy, particularly in the highrisk group, there does not appear to be any significant increase in foetal abnormalities, and 83% of women wishing to conceive are able to have at least one live birth. With the increasing numbers of long-term survivors following chemotherapy treatment for GTT, it has become apparent that intensive chemotherapy treatment with the EMA-CO and EP-EMA regimens can result in an increased risk of secondary malignancy. The information from the GTT patient database indicates that the
Gestational trophoblast tumours
lifetime risk of further malignancy is increased 1.5-fold, with the largest increase found for myeloid leukaemia (Rustin et al., 1996). Similar long-term risks have been seen in other malignancies that are routinely cured with chemotherapy such as Hodgkin lymphoma and non-Hodgkin lymphoma. In high-risk GTT patients it is unlikely that the intensity of chemotherapy can be reduced while achieving such high cure rates. However, earlier diagnosis would both reduce the duration of EMA-CO treatment and place some patients into the low-risk group. This concern over the use of combination chemotherapy reinforces the benefits of effective surveillance, allowing treatment to be commenced with single-agent methotrexate, which has not been demonstrated to have any long-term detrimental effects.
Summary GTTs form a rare, but fascinating, group of tumours. Their aetiology, biology and responsiveness to treatment are very different from those of any other form of cancer. In the UK, there is a centralised surveillance and treatment service that is a widely admired model in many parts of the world (Goldstein et al., 2004). As a result, most UK oncologists are not routinely involved in the care of these patients. However, each year a small number of patients with GTT present with disseminated malignancy, and a delayed diagnosis can result in increased treatment-related morbidity or the patient’s death before treatment can be started. We recommend that a formal serum HCG measurement is performed in all women presenting with cancer of unknown primary and, where appropriate, that cases are discussed promptly with the teams at Charing Cross or Sheffield.
REFERENCES Bagshawe, K. D., Dent, J., Newlands, E. S. et al. (1989). The role of low-dose methotrexate and folinic acid in gestational trophoblastic tumours (GTT). Br. J. Obstet. Gynaecol., 96, 795–802. Bower, M., Newlands, E. S., Holden, L. et al. (1997). EMA/CO for high-risk gestational trophoblastic tumors: results from a cohort of 272 patients. J. Clin. Oncol., 15, 2636–43. Bower, M., Rustin, G. J., Newlands, E. S. et al. (1998). Chemotherapy for gestational trophoblastic tumours hastens menopause by 3 years. Eur. J. Cancer, 34, 1204–7. Cole, L. A., Shahabi, S., Butler, S. A. et al. (2001). Utility of commonly used commercial human chorionic gonadotropin
immunoassays in the diagnosis and management of trophoblastic diseases. Clin. Chem., 47, 308–15. FIGO Oncology Committee (2002). FIGO staging for gestational trophoblastic neoplasia 2000. Int. J. Gynaecol. Obstet., 77, 285–7. Goldstein, D. P., Garner, E. I., Feltmate, C. M. et al. (2004). Comment on ‘The role of repeat uterine evacuation in the management of persistent gestational trophoblastic disease.’ Gynecol. Oncol., 95, 421–2. Kim, S. J., Lee, C., Kwon, S. Y. et al. (2004). Studying changes in the incidence, diagnosis and management of GTD: the South Korean model. J. Reprod. Med., 49, 643–54. McNeish, I. A., Strickland, S., Holden, L. et al. (2002). Low-risk persistent gestational trophoblastic disease: outcome after initial treatment with low-dose methotrexate and folinic acid from 1992 to 2000. J. Clin. Oncol., 20, 1838–44. Newlands, E. S., Bagshawe, K. D., Begent, R. H. et al. (1991). Results with the EMA/CO (etoposide, methotrexate, actinomycin D, cyclophosphamide, vincristine) regimen in high risk gestational trophoblastic tumours, 1979 to 1989. Br. J. Obstet. Gynaecol., 98, 550–7. Newlands, E. S., Mulholland, P. J., Holden, L. et al. (2000). Etoposide and cisplatin/etoposide, methotrexate, and actinomycin D (EMA) chemotherapy for patients with high-risk gestational trophoblastic tumors refractory to EMA/cyclophosphamide and vincristine chemotherapy and patients presenting with metastatic placental site trophoblastic tumors. J. Clin. Oncol., 18, 854–9. Newlands, E. S., Holden, L., Seckl, M. J. et al. (2002). Management of brain metastases in patients with high-risk gestational trophoblastic tumors. J. Reprod. Med., 47, 465–71. Papadopoulos, A. J., Foskett, M., Seckl, M. J. et al. (2002). Twenty-five years’ clinical experience with placental site trophoblastic tumors. J. Reprod. Med., 47, 460–4. Paradinas, F. J. (1998). The diagnosis and prognosis of molar pregnancy: the experience of the National Referral Centre in London. Int. J. Gynaecol. Obstet., 60 (Suppl. 1), S57–64. Pezeshki, M., Hancock, B. W., Silcocks, P. et al. (2004). The role of repeat uterine evacuation in the management of persistent gestational trophoblastic disease. Gynecol. Oncol., 95, 423–9. Powles, T., Young, A., Short, D. et al. (2006). The outcome of patients with relapsed gestational trophoblastic neoplasm (GTN) after the completion of chemotherapy. J. Clin. Oncol., 24 (Suppl.), 5033. Rustin, G. J., Newlands, E. S., Lutz, J. M. et al. (1996). Combination but not single-agent methotrexate chemotherapy for gestational trophoblastic tumors increases the incidence of second tumors. J. Clin. Oncol., 14, 2769–73. Savage, P. and Seckl, M. J. (2005). The role of repeat uterine evacuation in trophoblast disease. Gynecol. Oncol., 99, 251–2. Sebire, N. J., Foskett, M., Fisher, R. A. et al. (2002). Risk of partial and complete hydatidiform molar pregnancy in relation to maternal age. Br. J. Obstet. Gynecol., 109, 99–102. Sebire, N. J., Fisher, R. A., Foskett, M. et al. (2003). Risk of recurrent hydatidiform mole and subsequent pregnancy outcome following complete or partial hydatidiform molar pregnancy. Br. J. Obstet. Gynecol., 110, 22–6. Smith, H. O. and Kim, S. J. (2003). In Gestational Trophoblastic Diseases, ed. B. W. Hancock, E. S. Newlands, R. S. Berkowitz
311
Philip Savage
et al., 2nd edn. Sheffield: International Society for the Study of Trophoblastic Diseases.
FURTHER READING Hancock, B. W., Newlands, E. S., Berkowitz, R. S. et al. (2003). Gestational Trophoblastic Disease, 2nd edn. Sheffield: International Society for the Study of Trophoblastic Diseases.
312
International Society for the Study of Trophoblastic Diseases (ISSTD). http://www.isstd.org/index.html (accessed September 2006). Royal College of Obstetricians and Gynaecologists (RCOG). (2004). The Management of Gestational Trophoblastic Neoplasia. RCOG Guideline No. 38. London: Royal College of Obstetricians and Gynaecologists. Soper, J. T., Mutch, D. G. and Schink, J. C. (2004). Diagnosis and treatment of gestational trophoblastic disease: ACOG Practice Bulletin no. 53. Gynecol. Oncol., 93, 575–85.
28
LUNG Fergus Macbeth and Carys Morgan
Introduction Lung cancer has a significant impact on national mortality in the UK, accounting for 6% of all deaths and 22% of deaths from cancer. It has one of the lowest survival outcomes of any cancer with fewer than 25% alive at 1 year and only a 7% 5-year survival. These 5-year figures have changed little over the past 30 years, although there has been an improvement of around 10% in 1-year survival, probably due to more widespread use of palliative therapies. Over the past century, lung cancer has gone from a rare disease to being the most common cancer in men and the third most common in women, reflecting smoking habits. The direct link with tobacco use was made in the 1950s in the famous epidemiological study by Doll and Hill (1950); and although smoking prevalence in the population has continued to fall over the past five decades, lung cancer remains a major public health issue. At smoking’s peak in the 1940s, 82% of men smoked some form of tobacco; this number fell to a population prevalence of 26% in 2002. There has previously been a somewhat nihilistic view of lung cancer treatment both among health professionals and patients themselves, with referrals often late and only for palliation of symptoms. Significant changes have occurred more recently with the development of lung cancer services in the UK. In particular, multidisciplinary team (MDT) collaboration and rapid-access chest clinics have had an impact on referral patterns. Patients with early stage disease suitable for a radical approach are assessed by MDTs, and PET scanning is more widely available for this group and leads to more accurate pretreatment staging. Changes in radiotherapy (RT) techniques with conformal planning and the introduction of continuous hyperfractionated accelerated RT (CHART) have improved outcomes in patients treated with RT, and the use of adjuvant chemotherapy after surgery is becoming standard practice. Most patients still present with advanced disease and often have significant co-morbidities. Palliative RT can be very
beneficial in terms of symptom control, and palliative treatment with chemotherapy is now well established, with a significant if modest survival benefit for patients with a good performance status. More research to clarify optimum treatment strategies and to develop novel agents is essential; therefore, entry into clinical trials is recommended for all stages of disease. Although the improved survival outcomes seen in other cancers such as breast cancer have not been observed, the management of lung cancer is improving in the UK.
Classification Types of tumours affecting the lung are shown in Table 28.1.
Incidence and epidemiology The incidence of lung cancer in the UK (1997) is 75 in 100 000 for men, 35 in 100 000 for women. Approximately 38 800 new cases were diagnosed in 1998. Lung cancer is the most common cancer in men and the third most common in women, but it is the most common cause of cancer death in women. The incidence in men has decreased steadily over the past 20 years, from a rate of 115 in 100 000 in 1979, whereas the rate has increased in women, especially in those over the age of 65. Although the rate has decreased in those ages 45 to 64, it has increased in the 35 to 44 age group by 10% since 1979.
Carcinoma of the lung: non-small cell and small cell Risk factors and aetiology The major risk factor for lung cancer is undoubtedly cigarette smoking, which probably accounts for about 90% of cases. There is a clear relationship between risk and the age of smoking onset, the number of years of smoking, and the number of cigarettes smoked per day. 313
Fergus Macbeth and Carys Morgan
Table 28.1. Tumours affecting the lung Type
Examples
Benign
Papilloma Mucinous cystadenoma
Malignant
Squamous carcinoma
primary
Adenocarcinoma (various subtypes, including bronchioloalveolar) Large-cell carcinoma
tal bronchi which enter the bronchopulmonary segments. Each lung is attached to the mediastinum by its root, the hilum, which contains the bronchi, pulmonary vessels, lymphatics and nerves. The parietal pleura lines the thoracic wall, thoracic diaphragm and lateral mediastinum, and the visceral pleura completely covers the surfaces of the lungs extending into the fissures. These layers become continuous at the lung root (pleural reflection), and they are separated by a thin layer of fluid in a potential cavity (the pleural space).
Adenosquamous carcinoma Carcinosarcoma Adenoid cystic carcinoma Small-cell carcinoma Carcinoid tumours Unclassified Malignant
Bronchus
secondary
Kidney GI tract
Mediastinum The mediastinum contains the thymic remnant, trachea, oesophagus, heart and great vessels, the thoracic duct and lymph nodes, vagus and phrenic nerves and sympathetic trunks. All these structures can therefore be involved in disease spread by direct extension from lung cancer or by malignant lymphadenopathy.
Head and neck Sarcoma
Lymphatics
Bladder
The lymphatic drainage from the lungs is through channels running to the hilum along the bronchi. The hilar nodes drain to the tracheobronchial group and then up via mediastinal lymph trunks to the brachiocephalic veins. The regional lymph nodes of the thorax are divided into the intrathoracic (mediastinal and intrapulmonary nodes) and the scalene and supraclavicular nodes. The intrapulmonary nodes, which are hilar, peribronchial, and intrapulmonary, are classified as N1 if involved on the ipsilateral side. The mediastinal nodes include the para-, pre- and retrotracheal; aortic; subcarinal; peri-oesophageal; and inferior pulmonary ligament nodes. These nodes are classified as N2 if there is ipsilateral involvement and N3 if there is contralateral involvement. The subcarinal nodes are classified as N2. Positive supraclavicular or scalene nodes are classified as N3.
Uterus and cervix Ovary (typically pleural effusion)
Smoking cessation at any age reduces the risk of lung cancer, which drops to between 30 to 50% of that of continuing smokers after 10 years. With chronic exposure to tobacco smoke, the ciliated epithelium of the respiratory mucosa undergoes squamous metaplasia and eventually a field change can occur with the development of carcinoma in situ before the start of frankly invasive cancer. Industrial exposure to nickel, chromium and some arsenic compounds is associated with increased risk. Asbestos exposure appears to increase the risk of developing lung cancer by three-fold in smokers and is also a risk factor for mesothelioma. ‘Scar’ cancers occur in areas of lung fibrosis and there is an increased risk of adenocarcinoma in patients with fibrosing alveolitis.
Anatomy The trachea bifurcates into the right and left main bronchi at the carina. The bronchi further subdivide into the lobar bronchi (right upper, middle and lower, and left upper and lower) entering the lobes of the lung. Each lobar bronchus gives off branches called segmen314
Pathology The majority of bronchial carcinomas are ‘central’ in that they arise in the major airways – the main or segmental bronchi. Typically, an ulcerating mass grows in the wall of, and completely or partially blocking, the bronchus. Upper lobe tumours are slightly more common than middle and lower lobe and right-sided than left-sided tumours. ‘Peripheral’ tumours arising in the more distal airways and alveoli are more often
Lung
Table 28.2. Pathological features of lung cancer Features
Description
Macroscopic
Small cell Spreads in a submucosal fashion not usually intraluminal; crushes easily on biopsy Adenocarcinoma Peripheral location in 75%, and may arise in scar tissue; bronchioloalveolar subtype can present as multiple/bilateral nodules or an area of consolidation Squamous cell Often arise centrally, within main and lobar bronchi; often intraluminal and easily seen on bronchoscopy; bulky tumours that can cavitate
Microscopic
Small cell Small cells with round or oval nuclei, scant cytoplasm; lymphocyte-like cells; EM-dense core neurosecretory granules Usually cytokeratin +ve, also neuroendocrine markers (e.g. neuron-specific enolase) Adenocarcinoma Forms glands and papillae, mucous production seen Low-MW keratin +ve, EMA, CEA +ve, TTF-1 +ve; may be vimentin +ve; TTF-1 is useful in differentiating lung cancer from metastatic adenocarcinoma (only +ve in lung adenocarcinoma and thyroid carcinoma) Squamous cell Keratinisation ± intracellular bridges; epithelioid sheets; coarse chromatin and dense nucleus; high-MW keratin +ve Large cell No squamous or glandular patterns; large cells without glands, keratin pearls or bridges (Franklin, 2000)
EM = electron microscopy; EMA = epithelial membrane antigen; MW = molecular weight; TTF-1 = thyroid transcription factor 1.
adenocarcinomas. The pathological features of lung cancer are shown in Table 28.2. For clinical management, the important pathological distinction is between small-cell lung cancer (SCLC) and the remainder, which are lumped together as ‘non-small-cell’ lung cancers (NSCLCs). The distinction is relevant because, on the small bronchoscopic biopsies and cytology available from brushings and washings, SCLC can often be easily identified but distinction between the different types of NSCLC may be difficult and unreliable. SCLC accounts for about 15% of new cases and most of the rest are varieties of NSCLC. Squamous carcinoma is the most common NSCLC in the UK (about 50%) but the percentage of adenocarcinoma is increasing and it is now the most common subtype in the USA, which may in part be an artefact of changing diagnostic criteria with more sophisticated immunohistochemistry but is probably real and may be related to changes in cigarette tar content.
Spread Lung cancer can spread locally to the mediastinum, pleura (may produce effusion, especially adenocarcinoma), chest wall and ribs, vertebral body or diaphragm. The disease can spread via lymph nodes to the hilar nodes, the mediastinal nodes (pretracheal, paratracheal, para-aortic, subaortic and subcarinal) or the supraclavicular fossa (N3). It can also spread via the blood stream to the liver, adrenal gland, lung, brain, bone or skin.
Clinical presentation Respiratory symptoms include increasing cough, breathlessness, haemoptysis or unresolving pneumonia, all of which are symptoms that are not unusual in middle-aged smokers and, with a gradual onset, may be overlooked. Many patients are treated for asthma or with several courses of antibiotics before being referred 315
Fergus Macbeth and Carys Morgan
for a chest X-ray or further investigation. Severe breathlessness may be due either to significant obstruction to a major airway or to a pleural effusion. Superior vena caval obstruction (SVCO) results in distended neck and arm veins, facial oedema (especially in the mornings) and the development of collateral veins on the chest wall. Lung cancer is the most common cause of SVCO in middle-aged and older patients. A variety of different pain syndromes can occur with lung cancer: r Mediastinal pain is very variable in site and intensity but many patients complain of vague aching central chest pain, which can, however, be severe and become the dominant symptom. Severe pain is usually an indication of direct mediastinal involvement. r Chest wall pain – peripheral tumours may spread to involve the pleura and chest wall. Initially this may be poorly localised or pleuritic. With involvement of the ribs and intercostal nerve, a more characteristic severe and dysaesthetic pain may occur with dermatomal distribution. r Back pain – direct involvement of a vertebral body may result in back pain, and if the nerve roots are affected it may also be dermatomal. r Shoulder pain can result from direct involvement of the brachial plexus as part of the superior sulcus (Pancoast) syndrome, but it can also be due to referred pain from the diaphragm or mediastinal involvement of the phrenic nerve. r Facial pain is a not uncommon but poorly recognised symptom, indicating mediastinal involvement. It is characteristically vague, aching and persistent and located around the ipsilateral jaw, ear or maxilla. It may be the presenting symptom and patients may be diagnosed with trigeminal neuralgia or referred for dental care before the true cause is recognised. It is thought to be due to referred pain through the vagus. r Metastases are common in lung cancer and are associated with a wide variety of pain syndromes. Systemic symptoms include anorexia, fatigue, weight loss and sweats, which are quite common. They are often, but not always, indicators of extensive disease. Non-metastatic syndromes include the following: r Clubbing and HPOA – finger (and toe) clubbing is common. It is more usually (but not always) seen in patients with NSCLC rather than SCLC. When persistent it can be associated with hypertrophic pulmonary osteo-arthropathy (HPOA), resulting in joint pain (especially knees, ankles and wrists) and characteristic X-ray and bone scan appearances. 316
r SIADH – inappropriate ADH secretion is characterised by a low serum sodium and a urine osmolality that is inappropriately concentrated compared to that in the serum. SIADH is due to tumour production of an active peptide that mimics ADH and is quite commonly found in patients with SCLC. It can occasionally occur in patients with NSCLC. r Hypercalcaemia – high serum calcium can be caused by tumour secretion of an active peptide in the absence of bone metastases. Hypercalcaemia is unusual and most commonly occurs in patients with squamous carcinoma but it can occur in any tumour type. r Neurological – there are a variety of non-metastatic neurological syndromes (peripheral neuropathy, cerebellar ataxia, dementia) that can occur in patients with lung cancer of any histological type. Neurological symptoms may be associated with secretion of the anti-Hu antibody. The myasthenia-like Lambert Eaton syndrome is rare and typically occurs in patients with SCLC. r Prothrombotic tendency – many patients with lung cancer develop venous thrombosis and pulmonary embolism. There is good evidence that these patients have a prothrombotic tendency, which may be resistant to heparin. Metastases can occur at any site; the most common are the liver, adrenals, bone and brain.
Investigations Patients suspected of having lung cancer should be referred to a chest physician for assessment. The following referral recommendations have been made by NICE: r Urgent referral for a chest X-Ray should be offered when a patient presents with haemoptysis or any of the following unexplained or persistent (>3 weeks) symptoms or signs: cough, chest/shoulder pain, dyspnoea, weight loss, chest signs, hoarseness, clubbing, features suggestive of metastases or cervical/supraclavicular lymphadenopathy. r If CXR or CT scans suggest lung cancer (including pleural effusion and slowly resolving consolidation) patients should be offered an urgent referral to a member of the lung cancer MDT. Urgent referral prior to a CXR is recommended if there is persistent haemoptysis in a smoker/ex-smoker older than 40 years, signs of SVCO or stridor. r The patient should be seen within 2 weeks.
Lung
Patients should also have an FBC and serum biochemistry checked. Patients should have a CT scan of the thorax and upper abdomen with contrast to establish the extent of mediastinal disease and to look for liver and adrenal metastases. Scanning also helps to establish the best way to obtain histology. Unless they are very frail, an attempt should be made to establish a histological diagnosis either at bronchoscopy or by CT-guided lung biopsy. Routine imaging of the bone and brain is not indicated unless there are symptoms suggestive of problems. PET scanning is now recommended for all NSCLC patients being considered for radical treatment (surgery or RT). Patients being considered for surgery should have mediastinoscopy, unless both the PET scan and the CT scan show no mediastinal involvement. Potentially operable patients with a PET scan that shows abnormality in mediastinal nodes should still have mediastinoscopy because there is up to a 50% false-positive rate.
Table 28.3. TNM classification for non-small-cell lung cancer Stage
Description
Tx
Presence of malignant cells, no visible tumour
T0
No primary detected
Tis
Carcinoma in situ
T1
≤3 cm in largest dimension surrounded by lung or visceral pleura OR endobronchial tumour, proximal to one lobar bronchus (i.e. not in main bronchus)
T2
>3 cm, OR involves visceral pleura, OR involves main bronchus >2 cm from carina, OR atelectasis or obstructive pneumonitis ≤1 lung
T3
Direct invasion of chest wall (including superior sulcus tumours), diaphragm, mediastinal pleura, parietal pericardium, OR tumour ≤2 cm from carina, OR obstructive pneumonitis or atelectasis of 1 lung
Staging
T4
Direct invasion of mediastinum, heart, great
Non-small-cell lung cancer
vessels, trachea, oesophagus, vertebrae or
For NSCLC patients, the latest UICC staging system should be used. A 1997 amendment was introduced to the staging system to enable patients with similar prognoses to be grouped together, and stages I and II are now subdivided to reflect this change. The TNM and stage groupings for NSCLC are shown in Tables 28.3 and 28.4, respectively.
carina, OR separate nodules in the same lobe, OR malignant pleural effusion NX
Regional lymph nodes cannot be assessed
N0
No metastases to regional nodes
N1
Ipsilateral peribronchial, hilar nodes or intrapulmonary nodes including direct extension
Small-cell lung cancer
N2
Ipsilateral mediastinal ± subcarinal nodes
The staging system for SCLC is shown in Table 28.5. For SCLC there are also other prognostic factors (poor PS, low serum Na, elevated LDH) that can be used to divide patients into good- and poor-prognosis groups.
N3
Contralateral mediastinal or hilar nodes or any
Treatment: non-small cell lung cancer (stages I and II)
scalene or supraclavicular nodes MX
Cannot be assessed
M0
No distant metastases
M1
Distant metastases present (includes separate nodule in a different lobe)
Adapted from UICC (2002).
NSCLC patients should be considered for surgery and if this is possible they should have either lobectomy or pneumonectomy, depending on the site and extent of the tumour.
Radical radiotherapy In the UK, where the median age at presentation is over 70 and there is a significant risk of smoking-related (and other) co-morbidity, especially emphysema, car-
diovascular disease, and peripheral vascular disease, a number of patients may be considered to be medically inoperable and should be considered for radical RT. Standard UK practice involves a planned CT, three- or four-field, single-phase treatment without prophylactic RT to uninvolved mediastinal sites. Patients should be treated with CHART, 54 Gy in 36 fractions over 317
Fergus Macbeth and Carys Morgan
Table 28.4. Stage groups for non-small-cell lung cancer; 5-year survival by stage is also shown Five-year survival Stage
Definition
Stage I
No lymph node
(%)
Stage III
Stage IV
Description
Limited disease
Disease confined to one hemithorax including ipsilateral mediastinal
Extensive
IA = T1 N0 M0
IA 60%
IB = T2 N0 M0
IB 38%
disease
Any evidence of disease beyond the above
T1–2 N1 or T3 N0 IIA = T1 N1 M0
IIA 34%
Adjuvant radiotherapy
IIB = T2 N1 M0 or
IIB 24% (T2 N1)
T3 N0 M0
22% (T3 N0)
The PORT meta-analysis (PORT Meta-analysis Trialists Group, 1998) has shown that adjuvant RT has a modest effect on local control and a detrimental effect on survival. There was a reduction in OS of 7% at 2 years, which was greatest in stage I and II disease. This outcome may be due to late effects on lung and heart in long-term survivors. A subgroup analysis did not show this detrimental effect in stage III disease (N2 or 3). The meta-analysis has been criticised for including mainly old trials using RT techniques and fractionation that would not be considered acceptable today. Therefore, postoperative RT may be indicated for selected patients with known residual disease following surgery.
IIIA = T1–3 N2 M0 or
IIIA 13% (T1–3N2)
T3 N1 M0
9% (T3 N1)
IIIB = Any N3 or
IIIB 7% (T4 N0–2)
Any T4
3% (T1–4 N3)
Any T, Any N, M1
1%
Survival figures adapted from Mountain (1997).
12 days, if available. The CHART trial showed a significant 2-year survival benefit of 29% compared to 20% for patients given 60 Gy over 6 weeks (Saunders et al., 1999). If CHART is not available locally, alternative fractionation schedules include 52 to 55 Gy in 20 fractions over 4 weeks or 60 to 66 Gy in 30 to 33 fractions over 6 weeks. The overall 2-year survival should be in the range of 30 to 40%, and five-year survival should be 15 to 20%. Patients with pancoast tumours should be treated with radical radiotherapy if volume and lung function allow. Occasional patients may be operable and it is usual, but of uncertain benefit, to give them preoperative radiotherapy to a dose of 40 Gy in 20 fractions or equivalent. The dose to normal lung (especially the contralateral lung) should be kept to a minimum. The V20 (percentage total lung volume receiving >20 Gy) should be less than 35%. This is a very useful indicator of the risk of developing pneumonitis. The original paper (Graham et al., 1999) showed a 7% incidence of grade 2 or higher pneumonitis if the V20 was below 32%, but this incidence increased to 36% if the V20 was higher than 40%. There are no hard rules about the level of lung function that precludes radical RT, but any patient with an FEV1 of less than 1.2 l or 50% of predicted should be reviewed carefully. A small apical tumour will be more safely treated than a large central or lower lobe tumour. 318
Stage
and supraclavicular fossa nodes
involvement
Stage II
Table 28.5. Small-cell lung cancer staging system
Adjuvant chemotherapy following surgery The NSLC meta-analysis (Arriagada et al., 2004) suggested a 4% increase in 5-year survival following cisplatin-based adjuvant chemotherapy. This finding led to the question of whether this level of benefit (NNT = 25) is clinically useful and which groups of patients potentially gain the most benefit. A number of RCTs have since been reported, with varying results. However, a meta-analysis by the Lung Adjuvant Cisplatin Evaluation (LACE) Group presented in abstract form at ASCO 2006 (Pignon et al., 2006) looked at patient data from five large trials published since the 1995 metaanalysis and concluded that the benefit from adjuvant chemotherapy depends on the stage of disease and is greatest in patients with stage II and III disease. The 5-year absolute survival benefit overall was 4.2%. Currently the evidence suggests that the benefit from adjuvant chemotherapy is greater in more advanced disease and probably not significant for patients with stage IA disease. Further clarification is required in patients with stage IB disease, although subset analyses from some of the more recent trials have also failed to show a survival advantage for this group.
Lung
Neoadjuvant chemotherapy before surgery The value of neoadjuvant chemotherapy is undetermined because of conflicting evidence from randomised trials. The MRC LU22 trial (surgical resection ± preoperative chemotherapy in operable NSCLC) should provide further information.
Treatment: non-small-cell lung cancer (stage III) Non-small-cell lung cancer involves a large and heterogeneous group of patients, ranging from those with small-volume primary tumours and microscopic involvement of one group of mediastinal nodes to those with bulky tumours and widespread mediastinal disease. The treatment options are wide.
Surgery Surgery may be an option for some patients with resectable mediastinal disease.
Neoadjuvant chemotherapy (preoperative) Preoperative neoadjuvant chemotherapy is controversial and has no clear evidence of benefit.
Radical radiotherapy Radical RT should be standard treatment in suitable patients (good PS, ≤10% weight loss, normal biochemistry) with a volume of disease that can be treated to a radical dose without high risk of severe radiation pneumonitis (see earlier discussion). The 2-year survival should be in the range of 20 to 30%. Superior sulcus tumours should be treated with radical RT, if technically feasible, in two phases, including the whole of the adjacent vertebral bodies to the level of spinal cord tolerance because of the risk of nerve root involvement. No strong evidence exists to support combined modality approaches, though these are widely advocated.
Neoadjuvant chemotherapy with radical radiotherapy Evidence from a number of RCTs shows a significant survival benefit from giving cisplatin-based chemotherapy before conventional (60 Gy) radical RT, with some increase in radiation oesophagitis; however, the size of this benefit is unclear. Many consider sequential chemotherapy followed by RT to be standard treatment for stage III lung cancer in patients who are not treated with CHART. Although there has been some experience
of chemotherapy followed by CHART, this is currently the subject of the MRC LU23 (INCH) trial and has not previously been formally tested in trials.
Concurrent chemoradiotherapy Early results from RCTs suggested that concurrent chemoradiotherapy may be more effective but the advantage of concurrent treatment remains unclear. A recent meta-analysis and review of RCTs (Aup´erin et al., 2006) suggests that concurrent treatment may be more effective but there is an increase in toxicity and so treatment may be applicable to relatively few patients.
Palliative radiotherapy Palliative RT is a good option for patients who have either disease of too large a volume for radical RT or co-morbidity. The third MRC trial (Macbeth et al., 1996) showed that for patients with good PS (0 or 1) there is a significant survival advantage to giving 39 Gy in 13 fractions (or 36 Gy in 12 fractions) compared to 17 Gy in 2 fractions. The 2-year survival was 13% compared to 9%, but with no difference in symptom control. There was a small incidence of radiation myelitis in patients receiving 39 or 17 Gy. For patients with poor PS and local symptoms, lowdose RT is the best option (see later discussion).
Treatment: non-small-cell lung cancer (stage IV) Unfortunately many patients have obvious metastatic disease at presentation or develop it later. Clearly an important part of their management is good symptom control, but there are options for active treatment.
Palliative radiotherapy to metastases Single fractions of RT to symptomatic bone and skin metastases may be helpful. Palliative RT to brain metastases should only be considered for patients with good PS.
Palliative radiotherapy to chest For patients with significant local symptoms from the primary tumour or mediastinal disease, palliative RT is the treatment of choice. There is a 60% chance of significant improvement in the symptoms. RCTs have shown that there is no difference in symptom control or survival from radiobiologically equivalent regimens (e.g. 17 Gy in 2 fractions, 20 Gy in 5 fractions, 30 Gy in 319
Fergus Macbeth and Carys Morgan
10 fractions). Short regimens are generally preferable in that they involve fewer visits. For poor PS patients a 10 Gy single fraction is sufficient (Medical Research Council Lung Cancer Working Party, 1992). Use of 17 Gy in 2 fractions is associated with a risk of RT myelitis and either the spinal cord should be partly shielded or the dose should be reduced to 16 Gy. Large fractions are associated with acute side effects such as nausea, acute chest pain, fever and rigors. The patients should be warned and given antiemetics, analgesics, and paracetamol. Asymptomatic patients can be treated when symptoms develop without detrimental effect on outcomes or palliation (Falk et al., 2002). Patients with better performance status should be considered for chemotherapy.
Chemotherapy Chemotherapy for patients with symptomatic locally advanced and metastatic disease is an option that needs to be considered carefully. Large RCTs and two metaanalyses have shown that chemotherapy in advanced disease improves survival, with an objective response rate of around 20% and with 40 to 50% of patients experiencing symptomatic improvement. The survival benefit is around 2 months in median survival (from 6 to 8 months) with a 10% increase in 1-year survival (Non Small Cell Lung Cancer Collaborative Group, 1995). In the UK, the Big Lung Trial randomised 725 patients to cisplatin-based chemotherapy or best supportive care and the median survival was increased by 9 weeks (from 5.7 to 8 months) with an improvement in 1-year survival of 9% (29% compared to 20%) in keeping with these data (Spiro et al., 2004). The data on improvement in quality of life (QOL) are less robust but it seems that overall QOL does not deteriorate and it may improve a little for patients having chemotherapy. Newer agents are now available for use in NSCLC. These third-generation agents include taxanes, gemcitabine, and vinorelbine. Recent trials have suggested improved 1-year survival when these are used in combination with a platinum drug. A direct comparison of four third-generation regimens (cisplatin/paclitaxel, cisplatin/gemcitabine, cisplatin/docetaxel and carboplatin/paclitaxel; Schiller et al., 2002) showed no significant difference among the regimens in terms of survival but overall 1-year survival was 33%. However, the trial was amended to exclude patients with PS 2 after a high rate of serious adverse events in this group, which may have influenced the survival data. It does not seem to matter what drug com320
binations are used, provided that one agent is either a platin or a taxane. There is no strong evidence that any of the newer agents in combination are more effective than older agents, although the toxicity profiles are different. In the UK the BTOG2 trial compares carboplatin with cisplatin (50 mg/m2 or 80 mg/m2 ) in combination with gemcitabine in a three-arm RCT. Therefore, the decision of which regimen is used depends on issues of cost, convenience and toxicity. Some regimens in use include the following: r MIC – day 1, cisplatin 50 mg/m2 , mitomycin 6 mg/m2 , ifosfamide 3 g/m2 . r MVP – day 1, cisplatin 50 mg/m2 , mitomycin 6 mg/m2 , vinblastine 6 mg/m2 . r NP – day 1, cisplatin 80 mg/m2 , vinorelbine 30 mg/m2 ; day 8 vinorelbine 30 mg/m2 . r CG – day 1, cisplatin 80 mg/m2 or cisplatin 50 mg/m2 , gemcitabine 1250 mg/m2 ; day 8, gemcitabine 1250 mg/m2 . r GemCarbo – day 1, gemcitabine 1250 mg/m2 , carboplatin AUC 6; day 8, gemcitabine 1250 mg/m2 . Docetaxel monotherapy (75 mg/m2 ) should be considered as second-line chemotherapy in patients with a good PS who relapse. Pemetrexed has also been used in this situation and has been directly compared with docetaxel. Median survival and 1-year survival (29%) were equivalent in both regimens in this study with less toxicity in the pemetrexed arm (Hanna et al., 2004). This agent has not yet been approved by NICE.
Non-small-cell lung cancer: recurrent disease Patients relapsing after RT should be considered for first-line palliative chemotherapy. Re-irradiation is also an option in some cases but the risks must also be considered before retreatment. As discussed previously, docetaxel is an option for good PS NSCLC patients relapsing after first-line chemotherapy but has lower response rates than first-line treatment.
Erlotinib Erlotinib is an oral EGFR tyrosine kinase inhibitor that is licensed for use in locally advanced or metastatic NSCLC after failure of at least one previous chemotherapy regimen. Its use is based on the results of a placebocontrolled RCT that showed an increase in overall survival by 2 months when it was used in this group of patients (i.e. as second-line therapy on relapse; Shepherd et al., 2005). Subset analyses suggest an improved response in non-smokers, Asian patients and
Lung
patients with adenocarcinoma. Quality of life was also improved. The main toxicities of erlotinib are a rash and diarrhoea. Ongoing trials are looking at erlotinib as first-line therapy in patients not suitable for chemotherapy.
Treatment: small-cell lung cancer (general aspects) The management of patients with SCLC depends on an assessment of the extent of disease and other key prognostic factors. A widely used prognostic score is the Manchester score. The following factors each contribute a Manchester score of 1: r Extensive disease. r WHO PS ≥ 2 (KPS ≤ 60). r Serum Na ≤ 132 mM/l. r Bicarbonate ≤ 24 mM/l. r Alkaline phosphatase > 165 IU/l. r LDH > 450 IU/l. Patients with a total score of 0 to 1 have a good prognosis, 2 to 3 have an intermediate prognosis, and 4 to 6 have a poor prognosis.
Small-cell lung cancer: good-prognosis patients Unless there is some other contraindication, goodprognosis SCLC patients should be treated with curative intent with chemotherapy, thoracic RT, and prophylactic cranial RT. With appropriate treatment the median survival should be 18 to 20 months and 3-year survival around 15%. Patients who have not relapsed after 3 years are unlikely to relapse subsequently, although they have an increased risk of second primary lung tumours and death from other smoking-related disease.
Chemotherapy A number of different chemotherapy regimens are in use in the UK. There is RCT evidence that regimens containing cisplatin or carboplatin are more effective than those without either drug (Pujol et al., 2000). Whether the addition of ifosfamide to a regimen of carboplatin and etoposide is more effective is uncertain, but in phase II reports this combination is associated with high median survival. It is also associated with greater toxicity, especially in older patients. It is also not clear how many cycles are needed. Usually between four and six cycles are given, depending on
response and tolerance. Patients who achieve a complete response (CR) after two cycles are most likely to experience long-term survival. Patients who have not achieved CR after four cycles are unlikely to be cured and continuing to six cycles is probably unhelpful. Most patients relapse and, if fit enough, should be considered for more chemotherapy. If there has been a reasonable length of remission – 6 months or more – retreating with first-line chemotherapy is appropriate. Rapid relapse is associated with a poor prognosis but using a non-cross-resistant second-line combination may be appropriate. First-line chemotherapy regimens include the following: r Carboplatin/etoposide – day 1, carboplatin AUC 5 or 6 then etoposide 50 mg b.d. orally days 1 to 7. r Carboplatin/etoposide – day 1, carboplatin AUC 5 or 6, etoposide 120 mg/m2 , i.v.i. then etoposide 100 mg b.d. orally days 2 and 3. The second-line regimen is CAV: day 1, cyclophosphamide 700 mg/m2 , doxorubicin 40 mg/m2 and vincristine 1.2 mg/m2 .
Thoracic Radiotherapy There is good evidence that patients with good PS who achieve a CR to chemotherapy benefit from thoracic RT with a significant improvement in survival (Pignon and Arriagada, 1992). The timing, volume and dose/fractionation are all controversial. A number of RCTs have investigated the use of concomitant compared to sequential RT, with conflicting results. A recent meta-analysis did not show a definite improvement in survival with concomitant treatment but confirmed increased toxicity, in particular oesophagitis (De Ruysscher et al., 2006). Concomitant RT is probably only feasible if a regimen with neither anthracyclines nor alkylating agents is used. Concomitant RT will also generally mean the use of larger volumes with an increase in the risk of RT pneumonitis. The required target volume is also uncertain. In the past, the practice was to treat the prechemotherapy tumour volume, which in many cases was very large and sometimes hard to define because of collapse/ consolidation. Modern practice is to treat the volume after chemotherapy, including the whole mediastinum and the probable bronchial origin of the tumour. Although CT planning is preferred, it may not always be necessary. The dose to the lung should be kept to a minimum by the use of parallel-opposed fields 321
Fergus Macbeth and Carys Morgan
up to spinal cord tolerance, because of the risk of pneumonitis. In the UK a dose fractionation of 40 Gy in 15 fractions is often used. An alternative would be 50 Gy in 25 fractions. One RCT from the US suggests that b.d. fractionation is more effective but it is possible that acceleration may be the important factor rather than hyperfractionation itself (Turrisi et al., 1999).
Prophylactic cranial irradiation (PCI) Patients with LD and a complete response to chemotherapy will benefit from PCI. There is a 5.4% increase in 3-year survival associated with PCI in this group (Arriagada et al., 1997). It is not clear what dose is the most effective. Commonly used regimens are 20 Gy in 10 fractions, 30 Gy in 15 fractions and 36 Gy in 18 fractions. It is not clear whether or not PCI increases the risk of late effects such as cognitive impairment and ataxia, but it is probably wise to avoid giving PCI in patients older than 70 years and patients with known cerebrovascular disease.
Small-cell lung cancer: patients with intermediate and poor prognosis Treatment is palliative. Chemotherapy increases the median survival from around 2 or 3 months to 9 months, and it is usually associated with good symptom palliation. Chemotherapy, therefore, is recommended in patients with a good performance status (WHO 0 to 2). A combination of carboplatin or cisplatin with etoposide and a maximum of four cycles is probably the most effective regimen. Patients with poor PS (WHO 3 to 4) are not likely to benefit and the risk of treatment-related death is high. Palliative RT may be more appropriate for these patients.
External beam radiotherapy in lung cancer Radical radiotherapy for NSCLC 3D conformal treatment planning is usually recommended for patients undergoing radical treatment. Patients are treated in the supine position and immobilised, lying comfortably with arms above the head using an armpole or with hands behind the head and with the head on a low sponge. The patient is set up by reference to tattoos on the skin and bony landmarks, usually the sternum and mid-axillary lines. The patient is CT scanned at 5 mm intervals in the treatment position. Using the 3D planning system, the 322
GTV should be outlined on all CT slices. Defining the GTV may be difficult because the tumour may be hard to differentiate from areas of distal atelectasis and because of the uncertainties in identifying nodal involvement. The GTV should include the visible tumour and any clearly involved lymph nodes (size > 1 cm), and a recent contrast-enhanced CT scan and PET study should be available when planning the volume. A 5 mm margin is added to encompass microscopic spread (CTV) and, to produce a PTV, a margin of 1.5 to 2 cm around the GTV is recommended. Respiratory movement is a major factor, and the position of the tumour within the thorax can change the required margins. In general, the more peripheral and lower lobe tumours are likely to move more with respiration than upper lobe tumours or tumours adjacent to or invading the mediastinum. The proximity of the spinal cord may also influence the margins. A margin of 2 cm S–I is suggested due to the larger respiratory motion in this plane with 1.5 cm margins usually acceptable in other directions. There are sophisticated systems for ‘gating’ RT to fixed periods of the respiratory cycle and methods for ensuring the patients hold their breath during treatment, but they are not in wide use and are of uncertain clinical benefit. When reviewing the plan, the DVHs should be viewed and assessment will enable both maximum cord dose and V20 to be calculated. The spinal cord dose should not exceed 48 Gy in two Gy fractions. The dose received by the lungs is important because long-term lung function is likely to be affected by radiation fibrosis. The patient’s current lung function and the likely effects of the treatment need to be considered. In a fit patient with adequate lung function the V20 should not exceed 35%. Plan and field arrangement usually involves a threeor four-field plan with wedges but depends very much on the size of the PTV and its position within the thorax. Occasionally two fields are used, for example, in a Pancoast tumour. It is useful to verify all fields by simulation where respiratory movement can be visualised. Portal images on set should be checked by comparison with digitally reconstructed radiographs or simulator films to ensure the reproducibility of the setup. If a discrepancy is found, it can be adjusted and re-checked early on in treatment. Dose, fractionation and energy for the procedure are as follow: 6 to 10 MV photons are used to provide homogeneity across the PTV. Patients should have CHART (54 Gy in 36 fractions over 12 days) if available. Other options are 52 to 55 Gy in 20 fractions over 4 weeks or 60 to 66 Gy in 30 to 33 fractions over 6 to 6.5 weeks.
Lung
Isodose % 95 70 40 20 10
Table 28.6. Side effects from radiotherapy for lung cancer Side effect
Management
Acute Anorexia and
Antiemetics
nausea Dry cough
Simple linctus, codeine linctus
Oesophagitis
Supportive: analgesia, mucilage, rarely parenteral feeding
Figure 28.1. A radiotherapy treatment plan for a peripherally based non-small-cell carcinoma of the lung.
Late Radiation
Figure 28.1 shows an RT plan for a peripherally located carcinoma of the bronchus. Side effects from RT are shown in Table 28.6. Pancoast tumours should be treated with a two phase parallel opposed field technique covering the whole of the vertebral bodies for the first phase. The volume should be reduced once the tolerance dose to the spinal cord has been reached, to avoid the spinal cord but including the ipsilateral nerve roots.
pneumonitis
8–12 weeks after treatment: acute dyspnoea, cough and tightness; CXR can show hazy area in field Treat with oxygen, steroids: usually prednisolone up to 60 mg reduced over 2 weeks and antibiotics
Lung fibrosis
Usually asymptomatic but may cause increased breathlessness; oxygen if needed
Myelopathy
Rare, no specific treatment
Palliative radiotherapy A simple parallel-opposed pair is used to treat the thorax but care should be taken to consider spinal cord tolerance. Typical field sizes are less than 12 × 12 cm, especially when large fractions are used, and lead shielding can be used to reduce the lung volume that is irradiated.
Brachytherapy in lung cancer There are a few patients for whom bronchoscopic brachytherapy may be appropriate: r Patients with a very small, superficial and localised endoluminal T1 NSCLC who are medically inoperable. r Patients with a symptomatic endoluminal tumour who have had previous external beam RT to the limits of the lung and/or spinal cord tolerance. An RCT has compared EBRT with brachytherapy as a first-line palliative treatment of inoperable NSCLC and showed the superiority of EBRT, which exhibited better palliation of symptoms and less serious morbidity (Stout et al., 2000).
Technique Patients need to be fit enough and have lung function adequate enough to tolerate bronchoscopy and insertion of a fine-bore tube into the relevant bronchus. The patient is bronchoscoped and the tumour is identified
and its position related to the carina. A fine-bore polythene tube is passed via the suction channel into the relevant bronchus, past the site of the tumour, and the bronchoscope is then withdrawn over the catheter. The catheter is then fixed to the skin at the nasal orifice. The catheter position is confirmed by insertion of a dummy seed guide wire under fluoroscopy and the target area is outlined. An HDR afterloading technique is used with an 192 Ir source and therefore the treatment can be given as an outpatient procedure.
Dose For palliation, a single fraction is well tolerated. The usual dose is 10 to 15 Gy in a single-fraction HDR at 1 cm from the centre of the source.
Brachytherapy with radical intent The optimum regimen of radical-intent brachytherapy has not been determined. A number of fractionated regimes have been used, ranging between 22.5 and 42 Gy in 3 to 6 fractions, weekly. A series from the Christie Hospital also used single fractions in this group of 15 or 20 Gy with improved local control but more serious late morbidity (massive fatal haemoptysis) in the 20 Gy group. 323
Fergus Macbeth and Carys Morgan
Table 28.7. Survival following radical radiotherapy for
Table 28.8. Survival following palliative treatment for
non-small-cell lung cancer
non-small-cell lung cancer
Two-year
Three-year
Five-year
Stage
survival
survival
survival
I and II
30% (CHART)
20% (CHART)
21% (60 Gy/
13% (60 Gy/
30 F)
30 F)
Small vol. IIIA
One-year Stage
Treatment
survival
IIIA (bulky)
RT 39 Gy/13 F
36%
RT 17 Gy/ 2 F
31%
BSC
10%
IIIB/IV 10% (60 Gy/ 30 F)
Cisplatin-based chemotherapy
20%
Platinum + third generation
?35%
CHART = continuous hyperfractionated accelerated
BSC = best supportive care. Adapted from Medical
radiotherapy, F = fractions. Adapted from Aup´erin et al.
Research Council Lung Cancer Working Party (1996), Non
(2006) and Saunders et al. (1999).
Small Cell Lung Cancer Collaborative Group (1995), and Schiller et al. (2002).
Prognosis Non-small-cell lung cancer Clearly the most significant prognostic factor is stage of disease. Five-year survival drops off rapidly with increasing stage, as shown by the surgical series in Table 28.4 (Mountain, 1997). The addition of adjuvant chemotherapy improves overall survival in surgical patients by about 5%. Overall 5-year survival is increased from 40 to 45% and treatment is likely to be of more benefit in patients with stage II and III disease. Results for inoperable disease treated with radical RT are shown in Table 28.7, and results for palliative treatment for NSCLC are shown in Table 28.8. Other factors associated with a poorer prognosis include the following: r Poor performance status. r Greater than 10% weight loss. r Severe symptoms. r Large-cell histology. r Bone, liver or subcutaneous metastases. r Male gender.
Small-cell lung cancer Patients with a good prognosis experience a median survival of 18 to 20 months (15% 3-year survival rate), intermediate-prognosis patients experience a median survival of 9 months (≤5% 3-year survival), and poorprognosis patients experience a median survival of less than 3 months.
324
Areas of current interest and clinical trials Current areas of interests and controversy in NSCLC include the following: r Neoadjuvant chemotherapy. r Adjuvant chemotherapy for stage I disease. r Timing of chemotherapy and radical RT. r Dose escalation and IMRT for radical RT. r Most effective chemotherapy regimen in advanced disease. r Second-line chemotherapy. r Use of erlotinib and other new biological agents. These current trials may answer some of these questions. LU22 is an RCT involving surgical resection with or without preoperative chemotherapy in patients with operable NSCLC of any stage. SOCCAR is a phase III RCT of sequential chemotherapy followed by radical RT or concurrent chemoradiotherapy followed by chemotherapy in patients with inoperable stage III NSCLC and good PS. GRIN is a phase III RCT of radical RT with or without gemcitabine in stage T1–2 N0–1 M0 NSCLC. INCH is a phase II/III trial of induction chemotherapy followed by CHART compared to CHART alone in patients with inoperable NSCLC. BTOG 2 is a phase III RCT of gemcitabine with either cisplatin 80 mg/m2 , cisplatin 50 mg/m2 , or carboplatin AUC6 in stage IIIB/IV NSCLC. LLCG Study 14 is a phase II/III randomised, doubleblind placebo-controlled trial of gemcitabine/carbo-
Lung
platin with or without thalidomide in patients with advanced NSCLC. TOPICAL is a randomised placebo-controlled trial of erlotinib in patients with advanced NSCLC who are unsuitable for chemotherapy. FRAGMATIC is a randomised trial of 6 months of dalteparin or no dalteparin in newly diagnosed patients with lung cancer of all stages and all histologies. MUSICAL is a trial that uses mediastinal endoscopic ultrasound in the diagnosis and staging of carcinoma of the lung. QUARTZ is a phase III multicentre RCT that assesses whether BSC alone is as effective as BSC and wholebrain RT in the treatment of patients with inoperable brain metastases from NSCLC. Current interests with regard to small-cell lung cancer include timing of consolidation RT and hyperfractionation, and the use of chemotherapy in patients with PS 3 and 4. SCLC current trials are as follows: LLCG Study 10 is a phase III randomised comparison of gemcitabine/carboplatin with cisplatin/etoposide in SCLC. LLCG Study 12 is a phase III randomised, doubleblind placebo-controlled trial of carboplatin/etoposide plus thalidomide in SCLC. LU21 is a randomised trial of ifosfamide, carboplatin, and etoposide with mid-cycle vincristine (VICE) versus standard practice chemotherapy in limited SCLC and good PS. The Thalidomide SCLC trial studies thalidomide in SCLC. TR8SCLC is a randomised study of the timing of thoracic irradiation in SCLC.
Rare tumour types Bronchial carcinoid Bronchial carcinoids are rare neuroendocrine tumours (previously described as bronchial adenomas), which comprise fewer than 5% of primary lung tumours. They tend to occur in a younger age group (≤40 years) and are not associated with smoking. Three types are described but really represent a continuum: r Central carcinoid, the most common type, is polypoid and vascular and may be hormonally silent or produce ACTH. Lymph node metastases occur in 5% of patients
and distant spread is rare. The 5-year survival is 70 to 80%. r Peripheral carcinoid usually occurs as multiple tumours, has an excellent prognosis and rarely metastasises. r Atypical carcinoids have an increased mitotic rate and exhibit nuclear hyperchromasia and necrosis. It can be difficult to differentiate atypical carcinoids from small-cell carcinoma. Lymph node spread occurs in 60% of patients, and metastases are more common. In the absence of metastatic disease, the primary treatment for bronchial carcinoids is surgery. RT is rarely used except in palliation of metastatic disease. Combination chemotherapy with regimens similar to those used in SCLC (e.g. carboplatin and etoposide) has been used in metastatic disease but with relatively poor results. Carcinoid syndrome more commonly occurs with metastatic GI carcinoid but can rarely occur in bronchial carcinoid. It consists of paroxsysmal flushing, bronchoconstriction, diarrhoea, abdominal pain and right heart failure. Skin lesions can also occur. These are due to secretion of serotonin and are diagnosed by elevated 5-hydroxyindoleacetic acid levels in a 24-hour urine collection. Blood chromogranin A assay can be an additional confirmatory test, and octreotide (a somatostatin analogue) is useful for symptomatic relief in these cases. Carcinoid tumours are discussed further in Chapter 36 (see p. 418).
Thymoma Thymoma is an uncommon tumour that presents as an anterior mediastinal mass, and it arises from cells of epithelial and lymphocytic origin. The majority are benign but 30% of tumours are thymic carcinomas. Thymomas rarely metastasise but they can be locally invasive of the pericardium, myocardium, lung, sternum, and great vessels. ‘Degree of invasion’ is the best indicator of overall survival, rather than the cell type, and is reflected in the Masoaka staging system (see Table 28.9). Thymomas can be associated with the following paraneoplastic syndromes: r Myasthenia gravis – occurs in 30 to 50% of patients with thymoma and 70% improve after thymectomy. r Red cell aplasia – occurs in 5% of patients. r Acquired hypogammaglobulinaemia – occurs in 5 to 10%.
325
Fergus Macbeth and Carys Morgan
Table 28.9. The Masaoka staging system for thymoma Stage
Description
Stage 1
No microscopic capsular invasion
Stage IIa
Microscopic invasion of capsule
Stage IIb
Microscopic invasion of mediastinal fat/pleura
Stage III
Invasion of adjacent structures
Stage IVa
Pleural and pericardial metastases
Stage IVb
Distant metastases
Adapted from Masaoka et al. (1981).
r Rarely, ectopic Cushing’s syndrome, polymyositis/ dermatomyositis, SLE and hypertrophic pulmonary osteoarthropathy. The main modality of treatment is surgery and completeness of resection correlates with cure rate. Non-invasive (stage 1) tumours can be treated with surgery alone but postoperative RT should be considered for invasive tumours or incomplete excision. The usual dose is 45 to 50 Gy in 2 Gy fractions to the tumour bed. The role of chemotherapy is unclear but cisplatin-based combination chemotherapy has been used in metastatic disease with some partial responses reported.
REFERENCES Arriagada, R., Pignon, J. P., Laplanche, A. et al. (1997). Prophylactic cranial irradiation for small-cell lung cancer. Lancet, 349, 138. Arriagada, R., Bergman, B., Dunant, A. et al. (2004). Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N. Engl. J. Med., 350, 351–60. Aup´erin, A., Le P´echoux, C., Pignon, J. P. et al. (2006). Meta-Analysis of Cisplatin/carboplatin based Concomitant Chemotherapy in non-small cell Lung Cancer (MAC3-LC) Group. Concomitant radio-chemotherapy based on platin compounds in patients with locally advanced non-small cell lung cancer (NSCLC): a meta-analysis of individual data from 1764 patients. Ann. Oncol. 17, 473–83. De Ruysscher, D., Pijls-Johannesma, M., Vansteenkiste, J. et al. (2006). Systematic review and meta-analysis of randomised, controlled trials of the timing of chest radiotherapy in patients with limited-stage, small-cell lung cancer. Ann. Oncol., 17, 543–52. Doll, R. and Hill, A. B. (1950). Smoking and carcinoma of the lung. Preliminary report. B. M. J., 2 (4682), 739–48.
326
Falk, S. J., Girling, D. J., White, R. J. et al. (2002). Immediate versus delayed palliative thoracic radiotherapy in patients with unresectable locally advanced non-small cell lung cancer and minimal thoracic symptoms: randomised controlled trial. B. M. J.. 325, 465–72. Franklin, W. A. (2000). Diagnosis of lung cancer. Pathology of invasive and preinvasive neoplasia. Chest, 117 (4 Suppl. 1), 80S–89S. Graham, M. V., Purdy, J. A., Emami, B. et al. (1999). Clinical dose-volume histogram analysis for pneumonitis after 3D treatment for non small cell lung cancer (NSCLC). Int. J. Radiat. Oncol. Biol. Phys., 45, 323–9. Hanna, N., Shepherd, F. A., Fossella, F. V. et al. (2004). Randomized phase III trial of pemetrexed versus docetaxel in patients with non small cell lung cancer previously treated with chemotherapy. J.Clin. Oncol., 22, 1589–97. Masaoka, A., Monden, Y., Nakahara, K. et al. (1981). Follow-up study of thymomas with special reference to their clinical stages. Cancer, 48, 2485–92. Medical Research Council Lung Cancer Working Party. (1992). A Medical Research Council (MRC) randomised trial of palliative radiotherapy with two fractions or a single fraction in patients with inoperable non-small-cell lung cancer (NSCLC) and poor performance status. Br. J. Cancer, 65, 934–41. Macbeth, F. R., Bolger, J. J., Hopwood, P. et al. (1996). Randomised trial of palliative two-fraction versus more intensive thirteen fraction radiotherapy for patients with inoperable non-small cell lung cancer and good performance status. Clin. Oncol., 8, 167–75. Mountain CF. (1997). Revisions in the International System for Staging Lung Cancer. Chest, 111, 1710–17. Non Small Cell Lung Cancer Collaborative Group. (1995). Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised clinical trials. B. M. J., 311, 899–909. Pignon, J. P. and Arriagada, R. (1992). Role of thoracic radiotherapy in limited-stage small-cell lung cancer: quantitative review based on the literature versus meta-analysis based on individual data. J. Clin. Oncol., 10, 1819–20. Pignon, J. P. Tribodet, G. V., Scagliotti, J. et al. (2006). Lung Adjuvant Cisplatin Evaluation (LACE): a pooled analysis of five randomized clinical trials including 4584 patients. J. Clin. Oncol., 24, (18 Suppl.), 7008. PORT Meta-analysis Trialists Group. (1998). Postoperative radiotherapy in non-small cell lung cancer: systematic review and meta-analysis of individual patient data from nine randomised controlled trials. Lancet, 352, 257–63. Pujol, J. L., Carestia, L. and Daur`es, J. P. (2000). Is there a case for cisplatin in the treatment of small-cell lung cancer? A meta-analysis of randomised trials of a cisplatin-containing regimen versus a regimen without this alkylating agent. Br. J. Cancer. 83, 8–15. Saunders, M., Dische, S., Barrett, A. et al. (1999). Continuous hyperfractionated accelerated radiotherapy (CHART) versus conventional radiotherapy in non-small cell lung cancer: mature data from the randomised multicentre trial. Radiother. Oncol., 52, 137–48.
Lung
Schiller, J., Harrington, D., Belani, C. P. et al. (2002). Comparison of four chemotherapy regimens for advanced non-small cell lung cancer. N. Engl. J. Med., 346, 92–8. Shepherd, F. A., Rodriguez Pereira, J. R., Ciuleanu, T. et al. (2005). Erlotinib in previously treated non small cell lung cancer. N. Engl. J. Med., 353, 123–32. Spiro, S. G., Rudd, R. M., Souhami, R. L. et al. (2004). Chemotherapy versus supportive care in advanced non-small cell lung cancer: improved survival without detriment to quality of life. Thorax, 59, 828–36. Stout, R., Barber, P. V., Burt, P. A. et al. (2000). Clinical and quality of
life outcomes in the first United Kingdom randomised trial of endobronchial brachytherapy (intraluminal therapy) vs. external beam radiotherapy in the palliative treatment of inoperable non-small cell lung cancer. Radiother. Oncol., 56, 323–7. Turrisi, A. T. 3rd, Kim, K., Blum, R. et al. (1999). Twice-daily compared with once-daily thoracic radiotherapy in limited small-cell lung cancer treated concurrently with cisplatin and etoposide. N. Engl. J. Med., 340, 265–71. UICC. (2002). TNM Classification of Malignant Tumours, Ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 97–103.
327
29
MESOTHELIOMA Louise Hanna and Fergus Macbeth
Introduction Mesothelioma is a challenging disease for patients, relatives and doctors. It is strongly linked to exposure to asbestos and, despite a ban on the use of asbestos in the 1960s in the UK, its incidence is continuing to rise because of the long latent period between exposure and development of the disease. The incidence of mesothelioma is expected to peak around 2020 (Peto et al., 1995). High-quality clinical research evidence to guide treatment decisions is lacking, and there is an urgent need to identify and evaluate new chemotherapeutic agents and treatment strategies. Of all mesotheliomas, 94.5% affect the pleura, 5.1% affect the peritoneum and 0.4% affect the pericardium (Yates et al., 1997). Mesotheliomas may also arise in the tunica vaginalis.
Types of pleural tumour The most common tumours affecting the pleura are metastatic from other sites. Table 29.1 shows the types of tumour that affect the pleura.
Anatomy of the pleurae The pleurae surround both lungs and each is divided into two parts. The parietal pleura lines the cavity bounded by the thoracic wall, diaphragm and lateral mediastinum. The visceral pleura covers the surface of the lungs and extends into the interlobular fissures. The two layers are separated by a small amount of pleural fluid, which reduces friction during respiration. The parietal pleura is divided into regions: cervical, costal, diaphragmatic and mediastinal. The costal pleura is supplied by intercostal nerves and is sensitive to pain and touch.
328
Pleural mesothelioma Incidence and epidemiology The incidence of pleural mesothelioma is 6.2 in 100 000 per year for males and 1.1 in 100 000 per year for females (National Statistics, 2005). Approximately 2800 cases are diagnosed per year in England. The maleto-female ratio for the disease is 5.5:1, and the mortalityto-incidence ratio is 0.86 for males and 0.87 for females. Disease incidence is currently increasing in the UK but is expected to peak around 2020 and then decrease.
Risk factors and aetiology Asbestos exposure Exposure to asbestos is the single most important aetiological factor in mesothelioma. Around 90% of patients with mesothelioma report a history of occupational exposure (Yates et al., 1997). The risk of developing mesothelioma is related to the level of exposure and to the type of asbestos fibre; one in ten workers heavily exposed early in their working life may die of mesothelioma (Peto et al., 1982). But the latent interval between exposure and diagnosis is often very long; the mean latent interval between asbestos exposure and death is 41 years (Yates et al., 1997). The use of asbestos has now been banned in the UK for construction and refurbishment, but asbestos is still present in many buildings. There are two types of asbestos: amphibole and chrysotile. The most potent, amphibole fibres, include crocidolite (blue asbestos) and amosite (brown asbestos). The less potent type is chrysotile (white asbestos). Asbestos-containing products include the following: r Roofing materials. r Fire protection materials including blankets and curtains.
Mesothelioma
Table 29.1. Types of tumour affecting the pleura
Table 29.2. Pathological features of malignant mesothelioma
Type
Examples
Benign
Pleural fibroma (‘benign mesothelioma’) Adenomatoid tumour
Malignant primary
Features
Description
Macroscopic
Diffuse growth pattern affecting visceral and parietal surfaces as a thick rind;
Malignant mesothelioma
early involvement may be as discrete
Epithelioid
small nodules or plaques; rarely, a
Sarcomatoid Desmoplastic Biphasic Malignant secondary
solitary pleural-based mass Microscopic
Epithelioid – tubules, acini, papillae or sheets
Frequently adenocarcinoma (e.g. from
Sarcomatoid – spindled pattern
ovary, breast, lung)
resembling fibrosarcoma or malignant fibrous histiocytoma Desmoplastic – sarcomatoid with >50%
r Electrical casings. r Water pipe products. r Numerous others. Occupational exposure to asbestos can occur in construction, renovation and demolition workers; ship builders and dock workers; carpenters and electricians; or manufacturers of asbestos products. Non-occupational exposure can occur in women who washed their partner’s work clothes, from contaminated soil, from proximity to naturally occurring asbestos or from proximity to industrial plants where asbestos is used. Clinical effects of asbestos exposure include the following: r Asbestosis – lung fibrosis. r Pleural plaques. r Diffuse pleural fibrosis. r Pleural effusion. r Mesothelioma. r Lung cancer. r Asbestos is a possible factor in other cancers such as laryngeal, oropharyngeal, gastrointestinal and renal. The mechanism of carcinogenesis is not fully understood, but asbestos fibres are known to induce DNA and chromosome damage. Simian virus 40 (SV 40) may act as a co-factor via transforming activity of two viral proteins, large T and small t antigens (Jaurand and Fleury-Feith, 2005).
dense collagen stroma Biphasic – combined epithelial and sarcomatoid, each comprising at least 10% of the tumour; greater tumour sampling increases the ability to detect biphasic pattern Other types – tumours with heterologous elements; adenomatoid tumour-like; lymphocytoid; myxoid stroma deciduoid; multicystic; clear cell; small cell; poorly differentiated; anaplastic Adapted from Travis et al. (1999).
r Ionising radiation. r Pleural scars. r SV 40 that accidentally contaminated millions of polio vaccines in late 1950s and early 1960s.
Pathology The main pathological types of mesothelioma are epithelioid, sarcomatoid and biphasic (Travis et al., 1999). The term desmoplastic refers to sarcomatoid with dense collagen stroma. The pathological features of mesothelioma are shown in Table 29.2.
Other possible aetiological factors
Immunocytochemistry
Other possible aetiological factors for disease include (Hubbard, 1997): r Erionite rock (in central Turkey). r Sugar cane.
Epithelial mesotheliomas rarely stain positive for intracellular mucin or the carcinomatous epitopes CEA, B72.3 and CD-15 (Leu-M1) and, thus, may be distinguished from adenocarcinomas. Mesotheliomas tend to 329
Louise Hanna and Fergus Macbeth
Table 29.3. Spread of pleural mesothelioma Route of spread
Sites
Local
Chest wall, intercostal nerves, ribs Mediastinum, including pericardium
r Abdominal pain or ascites. r Cervical lymphadenopathy. r Haemoptysis. r Hoarseness of voice. r Superior vena cava obstruction. r Dysphagia. r Cardiac tamponade.
and oesophagus Vertebral body Diaphragm and liver Lymph node
Mediastinal nodes Internal mammary nodes Cervical nodes
Metastatic
Bone Contralateral lung Liver Brain
stain positive for keratin; thus, the sarcomatoid variant may be distinguished from chest wall sarcoma.
Spread The predominant mode of spread is local. However, distant metastases do commonly occur as a late event but are frequently asymptomatic. Routes of mesothelioma spread are shown in Table 29.3.
Clinical presentation Typically mesothelioma presents with symptoms and signs relating to the primary tumour (Parker and Neville, 2003): r Chest pain, which can be dull, diffuse or pleuritic and may have a neuropathic component. r Shortness of breath, often multifactorial, caused by pleural fluid, pleural thickening, thoracic restriction and lung encasement or co-morbid conditions (e.g. airflow obstruction, cardiac dysfunction). r Incidental finding of pleural effusion or pleural thickening. r Weight loss. r Finger clubbing. r Profuse sweating. Other, less common clinical features of mesothelioma include the following (BTS Statement, 2001): r Chest wall mass. 330
Investigation and staging Techniques for imaging and pathological confirmation of mesothelioma can be found in the BTS Statement (2001).
Imaging A chest X-ray can show pleural thickening, a pleuralbased mass, or effusion. Other features of asbestos exposure may be present. Ultrasound of the chest can be used to distinguish between solid and fluid lesions. A CT or MRI scan demonstrates invasion of the chest wall and enlarged mediastinal lymph nodes. Features suggestive of mesothelioma include nodular pleural thickening, thickening of the mediastinal pleura, constriction of the hemithorax and loss of lung volume. It can be difficult to quantify the amount of disease from a CT scan, which is relevant to assessing a treatment response.
Pathological confirmation Pathological confirmation of disease is performed via cytological examination of pleural fluid, blind percutaneous biopsy with an Abrams’ needle, an ultrasoundor CT-guided biopsy, thoracoscopic biopsy or open biopsy.
Staging classification The most commonly used staging system is the International Mesothelioma Interest Group (IMIG) staging system, which is shown in Table 29.4 (Patz et al., 1996).
Treatment overview Patients should be managed within the context of a specialist multidisciplinary team. Those patients with low-volume epithelioid or biphasic tumours may be suitable for radical surgery. For patients with sarcomatoid or more advanced tumours, the aim is palliation. For interventions that have a potential to cause
Mesothelioma
Table 29.4. The International Mesothelioma Interest Group (IMIG) staging system Stage
Description
T1a
Tumour limited to the ipsilateral parietal including mediastinal, diaphragmatic pleura, no involvement of the
T1b
Tumour involving the ipsilateral parietal including mediastinal and diaphragmatic pleura, scattered foci of tumour
T2
Tumour involving each ipsilateral pleural surface (parietal, mediastinal, diaphragmatic and visceral pleura) with at
visceral pleura also involving the visceral pleura least one of the following features: involvement of diaphragmatic muscle; confluent visceral pleural tumour (including fissures) or extension of tumour from visceral pleura into the underlying pulmonary parenchyma T3
Tumour involving each ipsilateral pleural surface (parietal, mediastinal, diaphragmatic and visceral pleura) with at least one of the following features: involvement of the endothoracic fascia; extension into the mediastinal fat; solitary; completely resectable focus of tumour extending into the soft tissues of the chest wall; non-transmural involvement of the pericardium
T4
Tumour involving each ipsilateral pleural surface (parietal, mediastinal, diaphragmatic and visceral) with at least one of the following features: diffuse extension or multifocal masses of tumour in the chest wall with or without associated rib destruction; direct transdiaphragmatic extension of tumour to the peritoneum; direct extension of tumour to the contralateral pleura; direct extension of tumour to one or more mediastinal organs; direct extension of tumour into the spine; tumour extending through to the internal surface of the pericardium with or without a pericardial effusion; tumour involving the myocardium
N1
Metastases in the ipsilateral bronchopulmonary or hilar lymph nodes
N2
Metastases in the subcarinal or the ipsilateral mediastinal lymph nodes, including the ipsilateral internal mammary
N3
Metastases in the contralateral mediastinal, contralateral internal mammary, ipsilateral or contralateral
nodes supraclavicular lymph nodes M0
No distant metastases
M1
Distant metastases present
From Patz et al. (1996).
significant morbidity, there is a lack of evidence from randomised controlled trials, resulting in controversy surrounding the balance between the potential benefit and toxicity. Patients and relatives should be informed of the possibilities for financial compensation. Deaths due to mesothelioma must be reported to the coroner, and a postmortem examination is usually required.
Surgery For more information on surgical procedures, see the BTS Statement (2001).
Extrapleural pneumonectomy Extrapleural pneumonectomy is radical surgery that involves the removal of lung, and parietal and visceral
pleura, together with the hemi-diaphragm and part of the pericardium.
Pleurectomy/decortication Pleurectomy/decortication is generally considered to be a palliative procedure used to debulk a tumour. It also prevents the recurrence of pleural effusion.
Video-assisted thoracoscopic surgery (VATS) VATS allows partial pleurectomy as a debulking procedure, and it can reduce fluid re-accumulation; it is associated with low operative morbidity and mortality.
Talc pleurodesis Talc pleurodesis is a well-tolerated procedure and reduces fluid accumulation. It may be carried out as part of a VATS procedure or following tube drainage of pleural fluid (BTS Statement, 2001). 331
Louise Hanna and Fergus Macbeth
Pleuroperitoneal shunt A pleuroperitoneal shunt is considered for patients with persistent pleural effusion, but it has a high incidence of shunt blockage and peritoneal seedlings.
plex and time-consuming. At the present time, however, radiotherapy in combination with surgery remains under investigation.
Role of chemotherapy
Radiotherapy Prophylactic radiotherapy to drainage or biopsy sites Invasive procedures used in obtaining a biopsy or draining fluid result in a 40% risk of seeding along the needle track. Local radiotherapy can effectively eliminate this risk (Boutin et al., 1995). A typical fractionation schedule is 21 Gy in 3 fractions using 250 to 300 KV photons.
Palliative radiotherapy Wide-field radiotherapy using parallel-opposed fields to a dose of 30 Gy in 10 fractions can improve pain in around 60% of patients (Bissett et al., 1991), but the effect is generally short-lived. Breathlessness does not improve. Obvious and symptomatic chest wall masses and localised pain may be treated with short courses of palliative radiotherapy (e.g. 8 to 10 Gy in a single fraction or 16 to 17 Gy in 2 fractions).
Postoperative radiotherapy following extrapleural pneumonectomy Rusch et al. (2001) reported a phase II trial of postoperative radiotherapy in 54 patients who had undergone extrapleural pneumonectomy. The local recurrence rate was observed to be much lower than expected in patients undergoing surgical resection alone. The radiotherapy technique has been discussed by Senan and van de Pol (2004). The CTV included the entire hemithorax, the ipsilateral mediastinal pleura, the ipsilateral pericardial surface, and the full thickness of the thorax at the site of the thoracotomy incisions and sites of chest drains. A margin of 1 cm was used to generate the PTV. The critical structures included the heart, spinal cord, oesophagus, contralateral lung, liver, kidney and stomach. It is likely that the ipsilateral kidney will receive a dose exceeding tolerance. Conformal planning using electrons, photons and blocking of critical structures achieved an adequate dose distribution. The delivered dose was 54 Gy in 30 fractions. The treatment was well tolerated; the toxicity of fatigue and oesophagitis was grade 0 to 2. More complex planning using IMRT may improve the dose distribution but it is highly com-
332
Palliative chemotherapy in mesothelioma has been disappointing; response rates have been between 10 and 20% with conventional therapeutic agents such as cisplatin, doxorubicin, bleomycin, mitomycin and vinorelbine. A randomised controlled trial of active symptom control with or without chemotherapy (MS01) has been completed recently and results will be reported in 2007. Recent interest has focussed on the novel antimetabolite pemetrexed, which has shown a 41.3% response rate when given in combination with cisplatin (Vogelzang et al., 2003) and a modest increase in survival compared to the use of cisplatin alone.
Palliative treatments For treatment of dyspnoea, consider breathing exercises, relaxation, home oxygen, opiates and benzodiazepines and exclude a pleural effusion. For chest pain, consider analgesics according to the WHO ladder, non-steroidal anti-inflammatory drugs, tricyclic antidepressants or anticonvulsants. For cough, consider simple linctus, codeine linctus, oral steroids and opiates or saline nebulisers for sticky sputum.
Prognosis Survival The median survival of patients with mesothelioma is around 8 months (Van Gelder et al., 1994) and the 5-year survival is 4.7% (Gatta et al., 2006).
Prognostic factors The following factors are associated with a poor prognosis (Van Gelder et al., 1994): r Sarcomatoid-type tumours. r Biphasic type tumours. r Advanced disease stage. r Old age.
Compensation Patients and relatives should be given advice about compensation and benefits for mesothelioma patients.
Mesothelioma
Information can be obtained via the specialist nurse, social worker or solicitor. Possibilities for benefits are listed here (see Cancerbackup; www.cancerbackup.org. uk, accessed September 2006): r Industrial Injuries Disablement Benefit – a weekly allowance for patients with mesothelioma who can show they were occupationally exposed to asbestos. r Civil law personal injury claim against the previous employer. r Pneumoconiosis Worker’s Compensation if the employer has gone out of business. r Statutory Sick Pay for those with adequate National Insurance contributions, paid for a maximum of 28 weeks. r Incapacity Benefit, paid after 28 weeks of Statutory Sick Pay. r Disability Living Allowance. r Attendance Allowance. r Constant Attendance Allowance. r Exceptionally Severe Disablement Allowance.
Areas of current interest Trimodality therapy is a combination of preoperative chemotherapy, extrapleural pneumonectomy and postoperative radiotherapy and is an attempt to improve upon the dismal results achieved by single modality treatment alone. A series reported by Sugarbaker et al. (1996) showed an overall survival of 45% at 2 years and 22% at 5 years, with a 5% operative mortality rate. The results were best in patients with early stage epithelioid tumours (Sugarbaker et al., 1995). Trimodality therapy is, therefore, feasible in a highly selected group of patients, but the results from randomised controlled trials are lacking.
Ongoing clinical trials The following trials were among those listed as open by the National Cancer Research Network (www.ncrn. org.uk, accessed September, 2006): MALCS is a population-based case-control study of mesothelioma and lung cancer and how they relate to occupation among British men and women under the age of 60. The Mesothelioma and Radical Surgery (MARS) trial is a pilot study to determine the feasibility and acceptability of performing a randomised trial comparing extrapleural pneumonectomy (EPP) against no-EPP
surgery within the context of trimodality therapy (chemotherapy, surgery and postoperative radiotherapy). The nebulised furosemide study is to determine whether nebulised furosemide improves severe breathlessness in patients with lung cancer. VATS is a prospective randomised controlled trial of video-assisted thoracoscopic (VATS) cytoreductive pleurectomy compared to talc pleurodesis in patients with suspected or proven malignant mesothelioma.
REFERENCES Bissett, D., Macbeth, F. R. and Cram, I. (1991). The role of palliative radiotherapy in malignant mesothelioma. Clin. Oncol. (R. Coll. Radiol.), 3, 315–7. Boutin, C., Rey, F. and Viallat, J. R. (1995). Prevention of malignant seeding after invasive diagnostic procedures in patients with pleural mesothelioma. A randomized trial of local radiotherapy. Chest, 108, 754–8. BTS Statement. (2001). Statement on malignant mesothelioma in the United Kingdom. Thorax, 56, 250–65. Gatta, G., Ciccolallo, L., Kunkler, I. et al. (2006). Survival from rare cancer in adults: a population-based study. Lancet Oncol., 7, 132–40. Hubbard, R. (1997). The aetiology of mesothelioma: are risk factors other than asbestos exposure important? Thorax, 52, 496–7. Jaurand, M. C. and Fleury-Feith, J. (2005). Pathogenesis of malignant pleural mesothelioma. Respirology, 10, 2–8. National Statistics. (2005). Cancer Statistics and Registrations, Series MB1 no. 34. London: Office for National Statistics. Parker, C. and Neville, E. (2003). Lung cancer 8: Management of malignant mesothelioma. Thorax, 58, 809–13. Patz, E. F., Rusch, V. W. and Heelan, R. (1996). The proposed new international TNM staging system for malignant pleural mesothelioma: application to imaging. Am. J. Roentgen, 166, 323–7. Peto, J., Seidman, H. and Selikoff, I. J. (1982). Mesothelioma mortality in asbestos workers: implications for models of carcinogenesis and risk assessment. Br. J. Cancer, 45, 124–35. Peto, J., Hodgson, J. T., Matthews, F. E. et al. (1995). Continuing increase in mesothelioma mortality in Britain. Lancet, 345, 535–9. Rusch, V. W., Rosenzweig, K., Venkatraman, E. et al. (2001). A phase II trial of surgical resection and adjuvant high-dose hemithoracic radiation for malignant pleural mesothelioma. J. Thorac. Cardiovasc. Surg., 122, 788–95. Senan, S. and van de Pol, M. (2004). Considerations for post-operative radiotherapy to the hemithorax following extrapleural pneumonectomy in malignant pleural mesothelioma. Lung Cancer, 45 (Suppl. 1), S93–6. Sugarbaker, D. J., Jaklitsch, M. T. and Liptay, M. J. (1995). Mesothelioma and radical multimodality therapy: who benefits? Chest, 107, 345S–50S. Sugarbaker, D. J., Garcia, J. P., Richards, W. G. et al. (1996). Extrapleural pneumonectomy in the multimodality therapy of
333
Louise Hanna and Fergus Macbeth
malignant pleural mesothelioma. Results in 120 consecutive patients. Ann. Surg., 224, 288–94. Travis, W. D., Colby, T. V., Corrin, B. et al. (1999). Histological Typing of Lung and Pleural Tumours, 3rd edn. Berlin, New York: Springer, pp. 51–4. Van Gelder, T., Damhuis, R. A., Hoogsteden, H. C. (1994). Prognostic factors and survival in malignant pleural mesothelioma. Eur. Respir. J., 7, 1035–8.
334
Vogelzang, N. J., Rusthoven, J. J., Simanowski, J. et al. (2003). Phase III study of pemetrexed in combination with cisplatin versus cisplatin alone in patients with malignant pleural mesothelioma. J. Clin. Oncol., 21, 2636–44. Yates, D. H., Corrin, B., Stidolph, P. N. et al. (1997). Malignant mesothelioma in south east England: clinicopathological experience of 272 cases. Thorax, 52, 507–12.
30
SOFT TISSUE AND BONE TUMOURS IN ADULTS Owen Tilsley
Introduction Soft tissue sarcomas are rare. Although there are more than 30 subtypes, the behaviour of most is similar. Tumour size and grade are the best predictors of survival. Surgery is the mainstay of treatment, and adjuvant radiotherapy (RT) is useful for high-risk tumours. The use of adjuvant chemotherapy remains controversial, and metastatic disease responds poorly to doxorubicin and ifosfamide. The most common bone cancer occurs as a secondary tumour, and the most common primary bone tumours are benign. Primary bone cancers are rare and include those derived from marrow elements, such as myeloma and lymphoma, which are discussed in Chapter 31. Osteosarcoma and Ewing’s sarcoma make up the majority of bone sarcomas: they occur most commonly in people in their teens and 20s. Both types of tumour are treated with neoadjuvant and adjuvant ‘sandwich’ chemotherapy and local surgery. The role of RT in osteosarcoma is uncertain, but RT is important in the local control and cure of inoperable Ewing’s sarcoma and as an adjunct to marginal surgery. Metastatic disease in both conditions carries a poor prognosis. Rarer bone sarcomas include chondrosarcoma, which responds poorly to either chemotherapy or RT. Other bone sarcomas that show morphological similarity to soft tissue sarcomas are treated in the same way as osteosarcomas. Aneurysmal bone cysts and giant cell tumours are of borderline malignancy. Sarcomas of the uterus are discussed in Chapter 23 and gastrointestinal stromal tumours are discussed in Chapter 15. Sarcomas in children are discussed in Chapter 37.
Soft tissue sarcomas Types of soft tissue tumour The great majority of soft tissue tumours are benign. The WHO classification of sarcomas is largely based on
an assumed cell of origin and is assessed by the majority component of the tumour stroma. In fact, there is little evidence that these tumours arise from such cells, and the anatomical distribution often bears no relationship to that of the associated mature tissue. For some sarcoma types there is no equivalent matching normal tissue. Table 30.1 is a short listing of the most common sarcoma types. Most oncology reference texts contain a larger table, which often extends to two pages. Although many of these subtypes are undoubtedly distinct entities, given the rarity of soft tissue sarcoma and the unsatisfactory nature of current treatment, their management is usually identical.
Incidence and epidemiology Soft tissue sarcomas can affect almost any part of the body; thus, it is difficult to collect accurate data about their incidence. There are approximately 1100 new cases of soft tissue sarcoma per year in England and Wales. The annual incidence of soft tissue sarcomas is 2.3 in 100 000 men and 1.8 in 100 000 women (National Statistics, 2005). However, the International Classification of Diseases is based on anatomical site, and as many as 50% of soft tissue sarcomas may be recorded as a primary malignancy of the anatomical site. The true incidence rate may be as high as 4 in 100 000. Incidence increases with age; peak incidence occurs in the fifth to seventh decades.
Risk factors and aetiology In most cases, the cause of soft tissue sarcoma is unclear. However, a number of inherited conditions are known to be associated with soft tissue sarcomas. These include the following: r Neurofibromatosis type 1 (von Recklinghausen’s disease) – the disease prevalence is 10 in 100 000, and this type carries a 5% lifetime risk of malignant peripheral nerve sheath tumours. 335
Owen Tilsley
Table 30.1. Types of soft tissue sarcoma Type
Notes
Leiomyosarcoma
Most common soft tissue sarcoma; most gastric leiomyosarcomas are now considered GISTs and they respond to imatinib
Liposarcoma
Probably a range of different tumours; grade 1 liposarcoma/atypical lipoma is a disease of borderline malignancy; round cell liposarcoma shows higher response rates to chemotherapy than other types; myxoid liposarcoma is relatively chemorefractory
Fibrosarcoma Malignant fibrous
Diagnosis of exclusion, made if no other type can be identified
histiosarcoma Rhabdomyosarcoma
Paediatric soft tissue sarcomas, 90% of which are rhabdomyosarcomas, are quite distinct from adult soft tissue sarcoma – they are chemoresponsive and discussed separately in Chapter 37; adult rhabdomyosarcoma is often treated with paediatric protocols, but the prognosis is not as favourable
Synovial sarcoma
Does not arise from synovium, but is so called because it is common around the knee; somewhat more chemoresponsive than other types of sarcoma
GIST = gastrointestinal stromal tumour.
r Li Fraumeni syndrome (associated with germline mutation in TP53), which is associated with a wide range of soft tissue and bone sarcomata. r Familial retinoblastoma – a germline mutation of the RB1 gene carries a 10% risk of sarcoma. r Tuberose sclerosis. r Gardner’s syndrome, a subtype of familial adenomatous polyposis – small bowel fibromas develop, which can undergo malignant change. r Gorlin’s syndrome. Other factors associated with sarcoma include chronic lymphoedema (Stewart Treves syndrome is the development of angiosarcoma in an area of lymphoedema), previous irradiation and chemicals (e.g. herbicides, chlorophenol).
Anatomy and pathology Malignant tumours may develop a pseudo- or false capsule via compression of surrounding tissue, but tumour cells may be found in the peritumoural oedema even some centimetres away from the visible primary. This oedema extends by the path of least resistance, usually axially in a limb, and it respects fascial and compartmental boundaries. Approximately half of soft tissue sarcomas arise in limbs, 20% are retro336
peritoneal, 15% are visceral and 10% occur in the head and neck. Completeness of excision, grade, size and, to a lesser extent, the depth of a tumour guide the choice of further treatment, admittedly without a strong evidence base. These indicators are often of more use than the subtype in determining treatment. Benign tumours outnumber malignant ones by at least 100 to 1, and even for lumps excised and sent for histology, the ratio is 20 to 1, which has been a challenge for the centralisation and optimisation of sarcoma services. A number of grading systems are in common use internationally. In Europe, including the UK, the Trojani system is widely used. This system divides sarcomas into grades 1 to 3 based on a scoring system of differentiation, mitoses and necrosis, with scores of 2 to 3 being classed as grade 1, 4 to 5 as grade 2, and 6+ as grade 3 (see Table 30.2).
Spread Local spread is described earlier in this chapter. Lymph node spread is unusual, except for synovial sarcoma and round cell liposarcoma. Haematological spread occurs to the lungs if the tumour is drained by the systemic
Soft tissue and bone tumours in adults
Table 30.2. The Trojani scale of grading Score
0
Table 30.3. The TNM classification of malignant soft
1
2
3
Differentiation
Well
Intermediate
Poor
Mitoses per
0–9
1–19
20+
10 hpf Necrosis
Nil
< 50%
> 50%
hpf = high power field. Adapted from Trojani et al. (1984).
circulation, and to the liver if drained by the portal circulation.
Clinical presentation A patient with soft tissue sarcoma may present with the primary, secondaries or as distant non-metastatic disease; most lumps are benign. Features suggestive of malignancy are tumour size greater than 5 cm, an increase in size, a lump deep to the deep fascia, and pain. Guidance has been issued to general practitioners to refer lumps meeting any of these criteria. Haematogenous metastases are common: 40% of sarcomas present with these metastases, usually to the lung. Paraneoplastic syndromes are uncommon, but unexplained pyrexia or cachexia is occasionally seen. Examination should note size, site and mobility, loss of function, and any stigmata of an underlying condition.
tissue tumours Stage
Description
T1
Tumour 5 cm or less in greatest dimension
T1a
Superficial tumour
T1b
Deep tumour
T2
Tumour more than 5 cm in greatest dimension
T2a
Superficial tumour
T2b
Deep tumour
N0
No regional lymph node metastasis
N1
Regional lymph node metastasis
M0
No distant metastasis
M1
Distant metastasis
Adapted from UICC (2002).
although they are seldom involved, the draining lymph nodes should also be imaged. MRI is usually more informative in the limb, CT in the thorax and abdomen. A CT scan to exclude lung secondaries, and for tumours drained by the portal circulation, a CT of the liver is required. Observational studies of referral delays in the UK have highlighted that considerable time is often spent in local investigation of a lump before a specialist referral is made. It is recommended that both tests be requested as soon as a diagnosis of soft tissue sarcoma is strongly suspected. CXR and routine blood tests should also be performed.
Investigation and staging A core biopsy is required for diagnosis and should be performed in all cases in which malignancy is considered in an extremity. This procedure is followed because excision margins of 1 cm are desirable for a sarcoma but would be excessive for benign conditions. The needle track should be excised at a definitive operation, and a poorly placed biopsy increases the extent and morbidity of surgery. Frozen tissue biopsy at the time of definitive surgery may also be acceptable, particularly when the diagnosis of soft tissue sarcoma is fairly certain and when excising the track of a wide-bore needle en bloc might be difficult (e.g. in the retroperitoneum). A fine needle aspirate is adequate for a confirmation of relapse, but it probably provides inadequate material for making the initial diagnosis. Computed tomography (CT) and/or magnetic resonance imaging (MRI) of the primary is required, and,
Staging classification The TNM classification of malignant soft tissue tumours is shown in Table 30.3. The regional lymph nodes are defined as those nodes most appropriate to the primary tumour site, and they vary with the location of the tumour (UICC, 2002).
Treatment overview Patients should be managed by a specialist multidisciplinary team. The aim of treatment is to maximise the chance of a cure while safely minimising treatment-related morbidity. Soft tissue sarcoma is incurable without surgery. Appropriate limb-sparing operations have the same local control and cure rates as amputation: surgery should be undertaken by specialists. RT halves the recurrence rate of large 337
Owen Tilsley
high-grade tumours. The role of adjuvant chemotherapy remains uncertain. Fairly high-dose doxorubicin and ifosfamide are the most active agents against advanced disease, but they are toxic and have response rates of only 20%, and so palliative care alone may be appropriate.
Local disease: surgical treatments The primary tumour should be excised with a 1 cm margin, although the extent of the margin has not been investigated in a clinical trial. The wider margins that have been historically recommended are no longer felt necessary and excision beyond fascial boundaries is not needed. A randomised trial of limb salvage surgery with RT versus amputation revealed a salvageable reduction in local control but no difference in disease-free or overall survival (Rosenberg et al., 1982). The functional loss from arm amputation is considerably more than that from leg amputation, and limb salvage should always be considered. Below the knee, the functional outcome from amputation and limb-sparing surgery are the same, but patient well-being and satisfaction, although high for both, is greater with limb salvage. A preoperative scan, good operative notes, clips at explained areas of concern, and photographs, diagrams, or annotated explanations of the operative procedure are helpful in planning RT.
Radiotherapy Postoperative radiotherapy Poor, possibly inadvertent, surgery results in a high local relapse rate, and, where feasible, re-excision should be undertaken. Sometimes limb salvage means accepting close or involved margins, for example, where a tumour abuts the sciatic nerve. For large high- and low-grade tumours, a randomised trial has shown a halving in the risk of local relapse with postoperative adjuvant external beam RT (Yang et al., 1998). Another trial has shown the same for brachytherapy for high- but not low-grade tumours (Pisters et al., 1996). Neither trial showed an effect on overall survival. However, quality of life and limb function depend on local control; if it is thought that there is an appreciable risk of local recurrence, postoperative RT should be given. An international consensus meeting advised re-excision or RT in all cases if the surgical margin was less than 10 mm, and RT for all T2b (>5 cm and deep) tumours and for high-grade
338
T1b (≤5 cm and deep) tumours with margins less than 20 mm. A survey of international practice has shown that 60% of oncologists apply a 5 cm margin to both the surgical scar and the tumour bed when planning preoperative treatment (M. H. Robinson, personal communication). Historically even wider margins of 7 to 15 cm have been advocated. However, the radial margin of 2 cm and longitudinal margin of 4 cm in the brachytherapy trial were acceptable. It is common practice to use a shrinking field, with a 5 cm margin to 50 Gy and a 2 cm margin to 60 to 66 Gy, but because the majority of local relapses occur in the high-dose volume, the shrinking field may not be necessary. In an attempt to clarify the size of the required margin, the Vortex Trial has been proposed: a randomised trial with the standard-arm first-phase CTV defined as the larger of either 5 cm around the preoperative GTV or 1 cm around the postoperative scar, and the phase 2 CTV being 2 cm around the preoperative GTV. A dose of 50 Gy is given to the phase 1 volume and 66 Gy to the phase 2 volume. The experimental arm treats only the phase 2 volume to 66 Gy. The optimum dose remains controversial. Local relapse appears unacceptably high with a dose of 50 Gy and the functional outcome is worse after 66 Gy than after 60 Gy. Most oncologists currently prescribe 66 Gy to the ICRU reference point. High-quality surgery is particularly important in areas such as the retroperitoneum, where adjuvant RT to a dose of even 50 Gy may be impossible. Postoperative adhesions can pose a problem of bowel toxicity. A spacer (e.g. a breast prosthesis) inserted at the time of operation and clips at particular areas of concern can be helpful.
Primary (preoperative) radiotherapy The optimum time for RT is also controversial, but preoperative RT is not currently in common use in the UK. Preoperative RT has not been compared with surgery alone in a randomised trial but has been compared with postoperative RT. This study was ended prematurely at a planned interim analysis, when a statistically significant increase in wound healing problems was found after 50 Gy preoperatively. Continued follow-up has shown a marginal but significant improvement in overall survival with preoperative RT but has shown no effect on local control, and longer-term poorer functional outcome with 66 Gy given postoperatively (O’Sullivan et al., 2002). Because of the lower dose and smaller volumes than those used for postoperative RT, this approach is
Soft tissue and bone tumours in adults
thought to be particularly useful in sites where wound healing is not usually a problem, and in sites such as the retroperitoneum, where normal tissue tolerance may limit the safe dose.
shielding may be required, for example, to shield pelvic structures during RT to the upper leg. Where possible, a strip of skin should always be spared to minimise the late complication of lymphoedema.
Definitive radiotherapy
Plan Frequently limbs can be treated with a parallel
Long-term local control can be achieved in as many as 30% of patients with RT alone and may be considered where surgery is not possible. Palliative RT may also be given.
opposed pair of fields, where the use of wedges and tissue compensators helps to achieve a uniform dose distribution. RT to other sites usually requires multiple beam arrangements.
Radiotherapy delivery technique: postoperative radiotherapy Patient preparation, position and immobilisation
Verification The plan should then be simulated to
For limbs, appropriate limb positioning is essential for RT planning; for example, placing the arm away from the trunk for a forearm sarcoma, abducting and externally rotating the hip into a ‘frog leg’ position for anterior thigh lesions, or raising one leg above another for a calf lesion. However, the patient needs to be comfortable to ensure reproducibility of the setup. Limb immobilisation, for example, with a plastic shell or vacuum bag, should always be considered but is not always needed for proximal tumours. A photograph of the position helps record the exact position. For other sites such as the head and neck region, an immobilisation shell is required.
Localisation and target volume Delineating the CTV can be difficult following surgery and requires a good understanding of limb anatomy. It is better to be able to plan the volume without the time pressure of a patient in discomfort on a simulator; and with the widespread availability of 3D planning and the increasing use of dose-shaping techniques such as IMRT, CT planning should be used. CT scanning does, however, constrain the positioning and immobilisation of the limb because the patient, the limb and the immobilisation device all need to fit through the aperture of the CT scanner. A diagnostic scanner typically has an aperture of 60 cm and CT simulators are usually wider, but even an aperture 100 cm wide can be limiting. With reference to all available information, such as preoperative scans and operation notes, the target volumes are planned. The phase 1 CTV for limbs typically comprises the operative bed plus 5 cm longitudinally and 2 cm radially and includes the scar and biopsy sites. The phase 2 CTV has only a 2 cm longitudinal margin. A 1 cm margin is added to form the PTV. Individualised
ensure that the planned beams can actually be delivered and that the patient’s position is reproducible. Given the possibility of quite large setup errors, often in excess of 1 cm, it is desirable to verify treatment delivery. However, the large size and the angle of the RT fields may make portal imaging difficult, and the lack of localising anatomy on the image may make its interpretation difficult.
Dose, energy and fractionation In phase 1, the dose is 50 Gy in 25 fractions over 5 weeks to the ICRU reference point. In phase 2, the dose is 10 to 16 Gy in 5 to 8 fractions over 1 to 1.5 weeks to the ICRU reference point. Sites such as the retroperitoneum and pelvis are generally prescribed a lower dose. Because of small bowel tolerance, the phase 1 dose is usually 45 Gy in 25 fractions over 5 weeks, and the phase 2 dose, if given, is individualised to stay within the patient’s tolerance. The use of preoperative rather than postoperative RT for retroperitoneal tumours should always be considered.
Preoperative radiotherapy technique The most obvious differences between preoperative and postoperative planning volumes are that the tumour is still in situ and, apart from the biopsy site, there is no scar. Because of this, the volumes tend to be smaller than those for postoperative RT. GTV to CTV margins are typically 5 cm longitudinally and 2 cm radially. The dose is 50 Gy in 25 fractions.
Adjuvant chemotherapy The use of adjuvant chemotherapy is controversial, despite a number of randomised trials involving adjuvant chemotherapy. The benefit is not likely to be large, and doxorubicin and ifosfamide, the drugs that show the most activity in advanced disease, are toxic. The largest
339
Owen Tilsley
trial of adjuvant postoperative chemotherapy and a 1997 meta-analysis, which included this trial, found similar results: disease-free survival and local control were both marginally improved with chemotherapy, but overall survival was not (Anonymous, 1997). These studies were undertaken with regimens that would not now be considered optimal. A more recent study of ifosfamide and epirubicin was ended prematurely at interim analysis as the primary end point, a significant improvement in disease-free survival, was achieved. Unfortunately on subsequent follow-up the statistically significant benefit has not been maintained (Frustaci et al., 2001). A large and negative trial of postoperative ifosfamide and doxorubicin has just been reported in abstract by the EORTC (Woll et al., 2007). A small randomised trial of neoadjuvant chemotherapy showed no benefit (Gortzak et al., 2001). Currently, with the two most active agents showing response rates of only around 20% in advanced disease, it seems unwise to prejudice the possibility of definitive curative surgery by giving them preoperatively. However, inoperable primary disease may occasionally be rendered operable and, hence, curable.
Metastatic disease Of patients with soft tissue sarcomas, 40% either present with or subsequently develop locally advanced or metastatic disease. Treatment is unsatisfactory; the only chance of a cure is surgery, which is occasionally possible for isolated lung metastases. The shorter the interval between the initial presentation and relapse, the higher is the likelihood of the patient developing further metastases. Most cytotoxic chemotherapy agents show little activity. Doxorubicin and ifosfamide appear to be the only drugs with response rates greater than 10% and both have significant toxicities. The response rate to the combination of both is higher than that reported with either alone, although it is possible that older patients, who generally have a lower response rate, may not have been included in the trial of the more toxic combination, thus skewing the results. An EORTC randomised trial of doxorubicin versus doxorubicin and ifosfamide versus CYVADIC (cyclophosphamide, vincristine, doxorubicin, and dacarbazine) showed no difference in progressionfree or overall survival (Santoro et al., 1995). Subsequent randomised dose intensification studies have shown no benefit from increased doses of single-agent doxorubicin with G-CSF support but have shown increased 340
response rates with 9 g/m2 ifosfamide given over 3 days compared with 5 g/m2 as a 24-hour infusion (van Oosterom et al., 2002). This outcome has led to argument that the doses tested in the combination study were too low. The superiority of the combination of 75 mg/m2 doxorubicin and 10 g/m2 ifosfamide over 4 days is currently being tested against the same dose of single-agent doxorubicin by the EORTC in a randomised trial. Pending the outcome of this trial, either or both agents can be considered appropriate. However, given the toxicity and low response rates of these drugs, supportive care and, where appropriate, palliative RT should also be considered.
Recurrent disease Surgical salvage of an isolated local relapse is usually possible, although salvage may involve amputation. The impact of local failure on survival is uncertain; series report either no or only modest effects. The treatment of relapsed metastatic disease after chemotherapy is unsatisfactory. The cumulative dose of previously administered doxorubicin may preclude giving more, and re-challenge with ifosfamide less than a year after administration has very poor response rates. No currently available drug is considered effective. ET-743 showed a response rate of only 4% but showed disease stabilisation in 25% of patients. ET-743 is moderately toxic and is currently unavailable. New agents are required.
Prognosis Prognostic factors Size and grade predict for survival. The subtype of the soft tissue sarcoma seems unimportant as a prognostic factor. Patients with metastatic disease have a median survival of around 12 months.
Prognosis Table 30.4 shows the prognosis for soft tissue sarcomas.
Clinical trials The EORTC is conducting a study of doxorubicin versus ifosfamide and doxorubicin, at optimal doses, in locally advanced and metastatic disease. The Vortex Trial in the UK is a randomised trial involving patients with soft tissue sarcoma in extremities.
Soft tissue and bone tumours in adults
Table 30.4. The prognosis for soft tissue sarcomas Size
Five-year survival
Ten-year survival
≤5 cm
80%
62%
5–10 cm
63%
56%
10–15 cm
58%
50%
15–20 cm
43%
39%
20–25 cm
37%
37%
25 cm +
22%
18%
Royal Orthopaedic Hospital, unpublished data; see NICE (2006).
The toxicity of ifosfamide and doxorubicin makes their use in palliation difficult in the elderly. The EORTC is conducting a trial of an oral ifosfamide analogue – trophosfamide – in advanced disease. The response rates in recurrent soft tissue sarcoma following doxorubicin and ifosfamide chemotherapy are poor, and newer agents are required. A number of agents are being tested in phase I studies.
Bone tumours Osteosarcomas typically arise from the metaphysis of a long bone such as the femur, tibia or humerus.
Range of tumours The WHO tumour classification is based on the assumed cell of origin as determined by the dominant tumour matrix. In Ewing’s sarcoma the cell of origin is unknown. There is a wide range of bone pathology, much of which needs no involvement from the clinical oncologist. Table 30.5 presents bone tumours and their clinical behaviour and treatment.
ber of rare inherited bone syndromes, is more common in Li-Fraumeni families and is dramatically increased in germ-line mutations of RB1. It is more common after exposure to radiation and occurs as a complication of long-standing Paget’s disease of bone. The risk factors for Ewing’s sarcoma are not known, but there is a rare familial association with neuro-ectodermal tumours and stomach cancer. Ewing’s sarcoma is rare in African Americans and people of Afro-Caribbean extraction.
Anatomy and pathology Osteosarcoma occurs most commonly in the diaphysis of long bones. The diaphysis is that part of the metaphysis (the middle) close to the epiphyseal growth plate from which the diaphysis/metaphysis and epiphyses (the ends of the bone) grow. Osteosarcoma occurs most commonly on either side of the knee, and away from the elbow; so with reducing frequency, the most common sites are the distal femur, proximal tibia, proximal humerus and distal radius or ulna. Ewing’s sarcoma occurs in both membranous and long bones. The incidence roughly mirrors the mass of bone, but it is more common in the diaphyseal regions of the long bones of the leg. Osteosarcoma is a spindle cell sarcoma characterised by the presence of osteoid in the tumour-associated stroma. Spindle cell sarcomas that do not produce osteoid cannot be called osteosarcoma, but they behave similarly and are treated identically. Ewing’s sarcoma is a ‘small round blue cell tumour of childhood’ and can look very similar to rhabdomyosarcoma, lymphoma, and, in younger patients, neuroblastoma. The presence of the t(11;22) EWS-FLI1 gene translocation in 85% of patients and the t(21;22) EWS-ERG gene translocation in 10% has become useful in identifying Ewing’s sarcoma. Extraskeletal Ewing’s sarcomas also contain the translocation.
Incidence and epidemiology Approximately 427 new cases of bone tumour occur annually in England and Wales. The annual incidence of disease is 0.9 in 100 000 men and 0.8 in 100 000 women (National Statistics, 2005).
Risk factors and aetiology Osteosarcoma seems to be associated with the pubertal growth spurt; peak incidence occurs about 4 years later. Osteosarcoma occurs as a complication of a num-
Spread Osteosarcoma commonly metastasises to the lung and, less commonly, to bone. Skip metastases may be seen as non-contiguous deposits in the same or neighbouring bone. Ewing’s sarcoma presents with metastases in around 25% of patients, but, with a dismal 5% cure rate in the era before chemotherapy, it is probably a systemic illness almost from the start. Metastases occur most commonly to the lungs, bone and bone marrow. 341
Owen Tilsley
Table 30.5. Tumours of bone Type
Examples
Treatment
Self-limiting, self-healing
Unicameral bone cyst, non-ossifying
No treatment required/curettage
fibroma, osteoid osteoma Persistent
Enchondroma, osteochondroma
No treatment required/curettage
Minimally locally aggressive
Aneuysmal bone cyst, chondroblastoma,
Definitive local surgery or curettage,
Locally aggressive
Giant cell tumour of bone, osteoblastoma
Definitive local surgery/curettage,
Malignant, low metastatic
Adamantinoma, parosteal osteosarcoma
Definitive local surgery, and also at
Osteosarcoma and other spindle cell
Chemotherapy, surgery, chemotherapy
chondromyxofibroma
radiotherapy at relapse (15–20 Gy) radiotherapy at relapse (50 Gy)
potential Malignant, sarcomatous
relapse sarcomas of bone (MFH/fibrosarcoma) Ewing’s sarcoma
Chemotherapy, surgery ± radiotherapy,
Chondrosarcoma
Local surgery and radiotherapy
Langerhans cell histiocytosis
Observation/local surgery/intralesional
chemotherapy Malignant, lymphoid
steroid; chemotherapy for systemic disease Myeloma, plasmacytoma, lymphoma
Chemotherapy, occasionally RT
Haemangiopericytoma, malignant
Definitive surgery
if localised Malignant, vascular
haemangioendothelioma Angiosarcoma
Treat as for Ewing’s sarcoma
MFH = malignant fibrous histiocytoma; RT = radiotherapy.
Clinical presentation Bone sarcomas present with a painful and enlarging mass or with a pathological fracture. Rarely patients may present with secondaries. Systemic malaise may also be seen.
Investigation and staging A biopsy is essential. The presentation of bone sarcoma, especially Ewing’s sarcoma, can be very similar to osteomyelitis. Biopsy of the involved bone should be undertaken by a specialist orthopaedic oncology service at a tertiary referral centre. Plain radiographs of the involved area may show the characteristic Codman’s triangle of osteosarcoma and are helpful in assessing the risk of pathological fracture. An MRI of the entire length of the involved bone is required. Special planar reconstructions may be helpful in planning conservative surgery if the tumour is close to 342
neurovascular structures. A CT scan may occasionally be helpful in providing additional anatomical information. Other staging investigations include the following: r CXR and CT of the lung for assessing for pulmonary secondaries. r Bone scan for bone secondaries. r Routine bloods; alkaline phosphatase has some limited use as a tumour marker. r Bone marrow in patients with Ewing’s sarcoma.
Staging classification The current 2002 TNM classification of bone sarcomas is shown in Table 30.6. The previous TNM classification was similar but with T1 defined as being confined to bone and T2 representing soft tissue extension. It is still in common use, but almost all tumours are classified as stage T2. Nodal disease is extremely uncommon. Neither system is used in staging Ewing’s sarcoma.
Soft tissue and bone tumours in adults
Table 30.6. The TNM classification of bone sarcomas Stage
Description
T1
Tumour 8 cm or less in greatest dimension
T2
Tumour more than 8 cm in greatest dimension
T3
Discontinuous tumours in the primary
N0
No regional lymph node metastasis
N1
Regional lymph node metastasis
M0
No distant metastasis
M1
Distant metastasis (M1a = lung; M1b = other
bone site
distant sites)
Treatment overview Osteosarcoma is treated with both neoadjuvant and adjuvant chemotherapy in a ‘sandwich fashion’ with surgery. The role of RT in osteosarcoma is uncertain. Elderly patients, possibly those over age 45, respond less well and are less capable of withstanding treatment. Chemotherapy is often omitted in patients over age 60. Extraskeletal osteosarcoma is a soft tissue sarcoma and should be treated as such. High-grade spindle cell sarcoma of bone is treated as an osteosarcoma. Chondrosarcoma responds poorly to either chemotherapy or RT, and its management is largely surgical. Ewing’s sarcoma is treated with sandwich chemotherapy; RT is useful for residual or inoperable disease.
Treatment of osteosarcoma Historically, surgery alone resulted in a 20% 5-year survival. Two small randomised trials of adjuvant chemotherapy conducted in the early 1980s reported 5-year-survival rates of around 60% (Eilber et al., 1987; Link et al., 1986). Modern trials report figures that are up to 15% better. It is traditional to give chemotherapy both before and after surgery, allowing time for the production of a custom-made endoprosthesis. A small, probably underpowered study showed no difference in overall survival or limb salvage with immediate versus delayed chemotherapy (Goorin et al., 2003). Neoadjuvant chemotherapy also allows for an assessment of histological response. Necrosis of greater than 90% is classed as a good response and is associated with an improved outcome compared to patients with a
poor response (75 versus 50% 5-year survival). Although not current UK practice, some oncologists intensify the postoperative chemotherapy for poor responders. The effectiveness of this practice is being investigated in the current Euramos 1 study. Poor responders will be randomised to continue methotrexate, doxorubicin, and cisplatin (MAP) or to receive ifosfamide and etoposide in addition. Cisplatin, doxorubicin, high-dose methotrexate, ifosfamide, and etoposide are active drugs against osteosarcoma and are used in various combinations. The reported response rates appear lower with just cisplatin and doxorubicin than the rates that include methotrexate as well, but a randomised trial showed equivalence (Bramwell et al., 1992). The previous UK/European trial showed the equivalence of three- and two-weekly cisplatin and doxorubicin with G-CSF support, but with poor response rates in both arms of the study (Lewis and Nooij, 2003). The role of RT is uncertain, and there is no consensus about its use. It may be given for involved surgical margins, poor histological response, pathological fracture and palliation. In the era before chemotherapy, a randomised trial showed a benefit equivalent to chemotherapy from adjuvant whole-lung RT (Burgers et al., 1998), but this procedure is felt to be currently unnecessary. The trial has not been repeated with chemotherapy, and the use of doxorubicin would make the radiation dose to the heart a concern.
Treatment of Ewing’s sarcoma In the 1970s, paediatric VAC, vincristine, dactinomycin and cyclophosphamide increased the dismal 5 to 10% cure rate in localised disease to around 60%. The first American Intergroup trial showed that the addition of doxorubicin or whole-lung radiation were both beneficial (Nesbit et al., 1990). Further dose escalation of doxorubicin and cyclophosphamide was also beneficial (Evans et al., 1991), as was the addition of ifosfamide and etoposide in the subsequent trial. Similarly, in the UK, the use of high dose doxorubicin and the substitution of cyclophosphamide with ifosfamide were found to be more effective (Craft et al., 1998). The results in patients with metastatic disease are disappointing. The current European study, EuroEWING 99, tests three questions in a randomised fashion: r In good-risk patients can the cumulative dose of ifosfamide (102 g in the standard arm) be reduced by its partial substitution with cyclophosphamide in the 343
Owen Tilsley
hope of reducing renal toxicity of treatment, and the risk of second malignancy? r Do patients with lung metastases benefit from wholelung radiotherapy with modern chemotherapy? r Is there is a role for high-dose chemotherapy with busulphan and melphalan and peripheral blood stem cell rescue for the poor-risk and metastatic patients?
Definitive radiotherapy in Ewing’s sarcoma Definitive surgery is preferable to RT, but definitive RT is required after chemotherapy in inoperable Ewing’s sarcoma. RT for Ewing’s sarcoma is difficult and specialised, and it can be very helpful to discuss difficult cases with specialist colleagues from other centres. The position and immobilisation technique used depends on which site is being irradiated. For limbs, a plastic shell or vacuum bag can be used. For the trunk, the patient is positioned prone or supine, depending on the site of the affected bone. Head and neck treatments require an immobilisation shell similar to those used for squamous carcinoma. The clinical target volume is usually defined as the prechemotherapy volume with a margin. In limbs, for phase 1, the margins are usually 5 cm longitudinally and 2 cm radially (2 cm in the trunk). For phase 2, the margins are reduced to 2 cm around the remaining tumour volume after induction chemotherapy. If there is extensive involvement within a bone, the whole bone should probably be treated. As with soft tissue sarcomas, a strip of skin should be spared to minimise late complications. The dose should be 55 to 60 Gy in 28 to 30 fractions over 5.5 to 6 weeks (in the postoperative situation, 45 Gy in 25 fractions for microscopic disease). Sometimes the dose needs to be reduced to avoid exceeding normal tissue tolerance levels.
Palliative treatments Palliation of a bone sarcoma is best achieved surgically, even if it requires amputation, because eventual pathological fracture is very likely and difficult to manage. When surgery is not possible, high-dose RT can be attempted.
Recurrent disease The treatment of recurrent osteosarcoma is curative only if surgical excision or possibly radical irradiation of the relapse sites is possible. Treatment may include amputation. Late relapse (occurring after more 344
than 2 years) carries a better prognosis than earlier relapse. Following surgical excision, further intensive systemic chemotherapy is usually given, if possible. If the metastatic disease is not amenable to radical local treatment, management is palliative. RT may help bone pain, and bone-seeking isotopes such as samarium-153 have been used. The treatment of recurrent Ewing’s sarcoma is unsatisfactory and largely palliative. Relapsed disease often remains chemosensitive, but it is not curable by chemotherapy. Agents in use include oral etoposide, topotecan and cyclophosphamide. Late relapse occurring after more than 2 years may occasionally respond completely to intensive re-induction chemotherapy. High-dose chemotherapy and peripheral blood stem cell rescue has also been tried, but with poor results.
Prognosis Prognostic factors Osteosarcoma in inoperable sites, in the elderly and in Paget’s disease carries a poorer prognosis than in other situations. More than 90% tissue necrosis at the time of definitive surgery after neoadjuvant chemotherapy carries a better prognosis, typically around 80%, but even patients with a poor response have cure rates, typically 45 to 55%, well above those reported in the era before chemotherapy. The prognosis of patients with Ewing’s sarcoma is determined by the presence of metastases, with bone, bone marrow and other distant disease carrying a poor 5-year survival rate of 15%. Patients with lung metastases seem to do a little better, with a 25% five-year survival. Patients over age 20, patients with primary tumours that are inoperable or in the pelvis, or patients who have a tumour volume greater than 200 ml all have a worse prognosis.
Five-year survival Modern series report 5-year survivals ranging from 55 to 70% for both osteosarcoma and Ewing’s sarcoma.
Areas of current interest NICE has published a guidance on the provision of sarcoma services in England and Wales, titled ‘Improving Outcomes for People with Sarcoma’ (NICE, 2006; www.nice.org.uk/page.aspx?o = csgsarcoma&c = 91496). This guide proposes the centralisation of sarcoma management in a sarcoma multidisciplinary
Soft tissue and bone tumours in adults
team, with a committed full-time sarcoma surgeon, and improved and defined referral and patient pathways in the hope of reducing referral delay and increasing the appropriateness of management.
Ongoing clinical trials At the time of writing, the following trials addressing soft tissue and bone sarcomas were registered with the National Cancer Research Network (National Cancer Research Network, www.ncrn.org.uk, accessed September 2006). EORTC 62012 is a randomised trial of single-agent doxorubicin versus doxorubicin plus ifosfamide in the first-line treatment of advanced or metastatic soft tissue sarcoma. Gemcitabine and docetaxel in leiomyosarcoma are studied in a phase II trial to assess the activity of gemcitabine and docetaxel as a first-line chemotherapy treatment in patients with unresectable leiomyosarcoma. Euramos is a randomised trial of the European and American Osteosarcoma Study Group to optimise treatment strategies for resectable osteosarcoma based on histological response to preoperative chemotherapy. EURO – EWING 99 is a randomised study that looks at autologous stem cell transplant, surgery, RT and chemotherapy in patients with Ewing’s sarcoma.
REFERENCES Anonymous. (1997). Adjuvant chemotherapy for localised resectable soft-tissue sarcoma of adults: a meta-analysis of individual data. Lancet, 350, 1647–54. Bramwell, V. H., Burgers, M., Sneath, R. et al. (1992). A comparison of two short intensive adjuvant chemotherapy regimens in operable osteosarcoma of limbs in children and young adults: the first study of the European Osteosarcoma Intergroup. J. Clin. Oncol., 10, 1579–91. Burgers, J. M. V., van Glabbeke, M., Busson, A. et al. (1998). Osteosarcoma of the limbs. Report of the EORTC-SIOP 03 Trial 20781 investigating the value of adjuvant treatment with chemotherapy and/or prophylactic lung irradiation. Cancer, 61, 1024–31. Craft, A. W., Cotterill, S. J., Malcolm, A. L. et al. (1998). Ifosfamide containing chemotherapy in Ewing’s sarcoma. The second United Kingdom Children’s Cancer Study Group and the Medical Research Council Ewing’s tumour study. J. Clin. Oncol., 16, 3628–33. Eilber, F., Giuliano, A., Eckhardt, J. et al. (1987). Adjuvant chemotherapy for osteosarcoma: a randomized prospective trial. J. Clin. Oncol., 5, 21–6. Evans, R. G., Nesbit, M. E., Gehan, E. A. et al. (1991). Multimodality therapy for the management of localized Ewing’s sarcoma of
the pelvic and sacral bones: a report from the second intergroup study. J. Clin. Oncol., 9, 1173–80. Frustaci, S., Gherlinzoni, F., De Paoli, A. et al. (2001). Adjuvant chemotherapy for adult soft tissue sarcomas of the extremities and girdles: results of the Italian randomized cooperative trial. J. Clin. Oncol., 19, 1238–47. Goorin, A. M., Schwartzentruber, D. J., Devidas, M. et al. (2003). Presurgical chemotherapy compared with immediate surgery and adjuvant chemotherapy for non-metastatic osteosarcoma. Pediatric Oncology Group Study POG-8651. J. Clin. Oncol., 21, 1574–80. Gortzak, E., Azzarelli, A., Buesa, J. et al. (2001). A randomised phase II study on neo-adjuvant chemotherapy for ‘high risk’ adult soft-tissue sarcoma. Eur. J. Cancer, 37, 1096–103. Lewis, I. J. and Nooij, M. (2003). Chemotherapy at standard or increased dose intensity in patients with operable osteosarcoma of the extremity: a randomised controlled trial conducted by the European Osteo-Sarcoma Intergroup (ISRCTN 86294690). Proc. Am. Soc. Clin. Oncol., 22, Abstr. 3281. Link, M. P., Goorin, A. M., Miser, A. W. et al. (1986). The effect of adjuvant chemotherapy on relapse free survival in patients with osteosarcoma of the extremity. N. Engl. J. Med., 314, 1600–6. National Statistics. (2005). Series MB1 no. 34. Cancer Statistics Registrations. Registrations of Cancer Diagnosed in 2003, England. London: Office for National Statistics. Nesbit, M. E. Jr., Gehan, E. A., Burgert, E. O. Jr. et al. (1990). Multimodal therapy for the management of primary, nonmetastatic Ewing’s sarcoma of bone: a long-term follow-up of the First Intergroup study. J. Clin. Oncol., 8, 1664–74. NICE. (2006). Improving Outcomes for People with Sarcoma. London: National Institute for Health and Clinical Excellence. Available at www.nice.org.uk/page.aspx?o = csgsarcoma&c = 91496. O’Sullivan, B., Davis, A. M., Turcotte, R. et al. (2002). Preoperative versus postoperative radiotherapy in soft-tissue sarcoma of the limbs: a randomised trial. Lancet, 359, 2235–41. Pisters, P. W., Harrison, L. B., Leung, D. H. et al. (1996). Long-term results of a prospective randomized trial of adjuvant brachytherapy in soft tissue sarcoma. J.Clin. Oncol., 14, 859–68. Rosenberg, S. A., Tepper, J., Glatstein, E. et al. (1982). The treatment of soft-tissue sarcomas of the extremities: prospective randomized evaluations of (1) limb-sparing surgery plus radiation therapy compared with amputation and (2) the role of adjuvant chemotherapy. Ann. Surg., 196, 305–15. Santoro, A., Tursz, T., Mouridsen, H. et al. (1995). Doxorubicin versus CYVADIC versus doxorubicin plus ifosfamide in first-line treatment of advanced soft tissue sarcomas: a randomized study of the European Organization for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group. J. Clin. Oncol., 13, 1537–45. Trojani, M., Contesso, G., Coindre, J. M. et al. (1984). Soft-tissue sarcomas of adults: study of pathological prognostic variables and definition of a histopathological grading system. Int. J. Cancer, 33, 37–42.
345
Owen Tilsley
UICC. (2002). In TNM Classification of Malignant Tumours. International Union Against Cancer, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 109–18. van Oosterom, A. T., Mouridsen, H. T., Nielson, O. S. et al. (2002). Results of randomised studies of the EORTC Soft Tissue and Bone Sarcoma Group (STBSG) with two different ifosfamide regimens in first- and second-line chemotherapy in advanced soft tissue sarcoma patients. Eur. J. Cancer, 38, 2397–406.
346
Woll, P. J., van Glabbeke, M., Hohenberger, P. et al. (2007). Adjuvant chemotherapy with doxorubicin and infosfamide in resected soft tissue carcinoma: interim analysis of a randomised phase III trial. J. Clin. Oncol., 2007 ASCO Annual Meeting Proceedings, Part I, vol. 25 (18S) 10008. Yang, J. C., Chang, A. E., Baker, A. R. et al. (1998). Randomized prospective study of the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcomas of the extremity. J. Clin. Oncol., 16, 197–203.
31
THE LYMPHOMAS AND MYELOMA Eve Gallop-Evans and Chris Poynton
Introduction The haematological malignancies make up a group of diverse diseases ranging from the very indolent, which may co-exist with the patient for many years, to the highly aggressive and rapidly fatal. They can occur anywhere in the body, and so this chapter is based on tumour type rather than anatomical site, although a detailed anatomical knowledge is required for treatment planning. Apart from making a tissue diagnosis, surgery rarely has a part in the management of these patients, thus, the development of curative treatment protocols for many of these patients has been a success story for chemotherapy and radiotherapy (RT). With more patients surviving longer, attention is focussing on minimising the unwanted late effects of treatment as much as on maximising the chances of a cure. Patients should be managed by multidisciplinary teams that bring together the appropriate expertise of haematologists, oncologists, radiologists, pathologists and specialist nurses. This chapter first considers the lymphomas, namely non-Hodgkin lymphoma and Hodgkin lymphoma. Then it considers the paraproteinaemias, which include multiple myeloma and solitary plasmacytoma. Lastly, it considers total-body irradiation, which is given as preconditioning prior to bone marrow transplantation in some lymphomas and leukaemias. Leukaemia in children is considered in Chapter 37 (see p. 431). However, a detailed discussion of adult leukaemias is outside the remit of this book.
Lymphomas: general aspects Introduction Non-Hodgkin lymphoma (NHL) is the seventh most common cancer in the UK, with a yearly incidence of about 9000 new cases, and occurs mostly in patients over age 65. This rate appears to be rising by 3 to 5% each
year, for reasons that are not clear; however, this finding has recently been challenged by a new survey of cancer registries (Morton et al., 2006). Hodgkin lymphoma (HL) is the 22nd most common cancer in adults, with a relatively stable yearly incidence of 1500. In children under 14 years of age, lymphoma is the third most common cancer, after leukaemia and CNS tumours, but in adolescents and young adults aged between 14 and 24 years, lymphoma is the most common cancer, accounted for mainly by HL (Birch et al., 2003). There are more than 60 types of lymphoid neoplasms with a spectrum ranging from some of the most indolent malignancies, such as mucosa-associated lymphoid tissue (MALT) lymphoma, to some of the most aggressive, such as Burkitt lymphoma. Historically, the use of multiple systems of classification has confounded the results of therapeutic trials. The Revised European-American Lymphoma (REAL) classification, adopted and updated by the World Health Organisation (WHO), has now become the standard throughout the world (Table 31.1).
Aetiology of lymphoma The aetiology of lymphoma is not clearly understood and appears to be multifactorial. Risk factors include r Immunodeficiency (HIV infection, post-transplant). r Autoimmune disorders. r Viruses – EBV, HIV, HTLV-1, hepatitis C, HHV-8. r Other infectious agents – Helicobacter pylori, Chlamydia, Borrelia, TB (usually MALT lymphomas). r Environmental carcinogens. r Family history.
Clinical presentation A patient presenting with lymphoma usually has painless lymphadenopathy. The patient may also have systemic ‘B’ symptoms in the 6 months before diagnosis: fever higher than 38◦ C, drenching night sweats and weight loss of more than 10% of the original body weight.
347
Eve Gallop-Evans and Chris Poynton
Table 31.1. WHO Lymphoma Classification (2001) Type
Examples
B-cell neoplasms
Precursor B-cell neoplasms
Type
Mature T-cell and NK-cell neoplasms
Precursor B lymphoblastic
T-cell prolymphocytic
leukaemia/lymphoma
leukaemia
Mature B-cell neoplasms
T-cell large granular
Chronic lymphocytic
lymphocytic leukaemia
leukaemia/small lymphocytic
Aggressive NK-cell leukaemia
lymphoma
Adult T-cell
B-cell prolymphocytic
leukaemia/lymphoma
leukaemia
Extranodal NK/T-cell
Lymphoplasmacytic lymphoma
lymphoma, nasal type
Splenic marginal zone
Enteropathy-type T-cell
lymphoma
lymphoma
Hairy cell leukaemia Plasma cell myeloma
Hepatosplenic T-cell lymphoma
Solitary plasmacytoma of bone
Subcutaneous panniculitis-like T-cell lymphoma
Extraosseous plasmacytoma
Mycosis fungoides
Extranodal marginal zone B-cell lymphoma of mucosa-
Sezary syndrome
associated lymphoid tissue
Primary cutaneous anaplastic large-cell lymphoma
(MALT lymphoma)
Peripheral T-cell lymphoma,
Nodal marginal zone B-cell
unspecified
lymphoma
Angioimmunoblastic T-cell
Follicular lymphoma
lymphoma
Mantle cell lymphoma
Anaplastic large-cell lymphoma
Diffuse large B-cell lymphoma
(large- and small-cell variants)
Mediastinal (thymic) large B-cell
T-cell proliferation of uncertain
lymphoma
malignant potential
Intravascular large B-cell
Lymphomatoid papulosis
lymphoma Primary effusion lymphoma Burkitt lymphoma/leukaemia B-cell proliferations of uncertain malignant potential Lymphomatoid granulomatosis Post-transplant lymphoproliferative disorder, polymorphic T-cell and NK-cell neoplasms
Precursor T-cell neoplasms Precursor T lymphoblastic leukaemia/lymphoma Blastic NK-cell lymphoma
Adapted from Jaffe et al. (2001).
348
Examples
Hodgkin lymphoma
Nodular lymphocyte predominant Hodgkin lymphoma Classical Hodgkin lymphoma Nodular sclerosis classical Hodgkin lymphoma Lymphocyte-rich classical Hodgkin lymphoma Mixed cellularity classical Hodgkin lymphoma Lymphocyte-depleted classical Hodgkin lymphoma
The lymphomas and myeloma
Table 31.2. Cotswolds modification of the Ann Arbor staging system Stage
Description
I
Involvement of a single lymph node region, lymphoid structure or extra-nodal site
II
Involvement of two or more lymph node regions on the same side of the diaphragm, or localised involvement of an
III
Involvement of lymph node regions on both sides of the diaphragm with or without localised involvement of an
IV
Diffuse organ involvement, any liver/bone marrow disease
X
Bulk disease > 10 cm
E
Extranodal extension or single isolated site of extranodal disease
A/B
B symptoms: weight loss > 10%, fever, drenching night sweats
extranodal site and one or more lymph node regions on the same side of the diaphragm extranodal site, or spleen, or both
Adapted from Carbone et al. (1971).
Symptoms of organ involvement or compression may also be present.
Investigations and staging It is important to emphasise to newly diagnosed patients that an accurate diagnosis and staging are very important in deciding their treatment. A clinical history should be taken and an examination performed to assess systemic symptoms, performance status and the presence of palpable disease. A full blood count and film; ESR; renal, liver and bone profile; urate; lactate dehydrogenase (LDH); and β2microglobulin should be taken. A CT scan of the neck (or ultrasound), CT scan of the thorax, abdomen and pelvis should be done. An MRI of the head and neck should be performed if relevant. The use of PET as a staging investigation is not yet standard practice. A biopsy is preferably excisional, or cutting needle, and requires expert pathological review. Bone marrow examination should be carried out for all patients with NHL and for patients with HL who have advanced disease or abnormal blood counts. Immunophenotyping of blood, bone marrow or tissue should be performed as appropriate. A CSF examination should be done in cases of suspected CNS involvement, or in cases of parameningeal or testicular disease.
Staging classification The Ann Arbor staging system was initially developed for HL but is also the basis for anatomic staging in NHL
(Table 31.2; Carbone et al., 1971). However, the system has limitations for NHL because the pattern of disease differs from that of HL, with more frequent extranodal and bone marrow involvement. The lymph node regions defined at the Rye Symposium in 1965 are as follow: r Right and left cervical (including cervical, supraclavicular, occipital and preauricular). r Right and left axillary. r Right and left infraclavicular. r Mediastinal. r Right and left hilar. r Para-aortic. r Mesenteric. r Right and left pelvic. r Right and left femoral-inguinal.
Prognostic index: NHL The Ann Arbor staging system does not give adequate prognostic information for many subtypes of NHL, nor does it always help to decide management. The International Prognostic Index (IPI), based on a study of more than 2000 patients with high-grade NHL treated with anthracycline-containing regimens, showed that five factors independently predict outcome. The use of the IPI has been further refined for patients older and younger than 60 years, and for patients with follicular NHL. Within the IPI, the spleen is defined as an extranodal site. However, although the IPI defines differing risk groups, individual patients with identical prognostic scores can still have very different outcomes. Recent research suggests that gene expression patterns can 349
Eve Gallop-Evans and Chris Poynton
be used as prognostic determinants up front and protein expression can provide new targets for treatment (Abramson and Shipp, 2005). The IPI scoring factors (score 1 point for each) include age > 60 years, performance status of 2 to 4, stage III or IV disease, LDH elevated above normal, and the patient having more than one extranodal disease site. The prognoses, based on IPI category (score) and 5-year survival (5YS), are low (score 0 to 1, 5YS = 73%), low intermediate (2; 5YS = 51%), intermediate high (3; 5YS = 43%) and high (4 to 5; 5YS = 26%). The age-adjusted IPI for patients age 60 years or more (score 1 point for each) includes a performance status of 2 to 4, stage III or IV disease, and elevated LDH. The prognoses, based on age-adjusted IPI category (score) and 5-year survival (5YS), are low (0; 5YS = 83%), low intermediate (1; 5YS = 69%), intermediate high (2; 5YS = 46%) and high (3; 5YS = 32%).
Pretreatment assessment Semen cryopreservation should be discussed with all postpubertal males. An echocardiogram or MUGA scan should be performed to assess cardiac function in patients with risk factors for ischaemic heart disease, or in all patients over 70 years in whom anthracyclines are being considered. Assess the disease bulk and the risk of tumour lysis before chemotherapy, and start allopurinol 300 mg p.o. o.d. as prophylaxis. When there is evidence of tumour lysis occurring with an elevated serum uric acid despite allopurinol prophylaxis, rasburicase should be given. Rasburicase may be indicated as prophylaxis in patients allergic to allopurinol or in patients experiencing renal failure prior to starting chemotherapy (serum creatinine >150 μM/l).
PET scanning in lymphoma Functional imaging with positron emission tomography using a radiolabelled glucose analogue, 2-18 F-fluoro-2deoxy-D-glucose (FDG-PET), identifies increased glycolytic activity associated with malignant tumours and complements conventional imaging. The degree of FDG uptake is expressed by the use of a semiquantitative measure called the standardised uptake value. Anatomical localisation of PET findings with concurrent CT scanning leads to fewer false-positive PET interpretations because physiological FDG uptake can vary among patients. In patients with lymphoma, FDG-PET imaging 350
can be used in initial staging, response evaluation and follow-up, although there are relatively few large RCTs assessing the use of FDG-PET imaging in these different clinical situations. Imaging may detect occult disease that has been missed by conventional staging, although the impact on outcome is not defined. FDG-PET imaging is used most commonly in the assessment of a residual mass at the end of treatment, and it is better than CT scanning in distinguishing between residual tumour and fibrosis. A meta-analysis of studies of PET scanning in 854 patients with lymphoma showed a median sensitivity of 90.3% and a median specificity of 91.1% compared to histological assessment (Isasi et al., 2005). Among the studies reporting changes in staging, between 8 and 17% of the patients were upstaged and 2 to 23% were downstaged. There is a good correlation between the results of post-treatment scanning and outcome. A study of 54 patients with lymphoma (HL and NHL) found that a positive PET scan after treatment was associated with a 1-year progression-free survival of 0% and an overall survival of 50% compared with 86 and 92% for patients with a negative PET scan, respectively (Jerusalem et al., 1999). For the prediction of prognosis, early interim PET scanning (e.g. PET scan after two cycles of chemotherapy) appears more promising, although the numbers of patients studied are small (Hutchings et al., 2006). A study of 85 patients with HL with a median followup of 3.3 years showed that PET scanning after two or three cycles of chemotherapy was highly predictive of outcome (Hutchings et al., 2005). Patients with positive interim scans had a 5-year progression-free survival of 38.5%, compared with 91.5% for patients with negative scans (p ≤ 0.0001). New trials for early and advanced-stage HL should use results of interim PET scans for randomisation between different treatment strategies.
CNS prophylaxis in lymphoma Patients with Burkitt lymphoma or patients with diffuse large B-cell lymphoma or peripheral T-cell lymphoma with high IPI scores (high LDH, bone marrow involvement, extranodal disease) or parameningeal or testicular involvement have a high risk of CNS relapse and intrathecal prophylaxis is recommended. No randomised trial has been performed, but a retrospective analysis of patients treated before and after the introduction of CNS prophylaxis suggests that this approach
The lymphomas and myeloma
is effective. A recent UK survey showed no clear consensus on either the number, the timing, the doses or the types of chemotherapy given for intrathecal prophylaxis. Practice throughout the world varies greatly. A reasonable approach would be four alternating injections of methotrexate 12.5 mg or ara-C 50 mg, given weekly with hydrocortisone 50 mg.
Diffuse large B-cell lymphoma (DLBCL) The annual incidence of DLBCL is 5 in 100 000 people and the median age of diagnosis is 64 years. The maleto-female ratio of disease occurrence is 1:1. Phenotypes CD19, CD20, CD22, CD79a are involved, and it has been determined more recently that BCL6, CD10 and MUM1 can differentiate germinal centre (GC) types from non-GC types. Some cases contain large numbers of T-cells (T-cell-rich B-cell lymphoma). The genetics of the disease involve the t(14;18) IgHBCL2 breakpoint for BCL2 and 3q27 breakpoint for BCL6.
Clinical presentation Patients usually present with widespread nodal or extranodal disease and B symptoms; 10% have bone marrow involvement.
DLBCL
Advanced and / or bulky*
Early stage (I and IIA)
R-CHOP x 3 IFRT 30 to 45 Gy
Cure
R-CHOP x 8
Relapse
Second-line chemotherapy ESHAP / R-ESHAP ICE / R-ICE
Complete response or very good partial response
High-dose chemotherapy / autologous stem cell transplant
Partial response or less
Palliative treatment
Figure 31.1. Flow chart for standard accepted treatment of diffuse large B-cell lymphoma (DLBCL). Clinical trials should always be offered
Treatment overview
where appropriate. *Bulky disease has been defined in a number of
A SWOG/ECOG trial found that chemotherapy with cyclophosphamide, doxorubicin, vincristine and prednisolone (CHOP) was as effective as, but less toxic than, multiple-drug dose-intense regimens (MACOP-B, mBACOD, ProMACE-CytaBOM). The 5-year survival was about 45% (Fisher et al., 1993), and a recent update showed 15-year survival to be about 40% (Fisher et al., 2004). The biggest therapeutic impact has been the introduction of rituximab, an anti-CD20 monoclonal antibody. The GELA study of patients ages 60 to 80 years compared eight cycles of rituximab with CHOP (RCHOP) with CHOP alone and showed 2-year survival rates of 76% compared to 63% in favour of R-CHOP, with similar benefits seen at 5 years (Coiffier et al., 2002). The MInT study, which used rituximab added to anthracycline-based chemotherapy in patients below the age of 60 with low-risk IPI scores, showed that the addition of rituximab improved 3-year progressionfree survival from 68 to 85%, and 3-year overall survival from 84 to 93% (Pfreundschuh et al., 2006). In
cyclophosphamide, doxorubicin, vincristine, prednisolone; ESHAP =
ways, ranging from 5 to 10 cm maximum diameter. CHOP = etoposide, methylprednisolone, cytarabine, cisplatin; ICE = ifosfamide, carboplatin, etoposide; IFRT = involved-field radiotherapy; R = rituximab.
this trial, RT (30 to 40 Gy) was given to sites of initial bulky disease larger than 5 cm. Bulky disease larger than 7.5 cm was found to be an independent poor prognostic factor. A flow chart for the standard accepted treatment of DLBCL is shown in Figure 31.1.
Treatment for early stage: stages IA/IEA/limited IIA A SWOG study, in which 70% of patients had low-risk IPI scores, compared eight cycles of CHOP to three cycles of CHOP with involved-field radiotherapy (IFRT) of 40 to 55 Gy (Miller et al., 1998). The CHOP plus IFRT arm had higher rates of 5-year progression-free (76 versus 67%) and overall survival (82 versus 74%). However, updated 351
Eve Gallop-Evans and Chris Poynton
results at 8.2 years median follow-up showed overlapping survival curves. An ECOG study randomised patients with early stage disease in complete remission between eight cycles of CHOP alone, or CHOP combined with 30 Gy IFRT (Horning et al., 2004). Patients in partial remission received 40 Gy RT. Ten-year follow-up shows that the addition of RT in complete responders improves 10-year disease-free survival (57 versus 46%) but not 10-year overall survival (64 versus 60%). A recent GELA study randomised low-risk patients between three cycles of CHOP with IFRT, and three cycles of dose-intensified doxorubicin, cyclophosphamide, vindesine, bleomycin and prednisolone (ACVBP) plus consolidation with methotrexate, etoposide, ifosfamide and cytarabine (Reyes et al., 2005). Event-free and overall survival were higher in the chemotherapy-alone group (82 versus 74%, and 90 versus 81%, respectively). In this setting, adding rituximab to short-course CHOP and IFRT is thought likely to give similar benefits and, although there is no RCT evidence, it has become an accepted treatment. Preliminary results from the BNLI dose trial of IFRT in high-grade NHL, comparing 40 Gy in 20 fractions with 30 Gy in 15 fractions, has shown that the lower-dose arm results in equally good local control (Hoskin et al., 2005). The current standard treatment is three cycles of RCHOP and IFRT, 30 to 45 Gy in daily 2 Gy fractions over 3 to 4.5 weeks (RCR, 2006).
Treatment for advanced stage There is currently no large prospective trial that looks at the role of consolidation RT for patients with initial bulky masses or for patients with residual masses, particularly in the current era when immunochemotherapy and PET scanning are used to detect residual active disease. Retrospective analyses suggest that RT improves disease-free, but not overall, survival. Dose intensity appears to be an important factor for treatment outcome. The German NHL-B1/2 trials randomised patients to six cycles of CHOP-21, CHOP14 (given two-weekly), CHOEP-21 (CHOP plus etoposide), or CHOEP-14. Patients in the two-weekly regimens received G-CSF from day 4. Patients in this trial also received RT (36 Gy) to sites of initial bulky and extranodal disease. In patients under 60 years of age, the addition of etoposide to CHOP reduced the rate of relapse (73 versus 63%, p = 0.001) but shorter regimens did not significantly improve relapse rates or 352
overall survival (Pfreundschuh et al., 2004a). In patients over 60 years of age, 5-year overall survival was better with CHOP-14 than with CHOP-21 (53.3 versus 40.6%, p < 0.001) but the addition of etoposide was more toxic and no more effective (Pfreundschuh et al., 2004b). With the success of rituximab therapy, current trials are now investigating the possibility of a reduction in the number of chemotherapy cycles, but keeping current (or even increasing) the number of rituximab doses given after the completion of chemotherapy. There is no evidence that high-dose therapy with stem cell transplantation improves survival when carried out after the first complete response. The current standard treatment is R-CHOP-21, every 3 weeks, for six to eight cycles. Consolidation RT that involves giving 30 to 40 Gy in 2 Gy daily fractions over 3 to 4 weeks could be considered for patients with residual disease if high-dose chemotherapy is not appropriate.
Current NCRI trial The current NCRI trial involves R-CHOP-14 versus R-CHOP-21: six cycles of R-CHOP-14 (8 × R in total) versus eight cycles of R-CHOP-21 – and looks at the role of dose intensity, with G-CSF support.
Treatment for relapsed/refractory disease High-dose chemotherapy (e.g. BEAM – carmustine, etoposide, cytarabine, melphalan) with an autologous stem cell transplant has been shown to improve progression-free and overall survival over those of conventional treatment. The PARMA study randomised patients responding to second-line chemotherapy with DHAP (dexamethasone, cytarabine, cisplatin) to receive either high-dose chemotherapy with autologous stem cell transplant (HDC/ASCT) or four further cycles of DHAP (Philip et al., 1995). At eight years, both eventfree and overall survival favoured the HDC/ASCT group (36 versus 11%, and 47 versus 27%, respectively). Poor prognostic factors include chemorefractory disease and a short duration of remission. The most effective conditioning regimen has not been established, and DHAP or ESHAP (etoposide, methylprednisolone, cytarabine, cisplatin) may be used, with or without rituximab. A current NCRI trial compares R-DHAP with R-ICE (rituximab, ifosfamide, carboplatin, etoposide) and randomises patients to maintenance rituximab or followup alone after transplant.
The lymphomas and myeloma
Prognosis DLBCL can be cured in about 50% of cases, but molecular profiling has identified two distinct subtypes with prognostic implications independent of clinical variables (Alizadeh et al., 2000): r Germinal centre B-cell type, which results in a 60% overall 5-year survival. r Activated B-cell type, which results in a 35% overall 5-year survival.
Primary mediastinal B-cell lymphoma The annual incidence of primary mediastinal B-cell lymphoma is 0.24 in 100 000 people. The median age of diagnosis is 40 years, and the male-to-female ratio is from 1:2 to 1:4. Phenotypes involved in the disease are CD20+ and CD30+.
Clinical presentation Patients usually present with a bulky mediastinal mass and pericardial and pleural effusions.
Treatment The current standard treatment is six to eight cycles of R-CHOP. Historically, mediastinal RT was recommended because of the bulk of disease at presentation, but the effectiveness of this approach is unproven; its use may be determined by the results of PET imaging after chemotherapy.
Prognosis The 5-year overall survival ranges from 0 to 65%, depending on the patient’s IPI score.
Primary CNS lymphoma (PCNSL) Of all PCNSLs, 90% are diffuse large B-cell lymphomas. PCNSLs account for fewer than 1% of all cases of NHL, and 5% of all brain tumours. The annual incidence of disease is 0.28 in 100 000 in immunocompetent patients and 4.7 in 100 000 in immunocompromised patients.
Clinical presentation Patients with PCNSL present in a similar way to patients with other primary CNS tumours. Patients usually
present with progressive neurological signs, cognitive and personality changes, fits, raised intracranial pressure or cranial neuropathies. There may be single or multiple lesions in the brain parenchyma, and leptomeningeal or ocular disease. Spread outside the CNS is rare.
Investigation The initial assessment should include HIV status and ophthalmology review. A biopsy is required to diagnose and classify the lymphoma, but there is no benefit from surgical resection. Therefore, possible investigations include MRI, stereotactic biopsy, lumbar puncture, ophthalmological examination and viral serology (HIV, CMV, HSV).
Management There is no clear consensus about the management of patients with these lymphomas. Management policies vary and there is little good research evidence on which to base management plans. The prognosis is usually very poor and the disease may be associated with serious neuropsychological problems, the most severe of which is dementia.
Principles of care Obtain input from the HIV team and the expertise necessary to supervise the use of high-dose chemotherapy regimes. Corticosteroids will reduce symptoms in the majority of patients. Chemotherapy regimens with high CNS penetration (particularly high-dose methotrexate; Bessell et al., 2001) are required in patients fit enough to tolerate them. There is a high relapse rate of 60% following treatment with RT alone, with a median survival of 12 to 18 months. High-dose methotrexate (HD-MTX > 3 g/m2 ) followed by RT can improve response rates (64 to 95%) and median survival (33 to 42 months). Using HD-MTX in combination with other agents that cross the blood-brain barrier can further improve median survival (40 to 60 months), but the optimum chemotherapy regimen has not yet been defined. For patients with residual disease after chemotherapy (three to six cycles), whole-brain RT is advised because of the risk of multifocal disease – 45 Gy in 25 to 30 fractions. Dose escalation above 45 Gy 353
Eve Gallop-Evans and Chris Poynton
does not improve local control. In practice, RT is usually given following a partial response to chemotherapy, or at relapse. The role of RT following a complete response to chemotherapy is unclear. Postchemotherapy RT may help to control the disease in those patients not achieving a complete radiological remission on imaging. But there is a very high risk of dementia or other neuropsychological problems in older patients (Bessell et al., 2004). Patients who are frail, elderly or immunosuppressed may only tolerate low-toxicity palliative treatment. The role of rituximab is unclear because its high molecular weight limits penetration through an intact blood-brain barrier. Intraventricular or intrathecal administration allows higher CSF levels, and responses have been seen in patients with leptomeningeal, but not parenchymal, disease.
Follicular lymphoma (FL) Follicular lymphoma accounts for 25% of all cases of NHL. The median age at diagnosis is 60 years. FL is a nodal lymphoma with a follicular growth pattern and arises from a mature B-cell in the germinal centre of the lymph node. Histologically, within the current WHO classification, the FLs are graded by the Berard criteria from 1 to 3, depending on the number of large cells per high-power field. Grade 3 can be further subdivided into 3a and 3b; the latter resembles de novo DLBCL and is generally treated as such.
Clinical presentation Follicular lymphoma patients usually present with stage III or IV disease, with progressive and widespread lymphadenopathy with bone marrow involvement. They rarely present with localised (stage I or II) disease.
Prognosis Prognostic factors The Follicular Lymphoma International Prognostic Index (FLIPI) allows for a score of 1 point for each of the following: r Age > 60 years. r Stage III or IV disease. r Five or more nodal sites involved (see Figure 31.2). r LDH elevated above normal. r Hb < 12 g/dl. 354
The prognosis based on FLIPI category (score) and 10year survival (10YS) can be good (0 to 1; 10YS = 71%), intermediate (2; 10YS = 51%) or poor (3 to 5; 10YS = 36%).
Overall management plan for FL Figure 31.3 shows a flow chart for treatment of patients with follicular lymphoma.
Treatment of localised FL Local IFRT results in a relapse rate of 55%, with no relapses after 7 years. Preliminary results of the BNLI dose trial show an equivalence in 24 Gy in 12 fractions to 40 Gy in 20 fractions. An EORTC trial randomised patients to IFRT alone or IFRT followed by CVP (cyclophosphamide, vincristine, prednisolone) but the addition of chemotherapy did not improve overall survival (Carde et al., 1984). Radical treatment of stage IA FL involves a dose of 24 to 40 Gy in 2 Gy fractions over 2.5 to 4 weeks (RCR, 2006).
Treatment of advanced FL For asymptomatic patients, there is still no clear survival benefit for immediate treatment compared with a ‘watch-and-wait’ approach. Indications for treatment include bulky disease, compromise of organ function, or progressive symptoms. Treatment options range from single-oral-agent chemotherapy with chlorambucil, to combination chemotherapy regimens with or without anthracyclines such as CVP, FMD (fludarabine, mitoxantrone and dexamethasone) and CHOP, to palliative RT. Rituximab as a single oral agent or in combination with chemotherapy gives superior results. An EORTC study comparing eight cycles of R-CVP with CVP initially showed a median time to treatment failure of 27 months for R-CVP compared to 7 months for CVP (Marcus et al., 2005). The latest analysis now shows a median duration of response of 38 months for R-CVP compared to 14 months for CVP (Solal-C´eligny et al., 2005). Maintenance therapy with rituximab is currently under investigation and may further improve progression-free survival. Figure 31.4 shows management questions at different stages of FL and emphasises that there is no consensus about managing patients with advanced FL. The efficacy of anti-CD20 monoclonal antibody therapy has led
The lymphomas and myeloma
Cervical Pre-auricular Upper cervical Median or lower cervical Posterior cervical Supraclavicular
Mediastinal Paratracheal Mediastinal Hilar Retrocrural
Axillary Axillary Mesenteric Coeliac
Para-aortic
Splenic (hepatic) hilar Portal
Para-aortic
Mesenteric
Common iliac External iliac
Inguinal Inguinal Femoral
Others: Epitrochlear, popliteal Figure 31.2. Nodal sites used in the Follicular Lymphoma International Prognostic Index (FLIPI) – each rectangle represents a nodal area. Reproduced ´ with permission from Solal–Celigny et al. (2004).
Complete staging and assign FLIPI
Localised* (mostly but not exclusively stage I) FLIPI 0-1
Advanced, with low tumour burden (mostly stage II-III) FLIPI 2
Advanced with high tumour burden (mostly stage III-IV) FLIPI 3+
Involved field RT 24 to 40 Gy
Observation
R-CVP or R-CHOP
Figure 31.3. Flow chart for treatment of follicular lymphoma, grades 1, 2 and 3a. *Localised means a true stage I by Ann Arbor criteria (i.e. a single node or area) or limited stage II (adjacent lymph node areas). CHOP = cyclophosphamide, doxorubicin, vincristine, prednisolone; CVP = cyclophosphamide, vincristine, prednisolone; FLIPI = Follicular Lymphoma International Prognostic Index; R = rituximab.
355
Eve Gallop-Evans and Chris Poynton
First-line treatment
Standard therapy R-CVP / R-CHOP 6 - 8 cycles?
Options R-chemo x 4 + R x 4? Standard therapy + consolidation with radioimmunotherapy Complete response: Maintenance retuximab? Good partial response: Monitor or next-line therapy? Poor response: Monitor versus next-line therapy versus HDC Palliative chemotherapy
Second-line treatment
Immunoconjugate
Standard therapy Fludarabine combination: FMD / FC ± R HDC / autograft versus reduced intensity allograft
Figure 31.4. Management questions at different stages of follicular lymphoma. CHOP = cyclophosphamide, doxorubicin, vincristine, prednisolone; CVP = cyclophosphamide, vincristine, prednisolone; FC = fludarabine, cyclophosphamide; FMD = fludarabine, mitoxantrone, dexamethasone; HDC = high-dose chemotherapy; R = rituximab.
to the suggestion that further reduction of chemotherapy cycles (CVP or CHOP), to four or fewer, coupled with more rituximab up front and maintenance rituximab after, should further prolong treatment-free intervals. By the end of 2007, two important studies, PRIMA (a four-arm trial comparing different chemotherapy regimens containing anthracyclines or purine analogues and varying amounts of rituximab for untreated FL) and RICOVER (comparing six or eight cycles of CHOP21 to CHOP-14 with and without rituximab), will have matured and may clarify whether it is appropriate to use less-intensive chemotherapy.
Treatment of relapsed disease Fludarabine monotherapy is effective in patients with relapsed FL, and response rates are further improved by using fludarabine in combination with anthracyclines or alkylating agents, as well as rituximab. Radioimmu356
notherapy (90 Y anti-CD20 or 131 I anti-CD20) can give overall response rates of about 75% and complete response rates of up to 40% in heavily pretreated patients. Stem cell transplantation may improve disease-free survival, but most patients eventually relapse. The only potentially curative treatment for patients with relapsed FL is allogeneic stem cell transplantation, which can be used in patients up to 70 years of age, using a nonmyeloablative, reduced-intensity approach, if a suitable donor (ideally a matched sibling) is available. The main cause of death of patients with FL is an uncontrolled tumour after pathological transformation to a higher grade (usually DLBCL, but occasionally Burkitt lymphoma). Appropriate treatment is given according to the pathological type, but the prognosis is poorer than for patients who present de novo. Repeat biopsies should always be done where there is clinical suspicion of transformation.
Current clinical trials The NCRI Watch and Wait trial tests surveillance compared to immediate rituximab, with a further randomisation to maintenance rituximab. The NCRI FoRT trial tests low-dose palliative RT, 4 Gy in 2 fractions over 2 days, compared to a control arm of 24 Gy in 12 fractions over 16 days.
Mantle cell lymphoma The annual incidence of mantle cell lymphoma is 0.72 in 100 000 patients, and makes up 5% of all cases of NHL. The median age of diagnosis is 65 years, and the male-to-female ratio is 4:1. Mantle cell lymphoma arises from CD5+ cells that populate the mantle zone of follicles. Morphologic subtypes include classical (87.5%), small-cell (3.6%), pleomorphic (5.9%) and blastic (2.6%) variants. A diffuse growth pattern and high proliferation rate (Ki-67 fraction) are associated with a worse prognosis (Tiemann et al., 2005).
Clinical presentation Most patients present with advanced disease that includes nodal and extranodal involvement (bone marrow in 90%, spleen, liver, gastrointestinal tract). There is a spectrum of pathology from low-grade (especially with a leukaemic presentation that seems to carry a better prognosis) to a blastoid variant that carries a very poor prognosis (Orchard et al., 2003).
The lymphomas and myeloma
Treatment Fludarabine, a purine analogue, can elicit good response rates when used in combination with cyclophosphamide (FC), particularly when used as a first-line therapy (Cohen et al., 2001). The advantages of FC are that it can be given orally, is generally well tolerated and does not usually cause alopecia. However, it can result in prolonged immunosuppression, and prophylactic cotrimoxazole is given concurrently. Anthracycline-based chemotherapy (e.g. CHOP) can improve complete response rates and time to progression compared with non-anthracyline regimens but does not affect overall survival. The role of rituximab is being investigated, as are the use of thalidomide and bortezomib, a proteasome inhibitor. There is some evidence that highdose chemotherapy with ASCT in first remission may improve progression-free and overall survival, but this, and allografting, are still experimental. The M. D. Anderson group has tested a dose-intense regimen of cyclophosphamide, vincristine, doxorubicin, and dexamethasone alternating with high-dose methotrexate and cytosine arabinoside (Hyper-CVAD) in a phase II study; a response rate of 93% was achieved, although toxicity was significant. The results have been updated, with the addition of rituximab to the regimen (Romaguera et al., 2005). This approach may be considered for young, fit patients with poor prognostic features. Combined modality treatment is not often used, but if patients cannot tolerate chemotherapy, or if local control of disease is required, 30 to 40 Gy in 2 Gy fractions may be appropriate. RT may be used for palliation of symptomatic nodal or extranodal disease (8 Gy in a single fraction, or 20 Gy in five fractions). The current standard treatment for mantle cell lymphoma is fludarabine-cyclophosphamide (FC).
Prognosis The 5-year overall survival of patients with mantle cell lymphoma is 25%; median survival is 3 years.
Extranodal marginal zone B-cell lymphoma (MALT type) Extranodal marginal zone B-cell lymphoma has an annual incidence of 0.6 in 100 000 patients and represents 8% of all NHL. The median age at diagnosis is 60 years, and the male-to-female ratio is 1:1.2.
MALT lymphoma is a clinicopathologically distinct entity arising at extranodal sites of acquired mucosaassociated lymphoid tissue in, for instance, the stomach, salivary glands, conjunctiva, orbits, lung, breast and thyroid. It is frequently associated with a pre-existing inflammatory or infective condition. Gastric MALT lymphoma is associated with H. pylori infection in up to 90% of cases. Thyroid lymphomas are considered in Chapter 35 (see p. 416).
Management The optimal treatment for patients with MALT lymphoma at many sites is uncertain. The response to locally directed therapy such as surgery or RT is similar to that with chemotherapy. Remission following RT may take up to 24 months and can be confirmed by repeat biopsies. Of patients with gastric MALT lymphoma, 70% achieve complete and prolonged histological remission following treatment to eradicate H. pylori. But the response to eradication alone appears to be less good in patients in whom the lymphoma has extended deeply through the gastric wall and in patients who have regional lymph node involvement or a t(11;18) genetic abnormality. However, patients with t(11;18) rarely transform to DLBCL.
H. pylori eradication therapy Helicobacter pylori eradication therapy involves omeprazole 20 mg b.d., metronidazole 400 mg b.d. and amoxicillin 1 g b.d. (clarithromycin 500 mg b.d. if allergic) for 14 days.
Local radiotherapy The local RT dose is 24 to 30 Gy in daily 2 Gy fractions. Future trials may investigate the use of lower doses. The gastric field includes the whole stomach and adjacent coeliac nodes with a margin of 2 cm. CT planning allows the left kidney to be shielded from the lateral field. The parotid gland can be treated using wide-angled, wedged anterior and posterior oblique fields to spare the contralateral salivary glands. For orbit treatments, the use of narrow-angled, wedged anterior oblique fields spares the contralateral eye.
Chemotherapy Chlorambucil or CVP are administered. The role of rituximab is being investigated. 357
Eve Gallop-Evans and Chris Poynton
Prognosis The disease remains localised for prolonged periods and prognosis is generally good; 5-year disease-free survival rates are higher than 90%.
Mycosis fungoides (MF) The annual incidence of mycosis fungoides is 0.36 in 100 000 patients, and the male-to-female ratio is 2:1. MF is a low-grade lymphoproliferative disorder of CD4+ T cells and is the most common type of cutaneous T-cell lymphoma. Distinct clinical stages of cutaneous disease are seen, with patches and plaques, progressing later to the development of cutaneous tumours. The clinical course is usually slow and can be indolent over several years. Nodal and visceral involvement occurs late. S´ezary syndrome is defined by the presence of erythroderma, lymphadenopathy and S´ezary cells in peripheral blood. Prognostic factors include disease stage, plaque thickness, age older than 60 years, serum LDH and response to treatment.
Management There have been few randomised controlled trials involving MF. PUVA (psoralen and ultraviolet A) is one of the most useful treatments for early disease. Lowdose superficial RT is used to palliate individual lesions and allows retreatment and overlapping fields. Doses depend on the size of the treatment field and may range from single 4 to 8 Gy fractions, to 15 Gy in daily 5 Gy fractions, to 30 Gy in daily 2 Gy fractions.
Total skin electrons (TSE) A meta-analysis of predominantly retrospective studies of TSE beam therapy has shown that response rates are related to cancer stage, dose and energy. Complete response rates vary from 96% in early stages to 60% in stage III disease. Higher skin-surface doses (32 to 36 Gy) and higher energies (4 to 6 MeV) improve response rates (Jones et al., 2002). Various techniques are used to deliver a homogeneous dose to the accessible skin surface. Eyes and nails are shielded and some areas of the skin may be underdosed, including the top of the scalp, soles, palms, perineum and inframammary areas, and may require additional treatment. The patient usually stands 3 to 4 m from the linear accel358
erator, either on a rotating platform or in several different treatment positions. Acute complications of TSE include erythema, dry desquamation, hair and nail loss, and reduced sweating. Late complications include secondary skin cancers, and skin changes such as hyperpigmentation and telangiectasia. Although TSE is not curative, it can produce prolonged palliation, and recurrences may be limited in penetration and extent. Adjuvant treatment with PUVA may preserve the remission and delay relapse. TSE can be repeated after 1 year or more if other disease-control measures fail and patients have had an initial good response. Retreatment is associated with increased acute and late complications. There is no consensus about the most appropriate chemotherapy regimen, and responses are usually short lived. Systemic chemotherapy (such as CHOP) may produce transient responses but advanced disease is usually resistant to therapy and has a poor prognosis (Siegel et al., 2000).
Prognosis Overall 5- and 10-year survival for MF is 80 and 57%, respectively, although patients may need multiple courses of treatment, with decreasing intervals between courses.
Hodgkin lymphoma The incidence of HL is bimodal. In developing countries, the first peak occurs in childhood, whereas in developed countries, it is seen in young adults. The second peak occurs in older adults, usually men. Epidemiological studies suggest an infectious aetiology, particularly involving EBV. The male-to-female ratio of disease incidence is 1.5:1. Of HL patients, 95% have classical HL, in which the tumour cells express CD30 and CD15, and lack B-cell antigens. In nodular lymphocyte predominant HL (5%), B-cell markers such as CD20 are expressed, and CD30 and CD15 are absent. However, in both types, the malignant Reed-Sternberg cells appear to be clonally derived germinal-centre B-cells.
Approach to treatment of Hodgkin lymphoma A flow chart for the management of Hodgkin lymphoma is shown in Figure 31.5. Individual situations are discussed in the following sections.
The lymphomas and myeloma
Hodgkin lymphoma
Early stage I-IIA
ABVD x 3-4 IFRT 30 Gy in 15 f
Advanced disease
IPFS < 4 ABVD x 8
IPFS > 4 Esc BEACOPP
Progressive / refractory / relapsed disease
Second-line chemotherapy (ESHAP / BEAM)
No / poor response
Complete / partial response
Clinical trial? Palliation
HDC / ASCT
Figure 31.5. Flow chart showing approach to treatment of Hodgkin lymphoma. ABVD = doxorubicin, bleomycin, vinblastine, dacarbazine; BEACOPP = bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisolone; BEAM = carmustine, etoposide, cytarabine, melphalan; Esc = escalated; ESHAP = etoposide, methylprednisolone, cytarabine, cisplatin; f = fractions; HDC/ASCT = high-dose chemotherapy with autologous stem cell transplant; IFRT = involved-field radiotherapy; IPFS = International Prognostic Factor Score.
Management of early stage (I to IIA) disease Patients with early stage HL may be divided into two prognostic groups to decide treatment: favourable and unfavourable. The prognostic factors are age, number of involved sites (three or more), B symptoms, ESR greater than 50 and bulky disease larger than 10 cm (usually mediastinal). Patients with one or more risk factors may be treated as patients with advanced disease. Extended-field RT alone gives high complete response rates, but up to 30% of the patients will relapse outside the radiation field. Combined-modality treatment reduces the risk of relapse and allows the use of smaller, involved fields, which should reduce the risk of long-term toxicity while maintaining cure rates. Doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD) chemotherapy has been shown to be more effective and less toxic than the combination of
mustine, vincristine, procarbazine and prednisolone (MOPP), which is no longer used. The GHSG HD10 trial has tested the number of cycles of ABVD chemotherapy (two to four) and radiation dose (20 to 30 Gy), and it appears possible to define a favourable prognostic group, without any of the aforementioned risk factors, in which treatment can be reduced to two cycles of ABVD followed by 20 Gy IFRT (Diehl et al., 2004). Longer follow-up is required to determine whether this approach can maintain high progression-free survival rates while reducing toxicity. Trials with a chemotherapy-only arm show that the omission of RT increases the relapse rate but, because these patients usually respond to salvage treatment, overall survival is usually similar for both arms (Meyer et al., 2005; Thomas et al., 2004). Because pulmonary toxicity from bleomycin is a problem with ABVD, alternative regimens are being investigated. Gemcitabine appears promising (Straus, 2006), and a new CALGB trial is testing gemcitabine in combination with doxorubicin and vinblastine (AVG) for patients with non-bulky stage I and II disease. RT is being omitted for these patients. Increasing use of interim PET scanning to assess early response to chemotherapy may be able to define a favourable cohort of patients in which IFRT can be safely omitted. For example, the current UK Children’s Cancer and Leukaemia Group (CCLG) treatment guidelines incorporate PET scanning for the assessment of response after two cycles of OEPA (doxorubicin, vincristine, etoposide, prednisolone) chemotherapy and to guide a decision on whether to give IFRT: if negative, there is no further treatment; if positive, IFRT of 20 Gy is given, with a boost of 10 Gy for significant residual masses. The treatment of HL in children is discussed further in Chapter 37 (see p. 432). The current standard treatment is three to four cycles of ABVD followed by IFRT 30 Gy in 15 fractions (RCR, 2006).
Management of advanced disease (stage IIB and higher) ABVD has replaced MOPP as the standard chemotherapy regimen; ABVD gives comparable overall survival rates with less toxicity. However, some patients still relapse, whereas others may be overtreated. A prognostic index has been defined, based on more than 5000 patients treated with ABVD or an equivalent regimen (Hasenclever and Diehl, 1998). 359
Eve Gallop-Evans and Chris Poynton
On the International Prognostic Factor Score (IPFS) for Hodgkin lymphoma, score 1 point for each of the following patient characteristics: r Age >45 years. r Male. r Serum albumin <40 g/l. r Hb <10.5 g/dl. r Stage IV disease. r Leucocytosis >15 × 109 /l. r Lymphopenia <0.6 × 109 /l or <8% WBC count. Prognosis based on the IPFS score, 5-year disease-free survival (5YDFS) and 5-year overall survival (5YOS) are as follows: r Score = 0 to 3 – 81% of patients; 5YDFS = 70%; 5YOS = 83%. r Score = 4 or more – 19% of patients; 5YDFS = 47%; 5YOS = 60%. This prognostic score can identify patients who are at a higher risk of relapse. Patients with a score of 0 to 3 are treated with six to eight cycles of ABVD, whereas patients with a score of 4 or more may be considered for more intensive treatment. The GHSG H9 study of advanced HL showed that patients with at least four risk factors had an improved 5-year progression-free survival with dose-escalated BEACOPP (bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisolone; 82%) when compared with patients receiving standard BEACOPP (74%) or alternating COPP (cyclophosphamide, vincristine, procarbazine, prednisolone)/ABVD (59%; Diehl et al., 2003). However, there is no evidence that high-dose chemotherapy and an autologous stem cell transplant (ASCT) improve survival when used as a part of first-line therapy. The current standard treatment for advanced HL is eight cycles of ABVD. Consider intensified chemotherapy, for example, escalated BEACOPP for patients with an IPF score of 4 or higher.
Role of radiotherapy in advanced HL A meta-analysis of 1740 patients treated in 14 randomised trials performed between 1972 and 1988 (MOPP or equivalent chemotherapy) did not show any significant survival benefit from the addition of RT (Loeffler et al., 1998). A GELA study compared two chemotherapy regimens, MOPP-ABV and ABVPP (doxorubicin, bleomycin, vinblastine, procarbazine, prednisolone), and randomised patients in CR or PR to RT or further chemotherapy (Ferm´e et al., 2000). The addition of RT to MOPP/ABV did not improve overall survival at
360
5 years (85 versus 88%). Overall survival was the highest for patients treated with ABVPP alone, without RT (94 versus 78%). An EORTC trial found no survival benefit with the addition of 24 Gy extended-field RT following a complete response to MOPP/ABV chemotherapy (Aleman et al., 2003). Patients with a PR received 30 Gy and had similar event-free and overall survival rates as patients with a CR. The role of RT in initially bulky disease is still unclear, and the increasing use of PET scanning for assessment of residual masses may inform multidisciplinary discussion. The latest trial from the German group, HD15, uses RT only for PET-positive residual masses measuring more than 2.5 cm. RT is still an integral part of shorter multidrug chemotherapy regimens such as Stanford V (meclorethamine, doxorubicin, vinblastine, prednisone, vincristine, bleomycin, etoposide), which is being compared with ABVD in a randomised clinical trial. RT is given to sites of bulk disease that measure 5 cm or more at diagnosis, or to residual disease, 34 to 36 Gy in 1.5 or 2 Gy fractions. The current CCLG treatment guidelines recommend a PET scan after two cycles of OEPA to decide on the need for IFRT. Only children with positive PET scans receive 20 Gy to involved fields, with a 10 Gy boost for residual disease, on completion of their planned chemotherapy regimen. In the current standard practice, RT is not given to patients who have a complete response to full-course chemotherapy. Patients with a partial response at the end of treatment, confirmed on PET scanning or with a biopsy, are usually referred for second-line chemotherapy followed by high-dose chemotherapy and an ASCT. Another option, if residual disease at the end of first-line therapy is encompassable within a reasonable radiation field, is to give consolidation RT, with a dose of 30 to 34 Gy in 15 to 20 fractions of 1.8 to 2 Gy over 3 to 4 weeks. If a patient is not a candidate for high-dose chemotherapy, then consolidation RT can improve complete response rates, though probably not overall survival.
Management of relapsed/refractory Hodgkin lymphoma Patients who relapse after first-line treatment or patients with primary refractory disease may be salvaged with high-dose chemotherapy and ASCT. Second-line chemotherapy (e.g. miniBEAM, ESHAP) with early stem cell collection priming is used to obtain
The lymphomas and myeloma
maximal cytoreduction before transplant. Published studies show progression-free survival of up to 61% at 5 years. Poor prognostic factors include relapse within a year, chemorefractory disease, extranodal relapse and B symptoms.
Nodular lymphocyte predominant Hodgkin lymphoma (NLPHL) NLPHL may present as slowly progressive lymphadenopathy, sometimes preceding the diagnosis by several years. It is more common in men, and systemic symptoms are rare. The prognosis of early stage disease is good, but current management protocols do not distinguish it from classical HL. A retrospective analysis of 131 patients with early stage low-risk disease treated within clinical studies of the GHSG showed no difference in outcome between extended- or involvedfield RT, or combined-modality treatment (Nogov´a et al., 2005). Paediatric treatment guidelines recommend a watch-and-wait strategy in patients who are PET-negative after surgical excision. The current standard treatment for NLPHL is the same as for classical HL.
Radiotherapy techniques in lymphoma Involved-field radiotherapy IFRT refers to irradiation of the entire lymph node regions initially involved with lymphoma as defined in the Rye and Ann Arbor staging system. However, this staging system was not designed to encompass radiation-field design, and there are no uniform definitions of the extent of the involved field in different clinical scenarios (Yahalom and Mauch, 2002). There are also different interpretations: for example, the mediastinal field may or may not include the hila, and the supraclavicular nodes may either be encompassed in the mediastinal field when there is upper mediastinal involvement, or in the neck field when there is ipsilateral cervical involvement. The volume of irradiated normal tissue can be reduced in certain areas such as the mediastinum and para-aortic regions, where the postchemotherapy volume is treated with lateral margins of 1.5 to 2 cm. All these field definitions offer guidance only; careful review should be performed of pre- and postchemotherapy imaging using contrastenhanced CT, and PET scans if available. In general, with
advances in imaging and planning technology, fields may simply encompass the postchemotherapy or PETpositive volumes (with predefined standardised uptake value [SUV]) with a margin. As far as possible, involved lymph node regions should be irradiated within a single field.
Neck IFRT of the ipsilateral neck and supraclavicular field (neck hyperextended) involves the following margins: r Superior – tip of mastoid process. r Inferior – 1 cm below clavicle and sternal notch. r Medial – midline or contralateral edge of vertebral body, depending on medial extent of disease. r Lateral – encompasses medial two-thirds of clavicle. Spinal shielding is rarely possible, and lower treatment doses make spinal shielding less critical. Shield the larynx anteriorly from the start if possible, or from 20 Gy.
Waldeyer’s ring The IFRT field should cover the lymphoid tissue of the nasopharynx, tonsil and base of tongue. Involvement of any site usually requires treatment of the whole ring, which should include the retropharyngeal, preauricular, submandibular and submental nodes, and be matched to the upper border of the neck field if cervical nodes are being treated. It should usually be CT planned and treated with two lateral fields.
Axilla The IFRT field should include the infraclavicular and supraclavicular regions (arm abducted) with the following margins: r Superior – C5/C6 interspace. r Inferior – 2 cm below the lowest extent of disease. r Medial – 2 cm medial to the extent of disease. r Lateral – inner edge of the humerus.
Mediastinum If the upper mediastinum is involved, both supraclavicular fossa fields should be included, and the lateral borders adjusted accordingly, using the following margins: r Superior – C5/C6 interspace. r Inferior – 5 cm below the carina, or 2 to 5 cm below the lowest extent of prechemotherapy disease; lowest extent of the T10/T11 interspace. r Lateral – postchemotherapy volume (or hila, if involved) with 1.5 cm margin.
361
Eve Gallop-Evans and Chris Poynton
Lung The lung is rarely treated as part of IFRT, although paediatric guidelines recommend treatment for cases of disseminated involvement of one or both lungs that is still detectable after two cycles of chemotherapy. Treatment of the whole of one lung (or both lungs) should be 3D planned, with lung correction, and the patient should be treated with a dose of 12 to 15 Gy in 1 to 1.2 Gy fractions. In children, the spine should be treated symmetrically (i.e. fully included when treating both lungs, or excluded when one lung is being treated). (For a discussion of late RT effects in children, see Chapter 37, p. 428). For localised lung involvement, residual disease following two cycles of chemotherapy is treated with a margin of 1.5 cm.
Spleen The whole spleen on the planning CT scan is considered the GTV, with a 1.5 to 2 cm margin for PTV that allows for movement with respiration. Splenic hilar nodes are only treated if involved.
Liver Treatment of the liver may be indicated under paediatric treatment guidelines. The whole liver is treated with a 1.5 to 2 cm margin for PTV. Portal nodes are included. The tolerance of the whole liver is 15 Gy in 1 to 1.2 Gy fractions.
For ipsilateral common and external iliac nodal involvement, treat the contralateral common iliac nodes as well, and the field should extend superiorly to the top of L4. For ipsilateral inguinal nodal involvement, treat the ipsilateral femoral and external iliac nodes using the following margins: r Superior – middle of the sacro-iliac joint. r Inferior – 5 cm below the lesser trochanter. r Lateral – greater trochanter, or 2 cm lateral to extent of disease. r Medial – medial border of the obturator foramen, or 2 cm medial to the extent of disease. In a standard pelvic field, the testicles are not included in the beam, although scattered irradiation may result in temporary oligospermia. The dose can be reduced with a special clam-shell shield, particularly if bilateral pelvic irradiation is required. The ovaries lie just medial to the external iliac nodes and surgical transposition or oophoropexy may be required to preserve ovarian function. With unilateral irradiation, one ovary remains outside the field, and it should retain normal function. Patients should be treated lying supine using anterior and posterior opposed 5 to 8 MV beams, with both fields treated daily. The dose is prescribed to the midplane at the field centre, except where the neck is treated with an asymmetric collimator, when the prescription point is 2 cm lateral to the midline at the central axis. Maximum and minimum doses across the volume should be measured, and if the dose gradient is greater than 10%, compensators should be used.
Para-aortic field IFRT treats all para-aortic nodes from the diaphragm to the aortic bifurcation, using the following margins: r Superior – T10/T11 interspace. r Inferior – L4/L5 interspace (bifurcation of the aorta) or 2 cm below the lowest extent of disease. r Lateral – postchemotherapy volume with a 1.5 cm margin. The position of the kidneys should be noted, so that no more than one-third of the total renal volume is included in the field.
Inguinal/pelvic fields For ipsilateral external iliac nodal involvement, treat the ipsilateral common iliac, external iliac and inguinal nodes. The field should extend superiorly to the top of L5 and inferiorly to the inferior border of the obturator foramen. 362
Bone involvement Treat the involved site with margins of 2 cm.
Extended field radiotherapy: technique The mantle field includes all lymph node regions above the diaphragm, whereas the inverted Y treats all nodal regions below the diaphragm. Total lymphoid irradiation combines both fields, which are usually treated sequentially. In the era of combined chemotherapy and RT, these large anterior and posterior opposed fields are rarely used as the sole treatment.
Mantle field Position The patient’s neck is extended to keep the oral cavity out of the field; arms are on hips, which moves the humeral heads laterally (arms up would move the
The lymphomas and myeloma
axillary nodes superiorly and reduce the lung volume in field if required).
Borders The superior border is the mastoid process, the inferior border is the T10/T11 interspace, and the lateral border is the outer border of the humeral heads.
Shielding Shielding of the lung involves placing the inferolateral border of lung blocks at the most lateral point of the sixth rib, and the upper and medial border curves, to spare as much lung tissue as possible while treating infraclavicular nodes, mediastinum and hilar nodes. The oral cavity and the larynx are shielded anteriorly from the start. The humeral heads can be shielded if there is no axillary disease. Spinal cord shielding from the posterior field should be considered after 20 Gy, if it does not shield the disease. Cardiac shielding should be considered after 20 Gy if possible. The patient should be treated with equally weighted 6 MV anterior and posterior opposed fields. The dose schedule is 36 Gy in 20 fractions over 4 weeks, or 30 Gy in 17 fractions after chemotherapy.
Acute side effects of mantle radiotherapy Acute side effects include mild oropharyngitis, dysphagia, fatigue, skin erythema and localised hair loss. L’Hermitte’s sign may occur in about 15% of patients approximately 6 to 12 weeks following RT. It is characterised by an electric shock sensation radiating down the backs of both legs when the neck is flexed, and is due to transient demyelination. It usually resolves spontaneously without any permanent sequelae.
Late complications of treatment for lymphoma Subclinical hypothyroidism develops in about a third of patients who undergo mantle RT. Loss of fertility is usually compounded by the use of chemotherapy, particularly high-dose chemotherapy. Sperm cryopreservation should be offered to all postpubertal males. Women who do not need urgent treatment may undergo a cycle of in vitro fertilisation before cancer treatment with cryopreservation of embryos, but this process delays the start of treatment (Wallace et al., 2006). Cryopreservation of mature, unfertilised oocytes is not yet an established technique, nor is cryopreservation of ovarian tissue. Another controversial area is whether the ovary can be protected during cancer treat-
ment by using GnRH agonists to create a temporary menopause. The major causes of late mortality, apart from HL, are second malignancies and cardiovascular disease. Mediastinal irradiation increases the risk of coronary artery disease, and patients should be warned about additional risk factors such as smoking, hyperlipidaemia, and hypertension. Second malignancies may include myelodysplasia and leukaemia, NHL and solid tumours, particularly of the breast and lung. It is clear that the increasing risk of breast cancer is inversely related to age, but directly related to increasing dose. The risk is particularly high for patients under the age of 25 years who are treated with radiotherapy (Swerdlow et al., 2000). The current strategy of using smaller field sizes and lower radiation doses should reduce this risk without compromising survival. Current guidelines recommend screening women from eight years after treatment, using either mammography or MRI, depending on the patient’s age and breast density.
Paraproteinaemias/myeloma Introduction The disorders associated with clonal immunoglobulin (paraprotein) secretion form a diverse group that ranges from benign monoclonal gammopathies to aggressive plasmablastic leukaemias.
Classification of paraproteinaemias Conditions that may be associated with paraproteinaemia include multiple myeloma (virtually never IgM), solitary plasmacytoma of bone, extramedullary plasmacytoma, lymphoplasmacytic lymphoma (Waldenstr¨om’s macroglobulinaemia, always IgM) and monoclonal gammopathy of undetermined significance (MGUS). Waldenstr¨om’s macroglobulinaemia is really a syndrome associated with IgM paraprotein that is usually considered to be a lymphoplasmacytic lymphoma. It is an indolent condition characterised by hyperviscosity and an absence of the lytic lesions typical of myeloma. In MGUS there is an accumulation of paraprotein in the serum, without other findings of malignancy. MGUS can develop into multiple myeloma (about 1% 363
Eve Gallop-Evans and Chris Poynton
incidence per year) and recent guidelines for its monitoring stratify high- and low-risk groups for transformation to myeloma (Kyle et al., 2002).
Myeloma incidence and epidemiology Approximately 3500 new cases of myeloma are reported per year in the UK (Cancer Research UK, http:// info.cancerresearchuk.org), where the annual incidence of disease is 5 in 100 000 patients. Approximately 2500 deaths occur from myeloma per year in the UK. Myeloma is very uncommon in patients less than 40 years old. Disease incidence rises steeply after the age of 50, with peak incidence occurring during age 70 to 80. Myeloma is slightly more common in men: the male-to-female ratio is 6:5.
Myeloma risk factors and aetiology Very few risk factors have been identified for myeloma: r Increasing age. r Family history of myeloma or MGUS. r Afro-Caribbean ethnic origin. r Environmental exposure to chemicals – pesticides, fertilisers, petrochemicals. Occasionally patients with MGUS develop myeloma.
Pathology of myeloma A neoplastic clone of plasma cells in the bone marrow produces monoclonal (M) protein. Most commonly this is IgG, IgA with excess light chains or free light chain only, and more rarely IgD. A few myelomas are completely non-secretory by standard assay. Light chains appear in the urine as Bence-Jones proteins. Within the myeloma cells, abnormalities of chromosomes 11 or 13 are common. Overproduction of IL-6 appears to be an important factor in the pathogenesis of myeloma and the high expression of vascular endothelial growth factor receptors (VEGFRs) makes them a useful target for therapy with anti-angiogenesis drugs (e.g. thalidomide and lenolinamide).
Clinical presentation of myeloma The main symptoms of patients with multiple myeloma are due to bone destruction, myelosuppression and paraprotein: r Back pain (80%). r Vertebral collapse/spinal cord compression. 364
r Myelosuppression leading to anaemia, recurrent infections or thrombocytopenia.
r Hypercalcaemia. r Renal failure. r Amyloid.
Investigation and staging in myeloma Initial investigations that should be used in patients suspected of having myeloma (UK Myeloma Forum, 2001) include the following: r FBC. r Serum urea and electrolytes, and calcium. r Serum protein electrophoresis. r Serum immunoglobulins. r ESR or plasma viscosity. r Urine sample for Bence-Jones proteins. r Standard X-ray imaging of long bones and axial skeleton. Further investigations include r Serum albumin. r Serum uric acid. r Bone marrow aspirate and trephine, to include cytogenetic studies. r β2-microglobulin. r LDH. r CRP. r Quantification of serum paraprotein and urinary light chain excretion. r Quantification of non-isotypic serum immunoglobulins.
Diagnosis and new definitions of myeloma and related diseases The definitions for myeloma are summarised in this section and are used in the Myeloma IX trial (International Myeloma Working Group, 2003). The conditions that define myeloma-related organ or tissue impairment due to the plasma cell proliferative process are as follow: r Calcium > 2.75 mM/l. r Creatinine > 173 μM/l. r Haemoglobin < 10 g/dl. r Lytic lesions/compression fractures. Asymptomatic myeloma is characterised by M protein ≥ 30 g/l, and/or bone marrow clonal plasma cells ≥ 10%, and no related end-organ or tissue impairment. Symptomatic myeloma is characterised by the presence of M protein in serum and/or urine, bone marrow
The lymphomas and myeloma
clonal plasma cells or plasmacytoma, and related endorgan or tissue impairment. There are other conditions such as non-secretory myeloma, solitary plasmacytoma of bone, and extramedullary plasmacytoma and multiple solitary plasmacytomas. The international staging system (ISS) for multiple myeloma has now superseded the Durie Salmon system (Greipp et al., 2005): Stage I, which comprises 28% of patients, gives a median survival of 62 months. Characteristics include: r β2-microglobulin < 3.5 mg/l. r Albumin > 35 g/l. Stage II, comprising 33% of patients, gives a median survival of 44 months. Characteristics of stage II include: r β2-microglobulin < 3.5 mg/l. r Albumin < 35 g/l. or r β2-microglobulin 3.5 to 5.5 mg/l. Stage III, which comprises 39% of patients, gives a median survival of 29 months. Stage III is characterised by β2-microglobulin > 5.5 mg/l.
The combination of idarubacin and dexamethasone (Z-DEX) is especially effective in the treatment of plasmablastic disorders. Plasmablastic multiple myeloma is a morphologic subset of myeloma, defined on a bone marrow aspirate slide as containing 2% or more of plasmablasts (immature blastic cells), and it is associated with a more aggressive clinical course and a worse prognosis (median survival of 10 months). For relapsed myeloma, stem cell transplant (usually autologous but sometimes an allogeneic minigraft procedure) is the evidence-based choice. Thalidomide maintenance is being assessed in numerous trials and more recently lenolinamide has been licensed in the UK and may be associated with higher response rates and less toxicity than thalidomide. The proteasome inhibitor bortezomib is effective in first relapse and it is licensed as monotherapy for the treatment of progressive multiple myeloma in patients who have received at least one prior therapy and who have already undergone or are unsuitable for bone marrow transplantation. Bortezomib has not yet been approved by NICE.
Treatment overview for myeloma
Radiotherapy in myeloma
Patients with asymptomatic myeloma (‘smouldering myeloma’) do not require chemotherapy but do require long-term monitoring for signs of worsening disease such as the appearance of lytic lesions or other organ or tissue damage. Patients with active disease require chemotherapy treatment. Patients may be very unwell at presentation – with dehydration, hypercalcaemia, renal failure, anaemia, infection and severe pain – and they may require full supportive care, including hydration, antibiotics, bisphosphates, red cell transfusion and analgesia. Patients with symptomatic hyperviscosity may require plasma exchange.
Indications for RT include a painful bone lesion, impending or actual pathological fracture (lesions at high risk should be referred for surgical stabilisation and treated postoperatively to improve pain and local control), and spinal cord compression. A single 8 Gy fraction can give good palliation of pain. For patients with spinal cord compression, 20 Gy in 5 fractions or 30 Gy in 10 fractions are acceptable regimens.
Chemotherapy in myeloma Standard approaches Melphalan and prednisolone should be given to patients who, by virtue of age or infirmity, cannot receive intensive therapy (although modified CTD may soon replace this approach). For other patients, the standard chemotherapy regimen, CVAD, is compared with cyclophosphamide, thalidomide and dexamethasone (CTD) in the myeloma IX trial.
Bisphosphonates in myeloma A Cochrane review of bisphosphonate use in myeloma looked at 11 randomised trials that included a total of 1113 patients. The use of bisphosphonates was found to prevent pathological fractures (OR = 0.59, CI = 0.45– 0.78, p = 0.0001). Patients taking bisphosphonates also experienced less pain, but there was no difference in survival. The authors also concluded that clodronate or pamidronate may be the preferred agents (Djulbegovic et al., 2001). A randomised trial comparing zoledronic acid to pamidronate (Rosen et al., 2001) showed that the two drugs were equivalent in terms of pain scores and skeletal-related events. 365
Eve Gallop-Evans and Chris Poynton
Solitary plasmacytoma Solitary plasmacytoma occurs more commonly in bone than in extramedullary sites. The male-to-female incidence ratio is 2:1, and the median age of patients is 50 to 55 years. Solitary plasmacytoma accounts for fewer than 5% of plasma cell malignancies. The investigations are the same as for the diagnosis and staging of multiple myeloma (Soutar et al., 2004). Criteria for diagnosis include the following: r Histologically confirmed single lesion. r Normal bone marrow biopsy (< 10% plasma cells). r Negative skeletal survey. r Normal haemoglobin, renal function and calcium. Low levels of a serum paraprotein may be present in 30 to 70% patients (a level greater than 20 g/l is suspicious of MM). The impact of improved diagnostic and staging techniques such as flow-cytometry studies, molecular detection of heavy and light chain gene rearrangements in clonal plasma cells, and MRI scans is not yet clear (Dimopoulos et al., 2000).
Solitary bone plasmacytoma Patients usually present with pain, or less commonly with spinal cord or nerve root compression, or with a pathological fracture. Solitary bone plasmacytoma primarily affects the axial skeleton, particularly the spine, and accounts for 30% of primary tumours of the spine. An MRI of the whole spine is required to delineate the extent of the tumour and to exclude possible disease at other sites. More than 75% of patients progress to multiple myeloma with a median time of 2 to 4 years. The median survival is 7 to 11 years. Radical RT improves local control rates if doses of at least 40 Gy are given (Mendenhall et al., 1980), but it does not reduce the risk of progression to MM.
Solitary extramedullary plasmacytoma Of all solitary extramedullary plasmacytomas, 90% occur in the head and neck, especially the upper respiratory tract, and produce local compressive symptoms. If complete surgical excision can be achieved with minimal morbidity, it may be curative, and postoperative RT is not recommended (Soutar et al., 2004). Incomplete excision is an indication for radical RT, which can significantly reduce local recurrence rates. Local lymph nodes are only included in the target volume if they are clinically involved. 366
Fewer than 30% of patients progress to multiple myeloma. Disease management involves a standard treatment of 40 Gy in daily 2 Gy fractions over 4 weeks. For tumours larger than 5 cm, consider giving 50 Gy in daily 2 Gy fractions over 5 weeks. Treatment fields should include all disease shown on imaging with a margin of at least 2 cm. The role of adjuvant chemotherapy is uncertain, but it may be considered in patients with highgrade extramedullary tumours, or for tumours larger than 5 cm.
Total-body irradiation Introduction Over the past 30 years, bone marrow transplantation (BMT) has changed from being an experimental procedure carried out as a last resort in terminally ill patients to being an integral part of curative treatment for an increasing number of patients with haematological malignancies. The effects of total-body irradiation (TBI) on the bone marrow provided the basis for BMT experiments in animals and man. TBI is used in combination with chemotherapy as the conditioning regimen before bone marrow transplantation. The purpose is to eradicate tumour cells, particularly from sanctuary sites, and to ablate the bone marrow prior to engraftment of the transplanted stem cells. Standard conditioning with cyclophosphamide/TBI has been compared to the use of busulphan/cyclophosphamide (Bu/Cy) in several studies, with no clear difference in overall survival but with more severe toxicity when using Bu/Cy. In acute lymphoblastic leukaemia (ALL), there appears to be a long-term advantage to using TBI-containing regimens (Gratwohl, 2004). Indications for BMT are increasing, particularly with the development of nonmyeloablative transplants, or ‘reduced-intensity conditioning’ transplants.
Technique The target volume is the entire body, treated within a single field at an extended SSD (usually 4 m). Setup techniques vary considerably, and it is important to be familiar with the one used in your centre. Important considerations are dose homogeneity, accurate delivery, reproducibility, ease of setup and local constraints on field and room size. Thermoluminescent dosimeters are placed at set sites along both sides of the body to calculate the dose received. Doses vary with the
The lymphomas and myeloma
patient’s body contour, and compensators are required. The dose-limiting toxicities are interstitial pneumonitis and hepatic veno-occlusive disease, which were more common with single-fraction regimens. Most parts of the body will receive within 10% of the dose delivered at the midplane. The standard dose regimen is 14.4 Gy in eight fractions over 4 days, with a minimum 6-hour interval between fractions. Acute side effects include fatigue, nausea, mucositis, parotitis and diarrhoea. Long-term side effects include cataracts, reduced pituitary and thyroid function, infertility and increased risk of second malignancy.
REFERENCES Abramson, J. S. and Shipp, M. A. (2005). Advances in the biology and therapy of diffuse large B-cell lymphoma: moving towards a molecularly targeted approach. Blood, 106, 1164–74. Aleman, B. M., Raemaekers, J. M., Tirelli, U. et al. (2003). Involved-field radiotherapy for advanced Hodgkin’s lymphoma. N. Engl. J. Med., 348, 2396–406. Alizadeh, A., Eisen, M., Davis, R. E. et al. (2000). Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature, 403, 503–11. Bessell, E. M., Graus, F., Lopez-Guillermo, A. et al. (2001). CHOD/BVAM regimen plus radiotherapy in patients with primary CNS non-Hodgkin’s lymphoma. Int. J. Radiat. Oncol. Biol. Phys., 50, 457–64. Bessell, E. M., Graus, F., Lopez-Guillermo, A. et al. (2004). Primary non-Hodgkin’s lymphoma of the CNS treated with CHOD/BVAM or BVAM chemotherapy before radiotherapy: long-term survival and prognostic factors. Int. J. Radiat. Oncol. Biol. Phys., 59, 501–8. Birch, J. M., Alston, R. D., Quinn, M. et al. (2003). Incidence of malignant disease by morphological type, in young persons aged 12–24 years in England, 1979–1997. Eur. J. Cancer, 39, 2622–31. Carbone, P. P., Kaplan, H. S., Musshoff, K. et al. (1971). Report of the committee on Hodgkin’s disease staging classification. Cancer Res., 31, 1860–1. Carde, P., Burgers, J.M, van Glabbeke, M. et al. (1984). Combined radiotherapy-chemotherapy for early stage non-Hodgkin’s lymphoma: the 1975–1980 EORTC controlled lymphoma trial. Radiother. Oncol., 2, 301–12. Cohen, B., Moskowitz, C., Straus, D. et al. (2001). Cyclophosphamide/fludarabine (CF) is active in the treatment of mantle cell lymphoma. Leuk. Lymphoma, 42, 1015–22. Coiffier, B., Lepage, E., Briere, J. et al. (2002). CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large B-cell lymphoma. N. Engl. J. Med., 346, 235–42. Diehl, V., Franklin, J., Pfreundschuh, M. et al. (2003). Standard and increased-dose BEACOPP chemotherapy compared with COPP-ABVD for advanced Hodgkin’s disease. N. Engl. J. Med., 348, 2386–95.
Diehl, V., Brillant C., Engert, A. et al. (2004). Reduction of combined modality treatment intensity in early stage Hodgkin’s lymphoma: interim analysis of the HD10 trial of the GHSG. Eur. J. Haematol., 73 (Suppl. 65), 37.E03. Dimopoulos, M. A., Moulopoulos, L. A., Maniatis, A. et al. (2000). Solitary plasmacytoma of bone and asymptomatic multiple myeloma. Blood, 96, 2037–44. Djulbegovic, B., Wheatley, K., Ross, J. et al. (2001). Bisphosphonates in multiple myeloma (Cochrane Review). In The Cochrane Library, vol. 4. Oxford: Update Software. Ferm´e, C., Sebban, C., Hennequin, C. et al. (2000). Comparison of chemotherapy to radiotherapy as consolidation of complete or good partial response after six cycles of chemotherapy for patients with advanced Hodgkin’s disease: results of the Group d’Etudes des Lymphomes de l’Adulte H89 trial. Blood, 95, 2246–52. Fisher, R. I., Gaynor, E. R., Dahlberg, S. et al. (1993). Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin’s lymphoma. N. Engl. J. Med., 328, 1002–6. Fisher, R. I., Miller, T. P. and O’Connor O. A. (2004). Diffuse aggressive lymphoma. In Hematology (American Society of Hematology Education Program). Washington, DC: American Society of Hematology, pp. 221–36. Gratwohl, A. (2004). Principle of conditioning regimens. In Haematopoietic Stem Cell Transplantation: the EBMT Handbook, ed. J. Apperley, E. Carreras, E. Gluckman et al. Genoa: Forum Service Editore, pp. 90–104. Greipp, P. R., San Miguel, J., Durie, B. G. et al. (2005). International staging system for multiple myeloma. J. Clin. Oncol., 23, 3412–20. Hasenclever, D. and Diehl, V. (1998). A prognostic score for advanced Hodgkin’s disease: International Prognostic Factors Project on Advanced Hodgkin’s disease. N. Engl. J. Med., 339, 1506–14. Horning, S. J., Weller, E., Kim, K. et al. (2004). Chemotherapy with or without radiotherapy in limited-stage diffuse aggressive non-Hodgkin’s lymphoma: Eastern Cooperative Oncology Group study 1484. J. Clin. Oncol., 22, 3032–8. Hoskin, P., Smith, P., Falk, S. et al. (2005). Radiation dose in non-Hodgkin’s lymphoma: preliminary results of UK NCRN randomised trial. Ann. Oncol., 16 (Suppl. 5), Abstr. 059. Hutchings, M., Mikhael, N. G., Fields, P. et al. (2005). Prognostic value of interim FDG-PET after two or three cycles of chemotherapy in Hodgkin lymphoma. Ann. Oncol., 16, 1160–8. Hutchings, M., Loft, A., Hansen, M. et al. (2006). FDG-PET after 2 cycles of chemotherapy predicts treatment failure and progression-free survival in Hodgkin lymphoma. Blood, 107, 52–9. International Myeloma Working Group. (2003). Criteria for the classification of monoclonal gammopathies, multiple myeloma and related disorders: a report of the International Myeloma Working Group. Br. J. Haematol., 121, 749–57. Isasi, C. R., Lu, P. and Blaufox, M. D. (2005). A meta-analysis of 18 F-2-deoxy-2-fluoro-D-glucose positron emission tomography in the staging and restaging of patients with lymphoma. Cancer, 104, 1066–74.
367
Eve Gallop-Evans and Chris Poynton
Jaffe, E. S., Harris, N. L., Stein, H. et al. (2001). World Health Organization Classification of Tumours. Pathology and Genetics. Tumours of Haematopoietic and Lymphoid Tissues. Lyon: IARC Press. Jerusalem, G., Beguin, Y., Fassotte, M. F. et al. (1999). Whole-body positron emission tomography using 18F-fluorodeoxyglucose for post treatment evaluation in Hodgkin’s disease and non-Hodgkin’s lymphoma has higher diagnostic and prognostic value than classical computer tomography scan imaging. Blood, 94, 429–33. Jones, G. W., Kacinski, B. M., Wilson, L. D. et al. (2002). Total skin electron radiation in the management of mycosis fungoides: consensus of the European Organization for Research and Treatment of Cancer (EORTC) Cutaneous Lymphoma Project Group. J. Am. Acad. Dermatol., 47, 364–70. Kyle, R. A., Therneau, T. M., Rajkumar, S. V. et al. (2002). A long-term study of prognosis in monoclonal gammopathy of undetermined significance. N. Engl. J. Med., 346, 564–9. Loeffler, M., Brosteanu, O., Hasenclever, D. et al. (1998). Meta-analysis of chemotherapy versus combined modality treatment trials in Hodgkin’s disease. International Database on Hodgkin’s Disease Overview Study Group. J. Clin. Oncol., 16, 818–29. Marcus, R., Imrie, K., Belch, A. et al. (2005). CVP chemotherapy plus rituximab compared with CVP as first-line treatment for advanced follicular lymphoma. Blood, 105, 1417–23. Mendenhall, C. M., Thar, T. L. and Million, R. R. (1980). Solitary plasmacytoma of bone and soft tissue. Int. J. Radiat. Oncol. Biol. Phys., 6, 1497–501. Meyer, R. M., Gospodarowicz, M. K., Connors, J. M. et al. (2005). Randomized comparison of ABVD chemotherapy with a strategy that includes radiation therapy in patients with limited-stage Hodgkin’s lymphoma: National Cancer Institute of Canada Clinical Trials Group and the Eastern Cooperative Oncology Group. J. Clin. Oncol., 23, 4634–42. Miller, T., Dahlberg, S., Cassady, J. et al. (1998). Chemotherapy alone compared with chemotherapy plus radiotherapy for localized intermediate- and high-grade non-Hodgkin’s lymphoma. N. Engl. J. Med., 339, 21–6. Morton, L. M., Wang, S. S., Devesa, S. S. et al. (2006). Lymphoma incidence patterns by WHO subtype in the United States, 1992–2001. Blood, 107, 265–76. Nogov´a, L., Reineke, T., Josting, A. et al. (2005). Lymphocyte-predominant and classical Hodgkin’s lymphoma – comparison of outcomes. Eur. J. Haematol., 75 (Suppl. 66), 106–10. Orchard, J., Garand, R., Davis, Z. et al. (2003). A subset of t(11;14) lymphoma with mantle cell features displays mutated IgVH genes and includes patients with good prognosis nonnodal disease. Blood, 101, 4975–81. ¨ Pfreundschuh, M., Trumper, L., Kloess, M. et al. (2004a). Two-weekly or 3-weekly CHOP chemotherapy with or without etoposide for the treatment of young patients with good prognosis (normal LDH) aggressive lymphomas: results of the NHL-B1 trial of the DSHNHL. Blood, 104, 626–33. ¨ Pfreundschuh, M., Trumper, L., Kloess, M. et al. (2004b). Two-weekly or 3-weekly CHOP chemotherapy with or without etoposide for the treatment of elderly patients with aggressive
368
lymphomas: results of the NHL-B2 trial of the DSHNHL. Blood, 104, 634–41. ¨ Pfreundschuh, M., Trumper, L., Osterborg, A. et al. (2006). CHOP-like chemotherapy plus rituximab versus CHOP-like chemotherapy alone in young patients with good prognosis diffuse large B-cell lymphoma: a randomised controlled trial by the MabThera International Trial (MlnT) Group. Lancet Oncol., 7, 379–91. Philip, T., Guglielmi, C., Hagenbeek, A. et al. (1995). Autologous bone marrow transplantation as compared with salvage chemotherapy in relapses of chemotherapy-sensitive non-Hodgkin’s lymphoma. N. Engl. J. Med., 333, 1540–5. RCR. (2006). Radiotherapy Dose-Fractionation. The Royal College of Radiologists, Board of the Faculty of Clinical Oncology. London: Royal College of Radiologists. Reyes, F., Lepage, E., Ganem, G. et al. (2005). ACVBP versus CHOP plus radiotherapy for localized aggressive lymphoma. N. Engl. J. Med., 352, 1197–205. Romaguera, J. E., Fayad, L., Rodriguez, M. A. et al. (2005). High rate of durable remissions after treatment of newly diagnosed aggressive mantle-cell lymphoma with rituximab plus hyper-CVAD alternating with rituximab plus high-dose methotrexate and cytarabine. J. Clin. Oncol., 23, 7013–23. Rosen, L. S., Gordon, D., Kaminski, M. et al. (2001). Zoledronic acid versus pamidronate in the treatment of skeletal metastases in patients with breast cancer or osteolytic lesions of multiple myeloma: a phase III double-blind, comparative study. Cancer J., 7, 377–87. Siegel, R. S., Pandolfino, T., Guitart, J. et al. (2000). Primary cutaneous T cell lymphoma: review and current concepts: J. Clin. Oncol., 18, 2908–25. Solal-C´eligny, P., Roy, P., Colombat, P. et al. (2004). Follicular lymphoma international prognostic index. Blood, 104, 1258–65. Solal-C´eligny, P., Imrie, K., Belch, A. et al. (2005). Mabthera (Rituximab) plus CVP chemotherapy for first-line treatment of stage III/IV follicular non-Hodgkin’s lymphoma (NHL): confirmed efficacy with longer follow-up. Blood, 106, Abstr. 350. Soutar, R., Lucraft, H., Jackson, G. et al. (2004). Guidelines on the diagnosis and management of solitary plasmacytoma of bone and solitary extramedullary plasmacytoma. Clin. Oncol. (R. Coll. Radiol.), 16, 405–13. Straus, D. (2006). Treatment of early-stage nonbulky Hodgkin lymphoma. Curr. Opin. Oncol., 18, 432–6. Swerdlow, A. J., Barber, J. A., Hudson, G. V. et al. (2000). Risk of second malignancy after Hodgkin’s disease in a collaborative British cohort: the relation of age at treatment. J. Clin. Oncol., 18, 498–509. Thomas, J., Ferme, C., Noordijk, E. et al. (2004). Six cycles of EBVP followed by 36 Gy involved-field irradiation vs. no irradiation in favourable supradiaphragmatic clinical stages I-II Hodgkin’s lymphoma: the EORTC-GELA strategy in 771 patients. Eur. J. Haematol., 73 (Suppl. 64), Abstr. 40. Tiemann, M., Schrader, C., Klapper, W. et al. (2005). Histopathology, cell proliferation indices and clinical outcome in 304 patients with mantle cell lymphoma (MCL): a clinicopathological study from the European MCL Network. Br. J. Haematol., 131, 29–38.
The lymphomas and myeloma
UK Myeloma Forum. (2001). Guideline. Diagnosis and management of multiple myeloma. Br. J. Haematol., 115, 522–40. Wallace, W. H., Anderson, R. A. and Irvine, D. S. (2006). Fertility preservation for young patients with cancer: who is at risk and what can be offered? Lancet Oncol., 6, 209–18. Yahalom, J. and Mauch, P. (2002). The involved field is back: issues in delineating the radiation field in Hodgkin’s disease. Ann. Oncol., 13 (Suppl. 1), 79–83.
FURTHER READING: LYMPHOMAS Anonymous. (1993). The International Non-Hodgkin’s Lymphoma Prognostic Factors Index Project. N. Engl. J. Med., 329, 987–94. Bessell, E. (2002). Radiotherapy for extranodal non-Hodgkin’s lymphoma. CME Cancer Med., 1, 9–12. Connors, J. (2005). State-of-the-art therapeutics: Hodgkin’s lymphoma. J. Clin. Oncol., 23, 6400–8. ¨ Kuppers, R. (2005). Mechanisms of B-cell lymphoma pathogenesis. Nature Rev. Cancer, 5, 251–62. Lee, S. J., Schover, L. R., Partridge, A. H. et al. (2006). American Society of Clinical Oncology recommendations on fertility preservation in cancer patients. J. Clin. Oncol., 24, 2917–31. Moskowitz, C. (2006). An evidence-based approach to the management of Hodgkin’s lymphoma. In Oncology: An Evidence-Based Approach, ed. A. Chang, D. Hayes, H. Pass et al. New York: Springer, pp. 1231–46. Quon, H. and Gliedman, P. (2002). Late complications after treatment of Hodgkin’s disease. In Atlas of Clinical Oncology: Malignant Lymphomas, ed. M. Grossbard. Hamilton: BC Decker, pp. 442–55. Yahalom, J. (2002). Radiation therapy for Hodgkin’s disease. In Atlas of Clinical Oncology: Malignant Lymphomas, ed. M. Grossbard. Hamilton: BC Decker, pp. 428–41.
Zelenetz, A. and Horwitz, S. (2006). The non-Hodgkin’s lymphomas. In Oncology: An Evidence-Based Approach, ed. A. Chang, D. Hayes, H. Pass et al. New York: Springer, pp.1247–75.
FURTHER READING: PARAPROTEINAEMIAS Attal, M., Harousseau, J.-L., Facon, T. et al. (2003). Single versus double autologous stem-cell transplantation for multiple myeloma. N. Engl. J. Med., 349, 2495–502. Barlogie, B., Shaughnessy, J., Tricot, G. et al. (2004). Treatment of multiple myeloma. Blood, 103, 20–32. Berenson, J. R., Hillner, B. E., Kyle, R. A. et al. (2002). American Society of Clinical Oncology clinical practice guidelines: the role of bisphosphonates in multiple myeloma. J. Clin. Oncol., 20, 3719–36. Dammacco, F., Castoldi, G. and R¨odjer, S. (2001). Efficacy of epoetin alfa in the treatment of anaemia of multiple myeloma. Br. J. Haematol., 113, 172–9. Dispenzieri, A., Kyle, R. A., Lacy, M. Q. et al. (2003). POEMS syndrome: definitions and long-term outcome. Blood, 101, 2496–506. Kuehl, W. M. and Bergsagel, P. L. (2002). Multiple myeloma: evolving genetic events and host interactions. Nature Reviews Cancer, 2, 175–87. Myeloma Trialists’ Collaborative Group. (1998). Combination chemotherapy versus melphalan plus prednisone as treatment for multiple myeloma: an overview of 6633 patients from 27 randomized trials. Myeloma Trialists’ Collaborative Group. J. Clin. Oncol., 16, 3832–42. Richardson, P., Barlogie, B., Berenson, J. et al. (2003). A phase 2 study of bortezomib in relapsed, refractory myeloma. N. Engl. J. Med., 348, 2609–17. Singhal, S., Mehta, J., Desikan, R. et al. (1999). Antitumor activity of thalidomide in refractory multiple myeloma. N. Engl. J. Med., 341, 1565–71.
369
32
CENTRAL NERVOUS SYSTEM Sean Elyan
Introduction This chapter describes the management of tumours in adults in the following anatomical areas: cerebral convexity and cerebral hemispheres, the skull base, the pituitary, the pineal region and the spinal cord. Central nervous system (CNS) tumours are heterogeneous. The terms malignant and benign are not very useful because: r Even small slowly growing tumours can cause severe symptoms because the brain is enclosed in a rigid skull. r Surgery can be difficult because many tumours are infiltrating and often lie close to critical structures. r Most of these tumours rarely, if ever, metastasise outside the CNS. r Slow-growing tumours may transform into a much more aggressive variant. This chapter does not deal with metastatic disease to the CNS that is considered in other relevant chapters, although management of cerebral metastases that require specialist neuro-oncology input is discussed briefly. CNS tumours in children are considered in Chapter 37 (see p. 432).
mit cranial nerves III, IV, VI and the maxillary branch of V.
Incidence and epidemiology Primary brain and CNS tumours are fairly uncommon. The incidence of disease is approximately 16 in 100 000 per annum. Between 1995 and 2000, 6500 primary tumours of the brain and CNS were registered annually in England and Wales, where brain tumours account for only 1.6% of all cancers. Approximately 3700 deaths occur per year in England and Wales that are attributable to primary brain and CNS tumours. On average, each GP in England sees one patient with a brain tumour every 7 years. The peak disease incidence occurs in the 70- to 80year-old age group. There has been an increase in the number of registrations of these tumours in recent years, probably because of increased investigation. There may be a significant under-registration of intracranial tumours; perhaps as many as half are not being recorded in cancer registries (Pobereskin and Chadduck, 2000).
Risk factors and aetiology Anatomy The tentorium separates the supratentorial from the infratentorial areas of the brain. The motor and sensory cortices lie at the central sulcus. Broca’s area (frontal, above the lateral sulcus) is responsible for expressive loss of speech, and Wernicke’s area (temporal, posterior end of the lateral sulcus) is responsible for receptive loss of speech. The ventricular system is lined with ependyma. CSF travels from the third to the fourth ventricle through the cerebral aqueduct and from the fourth ventricle to the subarachnoid space through the foramina of Magendie (median) and Luschka (lateral). The anterior and intermediate lobes of the pituitary arise from Rathke’s pouch. The cavernous sinuses trans-
370
The risk of developing a primary CNS tumour is related to increasing age, female gender and higher socioeconomic status. There is, in general, a lower incidence in less-developed countries. The acquired immune deficiency syndrome (AIDS) is a well-recognised cause of cerebral lymphoma (Beral et al., 1991). The only other certain causative factors are inherited cancer syndromes and ionising radiation. A number of familial autosomal dominant syndromes give rise to an increased risk of CNS tumours, which include neurofibromatosis types 1 (incidence 1 in 3000) and 2 (1 in 40 000), von Hippel-Lindau syndrome, tuberous sclerosis, Li-Fraumeni syndrome Cowden’s disease, Turcot’s syndrome and naevoid basal cell carcinoma
Central nervous system
Table 32.1. Inherited syndromes and their associated tumours Syndrome
Nervous system effects
Neurofibromatosis type 1
Neurofibromas, malignant nerve sheet tumours, optic nerve gliomas,
Neurofibromatosis type 2
Bilateral acoustic schwannomas, multiple meningiomas,
Von Hippel-Lindau syndrome
Haemangioblastomas
Tuberous sclerosis
Subependymal giant cell astrocytoma
Li-Fraumeni
Astrocytomas/PNET
astrocytomas astrocytomas, glial hamartomas
Cowden’s disease
Dysplastic gangliocytoma of the cerebellum
Turcot’s syndrome
Medulloblastoma, glioblastoma
Naevoid basal cell carcinoma syndrome
Medulloblastoma
(Gorlin syndrome) PNET = primitive neuroectodemal tumour.
(Gorlin) syndrome. Table 32.1 summarises these syndromes and their associated tumours.
Pathology For many patients, a tissue diagnosis is not possible. As a result, unlike most other tumour groups, the final diagnosis may depend on the results of imaging rather than a biopsy. The classification of CNS tumours is complex. The World Health Organisation (WHO) produced a classification in 1993, most recently updated in 2000, which is now widely used (Table 32.2). The term ‘high grade’ includes grades III and IV and ‘low grade’ includes grade I and II tumours. The terms ‘benign’ and ‘malignant’ are best avoided. r WHO grade I includes tumours with low proliferative potential, frequently discrete and the possibility of a cure following surgical resection alone. r WHO grade II includes tumours that are generally infiltrating and with few mitoses, but that can recur. Some tumour types may progress to higher grades of malignancy. r WHO grade III includes tumours with histological evidence of malignancy, generally in the form of mitotic activity, clear signs of infiltration and anaplasia. r WHO grade IV includes tumours that are mitotically active, necrosis-prone, and generally associated with a rapid pre- and postoperative growth.
The majority of pituitary tumours (95%) are adenomas. The remainder include craniopharyngiomas, Rathke’s cleft cysts and meningiomas. There is a wide variety of tumours of the base of the skull, ranging from slow growing to very malignant. The most common tumour occurring at this site is the schwannoma, a low-grade and usually slow-growing tumour that arises from the acoustic nerve. Tumours involving the pineal gland are very unusual. There are three main histological types: germcell tumours (GCTs), astrocytomas and pineal parenchymal tumours, which include pineocytomas and pineoblastomas. In this chapter, tumours are divided by their anatomical site and subdivided using the WHO histopathological classification.
Spread Tumours of the CNS rarely metastasise outside the CNS axis. Table 32.3 shows the main modes of spread.
Clinical presentation Patients with primary CNS tumours present with signs and symptoms attributable to the mass effect of the tumour. Patients may initially present to a general practitioner, as an acute medical admission, or through a range of specialities including ophthalmology, neurology and gynaecology.
371
Sean Elyan
Table 32.2. The WHO classification of CNS tumours Type
Examples
Neuroepithelial tumours
Gliomas Astrocytoma Oligodendroglioma Ependymoma Neuronal and mixed neuronal-glial tumours Gangliocytoma Paraganglionoma Nonglial tumours Pineal parenchymal tumour
Meningeal tumours
Meningioma
Germ-cell tumours
Germinoma Teratoma
Tumours of the sellar region
Pituitary adenoma Craniopharyngioma
Primary CNS lymphoma Tumours of peripheral nerves that affect the CNS
Schwannoma
Metastatic tumours From Kleihues and Cavenee (2000).
Table 32.3. Routes of spread of CNS tumours Route
Site
Examples
Local invasion
To involve contiguous structures
Gliomas
Local pressure
To compromise local structures
Pituitary adenomas
CSF spread
To ventricles and spine
PCNSL, medulloblastoma, ependymoma
Haematogenous spread (rare)
To lungs, liver, bone
Medulloblastoma
CSF = cerebrospinal fluid; PCNSL = primary central nervous system lymphoma.
For intracranial tumours, the main symptoms are headaches, seizures, changes of mental state, unilateral deafness and progressive neurological deficit and hormonal dysfunction (particularly for pituitary tumours). For spinal tumours, the main symptoms are pain and a progressive loss of neurological function. The finding of new neurological signs together with symptoms is more suggestive of pathology than symptoms alone. New neurological symptoms in patients with a past history of cancer suggest metastatic disease. Some tumours may cause cognitive, expressive and psychological problems and the patient may not be able 372
to explain his or her symptoms fully. Therefore, it may be very important to get a history from a friend or relative. Many patients need a lot of psychological, social and physical support and, because of the poor prognosis of many CNS tumours, management should be aimed at maximising the patient’s quality of life.
Investigation and staging All patients with CNS tumours should have their investigations and management plan coordinated by a neurooncology multidisciplinary team in the forum of a multidisciplinary meeting.
Central nervous system
The key investigations include routine blood tests, including HIV status where relevant; imaging, angiography, other laboratory tests and biopsy. Neuroradiological imaging is used, particularly CT scanning and MRI. Where biopsy would be too risky or in cases where the age or performance status of the patient would preclude biopsy or tumour resection, a presumptive diagnosis of a primary CNS tumour may need to be made on radiological grounds alone. A CT scan should reliably exclude a tumour in the majority of cases, but may miss early tumours, especially in the temporal lobes and posterior fossa. MRI should be performed as an initial investigation in a patient who has persisting symptoms despite a normal CT scan. MRI also gives useful information for planning surgery and radiotherapy. Angiography may be useful in planning surgical treatment for some CNS tumours (e.g. spinal tumours). Other laboratory tests may be performed, such as an assessment of pituitary function or serum tumour markers of primary GCTs. Imaging has a high sensitivity for identifying the presence of a tumour but is unreliable in identifying grade and type. Whenever possible, a histopathological specimen should be obtained to classify the tumour, plan appropriate treatment and determine prognosis. Sometimes a diagnosis can be made by cytological examination of the cerebrospinal fluid (CSF) or, for pineal GCTs, simply by finding raised tumour markers. Tissue can be obtained either by biopsy or tumour resection via one of the following methods: r Stereotactically guided biopsy. r Radiologically guided needle biopsy. r Radiologically guided open biopsy and/or resection. r Endoscopic biopsy. r Electrophysiologically guided resection. There should be a preoperative discussion among the neurosurgeon, neuropathologist and neuroradiologist about the best approach to surgery and the processing of tissue specimens. Intraoperative histopathological diagnosis is particularly valuable during needle biopsy and helps to ensure that enough of the right tissue is obtained. It can also provide information that will influence the course of the operation. Various other tests include the following: r Molecular diagnostic tests such as loss on chromosomes 1p and 19q and MGMT expression, which are expected to become more important (Cairncross et al., 1998; Stupp et al., 2005).
r Pituitary function. r Visual field assessment (pituitary). r Ophthalmology review (primary CNS lymphoma [PCNSL]).
r Tumour markers (primary GCTs). r Other imaging techniques, including MR spectroscopy, single photon emission computed tomography (SPECT) and positron emission tomography (PET), which are currently predominantly research tools and are not in day-to-day use. Certain groups of patients, for example those with neurofibromatosis types 1 and 2, are at a particular risk of developing intradural spinal tumours. These patients require monitoring and early resection if the tumours enlarge or cause symptoms. Patients with skull base tumours also may require the following assessments before treatment: r Audiological testing. r Auditory-evoked brain stem responses testing. r Vestibular testing. r Prosthetic assessment to establish the need for ocular, aural or skull bone replacement. r Speech and language therapy assessment and explanation to the patient of likely postoperative impairment. r Dietetic assessment. TNM staging for brain tumours has been withdrawn following publication of the fourth edition of the AJCC manual for staging of cancer and does not contribute to the management of most primary CNS tumours. Chang staging does determine treatment protocols in medulloblastoma.
Treatment of brain tumours The general principles of treating tumours in the various anatomical sites are described, followed by a detailed discussion of radiotherapy.
Treatment of low-grade glioma (WHO grades I and II) Up to 40% of low-grade gliomas (LGGs) diagnosed on imaging are found to have high-grade histopathological features and so all patients need to have a confirmed histopathological diagnosis unless a biopsy would be too risky for the patient or is otherwise inappropriate. The options for initial management are either watchful waiting or immediate surgery. The EORTC criteria (Pignatti et al., 2002) can help identify patients who are 373
Sean Elyan
at an increased risk of rapid deterioration and who may benefit from early intervention. Radiotherapy may be used in the following situations: r Patients with persisting neurological symptoms and significant residual tumour. r Patients with regrowth following surgery. r Patients with evidence of tumour progression from low to high grade. Radiotherapy improves survival but the benefit is evident, whether it is given initially or on progression (van den Bent et al., 2005). The exact role of chemotherapy is uncertain.
for patients with small, incidental meningiomas. In patients with skull vault meningiomas, surgical resection can prevent both further disease progression and the associated deterioration in neurological function, although recurrence may occur. The indications for radiotherapy (following histopathological confirmation) are as follows: r WHO histopathological grade II/III tumour. r Invasion by tumour of adjacent structures. r Relapsed disease. r Surgical contraindication.
Treatment of primary CNS lymphoma (PCNSL) Treatment of high-grade glioma (WHO grades III and IV) High-grade gliomas (HGGs) include glioblastomas, anaplastic astrocytomas, anaplastic oligodendrogliomas and anaplastic ependymomas. The important prognostic factors are age, performance status and co-morbidity, tumour type and grade and presence or absence of seizures (Bleehen and Stenning, 1991). The principles of treatment are to increase survival while maximising the patient’s functional capability and quality of life. It is important to identify patients who need urgent surgery (e.g. emergency decompression or shunt insertion for hydrocephalus) and patients suitable for elective surgery to debulk the tumour. Some patients are unfit for any intervention. Radical radiotherapy may be considered after a histopathological diagnosis has been confirmed (Laperriere et al., 2002). Adjuvant chemotherapy has been shown to have a small but significant survival advantage (Stewart, 2002) but is currently not widely used in the UK. Both the use of carmustine implants in combination with surgical resection and of adjuvant postoperative temozolomide (Stupp et al., 2005) concurrently with radiotherapy in patients with a newly diagnosed HGG have been shown to improve survival by approximately 2 to 3 months.
Treatment of meningioma Meningiomas most commonly arise in the skull vault, are usually low-grade tumours (WHO grade I) and have an indolent course. Disease management depends on signs, symptoms, the patient’s fitness and the tumour site and size. Watching and waiting are appropriate 374
Primary CNS lymphoma is discussed in Chapter 31 (see p. 353).
Treatment of medulloblastoma Medulloblastoma is a rare tumour found in adults, and probably fewer than 50 patients present annually in England and Wales. This type of tumour usually occurs in the posterior fossa and is associated with cerebellar symptoms and raised intracranial pressure. It commonly spreads through the craniospinal axis. Patients need to have MRI imaging of the brain and whole spine before surgery. Surgery to remove as much tumour as possible, followed by radiotherapy to the whole neuraxis, is the most appropriate treatment. The role of chemotherapy in addition to surgery and radiotherapy in the management of adults is not established but is increasingly being used (Brandes et al., 2003).
Treatment of ependymoma Ependymoma usually occurs in association with the ventricles and patients may present with an obstruction or a mass lesion. There is a tendency for ependymoma to spread via the CSF throughout the neuraxis but less commonly than for medulloblastoma. Patients need to have an MRI of the brain and whole spine before surgery. Standard treatment depends on the type and grade of the tumour and the details remain controversial. Surgery is commonly used to remove as much tumour as possible and this may be followed by radiotherapy, sometimes to the whole neuraxis. The role of chemotherapy following surgery and radiotherapy in the management of adults is not established.
Central nervous system
Treatment of pineal tumours Patients with pineal tumours most commonly present with symptoms and signs of raised intracranial pressure. Treatment varies according to the tumour type, the level of tumour marker and CSF cytology status. Many GCTs and pineal parenchymal tumours are curable with appropriate management. GCTs occurring elsewhere within the CNS are managed according to principles similar to those used in the pineal region. Surgical options include stereotactic or endoscopic biopsy, CSF diversion or resection for parenchymal tumours. The diagnosis of GCTs may be confirmed in some cases by the measurement of hormonal markers (αFP and βhCG). When positive, no biopsy is required; in marker-negative patients, a biopsy is necessary. Radiotherapy may be needed, especially for GCTs, but only once the diagnosis has been confirmed from the histopathological evaluation of biopsy material. Craniospinal axis irradiation may be needed for patients with pineoblastoma and metastatic GCTs. Stereotactic radiotherapy may be appropriate for low-grade pineocytomas. Chemotherapy forms part of the management plan of GCTs but has no proven role in the treatment of other pineal tumours.
Treatment of metastases Metastases in the brain occur in 20 to 40% of patients who have other primary cancers and are usually associated with a poor prognosis. The majority of patients require appropriate palliative support and treatments depending on the site of the primary tumour. In the following circumstances involvement of the neurooncology team should be considered: r When cerebral metastases are the first sign of malignant disease. r When the imaging findings suggestive of metastases are in doubt. r For patients with solitary metastases and with a prognosis that warrants considering neurosurgical intervention, in whom complete surgical excision should be considered but only when the risk of unacceptable complications is low (Patchell et al., 1990). r Following the resection of solitary metastases, when postoperative radiotherapy may reduce the likelihood of intracranial relapse (Patchell et al., 1998). r When stereotactic radiotherapy is considered as an alternative to surgery in patients with small (< 3 cm)
solitary (or occasionally multiple) tumours for which the histopathological diagnosis is known.
Treatment of pituitary and pituitary-related tumours All patients with pituitary mass lesions should be referred to an endocrinologist who specialises in pituitary dysfunction and the patient’s case should be discussed by a multidisciplinary team so that a management plan can be agreed on. Following the appropriate investigation and management of hyposecretion (with replacement therapy) or hypersecretion (e.g. somatostatin analogues and dopamine agonists for acromegaly), patients should be considered for surgery or radiotherapy. The role of both has decreased with the increased use of endocrinological treatments. The indications for surgery are to control the tumour mass effect and manage hypersecretion. Most operations are via the trans-sphenoidal route. Transcranial and subfrontal approaches are reserved for those patients with tumours extending above the optic apparatus. The indications for radiotherapy are in patients who are unfit for surgery, following tumour regrowth after surgery, and when there is persistent hormonal hypersecretion. Stereotactic approaches are now being used in patients with small functioning tumours (Mondok et al., 2005). Figure 32.1 shows a radiotherapy plan for treatment of a pituitary adenoma. Patients treated for pituitary tumours require followup in the form of regular assessment by an endocrinologist, regular assessment of the visual fields by perimetry with access to a neuro-ophthalmologist as clinically indicated and MRI imaging.
Treatment of skull base tumours All patients benefit at presentation from discussion by a specialist multidisciplinary team to agree on a management plan. Biopsy, either endoscopically or under CT guidance, is usually required depending on the site involved. Open intracranial biopsy is sometimes necessary. Surgery, often with a variety of expertise (e.g. neurosurgery, ENT, maxillofacial), is normally the treatment of choice. Complete excision may not be possible because of involvement of the surrounding structures. Ventricular shunting or draining may be required for large tumours. Some vascular tumours require embolisation. 375
Sean Elyan
Isodose % 95 70 45
(a)
Extramedullary tumours account for 70% of intradural tumours; with few exceptions, they are histologically benign. Extramedullary tumours include nerve sheath tumours (schwannoma, neurofibroma), meningiomas and myxopapillary ependymomas. Metastases are more unusual. Intramedullary tumours arise from glial cells in 80% of cases. The primary glial tumours include astrocytomas, ependymomas, gangliogliomas and oligodendrogliomas. Haemangioblastomas, metastases, nerve sheath tumours, melanocytomas and vascular tumours account for the remainder. In patients with low-grade tumours, the main aim of treatment is to prevent further neurological deterioration. Most intradural tumours can be treated with surgery alone and be completely excised. However, excision of intramedullary glial tumours can cause further damage to the spinal cord without any survival advantage if the lesion is high grade. For this reason, intraoperative pathology is essential. In addition intraoperative neurophysiological recording helps minimise the risk of spinal cord injury. Radiotherapy is an appropriate treatment for patients with incomplete resection and high-grade histology. Craniospinal axis irradiation should be considered for the treatment of high-grade spinal ependymoma. There is currently no proven role for the use of chemotherapy in adult patients with primary spinal tumours.
Radiotherapy techniques (b) Figure 32.1. A radiotherapy plan for a pituitary adenoma using a three-field technique. (a) Transverse section. (b) Sagittal section showing the angle of the anterior field to avoid the eyes.
Radiotherapy may be considered where complete excision is not possible, or for consolidation. External beam conformal radiotherapy is well established and there is increasing evidence of a role for stereotactic techniques, particularly for patients with small acoustic schwannomas.
Treatment of spinal cord tumours Intradural spinal cord tumours may either be within the spinal cord (intramedullary) or outside the cord (extramedullary).
376
Brain For patient preparation, positioning and immobilisation, the principles of radiotherapy for brain tumours are the same regardless of the primary tumour type. The patient is positioned supine or prone depending on the locality of the tumour. Immobilisation is achieved with a Perspex® or thermoplastic shell. Treatment for brain tumours is localised using CT planning, ideally with 3D conformal capabilities. The additional information from the preoperative crosssectional imaging is essential and, where available, pre- and postoperative image co-registration should be used. Axial slices at appropriate intervals, depending on the tumour size and proximity of surrounding critical structures, of 0.2 to 0.5 cm are taken through the area of interest. Radio-opaque markers in the sagittal and coronal plane facilitate localisation of fields from
Central nervous system
Table 32.4. Radiotherapy GTV, PTV and dose/fractionation for different CNS tumours Tumour type
GTV
PTV
Dose/fractionation
High-grade glioma (radical)
Enhancing preoperative
3 cm margin
60 Gy in 30 fractions
tumour High-grade glioma (palliative)
Enhancing tumour
3 cm margin
30 Gy in 6 fractions
Low-grade glioma
Preoperative tumour mass
2 cm margin
45 to 50.4 Gy in 25 to
Meningioma, low grade
Preoperative tumour mass
2 cm margin
45 to 50 Gy in 25 fractions
Meningioma, high grade
Preoperative tumour mass
3 cm margin
60 Gy in 30 fractions
Pineocytoma
Preoperative tumour mass
1 cm margin
45 to 50 Gy in 25 fractions
Pituitary adenoma
MRI mass
1 cm margin
45 Gy in 25 fractions
Skull base
As for histological type
28 fractions
Craniospinal axis (e.g. medulloblastoma, pineoblastoma,
Craniospinal axis
34 Gy in 18 fractions
Boost
2 cm margin
11 Gy in 6 fractions
Prechemotherapy enhancing
3 cm margin
45 Gy in 25 fractions
2–3 cm margin
45 to 50 Gy in 25 to
high-grade ependymoma) Primary cerebral lymphoma (postchemotherapy) Spinal cord
volume MRI tumour mass
30 fractions
the CT scan baseline marked on the shell. The GTV is defined on each slice and the PTV is grown manually or automatically. The GTV and PTV are defined for different situations in Table 32.4. The same treatment principles apply for patients receiving palliative hypofractionated radiotherapy for HGGs (Ford et al., 1997). Critical structures are outlined including the optic chiasm, eyes, brain stem and pituitary. Dose is prescribed to the ICRU 50 reference point and delivered using a three-to four-field technique. Each field should be treated daily, Monday to Friday. The PTV minimum should be no less than 95% and the PTV maximum should be no more than 107% of the dose prescribed to the ICRU 50 reference point. No point outside the PTV should receive more than 105%. The treatment is usually planned using a two- or three-field technique. Anterior and lateral field wedging is usually necessary for the two-field technique used in lateralised tumours. The plan should be verified in the simulator using lateral digital reconstructed radiographs (DRRs) reconstructed from CT scans before starting treatment; at least one portal image should be taken in the first three fractions on the linear accelerator and weekly throughout treatment thereafter.
The toxicity of brain radiotherapy is shown in Table 32.5.
Spine Patient preparation, positioning and immobilisation for spinal tumours involve the patient lying prone on a vacuum bag. For spinal tumours, treatment is localised using CT planning, ideally with 3D conformal capabilities as for brain tumours. Lateral and midline tattoos are placed to position the patient. The additional information from the preoperative cross-sectional imaging (particularly MRI) is essential and where available pre- and postoperative image co-registration should be used. Axial slices at an appropriate intervals, depending on the tumour size and proximity of surrounding critical structures, of 0.3 to 0.5 cm are taken through the area of interest. The GTV is defined on each slice and the PTV is grown manually or automatically. The GTV and PTV are defined for different situations in Table 32.4. Care must be taken in treatment planning to avoid exit beams passing through renal tissue or ovaries. Dose is prescribed to the ICRU 50 reference point and verification is as for brain tumours.
377
Sean Elyan
Table 32.5. Toxicity of radiotherapy to the brain Effect
Management
Acute Tiredness
Advice about fatigue (e.g. goal setting, treat anxiety/depression,
Skin reaction
Aqueous cream
Nausea and vomiting
Antiemetics
Hairloss
Prosthetic
moderate exercise if tolerable)
Late Memory loss
Neuropsychology support
Pituitary hypofunction
Endocrine review in relevant patients
Second malignancies (1% approx.; Bliss et al., 1994)
Warn patient
Optic chiasm damage (< 1%)
Warn patient
Table 32.6. Toxicity of radiotherapy to the spine and craniospinal axis Effect
Management
Acute Tiredness
Advice about fatigue (as in Table 32.5)
Skin reaction
Aqueous cream
Nausea and vomiting
Antiemetics
Hair loss
Prosthetic
Diarrhoea
Loperamide
Myelosuppression
Regular FBC –G-CSF
Late Memory loss
Neuropsychology support
Pituitary hypofunction
Endocrine review in relevant patients
Second malignancies (1% approx.)
Warn patient
Optic chiasm damage (< 1%)
Warn patient
Ovarian failure
Avoid ovaries with planning
Spinal cord damage
Warn patient
The treatment plan usually involves a two-field technique with wedged oblique fields. The toxicity of spinal radiotherapy is shown in Table 32.6.
Craniospinal axis Patient preparation, positioning and immobilisation for craniospinal axis irradiation are complex. The principles are the same regardless of the primary tumour type. The patient is positioned prone. Immobilisation 378
is achieved with a Perspex® or thermoplastic shell and with the patient lying in a vacuum-formed bag. Skull rotation is corrected by aligning outer canthus markers, and the spine is screened on the simulator with central tattoos to enable correct positioning. The patient is CT scanned in the treatment position. The whole brain is treated, down to the third cervical vertebra. Shielding to the orbit, nasopharynx and pharynx can be delineated using lateral films or from outlining on the CT planning scan. The spinal volume is defined on the planning CT scan. Where appropriate the boost
Central nervous system
PTV is outlined, as in Table 32.4. The composite target area is drawn on the central slice, and an outline is taken through the central volume. Dose is prescribed to the ICRU 50 reference point and verification is as for brain tumours. The treatment plan is done with lateral skull fields. Centring the field on the outer canthus avoids beam divergence through the contralateral eye. Gantry twist to match the beam divergence from the upper spinal field enables easier field matching. The spine is treated with two fields matched at the C3 level and normally the mid- to lower thoracic spine. The cervical and thoracic junction should be moved by 1 cm every 10 Gy, which is achieved by increasing or reducing the field size accordingly. Homogeneous dose distribution can often be best achieved by utilising a mixture of photons and electrons for the spinal fields. Boost volumes are planned as mentioned earlier. The plan should be verified in the simulator using lateral DRRs reconstructed from a CT scan before starting treatment, and at least one portal image should be taken in the first three fractions on the linear accelerator and weekly throughout treatment thereafter. The toxicity of radiotherapy is shown in Table 32.6
Recurrent disease The prognosis for patients with recurrent CNS tumours is generally poor. For patients with gliomas, a number of options are available: r Repeat surgery with consideration given to use of chemotherapy implants. r Administer chemotherapy using either combination treatment or single-agent temozolomide. Oligodendrogliomas appear to be more chemosensitive than other types of glioma (van den Bent et al., 2003). r Use either of these chemotherapy regimens: r Procarbazine 200 mg/m2 days 1 to 10. r CCNU 200 mg/m2 day 1. r Vincristine 1.4 mg/m2 (max 2 mg) day 1. r Repeated every 42 days. Or r Temozolomide 200 mg/m2 days 1 to 5. r Repeated every 28 days. Radiotherapy can be considered for small recurrences in previously irradiated patients although suitable patients are rare. Recurrent pituitary, skull base and pineal tumours may in some instances be managed surgically.
Table 32.7. Prognosis for CNS tumours Tumour
Five-year survival
All malignant brain tumours
17%
Grade IV glioma
5%
Grade III glioma
25%
Grade II glioma
65%
PCNSL
25%
PCNSL = primary CNS lymphoma.
Prognosis The prognosis for CNS tumours varies very widely. Some tumours such as pituitary adenomas may not affect long-term survival at all, whereas elderly patients with an HGG have a median survival of 3 months or less. Table 32.7 shows the prognosis for some CNS tumours. r Radical radiotherapy for HGGs prolongs survival by about 6 months. r Adjuvant chemotherapy prolongs survival by about 2 months. r Concurrent chemoradiotherapy for HGGs prolongs survival by about 2.5 months in addition to radiotherapy alone.
Prognostic factors The prognosis of gliomas is determined by the tumour grade, the patient’s age, the extent of surgery, the presence of fits and the patient’s performance status at presentation (Bleehen and Stenning, 1991). Similar factors have been shown to influence the prognosis in other tumour types. For medulloblastoma, stage, the presence of metastases, subtotal resection of less than 75% of the tumour and positive CSF cytology after surgery have been shown to affect prognosis (Laurent et al., 1985).
Areas of current interest and current clinical trials Prognosis Molecular indicators of prognosis in CNS tumours are being used more often. An association between a profile of loss on chromosomes 1p and 19q in anaplastic oligodendrogliomas has recently been identified as predictive of chemotherapy responsiveness. 379
Sean Elyan
Downregulation of the MGMT gene may also be a predictor of chemoresponsiveness to alkylating agents such as temozolomide in HGGs.
Radiotherapy Dose For HGGs in the brain the optimum dose of radiotherapy has been the point of investigation. Although there is some evidence that dose escalation can increase the control of tumours, it also increases the risk of radionecrosis. The data on stereotactic boosting are uncertain.
Stereotactic radiotherapy Stereotactic radiotherapy is increasingly being used as an alternative to surgery for solitary cerebral metastases and in the treatment of some pituitary and skull base tumours. This treatment has not been subjected to randomised controlled clinical trials.
Use of chemotherapy Adjuvant A meta-analysis of the use of adjuvant nitrosoureabased chemotherapy has shown a small but significant increase in median survival of approximately 2 months and an increase in 2-year survival from 10 to 15% (Stewart, 2002). This adjuvant therapy has not come into wide use except in young patients who have good performance status.
Concurrent chemotherapy Concurrent chemotherapy plus radiotherapy has recently been investigated in two randomised clinical trials (Stupp et al., 2005). Using concurrent temozolomide (75 mg/m2 ) and radiotherapy to 60 Gy in patients with grade IV tumours prolongs median survival by 2.5 months and increases 2-year survival from 10 to 26%. This approach is becoming the standard for patients with good performance status.
Implantable chemotherapy (carmustine implants) Implantable chemotherapy wafers, used at the time of the initial radical resection and on subsequent relapse after conventional first-line treatment, have been investigated in two small randomised clinical trials. The evidence suggests that these treatments delay relapse, but the impact on the patient’s quality of life remains uncertain. 380
Ongoing clinical trials The BR12 trial compares temozolomide with PCV chemotherapy in patients with recurrent malignant glioma. EORTC 22033–26033 is an LGG trial. Primary chemotherapy with temozolomide is compared to radiotherapy alone in patients with LGGs with stratification for genetic 1p loss. It is a phase III study. The Neo-adjuvant Carboplatin Trial Open Treatment is a multicentre, phase II study of carboplatin preirradiation in patients with primary glioblastoma multiforme following biopsy. EORTC 22952 is a randomised phase III trial that compares no radiotherapy with whole-brain radiotherapy for one to three brain metastases from solid tumours, following surgical resection or radiosurgery.
REFERENCES Beral, V., Peterman, T., Berkelman, R. et al. (1991). AIDS-associated non-Hodgkin lymphoma. Lancet, 337, 805–9. Bleehen, N. M. and Stenning S. P. (1991). A Medical Research Council trial of two radiotherapy doses in the treatment of grades 3 and 4 astrocytoma. The Medical Research Council Brain Tumour Working Party. Br. J. Cancer, 64, 769–74. Bliss, P., Kerr, G. R. and Gregor, A. (1994). Incidence of second brain tumours after pituitary irradiation in Edinburgh 1962–1990. Clin. Oncol. (R. Coll. Radiol.), 6, 361–3. Brandes, A. A., Ermani, M., Amista, P. et al. (2003). The treatment of adults with medulloblastoma: a prospective study. Int. J. Radiat. Oncol. Biol. Phys , 57, 755–61. Cairncross, J. G., Ueki, K., Zlatescu, M. C. et al. (1998). Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J. Natl. Cancer Inst., 90, 1473–9. Ford, J. M., Stenning, S. P., Boote, D. J. et al. (1997). A short fractionation radiotherapy treatment for poor prognosis patients with high grade glioma. Clin. Oncol. (R. Coll. Radiol.), 9, 20–4. Kleihues, P. and Cavenee, W. K. (2000). Pathology and Genetics of Tumours of the Nervous System. Lyon: IARC Press. Laperriere, N., Zuraw, L. and Cairncross, G. (2002). Radiotherapy for newly diagnosed malignant glioma in adults: a systematic review. Radiother. Oncol., 64, 259–73. Laurent, J. P., Chang, C. H. and Cohen, M. E. (1985). A classification system for primitive neuroectodermal tumors (medulloblastoma) of the posterior fossa. Cancer, 56, 1807–9. Mondok, A., Szeifert, G. T., Mayer, A. et al. (2005). Treatment of pituitary tumors: radiation. Endocrine, 28, 77–85. Patchell, R. A., Tibbs, P. A., Walsh, J. W. et al. (1990). A randomized trial of surgery in the treatment of single metastases to the brain. N. Engl. J. Med., 322, 494–500. Patchell, R. A., Tibbs, P. A., Regine, W. F. et al. (1998). Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. J. A. M. A., 280, 1485–9.
Central nervous system
Pignatti, F., van den Bent, M., Curran, D. et al. (2002). Prognostic factors for survival in adult patients with cerebral low-grade glioma. J. Clin. Oncol., 20, 2076–84. Pobereskin, L. H. and Chadduck, J. B. (2000). Incidence of brain tumours in two English counties: a population based study. J. Neurol. Neurosurg. Psychiatry, 69, 464–71. Stewart, L. A. (2002). Chemotherapy in adult high grade glioma: a systemic review and meta-analysis of individual patient data from 12 randomised trials. Lancet, 359, 1011–18. Stupp, R., Mason, W. P., van den Bent, M. J. et al. (2005).
Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N. Engl. J. Med., 352, 987–96. van den Bent, M., Chinot, O. L. and Cairncross, J. G. (2003). Recent developments in the molecular characterization and treatment of oligodendroglial tumors. Neuro. Oncol., 5, 128–38. van den Bent, M. J., Afra, D., de Witte, O. et al. (2005). Longterm efficacy of early versus delayed radiotherapy for low-grade astrocytoma and oligodendroglioma in adults: the EORTC 22845 randomised trial. Lancet, 366, 985–90.
381
33
SKIN CANCER OTHER THAN MELANOMA Sankha Suvra Mitra
Introduction
Benign tumours
Skin cancer is a common cancer, and it is often caused by ultraviolet radiation from chronic sun exposure in a fairskinned population. Basal cell carcinoma is the most frequent variety, and presents as a superficial tumour that very rarely metastasises. Squamous cell carcinoma can spread to regional lymph nodes. For the management of skin cancer, patients are usually discussed in a multidisciplinary team meeting. Surgical excision is the treatment of choice for the majority of tumours. Non-surgical treatment options include radiotherapy, curettage, imiquimod cream, photodynamic therapy, and topical 5-FU therapy. Radiotherapy is an important treatment option for older patients – it is preferred to surgery for large superficial tumours, multiple tumours, and in areas around the eye, nose, and ear where the cosmetic results may be better. The radiotherapy technique involves treatment with either superficial X-rays or electron therapy. Cure rates are about 95% at 5 years. The National Institute for Health and Clinical Excellence has recently published service guidance on the care of these patients (NICE, 2006).
Benign tumours include the following: r Benign naevus. r Sebaceous cyst. r Epidermal cyst. r Basal cell papilloma (seborrhoeic keratosis). r Vascular angioma. r Keratoacanthoma. r Dermatofibroma.
Range of tumours In the differential diagnosis of a skin tumour, the tumour could be benign or malignant, and a malignant tumour could be primary or secondary.
Premaligant conditions Premalignant conditions include the following (Soutar and Robertson, 2001): r Actinic keratosis. r Bowen’s disease. r Erythroplasia of Queyrat. r Paget’s disease.
382
Malignant tumours Primary malignant tumours include: r Basal cell carcinoma. r Squamous cell carcinoma. r Merkel cell carcinoma. r Malignant eccrine porocarcinoma. r Amelanotic melanoma. r Cutaneous T-cell lymphoma. r Primary cutaneous B-cell lymphoma. r Kaposi’s sarcoma. r Angiosarcoma. r Lymphangiosarcoma. r Atypical fibroxanthoma. Secondary malignant tumours arise from breast cancer, lung cancer, colon cancer, melanoma, and so forth.
Incidence and epidemiology The number of cases of skin cancer is highest for fairskinned males in Australia, where there is an annual incidence rate of 1372 males and 702 females per 100 000 population. More than 75% of patients are over 40 years of age at presentation, and more than 5% of the population over age 60 will develop a non-melanoma skin cancer in the UK (American Cancer Society, 2003).
Skin cancer other than melanoma
Risk factors and aetiology
Chronic scar
Risk factors for skin cancer include the following (Soutar and Robertson, 2001): r Chronic sun exposure and ultraviolet radiation. Premalignant conditions include actinic keratosis and Bowen’s disease. r Phototherapy with ultraviolet A/B radiation. r Immunosuppression in transplant patients (e.g. CML, CLL, and AIDS). r Ionising radiation from radiotherapy or atomic fallout. Cancers include squamous cell cancer, angiosarcoma, and lymphangiosarcoma. r Chemical carcinogenesis from nitrates, arsenicals, oral methoxsalen, soot and oils, and coal tar.
The cancer from a chronic scar is called Marjolin’s ulcer, and it may occasionally complicate chronic leg ulcers.
Genetic predisposition Naevoid basal cell carcinoma syndrome (Gorlin’s syndrome) is an autosomal familial cancer syndrome in which multiple basal cell carcinomas appear at an early age. The gene responsible (PTCH, patched gene) is on chromosome 9q22–31. Other features of the syndrome include bony cysts of the mandible that are visible on orthopantomograms, abnormalities of the ribs, short fourth metacarpal, coloboma at birth or cataracts in later life, and an increased risk of both medulloblastoma and meningioma. A skull X-ray shows calcification of the falx cerebri in 90% of cases by the age of 12 years. Bazex’s syndrome is a rare cross-linked syndrome that predisposes individuals to multiple basal cell carcinomas. This disease is linked to chromosome Xq24–27. Xeroderma pigmentosa is a rare autosomal recessive genetic disease associated with non-melanoma skin cancer due to defective DNA repair (nucleotide-excision repair). Ferguson-Smith disease is a familial disease presenting as self-healing squamous cell carcinoma. Muir-Torre syndrome is an autosomal dominant disease with sebaceous or adnexal gland tumours usually associated with colonic tumours (Soutar and Robertson, 2001).
Chronic inflammation The premalignant condition of chronic inflammation is erythroplasia of Queyrat, which can develop into squamous cell cancer of the glans penis. This is similar to Bowen’s disease.
Basal cell carcinoma and squamous carcinoma Basal cell carcinoma Skin cancer is the most common cancer, and basal cell carcinoma accounts for 80% of all non-melanoma skin cancers. The major cause is exposure to ultraviolet radiation. Recreational exposure to the sun during childhood and adolescence poses a significant risk. Fair complexion, red or blonde hair, and light eye colour are also independent risk factors (Rubin et al., 2005).
Clinical presentation Basal cell carcinoma characteristically arises in body areas exposed to the sun. The most common site is the head and neck, followed by the trunk. Basal cell carcinomas tend to grow very slowly, over years, and cause itching, bleeding, and discomfort in later years. They very rarely metastasise, unlike other cancers, but if neglected over several years, they tend to infiltrate locally; they are also called rodent ulcers. The classic form is the nodular basal cell cancer, which presents as a pearly papule, nodule, or cyst with overlying telangiectasia and a rolled border, with central crusting or ulceration. Occasionally, nodular basal cell cancers may resemble enlarged pores or pits on the sebaceous skin of the central portion of the face. Superficial spreading basal cell carcinoma presents as a scaly erythematous patch or plaque. Both of these forms can be brownish in colour due to melanin. The morphoeic form appears as an indurated, whitish, scar-like plaque with indistinct margins. Other variants include micronodular, mixed, basosquamous, adamantinoid, and clear cell.
Molecular pathogenesis Inappropriate activation of the hedgehog (HH) signalling pathway results in secreted sonic HH protein binding to a tumour-suppressor gene and preventing its suppression of transmembrane proteins and downstream targets. The GLI family of transcription factors are activated. Mutations in TP53 are found in about 50% of cases (Rubin et al., 2005). 383
Sankha Suvra Mitra
Squamous carcinoma Unlike basal cell carcinoma, there may be a clear progression through dysplasia, carcinoma in situ, or Bowen’s disease to frankly invasive squamous cell carcinoma. Solar damage causes red scaly patches that persist for several years and are termed actinic keratosis. Actinic keratoses are premalignant; however, only 1% are estimated to transform into a squamous cell carcinoma and up to 25% resolve spontaneously. The patches usually overlie a superficial ulcer; any induration or infiltration suggests that the tumour has progressed to frank malignancy. Another typical picture is a rapidly growing ulcer with a rolled edge and an indurated margin. Sometimes there can be exophytic growth resulting in a cauliflower-like mass (Soutar and Robertson, 2001). Secondary spread occurs to the regional lymph nodes. The ear, lip, and columella are sites that have a higher incidence of metastases (approximately 5 to 7%). Keratoacanthoma is histologically similar to squamous cell carcinoma but the history is of a tumour that grows rapidly over 4 to 6 weeks that subsequently undergoes spontaneous regression, leaving a small pitted scar on the surface.
Table 33.1. The TNM staging classification for carcinomas of the skin Stage
Description
T1
Tumour ≤2 cm
T2
Tumour >2 but ≤5 cm
T3
Tumour >5 cm
T4
Tumour invades cartilage, skeletal muscle, or bone
N0
No regional lymph node metastasis
N1
Regional lymph node metastasis
M0
No distant metastasis
M1
Distant metastasis
Adapted from UICC (2002).
If the tumour is present for years and slowly growing, then it is likely to be a basal cell cancer; if it has grown over months, then it is likely to be a squamous cell cancer; and if it has been growing rapidly over a few weeks, then it is likely to be a keratoacanthoma.
Clinical examination Poor prognostic factors Risk factors for extensive subclinical spread include a tumour size greater than 2 cm, tumour location on the central part of the face and ears, long-standing duration, incomplete excision, an aggressive histological pattern of growth, and perineural or perivascular involvement. Patients with known immunosuppression also tend to develop more aggressive tumours (Rubin et al., 2005).
Staging classification for skin carcinomas The TNM staging classification for skin carcinomas is shown in Table 33.1.
The tumour is examined with a magnifying glass under bright light. The margins of the tumour are established by careful palpation. The depth of the tumour is assessed to determine the type of radiation and energy to be used. Sites where it may be difficult to assess the depth include the embryonic folds like the inner canthus, nasolabial fold, ala nasi, tragus, and postauricular areas because palpation may underestimate the depth of the tumour. Regional lymph node areas are examined, particularly in patients with squamous cell carcinomas. Large ulcerative tumours of the scalp may need assessment of bony involvement by palpation.
Investigation Examination of the patient History Age, ethnic background, occupation, and geographic factors are important to consider (Diepgen and Mahler, 2002). Symptoms from the tumour, its duration, and the rate of growth are elicited. A history of sun exposure is usually found. The patient should be asked about topical treatment used in the past, history of chemical exposure, or previous irradiation to the involved area. 384
Tumour diagnosis is confirmed by scrapings, punch biopsy, or excisional biopsy. Large tumours may require assessment of deeper involvement (e.g. with X-ray and CT scans) if they appear fixed to underlying tissues.
Treatment overview Management is discussed in the MDM. The particular benefits of each treatment option are discussed next.
Skin cancer other than melanoma
Surgery Mohs’ micrographic surgery Mohs’ micrographic surgery is a complicated method that requires special training. It is indicated for sites that have a high initial treatment failure rate when traditional methods are used (e.g. periorbital area, nasolabial fold, etc.) and for recurrent tumours. It is a technique that allows rapid examination of horizontal frozen-section specimens, which are processed to include 100% of the peripheral and deep surgical margins. It is used to determine whether further excision is required. Cure rates at 5 years approach 99%. (Malhotra et al., 2004; Thomas and Amonette, 1998).
Surgical excision Surgery is indicated for small tumours that can be excised and then the defect closed directly because this one-off procedure results in a good cosmetic result and a high cure rate. A major advantage of surgery over radiation is that histological confirmation of removal is obtained, thereby avoiding long-term follow-up (Avril et al., 1997). Patients younger than 50 to 60 years are preferentially treated by surgery because the cosmetic results after 10 years of deterioration of the radiation scar are inferior to surgery. Recurrent tumours and tumours with uncertain margins are also treated by surgery. Tumours involving cartilage, tendon, bone, or joint and large, bulky tumours are also treated by surgery. Keratoacanthoma is usually treated by surgical excision or curettage/ cautery.
Electrodesiccation and curettage and cryosurgery Electrodesiccation and curettage and cryosurgery are indicated for small well-defined primary tumours with non-agressive features on the neck, trunk, arms, and legs. Close follow-up is required. Cure rates are 95% at 5 years.
Non-surgical treatments
Photodynamic therapy Both systemic and topical porphyrins can be used but photodynamic therapy is of value in treating superficial rather than thick tumours. Methyl aminolaevulinic acid is a photoactivated agent applied topically under occlusive foil to enhance tissue penetration and reduce bleaching and other systemic side effects. Deeper penetration is poor. It is indicated for small, superficial basal cell carcinomas at cosmetic sites such as the central face and the ‘V’ of the neck (National Cancer Institute, 2006, Health Professional Version, http://cancernet.nci.nih .gov./cancertopics/pdq/treatment/skin/).
Topical 5-fluorouracil therapy Topical 5-fluorouracil is available in strengths of 2 or 5% and is usually applied twice daily to the tumour and its margins (Dabski and Helm, 1988).
Radiotherapy Radiotherapy is an important treatment option for older patients who are less concerned about long-term cosmesis. It is preferred to surgery for large superficial or extensive tumours and for multiple tumours where the cosmetic results are better. Tumours that are in sites that make removal difficult or that are unresectable should also be treated by radiotherapy. Patients who refuse surgery, who are unfit, or who are on anticoagulant therapy should be offered radiotherapy. Cure rates are about 95% at 5 years (Caccialanza et al., 2003; Lovett et al., 1990). Radiotherapy is contraindicated for patients with xeroderma pigmentosum and the basal cell naevus syndrome because it may induce more tumours in the treated area. Sites of previous radiotherapy, areas of vascular insufficiency, skin overlying the shin and malleoli of the lower leg, and the middle third of the upper eyelid tolerate radiotherapy poorly. Squamous cell carcinomas of the lower lip can be treated by surgery, external beam radiation with electrons or X-rays, or with interstitial radiotherapy.
Imiquimod 5%
Radical radiotherapy
Imiquimod 5% is a topical immune modulator licensed for biopsy-proven, small, primary, superficial basal cell cancers on the trunk, neck, arms, or legs of adults with normal immune systems. A course of treatment involves once-daily administration of imiquimod 5 days per week for 6 weeks. Cure rates are approximately 85% (Lebwohl et al., 2004; Tran et al., 2003).
Informed consent is obtained after discussion with the patient. Acute effects include dry, red, inflamed skin like a mild sunburn, moist skin breakdown, scabbing and ulceration. Intermediate effects include slow healing over 4 to 8 weeks, tiredness, and watery eyes (epiphora due to nasolacrimal duct oedema and stenosis) if the inner canthus area is being treated. Late effects include 385
Sankha Suvra Mitra
pale thin skin with red dots (telangiectasis), fibrosis, permanent hair loss in the scalp, eyebrow and eyelash tumours, dry eyes due to lacrimal gland damage, and upper eyelid conjunctival keratinisation. Assessment of disease is by clinical examination with imaging if relevant.
Mould room preparation A customised lead cut-out is made, 1.5 mm thick for 90 to 150 superficial X-ray therapy (SXR) and 4 mm thick for electron therapy. Depending on the electron energy to be used and the field size, a certain thickness of wax bolus (in mm) is added over the cut-out area to increase the surface dose to the 90% isodose, at which the dose is prescribed.
Definition of target volume A margin is drawn around the gross tumour to encompass subclinical extension and penumbra of the beam. For a cystic, superficial, basal cell cancer with well-defined margins, a 0.5 cm margin is adequate for SXR. For large, morphoeic, or poorly defined basal cell cancers and squamous cell cancers, a 1 cm margin is required for SXR. For electron therapy an additional lateral 0.5 cm margin is required to account for bowing of the isodoses. Large infiltrative tumours with an unknown deep extension need to be CT planned (Solan et al., 1997).
Shielding A lead shield must be used to protect the eye when the eyelids and inner or outer canthus are treated. This shielding protects the conjunctiva, cornea, and lens. An internal eye shield, shaped like a large contact lens with a screw thread on the anterior surface, is used when the target volume involves both eyelids. An external spade-shaped eye shield is placed under the upper or lower eyelid when only one eyelid is included in the target volume. Internal eye shielding can only be done if the patient is not driving themselves to and from the appointment. The patient is warned that the local anaesthetic drops sting and a dose of one to two drops of 0.5% amethocaine eye drops is instilled into the eye. The eye shield is aseptically withdrawn from the sterile solution and is rinsed in sterile normal saline. The inner surface is coated with liquid paraffin to reduce friction with the cornea and the eyelid is pulled in one direction to introduce the eye shield. Both eyelids are usually taped together over the internal eye shield, whereas the external eye shield is usually taped over the other eyelid. The patient has to wear an eye pad for 2 hours afterwards, until the effects of the local anaesthetic have worn off and the corneal reflex is back to normal. 386
Intranasal shielding An intranasal lead shield is used to protect the nasal mucosa and cartilage of the nasal septum while treating tumours of the ala nasi. This shield is wrapped in wet gauze before it is inserted into the nostril. For electron treatments, any shielding that is placed internally (e.g. gum or nostril shielding) or, for example, behind the ear requires aluminium or wax coating to reduce the backscatter caused by secondary electrons produced in the lead.
Intraoral shielding An intraoral lead gum shield is used to protect the gums and mucosa when squamous cell cancers of the lower lip are treated with electron therapy. The lead is usually anteriorly coated with aluminium (surface facing the beam) to reduce the backscatter from secondary electrons released in the lead (Solan et al., 1997).
Choice of radiation: tumours smaller than 4 cm Tumours that are less than 4 cm in diameter and less than 5 mm thick are appropriate for superficial X-ray therapy. Tumours around the face, particularly in small concave areas like the inner canthus and ala nasi, obtain excellent cosmetic results from superficial X-ray therapy. Different superficial machines have different depth doses depending on the use of different filters in the head of the beam. Usually 90 to 150 kV is adequate to treat to a depth of a few millimetres. Tumours on both sides of the pinna have a thickness of about 6 mm and 160 kV is adequate to treat this through a single field. Radiotherapy should be withheld if the cartilage is directly involved. Reports of cartilage damage by superficial X-ray therapy are greatly exaggerated. Percentage depth dose tables for 95 and 195 kV photons are shown in Tables 33.2 and 33.3, respectively.
Choice of radiation: tumours larger than 4 cm Tumours that are more than 4 cm in diameter and more than 1 cm thick should be treated by electron therapy. With smaller tumours beam flatness is lost. Tumours on the scalp are treated by electrons to reduce the exit dose to the brain. Flat tumours on the trunks and limbs are suitable for electron therapy but lower-leg tumours are usually better treated with surgery. It is reasonable to avoid electron therapy near the eyes; wide margins of 1.5 cm are required because of bowing of the isodoses, and lateral scatter of radiation makes eye shielding difficult. However, if there is a 3-cm-thick tumour between the eyes, then there is no alternative. Electron therapy should also be avoided near complicated air spaces like
11.7
8.7
6.4
4.8
3.6
6
7
8
9
10
4.0
5.3
7.1
9.6
12.8
17.2
24.3
33.2
47.0
68.2
82.5
100.0
1.5
1.5 circle
4.4
5.8
7.8
10.4
13.8
18.5
26.2
35.4
49.7
70.7
84.1
100.0
2
2 circle
4.9
6.3
8.4
11.2
14.8
19.8
27.6
37.4
51.9
72.8
85.3
100.0
2.5
2.5 circle
5.3
6.8
9.0
11.9
15.8
21.1
28.9
39.4
54.1
74.8
86.5
100.0
3
3 circle
5.7
7.3
9.6
12.6
16.6
22.1
30.0
40.9
55.6
75.7
87.3
100.0
3.5
3.5 circle
6.1
7.8
10.2
13.2
17.4
23.1
31.1
42.3
57.1
76.6
88.1
100.0
4
4 circle
6.9
8.8
11.2
14.5
18.8
24.8
33.0
44.6
59.5
78.4
89.2
100.0
5
5 circle
9.2
11.4
14.2
18.0
23.0
29.7
38.4
50.0
64.8
81.8
91.2
100.0
6
6 circle
9.9
12.3
15.2
19.2
24.2
31.0
39.9
51.4
66.0
82.7
91.7
100.0
7
7 circle
10.6
13.2
16.1
20.3
25.3
32.3
41.3
52.7
67.1
83.5
92.1
100.0
8
8 circle
11.8
14.6
17.8
22.2
27.3
34.3
43.5
54.7
68.8
84.5
92.6
100.0
10
10 circle
13.4
16.6
20.4
24.9
30.0
37.0
46.1
57.1
71.1
86.2
93.5
100.0
14
14 circle
14.3
17.7
21.7
26.2
31.2
38.3
47.3
58.2
72.3
87.1
94.0
100.0
17
17 circle
10.8
13.5
16.4
20.7
25.7
32.7
41.7
53.1
67.4
83.7
92.2
100.0
8.4
5 × 15
cancer centre. It should not be used to guide treatment for any patient.
Physicist, Sussex Cancer Centre, and David Bloomfield, Head, Sussex Cancer Centre. Disclaimer: this table is for academic purposes only and applies to a specific machine and
HVL = half value layer = 3.0 Al; diameter ≤ 5 cm, SSD = source-skin distance = 20 cm; other applicators, SSD = 30 cm. Permission to print has been obtained from David Prior,
22.3
15.9
4
5
44.3
31.0
65.7
1
2
80.8
3
100.0
0.5
1
–
0
Depth (cm)
diameter (cm)
Equivalent
size (cm)
Applicator
Table 33.2. Percentage depth doses for 95 kV photons
Sankha Suvra Mitra
Table 33.3. Percentage depth doses for 195 kV photons Applicator size (cm)
–
5 circle
–
6 circle
–
7 circle
6×8
7×7
8 circle
4
4.5
5
5.4
6
6.3
6.9
7
7.2
Side of Eq. square (cm) Depth (cm) 0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
0.5
95.6
96.1
96.6
96.8
97.2
97.3
97.6
97.6
97.6
1
90.3
91.0
91.7
92.1
92.7
92.9
93.4
93.5
93.6
2
78.5
79.5
80.5
81.2
82.2
82.6
83.4
83.5
83.8
3
66.5
68.0
69.5
70.4
71.8
72.4
73.5
73.7
74.0
4
55.7
57.3
58.8
59.8
61.3
62.0
63.3
63.5
63.9
5
46.6
48.1
49.5
50.5
52.0
52.7
54.0
54.2
54.6
6
38.8
40.1
41.4
42.4
43.8
44.5
46.0
46.2
46.6
7
32.1
33.4
34.6
35.6
37.1
37.8
39.3
39.5
39.9
8
26.3
27.6
28.8
29.7
31.1
31.7
33.0
33.2
33.6
9
22.0
23.0
24.0
24.8
26.1
26.6
27.6
27.8
28.2
10
18.2
19.2
21.0
20.9
22.1
22.6
23.6
23.8
24.2
11
15.1
16.0
16.8
17.5
18.6
19.1
20.1
20.3
20.6
12
12.5
13.4
14.2
14.8
15.7
16.2
17.1
17.3
17.6
13
10.2
11.0
11.8
12.4
13.2
13.6
14.4
14.5
14.7
14
8.5
9.2
9.9
10.4
11.2
11.6
12.3
12.4
12.6
15
7.2
7.7
8.2
8.6
9.3
9.6
10.1
10.2
10.4
16
6.0
6.5
6.9
7.3
7.9
8.1
8.6
8.7
8.9
17
4.9
5.4
5.8
6.1
6.6
6.8
7.3
7.4
7.6
18
4.1
4.5
4.8
5.1
5.6
5.8
6.1
6.2
6.4
19
3.4
3.7
4.0
4.3
4.7
4.9
5.2
5.3
5.4
20
2.8
3.1
3.4
3.6
4.0
4.2
4.5
4.5
4.6
HVL = half value layer = 1.0 mm Cu; SSD = source-skin distance = 50 cm; closed applicators; Eq. square = equivalent square. Permission to print has been obtained from David Prior, Physicist, Sussex Cancer Centre and David Bloomfield, Head, Sussex Cancer Centre. Disclaimer: this table is for academic purposes only and applies to a specific machine and cancer centre. It should not be used to guide treatment for any patient.
the nasal cavity, sinuses, and mastoid air cells because of uncertainty of dose distribution due to tissue inhomogeneity.
Determining the electron energy Once the thickness and diameter of the tumour and diameter of the field are established, an equivalent square calculation is made for the field size. In conjunction with the physicist, the electron depth dose chart is consulted for that field size. If the tumour is 2 cm thick, then a margin of 0.5 cm is usually added posteriorly and a 0.5 to 1 cm wax/Perspex® bolus is required anteriorly to increase the skin dose to at least 90%. Multiplying this total thickness by 3 usually gives the approximate electron energy (9 MeV) that will 388
treat the PTV to a 90% isodose. The electron depth dose chart lists the appropriate skin dose for that energy and field size. Wax or Perspex® is used as bolus material to increase the skin dose to 90% because electron therapy is usually prescribed to the 90% isodose. Electron depth dose tables for 95 and 100 cm SSD are shown in Tables 33.4 and 33.5, respectively. Electron depth dose curves are shown in Figure 33.1.
Plan calculation The SSD is increased by the ‘stand off’ between the applicator and the tumour whereas the SSD is decreased by ‘stand in.’ In both cases, corrections must be made for the change in output with variation in SSD according to the inverse-square law.
63
72
62
72
30
10 96%
73
64
59
48
43
20
0
38 cm2
96%
73
64
59
48
43
23
0
50 cm2
c 8
84%
31
27
25
20
18
13
5
cm2
100
10
0
94%
73
65
59
49
44
25
92%
73
65
59
49
44
27
0
cm2
196
100 cm2
14
92%
40
35
31
25
22
15
0
cm2
12
c 4
10
94%
40
34
29
23
20
12
0
7 cm2
c 3
100%
86
74
64
48
39
4
0
20 cm2
c 5
91%
40
35
32
26
24
16
0
cm2
20
c 5
8 MeV
99%
87
76
67
52
43
7
0
28 cm2
c 6
90%
40
35
32
26
24
17
0
cm2
28
c 6
97%
48
41
35
26
22
11
0
7 cm2
c 3
18 MeV
98%
88
78
71
58
50
16
0
38 cm2
c 7
88%
41
36
33
27
24
17
4
cm2
100
10
98%
88
78
71
59
52
22
0
50 cm2
c 8
95%
49
43
38
30
27
17
0
cm2
12
c 4
96%
88
78
71
60
54
26
0
cm2
100
10
94%
50
44
39
32
29
19
0
cm2
20
c 5
10 MeV
94%
88
78
71
60
54
30
0
cm2
196
14
93%
50
44
39
33
29
21
0
cm2
28
c 6
100%
95
81
70
50
39
5
0
20 cm2
c 5
88%
50
44
40
33
30
22
3
cm2
100
10
97%
55
49
43
33
29
16
0
cm2
12
c 4
99%
96
83
73
55
43
9
0
28 cm2
c 6
98%
54
45
38
29
25
11
0
7 cm2
c 3
98%
97
86
77
62
52
13
0
38 cm2
c 7
20 MeV
97%
55
50
45
36
32
18
0
cm2
20
c 5
98%
97
86
78
63
54
17
0
50 cm2
c 8
95%
55
50
45
37
33
21
0
cm2
28
c 6
12 MeV
97%
97
87
79
65
57
22
0
cm2
100
10
91%
56
50
46
39
35
25
0
cm2
100
10
Disclaimer: this table is for academic purposes only and applies to a specific machine and cancer centre. It should not be used to guide treatment for any patient.
c = cm circle; = cm square. Permission to print has been obtained from David Prior, Physicist, Sussex Cancer Centre, and David Bloomfield, Head, Sussex Cancer Centre.
98%
57
55
50
99%
45
42
80
Surface dose
38
36
90
0
14
0
9
90
28 cm2
100
% Dose
86%
31
27
24
20
18
13
4
cm2
28
c 6
15 MeV
c 7
86%
31
20 cm2
31
31
Surface dose
27
24
c 6
27
26
30
c 5
24
23
50
20
18
87%
20
19
80
13
4
cm2
89%
18
16
90
10
4
13
2
11
cm2
20
12
90
7 cm2
c 5
c 4
100
% Dose
c 3
6 MeV
Table 33.4. Electron depth dose table for 95 cm source-surface distance (SSD), showing the depth in millimetres of percentage isodoses for different-sized applicators
94%
97
87
79
66
59
30
0
cm2
196
14
88%
56
50
46
39
35
25
3
cm2
196
14
57
64
72
56
63
72
50
30
10
6 6
82%
31
27
25
20
18
13
15 MeV
92%
72
64
58
48
43
23
0
38 cm2
c 7
82%
31
27
25
20
18
13
cm2
28
c 6
92%
72
64
58
48
43
23
0
50 cm2
c 8
82%
31
27
25
20
18
13
6
cm2
100
10
0
92%
72
64
59
49
44
25
92%
72
64
59
49
44
27
0
cm2
196
100 cm2
14
88%
40
35
32
26
23
16
3
cm2
12
c 4
10
91%
40
35
30
24
20
12
0
7 cm2
c 3
94%
86
75
67
51
44
18
0
20 cm2
c 5
85%
40
36
32
27
24
17
5
cm2
20
c 5
8 MeV
94%
87
77
69
55
47
23
0
28 cm2
c 6
85%
40
36
32
27
24
17
5
cm2
28
c 6
92%
49
42
36
27
24
15
0
7 cm2
c 3
18 MeV
94%
87
78
71
58
51
25
0
38 cm2
c 7
85%
41
36
32
27
24
17
5
cm2
100
10
94%
87
78
71
59
52
25
0
50 cm2
c 8
90%
49
43
39
31
27
18
0
cm2
12
c 4
94%
87
78
71
60
54
25
0
cm2
100
10
87%
50
44
39
33
29
20
3
cm2
20
c 5
10 MeV
94%
87
78
71
60
54
25
0
cm2
196
14
87%
50
44
40
33
30
21
4
cm2
28
c 6
95%
95
82
72
55
46
16
0
20 cm2
c 5
86%
50
44
40
33
30
21
5
cm2
100
10
92%
56
50
44
35
30
18
0
cm2
12
c 4
95%
96
84
75
59
49
17
0
28 cm2
c 6
93%
56
47
41
30
26
16
0
7 cm2
c 3
95%
96
86
78
63
54
22
0
38 cm2
c 7
20 MeV
90%
56
50
46
37
33
22
0
cm2
20
c 5
95%
96
86
78
64
55
22
0
50 cm2
c 8
89%
56
50
46
38
34
23
2
cm2
28
c 6
12 MeV
95%
97
87
79
65
57
22
0
cm2
100
10
89%
56
50
46
39
35
24
2
cm2
100
10
Disclaimer: this table is for academic purposes only and applies to a specific machine and cancer centre. It should not be used to guide treatment for any patient.
c = cm circle; = cm square. Permission to print has been obtained from David Prior, Physicist, Sussex Cancer Centre and David Bloomfield, Head, Sussex Cancer Centre.
93%
46
44
80
93%
40
38
90
Surface dose
0
22
0
19
90
28 cm2
100
% Dose
20 cm2
31
31
10
c 6
27
27
30
c 5
25
24
50
83%
20
19
80
87%
18
17
90
Surface dose
5
13
3
11
cm2
20
cm2
12
90
7 cm2
c 5
c 4
100
% Dose
c 3
6 MeV
Table 33.5. Electron depth dose table for 100 cm SSD, showing the depth in millimetres of percentage isodoses for different-sized applicators
94%
97
87
79
66
59
28
0
cm2
196
14
88%
56
50
46
39
35
25
3
cm2
196
14
Skin cancer other than melanoma
95 cm SSD 5cm Circle
100 90
6 MeV 8 MeV 10 MeV 12 MeV 15 MeV 18 MeV 20 MeV
80
Dose (%)
70 60 50 40 30 20 10 0 0
10
20
30
40
50
60
70
80
90
100
110
120
Depth (mm) 10 x 10 cm
100 90
6 MeV 8 MeV 10 MeV 12 MeV 15 MeV 18 MeV 20 MeV
80
Dose (%)
70 60 50 40 30 20 10 0 0
10
20
30
40
50
60
70
80
90
100
110
120
Depth (mm) Figure 33.1. Electron depth dose curves for a 5 cm circle and 10 × 10 cm square at 95 cm SSD. Permission to print has been obtained from David Prior, Physicist, Sussex Cancer Centre, and David Bloomfield, Head, Sussex Cancer Centre. Disclaimer: these graphs are for academic purposes only and apply to a specific machine and cancer centre. They should not be used to guide treatment for any patient. SSD = source-skin distance.
Dose Some radiotherapy dose schedules are as follow:
r For small basal cell cancers (<3 cm), give 35 Gy in five fractions over 5 days (National Comprehensive Cancer Network, 2005). (In very frail elderly patients, a single fraction of, e.g. 20 Gy may be used.) r For large basal cell cancers and squamous cell cancers, give 45 Gy in 10 fractions over 2 weeks. r For very large, thick squamous cell cancers, give 60 to 64 Gy in 30 to 32 fractions over 6 to 6.5 weeks.
r For interstitial radiotherapy, give 65 Gy to the 85% reference isodose according to the Paris system (Dobbs et al., 1999).
Skin reaction clinic The patient should be advised to wash the area with plain running water in a shower or bath every day and to pat it dry gently afterwards. Shaving and use of artificial deodorants and soaps should be avoided because they could sensitise the skin to
391
Sankha Suvra Mitra
radiation (heavy metals increase the photoelectric effect) and cause a radiation burn. Aqueous cream or E45® cream is applied and sun exposure is avoided by using sun barrier cream and hats.
Follow-up After the radiotherapy reaction has healed, patients with basal cell carcinomas are usually discharged back to the GP. Patients with squamous cell carcinomas should be followed up for 5 years.
Merkel cell carcinoma Merkel cell carcinoma is an aggressive cancer that develops satellite tumours and lymph node metastases, mainly in elderly patients. The cells are thought to be of neuroendocrine origin from the APUD (amine precursor uptake and decarboxylation) system. It can present as a painless, red, indurated nodule or an ulcer, usually in the region of the head and neck. Surgery with wide local excision is the main treatment, but local recurrence rates are high at 40% and lymph node metastases can occur in about 40 to 65% of cases. These tumours are radioresponsive. Therefore, wide local excision followed by postoperative radiotherapy is evolving as a combined-modality treatment with higher locoregional control rates. The use of chemotherapy is under investigation (Soutar and Robertson, 2001). Patients with metastatic disease often respond to chemotherapy with carboplatin and etoposide, but the duration of response is short.
Malignant porocarcinoma and eccrine carcinoma The adnexal structures include the eccrine sweat glands, hair follicles, sebaceous and apocrine glands, and the APUD system. The majority of tumours are benign. Eccrine carcinoma is histologically like an anaplastic squamous cell carcinoma and infiltrates locally and metastasises. Some tumours can resemble salivary gland tumours and show features of adenoid cystic carcinoma. Malignant sebaceous gland tumours are rare but most frequently found in the eyelids. Surgical excision with a wide margin is the preferred treatment (Soutar and Robertson, 2001).
Kaposi’s sarcoma There are four different types of Kaposi’s sarcoma (Safai, 1997): 392
r HIV-related Kaposi’s sarcoma is the most common malignancy in people with AIDS. Treatment is almost always palliative. Kaposi’s sarcoma can occur in the absence of HIV positivity. r Classical Kaposi’s sarcoma occurs as an indolent condition most commonly affecting the lower limbs in elderly Ashkenazi Jews. It also occurs in patients from Mediterranean countries. There may be a link to immunosuppression because patients also have an increased incidence of non-Hodgkin lymphoma. r Endemic Kaposi’s sarcoma is mainly found in subSaharan Africa, where it is more aggressive, occurs in children of both sexes, and causes fulminant lymphadenopathy. Some men present with a benign nodular form of the disease. r Kaposi’s sarcoma also occurs in immunosuppressed patients with a 400-fold increase in renal transplant recipients who are on long-term immunosuppressive therapy. The clinical presentation is with small, painful, reddish or purple papules, nodules, or plaques on the face, hard palate, gums, shins, lower legs, and soles of the feet. The oral tumours can cause ulceration, haemorrhage, and dental instability. Sometimes there can be marked oedema in the limbs, scrotum, or face. Pulmonary and gastrointestinal involvement is common and all internal organs can be affected except the brain. A symptomatic patient with Kaposi’s sarcoma referred for palliative radiotherapy should be immediately referred to colleagues who specialise in HIV for HIV counselling and testing. If the patient is found to be HIV positive, he or she is usually treated with highly active antiretroviral therapy (HAART). The patient should be reassessed in the radiotherapy department after 3 months of HAART, by which time some of the tumours may have resolved. If a tumour has not resolved by then, the patient should be offered palliative radiotherapy. Nodular localised disease in the limbs can be treated with a single 8 Gy fraction with SXR or electron therapy depending on the site, size, and thickness of the tumour. This procedure can be repeated if necessary and multiple tumours can also be treated. Mucosal tumours in the mouth, palate, and conjunctiva are treated with 20 Gy in 10 fractions over 2 weeks to avoid acute radiotherapy reactions, which are more severe in patients with HIV-associated Kaposi’s sarcoma. Classical non-HIV-related disease is usually treated with 16 Gy in four fractions given over 8 days.
Skin cancer other than melanoma
Palliative chemotherapy is used for more extensive disease. A combination of vincristine 1.4 mg/m2 i.v. plus bleomycin 15 mg/m2 i.v. infusion every 3 weeks or liposomal doxorubicin (Safai, 1997) can be used.
Skin lymphomas Mycosis fungoides Mycosis fungoides is discussed in Chapter 31 (see p. 358).
Primary cutaneous B-cell lymphoma SXR or electrons may be used as appropriate. The PTV is typically the GTV plus a 2 to 3 cm margin. The prescribed dose is 15 Gy in 5 fractions over 1 week or 20 Gy in 5 fractions.
Malignant fibrous histiocytoma Malignant fibrous histiocytoma is the most common soft tissue sarcoma in adults. It usually presents as a nodule but the histological appearance is of a high-grade sarcoma. The prognosis is related to the size of the primary tumour, and it is treated via wide local excision (Soutar and Robertson, 2001).
Dermatofibrosarcoma protuberans Dermatofibrosarcoma protuberans is a slowly growing multinodular skin tumour that typically affects the trunk in young black males. Local recurrence is a major problem, whereas metastases are very rare. Wide local surgical excision is the treatment of choice using Mohs’ micrographic surgery (Soutar and Robertson, 2001).
Angiosarcoma and lymphosarcoma Angiosarcomas without lymphoedema almost always occur on the face and scalp and in adults over 50 years of age. Nodules or ulcerated areas are present within blue-red discoloured skin. Local recurrence and distant metastases to the lungs are both common. Postoperative radiotherapy is usually offered after surgery but the prognosis is very poor. Angiosarcoma with lymphoedema, or lymphangiosarcoma, may arise in the chest and breast following radiotherapy or in the upper limb following a history of chronic lymphoedema.
Lymphangiosarcoma can occur in postmastectomy lymphoedema and it is known as Stewart-Treves syndrome. Rapid lymphatic and haematogenous metastases occur, with a poor prognosis. Radical surgery with amputation is often undertaken in conjunction with postoperative radiation. Isolated limb perfusion is also used (Soutar and Robertson, 2001).
Secondary metastases Secondary metastases can cause pain, bleeding, or fungation. A single fraction of 8 Gy or 20 Gy in 5 fractions over 5 days with superficial X-ray therapy or electron therapy usually provides prompt symptomatic relief. Single fractions can frequently be repeated after several months.
REFERENCES American Cancer Society. (2003). Cancer Facts and Figures. Atlanta: American Cancer Society. Avril, M. F., Auperin, A., Margulis, A. et al. (1997). Basal cell carcinoma of the face: surgery or radiotherapy? Results of a randomized study. Br. J. Cancer, 76, 100–6. Caccialanza, M., Piccinno, R., Moretti, D. et al. (2003). Radiotherapy of carcinomas of the skin overlying the cartilage of the nose: results in 405 lesions. Eur. J. Dermatol., 13, 462–5. Dabski, K. and Helm, F. (1988). Topical chemotherapy. In Skin Cancer: Recognition and Management, ed. R. A. Schwartz. New York: Springer-Verlag, pp. 378–89. Diepgen, T. L. and Mahler, V. (2002). The epidemiology of skin cancer. Br. J. Dermatol., 146 (Suppl. 61), 1–6. Dobbs, J., Barrett, A. and Ash, D. (1999). Practical Radiotherapy Planning, 3rd edn. London: Arnold, pp. 149–64. Lebwohl, M., Dinehart, S., Whiting, D. et al. (2004). Imiquimod 5% cream for the treatment of actinic keratosis: results from two phase III, randomised, double-blind, parallel group, vehicle-controlled trials. J. Am. Acad. Dermatol., 50, 714–21. Lovett, R. D., Perez, C. A., Shapiro, S. J. et al. (1990). External irradiation of epithelial skin cancer. Int. J. Radiat. Oncol. Biol. Phys. 19, 235–42. Malhotra, R., Huilgol, S. C., Huynh, N. T. et al. (2004). The Australian Mohs database, part II. Periocular basal cell carcinoma outcome at 5-year follow-up. Ophthalmology, 111, 631–6. National Comprehensive Cancer Network. (2005). Basal and squamous cell skin cancer guideline. In The Complete Library of NCCN Clinical Practice Guidelines in Oncology, version 1. Jenkintown, PA: National Comprehensive Cancer Network (CD-ROM). NICE. (2006). Improving Outcomes for People with Skin Tumours Including Melanoma. London: National Institute for Health and Clinical Excellence. Rubin, A. I., Chen, E. H. and Ratner, D. (2005). Basal-cell carcinoma. N. Engl. J. Med., 353, 2262–9.
393
Sankha Suvra Mitra
Safai, B. (1997). Kaposi’s sarcoma and acquired immunodeficiency syndrome. In AIDS: Epidemiology, Diagnosis, Treatment and Prevention, ed. V. T. DeVita, S. Hellman and S. Rosenberg, 4th edn. Philadelphia, PA: Lippincott-Raven, pp. 295–358. Solan, M. J., Brady, L. W., Binnick, S. A. et al. (1997). Skin. In Principles and Practice of Radiation Oncology, ed. C. A. Perez and L. W. Brady, 3rd edn. Philadelphia, PA: Lippincott-Raven, pp. 723–44. Soutar, D. S. and Robertson, A. G. (2001). Skin cancer other than
394
melanoma. In Oxford Textbook of Oncology, 2nd edn. Oxford: Oxford University Press, pp. 1235–43. Thomas, R. M. and Amonette, R. A. (1998). Mohs micrographic surgery. Am. Fam. Physician, 37, 135–42. Tran, H., Chen, K. and Shumack, S. (2003). Summary of actinic keratosis studies with imiquimod 5% cream. Br. J. Dermatol., 149 (Suppl. 66), 37–9. UICC. (2002). TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 123–5.
34
MELANOMA Tom Crosby, Louise Hanna and Dafydd Roberts
Introduction Melanocytes originate in the neural crest of the embryo and migrate widely during development to locations such as the basal layer of the epidermis and the uveal tract. As a result of this migration, malignant melanoma (MM) can affect sites in addition to the skin, including the central nervous system (e.g. meninges and uveal tract) and the aerodigestive and genitourinary tracts (e.g. nasopharynx, oral cavity and vagina). The incidence of cutaneous melanoma has increased by 50% in the past two decades, although there are significant geographical variations, and it is about ten-fold higher in white than in non-white populations. Despite the rise in incidence, mortality has plateaued and even fallen in some populations, for instance in women and young men in Australia (Giles et al., 1996). During the same period, there has been a six-fold increase in melanoma in situ, suggesting improvement in earlier detection. This chapter focuses on cutaneous and ocular melanoma.
Cutaneous melanoma Types of cutaneous melanoma The main clinicopathological varieties of MM are superficial spreading, nodular, acral lentiginous and lentigo maligna melanoma.
Incidence and epidemiology Annual disease incidence in the UK is 8 in 100 000 patients. Approximately 8000 new cases are diagnosed per year in England and Wales. Cutaneous melanoma accounts for 3% of all cancers, with approximately 1800 deaths occurring annually. There has been a 50% increase in the number of new cases diagnosed over the past four decades, although mortality has plateaued in some populations. This
Table 34.1. Annual melanoma incidence in different populations Incidence Population
(per 100 000)
Asian
0.2
United Kingdom
8
United States
12
White people in Queensland, Australia
40
Adapted from Armstrong and Kricker (1994).
plateau, together with an increase in the incidence of in situ disease, suggests a higher rate of earlier diagnosis due to vigilance among the public and health care professionals. There is a wide geographical variation in incidence (see Table 34.1), and the diagnosis is ten-fold more common among whites than non-whites. Melanoma incidence increases with age, rising to a peak in incidence in the fourth and fifth decades.
Risk factors and aetiology Sunlight Severe sunburn caused by exposure to excessive sunlight during childhood is associated with an increased risk of developing MM in adult life. The incidence is increased for fair-skinned people who live close to the equator. However, the complex association between genetic and environmental factors is shown by the fact that, unlike non-melanomatous skin cancers, patients do not necessarily develop tumours in sites of maximum sun exposure. It has been estimated that two-thirds of the cases worldwide may be related to sun exposure.
Skin type The risk of developing MM is higher in fair-skinned people with blond or red hair, who have a tendency to burn. 395
Tom Crosby, Louise Hanna and Dafydd Roberts
Table 34.2. Risk factors for developing melanoma Odds ratio for Risk factor
melanoma
11–50 common moles >2 mm
1.7–1.9
51–100 common moles >2 mm
3.2–3.7
>100 common moles >2 mm
7.6–7.7
Family history of melanoma
1.8
Presence of 1–4 atypical moles
1.6–7.3
Adapted from Roberts et al. (2002).
Melanocytic naevi The number of common, atypical (dysplastic) naevi on a person’s body correlates well with the risk of developing MM, as shown in Table 34.2.
Other risk factors Patients receiving immunosuppressive therapy following organ transplant are at a three-fold increased risk of developing MM. There have been reports of increased risk of patients developing MM during pregnancy and while on the oral contraceptive pill. These findings have not been confirmed in extensive studies, although reports of changes in moles during pregnancy have been recorded, and prompt diagnosis of suspicious lesions should be made. Extensive treatment with PUVA (psoralens and UVA therapy) is associated with the development of MM many years after treatment. A genetic predisposition probably accounts for 5 to 10% of all cases; chromosomes 1, 6 and 9 are implicated in a multistep malignant transformation process. Inherited conditions (e.g. xeroderma pigmentosum) are also risk factors.
Vitamin D There is some preclinical evidence that vitamin D may be protective against malignant transformation of melanocytes, and that lower levels contributed to by reduced sun exposure may predispose to MM. However, there is no conclusive evidence for this finding or for the use of vitamin D supplementation.
Screening and prevention Early detection There is no MM screening programme for the general population. For patients at a high risk of melanoma, 396
various strategies are available, such as regular selfexamination or examination by suitably qualified health professionals. Serial recording of images using conventional or digital photography can help in long-term monitoring.
Prevention Based on current evidence it seems sensible to avoid prolonged periods of direct sun exposure and to protect children from the same. The use of sunscreens to prevent MM is controversial. Some studies have found that their use is associated with more prolonged exposure to UVA by preventing the burning effect of UVB light and so people may spend longer in direct sunlight (Crosby et al., 1999). Therefore, it seems preferable to prevent sunburn primarily via sun avoidance and the use of clothing, reserving sunscreens for areas of the body that cannot be protected in other ways. Cancer Research UK’s SunSmart campaign recommends the following (SunSmart, www.sunsmart.org .uk./, accessed August 2006): r S pend time in the shade between 11 and 3. r M ake sure you never burn. r A im to cover up with a t-shirt, hat, and sunglasses. r R emember to take extra care with children. r T hen use factor 15+ sunscreen.
Pathology By definition MM involves atypical melanocytes that infiltrate into the dermis. There may be features of neighbouring dysplastic or benign pigmented lesions. There are four main types of MM with distinct clinicopathological characteristics.
Superficial spreading melanoma (SSM) Superficial spreading melanoma occurs in approximately 70% of cases; it accounts for the changing epidemiology in recent decades, usually arising from a dysplastic naevus. Macroscopically, SSMs are pigmented lesions that are often flat or with slight elevation. Typically they have an irregular border and irregular pigmentation. Microscopically, there is predominantly horizontal growth. Cytological variations occur (e.g. epithelioid, spindle cell, small naevoid-like, or giant cell).
Nodular melanoma (NM) Nodular melanomas make up approximately 15% of cases; they form raised, nodular lesions which may vary in colour from blue-grey to completely amelanotic, and
Melanoma
they are often associated with ulceration and/or bleeding. Microscopically they have no or minimal horizontal growth but extensive dermal invasion.
Acral lentiginous melanoma (ALM) Acral lentiginous melanomas make up approximately 10% of cases; they occur on the palms, soles, subungual regions (most commonly great toe or thumb) and mucosal surfaces. ALMs are probably genetically distinct from SSM and less related to UV light exposure. Mucosal lesions are often diagnosed later. Microscopically there is acanthosis of the epidermis, atypical melanocytes with branching dendritic processes, and frequent spindle cells with marked desmoplastic reaction.
Table 34.3. Clark’s classification for melanoma (level of invasion) Level
Definition
I
Lesions involving only the epidermis (in situ
II
Invasion of the papillary dermis but does not
III
Invasion fills and expands the papillary dermis
IV
Invasion into the reticular dermis but not into
V
Invasion through the reticular dermis into the
melanoma); not an invasive lesion reach the papillary-reticular dermal interface but does not penetrate the reticular dermis the subcutaneous tissue subcutaneous tissue Adapted from Clark et al. (1969).
Lentigo maligna melanoma (LMM) Lentigo maligna melanomas make up approximately 5% of cases; they occur in older patients, usually on the skin of the face, and they are almost certainly related to chronic sun exposure. The precursor lesion, lentigo maligna (Hutchinson’s freckle), is slow growing with a long radial growth phase. About 5% progress to invasive LMM, which may be indicated by localised thickening of the lesion. Microscopically, there is lentiginous proliferation of atypical melanocytes, which is often associated with solar atrophy of the epidermis.
Depth of invasion Cutaneous melanoma can be defined by its depth of invasion, which has important prognostic significance. The Breslow thickness is the depth in millimetres from the granular layer of the epidermis to the deepest identifiable tumour cells (Breslow, 1970). Clark’s classification also defines melanoma by the depth of invasion but is used less commonly than Breslow thickness (Clark et al., 1969; Table 34.3).
Spread The skin, subcutaneous tissues and lymph nodes are the first site of metastatic disease in 59% of patients.
Local spread After undergoing malignant transformation, melanoma becomes invasive by penetrating into and beyond the dermis. Initially there is a variable horizontal growth phase, followed by a vertical growth phase invading for a variable distance through dermis and into surrounding structures.
Lymphatic spread Lymphatic spread to regional nodes is the most common occurrence. In-transit metastases or satellite nodules are defined as cutaneous or subcutaneous nodules that are more than 2 cm from the primary tumour and not beyond the draining lymph nodes.
Metastatic spread Haematogenous spread to the lungs, liver, bone, brain and skin is not uncommon with more invasive or thicker lesions.
Clinical presentation The main clinical features of cutaneous melanoma are a pigmented lesion with an irregular edge and irregular pigmentation. Over 95% of patients report a change in size, shape, or colour of the lesion, whereas fewer than 50% describe a change in sensation or bleeding of the lesion (Roberts et al., 2002). Melanoma more often involves the extremities in females and the head, neck and trunk in males. In darkskinned people, malignant melanoma, when it does occur, tends to involve the palms and soles. Suspicious pigmented lesions are best examined in a good light with or without magnification and should be assessed using the seven-point checklist or ABCDE systems (see Table 34.4). The presence of any major feature in the seven-point checklist, or any of the features in the ABCDE system, is an indication for referral. The presence of minor features should increase suspicion. 397
Tom Crosby, Louise Hanna and Dafydd Roberts
Table 34.4. Clinical features suggestive of melanoma Scheme Revised seven-point checklist
Description Major features Change in size of lesion Irregular pigmentation Irregular border Minor features Inflammation Change in sensation Largest diameter 7 mm or greater Oozing, crusting, or bleeding
ABCDE rule
A Asymmetry B irregular Border C irregular Colour
be taken with a view to a subsequent wider therapeutic excision if required; clearly, specimen orientation is required for correct histological interpretation. The lymph nodes should be examined clinically in all cases. The role of sentinel lymph node biopsy (SLNB) as a staging investigation and treatment is discussed later in the chapter. Investigations for metastatic disease (e.g. CT scan of chest and abdomen; full blood count; renal, liver, and bone profile; and LDH) should be individually tailored, depending on any symptoms and the prognosis of the locoregional disease (see following). These types of investigation may, for instance, be appropriate for patients with high-risk local disease (stage IIB and higher) with or without regional disease where extensive primary surgery and/or adjuvant therapy is being considered.
D Diameter >5 mm E Elevation Adapted from MacKie (1989) and Fitzpatrick et al. (1988).
Differential diagnosis of melanoma The following can appear similar to MM: r Basal cell papilloma (seborrhoeic keratosis). r Pigmented basal cell carcinoma. r Thrombosed angioma. r Pyogenic granuloma. r Dermatofibroma.
Investigation and staging Dermatoscopy or epiluminescence microscopy (noninvasive ‘microscopic’ review of a suspicious lesion in vivo) has gained ground as a diagnostic aid. In experienced hands, it is useful for excluding benign lesions. The principal diagnostic investigation is the excision of the lesion, which is critical to further management and determines prognosis (see under treatment, staging, and prognosis). For smaller tumours, an excisional biopsy with a margin of 1 to 2 mm of normal skin is optimal. Where the primary lesion is large and/or the primary surgical therapy would be mutilating, an incisional biopsy is sometimes required. Although this approach may be non-representative of the whole specimen in terms of prognosis and there is a theoretical risk of disseminating MM cells, there is no evidence that this approach adversely affects the overall outcome (Roberts and Crosby, 2003). In either case, the biopsy should 398
Stage classification The TNM classification and stage groupings are shown in Tables 34.5 and 34.6, respectively.
Treatment overview The main aims of treatment are to detect the lesion as early as possible and to excise it with adequate margins but without unnecessarily mutilating the patient. Excisional surgery is very successful in the treatment of early or thin lesions. The extent of surgery and the role of routine lymph node dissection/biopsy have been areas of active research and are discussed later. MM is a relatively chemotherapy- and radiotherapyresistant tumour. When haematogenous spread to liver, bone, and brain occurs, the natural history of MM is that it can be one of the most aggressive of all malignant diseases. Response rates in patients with advanced disease are low. It is, therefore, of little surprise that there is no role for the use of these modalities in the adjuvant setting. Radiotherapy has been used for patients with incompletely excised disease or extracapsular nodal disease but its benefit in these situations is unproven. Responses to biological therapies are more variable although the value of their routine use, both in advanced disease and as adjuvant therapy, has not been established. An excellent summary of treatment recommendations can be located through the Scottish Intercollegiate Guidelines Network website (SIGN; www. sign.ac.uk/guidelines/fulltext/72/index.html, accessed September 2006).
Melanoma
Table 34.5. TNM staging classification for melanoma
Table 34.6. Stage groupings for melanoma
Stage
Description
Stage group
pTis
Melanoma in situ (Clark level 1)
0
Tis N0 M0
pT1
Tumour ≤1.0 mm in thickness with or
IA
T1a N0 M0
without ulceration
Definition
IB
T1b or T2a N0 M0
pT1a
Clark’s level II or III with no ulceration
IIA
T2b or T3a N0 M0
pT1b
Clark’s level IV or V or with ulceration
IIB
T3b or T4a N0 M0
Tumour >1.0 mm but ≤2.0 mm in thickness
IIC
T4b N0 M0
pT2a
Without ulceration
III
Any T N1–3 M0
pT2b
With ulceration
IV
Any T Any N M1
pT2
pT3
Tumour >2.0 mm but ≤4.0 mm in thickness
pT3a
Without ulceration
pT3b
With ulceration
pT4
Tumour >4.0 mm in thickness
pT4a
Without ulceration
pT4b
With ulceration
From UICC (2002).
Table 34.7. Excision margins recommended for cutaneous melanoma
Metastasis to one lymph node
Thickness of primary lesion
N1a
Clinically occult (microscopic) metastasis
(mm)
N1b
Clinically apparent (macroscopic) metastasis
<1
1
1–2
1–2
2–4
2
>4
2
N1
N2
Metastasis to 2 or 3 regional nodes or intralymphatic regional metastasis
N2a
Clinically occult (microscopic) metastasis
N2b
Clinically apparent (macroscopic) metastasis
N2c
Satellite or in-transit metastasis without nodal metastasis
N3
Metastasis in ≥4 regional nodes, or matted metastatic regional lymph nodes, or
Margin of excision (cm)
From British Association of Dermatologists, UK guidelines for the management of cutaneous melanoma; www.bad. org.uk/healthcare/guidelines/cm.asp, accessed November 2006.
satellite or in-transit metastasis with metastasis in regional node(s) M1
Distant metastasis
M1a
Metastasis to skin, subcutaneous tissues or
M1b
Metastasis to lung
M1c
Metastasis to all other visceral sites or distant
distant lymph nodes
metastasis at any site associated with an elevated serum lactic dehydrogenase Adapted from UICC (2002).
Surgery Surgery to primary lesion Once a diagnosis of MM has been made, and in the absence of any evidence of metastatic disease, definitive surgery should be planned. Surgery involves a wide excision of the primary lesion and its use depends on the thickness of the primary tumour. In recent years following the results of three large clinical trials, the mar-
gin of excision that is considered adequate has been reduced (Balch et al., 1993; Cohn-Cedermark et al., 2000; Heaton et al., 1998; Veronesi and Cascinelli, 1991). Table 34.7 shows recommended excision margins for different thicknesses of melanoma.
Routine elective lymph node dissection (ELND) Four RCTs including 1718 patients with no clinical evidence of lymph node metastases have investigated the role of ELND and concluded that there is no overall survival benefit for this procedure when it is used in addition to wide excision (Balch et al., 1996, 2000; Sim et al., 1986; Veronesi et al., 1982). Although ELND may be useful in patients with established lymph node metastases, lymph node dissection is not without risk, lymphoedema being the most frequent complication.
Sentinel lymph node biopsy (SLNB) SLNB involves the identification and biopsy of the firststation lymph node that drains an affected area by the 399
Tom Crosby, Louise Hanna and Dafydd Roberts
use of blue dye and/or radiolabelled colloid injected into the skin surrounding the primary lesion. The technique can successfully identify the sentinel node in up to 97% of cases. The current practice is that patients with a positive node proceed to radical node dissection. SLNB is generally agreed to be useful as a staging procedure in patients with primary cutaneous melanoma. Studies are under way to investigate whether SLNB should be used routinely in lesions with a thickness greater than 1 mm. If this role is established, quality assurance programmes should be used to ensure 90 to 95% sensitivity and specificity in identifying the appropriate lymph node. The SIGN guidelines recommend that SLNB should be considered as a staging technique in patients with primary melanoma 1 mm or thicker or primary melanoma smaller than 1 mm thick and Clark level 4, which gives staging and prognostic information. Whether SLNB and subsequent lymph node dissection improve outcome is the subject of an ongoing trial (see later discussion). Recently, Morton et al. (2006) reported a large trial of 1269 patients who were randomised to either wide excision of the primary plus observation of the regional lymph nodes with lymphadenectomy if nodal relapse occurred or wide excision and sentinel-node biopsy with immediate lymphadenectomy if nodal micrometastases were detected on biopsy. In the SLNB group, the presence of metastases in the sentinel node was the most important prognostic factor (5-year survival for the node-positive and node-negative groups was 72.3 versus 90.2%, HR for death = 2.48, p < 0.001). The 5-year melanoma-specific survival rates were similar in the two groups (87.1% in the biopsy group, 86.6% in the observation group). But a subgroup comparison showed that the rate of melanoma-specific deaths in the biopsy group was 26.2% among patients who underwent immediate lymphadenectomy compared with 48.7% in the observation group who underwent delayed lymphadenectomy. Overall, SLNB provides prognostic information, and immediate lymphadenectomy may improve survival for patients who are found to have positive nodes, but at present it is not a standard technique and it is not widely available.
Adjuvant therapy Immunotherapy At the present time, there is no evidence for routine use of any adjuvant therapy for melanoma. Current evidence and strategies are discussed later in the chapter. 400
Treatment for advanced disease Immunotherapy The immune system is important in metastatic MM, as shown by lymphoid infiltration into tumour and surrounding tissues, and well-reported spontaneous remissions. Interferon alfa and interleukin-2 have shown single-agent response rates of 15 to 20% and it has been suggested that such therapy produces a higher rate of durable remissions.
Chemotherapy Dacarbazine (DTIC) has been the most tested single chemotherapeutic agent. It is reasonably well tolerated and the dose-limiting toxicities are myelosuppression and nausea/vomiting. The latter has largely been overcome by the use of 5-HT3 antagonists. DTIC is still considered to be the ‘gold standard’ against which other therapies should be tested (Crosby et al., 2002). When used alone it gives partial response rates of about 20% and complete responses in 5 to 10%. It is given intravenously at 850 to 1000 mg/m2 every 3 weeks. Temozolomide is a novel alkylating agent with 100% oral bioavailability and good penetration of the bloodbrain barrier and cerebrospinal fluid. Its effectiveness is at least equal to that of DTIC in metastatic MM, giving a median survival of 77 months with temozolomide and 64 months compared with DTIC (HR = 1.18, CI = 0.92–1.52), and with improvement in some aspects of quality of life (Middleton et al., 2000). Given its similar mechanism of action to that of DTIC, it is not surprising that response rates are fairly similar, but in a disease with such a poor prognosis, ease of administration and quality of life are clearly very important. Many other drugs such as platinum agents, vinca alkaloids, nitrosoureas and, more recently, taxanes have been tried alone and in various combination regimens. Higher response rates are often seen in phase II studies, but none of the aforementioned drugs has shown superiority to single-agent DTIC in RCTs. An example of the false promise followed reports of a response rate of 55% with the combination of DTIC, cisplatin, carmustine and tamoxifen, known as the Dartmouth regimen. However, a multicentre randomised trial comparing this regimen with single-agent DTIC found no survival advantage and only a small, non-significant increase in tumour response in an intention-to-treat analysis (Chapman et al., 1999). The response rate to single-agent DTIC was 9.9% with a median survival of 6.3 months (CI = 5.4 to 8.7 months). The response rate
Melanoma
to the Dartmouth regimen was 16.8%, with a median survival of 7.7 months (CI = 6.3 to 8.9 months). Combination therapies should not be used routinely outside the context of clinical trials.
Combined palliative chemotherapy and immunotherapy One meta-analysis of single-agent DTIC and combination chemotherapy with or without immunotherapy was performed in metastatic MM (Huncharek et al., 2001). No overall survival advantage was found for combination treatment, although the addition of interferon alfa increased the response rate by 53% over DTIC alone. Outside clinical trials, standard therapy should remain single-agent DTIC, with temozolomide for selected patients such as those in whom intravenous therapy may particularly interfere with quality of life and possibly in patients with predominantly cranial metastases.
Isolated limb perfusion Isolated limb perfusion, using melphalan with or without tumour necrosis factor, can be useful for peripheral, usually lower-limb, melanomas where the primary lesion(s) is unresectable, are multiple, or has extensive in-transit metastases.
Endocrine treatment An early study in 117 patients suggested a benefit for the addition of tamoxifen to single-agent DTIC (RR = 28 versus 12%, p = 0.03, MS = 48 weeks versus 29 weeks, p = 0.02). Again, this result was not confirmed in a fourarm study in 258 patients with metastatic MM. Response rates were 19% (CI = 12–26%) for patients receiving tamoxifen and 18% in the non-tamoxifen group (CI = 12–25%). There is no consistent evidence to suggest a benefit for hormonal therapy (Falkson et al., 1998).
Radiotherapy Although MM is relatively radioresistant, radiotherapy can be used to control disease in patients with unresectable primary tumours. In particular, radiotherapy can prevent ulceration and bleeding, unresectable lymph node disease, and painful subcutaneous or bone metastases. In this situation, relatively high-dose and hypofractionated regimens should be used to overcome the ‘shoulder effect’ seen in radioresistant cell survival curves (Steel, 2002), for example, 6 Gy per fraction given weekly for 5 to 6 weeks.
Table 34.8. Approximate 5-year survival for cutaneous melanoma by stage Stage
Prognosis (approximate 5-year survival)
I
91%
II
64%
III
40%
Adapted from Ben-Porat et al. (2006).
Prognosis For all patients with metastatic disease, the median survival is approximately 7 months; 25% will be alive after 1 year and only 5% of patients will be alive 5 years after diagnosis. Patients with a better performance status and women have a better prognosis (p = 0.001 and 0.056, respectively). Survival is also better in patients with a longer duration of remission after primary disease, who have fewer metastatic sites involved, and in patients with non-visceral disease (Crosby et al., 2002). Table 34.8 shows the 5-year survival for stages I to III.
Prognostic factors Prognostic factors at time of presentation are as follow (Balch et al., 2004): r Tumour thickness – the principal prognostic factors for patients with early MM (stages I to III) relate to the depth of invasion of the primary lesion. The level of invasion, or Clark level, is strongly correlated with outcome but is not as reproducible among pathologists. r Ulceration – the absence of an intact epidermis diagnosed histologically is associated with a doubling of the risk associated with the depth of penetration. r Lymph nodes – the prognosis worsens with the number of nodes involved. A thin lesion (Breslow depth <0.76 mm) without lymph node involvement has a 3% risk of metastasising in 5 years, whereas if regional lymph nodes are macroscopically involved there is a less than 20 to 50% chance of surviving 5 years. r Gender – most studies have shown a better prognosis for females than males. r Anatomical location – tumours on the extremities appear to have a better prognosis than those on the trunk. r Age – older patients tend to have thicker lesions, are male with tumours on the head and neck and have ulcerated tumours at the time of diagnosis. 401
Tom Crosby, Louise Hanna and Dafydd Roberts
Areas of current interest Sentinel lymph node biopsy SLNB, discussed earlier, is known to provide important prognostic information. Uncertainty remains about the clinical benefit of immediate lymphadenectomy for patients who are found to have positive nodes. SLNB is not available for all patients at the present time.
Adjuvant interferon There is a plethora of evidence on factors that determine the prognosis of patients with primary MM. Patients with high-risk disease have up to a 50% chance of subsequent relapse and, once patients with MM develop distant metastatic disease, the prognosis is poor. Therefore, a clear need exists to develop a well-tolerated adjuvant therapy. The role of adjuvant treatments, mainly in the form of interferon alfa-2b, remains controversial. Several studies have shown that interferon alfa (IFNα) has a biologically modifying effect on MM but the effect on overall survival in the adjuvant setting has been variable. Side effects are a major problem with patients receiving high-dose IFNα. A meta-analysis of published trial data in patients receiving adjuvant IFNα suggests a significant reduction in local recurrence (HR = 0.83, 95% CI = 0.77–0.90, p = 0.000003) but no definite improvement in overall survival (HR = 0.93, 95% CI = 0.85– 1.02, p = 0.1; Wheatley et al., 2003). It is important to take into account the cost and toxicities associated with such treatment and, as such, the routine use of adjuvant therapy is not recommended. Treatment with interferon is expensive. In addition, interferon therapy commonly causes malaise, fevers and flu-like symptoms. In the ECOG 1684 study, high-dose IFNα also caused significant (>grade 3) myelosuppression in 24% of patients; hepatotoxicity in 15%, including two deaths; and neurotoxicity in 28% (Kirkwood et al., 1996). At 11 months, only 25% of participants were receiving greater than 80% of the planned dose. In the trial of low-dose interferon, 10% of patients suffered significant toxicity (Hancock et al., 2004).
High-dose interferon Intergroup E1684, an RCT of 287 people with tumours greater than 4 mm in depth, compared high-dose treatment (i.v. interferon alfa-2b 20 MU/day for 1 month, followed by 10 MU three times weekly for 11 months) with surgery alone. There was a significant improvement in disease-free and overall survival in those patients receiving interferon. The median 402
survival in the interferon group was 3.8 years, compared with 2.8 years for patients treated with surgery alone (Kirkwood et al., 1996). However, in a ‘confirmatory’ study of 642 patients randomised to receive high-dose interferon, low-dose interferon or no adjuvant therapy, there was no difference in the overall survival of patients with either high- or low-dose interferon alfa compared to patients who received surgery alone. It would appear that this result was due to a better-than-expected outcome with no adjuvant therapy (Kirkwood et al., 2000). A third study from the same authors compared highdose interferon alfa-2b with vaccine treatment (GM2KLH(QS-21) in 880 patients with resected stage IIB to III melanoma of the skin and randomised patients equally between the two interferon alfa and vaccine groups. There was a significant benefit for patients receiving interferon alfa-2b in median survival (HR = 1.52, CI = 1.07–2.15, p = 0.009). The outcome for patients receiving the vaccine seemed to be no worse than for similar patients receiving observation only (Kirkwood et al., 2001), which, according to the authors, seemed to confirm the benefits of high-dose interferon for relapse-free survival and overall survival reported earlier. Overall, adjuvant high-dose interferon has not become standard treatment for melanoma.
Low-dose interferon Three RCTs, including the UK AIM HIGH study investigating the benefit of low-dose interferon for 12 to 18 months in 674 patients with intermediate risk MM, found no benefit to overall survival (Grob et al., 1998; Hancock et al., 2004; Pehamberger et al., 1998). This treatment cannot be recommended routinely outside clinical trials.
Other adjuvant therapies Treatments such as BCG, coumarins, tamoxifen, retinoids and cytotoxic chemotherapy have also been attempted in order to improve results after surgical excision but have not demonstrated any consistent benefit (Sondak and Wolfe, 1997).
Vaccines Vaccines have been studied in melanoma in the adjuvant and metastatic settings. In advanced disease, vaccines have been shown to be capable of inducing specific T-cells, resulting in immune responses and disease stabilisation or regression. The MAGE-3 gene codes for a tumour-specific antigen that is expressed frequently in melanoma. The antigen has been combined with an adjuvant to stimulate an immune response and is now
Melanoma
the subject of a phase III study, the MAGE 3 study. Vaccination strategies are discussed in Chapter 3 (p. 20).
usually dome-shaped lesions, but they may infiltrate in a mushroom-like growth pattern through the Bruch membrane.
Current clinical trials At the time of writing, there were six open studies registered with the National Cancer Research Network (NCRN; www.ncrn.org.uk/, accessed September 2006). EORTC 18032 is a randomised phase II study of the EORTC Melanoma Group that studies an extendedschedule escalated-dose temozolomide versus DTIC in stage IV metastatic melanoma. MAGE 3 is a randomised, open phase II study of immunisation with the recombinant MAGE-3 protein combined with adjuvant ASO2B or AS15 in patients with unresectable and progressive metastatic cutaneous melanoma. The Melanoma Cohort Study is a melanoma followup and case-control family study. The Melanoma Family Study is a study of familial melanoma. PTK787 is a phase II study to evaluate the efficacy and safety of PTK787 in patients with metastatic cutaneous melanoma. The SLNB feasibility study in patients with melanoma is a study investigating the diagnosis and treatment of early lymph node involvement in patients with primary cutaneous melanoma by SLNB with or without completion lymphadenectomy and molecular markers.
Ocular melanoma Incidence and epidemiology After the skin, the eye is the second most common site of melanoma, but disease occurrence is still rare. The incidence is approximately 0.6 in 100 000 patients: fewer than 500 new cases are diagnosed per year in the UK, usually in patients in the fifth and sixth decades. For ocular melanomas, 85% of tumours arise from melanocytes within the highly vascular, subretinal choroid (tumours in the iris, ciliary body and conjunctiva are much less common). Disease occurrence is associated with the dysplastic naevus syndrome, neurofibromatosis, basal cell carcinoma (Gorlin’s) syndrome and BRCA2 +ve breast/ovarian cancer. The tumours are usually a mixture of spindle and epithelioid cells, a predominance of the latter being associated with a poorer prognosis. As with cutaneous tumours, their colouration may vary from deeply pigmented to amelanotic. They are
Clinical features and spread Presentation usually involves a patient with a visual field defect; in general, the farther away the tumour is from the optic nerve, the larger the defect before the patient notices a problem. Associated retinal detachment can cause more sudden vision loss. Other patients may present with a red eye due to anterior extension or a vitreous haemorrhage. The most common spread of the tumour is haematogenously; the liver is the most common site of secondary spread, followed by the lungs, bone, and brain. Distant spread has frequently occurred by the time of diagnosis. Secondary disease may appear up to 10 years after diagnosis of the primary but is most common in the first 12 months.
Investigation A diagnosis can usually be made by opthalmoscopy but fluorescein angiography and ultrasound/MRI can be valuable in confirming the diagnosis. In experienced institutions, a correct diagnosis can be made without histological confirmation in 98% of cases. Biopsies are occasionally taken, particularly for more accessible anterior tumours.
Treatment The optimal treatment for large choroidal tumours is enucleation. Smaller tumours may be treated with radiation, typically brachytherapy (scleral plaque radiotherapy) or charged particle (most commonly proton beam) external beam therapy, in an attempt to preserve vision. Radiotherapy is only performed in specialist centres.
Scleral plaque therapy Scleral plaques are sutured during a surgical procedure in which the tumour is located with transillumination. The dose is 80 to 100 Gy delivered to the apex of the lesion with low-energy gamma emitting isotopes, most commonly 125 I or ruthenium (106 Ru), shielded externally by gold. Treatment takes 2 to 10 days. The Collaborative Ocular Melanoma Study (COMS), which involved more than 1300 patients who were randomised to receive either enucleation or 125 I plaque brachytherapy, found 403
Tom Crosby, Louise Hanna and Dafydd Roberts
no difference in survival between treatments (DienerWest et al., 2001). For patients receiving brachytherapy, there was a 10% risk of local treatment failure, 85% retained their eye and 37% had normal visual acuity 5 years after treatment. Patients eligible for brachytherapy in the COMS trial had tumours larger than 2.5 mm and smaller than 10 mm in diameter and less than 16 mm thick.
Charged particle therapy Charged particle therapy (helium ions and proton beam) takes advantage of precisely focused pencilbeam radiation that has well-defined penetrating depths due to sudden decelerating energy dispersion (the Bragg peak) to deliver high doses to the target volume while sparing normal tissues. The target volume is defined by placing inert clips during surgery and accurately measuring tumour dimensions relative to these clips by studying retinal photographs and by performing ultrasonographic biometry of the globe prior to 3D planning. Patients require shell immobilisation, and a dose of 60 Gy is given in four daily fractions. This procedure achieves local control that is comparable to that of brachytherapy but the treatment is limited by the availability of equipment, which includes a cyclotron, to units such as Clatterbridge, Liverpool (Fuss et al., 2001), and is often reserved for patients with tumours thicker than 8 to 10 mm.
REFERENCES Armstrong, B. K. and Kricker, A. (1994). Cutaneous melanoma. Cancer Surv., 19–20, 219–40. Balch, C. M., Urist, M. M., Karakousis, C. P. et al. (1993). Efficacy of 2 cm surgical margins for intermediate-thickness melanomas (1 to 4 mm). Result of a multi-institutional randomized surgical trial. Ann. Surg., 218, 262–7. Balch, C. M., Soong, S. J., Bartolucci, A. A. et al. (1996). Efficacy of an elective regional lymph node dissection of 1 to 4 mm thick melanomas for patients 60 years of age and younger. Ann. Surg., 224, 255–66. Balch, C. M., Soong, S., Ross, M. I. et al. (2000). Long-term results of a multi-institutional randomized trial comparing prognostic factors and surgical results for intermediate thickness melanomas (1.0 to 4.0 mm). Ann. Surg. Oncol., 7, 87–97. Balch, C. M., Atkins, M. B. and Sober, A. J. (2004). Cutaneous melanoma. In Cancer: Principles and Practice of Oncology, Ed. V. T. Devita, Jr., S. Hellman and S. A. Rosenberg, 7th edn. Philadelphia, PA: JB Lippincott, pp. 1754–808. Ben-Porat, L., Panageas, K. S., Hanlon, C. et al. (2006). Estimates of stage-specific survival are altered by changes in the 2002 American Joint Committee on Cancer staging system for melanoma. Cancer, 106, 163–71.
404
Breslow, A. (1970). Thickness, cross-sectional areas and depth of invasion in the prognosis of cutaneous melanoma. Ann. Surg., 172, 902–8. Chapman, P. B., Einhorn, L. H., Meyers, M. L. et al. (1999). Phase III multicenter randomised trial of the Dartmouth regimen versus dacarbazine in patients with metastatic melanoma. J. Clin. Oncol., 17, 2745–51. Clark, W. H. Jr., From, L., Bernardino, E. A. et al. (1969). The histogenesis and biologic behaviour of primary human malignant melanomas of the skin. Cancer Res., 29, 705–26. Cohn-Cedermark, G., Rutqvist, L. E., Andersson, R. et al. (2000). Long term results of a randomized study by the Swedish Melanoma Study Group on 2 cm versus 5 cm resection margins for patients with cutaneous melanoma with a tumor thickness of 0.8 mm–2.0 mm. Cancer, 89, 1495–501. Crosby, T., Crosby, D. L. and Mason, M. D. (1999). Malignant melanoma: non-metastatic. Clin. Evid., 2, 659–68. Crosby, T., Fish, R., Coles, B. et al. (2002). Systemic treatments for metastatic cutaneous melanoma. In Cochrane Collaboration. Cochrane Library, Issue 2. Oxford: Update Software. Diener-West, M., Earle, J. D., Fine, S. L. et al. (2001). The COMS randomized trial of iodine 125 brachytherapy for choroidal melanoma, III: initial mortality findings. COMS Report No. 18. Arch. Ophthalmol., 119, 969–82. Falkson, C. I., Ibrahim, J., Kirkwood, J. M. et al. (1998). Phase III trial of dacarbazine versus dacarbazine with interferon-alpha2b versus dacarbazine with tamoxifen versus dacarbazine with interferon-alpha 2b and tamoxifen in patients with metastatic malignant melanoma: an Eastern Cooperative Oncology Group Study. J. Clin. Oncol., 16–17, 43–51. Fitzpatrick, T. B., Rhodes, A. R., Sober, A. J. et al. (1988). Primary malignant melanoma of the skin: the call for action to identify persons at risk; to discover precursor lesions; to detect early melanomas. Pigment Cell, 9, 110–17. Fuss, M., Loredo, L. N., Blacharski, P. A. et al. (2001). Proton radiation therapy for medium and large choroidal melanoma: preservation of the eye and its functionality. Int. J. Radiat. Oncol. Biol. Phys., 49, 1053–9. Giles, G. G., Armstrong, B. K., Burton, R. C. et al. (1996). Has mortality from melanoma stopped rising in Australia? Analysis of trends between 1931 and 1994. B. M. J., 312, 1121–5. Grob, J. J., Dreno, B., de la Salmoniere, P. et al. (1998). Randomised trial of interferon alfa-2a as adjuvant therapy in resected primary melanoma thicker than 1.5 mm without clinically detectable node metastases. Lancet, 351, 1905–10. Hancock, B. W., Wheatley, K., Harris, S. et al. (2004). Adjuvant interferon in high-risk melanoma: the AIM HIGH study – United Kingdom Coordinating Committee on Cancer Research randomized study of adjuvant low-dose extended-duration interferon alfa-2a in high-risk resected malignant melanoma. J. Clin. Oncol., 22, 53–61. Heaton, K. M., Sussman, J. J., Gershenwald, J. E. et al. (1998). Surgical margins and prognostic factors in patients with thick (> 4 mm) primary melanoma. Ann. Surg. Oncol., 5, 322–8. Huncharek, M., Caubet, J. F. and McGarry, R. (2001). Single agent DTIC versus combination chemotherapy with or without imunotherapy in metastatic melanoma: a meta-analysis of 3273 patients from 20 randomized trials. Melanoma Res., 11, 75–81.
Melanoma
Kirkwood, J. M., Strawderman, M. H., Ernstoff, M. S. et al. (1996). Interferon alfa-2b adjuvant therapy of high risk resected cutaneous melanoma: the Eastern Co-operative Oncology Group Trial. Est. 1684. J. Clin. Oncol., 14, 7–17. Kirkwood, J. M., Ibrahim, J. G., Sondak, V. K. et al. (2000). High- and low-dose interferon alfa-2b in high-risk melanoma: first analysis of Intergroup trial E1690/S9111/C9190. J. Clin. Oncol., 18, 2444–58. Kirkwood, J. M., Ibrahim, J. G., Sosman, J. A. et al. (2001). High-dose interferon alfa-2b significantly prolongs relapse-free and overall survival compared with the GM2-KLH/QS-21 vaccine in patients with resected stage IIB-III melanoma: results of Intergroup trial E1694/S9512/C509801. J. Clin. Oncol., 19, 2370–80. MacKie, R. M. (1989). Malignant Melanoma: A Guide to Early Diagnosis. Glasgow: University of Glasgow. Middleton, M. R., Grob, J. J., Aaronson, N. et al. (2000). Randomised phase III study of temozolamide versus dacarbazine in the treatment of patients with advanced metastatic malignant melanoma. J. Clin. Oncol., 18, 158–66. Morton, D. L., Thompson, J. F., Cochran, A. J. et al. (2006). Sentinel-node biopsy or nodal observation in melanoma. N. Engl. J. Med., 355, 1307–17. Pehamberger, H., Soyer, H. P., Steiner, A. et al. (1998). Adjuvant interferon alfa-2 a in resected primary stage II cutaneous melanoma. J. Clin. Oncol., 16, 1425–9.
Roberts, D. and Crosby, T. (2003). Cutaneous melanoma. In BMJ Books. Evidenced Based Dermatology, ed. H. Williams, M. Bigby, T. Diegpen et al. Oxford: Blackwell Publishing, Chap. 24. Roberts, D. L., Anstey, A. V., Barlow, R. J. et al. (2002). U. K. guidelines for the management of cutaneous melanoma. Br. J. Dermatol., 146, 7–17. Sim, F. H., Taylor, W. F., Pritchard, D. J. et al. (1986). Lymphadenectomy in the management of stage I malignant melanoma: a prospective randomized study. Mayo Clin. Proc., 61, 697–705. Sondak, V. K. and Wolfe, J. A. (1997). Adjuvant therapy for melanoma. Curr. Opin. Oncol., 9, 189–204. Steel, G. G. (2002). Basic Clinical Radiobiology, 3rd edn. London: Arnold. UICC. (2002). TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 126–30. Veronesi, U. and Cascinelli, N. (1991). Narrow excision (1 cm margin); a safe procedure for thin cutaneous melanoma. Arch. Surg., 126, 438–41. Veronesi, U., Adamus, J., Bandiera, D. C. et al. (1982). Delayed regional lymph node dissection in stage I melanoma of the skin of the lower extremities. Cancer, 49, 2420–30. Wheatley, K., Ives, N., Hancock, B. et al. (2003). Does adjuvant interferon-α for high-risk melanoma provide a worthwhile benefit? A meta-analysis of the randomised trials. Cancer Treat. Rev., 29, 241–52.
405
35
THYROID Laura Moss
Introduction Thyroid cancer consists of a diverse group of tumours with different clinical features and prognoses. Thyroid cancer can occur at any age, but it is rare in patients under the age of 25. Radiation exposure is the best-documented risk factor. Most thyroid cancers are carcinomas: these are papillary, follicular, medullary, and anaplastic, in order of frequency. Thyroid lymphomas and sarcomas are rarer. The overall prognosis is related to histological type; well-differentiated thyroid cancer (papillary and follicular) has the best prognosis. In contrast, anaplastic carcinoma progresses rapidly and has a very poor prognosis. There is a lack of evidence from prospective randomised controlled studies because differentiated thyroid cancer is an uncommon disease with a long natural history. Many areas of thyroid cancer management remain controversial, including the extent of surgery and the indications for radioiodine ablation and radiotherapy. Evidence-based guidelines published in 2002 relied on large retrospective and cohort studies (British Thyroid Association, 2002). This chapter focuses mainly on differentiated thyroid cancer, with shorter sections on medullary thyroid carcinoma (MTC), anaplastic thyroid cancer and thyroid lymphoma.
Types of thyroid tumour Thyroid tumours can be divided into benign, malignant primary and malignant secondary. Types of thyroid tumour are shown in Table 35.1. The relative proportions of patients with differentiated thyroid cancer in a geographic area depend on dietary iodine intake. The proportion of follicular cancers increases where there is dietary iodine deficiency.
Anatomy The thyroid develops from an endodermal outgrowth from the midline of the pharyngeal floor, which then 406
Table 35.1. The range of thyroid tumours Type of tumour
Examples (approximate incidence %)
Benign
Follicular adenoma – single or multiple;
Malignant
Differentiated thyroid cancer
¨ cell adenoma variant = Hurthle primary
Papillary (PTC: 80%) Follicular (FTC: 5–20%) Other carcinomas Medullary (8–12%, of which 75–80% are sporadic) Anaplastic (1–3%) Other malignant tumours Lymphoma (5%) Sarcoma
Malignant secondary
Melanoma Renal Breast Lung
becomes the thyroglossal duct and elongates before developing into two lobes. The solid cord joining the gland to the tongue then disappears. The thyroid consists of two lobes connected by the isthmus, and it weighs 15 to 20 g. It is very vascular and is surrounded by a sheath derived from the pretracheal fascia. The apex of each lobe can reach up to the oblique line on the thyroid cartilage and the base lies at the level of the fourth or fifth tracheal ring. The isthmus overlies the second, third and fourth tracheal rings. The pyramidal lobe is often present and extends up from the isthmus. The posterior aspect of each lobe is related to the four parathyroid glands, which lie within the fascial capsule of the thyroid gland behind the middle and inferior parts of the gland. The recurrent laryngeal nerve passes deep to the thyroid gland and is closely related to it, lying in the groove between the trachea and the oesophagus.
Thyroid
With respect to lymphatic drainage, the first-station nodes are paralaryngeal, paratracheal and prelaryngeal (central compartment/level VI).
mosome 10q, which codes for a receptor-like tyrosine kinase. C cell hyperplasia may be present.
Non-Hodgkin lymphoma
Pathology Microscopically, the majority of the thyroid gland is made up of follicles filled with colloid. The parafollicular or C cells originate from the neural crest, produce calcitonin, and are located outside the follicles. They account for 0.1% of thyroid cells and they lie at the junction of the upper and lower two-thirds of the lobes. Tumours can arise from the follicular epithelium (papillary, follicular and anaplastic), parafollicular or C cells (medullary), or non-epithelial stromal elements.
Many tumours are derived from mucosa-associated lymphoid tissue (MALT) and are therefore of low grade with a tendency for distant relapse. High-grade lymphomas can also occur.
Hurthle cell/oxyphil tumours ¨ ¨ The majority of Hurthle cell/oxyphil tumours are benign; if malignant, they are usually well differentiated. They produce thyroglobulin but rarely take up iodine.
Papillary cancers Papillary cancers are multifocal in up to 75% of cases (frequency of multifocality depends on method of pathological assessment). Orphan Annie nuclei and psammoma bodies are typical. The tall cell, columnar, and diffuse sclerosing variants are more aggressive.
Follicular cancers Cytology cannot distinguish adenomas from malignant tumours and, therefore, follicular cancers cannot be diagnosed with FNA. Histologically, capsular or blood vessel invasion is often the only feature of malignancy.
Anaplastic carcinoma Anaplastic carcinoma arises from follicular cells. There is disagreement regarding whether it originates de novo or from differentiated thyroid cancer.
Medullary thyroid cancer MTC arises from parafollicular cells or C cells (neural crest/neuroendocrine origin), which can secrete calcitonin and carcinoembryonic antigen (CEA). Amyloid may be present.
Hereditary types Tumours arising with MEN 2A and MEN 2B are often bilateral and multicentric. They are inherited in an autosomal dominant fashion and are associated with a germline mutation in the RET proto-oncogene on chro-
Screening There is currently no screening programme for the general population. For individuals with a strong family history of thyroid cancer or association with other cancers, genetic advice should be sought. For MTC, all newly diagnosed patients, regardless of their family history, should be referred for RET mutation testing, which tests for exons 10, 11 and 13 to 15. Adult gene carriers are at high risk and are therefore recommended to have total thyroidectomy with central lymph node dissection (after excluding phaeochromocytoma). As a general guide, child gene carriers of MEN 2B should undergo surgery at an early age; in practice, thyroidectomy is often performed soon after the first year. Children carrying MEN 2A mutations typically have surgery by the age of 3 so that they are well established on thyroxine therapy by school age. The precise timing and extent of surgery is now dictated by the specific type of codon mutation.
Stage classification TNM classification The TNM classification is shown in Table 35.2.
Stage groupings Papillary and follicular tumour subtypes are assigned to stage groupings depending on whether the patient is under age 45 or 45 years and older. All cases involving patients younger than 45 years, regardless of TNM 407
Laura Moss
Table 35.2. The TNM classification of thyroid cancer Anaplastic and Stage
Papillary, follicular and medullary
undifferentiated n/a
T1
≤ 2 cm, intrathyroidal
T2
> 2 to 4 cm, intrathyroidal
n/a
T3
> 4 cm or minimal extension
n/a
T4a
Subcutaneous, larynx, trachea, oesophagus, recurrent laryngeal nerve
Limited to thyroid
T4b
Prevertebral fascia, mediastinal vessels, carotid artery
Beyond thyroid capsule
N1a
Level VI
Level VI
N1b
Other regional (cervical, and upper/superior mediastinal)
Other regional
Adapted from UICC (2002).
Table 35.3. Stage groups for thyroid cancer Papillary and follicular:
Papillary or follicular: age ≥ 45
Anaplastic/undifferentiated
Stage
age < 45 years
years and medullary
(all cases are stage IV)
Stage I
Any T any N M0
T1 N0 M0
−
Stage II
Any T any N M1
T2 N0 M0
−
Stage III
−
T3 N0 M0 or T1–3 N1a M0
−
Stage IVA
−
T1–3 N1b M0 or T4a any N M0
T4a any N M0
Stage IVB
−
T4b, any N, M0
T4b any N M0
Stage IVC
−
Any T any N M1
Any T any N M1
Adapted from UICC (2002).
categories, are stage I or II. All cases of anaplastic or undifferentiated thyroid cancer are stage IV. The stage groupings for thyroid cancer are shown in Table 35.3.
Differentiated thyroid cancer Incidence and epidemiology Differentiated thyroid cancer makes up fewer than 1% of all malignancies but is the most common endocrine malignancy. The UK annual incidence is 2.3 per 100 000 females, 0.9 per 100 000 males. The female-to-male ratio ranges from 1.5:1 to 4:1. In England and Wales, 900 new cases are diagnosed per year, and approximately 250 deaths occur per year in the UK (British Thyroid Association, 2002). Recently, there has been an increase in incidence and decrease in mortality (9% die of their disease). High disease incidence occurs in Hawaii, Iceland, Finland, Israel and Colombia. 408
Risk factors and aetiology Risk factors for disease include the following: r Increasing age and female gender. r History of neck irradiation in childhood (particularly for papillary carcinomas). r Nuclear fallout (age at exposure is important). r Endemic goitre. r Family history or personal history of adenoma. r Gardner’s syndrome. r Cowden’s syndrome. r Familial adenomatous polyposis (FAP). r Familial differentiated thyroid cancer. r Turcot’s syndrome. r Carney complex.
Clinical presentation Possible clinical scenarios at presentation include the following:
Thyroid
r Asymptomatic thyroid nodule or cervical node. r Sense of fullness/pressure in neck. r Stridor, dysphonia, dysphagia, odynophagia, cough (more likely with lymphoma and anaplastic tumours).
r Unexpected finding after thyroidectomy for presumed benign disease.
r Distant metastases – dyspnoea, haemoptysis, bone pain and so forth.
Investigation and staging General investigation Fine needle aspiration cytology (FNAC) is categorised as Thy 1 to 5 (1 = inadequate, 2 = benign, 3 = follicular, 4 = suspicious, 5 = malignant). The malignant poten¨ tial of follicular and Hurthle cell tumours cannot be determined on FNA or frozen section because these investigations cannot demonstrate capsular and vascular invasion. Lymphoma often cannot be reliably diagnosed without incisional biopsy. US of the neck usually shows a hypoechogenic solid lesion. CT/MRI of neck and chest should be carried out to assess local disease extent and to plan treatment. CT/MRI is indicated if there is bulky or fixed neck disease in order to assess retrosternal extension, nodal status, and whether there are pulmonary metastases. It is necessary to avoid iodinated contrast media for approximately 2 months before radioiodine (because it interferes with iodine uptake). Serum calcitonin should be checked if MTC is suspected. Vocal cords should also be examined. Radionuclide imaging is of limited use in the diagnosis of thyroid cancer. Malignant nodules are classically stated to be ‘cold’ but they may also appear as ‘warm’ and ‘hot’ areas. Serum thyroglobulin measurement is of no diagnostic value unless the thyroid has been removed. 18 FDG-PET is not routinely indicated in the diagnostic stages (Intercollegiate Standing Committee on Nuclear Medicine, 2003). Consider imaging after TSH stimulation. 18 FDG-PET-positive metastatic disease is often radioiodine negative. Tetrofosmin, thallium and sestamibi imaging do not require thyroxine withdrawal and they allow immediate imaging, but they lack specificity compared with 131 I.
thyroid cells and its release is TSH dependent. Detection sensitivity is increased when the TSH level is elevated. The serum Tg level is more sensitive than an 131 I whole-body scan in detecting recurrent or metastatic disease. Serum Tg may be undetectable in 20% of cases with isolated lymph node metastases when the patient is on thyroxine. Thyroglobulin autoantibodies interfere with the ability to accurately measure and follow Tg. Autoantibodies are present in 4 to 27% of the general population and in 15 to 30% of thyroid cancer patients. In order to interpret the serum Tg level it is necessary to know the TSH level, recovery and Tg antibody level. Serum Tg is not useful preoperatively as a diagnostic test.
Treatment overview Early cases of differentiated thyroid cancer may potentially be cured with surgery alone, although the majority of cases are usually managed with a combination of surgery, radioiodine and TSH-suppressive doses of thyroxine. External beam radiotherapy may play a role in locally advanced and metastatic disease.
Surgery Surgery is the mainstay of treatment. There is controversy about the extent of surgery required for low-risk cases.
Lobectomy It can be argued that lobectomy is suitable for papillary cancers smaller than 1 cm in females younger than 45 years of age who have node-negative necks, and for follicular cancers smaller than 1 cm with minimally invasive features. The disadvantage of this approach is the difficulty of follow-up, because Tg measurements are inaccurate and imaging shows residual thyroid tissue. There are two reasons proposed to support the lobectomy approach: 1. Total thyroidectomy is associated with increased risks of hypoparathyroidism and recurrent laryngeal nerve damage. 2. This is a good prognosis group of patients and survival is not significantly increased with more ‘radical’ surgery.
Serum thyroglobulin (Tg) Serum thyroglobulin is a glycosylated protein, which is a key substrate for biosynthesis and storage of thyroid hormones. It is secreted by normal and cancerous
Total thyroidectomy Total thyroidectomy is indicated if the tumour is larger than 1 cm, multifocal, with extrathyroidal spread or if 409
Laura Moss
there is familial disease, positive lymph nodes, previous ¨ neck irradiation or Hurthle cell subtype. It could be argued that total thyroidectomy is indicated in all cases because this makes follow-up with Tg easier, papillary cancers are often multifocal, it eliminates the risk of a second thyroid cancer focus developing in the remaining lobe, it reduces the risk of local recurrence, and it makes radioactive iodine easier to administer when a negligible remnant is present. The British Thyroid Association guidelines (British Thyroid Association, 2002) recommend that all nodes within the central compartment (level VI) are removed if there is papillary thyroid cancer. Cervical nodes along the carotid sheath and internal jugular vein (levels II to IV, see Chapter 8, p. 93, for definitions) are palpated and, if the nodes are suspicious, a frozen section may be performed. If the nodes are malignant, selective neck dissection is performed. The rate of recurrent laryngeal nerve injury is 1 to 6% (higher if performed as a re-operation); 30% of patients require postoperative calcium supplementation. After 3 months, the rate of hypocalcaemia is only 2%.
Role of thyroxine The aim is to use a thyroxine dose sufficient to suppress TSH to between less than 0.1 mU/l (in high-risk cases) and 0.5 mU/l. The degree of suppression required has not been tested in prospective studies. It may be possible to relax the degree of TSH suppression in low-risk cases. r Free T4 (FT4) is often above the normal range when significant TSH suppression is achieved. r The average dose of thyroxine required is in the range 150 to 200 μg. The dose of thyroxine varies depending on the patient’s age and weight. r Thyroxine is assumed to reduce the risk of recurrence, tumour progression and death.
Radioisotope therapy Rationale Radioisotope therapy aids in the detection and earlier treatment of persistent/metastatic disease by destroying normal thyroid tissue, destroying microscopic foci of cancer in the thyroid remnant, and aiding in the interpretation of Tg results during follow-up. It may be used for ablation of the thyroid remnant or for treatment of residual or recurrent disease.
410
Indications Postoperative ablation should be performed for r papillary tumours > 1 cm. r papillary tumours with lymph node involvement or thyroid capsule invasion. r follicular tumours > 1 cm that are minimally invasive. r follicular tumours < 1 cm that are widely invasive. r follicular tumours with thyroid capsule invasion or involved lymph nodes. Treatment should also be used if the primary tumour is inoperable or if there is postoperative residual neck disease, distant metastases or recurrent disease.
Aim The aim of radioisotope therapy is to reduce local recurrence and increase survival.
Technique The usual 131 I ablation ‘dose’ is 3.7 GBq (the correct terminology is administered activity). Optimal activity is unknown because conflicting data exist between highand low-ablation activities. 131 I therapy activities are usually in the range of 5.5 to 7.4 GBq. Whole-body radioiodine scans are performed 2 to 10 days after giving radioiodine for ablation or therapy to determine the sites of radioiodine uptake. The timing of the scan depends on the administered activity, the method of patient preparation (rhTSH versus thyroid hormone withdrawal), and the clinician’s choice of the residual activity at which to image the patient. If imaging/therapy can be performed approximately 4 weeks after surgery, there is no need for thyroid hormone supplementation. If this is not possible, then start tri-iodothyronine 20 μg t.d.s. and withdraw 14 days before radioiodine use to allow a sufficient rise in TSH (> 30 mU/l). Elevated TSH level stimulates the sodium iodide symporter (membrane protein involved in active transport) and, hence, radioiodine uptake into thyrocytes. An alternative method is to administer rhTSH as an exogenous source of TSH. A postablation or posttherapy scan 4 to 10 days after radioiodine gives useful information about disease extent. Restart thyroxine administration 2 to 3 days after high-dose radioiodine administration. If the patient has been withdrawn from hormones, thyroxine can be started 2 to 5 days after radioiodine has been administered. If the patient has been given rhTSH, he or she should stay on thyroxine throughout.
Thyroid
Radioiodine side effects Side effects include neck discomfort and swelling (which are rare unless there is a large thyroid remnant), altered sense of taste, nausea (vomiting is uncommon), sialoadenitis, radiation cystitis, gastritis and bleeding/oedema in metastases. Male fertility may be affected; a rise in FSH is seen. High cumulative doses may reduce fertility and so sperm storage should be considered for high-risk cases. No significant difference has been found for female fertility rate, birth weights or prematurity rates. There is a slight increase in miscarriage rate in the first year after treatment. If there are miliary-type pulmonary metastases, the patient may develop pulmonary fibrosis. The risk of leukaemia and second cancer (salivary gland, breast, bladder, colon) is 0.5% (risk is highest with high cumulative dose, i.e. greater than 18.5 GBq, and after external beam radiotherapy; British Thyroid Association, 2002). Consider the use of high-dose corticosteroids before radioiodine if there is bulky neck disease or metastatic disease.
Radiation protection issues Radiation protection issues are as follow: r Exclude pregnancy and lactation before administration. r The patient should avoid pregnancy/conception for 6 months. r Visiting is restricted to non-pregnant adults. Visitors must not enter the patient’s room, must stay in a designated area, and may need to communicate by an intercom system. r Clothing must be washed separately after the patient returns home, unless it is heavily soiled, in which case onsite storage or disposal may be needed. r The patient should double flush the toilet; use separate cutlery and crockery. r The patient must sleep alone. r Restrict time and extend distance with personal contacts. r The duration of restriction is individualised for each patient. The timing of return to work depends on the type of work and work personnel involved.
External beam radiotherapy External beam radiotherapy is infrequently used. The main indications for its use are unresectable disease, non-iodine-avid disease, gross local invasion with
macro- or microscopic residual, recurrent neck disease not amenable to surgery, and palliation of inoperable metastatic disease. The treatment volume usually includes the thyroid bed, cervical and supraclavicular nodes, and the superior mediastinum. Intensity-modulated radiation therapy (IMRT) may allow a better dose distribution. It is possible that radiotherapy may reduce the uptake of radioiodine into residual thyroid tissue; therefore, consider giving radioiodine therapy before external beam radiotherapy.
Radiotherapy technique: thyroid bed only The patient is positioned using an immobilisation shell, supine, usually with chin extended. The PTV is U-shaped and surrounds the spinal cord. There is no ideal technique and there may be a need to compromise on the PTV dose to avoid exceeding the spinal cord tolerance if IMRT is not available. The technique is appropriate for (a) tumours confined to the thyroid bed (i.e. well-differentiated tumours that do not concentrate iodine) and (b) inoperable local recurrences. The field arrangement for the planned dose is obtained using an anterior oblique wedged pair (or three-field plan with anterior and paired anterior oblique wedged fields). The maximum spinal cord dose acceptable for a 20 cm length of cord is generally 46 Gy in 2 Gy fractions. The dose schedule is as follows: r Weighting is 100% for each beam. r Wedges are 30◦ for each beam. r Typical field size: 7 × 7 cm. r Phase 1 only – 60 Gy in 30 fractions over 6 weeks. r Energy: 6 MV photons.
Radiotherapy technique for thyroid bed and locoregional nodes including superior mediastinal nodes (more common) The patient is positioned using an immobilisation shell, supine, usually with chin extended. The CT planned volume should encompass the thyroid bed plus cervical, supraclavicular fossa and upper mediastinal nodes. Mandibular and apical lung shielding should be used. The PTV generally encompasses the area from the tip of the mastoid processes to the carina. However, if level II nodes do not need to be irradiated, the superior border may lie at the level of the hyoid, thereby reducing radiation dose to the parotids and risk of xerostomia.
411
Laura Moss
The field arrangement typically involves anterior and posterior parallel opposed fields to cover the PTV. Because the volume lies anteriorly in the neck and in the midplane in the mediastinum, consider using a wedge in the superior–inferior direction. Take three outlines because of the variation in contour along the volume. IMRT can reduce the dose to radiosensitive organs close to the PTV and may improve the PTV coverage; it reduces the volume of non-target tissue irradiated to greater than 90% of the prescribed dose but increases the volume of normal tissue irradiated to a low dose. The maximum spinal cord tolerance dose for a 20 cm length of spinal cord is generally 46 Gy in 2 Gy per fraction. The dose schedule is as follows: r Phase 1 – 46 Gy in 23 fractions over 4.5 weeks (maximum allowable spinal cord dose for 20 cm length). r Phase 2 – 14 Gy in 7 fractions over 1.5 weeks to the thyroid bed alone. r Total dose – 60 Gy in 30 fractions over 6 weeks. r Field arrangement – use a CT planned field arrangement. r Energy – 6 MV photons.
Indications for the use of rhTSH include hypopituitary function, functional metastases causing TSH suppression, severe IHD, a previous history of psychiatric disturbance precipitated by hypothyroidism, advanced disease or frailty. Side effects of rhTSH include flu-like myalgia, mild nausea and headache. There is possible stimulation of metastases resulting in local symptoms, and so the use of prophylactic corticosteroids before use of rhTSH should be considered.
Pregnancy It is essential to consider the risk to both mother and foetus. Many cases of differentiated thyroid cancer are not aggressive and therefore it is reasonable to allow the pregnancy to continue. Thyroidectomy in the first trimester is associated with a very high risk of abortion. Suppressive thyroxine may be considered until surgery is possible. The patient should avoid pregnancy for 6 months after receiving radioactive iodine (ARSAC, 2006). There may be an increased risk of miscarriage in the first year following radioactive iodine treatment.
Children
Chemotherapy Chemotherapy is not routinely used in the treatment of thyroid carcinoma. Its use is restricted to symptomatic progressive disease when surgery, radiotherapy and radioactive iodine have failed. Doxorubicin is the most frequently used drug, with a reported partial response rate of approximately 20 to 30%. There is no clear evidence that its use increases survival. It has been recently reported that response rates to chemotherapy increase in the presence of elevated TSH level (Santini et al., 2002).
Special clinical situations Use of recombinant human TSH (rhTSH) Injections of rhTSH (0.9 mg deep i.m. injection into buttock on days 1 and 2, radioiodine administered day 3, and thyroglobulin measured on day 5 when maximal response is seen) can be used as an alternative to thyroid hormone withdrawal. Currently, rhTSH is licensed for use before remnant ablation with 3.7 GBq 131 I in lowrisk cases, for diagnostic scans and for the assessment of serum Tg level during follow-up.
412
Papillary carcinoma is the most common form of thyroid cancer in children, with 30 to 40% of tumours multifocal; 40 to 90% are found to have involved cervical nodes at initial surgery (cf. 20% adults). At presentation, 10 to 20% have lung metastases; bone metastases are rare (< 1%). The recurrence rate of thyroid cancer is higher in children than it is in adults, especially young children. Fewer than 10% die of their disease.
Treatment for recurrent disease Early detection can lead to a cure or long-term survival, although one-third to one-half of patients who develop recurrence die of thyroid cancer.
Local recurrence in thyroid Debulking may be beneficial even if complete resection of the tumour is not possible. EBRT may also be beneficial: consider prolonged high-dose fractionation rather than the use of single fractions.
Metastatic disease Treatment depends on the site of the tumour: if lung and other soft tissue sites are not amenable to surgery then treat with 131 I 3.7 to 7.4 GBq at 3- to 12-month
Thyroid
intervals. There is no maximum cumulative dose but FBC and renal function should be monitored. The risk of a second malignancy increases with cumulative dose. Individual doses should not exceed 200 rem total-body exposure. For solitary metastases that do not concentrate radioiodine and are suitable for surgical resection, the 5-year postmetastectomy survival is 40 to 50%.
Rising thyroglobulin with negative diagnostic radioiodine scan An increase in Tg with a negative diagnostic radioiodine scan is a relatively common situation and there is controversy about optimal management. There are two possible explanations for this situation: (1) a small volume of thyroid cancer cells does not take up enough radioiodine to be shown on gamma camera images, or (2) thyroid cancer cells de-differentiate and are unable to take up radioiodine. Ensure that the imaging result is a true negative (i.e. optimal TSH level, no amiodarone, no iodinated CT contrast media, no high-iodine-content diet) and that the Tg result is a true positive. There are two main management options: First, use cross-sectional imaging and radioisotope functional imaging studies to find sites of disease. If an isolated lesion amenable to surgery or EBRT is found then this and RAI therapy can be considered. US/MRI of the neck and mediastinum should be performed. If negative, then CT of the lungs should be performed. If the CT is negative, an isotope bone scan should be performed. If the bone scan is negative, consider 18 FDG-PET imaging. Also, consider TSH stimulation before PET imaging because there is some evidence that suggests this increased sensitivity. Thallium, tetrofosmin and sestamibi imaging may also performed (these are nonspecific tracers). With regard to 111 In-labelled octreotide or other somatostatin analogue imaging, a small percentage of cases may have significant uptake and may be suitable for 90 Y or other radiolabelled somatostatin analogue therapy. The second management option is empirical highdose radioiodine therapy. Proponents of this procedure argue that a high proportion of patients will have positive postablation scans and/or Tg response and some may achieve a cure. Opponents argue that some of these patients have minimal disease and, hence, radioiodine is unlikely to improve survival and that treatment is associated with acute toxicity.
Follow-up The frequency and type of follow-up depend on an individual’s risk of recurrence (Mazzaferri et al., 2003; Schlumberger et al., 2004; Thyroid Carcinoma Task Force, 2001). The aims of follow-up are to detect recurrence early, to monitor TSH suppression and to detect and manage hypocalcaemia. The duration of followup should be life-long (long natural history, late recurrences, late side effects of radioactive iodine, consequences of supraphysiological thyroxine replacement).
Prognosis Of patients with well-differentiated thyroid cancer, 5 to 20% develop distant metastases (lung > bone > liver and brain). Older age is associated with worse outcomes. The most important predictors are the patient’s age; the tumour’s size, grade and extrathyroidal spread; and distant metastases. For children, however, the outcome is worse when the child is younger than 10 years because the disease behaves more aggressively and the risk of recurrence is higher. Recurrence is also more likely if the tumour is a tall cell/columnar/diffuse sclerosing subtype, is poorly differentiated, is large, or there is lymph node involvement. There are many prognostic scoring systems in use for well-differentiated thyroid cancer, for example, AGES (PTC), AMES (PTC and FTC), MACIS (PTC), Ohio State University (differentiated), and the University of Chicago (PTC). Approximately 75% of all patients with DTC would be classified as low risk. For example, using the AMES scoring system, which allocates patients to either high or low risk based on age, distant metastases and the extent and size of the primary, the mortality rates at 20 years are 39.5 and 1.2%, respectively.
Areas of current interest/controversy There is currently no prospective randomised trial involving the extent of initial surgery. Advocates for lobectomy in low-risk cases cite both increased morbidity with total thyroidectomy and low risk of death as reasons to support less radical surgery. The arguments for total thyroidectomy followed by thyroid remnant ablation include the high incidence of multifocal papillary cancer, a reduced risk of local recurrence, the possible improvement in overall survival, and easier
413
Laura Moss
follow-up using Tg. The risk of morbidity with total thyroidectomy performed by experienced surgeons is low. Other areas of current interest include the following: r Indications for routine radioiodine remnant ablation. r ‘Stunning’ – there is concern that higher 131 Iadministered activity for diagnostic scanning may diminish subsequent uptake of therapeutic 131 I by thyroid tissue or functioning metastases. Some use 123 I or 99m Tc for imaging to reduce this risk whereas others omit the uptake scan and ablate all cases with an empirical therapeutic activity. r The role of adjuvant EBRT. r Administered activity of radioactive iodine – optimal activity of 131 I to achieve remnant ablation (high versus low dose), empirical versus individualised dosing. r Management of patients with scan-negative Tgpositive differentiated thyroid cancer. r Lymph node dissection. r Degree of TSH suppression.
Medullary thyroid cancer Incidence and epidemiology The sporadic form of medullary thyroid cancer is most common in the fifth and sixth decades of life. The female-to-male incidence ratio is 1.5:1. There are different familial forms. In association with multiple endocrine neoplasia (MEN) type 2A, the disease presents in the first and second decades and is associated with parathyroid hyperplasia and phaeochromocytoma. In association with MEN type 2B, the disease presents in the first decade and is aggressive. It is associated with phaeochromocytoma, mucosal neuromas (especially of the lip and tongue), Marfanoid habitus, and a high-arched palate. Familial non-MEN MTC patients present in the sixth decade and beyond. Familial forms of medullary thyroid cancer have equal gender frequency.
Clinical presentation Possible clinical presentations include the following: r Detection during screening. r Typically with thyroid mass and lymphadenopathy. r Wheezing, secretory diarrhoea and flushing are involved with large-volume local or metastatic MTC. It is important to document the patient’s family history to ascertain whether there are any features to suggest MTC or MEN, including any history of sudden unexpected deaths. Specific investigations for MTC include the following: r Calcitonin should be measured preoperatively and during follow-up (it may be raised with chronic renal failure, pregnancy, pernicious anaemia, and in the neonatal period). r In some cases CEA is a useful marker. r Imaging with pentavalent 99m Tc DMSA, somatostatin analogues like octreotide or MIBG can be used in the assessment of MTC. r All new patients should be screened biochemically for phaeochromocytoma and hyperparathyroidism. A lack of family history does not exclude heritable disease because the disease can skip generations.
Treatment overview Surgery is the mainstay of treatment. Debate continues about the role of external beam radiotherapy in locally advanced disease. There is no role for radioiodine therapy. Thyroxine is required as a replacement dose only.
Surgery Surgery entails total thyroidectomy and central lymph node dissection (level VI). The lateral jugular nodes should be sampled; if they are positive, proceed to modified radical or selective neck dissection.
Role of thyroxine Risk factors and aetiology MEN types 2A and 2B are associated with MTC. They are inherited in an autosomal dominant fashion and associated with mutations in the RET proto-oncogene.
Spread MTC may spread to lymph nodes and via the blood stream. 414
Thyroxine is required as a replacement dose only.
External beam radiotherapy The role of postoperative RT is debated. (Some studies report improved local control whereas others report no difference and even a detrimental effect, probably because of selection bias.) Consider postoperative RT if there is locally advanced disease, multiple involved
Thyroid
lymph nodes, elevated postoperative calcitonin level, and bulky inoperable tumour.
Treatment of recurrent disease
Areas of current interest/controversy Areas of current interest include the optimal management of calcitonin positive cases and the role of postoperative radiotherapy.
Recurrent cervical nodal disease Recurrent nodal disease may be amenable to surgical excision.
Metastatic disease
Anaplastic thyroid cancer Incidence and epidemiology
Patients may experience frequent loose bowel actions, wheezing and flushing. The symptoms may respond to somatostatin analogue (e.g. octreotide) therapy. Chemotherapy is rarely helpful unless there is rapidly progressive symptomatic disease.
The incidence of anaplastic thyroid cancer is decreasing. Disease occurrence peaks in the seventh decade of life. The female-to-male incidence ratio is 1.5:1.
Raised calcitonin level
For anaplastic carcinoma, spread is predominantly local, to form a large bulky mass.
If the patient has a raised calcitonin level, imaging techniques may not be able to detect the site of recurrent disease until the serum calcitonin level is grossly elevated. Options for imaging include pentavalent DMSA, PET imaging (some advocate using DOPA as the tracer rather than 18 FDG), MIBG and octreotide. Some advocate laparoscopy to assess the liver surface because it is a common site of small superficial metastases. If significant 123 I MIBG or 111 In octreotide uptake is seen, high-dose radiolabelled therapies may be appropriate.
Follow-up Monitor calcitonin (and CEA in some cases). An elevated calcitonin level during follow-up often signals recurrence but it is often not possible to locate the site. Many advocate observation, with surgery being reserved until the site of recurrence is clinically detected rather than proceeding directly to further neck surgery (because further surgery is associated with a higher morbidity rate and untreated patients often survive long term), selective venous sampling, or routine annual imaging (both cross-sectional and functional/radioisotope modalities).
Spread
Treatment overview Patients are rarely suitable for surgery. Surgery carries a poor prognosis and the main decision is whether palliation is best achieved with symptom-control measures only or with the addition of radiotherapy.
Surgery Surgery is rarely possible. However, it may be indicated if there is small-volume disease or for airway maintenance in some cases.
External beam radiotherapy Anaplastic thyroid cancer is the least radioresponsive thyroid cancer type. A dose schedule of 50 to 60 Gy over 5 to 6 weeks achieves a response in fewer than 50% of patients. Concurrent chemoradiotherapy (with hyperfractionation) has been shown to improve patient survival in some studies but only in the short term and with significant morbidity.
Prognosis
Prognosis
Young age, male gender, positive nodes and incomplete initial surgery are adverse factors. The 10-year survival in patients presenting with clinical disease rather than screened disease is 65% overall (stage I = 100%, stage II = 93%, stage III = 71% and stage IV = 20%).
The median survival is 6 months from development of symptoms; 75% of patients still die from local progression. In practice the patient often spends much of their remaining life receiving treatment and recovering from acute radiation toxicity. 415
Laura Moss
Thyroid lymphoma Most thyroid lymphomas are either diffuse large B-cell lymphoma (‘high grade’) or MALT lymphoma (‘low grade’). Hashimoto’s thyroiditis is associated with thyroid MALT lymphoma.
rior fields from the mastoid tip to the carina. The oral cavity and lung apices should be shielded. The dose is 24 to 40 Gy, 2 Gy per fraction, in 12 to 20 fractions over 2.5 to 4 weeks. For high-grade NHL (e.g. diffuse large B-cell subtype), systemic chemotherapy (CHOP or R-CHOP) is given before involved field radiotherapy.
Incidence and epidemiology The mean age of thyroid lymphoma occurrence is 60 to 65 years. The disease predominates in women; the female-to-male incidence ratio is 3:1 (ratio increases above 60 years).
Prognosis The prognosis depends on the stage of disease. The majority of patients present with localised disease and have a favourable prognosis.
Clinical presentation
Current clinical trials
Possible clinical scenarios at presentation include the following: r The patient typically presents with thyroid mass ± lymphadenopathy. r B symptoms (although these are rare in thyroid lymphoma). See Chapter 31, p. 349, for the staging system. Most patients present with localised disease.
At the time of writing there was one clinical trial about to open. HiLo is a four-arm multicentre randomised phase III trial comparing high- (3.7 GBq) versus low- (1.1 GBq) administered activity of 131 I and thyroid hormone withdrawal versus recombinant TSH in differentiated thyroid cancer.
Treatment Overview Low-grade lymphoma is treated with radiotherapy alone; high-grade lymphoma is treated with a combination of chemotherapy and involved field radiotherapy.
Surgery Thyroidectomy is not indicated.
Chemotherapy Giving three cycles of CHOP plus involved field radiotherapy is standard therapy for high-grade lymphoma. For bulky disease up to six cycles might be given. RCHOP is often used; although there is no RCT evidence for its use in combination with involved field radiotherapy for early stage disease, a survival benefit would be assumed. Radiotherapy alone may, however, be indicated for stage IAE MALT positive lymphoma.
External beam radiotherapy For stage IE or IIE low-grade NHL (e.g. MALT type), the target volume should include the bilateral neck and superior mediastinum via opposed anterior and poste-
416
REFERENCES ARSAC. (2006). Notes for Guidance on the Clinical Administration of Radiopharmaceuticals and Use of Sealed Radioactive Sources. Chilton: NRPB. British Thyroid Association. (2002). Guidelines for the Management of Thyroid Cancer in Adults. London: Royal College of Physicians (2nd edition due 2007). Intercollegiate Standing Committee on Nuclear Medicine. (2003). Positron Emission Tomography. A Strategy for Provision in the UK. London: Royal College of Physicians. Mazzaferri, E., Robbins, R., Spencer, C. et al. (2003). A consensus report of the role of serum thyroglobulin as a monitoring method for low-risk patients with papillary thyroid carcinoma. J. Clin. Endocrinol. Metab., 88, 1433–41. Santini, F., Bottici, V., Elisei, R. et al. (2002). Cytotoxic effects of carboplatinum and epirubicin in the setting of elevated serum thyrotropin for advanced poorly differentiated thyroid cancer. J. Clin. Endocrinol. Metab., 87, 4160–5. Schlumberger, M., Berg, G., Cohen, O. et al. (2004). Follow-up of low-risk patients with differentiated thyroid carcinoma: a European perspective. Eur. J. Endocrinol., 150, 105–12. Thyroid Carcinoma Task Force. (2001). AACE/AAES medical/surgical guidelines for clinical practice: management of thyroid carcinoma. Endocr. Pract., 7, 202–20. UICC. (2002). TNM Classification of Malignant Tumours, ed. L. H. Sobin and Ch. Wittekind, 6th edn. New York: Wiley-Liss, pp. 52–6.
Thyroid
FURTHER READING Haq, M. and Harmer, C. (2004). Thyroid cancer; an overview. Nucl. Med. Comm., 25, 861–7. Harmer, C., Bidmead, M., Shepherd, S. et al. (1998). Radiotherapy planning techniques for thyroid cancer. Br. J. Radiol., 71, 1069–75. Mazzaferri, E. L., Ernest, L., Harmer, C. et al. (2006). An overview of the management of thyroid cancer. In Practical Management
of Thyroid Cancer – a Multidisciplinary Approach. New York: Springer, Chap.1. Schlumberger, M. J. (1998). Papillary and follicular thyroid carcinoma. N. Engl. J. Med., 338, 297–306. Sherman, S. I. (2003). Thyroid carcinoma. Lancet, 361, 501–11. Vini, L. and Harmer, C. (2002). Management of thyroid carcinoma. Lancet Oncol., 3, 407–14.
417
36
NEUROENDOCRINE TUMOURS Atul Kalhan and Aled Rees
Introduction Neuroendocrine tumours (NETs) constitute a heterogeneous group of neoplasms with significant variation in their mode of presentation and biological behaviour. They arise from neuroendocrine cells, which are widely distributed in the body; the spectrum of tumours that fall under this classification is accordingly diverse. Although tumours can arise within endocrine glands such as the pituitary and the parathyroids, tumours at these sites are typically benign, with limited growth potential, and are traditionally managed by endocrinologists. This chapter therefore focuses largely on NETs arising within the bronchial or gastroenteropancreatic systems, historically termed carcinoid tumours. Management of these rare tumours is improving largely due to advances in imaging (especially nuclear imaging) and biochemistry (notably chromogranin A) and to the increased use of multidisciplinary teams in specialist centres (Kaltsas et al., 2004; Ramage et al., 2005). Treatment options are diverse and should be tailored individually but they may include one or more of surgery, medical therapy (somatostatin analogues, interferon alpha), chemotherapy, radionuclide therapy, and embolisation. Current clinical trials are focused on defining the optimal use of existing therapies and exploring novel agents such as angiogenesis and mTOR inhibitors.
Tumour types NETs are classified according to (1) site of origin, (2) histological grade and (3) tumour stage. The site of origin is traditionally divided along embryological lines into foregut tumours (bronchus, thymus, stomach, pancreas, proximal duodenum), midgut tumours (distal duodenum, jejunum, ileum, appendix, proximal and transverse colon) and hindgut tumours (distal colon, rectum). The histological tumour grade is classified according to recently revised WHO criteria into four groups: 418
r Well-differentiated endocrine tumours of probable benign behaviour.
r Well-differentiated endocrine tumours of uncertain behaviour.
r Well-differentiated endocrine carcinomas. r Poorly differentiated endocrine carcinomas. The stage of the tumour can be localised, localised with regional (lymph node) involvement or occur with distant metastases.
Incidence and epidemiology The UK annual disease incidence is 3 in 100 000 patients, with a slight female predominance. Disease incidence is thought to be increasing overall, probably due to increased recognition of symptoms (Modlin et al., 2003). Autopsy studies indicate that small, clinically unrecognised NETs are relatively common (up to 10% for pancreatic NETs). There is no known geographical variation in tumour incidence. The average age at diagnosis is 61 years.
Risk factors and aetiology The risk factors for NETs are poorly understood. Most are sporadic but there is a small familial risk (4-fold increased risk with one affected first-degree relative, 12-fold with two affected first-degree relatives). A small proportion of NETs arise on a background of an inherited cancer syndrome, for example: r Multiple endocrine neoplasia type 1 (MEN1) – parathyroid, pituitary and pancreatic islet cell tumours, foregut carcinoids. r Multiple endocrine neoplasia type 2 (MEN2) – hyperparathyroidism, medullary thyroid carcinoma (MTC), phaeochromocytoma. MTC is discussed in Chapter 35 (p. 414). r Type 1 neurofibromatosis (NF1) – neurofibromas, caf´e-au-lait macules, optic glioma, phaeochromocytoma, and rarely duodenal somatostatinoma.
Neuroendocrine tumours
r Von Hippel Lindau (VHL) – renal cell carcinoma, phaeochromocytoma, cerebellar haemangioblastoma, retinal angioma, and renal and pancreatic cysts. r Carney complex – spotty skin pigmentation, cardiac myxomas, thyroid adenoma, nodular adrenocortical disease causing Cushing’s syndrome, sertoli cell tumours and ovarian cysts. The incidence of MEN1 in foregut NETs varies according to the tumour site: bronchus, 0 to 8%; stomach, 7 to 30%; insulinoma, 5%; and gastrinoma, 25%. A family history should be obtained and a clinical examination undertaken to exclude an inherited cancer syndrome in all cases of NET.
Table 36.1. Incidence of metastases at presentation according to primary site of NET Subsite and incidence of metastases Site/type Foregut
at presentation Bronchial well-differentiated < 15% Bronchial ‘atypical’ NET 30–50% Thymus 80% Stomach 80% (for sporadic type III)
Midgut
Duodenal 45% Small intestine 60–80% Appendix < 2 cm rare Appendix > 2 cm 30% Ileocaecal 70%
Pathology and spread
Hindgut
Pathology Macroscopically, NETs are typically solid and yellow in appearance, reflecting a high lipid content. Microscopically, NETs are often trabecular, glandular, or form rosettes. The tumour cells typically have a granular cytoplasm and round nuclei, and the majority display low proliferative potential with infrequent mitoses. However, a proportion of NETs are poorly differentiated. The neuroendocrine origin of tumours is confirmed via immunohistochemistry directed against a panel of general neuroendocrine markers, such as chromogranin A, synaptophysin and PGP9.5. The specific immunohistochemical analysis of hormone production (e.g. glucagon for glucagonomas, ACTH where ectopic ACTH production is suspected) depends on the tumour site and clinical presentation. All histological analysis should give an estimate of proliferative potential not only by mitotic rate but also by using an antibody to Ki-67 to generate a Ki-67 index. This procedure has prognostic relevance, especially in pancreatic tumours, and may influence choice of therapy (e.g. chemotherapy in tumours with a high Ki-67 index).
Spread NETs may be localised to the primary organ of origin, may invade regional lymph nodes, or may spread distantly to the liver, lungs or to bone. The tumour stage at presentation is dependent on the primary site (Table 36.1). When present in gastrointestinal NETs, bony metastases are generally a feature of advanced disease.
Rectal < 1 cm rare Rectal > 2 cm majority
Insulinoma
< 10%
Gastrinoma
70–80%
VIPoma
50–60%
Glucagonoma
80%
Somatostatinoma
60–70%
Nonfunctioning
60–80%
pancreatic
Clinical presentation In keeping with the widespread distribution of these tumours, their modes of presentation vary considerably. Many of them are asymptomatic and are discovered incidentally as a part of investigations performed for other reasons. Some patients present with symptoms suggestive of bowel obstruction even if abdominal radiology is normal.
Carcinoid syndrome Classical carcinoid syndrome is a presenting feature in fewer than 10% of patients and it occurs when vasogenic amines and peptides including serotonin gain access to the systemic circulation. This is usually seen in the context of midgut NETs with liver metastases, though similar syndromes can occasionally occur in the absence of measurable peptide hormones or their products. The features of carcinoid syndrome include flushing (90%), diarrhoea (70%), abdominal pain (40%), wheezing (25%), valvular heart disease (40%) and pellagra (5%). 419
Atul Kalhan and Aled Rees
The carcinoid flush is usually pink, lasts for a few minutes, and involves the face and upper trunk. Some patients are able to identify triggers such as alcohol, bananas, walnuts or chocolate. Atypical carcinoid syndrome is seen in a small proportion of patients with foregut NETs and consists of a protracted purplish flush leaving cutaneous telangiectasia in the face, upper trunk and limbs. Carcinoid crisis is an extreme presentation characterised by fluctuating blood pressure, tachycardia, arrhythmias, profound flushing and bronchospasm. It may be precipitated by tumour lysis (from embolisation, chemotherapy or radionuclide therapy), tumour handling or anaesthesia. Carcinoid heart disease classically affects the right side of the heart in patients with carcinoid syndrome; tricuspid regurgitation is the most common valvular lesion. In severe cases, valve replacement surgery may be necessary if disease elsewhere is controllable.
Table 36.2. Incidence and clinical features of ‘functioning’ pancreatic NETs Tumour type
Incidence
Insulinoma
1–2 per million
Symptoms Confusion Sweating Dizziness Relief with eating Weight gain
Gastrinoma
1 per million
Peptic ulceration Diarrhoea
Glucagonoma
1 per million
Diabetes Weight loss Necrolytic migratory erythema
VIPoma
1 per 5 million
Watery diarrhoea Marked hypokalaemia
Other presentations Pulmonary NETs present with symptoms suggestive of bronchial obstruction (pneumonia, dyspnoea, pleuritic pain), cough or haemoptysis. Many of these are discovered incidentally on chest X-ray (Fink et al., 2001). Foregut NETs may also cause ectopic hormone production (e.g. Cushing’s syndrome from ectopic ACTH secretion, acromegaly from ectopic growth hormone releasing hormone production or syndrome of inappropriate antidiuretic hormone secretion). Pancreatic NETs present with either a hypersecretory syndrome or with symptoms directly related to the tumour mass (abdominal pain, weight loss) or to metastases. Their clinical features are shown in Table 36.2.
Investigation and staging The diagnosis and staging of NETs are based on a combination of the following: r Clinical symptoms where present. r Biochemistry (general and specific neuroendocrine/ hormone markers). r Radiological and nuclear imaging. r Histological confirmation (the gold standard and mandatory whenever possible).
Biochemistry Biochemical tests comprise both general neuroendocrine markers and specific hormones for the underly420
Somatostatinoma
1 per 10 million
Steatorrhoea Cholelithiasis Weight loss Diabetes
ing tumour type. They are useful not only in establishing the diagnosis but also as a guide for determining prognosis and response to therapy. Plasma chromogranin A (CgA) is a useful general tumour marker (Eriksson et al., 2000). It retains high sensitivity in all types of NET with the highest levels often seen in metastatic midgut carcinoid tumours. Renal impairment and hypergastrinaemia are important causes of false-positive CgA elevation. 24-hour urinary 5-hydroxyindoleacetic acid (5HIAA): many midgut NETs secrete serotonin, and measurement of its metabolite 5-HIAA is useful, with up to 70% sensitivity. The tachykinin neurokinin A may be a good marker for prognosis and in evaluating a response to treatment in midgut NETs, but levels are not elevated in all patients and analysis is currently available in research laboratories only (Turner et al., 2006). Specific biochemical markers may also be useful in diagnosis depending on clinical presentation (e.g. hCG levels for hindgut NETs) and index of suspicion for inherited disease (e.g. prolactin, calcium and PTH where MEN1 is suspected).
Neuroendocrine tumours
A fasting gut hormone profile should complement plasma CgA measurement in suspected pancreatic NETs. Blood should be collected in a lithium heparin bottle containing trasylol, spun and frozen prior to subsequent analysis at one of currently two reference laboratories in the UK (Peptide Laboratory, Hammersmith Hospital, London; Regional Regulatory Peptide Laboratory, Royal Victoria Hospital, Belfast). To avoid false-positive gastrin elevation, patients should discontinue proton-pump inhibitors (for at least 2 weeks) and H2 blockers (for at least 3 days) prior to testing. Patients with suspected insulinoma require a supervised inpatient fast in a specialist endocrine unit.
Radiological and nuclear imaging A number of imaging modalities are employed in the evaluation of NETs. Determining an appropriate imaging strategy in an individual patient requires close liaison with radiologists and nuclear medicine physicians but is partly dependent on whether the imaging is undertaken for the initial detection of the disease in suspected cases or in determining the tumour extent where neuroendocrine disease is already confirmed. Somatostatin receptor scintigraphy (SSRS; ‘OctreoScan® ’; 111 In-octreotide) has high sensitivity in locating primary NETs and in assessing the extent of metastatic disease (Chiti et al., 2000; Kaltsas et al., 2001). The exception to this high sensitivity is for insulinomas, where sensitivity falls to less than 50%. Modern nuclear imaging centres now use SSRS in combination with single positron emission computed tomography and fusion imaging with CT (Schillaci et al., 1999). Demonstration of clear uptake on SSRS also predicts a response to somatostatin analogue therapy and determines suitability for targeted radiolabelled therapy. 123 I-MIBG has significantly lower sensitivity than 111 In-octreotide (Kaltsas et al., 2001) but may be useful where targeted radiolabelled MIBG therapy is being considered. CT scans are useful in identifying bronchial NETs, in assessing liver metastases (where triple-phase scanning enhances sensitivity), and in identifying primary abdominal NETs, especially where SSRS is negative. MRI and CT are both useful in imaging pancreatic NETs though EUS is particularly sensitive, especially in identifying lesions within the pancreatic head where it can also enable fine needle aspiration/biopsy for cytological/histological diagnosis (Anderson et al., 2000).
Studies examining the value of PET scanning in neuroendocrine disease are still in progress, although its widespread use is currently limited by scanner availability. However, 18 FDG (fluorodeoxyglucose)-PET is often negative in indolent disease, though where uptake is present this may correlate with a poorer prognosis. There is limited experience with other tracers at present, but 18 F-DOPA and 11 C tryptophan are under evaluation. Selective angiography with secretagogue (calcium) injection may be particularly useful in localising small gastrinomas or insulinomas, especially where other radiology has been normal. Direct palpation at laparotomy may occasionally be necessary in small functioning pancreatic NETs where imaging has been unhelpful. It is often combined with intraoperative ultrasound.
Treatment Treatment overview NETs are rare; consequently, randomised trial data to form a robust evidence base with which to guide management are limited. Few individual units have developed sufficient experience to evaluate this diverse range of tumours confidently and all cases should thus be managed in the context of a multidisciplinary team whose members vary according to local expertise and interests but should comprise representation from endocrinology, gastroenterology, surgery, oncology, radiology/nuclear medicine and histopathology. This facilitates accurate diagnosis and staging, develops consensus agreement on management, ensures individualised treatment planning, and improves interdisciplinary education. Increasingly, such groups should form part of a larger network of clinicians with interests in these tumours at national (UKNETwork) and international (ENET) levels, thereby facilitating multicentre studies and the development of consensus guidelines. Treatment should aim to cure the patient if possible but is often palliative because most patients have evidence of metastases at presentation. In patients with incurable disease, treatment should aim to maintain a good quality of life for as long as possible, recognising that this can be achieved for a number of years in many NETs, even in the presence of metastatic disease.
Surgery Surgery is currently the only treatment modality available that can achieve a cure. Although metastatic disease 421
Atul Kalhan and Aled Rees
at presentation frequently precludes a cure, surgery may still have a role in debulking the primary tumour and minimising the risk of intestinal obstruction. It should be recognised, however, that marked fibrosis around the tumour (a desmoplastic reaction) may make surgical resection technically challenging and can run the risk of vascular compromise in the small bowel if it is undertaken by an inexperienced surgeon. Preoperatively, patients with known functioning midgut NETs should be carefully assessed for the presence of carcinoid heart disease and should be treated in the pre-, peri- and postoperative periods with an intravenous infusion of octreotide (50 μg/hour) to minimise the risk of carcinoid crisis. Patients with pulmonary NETs should undergo lung or wedge resection with lymph node dissection. The surgical management of gastroenteropancreatic NETs is dependent on the primary site, mode of presentation (elective versus emergency), and extent of disease (Sutton et al., 2003). Emergency presentations usually demand a resection sufficient to correct the immediate problem but may need to be followed by a further, more definitive procedure, for example in patients presenting with appendicitis where prognosis can be improved by performing a subsequent right hemicolectomy in appendiceal tumours greater than 2 cm in size or in lesions 1 to 2 cm where the macro- or microscopic appearances are unfavourable. Small intestinal and colorectal NETs should be resected together with extensive locoregional lymph node dissection. This procedure should be considered even in patients with metastatic disease because it may improve the prognosis in addition to minimising the risk of future intestinal obstruction. Gastric NETs should be managed according to their type: type 1 (associated with chronic atrophic gastritis) and type 2 (associated with Zollinger-Ellison syndrome/MEN1) carcinoids occur in patients with hypergastrinaemia and require medical therapy (e.g. with proton-pump inhibitors for Zollinger-Ellison syndrome), perhaps with additional endoscopic surveillance in type 1 tumours. Type 3 gastric carcinoids occur sporadically, are more aggressive, and frequently show evidence of metastasis at presentation. Most require gastrectomy and regional lymph node clearance. Surgery for pancreatic NETs should only be performed by hepatobiliary specialists and the extent of disease can vary from enucleation only for superficial or easily localised insulinomas (which have low malig422
nant potential) to distal or total pancreatectomy or even pancreatoduodenectomy in more extensive disease. Liver resection of unilobar liver metastases may prolong survival and can additionally offer symptomatic relief in carefully selected cases in which a single hepatic lesion is dominant. Whenever abdominal surgery is contemplated for NETs, consideration should always be given to cholecystectomy in order to prevent gallstone formation in patients who subsequently commence somatostatin analogues.
Drug therapy: somatostatin analogue therapy Somatostatin analogue therapy forms the mainstay of symptomatic control in gastroenteropancreatic NETs. Endogenous somatostatin has a very short circulatory half-life, making it unsuitable for use. Octreotide has a half-life of several hours and is usually administered subcutaneously in doses of 50 to 100 μg, two to three times daily. More recently sustained-release preparations have become available, allowing dosing every 2 to 4 weeks. There are three drugs in routine use, namely Lanreotide (fortnightly injection), Lanreotide Autogel® (monthly), and Sandostatin Lar® (monthly), all of which have been shown to improve quality of life (including reduced frequency of flushing and diarrhoea) with equivalent or improved efficacy compared with short-acting octreotide (Rubin et al., 1999; Tomassetti et al., 1998). Patients are usually started on octreotide for 1 to 2 weeks to assess their tolerance and to enable symptom stabilisation prior to conversion to a long-acting preparation. In addition to good symptomatic control, which occurs in the vast majority of patients, biochemical response (inhibition of hormone production) occurs in 30 to 70% and tumour stabilisation (or rarely even shrinkage) may occur. Patients should be warned of side effects (diarrhoea, abdominal discomfort, flatulence, anorexia and nausea) at the start of therapy. These are usually mild and diminish with continued usage. In addition, patients with diabetes may need to adjust their insulin dosage. Long-term use can result in gallstone formation, though these are rarely symptomatic; hence, ultrasonographic surveillance is no longer considered necessary. Somatostatin analogues are also effective in controlling clinical syndromes associated with unresectable pancreatic NETs, including VIPomas, glucagonomas, and occasionally insulinomas, although 50% of the
Neuroendocrine tumours
latter demonstrate a paradoxical fall in blood glucose levels due to a suppression of counter-regulatory hormones such as glucagon. Proton-pump inhibitors form the mainstay of medical therapy in gastrinomas with no evidence for added benefit from somatostatin analogue therapy.
Drug therapy: interferon Interferon alfa in a dose of 3 to 5 MU, three to five times weekly, is employed in the treatment of NETs, either alone or more commonly in combination with somatostatin analogue therapy (Faiss et al., 2003). However, its use is not widespread at present, in part related to conflicting data as to its efficacy in addition to difficulties with tolerability and high cost. Nevertheless, biochemical and symptomatic response may be seen in 40 to 70% of patients. It may be particularly effective in combination with somatostatin analogues and in tumours which are slowly proliferative.
Drug therapy: chemotherapy The precise role of chemotherapy as well as the most effective chemotherapeutic regime in NETs remain uncertain. It is clear, however, that chemotherapy has little therapeutic value in well-differentiated, slowly proliferating gut NETs. A detailed histological profile of the proliferative potential of a tumour is therefore mandatory in selecting patients who are appropriate for treatment. Poorly differentiated or anaplastic NETs may demonstrate up to a 70% response rate with cisplatin and etoposide-based combinations, though the duration of response may not extend beyond 8 to 10 months (Mitry et al., 1999). Pulmonary NETs, which often share pathological and behavioural characteristics with small-cell lung cancers, are also often treated with combinations of a platinum and etoposide, though again treatment decisions should be based individually on tumour characteristics and radiological progression. Pancreatic NETs are often chemosensitive, with reported response rates of 40 to 70%, though the optimal regimen has not yet been determined and various combinations of streptozocin, 5-fluorouracil, dacarbazine, and adriamycin are in use. Where possible, such patients should be incorporated into clinical trials of new or existing agents, such as the current UKNETwork trial comparing capecitabine plus streptozocin with
or without cisplatin in unresectable/metastatic NETs (NET01).
Radionuclide therapy Indolent NETs are relatively radioresistant and so there is a limited role for conventional external beam radiotherapy in treating NETs other than for its analgesic benefits in bony metastases. However, by targeting radioisotopes directly at the tumour sites, higher doses of radiation can be administered than with beam irradiation, multiple sites can be treated simultaneously and non-target damage can be minimised. Targeted radionuclide therapy can be used for symptomatic patients with unresectable or metastatic disease when abnormally increased uptake of the corresponding imaging compound is evident, provided they are not pregnant or breastfeeding and do not have significant myelosuppression or renal failure. Gammaemitting radionuclides are replaced by beta emitters (e.g. 131 I-MIBG, 90 Y-octreotide or 90 Y-lanreotide). Therapy is only currently available at a limited number of centres and trials are still ongoing, though early results highlight the considerable promise of this technique with significant benefits in terms of symptom control, biochemical response and tumour stabilisation (Krenning et al., 2005). 131 I-MIBG can cause unstable blood pressure and can interact with other drugs such as butyrophenones, phenothiazines and sympathomimetics.
Embolisation and radiofrequency ablation Embolisation of the hepatic artery may be indicated for patients with multiple and hormonally active liver metastases that are not amenable to surgical resection (Eriksson et al., 1998). The aim of therapy is to control symptoms and reduce tumour size. The methods used vary considerably but the technique can use various combinations of embolising particles, chemotherapeutic agents and radionuclides, and in-house guidelines should be developed at each centre. A symptomatic response may occur in 40 to 80% of patients but the potential therapeutic benefit should be balanced against a recognised mortality of 4 to 7% and possible side effects including postembolisation syndrome (fever, abdominal pain and nausea), carcinoid crisis (the risk of which should be minimised by octreotide infusion, judicious use of fluids, antibiotics and allopurinol), and hepatic abscess formation (5%). 423
Atul Kalhan and Aled Rees
Radiofrequency ablation, which may be performed laparoscopically or percutaneously, is an emerging technique that has been used successfully in managing colorectal liver metastases. There is considerably less experience with its use in neuroendocrine liver metastases, although symptomatic improvement can occur when at least 90% of the visible tumour is ablated.
the QLQ-GINET21, to be used in conjunction with the EORTC QLQ-C30 generic questionnaire (Davies et al., 2006). With further validation, this supplement to the core questionnaire may be incorporated into routine practice and would enable the changes in quality of life in response to treatment or disease to be monitored.
Clinical trials
Prognosis Prognosis in NETs is dependent on histological features (worse with poorly differentiated tumours and those with high Ki-67 index), tumour size, location, and stage. The overall 5-year survival varies from 30 to 70% depending on these factors and the treatment employed. For any given site, the 5-year survival worsens with increasing stage of disease (10 to 40% with distant metastases, 30 to 60% with regional metastases, and 60 to 95% for localised disease). Appendicular NETs carry the best prognosis of tumours arising from the gut (80 to 90% 5-year survival). Bronchial NETs also carry a favourable prognosis where the histology is typical (80 to 90% 5-year survival), although tumours at this site with a less favourable histology (sometimes referred to as ‘atypical’ carcinoids) have 5-year-survival rates of 40 to 70% (Fink et al., 2001). The prognosis for pancreatic NETs is variable and depends on tumour type (Table 36.1), although overall 5-year survival is in the order of 50 to 80%.
Areas of current interest and clinical trials Areas of current interest The molecular biology of NETs is poorly understood at present. Searches for mutations, rearrangements or amplifications in known candidate oncogenes, or loss of tumour suppression gene activity (with the exception of MEN1) have largely been disappointing (Zikusoka et al., 2005). Comparative genomic hybridisation and microsatellite analytic methodologies have identified some new areas of chromosomal loss or gain in gastrointestinal and pancreatic NETs. It is hoped that these new approaches may identify potential candidate genes for novel diagnostic and treatment strategies. As in other cancer clinical trials, much current interest also focuses on quality of life in NETs. An important advance has been the recent development of a diseasespecific module for patients with gastrointestinal NETs, 424
As of July 2006 there are two trials in setup or recruiting within the UK Neuroendocrine Tumour group (UKNETwork) that are registered with the National Cancer Research Network: NET01 is a randomised phase II study comparing capecitabine plus streptozocin with or without cisplatin in the treatment of unresectable or metastatic gastroenteropancreatic neuroendocrine tumours. NET02 is a phase III, randomised, double-blind, stratified comparative, placebo-controlled, parallel group, multicentre study to assess the effect of deep subcutaneous injections of Lanreotide Autogel® 120 mg, administered every 28 days, on tumour progression-free survival in patients with non-functioning enteropancreatic endocrine tumour. Other current trials analyse the effects of histone deacetylase inhibitors, proteasome inhibitors, small molecule tyrosine kinase inhibitors, angiogenesis inhibitors and mTOR inhibitors in NETs.
REFERENCES Anderson, M. A., Carpenter, S., Thompson, N. W. et al. (2000). Endoscopic ultrasound is highly accurate and directs management in patients with neuroendocrine tumors of the pancreas. Am. J. Gastroenterol., 95, 2271–7. Chiti, A., Briganti, V., Fanti, S. et al. (2000). Results and potential of somatostatin receptor imaging in gastroenteropancreatic tract tumours. Q. J. Nucl. Med., 44, 42–9. Davies, A. H., Larsson, G., Ardill, J. et al. (2006). Development of a disease-specific Quality of Life questionnaire module for patients with gastrointestinal neuroendocrine tumours. Eur. J. Cancer, 42, 477–84. Eriksson, B. K., Larsson, E. G., Skogseid, B. M. et al. (1998). Liver embolizations of patients with malignant neuroendocrine gastrointestinal tumors. Cancer, 83, 2293–301. Eriksson, B., Oberg, K. and Stridsberg, M. (2000). Tumor markers in neuroendocrine tumors. Digestion, 62 (Suppl. 1), 33–8. Faiss, S., Pape, U. F., B¨ohmig, M. et al. (2003). Prospective, randomized, multicenter trial on the antiproliferative effect of lanreotide, interferon alfa, and their combination for therapy of metastatic neuroendocrine gastroenteropancreatic tumors – the International Lanreotide and Interferon Alfa Study Group. J. Clin. Oncol., 21, 2689–96.
Neuroendocrine tumours
Fink, G., Krelbaum, T., Yellin, A. et al. (2001). Pulmonary carcinoid: presentation, diagnosis, and outcome in 142 cases in Israel and review of 640 cases from the literature. Chest, 119, 1647–51. Kaltsas, G., Korbonits, M., Heintz, E. et al. (2001). Comparison of somatostatin analog and meta-iodobenzylguanidine radionuclides in the diagnosis and localization of neuroendocrine tumors. J. Clin. Endocrinol. Metab., 86, 895–902. Kaltsas, G. A., Besser, G. M. and Grossman, A. B. (2004). The diagnosis and medical management of advanced neuroendocrine tumors. Endocr. Rev., 25, 458–511. Krenning, E. P., Teunissen, J. J., Valkema, R. et al. (2005). Molecular radiotherapy with somatostatin analogs for (neuro-)endocrine tumors. J. Endocrinol. Invest., 28 (Suppl. 11), 146–50. Mitry, E., Baudin, E., Ducreux, M. et al. (1999). Treatment of poorly differentiated neuroendocrine tumours with etoposide and cisplatin. Br. J. Cancer, 81, 1351–5. Modlin, I. M., Lye, K. D. and Kidd, M. (2003). A 5-decade analysis of 13,715 carcinoid tumors. Cancer, 97, 934–59. Ramage, J. K., Davies, A. H. G., Ardill, J. et al. (2005). Guidelines for the management of gastroenteropancreatic
neuroendocrine (including carcinoid) tumours. Gut, 54 (Suppl. 4), iv 1–16. Rubin, J., Ajani, J., Schirmer, W. et al. (1999). Octreotide acetate long-acting formulation versus open-label subcutaneous octreotide acetate in malignant carcinoid syndrome. J. Clin. Oncol., 17, 600–6. Schillaci, O., Corleto, V. D., Annibale, B. et al. (1999). Single photon emission computed tomography procedure improves accuracy of somatostatin receptor scintigraphy in gastro-entero pancreatic tumours. Ital. J. Gastroenterol. Hepatol., 31 (Suppl. 2), S186–9. Sutton, R., Doran, H. E., Williams, E. M. et al. (2003). Surgery for midgut carcinoid. Endocr. Relat. Cancer, 10, 469–81. Tomassetti, P., Migliori, M. and Gullo, L. (1998). Slow-release lanreotide treatment in endocrine gastrointestinal tumors. Am. J. Gastroenterol., 93, 1468–71. Turner, G. B., Johnston, B. T., McCance, D. R. et al. (2006). Circulating markers of prognosis and response to treatment in patients with midgut carcinoid tumours. Gut, 55, 1586–91. Zikusoka, M. N., Kidd, M., Eick, G. et al. (2005). The molecular genetics of gastroenteropancreatic neuroendocrine tumors. Cancer, 104, 2292–309.
425
37
CANCER IN CHILDREN Sally Goodman
Introduction
Types of childhood cancer
Cancer in children is rarer than in adults. However, many cases are treated successfully, which means that an increasing number of adults are survivors of treatment for cancer in childhood with the long-term morbidity of treatment and an increased risk of further malignancy. Most childhood cancers differ significantly from cancer in adults in that childhood cancers involve more primitive cell types and a greater reliance on, and response to, chemotherapy for successful treatment. Considerable advances have been made in the management of childhood cancers over the past 30 years. These advances are at least partly due to the early establishment of multidisciplinary teams and collaboration between centres, thus allowing the development of protocols and the entry of many patients into trials or studies whereby treatment schedules could be refined and new ideas introduced. In the UK, most doctors involved in the diagnosis and treatment of childhood cancers are members of the United Kingdom Children’s Cancer Study Group (UKCCSG). The UKCCSG holds meetings twice a year as a group. It registers as many cases of childhood cancer as possible and oversees the management of trials and research. It also provides peer support for singlehanded practitioners (most radiotherapists), has a parallel nursing group and provides information and support for parents. It is difficult to define when children become adults in terms of malignant disease. The UKCCSG registers children age 15 and under, but some childhood malignancies continue into early adulthood, and there is an increasing movement to register teenagers and adults up to the age of 25 on the grounds that malignancies in this age group are more often akin to paediatric tumours in their histology and management.
The different histological subtypes fall into distinct age groups, with corresponding implications for treatment. Table 37.1 shows the UKCCSG registrations for children age 15 and under for the year 2002. Registration with the UKCCSG has increased steadily over the years, particularly for the common malignancies. Most malignancies have a registration rate in excess of 90%, but registration falls short for some conditions, particularly those apparently amenable to treatment with surgery alone. Older children are less likely to be registered. Table 37.1 is a reasonably accurate representation of incidence of the more common malignancies. The UKCCSG also has links with similar groups in Europe and the USA. International co-operative groups include the following: r Soci´et´e Internationale d’Oncologie P´ediatrique (SIOP). r European Organisation for Research and Treatment of Cancer (EORTC). r Children’s Oncology Group of North America (COG), which incorporates sites in USA, Canada and Australia. r St. Jude’s Children’s Research Hospital.
426
Classification by age group In the lists that follow, parentheses indicate overlap with the next group. Many tumours can occur across a wide age range but are most common in the groups shown. Carcinomas are rare, particularly in young children. Cancers that typically occur in infancy include r Retinoblastoma. r Sacro-coccygeal teratoma. r Neuroblastoma. r Juvenile chronic myeloid leukaemia.
Cancer in children
Table 37.1. UKCCSG registrations for children age 15 and under in 2002 Number of Cancer type
registrations in 2002
Leukaemias/myelodysplasias
565
Hodgkin lymphoma
99
Non-Hodgkin lymphoma
102
CNS
354
Neuroblastoma
104
Retinoblastoma
53
Wilm’s/renal
108
Osteosarcoma
44
Ewing’s sarcoma
30
Rhabdomyosarcoma
53
Germ cell tumours (includes
42
CNS, gonadal and extragonadal) Langerhan’s cell histiocytosis
48
Unspecified
< 20
Total
1785
r Ewing’s sarcoma. r Osteosarcoma. r Rhabdomyosarcoma. r CNS. r Germinomas. r T-cell NHL. r Hodgkin lymphoma.
Possible causes of cancer in childhood (See sections on individual tumours for more detail.) In most cases, the cause of the cancer is not known. Genetic causes include inherited cancer syndromes and ethnic differences. Environmental causes include ionising radiation (diagnostic, accidental), infection (EBV, hepatitis B) and parental factors (smoking, drugs, alcohol). Immunodeficiency, either congenital or acquired, can also be a factor in disease. Treatment for previous malignancy carries risks from chemo- and radiotherapy. In some cases, there is an increased genetic risk.
Management: general principles r Hepatoblastoma. r (Non-Hodgkin lymphoma [NHL]). r (Acute myeloid leukaemia). Cancers that typically occur in early childhood include: r Neuroblastoma. r Wilm’s tumour. r Rhabdomyosarcoma – embryonal. r Optic glioma. r ALL. r B-cell NHL. r (Medulloblastoma). r (Ependymoma). Cancers that typically occur in the prepubertal age group include: r Medulloblastoma. r Ependymoma. r Low-grade gliomas. r Rhabdomyosarcoma. r T-cell NHL. r Acute myelocytic leukaemia(AML). r Acute lymphoblastic leukaemia (ALL). Cancers that typically occur during puberty and in the teenage years are:
Children with cancer are managed by paediatric oncologists and haematologists. Surgery is usually carried out by paediatric surgeons, although neuro- and orthopaedic surgeons often cover all age groups. Clinical oncologists generally work as part of a team to provide radiotherapy. Improvements in survival have been considerable, but often at the expense of long-term morbidity. Important in the evolution of treatment protocols have been accurate staging and examination of prognostic indicators (clinical, histopathological, immunological, genetic, etc.), which have enabled patients to be divided into different risk groups. This has meant that treatment can be intensified for patients most at risk but reduced in intensity for those with a good prognosis. Management plans should follow discussion in the multidisciplinary meeting (MDM) and are usually in line with protocols or trials that describe the precise details of radiotherapy planning, dose and timing. Most protocols include intensive chemotherapy with or without surgery, with the radiotherapy fitted into the sequence at a specified time. Treatment should never be given without reference to the current protocols. For more detailed information on individual tumours, most protocols have comprehensive introductions and are well referenced for further reading. 427
Sally Goodman
The sections in this chapter on individual tumours do not attempt to go into detail but outline the relevance of radiotherapy and the usual doses used.
Adolescents and young adults Adolescents with cancer are increasingly recognised as a group requiring separate consideration, both in terms of cancer management and supportive care. There may be inconsistencies in management depending on the route of referral. For example, adult and paediatric protocols for the treatment of lymphomas and leukaemias may differ significantly, having been developed by different working groups. Adolescents are more likely than younger children to develop carcinomas, such as cancers of the thyroid, bowel, breast and ovary, and testicular teratomas. Some adolescents are still in full-time education, others in work, married or with young children. It can be difficult to decide whether they should be cared for by paediatric or adult teams, and where they should be nursed as inpatients. There is an ongoing movement to recognise adolescent/young adult malignancy as a specialty in its own right and to establish dedicated teams and facilities for 16- to 24-year-olds.
Radiotherapy in childhood malignancy Practical management The UKCCSG provides information sheets for parents and older children, and there are general information books written in very simple language for younger patients. The first consultation should be unhurried if possible, and the child should be involved in the explanation of treatment planning and delivery. Many children display considerable maturity and should be encouraged to ask questions and talk about their worries. Good technique is essential, because there is often little margin for error. Immobilisation devices are important, and the child’s co-operation is essential. If the child is too young, or unable or unwilling to co-operate, then sedation or general anaesthesia is necessary. However, most centres do all they can to avoid this for logistical as well as clinical reasons. The involvement of play specialists at the earliest possible stage is invaluable. The parents may be much more worried than the child, and time spent reassuring them may make the course of treatment easier. Often the most difficult step is leaving the child alone in the treatment 428
room with no direct or visual contact with their parents (parents are able to remain in the room with their child for most radiological investigations). Time spent playing on the set and making the child (and parents) familiar with the machine usually pays off. One of the reasons for recommending that the treatment of children be restricted to specialist centres is to ensure that all staff, not just the medical staff, are familiar with the particular problems encountered in treating children. Younger children require the support of paediatric nurses during treatment, and prescriptions of drugs need to be in appropriate formulations and doses. In most cases, this should remain the remit of paediatricians.
Radiotherapy side effects Children often cope with the acute side effects of treatment much better than adults. They are used to being supported through neutropenic episodes and may require prolonged parenteral nutrition. Mucosal and skin reactions are often less pronounced than in adults in spite of concurrent chemotherapy. Of much more concern are the late side effects. Late side effects vary with dose, but also with the age of the child at exposure, the relative maturity of the irradiated organs, and the concurrent treatment. Late effects may result from relatively low doses and it is important to remember to check doses to organs outside of the high-dose volume, for example in the line of the exit beam for spinal irradiation. It can be difficult to separate the effects of radiotherapy from those of chemotherapy. There may be psychosocial problems as well as the more specific organ-related toxicities. All children treated for cancer should be followed up, life-long, in ‘late effects’ clinics, with the aim of identifying problems when it may still be possible to reverse changes or at least limit further deterioration. This applies particularly to endocrine effects on growth and development. As more children survive, so more is being discovered and published about late consequences of cancer treatment in childhood. The Childhood Cancer Survival Study reviewed more than 10 000 survivors of treatment under the age of 21, and treated before 1986. Approximately two-thirds had received chemotherapy, and twothirds radiotherapy. Nearly half had multiple health problems, including congestive cardiac failure, myocardial infarction, second malignancies, gonadal failure, cirrhosis and loss of limb function. The worst affected were children treated for Hodgkin’s lymphoma and CNS
Cancer in children
tumours and children who had received both modalities of treatment. Since these patients were treated, the use of radiotherapy has declined in favour of chemotherapy, and techniques for delivering treatment have improved. On the other hand, drug treatment has intensified and anthracyclines are still central to the treatment of many cancers. Long-term follow-up of survivors of acute lymphoblastic leukaemia showed a higher death rate when radiotherapy had been part of therapy. This group also had a higher unemployment rate and were less likely to be married. Children treated with total-body irradiation (TBI) as part of conditioning for bone marrow transplant (BMT) may suffer more long-term side effects if the transplant is carried out late in the course of the disease, because of the greater exposure to both chemo- and radiotherapy. The risk of second malignancies increases with time. Lymphomas (post-transplantation lympho-proliferative disease, PLD) and haemopoietic disorders, particularly AML and myelodysplasia, are more common in the first few years, reflecting impaired immune function. For solid tumours the risk is increased by young age (< 10 years) at BMT and increasing doses of radiation. Carcinomas of the breast, salivary glands and thyroid; cancers of bone, connective tissue and brain; and melanomas are common (see also the following discussion).
Growth Direct effects of radiotherapy Radiotherapy can weaken mature bone and limit growth in developing bone. It does this by damaging the epiphyseal growth plates. The final effects depend on the age at the time of treatment, the stage of development of the particular bone, concurrent therapies, and other factors unrelated to treatment such as genetic influences. Doses below 20 Gy (fractionated) probably have little effect except in the very young child. More important than loss of height is asymmetrical growth, resulting from partial irradiation of a growth plate. This is most important in the spine, where failure to include the full width of the vertebral body can lead to scoliosis. Even when the vertebrae are uniformly irradiated, uneven growth in muscles and late fibrosis can still lead to development of scoliosis, particularly if the initial treatment included surgery. Irradiation of the femoral head to 25 Gy in children less than 4 years old can result in a slipped epiphysis. The femoral head should be shielded if possible. This also
helps reduce the risk of avascular necrosis seen particularly in the hips and shoulders in children treated for lymphoma. Chemotherapy and corticosteroids can impair bone mineralisation and enhance the effects of radiotherapy which, combined with endocrine disturbances, can lead to osteopenia and osteoporosis.
Indirect effects of radiotherapy Normal growth depends on a complex process mediated by the neuroendocrine axis. Growth can be disrupted centrally or peripherally, depending on the size and site of the radiation fields and the age of the child during treatment. Chemotherapy and corticosteroids contribute to the problem. Growth hormone deficit is common after brain irradiation, including TBI. Pituitary and hypothalamic function may also be affected by radiotherapy to tumours of the head and neck. Growth hormone secretion is reduced at doses greater than 18 Gy, or TBI 10 Gy (single fraction), 12 Gy (fractionated). TSH/TRH, ACTH, FSH, and LH are more resistant but are affected by doses in excess of 30 Gy. The testes are particularly sensitive to radiotherapy and chemotherapy (see later discussion). Leydig cells are more resistant than Sertoli cells but doses above 20 Gy lead to delayed or arrested puberty. Ovarian function is altered by doses above 6 to 10 Gy. Direct irradiation of the thyroid gland can lead to hypothyroidism, a condition more likely to occur in younger females. There is also an increased risk of developing thyroid nodules. Hypothalamic irradiation (>20 Gy), particularly in girls younger than 4 years, can lead to precocious puberty and obesity. Once puberty has been reached, attempts to improve growth by hormone supplements are not effective.
Neuropsychological/cerebral It has been difficult to separate the effects of radiotherapy from those caused by chemotherapy, high-dose steroids, surgery, damage from the tumour itself, disruption of education and of normal lifestyle and chronic institutionalisation. However, it is clear that radiotherapy on its own can cause significant problems. Complex psychometric testing is required to show the full extent of such problems, and the results vary with the age of the child and the time from treatment. However, it seems clear that lower doses of radiation, as used for leukaemia and lymphoma (<24 Gy cranial, 12 to 14.4 Gy 429
Sally Goodman
hyperfractionated TBI), can result in problems with information processing whereas higher doses, as used for brain tumours, can lead to more serious global deficits. The younger the child when treated, the greater the risk of significant damage. Leuko-encephalopathy can occur when radiotherapy is combined with methotrexate, particularly when the drug is given in high doses, and systemically, although standard dose/intrathecal therapy is not free from risk. The risk is highest for children under 2 years and with doses greater than 30 Gy, particularly in the presence of posterior fossa tumours and hydrocephalus. Radiotherapy to the region of the hypothalamus and optic chiasm can lead to occlusion of one or more major vessels, producing occlusive cerebral vasculopathy (Moyamoya syndrome). Risk factors include young age, higher doses, neurofibromatosis and a primary tumour involving the parasellar area.
Sexual development/fertility Sperm production is affected by doses to the testes as low as 1 Gy. Exposure to 6 Gy produces irreversible azoospermia in 100% of boys and the safe limit is 2 to 3 Gy. The testes do not have to be directly in the radiation field to be damaged. The age of the child at treatment does not affect sensitivity. Testosterone production is more resistant, and doses up to 20 Gy have little effect. Ovarian failure occurs at doses of 12 to 15 Gy, with the chance of permanent damage increasing with the child’s age and sexual maturity. When the spine is irradiated, the ovaries may receive a high enough dose from the exit beam to be affected. It is possible to mobilise the ovaries and to move them to a position where they are outside of the radiation beam or under lead shielding. Chemotherapy is also highly damaging and modern protocols aim to combine modalities but limit exposure, in the hope of better-preserving ovarian function. Even if the endocrine deficit is corrected and the patient continues to ovulate, the chances of carrying a pregnancy to term depend on the dose received by the uterus. Uterine vascular insufficiency is seen at doses of 20 to 30 Gy and the endometrium fails to respond to stimulation with sex steroids. Irradiation of the breast bud leads to impaired development of the breast, ranging from agenesis to modest loss of volume depending on the dose and stage of development at the time of irradiation. It may not be appropriate to shield the breast bud in all cases, but it is easy to forget that it is in the field and to therefore simply not consider shielding. It is also important to minimise 430
irradiation of breast tissue because of the risk of inducing malignancy.
Organ development The effects of radiotherapy on organ development depend on the age of the child and the relative maturity of the organ. Not all organs follow the same pattern of growth and maturation. For example, the brain and kidneys develop rapidly in the first 2 to 3 years of life, the gonads remain quiescent until just before puberty, and the musculoskeletal system goes through periods of great activity in early life and in adolescence, with slower growth in the interim. Organs are at their most vulnerable when the cells are still proliferating and differentiating. Once an organ is mature it is still subject to late radiation damage, with the same tolerance as an adult but with the proviso that the use of chemotherapy also influences late effects. Significant effects are seen in most organs at doses greater than 20 Gy to the whole organ. Kidneys and lungs are the most sensitive dose-limiting organs, where damage has potentially fatal consequences; 15 Gy to the whole kidney causes renal insufficiency. Doses to the lung need to be corrected for increased transmission through air. The lung is the dose-limiting organ in TBI, and fractionation and dose rate considerably influence toxicity. The dose rate is particularly important in singlefraction TBI.
Eyes/ears/teeth Children are more likely than adults to receive chemotherapy and high-dose steroids in addition to radiotherapy for tumours in the head because of the difference in tumour types between adults and children. Children are also more likely to be cured after such therapy so that the late effects of treatment have greater significance. In addition, the tumours themselves may contribute to organ damage and the frequent need for antibiotics for neutropenic sepsis also exposes them to toxicity. Remember that organs at risk from low doses of irradiation are affected by TBI, which includes the head. The eyes of children and adults are equally sensitive to radiation damage but combination with highdose steroids and chemotherapy increases the chance of cataract formation. Cataracts are almost inevitable after TBI. The developing ear is at risk of radiation damage to the sensorineural pathways at doses above 30 to 40 Gy. The risk is increased by the use of high-dose platinum compounds, aminoglycosides and the presence of a shunt.
Cancer in children
The size, shape and quality of teeth can be affected by doses as low as 20 Gy in children under age 5, and xerostomia has lasting effects on dental health. Irradiation of the facial bones in younger children leads to craniofacial abnormalities that may require surgery at a later stage, for cosmetic reasons but also to correct malocclusion. Chronic sinusitis can also be a problem following doses exceeding 30 Gy.
Genetic disorders with an increased incidence include trisomy 21, Bloom’s syndrome, Fanconi’s anaemia, Wiscott-Aldrich syndrome, ataxia telangiectasia, and myelodysplasia. Other risk factors include exposure to ionising radiation, alkylating agents and epidophyllotoxins.
Second malignancy
Chemotherapy is divided into four phases: induction, intensification, consolidation and maintenance. Regimens vary according to age and prognostic factors. Radiotherapy was used in the past to treat the CNS and testis, so-called sanctuary sites. As chemotherapy has become more intense, the risk of relapse at these sites has lessened, and very little prophylactic radiotherapy is now given. The main role of radiotherapy is in bone marrow transplantation (see later discussion) and in the treatment of relapsed disease in patients for whom BMT is not possible. It may also be used to palliate advanced disease. Doses for the treatment of leukaemia are as follows: r Brain – 24 Gy in 15 fractions (to cover whole brain and meninges, cribriform plate, optic nerve and upper cervical cord). r Testes – 24 Gy in 12 fractions (to cover testes, spermatic cord to inguinal ring). r Palliation – 4 to 6 Gy as a single fraction. Bone marrow transplantation is used increasingly in high-risk patients early in the disease, and after relapse. Stem cells are obtained from bone marrow or peripheral blood from matched or mismatched, related or unrelated donors. Conditioning schedules vary according to the degree of matching, the underlying malignancy, and the source of the stem cells. TBI is an integral part of conditioning for BMT, with techniques varying according to local facilities. A commonly used dose schedule is 14.4 Gy in eight fractions, treating twice daily with a 6-hour gap, which has a low risk of pulmonary toxicity, but there is some variation among centres. Acute pneumonitis is the doselimiting toxicity. TBI is also discussed in Chapter 31 (see p. 366).
Unfortunately, second malignancies are common in children who survive their first cancer, occurring in approximately 3 to 5% of survivors. Some children have a genetic predisposition to malignancy and are more likely to develop second sarcomas; all are vulnerable to the carcinogenic effects of chemotherapy and radiotherapy. Treatment-induced second malignancies fall into two main categories: leukaemias and solid tumours. The former are usually linked to chemotherapy, the latter with radiotherapy. Second malignancies are common in retinoblastoma, Wilms’ tumour, RMS and soft tissue sarcomas (especially those associated with neurofibromatosis). Patients who survive Hodgkin lymphoma are also at increasing risk of second malignancy with time. AML is more common in patients treated with chemotherapy alone and correlates with exposure to alkylating agents. When analysed, this relates to stage of disease at presentation, and to recurrent disease, requiring repeated and prolonged courses of chemotherapy. This risk appears to plateau at around 14 years after treatment. Thyroid nodules and cancers are common after irradiation of the neck, and breast cancer is an important late second malignancy, often not occurring until 15 to 20 years after treatment. Most breast cancers arise within, or on the edge of, the radiation field. Younger children (less than 10 years) appear less susceptible, with the risk rising through the teens and early twenties. The increase in risk appears to continue for at least 30 years.
Leukaemias Leukaemia is the most common malignant disease in children. Acute lymphoblastic leukaemia (ALL) is the most common form: more than 80% are of B-cell origin. The presence of the Philadelphia chromosome confers a worse prognosis. T-cell leukaemias merge into Tcell lymphomas and are more likely to show systemic involvement. They are also more common in older children.
Principles of treatment
Lymphomas Lymphomas make up 10% of childhood malignancies, with a male-to-female ratio of 3:1. Approximately twothirds of cases are non-Hodgkin lymphomas. 431
Sally Goodman
Possible causes of childhood lymphoma include ataxia telangiectasia, Wiskott-Aldrich syndrome, congenital and acquired immunodeficiency syndromes, prolonged immunosuppression, EBV infection and previous Hodgkin’s lymphoma.
Hodgkin lymphoma Hodgkin lymphoma has a high cure rate, even in cases that are advanced at presentation. Long-term follow-up studies have examined late morbidity, in particular the incidence of second malignancies (see later discussion). Repeated courses of treatment to salvage relapsed disease are associated with greater risks. In addition, the use of radiotherapy increases the risk of solid tumours, especially breast cancer. New protocols have been devised to reduce exposure to alkylating agents and to reduce the radiotherapy dose and treatment volume. However, it is important that the treatment minimises the risk of relapse and the need for salvage therapy. Radiotherapy remains a highly effective treatment for Hodgkin lymphoma, and it is associated with lower relapse rates, even if survival is not improved because of increased mortality from other causes. Trial designs aim to exploit the former property and reduce the latter. The new protocols also aim to preserve fertility. PET scanning is being incorporated into protocols to try and identify whether or not residual masses are active disease requiring consolidation radiotherapy.
Chemotherapy ABVD (doxorubicin, bleomycin, vinblastine and dacarbazine) has been the standard regimen for many years. New protocols also use OEPA (vincristine, doxorubicin, etoposide and prednisolone). The number of cycles is increased according to the stage of disease. COPP (cyclophosphamide, vincristine, procarbazine, prednisolone) is added in more advanced disease.
Radiotherapy The dose for involved field radiotherapy (IFRT) is 20 to 30 Gy in 12 to 18 fractions, depending on the response and the results of PET scanning. The latest UK protocol stipulates that radiotherapy should only be used for residual masses that are positive on a PET scan. All initially involved sites are treated with a margin of 1 to 2 cm unless they are close to critical organs, and a boost is given to poorly responding or bulky residual disease.
432
Non-Hodgkin lymphomas (NHLs) Main types The main types of NHL are as follow: r Anaplastic large cell lymphoma (extranodal sites common, including skin). r T-cell lymphoblastic lymphoma (affects adolescents, mediastinal disease –respiratory emergency). r B-cell lymphomas (abdominal mass [acute abdomen, intussusception], head and neck tumours). r Burkitt lymphoma (rare outside of equatorial Africa and is more likely to present with disease in the abdomen rather than the jaw). Bone marrow involvement occurs in up to 30% of children with NHL at presentation.
Treatment The high risk of second malignancies in addition to effects on growth and fertility have led to the development of protocols that limit exposure to radiotherapy and limit the dose and volume of radiation when it is indicated. Chemotherapy regimens that avoid alkylating agents are used as first-line therapy. Rituximab is now available for lymphomas that express CD20. The value of PET scanning in investigating residual masses is being assessed in current trials. Previously there was disagreement as to whether residual masses should be irradiated, with conflicting results from clinical trials around the world. PET scanning offers the possibility of distinguishing active residual disease from scar tissue and, thus, of sparing patients unnecessary radiotherapy. The most common problem site is the mediastinum. Radiotherapy requires careful planning with attention to heart, lungs, growth plates and the breast buds in girls. Radiotherapy can be given for CNS prophylaxis in high-risk disease, and in the treatment of CNS and testicular disease in the same way as for leukaemias.
Central nervous system Survival in this large group of children has improved more slowly than in most other groups, and the quality of life in survivors may be significantly affected by treatment-related problems. For many years, this group was not well represented in UKCCSG registrations and treatment was predominantly surgical and not necessarily by paediatric specialists. This situation is changing, and a multidisciplinary approach is now
Cancer in children
the norm, with an increasing number of patients being entered into trials. Trial design now includes biological studies and long-term quality-of-life assessments in addition to traditional end points such as survival. Assessment tools include the Health Utilities Index for comprehensive assessment of quality of life, looking at educational development, psychological problems, ability to form relationships, and other complex measures, and the Lansky Scale for assessment of performance status in 1- to 16-year-olds. The long-term effects of radiotherapy on growth, development and intellectual function have led to the development of protocols using chemotherapy in addition to surgery. This is particularly important in children younger than age 5, who are most at risk of longterm neurocognitive damage. However, radiotherapy remains an effective treatment for many CNS tumours, with or without surgery. Radiotherapy of CNS tumours requires the use of immobilisation devices for the head and sometimes the whole spine. Most CNS tumours are sporadic, but the following conditions are associated with an increased incidence: r Neurofibromatosis (15 to 20% of patients with optic gliomas, low-grade gliomas have NF1). r Tuberous sclerosis (subependymal giant cell astrocytoma). r Von-Hippel-Lindau disease. r Gorlin’s syndrome. r Turcot’s syndrome. r Li-Fraumeni syndrome (TP53 mutation). r Ataxia telangiectasia. r Previous radiotherapy (meningiomas after low-dose irradiation for benign conditions, gliomas in up to 10% of long-term survivors of ALL, depending on dose to the CNS). Patients present with signs of raised intracranial pressure, seizures and focal neurological signs indicating the site of the tumour. Midline tumours may lead to disturbances in growth, eating and drinking, diurnal rhythms, and behaviour. Infants may show a failure to thrive and feeding disturbances. Children with diencephalic syndrome show hyperactivity and asthenia, often with increased appetite. Parinaud’s syndrome consists of paralysis of upward gaze, limited pupillary response to light with normal accommodation, and diminished convergence. Suprasellar germ cell tumours may cause diabetes insipidus, visual defects and abnormal sexual development, usually precocious puberty.
Site/incidence Infratentorial disease sites comprise 60% of cases; 75% of these are in the cerebellum: r Cerebellum/fourth ventricle – medulloblastoma/ PNET (primitive neuroectodermal tumour), astrocytoma, ependymoma. r Brain stem – astrocytoma, glioma. Supratentorial disease comprises 40% of cases, with about 30% in the cerebral hemispheres and 40% in the parasellar region: r Cerebral hemispheres – astrocytoma, PNET, ependymoma (lymphoma, meningeal tumours). r Midline supratentorial – astrocytoma, craniopharyngioma, germinoma, non-germinomatous germ cell tumour. r Third/lateral ventricle – choroid plexus tumours, pineal region tumours, ependymoma. With regard to spinal cord tumours, intrinsic spinal tumours, astrocytoma and ependymoma, are extremely rare. Extrinsic tumours include neuroblastoma, Ewing’s tumour and lymphoma.
Staging There is no validated staging system, and so tumours are described in terms of site, structures involved, and number of lesions. Meningeal involvement is staged according to the classification by Chang, published in 1977: r M0 – no evidence of dissemination. r M1– malignant cells in the CSF more than 14 days postoperatively, with no evidence of meningeal enhancement on CT/MRI. r M2 – meningeal enhancement over the cerebral area, laminary thickening and nodular deposits. r M3 – as M2, over spinal cord. r M3 – extraneural disease.
Low-grade glioma Gliomas make up 30 to 40% of primary paediatric brain tumours and the majority are low-grade astrocytic and oligodendrocytic tumours with a mean age at presentation of 11 years. Complete resection is often possible for tumours in the cerebral hemispheres, with cure rates in excess of 90%. Even with incomplete resection, survival can be many years. These tumours also respond well to radiotherapy, although the response may be extremely
433
Sally Goodman
slow. Tumours may actually appear larger after treatment, and regress slowly over years. The aims of trials in this group are to: r Improve radiotherapy planning to minimise the dose to surrounding normal tissue. r Delay radiotherapy until there is evidence of progressive disease after surgery. r Use chemotherapy to delay the need for radiotherapy in younger children (with the minimum age for radiotherapy being increased as more confidence is gained in the effectiveness of chemotherapy). r Avoid radiotherapy in children with NF1, because this group seems to be more susceptible to neuropsychological deterioration and radiation-induced vasculopathy. Their tumours are also more likely to show spontaneous partial regression. But they are also at risk of more aggressive tumours for which radiotherapy may be essential. Chemotherapy with carboplatin and vincristine is effective, with induction and consolidation phases. Prolonged courses of treatment can delay the need for radiotherapy and often relieve symptoms even if there is no obvious tumour regression. Allergic reactions to carboplatin can be a problem. Radiotherapy should be carefully planned, and conformal or stereotactic techniques should be used where possible. There is evidence for a dose-response relationship, and so the emphasis is on reducing the volume rather than the dose. MRI assessment of the tumour volume seems to be reliable, and narrow margins are used in defining the PTV. The dose schedule is 54 Gy in 1.8 Gy fractions reduced to 50.4 Gy for spinal tumours.
High-grade glioma Grade 3 (anaplastic astrocytoma) and grade 4 (glioblastoma), as in adults, have a poor prognosis. Radiotherapy alone achieves a median survival of approximately 9 to 15 months. Current protocols aim to improve outcomes with chemotherapy such as the current UK/French cisplatin and temozolomide regimen. The radiotherapy dose is 54 Gy in 30 fractions.
Brain stem glioma Brain stem gliomas account for approximately 10% of intracranial tumours in childhood. The majority (approximately 85%) are diffuse pontine gliomas, usually high grade, and are diagnosed by typical radiology find434
ings because the risks of biopsy are generally considered to outweigh the benefits. Radiotherapy (54 Gy in 30 fractions, parallel opposed fields) is given with palliative intent. Altered fractionation and chemotherapy do not improve outcome. Current protocols investigate novel approaches (e.g. radiosensitisers, EGFR inhibitors).
Ependymoma Ependymomas are neuroepithelial tumours, making up 6 to 12% of brain tumours in children. More than half occur in children under age 5. The 5-year survival ranges from 30 to 50%, and progress in management has been slow. Much has been made of pathological grading in the past, but it is subject to considerable interobserver variation because the criteria for grading are quite subjective. Newer molecular approaches (Ki67 labelling index, DNA flow cytometry) are still being evaluated. One-third of cases are supratentorial, are more likely to be suitable for complete resection, and therefore have a better prognosis. Two-thirds are infratentorial, where complete resection is much harder to achieve. Spinal metastases are rare. Surgery offers the best chance of a cure, and newer surgical techniques such as ultrasound guidance, CTguided micronavigation, and second-look surgery after chemo- or radiotherapy are producing better survival rates. Traditionally, radiotherapy was given after surgery, and craniospinal irradiation was used for high-grade/ infratentorial tumours. However, there were no good data from randomised trials to support this approach, and CSI is now rarely used unless clearly indicated. Focal radiotherapy is still recommended. The standard dose is 54 Gy in 30 fractions, but hyperfractionated radiotherapy (70.4 Gy, 1.1 Gy fractions, minimum 6-hour gap) and 59.4 Gy in 33 fractions are being evaluated in several current protocols. Stereotactic radiotherapy can be used to boost the tumour bed. Radiotherapy is given within 4 weeks of complete resection, or after chemotherapy with or without second-look surgery. Children under 3 years of age are treated without radiotherapy. Cisplatin is the most active single cytotoxic agent. However, there are concerns over ototoxicity, and most groups are using combination chemotherapy even though the evidence for greater benefit is not strong. Current protocols are evaluating chemotherapy for incompletely resected disease.
Cancer in children
Primitive neuroectodermal tumours (PNETs) including medulloblastoma PNETs can occur at many sites within or outside the CNS. They are tumours with small, round blue cells and variable focal differentiation. In the CNS they show a tendency to seed through the CSF. They are reasonably responsive to chemotherapy and radiotherapy. However, PNETs at different sites in the CNS have very different outcomes and there are different protocols for treating medulloblastoma and supratentorial PNETs. In the latter group about 50% show some differentiation, enabling more specific identification as an embryonal tumour – medulloepithelioma, ependymoblastoma, cerebral neuroblastoma or pineoblastoma.
Medulloblastoma Medulloblastoma is the most common malignant brain tumour in children (20% of all primary CNS tumours). It usually arises from the cerebellar vermis in the roof of the fourth ventricle. One-third of children show evidence of CSF seeding at diagnosis. Medulloblastoma is more common in boys and, although the median age at diagnosis is 5 or 6 years, it can occur over a very wide age range. The prognosis depends on age and extent of disease, and patients with M2 and M3 disease have a less than 40% 5-year survival rate. M1 patients are now also classified as high risk. Resection is still the first-line treatment if possible but may lead to neurological problems caused by damage to the brainstem, with difficulties in swallowing and speech, and truncal ataxia. These problems often improve rapidly, but failure to do so should not prevent a curative approach because recovery can sometimes be prolonged. Shunting may be necessary for hydrocephalus. The fear that this increases the risk of extracranial metastases seems to be unfounded. Most tumours are incompletely resected. The value of radiotherapy in this situation has long been recognised. Postoperative radiotherapy (craniospinal irradiation; CSI) with doses of 35 to 36 Gy to the craniospinal axis and a boost of 18 to 20 Gy to the posterior fossa achieves long-term survival rates of 50 to 70% for early disease. However, this has been at the cost of considerable morbidity, particularly for younger children. Medulloblastomas are also chemosensitive. Good 5-year survival rates have been seen in young children (< 3 years) with reduced doses of radiotherapy and in older high-risk children with standard radiotherapy.
The regimen used is a combination of CCNU, vincristine, and cisplatin (known as the Packer or Philadelphia regimen). This combination, together with reduced-dose radiotherapy, is now standard in the USA and is being used increasingly in Europe. In spite of the combination with CSI, myelotoxicity is not a problem, but levels of renal and ototoxicity are high, and long-term neuropsychological effects still have not been fully evaluated. Current trials continue to use CSI but explore dose reduction in standard-risk patients, and hyperfractionation (HART). Whole-brain irradiation is more damaging than treating the posterior fossa alone, and there appears to be a dose/effect gradient between 24 and 36 Gy. However, the relapse rate is higher at the reduced dose. This effect may be countered by postradiotherapy chemotherapy using the Packer regimen. In patients with metastatic disease, the role and tolerability of concurrent chemotherapy with vincristine are being evaluated. There appears to be no survival advantage for chemotherapy before radiotherapy, which should start within 4 weeks of surgery if possible. Treatment planning must be meticulous and the PTV must cover all the meninges. For this reason, 6 MV is the maximum energy for CSI to ensure that the lateral surfaces of the cerebral meninges are adequately treated. Higher energies can be used for the posterior fossa boost, to spare the hair follicles and ears. Using oblique fields for the boost can also help to reduce the dose to the inner and middle ear. The cranial fields must be matched carefully to the spinal fields, which should have as few joins as possible, which may mean treating at an extended distance to obtain an adequate field length. Electrons or photons can be used to treat the spine. CT planning is essential, and trial protocols include strict quality assurance, involving pretreatment central review of planning films if possible. The HART protocol uses fraction sizes of 1.24 Gy with a gap between the two daily fractions of at least 8 hours. This protocol presents logistical difficulties for some departments, even more so if general anaesthesia is required.
Supratentorial PNETs In spite of histological similarities, children with supratentorial PNETs have a worse survival than children with medulloblastoma, with almost 90% of failures showing local recurrence with or without metastatic disease. Whether this represents a true biological 435
Sally Goodman
difference is unclear. Children with pineal tumours have a much better survival rate in spite of the fact that complete resection is rarely possible. Treatment for these tumours is identical to that for medulloblastomas, but supratentorial PNETs are randomised separately in clinical trials. Biological studies are very important for trying to understand the differences in behaviour, to facilitate stratification for risk in future studies.
Craniopharyngioma Craniopharyngiomas make up 10% of childhood brain tumours. These tumours are technically benign but they can invade locally and recur after surgery. These tumours arise from the pituitary stalk and the floor of the third ventricle and cause symptoms by involving the pituitary, hypothalamus and optic chiasm. They may cause hydrocephalus. They must be differentiated from pituitary tumours (prolactin levels), germinomas and germ cell tumours (βhCG, αFP), lowgrade gliomas and Langerhans cell histiocytosis. The diagnosis should be confirmed by biopsy, and pituitary and hypothalamic function should be carefully assessed before treatment. Surgery offers the best chance of a cure if complete resection is possible. If the tumour is less than 2 cm tall in the midline, then successful resection without undue morbidity is possible in most cases, but with a risk of relapse in up to 25%. Attempts to resect larger tumours may result in serious neuroendocrine disturbance (morbid obesity, adipsic diabetes insipidus, pituitary failure) with a late treatment-related mortality rate of 10%. A conservative approach is recommended. Similarly, repeated surgical intervention increases the risk of morbidity. Radiotherapy can control residual disease and reduce local recurrence rates. Timing is not critical, and salvage radiotherapy can achieve results equal to that of immediate postoperative treatment. If there is obvious residual tumour after surgery, radiotherapy should not be deferred except in very young children: in these cases, the risks of neurological damage from progressive disease, or from more extensive surgery, have to be weighed against the risks from radiotherapy (optic neuropathy, neurological and cognitive impairment, hypothalamic and pituitary failure, but low risk of mortality compared with surgery). CT planning using conformal techniques is recommended, or stereotactic guided radiotherapy if avail436
able, to reduce the dose to surrounding brain tissue. The recommended dose is 50 to 54 Gy in 1.8 Gy fractions. Prolonged follow-up is necessary to identify late morbidity, particularly endocrine, and to detect local recurrence early to minimise the damage from further treatment.
Neuroblastoma Neuroblastoma is the most common solid tumour in infants; its occurrence is exceptionally rare beyond the age of 10. It originates in foetal neuroblasts in the neural crest. Tumours can arise anywhere along the neural crest but are most common in the adrenal glands; 90% produce catecholamines or their metabolites, which may cause symptoms. The histology of neuroblastoma shows a small, blue, round cell tumour with a pseudo-rosette formation and neural features. Poor prognostic features include amplification of n-myc, and 1p deletion. Presentation is usually with an abdominal mass but depends on the site of the primary. Metastases are present in about two-thirds of cases at presentation and typically occur in bone and bone marrow (producing swelling, bruising and pain), liver and lymph nodes. Skin metastases are common in neonates – the ‘blueberry muffin’ sign. There are often non-specific generalised symptoms such as fever, weight loss and failure to thrive. Opsoclonus (rapid multidirectional eye movements) myoclonus-truncal ataxia syndrome is a rare paraneoplastic presentation of early disease. Radiology may show calcification in a suprarenal mass. Neuroblastoma is often very radiosensitive, but its widespread nature makes chemotherapy the more important treatment. Induction chemotherapy (rapid COJEC) is followed by resection of the primary tumour. The current protocol for high-risk disease includes radiotherapy to the primary site after high-dose chemotherapy with stem cell rescue. The dose to the tumour bed is 21 Gy in 1.5 Gy fractions. Radiotherapy can also provide useful and long-lasting palliation in metastatic disease. Cis-retinoic acid is given after radical therapy to encourage tumour differentiation. MIBG can be used to scan for metastatic disease, and to treat it. Treatment with MIBG is only available in a few centres. Stage IV-S disease is found in infants, with metastases to liver, bone marrow and/or skin; patients may have a
Cancer in children
survival in excess of 90% with minimal therapy. Research is under way to identify subgroups with a poor prognosis, for more intensive therapy.
Wilm’s tumour Wilm’s tumour is the most common intra-abdominal tumour in children and accounts for 6% of childhood cancers. It affects both sexes equally and 75% of cases present before the age of 5 years, mostly around 3 or 4 years. It is bilateral in 4 to 8% of cases and there may be associated congenital abnormalities in 10 to 15%, suggesting a genetic aetiology (see Knudson’s “two-hit” hypothesis under “retinoblastoma”). Associated syndromes include the following: r WAGR (Wilms’ tumour, aniridia, ambiguous genitalia, and mental retardation). r Beckwith-Wiedemann. r Aniridia/hemihypertrophy. r Denys-Drash syndromes. Most cases present with a large abdominal mass, and pain can indicate rupture; 20% have fever or haematuria. The histology shows a triphasic pattern with blastemal, epithelial and stromal elements. A favourable histology is present in 90% of cases; the remainder consist of anaplastic Wilm’s tumour, clear cell sarcoma and rhabdoid tumour. The latter two are highly malignant and are now treated on different protocols. Radiological investigations are important in determining the extent of the disease (both local and metastatic) and renal function. Tumour extension into the renal vein or IVC can be demonstrated on ultrasound scans, whereas CT scans allow evaluation of the extent of renal and nodal involvement. Bone metastases are uncommon except in clear cell sarcoma of the kidney and rhabdoid tumours.
Treatment Wilm’s tumour is very sensitive to both chemotherapy and radiotherapy. The development of protocols over the years is an elegant demonstration of how careful staging and treatment in well-run clinical trials can lead to refinements that spare patients with a good prognosis from much of the toxicity and can define a group at high risk of relapse who need intensive combined therapy. There is disagreement about whether the tumour should be biopsied at diagnosis or if chemotherapy should be started on the basis of clinical and radiological
features alone. The latter approach means that there is a risk of incorrect diagnosis and inappropriate treatment, but a reduced risk of tumour rupture and consequent upstaging. The current SIOP protocol requires preoperative chemotherapy. After surgery, the pathological stage and the histology are used to assign patients to a risk group. Radiotherapy is reserved for patients with higher-stage disease and/or high-risk histology. Simplified surgical staging is as follows (not comprehensive): r I – tumour limited to the kidney. r II – tumour extends to adjacent tissues/organs but is completely excised. r III – tumour involves abdominal lymph nodes/has ruptured/was incompletely excised. r IV – metastatic disease. r V – bilateral disease.
Histological subtypes (postchemotherapy): There are low-, intermediate- and high-risk subtypes of Wilm’s tumour: r Low risk – mesoblastic nephroma. r Intermediate risk – nephroblastoma, epithelial, stromal, or mixed. r High risk – nephroblastoma, blastemal type, diffuse anaplasia, clear cell sarcoma, rhabdoid tumour. Radiotherapy, where indicated, is given to the flank or whole abdomen within 2 weeks of surgery. For flank irradiation, the volume is defined with reference to the preoperative imaging, or from clips placed at the boundaries of the kidney/tumour during surgery. The medial border must cover the full width of the vertebral bodies (see section on late effects). Whole-lung irradiation is indicated where there is an incomplete response of pulmonary metastases to chemotherapy. Metastatic disease in the brain and bone may also be irradiated. Careful planning is essential, and sensitive structures must be shielded wherever possible. This includes the breast buds in girls. Radiation doses are as follow: r Flank – 14.4 Gy to 25.2 Gy plus 10.8 Gy boost with fraction size adjusted for age and volume (typically 1.8 Gy). r Whole abdomen – 21 Gy in 1.5 Gy fractions BUT remaining kidney 12 Gy max., whole liver 20 Gy max. r Lung – 15 Gy in 1.5 Gy fractions (lung-corrected) with or without a boost to bulky disease. r Treatment should not be attempted without reference to the current protocol. 437
Sally Goodman
Retinoblastoma Most retinoblastomas are detected before age 3 years (mean age 4 months); 20 to 35% are bilateral and are more likely to present at a younger age and to be familial. Sporadic tumours are more likely to present late and to be more advanced on presentation. A common presentation is with leukocoria (white pupillary light reflex), which is sometimes noticed on a photograph. The histology shows an anaplastic, small, round cell tumour of neuroepithelial origin. More differentiated areas may show rosette formation. Vitreous seedlings can occur. The 5-year survival is around 90%. The excellent survival and early diagnosis in familial cases means that many survivors go on to have children of their own who will also carry the gene. Family screening is therefore very important. Knudson’s hypothesis proposed that two somatic events were necessary for sporadic tumour development, but that in familial cases, only one somatic event was necessary because of an existing germ-line mutation. Subsequently, cytogenetic studies demonstrated a deletion on the long arm of chromosome 13. Treatment for all but advanced disease is at a very small number of specialist centres. Distant metastases are rare and treatment is usually localised, with the aim of sparing the eye and vision if possible. Organ-sparing techniques include photocoagulation, cryotherapy and the application of plaques of radioisotopes (60 Co, 125 I, 192 Ir, 109 Ru). External beam radiotherapy (EBRT) may be used for the treatment of locally advanced or recurrent tumours and for metastatic disease. Chemotherapy (e.g. ifosfamide, vincristine, doxorubicin) is used for refractory disease and for extraocular disease, in combination with EBRT. In disseminated disease it may be followed by high-dose therapy and autologous stem cell transplant. After enucleation, extensive invasion of the choroid, retrolaminar extension and involvement of the cut end of the optic nerve are all poor prognostic features. The benefit of adjuvant chemotherapy in these cases is not yet proven.
Bone tumours Osteosarcoma Osteosarcoma is the most common primary bone tumour; patients present with pain, swelling and occasionally pathological fracture. The majority occur in the 438
metaphyses of the long bones: femur (45%), tibia (18%) and humerus (11%). Most osteosarcomas arise on the surface of the bone and show periosteal new bone formation giving rise to ‘Codman’s triangle,’ a typical radiological finding. The most common histological types are osteoblastic, chondroblastic and fibroblastic. Parosteal osteosarcomas occur in older children and have a better prognosis. Osteosarcomas are associated with growth spurts and previous exposure to ionising radiation, particularly radiotherapy for retinoblastoma or rhabdomyosarcoma. More than 10% of patients have metastatic disease at presentation, usually involving lung (85% of those with metastases) or bone (20%). Chemotherapy has improved survival and resectability, sparing many patients amputation. The most active agents are doxorubicin and cisplatin, followed by methotrexate ifosfamide, and etoposide. The current trial protocol (EURAMOS 1) uses neoadjuvant chemotherapy with MAP (methotrexate in four courses, doxorubicin/cisplatin in two courses). Surgery is followed by consolidation chemotherapy with MAP and randomisations to include ifosfamide or interferon depending on the histological response. Limb-sparing surgery is carried out in supraregional centres. It is now possible to insert a prosthetic replacement that can grow with the limb. Osteosarcomas are relatively radioresistant. Radiotherapy may be used postoperatively for high-risk tumours, for inoperable tumours, and for palliation. The radical dose is 60 to 70 Gy. Survival is around 50 to 60% at 5 years.
Ewing’s sarcoma/peripheral PNETs These comprise 3% of childhood cancers. They occur across a wide age range, but are most common in early puberty. They are more common in white than black children in the USA. They can occur in a variety of sites, including long bones (usually metaphyseal or diaphyseal), with 25% in proximal extremities, 35% in distal, and 40% in flat bones. Twenty percent of patients have metastatic disease at presentation, mostly to lung or bone. Vertebral lesions can present with spinal cord or nerve root compression. Radiological features include Codman’s triangle and permeating destructive lesions with soft tissue masses, especially when flat bones are involved. Askin’s tumour of the rib is characterised by a large soft tissue mass displacing the lung.
Cancer in children
The histology shows small round cell tumours with varying degrees of neural differentiation. More than 95% show an 11;22 translocation. Although these tumours are very radioresponsive, survival is poor without systemic therapy. The primary treatment is with chemotherapy (vincristine, actinomycin D, cyclophosphamide, doxorubicin, etoposide and ifosfamide are all active drugs) followed by radical resection with prosthetic reconstruction wherever possible. Consolidation chemotherapy is used after surgery. Prolonged survival in metastatic disease has been achieved with myelo-ablative chemotherapy with busulfan and melphalan, and stem cell rescue. Radiotherapy is indicated for Ewing’s sarcoma/ peripheral PNETs in the following circumstances: r Before surgery, if there is tumour progression or anticipated marginal/incomplete resection. r After surgery, if there is marginal/incomplete resection or poor histological response. r For inoperable lesions. r For sites of metastases, including the whole lung. Patients receiving busulfan-melphalan must not receive doses greater than 30 Gy to the spinal cord or brain, and radiotherapy should not be given before busulfan. The dose schedule is 44.8 to 54.4 Gy in 1.8 Gy fractions or a hyperfractionated, accelerated split-course technique to enable concurrent chemotherapy. Five-year survival is up to 65% in patients with localised disease at presentation (best in younger patients, non-pelvic sites) and 35% in those with metastases.
Soft tissue sarcomas Rhabdomyosarcoma Soft tissue sarcomas account for 4 to 8% of childhood cancers. Half of these are rhabdomyosarcomas. Possible causes for disease include neurofibromatosis, Li-Fraumeni syndrome and, possibly, smoking in parents. The disease is less common in black than in white children. A wide range of associated congenital abnormalities are seen. Tumours can arise almost anywhere and they tend to metastasise early to lung, lymph nodes, bone and bone marrow. The Intergroup Rhabdomyosarcoma Studies (IRS) in the USA and SIOP in Europe have developed protocols that have brought about considerable improvements in survival over the past 30 years.
Overall 5-year survival is approximately 70%, but this rate varies considerably according to tumour site, stage, and histology. Embryonal, alveolar, pleomorphic and mixed subtypes vary histologically as well as molecularly; the first two show features similar to different stages of foetal muscle development. The later IRS trials divide tumours into unfavourable histology (alveolar) and favourable histology (the rest). Diploid tumours have a worse prognosis than hyperdiploid and the t(2;13) translocation carries a very poor prognosis. Alveolar tumours are more likely to occur in limbs and the perineum, whereas orbital tumours are mostly embryonal. Common sites of disease are: r Genitourinary – approximately 25%, the site of sarcoma botryoides (‘like a bunch of grapes’). r Head and neck – approximately 25%, divided into parameningeal (e.g. nasopharynx, middle ear) and non-parameningeal. r Orbital – approximately 8%. Treatment is with a combination of surgery, chemotherapy, second-look surgery and radiotherapy according to prognostic factors. As with Wilm’s tumour, a combination of pathological staging and histological risk grouping can be used to allocate patients to groups with differing risks of local recurrence. There are several different staging systems, mostly postoperative. A simplified staging system is as follows: r IRS group I – initial complete resection, no lymph node involvement. r IRS group II – gross resection with microscopic residual disease or regional nodes involved. r IRS group III – initial incomplete resection with gross residual disease. r IRS group IV – metastatic disease at presentation. The prognosis depends on the stage, but the anatomic site is likely to determine resectability and, hence, stage. Parameningeal tumours invading the skull base or meninges are harder to control.
Chemotherapy Chemotherapy involves giving vincristine/actinomycin D/cyclophosphamide or ifosfamide before and after surgery, with the extent of surgery depending on response.
Radiotherapy Indications for radiotherapy include incomplete response to chemotherapy with or without surgery, parameningeal tumours, orbital tumours, relapsed 439
Sally Goodman
disease and lymph node involvement. The timing of radiotherapy depends on the site of disease and the patient’s response to chemotherapy. Treatment volumes are strictly defined and it is essential to have all relevant imaging available together with the operation note and histology report if the tumour has been resected. Radiotherapy to tumours in parameningeal sites which have caused bony erosion or cranial nerve palsies must include the skull base in the involved fossa and any intracranial component. Radiotherapy is essential for high-risk patients. It may be omitted in patients who have a favourable histology in IRS group I, in very young patients who have gone into complete remission with chemotherapy, and in patients with tumours in vulnerable sites where the late radiation damage may be considerable. Brachytherapy may be used for vaginal tumours, and some other sites. Referral to a specialist centre is essential. The dose schedule is 36 to 50.4 Gy in 1.8 Gy fractions or smaller according to protocol/risk group/age and treatment volume. There should be a dose adjustment for breaks, but hyperfractionation is not included in current European protocols.
Non-rhabdomyosarcomas Paediatric soft tissue sarcomas cover a wide range of histologies with differing patterns of behaviour. A new European protocol attempts to improve staging and classification, explore biological markers, and standardise management.
Germ cell tumours Germ cell tumours are rare tumours derived from primordial germ cells and they can be found anywhere along the pathway of germinal tissue migration. For the purposes of treatment they can be divided into cranial (suprasellar) and extracranial tumours (gonads, vagina, uterus, sacrococcygeal, mediastinum, thorax). They include germinoma, yolk sac tumour, choriocarcinoma, gonadoblastoma, embryonal carcinomas and teratoma. The incidence is bimodal with peaks before 3 years of age and over the age of 12. Tumour-marker assays (αFP, βhCG) are essential for staging and follow-up. The treatment of extracranial tumours involves surgery if a complete resection is possible or biopsy 440
followed by chemotherapy. The standard UK regimen is JEB (etoposide, carboplatin, bleomycin). Radiotherapy is reserved for the palliation of metastases. Survival is usually excellent and new strategies are aimed at reducing treatment-related morbidity. In the UK, suprasellar germinomas are usually treated with CSI although some European centres favour chemotherapy with localised radiotherapy. The dose schedule is 24 Gy CSI, with a boost of 16 Gy to primary site.
Hepatoblastoma/hepatocellular carcinoma Hepatoblastoma (HBL) is rare in Europe and the USA and more common in Japan and Southeast Asia. HBL is more common in boys, and the median age at presentation is 12 months. Presenting features of HBL include failure to thrive, fever, anorexia, anaemia and abdominal mass; a massively raised alpha-fetoprotein level also occurs. Of HBL patients, 66% show a mixture of foetal and embryonal hepatoblasts, and have tumours that are still resectable at presentation; 30% have anaplastic tumours. Resection is often impossible and long-term survival is rare in the anaplastic group. Hepatocellular carcinoma (HCC) is even less common, occurs in older children, and is usually in association with cirrhosis. Resection offers the best chance of a cure (30 to 70%) in both tumour types. Preoperative chemotherapy with cisplatin or PLADO (cisplatin + doxorubicin) in standard-risk patients and with SuperPLADO (PLADO + carboplatin) in high-risk cases may improve survival in HBL. The addition of anti-angiogenesis drugs may improve the outcome in HCC. These platinum-intense regimens carry a high risk of ototoxicity.
Footnote The UKCCSG has recently amalgamated with the UK Childhood Leukaemia Working Party to become the CCLG – the Children’s Cancer and Leukaemia Group.
Appendix NICE has developed guidance on the health care that should be provided to children and young people with cancer (see Further Reading section). NICE recognises that the needs of children and young people with cancer
Cancer in children
are different from the needs of older adults with cancer. For example, during childhood and adolescence, physical changes occur that may affect the type of cancer present, how it needs to be treated, and how well the treatment works. Children and young people also undergo psychological changes that may influence their emotional response during and after treatment. The aim of this guidance is to ensure that children and young people with cancer get the best possible care and the care that is most appropriate for their age. The key recommendations are as follow: r Care should be coordinated across the whole of the NHS and be available as close to the patient’s home as possible. r Cancer networks should ensure that they meet the needs of children and young people with cancer. r Multidisciplinary teams should provide cancer care. r Each child or young person with cancer should have a key worker. r Care should be appropriate to the child’s or young person’s age and type of cancer. r Time in the operating theatre and a children’s anaesthetist should be available when needed. r Children and young people with cancer should be offered the chance to take part in research trials. r Treatment should be based on agreed protocols.
r A register of all cancers in people ages 15 to 24 should be urgently considered.
r Cancer networks should ensure that there are sufficient specialist members of staff. FURTHER READING Further details about tumours can be found in any textbook of paediatric oncology. The trial protocols are an excellent source of up to date information. British Society of Paediatric Endocrinology and Diabetes and The UK Children’s Cancer Study Group. (2005). Paediatric Endocrine Tumours, Recommendations for the Investigation, Diagnosis and Treatment of Patients and Their Families: A Multi-Disciplinary Consensus Statement of Best Practice from a Working Group Convened Under the Auspices of the BSPEO and UKCCSG, ed. A. Spoudeas. Crawley: Novo Nordisk Ltd. Children’s Oncology Group evidence-based website with recommendations for long-term follow-up and links with pages covering in detail the many possible late problems resulting from the treatment of children and adolescents: www.survivorshipguidelines.org. NICE. (2005). Guidance on Cancer Services. Improving Outcomes for Children and Young People with Cancer. London: National Institute for Health and Clinical Excellence. Royal College of Paediatrics and Child Health. (1997). Guidance for Services for Children and Young People with Brain and Spinal Tumours – Report of a Working Party of the UKCCSG and the Society of British Neurological Surgeons (SBNS). London: Royal College of Paediatrics and Child Health. CCLG website for patients and professionals (members’ section password protected): www.cclg.org.uk.
441
38
CANCER OF UNKNOWN PRIMARY Paul Shaw and Tom Crosby
Introduction
Incidence and epidemiology
Cancer of unknown primary (CUP) is metastatic disease for which a primary site is undetectable at presentation (Varadhachary et al., 2004). It represents 2 to 6% of all invasive malignancies (Muir, 1995; Greco and Hainsworth, 2001). It is important to consider the possibility of potentially curable malignancies, such as germ cell tumours or lymphoma, and the workup for patients with CUP is therefore likely to include biopsy with immunohistochemistry to identify the cell lineage, if possible, and tumour markers. Investigations such as cross-sectional imaging or endoscopy are determined by the site of cancer, the patient’s symptoms, and the general condition of the patient; however, an exhaustive diagnostic workup is not usually justified because it is unlikely to influence the outcome of treatment. For those patients whose primary tumour is identified, treatment should continue as for that individual tumour site. For those patients whose primary tumour has not been identified after the initial workup, treatment is often empirical and based on research in patients whose primary tumour is known. Definitions in the medical literature of such tumours have been published. Examples include a biopsy-proven malignancy for a cancer that could not have originated at the biopsy site; no primary tumour found after a thorough medical history or physical examination, including breast and pelvic examination in women and testicle and prostate examination in men; and normal laboratory test results including the results of a complete blood count, blood chemistry, chest X-ray, CT scan of the abdomen and pelvis, and mammography or prostate-specific antigen (PSA) test (Varadhachary et al., 2004). This chapter focusses on the investigation of patients presenting with CUP and the treatment possibilities for those whose primary tumour is not identified after initial workup.
CUP accounts for 2 to 6% of all invasive malignancies (Muir, 1995; Greco and Hainsworth, 2001) and ranks in the top ten most common cancer diagnoses. The exact incidence is uncertain (Greco and Hainsworth, 2001), but it is roughly equivalent to the number of new cases of non-Hodgkin lymphoma, melanoma, leukaemia, head and neck, oesophagus, pancreas, ovary or kidney cancer (Cancer Research UK, 2004), which is approximately 7000 new cases per annum. Published figures for annual mortality are not available but, given that carcinoma of the pancreas and lung are the most common primary cancers in patients who eventually had their primary site defined (Neumann and Nystrom, 1982), CUP possibly accounts for about 6000 deaths per year. With better pathological processing, imaging and endoscopy, the incidence of CUP has probably reduced slightly over the past decade (Mintzer et al., 2004). The average age of onset is 58 years, but patients presenting with a midline distribution of poorly differentiated carcinoma have a median age of 39 years (Casciato, 2006). Around 10 to 20% of patients have their tumour site identified via antemortem investigation; with the addition of postmortem studies, this rises to around 50% (Abbruzzese et al. 1995; Neumann and Nystrom, 1982). In patients who have their primary tumour site identified, either by investigation or postmortem analysis, the most common sites are the pancreas (20%), lung (18%), liver (11%), large bowel (10%), stomach (8%), kidney (6%), ovary (3%) and prostate (3%) (Neumann and Nystrom, 1982).
442
Risk factors and aetiology Possible explanations for CUP include the following: r Previous excision of unrecognised primary sites.
Cancer of unknown primary
r Spontaneous regression of the primary site following dissemination. r Undetected primary site due to small size. r Site of origin obscured by metastatic deposits.
Table 38.1. Histological subtypes of CUP diagnosed by light microscopy and their frequency Approximate Histological type (light microscopy)
Clinical presentation Presentation is variable and depends on the site of metastasis and may be accompanied by constitutional symptoms of weight loss, lethargy and malaise. The most frequent presenting features include pain (60%), liver mass/abdominal symptom (40%), nodal disease (20%), bone pain or fracture (15%), chest symptoms (15%) and central nervous system abnormality (5%) (Casciato, 2006). The site of CUP metastases offers some clues to the likelihood of whether the primary cancer originates above or below the diaphragm; lung metastases are twice as likely to arise from a primary above the diaphragm (thyroid, breast, parotid gland and lung), whereas liver metastases are more common from primary malignancy below the diaphragm (pancreas, liver colon and rectum, kidney, ovary, prostate, stomach and adrenal gland) (Neumann and Nystrom, 1982). The pattern of metastatic disease in cancers presenting as CUP differs from the expected distribution of a known primary cancer (Neumann and Nystrom, 1982). In a study of 266 patients, 48% eventually had their primary site identified, mostly postmortem. It was found that bone metastases occur in 4% of cases of lung cancer presenting as CUP compared to 30 to 50% of cases where the primary lung tumour was known at the outset, whereas they are more frequent in pancreatic cancer presenting as an unknown primary malignancy. Prostate cancer presenting as CUP is three times less likely to involve bones compared to primary prostatic carcinoma and more frequently involves lung (75%), liver (50%) and brain (25%). Patients require a thorough examination, including head and neck, breast, rectal and pelvic examination (European Society for Medical Oncology, 2005).
Investigation Biopsy Patients with CUP should undergo a biopsy if possible. Following the initial biopsy, the most commonly
frequency (%)
Adenocarcinoma Squamous carcinoma Poorly differentiated types: PDN, PDC,
60 5 35
and PDA PDA = poorly differentiated adenocarcinoma; PDC = poorly differentiated carcinoma; PDN = poorly differentiated neoplasm. Adapted from Greco and Hainsworth (2001).
encountered light microscopic appearances of the tumour are adenocarcinoma, squamous carcinoma, and poorly differentiated types: poorly differentiated neoplasm (PDN), poorly differentiated carcinoma (PDC) and poorly differentiated adenocarcinoma (PDA), as shown in Table 38.1. Melanoma, sarcoma and neuroendocrine tumours are infrequent. For PDN, PDC or PDA, which account for approximately 35% of histological diagnoses, further tests can provide additional information. For example, 20 to 70% of patients with PDN, PDC or PDA may have unsuspected non-Hodgkin lymphoma that may be amenable to potentially curative therapy (Hainsworth and Greco, 1993). It is also necessary to consider germ cell tumours or neuroendocrine neoplasms where treatment may be curative or to provide useful palliation with a survival advantage. In these circumstances immunohistochemistry can be helpful in getting a more precise diagnosis, but it is not often definitive because of false positives (Table 38.2; Casciato, 2006; Hainsworth and Greco, 2001). Electron microscopy can help identify secretory granules in small-cell carcinoma, melanoma and lymphoma that would otherwise be reported as undifferentiated. Chromosome analysis is possible when immunohistochemistry and electron microscopy are unhelpful and may identify translocations such as t(11;22) in Ewing’s sarcoma and PNET. Communication between the oncologist and the pathologist is vital to help interpret clinicopathological findings in the most accurate way.
443
Paul Shaw and Tom Crosby
Table 38.2. Immunohistochemical stains that help tumour diagnosis Tumour type
Immunohistochemical staining
Lymphoma
Common leukocyte antigen (CLA)
Carcinoma
Cytokeratin (CK)
Breast
ER, PgR (+) Gross cystic fluid protein
Prostate
Prostate-specific antigen (PSA)
Germ cell
Human chorionic gonadotrophin (βhCG) Alfa-fetoprotein (αFP)
Sarcoma
Vimentin
Melanoma
S-100
Desmin Vimentin HMB-45 Neuroendocrine
Neuron-specific enolase Chromogranin Synaptophysin
Thyroid
Thyroglobulin Thyroid transcription factor (TTF-1)
Lung
TTF-1
ER = oestrogen receptor; HMB = human melanoma black; PgR = progesterone receptor; S-100 = S-100 calcium binding protein; TTF-1 = thyroid transcription factor. Adapted from Casciato (2006), Greco and Hainsworth (2001).
Other investigations Blood tests, including FBC, renal, liver and bone profiles, should be performed. Tumour markers may be helpful in specific situations. Men with CUP should have serum PSA measured to exclude the possibility of prostate cancer. A normal serum βhCG and αFP exclude extragonadal germ cell tumours. CA15-3 can be of help in women with isolated axillary metastases and CA125 in women with peritoneal papillary adenocarcinomatosis (Pavlidis et al., 2003). Women with gestational trophoblastic tumours have a raised βhCG. Abruzzese et al. (1995) have used a simplified diagnostic strategy and demonstrated that it has a diagnostic yield of 20% when applied to patients with suspected CUP referred to a tertiary centre. They showed 444
that pathological review, physical examination, chest radiography, and mammography often provide the most information. CT imaging increases the number of primary diagnoses (e.g. non-small-cell lung cancer or pancreatic carcinoma) for which there are only relatively ineffective treatments in patients with metastases. Nevertheless, multiple-slice CT scanning of the thorax, abdomen, and pelvis is quick and relatively noninvasive, and it detects tumours of a significant size, which may help to determine the treatment and prognosis and to identify primary tumours that may require specific palliation in the future. CT scanning of the thorax, abdomen and pelvis is therefore frequently performed in patients with CUP. Figure 38.1 shows an algorithm for a suggested investigation strategy in patients with CUP. Serum thyroglobulin measurement may be a useful adjunctive test, in association with fine needle aspiration biopsy of thyroid tissue, in identifying the origin of distant metastases in patients with a cold thyroid nodule and metastases from an unknown primary site (Panza et al., 1987). Further investigations may be indicated, particularly if patients have symptoms or signs that suggest a primary site, but patients with CUP are probably often overinvestigated. For example, invasive endoscopic and laparoscopic procedures are rarely warranted in the absence of clinical symptoms. When a head and neck primary is suspected (e.g. in patients with malignant neck nodes), panendoscopy should be combined with blind biopsies of the nasopharynx and, if negative, tonsillectomies. Patients with occult head and neck cancers typically have upper neck nodes, but isolated level IV or V nodes can also occur. Routine measurement of the tumour markers CEA, CA125, CA19-9 and CA15-3 are usually unhelpful in establishing the primary tumour site because they are not truly tumour-site specific (Pavlidis et al., 2003).
Stage classification Patients presenting with CUP should be staged according to the AJCC cancer staging manual (AJCC, 2002) for the most likely primary site affected if feasible. As such not all patients are immediately classified as having stage IV disease. It is important to note, for example, that patients with cervical lymphadenopathy are not classified as having M1 disease unless there is mediastinal lymphadenopathy or distant metastasis.
Initial presentation
History and examination Tumour markers: PSA, CEA, CA19-9, CA125, βhCG, αFP Mammogram Chest X-ray CT thorax, abdomen and pelvis Biopsy* Immunohistochemistry: IHC minimum = CLA (lymphoma), βhCG, αFP.
Further investigations: GI endoscopy if symptoms / CT abnormal / iron deficient Testicular ultrasound if markers / biopsy are suggestive Thyroid ultrasound if raised thyroglobulin ± thyroid mass Panendoscopy ± PET if cervical lymphadenopathy Other IHC stains depend on presentation: TTF-1 (lung), PSA (prostate), CK7 and CK20 (GI), ER and PgR (breast)
CUP diagnosis
PS > 2
Supportive care
Unfavourable subgroup
Favourable subgroup
PS 0 or 1
See Table 38.3
Consider chemotherapy in exceptional cases
Predominant lung metastases
Platinum-based regimen (if breast or thyroid cancer thought unlikely)
Brain metastases
Isolated tumour: Neurosurgical or stereotactic RT >1 tumour: Whole brain RT (+/− stereotactic boost) and supportive care
Bone metastases
RT. Endocrine therapy Consider combination chemotherapy
Predominant liver metastases or malignant ascites
Treat with 5-FU-containing regimen Consider possibility of liver resection if isolated tumour
Figure 38.1. Diagnostic and treatment algorithm for CUP. ∗ Consider the value of obtaining a tissue diagnosis if epithelial markers are positive with radiological evidence of malignancy, PS > 2, more than three metastatic sites, and liver metastases. 5-FU = 5-fluorouracil; αFP = alpha feto-protein; βhCG = beta human chorionic gonadotrophin; CA = cancer antigen; CEA = carcino-embryonic antigen; CK = cytokeratin; CT = computed tomography; CLA = common leukocyte antigen; ER = oestrogen receptor; GI = gastrointestinal; IHC = immunohistochemistry; PET = positron emission tomography; PgR = progesterone receptor; PS = WHO performance status; PSA = prostate-specific antigen; RT = radiotherapy; TTF-1 = thyroid transcription factor 1.
445
Paul Shaw and Tom Crosby
Table 38.3. Standardised treatments for ‘favourable’ CUP subsets CUP subset
Recommended treatment
Poorly differentiated carcinoma in midline (extragonadal
Platinum-based regimen (germ cell)
germ cell syndrome) Woman with papillary serous or serous adenocarcinoma of
As for FIGO stage III ovarian cancer
peritoneal cavity Woman with adenocarcinoma involving axillary nodes
As for breast cancer
Squamous cell carcinoma of cervical nodes
Radical radiotherapy/surgery/adjuvant chemotherapy
Isolated inguinal lymphadenopathy (squamous cell
Surgical dissection ± radiotherapy
carcinoma) Poorly differentiated neuroendocrine carcinoma
Platinum/etoposide regimen
Man with blastic bone metastases and elevated PSA
Endocrine treatment as for prostate cancer
Single metastasis only
Definitive local treatment (surgery or radiotherapy)
Adapted from Pavlidis et al. (2003).
Treatment Table 38.4. ‘Unfavourable’ CUP prognostic subsets
Little has changed since Neumann and Nystrom’s statement more than 20 years ago that patients with metastatic cancer of unknown origin represent a therapeutic dilemma (Neumann and Nystrom, 1982). The US National Cancer Institute suggests that, when a probable primary site is found in patients who presented initially with CUP, standard treatment for that site should be given (National Cancer Institute, www. cancer.gov/cancertopics/types/unknownprimary, accessed September 2006). In an attempt to standardise the treatment of patients with CUP for whom a primary site is not discovered, guidelines have been written based, where possible, on available evidence (European Society of Medical Oncology, 2005; Pavlidis et al., 2003). Specific subsets of CUP have been identified that are described as having a ‘favourable’ or ‘unfavourable’ prognosis. For patients within the favourable group, there are suggested treatments (see Table 38.3 and Figure 38.1). However, most CUPs fall into the unfavourable subsets (see Table 38.4 and Figure 38.1). For this group, the most effective chemotherapy regimen is unknown.
Surgery Surgery may form an important component of the management of CUP, particularly in the treatment of favourable subsets (Table 38.3). It may also have a role 446
CUP ‘unfavourable’ subgroups Metastatic adenocarcinoma in liver or multiple sites Malignant ascites (non-papillary serous adenocarcinoma) Multiple cerebral metastases (adeno- or squamous carcinoma) Multiple lung/pleural metastases (adenocarcinoma) Multiple metastatic bone metastases (adenocarcinoma) Adapted from Pavlidis et al. (2003).
in the management of isolated brain metastases or be required for optimal palliation (e.g. prophylactic surgery for bone metastases or epidural spinal cord decompression). Multidisciplinary team work is vital between medical and surgical specialities to optimize patient care.
Radiotherapy Radiotherapy treatment may be indicated in the following circumstances: r As part of standard treatment for a probable primary cancer presenting as CUP. r Squamous cell cancer of cervical lymph nodes or isolated inguinal lymphadenopathy. r If there is a single metastatic site. r Palliative treatment (e.g. bone pain, epidural spinal cord compression, SVCO).
Cancer of unknown primary
Chemotherapy A wide variety of chemotherapy regimens for CUP has been reported in phase II studies, with an overall response rate of around 30%. It is not surprising that a recent audit showed 18 different chemotherapy regimens were used to treat 37 patients with CUP (Shaw et al., 2006). Of the few published randomised studies, one showed an improved response rate when cisplatin and epirubicin were added to mitomycin C in patients with adenocarcinoma of unknown primary (Falkson and Cohen, 1998). Pavlidis et al. (2003), suggesting that treatment should be individualised according to clinical subsets. For patients with favourable CUP subsets, specific chemotherapy is advised as shown in Table 38.3 and Figure 38.1. For patients with unfavourable CUP, treatment is palliative and may include chemotherapy if the patient’s performance status is adequate. There is little evidence to suggest a specific regimen; choice of regimen is undoubtedly influenced by local experience and practice, as well as, rightly or wrongly, the likely site of primary disease. Commonly used regimens include epirubicin, cisplatin and 5-FU (ECF) or other infusional 5-FU-based regimens. More recent combinations include newer agents such as capecitabine, gemcitabine and the taxanes. Patients with unfavourable CUP, and performance status 0 or 1, can be divided into one of four groups: 1. Predominant lung metastases – consider platinumbased treatment regimen (assuming breast and thyroid cancer thought unlikely). 2. Brain metastases – consider whole brain radiotherapy ± stereotactic boost for patients with two to three brain metastases (Andrews et al., 2004). 3. Bone metastases – consider radiotherapy for painful sites and a trial of hormone treatment (breast or prostate cancer) or possible combination chemotherapy. Bisphosphonates can be useful to palliate widespread bone pain when there is no single site amenable to radiotherapy. 4. Predominant liver metastases or malignant ascites – consider a 5-FU-based regimen.
toxicity. Patients with unfavourable CUP and performance status of 2 or higher with either liver metastases or more than three metastastic sites of disease are probably best treated with supportive care alone, reserving radiotherapy for symptomatic control.
Prognosis The overall median survival in a study of 657 CUP patients in whom just fewer than 50% underwent chemotherapy was 11.0 months with only 1.5% alive at 5 years (Abbruzzese et al., 1994). Patients in a good prognostic group (PS 0 or 1 without liver metastases), intermediate prognostic group (PS > 1 or liver metastases), or poor prognostic group (PS > 1 and liver metastases) had a median survival of 10.8, 6, or 2.4 months, respectively (Culine et al., 2002; van de Wouw et al., 2004). Perhaps not surprisingly, patients treated with radiotherapy do not have a significantly better survival (MS = 3.0 months; 95% CI = 1.8 to 4.2 months) than those who had no radiotherapy (MS = 2.0 months; 95% CI = 1.2 to 2.7 months, p = 0.135; Shaw et al., 2006).
Prognostic factors Poor prognostic factors include the following (Abbruzzese et al., 1994; Culine et al., 2002; Pasterz et al., 1986; van de Wouw et al., 2004): r More than three metastatic sites. r Liver metastases. r Performance status 2 or worse. r Male gender. r Non-lymph-node metastases. r Raised LDH. r ‘Unfavourable’ CUP subgroup (see Table 38.4).
Current trials There is currently no trial in the UK recruiting from this patient population but planned clinical and translational studies are planned in the near future and should be supported.
Areas of current interest Palliation Unfortunately, the majority of patients with CUP have incurable disease at diagnosis, and the focus of treatment should be to ensure useful palliation with the least
In the future, the treatment for CUP patients is likely to move away from that based on most likely anatomical site of origin to one based on molecular profiling predicting for treatment response. 447
Paul Shaw and Tom Crosby
REFERENCES Abbruzzese, J. L., Abbruzzese, M. C., Hess, K. R. et al. (1994). Unknown primary carcinoma: natural history and prognostic factors in 657 consecutive patients. J. Clin. Oncol., 12, 1272–80. Abbruzzese, J. L., Abbruzzese, M. C., Lenzi, R. et al. (1995). Analysis of a diagnostic strategy for patients with suspected tumors of unknown origin. J. Clin. Oncol., 13, 2094–103. AJCC. (2002). AJCC Cancer Staging Handbook, ed. F. L. Greene, D. L. Page, I. D. Fleming et al., 6th edn. New York: Springer. Andrews, D. W., Scott, C. B., Sperduto, P. W. et al. (2004). Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase 3 results of the RTOG 9508 randomised trial. Lancet, 363, 1665–72. Cancer Research UK. (2004). CancerStats Monograph 2004. London: Cancer Research UK. Casciato, D. A. (2006). In Manual of Clinical Oncology, ed. D. A. Casciato, 5th edn. Philadelphia, PA: Lippincott Williams and Wilkins, pp. 402–14. Culine, S., Kramar, A., Saghatchian, M. et al. (2002). Development and validation of a prognostic model to predict the length of survival in patients with carcinomas of an unknown primary site. J. Clin. Oncol., 20, 4679–83. European Society for Medical Oncology. (2005). ESMO minimum clinical recommendations for diagnosis, treatment and follow-up of cancers of unknown primary site (CUP). Ann. Oncol., 16 (Suppl. 1), i75–6. Falkson, C. I. and Cohen, G. L. (1998). Mitomycin C, epirubicin and cisplatin versus mitomycin C alone as therapy for carcinoma of unknown primary origin. Oncology, 55, 116–21. Greco, F. A. and Hainsworth, J. D. (2001). In Cancer: Principles and Practice of Oncology, ed. V. T. DeVita Jr., S. Hellman and
448
S. A. Rosenberg, 6th edn. New York: Lippincott Williams and Wilkins, pp. 2537–60. Hainsworth, J. D. and Greco F. A. (1993). Treatment of patients with cancer of an unknown primary site. N. Engl. J. Med., 329, 257–63. Mintzer, D.M, Warhol, M., Martin, A. et al. (2004). Cancer of unknown primary: changing approaches. A multidisciplinary case presentation from the Joan Karnell cancer centre of Pennsylvania hospital. The Oncologist, 9, 330–8. Muir, C. (1995). Cancer of unknown primary site. Cancer, 75 (1 suppl.), 353–6. Neumann, K. H. and Nystrom, J. S. (1982). Metastatic cancer of unknown origin: nonsquamous cell type. Sem. Oncol., 9, 427–34. Panza, N., Lombardi, G., De Rosa, M. et al. (1987). High serum thyroglobulin levels. Diagnostic indicators in patients with metastases from unknown primary sites. Cancer, 60, 2233–6. Pasterz, R., Savaraj, N. and Burgess, M. (1986). Prognostic factors in metastatic carcinoma of unknown primary. J. Clin. Oncol., 4, 1652–7. Pavlidis, N., Briasoulis, E., Hainsworth, J. et al. (2003). Diagnostic and therapeutic management of cancer of an unknown primary. Eur. J. Cancer, 39, 1990–2005. Shaw, P., Adams, R., Jordan, C. et al. (2007). A clinical review of the investigation and management of carcinoma of unknown primary in a single cancer network. Clin. Oncol., 19, 87–95. van de Wouw, A. J., Jansen, R. L., Griffioen, A. W. et al. (2004). Clinical and immunohistochemical analysis of patients with unknown primary tumour. A search for prognostic factors in UPT. Anticancer Res., 24, 297–301. Varadhachary, G. R., Abbruzzese, J. L. and Lenzi, R. (2004). Diagnostic strategies for unknown primary cancer. Cancer, 100, 1776–85.
39
THE USE OF RADIOTHERAPY IN THE TREATMENT OF BENIGN CONDITIONS Alison Brewster
Introduction
Ankylosing spondylitis
Early in the past century, radiation therapy was applied to a whole range of non-malignant conditions largely because of its immunosuppressive and antiproliferative properties. With the evidence of its potential to cause serious and irreversible long-term side effects and the development of more effective medical antiinflammatory therapies, the place of radiotherapy in the treatment of benign conditions has declined. Although described as benign, many non-malignant conditions can have grave consequences for the patient in terms of symptoms and quality of life. In certain clinical conditions and in carefully selected patients, lowdose radiation therapy should be considered, provided that the patient is adequately informed of the potential risks and benefits. All patients should be managed in conjunction with other specialists to ensure that all appropriate therapeutic options have been considered. There is a lack of published data, but overall, the risks are small as long as low doses are combined with careful technique. Sensitive normal tissues such as the eye, gonads, thyroid and bone marrow should be avoided, and radiation therapy in general should not be administered to children and young adults. CourtBrown and Doll (1965) demonstrated a 9.5-fold increase in leukaemia in patients with ankylosing spondylitis who were treated with radiotherapy. The risks were greater when larger radiation fields and higher doses were involved. There is also documented evidence of an increased incidence of thyroid cancer and soft tissue or bone sarcoma.
Description. Ankylosing spondylitis is inflammatory arthritis of the spine and sacro-iliac joints, which predominantly affects young males. Figure 39.1 shows ankylosing spondylitis affecting the lumbar spine. Administration. Radiotherapy should be administered in small fields to affected joints, or in 12 × 8 cm fields to the whole spine in stages. The recommended dose is 10 Gy in ten fractions with orthovoltage. Response. Patient response is dose-dependent, but 60 to 90% of patients report a good response. Comments. The incidence of leukaemia is increased 9.5-fold, with 52 reported cases out of 14 000 treated. There are some reports of re-activated TB.
Conditions in which radiation treatment could be considered This list is not exhaustive, but it includes those conditions in which there is a reasonable consensus that radiation could be considered. For information on other conditions, refer to Order and Donaldson (1998) or Massullo (1999).
Arterial restenosis Description. Restenosis of coronary arteries occurs after balloon angioplasty and occurs in about 36 to 44% of cases. Administration. Postangioplasty irradiation should be administered using 15 to 20 Gy with 192 Ir. Response. Radiotherapy gives a reduction of restenosis incidence from 54% to between 17 and 25%. Comments. Several clinical trials are in progress to confirm preliminary findings.
Cerebral arteriovenous malformation (AVM) Description. Cerebral AVM occurs when a network of arterial channels bypasses the capillaries and shunts oxygenated blood into the venous system. Intraluminal hypertension can occur and result in aneurysm formation. AVM accounts for 1 to 2% of strokes and can cause headache, epilepsy, cranial nerve palsy and raised intracranial pressure. Administration. The dose should be 15 to 30 Gy to the 50% isodose using stereotactic radiosurgery. Response. Complete obliteration of AVM is seen in 71 to 90% of cases within 2 years. 449
Alison Brewster
Isodose % 103 95 10
Figure 39.2. Radiotherapy plan for thyroid eye disease. Parallel-opposed lateral fields have been used because the disease is bilateral. The anterior border of the radiation fields has been set specifically to allow the 10% isodose to lie at the back of the lenses.
(a)
(b)
Figure 39.1. Radiographs of the lumbar spine showing ankylosing spondylitis. (a) A–P. (b) Lateral. Used with permission from eMedicine.com, Inc., 2006.
Comments. Results are best when the lesion is 1 to 2 cm in size and all feeder vessels are irradiated.
Desmoid tumours Description. Desmoid tumours are also known as aggressive fibromatosis, where low-grade invasive nonmetastasising tumours arise in connective tissue. They tend to merge into local structures and resection margins are often positive, resulting in high rates of local recurrence. Administration. The recommended dose is 50 Gy in 2 Gy fractions after local resection. Response. There is a 70% local control rate. Comments. Regression can continue to occur up to 2 years after radiotherapy.
Thyroid eye disease Description. Thyroid eye disease occurs as an autoimmune response in which activated T cells invade the orbit and stimulate the production of glycosaminoglycan in fibroblasts. Administration. The patient should be treated with a single lateral field angled 5◦ away from the lens or 450
Half-beam blocking has been used anteriorly.
with half-beam blocking, or using an anterior field with central-axis beam blocking. The dose is 20 Gy in 8 to 10 daily fractions, 5 fractions a week, using 4 to 6 MV photons. Figure 39.2 shows a CT-planned radiotherapy treatment for thyroid eye disease. Response. Of patients treated in this way for thyroid eye disease, 75% report an improvement. Comments. Surgery is indicated in the presence of rapidly progressive optic neuropathy. Steroids are used as an alternative to radiotherapy but the benefits are short lived. A 10% incidence of cataracts has been found in long-term follow-up studies. The use of radiotherapy is examined in NICE Interventional Procedure Guidance 148, which supports its use in patients for whom other treatments are inadequate or associated with significant side effects.
Heterotopic bone formation (HBF) Description. HBF is postoperative ossification that can develop 3 to 6 weeks after surgery and calcify within about 8 weeks. This condition occurs in up to 43% of patients who have a hip replacement but is only clinically significant in 9% (Balboni et al. 2006). Administration. The dose is 8 Gy in a single fraction within 4 days of surgery or 4 to 6 hours before reoperation. Response. Radiotherapy prevents recurrent HBF in 52 to 98% cases.
The use of radiotherapy in the treatment of benign conditions
Comments. The long-term risks are quoted as 1 in 3000 to 5000, but the risks depend on age, treatment site and reproductive status.
Keloid Description. A keloid is a fibrous overgrowth associated with scar formation. Administration. Patients should be treated with a 6 Gy single exposure with 100 kV photons if the keloid is less than 2 cm. Larger keloids have a higher risk of recurring and so in these cases a dose schedule of 12 Gy in three fractions with 6 to 8 MeV electrons is recommended. Consider using 1 cm wet gauze to increase the dose to the skin. Use a minimal margin around the scar and include sutures. Alternatively use an iridium implant that gives 20 Gy at 2.5 mm over 24 hours. Response. Radiotherapy is effective in 80% of keloid cases. Comments. The procedure first requires a careful excision of the existing scar, followed by radiotherapy within 1 to 10 days.
Response. Control is achieved in 75% of cases. Comments. CT scanning is helpful, but biopsy is required to exclude lymphoma. Steroids should be considered as first-line therapy, followed by radiotherapy if the patient fails to respond to steroids.
Peyronie’s disease Description. Peyronie’s disease is an idiopathic deformity caused by single or multiple plaques within the corpora cavernosa, which results in sexual dysfunction and pain. Administration. The recommended dose is 1 Gy weekly for 4 weeks with either orthovoltage or 3 to 5 MeV electrons. Response. Patient responses of between 38 and 84% have been reported. Comments. If possible the treatment should be given during the active (inflammatory) phase of the disease, because radiotherapy appears to be less beneficial if the disease has been present for a long time (Van Helvoirt et al., 1999).
Pterygium Macular degeneration Description. Age-related macular degeneration is a leading cause of blindness in developed countries and affects up to 28% of patients over 75 years old. Proliferation of choroidal vessels causes subretinal haemorrhages and retinal detachment. Administration. Consider giving the patient 15 Gy in five fractions in 1 week using a unilateral 6 MV field with half-beam blocking. Avoid the contralateral eye and pituitary gland. Response. Visual acuity is improved or stabilised in 66% of cases at 12 month follow-up. Comments. Initial trials show promise, but NICE Interventional Procedure Guidance 49 concludes that there is insufficient evidence to recommend use outside of clinical trials.
Orbital pseudotumour Description. An orbital pseudotumour is a benign idiopathic orbital inflammation causing periorbital swelling, decreased orbital mobility and pain. Administration. The recommended dose is 25 Gy in 12 fractions with 4 to 6 MV photons and half-beam blocking.
Description. Pterygium is a growth of fibrovascular tissue on the cornea, which can impair visual acuity. Administration. Use a beta-emitting 90 Sr applicator. An energy of 2.2 MeV gives 50% at about 1 mm and 10% at 3.3 mm. The recommended dose is 7 Gy a week for three fractions, to start within 24 hours of surgery. Response. Local recurrence rates vary from 3 to 16% compared with 8 to 50% in patients without postoperative radiotherapy. Comments. Radiation appears to reduce vascularisation at the operative site and so decrease recurrence.
Synovitis Description. Synovitis is an inflammatory reaction that affects the synovial lining of the joints and is of variable aetiology. Administration. The patient should be given an intraarterial injection of 90 Y to give 185 MBq. Response. Response rates appear to be comparable to that of surgical synovectomy. Comments. Radiotherapy for synovitis should be restricted to patients in whom other measures have failed. 451
Alison Brewster
Thyrotoxicosis Description. Thyrotoxicosis is hyperthyroidism due to Graves’ disease or toxic multinodular goitre. Administration. The recommended dose is 400 to 550 MBq of 131 I 3 days after stopping antithyroid medication, which can be restarted 3 days later. The patient should be advised to avoid close contact with young children or pregnant women for 21 to 24 days depending on the dose prescribed. Response. Normal thyroid function returns in 50 to 70% of patients within 6 weeks; 15 to 20% develop hypothyroidism after 2 years and so these patients need long-term follow-up of thyroid function. Comments. Severe thyrotoxicosis should be controlled first. Radioiodine is indicated after relapse on carbimazole or propylthiouracil. Prolonged use of antithyroid medication can be associated with irreversible agranulocytosis and hepatic dysfunction. Radioiodine should be used with great caution in patients with thyroid eye disease because of the risk of exacerbation (Bartalena et al., 1998). The patient should be provided with written information about precautions following radioiodine dosage. Precautions include restricting time spent in contact with people, especially children and pregnant women, avoiding sharing a bed, reducing time spent on public or shared private transport, avoiding sharing crockery or cutlery and taking time off work. The length of time for these restrictions to apply is individualised depending on the dose of radioiodine, the nature of work and so on.
stem cells. It has a relatively long and mild clinical course. Administration. The recommended dose is 100 to 150 MBq i.v. of 32 P, providing that the patient’s platelet count is greater than 100 × 109 /l and the WBC is greater than 3 × 109 /l. Response. Response rates of around 80% have been reported that may be sustained for 6 months or more. Comments. The patient may require repeated doses of 32 P, which should be reserved for patients in whom venesection has failed and hydroxyurea has been poorly tolerated, because there is evidence that 32 P may increase the risk of leukaemic transformation.
REFERENCES Balboni, T.A, Gobezie, R. and Mamon, H. J. (2006). Heterotopic ossification: pathophysiology, clinical features and the role of radiotherapy for prophylaxis. Int. J. Radiat. Oncol. Biol. Phys., 65, 1289–99. Bartalena, L., Marcocci, C., Bogazzi, F. et al. (1998). Relation between therapy for hyperthyroidism and the course of Graves’ ophthalmopathy. N. Engl. J. Med., 338, 73–8. Court-Brown, W. M. and Doll, R. (1965). Mortality from cancer and other causes after radiotherapy for ankylosing spondylitis. B.M.J., 2, 1327–32. Massullo, V. ed. (1999). Treatment of benign disease. Sem. Rad. Oncol., 9 (2). Order, S. E. and Donaldson, S. S. (1998). Radiation Therapy of Benign Diseases – A Clinical Guide, 2nd edn. Berlin: Springer. Van Helvoirt, R. P., Battermann, J. J., Scheijmans, L. J. E. E. et al. (1999). Radiotherapy in Peyronie’s disease: indication for a dose-response relationship. Radiother. Oncol., 53 (Suppl. 1), S12.
Polycythaemia vera
FURTHER READING
Description. Polycythaemia vera is uncontrolled proliferation of erythroid, granulocyte and megakaryocyte
Royal College of Physicians. (1995). Guidelines on the Use of Radioiodine in the Management of Hyperthyroidism. London: Royal College of Physicians.
452
MULTIPLE CHOICE QUESTIONS
1. A 20-year-old man presents with a metastatic testicular tumour. Which of the following features would put him into the poor-risk category? a) Pure seminoma with 20 lung metastases b) Pure seminoma with hCG 50× upper limit of normal (ULN) c) Choriocarcinoma with 20 lung metastases d) Embryonal germ cell tumour (GCT) with αFP 50× ULN e) Mixed GCT with 10 lung and 2 liver metastases 2. With regard to cancer-related emergencies, which of the following statements is correct? a) Syndrome of inappropriate antidiuretic hormone (SIADH) gives rise to hyponatraemia and reduced plasma osmolarity in the presence of inappropriately dilute urine b) In patients with SIADH who do not respond to fluid restriction, cautious administration of intravenous sodium chloride should be considered c) The addition of surgery to radiotherapy has been shown to improve functional outcome compared to surgery alone in the treatment of spinal cord compression d) In good-performance-status patients with solitary intracerebral metastases, radiotherapy with 12 Gy in two fractions is equivalent to 30 Gy in ten fractions e) Patients at high risk from neutropenic sepsis should receive intravenous ciprofloxacin and co-amoxiclav on admission to hospital 3. A 45-year-old man presents with an enlarging mass on the penis. Biopsy reveals invasive squamous carcinoma. The tumour is invading the corpus cavernosa and there are two superficial inguinal node metastases. What is the stage? a) T3 N1 MX b) T1 N1 MX c) T3 N2 MX d) T2 N1 MX e) T2 N2 MX
4. Xeroderma pigmentosa is associated with which of the following? a) Autosomal dominant inheritance b) Jaw cysts c) Medulloblastoma d) A defect of nucleotide excision repair e) A decreased risk of skin cancer 5. A patient presents with multiple bilateral pulmonary opacities suggestive of malignancy but with no obvious underlying primary site of disease. A CT-guided biopsy reveals adenocarcinoma on which the following immunohistochemical tests are performed: cytokeratin-7 (CK7) positive, cytokeratin-20 (CK20) negative, thyroid transcription factor (TTF-1) positive, thyroglobulin negative. The most likely underlying primary site of disease is which of these? a) Colon b) Bladder c) Lung d) Thyroid e) Pancreas 6. A 53-year-old woman presents with a painless rightsided thyroid nodule and normal TFTs. There is no family history of thyroid disease or endocrine malignancy. US confirms a solitary 3 × 2 cm hypoechoic nodule with no abnormal nodes. FNA is reported as showing a follicular neoplasm. The next most appropriate step is: a) Serum thyroglobulin measurement b) Repeat FNAC c) To request CT scan of the neck, thorax and abdomen with intravenous contrast d) To advise a right-sided thyroid lobectomy e) Total thyroidectomy and central lymph node dissection 7. A 54-year-old man presents with a 6-month history of a hoarse voice and no other ENT symptoms. Clinic room examination reveals a proliferative lesion on the left false cord with normal true cord mobility and palpable lymphadenopathy in level II on the
453
Multiple choice questions
8.
9.
10.
11.
454
ipsilateral side. EUA and CT scan confirm involvement of the false cord with extension along the aryepiglottic fold towards the epiglottis and involvement of the pre-epiglottic space. There is a left level II lymph node measuring 3 cm. There is no evidence of distant metastases. Biopsy from the false cord shows moderately differentiated squamous carcinoma. According to the 6th edition TNM classification, what stage disease is this? a) T2 N1 M0 b) T2 N2a M0 c) T2 N2b M0 d) T3 N1 M0 e) T3 N2a M0 A 54-year-old patient has hepatocellular carcinoma associated with hepatitis C infection. Investigations reveal three tumours in the right lobe of the liver: 2 cm, 2 cm and 3 cm in size. There is no evidence of metastatic disease. Which of the following options is the treatment of choice? a) Radiofrequency ablation b) Liver transplantation c) Transarterial chemoembolisation d) External beam radiotherapy e) None of the above is superior to the others In electron therapy, which of the following statements is correct? a) The therapeutic range (on the central axis) is approximately E/2 cm (where E is the energy of the incident electron beam in MeV) b) Unlike kV photons, electron treatments rarely require the use of bolus c) The 80% isodose is wider at depth than at the surface d) Hot and cold spots caused by inhomogeneities can be easily corrected for, provided their position is known e) The 10% isodose is wider at depth than at the surface Which of the following conditions is most typically associated with an increase in colorectal adenocarcinomas? a) Neurofibromatosis b) Xeroderma pigmentosa c) A germline mutation in BRCA1 d) Turcot’s syndrome e) A genetic defect in the Hedgehog pathway A 62-year-old woman presents with postmenopausal bleeding. A hysteroscopy shows a tumour in the uterus which on biopsy is found to be adeno-
12.
13.
14.
15.
carcinoma. An MRI scan suggests this is a FIGO stage IB tumour. What initial treatment would you recommend? a) TAH, BSO and pelvic lymphadenectomy b) TAH, BSO c) Radical radiotherapy d) Intrauterine progesterone e) TAH, BSO and infracolic omentectomy A 54-year-old woman presents with an enlarging mass on the vulva that has been present for 12 months. A biopsy confirms this is squamous carcinoma. There is no other finding on examination. The tumour involves the right side of the vulva, measures 8 cm in maximum diameter and extends to within 3 mm of the anal margin. She does not wish to have an anovulvectomy. What is the most appropriate treatment? a) Local excision b) Radical vulvectomy and groin node dissection c) Radical radiotherapy to include the primary tumour with an electron field to the perineum d) Radical chemoradiotherapy to the vulva and groin nodes e) Symptomatic treatment A 65-year-old man with pancreatic carcinoma undergoes a Whipple’s operation. Final histology shows a T3 N1 (5 of 15 nodes positive) grade 3 adenocarcinoma with vascular invasion, distance to closest resection margin is 1 mm. Based on current evidence (and outside the context of a clinical trial), what is the most appropriate management? a) No adjuvant therapy b) Adjuvant 5-FU chemotherapy c) Adjuvant gemcitabine chemotherapy d) Adjuvant chemoradiation e) Re-operation A patient presenting with SVCO is given oxygen, commenced on high-dose steroids and his condition stabilises. CT scanning confirms the presence of a large superior mediastinal mass. What is the most appropriate management at this stage? a) Radiotherapy to the mediastinum using parallel opposed anterior and posterior fields; dose 20 Gy in five fractions over 5 days b) Chemotherapy with BEP c) Referral for biopsy d) Anticoagulation e) Insertion of a percutaneous stent into the IVC For someone on MST 60 mg b.d., what should the breakthrough dose of oral morphine be?
Multiple choice questions
16.
17.
18.
19.
a) 20 mg b) 10 mg c) 30 mg d) 40 mg e) 60 mg An otherwise fit and well 61-year-old man presents with dysphagia and is found to have an adenocarcinoma of the oesophagus. A CT scan shows no evidence of metastatic disease. An EUS confirms the tumour to be from 36 to 39 cm ab oral, extending into the right crus of the diaphragm together with malignant lymphadenopathy in the peritumoral region, and along the lesser curve (left gastric). What stage is this and what is the standard treatment in the UK? a) Stage T3 N1 M0: 4 cycles epirubicin, cisplatin and capecitabine (ECX) chemotherapy prior to surgery b) Stage T3 N1 M1b: definitive chemoradiotherapy c) Stage T4 N1 M0: 2 cycles cisplatin and 5-FU followed by surgery d) Stage T4 N1 M1b: 2 cycles cisplatin and 5-FU followed by surgery e) Stage T3 N1 M0: 2 cycles cisplatin and capecitabine followed by surgery The age-specific incidence in England and Wales is decreasing for which of these cancers? a) Pancreatic cancer b) Gastric cancer c) Colorectal cancer d) Oesophageal cancer e) Testicular germ cell tumours Criteria for urgent referral with suspicion of colorectal cancer include: a) Rectal bleeding with anal symptoms in a 62-yearold b) Eight weeks change in bowel habit without rectal bleeding in a 45-year-old c) Persistent rectal bleeding without anal symptoms in a 35-year-old d) Unexplained iron deficiency anaemia in men Hb < 11 g/dl e) Left iliac fossa mass A previously fit 46-year-old woman is found to have a 4 cm basaloid carcinoma of the anal canal. On examination she has a palpable lymph node in her left groin. There is no evidence of distant metastatic disease. An FNA of the groin node is negative for malignancy. What is the correct initial treatment? a) Surgery alone because basaloid tumours are not radiosensitive
20.
21.
22.
23.
b) Cisplatin/5-FU concurrent with radiotherapy to the GTV plus 3 cm c) 5-FU plus mitomycin concurrent with radiotherapy to the GTV plus 3 cm d) 5-FU and mitomycin concurrent with radiotherapy to the stage-defined CTV e) Cisplatin and 5-FU concurrent with radiotherapy to the stage-defined CTV Regarding radical radiotherapy for prostate cancer, which of the following statements is correct? a) The GTV to PTV margin is usually 2 cm b) 3D conformal radiotherapy allows dose escalation to 64 Gy c) If included in the treatment, an appropriate seminal vesicle dose is 56 Gy in 28 fractions d) The use of neoadjuvant hormone therapy causes greater rectal toxicity e) A three-field plan with an anterior field and two laterals causes greater rectal toxicity than a threefield plan with an anterior field and two posterior obliques An otherwise fit 45-year-old woman presents with epigastric discomfort. An endoscopy reveals a smooth 3 cm submucosal lesion at the gastric cardia. A CT scan is consistent with a gastrointestinal stromal tumour. What would you recommend next? a) Surgical resection b) Endoscopic ultrasound guided FNA c) Imatinib d) A D2 total gastrectomy e) A PET scan The MRC CR07 Study has demonstrated: a) An improvement in overall survival with preoperative radiotherapy b) A reduction in CRM positivity with pre- over postoperative radiotherapy c) A significant reduction in local recurrence confined to lower-third tumours d) A significant reduction in local recurrence confined to lymph-node-positive tumours e) A significant reduction in local recurrence confined to CRM negative tumours An 84-year-old woman presents to an ophthalmologist with diplopia and reduced visual acuity in her left eye. A CT scan shows a left retro-orbital mass. She had a mastectomy 10 years previously for a left breast cancer, followed by tamoxifen for 5 years. Other staging scans are negative. The CA15-3 is 310 u/ml. The original tumour was ER+ve. The most appropriate management now is: 455
Multiple choice questions
a) Anastrazole b) Urgent palliative radiotherapy to orbit c) Urgent palliative radiotherapy to orbit and anastrazole d) Palliative chemotherapy e) Surgical excision of the mass 24. Which of the following statements about opioids is correct? a) Morphine is glucuronated in the liver, so renal impairment does not lead to opioid toxicity b) Tachycardia is a characteristic feature of opioid toxicity c) Patients with opioid toxicity may think they have seen animals running under their bed d) Low-dose morphine is classed as a ‘weak opioid’ on the WHO analgesic ladder e) Opioid toxicity typically causes pupil dilatation 25. A fit 60-year-old woman undergoes surgery for a carcinoma of the tranverse colon. Histology shows a grade 2 adenocarcinoma, T3 N2 (4 out of 14 nodes involved) with vascular invasion and perineural invasion. There is no evidence of metastatic disease. What is the most appropriate management? a) Weekly 5-FU/folinic acid b) Modified de Gramont 5-FU c) Oxaliplatin – modified de Gramont d) Oxaliplatin – capecitabine e) Irinotecan – modified de Gramont 26. A 45-year-old man presents with gradual onset of right-sided weakness. An MRI scan of the brain shows an enhancing tumour in the right frontal and parietal region. The tumour is excised surgically and histology shows a glioma with frequent mitoses and necrosis. It is decided to treat with radical postoperative radiotherapy. Which of the following statements is correct? a) Treating each field each day, Monday to Friday, 5 days a week, the dose to the ICRU reference point should be 54 Gy in 30 fractions b) Craniospinal axis radiotherapy improves survival by reducing the risk of recurrence elsewhere in the central nervous system c) CT and MRI image co-registration is not usually helpful in planning because of difficulties correcting for scale, rotation and lateral translation d) The GTV is the contrast-enhancing tumour as seen on imaging e) If megavoltage photons are used, alopecia is unlikely because of the skin sparing in the buildup region
456
27. Which of the following is associated with a poor prognosis in neuroendocrine tumours? a) Well-differentiated tumour b) High Ki-67 index c) Origin in the appendix d) Bronchial origin e) Absence of atypical cells 28. Which of the following statements about radiotherapy for thyroid eye disease is correct? a) It should be given prophylactically b) Proptosis usually pushes the lenses so far forward that their radiation dose can be ignored c) NICE recommends avoiding radiotherapy because of the risk of long-term side effects d) The long-term risk of cataract is 10% e) The usual dose is 40 Gy in 20 fractions over 4 weeks 29. A 2-year-old girl with a family history of retinoblastoma undergoes screening. A tumour is found in the anterior part of the retina that is three disc diameters in size. Out of those listed, what is the best treatment option? a) Cryotherapy b) External beam radiotherapy c) Chemotherapy with ifosfamide, vincristine and doxorubicin d) Enucleation e) Surveillance 30. A 43-year-old man presents with a swelling in the right thigh. Investigations show a soft tissue mass within the quadriceps muscle. A biopsy is undertaken, followed by surgery to remove the tumour. Histology shows a leiomyosarcoma with 20 mitoses per high-power field and more than 50% necrosis. The tumour measured 3 cm diameter and had an excision margin of 25 mm. According to the TNM 6th edition classification, what stage is the tumour? a) pT1 NX MX b) pT2 N0 M0 c) pT1 N0 M0 d) R1 NX MX e) G3 T2 N0 M0 31. A 63-year-old male is being planned for potentially curative definitive platinum-based chemoradiation for a squamous cell carcinoma starting at 30 cm ab oral using a two-phase technique: phase 1 with anterior–posterior parallel opposed fields; phase 2 with three fields, one anterior and two posterior obliques. You are asked to review the plans: PTVmax = 106%; PTVmin = 95%; the spinal cord V45 = 20%;
Multiple choice questions
combined lung V20 = 10%; and heart V30 = 70%. What action do you recommend? a) Accept the current plan b) Increase the dose contribution from phase 2 c) Increase the contribution of the anterior field during phase 1 d) Increase the dose contribution from phase 1 e) Increase the contribution of the posterior field during phase 1 32. A 45-year-old maths teacher has just completed adjuvant treatment for a right breast cancer. She had a 2.0 cm grade 3 tumour, with 1 of 17 nodes involved. The tumour was ER+ve but HER-2 −ve. She has had 6 cycles of FEC chemotherapy and radiotherapy to the conserved breast. She has started tamoxifen. She wants to know what her prognosis is. According to the Nottingham prognostic index, what is her 10year survival likely to be? a) 51% b) 41% c) 61% d) 71% e) 31% 33. A 57-year-old fit woman presents with abdominal distension, poor appetite and weight loss. CT scan shows bilateral ovarian masses and an omental ‘cake.’ Her CA125 is 763 U/ml. She undergoes laparotomy and debulking. Tumour deposits are found in the ovaries and on the omentum and surface of the liver measuring 2.5 cm. What stage is her ovarian cancer? a) FIGO stage IIIC b) FIGO stage IVA c) FIGO stage IIC d) FIGO stage IIIB e) FIGO stage IIIA 34. Which of the following statements about radiotherapy for vulval cancer is correct? a) The dose of concurrent weekly cisplatin is 50 mg/m2 b) Moist desquamation of the vulval skin can occur, but it is not usually painful c) The vulva does not tolerate radiotherapy well, so radiation doses in excess of 60 Gy cannot be given d) The clinical target volume in advanced vulval cancer includes the primary tumour, groin and pelvic nodes e) Postoperative radiotherapy does not reduce the local recurrence rate in patients with a resection margin of less than 8 mm
35. Which of the following statements about radiotherapy for mesothelioma is correct? a) Radiotherapy to the needle track or drain site is an accepted treatment which is aimed at reducing the risk of tumour seeding b) Palliative radiotherapy to the chest for mesothelioma usually alleviates shortness of breath c) The dose for needle-track radiotherapy can be kept as low as 50 Gy in 25 fractions because only microscopic disease is being treated d) In postoperative radiotherapy following extrapleural pneumonectomy, the GTV is the entire hemithorax e) Symptomatic disease in the thorax should be treated with high-dose palliation 36. Which of these is the correct estimated 5-year overall survival following an Ivor-Lewis oesophagectomy on a fit patient with a T3 N1 adenocarcinoma? a) 60% b) 40% c) 5% d) 20% e) 0% 37. In megavoltage photon therapy, which of the following statements is correct? a) The width of the radiation beam increases linearly with distance from the treatment head because of the inverse-square law b) The beam penumbra is the distance between the 30 and 70% depth doses c) As the energy increases, the dmax decreases d) When the distance from the treatment machine to the patient increases, the inverse-square law causes percentage depth doses below the dmax to increase e) As the field size increases, the central axis receives less radiation per monitor unit because of reduced scatter from the machine head and within the patient 38. Which of the following aspects of reporting trials can actually lead to less reliable conclusions? a) Ensuring as complete a follow-up as possible b) Reporting trials whether or not they show a ‘positive’ or ‘negative’ result c) Stressing to readers those outcomes where there is a difference between subgroups d) Performing a meta-analysis e) Reporting treatment effects and confidence intervals
457
Multiple choice questions
39. Which one of the following cytotoxic drugs is a vesicant when extravasated – that is, is associated with a high risk of causing inflammation and blistering of local skin and underlying tissue leading to tissue death and necrosis? a) Carboplatin b) Bleomycin c) Paclitaxel d) Etoposide e) Methotrexate 40. Which of the following statements relating to the cervical lymph nodes is correct? a) Level II contains the submandibular nodes b) Preauricular and intraparotid nodes are included in level I c) Lower jugular nodes lie in level III d) Level III extends from the hyoid bone superiorly to the lower border of the cricoid cartilage e) Level VI contains the anterior central compartment nodes and extends between the hyoid bone superiorly and the carina inferiorly 41. Whose law states that a palpable non-tender gallbladder is a typical feature of an obstructive tumour of the lower biliary tree? a) Kruckenberg b) Courvoisier c) Trousseau d) Bismuth e) Virchow 42. Which of the following genetic defects is most commonly associated with the adenoma-carcinoma sequence for colorectal carcinoma? a) A loss of heterozygosity in the RB gene b) DNA mismatch repair c) Chromosome 22 loss d) HER-2 amplification e) DNA hypomethylation 43. Regarding carcinoma of the anus, which of the following statements is correct? a) Anal-margin tumours have a worse prognosis than tumours in the anal canal b) Patients with HIV/AIDS are at reduced risk of toxicity from treatment c) Anal fissure is a risk factor for anal cancer d) A history of intraepithelial neoplasia of the cervix, vagina or vulva is a risk factor for anal cancer e) Patients who are treated with radical chemoradiotherapy have a significantly better life expectancy than patients treated with radical surgery 458
44. Osteolytic bone metastases are a typical feature of: a) Prostate cancer b) Gastric cancer c) Renal cell cancer d) Carcinoid tumours e) Bladder cancer 45. A man presents with a grade 3, pT1 bladder cancer. The most appropriate management is transurethral resection of bladder tumour (TURBT) followed by: a) Radical radiotherapy with regular cystoscopic follow-up b) Radical cystectomy c) Neoadjuvant chemotherapy and radical cystectomy d) Systemic chemotherapy with regular cystoscopic follow-up e) Intravesical BCG with regular cystoscopic followup 46. Which one of the following statements about bone scans is true? a) Bone scans are very useful for diagnosing myeloma deposits b) Bone scans are not useful for showing metastases from bladder cancer c) The isotope used is 123 I d) Bone scans show osteoblast activity e) Bone scans show high activity in the brain and heart 47. A 58-year-old man presents with an enlarging mass on the penis. A biopsy confirms this is squamous carcinoma. Clinical examination and staging investigations are otherwise normal. The tumour is staged as T2 grade 3. What is the standard treatment? a) Brachytherapy b) External beam radiotherapy to the penis and groin nodes c) Partial amputation of the penis and bilateral elective groin node dissection d) Partial amputation of the penis and prophylactic radiotherapy to the groin nodes e) Topical chemotherapy with 5-FU 48. A 25-year-old man with stage II non-seminomatous germ cell cancer undergoes treatment with four cycles of BEP chemotherapy. After treatment, a CT scan shows a significant response, but the presence of a residual mass measuring 4 cm diameter. The treatment should be: a) Give a further two cycles of BEP and re-scan b) Observe with serial CT scans c) Surgery to remove the mass d) Observe with serial CT scans and tumour markers e) Radiofrequency ablation of the mass
Multiple choice questions
49. A 57-year-old fit woman presents with abdominal distension, poor appetite and weight loss. A CT scan shows bilateral ovarian masses and an omental ‘cake.’ Her CA125 is 763 U/ml. She undergoes laparotomy and debulking. Tumour deposits are found in the ovaries and on the omentum and surface of the liver measuring 2.5 cm. What chemotherapy would you recommend now? a) Offer single-agent carboplatin chemotherapy b) Discuss options of platinum-based chemotherapy alone or platinum-based chemotherapy with paclitaxel c) Discuss options of platinum-based chemotherapy alone or paclitaxel alone d) Offer cisplatin, doxorubicin and cyclophosphamide chemotherapy e) Offer single-agent cisplatin 50. A 65-year-old woman presents with postmenopausal bleeding. A hysteroscopy shows polyps and biopsy confirms adenocarcinoma. She undergoes TAH and BSO and histology shows a FIGO stage IA grade I tumour and there is adenomyosis extending into the deep myometrium. What management would you recommend? a) Adjuvant progestin b) Radiotherapy with external beam and obturator c) External beam radiotherapy alone d) Clinical follow-up only e) Radiotherapy with obturator alone 51. A patient with cervical cancer is undergoing radical chemoradiotherapy and has received five cycles of cisplatin chemotherapy. She developed paraesthesiae, muscle cramps, tremor and hyperreflexia. Her calcium level was 1.89 mmol/l. She was given calcium supplements and her most recent serum calcium is within the normal range but her symptoms have not resolved. What is the most likely biochemical cause of her symptoms? a) Hypercalcaemia b) Hypokalaemia c) Hypomagnesaemia d) Hyponatraemia e) Hyperphosphataemia 52. Which statement regarding patients with small-cell lung cancer and a good performance status (PS 0–1) is correct? a) Platinum-based systemic chemotherapy in patients with limited disease does not prolong survival b) Single-agent chemotherapy is as effective as combination chemotherapy
c) Prophylactic cranial irradiation in patients with limited disease and a complete response to chemotherapy does not prolong survival d) Consolidation thoracic radiotherapy in patients with extensive disease and a complete response to chemotherapy prolongs survival e) Consolidation thoracic radiotherapy in patients with limited disease and a complete response to chemotherapy prolongs survival 53. A 43-year-old man presents with a swelling in the right thigh. Investigations show a soft-tissue mass within the quadriceps muscle. A biopsy is undertaken, followed by surgery to remove the tumour. Histology shows a leiomyosarcoma with 20 mitoses per high-power field and with more than 50% necrosis. The tumour measured 3 cm diameter and had an excision margin of 25 mm. What is the most appropriate management? a) Clinical follow-up with chest X-ray every 3 months b) Adjuvant chemotherapy with ifosphamide and doxorubicin c) Re-excision d) Postoperative radiotherapy e) Sentinel node lymph node biopsy followed by groin-node dissection if positive 54. A 30-year-old woman presents with an enlarging mass of lymph nodes in the right neck. It is mobile, measures 5 cm in diameter, and has a rubbery feel. She has no ‘B’ symptoms. A lymph node biopsy confirms classical Hodgkin lymphoma. There is no evidence of any other sites of disease. Out of the options listed, what is the most appropriate treatment? a) Six cycles of CHOP chemotherapy b) Mantle radiotherapy with a radiotherapy dose of 40 Gy in 20 fractions c) Six cycles of ABVD chemotherapy d) Four cycles of ABVD chemotherapy followed by involved field radiotherapy (30 Gy in 15 fractions) e) Surgery to remove the rest of the mass followed by involved field radiotherapy 55. A 75-year-old woman develops small-volume lymphadenopathy in the left axilla and left neck. A CT scan shows no other sites of disease. Her haemoglobin is 12.5 g/dl and her LDH is in the normal range. Biopsy of one of the lymph nodes shows grade 1 follicular lymphoma. What is the best management? a) CHOP chemotherapy b) Single-agent chlorambucil c) Surveillance 459
Multiple choice questions
56.
57.
58.
59.
460
d) ABVD chemotherapy e) Single-agent fludarabine A 70-year-old man, who is on warfarin for atrial fibrillation, presents with a slowly enlarging lesion at the inner canthus of his left eye. On examination, the lesion diameter is 0.75 cm and depth is 0.5 cm. Biopsy confirms a basal cell carcinoma. The most appropriate management is: a) Surgical excision b) Radiotherapy with 90 kV X-rays c) Radiotherapy with 6 MeV electrons d) Radiotherapy with 170 kV X-rays e) Radiotherapy with 9 MeV electrons A patient has an excisional biopsy for an ulcerated pigmented cutaneous lesion. The histology is of a 3.7-mm-thick nodular melanoma completely excised. Which of the following options would be considered as standard to offer the patient? a) A further 5-cm-wide excision b) Sentinel lymph node biopsy followed by lymph node dissection c) Elective lymph node dissection d) Adjuvant low-dose interferon α e) A CT scan of thorax and abdomen A 39-year-old man with medullary thyroid cancer and lung metastases has intense uptake in known sites of disease on diagnostic MIBG whole-body imaging and is being admitted for therapeutic MIBG. Past medical history includes total thyroidectomy and bilateral neck dissection. Which of the following is correct? a) The radioisotope used for therapy is 123 I MIBG b) He needs potassium iodate prior to treatment c) The therapeutic isotope is given orally d) He requires regular BP monitoring during the treatment e) The prophylactic antiemetic of choice is haloperidol A 51-year-man old undergoes an ileal resection and right hemicolectomy for a locally advanced, nodepositive, well-differentiated carcinoid tumour. Five years later the patient presents with flushing and palpitations and a CT scan reveals three tumours in segments 2 and 3 of the liver. What is the most appropriate next investigation? a) Twenty-four-hour urine 5-HIAA b) Serum chromogranin A c) PET scan d) Somatostatin receptor scintigraphy (OctreoScan® ) e) 123 I-MIBG scan
60. Which of the following is an ‘unfavourable’ subset of carcinoma of unknown primary site? a) Poorly differentiated carcinoma in the midline b) Female patient with papillary serous adenocarcinoma of the peritoneal cavity c) Female patient with axillary lymphadenopathy d) Adenocarcinoma of the pleura e) Squamous cell cervical lymphadenopathy 61. Regarding heterotopic bone formation, which of the following statements is correct? a) It causes clinical problems in 43% of patients after a hip replacement b) It typically develops 6 months after hip surgery c) Prophylactic radiotherapy is effective if given up to 4 days after re-operation of the hip d) Prophylactic radiotherapy is ineffective if given just before re-operation e) Fractionated radiotherapy is more effective than a single fraction 62. Which of the following statements is correct? a) The oesophageal pharyngeal syndrome of oxaliplatin is made worse by exposure to heat b) Palmar-plantar erythrodysaesthesia caused by liposomal doxorubicin is made worse by exposure to cold c) Scalp warming can reduce alopecia caused by some chemotherapy drugs d) An anthracycline extravasation is best treated with a cold compress e) A vinca alkaloid extravasation is best treated with a cold compress 63. A 74-year-old woman presents with a painless smooth swelling in the right neck (3 × 4 cm). There are no ENT symptoms and no abnormalities detectable in clinic. She has been a smoker of 20 cigarettes a day for 40 years. There is no co-morbidity and her WHO PS = 0. US of the neck shows a solitary cystic mass at level II only. FNAC shows squamous carcinoma cells. What would you do next? a) Contrast enhanced CT neck b) FDG-PET scan c) Panendoscopy of upper aerodigestive tract d) Right modified radical neck dissection e) Excision of cystic mass for definitive pathology 64. A 65-year-old man underwent abdominoperineal resection in January 2004 for a T3 N0 M0 rectal carcinoma. At routine follow-up in January 2005, his CEA was found to be elevated at 25 U/ml and a CT scan showed a solitary liver metastasis that was resectable on radiological grounds. Based on current
Multiple choice questions
evidence, the standard treatment outside of a clinical trial is: a) Liver resection only b) Preoperative modified de Gramont chemotherapy followed by liver resection c) Preoperative oxaliplatin based chemotherapy followed by liver resection d) Preoperative irinotecan based chemotherapy followed by liver resection e) Radiofrequency ablation 65. A fit 66-year-old man presents with a bilateral cT3a prostate cancer, PSA = 25 ng/ml and Gleason 4 + 4 = 8 in all eight cores, maximum core length = 12 mm. MRI/bone scan confirm T3a N0 M0 disease. The most appropriate management is: a) Radical prostatectomy, lymph node dissection and adjuvant radiotherapy b) Radical prostatectomy, lymph node dissection and 2 years of adjuvant hormonal therapy c) Radical external beam radiotherapy with a highdose-rate brachytherapy boost d) Neoadjuvant androgen deprivation and radical external beam radiotherapy e) Neoadjuvant and adjuvant androgen deprivation and radical external beam radiotherapy 66. A patient is to receive a course of definitive chemoradiation for an inoperable carcinoma of the tail of the pancreas. Both kidneys are functioning. You are given a radiotherapy plan for approval that has an anterior and two lateral fields. The liver V30 = 40%, small bowel Dmax = 45 Gy, left kidney V30 = 90%, the right kidney V20 = 25%. What do you do? a) Reduce the dose to the PTV b) Accept the plan c) Reduce the margin given to the GTV d) Add a posterior field e) Increase the weight to the anterior field 67. With regard to brachytherapy for carcinoma of the cervix, which of the following statements is correct? a) For practical purposes, Manchester point A is sometimes defined as 1 cm above and 1 cm lateral to the flange at the lower end of the uterine tube b) Manchester point B is 3 cm from the midline c) The ICRU 38 bladder point is the anterior surface of the bladder balloon d) The ICRU 38 rectal point is 5 mm behind the posterior vaginal wall at the level of the lower end of the intrauterine tube e) The rectal dose should be less than half of the dose to point A
68. You are shown a plan for a radical lung treatment for CHART radiotherapy (54 Gy in 36 fractions over 12 days). The V20 for the lung is 45%. The dose to the PTV lies between 95 and 107%. The maximum spinal cord dose is 15 Gy. What action should you take? a) Accept the plan b) Accept the plan but reduce the prescribed dose of radiotherapy c) Treat the patient with 60 Gy in 30 fractions over 6 weeks instead d) Move one or more beams so they pass through the spinal cord rather than the lung e) Move one or more beams so they pass through lung rather than spinal cord 69. Which of the following is important for minimising measurement bias in clinical studies? a) Meta-analysis b) Adjusted analyses c) Blinding d) Per protocol analysis e) A transparent randomisation process 70. The ICRU reference point should be: a) Always at the centre of the tumour b) Representative of the PTV dose c) At the isocentre if there is one d) At the point of maximum dose within the PTV e) At the point of minimum dose within the PTV 71. A patient has a 5 cm squamous cancer of the lower anal canal. An MRI demonstrates evidence of leftsided internal iliac lymphadenopathy. No other disease is found. The correct stage is: a) T2 N1 M0 b) T2 N2 M0 c) T3 N1 M0 d) T3 N1 M1 e) T2 N1 M1 72. Which of the following statements about gastrointestinal stromal tumours is correct? a) They arise from the intestinal cells of Jacal b) They are associated with dysregulation of the tumour suppressor gene, KIT c) An increase in size of the tumour can occur when responding to imatinib d) Standard treatment for patients whose tumours have been resected is adjuvant imatinib e) GIST tumours stain positively with CD119 73. A 56-year-old woman undergoes breast screening. A 1.5 cm abnormality is detected in the left breast, which is biopsied and shows lobular carcinoma 461
Multiple choice questions
74.
75.
76.
77.
462
in situ. She has a very strong family history of breast cancer and does not wish to accept observation alone. What surgical treatment would you recommend? a) Wide local excision followed by postoperative radiotherapy b) Wide local excision alone c) Unilateral mastectomy d) Bilateral mastectomy e) Bilateral mastectomy with unilateral axillary lymph node dissection A 62-year-old man undergoes a radical nephrectomy for renal cell carcinoma. Histology confirms an 8 cm tumour confined to the kidney with three positive regional lymph nodes. What stage is this? a) T2 N1b b) T3 N2 c) T3 N1b d) T2 N2 e) T2 N3 A man is discussed at the multidisciplinary meeting following TURBT. A large papillary mass was fully resected. Pathology shows a grade 3 transitional cell carcinoma with invasion of the muscularis propria. What stage is he? a) T2a b) pT2a at least c) pT1b d) pT2b e) pT3a A man with metastatic NSGCT is being treated with his first course of chemotherapy with standard 5-day BEP. When he attends for his third cycle of chemotherapy, he mentions some mild exertional dyspnoea. What is the most appropriate management? a) Delay chemotherapy for 1 week to allow the patient time to improve b) Stop chemotherapy after two cycles c) Stop BEP and change to TIP for the remaining cycles d) Treat with oxygen, antibiotics and high-dose intravenous steroids and continue with the third cycle of BEP without delay. e) Investigate with CXR, CT of the chest and pulmonary function tests, proceeding with EP for a total of four cycles if bleomycin toxicity is suspected Which of the following is a risk factor for ovarian cancer? a) Breast feeding b) Tubal ligation
78.
79.
80.
81.
c) Use of the oral contraceptive pill d) A germline mutation in the HNPCC gene e) Early menopause A 45-year-old woman presents with FIGO stage IB1 carcinoma of the cervix. Following radical hysterectomy and lymph node dissection, she is found to have a 6 cm tumour with extension into the parametrium with negative lymph nodes. There is no evidence of distant metastases. What stage is the tumour? a) FIGO IIA b) pT2b pN0 c) FIGO IIB d) pT1b2 pN0 e) FIGO IIIB In vaginal cancer, which of the following statements is correct? a) Vaginal cancer is a rare disease, so treatment should always follow the standard protocol b) Radiotherapy phase 1 includes the pelvic and paraaortic lymph nodes c) The radiotherapy field for carcinoma of the upper third of the vagina should include the inguinal nodes d) Vaginal cancers in the lower vagina usually have a worse prognosis than those in the upper vagina e) There is a strong evidence base for the use of concurrent chemoradiotherapy for squamous carcinoma of the upper vagina A 53-year-old man had a left upper lobectomy for a moderately differentiated pT2 pN1 M0 squamous cell carcinoma of the lung. Resection margins were reported as clear, with a medial resection margin of 6 mm. What further therapy is it appropriate to offer him (outside a clinical trial)? a) Adjuvant radiotherapy to the mediastinum b) Adjuvant chemotherapy c) Adjuvant radiotherapy to the mediastinum followed by adjuvant chemotherapy d) Further surgical resection e) Concomitant chemoradiotherapy Which of the following statements about Ewing’s sarcoma is correct? a) Relapsed disease can often be cured with secondline intensive chemotherapy b) The incidence of distant metastases is low c) It is a chemoresistant tumour d) Radiotherapy should not be given in inoperable disease because of the risk of fracture when the tumour responds
Multiple choice questions
82.
83.
84.
85.
e) Radiotherapy is indicated for sites of metastases, including the whole lung A 59-year-old woman presents with vague abdominal symptoms of nausea, weight loss, leg oedema and urinary pressure. Her serum CA125 level is 2537 U/ml. A CT scan shows bilateral pelvic masses with para-aortic lymphadenopathy. A biopsy and immunohistochemistry are performed. The tumour is found to be cytokeratin negative, CD19 positive, and CD20 positive and Ki-67 of 95%. What is the most appropriate initial treatment? a) Laparotomy, debulking, total abdominal hysterectomy, bilateral salpingo-oophorectomy, omentectomy b) Chemotherapy with carboplatin and paclitaxel c) Rasburicase then chemotherapy with R-CHOP d) Radiotherapy to sites of bulky disease e) Allopurinol then chemotherapy with ABVD Which statement best describes the role of chemotherapy in primary tumours of the central nervous system? a) Concurrent chemotherapy with radical radiotherapy has been shown to more than double the 2-year survival in patients with grade IV gliomas b) The improvement seen with concurrent chemotherapy and radical radiotherapy is greatest in patients with a poor performance status c) Radiological features of primary CNS lymphoma are typical, so that patients should receive chemotherapy without the need for histological confirmation d) Adjuvant chemotherapy has been shown to double median survival in grade gliomas e) The usual dose of temozolamide is 20 mg/m2 An 85-year-old lady presents with a non-healing ulcer on the left lower leg. It measures 6 cm in maximum diameter, 1 cm in depth, and has a rolled edge. Clinical examination reveals no other tumours. A biopsy is undertaken that shows moderately differentiated squamous carcinoma. What is the optimal management? a) Observation b) Radiotherapy with 170 kV photons, 40 Gy in 8 fractions over 8 days c) Radiotherapy with 9 MeV electrons and 1 cm bolus, 60 Gy in 30 fractions over 6 weeks d) Surgical excision and healing with secondary intention e) Surgical excision and skin grafting Concerning malignant melanoma, which of the following statements is true?
86.
87.
88.
89.
a) Nodular melanoma is the most common subtype b) Sunscreens have been shown to reduce the risk of malignant melanoma c) A 10 mm choroidal melanoma can be treated with scleral plaque therapy instead of enucleation d) Adjuvant low-dose interferon α has been shown to improve survival e) The addition of carmustine, tamoxifen and cisplatin to dacarbazine improves survival over dacarbazine alone in advanced melanoma A 51-year-old woman develops appendicitis. During surgery, it is noted that the appendix forms a mass that is locally invasive. The mass is excised and shows a carcinoid tumour with atypical cells. Four years later she develops some mild right upper quadrant discomfort and an ultrasound shows a 3 cm mass in segment 8 of the liver. Further investigations show no other sites of disease. What is the optimal management? a) Capecitabine and streptozocin b) Interferon α c) 90 Y octreotide isotope therapy d) 131 I MIBG isotope therapy e) Surgical resection A 4-year-old boy presents with painless swelling of the abdomen. An abdominal ultrasound shows a large tumour in the right flank. A CT scan shows that this is arising from the right kidney and is invading the perinephric structures. A biopsy is not undertaken because of the risk of tumour rupture. What is the most likely diagnosis? a) Neuroblastoma b) Wilm’s tumour c) Rhabdomyosarcoma d) Non-Hodgkin lymphoma e) Teratoma Which of the following statements about keloids is correct? a) Keloids should not be excised before radiotherapy because this would increase the risk of recurrence b) Radiotherapy should be given within 1 month of excision c) Radiotherapy is effective in about half of all cases d) Keloids are unsuitable for treatment with brachytherapy e) A single fraction of 6 Gy with 100 kV photons is adequate for small keloids A 59-year-old fit woman has a 10 mm mass in the upper outer quadrant of the right breast. Biopsy shows poorly differentiated cancer; CK7+, CK20−, TTF−1 focally+, ER−. CT scan shows multiple lung 463
Multiple choice questions
90.
91.
92.
93.
464
and liver metastases. Which of the following is the most appropriate chemotherapy regimen? a) FEC b) gemcitabine and carboplatin c) modified de Gramont d) R-CHOP e) Ifosphamide and doxorubicin A 56-year-old man in previously excellent heath has been referred to you for a second opinion about management of his epithelioid malignant mesothelioma. Which statement is correct? a) Pleuropneumonectomy followed by postoperative chemotherapy and radiotherapy has been shown to improve survival compared to no surgery for stage I disease b) The median survival for all patients is 18 months c) The epithelioid variant has the best prognosis d) The response rate to chemotherapy is more than 50% e) The presence of pleural plaques significantly increases the chance of developing mesothelioma in patients exposed to asbestos A 36-year-old woman is undergoing surveillance after a molar pregnancy. Her hCG fails to fall into the normal range and her pretreatment level was 913 IU/l. What management would you recommend? a) Chemotherapy with intramuscular methotrexate b) Chemotherapy with actinomycin D, methotrexate and etoposide c) Chemotherapy with actinomycin D, methotrexate, etoposide, vincristine and cyclophosphamide d) Observation e) Hysterectomy A 45-year-old man has had a right hemicolectomy for a mucinous carcinoma of the caecum as an emergency. The stage was T2 N0, grade II, completely resected with no lymphovascular infiltration; what further management would you advise? a) A complete total colectomy b) An adjuvant course of oxaliplatin and 5-FU chemotherapy c) An adjuvant course of capecitabine chemotherapy d) Postoperative staging including colonoscopic review of the remaining bowel e) Postoperative CEA Which of the following statements is true about choosing wide or restrictive entry criteria to a trial? a) With wide entry criteria it is impossible to work out what the trial population was b) With restrictive entry criteria there is more chance of a significant result
94.
95.
96.
97.
c) Wide entry criteria make it more difficult to recruit patients to the trial d) Restrictive entry criteria give more clinically meaningful results e) Restrictive entry criteria mean that it is often impossible to determine if a treatment works for all people, or just some What is the definition of high-dose-rate brachytherapy? a) Dose rate greater than 6 Gy/hr b) Dose rate greater than 8 Gy/hr c) Dose rate greater than 10 Gy/hr d) Dose rate greater than 12 Gy/hr e) Dose rate greater than 14 Gy/hr Which of the following statements about hormonal treatments is correct? a) Aromatase inhibitors should not be given to premenopausal women after ovarian function suppression b) Tamoxifen is less likely than an aromatase inhibitor to cause thromboembolic events c) Aromatase inhibitors are less likely than tamoxifen to cause osteoporosis d) Tamoxifen is less likely than aromatase inhibitors to cause endometrial hyperplasia e) Tamoxifen is less likely than aromatase inhibitors to cause arthralgia Which of the following statements about postoperative adjuvant radiotherapy for stage IC, grade 3, carcinoma of the endometrium is correct? a) It should be combined with concurrent chemotherapy with weekly cisplatin b) A reasonable dose is 45 Gy in 25 fractions to the whole pelvis followed by 9 to 18 Gy as an external beam boost to the GTV c) It should be followed by adjuvant hormonal therapy with a progestin d) An acceptable dose for the boost would be 20 Gy LDR to the whole vagina e) The lower border of the radiation field should cover the obturator foramina Which of the following statements regarding carboplatin and paclitaxel is correct? a) Carboplatin and paclitaxel have different routes of excretion so the order in which they are given does not matter b) Paclitaxel is given before carboplatin; otherwise there is a 20% reduction in paclitaxel clearance c) Paclitaxel is given after carboplatin to reduce the risk of myelosuppression
Multiple choice questions
98.
99.
100.
101.
102.
d) Paclitaxel and carboplatin are given simultaneously to reduce the treatment time for the patient e) Scalp cooling should be offered because it reduces the risk of alopecia due to carboplatin Which of the following statements about trastuzumab is correct? a) It has been shown to improve overall 10-year survival when used in patients with HER-2-positive breast cancers as adjuvant therapy in combination with chemotherapy b) It is a chimeric IgG antibody c) It causes symptoms of heart failure in approximately 10% of patients d) It should always be preceded by the administration of antihistamines and steroids e) It is recommended with docetaxel as first-line therapy in patients with HER-2-positive mestastatic breast cancer who are unsuitable for an anthracycline Regarding pineal tumours, which of the following statements is correct? a) Due to the site of these tumours, biopsy is rarely indicated before treatment b) Germ cell intracranial tumours rarely produce elevated serum βhCG and/or αFP c) Pineocytomas are less chemosensitive than oligodendrogliomas d) Pineoblastomas should be considered for stereotactic radiotherapy e) The dose recommended for craniospinal axis radiotherapy is lower for germ cell tumours than for parenchymal tumours A 55-year-old woman with stage 3C ovarian cancer relapses 4 months after carboplatin and paclitaxel chemotherapy. What would be the recommended chemotherapy treatment now? a) Liposomal doxorubicin b) Topotecan c) Single-agent carboplatin d) Carboplatin and paclitaxel e) Carboplatin and gemcitabine Which cancer is the most common in the UK? a) Uterus b) Oesophagus c) Larynx d) Bladder e) Carcinoma of unknown primary site A patient has recently been diagnosed with carcinoma of the pancreas. Investigations reveal a 3 cm mass in the head of the pancreas invading the sec-
103.
104.
105.
106.
ond part of the duodenum. There is evidence of coeliac lymphadenopathy but no other evidence of metastatic disease. The stage is: a) T2 N1 M0 b) T3 N1 M0 c) T4 N1 M0 d) T3 N1 M1a e) T4 N1 M1a A 45-year-old woman weighs 73 kg and her serum creatinine is 95 mM/l. Using the Cockcroft-Gault formula, what is the best estimation of her creatinine clearance in ml/min? a) 36 b) 56 c) 76 d) 96 e) 116 Radiotherapy doses as low as which level can affect sperm production? a) 3 Gy b) 16 Gy c) 10 Gy d) 5 Gy e) 1 Gy A 66-year-old electrician presents with shortness of breath. His chest X-ray shows pleural thickening and an ultrasound shows this to be solid with an associated pleural effusion. A CT scan shows thickening of the mediastinal pleura and loss of lung volume. Out of the following, what investigation would be most likely to give the diagnosis? a) Bronchoscopy b) CT-guided fine needle aspiration cytology c) Thoracoscopic biopsy d) Cytology of pleural fluid e) Sputum cytology Which of the following statements about hormonal treatments is correct? a) Tumours that develop resistance to tamoxifen are very unlikely to respond to an aromatase inhibitor b) Aromatase inhibitors act primarily on steroid synthesis in the adrenal gland c) In patients with prostate cancer who are starting treatment with an LHRHa, an antiandrogen is also required to prevent tumour ‘flare’ d) Ovarian cancer is not a hormonally sensitive disease e) Tumours that are oestrogen-receptor negative and progesterone-receptor positive should not be treated with tamoxifen 465
Multiple choice questions
107. A 54-year-old man undergoes a left nephrectomy for renal cell cancer. Eight months later he feels short of breath and has back pain. CXR and CT scan confirm the presence of a local recurrence and two peripheral lung metastases. His Karnofsky performance status is 50. His Hb is 10.5 g/dl and his serum corrected calcium is 2.7 mM/l. What management would you recommend? a) Symptomatic treatment only b) Interferon α c) Interleukin-2 d) Chemotherapy with the Atzpodien regimen e) Attempted surgical resection of the lung metastases in an attempt to control the local recurrence 108. The MRC ST02 (MAGIC) trial was recently published. This trial demonstrates that: a) Preoperative epirubicin, cisplatin and capecitabine (ECX) followed by surgery is superior to surgery alone b) Postoperative epirubicin, cisplatin and 5-fluorouracil (ECF) has no proven role in gastric cancer c) Preoperative chemotherapy is superior to postoperative chemotherapy in gastric cancer d) Pre- and postoperative ECF chemotherapy is superior to surgery alone e) Capecitabine is equivalent to 5-FU in the adjuvant therapy of gastric cancer 109. A 49-year-old woman had radical chemoradiotherapy for cervical carcinoma 2 years ago. She now presents with back pain radiating down the right leg with pain and numbness of the left knee. What is the most likely site of recurrence? a) L3 vertebra b) L4 vertebra c) Left psoas muscle d) Left para-aortic nodes e) Pelvic nodes 110. A 60-year-old male smoker presents with a 6-week history of cough and swelling of the face and upper limbs. Examination findings are consistent with SVCO and CXR shows right paratracheal lymphadenopathy. What is the most likely underlying diagnosis? a) Thymoma b) High-grade NHL c) Lung cancer d) Metastatic germ cell tumour e) Hodgkin lymphoma 111. A 55-year-old man complains of back pain. Examination is unremarkable, but spinal MRI reveal a 3 cm extradural soft tissue mass at the level of T12. Biopsy 466
112.
113.
114.
115.
shows malignant plasma cells, but staging investigations for multiple myeloma are negative. He is referred for radiotherapy. What is the optimum PTV and dose schedule? a) GTV plus 10 cm margin, 50 Gy in daily 2 Gy fractions b) GTV plus 2 cm margin, 60 Gy in daily 2 Gy fractions c) GTV plus 2 cm margin, 40 Gy in daily 2 Gy fractions d) GTV plus 5 cm margin, 30 Gy in daily 3 Gy fractions e) GTV plus 2 cm margin, 20 Gy in daily 4 Gy fractions A laboratory assessment of the α/β ratio for a malignant melanoma cell line demonstrated an average value of approximately 2. Taking this information into account, which of the following radiotherapy regimens is likely to be the most effective in cancer control of melanoma? a) 30 Gy in 10 fractions b) 21 Gy in 3 fractions c) 10 Gy in a single fraction d) 30 Gy in 5 fractions e) 35 Gy in 15 fractions A 35-year-old man with a smoking history of 20 pack years presents with chest discomfort and shortness of breath. He has lost weight and has been having night sweats. He is found to have a poorly differentiated carcinoma in the mediastinum. What is the most appropriate chemotherapy regimen? a) BEP b) R-CHOP c) FOLFOX d) ABVD e) ICE Which of the following statements about erlotinib is correct? a) It is a monoclonal antibody directed against EGFR b) It is given intravenously c) Patients who develop a rash typically have a worse response than those who do not d) The recommended dose is 200 mg daily e) The rash responds to topical clindamycin Which of the following statements about oestrogen is correct? a) In premenopausal women, oestrogens are mainly synthesised in the theca cells of the ovary b) In women, oestrogens are produced in the adrenal gland by the action of the aromatase enzyme c) Oestrogen-receptor-positive breast cancers invariably respond to hormonal treatment d) Oestrogens reduce low-density lipoproteins e) Oestrogens are anticoagulants
Multiple choice questions
116. In radiotherapy planning, which of the following statements is correct? a) The internal margin accounts for the uncertainties and lack of reproducibility in setting up the patient day by day b) The treated volume is the tissue volume that receives a dose considered significant in relation to normal tissue tolerance c) Reducing the planning CT-scan slice thickness reduces the distance required to be added to the gross tumour volume to make the clinical target volume d) CTV + IM + SM = PTV e) The function of ‘parallel’ organs may be seriously affected even if a small portion is irradiated above a tolerance dose, and the effect of radiation on the function of ‘serial’ organs is more dependent on the volume irradiated. 117. A 64-year-old man develops jaundice with pale stools and dark urine. He also has some vague right upper quadrant pain and weight loss. There are no masses palpable on examination. An ultrasound scan suggests the presence of a distal tumour in the biliary tract. Out of the following options, what is the most appropriate investigation to perform next? a) Percutaneous transhepatic cholangiograph b) MRI scan c) ERCP d) CT scan e) Doppler ultrasound 118. In vaginal cancer, which of the following statements is correct? a) The most common malignant tumour is squamous carcinoma of the vagina b) The lymphatic drainage from the lower two-thirds
of the vagina is to the pelvic nodes and from the upper third to the inguinal nodes c) Carcinoma of the vagina is associated with procidentia d) Previous abnormal cervical cytology is not associated with an increased risk of vaginal cancer e) Vaginal cancer accounts for 10% of all gynaecological malignancy 119. Regarding malignant forms of gestational trophoblastic tumour, which statement is correct? a) Choriocarcinoma occurs with a frequency of about 1 per 500 000 live births b) The most widely used standard therapy for low-risk patients is intravenous actinomycin D c) The overall survival in the low-risk group is good at around 75% d) Patients with high-risk disease should be offered EMA-CO chemotherapy (etoposide, methotrexate, actinomycin-D, cisplatin and vinorelbine) e) The rare placental site trophoblast tumour can usually be cured if patients present within 4 years of pregnancy 120. A 45-year-old homosexual man presents with a 2week history of multiple, painless, purple macules on his hard palate. He is found to be HIV positive on testing. What is the most appropriate management? a) Intralesional chemotherapy with vinblastine b) Optimise antiretroviral therapy for 3 months and review c) Radiotherapy to hard palate using parallel opposed 6 MV photons, 20 Gy in 10 fractions over 2 weeks d) Radiotherapy to hard palate using parallel opposed 6 MV photons, 16 Gy in 4 fractions over 4 days e) Systemic chemotherapy with liposomal doxorubicin 20 mg/m2 every 3 weeks
467
MULTIPLE CHOICE ANSWERS
1. e
21. a
41. b
61. c
81. e
101. d
2. c
22. e
42. e
62. d
82. c
102. b
3. e
23. c
43. d
63. c
83. a
103. c
4. d
24. c
44. c
64. a
84. e
104. e
5. c
25. c
45. e
65. e
85. c
105. c
6. d
26. d
46. d
66. b
86. e
106. c
7. d
27. b
47. c
67. d
87. b
107. a
8. b
28. d
48. c
68. d
88. e
108. d
9. e
29. a
49. b
69. c
89. b
109. a
10. d
30. a
50. d
70. b
90. c
110. c
11. b
31. b
51. c
71. b
91. a
111. c
12. d
32. a
52. e
72. c
92. d
112. d
13. b
33. a
53. d
73. d
93. e
113. a
14. c
34. d
54. d
74. d
94. d
114. e
15. a
35. a
55. c
75. b
95. e
115. d
16. c
36. d
56. b
76. e
96. e
116. d
17. b
37. d
57. e
77. d
97. b
117. c
18. d
38. c
58. d
78. b
98. e
118. c
19. d
39. c
59. d
79. d
99. c
119. e
20. c
40. d
60. d
80. b
100. a
120. b
468
INDEX
Page numbers followed by “t ” indicate tables.
abarelix, 36 abiraterone acetate, 36 accelerated partial breast irradiation (APBI), 200 acquired immune deficiency syndrome (AIDS), and cerebral lymphoma, 370 acral lentigenous melanoma (ALM), 396 actinic keratosis, 384 actinomycin-D, for gestational trophoblast tumours (GTT), 310 activating functions on oestrogen receptor, 25 activator protein-1 (AP-1), 30 active surveillance, for prostate cancer, 234 acute lymphoblastic lymphoma (ALL), in children, 431 acute myeloid leukaemia acute nausea and vomiting after chemotherapy, 7 adenocarcinoma, 180 in bladder, 223 of lung, 315 adenoma-carcinoma sequence, 161 adenomatous polyposis coli (APC), 161 adipose tissue oestrogen synthesis, 23 adolescents. See also childhood cancer cancer management, 428 cancers typically occurring, 427 adult granulosa cell tumour, of the ovary, 264 adverse events report standardisation, 64 αFP (alpha feto-protein) as marker for testicular cancer, 242 as potential vaccine target, 21 age, and breast cancer incidence, 190 aggressive fibromatosis, radiotherapy for, 450 airway acute obstruction, 74 alcohol, and liver cancer, 142
alkaline phosphatase evaluating before chemotherapy, 5 alkylating agents, 2 for breast cancer, during pregnancy, 208 phlebitis from, 8 allocation concealment, 58 allopurinol, for lymphomas, 350 alopecia from chemotherapy, 8 from non-steroidal antiandrogens, 34 alpha feto-protein (αFP) as marker for testicular cancer, 242 alveolus, lower cancer of radical radiotherapy for, 105 American Joint Committee on Cancer, 195 Cancer Staging Manual for carcinoma of stomach, 135t American Society of Clinical Oncology (ASCO) aminoglutethimide, 27, 28 aminoglycoside for neutropenia, 6 AML14 trial, 63 amoxicillin, for H.pylori infection eradication, 357 amputation, for soft tissue sarcoma, 338 anaemia chemotherapy and, 5, 7 erythropoietin for, 20 and hypoxia, 287 anal canal, anatomy, 174 anal cancer, 174 anal intra-epithelial neoplasia (AIN), 175 areas of current interest, 181 clinical presentation, 176 current clinical trials, 181 human papilloma virus (HPV) and, 174 incidence and epidemiology, 174 investigation and staging, 176 palliative care, 180 pathology, 175 patients with HIV/AIDS, 180
prognosis, 180 risk factors and aetiology, 174 spread, 175 as squamous carcinoma, 175 pathological features, 175t TNM classification and stage grouping, 176t treatment follow up after radical therapy, 180 for locally-recurrent, 179 overview, 176 radical radiotherapy and chemoradiotherapy, 177 radical surgery, 177 radiotherapy dose, fractionation and energy, 178 radiotherapy side effects, 178, 179 radiotherapy technique, 177 tumour types, 174 analgesics adjuvant, 84t anaphylaxis chemotherapy and, 9 related to anticancer drugs, 79 anaplastic large cell lymphoma, in children, 432 anaplastic thyroid cancer external beam radiotherapy (EBRT), 415 incidence and epidemiology, 415 prognosis, 415 spread, 415 treatment, 415 anastrazole, 28, 34, 203 for breast cancer treatment, 27 vs. tamoxifen, 29 androgen deprivation therapy (ADT), 31–34 adverse effects, 34 androgen receptor (AR), 25 androgens effects, 26 synthesis, 24 androstenedione, 24
469
Index
angiography, for neuroendocrine tumours (NETs), 421 angiosarcoma, 393 ankylosing spondylitis, radiotherapy for, 449 anorexia, 87 anthracyclines, antidotes for extravasation, 81t for breast cancer, 202 metastatic disease, 206 during pregnancy, 208 cardiotoxicity, 4, 7 for mantle cell lymphoma, 357 reaction from, 81t antiandrogens for prostate cancer treatment, 237 steroidal, 35 withdrawing therapy, 34 antibiotics for febrile neutropenia, 78 prophylactic, 79 anticipatory nausea and vomiting, 7 antidiuretic hormone (ADH) excess levels, 71 antidotes for extravasation, 81t anti-EGFR agents skin rash with, 17 anti-emetics, 86t prescribing, 84 antigen vaccines, 20 antimetabolites, 2 antitumour antibiotics, 2 anxiety, 88 aprepitant for treating nausea from chemotherapy, 7 area under the curve (AUC) dosage, 4 aromatase inhibitors (AIs), 25, 203 for breast cancer treatment, 28, 30 vs. tamoxifen, 29 arterial restenosis, radiotherapy for, 449 arteriovenous malformation (AVM), cerebral, radiotherapy for, 449 artificial hips and radiotherapy planning, 45 asbestos, 329 clinical effects, 329 and mesothelioma, 328 types, 328 ascites, 88 in ovarian cancer, palliative care for, 263 ASCO, 19, 155 Askin’s tumour of the rib, 438 asthenia from non-steroidal antiandrogens, 34 L-asparaginase anaphylaxis from, 79 ASTRO consensus definition, 237
470
ataxia-telangiectasia (A-T), and breast cancer, 192 attendance allowance, 91 attenuated FAP, 160 attrition bias, 61 atypical carcinoid, 325 autonomy of patient decision making and, 91 Avastin® . See bevacizumab (Avastin® ) axilla, radiotherapy for lymphoma, 361 axillary node clearance, 198 βhCG (beta human chorionic gonadotrophin), as marker for testicular cancer, 241 B catenin as potential vaccine target, 21 B-cell lymphoma in children, 432 diffuse large clinical presentation, 351 prognosis, 353 risk of relapse, 350 treatment flow chart, 351 treatment for advanced stage, 352 treatment for early stage, 351 treatment for relapsed/refractory disease, 352 treatment overview, 351 back, pain from lung cancer, 316 balanitis xerotica obliterans, 252 Barrett’s oesophagitis, 122 and gastric cancer, 133 basal cell carcinoma, 382, 383 radiotherapy dose calculation, 391 Bazex’s syndrome, 383 BC Cancer Agency, 1 BCG therapy, for bladder cancer treatment, 225 BCR-ABL fusion protein, 15 beam data arrangements, 43 concepts of basic, 42 matching adjacent in complex treatment, 45 beam divergence, 42 beam penumbra, 42 Beatson, G.T., 23 benign conditions, radiotherapy for, 449–452 benign proliferative breast disease, and breast cancer incidence, 190 benzodiazepines for anticipatory nausea, 7 for anxiety, 88 BEP regimen for testicular cancer, 2 practical management, 247 versions
bevacizumab (Avastin® ), 19 for liver cancer treatment, 145 for metastatic breast cancer, 206 for metastatic colorectal cancer, 169 for pancreatic cancer treatment, 157 bicalutamide, 31, 32, 34 for prostate cancer, 31, 237 bile ducts anatomy, 146 carcinoma of, 141 blood tests, 147 chemotherapy, 148 clinical presentation, 146 clinical trials, 149 imaging, 147 incidence and epidemiology, 145 palliative care, 148 photodynamic therapy (PDT), 148 prognosis, 149 radical surgery for, 148 stents, 148 treatment overview, 147 tumour types, 146t, 188t bilirubin evaluating before chemotherapy, 5 Billroth II partial gastrectomy surgical scheme, 136 biochemical renal, liver, and bone profile before chemotherapy, 5 biological agents, 13 cytokines, 13–15 biopsy for breast cancer assessment, 194 for cancer of unknown primary, 443 for mesothelioma, seeding risk from, 332 bisacodyl for constipation, 85 Bismuth classification, for perihilar tumours, 146t bisphosphonates, 34 for metastatic breast cancer, 207 for myeloma, 365 for prostate cancer, 238 side effects, 71 bladder anatomy, 222 tumour types, 222 bladder cancer, 222 areas of current interest, 229 chemoradiotherapy for, 227 chemotherapy for, 228 clinical presentation, 223 current trials, 229 incidence and epidemiology, 222 invasive, managing nodes, 227 investigation, 223 pathology, 223
Index
radical radiotherapy for, 226 applications, 226 beams, 227 doses, 227 setup, 226 target volume, 226 use of CRT/IMRT, 227 radiotherapy plan, 227 risk factors and aetiology, 222 screening, 222 staging, 224 TNM classification, 224t treatment of invasive, 225–227 overview, 224 patient fitness and attitude, 226 bleeding, 75 post menopausal, and endometrial carcinoma, 269 bleomycin, 12 in BEP regimen for testicular cancer, 2 for diffuse large B-cell lymphoma, 352 flat dosing of, 4 for Hodgkin lymphoma, 359 in children, 432 reaction from, 81t for testicular cancer, metastatic disease, 247 for vulval cancer, 300 bleomycin lung, 247 blood count before chemotherapy, 5 blood tests, for pancreatic cancer investigation, 153 body surface area (BSA) for chemotherapy dose calculation, 3 body weight dosing, 4 bone breast cancer spread to, 207 heterotropic formation, radiotherapy for, 450 radiotherapy impact on growth in child, 429 and radiotherapy planning, 45 solitary plasmacytoma, 366 bone marrow transplantation (BMT), 366 for childhood leukaemia, 431 bone mineral density (BMD), 28 bicalutamide impact, 34 prostate cancer impact, 34 bone scans, for prostate cancer, 233 bone tumours, 335, 341–345 anatomy and pathology, 341 areas of current interest, 344 chemotherapy, 343 in children Ewing’s sarcoma/peripheral PNETs, 438 osteosarcomas, 438
clinical presentation, 342 clinical trials, 345 incidence and epidemiology, 341 investigation and staging, 342 metastatic disease, treatment, 447 palliative care, 344 prognosis, 344 recurrent disease, 344 risk factors and aetiology, 341 spread, 341 staging classification, 342 surgery, 343 treatment overview, 343 tumour types, 341, 342 bortezomib for mantle cell lymphoma, 357 for myeloma, 365 bowel obstruction, in ovarian cancer, palliative care for, 263 Bowen’s disease, 252 brachytherapy, 51–53 for anal cancer, 178 areas of current interest, 53 for cervical cancer treatment, 282, 284, 287 post operative, 286 dosimetry, 51 for endometrial carcinoma, 271, 273 gynaecological, Manchester system for, 52 for head and neck tumours, 98 for lung cancer, 323 for oral cavity tumours, 104 for penile cancer, 254 for prostate cancer, 235 high dose rate afterloading, 235 morbidity from, 235 quality assurance, 53 radiographs from intracavitary treatment of cervical carcinoma, 285 radionuclide properties, 51 reference point for bladder and rectum during gynaecological, 284 treatment, 52 brain metastatic disease, treatment, 447 radiotherapy, 376 toxicity, 378 tumour treatment, 373–376 of ependymoma, 374 of high grade glioma, 374 of low grade glioma, 373 of medulloblastoma, 374 of meningioma, 374 of metastases, 375 of pineal tumours, 375 of pituitary, 375 brain stem
shielding in radiotherapy, 113 breast female anatomy, 190 radiotherapy impact in children, 430 tumours affecting, 191t breast cancer adjuvant treatment chemotherapy, 201 dose-dense chemotherapy regimens, 203 endocrine treatment, 203 guidelines, 201 radiotherapy, 199 risk categories, 201 trastuzumab (Herceptin® ), 204 antioestrogens for, 26 areas of current interest, 30 bilateral, 208 clinical presentation, 194 clinical trials, 210 endocrine responsive, 201 hormonal therapies, 25, 27 incidence and epidemiology, 190 inflammatory, 208 lapatinib against, 17 lifetime risk, 190 locally-advanced chemotherapy, 205 management, 204 male, 208 mechanisms of resistance, 29 metastatic disease, 205–207 bisphosphonates, 207 chemotherapy, 206 hormone therapy, 205 targeted therapy, 206 N and M staging, 196 ovarian function suppression, 26 Paget’s disease, 208 pathology, 192 and pregnancy, 207 prognosis five-year survival rate, 209, 210t histological type, 209 hormone receptor status, 209 lymphatic or vascular invasion, 209 regional lymph node status, 209 ten-year survival, 210t recurrent disease, 207 risk factors and aetiology, 190 screening and prevention, 193 spread, 193 stage groupings, 196t staging, 195 studies in early, 29 surgery axillary surgery, 198 breast conserving surgery, 197
471
Index
breast cancer (cont.) breast reconstruction, 198 modified radical mastectomy, 197 TNM classification, 195t trastuzumab for, 3, 18, 204 treatment adjuvant choices, 201 ductal carcinoma in situ (DCIS), 195 lobular carcinoma in situ (LCIS), 196 radiotherapy to regional lymphatics, 200 role of radiation boost, 200 treatment protocol importance, 1 triple assessment, 194 breast conserving surgery, vs. mastectomy, 204 bronchial carcinoid, 325 bronchoscopy interventional, 75 bronchus anatomy, 314 blockage of, 74 radical radiotherapy plan for, 49 BSA. See body surface area (BSA) buccal mucosa cancer of radical radiotherapy for, 105 Burkitt lymphoma in children, 432 risk of relapse, 350 and tumour lysis syndrome (TLS), 80 caesium-137 properties, 51 calcitonin, 71 and recurrent thyroid cancer recurrence, 415 Calvert equation, 4 cancer antigen (CA) 19-1 tumour marker, 153 cancer of unknown primary (CUP) basics, 442 clinical presentation, 443 diagnostic and treatment algorithm, 445 histological subtypes, 443t immunohistochemical stains for diagnosis, 444t incidence and epidemiology, 442 investigation, 443 palliative care, 447 prognosis, 447 unfavourable subsets, 446t risk factors and aetiology, 442 staging classification, 444 treatment, 446 chemotherapy, 447 radiotherapy, 446 surgery, 446
472
canertinib, 16 capecitabine (Xeloda® ), 2, 11 capecitabine palmar–plantar erythrodysaesthia (PPE) from, 8 in chemotherapy for colorectal cancer, 164, 169 for colorectal cancer (CRC), 173t diarrhoea from, 8 for liver cancer treatment, 145 for pancreatic cancer treatment, 156 carboplatin, 12 anaphylaxis from, 79 for cervical cancer treatment, palliative care, 286 dose calculation based on renal function, 4 for endometrial carcinoma, 273 for ovarian cancer treatment, 3, 261 for small cell lung cancer, 321 for testicular cancer treatment, 244 renal function measurement when using, 4 carcino-embryonic antigen (CEA) as potential vaccine target, 21 carcinoid syndrome, 325 and neuroendocrine tumours, 419 carcinoid, bronchial, 325 ‘carcinoma in situ’, 223 carcinosarcoma, 275 cardiac failure congestive, 35 cardiac function assessment before chemotherapy, 4 cardiac tamponade, 77 cardiotoxicity from chemotherapy, 7 carer’s allowance, 91 carina blockage of, 74 case studies in research, 56 caspase 8 as potential vaccine target, 21 castration, 31 cauda equina compression and malignant spinal cord, 72–73 celecoxib, 31 cells interferon alpha (IFNa) production by, 14 cellular vaccines, 20 central carcinoid, 325 central lines before chemotherapy, 5 central nervous system. See also brain anatomy, 370 primary lymphoma (PCNSL) clinical presentation, 353 investigation, 353
management, 353 central nervous system tumours areas of current interest, 379 characteristics, 370 in children, 432 brain stem glioma, 434 craniopharyngioma, 436 ependymoma, 434 high grade glioma, 434 low grade glioma, 433 medulloblastoma, 435 primitive neuroectodermal tumours, 435 site/incidence, 433 staging, 433 clinical presentation, 371 clinical trials, 380 incidence and epidemiology, 370 inherited syndromes and, 371 investigation and staging, 372 pathology, 371 prognosis, 379t recurrent disease, risk factors and aetiology, 370 spread, 371, 372t treatment, 373–377 WHO classification, 372t cephalosporin for neutropenia, 6 cerebral arteriovenous malformation (AVM), radiotherapy for, 449 cerebrospinal fluid flow obstruction, 74 Cervarix® , 20 cervical cancer areas of current interest, 287 as cause of death, 278 clinical presentation, 280 clinical trials, 288 HPV and, 20 incidence and epidemiology, 278 investigation and staging, 280 pathology, 280t in pregnancy, 286 prognosis, 287 radiographs from intracavitary bracytherapy treatment, 285 risk factors and aetiology, 278 screening for, 279 small cell carcinoma, 287 spread, 280 staging classification, 280, 281t survival, 287t treatment chemotherapy toxicities, 286 overview, 280 radiotherapy, 282–286 surgery, 282 tumour types, 278, 279t
Index
cervical intraepithelial neoplasia (CIN), grading, 280 cervical lymphadenopathy, 119–120 clinical presentation of, 119 investigation and staging, 119 staging classification, 119 treatment overview, 119 cervix, anatomy, 278 cetrorelix, 36 cetuximab (Erbitux® ), 18, 100 for metastatic colorectal cancer, 169 for pancreatic cancer treatment, 157 skin rash from, 17 charged particle therapy, for ocular melanoma, 404 CHART (Continuous Hyperfractionated Accelerated RT), 318 chemical agent exposure, and liver cancer, 142 chemoradiotherapy for cervical cancer treatment, 283 toxicity, 285 for non-small cell lung cancer stage III, 319 for vaginal carcinoma, 294 for vulval cancer, 300 chemotherapy adjuvant for breast cancer, 201 for rectal carcinoma, 165 administration, 8, 9 for advanced cutaneous melanoma, 400 palliative care, 401 aims of, 1 for anal cancer combined with radiotherapy, 178 palliative care, 180 anaphylaxis related to, 79 for bile tract cancer, 148 for bladder cancer, 228 for breast cancer, during pregnancy, 207 for cancer of unknown primary (CUP), 447 cell-cycle specificity of drugs, 2 for cervical cancer treatment, toxicity, 286 for childhood cancer, leukaemia, 431 for CNS tumours, 380 recurrence, 379 for colorectal cancer (CRC), 173t dose calculation, 3, 4 for endometrial carcinoma, 273 high risk, 274 for epithelial-stromal ovarian cancers, 261–262 extravasation of, 80 for gastric cancer adjuvant treatment, 137 palliative care, 137
perioperative, 137 with radiotherapy, 138–139 for gastrointestinal stromal tumours (GIST), 186 for gestational trophoblast tumours (GTT), 307–309 risk of secondary malignancy, 310 for head and neck tumours, 98 concurrent with radiotherapy, 98 of nasopharynx, 112 palliative care, 99 postoperative concurrent with radiotherapy, 98 for Hodgkin lymphoma, in children, 432 indications for, 1 for liver metastases from colorectal cancer, 168 for mesothelioma, 332 for myeloma, 365 for neuroendocrine tumours (NETs), 423 for non-small cell lung cancer, 320 with radiotherapy for stages III, 319 after surgery for stages I and II, 318 for stage III, 318 for stage IV, 320 for oesophageal cancer palliative care, 129 for oesphageal cancer preoperative, 125 for penile cancer, 256 pretreatment investigations and checks, 4, 5 protocols and guidelines, 3 for prostate cancer, palliative care, 238 scheduling and administration, 2, 3 for seminoma, 244 for small cell lung cancer, 321 for soft tissue sarcomas, 339 for testicular cancer relapse, 246 for thyroid cancer, 412, 416 for trophoblast disease, 309t for vulval cancer, 300 chemotherapy toxicities alopecia, 8 anaemia, 7 cardiotoxicity, 7 diarrhoea, 8 fertility and foetal abnormalities, 8 nausea and vomiting, 7 neutropenic fever and sepsis, 6 palmar-plantar erythrodysaesthesia (PPE), 8 phlebitis, 8 renal toxicity, 7 chest wall, pain from lung cancer, 316 childhood cancer basics, 426
bone tumours Ewing’s sarcoma/peripheral PNETs, 438 osteosarcomas, 438 central nervous system tumours, 432 brain stem glioma, 434 craniopharyngioma, 436 ependymoma, 434 high grade glioma, 434 low grade glioma, 433 medulloblastoma, 435 primitive neuroectodermal tumours, 435 site/incidence, 433 staging, 433 germ cell tumours, 440 hepatoblastoma, 440 hepatocellular carcinoma, 440 international co-operative groups, 426 leukaemias, 431 treatment, 431 lymphomas, 347, 431 Hodgkin lymphoma (HL), 431 non-Hodgkin lymphoma (NHL), 432 management basics, 427 neuroblastoma, 436 NICE guidelines, 440 possible causes, 427 radiotherapy, 428 effect on eyes/ears/teeth, 430 effect on growth, 429 effect on organ development, 430 late consequences, 428 neuropsychological/cerebral effects, 429 risk of second malignancies, 431 sexual development/fertility, 430 side effects, 428 retinoblastoma, 438 soft tissue sarcomas non-rhabdomyosarcomas, 440 rhabdomyosarcoma, 439 thyroid cancer, 412 types, 426 classification by age group, 426 Wilm’s tumour, 437 Child-Pugh grading system for cirrhosis, 145 Children’s Cancer and Leukaemia Group (CCLG), treatment guidelines for Hodgkin lymphoma, 360 chlorambucil, for follicular lymphoma, 354 cholangiocarcinoma, 146 cholinergic syndrome, 9 diarrhoea as component, 8 chondrosarcoma, managing, 343 CHOP chemotherapy for non-Hodgkin lymphoma, 2
473
Index
choriocarcinoma, 304, 306 chronic myelogenous leukaemia IFNa for, 14 circumcision, and penile cancer, 252 cirrhosis, Child-Pugh grading system for, 145 cisplatin, 12 antidotes for extravasation, 81t in BEP regimen for testicular cancer, 2 for bladder cancer treatment, 227 for cervical cancer, 281, 284 palliative care, 286 for cholangiocarcinoma palliative care, 148 for gestational trophoblast tumours (GTT), 309 for non-small cell lung cancer, 320 for oesophageal cancer, 127 for osteosarcoma, 343 renal function measurement when using, 4 renal toxicity, 7 for small cell lung cancer, 321 for testicular cancer metastatic disease, 247 relapsed disease, 249 for vulval cancer, 300 classical seminoma, 240 clinical target volume (CTV), 41 growth from GTV to, 41 clinical trial authorisations (CTA), 64 clinical trials, 55–63 blinding, 58 choice of endpoints, 57 choice of subjects, and how many, 59 in context, 67 data analysis - intention-to-treat principle, 60 data collection, 60 factors in success, 55 with negative or inconclusive results, 66 randomised controlled trial (RCT), 56 size of, 62 subgroups in, 61 Clinical Trials Toolkit, 65 clodronate, for bone metastases, 207 clostridium difficile and diarrhoea, 87 clubbing, finger and/or toe, 316 CMF (cyclophosphamide, methotrexate, 5-FU), for breast cancer treatment, 202 cobalt-60 properties, 51 Cockcroft Gault formula, 247 Cockcroft-Gault formula for calculating creatinine clearance, 4 co-codamol, 85t
474
co-danthramer for constipation, 86 codeine for diarrhoea, 87 colon, risk factors for malignant potential within polyps, 162t colonoscopy, for colorectal cancer screening, 161 colorectal cancer (CRC), 159 adjuvant chemotherapy for, 164 areas of current interest, 170 bevacizumab for, 19 chemotherapy, 173t clinical presentation criteria for referral to specialist, 162 symptoms and signs from primary tumour, 162 current clinical trials, 170 Dukes’, TNM and AJCC staging classifications, 163 follow up after radical therapy, 170 incidence and epidemiology, 159 investigation, 163 irinotecan-refractory, 18 locoregional recurrence, 168 morphology, 161 pathological staging, 163 pathology, 161, 162t range of tumours, 160t risk factors and aetiology associated conditions, 160 environmental factors, 160 family history, 159 screening and prevention, 159, 160 spread, 161 TNM classification, 164t treatment for advanced/inoperable, 167 treatment for metastatic disease, isolated liver metastases, 168 treatment for non-metastatic disease, 163 types, 159 common bile duct, 151 communication issues, 89–91 speed of information sharing, 90 compensation, for mesothelioma patients, 332 confidence intervals, 59 confusion and delirium, 88 congestive cardiac failure from DES, 35 CONSORT guidance, 67 constant region of antibodies, 18 constipation, 85 continent urinary diversion, as cystectomy option, 226 cost effectiveness research evidence of, 65
Cowden’s syndrome (PTEN), and breast cancer, 192 craniopharyngioma, in children, 436 craniospinal axis, radiotherapy of, 377, 378 toxicity, 378t creatinine clearance Cockcroft-Gault formula for calculating, 4 Crohn’s colitis, and colorectal cancer, 160 crying by patient, 90 cryosurgery, for skin cancer treatment, 385 cryptorchidism, and testicular cancer risk, 240 CT scan for neuroendocrine tumours (NETs), 421 for pancreatic cancer, 153 for prostate cancer, 233 CT simulation in radiotherapy planning, 40 curettage, for skin cancer treatment, 385 cutaneous B-cell lymphoma, 393 cutaneous melanoma areas of current interest, 402 Clark’s classification, 397t clinical features suggestive of, 398t clinical presentation, 397 clinical trials, 403 depth of invasion, 397 differential diagnosis, 398 incidence and epidemiology, 395 investigation and staging, 398 pathology, 396 prognosis, 401 five-year survival rate, 401t risk factors and aetiology, 395, 396t screening and prevention, 396 spread, 397 stage groupings, 399t TNM classification, 399t treatment adjuvant therapy, 400 for advanced disease, 400 endocrine treatment, 401 overview, 398 palliative care, 401 surgery, 399 surgery excision margins, 399t types, 395 cyclizine, 86t for treating nausea from chemotherapy, 7 cyclophosphamide, 11 for breast cancer, 202 for diffuse large B-cell lymphoma, 351, 352 for Ewing’s sarcoma, 343 for gestational trophoblast tumours (GTT), 310
Index
for Hodgkin lymphoma, in children, 432 for mantle cell lymphoma, 357 for metastatic soft tissue sarcoma, 340 for myeloma, 365 for non-Hodgkin lymphoma treatment, 2 reaction from, 81t renal toxicity, 8 CYP1B1 Inhibitors, 36 cyproterone acetate, 31 for prostate cancer, 31, 35 cystectomy, 225 patient fitness and attitude, 226 radical, 226 cytarabine, for diffuse large B-cell lymphoma, 352 cytogenetics, for renal cell tumour analysis, 215 cytokines, 13–15 combination regimens, 15 interferons, 14 interleukin-2 (IL-2), 13 for renal cell cancer, palliative care, 218 cytology, for breast cancer assessment, 194 cytosine arabinoside, for mantle cell lymphoma, 357 cytotoxic chemotherapy, 1 agents classified by reaction, 81t for prostate cancer, palliative care, 238 dacarbazine, 11, 12 for advanced cutaneous melanoma, 400 for Hodgkin lymphoma, 359 in children, 432 for metastatic soft tissue sarcoma, 340 phlebitis from, 8 dactinomycin, for Ewing’s sarcoma, 343 darbepoetin alpha, 20 dasatinib, for gastrointestinal stromal tumours (GIST), 188 data monitoring committees (DMC), 65 de novo resistance to endocrine therapy, 29 decision making ethical, 91 degarelix, 36 deleted in colon cancer (DCC) tumour suppressor gene, 161 dendritic cell vaccines, 21 denial as patient response, 89 density of organs inhomogeneities in radiotherapy planning, 45 depression, 88 as ADT adverse effect, 34 dermatofibrosarcoma protuberans, 393 dermoid cysts, 265 desmoid tumours, radiotherapy for, 450
dexamethasone, 35 for mantle cell lymphoma, 357 for myeloma, 365 for prostate cancer, 31 for treating nausea from chemotherapy, 7 DHA sulphate, 24 diamorphine, 76 diaphysis of long bone, 341 diarrhoea, 87 from chemotherapy, 8 dietary factors, and breast cancer incidence, 191 diethylstilboestrol for prostate cancer, 31 diethylstilboestrol (DES), 31, 35 and breast cancer incidence, 191 differentiated thyroid cancer areas of current interest, 413 chemotherapy, 412 clinical presentation, 408 external beam radiotherapy (EBRT), 411 follow up, 413 incidence and epidemiology, 408 investigation and staging, 409 prognosis, 413 radioisotope therapy, 410 risk factors and aetiology, 408 serum thyroglobulin, 409 surgery, 409 thyroxine for treating, 410 treatment overview, 409 diffuse large B-cell lymphoma clinical presentation, 351 prognosis, 353 risk of relapse, 350 treatment for advanced stage, 352 for early stage, 351 flow chart, 351 overview, 351 for relapsed/refractory disease, 352 digital rectal examination (DRE), for rectal cancer staging, 163 dihydroxyandrostenedione (DHA), 24 disability living allowance, 91 DNA-based vaccines, 21 docetaxel, 12, 31 anaphylaxis from, 79 for breast cancer treatment, 202 neutropenia from, 6 for non-small cell lung cancer, 320 for pancreatic cancer treatment, 156 domperidone for treating nausea from chemotherapy, 7 dose calculation for chemotherapy area under the curve (AUC) dosage, 4 body surface area (BSA), 3
body weight dosing, 4 dose capping, 4 flat dosing, 4 dose calculation for radiotherapy dose verification, 46 double-blind trials, 58 Down’s syndrome, and testicular cancer risk, 240 doxorubicin, 2, 11 for breast cancer treatment, 202 cardiotoxicity, 4, 7 for diffuse large B-cell lymphoma, 351, 352 for Hodgkin lymphoma, 359 in children, 432 for liver cancer palliative care, 144 for mantle cell lymphoma, 357 for metastatic soft tissue sarcoma, 340 for osteosarcoma, 343 for soft tissue sarcomas, 339 for thyroid cancer, 412 hepatic impairment and, 5 drop-outs from trial, 61 ductal carcinoma in situ (DCIS), 195 managing, 197t radiotherapy for, 195 role of axillary nodal surgery, 195 scoring index, 195, 197t tamoxifen for, 196 Dutch TME Trial, 166 dysgerminoma, 265 dysphagia, 87 grading system, 123 ear. See also middle ear anatomy, 118 Early Breast Cancer Trialists’ Collaborative Group (EBCTCG), 66 Early Prostate Cancer Trials, 34 ears, radiotherapy impact in children, 430 eccrine carcinoma, 392 effectiveness of treatment research on, 55 efficacy of treatment research on, 55 EGFR tyrosine kinase inhibitors, 16 elective lymph node dissection (ELND), 399 elective nodal treatment for head and neck tumours, 96 electrodesiccation, for skin cancer treatment, 385 electron depth dose curves, 391 electron depth dose table for 95 cm source-surface distance, 389 for 100 cm source-surface distance, 390
475
Index
embolisation, for neuroendocrine tumours (NETs), 423 emergencies in oncology, 70 metabolic, 70 obstruction from GIST, 186 structural/obstructive, 72–77 acute airway obstruction, 74 bleeding, 75 cardiac tamponade, 77 malignant spinal cord and cauda equina compression, 72–73 raised intracranial pressure, 74 superior vena caval obstruction (SVCO), 73–74 urinary obstruction, 76 treatment overview, 70 treatment-related anaphylaxis related to anticancer drugs, 79 extravasation of chemotherapy, 80 neutropenic fever and sepsis, 77 tumour lysis syndrome (TLS), 79 end of life palliative care at, 89 endocrine responsive breast cancer, 201 endocrine treatment for advanced cutaneous melanoma, 401 for breast cancer treatment, 203 endodermal sinus tumour, 265 endometrial carcinoma, 267 areas of current interest, 274 clinical presentation, 269 clinical trials, 275 investigation and staging, 269 pathology, 268t prognosis, 274 recurrent and metastatic disease, 274 risk factors and aetiology, 267 screening, 269 spread, 269 staging classification, 270t treatment brachytherapy, 271, 273 chemotherapy, 273, 274 hormonal therapies, 273, 274 overview, 270 radiotherapy, 271–273 radiotherapy toxicity, 272 surgery, 270 endometrial stromal sarcoma, 275 endoscopic laser photo-coagulation for gastric cancer palliative care, 139 endoscopic retrograde cholangiopancreaticogram (ERCP), 153 endoscopic therapy for inoperable colorectal cancer, 168 for oesophageal cancer, 129 endoscopic ultrasound (EUS), 153
476
for rectal cancer staging, 163 endpoint in cancer, 57 energy level reduction as ADT adverse effect, 34 environmental factors in gastric cancer, 133 ependymoma in children, 434 treatment, 374 EPIC (European Prospective Investigation into Cancer and Nutrition), 160 epidermal growth factor receptor (EGFR), 15 signalling, 100 epirubicin, 11 for soft tissue sarcomas, 340 epithelial growth factor receptor (EGFR), inhibitors of, 159 epithelial-stromal ovarian cancers, 257–264 areas of current interest, 263 chemotherapy, 261–262 classification, 257 clinical presentation, 259 clinical trials, 264 fertility sparing treatment, 260 hormone therapy, 263 investigation and staging, 259 palliative care, 263 pathology, 258 prognosis, 263 recurrent disease, 262 risk factors and aetiology, 257 screening and prevention, 259 spread, 258 surgery, 260 treatment overview, 259 ERBB2 HER-2 as potential vaccine target, 21 Erbitux® (cetuximab). See cetuximab (Erbitux® ) erlotinib (Tarceva® ), 16 for liver cancer, 145 for pancreatic cancer, 157 for recurrent disease, non-small cell lung cancer (NSCLC), 320 skin rash from, 17 erythroplasia of Queyrat, 252, 383 erythropoietin, 7, 20 ethanol injection, for liver cancer treatment, 144 ethical decision making, 91 ethical issues in research, 63 data monitoring committees (DMC), 65 good clinical practice and ethics committees, 64 informed consent, 64 study protocol, 63 ethics application process, 64
ethinyloestradiol, 35 ethmoid sinus carcinoma of TNM classification, 116t etoposide, 12 in BEP regimen for testicular cancer, 2 for diffuse large B-cell lymphoma, 352 for Ewing’s sarcoma, 343 for gestational trophoblast tumours (GTT), 310 for osteosarcoma, 343 reaction from, 81t for testicular cancer, metastatic disease, 247 European Prospective Investigation into Cancer and Nutrition (EPIC), 160 European Union (EU) Directive 64 Ewing’s sarcoma, 335, 341 anatomy and pathology, 341 in children, 438 metastases, 341 radiotherapy, 344 risk factors and aetiology, 341 treatment, 343 exemestane, 27, 28, 203 vs. tamoxifen, 29 extensive intraduct component (EIC), 209 external beam radiotherapy (EBRT) for liver cancer treatment, 144 for medullary thyroid cancer, 414 for oesophageal cancer palliative care, 129 for thyroid cancer, 411 for thyroid lymphoma, 416 extranodal marginal zone B cell lymphoma, 357 managing, 357 prognosis, 358 extravasation, 8 antidotes for, 81t risk factors for, 81t eyes, radiotherapy impact in children, 430 for macular degeneration, 451 shield for, 386 face, pain from lung cancer, 316 factorial design of clinical trials, 62t faecal occult blood (FOB), and colorectal cancer screening, 161 familial adenomatous polyposis (FAP), 160 familial retinoblastoma, and soft tissue sarcomas, 336 family history. See heredity febrile neutropenia treatment, 78 Federation Internationale de Gynaecologie et d’Obstetrique (FIGO)
Index
cervical cancer staging, 281t endometrial carcinoma staging, 270t GTT prognostic scoring system, 308 GTT staging classification, 307t ovarian cancer staging, 260 staging classification, 259 vaginal tumour classification, 290 vulval cancer stage groupings, 299t female. See also breast cancer; ovaries; uterus age, and molar pregnancy risk, 305 bladder anatomy, 222 hypothalamic-pituitary-gonadal axis in, 24 Ferguson-Smith disease, 383 fertility from chemotherapy, 8 after chemotherapy for GTT, 310 endometrial carcinoma treatment and, 274 mantle radiotherapy and, 363 ovarian cancer surgery and, 260 testicular cancer treatment and, 249 fever neutropenic, 6 fibre in diet, and colorectal cancer, 160 fibromatosis, aggressive, radiotherapy for, 450 fibrosarcoma, 336 fibrous histiocytoma, malignant, 393 financial considerations in palliative care, 91 fingers, clubbing, 316 flat dosing, 4 fludarabine, 11 for mantle cell lymphoma, 357 reaction from, 81t for relapsed follicular lymphoma, 356 fluid retention from steriodal antiandrogens, 35 fluoropyrimidines in chemotherapy for colorectal cancer, 164 diarrhoea from, 8 for liver metastases from colorectal cancer, 169 fluoroquinolone, 79 5-fluorouracil 2, 11 for breast cancer treatment, 202 cardiotoxicity, 7 for cholangiocarcinoma, 148 for colorectal cancer (CRC), 173t diarrhoea from, 8 for gastric cancer, 137 for oesophageal cancer, 127 reaction from, 81t for skin cancer treatment, 385 flutamide, 34 for prostate cancer, 31, 237
foetal abnormalities from chemotherapy, 8 folinic acid, 63 follicular cancers, of thyroid, 407 follicular lymphoma, 354 clinical presentation, 354 clinical trials, 356 management, 354, 356 prognosis, 354 treatment of advanced, 354 flow chart, 355 of localised, 354 of relapsed, 356 forward planning in IMRT, 47 5-FU folinic acid based chemotherapy, 63 Fuhrman nuclear grading system, for renal cell tumours, 215 fulvestrant, 27, 30 for breast cancer treatment, 27 furosemide, 71 gall bladder anatomy, 146 blood tests, 147 carcinoma incidence and epidemiology, 145 photodynamic therapy (PDT), 148 risk factors and aetiology, 146 tumour types, 145 clinical presentation, 146 imaging, 147 treatment overview, 147 surgery, 148 tumours, 141 types, 146t gangliosides, 20 as potential vaccine target, 21 Gardasil® , 20 Gardner’s syndrome, 160 and soft tissue sarcomas, 336 gastrectomy, 136 surgical scheme, 136 Billroth II partial, 136 gastric cancer, 132 clinical presentation, 134 incidence and epidemiology, 132 investigation and staging, 134 pathological features of diffuse-type, 134t pathological features of intestinal-type, 134t pathology, 133 prognosis, 139 risk factors, 133 spread, 134 stage classification, 135 stage groupings, 135t
treatment chemoradiotherapy side effects, 139t chemotherapy, 137–138 concurrent chemoradiotherapy, 139 overview, 135 palliative radiotherapy, 139 radiotherapy and chemoradiotherapy, 138–139 surgery, 135 surgery, patterns of failure after, 136t tumour types, 132, 133t gastrografin, 166 gastrointestinal autonomic tumour (GANT), 183 gastrointestinal stromal tumours (GIST), 15, 183 clinical presentation, 185 clinical trials, 188 dasatinib, 188 sunitinib (Sutent® ), 188 follow up after surgery, 187 histogenesis, 183 imatinib (Glivec® ) for, 186–187 contrast enhanced CT for assessing response, 187 response rate, 188 toxicities, 187 immunohistochemical features, 184t incidence and epidemiology, 183 investigation and staging, 185 KIT-negative, 187 molecular classification, 184t palliative care, 187 pathology, 184 prognosis, 188t risk factors and aetiology, 183 special clinical situations, 187 spread, 184 sunitinib activity against, 17 treatment chemotherapy, 186 overview, 185 radiotherapy, 186 surgery, 185 tumour types, 183 gastro-oesophageal reflux disease (GORD), 121 gefitinib (Iressa® ), 16 skin rash from, 17 gemcitabine, 11 for bladder cancer treatment, 227 for cholangiocarcinoma palliative care, 148 for Hodgkin lymphoma, 359 for liver cancer treatment, 145 for metastatic or inoperable pancreatic tumour, 156 for non-small cell lung cancer, 320
477
Index
gemcitabine (cont.) for pancreatic cancer treatment, 156 reaction from, 81t gene array technology, 66 gene mutation and breast cancer, 192 and gastric cancer, 133 genetics, in renal cell carcinoma, 214 geographic variation, in breast cancer occurrence, 190 germ cell tumours in children, 440 in ovaries, 265 classification, 257 in testicular cancer clinical presentation, 241 incidence and epidemiology, 240 investigation and staging, 241 pathology and classification, 240 risk factors and aetiology, 240 screening for, 240 spread, 241 testicular cancer types, 241t gestational trophoblast tumours (GTT), 304 chemotherapy, 307–309 investigation, 307 malignant forms, 306–307 choriocarcinoma, 306 invasive mole, 306 placental site trophoblast tumour (PSTT), 307 risk of relapse and late treatment complications, 309 serum hCG graphs during treatment, 310 staging classification and prognostic classification, 307t types, 304, 305t. See also molar pregnancy complete molar, 304 partial molar, 304 pre-malignant forms, 304 Gleason system, 232t glioma brain stem, in children, 434 high grade in children, 434 treatment, 374, 377 low grade in children, 433 treatment, 373, 377 glomerular filtration rate (GFR), 4 glottis carcinoma of radiotherapy for, 102 TNM stages, 101 treatment, 100 glucocorticoids, 35
478
glycolytic activity, identifying increased, 350 gonadotropin releasing hormone (GnRH/LHRH) agonists, 26 good clinical practice (GCP) aims of, 64 defining, 63 in randomisation, 58 Gorlin’s syndrome, 383 and soft tissue sarcomas, 336 goserelin, 26, 31 for ovarian function suppression in breast cancer treatment, 203 for prostate cancer, 31 granisetron, 86t for treating nausea from chemotherapy, 7 granulocyte-colony stimulating factor (G-CSF), 19 for febrile neutropenia, 78 prophylaxis, 6 granulosa cell tumours, in ovaries, 264 gross tumour volume (GTV), 41 delineating, 41 imaging to determine, 40 growth hormone, deficit from radiotherapy, 429 gynaecological brachytherapy Manchester system for, 52 gynaecomastia from DES, 35 from non-steroidal antiandrogens, 34 H. pylori infection eradication therapy, 357 and MALT lymphoma, 357 5-HT3 (5-hydroxy-tryptamine 3) antagonist for treating nausea from chemotherapy, 7 haematemesis treatment, 75, 76 haematogenous metastatic spread of oesophageal tumour, 123 haematological malignancies, 347. See also lymphomas IFNa for, 14 paraproteinaemias/myeloma, 363–366 chemotherapy, 365 classification, 363 clinical presentation, 364 diagnosis and new definitions, 364 incidence and epidemiology, 364 investigation and staging, 364 pathology, 364 radiotherapy, 365 risk factors and aetiology, 364 treatment overview, 365
haematuria, 75 and bladder cancer, 223 screening for, 222 treatment, 76 haemopoietic colony stimulating factors, 19 haemoptysis treatment-related, 75 hairy cell leukaemia IFNa for, 14 haloperidol, 86t head and neck tumours, 93. See also larynx; maxillary antrum; middle ear; nasal cavity; nasopharynx; oral cavity; salivary gland anatomy, 93 areas of current interest, 99 chemotherapy for, 98 postoperative concurrent with radiotherapy, 98 clinical presentation of, 94 clinical trials, 100 concurrent chemoradiotherapy for, 98 incidence and epidemiology, 93 investigation and staging, 95 pathology, 94 prognosis, 99 radiotherapy for, 96–98, 361 fractionation, 98 indications for postoperative, 97 management of acute reactions, 97 rehabilitation following radical treatment, 99 risk factors and aetiology, 93 screening for, 94 solitary extramedullary plasmacytoma, 366 spread, 94 surgery for, 96 treatment overview, 95 elective nodal treatment, 96 types, 94t health economics and research, 65 heart. See cardiac . . . heat shock protein 90 (HSP-90) and androgen receptor, 25 hedgehog (HH) signalling pathway, 383 height before chemotherapy, 5 Helicobacter pylori infection and gastric cancer, 133 hemoptysis treatment, 76 hepatic arterial chemotherapy, for liver cancer treatment, 144 hepaticojejunostomy, 148 hepatitis, and liver cancer, 142 hepatoblastoma, in children, 440
Index
hepatocellular carcinoma. See liver cancer hepatoxicity DES and, 35 from steriodal antiandrogens, 35 HER-2 oncogene, and breast cancer, 209 Herceptin® (trastuzumab). See trastuzumab (Herceptin® ) heredity and breast cancer incidence, 192 and medullary thyroid cancer, 407 and neuroendocrine tumours, 418 and ovarian cancer risk, 258 and skin cancer risk, 383 L’Hermitte’s sign, 363 heterotropic bone formation (HBF), radiotherapy for, 450 Hickman line, for chemotherapy, 5 high dose chemotherapy (HDCT), for relapsed testicular cancer, 249 histiocytoma, fibrous, 393 histiosarcoma, malignant fibrous, 336 HIV/AIDS patients, anal cancer risk, 180 Hodgkin lymphoma (HL), 347, 358–361 in children, 431 managing advanced disease, 359 radiotherapy in, 360 managing early stage, 359 managing relapsed/refractory, 360 nodular lymphocyte predominant, 401t treatment approach, 358 treatment flow chart, 359 homogenous dose to the PTV in critical analysis of treatment plans, 48 hormonal therapies, 23 as adjunct to radical radiotherapy, 32 for breast cancer, 25, 27 for endometrial carcinoma, 273, 274 for epithelial-stromal ovarian cancers, 263 historical landmarks in anticancer, 23 for metastatic breast cancer, 205 for prostate cancer, 31 palliative care, 238 for renal cell cancer, 219 hormone positive tumours resistance to endocrine therapy, 29 hormone receptor status, and breast cancer prognosis, 209 hormone replacement therapy (HRT), and breast cancer incidence, 191 hormones classification, 23 receptors, 24 synthesis, 23 human epidermal growth factor receptor 2 (HER-2) protein as trastuzumab target, 18 human papilloma virus (HPV) and anal cancer, 174
and cervical cancer, 278 and Kaposi’s sarcoma, 392 and penile cancer, 252 vaccine against, 20, 287 and vaginal cancer, 290 ¨ Hurthle cell/oxyphil tumours, 407 hydrocephalus obstructive, 74 hydronephrosis, 226 5-hydroxyindoleacetic acid (5-HIAA), neuroendocrine tumours (NETs) and, 420 hypercalcaemia, 70 lung cancer and, 316 hypernephroma, 214 hypersensitivity chemotherapy and, 9 hypertrophic pulmonary osteo-arthropathy (HPOA), 316 hypocalcaemia, 71 hypogammaglobulinaemia, thymomas and, 325 hypopharynx anatomy, 108 carcinoma of, 108 clinical presentation of, 108 five-year survival rate, 110t investigation and staging, 108 prognosis, 110 radical radiotherapy for, 110 surgery for, 110 TNM classification, 109t treatment, 109 hypothalamic-pituitary-gonadal axis in female, 24 in male, 24 hypothyroidism, from mantle radiotherapy, 363 hypoxia, cervical cancer and, 287 hysterectomy, 270 for cervical cancer treatment, 282 ibandronate, for bone metastases, 207 idarubicin, 11 for myeloma, 365 IFNAR, 14 ifosfamide, 11 for diffuse large B-cell lymphoma, 352 for metastatic soft tissue sarcoma, 340 for osteosarcoma, 343 renal toxicity, 8 for small cell lung cancer, 321 for soft tissue sarcomas, 339 for testicular cancer, relapsed disease, 249 ifosphamide reaction from, 81t ileal conduit, as cystectomy option, 226 imaging the patient
in radiotherapy planning, 40 imatinib (Glivec® ), 15, 16 acute gastrointestinal haemorrhage from, 186 for gastrointestinal stromal tumours (GIST), 186–187 contrast enhanced CT for assessing response, 187 preoperative, 185 response rate, 188 toxicities, 187 imiquimod 5%, for skin cancer treatment, 385 immunotherapy after radical nephrectomy, 218 for advanced cutaneous melanoma, 400 palliative care, 401 implantable chemotherapy, for CNS tumours, 380 impotence as ADT adverse effect, 34 from non-steroidal antiandrogens, 34 incapacity benefit, 91 income support, 91 induction chemotherapy for head and neck tumours, 98 industrial exposure to chemicals, as bladder cancer risk factor, 222 infants, cancers typically occurring, 426 infertility. See fertility inflammatory breast cancer, 208 information speed of sharing with patient, 90 informed consent, 64 before chemotherapy, 4 inguinal/pelvic fields, radiotherapy for lymphoma, 362 Inhibin, 23 intensity-modulated radiation therapy (IMRT), 47, 99 for bladder cancer treatment, 227 for breast cancer treatment, 200 for cervical cancer treatment, 287 forward planning, 47 inverse planning, 47 for liver cancer treatment, 144 for pancreatic cancer treatment, 157 for prostate cancer, 239 intention-to-treat principle, 60, 61 interferon alpha (IFNa), 14 interferon therapy for cutaneous melanoma treatment, 402 for liver cancer treatment, 144 for neuroendocrine tumours (NETs), 423 for renal cell cancer after radical nephrectomy, 218 palliative care, 218
479
Index
interferons, 14 interleukin-2 (IL-2), 13 after radical nephrectomy, 218 for renal cell cancer, palliative care, 219 internal margin (IM), 41 International Classification of Diseases, 335 International Commission on Radiation Units and Measurements (ICRU), 41, 51 reference point selection, 45 International Committee on Harmonisation (ICH) guidelines, 64 International Federation. See Federation Internationale de Gynaecologie et d’Obstetrique (FIGO) International Germ Cell Consensus Classification, 242 International Mesothelioma Interest Group (IMIG), staging system, 331 International Prognostic Factor Score (IPFS), for Hodgkin lymphoma, 359 International Prognostic Index (IPI), for non-Hodgkin lymphoma, 349 intracranial pressure raised, 74 intraepithelial carcinoma, 252 intra-luminal brachytherapy (ILT) for oesophageal cancer palliative care, 129 intrathecal chemotherapy, 8 intratubular germ cell neoplasia (IGCN), 240 managing, 242 intravenous contrast in CT scan, 40 invasive mole pregnancy, 306 inverse planning in IMRT, 47 investigational medicinal product (IMP), 64 involved field radiotherapy, for lymphomas, 361 iodine-125 properties, 51 Iressa® (gefitinib). See gefitinib (Iressa® ) iridium-192 properties, 51 wires, 51 irinotecan, 5, 12 cholinergic syndrome from, 9 for colorectal cancer (CRC), 164, 173t for palliative care, 169 for liver cancer treatment, 145 for liver metastases, 169 for pancreatic cancer treatment, 156 reaction from, 81t irinotecan-refractory colorectal cancer, 18 irradiated volume, 41
480
isodose plan - photons, 42 production, 44 Japanese Gastric Cancer Association nodal stations, 133 jargon in patient communications, 90 jaundice, obstructive, investigation strategy for, 147 jugulo-digastric (JDG) node, 108 justice decision making and, 91 juvenile granulosa cell tumour, of the ovary, 264 keloid, radiotherapy for, 451 keratoacanthoma, 384 ketoconazole, 31 kidney anatomy, 214 cancer of. See also renal cell carcinoma areas of current interest, 220 biological treatment, 218 chemotherapy, 219 clinical presentation, 215 clinical trials, 220 hormonal therapies, 219 incidence and epidemiology, 214 laparoscopic nephrectomy, 217 metastectomy, 217 occurrence, 214 palliative nephrectomy, 217 pathology, 215, 216 prognosis, 219t, 220 radical nephrectomy, 216 radiotherapy, 217 spread, 215 TNM stages, 217t treatment overview, 216 tyrosine kinase inhibitors, 220 chemoradiotherapy tolerance dose for pancreatic tumour, 155 cisplatin impact on function, 247 function assessment before chemotherapy, 4 radiotherapy impact in children, 430 toxicity from chemotherapy, 7 tumour types, 214 differentiating benign from malignant, 216 Heidelberg classification, 215t staging classification, 216 KIT tyrosine kinases imatinib activity against, 15 KIT-negative GIST, 187 Klinefelter’s syndrome, and testicular cancer risk, 240 knowledge patient need for, 89–91
laboratory research, 66 lactulose for constipation, 85 laparoscopic hysterectomy, 271 laparoscopic nephrectomy, 217 laparoscopy, for pancreatic cancer investigation, 153 lapatinib, 16, 17 skin rash from, 17 larynx anatomy, 100 blockage of, 74 carcinoma of, 48, 100–103 clinical presentation of, 100 five-year survival rate, 103t investigation and staging, 100 nodal and TNM stages, 101 radiotherapy dose, 103 radiotherapy for, 102 treatment, 100 laser therapy for oesophageal cancer, 129 laxatives, 85 lead shields for skin cancer treatment with radiotherapy, 386 leiomyosarcoma, 275, 336 lentigo maligna melanoma (LMM), 396 letrozole, 28, 203 for breast cancer treatment, 27 vs. tamoxifen, 29 leukaemia and tumour lysis syndrome (TLS), 80 in children, 431 treatment, 431 leuprorelin, 31 for prostate cancer, 31 levamisole, 63 levofloxacin, 79 Leydig cells of the testis, 250 testosterone production, 24 Li Fraumeni syndrome, and soft tissue sarcomas, 336 ligand binding domain, 25 limb salvage surgery, for soft tissue sarcoma, 338 liposarcoma, 336 liposomal doxorubicin, 11 PPE from, 8 lips cancer of, 104 five-year survival rate, 105t radical radiotherapy for, 105 liver anatomy, 141 chemoradiotherapy tolerance dose for pancreatic tumour, 155 function assessment before chemotherapy, 5
Index
isolated metastases from colorectal cancer in-situ destructive therapies, 168 palliative chemotherapy, 169 surgery for, 168 radiotherapy for lymphoma, 362 liver cancer, 141–145 angiosarcoma, polyvinyl chloride monomers and, 145 in children, 440 clinical presentation, 142 diagnostic and staging investigations, 142 incidence and epidemiology, 141 metastatic disease, treatment, 447 pathology, 142t prognosis, 145 radiotherapy, 144 risk factors and aetiology, 141 systemic therapy, 144 TNM classification, 143t treatment local non-surgical therapy, 144 overview, 143 surgery, 143 tumour types, 142t lobectomy, for thyroid cancer, 409 lobular carcinoma in situ (LCIS), 196 lomustine, 11 loperamide for diarrhoea, 87 lorazepam for anticipatory nausea, 7 lung radiotherapy for lymphoma, 362 radiotherapy impact in children, 430 lung cancer. See also non-small cell lung cancer (NSCLC); small cell lung cancer areas of current interest and clinical trials, 324 brachytherapy for, 323 clinical presentation, 315–316 endobronchial intervention, 75t external beam radiotherapy (EBRT), 322 incidence and epidemiology, 313 investigations, 316 metastatic disease, treatment, 447 mortality statistics, 313 non-small cell (NSCLC), 16 pain from, 316 palliative care for, 313, 323 pathology, 314, 315 prognosis, 324 radiotherapy, side effects, 323t rare tumour types, 325 bronchial carcinoid, 325 thymoma, 325 risk factors and aetiology, 313
spread, 315 staging, 317 tumour types, 313 lungs anatomy, 314 tumour types, 314t luteinising hormone releasing hormone (LHRH) agonists, 31 new options, 36 lymph nodes in bladder cancer, 227 in breast cancer dissection in treatment, 198 histopathological assessment, 193 radiotherapy for, 200 drainage from lungs, 314 elective dissection (ELND), 399 and endometrial carcinoma spread, 269 melanoma spread to, 397 oesophageal tumour spread to, 122 para-aortic, radiotherapy for cervical cancer, 285 prostate cancer spread to, 233 regional status and breast cancer prognosis, 209 regions, 349 sentinel lymph node biopsy (SLNB), 399 and spread of anal cancer, 176 surgery for, 96 lymphadenectomy for cervical cancer treatment, 282 in endometrial carcinoma treatment, 270, 271 lymphadenopathy cervical, 119–120 lymphatic drainage from breast, 190 lymphoedema, and soft tissue sarcomas, 336 lymphoma treatment protocol importance, 1 and tumour lysis syndrome (TLS), 80 lymphomas. See also Hodgkin lymphoma (HL) aetiology, 347 basics, 347 in children, 431 clinical presentation, 347 CNS prophylaxis in, 350 diffuse large B-cell clinical presentation, 351 prognosis, 353 risk of relapse, 350 treatment flow chart, 351 treatment for advanced stage, 352 treatment for early stage, 351 treatment for relapsed/refractory disease, 352 treatment overview, 351
extranodal marginal zone B cell, 357 managing, 357 prognosis, 358 follicular, 354 clinical presentation, 354 clinical trials, 356 management plan, 354 managing, 356 prognosis, 354 treatment flow chart, 355 treatment of advanced, 354 treatment of localised, 354 treatment of relapsed, 356 investigation and staging, 349 mantle cell clinical presentation, 356 prognosis, 357 treatment, 357 mycosis fungoides (MF), 358 managing, 358 prognosis, 358 PET scanning for, 350 pretreatment assessment, 350 primary CNS clinical presentation, 353 investigation, 353 management, 353 primary mediastinal B-cell clinical presentation, 353 prognosis, 353 treatment, 353 radiotherapy for extended field, 362 involved field, 361 staging classification, 349t thyroid, 416 treatment late complications, 363 WHO classification, 348 lymphosarcoma, 393 lymphovascular invasion, and risk of testicular cancer relapse, 246 MabThera® (rituximab). See rituximab (MabThera® ) macular degeneration, radiotherapy for, 451 magnesium cisplatin impact on excretion, 5 magnesium hydroxide for constipation, 85 male. See also prostate cancer bladder anatomy, 222 breast cancer, 208 hypothalamic-pituitary-gonadal axis in, 24 malignant fibrous histiosarcoma, 336 malignant spinal cord and cauda equina compression, 72–73
481
Index
mammography for breast cancer screening, 193 finding in DCIS, 195 Manchester Interstitial System for brachytherapy, 52 Manchester score, 321 mantle cell lymphoma clinical presentation, 356 prognosis, 357 treatment, 357 mantle field, radiotherapy for lymphoma, 362 margins for skin cancer, 386 Marjolin’s ulcer, 383 Masoaka staging system, 325, 326t mastectomy vs. breast conserving surgery, 204 breast reconstruction after, 198 modified radical, 197 radiotherapy after, 199 maxillary antrum anatomy, 113 carcinoma of, 113–115 clinical presentation of, 113 CT-planned radical radiotherapy treatment, 115 five-year survival rate, 115t investigation and staging, 114 pathology, 113 prognosis, 115 radiotherapy, 114 surgery for, 114 TNM summary of staging, 114t treatment overview, 114 maximally tolerated dose (MTD), 56 mechlorethamine reaction from, 81t mediastinal pain, from lung cancer, 316 mediastinum anatomy, 314 radiotherapy for lymphoma, 361 Medicines and Healthcare Products Regulatory Agency (MHRA), 64 Medicines Compendium, 13 medroxyprogesterone acetate for breast cancer treatment, 29 for prostate cancer, 35 medullary thyroid cancer, 407 clinical presentation, 414 external beam radiotherapy (EBRT), 414 follow up, 415 incidence and epidemiology, 414 prognosis, 415 risk factors and aetiology, 414 spread, 414 treatment, 414 treatment of recurrent, 415
482
medulloblastoma in children, 435 treatment, 374 megestrol acetate, 27 for breast cancer treatment, 29 for prostate cancer, 35 melanocytic naevi, and cutaneous melanoma risk, 395 melanoma. See also cutaneous melanoma annual incidence in different populations, 396t basics, 395 IFNa in adjuvant management of, 14 ocular clinical features and spread, 403 incidence and epidemiology, 403 investigation, 403 treatment, 403 melanoma antigen expression family (MAGE) as potential vaccine target, 21 melphalan, for myeloma, 365 meningioma, treatment, 374, 377 Merkel cell carcinoma, 392 mesothelioma basics, 328 pleural areas of current interest, 333 chemotherapy, 332 clinical presentation, 330 clinical trials, 333 compensation, 332 incidence and epidemiology, 328 investigation and staging, 330 palliative care, 332 pathology, 329t prognosis, 332 radiotherapy, 332 risk factors and aetiology, 328 spread, 330t staging classification, 330, 331 surgery, 331 treatment overview, 330 meta-analysis, 66 metabolic emergencies, 70 metastatic disease. See also cancer of unknown primary (CUP) from anal cancer, 176 from bladder cancer, 224 in brain, treatment, 375 breast cancer and, 194 from colorectal cancer (CRC), isolated liver metastases, 168 endometrial carcinoma, 274 pancreatic carcinoma, chemotherapy for, 156 from prostate cancer, 233 hormonal therapies for palliative care, 238
radiotherapy for palliative care, 238 renal cell carcinoma, 14 soft tissue sarcomas, 340 testicular cancer, 249 from thyroid cancer, 412 metastatic melanoma IFNa for, 14 IL-2 for, 14 methotrexate, 11 for breast cancer treatment, 202 for childhood cancer, neuropsychological/cerebral effects, 430 for diffuse large B-cell lymphoma, 352 for gestational trophoblast tumours (GTT), 307, 310 for mantle cell lymphoma, 357 for osteosarcoma, 343 for primary CNS lymphoma, 353 reaction from, 81t renal toxicity, 8 for vulval cancer, 300 metoclopramide, 86t for treating nausea from chemotherapy, 7 metronidazole, for H.pylori infection eradication, 357 midazolam, 76 middle ear anatomy, 118 carcinoma of clinical examination, 118 investigation and staging, 118 prognosis, 119 radical radiotherapy for, 119 spread, 118 surgery for, 119 treatment overview, 119 mitomycin, 11 antidotes for extravasation, 81t for bladder cancer treatment, 225 for cholangiocarcinoma palliative care, 148 reaction from, 81t mitomycin C neutropenia from, 6 mitozantrone, 11 ¨ mixed mullerian tumour, 275 moderate benefits expectations from treatment, 56 Mohs’ micrographic surgery, 385 molar pregnancy, 304 incidence, 305 indications for further treatment, 306 invasive, 306 risk factors, 305 screening after, 305 monoclonal antibodies classification and nomenclature, 18t for metastatic colorectal cancer, 169
Index
monoclonal gammopathy of undetermined significance (MGUS), 363 mood changes as ADT adverse effect, 34 morphine, 83 dose calculation, 85 potencies, 85t mouth. See oral cavity mouth floor cancer of radical radiotherapy for, 105 MRI scans for neuroendocrine tumours (NETs), 421 for prostate cancer, 233 for radiotherapy planning, 40 for rectal cancer staging, 163 mucin 1 (MUC1) as potential vaccine target, 21 Muir-Torre syndrome, 383 and breast cancer, 192 mustine, for Hodgkin lymphoma, 359 myasthenia gravis, thymomas and, 325 mycosis fungoides (MF), 358 managing, 358 prognosis, 358 myelogenous leukaemia chronic, 14 myeloma clinical presentation, 364 diagnosis and new definitions, 364 incidence and epidemiology, 364 investigation and staging, 364 pathology, 364 risk factors and aetiology, 364 treatment chemotherapy, 365 overview, 365 radiotherapy, 365 naevoid basal cell carcinoma syndrome, 383 nasal cavity anatomy, 115 carcinoma of clinical presentation of, 115 investigation and staging, 115 pathology, 115 radical radiotherapy for, 116 spread, 115 surgery for, 115 TNM classification, 116t nasopharynx anatomy, 110 carcinoma of, 110–113 chemotherapy, 112 clinical presentation, 111 diagnosis and staging, 111
five-year survival rate, 113t incidence and epidemiology, 111 prognosis, 113 radiotherapy, 112–113 risk factors and aetiology, 111 TNM classification, 111t treatment overview, 111 tumour types, 110 National Cancer Institute (USA) on cancer vaccines, 21 National Cancer Research Network (NCRN), 65 National Comprehensive Cancer Network (NCCN), 19, 201 National Institute for Health and Clinical Excellence (NICE), childhood cancer guidelines, 440 National Institute of Health (NIH), 201 National Translational Cancer Research Network (NTRAC), 66 nausea and vomiting from chemotherapy, 7 managing, 84 neck. See also head and neck tumours radiotherapy for lymphoma, 361 neoadjuvant chemotherapy for head and neck tumours, 98 neobladder, orthotopic, 226 neuroblastoma, 436 in children, 436 neuroendocrine tumours (NETs) areas of current interest, 424 basics, 408t clinical presentation, 419 carcinoid syndrome, 419 clinical trials, 424 incidence and epidemiology, 418 investigation and staging, 420 biochemistry, 420 radiological and nuclear imaging, 421 metastases at presentation by primary site, 419t pathology, 419 prognosis, 424 risk factors and aetiology, 418 spread, 419 treatment, 421 chemotherapy, 423 drug therapy, 422 embolisation and radiofrequency ablation, 423 radionuclide therapy, 423 surgery, 421 tumour types, 418 neurofibromatosis type 1, and soft tissue sarcomas, 335 neurokinin A, neuroendocrine tumours (NETs) and, 420
neutropenia chemotherapy and, 5, 6 neutropenic fever and sepsis, 77 Nexavar® (Sorafenib). See sorafenib (Nexavar® ) NHL. See non-Hodgkin lymphoma (NHL) nipple, changes in breast cancer, 194 nodular lymphocyte predominant Hodgkin lymphoma (HL), 401t nodular melanoma (NM), 396 non-coplanar beams in radiotherapy, 43 non-Hodgkin lymphoma (NHL), 347 CHOP chemotherapy for high-grade B-cell, 2 in children, 432 prognostic index, 349 rituximab for, 19 in thyroid, 407 non-maleficence decision making and, 91 non-receptor tyrosine kinases, 15 non-rhabdomyosarcomas, in children, 440 non-small cell lung cancer (NSCLC), 16, 315 areas of current interest and clinical trials, 324 external beam radiotherapy (EBRT), 322 prognosis, 324 stage groups with five-year survival, 318t staging, 317 treatment, 317–321 chemoradiotherapy for stage III, 319 chemotherapy after surgery for stages I and II, 318 chemotherapy for stage III, 318 chemotherapy for stage IV, 320 chemotherapy with radiotherapy for stages III, 319 palliative radiotherapy for stage III, 319 palliative radiotherapy for stage IV, 319 radiotherapy for stage III, 319 radiotherapy for stages I and II, 317 recurrent disease, 317–321 survival after palliative treatment, 324t survival after radiotherapy, 324t non-steroidal antiandrogens (NSAAs), 34 for prostate cancer treatment, 237 non-steroidal hormones, 23 North London Cancer Network, 4 nose. See nasal cavity Nottingham prognostic index (NI), 209 nuclear medicine in radiotherapy planning, 40
483
Index
obesity, and breast cancer incidence, 191 occupational factors, in renal cell carcinoma, 214 octreotide for diarrhoea, 87 for neuroendocrine tumours (NETs), 422 ocular melanoma clinical features and spread, 403 incidence and epidemiology, 403 investigation, 403 treatment, 403 odynophagia managing, 97 oesophageal cancer adenocarcinoma increase, 121 areas of current interest dose of radiotherapy, 130 clinical presentation, 123 clinical trials, 130 incidence and epidemiology, 121 investigation and staging, 123 pathology, 122, 123t prognosis, 130 radiotherapy beam direction, 43 risk factors and aetiology, 121 small cell carcinoma, 130 spread, 122 staging classification, 124t treatment definitive chemoradiotherapy, 126 overview, 124 palliative care, 129 postoperative adjuvant therapy, 125 preoperative chemoradiotherapy, 126 preoperative chemotherapy, 125 radiotherapy and chemoradiotherapy technique, 126–128 radiotherapy side effects, 129t of recurrent, 128 surgery for non-metastatic, 125 tumour types, 121, 122t oesophageal-pharyngeal syndrome, 9 oesophagus anatomy, 121, 123 oestrogen effects, 25 receptors (ERs), 24, 25 activation domains, 203 synthesis, 23 omeprazole, for H.pylori infection eradication, 357 Oncotype DX Recurrence Score, 202 ondansetron, 86t for treating nausea from chemotherapy, 7 opioids toxicity, 83 optic nerve shielding in radiotherapy, 113
484
oral cavity anatomy, 103 carcinoma of, 103–106 five-year survival rate, 105t investigation and staging, 103 radical radiotherapy for, 104 TNM classification, 104t treatment, 104 oral chemotherapy and over-compliance, 9 oral contraceptives (OCs), and breast cancer incidence, 191 oral mucositis managing, 97 orathecin, for pancreatic cancer treatment, 156 orbital pseudotumour, radiotherapy, 451 organs movement during radiotherapy, 46 organs at risk (ORs, OARs), 41 chemoradiotherapy doses for pancreatic tumour, 155 delineating, 42 doses to, 45 in critical analysis, 48 oropharynx anatomy, 106 carcinoma of, 106 five-year survival rate, 108t investigation and staging, 106 radical radiotherapy for, 107 surgery for, 106 TNM summary of staging, 106t orthotopic neobladder, as cystectomy option, 226 osteopenia, from radiotherapy for childhood cancer, 429 osteoporosis from radiotherapy for childhood cancer, 429 risk factors for, 28 osteosarcomas, 335. See also bone tumours in children, 438 outcome measure choosing for research, 58 ovarian cancer paclitaxel and carboplatin for treating, 3 ovaries, 260 borderline tumours, 264 cancer, 257 incidence and epidemiology, 257 epithelial-stromal tumours, 257–264 areas of current interest, 263 chemotherapy, 261–262 classification, 257 clinical presentation, 259 clinical trials, 264 fertility sparing treatment, 260
hormone therapy, 263 investigation and staging, 259 palliative care, 263 pathology, 258 prognosis, 263 recurrent disease, 262 risk factors and aetiology, 257 screening and prevention, 259 spread, 258 surgery, 260 treatment overview, 259 function suppression, 26 germ cell tumours, 265 classification, 257 granulosa cell tumours, 264 pseudomyxoma peritonei, 264 radiotherapy impact in children, 430 sex cord-stromal tumours, classification, 257 tumour types, WHO classification, 257 over-compliance with oral chemotherapy, 9 oxaliplatin, 12 for cholangiocarcinoma palliative care, 148 for colorectal cancer (CRC), 164, 173t for palliative care, 169 for liver metastases, 169 oesophageal-pharangeal syndrome from, 9 for pancreatic cancer treatment, 156 reaction from, 81t oxycodone, 85t paclitaxel, anaphylaxis from, 79 antidotes for extravasation, 81t for breast cancer treatment, 202 for cervical cancer treatment, palliative care, 286 for endometrial carcinoma, 273 for epithelial-stromal tumours, trials, 261–262 for non-small cell lung cancer, 320 for ovarian cancer treatment, 3, 261, 262 for recurrent disease, 262 summary of trials, 261 peripheral neuropathy from, 263 reaction from, 81t paediatrics. See childhood cancer Paget’s disease, 208, 341 prognosis, 344 pain from lung cancer, 316 in breast, 34 pain control, 83 non-pharmacological methods, 84, 86t pharmacological methods, 83
Index
palladium-103 properties, 51 palliative care for anal cancer, 180 anorexia, 87 anti-emetic prescribing, 84 areas of current interest, 91 ascites, 88 bone tumours, 344 changing model, 83 changing role of, 83 chemotherapy, 1 for bile tract cancer, 148 for bladder cancer, 228 for cervical cancer, 286 for colorectal cancer, 169 for gastric cancer, 137 for head and neck tumours, 99 for liver metastases from colorectal cancer, 168 for metastatic or inoperable pancreatic tumour, 156 for penile cancer, 256 communication issues, 89–91 speed of information sharing, 90 constipation, 85 diarrhoea, 87 doxorubicin for liver cancer, 144 dysphagia, 87 for epithelial-stromal ovarian cancers, 263 ethical decision making, 91 financial considerations, 91 last 48 hours of life, 89 for lung cancer, 313, 323 for mesothelioma, 332 radiotherapy, 332 for nasopharynx tumours, 113 for non-small cell lung cancer, 319 radiotherapy, 319 for oesophageal cancer, 129 ongoing research, 92 pain control, 83 non-pharmacological methods, 84, 86t pharmacological methods, 83 for pancreatic cancer, surgery for, 154 pleural effusions, 87 for prostate cancer hormonal therapies for metastatic disease, 238 radiotherapy for metastatic disease, 238 psychological symptoms, 88 anxiety, 88 confusion and delirium, 88 depression, 88 radiotherapy for bladder cancer, 228
for cervical cancer, 286 for gastric cancer, 139 for pancreatic tumour, 156 for rectal cancer, 168 for relapsed testicular cancer, 250 for renal cell cancer cytokine therapy, 218 nephrectomy, 217 radiotherapy, 218 surgery, for advanced rectal cancer, 168 syringe drivers, 89 palmar-plantar erythrodysaesthesia (PPE) from chemotherapy, 8 pamidronate, 207 Pancoast tumour, 322 pancreas anatomy, 151 carcinoma, 151 areas of current interest, 156 chemoradiotherapy for locally advanced disease, 155 chemotherapy and chemoradiation, 154 chemotherapy and chemoradiation for locally-advanced disease, 155 chemotherapy for metastatic or inoperable pancreatic tumour, 156 clinical features, 152 endoscopic relief of obstruction, 156 imaging, 153 incidence and epidemiology, 151 investigation and staging, 153 ongoing/planned trials, 157t pathological features, 152t radiotherapy for palliative care, 156 stage groups, 154t staging classification, 153 surgery for, 154 TNM classification, 154t treatment overview, 153 types, 151, 152t carcinoma of exocrine clinical presentation of, 152 pathology, 151 risk factors and aetiology, 151 neuroendocrine tumours (NETs), 420 chemotherapy, 423 incidence and clinical features, 420t surgery, 422 pancreatic duct, 151 pancreatico-duodenectomy (Whipple’s procedure), 154 Papanicolau (Pap) smear, 279 treatment following, 279t papillary cancers of thyroid, 406 para-aortic field, radiotherapy for lymphoma, 362 parallel organs and radiation therapy planning, 42
paraproteinaemias, classification, 363 parietal pleura, 328 Paris system for brachytherapy dosimetry, 51 parotid gland anatomy, 116 carcinoma of CT-planned radiotherapy treatment, 118 tumours post operative radiotherapy, 117 treatment overview, 117 partial breast irradiation (PBI), 200 patents rights in clinical trials, 63 Paterson-Parker system for brachytherapy, 52 patient communication with, 89–91 patient information sheet, 64 PDGFR tyrosine kinases Imatinib activity against, 15 pegylated G-CSF (Neulasta® ) for neutropenia, 6 pegylated liposomal doxorubicin hydrochloride (PLDH), for ovarian cancer treatment, for recurrent disease, 262 pelvis carcinoma of renal, 220 pain in ovarian cancer, palliative care for, 263 treatment in prostate cancer, 236 pemetrexed for mesothelioma, 332 penis cancer chemotherapy, 256 diagnostic and staging, 252 incidence and epidemiology, 252 pathology, 252 risk factors and aetiology, 252 TNM classification, 253t treatment of carcinoma in situ, 253 treatment overview, 253 invasive squamous cell carcinoma radiotherapy, 254 treatment, 253 management of groin nodes, 254 clinically and radiologically negative groin, 254 clinically involved nodes, 255 management following surgery, 255 tumour types, 253t peptide growth factor receptor pathways in breast cancer cells, 30 peptide-based vaccines, 20 percentage depth dose, 42
485
Index
pericardium, involvement by mesothelioma, 328 perihilar tumours, Bismuth classification for, 146t peripheral carcinoid, 325 peripheral PNETs, in children, 438 peritoneum, involvement by mesothelioma, 328 permuted block randomisation, 57 PET. See positron emission tomography (PET) Peutz-Jeghers syndrome, and breast cancer, 192 Peyronie’s disease, radiotherapy, 451 pharynx carcinoma of posterior wall radical radiotherapy for, 108 Phase I trials, 55 Phase II trials, 56 Phase III trials, 56 Phase IV trials, 56 phlebitis from chemotherapy, 8 photodynamic therapy (PDT) for oesophageal cancer, 129 for skin cancer treatment, 385 photons percentage depth doses fo 195kV, 388t percentage depth doses fo 95kV, 387 PIAF regimen, for liver cancer treatment, 144 PICC line for chemotherapy, 5 pineal tumours, 371 treatment, 375 pineocytoma, treatment, 377 piperacillin for neutropenia, 6 pituitary tumours, 371 radiotherapy plan for adenoma, 376 treatment, 375, 377 placebo and blinding, 59 placental site trophoblast tumour (PSTT), 304, 307 managing, 309 planning organ at risk volume (PRV), 41 planning target volume (PTV), 41 forming, 42 plasma chromogranin A, neuroendocrine tumours (NETs) and, 420 plasmacytoma, solitary, 366 platinum compounds reaction from, 81t platinum-based chemotherapy allergy to, 263 for cervical cancer, 281 for ovarian cancer treatment, 261, 262t for recurrent disease, 262
486
platinums, hypersensitivity and anaphylaxis from, 9 pleura anatomy, 328 tumour types, 328, 329t pleural effusions, 87 in ovarian cancer, palliative care for, 263 pleural mesothelioma areas of current interest, 333 chemotherapy, 332 clinical presentation, 330 clinical trials, 333 compensation, 332 impact, 328 incidence and epidemiology, 328 investigation and staging, 330 palliative care, 332 pathology, 329t prognosis, 332 radiotherapy, 332 risk factors and aetiology, 328 spread, 330t staging classification, 330, 331 surgery, 331 treatment overview, 330 pleuroperitoneal shunt, 332 Plummer-Vinson syndrome, 122 pneumonectomy, in mesothelioma treatment, 331 pneumonitis, bleomycin and, 247 polycystic kidney disease, 214 polycythaemia vera, radiotherapy, 452 polyvinyl chloride monomers, and angiosarcoma of the liver, 145 porocarcinoma, malignant, 392 positron emission tomography (PET) for assessing impact of imatinib in GIST, 187 in Hodgkin lymphoma treatment, 359 in children, 432 for neuroendocrine tumours (NETs), 421 for prostate cancer, 233 postmenopausal patients with breast cancer adjuvant endocrine therapy for, 203 hormone therapy, 205 potassium cisplatin impact on excretion, 5 power, 59 PPE. See palmar-plantar erythrodysaesthesia (PPE) prednisolone, 2, 35 for diffuse large B-cell lymphoma, 351, 352 for Hodgkin lymphoma, 359 in children, 432 for myeloma, 365 for prostate cancer, 31
pregnancy and breast cancer, 207 cervical cancer in, 286 thyroid cancer and, 412 premenopausal patients with breast cancer adjuvant endocrine therapy for, 203 hormone therapy, 205 pre-pubertal age group, cancers typically occurring, 427 primary CNS lymphoma (PCNSL) clinical presentation, 353 investigation, 353 management, 353 primary mediastinal B-cell lymphoma clinical presentation, 353 prognosis, 353 treatment, 353 primitive neuroectodermal tumours (PNETs), in children, 435, 438 procarbazine, 12 for Hodgkin lymphoma, 359 in children, 432 progestagens, for endometrial carcinoma, 273 progesterone receptor (PgR), 25 role in hormone resistance, 30 status measurement in breast cancer, 25 progesterones for renal cell cancer, 219 intrauterine, in endometrial carcinoma treatment, 270 progestins for breast cancer treatment, 29 prophylactic antibiotics for febrile neutropenia, 79 prophylactic cranial irradiation for small cell lung cancer, 322 prostate cancer adenocarcinoma pathological features, 232t TNM classification, 233t androgen deprivation therapy for, 31 areas of current interest, 36, 239 basics, 231 clinical presentation local symptoms, 233 lymph node spread or metastatic disease, 233 clinical trials, 239 hormonal therapies, 31, 237 incidence and epidemiology, 231 investigation and staging, 233 mechanisms of hormone resistance, 36 palliative care for bisphosphonates, 238 hormonal therapies for metastatic disease, 238
Index
radiotherapy for metastatic disease, 238 pathology, 231 prognosis, 239 radical radiotherapy plan for, 48 risk factors and aetiology, 231 screening for, 232 spread, 232 steroidal antiandrogens for, 35 TNM T-staging classification, 233 treatment brachytherapy, 235 external beam radiotherapy (EBRT), 236–237 follow up after radiotherapy, 237 observation alone, 234 overview, 233 post operative adjuvant radiotherapy, 237 surgery, 234 tumour types, 232t prostate gland, anatomy, 231 prostate specific antigen (PSA) as potential vaccine target, 21 after radical prostatectomy, 235 after radiotherapy, 237 screening, 232 surveillance, 234 prostatectomy, radical, 234 proto-oncogene KIT, and gastrointestinal stromal tumours (GIST), 183, 184 pseudomyxoma peritonei, in ovaries, 264 psychological symptoms, 88 anxiety, 88 confusion and delirium, 88 depression, 88 pterygium, radiotherapy, 451 puberty, cancers typically occurring, 427 published research reports assessing, 67 pulmonary neuroendocrine tumours, 420 chemotherapy, 423 surgery, 422 pulsed dose rate (PDR) for brachytherapy, 53 p-value, 59 quality assurance in brachytherapy, 53 in radiotherapy planning, 47 quality of life palliative chemotherapy and, 1 quinolone antibiotic for neutropenia prophylaxis, 6 radial profile of beam, 42 radical prostatectomy, 234 radical radiotherapy hormone therapy as adjunct, 32
radiofrequency ablation (RFA) for liver cancer, 144 for neuroendocrine tumours (NETs), 423 radioisotope therapy for thyroid cancer, 410 radiation protection issues, 411 side effects, 411 radiolabelled antibodies, 19 radionuclide therapy, for neuroendocrine tumours (NETs), 423 radionuclides for brachytherapy properties, 51 radiotherapy, 244, 246 for anal cancer chemotherapy combined with, 178 dose, fractionation and energy, 178 palliative care, 180 side effects, 178, 179 technique, 177 for bladder cancer, 226 applications, 226 beams, 227 doses, 227 palliative care, 228 setup, 226 use of CRT/IMRT, 227 for brain tumours, 376 toxicity, 378 for breast cancer treatment, 199 lymph nodes, 200 after mastectomy, 199 for cancer of unknown primary (CUP), 446 for cervical cancer treatment, 282–286 post operative, 285 for childhood cancer, 428 late consequences, 428 leukaemia, 431 side effects, 428 for CNS tumours, 377, 380 of craniospinal axis, 378 for diffuse large B-cell lymphoma, 352 for endometrial carcinoma, 271–273 post operative, 273 toxicity, 272 for Ewing’s sarcoma, 344 for gastric cancer with chemoradiotherapy, 138–139 for gastrointestinal stromal tumours (GIST), 186 for head and neck tumours, 96–98 concurrent with chemotherapy, 98 fractionation, 98 indications for postoperative, 97 intensity modulated (IMRT), 99 management of acute reactions, 97 nasal cavity, 116 nasopharynx, 112–113 of hypopharynx, 110 tongue, 105, 107, 108
for Hodgkin lymphoma advanced, 360 in children, 432 hormone therapy as adjunct to radical, 32 intensity modulated (IMRT), 47 for invasive squamous cell carcinoma of penis, 254 for liver cancer treatment, 144 for lung cancer, side effects, 323t for lymphomas, involved field, 361 for malignant spinal cord compression, 72 dose calculation, 73 for MALT lymphoma, 357 for mesothelioma, 332 for middle ear carcinoma, 119 for mouth floor cancer, 105 for non-small cell lung cancer, 322 palliative for stage III, 319 palliative for stage IV, 319 for stage III, 319 for stages I and II, 317 for oesophageal cancer dose, 130 palliative care, 129 side effects, 129t for osteosarcoma, 343 for primary CNS lymphoma, 353 for prostate cancer, 236–237 definition of biochemical failure after, 237 dose calculation, 236 palliative care, 238 post operative adjuvant, 237 for rectal carcinoma, 165, 166 for renal cell cancer, 217 for seminoma, 243 strips and doglegs technique, 245 toxicity, 246 for skin cancer, 382, 385 dose calculation, 391 percentage depth doses fo 195kV photons, 388t percentage depth doses fo 95kV photons, 387 for small cell lung cancer prophylactic cranial, 322 thoracic, 321 for soft tissue sarcomas, 338–339 in children, 439 for spinal tumours, 377 for superior vena caval obstruction (SVCO), 73 for thyroid cancer, 411 for treating benign conditions, 449–452 for vaginal carcinoma, 293–294 complications, 293
487
Index
radiotherapy (cont.) for vulval cancer post operative, 299 preoperative, 300 primary for inoperable disease, 300 radiotherapy planning beam arrangements, 43, 44 brachytherapy, 51–53 calculation cycle, 44 crticial analysis, 48 doses to organs at risk (OARs), 45 electron beam isodoses, 49 examples requiring improvement, 48 for gastric cancer, 138 for head and neck tumours, 96 indications for postoperative, 97 imaging the patient, 40 inhomogeneities of density, 45 isodose plan - photons, 42 matching adjacent beams in complex treatments, 45 overview, 39 patient position and immobilisation, 39 quality assurance, 47 surface obliquity target volumes definition, 41–42 verification, 46 raloxifene, 30 and breast cancer prevention, 193 randomisation good practice in, 58 randomised controlled trial (RCT), 55, 56 rasburicase, 80 for lymphomas, 350 receptor tyrosine kinases, 15 recombinant human TSH (rhTSH), for thyroid cancer, 412 recombinant viruses, 21 rectal carcinoma. See also colorectal cancer (CRC) adjuvant chemotherapy for, 165 early-stage, 165 radiotherapy and chemoradiotherapy for, 165 dose, energy, fractionation, 167 patient instructions, 167 post operative, 166 practicalities, 166 preoperative, 166 side effects, 167 treatment for advanced/inoperable, 167 treatment for non-metastatic disease, 165 red cell aplasia, thymomas and, 325 rehabilitation following radical treatment of head and neck tumours, 99 relative malignancy index (RMI), 259
488
renal antigen expression family as potential vaccine target, 21 renal cell cancer, 17, 214. See also kidney IL-2 for, 13 metastatic disease, 14 radical nephrectomy for, 216 risk factors and aetiology, 214 sorafenib for treating, 17 renal pelvis, carcinoma of, 220 reproductive factors, and breast cancer incidence, 190 research in cancer assessing published reports of research, 67 clinical trials, 55–63 ethical considerations, 63–65 health economics, 65 randomised controlled trial (RCT), 55 research networks, 65 synthesising results, 66–67 translational research, 66 research networks, 65 residual para-aortic disease, 248 restenosis, arterial, radiotherapy for, 449 retinoblastoma, in children, 438 retromolar trigone cancer of radical radiotherapy for, 105 Revised European-American Lymphoma (REAL) classification, 347 rhabdomyosarcoma, 336 in children, 439 rituximab (MabThera® ), 2, 19 for diffuse large B-cell lymphoma, 351 for follicular lymphoma, 354 for mantle cell lymphoma, 357 for primary CNS lymphoma, 354 Roach formula, 236 Rockall score, 75, 76 rodent ulcers, 383 Royal Marsden Hospital (RMH) staging system, for testicular cancer, 242 ruthenium-106 properties, 51 SAB (Same As Before) in acute myeloid leukaemia, 57 St. Gallen guidelines, 201 salivary gland, 116–118 anatomy of parotid gland, 116 tumours clinical presentation of, 117 five-year survival rate, 118t investigation and staging, 117 pathology, 116 post operative radiotherapy, 117 surgery for, 117 TNM classification, 117t treatment overview, 117
salpingo-oophorectomy, 270 Same As Before (SAB) in acute myeloid leukaemia, 57 sample size factors impacting, 59 sarcoma. See also soft tissue sarcomas uterine, 275–276 treatment, 275 types, 275 WHO classification, 335 Schwann cell tumours, 183 schwannoma, 371 scleral plaque therapy, for ocular melanoma, 403 scoliosis, from radiotherapy for childhood cancer, 429 screening for breast cancer, 193 for colorectal cancer (CRC), 159, 160 for cutaneous melanoma, 396 for head and neck tumours, 94 after molar pregnancy, 305 selection bias, 56 selective serotonin reuptake inhibitors (SSRIs), 88 seminoma, 240 adjuvant chemotherapy, 244 adjuvant radiotherapy, 243 radiotherapy strips and doglegs technique, 245 residual mass postchemotherapy, 249 spermatocytic, 241, 250 stage II, 244 surveillance, 243 treatment overview, 243 senna for constipation, 85 sentinel lymph node biopsy (SLNB), 198, 399, 402 sepsis neutropenic, 6 septic shock initial management, 78 serial organs and radiation therapy planning, 42 serial tomotherapy, 47 serious adverse events reporting, 64 Sertoli cells, 250 set-up margin (SM), 41 sex cord stromal tumours in ovaries, classification, 257 in testis, 250 S´ezary syndrome, 358 sham surgery in research, 59 shields for radiotherapy for skin cancer treatment, 386 shoulder, pain from lung cancer, 316
Index
shunt, pleuroperitoneal, 332 SIADH (syndrome of inappropriate antidiuretic hormone), 71 side effects of chemotherapy, 1 sigmoidoscopy, for colorectal cancer screening, 161 signal to noise ratio, 59 signal transduction inhibitors, 31 significance level of statistical testing, 59 Significant trial, 6 silence in patient communications, 90 simple randomisation, 57 single beam, 43 single-blind trials, 58 skin changes in breast cancer, 194 reaction to radiotherapy, 97 skin cancer angiosarcoma, 393 basal cell carcinoma, 383 clinical presentation, 383 molecular pathogenesis, 383 basics, 382 dermatofibrosarcoma protuberans, 393 eccrine carcinoma, 392 incidence and epidemiology, 382 Kaposi’s sarcoma, 392 lymphomas, 393 lymphosarcoma, 393 malignant fibrous histiocytoma, 393 malignant porocarcinoma, 392 merkel cell carcinoma, 392 patient examination, 384 risk factors and aetiology, 383 genetic predisposition, 383 squamous cell carcinoma, 384 TNM classification, 384t treatment 5-fluorouracil, 385 imiquimod 5%, 385 photodynamic therapy (PDT), 385 radiotherapy, 385 radiotherapy dose calculation, 391 surgery, 385 tumour types, 382 skin rash from non-steroidal antiandrogens, 34 with anti-EGFR agents, 17 skull, tumours of base treatment, 375 types, 371 small bowel, chemoradiotherapy tolerance dose for pancreatic tumour, 155 small cell carcinoma, of cervix, 287 small cell lung cancer, 315
areas of current interest and clinical trials, 325 good prognosis patients, 321 chemotherapy, 321 prophylactic cranial irradiation, 322 thoracic radiotherapy, 321 intermediate and poor prognosis patients, 322 pathology, 315 prognosis, 324 staging, 317 staging system, 318t treatment, general aspects, 321 smoking as bladder cancer risk factor, 222 statistics, 313 sodium docusate for constipation, 85 sodium retention from DES, 35 soft palate carcinoma of radical radiotherapy for, 108 soft tissue sarcomas, 335 anatomy and pathology, 336 chemotherapy for, 339 in children germ cell tumours, 440 hepatoblastoma, 440 hepatocellular carcinoma, 440 non-rhabdomyosarcomas, 440 rhabdomyosarcoma, 439 clinical presentation, 337 clinical trials, 340, 345 incidence and epidemiology, 335 investigation and staging, 337 metastatic disease, 340 prognosis, 340, 341t radiotherapy, 338–339 recurrent disease, 340 risk factors and aetiology, 335 spread, 336 staging, 337 surgery, 338 TNM classification, 337t treatment overview, 337 tumour types, 336 solitary plasmacytoma, 366 somatostatin analogue therapy, for neuroendocrine tumours (NETs), 422 somatostatin receptor scintigraphy, for neuroendocrine tumours (NETs), 421 sorafenib (Nexavar® ), 16, 17 for renal cell cancer, 220 sperm storage, 8 before mantle radiotherapy, 363 before testicular cancer treatment, 249
spermatocytic seminoma, 241, 250 spinal cord chemoradiotherapy tolerance dose for pancreatic tumour, 155 compression, 70 tumour treatment, 376, 377 spindle cell sarcomas, 341 spine radiotherapy, 377 toxicity, 378t solitary plasmacytoma, 366 spleen, radiotherapy for lymphoma, 362 squamous carcinoma anal cancer as, 175 pathological features, 175t of lung, 315 of penis radiotherapy, 254 treatment, 253 of skin, 382, 384 radiotherapy dose calculation, 391 squamous metaplasia, as bladder cancer risk factor, 222 statutory sick pay (SSP), 91 stem cell transplantation for follicular lymphoma, 356 for mantle cell lymphoma, 357 for myeloma, 365 stents for oesophageal cancer, 129 stereotactic radiotherapy, for CNS tumours, 380 steroidal antiandrogens for prostate cancer, 35 steroidal hormones, 23 Stewart-Treves syndrome, 393 and soft tissue sarcomas, 336 stomach. See also gastric cancer anatomy, 132 neuroendocrine tumours (NETs), surgery, 422 stopping rules for research trials, 65 streptozocin reaction from, 81t structural/obstructive emergencies, 72–77 acute airway obstruction, 74 bleeding, 75 cardiac tamponade, 77 malignant spinal cord and cauda equina compression, 72–73 raised intracranial pressure, 74 superior vena caval obstruction (SVCO), 73–74 urinary obstruction, 76 study protocol, 63
489
Index
subglottis carcinoma of radiotherapy for, 103 TNM stages, 101 treatment, 102 subgroups in research, 61 sunitinib (Sutent® ), 16, 17 for gastrointestinal stromal tumours (GIST), 188 for renal cell cancer, 220 sunlight, and cutaneous melanoma risk, 395 sunscreen, 396 superficial spreading melanoma (SSM), 396 superior vena caval obstruction (SVCO), 73–74 lung cancer and, 316 supraglottis carcinoma of radiotherapy for, 102 TNM stages, 101 treatment, 102 surface epithelial-stromal tumours, in ovaries, 257 surveillance for prostate cancer, 234 for seminoma, 243 for testicular cancer relapse, 246 suspected unexpected serious adverse reactions (SUSAR), 64 Sutent® (Sunitinib), 17 sweats as ADT adverse effect, 34 Swedish Rectal Cancer Trial, 166 syndrome of inappropriate antidiuretic hormone (SIADH), 71 synovial sarcoma, 336 synovitis, radiotherapy, 451 syringe drivers, 89 talc pleurodesis, 331 tamoxifen, 25, 30 vs. aromatase inhibitors (AIs), 29 and breast cancer prevention, 193 for breast cancer treatment, 26, 27, 203 ductal carcinoma in situ (DCIS), 196 IBCSG trials, 26 for liver cancer treatment, 145 for ovarian cancer treatment, 263 and radiation therapy, for breast cancer treatment, 199 randomised trials, 66 for renal cell cancer, palliative care, 219 toxicity data, 28 Tarceva® (erlotinib). See erlotinib (Tarceva® )
490
taxanes, for non-small cell lung cancer, 320 hypersensitivity and anaphylaxis from, 9 T-cell lymphoma in children, 432 peripheral, risk of relapse, 350 teenagers. See adolescents teeth, radiotherapy impact in children, 430 tegafur-uracil (Uftoral® ), in chemotherapy for colorectal cancer, 164 temozolamide, for advanced cutaneous melanoma, 400 temsirolimus, for renal cell cancer, 220 teratoma, 240 ovarian, 265 testicular British and American classification systems, 242 residual mass postchemotherapy, 248 stage I treatment, 246 treatment overview, 243 ‘terminal care’ service vs. palliative care, 83 testes, radiotherapy impact in children, 430 testicular cancer basics, 240 BEP regimen, 2 clinical trials, 250 flat dosing of bleomycin for, 4 germ cell tumours clinical presentation, 241 incidence and epidemiology, 240 investigation and staging, 241 pathology and classification, 240 risk factors and aetiology, 240 screening for, 240 spread, 241 intratubular germ cell neoplasia (IGCN), managing, 242 metastatic disease, 247–249 relapsed disease, 249 high dose chemotherapy (HDCT), 249 palliative care, 250 seminoma adjuvant chemotherapy, 244 adjuvant radiotherapy, 243 radiotherapy strips and doglegs technique, 245 residual mass postchemotherapy, 249 spermatocytic, 250 stage II, 244 surveillance, 243 treatment overview, 243 sex cord stromal tumours, 250 staging/prognostic grouping, 242 teratoma British and American classification systems, 242
residual mass postchemotherapy, 248 stage I treatment, 246 treatment overview, 243 treatment protocol importance, 1 tumour types, 241t testosterone, 23 preventing rise in, 32 TGF-β receptor II as potential vaccine target, 21 thalidomide for mantle cell lymphoma, 357 for myeloma, 365 thoracic radiotherapy, for small cell lung cancer, 321 thorotrast, 145, 214 thrombocytopenia chemotherapy and, 5 thromboembolic disease from steroidal antiandrogens, 35 thymidine phosphorylase, 2 thymoma, 325 thyroglobulin, 413 thyroid anatomy, 406 thyroid cancer anaplastic external beam radiotherapy (EBRT), 415 incidence and epidemiology, 415 prognosis, 415 spread, 415 treatment, 415 basics, 406 clinical trials, 416 differentiated areas of current interest, 413 chemotherapy, 412 clinical presentation, 408 external beam radiotherapy (EBRT), 411 follow up, 413 incidence and epidemiology, 408 investigation and staging, 409 prognosis, 413 radioisotope therapy, 410 risk factors and aetiology, 408 serum thyroglobulin, 409 surgery, 409 thyroxine for treating, 410 treatment overview, 409 lymphomas, 416 medullary clinical presentation, 414 external beam radiotherapy (EBRT), 414 follow up, 415 incidence and epidemiology, 414 prognosis, 415 risk factors and aetiology, 414
Index
spread, 414 treatment, 414 treatment of recurrent, 415 pathology, 406 screening, 407 special clinical situations, 412 stage classification, 407 stage groupings, 408t TNM classification, 408 treatment for recurrent, 412 tumour types, 406, 407t thyroid eye disease, radiotherapy for, 450 thyrotoxicosis, radiotherapy, 452 thyroxine, for thyroid cancer, 410 tipafarnib, for pancreatic cancer treatment, 157 tiredness as ADT adverse effect, 34 tissue leakage of intravenous drugs from vein into, 80 tissue maximum ratio (TMR), 42 tissue phantom ratio (TPR), 42 tissue samples consent for, 64 tobacco use as bladder cancer risk factor, 222, 226 statistics, 313 toes clubbing, 316 tongue carcinoma of radical radiotherapy for, 105, 107, 108 surgery for, 106 MRI of, 103 tonsil carcinoma of radiation fields for postoperative therapy, 107 surgery for, 106 topoisomerase 1 inhibitors, topotecan, 12 for endometrial carcinoma, 274 for ovarian cancer treatment, for recurrent disease, 262 reaction from, 81t total body irradiation (TBI), 366 for childhood cancer leukaemia, 431 long-term side effects, 429, 430 total mesorectal excision (TME), 167 total skin electrons (TSE) beam therapy, for mycosis fungoides, 358 trachea, 314 blockage of, 74 tramadol, 85t transaminases evaluating before chemotherapy, 5
trans-arterial chemo-embolisation (TACE), for liver cancer treatment, 144 transitional cell carcinomas (TCCs), 223 in prostate, 232 superficial, 224 translational research, 66 transplantation of liver, 143 transrectal ultrasound (TRUS) guided systematic sampling, for prostate cancer, 233 transurethral resection (TURBT), 224 trastuzumab (Herceptin® ), 3, 18 for breast cancer treatment, 3, 18, 204 clinical trials summary, 205 metastatic, 206 treated volume, 41 treatment stages of development, 55 treatment-related emergencies anaphylaxis related to anticancer drugs, 79 extravasation of chemotherapy, 80 neutropenic fever and sepsis, 77 tumour lysis syndrome (TLS), 79 trials. See clinical trials Trojani grading system, 336, 337t trophoblast disease chemotherapy, 309t investigation, 307 true smooth muscle tumours, 183 TSE (total skin electrons) beam therapy, 358 tuberous sclerosis, 214 and soft tissue sarcomas, 336 tumour baseline assessment, 5 tumour lysis syndrome (TLS), 79 tumour markers in cancer of unknown primary, 444 for testicular cancer, 241 measuring before chemotherapy, 5 tumour necrosis factor a, 15 tumour-associated antigens, 20 as potential vaccine targets, 21 Turcot’s syndrome, 160 tyrosine kinase inhibitors, 15 for renal cell cancer, 220 summary of clinical use and trials, 16 vasular endothelial growth factor receptor, 17 tyrosine kinases, 15 UK MRC Guidelines, 64 ulcerative colitis (UC), and colorectal cancer, 160 ultraviolet radiation, and skin cancer, 382 United Kingdom Children’s Cancer Study Group (UKCCSG), 426
registrations for children 15 and under in 2002, 427t urethra, indications for removal in radical cystectomy, 226 urinary obstruction, 76 urothelial tumours, 223 uterus. See also endometrial carcinoma cancer, incidence and epidemiology, 267 radiotherapy impact in children, 430 sarcoma, 275–276 treatment, 275 types, 275 tumour types, 267, 268t vaccines, 20 for melanoma, 402 after radical nephrectomy, 218 vagina anatomy, 290 carcinoma, 290 areas of current interest, 294 clinical presentation, 291 clinical trials, 294 histological features of squamous, 292t incidence and epidemiology, 290 investigation and staging, 291 palliative care, 294 pathology, 290 prognosis, 294 recurrent disease, 294 risk factors and aetiology, 290 spread, 290 staging classification, 291, 292t carcinoma treatment follow up, 294 overview, 292 radiotherapy, 293–294 surgery, 292 tumour types, 290, 291 vaginal intraepithelial neoplasia (VAIN), 290 vaginal trachelectomy, for cervical cancer treatment, 282 vancomycin for neutropenia, 6 vascular endothelial growth factor (VEGF), 159 vatalanib, 16 verrucous tumours, 252 video assisted thoracoscopic surgery (VATS), 331 vinblastine, 11 for Hodgkin lymphoma, 359 in children, 432 vinca alkaloids, antidotes for extravasation, 81t fatalities from, 8
491
Index
vinca alkaloids (cont.) phlebitis from, 8 reaction from, 81t vincristine, 2, 11 dose capping, 4 for diffuse large B-cell lymphoma, 351 for Ewing’s sarcoma, 343 for gestational trophoblast tumours (GTT), 310 for Hodgkin lymphoma, 359 in children, 432 for mantle cell lymphoma, 357 for metastatic soft tissue sarcoma, 340 vindesine, for diffuse large B-cell lymphoma, 352 vinorelbine, 12 for non-small cell lung cancer, 320 viral-protein based vaccines, 20 visceral pleura, 328 Vitamin D, and protection from melanoma, 396 Vogelstein’s model of carcinogenesis, 161 vomiting. See nausea and vomiting Von Hippel-Lindau disease, 214 vulva anatomy, 296 carcinoma areas of current interest, 301 chemoradiotherapy toxicity, 300 chemotherapy, 300
492
clinical presentation, 297 clinical trials, 302 five-year survival rate, 302t investigation and staging, 297 palliative care, 301 pathological features of squamous, 298t pathology, 296 post operative radiotherapy, 299 preoperative radiotherapy or chemoradiotherapy, 300 prognosis, 301 recurrent disease, 301 risk factors and aetiology, 296 spread, 296 staging classification, 297 surgery, 298 TNM classification, 298t treatment overview, 297 incidence and epidemiology, 296 tumour types, 296, 297 Waldenstr¨om’s macroglobulinaemia, 363 Waldeyer’s Ring, radiotherapy for lymphoma, 361 watchful waiting, for prostate cancer, 234 web resources BC Cancer Agency, 1
Medicines Compendium, 13 North London Cancer Network, 4 wedges, 46 for beam arrangement, 43, 44 weight checking before chemotherapy, 5 weight gain as ADT adverse effect, 34 from non-steroidal antiandrogens, 34 Wilm’s tumour, in children, 437 World Health Organisation analgesic ladder, 83, 84 pain ladder, 70 palliative care, defined, 83 tumour classification for CNS, 371, 372t for ovaries, 257 tumour classification for nasopharynx, 110 Xeloda® (capecitabine). See capecitabine (Xeloda® ) Xeroderma pigmentosa, 383 xerostomia, 97 yolk sac tumour, 265 zactima, 16 zoledronic acid, 31, 71 for bone metastases, 207